^* I should like to acknowledge here the valuable help given to me by Hans-Joachim Newiger, Michael Silk, and Alan Sommerstein, who were kind enough to read this paper and offer their comments and constructive criticism; my thanks also go to the anonymous referee of the Journal. Needless to add that they do not share all my views, and have no responsibility for my mistakes.

¹ Aristotle mentions but does not define these terms in relation to comedy, Poet. 1449b 4, Rhet. 1415a 9–22, and EN Viv. 2, 1123a 23.

² Using various forms of the verb: Ar. Ach. 629, Eq. 508, Pax 735, Thesm. 785; Piaton, fr. 92 Kock.

³ Hephaistion (p. 72) Consbruch; Heliodoros survives in metrical scholia to Aristophanes: Ach. 626, 659, Eq. 498, 503, Nu. 518, 1115, Vesp. 1009, Pax 733, Av. 1058, Ran. 675, 686, reproduced by White, J. W., The verse of Greek comedy (London 1912) 397–421Google Scholar (White on Heliodoros, 384–395); cf. Pollux, iv 111; Platonios, p. 4 Kaibel; Plutarch, Mor. 711f.

⁴ The Tractatus Coislinianus already mentions prologue, khorikon, episode, exodos (xvii, p. 40 Janko); episode is defined by Suidas (E 2144 Adler), and other lexica (Et.M. 356. 28; Bekker, Anecd. Cr. i. 253, 19) as ‘what is adduced to the play for the sake of laughter, being outside the plot;’ and by Suidas (E 2143) and Et.M. 356. 34, as ‘what is introduced to the plays as an addition and an expansion of the play.’ This recalls what Aristotle says about epeisodiodes at Poet. 1451b 33–34, and episode at 1455b 23, but the definition is hardly Aristotelian for that.

⁵ First observed by Westphal, R. (Rossbach, A. and Westphal, R., Metrik der Griechen im Vereine mit den übrigen musischen Künsten, ii: Griechische Metrik [Leipzig² 1868] 401 ff.)Google Scholar, the epirrhematic agon was extensively treated by Zieliński, T. in his book on Die Gliederung der altattischen Komödie (Leipzig 1885)Google Scholar, which justly became the foundation of all subsequent work on the ‘Bauformen’ of Old Comedy.

⁶ Any bibliographical selection here, however limited, would be far too long to be contained in a footnote. I can do no better than refer to Newiger's, bibliography in Aristophanes und die alte Komödie, (Darmstadt 1975).Google Scholar A. C. Quicke's unpublished thesis, Aristophanes and Athenian Old Comedy. A survey and bibliography of twentieth-century criticism, with an essay on the current state of bibliography in classical studies, submitted for the Fellowship of the Library Association, 1984, consists of 533 pages and lists 2398 items, with a brief and usually very clear summary of their contents (it is available for consultation in the Institute of Classical Studies, London).

⁷ I should like to mention in this connexion Landfester's, M. study on Handlungsverlauf und Komik in den frühen Komödien des Aristophanes (Berlin 1977)Google Scholar, and A. H. Sommerstein's felicitous concept of ‘functional structure’ of Aristophanic comedy, which was put forward in the introduction to his edition of Acharnians ([1980] 11–13) and anticipates the sequence of ‘functions’ discussed in this paper.

⁸ ‘Towards a modern poetics of Athenian Old Comedy,’ Métis, iii (1988), 53–67.

⁹ Perhaps I should not fail to declare at the outset my allegiance to Aristotle, both as a historian and a theoretician of drama. I believe that only when we have made every effort to understand the implications of what he says about poetry and drama (not only in the Poetics) should we introduce our own conjectures and theorizing.

¹⁰ 1st ed. trans, by L. Scott (1958), 2nd ed. rev. by L. A. Wagner, with an introduction by A. Dundes, (Austin 1968). Propp's book has now been translated into eleven European languages and Japanese.

¹¹ A modern definition of morphology as a branch of biological sciences can be found in Encycl. Britannica (15th ed., vol. 11, s.v.): ‘Morphology is a term used in biology for the study not only of shape and structure in plants, animals, and microorganisms but also of the size, structure, and relationships of their parts. The term morphology connotes the general aspects of biological form and arrangement of the parts of a plant or an animal.’

¹² ‘La structure et la forme. Reflexions sur un ouvrage de Vladimir Propp,’ Cahiers de l'Institut de Science Économique Appliquée, série M, vii (1960) 1–36, reprinted in the author's Anthropologie structurale, ii (Paris 1973) and its English translation, Structural anthropology, ii (New York, 1976).

¹³ Propp's reply was written at the instigation of the publisher of Morphology's Italian translation (1966). The quotation is from the English translation of this text, published in Propp, V., Theory and history of folklore, trans. Martin, and Martin, , ed. Liberman, A. (Minneapolis 1984) 69–70.Google Scholar

¹⁴ Ibid., 73–74.

¹⁵ Logique du récit (Paris 1973).

¹⁶ Reference must be made here to E. M. Meletinskij's study, ‘L'étude structurale et typologique du conte,’ published in Propp, V., Morphologie du conte (Paris 1970)Google Scholar and, in English, in Genre iv (1971) 249–279 (originally, a supplement to the second Russian edition of Morphology [1969]). Also, very useful is Liberman's Introduction to Propp's Theory… (see n. 13). Among recent contributors to the study of ancient mythology, Burkert, W. has made good use of Propp's analytical principles in his Structure and history in Greek mythology (Berkeley 1979)Google Scholar.

¹⁷ Strangely, most of Propp's critics and successors tried to simplify his scheme and make it more abstract rather than discover variations or different function patterns, which might have been a more productive line of research.

¹⁸ Plato, Rep. 392d ff., Aristotle, Poet. 1447a 16 ff.

¹⁹ It may be useful to recall in this connexion that translation from narrative to drama has been extremely common throughout the history of dramatic art, from Greek tragedy to modern cinema and TV dramas (most often based on modern novels, short stories, old tales, etc.). The opposite is also true, though much less common, witness Charles and Mary Lamb's Tales from Shakespeare or Brecht's Threepenny novel. The affinity of tale and drama is already highlighted by Plato's calling Homer the first of tragic poets (Rep. 607a).

²⁰ Theory and history of folklore, 81.

²¹ The term syntagmatic is used here in both its linguistic and archaeological senses (a linear order of independent forms standing in relation to each other).

²² In naming the functions of comedy I have tried to keep as close as possible to Propp's terminology. ‘Villainy,’ of course, in relation to comedy should be taken with a grain of salt; but being an active term it is usefully differentiated from ‘lack’ and ‘misfortune’ (or ‘predicament,’ suggested to me by Sommerstein), which are passive.

²³ ‘Popular fiction is fond of far away places; and fifth-century Greek comedy is no exception. Birds, Thesm., Frogs, Plut., all begin with two people going somewhere,’ Handley, E. W. in The Cambridge history of classical literature i (1985) 379.Google Scholar

²⁴ Structurally speaking, a negative triumph is the exact opposite of the canonical ending. It is not the same as the punishment (of villain) on Propp's scheme, but occurs in comic forms of folk drama, such as the modern Greek shadowpuppet theatre, in which the comic hero, Karagiozis, is often beaten up at the end of plays (but he may also, though less often, be treated to a sumptuous dinner; see my Παραδοσιακὴ δραματουργία τοῦ Καραγκιόζη, Athens, 1984). It should be noted, however, that the ending of Clouds, as we have it, is that of the revised (and apparently never produced) version of the play; cf. Dover who suggests that the new ending, too, provided ‘the customary noise and movement’ (Clouds, pp. xxiv f., xciii f.)

²⁵ Kleinknecht, H., ‘Die Epirphanie des Demos in Aristophanes' “Rittern”’, Hermes lxxvii (1939) 58–65Google Scholar (= Newiger [ed.], Aristophanes und die alte Komödie 141–154). See also Kleinknecht's treatment of Av., ‘Zur Parodie des Gottmenschentums bei Aristophanes,’ ARW xxxiv (1937) 294 ff.

²⁶ Cf. Dover, Clouds liii.

²⁷ In a discussion of dramatic abstractions, reference should be made to Souriau, Étienne, Les deux cent mille situations dramatiques (Paris 1950).Google Scholar Souriau's strange title refers to the theoretically possible situations resulting from the interactions of what he calls ‘fonctions dramaturgiques,’ or roles, represented by dramatic characters and forming a ‘system of forces.’ All possible characters, he suggests, in all kinds of drama, correspond to six such functions or forces (to which he attributes astrological names and symbols): ‘Le Lion: la Force thématique,’ ‘le Soleil: le Représentant du bien souhaité, de la Valeur orientante,’ ‘la Terre: l'Obtenteur virtuel de ce bien,’ ‘Mars: l'Opposant,’ ‘la Balance: l'Arbitre de la situation, attributeur du bien,’ ‘la Lune: l'Adjuvant, redoublement d'une des forces précédentes” (83–117).

²⁸ Tract. Coisl. xii; Aristotle, EN 1108a 21–26, 1127a 13–1128b 3, Rhet. 1419b 2 ff.

²⁹ In his well known book on Aristophanes and the comic hero (Cambridge, Mass. 1964). The comic hero has found considerable acceptance ever since, cf. Henderson, J., The maculate muse (New Haven 1975) 13Google Scholar; Janko, R., Aristotle on comedy. Towards a reconstruction of Poetics II (London 1984) 218.Google Scholar

³⁰ I hope to deal with comic types and characters more fully on another occasion.

³¹ Or else we shall continue to believe, with Dover, that Philokleon ‘seems to be the kind of man (fortunately, there are not many of them) who as a soldier loots from unarmed civilians but keeps clear of an armed enemy, and in old age spends his days in the infliction of pain on others and his evenings in running his hand up his daughter's skirt. If we still like him, why do we? Is it that Aristophanes, by some dramaturgical skill which resists analysis, has compelled us to like him? Or is it that we, through a deficiency of imagination which prevents us from reacting to fictitious creations as we do to the people and situations encountered in our own lives, have misunderstood Philokleon and his creator's intentions?’ (Aristophanic comedy, 127). What seems to me to be Dover's difficulty here is that he judges Philokleon as if he were a psychologically consistent creation by a realist author such as Ibsen. The reason he and we like Philokleon is that the man belongs to such an established and likeable type that his aberrations, mentioned in various parts of the play, cannot upset this basic fact nor amount to a negative moral portrait. He is much closer to Demos than to Dikaiopolis or Trygaios; he has no moral purpose but is a misled old man on the verge of dotage; hence he is naughty and in need of being reformed. In other words, all these characters are variants of the same type but carriers of different functions and roles: Dikaiopolis and Trygaios are seekers while Demos and Philokleon are objects of quest. Philokleon is likeable by virtue of belonging to the same type as many a comic hero but as his sphere of action is different from that of the imaginative and energetic seekers he shows certain ugly sides of his character, which are not enough to differentiate and detach him from the company of other type variants and their generally congenial behaviour.

³² Dearden, C. W., The stage of Aristophanes (London 1976) 128–42Google Scholar, has tried to correlate Aristophanic characters with the archaeological evidence but his results are very tentative because he did not employ a strict method to elicit the type variants from the texts. Largely impressionistic is the classification of types by McLeish, K., The theatre of Aristophanes (London 1980)Google Scholar, who uses such descriptions as spoudaios and poneros, the latter in its Modern Greek sense, without reference to Whitman — which is amazing.

³³ First published as a supplement to Jahresbericht der Deutschen Schulen zu St. Annen in St. Petersburg, reprinted in the author's collected papers under the title Iresione, I. Dissertationes ad comoediam et tragoediam spectantes continens (Eos, Suppl. 2 [Lwów 1931]) 8–75.

³⁴ The hero marrying the king's daughter is a canonical folktale ending but Basileia is nowhere explicitly called daughter of Zeus. To Pgrstheegriros' question, ‘Who is Basileia?’ (1537), Prometheus replies that she is ‘A maid most fair, who keeps in charge for Zeus / his thunder and all his royal attributes, / law, counsel, discipline, naval estimates, etc.’ (Murray's translation, published in 1950), and the chorus call her Zeus’ πάρεδρος (1753). So she does sound like a clear case of personification of ‘die Weltherrschaft des Zeus,’ in Kock's words (his ed. of Av., on 1864), although the form Βασίλεια means Queen, Princess, ‘Sovereign Bride’ (Murray) and not Monarchy or Sovereignty (Βασιλεία, with a long ending). An enormous amount of effort has been directed to speculating about her mythological identity, and Zieliński was neither the first nor the last scholar to support Athena's candidacy for serving as Pisthegriros' bride (cf. Newiger, , Metapher und Allegorie [Mgrnich 1957] 100–102Google Scholar), which seems to me impossible for two reasons: (a) To have Athena, the virgin goddess, appear on the stage in a role that should feature or connote, according to the expectations of the audience, a courtesan would amount to ridiculing the protectress of the city and committing a religious offence that was beyond comedy's licence; (b) Basileia is not called daughter of Zeus, and has been taken as such only because our expectations demand a king's daughter as the proper reward of the tale's triumphant hero, who is to become the new king.

³⁵ Such comedies were Kratinos' Ploutoi, Telekleides' Amphiktyones, Krates' Theria, Pherekrates' Metallês, Metagenes' Thouriopersai, and Nikophon's Seirenes.

³⁶ Mythical fantasy in comedy, then, is the same as in myth, described by Kirk as follows: ‘[Mythical] fantasy deals in events that are impossible by real-life standards; but in myths it tends to exceed the mere manipulation of the supernatural and express itself in a strange dislocation of familiar and naturalistic connexions and associations. That is what Cassirer meant by referring to the mythical concept of causality (…) The dynamic aspect of mythical fantasy is both more revealing and more typical. It includes more than causality; indeed, all the rules of normal action, normal reasoning and normal relationships may be suspended and distorted’ (Myth, 268–269). If this description does not sound quite clear it is because of the author's effort to reduce fantasy to some kind of principle. The same is attempted by Burkert (n. 16) 19–20, who even denies that the element of the fantastic (‘in the sense of the “impossible”, from our point of view’) is indispensable in myth and even in fairy-tale; but he goes on to speak of ‘action inverted by logic,’ and of ‘logic disregarding reality;’ ‘a good tale,’ he claims, ‘overdetermined and “crystallized”, may just be too logical to be true.’ I am afraid Burkert's ‘logic disregarding reality’ is too much of a simplification, which will not explain away the constant of mythical causality in folk narratives. That this causality may not be unprincipled or haphazard is very likely, but scholarship has still a long way to go before it can discover the ways it works.

³⁷ For an application of various kinds of structural analysis to Greek myths and discussions of methodological matters see Gentili, B. and Paioni, G. (eds.), Il mito greco (Rome 1977).Google Scholar

³⁸ The legend of Perseus (London 1894–1896).

³⁹ Theory and history of folklore, 80.

⁴⁰ In Pickard—Cambridge, Dithyramb, tragedy, and comedy (²1962) 193.