©Damen, 2012

Classical Drama and Society


Course Description
Class Grading and Projects
Chapters
Syllabus
Slides
A Guide to Writing in History and Classics

 

SECTION 1: THE ORIGINS OF WESTERN THEATRE

Chapter 4: The Origins of Greek Theatre, Part 1


I. Introduction: Standards Views of the Origin of Greek Drama

The standard views of the origin of Greek drama and theatre center for the most part around three distinct and incompatible pieces of data: (1) accounts concerning Thespis who is the purported "inventor" of tragedy, (2) the meaning and evolution of the Greek word tragoidia ("tragedy" ) and (3) the historical account of early Greek theatre found in the fourth chapter of Aristotle's Poetics.

• That is, some ancient sources report that tragedy was the invention of a person named Thespis who was famous for riding around in a cart and performing his dramas.

• Others see an important clue to tragedy's rise in the Greek word tragoidia, meaning literally "goat-song," and from that assume a goat was the prize for the winning playwright at primitive Greek festivals featuring drama. In the words of Oscar Brockett, one of the pre-eminent theatre historians of our age, "the chorus danced either for a goat as a prize or around a goat that was then sacrificed." (note)

• Finally, the renowned fourth-century (i.e. post-Classical) philosopher Aristotle, the first person we know to have researched the origins of drama, concluded in his famous work on drama and theatre, The Poetics, that tragedy evolved out of another form of performance popular in the day, a type of choral singing called dithyramb.

As we will see, there is much that can be said in opposition to all these assertions, so much that no real hope exists of achieving consensus by using these data.

A. Thespis

Greek vase painting depicting a Dionysiac procession riding in a wagon (click to see larger image)First, except for what we are told concerning the origin of drama, Thespis is entirely unknown, a name that means essentially nothing to us except as the purported founding father of Greek drama. Worse yet, all sources which talk about him are late, none even vaguely contemporary. For instance, our information about his cart comes primarily from Horace, a Roman poet who wrote more than half a millennium after the age when Thespis would have to have lived. Moreover, that very few earlier sources mention him is particularly troubling if indeed we are to subscribe to him the invention of tragedy.

All this makes him sound more like a fabrication of later ages attempting to simplify theatre history by assembling what scant data there are under one name, in much the same way that young children in America are taught "George Washington was the founding father of the United States." Yet historians will note that the reality of early America is vastly more complicated, involving hundreds, if not thousands, of "founding fathers." The aggrandisement of George Washington is a way of simplifying the complex history of the America Revolution for those who have little room in their lives for the actual and complex realities of the past. Is Thespis' role in history comparable?

The historicization of George Washington can be telling in other ways as well. For one, the American general actually existed and is not a fictitious icon. One might conclude the same about Thespis except that there is no primary evidence for his existence as there is for the American founding fathers—the records of the Pre-Classical Age were scant, even in antiquity—thus, it would have been much easier back then to concoct a Thespis than it would be today to counterfeit a George Washington. Moreover, that very lack of evidence would also have driven the need to explain somehow tragedy's origin. One simple solution to the innocent question, "Where did Greek drama begin?," would have been to personify the complex evolution of early Greek theatre by forging a "founder," making it easier in general to assimilate a long and convoluted past.

Whatever the truth, by all fair standards of history, Thespis is undated, unattested, and unassociated by any credible primary source with any particular practices in early theatre. Except for the legends surrounding his name, we have no outside corroboration even of his existence, much less his contribution to drama. If he resembles any personage in the "forest primeval" of America, it is Evangeline or Paul Bunyan, not George Washington.

B. Tragoidia

Second, the Greek word tragoidia presents even more of a mystery. That it means "goat (trag-) song (-oidos)" is certain because there is no other way to interpret the word satisfactorily in Greek, but to what the "goat" refers is not at all clear. Goats are not known as prizes for winning playwrights in any other ancient venue, nor would one make a very attractive reward. Frankly, one hopes ancient playwrights competed for something a bit more becoming. Goats, in fact, feature nowhere in the extant primary records of ancient Greek theatre, so to conclude that they were sacrifices, as Brockett and others do, is pure speculation. Still there is no writing off the goats entirely, the way one can with Thespis, because "goat" is firmly anchored in the very name of the genre and that name goes back to the origin itself of the art form.

The desperation of this situation has led otherwise conservative and judicious critics to make uncharacteristically wild guesses at the reason goats are found grazing around the ancient stage. For instance, Margarete Bieber suggests that "goats" may be a nickname for the worshipers of Dionysus, the god in whose honor drama is performed at Athens. (note) While it is true that such nicknames exist for the adherents of other gods, for example, "bears" or "bees" for devotees of Artemis, there is no such recorded appellation for the celebrants of Dionysus. In fact, few gods beside Artemis have worshipers with nicknames—it seems to be something largely peculiar to her cult—so, on closer examination, the extrapolation has little merit.

Another avenue leads to more productive results. It has been suggested by another scholar that "goat-song" may refer not to goats as such, but to an ancient Greek slang term, "to goat," referring to the breaking or cracking of a young man's voice during puberty. (note) If ancient tragic choruses were performed by young men, this could make an odd sort of sense. (note) The term tragoidia would then be a joke, based on the cracking of the young men's voices—though not during an actual performance one hopes!

Names for dramatic genres are, in fact, known to be based on jokes elsewhere, for instance, the modern performance genre called "soaps." After all, could the reason for the name "soap" be easily reconstructed two millennia from now, if it were not known that there was once a connection between daytime drama and the detergent industry? Without such knowledge, future scholars might go for simplistic but reasonable-sounding explanations, such as, "These emotional melodramas were seen to ‘cleanse' the soul, and thus the genre came to be called ‘soaps'." It would be much harder to recreate the real reason for the name, that it is essentially an ironic jab at the early commercialism of this art form. All in all, the reason why there are goats in tragedy is an unresolved conundrum, though in light of theatre and ancient history the "goating voice" explanation makes better sense than suggesting that goats once served as prizes.

C. Aristotle's Poetics

Third, Aristotle and the theory of the origin of Greek drama presented in his treatise on theatre, The Poetics, deserves to be read in full.

Aristotle, The Poetics, Chpt. 4.1-6 (1449a) [The Poetics, according to one modern commentator, "is often elliptical in expression, and some of its ideas seem inadequately, at times almost incoherently, developed." (note) Thus, it is hardly possible to situate Aristotle's analysis of the origins of tragedy within the larger context of what Aristotle is saying prior to this passage. Suffice it to say, he turns abruptly from a description of mankind's "instinct for imitation (mimesis)" out of which he claims tragedy arises, to the origin of Greek dramatic forms. After that, without stating any reason for making the transition, he moves on to the general nature of tragedy and comedy. In both tone and style The Poetics reads more like lecture notes than a polished critical thesis. In light of all this, I have translated the text with the aim of capturing both Aristotle's exact words and also his punctuated and compressed style. If the English below makes difficult reading, believe me, so does the Greek.]

Arising from a beginning in improvisation, both itself (i.e. tragedy) and comedy, the former (arising) from those leading the dithyramb, and the latter from those (leading) the phallic songs which still even now in many of our cities remain customary, little by little it (i.e. tragedy) grew making advances as much as was obvious for it to do, and after having undergone many changes, tragedy came to a stop, when it attained its own nature. Aeschylus increased the number of actors (literally, "interpreters" or "answerers" ) from one to two for the first time and he reduced the chorus' business and prepared the dialogue to take prominence. Sophocles (introduced?—there is no verb here) three (i.e. actors) and scene-painting. And also the grandeur (or "length" , i.e. of tragedy; was increased? by Aeschylus? Sophocles?—again, no verb!). From slight (or "short" ) stories and joking expression, since it evolved out of satyric forms, it became reverent (only) rather late, and the meter changed from tetrameter (i.e. comical, fast-paced) to iambic (i.e. normal, conversational). At first they used tetrameter since drama was satyric and more dance-related, but with the rise of speech (i.e. as opposed to "song" ) the nature (i.e. of tragedy) on its own found its proper meter. Indeed, the most conversational of meters are iambics. The evidence of this, we speak iambs (i.e. daDUM daDUM) most of all in conversation with one another; (we speak) hexameters (i.e. the meter of epic, DUMdada DUMdada), on the other hand, infrequently and when we depart from a conversational tone. And also the number of episodes (or "acts" ; was increased? –no verb). And as to the other matters, as each is said to have been set in order, let that be said by us. For it would be perhaps a great task to explain each thing individually.

Consider the comments of D.W. Lucas, a modern classicist, on Aristotle's views of early Greek drama. (note)

Finally, it is worth observing that Aristotle's account of the origin of tragedy from a basically ludicrous form fits so badly with the scheme of development presented in the first part of the chapter that he would not have been likely to offer it unless he had been reasonably confident that it was true. Here again it is important that he knew more than we do about the early satyr play.

Aristotle by Rembrandt (click to see larger image)But did he? Lucas raises a central question faced by all who confront Aristotle's conclusions here: "How much better was his information than ours?" Aristotle clearly had a few advantages we do not. He lived much closer in time to early Greek drama than we do and, to judge from the dramas he cites and quotes, had access to fifth-century (classical) plays we do not. Furthermore, he spent at least some of his life in ancient Athens and was personally involved in Athenian culture. Whether or not he did, he certainly could have gone to the ancient theatre, so he speaks from at least the assumption of having seen Greek tragedy in action in its day, something beyond our grasp.

While these do, in fact, seem like overwhelming advantages, on close inspection none are so imposing that we cannot question his thesis. Aristotle lived from 384 to 322 BCE, which is about two hundred years after the period he is discussing. If that does not seem like a long time, especially in the great sweep of history, one should remember how long two centuries can be. By our standards that is like going back to the early days of the United States. How much would a person today be able to remember about that period without historical records to go on? Credible oral histories cannot exist at such a remove. So it is fair to assume Aristotle is dependent on what data he can collect from that period or, to put it another way, The Poetics is by definition secondary evidence —not primary!—about early Greek theatre.

Nor does it help those who would defend him as an authority on primordial Hellenic drama that in all of The Poetics he does not quote from one piece of primary historical evidence written in the sixth century BCE: no playwright's work, no dramatic commentary, no record of audience reactions, nothing! He either did not have access to such information or chose not to reference his sources in The Poetics. If he did have access to data about early drama, it seems likely a researcher who is otherwise so meticulous would have explored those avenues and included them in his finished work. The greater likelihood is that all but no records of early drama had survived to his day.

To see the point here, it may help to think of this situation in modern terms. For instance, in several centuries or more from now there will probably no longer exist comprehensive records of the very earliest phases of film-making: Edison, D.W. Griffith, Cecil B. De Mille's first silent version of The Ten Commandments. (note) How will scholars in this future age piece together the origins of that all-important watershed in art, the invention of the movie? In the absence of clear data, will they be able to see its dependence on theatre, the novel, opera and artistic movements like impressionism? So, any way you look at it, Aristotle's primacy as a researcher of early Greek theatre is not as great an advantage as it might appear at first.

Nor is his cultural advantage as great as it might seem on the surface. The Athens he knew was very different from that into which drama was born. Between the days of early tragedy and Aristotle's lifetime, the Classical Age had come and gone, leaving in its wake a very changed civilization and view on life. Aristotle, without doubt, was acculturated to see the world through eyes quite different from those which had directed and witnessed the birth of Greek drama.

Moreover, lacking immediate experience with the full cultural framework of the time in which drama first came to light, he was as prone as anyone to make misassessments about the inclinations and motivations of his remote ancestors in the Pre-Classical Age. Granted, even if faulty, his reconstructions of the past probably appeared sensible to him and his peers—and perhaps to many today, too—nevertheless, his conclusions will have diminished validity if they do not address directly the age in question. In other words, Aristotle was at some risk of making the same mistake all historians are: his conclusions possibly say more about himself and his own times than the subject he is studying.

D. Conclusion: How right is Aristotle about the origin of tragedy?

With all that in mind, let us examine Aristotle's hypothesis about the origin of Greek drama. In essence, Aristotle looked at theatre-like entertainments and ritual celebrations that were not tragedy as such and that had survived to his day and seemed "primitive" to him—remember his words: "which still even now in many of our cities remain customary"—and concluded that these celebrations must somehow have played a direct role in the evolution of tragedy. The postulates underlying The Poetics—that is, Aristotle's assumptions about what constitutes historical data in the case of early theatre and how it should be ordered—are remarkably similar to those adopted by Frazer in The Golden Bough.

And that opens Aristotle up to the same criticisms as those which have been directed at Frazer. Basically, Aristotle assumes that something which looks "primitive" to him must be "early," that ritual leads to theatre and that there is inevitable "progress" toward more complex forms: "[tragedy] grew making advances as much as was obvious for it to do, and after having undergone many changes, tragedy came to a stop, when it attained its own nature." But who's to decide what is "obvious" or "its own nature"? Without a clear and documented basis in data, such assumptions cannot carry much weight.

Greek vase depicting a dithyrambic chorus and flute-player (click to see larger image)The dithyramb, what Aristotle cites as the art form from which tragedy arose, also poses several obstacles to the construction of a coherent case for some sort of linear development in the performing arts of early Greece. To begin with, no early dithyrambs survive from antiquity. Indeed, until recently we did not have any dithyrambs at all, but in the last century a few have emerged from the sands of Egypt. Unfortunately for our purposes here, they are later dithyrambs by a classical—not pre-classical!—poet named Bacchylides, and their connection to the earlier dithyrambs about which Aristotle speaks is unclear (see Reading #1). It is enough to note that Bacchylides' dithyrambs do not employ spoken language—they are meant to be sung entirely—nor do they entail elaborate characterization, as tragedy usually does. That means it is questionable even whether dithyrambs represented institutional theatre, much less served as the predecessor of tragedy.

Indeed, one of Bacchylides' dithyrambs includes only a single character, while another has no characters at all, only a chorus, something unheard of in extant tragedy. Moreover, these dithyrambs are episodic, meaning they do not have conventional plots with a clear beginning, middle and end, again unlike all known Greek tragedies. What they seem to be are short "epics" cast in the form of lyric poems to be performed by a chorus, with language poetic and elevated, focusing on the genealogies and epithets of heroes and, as such, peopled by characters huge and lofty, mostly gods and heroes, not the desperate, stricken mortals who dominate the tragedies available to us. (note)

In sum, none of this adds up to a compelling case, at least superficially, for dithyramb as the precursor of tragedy—put simply, the dithrambs we have simply do not look much like the tragedies which are extant—so is it possible, then, that Aristotle is wrong about the origin of tragedy? Before entertaining such a notion, we must admit that it would be foolish to cast away lightly the opinion of one of the finest minds ever and, even if his report constitutes secondary evidence, a researcher who stands much closer to the actual event in question than we are. But let us assume for a moment that Aristotle is mistaken. It is still incumbent on his prosecutors to show how and why, and to present some better case than he does—no small task!

Suppose, then, that we had access to all the dithyrambs ever written in early Greece and we could see for ourselves that there were, in fact, no dithyrambs which resembled tragedy closely enough to posit a cause-and-effect relationship, as Aristotle does. He is not an idiot or liar, so it is incumbent on us to find some reasonable explanation for his misconstruction of the data. Fortunately, that is not an insurmountable challenge.

The Poetics does not focus on the issue of origins with nearly the sort of attention modern theatre historians might wish for. There is, in fact, very little in this work about the origin of drama and, to judge from the fractured density of Aristotle's language—if it is even his language and not the notes of someone listening to his lectures, as some scholars suppose!—he did not revise the work to the degree typical of his other works. Thus, it seems fair to say that the question of the origin of drama was not central in his mind. To wit, he does not cite the sources on which his opinion is based, including not even a single quotation from an early dithyramb or anything which would ground his argument in compelling, primary data.

Now let us assume the converse, that Aristotle is correct and evidence actually once existed that there were dithyrambs that looked like tragedies, at least on paper. He could, after all, never have seen such dithyrambs performed since he lived so long after the fact and so it still does not argue for an evolutionary link between these types of performance. Aristotle could be making an error to which many historians of theatre are susceptible. That is, he has assumed a connection between theatrical genres which happen to look alike in written form—in this case, dithyrambs and tragedies both include choruses, lyrics, characters, scenes, dramatic tension, climax and so on—but, when seen in theatres in performance, they probably looked and were very different. Much the same could be said for opera and oratorio.

Perhaps, a more modern analogy will help clarify the situation. For instance, compare the screenplays of animated films and live-action movies. They look, in fact, very similar, but the finished products in performance are worlds apart and, as we know, grew out of vastly different artistic milieux. All in all, if it is true that Aristotle is projecting a hypothesis based primarily on evidence from his own day and not the Pre-Classical Age, he shows himself to have been a purveyor—but not necessarily a spectator—of plays who has made a classic "reader's error." He saw what looked to him like a similarity between different forms of performance art, when, in fact, it was actually only a superficial, "papyrus" likeness, and from that he assumed there was a direct evolutionary relationship between the genres.

It is possible to see that same sort of error elsewhere in The Poetics. Aristotle claims—or seems to claim since the text is gravely abbreviated—that "Sophocles (introduced?) three (actors) and scene-painting." (note) But it is obvious even at our remove that, if he did, Sophocles must have done this very early in his career, since Aeschylus utilizes three speaking actors in The Oresteia, a trilogy produced in 458 BCE. Sophocles' career had begun only a decade earlier (ca. 468 BCE). It seems unlikely that a novice, before even having earned his dramatic stripes, would have been allowed to reformulate the rules in as rigidly controlled an environment as the prestigious, award-granting religious festival of the City Dionysia. That increases the probability that Aristotle—if he actually wrote these strangled words—may be incorrect on this point, and the reason for this is also not altogether unfathomable.

Production photograph of a modern performance of Aeschylus' Agamemnon, with Clytemnestra, Cassandra and Agamemnon together on stage (click to see larger image)Though Aeschylus' later plays do, in fact, utilize three speaking actors—and perhaps at one point even four (Libation-Bearers 900ff.)—never at any moment is there a three-way conversation on the stage, that is, no "trialogues" anywhere. If one were to read Aeschylus' dramas quickly, not following closely the assignment of parts demanded by the text which involves imagining the actors' movements backstage, it would be very easy to conclude that Aeschylus never employed more than two speaking actors in the execution of his drama. Careful scrutiny of Aeschylus' plays, however, belies this presumption.

For instance, in the great confrontation between Agamemnon and his wife Clytemnestra in Aeschylus' Agamemnon, the only characters who speak are Agamemnon and Clytemnestra (and the chorus, of course). Yet there is another character on stage who speaks later, though not in that scene, Cassandra. Both Agamemnon and Clytemnestra refer to her presence on the stage and the latter even attempts to speak to her but Cassandra never says a word to either of them. In a later scene, however, she does finally talk, and without having exited the stage since her original appearance on stage. So, it must be the same actor portraying this character throughout. (note)

It is clear, then, Aeschylus' Agamemnon does, in fact, demand three speaking characters, though a cursory glance at the text would seem to indicate not. In sum, the case is strong that Aristotle has made a "reader's error"—if, in fact, he wrote the words "Sophocles three" and in saying this meant that the tragedian had introduced the third actor—and has assumed from a cursory overview of Aeschylus' drama what the playwright's scripts called for in terms of performers. That is, Aristotle did not fully envision the requirements which theatrical production requires. It would accord well with his complete neglect of any discussion regarding rehearsal and the physical dimension of stage productions in his day.

All this casts Aristotle's assessment of early theatre in, at best, a mottled light. This type of misconstruction suggests he did not seek out the theatre recoverable in drama. In other words, he reads and perhaps listens to drama in the "theatre" of his mind, but he does not watch it as carefully as the theatre demands, which makes him liable to certain types of critical errors. In the words of Ibn Kahldun, he shows an "inability rightly to place an event in its real context, owing to the obscurity and complexity of the situation. The chronicler contents himself with reporting the event as he saw it, thus distorting its significance."

If so, it opens the possibility of consigning Aristotle's theory linking early tragedy with dithyramb to the same family of oversights as his purported attribution of the third (actor) to Sophocles. To wit, the philosopher saw apparent similarities in the written texts of dithyramb and tragedy and from that assumed some sort of evolutionary connection—it is not an illogical conclusion by a "reader's" standards since these texts would have looked alike on "papyrus"—nevertheless, the graphic similarity may not have an equivalent validity in the theatre of pre-classical Athens. Or, to put it another way, Aristotle is a "lumper" who postulates a direct line of development in the data by linking together forms he knows to have existed early on. On the other hand, the data—what few there are!—leave much room to be a "splitter," too.

That opens the door to supposing that dithyramb was not the "father" of tragedy, but a similar-looking form of entertainment that grew from the same stalk of the "family tree" as tragedy did. In other words, dithyramb was not necessarily a direct ancestor of tragedy or even a close relative, but rather a sibling or cousin of some sort. All in all, the standard view of the origin of Greek drama in the Pre-Classical Age—whether we orchestrate our hypotheses around Aristotle, Thespis or goats—is hardly unassailable. It is limited at best, and a real possibility exists that it is simply wrong.


II. Those Few Known Facts about Early Greek Drama

In such a situation, it is wise to step back and re-assess the data. What do we know about early Greek drama, where is there room to speculate and what limitations should there be to any speculation? This much at least is certain:

• Greek theatre must have arisen in the early or mid-500's BCE, because drama is simply not mentioned in texts or credible sources prior to that time.

• By 534 BCE theatre is clearly under way because that was when Pisistratus, the tyrant of Athens, inaugurated the urban ceremonies honoring the god Dionysus, a festival that for the first time in recorded history included theatrical performances. Thus, all reasonable evidence suggests that Athens was the cradle, if not the birthplace, of early drama.

• That innovative festival, called the City Dionysia, is certainly worth looking at in detail, because from what is known about its general mood and structure it may be possible to speculate constructively about Pisistratus' reasons for making drama part of this festival.

• Crucial, too, in all this is Dionysus, a god seen by the Greeks as a foreign import. His worship was said to have originated in Asia Minor (modern Turkey) and to have entailed several non-Greek elements, such as "orgiastic" rituals.

So, amidst all the conflicting and confusing data, there do exist at least some things we can rely on. If these data do not form a seamless bridge to the truth, they are at least stepping stones on which to cross a very treacherous torrent of conflicting information.

Greek vase painting of a maenad, a female worshipper of Dionysus (click to see larger image)Let us begin, then, by investigating the worship of Dionysus because it may shed light on the nature of early theatre and, perhaps, even help sort out the evidence. The alien nature of Dionysiac worship is reported to have precipitated more than one crisis in Greek society. As many today would also, the ancient Greeks initially distrusted any social force that preached release of inhibitions, promotion of the downtrodden in society—and women, in particular, clearly one audience at whom the cult was directed—and general dancing, drinking and cavorting. (note)

The form of worship sanctioned in Dionysian religion was ecstasy, literally "the act of standing outside oneself." At the highest pitch of the celebration, it was believed that worshipers left themselves, as the god entered their bodies, and they could then perform miracles and wondrous acts beyond normal human capabilities. That the "impersonation" of a human by the god bears some resemblance to what an actor does in performing a drama has led many a scholar to assert an evolutionary connection between this sort of worship and the performance of drama. For this and other reasons, Bieber asserts: "The religion of Dionysus is the only one in antiquity in which dramatic plays would have developed." (note)

There is much to applaud in Bieber's approach. To begin with, the festivals of Dionysus regularly included dance, and at the same time the impersonation of deities, the use of masks and parades of celebrants who can be seen to resemble tragic choristers. Besides its innate theatricality, Dionysian religion was also a later import to Greece and thus, unlike better established cults with age-old rituals, it was open to new formulations in its worship, celebrations that might involve all sorts of "entertainment" (epic narration, lyric singing, choruses, and so on). Furthermore, the story of Dionysus as told in myth is varied and full of different events, indeed a rich storehouse of different-looking episodes that would make a fine arena for theatre, if a playwright were inclined to dramatize them. In sum, looking on the surface of things, one is forced to agree with Bieber that Dionysus' world in early Greece seems like the perfect womb for fetal drama. Now all that's needed is strong evidence corroborating that was what actually happened.

Unfortunately, for all the sense Bieber's argument makes to many today, the corroborating evidence that what she proposes did, in fact, take place is simple not there. Moreover, when one looks below the surface, there are powerful counter-arguments to the thesis that drama arose directly or smoothly from Dionysiac cult practices. While the rituals attested as Dionysian are indeed theatrical, they are not by any means institutional or autonomous theatre. Yes, there is impersonation, which assumes that there are also watchers and watched, but the nature of the words spoken in these performances is unknown. If the scripts of these ceremonies did not vary but comprised basically the same hymnic praise of the god year after year, it is less like a play than an Easter mass, which is certainly dramatic but not drama as such. Finally, for all the possibilities of creating new and innovative scripts which the mythic sagas of Dionysus inherently contained, there is little evidence that such play-texts were ever created in classical drama. It's a fine suggestion but it doesn't seem to have played out that way in antiquity.

Roman mosaic of Dionysus in triumph, riding a leopard (click to see larger image)And there is another problem. A well-known Athenian maxim from the day claimed that Greek tragedy had "Nothing to do with Dionysus." This popular saying dates back to at least the Classical Age, where it appears to have been assumed, as a joke almost, that, although tragic drama was performed at the Dionysia in honor of its eponymous deity, the plays rarely revolved around Dionysus or his worship explicitly, which is something we can see for ourselves. All but one of the thirty-three preserved classical tragedies deal with myths which do not center around Dionysus. The question, then, is not about the validity of this adage, but how far back in time it applies. Does the tendency well-evidenced in the Classical Age not to perform plays about Dionysus at the Dionysia go back into the Pre-Classical Age, perhaps even to the origin of drama itself?

That, in turn, brings up another problem, the very different modes in which Dionysiac ritual, based as it is on ecstatic revelry, and tragedy are conducted. Almost all the tragedies we have access to are fairly serious in tone, many very serious. If tragedy arose from the riotous, grape-stomping, orgiastic rituals of Dionysus worship, how and when and why did this total about-face in tone and attitude take place? The result of these questions is that for all its apparent validity the quick-and-easy association of Dionysiac ritual and early tragedy is much more problematical than it may seem at first glance, yet another lesson in the dangers of applying what to us makes sense to a past culture where sensibilities were very different from our own.


III. Two Modern Theories about the Origin of Greek Drama: Murray and Else

Given that, theatre historians must seriously consider the possibility that tragedy did not naturally grow out of Dionysiac ritual but was somehow layered onto it. Indeed three important theories have been proposed building upon that premise: Murray's suggestions surrounding the "year-spirit," Ridgeway's "Tomb Theory," and Else's proposal that Greek drama is the product of Thespis' and Aeschylus' genius. Let us review them now.

A. Murray, Cornford and the Eniautos Daimon

First, Gilbert Murray, and later F.M. Cornford, posited that drama evolved out of a celebration of the year-spirit—in Greek, eniautos daimon ("annual spirit")—a ritual tied to the changing of seasons. (note) In Murray's view, comedy began as a celebration of the marriage of gods and the fertility that results of their happy union; conversely, tragedy originally mourned some god's death and usually took place in autumn. Much of this work built upon Frazer's theories with all their concomitant biases and positivistic fallacies. When that attitude came under fire, Murray's theoretical infrastructure also began to crumble.

From another quarter, critics of Murray's work pointed out that, while the term is attested, there is remarkably little evidence for the celebration of the eniautos daimon in Greece. On top of that, year-spirit celebrations do not, at least on the surface, resemble tragedy much. To cap things off, it was noted that Greek tragedy was regularly celebrated in early spring, just as the year was renewing itself—the wrong time of year, to judge by seasonal change—and comedy was performed at the same festival. According to many scholars, the congruence of diametrically opposed rituals according to the year-spirit's religious calendar was a fatal blow to this theory. Except for stopping a show here and there, the weather just doesn't seem to have played a big part in Greek theatre.

B. Ridgeway and the Tomb Theory

Along similar lines emerged a second theory, one rooted in very different data. It set out to explain why tragedy had "nothing to do with Dionysus." This proposition began with a passage in Herodotus' Histories which offers, if rather succinctly, another possible avenue for the distillation of drama out of non-Dionysian ritual:

Herodotus, The Histories, Book 5.67.4-5 [Herodotus is discussing the rise of Athenian democracy which he sees as the fallout of a conflict between two men, Cleisthenes and Isagoras. In reporting about the former, he digresses into a discussion of Cleisthenes' grandfather, also named Cleisthenes. This older Cleisthenes had tried, according to Herodotus, to rid Sicyon, a city in southern Greece, of the worship of the hero Adrastus and impose on it that of a hero imported from Thebes, Melanippus. Adrastus had lived much earlier (in mythological times) and later his spirit was seen to protect the city. Thus, the Sicyonians worshiped Adrastus as a demigod and performed certain rituals in his behalf, which Herodotus now describes.]

But the Sicyonians by tradition very excessively worshiped Adrastus, the reason being that the country (i.e. Sicyon) was once Polybus' very own and Adrastus was Polybus' maternal grandson, but childless at the time of his death Polybus gave to Adrastus the realm. So, in other respects the Sicyonians used to honor Adrastus but particularly with respect to his sufferings (or "experiences" ) they held celebrations with tragic choruses, honoring not Dionysus but Adrastus. Cleisthenes (i.e. the older) returned (or "delivered over" ) the choruses to Dionysus and the other sacrifices to Melanippus.

This has led some scholars, foremost among them William Ridgeway, to propose the theory that this sort of hero-worship was the font from which Greek tragedy came, in Ridgeway's words: (note)

Tragedy proper did not arise in the worship of the Thracian god Dionysus; but sprang out of the indigenous worship of the dead . . . the cult of Dionysus was not indigenous in Sicyon but had been introduced there . . . and had been superimposed upon the cult of the old king; . . . even if it were true that Tragedy proper arose out of the worship of Dionysus, it would no less have originated in the worship of the dead since Dionysus was regarded by the Greeks as a hero (i.e. a man turned into a saint) as well as a god.

This theory, called the "tomb theory" or "hero-cult theory," has certain advantages over the supposition that tragedy arose from Dionysian worship. Principally, it addresses directly the content and form of all later, extant classical tragic dramas which invariably include choruses, most often focus on human heroes over immortals, and frequently center around their mortality. The performance of a hero's life at his "tomb," presumably a narrative recreation of his life and death accompanied by a chorus, does indeed bear a strong, if superficial, resemblance to later tragedy.

But there are serious drawbacks to the "tomb theory." Primarily, this evidence is so scant that, if the theory is correct, the discussion about the origin of Greek tragedy is over. This leaves nothing else to say, for there is no other information about such rituals to be found in the historical record. Herodotus' account constitutes our one and only source of information on this custom. No other historian, no later commentator, no vase painting, nothing else in antiquity refers to this sort of entertainment, as far as we know.

Close attention to Herodotus' language also raises some serious issues about the validity of this perspective. The historian appears to be pushing the very same point as Ridgeway, that tragedy arose from these sorts of celebrations ("the Sicyonians . . . held celebrations with tragic choruses, ..." ). To call them "tragic" choruses is essentially to retroject a later cultural term onto an earlier custom and begs the point that these songs in some way constitute "proto-tragedy." Note also that Herodotus seems to be denouncing the cult of Dionysus as the font of tragedy when he says "honoring not Dionysus but Adrastus." The great "Father of History" looks to be asserting the same argument as Ridgeway and his modern followers.

In other words, he does not just happen to record luckily for us the true origin of drama, but is making the very point on very slender evidence. If so, and if he'd had more data to this effect, Herodotus would surely have cited it. But he does not, and that speaks poorly for the credibility of this thesis. All in all, given the scarcity of information here and Herodotus' potential bias, we must take the "tomb theory" with a grain of salt, if not a whole shaker. After all, when the facts lead to a dead end, we're at "Thespis" all over again.

C. Else's Theory

Speaking of Thespis, a third theory also downplays the role of Dionysian worship in the formation of early Greek drama and has won generally wider support. Gerald Else, a twentieth-century classicist, suggested looking at drama, not as the product of some other type of ritual or celebration nor even the evolution of older forms into something new and different, but as a unique event without true forebears. Else calls Greek drama "the product of two successive creative acts by two men of genius," meaning Thespis and Aeschylus. (note) In other words, tragedy simply "happened" because two men in fairly close succession saw possibilities in the art form and in the process of reshaping it created drama.

Else's theory has several clear advantages over the others. First and foremost, it explains the chronology and locality of tragedy's genesis. There can be little doubt that it happened in Athens at some point during the sixth century BCE. Even if there are theatre-like entertainments elsewhere in the Greek world prior to tragedy, there is no competent ancient source that does not recognize the Athenians in some way as the inventors of drama—there is not even a shred of credible evidence to suggest otherwise—so to Else, the dynamic combination of Thespis and Aeschylus, not some poorly attested social institution, was the real force behind the drive to create this art form. Tragedy is the unique concoction of men who lived at the right moment in the right place to bring theatre to life.

The form and content of tragedy, then, do not matter in the same way they do if we suppose that tragedy is an elaboration of certain celebrations or rituals. What tragedy became is simply what Thespis and Aeschylus saw as practicable entertainment, because in manufacturing tragedy, they simply used what they wanted to of the celebrations and arts around them. If tragedy looks like it has pieces of Dionysiac ritual or hero-worship or lyric poetry or epic, it is because that is what Thespis and Aeschylus selected at will and from these elements forged drama. It must also have been to some extent what their public wanted, in the language of modern advertising, what "sold."

Essentially, what Else is doing is subverting the whole mentality of "evolution," the sense that art entails cause and effect which over time lead to visible changes in its presentation. Instead, he has posited a theory of spontaneous generation for tragedy—a student of mine aptly called this the "creationist theory" of Greek drama—inasmuch as Else is essentially saying that early Greek theatre did not evolve, but was "created," born several times, in fact: in Thespis' early touring shows, and again in 534 BCE when Pisistratus instituted the City Dionysia in Athens and decided to include drama among its celebrations, and finally in Aeschylus' masterful hands during the early Classical Age when it took on the shape in which we know it. It's hard to argue against Else's contentions, and thus many scholars today follow his thesis.

But, like all the others, his theory is not without its drawbacks. It leaves little room for universal application, because it applies by definition only to Athens in the sixth and fifth centuries BCE. In premise, Else's thesis cannot be expanded to include China or India, where we know theatre also developed. There is no room for a common human element when one sees Greek drama as "unique." While this may be true, one should hope it is not, because, just as with Ridgeway's Herodotus-based theory, the argument would be at a dead end.

To many scholars, that alone is unsatisfying. Furthermore, although there is much to question in Aristotle's thesis, many scholars are ill at ease throwing the great philosopher's notions out the window altogether, even if there seem to be good reasons for doing so. All in all, it would be much more satisfying if we could incorporate Aristotle's ideas in some form at least. And finally, to say that in the end "geniuses invented drama" seems somewhat obvious. As another student of mine once blurted out in class when I was outlining Else's thesis, "Well, duh!" Unfortunately, I must concur.

D. Conclusion

In sum, the facts about early Greek theatre appear to lead to shaky or unsatisfying theories at best, and the whole situation looks lamentable. Facing a seemingly insoluble problem, we should probably throw up our hands and surrender, especially given the underwhelming body of evidence we bring into the arena. With little hope of uncovering more data—remember that Aristotle seems to have had not much more to go on than we do, and if he didn't, is there any reasonable expectation we ever will?—the gavel should probably come down on this mistrial and we should all leave the courthouse. Unless, of course, the reason there isn't better evidence is not because important data have failed to survive but because the evolution of drama did not take place the way any of these theories suggest.

In other words, perhaps we lie at the heart of the problem. Perhaps we are expecting to find something that is not there because it never was there, or there was so little of it that none is likely to have survived the ravages of time. To put it another way, perhaps we are looking for a preponderance of data that never existed. In that rests the only real hope of finding credible answers about early Greek drama, especially in light of the grave improbability that we will stumble across new data about the events surrounding the origin of Greek drama. Perhaps we already have all the evidence we need and all we need to do is to recognize it as such.

Terms, Places, People and Things to Know
Thespis
Tragoidia
Aristotle
The Poetics
Dithyramb
Bacchylides
Trialogue(s)
Dionysus
City Dionysia
Ecstasy
"Nothing To Do With Dionysus"
Year-Spirit
William Ridgeway
Tomb-Theory
Hero-Cult Theory
Gerald Else

Course Description
Class Grading and Projects
Chapters
Syllabus
Slides
A Guide to Writing in History and Classics

 

Creative Commons License
This work is licensed under a Creative Commons Attribution-No Derivative Works 3.0 United States License.