PARA PROSDOKIAN AND THE COMIC BIT IN ARISTOPHANES

Abstract

This article bridges a gap in the study of Aristophanic humour by better demonstrating how individual jokes (in this case, the para prosdokian ‘contrary to expectation’ joke) contribute to the wider comic scenes in which they are embedded. After analysing ancient and modern explanations and examples of para prosdokian jokes, this paper introduces the concept of ‘comic bit’, a discrete unit of comedy that builds humour around a central premise, and establishes how para prosdokian jokes contribute to comic bits in a way that recent theories of para prosdokian cannot account for.

Over the past few decades, our understanding of how humour operates in ancient Greek comedy has improved dramatically. One strand of scholarship applies modern humour theory to the study of Aristophanes and the other poets of Old Comedy, often but not exclusively focussing on the mechanics of the joke.¹ A different strand of scholarship tends to study larger topics related to humour, but not humour itself per se; here we have analyses of comic issues such as laughter,² nonsense,³ surprise⁴ and parody,⁵ all of which have strong implications for our understanding of humour in Greek comedy. However, there remains a gap in the literature for research that blends these two strands of scholarship—the small-scale analysis of individual jokes and the large-scale analysis of the wider context of humour.⁶ In a recent edited volume on Aristophanic humour in theory and practice,⁷ Lowe notes the problem: ‘while the cognitive element of joke form is now fairly well understood, other aspects of the larger phenomenology of humour remain work-in-progress. […] Among the particular challenges for joke-based models are […] the workings of extended comic sequences, routines, performances, and narratives, and the question of their reducibility to sequences of atomized jokes’.⁸ Thus the objective is not only to study the mechanics of individual jokes, but to examine the relationship between jokes and larger units of humour. This paper adds to this conversation by investigating as a case study the para prosdokian ‘contrary to expectation’ joke. This type of joke was selected not only because of its familiarity to scholars of Old Comedy,⁹ but also because of its recent attention by Kanellakis, who in three publications has suggested a new, comprehensive theory of para prosdokian jokes.¹⁰ After analysing this account and drawing attention to some areas where I disagree, I introduce the concept of ‘comic bit’, a discrete unit of comedy that builds humour around a central premise, and establish how para prosdokian jokes contribute to comic bits in a way that previous theories cannot account for.

1. DEFINITIONS OF PARA PROSDOKIAN

Since this paper concerns para prosdokian jokes, it is worth considering a few definitions and examples of the concept, both ancient and modern. Silk defines para prosdokian as ‘a “straight” sequence interrupted by a sudden explosive joke’,¹¹ a definition which highlights the component of surprise. More detailed is the definition of Sommerstein, who says that para prosdokian and its synonym par’ hyponoian ‘refer to a comic device consisting in the substitution, for a word or phrase that the audience had been led by the context to expect, of an entirely different word or phrase. The substituted expression normally makes sense in context, but a sense quite unlike the one that had been expected, and the surprise effect is calculated to excite laughter’.¹² Modern examples of para prosdokian jokes illustrate this process whereby expectations are set and then subverted with a surprising twist:

• ‘Take my wife—please!’ – Henny Youngman¹³

• ‘I haven't slept for ten days, because that would be too long.’ – Mitch Hedberg¹⁴

• ‘I want to die peacefully in my sleep like my grandfather, not screaming in terror like his passengers.’¹⁵

• ‘I tell you what's a bad job is a cop. That's a tough job, right, you got all that power but they don't pay you enough. You got a badge and a gun but you make sometimes less than a teacher … Can you imagine risking getting shot every day and you only make a little more than a cop?’ – Sam Morril¹⁶

Sam Morril follows his joke by telling the audience, ‘We call that the ol’ switcheroo in show biz’, and while a little imprecise, we can do far worse than defining para prosdokian jokes as ‘the ol’ switcheroo’.

Ancient commentators give definitions and examples of para prosdokian jokes as well, and their picture largely corresponds with our modern conception.¹⁷ In his work On Style, Demetrius analyses para prosdokian as a particular kind of ‘charm’ (152):¹⁸

ἔστι δέ τις καὶ ἡ παρὰ προσδοκίαν χάρις, ὡς ἡ τοῦ Κύκλωπος, ὅτι “ὕστατον ἔδομαι Οὖτιν”. οὐ γὰρ προσεδόκα τοιοῦτο ξένιον οὔτε Ὀδυσσεὺς οὔτε ὁ ἀναγινώσκων. καὶ ὁ Ἀριστοφάνης ἐπὶ τοῦ Σωκράτους, “κηρὸν διατήξας”, φησίν, “εἶτα διαβήτην λαβών, ἐκ τῆς παλαίστρας ἱμάτιον ὑφείλετο”.

There is also a sort of charm from the unexpected, as in the Cyclops’ words, ‘No-man I will eat last’. Neither Odysseus nor the reader was expecting this kind of hospitality gift. Similarly Aristophanes says of Socrates, ‘He melted some wax first, then grabbed a pair of compasses, and from the wrestling school—he stole a coat’.¹⁹

The first example is adapted from Homer's Odyssey (9.369), right after Odysseus had asked for a hospitality gift from the Cyclops and told him his name was ‘No-man’; we are led to believe that the Cyclops will give a typical hospitality gift, but the para prosdokian is that the gift Odysseus receives will be the gift of dying last. The second example is adapted from Aristophanes’ Clouds (149, 179), where the sentence establishes a set of high-class, philosophical expectations that are then subverted by Socrates stealing the coat. Hermogenes’ On Method of Forceful Speaking lists para prosdokian as one of three methods of speaking in comic style (34):

τὸ δὲ παρὰ προσδοκίαν τοῦτο “βδελυρὸν μὲν οὖν τὸ πρᾶγμα καὶ οὐκ ἂν ἐβουλόμην λαχεῖν, ἐπειδὴ δὲ ἔλαχον”⋅ προσδοκᾷ μὲν ὁ ἀκροατὴς ἀκοῦσαι “ὑπομενῶ”, φησὶ δὲ “οὑκ ἂν ἐβουλόμην”.

τούτοις πᾶσι χρῆται Δημοσθένης ἐν τῷ Περὶ τοῦ στεφάνου⋅ ὅθεν δηλοῦται, ὅτι κωμῳδεῖν ἐπίσταται. […] τῷ δὲ παρὰ προσδοκίαν οὕτω περὶ Αἰσχίνου λέγων “οὐδὲ γὰρ ὧν ἔτυχεν ἦν, ἀλλ᾽ οἷς ὁ δῆμος καταρᾶται”.

An example of para prosdokian is this, ‘Now the thing is disgusting, and I would not want to get it, but since I did…’ The hearer expects to hear ‘I put up with getting it’ but he says ‘I would not want.’

Demosthenes uses all of these in On the Crown, from which it is clear that he knew how to speak in a comic style … He uses para prosdokian when speaking about Aeschines thus (18.130), ‘He was not whatever he happened to be, but what the people curse’.²⁰

Neither example is particularly funny, and the first example does not even illustrate the point well—Kennedy writes that ‘a better example could have been found’.²¹ Still, we can recognize the objective of para prosdokian to establish and subvert audience expectations. The fourth-century c.e. rhetorician Tiberius provides another example (De Figuris Demosthenicis 16):

παρὰ προσδοκίαν δ᾽ ἐστὶν ὅταν, ἄλλο τοῦ ἀκροατοῦ προσδεχομένου, ἄλλο μετά τινος χάριτος ἐπενέγκῃ, οἷον⋅ “τὰ μέντοι Φιλίππου εὐεργετήματα τοιαῦτα ἔσται⋅ οὔτε τὰ ὑμέτερα ὑμῖν αὐτοῖς ἀποδώσει, οὔτε ἐν τῇ οἰκουμένῃ αἱ τιμαὶ ἔσονται”.

para prosdokian is when, despite the hearer expecting one thing, something different is emerging, with some charm, like this (7.35): ‘Such will be the benefactions of Philip: he will neither restore your possessions, nor will he bring glory to the world.’

We are led to believe that Philip's benefactions will be a good thing, but the twist is that they promote bad outcomes—not restoring the audience's possessions and not bringing glory to the world. The Tractatus Coislinianus, a tenth-century manuscript that analyses comedy in the style of Aristotle, utilizes the term para prosdokian as well. Some have posited that this manuscript reflects material from the lost second book of Aristotle's Poetics; however, the issue of whether this is the work of Aristotle or someone following in the Aristotelian tradition remains unresolved.²² While the Tractatus does not define the term or give an example, it places para prosdokian under the category ‘laughter from actions’ (ὁ γέλως ἀπὸ τῶν πραγμάτων) as opposed to ‘laughter from diction’ (ὁ γέλως ἀπὸ τῆς λέξεως).²³ It might be surprising that para prosdokian, which seems restricted to verbal humour, might be categorized under action, but this is consistent with other Aristotelian references to things ‘contrary to the expected’.²⁴ In Aristotle's own works, he reveals familiarity with the concept of things arising contrary to expectation, though he tends to use slightly different phrasing, such as παρὰ τὴν δόξαν (Poet. 1452a), παρὰ δόξαν (Rh. 2.1379a) or παράδοξος (Rh. 3.1412a). In this last passage, Aristotle includes an example of a para prosdokian joke:

οὐ γὰρ ὥσπερ ὁ ἀκούων ὑπέλαβεν· “ἔστειχε δ᾽ ἔχων ὑπὸ ποσσὶ χίμεθλα”· ὁ δ᾽ ᾤετο πέδιλα ἐρεῖν.

For instance, [the following verse] does not finish as the listener expected: ‘He strode on, wearing on his feet—chilblains’, whereas the listener thought he was going to say ‘sandals’.²⁵

As Destrée notes, the humour derives from the incongruity of the word in its context, enhanced by its unexpectedness, though additional humour can be found in its reference to something repugnant and the fact that it parodies Homeric language, where walking with fine sandals typically applies to divinities.²⁶ The other important ancient source that uses the term para prosdokian (or its synonym par’ hyponoian) are the scholiasts, who employ it to gloss lines where the commentator detects a para prosdokian joke (twenty instances in the scholia to Aristophanes; nineteen instances in the scholia to other authors).²⁷ While we might disagree with a scholiast whether a particular instance is truly an example of para prosdokian, there is nothing drastically different about their usage to suggest they have an alternate conception of para prosdokian in mind.²⁸

From this summary, we can see that para prosdokian is conceived of as operating by establishing expectations and then violating them with a surprising twist. If we analyse this in terms of humour theory, para prosdokian jokes exemplify the incongruity theory of humour, which operates via the same principle.²⁹ Incongruity theory suggests that the hearer interprets the sentence in one way (in technical language, she follows a ‘script’) but then, at the moment of the ol’ switcheroo, suddenly realizes that she was supposed to be following a second and heretofore latent second script, or alternate way of interpreting the sentence. The resolution of the incongruity between these two scripts produces the humour. One thing to note about these ancient sources is that para prosdokian is not restricted to comic passages—we have examples from Homer and Demosthenes in addition to Aristophanes and the unnamed comic poet Aristotle is referencing. Still, para prosdokian is most associated with Aristophanes.

Several Aristophanic para prosdokian jokes are typically presented as canonical examples of the form. In Acharnians, when the Megarian is addressing his starving daughters and telling them to pay attention, the standard phrase would be ‘give me your attention (τὸν νόον).’ Instead, the Megarian replaces ‘attention’ (τὸν νόον) with ‘belly’ (τὰν γαστέρα)—since his daughters are starving, the only way he can save their lives is to sell them into slavery, where at least they will be fed (733):³⁰

ἀκούετε δή, ποτέχετ’ ἐμὶν τὰν γαστέρα⋅

Now listen, give me your undivided bellies.³¹

In Knights, an initial set of promises to the Sausage-Seller about his future power (being able to trample the council and imprison the generals) shifts to the assertion that, rather than dining in the Prytaneum, he would be giving oral sex there (166–7):³²

βουλὴν πατήσεις καὶ στρατηγοὺς κλαστάσεις,

δήσεις, φυλάξεις, ἐν πρυτανείῳ λαικάσεις.

You'll trample the council, dock the generals,

Put people in chains and lock them up, and in the Prytaneum you'll suck their cocks.

Many other examples have been proposed in the scholarly commentaries as well as in various publications on Aristophanes, but this should suffice to illustrate the point. With this baseline knowledge about para prosdokian, let us turn to Kanellakis’ views on the subject.

2. KANELLAKIS’ INTERNAL EXPECTATIONS ACCOUNT OF PARA PROSDOKIAN

Kanellakis’ account revolves around the idea that our discussion of para prosdokian should consider various contexts that are internal to the play; therefore, I will call this view the ‘internal expectations account of para prosdokian’. Kanellakis’ starting point is Silk's definition, ‘a “straight” sequence interrupted by a sudden explosive joke’, and he writes that while para prosdokian most obviously applies to subversions of fixed phrases such as idioms or proverbs, it equally applies to any verbal sequence with potential for logical and stylistic cohesion. Kanellakis’ goal is to dive more deeply into the details of the semantic contradictions that underlie para prosdokian jokes. He points out that para prosdokian jokes need not locate the unexpected twist at the end of the sentence, since easily recognizable phrases can be subverted at the beginning, such as ‘meowing up the wrong tree’ which plays on ‘barking up the wrong tree’ or Aristophanes’ δειλὸν καὶ μέγα ‘cowardly and great’ (Av. 1477) which plays on the formulaic phrase δεινὸν καὶ μέγα ‘powerful and great’.³³ Above all, Kanellakis encourages us to examine the expectations in a more nuanced manner—if para prosdokian jokes work by moving contrary to expectation, then we must first know what the prosdokia ‘expectations’ are.³⁴

The precise nature of the expectations involved in a para prosdokian is something that previous accounts took for granted, and it is worth contemplating more closely. If we do not consider the nature and extent of the relevant expectations, we might be led into error and judge lines to be para prosdokian when they are not. For example, at Peace 95, Trygaeus is about to fly his dung-beetle to Olympus, and a servant asks him:

τί πέτει; τί μάτην οὐχ ὑγιαίνεις;

Why do you fly? Why act crazy for nothing?

The scholiast on this line calls τί πέτει a para prosdokian for τί κάμνεις; but Kanellakis points out that τί πέτει follows perfectly from what Trygaeus had just said, ‘I'm flying for the sake of all the Greeks, trying my hand at a novel adventure’ (Peace 93–4); thus there is no reason to consider this a para prosdokian.³⁵ This is certainly plausible; however, there is another way to interpret the line in the broader context of the passage, where Trygaeus’ madness is pathologized as illness.³⁶ We might be able to defend the scholiast's claim as follows: what the scholiast thinks is expected is yet another direct comment about Trygaeus’ madness such as τί κάμνεις; ‘what are you suffering from (sc. that you are raving in this way)?’, or perhaps ‘why are you ill?’, but Aristophanes plays against this expectation by having the slave use language that makes both literal sense and metaphorically suits the situation, since πέτομαι means both ‘I am flying’ and ‘I am mentally unstable’. This accords with what the scholiast says (on Peace 95):

τί πέτῃ· παρὰ προσδοκίαν, ὡσανεὶ ἔλεγε “τί κάμνεις;” ἀπὸ κοινοῦ δὲ τὸ “μάτην”, ἵνα ᾖ “τί μάτην πέτῃ;”

τί πέτῃ: para prosdokian, as if he was saying τί κάμνεις; The μάτην is applicable to both phrases, so that it means τί μάτην πέτῃ;³⁷

Kanellakis’ enterprise is to explore the logic behind alleged cases of para prosdokian to see how exactly the expectations are subverted; however, this example shows just how difficult it can be to navigate the complexities of the situation. This is because expectations come in degrees, and judging a line to be para prosdokian involves converting those degrees of expectation into something binary—one must assert that it either is, or is not, a para prosdokian. Since para prosdokian works best when there is a clear, fixed phrase that the line is being positioned against, these judgement calls are sometimes difficult to make, especially when there are other similar phenomena in question, such as metaphor. Consider the following lines (Peace 62–3):

ὦ Zεῦ, τί δρασείεις ποθ’ ἡμῶν τὸν λεών;

λήσεις σεαυτὸν τὰς πόλεις ἐκκοκκίσας.

Zeus! What on earth are you trying to do to our people?

Before you know it, you'll have pitted and pulped our cities!

Is this a para prosdokian or merely a metaphor? If any subversion of expectations exists, it lies with the very last word, ἐκκοκκίζω ‘pit, shell (as in seeds)’, which comes in final position, as it tends to with para prosdokian. We can also imagine some word that ἐκκοκκίζω reasonably subverts, such as ‘destroy’, ‘sack’ or ‘demolish’ (e.g. ἐκπέρθω, ἐκπορθέω, ἐκφθείρω). There is certainly room for differences of opinion here—some might claim it as a para prosdokian and others might not.³⁸ Separate from the issue of degree of expectation, the key point is that the frame of reference for both examples (τί κάμνεις; ‘destroy’) is consistent, lying with some phrase external to the play.

On the other hand, Kanellakis suggests that various explanations about para prosdokian, both ancient and modern, are erroneous because they overlook some important internal context. These contexts are categorized into different types (verbal, literary, scenic, and cultural), but there is no functional difference between them—all of them, according to Kanellakis, provide some internal context that makes the proposed example of para prosdokian not surprising, not contrary to expectation, and consequently, not actually a para prosdokian. Here are a few examples, to which I will be returning:

Case 1: In Women at the Thesmophoria, Aristophanes parodies the Athenian political system by creating an assembly of women at the Thesmophoria religious festival, and Critylla opens the assembly with a parodic version of a standard Athenian curse against their enemies, the Persians (335–7):

εἴ τις ἐπιβουλεύει τι τῷ δήμῳ κακὸν

τῷ τῶν γυναικῶν, ἢ ’πικηρυκεύεται

Εὐριπίδῃ Μήδοις τ’ ἐπὶ βλάβῃ τινί…

If anyone conspires in any way to harm

the dêmos of the women; or negotiates secretly

with Euripides and the Medes in any way to the women's harm …

According to Austin and Olson, these lines contain two para prosdokian jokes. The first is the substitution of τῶν γυναικῶν ‘of the women’ for the expected phrase τῶν Ἀθηναίων ‘of the Athenians’.³⁹ The second is the insertion of Euripides’ name into a standard Athenian curse against the Persians, thus making Euripides a civic enemy of the Athenians on the same level as them.⁴⁰ Neither of these does Kanellakis count as para prosdokian: ‘That the assembly would consist of women had already been announced in 83, and more importantly, this is the scenic “reality” now: women are addressing each other as women. That Euripides is hostile to women and vice versa is explicitly said already in 82–5 and 181–2. Therefore, there is nothing surprising in the passage when seen within its dramatic context.’⁴¹

Case 2: A second example comes earlier when Euripides asks his Kinsman to infiltrate the women at the Thesmophoria, and to look the part, Euripides needs to remove the Kinsman's hair so he can pass as a woman. Not only does Euripides shave the Kinsman's face, but he also uses a torch to singe the hair off his buttocks, with the Kinsman exclaiming (Thesm. 241–2):

οἴμοι τάλας. ὕδωρ ὕδωρ, ὦ γείτονες,

πρὶν ἀντιλαβέσθαι πρωκτὸν ἕτερον τῆς φλογός.

Oh no, no! Water, water, neighbours,

before somebody else's arse catches fire!

Austin and Olson suggest that πρωκτόν ‘arse’ is a para prosdokian for οἰκίαν ‘house’, as a parody of the call to one's neighbours for help in fighting a fire.⁴² Kanellakis disagrees: ‘What happens onstage, however, is an arse singeing rather than a house burning, hence the reference to arses is anything but surprising.’⁴³

These two examples illustrate a crucial aspect of Kanellakis’ theory. Austin and Olson highlight a connection between the lines in question and some relevant comparison in the external world—the parodic curse within Women at the Thesmophoria is compared to the real-world curse spoken by the Athenians and the Kinsman's parodic shout for help that his arse is on fire is compared to what a real-world Athenian would say if his house were on fire. Kanellakis, on the other hand, considers the internal context more important. These concepts, after all, have already been introduced—the audience already knows that there will be an assembly of women, that Euripides has been characterized as an enemy, and that the Kinsman will undergo an arse-singeing. How can these jokes possibly be surprising if the play had already set the stage for them? Kanellakis concludes that these jokes are not para prosdokian.

Case 3: This example illustrates how this view is problematic and elucidates the source of the problem. In Acharnians, there is a scene where Dicaeopolis takes the character Euripides’ tragic paraphernalia away from him—some rags, a hat, a cane, a basket, a jug and a pot. Euripides responds by exclaiming (464):

ἄνθρωπ’, ἀφαιρήσει με τὴν τραγῳδίαν

Fellow, you will deprive me of my tragedy!

Here, ἀφαιρήσει με τὴν τραγῳδίαν is used instead of some other expected phrase, and there are a few possibilities for what this expected phrase might be. Mitchell's 1835 edition prints the line with a dash, ἀφαιρήσει με τὴν—τραγῳδίαν and writes: ‘The reader expects the poet to say σκευὴν, or some such word, you will rob me of my whole wardrobe […]. The poet by a bitter piece of satire substitutes the word τραγῳδία, as if the whole substance of one of these dramas of Euripides consisted in these beggarly externals’.⁴⁴ Kanellakis favours ἀφαιρήσει με τὴν οὐσίαν/τὰ χρήματα ‘you will deprive me of my property’ on the grounds that this is a well-attested phrase in classical Greek.⁴⁵ Unlike the two previous examples, this time he suggests that it is a para prosdokian, adding, ‘This para prosdokian epitomises the central, parodic concept of the scene, i.e., that Euripides’ poetry is just a filthy trivial mishmash.’⁴⁶

In Case 3, Kanellakis focusses on the external context—the fact that there exists a similar phrase in Greek more broadly, attested in such authors as Xenophon, Plato, Demosthenes, Aristotle, Herodotus and Aristophanes himself. Yet we could apply the same logic from the previous two cases and argue that there is no surprise here, since it would be entirely consistent to expect punchlines about tragedy in a scene where Aristophanes had already mocked Euripides and Euripidean tragedy at length. This passage, when considered in conjunction with the previous two passages from Women at the Thesmophoria, reveal something important about para prosdokian jokes—two sets of expectations need to be weighed, and prioritizing one over the other might lead to peculiar results.

Earlier, I said that the task of identifying an example of para prosdokian consisted of determining cases where some set of expectations (prosdokia) is surprisingly subverted. However, there are two types of expectations potentially operative in para prosdokian jokes—external expectations deriving from the broader linguistic and cultural context and internal expectations emerging from the immediate literary and dramatic context of the play. A further question arises: which set of expectations should we be using to determine para prosdokian jokes? Should we pay attention only to external expectations or should we consider the internal ones as well? There are two possibilities to consider. Definition (1) is that para prosdokian refers to cases where there are subversions of external expectations only, and definition (2) is that para prosdokian refers to cases where there are subversions of both external and internal expectations. (I omit the possibility that para prosdokian refers to cases where there are no subversions of external expectations at all, since this seems to be a sine qua non for the concept). I submit that Kanellakis is shifting between these different senses of expectation in his presentation of the three cases above. In Case 1 and Case 2, Austin and Olson classify them as para prosdokian based on external expectations, and Kanellakis criticizes Austin and Olson for failing to consider the internal ones, from which it seems that Kanellakis subscribes to definition (2) above. In Case 3, Kanellakis classifies it as a para prosdokian based on external expectations while failing to consider the internal ones himself, from which it seems that Kanellakis subscribes to definition (1) above.

In principle, either definition is possible, since we can use terms to refer to whatever we would like; however, there are some seriously problematic consequences if we accept definition (2). First, something considered to be para prosdokian in the outside world would lose its status as a para prosdokian once embedded within a play, so long as the play contained some previous reference to the joke's topic. For example, consider the English phrase ‘meowing up the wrong tree’, which many would consider a paradigmatic case of para prosdokian. If we were to embed that same phrase in a play that featured anthropomorphic cats—and thus provided some internal expectations that we might expect jokes about cats—then ‘meowing up the wrong tree’ would no longer qualify as a para prosdokian according to definition (2). A second corollary is that para prosdokian jokes cannot exist within larger scenes if they have anything to do with the topic of that scene. For, according to definition (2), once the topic of the scene was established, then any subsequent reference to that topic, including a para prosdokian joke, would no longer be surprising or subvert expectations. Therefore, if a play establishes that a political assembly of women will be gathered at a festival to decide Euripides’ fate, then no further jokes in that scene (or indeed the rest of the entire play) about politics, assemblies, women, religion or tragedy could ever be deemed para prosdokian. A third consequence arises if we truly consider what it would mean for a play to have a para prosdokian joke that did not cohere with the internal expectations in some capacity, but instead subverted them. Given that Kanellakis has a fairly broad conception of the sorts of internal expectations that would make a potential case of para prosdokian not surprising (not just the topic of the scene but also certain types of imagery or vocabulary),⁴⁷ the sort of joke that could accomplish this would feel utterly random.⁴⁸ Think back to Aristophanes’ substitution of τῶν γυναικῶν ‘of the women’ for the expected phrase τῶν Ἀθηναίων ‘of the Athenians’ in the women's political assembly at Women at the Thesmophoria 335–6. The external expectation would be ‘of the Athenians’, which is what the real-world curse contains. If we subverted those external expectations but made it cohere with the internal expectations of the scene, we would get ‘of the women’, which is what Aristophanes writes. But for something to subvert both the external expectations and the internal expectations—and be a true para prosdokian according to definition (2)—we would have to supply something like ‘of the pickles’ or ‘of the rabbits’ or ‘of the llamas’, where there is no connection at all. Sudden references to pickles or rabbits or llamas would indeed be surprising, I admit, and they might even be funny, but they do not seem to be para prosdokian, at least according to how the term is conventionally used.⁴⁹ And fourth, if jokes such as Case 1 and Case 2 fail to count as para prosdokian according to definition (2), then it is unclear what they are instead. This is especially noteworthy given that Kanellakis frequently contrasts para prosdokian with similar concepts such as metaphor, simile, oxymoron, hyperbole, metonymy, synecdoche, antithesis, and garden-path sentences in his analyses.⁵⁰ Are Case 1 and Case 2 no different from normal speech within a play? Is it not significant at all that they subvert some set of external expectations?

Given these problems with definition (2), I submit that we should return to the standard definition for para prosdokian of Sommerstein cited above, with the added caveat that these expectations are only to be derived externally, not internally. This definition has some advantages: a joke that would be considered a para prosdokian joke if told in some non-dramatic situation would not lose its status if told in a dramatic situation, and para prosdokian jokes would be allowed to contribute to the topics under discussion in the play. Not only does this conception more correctly assess particular cases of para prosdokian in Aristophanes, it also has important consequences for understanding how these jokes operate within the larger comic sequences they are embedded within—what I call ‘comic bits’.

3. THE COMIC BIT IN ARISTOPHANES

In long-form comic works, jokes are typically embedded within larger, discrete units of humour. Contemporary comedians often use the term ‘bit’ to refer to these larger comic sequences, as in ‘Last night I did a bit about X’ or ‘I loved that comedian's bit on Y’. In her stand-up special Happy To Be Here (2018), the comedian Tig Notaro has a bit where she describes what she calls her ‘party bits’. For example, at a crowded party she will do a bit where she's looking for an imaginary dog:

Another real joy for me is to … look for my dog at a party. And I don't own a dog. Okay, so what happens is: huge crowd of people, I'm looking under tables, chairs, people's legs, and I'm like ‘Mitzi. Mitzi, come on!’ Then people start asking questions. You know, ‘What does the dog look like?’, ‘How can I help?’ I'm like ‘She's a tiny black dog. Her name's Mitzi’. Then as soon as I start seeing a lot of people's heads ducking over like that, I go ‘There she is. Come on, let's go, Mitzi’. And I walk out of the party without a dog behind me. I love looking insane.

This example uses the concept ‘bit’ in two ways: Notaro's stand-up contains a bit where she describes a situation in which she (allegedly) uses bits in real life. Improv comedy, especially short-form improv like the television show Whose Line Is It Anyway?, utilizes bits as well. Here, the audience's suggestions before the scene are important for delimiting the parameters of the bit that will be improvised on stage. For example, the audience might be asked to shout out an occupation (‘dentist!’) and a feeling (‘boisterous!’), and then the improvisors create a bit about a boisterous dentist. Even without prompts from the audience, improv bits can emerge naturally as soon as multiple jokes arise on a topic, and the improvisors build upon each other's contributions to advance the bit. Bits are equally at home in sketch comedy (e.g. Monty Python's Flying Circus, Saturday Night Live), which involves a comic exploration of a particular concept, character, or situation, where the distinctive nature of that concept, character, or situation forms the substance of the bit. Despite the prevalence of the term in contemporary comedy, no standard definition exists. But preliminarily, we can define a bit as follows—a bit is a discrete unit of comedy, situated within a larger comic performance or literary work, in which the material is based on some central premise, where the humour revolving around that central premise develops and escalates over the course of the bit, often through a series of interconnected jokes.⁵¹

Before considering some examples of bits in Aristophanes and how this concept relates to para prosdokian, I would like to explore some aspects of this definition more fully. As can be seen from the semantics of the word ‘bit’, bits are smaller portions of a larger comic work. Contemporary stand-up comedians often describe the smallest unit of their material as a joke, the intermediate unit as a bit and the entire unit as a set. A long-form piece of comedy might contain dozens of bits woven together to comprise the whole, and each bit might itself be comprised of multiple jokes. This conceptualizes the structure of a comic work in terms of units of humour, and this structure might differ from other ways of conceptualizing the structure of a comedy. For example, Aristophanic comedies are often given a formal structure, divided into scenes based on various metrical and theatrical criteria, including the entrances and exits of characters. Conceiving of comic bits as structural units of humour is compatible with certain other narrative approaches to Aristophanic structure, such as the so-called Great Idea, the ‘fantastic project at the heart of the play’.⁵² In a play like Women at the Thesmophoria, the Great Idea is that Euripides must acquire help in infiltrating the women's festival, since the women are angry at him for slandering them in his plays. From this initial premise, various comic bits emerge—to infiltrate the women's festival, Euripides requires someone effeminate to help, which leads to the bit about Agathon, and when Agathon says no, Aristophanes pivots to a new bit where the masculine Kinsman volunteers to infiltrate the festival. If we map the framework of comic bits onto the framework of formal comic scenes, sometimes they are coterminous and sometimes they are not—a comic scene might have multiple bits within it. This paper is not concerned with how bits contribute to larger scenes or how bits are woven together to create an Aristophanic plot; rather, I have the more modest goal of examining the internal structure of bits and exploring strategies that Aristophanes uses to develop bits, including the inclusion of para prosdokian jokes.

Bits have a beginning and an end, where the beginning of a bit is marked simply by starting to discuss the central premise, and the end of a bit is marked simply by ceasing to discuss the central premise.⁵³ Because of this, comic bits are excerptable; that is, they are able to be removed from their context within the larger work and still retain much, if not all, of their humour.⁵⁴ Since what defines a bit is the central premise, how bits are conceptualized might lead to different opinions about where they begin and end. For example, at Wasps 1–53, Aristophanes portrays Sosias and Xanthias discussing the symbolism of their dreams: Xanthias’ dream involves an eagle swooping down, grabbing a shield, and then dropping it, becoming Cleonymus in the process (Cleonymus was a punchline of many an Aristophanic joke because he famously dropped his shield in battle),⁵⁵ and Sosias’ dream involves a dragon haranguing a bunch of sheep on the Pnyx, where the dragon is not-so-subtly identified as the Athenian demagogue Cleon. Is this one bit or two? I regard this as a single bit where the premise is two slaves talking about their hilariously symbolic dreams. But someone else might treat this as two separate bits, especially if they wished to focus only on the humour of one of the dreams. There are no objective criteria for labelling a certain comic sequence a bit—either explanation of the passage above would work.

Bits develop their humour by starting with the premise and following the associations that naturally emerge from it.⁵⁶ If the premise is that women at the Thesmophoria religious festival are holding a political assembly where they will prosecute Euripides, then we would expect comic material exploring gender, religion, politics, tragedy, and the like. In addition to the associations that emerge more or less naturally from the premise, another way that humour arises is by inserting the perennial favourites of comedy—obscenity,⁵⁷ sex,⁵⁸ food and drink,⁵⁹ and personal attacks.⁶⁰ These can either coalesce with the existing humour (making an obscene or sexual joke that is on topic) or intrude more disjunctively (making an obscene or sexual joke that comes out of nowhere).⁶¹ Since this is all a little abstract, let us see how the three cases discussed above contribute to their comic bits.

Case 1 occurs within the larger scene of Women at the Thesmophoria 295–530, in which the women hold their assembly. The premise of the bit is outlined earlier when Euripides claims that the Women at the Thesmophoria have plotted against him and will hold an assembly on the question of his destruction, since he slandered them in his tragedies (82–92). The associations we would expect from this premise include religion, politics, gender, and tragedy. The bit begins with Critylla's short introductory speech (295–311), which introduces the premise through a blend of religious language (the call for ritual silence, 295) and political language (the discussion of the assembly and its benefits for Athens, 301–5). The theme of women is introduced moments later (‘concerning the demos of the Athenians and the demos of the women’, 307–8). So far, nothing terribly funny has happened; Aristophanes is merely setting the stage by blending these three strands of the premise together. The next part of the bit expands on the premise through a cletic hymn that summons various gods to grant success and favour to the women in their political enterprise (312–30). Next, we have the parody of the curse that includes the two para prosdokian jokes where ‘of the women’ is substituted for ‘of the Athenians’ (336) and ‘Euripides’ is added among the enemies of Athens (337). The function of these two jokes is to build the humour by further combining the themes associated with the premise—‘of the women’ perpetuates the blend of religion, politics, and gender that has already been established, and ‘Euripides’ introduces the theme of tragedy which had been lacking from the bit until this point. The strands of the premise are blended again and again throughout the rest of the curse, which denounces people who help establish tyrants (338–9), women who pass off another's child as their own (339–40), slave-girls who reveal their mistress's adultery and tell lies (340–2), adulterers who lie (343–4), old women who give gifts to young lovers (345), courtesans who cheat (346), and barkeepers who underpour their drinks (347–8). Note the gradual but steady intrusion of the standard comic tropes of sex and drinking into the mix, to such an extent that the final part of the curse avoids mention of the premise's associations at all and only discusses the comic trope of drinking (the barkeepers who underpour their drinks). Overall, in Case 1, the two instances of para prosdokian serve as relatively safe jokes that naturally follow from the premise, expanding that line of humour while setting the stage for humour that moves further afield (sex, drinking).

Case 2 occurs within the larger scene of Women at the Thesmophoria 209–79.⁶² The premise of the bit is that Euripides must prepare the hyper-masculine Kinsman to pass as a woman so that he can infiltrate the women's festival of the Thesmophoria. The associations we would expect from this premise include jokes about masculinity vs femininity and gender expression, and to a lesser extent, tragedy and religion. The bit begins by Euripides removing from the Kinsman the obvious signs of his masculinity. First, the Kinsman discards his cloak (213–15), signifying a demeaning loss of status.⁶³ Second, in a more extended section, the Kinsman shaves his beard (215–35), which involves four levels of escalating humour: the request for a razor from the feminine Agathon, the shaving of half the Kinsman's beard, the shaving of the rest of his beard and the examination of his clean-shaven face in a mirror. At this point, the Kinsman remarks that in the mirror, he sees not himself but Cleisthenes, a personal attack upon a politician famous in comedy for his femininity. Next, the humour moves towards removing the non-obvious signs of the Kinsman's masculinity—the hair on his buttocks needs to be singed away. This scene prompts the audience to undergo what Lowe calls ‘mentalizing’,⁶⁴ through which we not only parse the semantic puzzles contained in the jokes in the scene, we also construct narrative scenarios in our minds (under what circumstances would the Kinsman's singed buttocks be the make-or-break moment that preserved his disguise?) that include the mental states of the various characters within the play (why is the Kinsman submitting to this humiliation, and why does Euripides seem to take delight in it?) and even the state of mind of the comedian himself (what kind of twisted mind would come up with this, anyway?). Within this context we get jokes that blend gender and food (‘damn the luck, I'm going to be roast pig!’, 237) or gender and obscenity (‘watch out for the tip of your dick’, 239). Finally, the moment comes when the Kinsman's buttocks is singed, and he exclaims, ‘I'll watch out, all right—only I'm on fire! Oh no, no! Water! Water, neighbours, before somebody else's arse catches fire!’ (241–2). This complex moment brings together several different strands of humour that had been developing. It continues the feminization and humiliation of the masculine Kinsman and escalates it to the point of physical pain. This point is punctuated by the Kinsman's cries for water to help alleviate his pain, though, of course, this help will never come. The call for help is addressed to his ‘neighbours’, which presumably refers to the audience itself; this breaks the fourth wall and adds a metatheatrical immediacy to the moment. And once Aristophanes has established that the Kinsman's arse is equivalent to a house in that both can be on fire, the incongruity between the two can be expanded and elements of one can be imported onto the other, which results in the fear that the fire will spread to other people's arses, on the model of fire spreading to other people's houses. Furthermore, the para prosdokian joke itself includes an obscenity (πρωκτόν), which blends the two types of humour I described above—humour following from the premise and humour from stock comic tropes. Here, the para prosdokian contributes to the humour by capping this portion of the bit and paving the way for further expansions of the humour, such as forcing the Kinsman to wear women's clothes (249–65) and having him effect a woman's voice (266–8).

Case 3 occurs within the larger scene of Ach. 393–488. The premise is that Dicaeopolis must speak publicly to defend himself against Cleon's slanders and to improve his chances of success, he decides to borrow a ‘guise most piteous’ (384) from Euripides. The associations we would expect from this premise include tragedy, costume, and rhetoric. The bit begins with Dicaeopolis knocking on Euripides’ door, which is answered not by Euripides but his slave, who speaks in highly Euripidean tragic language (‘he's home and not at home, if you get my point’, 396).⁶⁵ When Euripides comes out to speak with Dicaeopolis, he emerges not through a door but upon the eccyclêma (407–9). Dicaeopolis takes this humour focussing on tragedy and incorporates costume (‘And why do you wear those rags from tragedy, a raiment piteous? No wonder you create beggars!’, 412–13) and rhetoric (‘I've got to make a long speech to the chorus, and if I speak poorly, it means my death’, 416–17). Next comes a joke-sequence about various rag costumes from his previous plays that Euripides possesses. Blending humour about tragedy, costume, and rhetoric, Dicaeopolis rejects the rags of Oeneus, Phoenix, Philoctetes, and Bellerophon as not piteous enough before finally selecting the rags of Telephus (418–34). Here, Aristophanes pushes the relationship between tragic costumes and rhetoric. Previously, the idea was that a piteous tragic costume would lend an air of authority to Dicaeopolis’ speech that might make his speech more effectual; now, the idea is that the piteous tragic costume actually transforms Dicaeopolis into a better speaker (‘how I'm filling up with phraselets already!’, 447). And if the tragic costume can provide such benefits, then Dicaeopolis realizes that he might be even more rhetorically effective if he acquires other tragic paraphernalia: a cane, a basket, a goblet, and a bottle (448–63). At this point, Euripides speaks the para prosdokian joke: ‘Fellow, you will deprive me of my tragedy!’ (464), and Dicaeopolis leaves after taking one last prop, ‘some withered greenery for my basket’ (469). The effect of the para prosdokian in this passage is to drive home the main point—tragic costumes and props are the substance of Euripidean tragedy, and without them, Euripides has been deprived of what makes his art successful. Dicaeopolis has finally acquired exactly what he was hoping for, a guise most piteous that would make him a better speaker. The extra joke in all of this is that by acquiring these things from Euripides, he has also deprived Euripides of their use, reducing his artistic abilities down to nothing or, as Euripides says, ‘Gone are my plays!’ (470).

In these three cases, para prosdokian jokes occupy different locations within the bit. In Case 1, it appeared towards the beginning and helped frame the rest of the bit about the women holding a political assembly; in Case 2, it appeared in the middle, capping the portion of the bit dealing with the arse-singeing and leading into the cross-dressing; and in Case 3, it appeared at the end, highlighting the main parodic critique of the entire bit, that Euripides’ tragedies are nothing but fluff once deprived of their costumes and props. However, all of them serve an important purpose within their bit—they contribute to the humour by conjoining various themes associated with and emerging from the premise. para prosdokian jokes work by pointing to the external world, and by setting up a relationship between something within the play and something out in the world, the comedian can utilize this relationship to increase the humour. In Case 1, it reinforced some of the primary associations of the premise (gender, religion, politics, and tragedy). In Case 2, it reinforced the premise's primary associations while also adding metatheatrical (‘neighbours’) and obscene (‘arse’) components to augment the humour. In Case 3, it clarified a latent metaphor underlying the humour—Euripides’ costumes and props are his personal property, and the art of his tragedy consists entirely in these tragic paraphernalia.

If we adopt my conception of para prosdokian, then two other aspects of Kanellakis’ view must be modified. First, after surveying numerous examples, Kanellakis explores the norms of para prosdokian in its typology, function, and usage. Since these conclusions, which are often statistical in nature, depend on his analysis of what counts as para prosdokian, many of these conclusions would have to be revised.⁶⁶ Second, Kanellakis distinguishes between ‘standard’ and ‘non-standard’ uses of para prosdokian, where his own view is standard and the views of those with whom he disagrees is non-standard.⁶⁷ Kanellakis is careful to state that by these terms he is only referring to the ‘statistically dominant model’, though with words such as ‘standard’ and ‘non-standard’, it is difficult to avoid making additional implicit normative and evaluative claims. Even if we accept this, it is unclear that Kanellakis’ account remains the statistically dominant model if some of his data about para prosdokian need to be reclassified.

In this paper, I have suggested, first, that we need to consider the nature of the expectations behind para prosdokian jokes; second, that there is a substantive difference between internal and external expectations; and third, that para prosdokian jokes should be understood as only subverting external expectations. Furthermore, I have introduced the concept of the comic bit, which, I submit, is applicable and useful for the analysis of Aristophanic humour more broadly. para prosdokian jokes, when embedded within comic bits, cohere with the internal expectations of the play. This does not disqualify them as para prosdokian jokes as definition (2) might suggest; rather, this cohesion is crucial to how para prosdokian jokes function within larger sequences of humour. para prosdokian jokes can combine various topical strands that emerge from the premise, weave them together in new and exciting ways, incorporate other sources of humour such as obscenity or drinking, and contribute to the narrative plot of the play or larger extra-dramatic discourses. These are no mere jokes, but important components of Aristophanes’ comedy that highlight his artistic skill and dramatic technique.