¹ The bibliography on this subject is quite extensive; see Barsby, J., Plautus: Bacchides (Warminster and Oak Park, Illinois 1986) 25–27Google Scholar. Those works that have direct bearing on this article are: Anderson, W.S., Barbarian Play: Plautus' Roman Comedy (Toronto 1993), esp. 3–29Google Scholar; Arnott, W.O., Menander, Plautus, Terence (Oxford 1975)Google Scholar; and Bain, D., ‘Plautus Vortit Barbare: Plautus, Bacchides 526-61 and Menander, Dis Exapaton 102-12’, in West, D. and Woodman, A.J. (eds). Creative Imitation and Latin Literature (Cambridge 1979) 17–34CrossRefGoogle Scholar. Henceforth, I will refer to the works above by the author's last name alone.

² See Barsby, 95; Bain, D., ‘Female Speech in Menander’, Antichthon 18 (1984) 40–41CrossRefGoogle Scholar.

³ See Barsby, 93.

⁴ Menander usually employs more than seven speaking roles in a comedy. Only one, Samia, calls for fewer (six). All other reasonably well-preserved plays involve more: Aspis includes at least eight, Perikeiromene and Sikyonios at least nine, Epitrepontes at least eleven and Dyskolos thirteen.

⁵ Cf. also Grumio in Mostellaria and Sosia in Andria.

⁶ See Barsby, 152.

⁷ See Fraenkel, E., Elementi Plautini in Plauto (Florence 1960) 21–54Google Scholar.

⁸ For Menander's regular use and Plautus' general avoidance of names like Parmeno and Syrus, see Barsby, 153.

⁹ Lesser features in the song abet in advancing the ‘double’ pun. First, the opening line of the song also includes a pun, stated twice, on the root meaning of Chrysalus' name in Greek, ‘golden’ (auro). This sets the stage for the more complicated word-play on duplex, also repeated, in the next line, where again the Greek cognate is not stated but implied. So, from its very outset the song sets a tone of bilingual paronomasia centring around Chrysalus. Second, ‘doubling’ itself is a pervasive motif in the earlier part of this song. There are no fewer than ten pairs of identical or similar words: hunc/huic. decet/decet, auro/auro, statuam/statui, duplex/duplicibus, hodie/hodie, feci/adfectus, lusi/ludificatus, callidum/callidis, compuli/perpuli, quicum/quicum. Thus, in more ways than one, Plautus advertises Chrysalus' prowess as a master ‘doubler’.

¹⁰ This idea is supported by the observation that most of the rest of the aria seems to stem from Plautus. At least three other passages must derive from his, not Menander's, hand: Chrysalus’ anticipation of rewards (646), his advice on how to be a conniving slave (651-62) and his allusion to Hercules' tithe, a Roman custom (665-6). Adding these to the deprecation of Greek comic slaves at 649-50, one can see that so much of this canzonetta is Plautus' concoction that it becomes tempting to say that all of it is; see Barsby, 152: ‘There is therefore every justification for seeing the whole monody as essentially an invention by Plautus, based perhaps on a brief monologue in Menander where the slave simply referred to his successful deception of the father and wondered what use his young master had made of it …’. If Menander is not responsible to any significant extent for the content of the aria, it is then possible for it to contain deprecatory remarks about the original at any point including the beginning.

¹¹ For the base duplic- used in reference to the intensification of a deed or crime, cf. Poen. 564 (id duplicabit omne furtum). Cf. also Men. 546, where duplex refers to gold and means ‘double in quantity’. The middle part of the song (651-60), where Chrysalus praises the confident trickster who has a multipotens pectus (652), may also contain implied criticism of the original, if the Greek slave were less brazen and self-assured than the Roman in the execution of his deception; see Damen, M., ‘Translating Scenes: Plautus' Adaptation of Menander's Dis Exapaton’, Phoenix 46 (1992) 227CrossRefGoogle Scholar: ‘… the Greek playwright's depiction of a desperate, reactive slave would have run counter to the cool-headed, manipulative servi callidi Plautus preferred to present’. Thus, it may be that most of the song deals, albeit rather elliptically, with differences in the Greek and Roman treatments of the plot.

¹² Close scrutiny of Plautus' language at Bacch. 649-50 controverts the thesis that this passage can be construed as providing an unambiguous indication of the sum swindled in Menander's play. When Plautus not only uses the Greek names in the plural (Par-menones, Syri 649) but also does not indicate one clear sum (duos aut tris minas), he shows clearly that he is generalising, not discussing the particulars of any one play. His allusions to Dis Exapaton are thus at once conspicuous and indirect, no doubt because he wishes to confront his model without inviting literary or scholarly debate. In this context, ‘two or three minas’ is patently not intended to recall the precise amount swindled in Dis Exapaton. Furthermore, it is far from certain that Plautus' generic statement about the amount of money slaves in Greek comedy abstracted from their masters is meant to be taken at face value, either. Plautus' mocking stance toward Greek comedy here undermines his credibility concerning the usual range of those sums (which is to assume there was a usual sum!), especially when the remains of Menander's work are too scant to allow us to check Plautus' claim. If anything, it seems more likely that Plautus has deflated the typical Greek sum, if such a thing actually existed, in order to magnify the grandeur of his slave's malfeasance; see Barsby, 153, where he notes that in Hauton Timorumenus a different Syrus swindles ten minas.

¹³ Plautus' principal goal here is not to present a rigorous or legalistic defence of the superiority of his adaptation to Menander's original, but to delight and entertain his audience. In such a milieu, precise calculations have no more place than a ‘smooth’ flow of logic; see Barsby, 144-5, on the integrity of 540-51: ‘It is true that the play would proceed smoothly if 552 directly followed 539, but smoothness is not a primary characteristic of Plautus' work.’

¹⁴ As a rule, monetary sums in Plautus are substantially larger than those in Menander or Terence; see Barsby, 147, 153.

¹⁵ From this it is apparent Plautus trusted his audience to have sufficient familiarity with Dis Exapaton for them to know (or, at least, to gather from the title) that the Greek play contained two deceptions. That comes as no surprise, since the presence of two deceptions in a play entitled ‘The Double Deceiver’ is not very difficult to surmise. It also assumes he expected that they could recognise a word from its title translated into Latin, but this is hardly startling either. Clearly he thought that at least some part of his audience would know an even more refined bit of information about Dis Exapaton, that the cast of the Greek play included a slave named Syros. Elsewhere in Bacchides Plautus' jokes require knowledge of rudimentary Greek vocabulary, e.g. Chrysalus' earlier pun at Bacch. 240 on the Greek base of his name; see Barsby, 119.

¹⁶ Bain, 29: ‘What follows [538], however, is at least at first glance surprising since it shows Plautus making a change which does not quite fit into one of the expected categories of Plautine activity and reveals as Plautine a passage which no one had thought of denying Menander.’

¹⁷ Leo, F., Plautinische Forschungen² (Berlin 1912) 131Google Scholar. Arnott, 39: ‘Alone of editors here Leo is right two out of three times: vindication from an unexpected source of the quality of his judgement.’

¹⁸ More than that, the confrontation of the friends here constitutes a subtle play on character types typical of Greek New Comedy. Sostratos so readily accepts the conclusion that Moschos' lust and his beloved's charms have overcome his friend's sense of propriety and philia (ϕιλοῦντα, Dis Ex. 109), it is almost as if the comic character had seen enough New Comedy to expect a young man like Moschos to fall prey to his desire just as so many other young lovers in New Comedy do. Menander elsewhere generates irony by contrasting a character's real actions and the actions expected of him or her as a type. For instance, Smicrines in Epiirepontes incorrectly assumes that his son-in-law Charisios is acting like a typical, faithless young man when he abandons his new bride, Smicrines' daughter Pamphile. After Demeas in Samia overhears a nurse talking about his purported baby, he leaps to the false conclusion that his adopted son Moschion has fathered the child illegitimately, a speculation grounded at least to some extent in the frequency with which young men (and Moschions in particular) do this in Greek New Comedy. Because Polemon sees Glycera in the arms of her brother Moschion, he believes she is being unfaithful to him as soldiers' lovers often are in New Comedy (e.g. Miles Gloriosus) when, in fact, she is not. With all this, it is hard to believe Menander did not see or utilise the opportunity during the young men's confrontation in Dis Exapaton for playing the ironies built into Sostratos' misunderstanding, since they are so clearly based on assumptions about New Comedy character types.

¹⁹ Owens, W.M., ‘The Third Deception in Bacchides: Fides and Plautus' Originality’, AJPh 115 (1994) 385-7Google Scholar, reviews the evidence against Williams' thesis that Plautus derived all three deceptions from Menander.

²⁰ The Dis Exapaton papyrus shows at least one instance in which Plautus reshuffled the order of the scenes as they appeared in the original. He brings Pistoclerus on stage before Mnesilochus to cover the return of the money (Bacch. 526 ff), whereas in Menander their counterparts entered in the reverse order (Dis Ex. 91 ff.). Earlier in the play a similar transposition of scenes in order to cover an act-break probably resulted in Lydus' illogical but comical ‘extended visit’ within Bacchis' house; see Damen (above, n.11) 213 ff.

²¹ Dis Ex. 104 () = Bacch. 669-70 (quid vos maestos tam tristisque esse conspicor? … quin respondetis mihi?).

²² Contrary to the notion that Bacch. 540-51 is not genuine (see Barsby, 144 f.; also below, n.26), these data support Barsby's insightful suggestion that Plautus may have ‘contaminated’ the original at this point (ibid.), only they show that it was not with a piece of another play, rather with something from the same play itself. Barsby's basic sense that the ironies are ‘out of place’ is perceptive and still sound.

²³ Plautus has almost certainly made other adjustments too, mainly in the language and tone of the ironies. Owens (above, n.19) 403-4, points to notably Roman features in this passage, in particular the Roman political language and the use of fides. We are here, however, concerned with the larger structure of the scene, not details of phrasing or precisely how the change was engineered.

²⁴ That Sostratos had earlier stated it was not Moschos but his girlfriend who is principally to blame is hardly relevant to this case. As the story works out, Sostratos runs into Moschos first and vents his anger on him, just as Demeas in Samia (324-56) faults Chrysis and exculpates Moschion in a monologue much like Sostratos' two (Dis Ex. 18-30, 91-102), but later denounces Moschion for his purported crime anyway (Sam. 477-91). See Goldberg, S.M., ‘Act to Action in Plautus' Bacchides’, CPh 85 (1990) 197Google Scholar, who astutely notes that Sostratos is ‘probably on the verge of a nervous loquacity as the papyrus breaks off’.

²⁵ See Arnott, W.G., ‘A Study in Relationships: Alexis’ Lehes, Menander's Dyskolos and Plautus' Aulularia’, QUCC 33.3 (1989) 32Google Scholar: ‘The irony arises when he is confronted by Gorgias in the second act, and Gorgias lectures him on the sexual immorality of rich young men with his background: him, Sostratos, the one rich young man in Menander who does not need such a lecture.’

²⁶ This may help explain a curious detail of Plautus' play. Pistoclerus eagerly denounces those who are jealous of others' success and are lazy in order to avoid others' jealousy (543-4) without knowing that his purported usurpation of Bacchis' attentions is the very thing motivating Mnesilochus' anger. Why does he cite this specific possibility, an all-too-appropriate corollary of a false friend's infidelity, when he has no obvious reason to be thinking along these particular lines? In general, the words would fit better in Mnesilochus' mouth who knows the supposed truth. But if Plautus has given to Pistoclerus some of the ironies that Menander had originally put in Sostratos' mouth, we would expect a denunciation of just this sort. Note, however, that 550-1 are a different matter altogether. They do not suit the context of the scene well at all. They interrupt the flow of the logic and thus may indeed be spurious.

²⁷ See Damen (above, n.11) 228; see also Anderson, 116, who discusses the manner in which Plautus has most likely expanded on Syros' language at Bacch. 816-21. For a more general assessment, see Fraenkel (above, n.7) 105-34 (‘Ampliamenti del dialogo’).

²⁸ See Hunter, R.L., ‘The Aulularia of Plautus and its Greek Original’, PCPhS n.s. 27 (1981) 45Google Scholar: ‘It is my impression that the variety of approaches which Plautus adopted when reworking Greek plays is at least as striking as the common thread which runs through all his work.’

²⁹ posthac etiam illud quad scies nesciveris / nec videris quod videris (MG 572-3).