Is Galatians an Ironic Letter? Ancient Letter Writing Handbooks, and Galatians 1:6

This article queries whether Paul wrote Galatians with reference to epistolary conventions for ironic letters. First, the author explores the use of the θαυμάζω + conjunction “epistolary formula” in the non-literary papyri to determine the relationship between this expression, irony, and Gal 1:6. Then, he weighs the evidence for an ironic reading of Gal 1:6 itself before turning to the extant ancient letter writing handbooks to assess the extent to which Gal 1:6 meaningfully parallels the ironic letters in the handbooks. The author argues that while an ironic reading of Gal 1:6 is plausible, there is no evidence that Paul has crafted Galatians with reference to epistolary conventions for ironic letters.

Ever since Betz's seminal commentary on the letter, ancient rhetorical theory has featured significantly in scholarship on Galatians. While forensic rhetoric has long held pride of place in these discussions, in more recent years epistolary theory has come to occupy an important position in rhetorically-informed research on Galatians.1 This work has called rightful attention to how Galatians functions as an actual letter, arguing that the epistolary genre has a role to while Galatians is certainly a letter that contains irony, there is no evidence that Paul crafted its opening with reference to ancient epistolary theory.

The θαυμάζω Formula in Scholarship
The fact that Paul does not open Galatians with his usual thanksgiving but an expression of frustration at his congregation for turning aside from their calling has been much remarked on in scholarship.7 Dahl considers Paul to have exchanged his standard expression of thanksgiving for the ironic rebuke formula.8 While the Galatians would not have had other Pauline letters for comparison-and would probably not have been expecting a thanksgiving section in the first place9-by the end of 1:6 they could not have missed the intensity of Paul's tone.10 In this section we will assess the extent to which the θαυμάζω "ironic rebuke" formula in Gal 1:6 actually conveys irony before going on to address whether it can meaningfully be considered an epistolary formula.
White, citing four ancient examples, argues that θαυμάζω followed by ὅτι or πῶς is a typical epistolary introduction that points out a failing of the addressee, most typically a failure to write to the sender.11 Adding, by my count, six new examples to the discussion, Mullins argues that θαυμάζω ὅτι/πῶς can only be considered a true epistolary convention when it is being used ironically or sarcastically to suggest a tone of wonder and surprise that is not actually there. When used in this fashion, the clause expresses frustration while "rebuking, even scolding, the addressee."12 Without these elements, there is no formula, only an indicative verb followed by a conjunction: "If the verb is used in the normal sense, the form is not there. If the verb is used in an ironical sense to scold, then the form may be there."13 Although we owe the coinage "ironic rebuke" to Mullins, it is worth noting that he is not entirely convinced that 7 For a summary, see R. Van Voorst, "Why Is There No Thanksgiving Period in Galatians? An Assessment of an Exegetical Commonplace," jbl 129.1 (2010) 154-159. 8 Dahl, "Genre," 117-118. 9 Van Voorst, "Thanksgiving Period," 160-166. 10 Moo writes concerning 1:6: "it seems justified to conclude, with most commentators, that Paul is signaling his extreme distress at the situation of the Galatian Christians" (D.J. Moo, Galatians [3 vols.; becnt; Grand Rapids: Baker Academic, 2013] 1:75). 11 White, "Formluae," 94, 96. 12 Mullins, "Formulas," 385-387. 13 Mullins, "Formulas," 385.

Pawlak
Novum Testamentum 63 (2021) 249-270 θαυμάζω ὅτι/πῶς should be considered a proper epistolary formula: "I suggest that if this is accepted as a form, it should be called ironic rebuke."14 More than a decade later, G.W. Hansen takes up the subject of θαυμάζω as an epistolary convention.15 He departs from Mullins in at least one significant way. Hansen states that, when it comes to θαυμάζω, Mullins's "denial of surprise or astonishment in the term is an overstatement" and argues that some uses indicate both genuine surprise and irritation.16 Where this surprise is not feigned, there is no irony, and therefore Hansen prefers the term "astonishment-rebuke" to "ironic rebuke." 17 We see a jump in the number of examples cited with J. Robert's work in the early 1990s. Roberts explicitly sets out to argue against an ironic reading of θαυμάζω in Gal 1:6, preferring to see the verse as an expression of perplexity.18 To this end, he marshals 25 examples of θαυμάζω plus conjunction in the papyri, ultimately arguing that while irony is possible in such expressions, it is not inherent to the form and its presence must be established from context.19 Roberts identifies several potential meanings of θαυμάζω, finding perplexity to be the most common.20 D. Kremendahl's survey is the most systematic to date, overviewing 30 uses of θαυμάζω ὅτι/πῶς/εἰ occurring in the introductions of papyrus letters. Kremendahl overturns White's observation that this opener is usually a comment on the addressee's failure to write, as this occurs in only 11/30 cases.21 Ultimately, for Kremendahl, θαυμάζω ὅτι/πῶς/εἰ is a way of signalling rebuke, but not direct rebuke. The speaker's comment about their wonder or confusion in the first person implies a second person complaint ("Ich wundere mich, daß" implies, "Du hast in der und der Sache gefehlt" With θαυμάζω ὅτι/πῶς/εἰ sometimes ironic but more often unironic in the papyri, the use of θαυμάζω ὅτι in Gal 1:6 does not give the reader any information regarding whether the verse is ironic, or whether the letter has been written in accordance with ancient epistolary conventions for ironic letters. With the papyri essentially irrelevant to the issue of Galatians as an ironic letter, we must turn to Gal 1:6 itself for evidence of irony. Following an exegesis of the text, we may then assess the extent to which what Paul is doing in Galatians has parallels in ancient letter writing handbooks. Before moving on, however, I would like to push back on an assumption underlying scholarship on θαυμάζω ὅτι/πῶς/εἰ, in order to promote further critical thinking about the relationship between Gal 1:6 and the papyri. and his congregation. If we, still allowing a considerable window, constrict our search only to the first two centuries bce and ce, we are left with 8-11 instantiations of this so-called epistolary formula.31 Comparison with greeting formulae, which perhaps more than any other feature deserve the title epistolary formulae in its most technical sense, can help refine our understanding of how strong the evidence for θαυμάζω ὅτι/πῶς/ εἰ as a proper convention is. Between 200 bce and 200 ce, there are thousands of ancient papyrus letters extant.32 A ratio of circa 10 instantiations of θαυμάζω ὅτι/πῶς/εἰ out of 3747 letters should already suggest caution in using the technical designation "epistolary formula" to refer to an indicative verb plus conjunction that occurs so infrequently. To add further perspective, a search for χαίρειν between these dates in papyrus.info turns up 4854 hits. These are not all letters, but include contracts, receipts, and other types of writing. ἀσπάζομαι/ ἀσπάζεται occur 364 times within this date-range and, unlike χαίρειν, occur predominantly in letters.
Of course, one must expect greeting formulae to be more prevalent than other conventions that deserve the designation epistolary formulae in their own right. However, only about 10 examples within a timeframe relevant to Paul's historic and linguistic context are too few to suggest that the use of θαυμάζω ὅτι/πῶς/εἰ in letter writing was formalised enough to merit reference as an "epistolary formula." Pragmatically speaking, this means that we cannot assume that it would be recognizable to an audience as a common convention or turn of phrase in letter writing with stereotyped features. We cannot assume that Galatian Christians would hear θαυμάζω ὅτι and immediately recognise 31 Nine letters date to the first and second centuries ce, and three could be second or third century ce-for a maximum Paul as expressing perplexity, rebuke, or whatever. This does not mean that Galatian Christians would hear θαυμάζω ὅτι with no preconceptions at all. θαυμάζω is a common, versatile verb capable of being used positively, neutrally, and negatively,33 literally and ironically.34 Doubtless Paul's audience in Galatians would have used or heard θαυμάζω used in all of these ways. It is therefore exegetically inappropriate to use the papyri as an interpretive key for understanding Gal 1:6. Without sufficient evidence for θαυμάζω ὅτι/ πῶς/εἰ as an epistolary formula, we cannot assert that Paul's audience would be biased towards a specific interpretation of θαυμάζω in Gal 1:6 just because Paul uses the verb in the context of a letter. This is not to say that comparison between Paul and the papyri is not worth making, only that Paul's use of θαυμάζω must be established in its own right. Then, when compared to other ancient letters, the interpreter must be prepared to explain the presence of difference as much as similarity.
With θαυμάζω ὅτι/πῶς/εἰ neither inherently ironic nor an established epistolary formula, our determination of whether Galatians opens ironically must be made on the basis of Galatians itself rather than on the papyri. We turn to this analysis now, with an aim to lay out the evidence either for or against an ironic reading of Gal 1:6. To facilitate our exegesis, and because we are interested in assessing evidence that Galatians has been crafted as an ironic letter, it will be necessary to first lay out briefly how ancient authors conceived of irony at the time.

Irony in the First Century ce: Dispraise through Fake-Praise
In discussing the presence of irony in θαυμάζω ὅτι/πῶς/εἰ, Kremendahl and Roberts define irony as saying one thing and meaning something elsegenerally the opposite.35 H. Colston has recently called this a "lay account" of irony, which has long since been surpassed in modern scholarship.36 In more 33 Kremendahl, Botschaft, 103. See also lsj, s.v. θαυμάζω. 34 For ironic examples, see Pawlak, "Sarcastic," 559 n. 27. 35 Kremendahl, Botschaft, 104; Roberts, "θαυμαζω," 116-117. Helpfully, Nanos cites a selection of ancient texts discussing irony (Irony, 35-36). 36 For a brief discussion of problems with this lay account, see H.L. Colston, "Irony as Indirectness Cross-Linguistically: On the Scope of Generic Mechanisms," in Indirect Reports and Pragmatics in the World Languages (ed. A. Capone, M. García-Carpintero, and recent years a number of theoretically sophisticated paradigms have been advanced by scholars to explain irony in its different forms.37 Fortunately, because the first century is our target, it will not be necessary to review the last half-century of theoretical work on irony-although it is very helpful and biblical scholars interested in the subject of irony certainly should. Instead, we will use ancient treatments of irony to ground our discussion in terminology relevant to Paul's day. Writing around the first century bce, the grammarian Tryphon-or someone writing in his name-describes irony as follows: "Irony is a stylistic device that uses what is expressed literally to hint at an oppositional meaning, with pretence" (Εἰρωνεία ἐστι φράσις τοῖς ῥητῶς λεγο<μένοις αἰνιττο>μένη τοὐναντίον μεθ' ὑποκρίσεως, [Greg. Cor.] Trop. 15).38 This terse definition gives the reader three important pieces of information about irony. (1) Irony expresses something that conflicts with the speaker's literal utterance (τοῖς ῥητῶς λεγομένοις αἰνιττομένη τοὐναντίον).
(2) It does so implicitly, through some sort of hinting (αἰνιττομένη), (3) and pretending (μεθ' ὑποκρίσεως). These three features recur in other, roughly contemporaneous ancient definitions as well.39 Ancient authors also subdivided irony into different species, enabling us to be more specific about what sort of εἰρωνεία we are looking for in Gal 1:6. These subcategories included sarcasm (σαρκασμός), self-deprecating irony (ἀστεϊσμός), and different forms of mockery (μυκτηρισμός and χλευασμός).40 If Paul's use of θαυμάζω in Gal 1:6 is ironic, it is so because it uses ostensible surprise/wonderment (positive affect) to express rebuke (negative affect). This fits best with ancient definitions of sarcasm. Tryphon's definitions of sarcasm and irony differ by only two Greek words. The elements of contrast and implicitness are common to both, but while irony is delivered "with pretence" (μεθ ὑποκρίσεως), sarcasm is spoken "with mockery" (μετὰ χλευασμοῦ, [Greg. Quintilian's master work on rhetoric, Institutio Oratoria, can reinforce and extend this understanding of ancient irony and sarcasm. Quintilian situates irony as a subcategory of allegory: "On the other hand, that class of allegory in which the meaning is contrary to that suggested by the words, involve an element of irony, or, as our rhetoricians call it, illusio" (eo vero genere quo contraria ostenduntur, ironia est; illusionem vocant, Inst. 8.6.53-54 [Butler, lcl]; cf. Inst. 8.6.44).
It is important that we do not read Quintilian's "contrary meaning" (contraria) or Tryphon's "oppositional message" (τοὐναντίον) too literally, as I have sought do by avoiding the more restrictive translation, "the opposite." The appropriateness of this softer notion of contrast becomes clear as Quintilian moves on to define irony's subspecies. For Quintilian, sarcasm requires nothing more than "censur[ing] with counterfeited praise" (laudis adsimulatione detrahere) or "disguis[ing] bitter taunts in gentle words" (tristia dicamus mollioribus verbis, Inst. 8.6.55, 57, respectively [Butler, lcl]). Such sentiments are certainly oppositional to the literal message, but not necessarily its opposite.
Lay definitions, which conceive of irony as saying one thing and meaning the opposite, therefore, are misleading from both an ancient and modern standpoint. Ancient definitions are neither that restrictive nor that specific. The evidence we have suggests that in the first century, irony was generally considered to occur when a speaker pretends to express one evaluation, but ultimately implies a different evaluation. What is contrasted is not information, but affect (praise versus dispraise). Whether the speaker's utterance is true or false is not determinative for ironic interpretation. Since we know that in exegeting Gal 1:6 we are looking for a specific type of irony, σαρκασμός, we can be even more specific. In sarcasm the literal affect is always positive, but the actual evaluation is negative. Thus, the exegetical question is whether Paul intends dispraise through insincere praise, or as Rhetorica Anonyma puts it: 41 It is best to view this difference as additive. It is not that Tryphon considers sarcasm to lack pretence, but to involve mockery in addition to ὑπόκρισις. "expresses dishonour through kind words" (διὰ χρηστῶν ῥημάτων τὴν ἀτιμίαν ἐμφαίνων, Trop. 23).43

2.2
Irony and Gal 1:6 We now explore evidence of irony in Gal 1:6 itself.44 Here we are not concerned with whether Paul really is or is not surprised, that is, whether the statement is true or false.45 Instead, we are looking for evidence of dispraise through ostensible praise. Multiple lines of evidence support an ironic reading of Gal 1:6, although Paul's use of ambiguous language makes this interpretation ultimately uncertain.

2.2.1
An Ironic Reading Exaggeration is a common way for speakers to indicate insincerity, and thereby functions as a common signal of irony in both modern English and ancient Greek.46 The opening of Gal 1:6 is certainly emphatic. Paul begins: θαυμάζω ὅτι οὕτως ταχέως. θαυμάζω can indicate amazement, fascination, and awe at something wonderful that has just transpired.47 The double adverbs οὕτως ταχέως also create a sense of emphasis, which is underlined by their placement early in the clause, just after the ὅτι.48 Paul expresses amazement at the Galatians celerity,49 and when we find out later in the sentence that this is nothing less than the blinding speed with which they have turned from God, it is certainly plausible to read a measure of irony in Paul's expression of awe. 43 This understanding of sarcasm also accords well with the ancient letter writing handbooks, as we shall see in § §3.2-3 below. 44 We begin our exegesis with 1:6, as I do not see Paul as having provided significant indication of his tone in 1:1-5. Contra Van Voorst, who argues that in 1:1-5: "Paul has been implicitly but clearly critical of the Galatians in this prescript" ("Thanksgiving Period," 171, 166-72). Paul's apostolic self-description in 1:1 does not imply that the Galatians have done anything wrong. Likewise, the benediction and doxological aside in 1:3-4 are mildly positive where they touch on the Galatians at all-Paul wishes χάρις and εἰρήνη upon them (1:3) and they have been set free from "the present evil age" (1:4 nrsv). There is no reason why the Galatians should have been biased to expect criticism in 1:6, or praise for that matter. 45 Roberts sees "genuine perplexity" ("Perplexity," 337 The strongest evidence for an ironic reading of Gal 1:6 lies in the element of pretence-ὑπόκρισις as Tryphon would put it-in the verse as a whole. An example of sarcastic encouragement from Mark's Gospel can provide a helpful illustration. In a dispute over handwashing, Jesus tells the Pharisees and scribes, "You do a good job of rejecting God's command so you can hold onto your tradition" (καλῶς ἀθετεῖτε τὴν ἐντολὴν τοῦ θεοῦ, ἵνα τὴν παράδοσιν ὑμῶν στήσητε, Mark 7:9). Although the audience knows almost instantly that Jesus does not mean to praise his opponents, his comment is still clearly sarcastic. With his ostensibly positive congratulations (καλῶς), Jesus creates the sort of "counterfeited praise" discussed in Quintilian. This contrasts sharply with the obvious dispraise he means to communicate (ἀθετεῖτε τὴν ἐντολὴν τοῦ θεοῦ). The absurdity of this juxtaposition makes his irony obvious; in a discussion about keeping the law, Jesus tells others that they have done well in breaking it! The same sort of ironic congratulations for doing something obviously wrong would also be present in an ironic reading of Gal 1:6. Here Paul pretends to complement the Galatians, "amazed" at how quickly they have defected from their calling to an entirely different gospel: "I marvel at just how quickly you've managed to abandon the one who called you in the grace of Christ for a different gospel!" This pretence (ὑπόκρισις) would be very thin, the dispraise and rebuke implied being far and away the dominant sentiments communicated.
In addition to this sarcastic congratulations, there may be further irony later in the verse as well. Paul writes that the Galatians are turning to "a different gospel, which is not another" (εἰς ἕτερον εὐαγγέλιον, ὃ οὐκ ἔστιν ἄλλο, Gal 1:6-7a). Some scholars argue that Paul uses "gospel" ironically here, an inversion of its normal positive sense.50 In a recent study on the communication of sarcasm in ancient Greek, I found that following an ostensibly positive message with a conflicting statement that expresses the speaker's real attitude is one of the most common signals of sarcasm in ancient Greek texts, occurring in 26 percent of 340 examples of sarcasm across Lucian and the New Testament.51 This is precisely what we see in Galatians. ὃ οὐκ ἔστιν ἄλλο immediately undercuts ἕτερον εὐαγγέλιον, clarifying that Paul does not mean "gospel" in a positive sense, but intends to communicate a negative evaluation of his opponents' message, which distorts the gospel (μεταστρέψαι τὸ εὐαγγέλιον τοῦ Χριστοῦ, 1:7) and is worthy of anathema (1:8-9). Therefore, to imply such strong negative affect through the positive term "gospel" is textbook ancient sarcasm, well captured by Rhetorica Anonyma's "dishonour through kind words" (Trop. 23).
Recognizing this typical means of communicating sarcasm supports the exegesis of scholars who see ἕτερον and ἄλλο here as essentially synonymous.52 Schröter objects, arguing that if ἕτερον and ἄλλο are synonymous, ἄλλο becomes unnecessary. He also sees it as a problem that they "stand in syntactical opposition to each other here," and that with the ἄλλο clause Paul immediately retracts his initial statement that his opponents' message is a gospel.53 Such repetition and opposition do not however indicate inconsistency on Paul's part or create an exegetical problem. Instead, they constitute the common signal of sarcasm discussed above in which the sarcastic statement is juxtaposed with a literal, negative message.
By way of clarification, we should also note that an ironic reading does not necessitate the interpretation that Paul means the opposite of what he says. Kremendahl writes that for Paul, his opponents' message "gerade das Gegenteil eines Evangeliums ist," and therefore Paul's ὃ οὐκ ἔστιν ἄλλο makes it clear that "der Terminus εὐαγγέλιον in 1,6 nicht im eigentlichen Wortsinn gemeint gewesen sein kann."54 This is close to what I have argued, but it is worth being clear about the distinction between meaning and evaluation in irony. Since, as we saw above ( §2.1), irony inverts ostensible affect rather than meaning, an ironic reading of εὐαγγέλιον does not commit the interpreter to a stance on whether Paul admits the possibility of other legitimate gospels,55 or whether Paul's opponents were in the practice of referring to their message as a gospel.56 Such exegetical questions must be decided on other grounds. What is essential to an ironic reading is oppositional evaluation. Paul negatively evaluates his opponent's message-whatever they were in the habit of calling it-regardless of whether he really means "false-gospel" or "misused (legitimate) gospel."

2.2.2
An Unironic Reading Although multiple lines of evidence support the plausibility of an ironic reading of Gal 1:6, we must also acknowledge a level of ambiguity. θαυμάζω is a polyvalent term, and it is often difficult to determine whether its positive sense is being inverted ironically, or whether it is being used literally in a negative sense. A literal, negative reading of θαυμάζω remains a possibility, communicating emphatic rebuke rather than sarcastic congratulations: "I'm shocked that you have so quickly abandoned …" It is also possible that Paul's addition of ἕτερον to εὐαγγέλιον is itself meant to convey negative evaluation, leaving no "counterfeited praise" or "kind words" to invert ironically. At the same time, a straightforwardly negative, unironic reading of 1:6 leaves little reason for Paul to qualify what he says so immediately (ὃ οὐκ ἔστιν ἄλλο, 1:7a). As we have seen, this sort of immediately contrastive statement is very common in ancient irony and would be somewhat out of place here if there was no element of ostensible praise or positive affect to undercut in either Paul's θαυμάζω ὅτι opening or use of εὐαγγέλιον. Ultimately, the difference between the ironic "I marvel that …" and the straightforwardly negative "I am shocked that …"-as well whether "gospel" belongs in inverted commas-would come down to tone and delivery. Without access to these elements, we must speak in terms of plausibility rather than certainty.

2.3
Conclusion: Galatians 1:6 Compared to the Papyri Having established that an ironic reading of Gal 1:6 is both coherent and plausible, if uncertain, it remains to determine the extent to which what Paul is doing in Galatians meaningfully parallels other uses of θαυμάζω ὅτι/πῶς/εἰ in the papyri-I am intentionally avoiding the term "epistolary formula." The strongest overlap here is the presence of rebuke. While this is a meaningful connection, there remains considerable variation in both the degree and subject matter of the rebuke in Gal 1:6 compared to the papyri. 11 of the 30 instances of θαυμάζω ὅτι/πῶς/εἰ that Kremendahl cites are reproaches for failing to write the sender; these also occur in four The degree of rebuke also sets Galatians apart from many of these other ancient letters. Kremendahl aptly observes that θαυμάζω ὅτι/πῶς/εἰ often makes the speaker's reproach less direct and more polite (recall P.Mich. 8.500 in §1.1.1 above; see also n. 24).59 Regardless of whether one reads it ironically or 57 See Kremendahl, Botschaft, 101-102. 58 The tone of Gal 1:6 is probably most similar to P.Bad. 2.35, although this letter still concerns financial affairs. 59 See n. 23.
as an unironic rebuke, Paul's use of θαυμάζω in Gal 1:6 simply does not do this. It is a strong, vehement criticism in a highly emotive passage that crescendos through to Paul's pronunciation of anathema in 1:8-9. Finally, the fact that Gal 1:6 can be coherently read as irony is another departure from the majority of the papyri. As discussed above ( §1.1.1), irony is not inherent to the use of θαυμάζω ὅτι/πῶς/εἰ in the papyri, and only a minority of examples are potentially ironic.60 In terms of its subject matter and rhetorical function, Gal 1:6 has little in common with most relevant uses of θαυμάζω ὅτι/πῶς/εἰ in the papyri beyond a semantic parallel and the presence of rebuke, broadly construed. These differences attest to the rhetorical versatility of expressions of surprise in epistolography, which can convey anything from polite concern to cutting irony.

Epistolary Handbooks
In order to determine whether an ironic reading of Gal 1:6 qualifies the text as an ironic letter, we turn to the extant ancient epistolary handbooks. Two of such handbooks exist, both falsely ascribed to Demetrius and Libanius, respectively. These handbooks are written to provide "practical instruction in letter writing" for students who already have significant background in rhetoric and composition.61 They provide short definitions of different letter types with examples, in addition to general discussion of style.62 For our purposes, these handbooks provide four pieces of relevant data: two definitions of the ironic letter and one example each.63 Nanos concludes his discussion of the handbooks by stating that, "Although no connection with these particular handbooks is implied, it appears that either Paul or the secretary (amanuensis) selected [the θαυμάζω] introductory formula on the basis of its ability to best communicate the sentiments from among the known letter types."64 Our task is therefore to determine whether there is a meaningful relationship between Gal 1:6 and the ironic letters discussed in Pseudo-Demetrius and Pseudo-Libanius suggesting that Galatians was crafted with an eye on epistolary convention. We shall discuss the evidence systematically, beginning with our earliest handbook, Pseudo-Demetrius. However, before turning to the text, it is important to raise certain cautions regarding the utility of our data.

Caution: Date and Letter Writing Education
Pseudo-Demetrius's handbook in its final form dates broadly within a 200 bce to 300 ce range. Multiple revisions of the text are likely, and a pre-Christian original is plausible.65 It is therefore a relevant source for thinking about letter writing in the first century. Pseudo-Libanius is however very late, dating to the time of Christianity's rise to prominence somewhere in the 4th to 6th centuries ce.66 D. deSilva argues that its late date makes Pseudo-Libanius of limited utility for understanding letter writing in the first century.67 It will therefore be important to reflect on provenance when assessing the relationship between Galatians and this text.
It is also important to recognise that the level of letter-writing training represented in these handbooks is not a given even for those who spent significant time writing letters. Malherbe argues that, although students were likely taught to write letters in their teen years during their education, these letter writing handbooks were likely designed to train professional writers of letters, not students.68 Paul's engagement with epistolary theory, then, depends on his having specific education in letter writing, not merely a standard rhetorical education. In his work on 2 Cor 10-13, R. Schellenberg provides a more sustained argument for the implausibility of Paul being "a participant in the epistolary tradition to which Pseudo-Demetrius attests."69 Schellenberg bases this conclusion on an assessment of where the sort of letter writing represented in the handbooks fits into the ancient curriculum, and notes that previous scholarship on Second Corinthians has not engaged with these handbooks critically, being quick to draw comparisons and ignoring telling differences. 70 Without rehashing Schellenberg's discussion in full, it will suffice to recognise that these cautions must impact how we weigh the evidence of the rhetorical handbooks. In order to support a designation of Galatians as an ironic letter, there must be significant meaningful parallels between Gal 1:6 and the handbooks, especially Pseudo-Demetrius, which is by far our earliest and most relevant comparative text.

3.2
Pseudo-Demetrius Pseudo Demetrius defines the ironic letter as follows: "It is the ironic type when we speak of things in terms that are their opposites, and we call bad men noble and good" (Εἰρωνικὸς δε ἐστιν, ὅταν ἐναντίοις πράγμασιν ἐναντία λέγωμεν καὶ τοὺς κακοὺς καλοὺς καὶ ἀγαθοὺς λέγωμεν, Ps.-Dem. Epist. Typ. 20 [Malherbe]). In terms of content and vocabulary, this is very close to the definitions of irony found in Greek grammatical treatises (see Alex. Fig. 18 ). Since Pseudo-Demetrius's definition is not specific to the epistolary genre-it is simply a definition of irony-we can expect that any ancient use of irony will correspond with it. Thus, while Pseudo-Demetrius's definition fits well with an ironic reading of Gal 1:6, this connection does not provide significant evidence for or against Paul's use of epistolary theory. The use of irony does not imply familiarity with technical discussions of irony, nor does it rule it out.
ps.-dem. Epist. Typ. 20 You have (now) shown the goodwill toward us that you have had for a long time. For your noble and good policy (toward us) has not gone unnoticed, since, as far as you're concerned, you have ruined us. But if you should get what you deserve, do not be distressed, for we are confident, the gods willing, that we shall have such an opportunity (to repay you) as you will never have against us. malherbe This letter essentially consists of mockery and a veiled threat. The major elements that we see in an ironic reading of Gal 1:6, mock astonishment, ironic encouragement, and rebuke, are all absent here. Besides the fact that both contain irony, which should be a given, there is little comparison worth making between this letter and Gal 1:6.71 Without strong parallels between the only extant epistolary handbook written by Paul's day and Gal 1:6, the case for Galatians as an ironic letter is seriously jeopardised.

3.3
Pseudo-Libanius Despite its late date, Pseudo-Libanius has attracted more attention in discussions of Galatians. While its definition of the ironic letter is interesting, like the definition of Pseudo-Demetrius, it is largely a description of irony and therefore not particularly helpful for determining whether Paul has written Galatians with reference to letter writing handbooks: εἰρωνικὴ δι᾽ ἧς ἐπαινοῦμέν τινα ἐν ὑποκρίσει περὶ τὴν ἀρχήν, ἐπὶ τέλει δὲ τὸν ἑαυτῶν σκοπὸν ἐμφαίνομεν, ὡς τὰ ῥηθέντα καθ᾽ ὑπόκρισιν εἰρήκαμεν.
Ps.-Lib. Epist. Styl. 9 The ironic style is that in which we feign praise of someone at the beginning, but at the end display our real aim, inasmuch as we had made our earlier statements in pretense. malherbe We have already observed this pattern of juxtaposing ironic praise with literal negative evaluation in Gal 1:6-7a (εἰς ἕτερον εὐαγγέλιον, ὃ οὐκ ἔστιν ἄλλο), and seen the prevalence of this signal of irony in ancient Greek more generally ( §2.2.1). With this feature being so ubiquitous, we cannot say that anyone whose irony follows the pattern [ironic evaluation] → [literal evaluation] is following a model like Pseudo-Libanius. Pseudo-Libanius has made an astute observation about a typical feature of ancient irony, which also happens to be true of Gal 1:6-7.
The section from the handbooks most drawn into comparison with Galatians is Pseudo-Libanius's exemplar ironic letter. It is easy to see why this text would be relevant: I am greatly astonished at your sense of equity, that you have so quickly rushed from a well-ordered life to its opposite-for I hesitate to say wickedness. It seems that you have contrived to make, not friends out of your enemies, but enemies out of your friends. For your action has shown 71 Nanos does not cite this example. Note its absence from the discussion in Nanos, Irony, 50-51.
itself to be unworthy of friends, but eminently worthy of your drunken behavior.
ps.-lib. Ep. Styl. 56 Θαυμάζω ὅτι οὕτως ταχέως μετατίθεσθε ἀπὸ τοῦ καλέσαντος ὑμᾶς ἐν χάριτι χριστοῦ εἰς ἕτερον εὐαγγέλιον, ὃ οὐκ ἔστιν ἄλλο. . From here, the fact that this letter "corresponds with many features in Galatians" is essentially taken as obvious.74 However, considering that we are dealing with a text composed three to five centuries after Galatians, moving from this example to the idea that Galatians was intentionally constructed as an ironic letter requires something like the following to be true: (1) Pseudo-Libanius's example ironic letter represents a widespread, influential tradition codified in much earlier epistolary theory, (2) Paul is drawing directly on this tradition in Gal 1:6, resulting in both thematic and linguistic parallels. However, Pseudo-Demetrius, the only direct evidence that we have concerning (1), suggests the contrary. Pseudo-Libanius's 72 Dahl also sees a connection between the second sentence, the mention of friends becoming enemies, and Gal 4:16. He writes, "the theme of friends becoming foes is also attested in other letters and is an indication of the deteriorated relationships that are often part of the general background of letters that contain ironic rebukes (Dahl, "Genre," 124). He does not cite any examples. For this reason, and because ironic rebuke as an epistolary formula is itself problematic, this parallel cannot be established as significant. 73 Dahl, "Genre," 118. 74 "[T]he affinity of the two subgenres, ironic and oneidistic letter, is obvious, as is the affinity of Galatians to both of them" (Dahl, "Genre," 118; cf. Nanos, Irony, 51). While much is also made of Pseudo-Libanius's letter of reproach (Ὁνειδιστική), which also features θαυμάζω (see Ps.-Lib. Ep. Styl. 64), we shall not discuss this letterform at present, as irony is our primary concern. epistolary theory on Paul's part. Second, Pseudo-Libanius provides evidence that Galatians has been and can be read as an ironic letter. If Pseudo-Libanius has used Gal 1:6 as a model for his exemplar ironic letter, his reading represents ancient support for the ironic reading argued earlier in this article ( §2.2.1). His identification of Gal 1:6 as ironic also suggests that scholars who have designated the letter as such are within the realm of ancient interpretive possibilities, even if there is no evidence that Paul had recourse to letter writing handbooks in penning the irony of Gal 1:6.

Conclusions
Investigation of whether Galatians can appropriately be deemed an ironic letter has taken us from the papyri, through ancient discussions of irony, and into epistolary theory. A review of scholarship on the use of θαυμάζω ὅτι/πῶς/εἰ in papyrus letters has demonstrated that this feature cannot be appropriately termed a formula of "ironic rebuke," as irony is simply not inherent to it. I have also argued that there are too few instances of θαυμάζω ὅτι/πῶς/εἰ in the papyri for this feature to be meaningfully designated as an epistolary formula. The data is too sparse and much of it is too late to be relevant to Paul's context. This observation suggests that the use of θαυμάζω in Gal 1:6 would not immediately signal a specific interpretation to Paul's audience that they would be familiar with from their day-to-day correspondence. θαυμάζω is a polyvalent verb, and all of its positive, negative, and ironic resonances would be fair game. Furthermore, without enough data to suggest that θαυμάζω ὅτι/πῶς/εἰ was an epistolary formula around the first century, the interpreter cannot use occurrences of this feature in the papyri as an interpretive key for Gal 1:6. There may be some affinities to some ancient letters, but Paul's rhetoric must be assessed in its own right. Exegesis of Gal 1:6-7a has itself yielded more positive results. A brief review of grammatical and rhetorical discussions of irony roughly contemporaneous with the first century has demonstrated that the sort of irony we are looking for in Gal 1:6 is σαρκασμός, which is best defined as the expression of dispraise through feigned praise. Evidence of exaggeration, pretence (ὑπόκρισις), and the ironic use of εὐαγγέλιον all support an ironic reading of Gal 1:6. This reading sees Paul ironically congratulating the Galatians for so quickly rejecting their God. At the same time, this pretence is transparent enough that the primary sentiment communicated is disapproval and rebuke. The sharpness of this rebuke contrasts significantly with many uses of θαυμάζω ὅτι/πῶς/εἰ in the papyri, which often soften reproaches and seek to avoid conflict. While the Pawlak Novum Testamentum 63 (2021) 249-270 ambiguity of θαυμάζω does not allow certainty in an ironic reading of Gal 1:6, this reading is plausible and supported by multiple lines of evidence.
Critical assessment of the ancient epistolary handbooks was then necessary to demonstrate that an ironic reading of Gal 1:6 does not mean that Galatians was crafted as an ironic letter. From the outset, the late date of Pseudo-Libanius and the limited influence of epistolary theory in ancient rhetorical education reduce the probability that Paul or his secretary would be familiar with examples of ironic letters like those preserved in the handbooks. This difficulty is compounded by the fact that Gal 1:6 is quite unlike the exemplar ironic letter in Pseudo-Demetrius, our earliest extant letter writing handbook.
There are however strong parallels between Gal 1:6 and Pseudo-Libanius's ironic letter, which dates somewhere in the 4th to 6th centuries ce. I have argued that the most likely explanation for these parallels is that Pseudo-Libanius has read Paul, rather than Paul having read an ironic letter very like Pseudo-Libanius's, for which we have no evidence. The widespread influence of Christianity and Paul's letters in Pseudo-Libanius's day makes this connection quite reasonable. Pseudo-Libanius, therefore lends ancient support to an ironic interpretation of Gal 1:6. He also provides evidence that reading irony in Galatians is a part of the letter's early reception.
Evidence from the papyri to ancient epistolary theory therefore does not support the contention that Paul has crafted Galatians as an ironic letter. As far as it relates to irony, Paul neither appears to be using an established epistolary formula in Gal 1:6 nor drawing on letterforms peculiar to specialised training in letter writing.
This does not mean that irony is unimportant for interpreting Galatians. There is evidence to suggest that Gal 1:6 is itself ironic, and I do not think this is the only instance of irony in the letter. Exploration of Paul's use of irony in Galatians is a worthwhile endeavour that promises to shed further light on the sentiments Paul seeks to communicate between the lines of the text.