[Aristotle's Poetics: Introduction] [Aristotle's Poetics: Seminar Notes]
Aristotle makes the following references to Euripides' play:
(i) ch. 11 (52a5-8): example of
reciprocal recognition;
(ii) ch. 14 (54a4-7): example of the
best kind of tragic plot;
(iii) ch. 16 (54b30-36): the recognition
of Orestes is an example of the second least artistic kind of
recognition;
(iv) ch. 16 (55a18-20): the recognition
of Iphigeneia is an example of the most artistic kind of recognition;
(v) ch. 18 (55b2-15): Aristotle uses this play to illustrate his suggestion that a dramatist should start with a universalised abstract of the plot, and then turn it into 'episodes';
(vi) at Rhetoric 1407b35 Aristotle quotes line 727 to illustrate a stylistic device.
There are also some more general references:
(i) ch. 16 (55a6-8), cf. ch. 18 (55b9-12): an alternative way of bringing about the recognition, suggested by a sophist named Polyidus.
1. Iphigeneia's opening speech gives an account of some of the events in the plot that precede the beginning of the plot. These include one spectacularly strange event: Iphigeneia, on the point of being sacrificed, is spirited away; a deer is substituted; and none of the people present notice what has happened. This would be difficult to stage without it seeming ludicrous: compare what Aristotle says about the pursuit of Hector in the Iliad (ch. 24, 60a11-18: Homer can only get away with it because it is narrated, not seen); and it is, of course, outside the play (cf. ch. 24, 60a27-34).
2. Orestes also tells us something of his earlier history in the opening scene; but a detailed exposition is reserved until much later (939-82). Here, too, there are details which might, on reflection, seem puzzling. In particular, why do some of the Furies refuse to accept the verdict of the trial at Athens and continue to pursue Orestes? This (which is not, so far as I am aware, found in other versions of the story: it seems to have been invented by Euripides for this play) is just presented to us as an unexplained datum. Here, too, the fact that the event is outside the play may help to make the oddity less obtrusive.
3. The sacrifice of Iphigeneia is not an isolated event in the history of this family: Orestes has killed his mother; Clytaemnestra killed her husband; Atreus killed his nephews and fed them to his brother Thyestes; Tantalus killed his own son Pelops and tried to feed him to the gods. The many allusions to this background in the play help to set what happens within the play in a context which gives it added significance and emotional force: the death of Orestes at his sister's hands would be an appallingly apt continuation of this family's self-destructive history.
4. The fact that the plot centres on the potentially fatal interaction of two close relatives fits with Aristotle's analysis of things that evoke fear and pity in ch. 14 (53b15-22): tragedy should aim at plots in which someone kills, or does something similarly terrible, to a person closely related to them (rather than to an enemy or a neutral).
5. But the family's past history is relevant here too. When Iphigeneia tells us about the dream which has convinced her that Orestes is dead, she reminds us that Orestes (the last surviving male member of the family) is the only remaining pillar of the house (50). His death will therefore not just be the next step in the family's self-destructive history; it would be the last one. So what is put at risk in the play is not only the fortunes of two individuals, but the existence of the family itself. It is not clear how alert Aristotle would have been to this dimension of the play: he seems to be interested in what happens to individuals, rather than to families (he refers to families twice: ch. 13, 53a11; ch. 14, 54a12; but even here his focus is arguably on the individual family members).
6. We have spoken of Orestes' death at his sister's hands: and Euripides sometimes uses similar language (e.g. 704f.). But in fact it is clear that Iphigeneia herself would not do the killing: her role is to prepare the sacrificial victims, but others will actually wield the knife (40f., 621-4). The complicity in the death which this involves may be equivalent to her doing the deed herself. After all, we would not absolve Agamemnon from blame for the death of his daughter simply because he did not use the knife in person.
7. On the other hand, at Iliad 3.292 Agamemnon does act in person in an animal sacrifice; so we should not take it for granted that he would not have sacrificed his daughter with his own hand. And that would make his guilt much or more dreadful. So does Euripides go out of his way to make it clear that Iphigeneia does not perform the sacrifice in person in order to distance her from the worst part of the sacrifice? Otherwise, the audience might view her with distaste - even if she is not to blame, she would still be implicated in something revolting. (Or is she even so still implicated in human sacrifice to such a degree as to compromise her innocence? It may be relevant that her involvement in sacrifice is something outside the play, and therefore less salient for the audience than the role she plays within the play, which is one of sympathetic concern for the strangers - a fact highlighted by her prediction that she will henceforth be more ruthless: the fact that this claim is made and disproved throws light on her sympathetic character.)
8. Another question arises about Iphigeneia's role later in the play: when she devises the plan of escape, does she display a cleverness that Aristotle would regard as inappropriate in a woman (see ch. 15, 54a22-24)? Orestes says that women are clever at this kind of contrivance (1032); and the resourceful women are not uncommon in tragedy (or comedy). Perhaps we should distinguish this kind of practical intelligence from the intellectualism which Aristotle objects to in his example of an inappropriately clever woman, Melanippe, who made a speech that involved sophisticated philosophical argument (cf. 54a31).
9. The play contains two recognitions (Orestes recognises Iphigeneia; Iphigeneia has then to be made to acknowledge his identity); the care with which Euripides prolongs the process of recognition shows how important this element was in his design.
10. Aristotle regards the recognition of Iphigeneia (her identity is revealed as a consequence of her plan to send a letter to her brother) as an example of the best kind of recognition, i.e. the kind which comes about as part of a probable sequence of events (ch. 16, 55a16-19). He is less enthusiastic about the way Orestes proves his identity (ch. 16, 54b30- 36): he regards this as inartistic, on the grounds that Orestes says what the poet requires, not what the plot requires; thus it is 'contrived by the poet'. This evaluation seems strange. In these circumstances, surely it is entirely necessary or probable that Orestes would seek to establish his identity; and the way he does it is natural enough. So in that sense the recognition does comes about as part of a probable sequence of events.
11. This is not to deny that the recognition of Iphigeneia is dramatically more interesting. It is the more crucial turning-point: up to that point, the trend of events is towards disaster; but once Iphigeneia's identity has been disclosed, the revelation of his Orestes' identity is inevitable. The recognition of Iphigeneia, therefore, marks the beginning of the change of fortune. (On the other hand, it is also true that until the second recognition is achieved, the threat to Orestes' life still exists: so the dramatic tension is not completely dissipated with the first recognition.)
12. Furthermore, the recognition of Iphigeneia does have a distinctive element. Even if Aristotle is unfair when he says that the recognition of Orestes is contrived by the poet, it is certainly contrived by Orestes; he knows what he is doing, and does it deliberately. By contrast, the recognition of Iphigeneia is an unforeseen result, not brought about intentionally by any of the characters; the course of events brings about this result without the deliberate contrivance of any of the characters. This does seem to support Aristotle's claim that the recognition of Iphigeneia is technically superior. Moreover, the unexpectedness of the outcome involves an element of astonishment, which (on Aristotle's view: ch. 9, 52a1- 11) enhances its emotional impact. When Aristotle cites this recognition as an example of the best kind, he refers to its emotional impact as well as its probability (ch. 16, 55a17).
13. The recognition of Iphigeneia, which is not brought about intentionally but is an unforeseen by-product of the characters' actions, involves reversal (see ch. 11, 52a22-29). Compare Aristotle's comments on the two recognitions in the Odyssey which Odysseus' identity is established by his scar (ch. 16, 54b25-30): the recognition by the Nurse, which involves reversal, is better than Odysseus' deliberate use of the scar to confirm the declaration of identity which he makes to the herdsmen.
14. Aristotle also refers to another way in which the recognition of Orestes could be handled (proposed by the sophist Polyidus: ch. 16, 55a6-8): Orestes, facing sacrifice, might comment on how fitting it was that he should die in the same way as his sister - prompting Iphigeneia to infer his identity. This kind of recognition through inference is ranked second-best by Aristotle. It, too, has the advantage that it is not brought about intentionally, and therefore has an element of astonishment. (Polyidus' recognition, if adopted, would of course require other changes to the plot. For example, it would now be Iphigeneia who, having recognised her brother, has to establish her identity: since Orestes 'knows' that his sister is dead, that might not be easy. One possible advantage of Polyidus' proposal would be that we could get much closer to the actual moment of sacrifice before the recognition intervenes; so the tension might be built up more.)
15. The third element of plot which Aristotle identifies is suffering (ch. 11, 52b9-13). Suffering is absent from this play, since the disaster is averted. Or would it be better to say that suffering is present in the audience's anticipation of disaster - which is then averted?
16. The fact that the disaster is averted gives the play a happy ending. When Aristotle introduces the concept of 'change of fortune' (ch. 7, 51a11-15) he envisages the possibility of a change from bad to good fortune, as well as a change from good to bad fortune, as the basis for a tragic plot.
17. We must remember that the word 'tragedy' in Greek does not carry all the connotations which it has acquired in post-classical literature. In Greek it refers primarily to the plays which were performed in a particular slot in the dramatic festivals (by contrast with satyr plays and comedies). The notion of an unhappy ending is not built into 'tragedy', when the word is understood in this sense.
18. Nevertheless, Aristotle defines tragedy with reference to the emotions of pity and fear (ch. 6, 49b27; ch. 10, 52a2-11 etc). In what sense is the change from bad fortune to good suitable to tragedy, defined in those terms? The miserable state in which both Iphigeneia and Orestes start out evokes pity; and the imminence of a still worse disaster evokes fear. So this plot-pattern does have the essential tragic emotions.
19. Iphigeneia is far from being the only tragedy with a plot of this kind. Eumenides is an obvious example. Among the works of Euripides the Ion or the lost Cresphontes (see ch. 14, 54a5-7); and there is the Lynceus of Aristotle's friend Theodectes (ch. 11, 52a27-29; cf. ch. 18, 55b29-32). The Lynceus, in which the hero is saved from imminent death and his enemy Danaus is killed, is an example of the 'double' plot which Aristotle describes in ch. 13 (53a30-39); the example he gives there is the Odyssey.
20. Aristotle's discussion of the 'double' plot provides evidence that fourth-century audiences preferred this kind of play. But he thinks that this is evidence of their poor taste and 'weakness': the pleasure given by this kind of play is more appropriate to comedy than tragedy. Hence he rates this kind of plot less highly than the kind of plot in which someone (who is neither outstandingly virtuous nor wicked) falls into misfortune: it is wrong to object to plays which end in misfortune, because this is what is most tragic (ch. 13, 53a23-30: Aristotle cites audience perceptions in support of this claim - is this inconsistent with the attack on the 'weakness' of audiences a few lines later? Or is he saying that people recognise that unhappy endings are most tragic but still prefer happy endings?)
21. The considerations which lead Aristotle to downgrade the 'double' plot in ch. 13 would seem to apply to Iphigeneia as well. Admittedly, this play is not an example of the 'double' plot, since there are no characters, like Danaus in the Lynceus, or the suitors in the Odyssey, who end unhappily; but it shares with the 'double' plot the movement from bad to good fortune. In fact, Aristotle says that the pleasure given by 'double' plot is more characteristic of comedy than tragedy, because in comedy no one is killed; so the fact that no one ends unhappily in Iphigeneia seems to imply that its pleasure is even more characteristic of comedy than that of the 'double' plot - making it even less successful as tragedy.
22. It is therefore surprising to find that in ch. 14 he concludes that the best kind of tragic plot is that found in Iphigeneia (in which someone unwittingly comes close to doing something terrible to a person closely related to them, but discovers their identity in time, so that the disaster is averted). This is ranked higher than the kind of plot in which the truth is discovered only after the terrible deed has been performed.
23. Ch. 14 therefore presents us with two problems: (i) why does Aristotle (apparently) present us with two incompatible analyses of the best kind of tragic plot? Answering this would be pure guesswork; so it is better to concentrate here on: (ii) how is the analysis in ch. 14 to be explained or defended?
24. The kind of plot approved in ch. 14 does have some obvious advantages. It excites a wider range of emotions (pity and fear, but also relief, joy etc); and it leaves the audience in a more positive mood. It is easy to see how this might appeal to some temperaments. However, Aristotle has already condemned the kind of temperament to which this kind of plot appeals as 'weak'. So, though it is easy to see how some people might have expressed a preference for this kind of plot, it is much less easy to see why Aristotle - given his assumptions about the nature of tragedy - should have done so.
25. Aristotle does not give us any help here: the claim that this kind of plot is best is simply stated, without any argument or explanation (54a4-5). Can we work out how he might have defended this position? (Remember that in the Poetics we seem to have lecture notes; when Aristotle was giving the lecture he might have expanded and explained his point.) The answer seems to be that we cannot. But some inconclusive suggestions can be made.
26. One possible line of argument would be that the killing of someone closely related is so shocking that there is a risk that the audience's enjoyable sense of pity and fear will be swamped by a contrary reaction of revulsion; if so, it is better to work on the audience's emotions by confronting them with the imminence of such a horrible event - but to avert it at the last moment, so that they are spared intolerable shock and distress. Aristotle certainly does believe that there are limits to what tragedy can do: if the dramatist goes beyond those limits, the audience will feel revulsion rather than pity or fear. In ch. 13 (52b34-36) he says this about fall of a virtuous character into misfortune; and in ch. 14 (53b37-38) he says the same about killing or injuring a close relative knowing who they are.
27. But there is a problem with this line of argument. Aristotle goes on to say that the element of revulsion is absent when a close relative is killed or injured in ignorance (54a2-4). The implication is that what is revolting is the witting intention, not the act of violence itself. So this line of argument seems to fail. (The best one could do to rescue it is to that Aristotle's sensitivity to the limits of what tragedy can do may have influenced him unconsciously.)
28. A refinement on this line of argument might be considered. The fall of a very virtuous character into misfortune would cause revulsion: but what if a very virtuous character seems to be about to fall into misfortune, but escapes at the last minute? Perhaps this excites a greater degree of emotion (because it is a very virtuous character who is in danger), while avoiding revulsion (because the misfortune is averted). To make this work, you will have to argue that characters such as Iphigeneia and Orestes (or, e.g., Ion in Ion) are virtuous to a degree that (e.g.) Oedipus is not. Is this a plausible position?
29. Another angle of approach to this problem would be to compare the Iphigeneia with a play of the kind which Aristotle prefers in ch. 13 but ranks second in ch. 14, to see whether there is any respect in which Iphigeneia is superior.
30. The obvious move is to compare Iphigeneia with Oedipus. But this would be unfair: the Oedipus is such a remarkable and rich tragedy that any other play is going to look unimpressive by comparison. (Note that what Aristotle claims in ch. 14 is that the type of plot used in Iphigeneia is better in principle than the type of plot used in Oedipus; this does not mean that Iphigeneia itself is a better play than Oedipus - it might be that Sophocles has made more effective use of material that is in principle inferior. As always, we must remember that Aristotle's focus is on the plot, as distinct from the play; and, as always, we must ask ourselves whether this abstraction is helpful.)
31. So it might be fairer to compare the Iphigeneia that Euripides wrote with a revised Iphigeneia, in which Orestes is not recognised (e.g. Iphigeneia postpones giving her message to Pylades until later); so the sacrifice of Orestes takes place; when it is reported to Pylades he exclaims on what an unhappy life (and death) Orestes had; and it is at that point, when she hears her brother's name, that Iphigeneia realises what she has done, and is overwhelmed with grief. Is there any respect in which this would be a less satisfactory play than the one we have? (Of course, a lot of other changes would have to be made to make this adapted plot work: for example, we would not want to have Apollo sending Orestes to his death.) Several points could be made:
(a) There is no change of fortune - or there is only a change from bad fortune to worse fortune; the less pronounced contrast makes this kind of plot less dynamic and interesting;
(b) Although this play would have a recognition (and quite a good one: it would be as unexpected and astonishing as the recognition of Iphigeneia in the play we have), there is no reversal; so the element of astonishment would be diminished, again making for a less interesting plot. (One the other hand, it could be argued that the reconstructed play would have an even greater astonishment and impact, because the audience's anticipation of a last-minute escape is thwarted.)
(c) The last part of the play would contain lamentation, but not much more; the excitement of the escape-plan would be lost, again making for a less interesting and varied plot.
32. One further observation - though it is not clear what implications it has for our current problem. In Iphigeneia the act of violence against a close relative is averted; in Oedipus it happens outside the play. These can be seen as different ways of distancing the audience from that shocking and potentially revolting event (if it is a shocking and potentially revolting event: but see above for the problem with this view).
33. Another question that could be asked, relevant to the relationship between the analyses of the best kind of tragic plot in chs. 13 and 14, concerns the role of 'error' (hamartia) in this plot. Right at the beginning of the play we learn of one past error: Agamemnon rashly vowed to sacrifice to Artemis the most beautiful thing produced in a given year (20f.) - and this was Iphigeneia. The bad fortune in which the play begins results from this error. But that is before the play, and of course is not what the play or the audience focus on. Within the play, the crucial thing is the error which Iphigeneia almost makes - unwittingly sacrificing her brother. The error here is potential; removing the ignorance that would bring about this error is what averts the disaster.
34. The end of the play raises a number of questions. In ch. 15 (54a34-b8) Aristotle accepts the use of a god to convey information about subsequent events, but is critical of divine intervention used to provide the resolution of the plot - or (on an alternative interpretation) of the arbitrary use of divine intervention (i.e. where it is not part of the necessary or probable sequence of events).
35. Athene gives information about the future, in line with what Aristotle allows. But she also tells us that Poseidon has put an end to the storm which has been preventing the escape of Orestes and Iphigeneia as a favour to her (1442-5); and this looks like an intervention to bring about the resolution of the plot. Is that necessary or probable? There is an obvious motivation: as the patron goddess of Athens, Athene does have a vested interest in Orestes' escape to Athens with the image of Artemis.
36. But why does the last-minute crisis which requires Athene's intervention arise at all? Are we to assume that Poseidon has tried to obstruct the escape? If so, his motive is not made clear; so this would not be a necessary or probable development in the plot. And if Poseidon is not trying to obstruct the escape (Athene says that Poseidon is not as you suppose seeking Orestes' death, perhaps implying that this assumption is wrong) then the obstacle is just a storm that happens to blow up at an inconvenient time - a chance event, and thus not necessary or probable. It is easy to see why Euripides wanted the storm (the obstacle adds to the excitement of the end of the play; and note too that the adverse winds preventing the escape thematically echo the adverse winds which brought about Iohigeneia’s sacrifice at Aulis); but it is not something that results from the preceding events in the plot.
37. We can try to explain any event in a plot on two levels. One is to ask why the poet wanted to include it; the other is to ask why that event happens, in the sense of how it results from preceding events in the plot. These two kinds of explanation are what Aristotle has in mind when he says (see above) that in the recognition says Orestes what the poet requires, not what the plot requires. That criticism seems unjustified with regard to the recognition; but it may have more force here. It is easy to see why Euripides wanted the storm (the obstacle adds to the excitement of the end of the play; and note too that the adverse winds preventing the escape thematically echo the adverse winds which brought about Iohigeneia’s sacrifice at Aulis); but it is not something that results from the preceding events in the plot.
38. So would Aristotle disapprove of this? It is easy to imagine an alternative way of handling the ending: Euripides could have had the escape-plan succeed without a last-minute obstacle; the messenger would report it as an accomplished fact, Thoas would be angry, and Athene would appear to calm him down and tell us about the future. Perhaps this would be less dramatically effective: it would lack the astonishment that comes from the unexpected crisis, and the excitement while the outcome is still in doubt; and Athene's appearance would be less interesting, since the tension had already been dissipated. So if Aristotle were to object to what Euripides has done, on the grounds that it violates necessary or probable connection, one could argue that this just proves his poor sense of what makes good drama. But perhaps Aristotle can defend himself: in ch. 25 (60b23-26) he acknowledges that what would otherwise be an error may be justified if it enhances the impact of that (or some other) part of the composition. In other words, although Aristotle thinks that in principle and in general it is best for a poet to adhere to probability, he recognises that in practice and in particular cases there may be something to be gained from sacrificing strict adherence to probability. This element of flexibility in Aristotle's position may be enough to make what Euripides has done at the end of Iphigeneia acceptable.
39. This relatively loose handling of plot towards the end of the play contrasts with the care taken over earlier stages of the plot (note the interconnectedness of events: the pursuit of Orestes by the Furies - itself a consequence of the killing of Clytaemnaestra to avenge the murder of his father, which was partly motivated by the sacrifice of Iphigeneia - brings Orestes to Tauris, and also makes possible his capture; the fact that he is a matricide is also made the basis of the escape-plan). At the same time, there is a strong interest in the details Athenian religious cults; this is something which Euripides' Athenian audience would no doubt have appreciated (the play anchors their own experience in the heroic past), but it is something which Aristotle's theory of tragedy takes no account of. Is he justified in ignoring what is (it might be argued) a parochial and gratuitous appeal to the audience that has nothing to do with the real nature of tragedy? Or is it to be deplored as evidence of his lack of sensitivity to the social context of tragedy?
40. Aristotle's discussion of the chorus (ch. 18, 56a25-32) implicitly approves of Euripides (by contrast with Agathon and later tragedians) for making the choral songs relevant to the plot; but he also contrasts him unfavourably with Sophocles for failing to use the chorus as one of the actors within the scenes. Is this the case in Iphigeneia? It is, perhaps, a little awkward that the chorus have to be left behind when Orestes and Iphigeneia escape - a loose end which has to be tied up by Athene (1467f.). On the other hand, the chorus does intervene in a striking way when the messenger comes to tell Thoas that Orestes and Iphigeneia are escaping (1293f.: they pretend that Thoas has left the temple, in the hope that misdirecting him will gain time for the fugitives); in the event, the ruse has no effect - but the chorus are not just standing about as passive onlookers here.
41. There is an interesting discussion of the play in Aristotelian terms in:
[Aristotle's Poetics: Introduction] [Aristotle's Poetics: Seminar Notes]