Hostname: page-component-797576ffbb-pxgks Total loading time: 0 Render date: 2023-12-07T08:02:15.262Z Has data issue: false Feature Flags: { "corePageComponentGetUserInfoFromSharedSession": true, "coreDisableEcommerce": false, "useRatesEcommerce": true } hasContentIssue false

The number of speaking actors in Old Comedy

Published online by Cambridge University Press:  11 February 2009

Douglas M. MacDowell
University of Glasgow


The number of speaking actors in Old Comedy has been much discussed, but no consensus has been reached. The old assumption that the number was three, as in tragedy, was shaken when it was realized that some scenes of Aristophanes have four characters on-stage at once, all taking part in the dialogue: for example, in Lys. 77–253 we have Lysistrate, Kalonike, Myrrhine, and Lampito, and in Frogs 1414–81 we have Dionysos, Aiskhylos, Euripides, and Plouton. Rees therefore argued that there was no fixed number, but that view was not generally accepted. A more widely held view is that there were three principal actors with additional performers for small parts. However, there is no evidence contemporary with Aristophanes which distinguishes three actors from the others in this way, and it is probable that writers of later periods who mention three actors are referring to their own times and did not have authentic information about the fifth century. The passage which DFA, p. 149, seems to regard as the most trustworthy is in a brief account of comedy attributed to Tzetzes: πιγενμενος δ Κρατῖνος κατστησε μν πρτον τ ν τῇ κωμῳδᾳ πρσωπα μεχρ τριν, στσας τν ταξαν4DFA paraphrases this as ‘Cratinus reduced the disorderliness and, in some sense, fixed the number of regular actors at three’. But πρσωπα means ‘masks’ or ‘characters’; it does not mean ‘actors’ (for which the Greek word is ὑποκριτα). What the writer meant by saying that Kratinos settled the masks or characters in comedy at ‘up to three’ is not clear, but his statement is useless as evidence for the number of actors.

Copyright © The Classical Association 1994

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


1 I use the following abbreviations for the principal recent discussions. DFA = Pickard-Cambridge, A. W., The Dramatic Festivals of Athens, second edition revised by Gould, J. and Lewis, D. M. (Oxford, 1968Google Scholar; reprinted with addenda, 1988). Dover, Clouds = Dover, K. J., Aristophanes: Clouds (Oxford, 1968CrossRefGoogle Scholar. Dover, Comedy = Dover, K. J., Aristophanic Comedy (London, 1972Google Scholar). Henderson, Lys. = Henderson, J., Aristophanes: Lysistrata (Oxford, 1987)CrossRefGoogle Scholar. Russo, Aristofane = Russo, C. F., Aristofane autore di teatro (Florence, 1962Google Scholar; reprinted with addenda, 1984). Thiercy, Aristophane = Thiercy, P., Aristophane: fiction et dramaturgic (Paris, 1986).Google Scholar

2 Rees, K., The So-called Rule of Three Actors in the Classical Greek Drama (Chicago, 1908).Google Scholar

3 Russo, , Aristofane, pp. 150–5Google Scholar; DFA, pp. 149–53. This view is held in its most elaborate form by Thiercy, , Aristophane, pp. 4067Google Scholar. He maintains that the three actors formed a hierarchy, with the Protagonist playing the most important parts, the Deuteragonist the next most important, and the Tritagonist the less important, and that a character who speaks much in one scene and little in another scene (such as Pheidippides in Clouds) will accordingly have been played by different actors in the different scenes. No evidence supports this.

4 Now printed, without the attribution to Tzetzes, in Scholia in Aristophanem, pars I fasc. IA: Prolegomena de Comoedia (ed. Koster, W. J. W.; Groningen, 1975), p. 14Google Scholar. On this passage see also Hourmouziades, N. C., GRBS 14 (1973), 186–7.Google Scholar

5 Dem. 21.56–61. For discussion of the law see MacDowell, D. M., Demosthenes: Against Meidias (Oxford, 1990), pp. 276–7.Google Scholar

6 On the question of when the Envoy exits, see Dover, K. J., Maia 15 (1963), 89Google Scholar, reprinted in his Greek and the Greeks (Oxford, 1987), p. 290.Google Scholar

7 DFA, p. 150; cf. Dover, , Comedy, p. 27Google Scholar and Maia 15 (1963), 9Google Scholar n. 8, reprinted in Greek and the Greeks, p. 290 n. 8.

8 Russo, , Aristofane, pp. 226–7Google Scholar, suggests that the roles of children were taken by boys who were specialists in singing. But the Megarian's Daughters do no singing.

9 Henderson, Lys. p. 177.

10 Russo, , Aristofane, pp. 139–40Google Scholar, and DFA, p. 150 n. 1, comment on lines 1254–6; Dover, , Comedy, p. 94Google Scholar, comments on line 234; Thiercy, , Aristophane, p. 43Google Scholar, comments on both those passages.

11 Here for convenience I use the names Demosthenes and Nikias for the two slaves who appear at the beginning of the play, without entering into discussion of the question how far they are to be identified with the generals of those names. At any rate 54–7 implies some degree of identification of one of them with Demosthenes.

12 Rogers, B. B., The Knights of Aristophanes (London, 1910), p. 168.Google Scholar

13 Russo, , Aristofane, pp. 139–40.Google Scholar

14 Dover, , Clouds, pp. lxxvii, xcii–xciii, 208Google Scholar, and Comedy, pp. 26–7.

15 Thiercy, , Aristophane, p. 45Google Scholar, suggests that Strepsiades exits at 888 and reappears at 1105, enabling this actor to play the other Argument; this means postulating the loss of another choral song between 1104 and 1105. The partly-revised state of Clouds makes it impossible to rule out this suggestion, but the text as we have it does not indicate that Strepsiades exits here.

16 Dover, , Clouds, pp. lxxvii, 266–7.Google Scholar

17 See the commentaries of Platnauer, M. (Oxford, 1964)Google Scholar and Sommerstein, A. H. (Warminister, 1985) at line 1210.Google Scholar

18 DFA, p. 151, is wrong to call them ‘nonsense’

19 An anonymous referee for CQ objects that Skythian archers could not do this work because they would be carrying weapons: bows and arrows, and possibly whips and swords (cf. Thesm. 933, 1125–7). But this is unconvincing; they can easily lay their weapons down, and it is much more likely that they do so than that the Proboulos labours to save them the trouble.

20 This expression, almost exactly repeated in 441 and 445, does not mean that he does not know where the archers are. πο is used with a person in the nominative as an order, meaning ‘Come here!’, e.g. Lys. 184, 1114, Clouds 633, Wasps 935, 976, Peace 1295, Birds 353, 863, Frogs 1305, Ekkl. 734. With a thing it means ‘Bring–!’, e.g. Wasps 995, Peace 1059.

21 There is no strong reason why the speakers of 439–40 and 443–4 should not be respectively Kalonike and Myrrhine, the same speaking characters who entered the Akropolis with Lysistrate at 253. It is no real objection that one of them is later addressed as an old woman (506 ὦ γρα), since the Proboulos there is being rude to her. We can, if we wish, imagine Kalonike as being in her forties; she still enjoys sex and smart dressing (51, 133–5, etc.), but she is probably older than Lysistrate, whom she addresses affectionately as ‘child’ (7). So I assume that the speakers of 439–40 and 443–4 are Kalonike and Myrrhine, as do Rogers and Coulon in their editions. But it makes no difference to the rest of my argument if any reader prefers to assume, with Henderson and Sommerstein in their editions, that they are two other women.

22 Henderson, , Lys. pp. xlii, 117, 123Google Scholar. On p. 126, and earlier in ZPE 34 (1979), 31Google Scholar, Henderson even envisages the presence of a ‘troop’; but that is incompatible with the Proboulos' complaint in 449 that he has run out of archers.

23 Dover, , Comedy, p. 167Google Scholar, rightly rejects the possibility that the herald is the leader of the chorus.

24 Cf. Russo, , Aristofane, pp. 153–4Google Scholar; DFA, p. 152; Dover, , Comedy, p. 27.Google Scholar

25 Plouton must appear by 1414 at the latest. In CQ 9 (1959), 261–2Google Scholar, I postulated the loss of about three lines including an announcement of his arrival at that point, but Dover, , Aristophanes: Frogs (Oxford, 1993), p. 295Google Scholar, argues that he is present from 830 onwards.

26 Thiercy, , Aristophane, p. 49Google Scholar, accepts Bergk's conjecture that a choral interlude has been lost from the text after 1111, enabling one of the actors who exit at that point to change into the Slave. But there is no manuscript evidence for this, and it should probably be rejected; cf. Vetta, M., Aristofane: Le Donne all'assemblea (Milan, 1989), p. 267.Google Scholar

27 DFA, p. 153, attempts a distribution of all the parts in this play to only three actors, assuming that the part of Wealth was divided between two actors, but admits that a four-actor distribution is more likely.