It was a delight to discover an article placing great importance for Aristotle on the too-often-ignored aspect of tragedy in the Dramatics, opsis, which Sifakis says means either “spectacle or the appearance of the actors” (p. 2). He notes the difficulties of the last 90 years (which is surely an understatement because he could have said 450-1000 years) in understanding the treatise and in part blames (correctly in my view) “misunderstandings firmly established by unreflective repetition” (p. 2). He adds:
One such well-entrenched notion is the often repeated or implied assumption, shared by Aristotelians as well as by historians of drama, that Aristotle did not care about and actually underestimated the theatrical performance of tragedy. The basis of this assumption is the apparent contradiction between the inclusion of opsis...in the six qualitative or formative parts of tragedy (Poet. ch. 6, 1449b 31-32) and the fact that a little later, in the same chapter, opsis is called atechnotaton (totally nontechnical) and unrelated to the art of poetry. This statement is further qualified by Aristotle's assertion that the potential of tragedy exists even without performance and actors, not to say that as far as the execution of the visual aspects of performance is concerned “the art of the mask-maker is more essential than the art of the poets” (1450b 16-20).
Sifakis then, for practical reasons, focusses only on two competing interpretations, by Stephen Halliwell (1998) and earlier publications by (Sir) Oliver Taplin. Halliwell defends the appreciation of performance for Aristotle to a certain extent and has defended it more against Taplin in his newer work “Aristotelianism and anti-Aristotelianism in Attitudes to Theatre” (2003), which Sifakis seemingly is unaware of (cf. ADMC pp. 285ff). However, Halliwell says in the newest work that the performance is purely imaginary for the Northern Greek, keeping the traditional view that tragedy is merely or primarily literature (a view rigorously rebutted in ADMC Chapters 2-3). Like so many others over centuries, Taplin castigates Aristotle for not appreciating performance at all, based primarily on the end of Dramatics 6, in a passage that Sifakis examines and that we re-examine below. Suffice it to say here concerning Halliwell's newest work that in no way does Halliwell rebut Taplin when Halliwell says that the performance is merely imaginary. Taplin is not displeased with the ancient Northern Greek because performance could be imaginary but because Taplin believes that the Dramatics denigrates the real attributes of dramatic performance, which Greek culture clearly considered very important.
Sikafis, as he begins to evaluate Taplin and the related topics, writes:
And surely I cannot deny the obvious, namely, that Aristotle clearly states that “the potential of tragedy exists without public performance and actors” (p. 4).
We can, however, and should deny that the obvious is obvious. The Greek is agōnos and hupocritōn, and the better translation is “competition and the (special) actors (who were given certain privileges if they had previously won).” The real meaning of the passage is that the potential of tragedy exists without the competitions and all the expensive spectacle that the state/producers paid for. The effect could be given by the other five elements in modest performance or with Thespian-type travelling “drama wagons.” This is a far cry from interpreting the Greek so that no performance is necessary in Aristotle's conception of tragedy as defined in Chapter 6, especially considering that at least twice the Stagirite says in the chapter that all tragedies have the six necessary conditions! In brief, Sifakis yields on a point that completely supports his view, which is that spectacle is important for Aristotle, just not as important as the five other “qualitative” parts.
Starting with p. 5, Sifakis translates melopoiia as “song composition” and lexis as “diction.” Unfortunately, this translation is another of those “misunderstandings firmly established by
unreflective repetition.” The terms, which are very equivocal, in this chapter really mean “music-dance composition” and “speech (or language)” respectively, and it is either shocking or sad or both that dance just disappears from the chorus for Sifakis, although he redeems himself if only a little at the end of his article in his discussion of Chapter 17, where the dramatist should be concerned also with schêmata, which Sifakis acknowledges could be dance. A cognate of this term is used in Chapter 1 when Aristotle speaks of the corps de ballet impersonating even character, passions, and actions with their gestured “ordered body movement,” or, for short, gestured dances (schēmatizomenōn rhuthmōn).
How does Sifakis glean on p. 5 that what I call the 2-1-3 pattern (for the means, manner, and objects of mimesis) refers to what he states—melopoiia and lexis are the two means, spectacle is the single manner, and the other three necessary conditions of all tragedy are the objects, namely, plot, character, and thought? Determining for sure why, e.g., plot is not the manner and proving manner is spectacle took me part of a PhD dissertation (1992) and then a summary in my article “The Poetics of Performance” (Cambridge University Press, 1999). Still, to his credit, Sifakis arrives at the right answer, however he determined it.
On p. 5, Sifakis also mentions in passing the Stagirite's “explanation of the strange (for us) metaphor of 'seasoned language' used in the preceding definition” and he gives an enlightening bit of history (in Endnote 14) that I myself had missed, to wit, “The same metaphor had already been used by Plato: hēdusmenēn mousan (Rep. 607a 5).” Rather strangely, Sifakis never follows up on Aristotle's meaning of the phrase of explanation, namely, that seasoned language is “language having rhuthmos kai harmonia kai melos.” Getting this phrase right is crucial to understanding Aristotle adding a cognate of poiēsis to melos to arrive at melopoiia later in the chapter. Diotima had explained that poiēsis had 2 senses, “composition” (or making) in general and more appropriately (at least in the context of orchestral art) mousikē kai metra. Here the -poiia just means “the composition of,” and hence melopoiia is “the making of the music-dance.”
Sifakis astutely recognizes that the three objects of mimesis are the top ranked elements of the six (necessary) ones: Plot, character, and thought. Then come the two means, speech and melopoiia, and finally opsis (but it is still necessary). He properly is surprised (given the traditional interpretation that tragedy is “poetry”), because speech is only fourth in the ranking. As he explains very briefly or, one might say, too briefly:
What is remarkable at first sight is not that opsis gets the lowest ranking, but the fact that, in a discussion of the qualitative elements of tragedy as a poetic genre, language is ranked fourth (50b 12) (p. 6).
This poses a dilemma for anyone taking the traditional, literary view of tragedy but not for my own position. Because plot can be done with mere action (that is, performers on a stage), language has to be ranked further down the list. What is completely open, though, for further examination and publication by future scholars is why speech is below character (and thought), despite the attention that has been paid to this chapter for over 450 years. Again, Sifakis's perspicacious phrase comes to mind: “misunderstandings firmly established by
unreflective repetition.”
Sifakis then gives the wrong reason for why character is second to plot:
tragedy dealt with traditional myths in which things were set in motion by the will of gods, with which the human characters tried to cope; the latter, however, did not instigate the action as they normally do in modern drama (p. 7).
I believe there is a better reason, in part because the Stagirite did not care a jot about modern drama and in part because he cites with approval the play that Agathon created, Antheus, which was completely original. Aristotle adds immediately that dramatists need not stick to the traditional stories (9.1451b21-22). The better reason why plot is above character is that
character requires choice for the Northern Greek of a particular sort (dealing with values and ethics). Just as one could have a film with a plot, but with car-chases and other types of action that show no real choice on the part of the actors (choosing to turn right or left is not a choice of character but of driving), likewise with drama. The reason plot could exist on its own is that it could be done with acting, pantomime, or dance. That does not entail in and of itself that plot has the other necessary elements for the type of tragedy that satisfies all six necessary conditions. Again, Aristotle does not state the reasons why the 3rd-5th elements are given their precise rank, and I myself occasionally wonder about the issue, with no settled opinion. Language may be above music-dance because of what Aristotle says in Chapter 4 about Aeschylus diminishing the chorus and giving more priority to dialogue, but why would thought be above language unless the Stagirite appeals to his view that language is an expression of thought (which then takes priority and which can be expressed with gesture as well, if not better at times, especially in a loud environment where one cannot hear another speaking)?
Sifakis continues:
The stories represented by tragedy may have been traditional, but tragedy as a dramatic genre had moved away from its ritual/oral beginnings and evolved at a fast pace, alongside prose and the visual arts, to become the best kind of poetry—superior to epic, as Aristotle argues in the last chapter of his treatise—and the most characteristic art form of Classical Athens. Clearly, the poetic masterpieces of the great tragedians were not 'scripts' for actors to display their skills like the 'scenarios' of the Commedia dell'arte, or the 'sequences' (dan) of Nô drama in which actors perform long typical routines (p. 7).
No! (if I may pun on Nô). The beginning of “tragedy” was performance with dance (Dramatics 4, 1449a19-24). The text was the script, once it took priority with Aeschylus and was no longer "silly," as the Northern Greek calls the language there. Speech was until Aristotle's time always part of the whole performance with a singing and dancing chorus, at least concerning the type of tragedy analyzed in the Dramatics. Sifakis has missed that poiēsis is used as Diotima explained it (as noted throughout this website and as demonstrated in ADMC 1-2), as mousikē kai metra (“music-dance” and verse), not simply as verse. It is another of those “misunderstandings firmly established by unreflective repetition.”
Sifakis adds very wisely, though, that:
Still, he [Aristotle] includes opsis in the parts of tragedy, because tragedy is intended for the stage; but that does not make spectacle a part of tragic poetry, and Aristotle is careful to allow no uncertainty about it (p. 10).
Why Sifakis is correct on this particular point is that, again, poiēsis for the Stagirite is based on “the Diotiman narrow sense,” comprising music, dance, and verse. Aristotle adds plot as a fourth necessary and sufficient condition. That is, poiēsis is a technical term for him in the treatise and in the Lyceum (cf. ADMC pp. 205ff). In one way poiēsis is broader than tragedy because “musical” composition with plot also includes comedy, the satyr play, and epic insofar as the rhapsode performed it with music/chanting and gesturing, the latter of which for Plato and Aristotle counts as the “ordered body movement” of Laws II 665a, that is, dance. However, poiēsis is an equivocal term and at other times used more narrowly than tragedy. As just described concerning Diotima, it only refers to the four necessary elements that make up the technical sense of poiēsis, ultimately compressed into three of the six necessary conditions within the 2-1-3 pattern: Plot, speech, and melopoiia (the last of which combines music and dance). The compression occurs at the phrase mentioned above, rhuthmos kai harmonia kai melos, which has exercised every commentator historically, because, amazingly, not one realized that rhuthmos simply means “dance.” The phrase therefore means “dance and music, that is, music-dance.” Aristotle thereby combines two elements of the Diotiman “composition” into one term, melos (or its substitute later in the chapter, melopoiia).
So from this perspective the spectacle is outside the realm of the dramatist qua poiētēs, but from the former perspective, namely, what tragōidia is, spectacle is indeed included. Note how Sophocles in Chapter 4 is the one who creates the scene-painting.
Given the reasons above, I would quibble, and have quibbled in ADMC, about the two phrases in italics in the following paragraph of Sifakis, all of which leads us to the final topic (and, fortunately, regarding the rest of the paragraph, he arrives at roughly the correct solution):
It should be kept in mind that spectacle is a part of tragedy as much as plot, character drawing, thought, poetic language and music are, but it is hardly related to poetic art,” which is actually the subject of Poetics. Once this distinction between tragedy as drama—implied by “the definition of [tragedy's] essence”—and “the poetry of tragedy” is made, Aristotle's reputed ambivalence about, or even dismissal of, theatrical performance disappears. Opsis belongs neither to the art of poetry nor to any other technê acknowledged as such in antiquity (p. 14).
Regarding the first dubious, italicized phrase, it suffices to point out that “poetics” cannot be the subject of the Northern Greek's treatise because not one poem exists in it (rather the treatise is about dramatic theory). By "poem" I mean a complete poem, (only) language in meter, be it a 2- to 4-line epigram by Simonides or a 7-stanza, 28-line composition by Sappho, like "The Ode to Aphrodite," which, by the way, is often called a "lyric poem" and which, according to prestigious classicists like Gregory Nagy, was meant to be sung. So, for us the compositions should be called songs and not (pure) poems, unless one likes the two-word description, which, however, in and of itself is ambiguous: "Lyric" comes from the lyre, but songs can be sung with an instrument other than the lyre or without instrumental accompaniment. Consider a composition by Simonides, in which he speaks of himself:
Six and fifty bulls and as many tripods you did win, Simonides, before you dedicated this tablet. Even so many times, after teaching your odes to the delightful chorus of men, you mounted the splendid chariot of glorious victory.
At least some of the composer's works are odes, not poems per se.
In the Dramatics, just because the speech was done in verse for ancient Greek "serious drama" does not mean the book is about poetry; rather the snippets of verse that Aristotle presents as examples are parts of scripts that are intended to be performed. The outlier Chaeremon is mentioned in Chapter 1 as an (odd) composer who uses a medley of meters, and in the Rhetoric he is described in effect as being a new breed, composing plays only to be read or read out loud. Nevertheless, already by the end of Dramatics 1 we realize that the treatise cannot be about "plays (only) in verse," whether a single meter or a medley, because Aristotle adds that they also have music and dance (rhuthmos). This is much more meaningful than typically recognized, because in Chapter 25 Aristotle says that each art form has its own principles, and it can be as foolish at times to use dramatic rules for poetry (or vice-versa) as to use rules of painting for music (or vice-versa). The stricture of Chapter 25 is confirmed when Aristotle describes the respective advantages and disadvantages of epic versus tragedy in Chapter 26.
Regarding the second dubious phrase, let us examine now the support Sifakis gives for the claim that opsis does not belong
“to any other technê.”
There is a useful parallel between Poetics and Rhetoric in that both treatises deal with the composition of literary works intended for public performance, but their author refrains from discussing the latter because performance relies on different arts (separate from the art of composition) which had not been methodically examined by Aristotle's time (p. 10).
Sifakis is right about the Northern Greek being the first to treat of a part of public performance, such as delivery, in the context of rhetoric (the law courts and public assemblies) but any implication that this applies to drama is contradicted by the very end of Chapter 15, as commentators like Bywater, Tarán and Gutas get right: Aristotle says there that the issues of stagecraft are necessarily connected to the dramatist's business but he adds that enough has been said about those issues in his already published work. (Janko mistranslates aisthēseis as “reactions,” thereby completely stripping out any reference to performance.) Part of that published work may be Chapter 17, which seems to have come from a different treatise and which was probably interpolated into the agglomeration that we now possess. That chapter deals exactly with these issues of stage-performance that the dramatist qua full composer has to be aware of (cf. A Primer, pp. 199-202 and 261). Sifakis gives an informative discussion of Chapter 17, also showing the importance of stage-effects for the composer (usually badly translated as “poet”) qua, in Aristotle's technical sense, “musical” dramatist (pp. 16-19).
For our purposes, this concludes the discussion of Sifakis's original and illuminating article.
Return to ADMCupdates.