Monday, 6 October 2008

Aristotle's Unities

Greek and Latin plays were very different from the native traditions of drama that the young Shakespeare might have come across if he had seen a mystery cycle, or watched the travelling troupes of actors who came to Stratford, performing moralities or the various types of drama which developed from them. (Picture on Left-above: Roman theatre, Kermessos. Photograph Peter Smith.)

Greek and Latin drama were strict in form. The stage represented a single place throughout the action; the plot recounted the events of a single day; and there was very little irrelevant by-play as the action developed. Aristotle described the drama of an earlier age in his important work On the Art of Poetry; those who followed his precepts called this disciplined structure the three "unities": unity of place, unity of time and unity of action.

The "Rules"
Neo-classical Renaissance critics codified Aristotle's discussion, claiming that all plays should follow these three precepts:
  • Place. The setting of the play should be one location: in comedy often a street, in Oedipus Rex the steps before the palace.
  • Time. The action of the play should represent the passage of no more than one day. Previous events leading up to the present situation were recounted on stage, as Prospero tells Miranda of the events which led to their abandonment on the island.
  • Action. No action or scene in the play was to be a digression; all were to contribute directly in some way to the plot.

Compare this structure with the episodic, wide-ranging plots of romantic comedy like Shakespeare's Winter's Tale.

The classical unities or three unities are rules for drama derived from a passage in Aristotle's Poetics. In their neoclassical form they are as follows:

  1. The unity of action: a play should have one main action that it follows, with no or few subplots.
  2. The unity of place: a play should cover a single physical space and should not attempt to compress geography, nor should the stage represent more than one place.
  3. The unity of time: the action in a play should take place over no more than 24 hours.

Aristotle dealt with the unity of action in some detail, under the general subject of "definition of tragedy", where he wrote:

Now, according to our definition, Tragedy is an imitation of an action that is complete, and whole, and of a certain magnitude(Aristotle's Poetics, XVII.) … As therefore, in the other imitative arts, the imitation is one when the object imitated is one, so the plot, being an imitation of an action, must imitate one action and that a whole, the structural union of the parts being such that, if any one of them is displaced or removed, the whole will be disjointed and disturbed. For a thing whose presence or absence makes no visible difference, is not an organic part of the whole.(Aristotle's Poetics, XVIII.)

His only reference to the time in the fictive world is in a distinction between the epic and tragic forms:

Epic poetry agrees with Tragedy in so far as it is an imitation in verse of characters of a higher type. They differ, in that Epic poetry admits but one kind of metre, and is narrative in form. They differ, again, in their length: for Tragedy endeavours, as far as possible, to confine itself to a single revolution of the sun, or but slightly to exceed this limit; whereas the Epic action has no limits of time.(Aristotle's Poetics, V.)

Unlike his prescriptive attitude regarding the plot (unity of action), Aristotle here merely remarks on the typical duration of a tragedy's action, and does not suggest any kind of imperative that it always ought to be so. He was writing after the golden age of Greek drama, and many Greek playwrights wrote plays that do not fit within these conventions.

Even more tellingly, Aristotle does not mention the neoclassical unity of place at all. So Aristotle suggested only one unity -- that of action -- but the prevalent interpretation of his Poetics during the Middle Ages already inclined toward interpreting his comment on time as another "unity".

Any discussion of Aristotle's Unities of Time, Place and Action must start from the acknowledgement that his Poetics from which we receive his ideas about the drama deals only with tragedy: we do not know whether he recommended the same canon of rules for comedy [1], or indeed for history plays, of which he would certainly have known at least one, namely The Persians by Aeschylus. I shall also be proposing that, as a man of his times, his concepts arose from, and pertained to, the Greek drama of the 5th-4th centuries BC, and accordingly they do not represent necessary and universally applicable rules. I shall argue moreover that the plays with which Aristotle was familiar, and on which he based his views, were themselves products of the performance conditions that prevailed in his era. His Unities, in short, are the unacknowledged children of ancient Greek economics and theatrical technology.

Theatre in Aristotle's day typically had a civic and/or religious significance: theatrical performances were presented to honour a god or a city or a city's god, and they were therefore state occasions. Everybody turned out for the day, even slaves and (for the tragedies at least) women. Moreover, entrance to the shows was by and large free, albeit it was possible to purchase or hire cushions from quick-off-the-mark entrepreneurs. Who paid for it all? Well, although the temples and city magistrates might put up prize money if plays were presented in competition, it was not generally the god or the city that financed the productions. On the contrary, each show was under-written by a single wealthy sponsor who hired a troupe of two (later on three) actors and also coughed up for the chorus, the choreographer, the singing teacher, the pit musicians, the costumes, the stage machinery and the set. With no prospect of getting his money back, the sponsor's motivation was 1) to accumulate heavenly brownie points, 2) to accrue some kudos as a devout and civic-minded patron of the arts, and 3) by so doing, to establish a lead over his rivals at the next election.

Besides the economic factors I have just described, other features of Greek production tradition also need to be taken into account when evaluating the origin and validity of Aristotle's Unities. Chief among these are the fact that plays were presented in the open air against a permanent architectural background, and were illuminated only by sun- or torchlight. I believe that these characteristics, taken together, are sufficient to explain Aristotle's stress on the Unities of Time, Place and Action.

Unity of Place is straightforward enough: there was only a single permanent set, and obviously it could not be subjected to the quick scene changes that are familiar to us in modern theatre. Equally obviously, it would have been impossible to suggest changes of locale by lighting changes (as we can do today), even if the script called for it. At this point in my argument, however, an astute critic will instantly object that Shakespeare too had to work with a fixed architectural background a limitation that did not prevent him from changing location from scene to scene. Clearly, a permanent set is by no means as inflexible as I seem to imply. To this objection I observe that Shakespeare could persuade us he had changed location by virtue of the fact that each new scene would begin with a new grouping of characters sweeping onto the stage. Ancient Greek casts, by contrast, were too tiny to allow that. And the size of the cast also accounts, I suspect, for Aristotle's stress on Unity of Action. With only two or three actors (who were in any case doubling up to play all the characters), there would not have been much energy left over for subplots.

In this connection it is not clear whether the small cast was merely a matter of convention, or whether it too was demanded by the economics of production. For my own part, I am inclined to suspect that money spoke as eloquently in the fourth century BC as it does in the twenty-first AD. I rather fancy that Greek sponsors simply balked and told the playwright, "We was pushed enough to pay for Aeschylus's Actor Number Three, there is no way we are going to pay for a fourth or a fifth or a sixth bleedin' poncy thespian, not even if you was bleedin' 'Omer 'imself, which you are not!" Impresarios of West End or Broadway productions will be familiar enough with the economics of mounting a show, and they too will go for a single set, small-cast plays over an epic any day contemporary capitalism's own bow in the direction of Aristotle's Unities.

Alternatively, the economic impetus may have come inadvertently from the actors themselves. Theatrical productions, as I've hinted, were seasonal, and were mounted to celebrate some local religious festival. So when the Athenian shows were over, the actors had to go on the road to cities that had different calendars of festive days. And since, however good the plays were, or however outstanding the cast, there was only the one performance (revivals apart), whatever fees the actors earned had to be stretched pretty thinly to cover the "resting" and rehearsal time (Actors Equity please note!). Under those conditions it's easier to keep a small troupe of two or three on the road than even a moderately sized repertory company [2].

We come at last to Unity of Time - an Aristotelian stricture that was almost always more honoured in the breach than the observance, even by the Greek tragedians themselves. I for one, at any rate, find it impossible to believe, for example, that the action of Aeschylus's Agamemnon can be compassed by a single span of twenty-four hours. The play opens in the early morning, with beacon fires announcing the end of the Trojan War in Asia Minor: it takes the most rigorous suspension of disbelief to accept that Agamemnon returns the 500-odd miles to Greece and gets himself bumped off in the bath before nightfall! No; what Aeschylus did was to create an illusion of Unity of Time a feat that was accomplished, as it is today, by focussing on the central story. The playwright does not show us the victorious Greeks embarking for home at the port of Tenedos, the gangplanks that collapsed, the horses and men that slipped and fell into the sea, the four or five days of tedious rowing, the squabbles at mealtimes, or Agamemnon's dusty ride up from the sea to Argos.

What I am saying, in effect, is that theatrical time was not then, and is not now, "real" time; on the contrary, it was and is "speeded up" time. But the illusion of Aristotelian Unity of Time was promoted then, as it is today, by the familiar technique of "cutting to the chase," i.e. omitting all distractions, which is another way of understanding Unity of Action. The illusion was perhaps also fostered by a dramatic structure in which five acts were punctuated by choral odes that implied the passage of time. Whether the passage of time represented by the choral interlude represented minutes or months was never stated, however [3].

Aristotle, it seems, further believed that observance by the playwright of the Dramatic Unities contributed to the intensity of the audience's experience and particularly to the strength of their cathartic response to the play. The spectator, in Aristotle's view, came away wrung out by the emotions of pity and fear he or she had undergone in the theatre - and they probably did. In fact you can get wrung out by a Greek tragedy in less than an hour, as I recently discovered at a stripped-down production of Euripides's Trojan Women, that ran a mere fifty-five minutes! But (pace Aristotle) who is to say that the intensity of feeling you experience at a classical Greek tragedy that obeys the Unities is greater than what you experience at a good production of the very un-classical Richard II that doesn't?

If there were any European plays in the thousand years after the Christians took over the Roman Empire, we don't know of them. Theatre, it seems, died out or went deep underground. It only began to re-emerge in the Middle Ages, when it took the form of small-scale religious dramas, and later developed into large-scale cycles of Mystery Plays. But a sequence of episodes that re-tell bible stories from the Creation to the Resurrection is not a form that lends itself to being structured by Aristotle's canon of Unities, and it was not until the rediscovery of ancient learning during the Renaissance that Aristotle's ideas began to influence playwrights once again. Even then, however, his impact remained indirect, via the tragedies of Seneca that became models for the earliest dramatists of the sixteenth century.

Then came a fork in the road. The most exciting drama of the Renaissance was undoubtedly the theatre that developed in England and Spain during the late sixteenth and early-to-mid seventeenth centuries and there, writers like Shakespeare, Marlowe, Webster, Tourneur, Jonson, Cervantes and Lope de Vega were busy disregarding the Unities all over the place. All that remained of the Greek tradition was the five-act structure. It was only in France where the drama was intended more for court than courtyard that Aristotle ruled supreme. For a brief, shining moment the Unities were rescued from oblivion by Corneille and Racine. Their tightly-structured five-act plays, their method of telling stories by means of scenes involving only two or three characters, and their elegant, formal alexandrines are still capable of delivering a powerful emotional punch to this very day as anyone who saw the recent productions of Phèdre and Britannicus with Diana Rigg will testify. But the new model English and Spanish drama was winning the day, and Lope de Vega and Cervantes were even abandoning the five-act structure in favour of three acts (the English didn't get round to that innovation till the Restoration).

It is, of course, still possible to write dramas that adhere to Aristotle's Unities [4], but developments since the Renaissance have completely put paid to any notion that it is obligatory to do so, while the evolution of a brand new dramatic narrative form in the shape of the motion picture has undermined the Unities still further. We have become so accomplished at telling and understanding stories visually that we could probably make a film of War and Peace that ran no longer than the 17 minutes of the 1812 Overture soundtrack, we could do an average 80,000-word novel in no more time that it takes to run an MTV video, and we routinely pack soap opera episodes into the 30 seconds of a TV commercial. This is so commonplace that it doesn't even register, but if you scrutinise a few good commercials with the eye of a playwright you'll see what I mean and you'll be astonished.

And yet, despite filmic narrative techniques (brief scenes, multiple locations, jump cuts, flashbacks, flashbacks within flashbacks, etc) techniques which are also impacting on writing for the stage there is no question but that we still experience even a non-Aristotelian movie or play as a unity. This phenomenon, which would doubtless have set the Greek philosopher scratching his head in perplexity, stems from the fact that the sense of unity does not arise from some set of Aristotelian rules imposed on the play from the outside, but from within ourselves. If we hadn't known this intuitively, we were persuaded of it by a number of ingenious experiments on visual perception that were carried out by Gestalt psychologists in the 20s and 30s of the last (i.e. twentieth!) century. Unities 'r' Us. And this means that the Aristotelian Unities, when we do encounter them in a play, do not create but only enhance the sense of a whole, complete experience.

Now all this has led me to ponder the question: If Aristotle is indeed wrong, what would a modern Theory of the Unities look like? My initial impulse was to attack this question negatively, by asking what factors created a sense that a play or film is incoherent and at sixes-and-sevens. To which the first answer that popped into my head was that a mixture of genres would be confusing in just this manner. On second thoughts and after a few heated arguments with friends, I hasten to add I have to confess to being in error. Shakespeare had his own genres “comical, tragical, historical and pastoral” but that didn't stop him mixing them sometimes. Plays like The Merchant of Venice, Measure for Measure and All's Well That Ends Well the so-called "problem" plays in other words are tragi-comedies, a mode that developed into a genre in its own right, in which "many come close to death, but none die [5]." Nowadays, what's more, we are familiar with comedy-thrillers, comedy-westerns, science fiction-noir, comical-historical-romantic-adventure (The Three Musketeers) etc, while one of the most glorious operas of the twentieth century Richard Strauss's Ariadne auf Naxos even succeeds in mixing tragi-romantic drama with high camp commedia del arte farce!

So if not mixed genres, then what? - for it is still the case that most dramatic productions, in the theatre as well as the cinema, carry a subliminal label, "Unities At Work". One factor contributing to the sense of unity, albeit a relatively minor one, is a coherently and consistently designed show: there is something very satisfying about a production that simply looks as if it's all of a piece. Far more important, however, is the contribution made to the sense of wholeness by the story arc. Call me old fashioned if you like, but I only respond fully to stories with a beginning, middle and an ending a sequence I experience as a unity. And more important still than the movement of the plot is the sense of unity that comes with a traditional emotional arc for the story. By this I mean that the drama should end with the resolution for the protagonist of some kind of internal conflict.

The elements sketched out in the last couple of paragraphs seem, if I'm not being unduly immodest, to constitute at least the beginnings of a modern Theory of the Unities. I would be interested to read other people's contributions to the topic.


  1. It will be recalled that the rediscovery and subsequent destruction of Aristotle's lost book on comedy forms the background to the monastic murders in Umberto Eco's novel, The Name of the Rose.
  2. Those interested in a lively fictional account of the Greek actor's life can do no better than turn to The Mask of Apollo by Mary Renault.
  3. We reverse this illusion in our own practice of experiencing theatrical and cinematic spectacles in a darkened auditorium. The period "in the dark" exists outside normal time, and so can encompass years, if necessary, of dramatic time.
  4. A play by Ian Mandleberg, Passover, posted to MontageShowcase is a case in point.
  5. Still, not even Shakespeare mixed genres to the extent that the Players in Hamlet did, with their "tragical-comical, historical-pastoral, and comical-tragical-historical-pastoral."

1 comment:

Anonymous said...

great post, i enjoyed reading it.