Introduction
In undertaking to treat of a subject concerning hardly a detail of which can any statement be made without the possibility of dispute, the unfortunate necessity rests upon me of beginning with three topics which are the most controversial of all—the origin of tragedy, the origin of comedy, and the Greek theater. Instead of trying to conceal our ignorance on these matters by vague generalities, I shall set forth such data as are known, and attempt, clearly and frankly, to erect hypotheses to answer the questions that most naturally arise, even though this very striving for clearness and frankness will expose me to attack. I believe with Bacon that “truth emerges sooner from error than from confusion,” or, as a recent writer has expressed it, that “the definitizing of error is often the beginning of its disappearance.” Limits of space will require, at many points, a dogmatic statement of my views without stopping to examine the evidence from every angle. It must be understood, however, that no account of these subjects, whoever its author or however detailed his treatment, could find universal acceptance or anything approaching it.
The Origin of Tragedy.12—It is still the canonical doctrine, though its modern history goes back no farther than Welcker’s book on the Satyrspiel in 1826 and though no conclusive testimony for this view can be cited more ancient than Byzantine times, that satyric drama was the intermediate stage in the derivation of tragedy from the dithyramb. The argument runs somewhat as follows: The dithyramb was an improvisational song and dance in honor of Dionysus (Bacchus), the god of wine, and was performed by a band of men provided with goatlike horns, ears, hoofs, and tails and clad in a goatskin (or in a goat-hair loin-band) in imitation of Dionysus’ attendant sprites, the satyrs; on account of this costume the choreutae (members of the chorus) were sometimes called tragoi, which is the Greek word for “goats”; in certain localities, as the dithyramb became quasi-literary and took on a dramatic element, its name was changed to satyric drama; still later, as these tendencies increased, especially through the addition of an actor, the satyr-play came to be called tragoidia (“goat-song”), derived from the nickname applied to the caprine choreutae; the chorus still consisted of satyrs and, since these were licentious, bestial creatures, the performance was yet crude and undignified; Aeschylus (525-456 B.C.) was possibly the first to abandon satyric choreutae and was certainly the first to raise tragedy to the rank of real literature; during the fifth century each poet was required to follow his group of three tragedies at the dramatic festival with a satyr-play as a concession to the satyric origin of the performance.
In recent years, essential supports of this doctrine have slowly crumbled away before searching investigation; at present, scarcely a single clause in the foregoing sketch would escape unchallenged by some scholar of deserved standing. An ever-increasing number of students believe that tragedy is not the child of the satyr-play, but that the two are separate in their origin. Unfortunately, however, these dissenters, including such men as Dr. Emil Reisch of Vienna, Mr. Pickard-Cambridge of Oxford, Professor Wilhelm Schmid of Tübingen, and Professor William Ridgeway of Cambridge, though they are unanimous in rejecting Welcker’s hypothesis, cannot agree among themselves as to a constructive policy. My own view is that tragedy and satyric drama are independent offshoots of the same literary type, the Peloponnesian dithyramb. The former came to Athens from Corinth and Sicyon by way of Icaria. Somewhat later the latter was introduced directly from Phlius by Pratinas, a native of that place. My reasons for these opinions will develop in the course of the discussion.
Very recently, notable efforts have been put forth to interpret the religious practices of the Greeks, partly in the light of anthropology and partly in accordance with the new psychological method which inquires, not what the god is, but what are the social activities and the social organization of his devotees. Whatever may be said for these avenues of approach in other respects, in practice those who employ them have shown more eagerness to assemble data which might be considered confirmatory of their theories than to reach an unprejudiced interpretation of the whole body of ancient evidence. Thus, much has been made of present-day carnivals in Thessaly, Thrace, and Scyrus,13 and these ceremonies are employed as if they were assured survivals of the primitive rites from which Greek drama developed and as if their evidence were of greater value than the most firmly established data in the ancient tradition. Now the a priori possibility that these carnivals should retain their essential features unchanged through two and a half millenniums amid all the vicissitudes which have come upon these regions must be pronounced infinitesimal. And an examination of the details confirms this impression. Certain parts of the ceremonies are parodies of the Christian rites of marriage and burial. Not only an Arab but also a Frank appear in the cast of characters. Though Phrynichus is said to have been the first to represent female rôles,14 such rôles abound in these modern plays. Yet there is another defect in this assumption which is still more serious. If there is one well-authenticated fact in the history of Greek drama, expressly stated in ancient notices and fully substantiated by the extant plays, it is that tragedy arose from a choral performance and only gradually acquired its histrionic features. On the contrary, these carnivals are predominantly histrionic; there is either no chorus or its rôle is distinctly secondary. Had Aristotle been guilty of such a faux pas, we can easily imagine the derisive comments in which modern investigators would have indulged at his expense.
Of course, our evidence is far from being as complete as we could wish, and must therefore be supplemented at many points by conjecture pure and simple; but this fact does not justify us in throwing all our data overboard and in beginning de novo. In this matter we have been too prone to follow a practice which the late Professor Verrall characterized, in a different connection, as follows: “We are perhaps too apt, in speculations of this kind, to help a theory by the convenient hypothesis of a wondrous simpleton, who did the mangling, blundering, or whatever it is that we require.”15 Now, whatever may be true in other cases, Aristotle at least was no “simpleton,” competent only to mangle his sources of information; and furthermore, apart from certain ethnographic parallels which are of only secondary importance after all,16 our fund of knowledge in this field is in no wise comparable with his. In fact, except for the extant plays our information is almost confined to what we derive, directly or indirectly, from him. Since this is so, what can be more absurd than to reject his conclusions and have recourse to unhampered conjecture?
But if we are to hold fast to Aristotle, one precaution is necessary—we must be sure that we do not make him say more or less than he does say. He wrote for a very different audience from that which now reads his words and with a very different purpose from that to which his book is now put. And these factors often render him enigmatical. This resulted also from his frequently assuming a familiarity with things which now cannot always be taken for granted. As Professor Bywater expressed it: “It is clear from Aristotle’s confession of ignorance as to comedy that he knows more of the history of tragedy than he actually tells us, and that he is not aware of there being any serious lacuna in it.”17 Thus, Aristotle says that tragedy was “improvisational by origin” and, more specifically, was derived “from the leaders of the dithyramb.”18 Though this expression unhappily is somewhat lacking in precision, the main item, that the dithyramb is the parent of tragedy, emerges from any interpretation. Ridgeway may proceed to dissociate the dithyramb from Dionysus and to derive it from ceremonies at the tombs of heroes if he choose; however unwarranted, that is at least logical. But to ignore this statement of Aristotle’s and to seek, as many do, to trace tragedy back to δρώμενα (“ritual acts”) of various kinds by another line of development transgresses good philological practice.
There is an unfortunate facility in such attempts. Tragedy embraced many diverse elements in its material and technique. Accordingly, whatever anyone sets out to find, he can be almost certain of discovering there. Thus, Dieterich with his theory of the development of tragedy from funeral dirges, the Eleusinian mysteries, and various aetiological sources; Ridgeway with his tomb theory; Miss Harrison with her “Year Spirit” (the Eniautos-Daimon) and sympathetic magic; and Murray with his attempt to reconcile and expand the Dieterich-Harrison theories, all find confirmation for their views in the same body of dramatic literature. The very facility of such analyzing is its undoing.
Moreover, despite numerous attempts to the contrary, the real nature of the primitive dithyramb can scarcely be a matter of doubt. Plato, who was also no “simpleton,” defined it as a song in celebration of the birth of Dionysus.19 Now since the dithyramb is known to have been opened up to a wider range of themes considerably before Plato’s time, his definition must apply to the original meaning of the term. This interpretation does not remain unsupported. Thus, the first extant instance of the words occurs in a fragment of Archilochus (ca. 680-640 B.C.), who declares that he “knows how, when his heart is crazed with wine, to lead lord Dionysus’ dithyramb.”20 It should be observed that Archilochus does not say that he knows how to write a dithyramb, but how to take part in one as a drunken ἐξάρχων (“leader”). Such a performance was doubtless, as Aristotle said, largely improvisational, being perhaps coupled with the rendition of some ritual chant (καλὸν μέλος). Dionysus is characterized as θριαμβο-διθύραμβος (“celebrated in dithyrambs”) by Pratinas,21 and addressed as διθύραμβος by Euripides in his Bacchanals, vs. 526. In an ode in honor of the victories which were won by Xenophon of Corinth in 464 B.C. Pindar inquires, “Whence appeared the charms of Dionysus in connection with the ox-driving dithyramb?”22 Here, also, the author is not referring to the Corinthian dithyramb of his own day but to the period when it was put upon a quasi-literary level by Arion (see below). Finally, Epicharmus went so far as to declare that “when you drink water, it isn’t a dithyramb,”23 showing that the more primitive meaning of the term was not crowded out by later developments. These passages are sufficient to show that the dithyramb was at all times intimately associated with Dionysus and at the beginning belonged to him exclusively; their force is not invalidated by the acknowledged fact that at an early period (see p. 11, below) the restriction was broken down.
It was not until after the middle of the seventh century that the dithyramb became “poetized.” This step was taken by Arion of Methymna in Lesbos, then resident in Corinth. His connection with the dithyramb and early tragedy is vouched for by irrefutable evidence. Solon of Athens (639-559 B.C.) is said in a recently discovered notice24 to have declared in his Elegies that “Arion introduced the first drama of tragedy.” The question immediately arises as to exactly what language Solon had employed. The words τῆς τραγῳδίας πρῶτον δρᾶμα are, of course, only a paraphrase, for no form of the word τραγῳδία can be used in elegiac verse. This objection does not lie against the word δρᾶμα, however, and it will be remembered that the Dorians based their claims to tragedy partly upon this non-Attic term.25. Thus, we obtain an explanation of the cumbersome circumlocution “the first drama of tragedy.” In Solon’s Elegies the author of this notice (or his source) found only the ambiguous term δρᾶμα. A desire to retain the terminology of the original prevented his frankly substituting τραγῳδία. Accordingly, he kept δρᾶμα but inserted the qualifying genitive τῆς τραγῳδίας. I do not understand that Aristotle either indorses or rejects the Dorian pretensions with respect to this word; but in view of our present evidence I am of the opinion that Arion called his performances “dramas” and was the first to use the word in this sense and that there is so much of justice in the Dorian claims. It is not necessary to believe, however, that they were ever called satyric dramas, see p. 22, below.
Now, Dr. Nilsson has objected that Solon would have had no occasion to express his opinion upon a matter of this kind (op. cit., p. 611, note). But the mention of the title of the work from which the citation purports to come goes far to substantiate its genuineness. Furthermore, Solon was incensed at Thespis (see pp. 17 f., below), and therefore it was only natural that he should take an interest in the matter, assign the distinction to another, and state his opinion in as public a manner as possible. The fact that he lived in the days before real (Aeschylean) tragedy and before the importance of Thespis’ innovations was understood explains the error in his judgment. But at the very least, this notice proves that the tradition of Arion’s connection with tragedy was current as early as the first half of the sixth century.
Pindar’s reference to the development of the dithyramb at Corinth has already been mentioned. In the next generation Herodotus characterized Arion as follows: “Arion was second to none of the harpists of that time and was the first of the men known to us to compose (ποιήσαντα) a dithyramb and to give it a name (ὀνομάσαντα) and to represent it at Corinth” (I, 23). It is customary nowadays to seek to explain such notices as arising from the rival claims of jealous cities; but be it noted that here are two Attic sympathizers, Solon and Herodotus, granting full recognition to the literary achievements of a neighboring city. In fact, Herodotus is apparently too generous, for Arion could not have been the inventor of the dithyramb, broadly speaking. But ποιεῖν denotes not only “to compose” but also “to poetize,” and the latter translation is in better accord with what else we know of Arion’s contribution to the history of the dithyramb. On the other hand, ὀνομάσαντα probably means that in Herodotus’ opinion Arion was the first to give names (titles) to his performances.26
A Byzantine writer repeats and amplifies Herodotus’ statements but adds one interesting clause to the effect that Arion “introduced satyrs speaking in meter.”27 In this there is nothing surprising. In the Peloponnesus caprine satyrs were regular attendants upon Dionysus, and in consequence the dithyrambic choreutae must usually have been thought of as satyrs. Their improvisations, also, must always have engaged the speaking as well as the singing voice. This fact, however, did not at this time involve histrionic impersonation (μίμησις) for the reason that they would not attempt to say what was appropriate to satyrs but to themselves in propria persona as revelers and worshipers. The word ἔμμετρα (“in meter”), therefore, is the important one. The use of meter marked the coming of artistic finish and the passing of a performance largely extemporaneous. Some idea of the technique of Arion’s productions may be drawn from a dithyramb by Bacchylides (first half of the fifth century) in honor of Theseus. This is in the form of a lyric dialogue and was doubtless influenced somewhat by contemporaneous tragedy. The chorus of Athenians, addressing Aegeus, king of Athens, inquires why a call to arms has been sounded (vss. 1-15), and the coryphaeus (“chorus-leader”) replies that a herald has just arrived and summarizes his message (vss. 16-30). The chorus asks for further details (vss. 31-45), and once more the king’s reply is borrowed from the herald (vss. 46-60). Here Theseus, not Dionysus, is the theme of the poem; the choreutae do not represent satyrs, but appear in their true character as plain citizens of Athens; and the coryphaeus is given a dramatic character, that of Aegeus. These are all developments later than the time of Arion; nevertheless, the general effect must have been much the same.
Before the close of the sixth century the dithyramb had become a regular form of literature—a chorus of fifty, dancing and singing formal compositions. In 508 B.C. a contest of dithyrambic choruses of men was made a standing feature of the program at the City Dionysia in Athens. Simonides (556-467 B.C.) is known to have composed a dithyramb entitled Memnon, the exclusively Dionysiac character of the genre being then, if not earlier, abandoned. But it is important to remember that originally the dithyramb was extemporaneous and confined to the worship and exaltation of Dionysus.
In the new notice concerning Solon and Arion, von Wilamowitz finds “die Bestätigung dass die τραγῳδοί vor Thespis bestanden” (cf. op. cit., p. 470). This development could scarcely have taken place at Corinth in Arion’s time, for there was no need of coining a new word to designate the performers so long as they appeared as satyrs. And if a term had then been derived from the choreutae to designate their performance, it must have been *σατυρῳδία and not τραγῳδία. Neither could the new term have been derived at this period from the prize, for then the goat was only the third award.28 Let us therefore turn to Sicyon.
In a well-known passage (v. 67) Herodotus tells how the Sicyonians used to honor their former king, Adrastus, in other ways, and in particular celebrated his sorrows with “tragic” (or “goat”) choruses (τραγικοῖσι χοροῖσι) and how their tyrant Clisthenes in anger at Adrastus assigned these choruses to Dionysus and the other features of the rites to Melanippus. Melanippus in his lifetime had killed Adrastus’ brother and son-in-law, and Clisthenes had brought his bones from Thebes and transferred to him part of the honors which had previously been paid to Adrastus, in order to insult the latter as outrageously as possible. The superimposition of the worship of Dionysus upon that of the local hero and the reference to tragic choruses have furnished Ridgeway a foundation upon which to rear his theory that tragedy developed from ceremonies at the tombs of heroes. In this passage the meaning of the word τραγικοῖσι has provoked much discussion. I believe that Herodotus meant τραγικός here in the sense current in his own day, viz., tragic, but I do not believe that he stopped to consider whether these Sicyonian dances “were sufficiently like the choruses in the tragedies of his contemporaries to be called ‘tragic.’”29 I think he employed that adjective simply because τραγικοὶ χοροί was the Sicyonians’ own designation for their performances. If so, whatever τραγικοῖσι χοροῖσι connoted to Herodotus, or even to contemporaneous Sicyonians, originally τραγικός in this phrase must have meant “goat,” and these choruses must originally have been, for whatever reason, “goat” choruses.
Some considered Epigenes of Sicyon the first tragic poet, Thespis being second (or as others thought, sixteenth) in the list.30 In connection with Epigenes another tradition must be mentioned. Several explanations are preserved of the proverb oὐδὲν πρὸς τὸν Διόνυσον (“nothing to do with Dionysus”). These are somewhat vague in details and need not be taken too seriously; but at least they are valuable as showing the general periods in which their authors thought that the proper situation for the rise of such a proverb had existed. According to one account, this expression was uttered “when Epigenes had composed a tragedy in honor of Dionysus.”31 In just what particular Epigenes’ performance seemed alien to the worship of Dionysus the retailers of the anecdote do not specify. Ridgeway supposes that Epigenes “did not confine himself to Dionysiac subjects.”32 But surely that development came much later. In my opinion, the explanation is simpler. We have no information as to the costume which the choreutae wore in honoring the sorrows of Adrastus. There was, of course, no reason for their appearing as satyrs. But were satyric choreutae introduced at the same time that the dances were given over to Dionysus? If we answer this question in the negative, the situation becomes clear. The audience, or part of it, was sufficiently acquainted with the performances instituted by Arion at Corinth to expect a chorus of satyrs in the Sicyonian dances after they were transferred to Dionysus. And when Epigenes brought on his choreutae in the same (non-satyric) costume as had previously been employed, they naturally manifested their surprise with the ejaculation: οὐδὲν πρὸς τὸν Διόνυσον. By this they meant: “Why, these choreutae are just what we have had all the time; there is nothing of the satyrs about them. They have nothing to do with Dionysus.”
Practically everyone is convinced that τραγῳδία means “goat-song.” The only difficulty consists in explaining how this name came to be applied. We have already noted (see p. 2, above) that Welcker explained it on the basis of costume, and this is now the prevailing view. But though the choreutae at Corinth were satyrs, there were good reasons why no new term should be coined there to designate them (see p. 11, above), and in fact, τραγῳδία, τραγῳδός (“goat-singer”), and τραγικός (in a technical sense) apparently did not originate there. On the other hand, in Sicyon (where at least the expression τραγικοὶ χοροί, if not the others, seems to have been in use at an early day) the costume of the choreutae was assuredly not caprine before the dances were transferred from Adrastus to Dionysus and probably was not thereafter. Consequently, Welcker’s explanation must be rejected.
But the earliest and favorite explanation of these terms in antiquity derived them from the fact that a goat was given to the victorious poet as a prize.33 Knowledge and approval of this interpretation can be traced almost uninterruptedly from the high authority of the Parian Chronicle34 in the third century B.C. onward, and there is no cogent reason for doubting its truth. The other suggestion that the name was derived from the goat which was offered in sacrifice in connection with the performances will be seen not to conflict with this view when it is remembered that in the later dithyrambic contests the prize (a tripod) was not regarded as the personal possession of the victor but was customarily consecrated in some temple or other public place. In my opinion, these explanations have been most unwarrantably abandoned in modern times, and I think a reaction in their favor has set in. They are spoken of respectfully by Dr. Reisch,35 and Mr. Pickard-Cambridge mentions them exclusively.36
Now the transfer of the Sicyonian dances from Adrastus to Dionysus would probably happen early in the reign of Clisthenes (ca. 595-560 B.C.), and for this very period Eusebius preserves a notice to the effect that “a goat was given to contestants among the Greeks, and from this fact they were called τραγικοί.”37 I therefore believe that Herodotus, Eusebius (Jerome), and Suidas all refer to the same event: that Clisthenes of Sicyon established the goat prize about 590 B.C. when he surrendered to Dionysus the dances which had previously been performed in honor of Adrastus,38 that Epigenes was the poet whom Clisthenes employed to initiate this innovation, and that non-satyric choreutae and the terms τραγικός, τραγῳδός, etc., arose in this manner, time, and place. The neatness with which these notices fit together to produce this result renders them comparatively secure from the critical assault which might more successfully be directed against them individually. In any case, it is incumbent upon any skeptic, not merely to reject the later authorities, but also to provide a more satisfactory explanation of Herodotus.
If this series of conclusions is accepted, we have an answer to the question under consideration—the occasion of the term τραγῳδοί. We must conclude that honoring Adrastus with choruses either did not involve the giving of a prize or that the prize was other than a goat. With the transfer to Dionysus, a goat (for some reason) was chosen as the object of competition, and was doubtless immediately consumed in a sacrificial feast. We have seen that at Corinth, where the choreutae were satyrs, there was no reason to coin a new term to designate them. But at Sicyon the situation was different. What more natural than that from the new prize should be derived new names (τραγικοὶ χοροί and τραγῳδοί respectively) for the new-old performances and their choreutae.39 It is not enough to pass this tradition of Sicyonian tragedy by in silence or to brand it as aetiological or as arising from the partisanship of rival cities. It must first be shown to be inconsistent, either with itself or with other established facts.
Hitherto we have dealt with the Peloponnesus, which was inhabited by the Dorian branch of the Greek stock; at this point we pass to Attica, which was Ionic. We are indebted to the late Professor Furtwängler (op. cit., pp. 22 ff.) for having pointed out that among the Dorians the attendant sprites of Dionysus were caprine satyrs, but that among the Ionians he was attended by sileni, creatures with equine ears, hoofs, and tails. Caprine satyrs do not appear upon Attic vases until about 450 B.C. (see p. 24, below). Although the sort of dances from which tragedy developed had existed in Attica from time immemorial,40 yet they did not emerge into prominence and literary importance until the age of Thespis and in Icaria. Evidently Thespis’ innovations were partly borrowed from the Peloponnesus and partly his own. Included among the former would be the dropping of improvisation, the use of meter, the goat prize, and such terms as δρᾶμα and τραγῳδός. Most distinctive among the latter was his invention of the first actor. In early choral performances it was customary for the poet himself to serve as coryphaeus, and in Bacchylides’ dithyramb we have seen how the coryphaeus was set apart from the other choreutae, answering the questions which they propounded. It was inevitable that to someone should come the happy thought of developing this rôle still further and of promoting the coryphaeus to a position independent of the chorus. It is significant that the verb which was first used to designate the actor’s function was ἀποκρίνεσθαι (“to answer”), and that until the time of Sophocles all playwrights were actors in their own productions. We are now in a position to realize the true inwardness of Aristotle’s phrase: he does not say merely that tragedy was derived from the dithyramb but from the “leaders” of the dithyramb.
We have noted that the early dithyramb did not require impersonation (see p. 10, above). Even at an advanced stage it was probably much like a sacred oratorio of modern times in which the performers may sing words which are appropriate to characters and yet make no attempt by costume, gestures, or actions to represent those characters. Thespis changed all this. Since he assumed an actor’s rôle himself, first of all probably that of Dionysus, the choreutae could no longer conduct themselves as worshipers in disguise, but must now not merely look like real attendants of Dionysus but also behave as such. This is a fundamental matter. Only after this step had been taken could real drama in the modern sense become possible. Neither honoring the sorrows of Adrastus nor the “fore-doing” of imitative magic, not even the primitive δρώμενα at Eleusis or elsewhere demanded or presupposed actual impersonation. This development took place at Icaria and by the agency of Thespis. I cannot do better than to quote certain sentences of Miss Harrison’s:
We are apt to forget that from the epos, the narrative, to the drama, the enactment, is a momentous step, one, so far as we know, not taken in Greece till after centuries of epic achievement, and then taken suddenly, almost in the dark, and irrevocably. All we really know of this momentous step is that it was taken sometime in the sixth century B.C. and taken in connection with the worship of Dionysus. Surely it is at least possible that the real impulse to the drama lay not wholly in “goat-songs” and “circular dancing places” but also in the cardinal, the essentially dramatic, conviction of the religion of Dionysus, that the worshipper can not only worship, but can become, can be, his god. Athene and Zeus and Poseidon have no drama, because no one, in his wildest moments, believed he could become and be Athene or Zeus or Poseidon. It is indeed only in the orgiastic religions that these splendid moments of conviction could come, and, for Greece at least, only in an orgiastic religion did the drama take its rise.41
Thespis’ invention of impersonation probably provides the clue for understanding the clash between him and Solon:
Thespis was already beginning to develop tragedy, and on account of its novelty the matter was engaging general attention but had not yet been brought into a public contest. Now Solon, who by nature was fond of hearing and learning, to a still greater extent in old age gave himself up to leisurely amusement and even to conviviality and music. Therefore, he went to see Thespis himself act, as was customary for the earlier poets. And when the spectacle was over, Solon addressed him and inquired if he had no sense of shame to lie so egregiously before so many. Moreover, when Thespis said that it was no crime to say and enact such things in sport, Solon struck the ground violently with his staff and said: “Yet if we praise and honor this ’sport’ under these circumstances, it will not be long before we discover it in our contracts.”42
To so straightforward a man as Solon such a facile abandonment of one’s own personality might well seem like barefaced lying, and to augur and even encourage similar shuffling prevarications in the more serious affairs of life.
To Ridgeway, however, all this appears in a different light. In the first place, after citing Diogenes Laertius to the effect that “in ancient times the chorus at first carried on the action in tragedy alone, but later Thespis invented an actor in order to allow the chorus intervals of relief,”43 he declares flatly: “But this cannot mean, as is commonly held, that Thespis first separated in some degree the coryphaeus from the chorus and made him interrupt the dithyramb with epic recitations, for, as we have seen above, before his time the poet or coryphaeus used to mount a table and hold a dialogue with the chorus.”44 In the cross-reference Ridgeway had quoted Pollux iv. 123: “The ἐλεός was a table in the olden days upon which in the period before Thespis some one mounted and made answer to the choreutae,” and Etymologicum Magnum, s.v. “θυμέλη”: “It was a table upon which they stood and sang in the country when tragedy had not yet assumed definite form.” These late notices are manifestly vague and inexact references to rudimentary histrionicism among the choreutae themselves or between them and their coryphaeus. The first of them is probably due to a false inference from a scene in some comedy.45 It is true that the invention of the first actor is expressly attributed to Thespis only by Diogenes, yet it may be inferred in several other connections. Evidently the matter is largely one of definition. Ridgeway himself concedes all that is important, when he continues: “There seems no reason to doubt that Thespis in some way defined more exactly the position of the actor, especially by the introduction of a simple form of mask.”
In the second place, Ridgeway considers that Thespis made the “grand step” in the evolution of tragedy when he
detached his chorus and dithyramb from some particular shrine, probably at Icaria, his native place, and taking his company with him on wagons gave his performances on his extemporised stage when and where he could find an audience, not for religious purposes but for a pastime. Thus not merely by defining more accurately the rôle of the actor but also by lifting tragedy from being a mere piece of religious ritual tied to a particular spot into a great form of literature, he was the true founder of the tragic art. This view offers a reasonable explanation of Solon’s anger on first seeing Thespis act. A performance which he would have regarded as fit and proper when enacted in some shrine of the gods or at a hero’s tomb, not unnaturally roused his indignation when the exhibition was merely “for sport,” as Thespis himself said (and doubtless also for profit), and not at some hallowed spot, but in any profane place where an audience might conveniently be collected [op. cit., p. 61].
Not only does such an interpretation find no support in Plutarch’s anecdote but it is highly improbable as well. It may be granted that after long neglect Thespis’ “wagon”46 seems to be enjoying a recrudescence of favor. Dieterich and von Wilamowitz have referred to it in all seriousness.47 There is nothing improbable about the tradition nor any compelling reason for supposing it borrowed from the history of early comedy. It is natural to suppose that Thespis did not restrict his activities to Icaria, but extended them to such other demes as were interested or found them appropriate to their festivals. In that case, means of transportation for performers and accessories became imperative. The use of such a vehicle in the Prometheus Bound of Aeschylus shows that it need not necessarily have served also as a stage, as has sometimes been thought. Now, as a matter of fact, several Attic vases, dating from the close of the sixth century B.C., represent the “wagon-ship” of Dionysus (Fig. 65). Just what relationship subsisted between primitive drama and the scenes depicted upon these vases has yet to be definitely established. Dr. Frickenhaus would associate them with the preliminary procession at the City Dionysia (see p. 121, below). But at least, until such time as any connection with Thespis’ wagon has been shown to be impossible, the suggestion can scarcely be laughed out of court as utterly ridiculous. On the other hand, to suppose that Thespis entirely dissociated his performances from shrines and festivals not only rests upon no evidence but is so out of harmony with other data as to be incredible.
Whether the innovation of treating non-Dionysiac themes in tragedy must also be credited to Thespis before he brought his career to a close must remain a matter of doubt, though personally I am inclined to suppose so. Suidas48 reports Phorbas or the Prizes of Pelias, Priests, Youths, and Pentheus as the titles of four of his plays. Of these the last is clearly Dionysiac, the first probably is not, and the other two are noncommittal. This evidence, however, cannot be relied upon, for the reason that Aristoxenus is said to have declared that Heraclides Ponticus wrote tragedies and attributed them to Thespis.49
But as we are not told that these plays bore the same titles as those ascribed to Thespis by Suidas, it does not by any means follow that the latter are spurious. But even if the titles were the same, it is not unlikely that Heraclides would have chosen as titles for his spurious compositions names declared by tradition to be those of genuine works of the Father of Attic Tragedy. The titles as they have reached us indicate that the ancients most certainly did not believe that Thespis confined himself to Dionysiac subjects.50
In any case, this development could not have been long deferred after 534 B.C. To the more conservative it is said to have given offense; according to some authorities, the expression “Nothing to do with Dionysus” took its rise at this juncture.51 Simultaneously, or at least only a little subsequently, the tragic choreutae were no longer dressed to represent sileni but whatever the needs of the individual play demanded, often plain citizens of Athens, Corinth, Thebes, etc.
Even after all that Thespis did for it tragedy must still have been a crude, coarse, only semi-literary affair. Nevertheless, in 534 B.C., when Pisistratus, tyrant of Athens, established a new festival called the City Dionysia, in honor of Dionysus Eleuthereus,52 he made a contest in tragedy the chief feature of its program. As was but fitting, Thespis won the first goat prize ever awarded in this Athenian festival.53 It is unnecessary to enlarge upon this recognition except to protest against a not uncommon tendency to assume that terms like τραγῳδία and τραγῳδός were not in use before this date. Of course, the matter can not be definitely proved, but the evolution which I have been tracing at Sicyon and Icaria distinctly favors the other view.
We have seen that Aristotle’s statements ought not to be ignored or lightly rejected. On the other hand, it is no less important to read nothing into his language which does not belong there. Thus, when he declares: “Discarding short stories and a ludicrous diction, through its passing out of its satyric stage, tragedy assumed, though only at a late point in its progress, a tone of dignity,”54 the phrase διὰ τὸ ἐκ σατυρικοῦ μεταβαλεῖν ὀψὲ ἀπεσεμνύνθη has generally been taken to mean that tragedy developed out of a form like the satyric dramas known to us, in the next century, from Sophocles’ Trackers and Euripides’ Cyclops. For such a historical development no other testimony can be cited until Byzantine times (see p. 29 and n. 2, below). Now this interpretation of Aristotle’s phrase has always involved certain difficulties and has been pronounced inconsistent with his other statement that tragedy developed “from the leaders of the dithyramb.” But in my opinion we must accept Reisch’s interpretation: “We are certainly not warranted in translating ἐκ σατυρικοῦ baldly as ‘from the satyr-play.’ On the contrary, Aristotle is speaking only of the ‘satyr-play-like origin’ and of the ’satyr-like poetry’ (as Theodor Gomperz suitably renders it in his translation); and from this, first of all, only a family relationship between primitive tragedy and the satyr-play, not an identity, may be inferred.”55 The same thought recurs in Aristotle’s next sentence, when he says: “The iambic measure then replaced the trochaic tetrameter, which was originally employed when the poetry was of the satyric order, and had greater affinities with dancing.”56 In other words, though early Attic tragedy never received the name of “satyric drama,” and though its choreutae were probably sileni and not satyrs, nevertheless, since the Thespian and pre-Thespian performances, by reason of their obscenities, grotesque language, ludicrous and undignified tone, the predominance of choral odes, etc., bore a certain resemblance to the contemporaneous exhibitions of satyrs in the Peloponnesus and to Pratinas’ satyric drama in Athens at a later period, it can truthfully be said that tragedy had passed through a “satyric stage” and had had a “satyric” tinge which it was slow to lose.
What, then, was the origin of the performance which in the fifth century constituted the final member of tetralogies? Such tetralogies cannot be made out for any playwright before Aeschylus; and the number of plays attributed to Pratinas, eighteen tragedies and thirty-two satyric dramas, throws additional doubt upon the probability that the early poets were required to present four plays together.57 We have thus far considered three types of performances: the improvisational dithyramb, which was still continued in rural and primitive districts; the improved dithyramb (in 508 B.C. dithyrambic choruses of men were added to the program of the City Dionysia at Athens), and tragedy. The last two had by this time become semi-literary types. Now we are expressly told, and there is no reason to discredit the information, that Pratinas of Phlius in the Peloponnesus was “the first to write satyr-plays.”58 The general situation is clear. After tragedy had lost its exclusively Bacchic themes and had considerably departed from its original character, Pratinas endeavored to satisfy religious conservatism by introducing a new manner of production, which came to be called satyric drama. This was a combination of the dramatic dithyramb of his native Phlius, which of course had developed somewhat since the days of Arion and Epigenes, and of contemporary Attic tragedy; and it had the merit of continuing, at least for a while, the Dionysiac subjects which were so appropriate to the god’s festival. It appears that at first satyr-plays were brought out independently of tragedy and in greater numbers, comparatively, than was afterward the case. But about 501 B.C. the City Dionysia was reorganized: the goat prize was abandoned; κῶμοι, i.e., the volunteer performances from which comedy was later to develop, were added to the program; and, in particular, the regulation was established that each tragic poet must present three tragedies and one satyr-play in a series. Pratinas is known to have competed against Aeschylus about 499 B.C. His innovation doubtless fell somewhere between the institution of the tragic contest in 534 B.C. and the reorganization of the festival program in 501 B.C., possibly about 515 B.C.
There remains the difficult problem as to the appearance of the choreutae in the satyric drama at different periods in Athens. Fortunately the aspect of non-dramatic sileni and satyrs is fairly certain. Already on the François vase, an amphora signed by Clitias and Ergotimus and belonging to about 600-550 B.C., there are representations of three ithyphallic creatures with equine ears, hoofs, and tails (Fig. 3).59 An inscription ΣΙΛΕΝΟΙ leaves no doubt as to the identity of the figures. Mr. A. B. Cook lists six other inscribed vases from Attica which tell a similar story.60 None of these seven vases, however, betrays any relationship to the theater.
On the other hand, a list61 of fifteen Attic vases has been drawn up on which goat-men appear. None of these antedates 450 B.C., so that it is clear that such figures did not go back to a remote period in Athenian history. In fact, they can hardly be conceived of as preceding Pratinas’ introduction of the satyric drama toward the close of the sixth century. Unfortunately none of these vases is inscribed, but the caprine ears, hoofs, horns, and tails scarcely leave room for doubt that these creatures, like similar figures of Hellenistic and Roman times, were known as satyrs. With one possible exception (Fig. 9), which will be discussed presently, these representations also have no direct relationship to the theater. It would thus appear that from first to last a clear distinction was drawn, outside the sphere of theatrical influence, between the equine sileni and the caprine satyrs.
Of the vases which may certainly be regarded as representing scenes from satyric drama the best known and most pretentious is a crater in Naples (Fig. 4).62 This and a crater at Deepdene were painted about 400 B.C. Somewhat earlier are another crater at Deepdene, a dinos at Athens (Figs. 5 and 6), and fragments of two dinoi at Bonn (Fig. 7).63 The last three are derived from the same original. On the Naples crater preparations for a satyr-play are being made in the presence of Dionysus and Ariadne, who are seen in an affectionate embrace in the center of the top row. The names of the figures are made known by inscriptions in most cases but are not always significant. Just beyond Ariadne, Love (Ἵμερος) hovers above an uninscribed actor in women’s costume, whose mask is provided with a Scythian cap. The next figure is Heracles (inscribed) and the next is thought to be Silenus. Beyond Dionysus is an uninscribed actor in royal costume. Except Love, all these figures carry masks and constitute the histrionic personages in the drama. It has been claimed with great plausibility that the play dealt with Heracles’ exploits at Troy.64 In that case the king is Laomedon and the maiden is Hesione, his daughter, who was rescued from the sea monster by Heracles. To the right of the dancing choreutes in the lower row is the flute-player (Pronomus), who will furnish the accompaniment for the lyrical portions of the play; to the left is Demetrius with a roll in his hand, probably the poet. The remaining twelve figures are probably choreutae and bear more directly upon our present investigation. Most of them carry masks, and they have human feet and no horns. They resemble sileni in having long equine tails. The sole resemblance to satyrs is found in the fact that nine of them wear a shaggy covering about the loins, supposedly a goatskin. The waistband upon the choreutes in the extreme upper left-hand corner, however, resembles cloth trunks more than a skin. Yet this divergence is probably to be explained as due to carelessness or a whim on the part of the draftsman instead of to an essential difference in material. This appears plainly from a study of the other vases in this series, on which the loin-bands resemble the trunks of the last-mentioned choreutes on the Naples crater rather than the skins of his nine companions. None the less, a multitude of short dashes on the waistbands in one of the Bonn dinoi (Fig. 7) is plainly intended to characterize them as skins, and the bands on the Deepdene craters are “patterned in such a way as to suggest a fringed or shaggy edge.” An illuminating side light upon the freedom which the painter exercised is afforded by a comparison of the left-hand choreutae in Figs. 6 and 7. These are identical figures in different copies of the same original; yet the shagginess of the loin-band is clearly indicated in the one and entirely omitted in the other. Moreover, the choreutes on the other dinos at Bonn seems to wear no waistband at all!65 In conclusion, it will be observed that, except for variations in the representation of the conventionalized goatskin, the choreutae upon all these vases are exactly alike:66 they all have human feet, no horns, and equine tails. It is evident that by 400 B.C. or a little earlier this type had become standardized for theatrical purposes. That it suffered no material modification thereafter appears from a Pompeian mosaic (Fig. 8).67
It is plain that this was the type of satyr which the unknown source of the notice in Etymologicum Magnum had in mind when attempting to explain the etymology of τραγῳδία: “... or because the choruses generally consisted of satyrs whom they called ‘goats’ in jest either on account of the shagginess of their bodies or on account of their lasciviousness, for the animal is of such a sort; or because the choreutae plaited their hair, imitating the form of goats.”68 This passage has been used to support the canonical doctrine that tragedy was the child of satyric drama (see pp. 2 and 22 f., above), but is far from adequate for that purpose. The words after δασύτητα (“shagginess”) are often ignored or even omitted. But it is necessary to interpret the final phrase, “imitating the form of goats,” in terms of the details stated in the context. So far as we are now concerned, the only point of resemblance mentioned is their “shagginess.” This and Horace’s expression about the tragic poet “stripping his satyrs” for the satyr-play69 would be entirely suitable in describing the choreutae on the Naples crater. Furthermore, it will be noted that this explanation occurs only in a late Byzantine notice and that no earlier source is mentioned. The only way in which a respectable antiquity can be claimed, by means of literary evidence, for this interpretation consists in maintaining that it is implicit in Aristotle’s phrase ἐκ σατυρικοῦ μετέβαλεν. But we have already seen (see p. 22, above), that this expression need not, and probably does not, support this view. The only other passage which can be cited in this connection occurs in three other Byzantine writers.70 The conclusion is irresistible that both the goat-men explanation of the word τραγῳδία and the supposed development of tragedy from satyric drama are due to “reconstructions” of literary history at an extremely late period.
Evidently this standard type of theatrical satyr took its genesis from an amalgamation of the caprine satyrs and the equine sileni. It is significant that in Euripides’ Cyclops and Sophocles’ Trackers Silenus is one of the characters and is the father of the chorus. These satyr-plays were brought out in the vicinity of 440 B.C.71 The question now arises: Was this conventional type the invention of Pratinas or did it develop later? It will be remembered that in the list of fifteen fifth-century vases from Attica on which representations of goat-men occur (see p. 25, above), one was mentioned as having a possible connection with the theater. The single exception is a crater in the British Museum of about 450 B.C. (Fig. 9).72 The larger design on the same side of the vase represents the decking of Pandora, and it is commonly thought that the two scenes belong together and are derived from a satyr-play dealing with Pandora. However that may be, the presence of a flute-player would seem to indicate that at least Fig. 9 is theatrical. If so, the choreutae are not of the type which we have been studying, but true satyrs with caprine hoofs, horns, and tails.73 About their loins they wear trunks, which in three cases are painted black (to represent a goatskin?) but in one case are left unpainted. Now from Aeschylus’ satyric drama entitled Prometheus the Fire-Kindler is preserved a line “O goat, you will mourn (lose) your beard,” which was addressed by Prometheus to a satyr who wished to kiss a flame and which has been used as proof that the choreutae were caprine in appearance.74 Again, in Sophocles’ Trackers occur the words: “For though you are young with a flourishing beard, you revel as a goat in the thistles.”75 Finally, in Euripides’ Cyclops the chorus speak of wandering about “with this poor goatskin cloak.”76 Although these passages do not constitute proof that the dramatic satyrs were of caprine appearance, they gain considerably in point if we may suppose that they were, and to that extent they confirm the evidence of the British Museum crater.
Such, then, is the penultimate stage in the evolution of the satyric chorus, and many authorities are content to stop here. But there remains evidence for a still earlier stage. A British Museum psykter by Duris (Fig. 10)77 represents ten “choreutae” and a herald, and a British Museum cylix by Brygus contains two scenes, in one of which three “choreutae” are attacking Iris before Dionysus and his altar and in the other Hermes and Heracles are protecting Hera from four “choreutae.”78 These vases belong to about 480 B.C., and the “choreutae” upon them have human feet, no horns, no loin-bands, and equine ears and tails. Reisch is undoubtedly correct in recognizing in these scenes at least the indirect influence of the satyr-play.79 Furthermore, a similar figure appears upon a Würzburg cylix of about 500 B.C. (Fig. 11).80 This bears the inscription ΣΑΤΡΥΒΣ, a manifest mistake for σάτυρος. Here we have the earliest representation of a satyr in Attica. And though it does not belong to a theatrical scene, its divergence from contemporaneous satyrs of the Peloponnesus and from Attic satyrs of a later period can be explained only on the basis of the appearance of the choreutae in contemporaneous satyr-plays. The Duris psykter and the Brygus cylix show that this type did not at once disappear.
To my mind the meaning of all this is fairly clear. When Pratinas attempted to restore the Dionysiac element to contemporaneous drama at Athens, he kept the Peloponnesian name but did not venture to shock conservatives still further by disclosing to their eyes creatures so foreign and strange as the Dorian goat-men would have been. Accordingly, he transformed his satyrs so as to approximate the sileni of native tragedy.81 After fifty or sixty years, however, satyric drama had become so thoroughly at home in Athens that the experiment was tried of imposing the Peloponnesian type unchanged upon the Attic choruses. But the reaction could not and did not endure. In two or three decades the final type had emerged, such as we see it in the Naples crater. Except for the goatskin about the loins, which is often highly conventionalized, the native sileni are at every point victorious.
The Greeks were inordinately fond of associating every invention or new literary genre with some one’s name as discoverer (εὑρετής). In the case of tragedy the problem was unusually complicated. In later years Arion, Epigenes, and Thespis all had their partisans. The last named is the one most frequently mentioned, and strictly speaking this view is correct. But more broadly considered, the question largely depends upon the stage of development to which one is willing to apply the word “tragedy.” To many moderns, with almost two and a half millenniums of dramatic history as a background, Aeschylus will seem the first tragic playwright. At least, in his hands tragedy became for the first time real literature.
The foregoing treatment will show that I do not believe a study of the origin of religion to be indispensable for a discussion of the origin of Greek tragedy. Prior to Arion and Epigenes there was nothing which the most fanciful could recognize as akin to modern tragedy. After the work of Thespis and Aeschylus no one can fail to note its presence. To trace, so far as we may, the gradual unfolding of the new genre from a state of nonexistence to a period of vigorous growth seems to me a concrete problem and distinctly worth while. The songs and dances from which tragedy and the satyr-play developed were associated, at the period when they became truly dramatic, with the worship of Dionysus, and at that same period Dionysus was as truly a “god” (as distinct from a “hero”) as any that the Greeks ever knew. To abandon these plain facts and others like them in favor of vague theorizing on religious origins will never bring us satisfactory results. Now, in his Origin of Tragedy Ridgeway, who may serve as a protagonist of this method, recognized only the satyr-play as Dionysiac in origin, and attempted to dissociate tragedy and the dithyramb from that deity and to derive them from ceremonies at the tombs of heroes, i.e., from ancestor worship. I cannot conceive that many classical scholars will believe him to have succeeded in this attempt. Ridgeway evidently foresaw this and tried to forestall it by saying that “as Dionysus himself had almost certainly once been only a Thracian hero, even if it were true that Tragedy had risen from his cult, its real ultimate origin would still be in the worship of the dead” (op. cit., p. 93). What, then, was the point in his conceding that satyric drama was Dionysiac in origin? In that case the ultimate origins of tragedy and satyric drama must, after all, have been identical, and the differences in their origins must have consisted only of the minor divergencies in the final stage of their development. In practice, how does this result differ from the more usual procedure, which ignores the ultimate sources and concentrates attention upon the last stage of development? So far as I can see, it would differ only to the extent that the underlying religion of both genres would now be understood to be ancestor worship. But this distinction loses all meaning, for the reason that in his last volume Ridgeway maintains that “Vegetation, Corn, and Tree spirits, as well as those of rocks, mountains, and rivers, and what are collectively termed Totemistic beliefs,” fertility-rites, initiation-rites, mana, “the worship of Demeter and almost82 all other Greek deities” are “not primary phenomena but merely secondary and dependent on the primary belief in the immortality and durability of the soul,” and consequently that tragedy and serious drama (being everywhere associated with some form of religion) not only in Greece but “wherever they are found under the sun have their roots in the world-wide belief in the continued existence of the soul after the death of the body.”83 How much of truth there may be in Ridgeway’s contention that ancestor worship is prior to and the ultimate source of other forms of religion I shall not stop to discuss. But the practical value of so universal a generalization has been well expressed by another: “Even if it can be shown that your far-off ancestor was an ape, it does not follow that your father was an ape.”84 In other words, in spite of any resemblance which may have obtained between the ultimate forms of Dionysiac worship and the true veneration of heroes, at the time when tragedy actually came into being the existing differences between them were of much greater significance than any alleged identity of origin in the far-distant past could have been. If it were possible for Ridgeway to substantiate his first position, viz., that tragedy arose directly from the worship of the hero Adrastus at Sicyon, or the like, there would be some meaning in his work. But his doctrine of ultimate derivation loses itself in primeval darkness.
The Origin of Comedy.85—The difficulty of this problem was recognized as early as Aristotle:
Now the successive changes in tragedy and the persons who were instrumental thereto have not passed into oblivion, but comedy did suffer oblivion for the reason that it was not at first taken seriously. And a proof of this is found in the fact that it was relatively late [viz., 486 B.C.] before the archon granted a chorus of comic performers; they used to be volunteers. And comedy already had certain forms when the aforementioned comic poets [i.e., Chionides and Magnes, the first comedians after official recognition was granted] appear in the records. Who furnished it with “characters” (πρόσωπα)86 or prologues or number of actors and the like remains unknown. Developing a regular plot was a Sicilian invention, but of the Athenians the first to abandon the “iambic” or lampooning form and to begin to fashion comprehensive themes and plots was Crates.87
But whatever uncertainties may obscure the various stages in the history of comedy, fortunately there is little doubt as to the source from which it came. Aristotle states that “comedy also sprang from improvisations, originating with the leaders of the phallic ceremonies,88 which still survive as institutions in many of our cities.”89 Mr. Cornford (op. cit., pp. 37 ff.) finds the best illustration of these ceremonies in the well-known passage in Aristophanes’ Acharnians, vss. 237 ff. Dicaeopolis has just concluded a private peace with Sparta and prepares to celebrate a festival of Dionysus on his country estate. He marshals his meager procession as if it contained a multitude, his daughter carries upon her head a sacred basket with the implements of sacrifice, two slaves hold aloft a pole which is surmounted by the phallic symbol, and Dicaeopolis himself brings up the rear with a large pot in his arms, while the wife and mother constitutes the watching throng. At vss. 246 ff. a sacrifice is offered to the accompaniment of an invocation to Dionysus. Finally Dicaeopolis re-forms his procession with various coarse remarks and starts up a phallic ballad of an obscene nature in honor of Phales, “mate of Dionysus and fellow-reveller” (ξύγκωμε). The proceedings thus consist of a procession to the place of sacrifice, the sacrifice itself, and the phallic song or comus (κῶμος). The last is important for our present purpose because comedy (κωμῳδία) etymologically means “comus-song” (κῶμος + ᾠδή). Κῶμος denotes both a revel and the band of masqueraders participating therein. The comus was the particular type of phallic ceremony from which comedy developed.
The comus in Aristophanes’ Acharnians is sung by Dicaeopolis alone for the reason that the lack of suitable helpers compelled him to act as both priest and congregation. But Cornford is right (op. cit., pp. 38 ff.) in recognizing this song as belonging to a widely spread type in which the improvisations of one or more leaders (ἐξάρχοντες) are interrupted at more or less regular intervals by a recurrent chantey on the part of the chorus. In this instance the song is not continued to a length natural to the type, but is cut short by the real chorus of the play which has been hiding but now bursts forth and stops proceedings with a shower of stones. From the standpoint of contents Cornford detects two elements in the comus: an invocation to the god to attend his worshipers in their rites, and an improvisational “iambic” element of obscene ribaldry, which often took the form of satire directed against individuals by name (ibid., p. 41). These two elements exactly correspond to the double object of all phallic ceremonies, which were both a “positive agent of fertilization” and a “negative charm against evil spirits.” The former result was obtained by the invocation of friendly powers; as to the latter,
the simplest of all methods of expelling such malign influences of any kind is to abuse them with the most violent language. No distinction is drawn between this and the custom of abusing, and even beating, the persons or things which are to be rid of them, as a carpet is beaten for no fault of its own, but to get the dust out of it.... There can be no doubt that the element of invective and personal satire which distinguishes the Old Comedy is directly descended from the magical abuse of the phallic procession, just as its obscenity is due to the sexual magic; and it is likely that this ritual justification was well known to an audience familiar with the phallic ceremony itself [ibid., pp. 49 f.].
It is possible to cite many examples of ritualistic scurrility among the Greeks, such as that indulged in by the Eleusinian procession as it approached “the bridge,” that of the riders upon the carts on the Day of Pots (χόες) at the Anthesteria, that at the Stenia festival, and many others. Sometimes these involved physical violence as well as mere abuse, and this element (or the threat of it) frequently recurs in Old Comedy. Perhaps the most interesting parallel is afforded by Herodotus v. 82 f. In the sixth century B.C., in order to avert a famine, the Epidaurians set up wooden statues of Damia and Auxesia, goddesses of fertility.90 Somewhat later, the Aeginetans stole these images and set them up in their own country; “they used to appease them with sacrifices and female satiric choruses, appointing ten men to furnish the choruses for each goddess; the choruses abused no man but only the women of the country; the Epidaurians also had the same rites.”
The comus frequently took the form of a company marching from house to house to the music of a flute-player and rendering a program of singing and dancing at every dwelling. From what has already been said it will be understood that the improvisations of the comus leaders would rarely redound to the credit of the householders. These scurrilous attacks upon their neighbors combined with other motives to induce the comus revelers to assume disguises, which varied from year to year. Now, according to the Parian Chronicle, comic choruses were the invention of Susarion and were first performed at Icaria. This doubtless means that Susarion transformed the ceremonies of an old ritual procession in the country into a “stationary” performance in an orchestra. The same authority informs us that this innovation was introduced into Athens between 580 and 560 B.C.91 This notice must refer to the Lenaean festival, since the program of the City Dionysia did not receive this addition until about 501 B.C. At both festivals the performances still continued for some time to be called comuses (κῶμοι), comedy being a name of later date, and were produced by “volunteers.” Five Attic vase paintings of about 500 B.C. depict comus revelers as cocks, birds, or as riding upon horses, dolphins, or ostriches (Figs. 12-16).92 The state did not assume official supervision of comedy until 486 B.C. at the City Dionysia and about 442 B.C. at the Lenaea.93
Before we can proceed further, it will be necessary to consider the nature of ancient comedy. In the time of Hadrian the history of literary comedy at Athens was divided into three periods, called Old, Middle, and New Comedy, respectively. Old Comedy came to a close shortly after the beginning of the fourth century B.C. Politics and scurrilous attacks upon contemporaneous personages made up the bulk of its subject-matter. Living men, such as Pericles, Socrates, Euripides, and Cleon were represented by actors on the stage and were lampooned with the utmost virulence. Sometimes their identity was thinly disguised under a transparent pseudonym, but oftentimes the very name of the victim was retained along with the other marks of identification. Middle Comedy was a transitional period of about half a century’s duration between Old and New. It renounced the political and personal themes of its forerunner and was largely given up to literary criticism, parodies, and mythological travesty. New Comedy, in turn, abandoned such subjects for the most part and devoted itself to motives drawn from everyday life. Except for the occasional presence of the chorus, it does not greatly differ in structure, theme, or technique from the comedy of manners today, mutatis mutandis.
For the study of origins, however, we must turn back to the earliest type, Old Comedy, which is entirely unlike any present-day genre. We are fortunate in possessing eleven complete plays of Aristophanes, the chief poet of Old Comedy; and though no two of them are exactly alike in the details of their structure, yet the general outline is clear. The leading features are as follows:94
1. The prologue (πρόλογος) spoken by the actors and serving both as an exposition and to set the action of the play in motion.
2. The parodus (πάρoδος), or entrance song of the chorus. Originally this division must have been exclusively choral, but by Aristophanes’ time it has been developed so as sometimes to include lines spoken by actors.
3. The agon (ἀγών, “contest”), a “dramatized debate” or verbal duel between two actors, each supported by a semi-chorus; see p. 43, below.
4. The parabasis (from παραβαίνω, to “come forward”), a “choral agon” in which the chorus, the actors being off stage, march forward to address the audience. When complete, the parabasis consists of seven parts which fall into two groups: the first group contains three single parts, which were probably rendered by the first coryphaeus. Dropping all dramatic illusion and all connection with the preceding events of the play, he sets forth the poet’s views concerning his own merits and claims upon the public, ridicules the rival playwrights, announces his opinions on civic questions, etc. The second group contains four parts in the form of an epirrhematic syzygy, i.e., a song (ᾠδή) and epirrheme (ἐπίῤῥημα, “speech”) by one semi-chorus and its leader, respectively, are counterbalanced by an antode (ἀντῳδή) and an antepirrheme (ἀντεπίῤῥημα) by the other semi-chorus and its leader; here the chorus usually sing in character once more, the knights praising their “horses,” the birds their manner of life as compared with men’s, etc.95
5. There follows a series of episodes (ἐπεισόδια), histrionic scenes separated (6) by brief choral odes (στάσιμα or χορικά). The episodes portray the consequences of the victory won in the agon (3). For example, in the Acharnians the subject of controversy is whether Dicaeopolis shall be punished for the alleged treason of having made a private peace with Sparta, and part (5) represents him, in a succession of burlesque scenes, as enjoying the fruits of that peace.
7. The exodus (ἔξοδος), or recessional of the chorus. Properly speaking, this should contain only the final, retiring song of the chorus (the ἐξόδιον), but the term came to include the histrionic passage just preceding it, also.
This is a very incomplete sketch of a highly complicated subject, but it will suffice for present purposes.
Now in the scurrility of the primitive (non-literary) comus Professor Navarre (op. cit., p. 248) would recognize three stages. In the first, the ribaldry of the comus received no answer from the crowd of spectators. This is doubtless to be explained by supposing that all who were competent to participate were already members of the comus; the spectators consisted only of women and children, who frequently had no more right of speech in religious ritual than in law. So Dicaeopolis’ wife is present but speechless in Aristophanes’ Acharnians (see p. 36, above). In the second stage, the bystanders retorted to the assaults of the comus revelers. This probably indicates that membership in the comus has been restricted in some way, leaving others free to retaliate in kind from the crowd. The third stage was reached when this new element was formally recognized and brought within the comus itself, which was thus divided into antagonistic halves for mutual recrimination. Thus may be explained a peculiar feature of Old Comedy. Its chorus was a double chorus of twenty-four members, always divided into two semi-choruses, which often were hostile during a large portion of the play. Sometimes this division between them was shown by their masks or costumes, as when the chorus represented men and women, horses and their riders, etc. But sometimes the division was one of sentiment—one semi-chorus, for example, favoring peace and the other being opposed to it. The result of this division of the early comus revelers into semi-choruses is a parallelism of structure in certain parts of comedy, ode being matched by antode, and the epirrheme of one chorus leader by the antepirrheme of the other. It is clear that all the divisions which show this duality of arrangement descend from the comus.96
One of these divisions is the parabasis (4). Though one of the most ancient features of Old Comedy, it was also one of the first to decay: complete in Aristophanes’ earlier plays, it is always mutilated in some way during his middle period and in his last two comedies has disappeared entirely. We have seen (p. 37, above) that the essential characteristics of the phallic ceremonies were the induction of the good influences by invocation and the aversion of the bad by vituperation. Now in the epirrhematic syzygy which constituted the second half of the parabasis, even as late as Aristophanes, when it naturally must have changed considerably in function, “the ode and antode normally contain an invocation, either of a muse or of gods, who are invited to be present at the dance, the divine personages being always selected with reference to the character of the chorus. The epirrheme and antepirrheme often contain the other element of satire or some milder form of advice and exhortation.”97
Another division of Old Comedy which was carefully balanced and which ought, therefore, to be a derivative of the comus is the agon (3). Normally this division was epirrhematic in structure and fell into nine parts, as follows: First comes the ode sung by one half-chorus, then the cataceleusmus (κατακελευσμός, “encouragement”) in which their leader exhorts one of the actor contestants, thirdly this actor delivers his speech (epirrheme), concluding with a peroration (πνῖγος, “choke,” so called because it was all to be delivered in one breath and left the performer speechless). Next came the antode, anticataceleusmus, antepirrheme, and antipnigus rendered by the other half-chorus, their leader, and the second actor, respectively. Finally, in the sphragis (σφραγίς, “seal”) is given the unanimous verdict of the whole chorus. At first glance it would seem that too important a rôle is here played by actors for the agon ever to have been derived from the comus, which was purely choral. The comus consisted of an undifferentiated band of revelers and its choreutae assumed no distinct parts. In fact, there is no reason to suppose that their performances involved dramatic impersonation (μίμησις) at all. They might be dressed to represent birds or animals, but with few or no exceptions they sang and spoke and conducted themselves as would be appropriate for men engaged in such a rite to do. As we have already seen (p. 38, above) their costumes were for disguise.
Nevertheless, the situation is not so impossible as it seems. The fact that the masks and costumes of the choreutae were all alike, or at most of two types to correspond to the two semi-choruses, did not prevent each member of the chorus from speaking, or singing, apart from the rest. This was sometimes done even in fully developed tragedy, where the line of distinction between chorus and actors was usually a sharp one. Thus, in Aeschylus’ Agamemnon, vss. 1348 ff., each of the choreutae in turn pronounces two iambic lines. In particular, the rôles of the two chorus leaders must have been developed in the comus and early comedy so as partly to compensate for the lack of actors. Note that Aristotle does not state merely that comedy sprang from phallic ceremonies but from the leaders (ἐξάρχοντες) of the phallic ceremonies. An illustration of what may result from participation in the action on the part of individual choreutae is afforded by Aristophanes’ Women in Council. I believe that the “First Woman” and the “Second Woman” who appear in our editions as uttering brief remarks at the beginning of this play are not actors but the leaders of the two half-choruses.98 In function they are not at first distinguishable from Praxagora. Indeed, it does not transpire until later that Praxagora herself is an actor, not the coryphaeus. The fact is that in all his plays Aristophanes seems to have assigned his two chorus leaders more extensive participation both in lyrics and in recitative than has been generally recognized (cf. White, op. cit., passim). In my opinion this sort of thing was even more common at an earlier period, and in this way it was possible for the comus to have a quasi-agon from which the later histrionic agon could easily develop. Of course, the chorus leaders could not appear in individualized rôles, as the actors did in the Aristophanic agon, for characters had not yet been introduced into comedy; but they could engage in a contest of perfectly general, depersonalized billingsgate or, at a later period, speak as the poet’s mouthpiece for the pros or cons of any question. Thus, they would not represent individual men, with an individual’s name and characterization, but any men. Their sentiments would have been equally appropriate in the mouths of any of the other choreutae.
The agon and parabasis must necessarily have been flanked on either side by a processional and a recessional. In their simplest form, these need not have involved more than silent marching in and out again; but probably the flute accompaniment was always present, and singing would soon be added. Even when words and singing were employed, there was no necessity of these being newly composed for each occasion or even original at all. It will be remembered that in Aristophanes’ earliest and latest plays he did not write special exodia but borrowed from earlier poets any popular airs that suited his purpose.99 Moreover, Aristophanes’ exodi lack the balanced structure which is characteristic of all divisions which descend directly from the primitive comus; but in this instance that fact has no significance, for the reason that by the end of a comedy (or comus) the two half-choruses would always be reconciled and go marching off together. Nevertheless, the intrusion of the histrionic element, the comparative rarity of the earliest dramatic meter (the trochaic tetrameter), and the absence of a canonical structure make it plain that the recessional of the primitive comus never developed into a regular division—in other words, that the exodus of Aristophanic comedy was the product of a later period.
On the other hand, the Aristophanic parodus resembles the agon and the parabasis in making a large use of the tetrameter (op. cit., p. 185). Moreover, it contains distinct survivals of epirrhematic composition (ibid., pp. 159 and 366), so that, in spite of its histrionic elements and the absence of a canonical form, the parodus ought to be considered as having been exclusively choral by origin and as having developed out of the simple processional before the comus became histrionic.
The theatrical comus, then, must have been something as follows: first a choral parodus, next a semi-histrionic agon, then a parabasis, and finally a recessional which ultimately developed into an exodus. A late notice,100 if correctly emended, informs us that at one time comedies contained no more than three hundred verses. I am of the opinion that this is the type of performance alluded to and that comedy did not, in essence, greatly depart therefrom until actors, as distinct from the chorus, were added.
How did this addition come to be made? It is impossible that the comic playwrights, with the actors of tragedy ever before them, should never have thought of taking this step. Nevertheless, the main impulse seems to have come from another direction. We have seen (p. 36, above) that in the non-theatrical comus the phallus was borne on a pole in the ritual procession with which the comus was originally associated; it was not worn. Neither is it worn by the comus choreutae as represented on Attic vase paintings (Figs. 12-16). But in Old Comedy it is clear that at least some of the characters wore the phallic emblem. That this was in fact the general practice appears from the language in which Aristophanes boasts of the modesty of his Clouds:
And observe how pure her morals: who, to notice first her dress,
Enters not with filthy symbols on her modest garments hung,
Jeering bald-heads, dancing ballets, for the laughter of the young.101
And Dr. Körte (op. cit., pp. 66 ff.) has collected ten passages in other plays of our poet which indicate that Aristophanes was not always so puritanical as he claims to be here. These conclusions are confirmed also by numerous representations, of Attic workmanship, which are plausibly thought to depict actors in Old and Middle Comedy (Figs. 17-19).102 By the time of New Comedy, on the contrary, the phallus was apparently no longer worn, and the characters were garbed in the dress of everyday life. Now the Dorian mime or farce was widely cultivated in the Peloponnesus and Magna Graecia. The performers were individualized actors, not welded into a chorus. They wore the phallus, had their bodies stuffed out grotesquely both in front and behind by means of copious padding, and in general bear a very close resemblance to the comic actors at Athens (Figs. 20 and 21).103 Their performances were loosely connected, burlesque scenes, abounding in stock characters and enlivened by obscenity and ribald jests. Most authorities agree that the burlesque episodes (5) of Old Comedy are derived from this source. According to Aristotle,104 the Megarians claimed that comedy originated with them about 600 B.C. when a democracy with its resultant freedom of speech was established among them. It was even asserted that Susarion, the reputed founder of Attic comedy (see p. 38, above), came from Megara, but this claim is apparently unwarranted.105 The fact remains, however, that Aristophanes and his confrères often speak of stupid, vulgar scenes or jokes as being “stolen from Megara.”106 Though these words have been otherwise explained,107 I believe that Megara, which is the nearest Dorian city to Attica, had something to do with the introduction of the histrionic element into Attic comedy. Of course, this does not mean that Megara is to be regarded as the inventor of Athenian comedy, for the comus was indigenous and received its development on Attic soil and the type of performance which came into being after the introduction of actors was quite unlike anything in Megara or any other part of the Dorian world.
With actors, impersonation became possible for the first time in Attic comedy. Besides the nondescript chorus and chorus leaders, there were now performers who could assume the identity of real or imaginary characters and carry a rôle or, by a change of mask, several rôles through the play. The importance of all this is too obvious to require amplification. It marked the birth of dramatic comedy at Athens. Through the introduction of actors, comedy became amenable to several other influences. Tragedy could at once make itself felt. A histrionic prologue could now be added, the comic prologue corresponding in length and function to the tragic prologue and first episode combined.108 A real agon of actors now became possible, whatever use may have been made previously of the chorus leaders for this purpose. Furthermore, the new Megarian burlesque episodes (5) would naturally be separated by stasima (6) in imitation of tragedy. It would also be possible to insert an episode109 between the parodus and the agon, as is done in Aristophanes’ Plutus, vss. 322-486; or between the agon and the parabasis, as in Aristophanes’ Knights, vss. 461-97; or to compose a second parabasis and to insert an additional episode between them, as in Aristophanes’ Peace, vss. 1039-1126, etc. In addition to all this, tragedy would exert a constant influence in elevating and standardizing all parts of comedy alike.
But the restricted and even disconnected method of elaboration employed in earlier comedy, with its invective, lampoons, and obscene jests, would not suffice to fill so ample a framework. Therefore, it became necessary to broaden and deepen the plots; in fact, now for the first time in Attic comedy was it possible to have a plot worthy of the name. All this is implied in the words which have already been quoted from Aristotle (p. 35, above): “Developing a regular plot was a Sicilian invention, but of the Athenians the first to abandon the ‘iambic’ or lampooning form and to begin to fashion comprehensive themes and plots (καθόλου ποιεῖν λόγους καὶ μύθους) was Crates.” The reference in the first half of this sentence is to Epicharmus, whose name actually appears in Aristotle’s text at this point but without grammatical construction. Epicharmus was a resident of Megara Hyblaea in Sicily, whence he migrated to Syracuse about 485 B.C. Like the Megarians on the Greek mainland, also the Sicilian Megarians laid claim to the honor of having invented comedy.110 They based their pretensions on the fact that Epicharmus flourished and won his reputation before 486 B.C., which was the terminus post quem for the beginning of the official careers of Magnes and Chionides, who were the first poets of state-supported (as opposed to volunteer) comedy, at the City Dionysia in Athens. Epicharmus raised the Dorian mime in Sicily to literary importance, and seems to have improved upon the detached or but loosely connected scenes of his predecessors by stringing them together upon the thread of a common plot-interest. His plays had no chorus and did not touch upon his contemporaries or politics. Now Aristotle’s words concerning Crates must certainly be understood as indicating a resemblance between him and Epicharmus in at least some of these particulars. The expression which I have translated “to fashion comprehensive themes and plots” has been rendered “generalized his themes and plots” by Butcher, “to frame stories of a general and non-personal nature, in other words, Fables or Plots” by Bywater, and “composed plots or fables of a ‘universal’ character” by Cornford (op. cit., p. 217). Whatever other meaning may inhere in this phrase, I think that it must be taken to mean, first of all, that Crates, like Epicharmus, made all or, at least, most of the parts of his plays subservient to one connecting idea or plot; and it seems to me that the previous clause which refers to his abandonment of the “iambic” or lampooning form looks in the same direction. In my opinion, the invective of his predecessors had been episodic and unrelated to its context by any sequence of thought, often being expressed in passages like the following:
Shall we all a merry joke
At Archedemus poke,
Who has not cut his guildsmen yet, though seven years old;
Yet up among the dead
He is demagogue and head,
And contrives the topmost place of the rascaldom to hold?
And Clisthenes, they say,
Is among the tombs all day,
Bewailing for his lover with a lamentable whine.
And Callias, I’m told,
Has become a sailor bold,
And casts a lion’s hide o’er his members feminine.111
Here this abuse is dragged in a propos of nothing, and the three citizens who are assailed within a score of lines have no connection with the main theme of the play. It was this sort of thing, I venture to believe, that Crates discontinued; and Aristotle’s language does not require us to conclude that he relinquished scurrility altogether. It is usually thought, however, that Crates made no assaults of any kind upon his contemporaries but “generalized” his plots by treating imaginary, “ideal” characters in his plays. In other words, he is supposed to have anticipated to some extent the manner and material of New Comedy. I have no desire to combat this view, which simply advances a step beyond my own. The main fact, that of Crates’ having invented plot sequence in Attic comedy, can hardly be made a matter of dispute.
We are indebted to a late authority, Tzetzes, for the following statements:
But also Old Comedy differs from itself [i.e., falls into two types], for those who first established the institution of comedy in Attica (and they were Susarion and his successors) used to bring on the characters (πρόσωπα) in an undifferentiated crowd (ἀτάκτως), and laughter alone was the object sought. But Cratinus [a contemporary of Crates], succeeding them, put a stop to the confusion (ἀταξίαν) and set the characters (πρόσωπα) in comedy for the first time at three; and he added profit to the pleasure of comedy, lampooning the evildoers and chastising them with comedy as with a public scourge. But even he still shared in the archaic qualities and, slightly, in the confusion (ἀταξίας).112
Whatever the ultimate source of this notice, it contains much of value. In the first place, a distinction is correctly drawn between primitive comedy (Susarion to Cratinus; ca. 565 to ca. 450 B.C.) and Old Comedy (450 to ca. 385 B.C.). The earlier period is marked by ἀταξία, which I refer to the practice of having characterless choreutae take part singly as if they were actors (see p. 44, above). Though still occasionally guilty of this practice, as even Aristophanes sometimes was, Cratinus regularly withdrew his choreutae from participation in the dialogue and reduced the performers to three. These three, however, were now real actors, as distinct from the chorus and chorus leaders, and played individualized rôles which demanded dramatic impersonation. The number three was doubtless due to contemporaneous tragedy in which the number of actors had recently been increased by Sophocles from two to three (see p. 167, below).113
A second difference between primitive comedy and Old Comedy is found in the use which was made of invective. If this development had not taken place, Old Comedy would not occupy the unique place which it now holds in the dramatic literature of the world. As we have just seen, the lampooning of primitive comedy was probably episodic and detached from the context, like that in Aristophanes’ Frogs, vss. 416-30; a whole play was not devoted to one person, and no citizen was impersonated by an actor. Its object was merely to cause a laugh and it rarely served any useful purpose, certainly none for the public interests of the state. It was a natural outgrowth of the magical abuse of the old phallic processions. Now Old Comedy, on the whole, was just the reverse of this, and Cratinus seems to have been the innovator who, “generalizing” his plots by giving them a single theme, after the fashion set by Crates, devoted them solely or mainly to political and social questions and dragged his victims in person upon his stage.
When did these changes take place? First let it be noted how they mutually depend one upon another: neither tragedy nor the Sicilian mime could greatly influence early Attic comedy until actors, as distinct from a chorus, were introduced, nor could their influence be long delayed after the actors came. I think that these factors came to fruition not long before 450 B.C.
a) Reverting to Aristotle’s words (quoted on p. 35, above), when are we to suppose that the Athenians began to “treat comedy seriously”? The most obvious answer would be, “486 B.C., when comedy first received official recognition.” Chionides and Magnes are the poets of this period, and there is no reason to believe that they improved upon their immediate predecessors of the “volunteer” comedy otherwise than in a more worthy literary treatment of their plays. Aristophanes describes Magnes’ efforts in the following terms:
All voices he uttered, all forms he assumed, the Lydian, the fig-piercing Fly,
The Harp with its strings, the Bird with its wings, the Frog with its yellow-green dye.114
It is plain that these words refer to plays by Magnes which were called The Lydians, The Gall-Flies, The Harpists, The Birds, and The Frogs. These titles at once remind us of the animal masks which were so common in the comus (Figs. 12-16). Of course, state supervision implies a certain amount of serious attention. Nevertheless I think that in this passage Aristotle had a later period in mind.
It was long ago pointed out that Attic comedies were not published before the time of Cratinus. The fact of publication shows that comedy was at last being treated with true seriousness and helps to explain the ignorance, in later times, with respect to certain points. Though the state records gave the names of comic victors from 486 B.C. on, they did not include information upon matters of mere technique. For knowledge of this sort Aristotle (the ultimate source of Tzetzes) and all other ancient investigators were almost entirely dependent upon what they could glean from the editions of Cratinus, Crates, and their successors. Now the earliest texts available revealed the use of characters, prologues, and three actors as well as of the parodus, agon, parabasis, and exodus. Why did Aristotle specifically name the first group and not the second?
In my opinion, Professor Capps115 has provided the correct answer. He maintains that Aristotle distinguished two kinds of ignorance concerning the history of comedy. In the first place, there was the Egyptian darkness which covered the period previous to 486 B.C. For example, when Aristotle declared that comedy “already had certain forms” (σχήματά τινα) at this time, he could not have specified what these forms were; he was merely surmising that the fact of state supervision presupposed more or less definiteness of form. In the second place, there was the period of semi-darkness immediately after 486 B.C. Tradition must have placed in this period the introduction of characters, prologues, and three actors, and so Aristotle singled them out for mention. But tradition had not handed down also the names of the innovators, and in the absence of texts it was impossible to probe the matter further. Needless to state, the situation regarding the other innovations, whether of this period or earlier, was much worse.
b) Though Thespis is said to have invented the prologue in tragedy, this statement is justly discredited (see p. 298, below); and no tragedy is actually known to have had one before Phrynichus’ Phoenician Women (476 B.C.). Aeschylus’ Suppliants (about 490 B.C.) and Persians (472 B.C.) have none. It is most unlikely that comedy should have anticipated tragedy in this feature.
c) Capps116 has plausibly suggested that knowledge of Epicharmus’ achievements in comedy was brought to Athens by Aeschylus, who is known to have been in Sicily ca. 476 B.C., shortly after 472 B.C., and for about two years before his death there in 456 B.C.
d) The third actor was introduced into tragedy between about 468 and 458 B.C., and it is more probable that the use of three actors in comedy was borrowed from tragedy than vice versa.
e) Cratinus won his first victory at the City Dionysia of 452 B.C. and (f) Crates at that of 450 B.C. Doubtless the activity of both men began somewhat earlier.
g) It is incredible that the state should have postponed official control of comedy at the Lenaean festival until about 442 B.C., if the developments which we have been sketching had taken place long before.
h) The earliest comedian to refer to Megarian comedy is Ecphantides, whose first victory was won between 457 and 453 B.C. Whenever Aristophanes “names any writers of ‘vulgar comedy’ who used the stale antics which he repudiates, these writers are his own predecessors and contemporaries of the Attic stage.”117 This implies that the borrowing was a fairly recent occurrence.
i) Finally, Megara was actually under the sway of Athens during 460/59-446/45 B.C. The opportunity for the exchange of ideas between Megara and Athens would naturally be most favorable at that time.
In view of the preceding considerations, I am of the opinion that actors were introduced into Athenian comedy shortly before 450 B.C.
The Greek Theater.118—Since, as we have seen, both tragedy and comedy among the Greeks were choral by origin, the center of their theaters was a circular “dancing place” called an orchestra119 (ὀρχήστρα), in the middle of which stood a thymele (θυμέλη) or “altar” (Figs. 22 f.).120 When an actor was added to the tragic choreutae, it became necessary to provide a dressing-room where he might change his mask and costume. This temporary structure was called a σκηνή (“hut”: our English word “scene”), and at first stood outside the spectators’ range of vision. Afterward it was brought immediately behind the orchestral circle and then served also as a background in front of which the dramatic action was performed. Its face was pierced by doors, usually three but sometimes only one, which were conventionally thought of as leading into as many different houses. The scene-building often had two projecting side wings called parascenia (παρά, “beside” + σκηνή). The front of the scene-building and of the parascenia came to be decorated with a row of columns, the proscenium (πρό, “before” + σκηνή). The top of this proscenium was used by actors when they had occasion to speak from the housetop or were thought of as standing upon some elevation. In the course of time it was employed also for divinities, especially in epiphanies at the close of tragedies (see p. 292, below). Since this spot was never invaded by the singing or dancing of the chorus and was the only place reserved for actors exclusively, it came to be called the logium (λογεῖον, from λέγειν to “speak”) or “speaking place.”121 Behind the logium was the second story of the scene-building, known as the episcenium (ἐπισκήνιον; ἐπί, “upon” + σκηνή); its front wall was pierced by one or more large doorways. Past each parascenium a “side entrance” or parodus (πάροδος; παρά, “beside” + ὁδός, “passage”) led into the orchestra. These entrances were used by the audience before and after the play, and during it by the actors (who could use also the doors in the scene-building) and the chorus. The parodi were often framed by beautiful gateways (Figs. 51 f.). The remainder of the orchestral circle was surrounded by the auditorium, the “theater” proper.122 Chorus and actors stood on the same level in the orchestra or in the space between it and the scene-building. There was no stage in the Greek theaters until about the beginning of the Christian era.
But when the Greek theaters came under Roman influence and were provided with a stage, these technical terms naturally acquired a somewhat different significance (Figs. 24 and 62-64).123 The proscenium was still the columned wall in front of the scene-building, but it now stood upon the stage (at the rear), and the stage itself was the logium. Whenever theophanies required a still higher level, this was furnished by the top of the proscenium,124 which was called the theologium (θεολογεῖον; θεός, “god” + λογεῖον) or “speaking place of divinities.”125 The space beneath the stage, or its front wall alone, was known as the hyposcenium (ὑποσκήνιον; ὑπό, “beneath” + σκηνή).126 There were now two sets of parodi, leading upon the stage and into the orchestra respectively. These two paragraphs are meant for purposes of orientation and are written from the standpoint of one who believes with Dörpfeld that in Greek theaters of the classical period actors and chorus normally moved upon the same level.127
A Greek town could hardly be so small or so remote as not to have its own theater and dramatic festival (Figs. 25 and 70 f.).128 The Greek theaters were regularly built upon a hillside and often commanded an outlook over a scene of great natural beauty and picturesqueness (Figs. 26-28).129 So far as such structures have come down to us, the oldest is the theater of Dionysus Eleuthereus at Athens, and this is also the one of greatest interest to us, for the reason that in it were produced practically all the masterpieces of the greatest Greek dramatists (Figs. 1 and 31-41).130 It seems strange that this building should not have remained continuously known to men from ancient times until the present hour, but in fact its very location passed into oblivion for centuries. During mediaeval times and until well into the modern era it was thought that the theater or odeum of Herodes Atticus, a Roman structure of the second century A.D. and situated at the opposite end of the Acropolis, represented the Dionysiac theater of the classical period (Fig. 29).131 The correct site was first pointed out by R. Chandler in 1765, and is clearly indicated by a bronze coin of imperial times which shows the relation subsisting between the theater of Dionysus and the Parthenon (Figs. 30 f.).132 Excavations were conducted desultorily from time to time, beginning in 1841, but were not completed until the work under Dörpfeld’s direction in 1886, 1889, and 1895.
The oldest structure in the precinct of Dionysus Eleuthereus is the earlier temple (Fig. 32).133 This was built in the sixth century B.C., possibly in 534 B.C., when Pisistratus established the tragic contest. Here was housed the cult image of Dionysus which had been brought from Eleutherae.
Somewhat later are the remains of the early orchestra. According to late notices,134 the original place of holding theatrical performances in Athens was an orchestra in the old market place, the location of which has not yet been determined. At that period the audience sat upon “wooden bleachers” (ἴκρια), which are said135 to have collapsed on the occasion of a contest between Aeschylus, Pratinas, and Choerilus in the seventieth Olympiad (about 499 B.C.). In consequence, a new theater was constructed in the precinct of Dionysus, where the seats, though still of wood, could be supported in part by the south slope of the Acropolis. When the stone theater on this site was first brought to light, it was erroneously supposed that this was the structure which had been erected as a result of the accident just mentioned. As a matter of fact, practically all that remains of the first theater are certain fragments of the orchestra (Figs. 33 f.).136 These are sufficient to indicate that this orchestra was over seventy-eight feet in diameter and stood nearly fifty feet farther south than the later orchestra (Figs. 32 and 32a).137 As it receded from the Acropolis it was banked up to a maximum of about six and a half feet, leaving a declivity immediately behind it. The extant plays of this period show that for about thirty years no background of any kind stood in this declivity (see p. 226, below). Theatrical properties, such as a tomb, might be temporarily built at the center or to one side of the orchestra. If dressing-rooms were then provided for the actors and chorus they must have stood some distance away. In the absence of a back scene, the performers could enter only at the sides. These same entrances were used also by the spectators in assembling. The seats, being of wood until the fourth century, have left no trace; but there can, of course, be no doubt of their position on the slope. Well up the side an ancient road cut the auditorium into an upper and lower section138 and permitted ingress and egress for the audience at two additional points. The Athenian theater was somewhat unusual in having these upper entrances.
About 465 B.C., as the plays indicate,139 a wooden scene-building was set up behind the orchestra, where the declivity had been.140 The front of this was probably pierced by three doors, which might be conventionally thought of as leading to as many different buildings, and thus the number of entrances available for the actors’ use was more than doubled. This seemingly simple alteration produced profound changes in dramatic technique (see pp. 228-31, below). The scene-building of this period must be thought of as quite unpretentious: its material was wood, it probably consisted of but a single story, and I think it had neither parascenia nor a columned proscenium (Fig. 74; see p. 235, below). Its construction was flimsy enough for it to be capable of being easily rebuilt or remodeled to meet the scenic requirements of each drama, for of course it was not until long after the introduction of a scenic background that the plays were uniformly laid before a palace or temple. According to Aristotle, Sophocles was the inventor of scene-painting, and this is also said to have been invented during the lifetime of Aeschylus.141 If these notices are correct, we must suppose that scene-painting was invented in the decade ending in 458 B.C. and so under theatrical conditions such as have just been described. This would mean that at first the scenery must have been attached directly to the scene-building itself and not inserted between the intercolumniations of the proscenium columns.
The next building in the precinct seems to have been the later temple, slightly south of the earlier one (Fig. 32). Its substructure was of breccia (conglomerate), and its erection must be assigned to about the last quarter of the fifth century B.C.142 An image of Dionysus by Alcamenes found its home here.
Of the same material are the foundations of the parascenia and of the front and back walls of the scene-building (Fig. 35),143 and perhaps they are to be assigned to the same period as the temple which has just been mentioned.144 The superstructure was still of wood, since the wide variation of scenic setting called for a background which could readily be adapted to changing needs. It is likely that the ten square holes in the rear foundation wall (Fig. 38) were intended to receive the supporting beams of such an adjustable structure.145 Probably the scene-building now rose to a second story, a supposition which is confirmed by the use of the crane or μηχανή (“machine”) in the extant plays of this period (see pp. 289 and 292 f., below). At about the same time a proscenium (also of wood) was erected before the parascenia and the intermediate front of the scene-building (see pp. 235 f., below), and painted panels of scenery could be fastened between its intercolumniations. In my opinion, we must suppose that such a proscenium stood far enough removed from the front of the scene-building146 so that, when there was no occasion to fill the intercolumniations with panels, a porch or portico was automatically produced (its floor probably raised a step or two above the orchestra level), in which semi-interior scenes might be enacted (see pp. 238 f., below). It has even been maintained that a projecting vestibule was sometimes built out from the center of the proscenium in order to provide additional space of a semi-private sort (see pp. 236 f., below and Fig. 73). Of course, no foundations for such a structure are found either at this period or subsequently, for the reason that permanent foundations for something which was only occasionally employed would have been unsightly and in the way for the greater part of the time. No fragments belonging to the orchestra of this period have been discovered (see next paragraph and p. 73). Moreover, the seating arrangements belong to the Lycurgus theater of the next century. Fortunately, however, there can be no doubt as to the relative position of these parts: it is apparent that the whole theater has been pushed some fifty feet farther north (Fig. 32), and the causes of this alteration are not hard to guess. In the first place, room was thus secured for the scene-building without occupying the space immediately in front of the earlier temple of Dionysus. In the second place, the slope of the Acropolis could now be employed more extensively as a support for the seats of the spectators. There are no means of determining whether this slight change in site was made at this period or about 465 B.C., when the first scene-building was erected.
Slight as may seem the theater remains which have been discussed up to this point, it must be noted before proceeding that they entirely exhaust the field. There is not a stone outside of Athens which can be assigned to any Greek theater before 400 B.C.147 Yet all the plays of Aeschylus, Sophocles, and Euripides, and all the extant comedies of Aristophanes, except two, were performed before this date! In the latter half of the fourth century Lycurgus, who was finance minister of Athens between 338 and 326 B.C., “completed”148 the theater which is reproduced so clearly in Dörpfeld’s plan (Fig. 32) that it is unnecessary to describe it at length. Most of the stone remains now upon the site belong to this structure. So far as the auditorium is preserved, its arrangements and furnishings are almost entirely those of Lycurgus’ time. Most of the inclosing walls, the stone thrones in the front row for the use of dignitaries, and the stone seats for the rest of the audience all belong to this period (Fig. 36). The only part of the present orchestra which goes back to the fourth century is the gutter just inside the balustrade (Fig. 37), but this is sufficient to show that the Lycurgus orchestra was sixty-four feet and four inches in diameter or exactly sixty Greek feet. This figure is significant as showing that the orchestra was the starting-point in the measurements and not incidentally derived from some other part of the theater. Behind the orchestra and upon the old foundations was now erected a scene-building of stone, one hundred and fifty-two feet in breadth and twenty-one feet deep at its shallowest part. About its parascenia stood a row of stone columns, from which it can be estimated that the first story was about thirteen feet in height. But the stone connecting columns which Dörpfeld restored before the central part of the scene-building (Fig. 32) have been assailed on every hand and have now been relinquished by their sponsor.149 This part of the proscenium was still of wood, for though the scenic requirements by this time were fairly standardized for each genre, the conventional setting for tragedy was quite different from that for comedy or satyric drama. Furthermore, the Greeks seem to have been slow to lose the notion that a wooden background was necessary in order to secure the best acoustic results.150 This wooden proscenium probably did not stand so close to the scene-building as the drawing would indicate, but formed a portico as in the Hellenistic theater (Fig. 38). At the same time, or possibly at the close of the fifth century, a colonnade was built just behind the scene-building as a place of refuge from heat and sudden showers. There are two considerations which make the Lycurgus theater highly important to us: in the first place, here were produced the plays of the Greek New Comedy which furnished the originals of Plautus’ and Terence’s Latin plays and which has partially been restored to us by the recent discovery of large fragments of Menander’s comedies; and in the second place this fourth-century structure probably reproduced in stone the main outlines of the earlier theater in which the later tragedies of Sophocles and Euripides and all the plays of Aristophanes were performed. This supposition is strengthened by the fact that the extant fifth-century dramas could readily be “staged” in the Lycurgus theater.
Further alterations were made in the Athenian theater during the first or second century B.C. (Fig. 38).151 So far as can now be established, this Hellenistic theater differed from its immediate predecessor only in two particulars. The front of the parascenia was moved back about six and a quarter feet,152 the parodi being thereby enlarged to the same extent. What advantage was gained by this alteration has not yet been discovered. The other change consisted in the erection, at last, of a stone proscenium, about thirteen feet in height, between the parascenia and about six and a half feet in front of the central fore wall of the scene-building. At Epidaurus, Eretria, Delos, etc., the supports of the proscenium were only half-columns, and sometimes they had grooves or rims running vertically along their sides or had the rear half of the column cut into an oblong for the purpose of providing a firmer fastening for the painted panels (πίνακες) in the intercolumniations (Fig. 72). But at Athens the proscenium columns were whole and were not equipped with any of these devices.
We have already passed far beyond the time when masterpieces of Greek drama were receiving their premier performances in the Athenian theater; after the third century the dramatic productions in Attica were no longer of consequence. Yet for the sake of completeness it will be necessary to record briefly two later periods in the history of this structure.
The result of the earlier of these remodelments is commonly known as Nero’s theater, for the reason that its façade originally bore an inscription of dedication to Dionysus and Nero. The motive for the alteration and dedication is doubtless to be found in the Emperor’s visit to Greece and “artistic” triumphs there in 67 A.D. Under the circumstances it is not surprising that two features of Roman theaters were now for the first time introduced into Athens: a stage was built before the scene-building, and the hitherto full orb of the orchestral circle was thereby infringed upon. At the back of the stage rose a new proscenium, probably no longer in the form of a straight and simple colonnade but an elaborate façade with projecting and receding members, such as was common in the Roman and Graeco-Roman theaters (Figs. 40 and 59). The depth of the stage cannot be exactly determined,153 but its front wall is usually thought to have coincided with that of the stage now standing, which belongs to the next period. But we shall presently find reasons for believing that, though the Nero stage was deeper than the Hellenistic proscenium, it was shallower than the later (Phaedrus) stage (see pp. 75 and 99, below). Space would thus be left for the parodi still to lead directly into the orchestra. Dörpfeld first estimated the height of the Neronian stage at about four feet nine and a half inches (see next paragraph), but is now inclined to think that it belonged to the high Graeco-Roman type.154 In my judgment, however, his earlier position is to be preferred. I consider it probable that stone steps led from the orchestra to the center of the stage, as in the Phaedrus theater (Fig. 40). Just outside the gutter of the Lycurgus theater was erected a marble balustrade (Fig. 39),155 which stood about three and a half feet above the orchestra level and protected the spectators from accident when gladiatorial combats (another Roman institution) or the like were being exhibited in the orchestra. In order to compensate for the curtailment of the orchestra by the stage, the gutter, which had been left open except opposite the vertical aisles of the auditorium, was covered over, except for occasional rosette-shaped openings. Up to this time the orchestra seems to have had no covering but hard-pressed earth, but it was now paved with marble slabs. In the middle of the pavement is a rhomboid design (Fig. 40), and in its central block is a depression about twenty inches in diameter, by means of which an altar of Dionysus (the thymele) was doubtless held in place.
The final alterations in the Athenian theater (Fig. 40)156 were made in the third or fourth century A.D. by Phaedrus, governor of Attica (Ἀτθίδος ἀρχός), who dedicated the “platform of the theater” (βῆμα θεήτρου) to Dionysus in an inscription which still stands on the uppermost of the stone steps leading from the orchestra to the stage. The gutter was now filled up with earth and refuse, and the rosette-shaped openings in its covering were carefully closed. Plaster was used as needed, and the balustrade and the front wall of the stage (the hyposcenium) were reinforced and made water-tight by supporting walls. The intention was plainly to enable the orchestra to be flooded for the representation of mimic sea fights. The stage was partially rebuilt and was lowered. The hyposcenium was adorned with a frieze (Figs. 39 and 41),157 the extant portion of which is interrupted at three points by two blank spaces and a recess. The latter is filled by a kneeling Silenus. It is clear that the frieze had been used before and that its slabs had originally been placed in immediate juxtaposition. Moreover, the heads of the figures have been cut away, so that the frieze, when complete, must have been about half a foot higher than at present. The Phaedrus stage is four feet three and a half inches high; and as Dörpfeld was originally inclined to believe that this same frieze had at first stood before the Neronian stage, he estimated the height of the latter at about four feet nine and a half inches. In my opinion, this estimate ought to be retained. But though Dörpfeld now considers the Nero stage to have been higher than this, he has not indicated whether he still believes its front wall to have been the original position of the frieze.
It has been suggested that after the lapse of two centuries or more the Neronian stage was perhaps in need of repair or renewal and that the changes for which Phaedrus was responsible are thus to be explained. However that may be, other influences were plainly at work. I think that at this period the Athenian theater was at last thoroughly Romanized. That is to say, I think that the Nero stage did not project so far into the orchestra (see p. 72, above), but was now enlarged so as to accommodate all the performances, and that at the same time the Roman custom of placing seats in the orchestra was for the first time introduced into Athens. But in order that the orchestra might find occasional continuance of its function as a place of exhibition, or possibly because of interest in the sport per se, all openings were closed up and the old dancing place was made capable of being flooded. It follows that the parodi no longer debouched into the orchestra but led to steps at either side of the stage, as shown in Fig. 40. The participants in the mimic sea fights and gladiatorial combats and the spectators at other performances could enter the orchestra only by passing over the stage and down the front steps. Of course, the presence of spectators so close to the performers would permit no type of stage except one of moderate height; evidently even the low Nero stage was a little too high under these conditions.
The foregoing account of the Athenian theater is founded, in the main, upon Dörpfeld’s conclusions, but the reader needs to be warned that not all of his conclusions are acceptable to everyone. Until about half a century ago our information concerning Greek theaters was largely restricted to literary tradition. There was no theater of the earlier Greek types above ground, and even the exact location of the Athenian theater had been, during many centuries, forgotten. The literary tradition was mainly derived from Vitruvius, a Roman architect at the beginning of the Christian era, who devoted two chapters of Book V in his work On Architecture to a description of Greek and Roman theaters. According to him, the front and back walls of the Roman stage were determined by the diameter of the orchestral circle and one side of an inscribed equilateral triangle; in other words, its depth would be one-half the radius of the orchestra (Fig. 42).158 Its height was not to exceed five feet,159 since all the performers stood on the stage and the unelevated front half of the orchestral circle was reserved for the seats of senators. In the Greek theater, on the other hand, Vitruvius asserted that the front wall of the stage was marked by one side of an inscribed square, and its back wall, which he calls the scaenae frons, by the parallel tangent, its depth being thus about three-tenths of the radius (Fig. 43).160 Its height was to range between ten and twelve feet. Vitruvius expressly states that this stage in the Greek theater was called a logium, that the tragic and comic actors performed in scaena161 and the “other artists” per orchestram, and that for this reason the Greeks drew a distinction between the adjectives “scenic” and “thymelic” as applied to performances and performers.162 The differences between the two types of structure are obvious: (1) the auditorium and orchestra in Vitruvius’ Roman theater occupied exactly a semicircumference, in his Greek theater distinctly more than this; (2) the Roman stage was deep and low, the Greek high and comparatively shallow; (3) in the Greek theater both orchestra and stage were employed (separately) by different forms of entertainment; in the Roman theater all performers stood on the stage and the semicircular orchestra was occupied by the seats of senators.
Moreover, Pollux (second century A.D.) states that in the Greek theater “the σκηνή belongs to the actors and the orchestra to the chorus.”163 Everyone used to think (and some still do) that σκηνή here signified “stage” and that Vitruvius’ reference to scaenici and thymelici was to be interpreted in a similar fashion. Accordingly, it was supposed that Greek actors performed (and had always performed) upon a ten- or twelve-foot Vitruvian stage and the dramatic chorus in the orchestra below. Confirmation was found for this theory in Pollux’ further mention of ladders rising from the orchestra to the σκηνή.164 The use of both orchestra and stage is mentioned a few times also in scholia (ancient commentaries) upon the Greek plays. The possibility of other interpretations of these passages will be considered later (see pp. 97 ff., below). For the present this should be said: We are interested in the Greek theater mainly because of Aeschylus, Sophocles, Euripides, and Aristophanes, all of whom lived in the fifth century B.C., and Pollux and Vitruvius, who flourished many centuries later, nowhere assert that they are attempting to describe the theater of this earlier period. Nevertheless, this initial assumption used tacitly to be taken for granted, and these Procrustean conditions were arbitrarily imposed upon the extant Greek dramas by all editors and commentators alike. As a matter of fact, such a difference of level between orchestra and stage, chorus and actors, with no convenient connection between the two, presented an insuperable obstacle to the (imaginary) “staging” of the fifth-century plays. Various expedients were proposed to evade the difficulty. One of the most popular was that of G. Hermann, who in 1833 suggested that the Greek orchestra was covered with a wooden platform to within a few feet of the stage level and that thus a more intimate connection between the two was established, and Wieseler (1847) proposed to identify this platform with the thymele. Nonsensical as this suggestion appears to everyone without exception now, it enjoyed a tremendous vogue for some time. In the eighties the news began to seep through to Western Europe and this country that Dörpfeld had evolved a new theory, to the effect that actors and chorus had performed in the orchestra on the same level until Roman times.165 Again, Mr. A. E. Haigh (1889) maintained that a low stage was employed uninterruptedly until the fourth century B.C., when a high Vitruvian stage was introduced. Dr. Bethe (1896) contends that at first actors and chorus performed in the orchestra but that about 427 B.C. a low stage was introduced, which in the fourth century was raised to the Vitruvian level. On the other hand, Dr. Puchstein (1901), who stated in his Preface that he ignored the literary evidence, argued for a Vitruvian stage already in the fifth century. And now Professor Fiechter (1914) has given his adherence to Bethe’s hypothesis that a low stage at the end of the fifth century was raised to a high one in the fourth. It will be seen that all authorities are in substantial agreement that the Greek theater had a stage, even a high Vitruvian stage, but they are hopelessly divided with regard to the important detail as to when this stage was introduced—at the very first, at the close of the fifth century, in the time of Lycurgus, in the Hellenistic period, or in the reign of Nero.
But before taking up the question of the stage in the Greek theater, it will first be necessary to determine Vitruvius’ relationship to the matter. The Roman architect’s description of the Roman theater does not coincide precisely with any extant Roman theater. Nevertheless, there has never been any doubt as to the general type of structure which he had in mind. It is evident, however, that he is describing no particular, actually existent, theater but is giving directions for an ideal structure. Indeed, he declares: “Whoever wishes to use these directions will render the perfect qualities of theaters faultless.”166 There is, therefore, no reason to expect that his directions for Greek theaters would agree any more closely with any extant Greek theater, and in fact they do not. During the last two decades of the nineteenth century the ancient theaters at Epidaurus, Oropus, Thoricus, Eretria, Sicyon, Megalopolis, Delos, Assus, Pergamum, etc., were unearthed. The first result of this activity was to show that no two of these structures were entirely alike and that none exactly corresponded to Vitruvius’ directions. Furthermore, it has become evident that all ancient theaters are no longer to be classified under the two general Vitruvian types, “Greek” and “Roman,” but rather under a larger number of categories according to time, place, and conditions of use. But the question which one of these types Vitruvius had in mind still remains, and unfortunately the answer has not been so clear as to compel everyone’s acceptance. In Vitruvius’ day many Hellenistic, stageless theaters were still standing, and the modern attempt to identify these with Vitruvius’ Greek type and to force them into conformity with his prescriptions has wrought great confusion in the field of scenic antiquities. But Vitruvius nowhere professes to be writing a history of Greek theaters nor had he any intention of presenting antiquarian lore. His book was planned for distinctly practical purposes. Now in his day only two kinds of new theaters were being erected, the Roman and what Dörpfeld has christened the Graeco-Roman.167 Dörpfeld supposes the latter type to have originated with the theater which Pompey had built in Rome in 55 B.C. This is said to have been modeled upon the Greek theater at Mitylene in the island of Lesbos,168 and Dörpfeld supposes that the orchestra of Pompey’s theater was kept free of seats, after the Greek fashion, and devoted to thymelic performances, but that the top of the proscenium, despite its height and narrowness, was converted into a stage, to which, according to Roman practice, the comic and tragic actors were now elevated. However this may be, the fact remains that from about this time theaters of this type were so extensively built or created by a remodeling of Hellenistic theaters that they became the only rivals of purely Roman structures. Such theaters are found in the Nero theater at Athens (according to Dörpfeld’s present but questionable view), Pompeii, Segesta, Syracuse, Taormina, and extensively in Asia Minor. Early in the nineteenth century Schönborn and Wieseler correctly recognized buildings of this type as representing Vitruvius’ Greek theater.169 But later on, when the earlier Greek theaters were revealed by new excavations at Athens and elsewhere, an attempt was made to identify these with Vitruvius’ Greek type. Dörpfeld himself fell into this error and in Das griechische Theater maintained that Vitruvius had misunderstood the function of the Hellenistic proscenium, interpreting as a stage what in fact was only a background. But though Dörpfeld thus incurred a large share of blame for confusing the situation, he soon came to recognize his error and frankly recanted.170 Unhappily the pro-stage writers still persist in it.
It might be supposed that Vitruvius’ Greek theater could readily be identified by comparing his directions for the height and depth of the stage with the actual measurements of various Greek theaters. Dörpfeld and Fiechter have both attempted this but without any great success.171 For the sake of convenience and clearness I have drawn up their figures in the form of tables. Dörpfeld cited six Graeco-Roman structures as affirmative arguments and two Hellenistic buildings as negative arguments. Of course, the figures for the Hellenistic theaters refer to the proscenium, in which some would recognize a stage. The problem, therefore, is not merely as to what type of Greek theater Vitruvius was describing, but the function of the proscenium in Hellenistic theaters is also involved. On the other hand, Fiechter, whose object is diametrically opposed to Dörpfeld’s, cites four Hellenistic and six Graeco-Roman theaters as positive and negative arguments respectively.
Table I (Dörpfeld)
| Buildings | Radius of Orchestra | Three-tenths of Radius | Depth of Stage or Proscenium | Height of Stage or Proscenium |
|---|---|---|---|---|
| Graeco-Roman: | ||||
| Termessus | 11.00 m. | 3.30 m. | about 4.00 m. | 2.45 m. |
| Sagalassus | 12.75 m. | 3.80 m. | 5.70 m. | 2.77 m. |
| Patara | 11.85 m. | 3.55 m. | 3.50 m. | 2.50 m. |
| Myra | 17.50 m. | 5.20 m. | 3.50 m. | |
| Tralles | about 3.00 m. | |||
| Magnesia (rebuilt) | at least 2.30 m. | |||
| Hellenistic: | ||||
| Eretria | 2.40 m. | |||
| Oropus | 1.95 m. |
Table II (Fiechter)
| Buildings | Radius of Orchestra | From Center of Orchestra to Scaenae Frons | Three-tenths of Radius | Depth of Stage or Proscenium | Height of Stage or Proscenium |
|---|---|---|---|---|---|
| Hellenistic: | |||||
| Priene | 9.32 m. | 9.31 m. | 2.79 m. | 2.74 m. | 2.72 m. |
| Ephesus | 12.33 m. | 12.25 m. | 3.69 m. | 2.62 m. | |
| Delos | about 10.55 m. | 10.60 m. | 3.16 m. | 3.60 m. | 3.00 m. |
| Magnesia | more than 2.30 m. | ||||
| Graeco-Roman: | |||||
| Termessus | 9.90 m. | 12.60 m. | 2.97 m. | 4.00-5.5 m. | |
| Sagalassus | 12.73 m. | 17.94 m. | 3.80 m. | 7.54 m. | 2.77 m. |
| Patara | 11.85 m. | 14.50 m. | 3.55 m. | 6.00 m. | 2.50 m. |
| Tralles | 13.20 m. | 3.96 m. | 6.50 m. | at least 2.50 m. | |
| Magnesia (rebuilt) | 10.65 m. | 3.20 m. | 6.00 m. | more than 2.30 m. | |
| Ephesus (rebuilt) | 14.47 m. | 12.50 m. | 4.34 m. | 6.00-9.00 m. | 2.62 m. |
It will be observed that five theaters appear in both tables, and that for three of them the figures do not altogether agree. This is to be explained as due to differences in the manner of taking the measurements. Thus, for Termessus, Fiechter gives for the depth 4 m. (Dörpfeld’s figure) and 5.5 m. Similarly, for Ephesus he gives 6 m. and 9 m., and explains that the former does not include the socle projections. Evidently Fiechter still believes that the scaenae frons in Vitruvius’ description of the Greek theater ran behind the proscenium and did not include it (see p. 76, n. 2, above). The same difference of interpretation probably accounts for 6 m. (Fiechter) and 3.50 m. (Dörpfeld) being reported as the depth of the stage at Patara.
A similar opportunity for variance of measurement occurs also in connection with the orchestra. In my opinion, Vitruvius used this term in its broadest sense, viz., as including all the space between the lowest tier of seats172 (Fig. 43). Fiechter’s measurement of the Hellenistic orchestra at Priene is given on this basis. Sometimes, however, the term is used with reference to the space bounded by the gutter.173 Fiechter states that this was his method in measuring the Hellenistic orchestras at Ephesus and Delos. The discrepancy in the reports concerning the orchestra at Termessus (9.90 m. and 11 m.) is also to be explained thus.
But whatever allowance may be made for variations of this sort, I think that whoever impartially examines these figures with the expectation of obtaining a clear answer to the problem involved will be doomed to disappointment. Vitruvius’ Greek stage should range between ten and twelve feet (Roman) in height, or 2.959 m. and 3.55 m., respectively. Only one Graeco-Roman stage and one Hellenistic proscenium in both tables fall within these limits.174 On the other hand, though Dörpfeld is clearly right in maintaining that the proscenia at Eretria and Oropus are too shallow to accommodate the entire histrionic action of a play, Fiechter makes it appear that Vitruvius’ rule that the stage of the Greek theater should be about three-tenths of the orchestra radius in depth is satisfied more closely by the Hellenistic proscenium than by the Graeco-Roman stage. It should be emphasized, however, that he obtains this result only by shifting the value of the word “orchestra,” taking it now in the largest and now in a narrower sense.
Fiechter has tried to utilize Vitruvius’ diagram still further by pointing out that in Vitruvius’ Greek theater the distance from the center of the orchestra to the front wall of the stage (the hyposcenium) plus the depth of the stage, i.e., the distance from the center of the orchestra to the scaenae frons, ought to equal one radius (Fig. 43). The figures in the first two columns of his table apparently show that this condition is met by the Hellenistic theaters and is not met by the Graeco-Roman theaters. But here again we encounter a variable quantity caused by a dispute as to whether the proscenium is to be counted a part of the scaenae frons (see above). In the Patara theater the distance from the center of the orchestra to the hyposcenium is 8.50 m. (14.50 m. - 6.00 m., Fiechter’s figures), and the depth of the stage according to Dörpfeld, who measures from the proscenium, is 3.50 m. Therefore, the total distance is 12 m. as against a radius of 11.85 m. Again, in the Termessus theater the distance from the center of the orchestra to the hyposcenium is 7.10 m. (12.60 m. - 5.50 m., Fiechter’s figures), and the depth of the stage is 4 m. according to Dörpfeld, measuring as before. Therefore, the total distance is 11.10 m. as against a radius of 11 m. according to the largest (Vitruvian) measure of the orchestra. These correspondences are close enough so as not to be unworthy of comparison with those obtained by Fiechter.
In my opinion, the net result of the above must be the frank recognition that such data concerning the Greek theaters as are at present known to us do not afford convincing proof as to the type which Vitruvius was describing. Nor need this conclusion surprise us, if we accept Dörpfeld’s theory that Pompey’s theater was the first example of the Graeco-Roman type. We have no information concerning the Mitylene theater, upon which Pompey’s building was modeled, nor concerning the number or extent of its departures from that model. But any theater in Asia Minor at that time must have belonged to the Hellenistic type. Consequently, a certain resemblance between Hellenistic and Graeco-Roman theaters was inevitable. If Vitruvius was describing an old type, viz., the Hellenistic, its variations in regard to the particulars just discussed must have been too great for him to be able to find any single formula which would comprehend them all, and he had to content himself with recording a theoretical ideal. Or if he was describing a contemporaneous but developing type, viz., the Graeco-Roman, we must suppose that his authority was not sufficient to secure the adoption of his rules by later architects.
Are we, then, unable to determine which type of Greek theater was the subject of Vitruvius’ discussion? I think that we can, but that we must depend upon other arguments. I mention a few of the many which have been advanced: (a) In the Hellenistic and earlier Greek theaters the orchestra, in the narrowest sense (see p. 83, n. 2, above), usually formed a complete circle, or at least, if its boundary was not actually continued into a complete circle, there was room for one without infringing upon the proscenium. Examples of this are found at Epidaurus (Fig. 46), Athens (Fig. 38), Eretria (Fig. 53), Oropus (Fig. 56), Magnesia, Piraeus, etc. Fiechter denies this (op. cit., p. 65), but only because he chooses to understand the word “orchestra” in a larger sense. Now though Vitruvius used the term in the largest sense (measured from the lowest seats, see p. 83, above) he nowhere informs us what relative size the most restricted orchestra should or might have as compared with the largest space passing under that name.175 But his directions require the stage to intrude so far upon his orchestra that it is apparent that, if the same proportions were to be observed as in the Hellenistic theaters, there could be no such full orchestra with a smaller diameter. This is also true of Graeco-Roman structures, and in this important respect they resemble Vitruvius’ Greek theater and the Hellenistic theaters do not.
b) The logium of Graeco-Roman theaters is never supported by columns along its front wall. The only exception to this statement is found at Priene (Figs. 63 f.), where the columns of the Hellenistic proscenium were left standing when the theater was remodeled. The reason why columns were not set in this place is obvious—the floor of the Graeco-Roman stage naturally was thought of as representing earth or a street and it was manifestly improper for either to be supported on columns.176 On the contrary, so fundamental an aesthetic principle would have been violated if the actors had regularly appeared upon the top of the Hellenistic proscenium. But there is no doubt that Vitruvius’ Greek theater had a stage for actors. It is, therefore, more likely that this corresponds to the Graeco-Roman logium than to the colonnade-like proscenium of the Hellenistic theaters. Moreover, the columns of the Hellenistic proscenia were in some cases unmistakably equipped to hold painted panels. But if the actors had stood on top of the Hellenistic proscenium, this scenery would have been beneath their feet and not behind them!
c) Vitruvius discussed the theatrum Latinum in chapter 6 of his fifth book and his theatrum Graecorum in chapter 7. The former chapter is longer than the latter by more than a half, and the latter begins with these words: “In the theaters of the Greeks not all things are to be done in the same way” (as in the Roman theaters). The implication is plain that some of the directions in chapter 6 are to be understood as applying also to the Greek theater of chapter 7, and of course the particulars involved would be those which are not modified by the discussion in chapter 7. One of these is the injunction that, for acoustic reasons, the roof of the portico at the top of the auditorium shall be of the same height as the scene-building (v. 6. 4). The scene-building is never built so high as this in Hellenistic theaters, but the rule is often observed in Graeco-Roman and purely Roman theaters.177
Dörpfeld has advanced several other arguments bearing upon this problem,178 but in my opinion those just mentioned are sufficient. Now if Vitruvius’ Greek theater is to be identified with the Graeco-Roman structures dating from just before the beginning of the Christian era, it becomes impossible to cite Vitruvius in support of a stage or the use of the proscenium as a stage in Greek theaters of Hellenistic or earlier times. It will be necessary, therefore, to turn back to the fifth century and examine without prejudice the conflicting claims with reference to the presence or absence of a stage at that period. Our discussion of the extant theatrical remains of that century has already made it plain that there is nothing in them which can be employed to prove that there was a stage for the exclusive use of actors. But fortunately the paucity of such evidence is compensated for by the preservation of forty-odd tragedies and comedies of this period. A leading by-product of the stage controversy has been the recognition of the fact that these plays are not only to be taken into consideration together with other evidence but that they must be the final test of all theories based on evidence drawn from other sources. If a given theory will not permit these plays to be “staged” easily and naturally, that theory ipso facto falls to the ground. As von Wilamowitz wrote: “Von dem, was in den Stücken selbst steht, lässt sich nichts abdingen.”179 Whatever judgment may ultimately be formulated with respect to Dörpfeld’s contributions to scenic antiquities, one of his principal achievements must ever be recognized as the minute, searching, and unprejudiced re-examination of the plays themselves which he provoked.
An illuminating exemplification of the use that may be made of the plays in the study of such problems has been given by Professor Edward Capps.180 He showed that if chorus and actors be thought of as separated by a clearly marked line such as the edge of a ten-foot stage would afford, the action of the forty-four extant dramas requires the chorus alone to pass over this boundary at least sixty-eight times, the chorus and actors together nine times, and the actors alone thirty-nine times. Actors and chorus are repeatedly brought into the closest possible contact. For example, in Euripides’ Iphigenia among the Taurians, vss. 1068-70, Iphigenia appeals to each member of the chorus in turn, touching the hand of one and the chin and knees of another, begging for their help.
Again, the incidents of many plays come into harmony with theatrical conditions only if we suppose that there was no stage. Perhaps the best and clearest illustration of this is afforded by Aristophanes’ Frogs (405 B.C.). Xanthias and Dionysus, engaged in conversation, enter the orchestra at one of the side entrances (Fig. 44A). At vs. 35 the latter calls attention to the nearest of the three doors in the proscenium, saying: “I am already near this door where I must turn in.” It transpires that this is the house of Heracles (Fig. 44B), and Dionysus’ knock brings his brother in person to the door. From him they receive directions for their trip to the lower world—that first they will come to a large lake which they must cross in a tiny boat, then they will see perjurers, thieves, and criminals of the deepest dye, and finally will be received by happy bands of initiates (the chorus), who “dwell alongside the very road at the doors of Pluto” (vss. 162 f.). Scarcely have they left Heracles’ door when they behold a trundle-boat pushed from the opposite parodus into the orchestra (CC′) and hear Charon’s “Yo-heigh, Yo-ho” (vs. 180). He approaches the edge of the orchestra where they now stand, but when they prepare to embark Charon refuses to receive a slave on board and poor Xanthias is ordered to run around the lake (C′C″D; vs. 193). Meanwhile Dionysus and Charon direct their boat across the orchestra (C′D) to where, in the center of the front row of seats, the priest of Dionysus and other functionaries always sat (Fig. 45);181 and from behind the scenes, to accompany their rowing, the choreutae sing a “frog” chorus as if from the bottom of the lake (vss. 209-69). Upon disembarking (at D) Dionysus calls for his slave and catches his faint reply as he comes into sight (!) from his “arduous” trip around the orchestra’s semicircumference. Xanthias now points out to his master the perjurers, etc., in the nearby audience (vs. 275). Presently they are badly frightened and Dionysus appeals to his priest, who is within arm’s length of him, to protect him (vs. 297). Now the sound of flutes is heard and the chorus of initiates enter. Dionysus and Xanthias crouch down, where they are, to listen (vs. 315). Immediately the orchestra, which has just been a subterranean lake, is changed to the imagination into a flowery meadow (vss. 326, 351, etc.). At vs. 431 Dionysus starts up from his lurking-place and inquires of the chorus, “Could you tell us where Pluto dwells hereabouts?” and the coryphaeus promptly replies: “Know that you have come to the very door” (vs. 436). Dionysus orders his slave to pick up the baggage, walks across the orchestra (DE), and raps at the central door (E), which represents the palace of Pluto (vss. 460 ff.). We need continue no further, for the remainder of the play contains nothing that is noteworthy for our present purpose; but it is already evident how closely the successive situations of the comedy correspond to the physical conditions and arrangements of a stageless theater. To those who would apply Vitruvius’ account to the fifth-century theater, this play presents ineluctable difficulties; there is insufficient room for Charon’s boat on a Vitruvian or any other kind of a Greek stage, Dionysus must appeal to his priest who is some eighty feet away,182 Xanthias has no lake to run around, and Dionysus must inquire the way to Pluto’s palace when he would be standing considerably nearer to it than the chorus.
It was a convention in the earlier fifth-century plays that if the chorus and one actor were before the audience, an incoming actor should speak first to the chorus and ignore the other actor for the time being (see pp. 165 f., below). This convention was oftentimes extremely awkward and unnatural; but if both actors had stood on a stage several feet above the chorus it surely would have been altogether impossible.183
The only tangible argument for a stage of any height in the fifth century is afforded by the occurrence of the words ἀναβαίνειν (“to ascend”) in Aristophanes’ Acharnians (vs. 732), Knights (vs. 149), and Wasps (vs. 1342), and καταβαίνειν (“to descend”) in his Wasps (vs. 1514) and Women in Council (vs. 1152). All of these plays, except the last, were performed prior to Aristophanes’ Frogs, which we have already seen to be incapable of presentation in a staged theater. In my opinion, then, these words are best explained on the basis of the slight difference in level between the orchestra and the floor of the proscenium colonnade, which was probably elevated a step or two above the orchestra and was often used by the dramatic performers (see p. 68, above, and pp. 238 f., below).184 Since the Acharnians was produced in 425 B.C., the appearance of ἀναβαίνειν in that play is valuable as affording a terminus ante quem for the introduction of a wooden proscenium at Athens.
The chorus of the fifth-century plays is fatal to any suggestion of a Vitruvian stage, and except Puchstein, who frankly ignored the literary evidence, no recent writer has advocated a high stage for the theater of that period. The advocates of a high stage have clearly seen that they can make headway only by the sacrifice of the dramatic chorus. They are assisted in this attempt by the fact that only three complete plays of the fourth century are extant, the pseudo-Euripidean Rhesus and two comedies of Aristophanes, and that the rôle of the chorus in the latter happens to be curtailed. Aristotle,185 also, speaks of irrelevant embolima in the work of Agathon, who won his first victory in 416 B.C. From these facts it has been declared that at the close of the fifth century or early in the fourth the chorus was either given up altogether or “its functions were merely those of the modern band” or “of mere interlude-singers.” Accordingly, it has been argued that the actors at the end of the fifth century stood upon a low stage (which for the kind of plays then exhibited was only less impracticable than a Vitruvian stage) and that they were suddenly elevated to the full height of the proscenium before the close of the fourth century. It must be added that even among those who accept Dörpfeld’s theory for the fifth century there is a tendency to go over to Vitruvius for the period represented by the Lycurgus theater at Athens and by the theater at Epidaurus—the last quarter of the fourth century.186 So far as Vitruvius himself is involved in this, the matter has already been disposed of. The alleged disappearance or waning of the chorus, however, furnishes no better ground of support for pro-stage writers. To trace the history of the chorus in detail will not be feasible at this point.187 It will be sufficient to state that there is no reason to believe that the tragic chorus failed to participate in the action or to bear a respectable share of the spoken lines until Roman times. Even in New Comedy, in which the chorus is now known to have appeared only for the entr’actes, its on-coming is often used to motivate the withdrawal of the actors. Such a motivation could scarcely have become common if the actors stood so far above the choreutae as to be safe from their drunken words and acts.188
Another argument in favor of a stage has been drawn from the phrases ἐπὶ τῆς σκηνῆς and ἀπὸ τῆς σκηνῆς, which occur in two fourth-century authors, Aristotle and Demosthenes.189 It has been claimed that ἐπί “naturally means ’on’ and implies elevation” and that σκηνή means “stage.” If this exegesis were correct, there could be no doubt as to the presence of a stage in the fourth-century theater; but as a matter of fact neither claim is warranted. Everyone would concede that the primary, untechnical meaning of σκηνή is “hut” or “tent,” and that the word was applied to the scene-building, which was erected back of the orchestra and which came to be increasingly substantial in construction. Though the term acquired a variety of other theatrical meanings, I agree with those who maintain that at no period did it mean “stage” in classical Greek. It is manifestly impossible to discuss the matter here, but I shall presently have occasion to show that even in Pollux, who lived in the second century A.D., it had not gained this meaning (see p. 98, below). If σκηνή does not mean “stage,” it is unnecessary to argue that ἐπί does not mean “on,” for actors could speak from the porch or from between the columns of the proscenium, and so could be said to speak “from the scene-building” (ἀπὸ τῆς σκηνῆς) or to be standing “on the scene-building” (ἐπὶ τῆς σκηνῆς) without being “on top of the scene-building.” Just so the teachings of the Stoic philosophers are referred to as οἱ ἀπὸ τῆς στοᾶς λόγοι190 without any implication that the Stoics spoke from a platform, let alone from the top of the stoa. Nevertheless, it is a fact that ἐπί does not always mean “on.” For example, Diodorus and Plutarch both employ ἐπὶ σκηνῆς in a non-technical sense with reference to an occurrence “before” or “at the quarters” of a commander. And Lucian’s metamorphosed ass was mortified at being shown to be a thief and glutton “before his master” (ἐπὶ τοῦ δεσπότου)191—surely there was no superposition there. Such passages, however, come from later Greek, when the prepositions were less clear-cut in meaning, and it is better, as Professor Gildersleeve has suggested192, to “repose quietly on the phraseological use of ἐπί; ‘on the playhouse side’ is all the Dörpfeld theory demands.”
This being the theoretical situation with regard to the original meaning of ἐπὶ τῆς σκηνῆς, it is important to observe that already in its fourth-century usage the phrase was employed vaguely, often meaning little more than “in the theater” or “in a play.” In fact, in one Aristotelian passage, as frequently in later writers, it clearly includes both chorus and actors within its scope. “We ought, therefore, to represent the marvelous in tragedy, but in epic there is greater room for the improbable (by which the marvelous is most often brought to pass) on account of our not actually beholding the characters. For example, Achilles’ pursuit of Hector, if enacted in a play (ἐπὶ τῆς σκηνῆς), would appear absurd—the Greeks (οἱ μὲν) standing still instead of joining in the pursuit and Achilles (ὁ δ’) motioning them back—but in epic verse the absurdity escapes notice.”193 It is evident that Aristotle was thinking of Homer’s Iliad xxii, vss. 205 f.: “But Achilles shook his head to the people in refusal and did not permit them to cast their sharp weapons at Hector,” and was trying to show why a scene that was excellent in an epic could not be dramatized with success. In Homer there are two groups of characters: (a) Achilles and Hector, and (b) the Greek army. In Aristotle’s imaginary dramatization of the incident these groups are represented by the actors (ὁ δέ) and the chorus (οἱ μέν), respectively. Consequently, if σκηνή here means an elevated stage, chorus as well as actors must have stood thereon. Nor did the incongruity consist in the mere position of the chorus inactive in the orchestra and the actors running on the stage, but in the action itself, since the action is equally irrational in the epic (where orchestra and stage assuredly play no part) but is there more tolerable because the scene is not distinctly visualized. I do not insist upon σκηνή here meaning “play” or “performance,” though that is a frequent use and gives the indefinite sense required; but at least until this passage can be shown capable of another interpretation, believers in a stage cannot fairly cite Aristotle’s use of ἐπὶ τῆς σκηνῆς in support of their opinion.
But though ἐπὶ (ἀπὸ) τῆς σκηνῆς was broad enough to comprise both chorus and actors, it naturally did not always include them both. Particularly, if it were desired to distinguish between the two kinds of dramatic performers, since οἱ ἐπὶ (ἀπὸ) τῆς θυμέλης could be used of the dithyrambic choruses and other “thymelic” (i.e., orchestral) performers, and could not possibly be applied to the actors, that phrase would naturally be used to designate the dramatic chorus as well, and οἱ ἐπὶ (ἀπὸ) τῆς σκηνῆς would be used in the restricted sense for the actors alone, even in opposition to the dramatic chorus. This was especially common in the case of οἱ ἀπὸ σκηνῆς, doubtless because the scene-building was thought of as the home of the characters “from” which they came, as the choreutae, whether dramatic or dithyrambic, did not. Thus, a lyrical duet between the dramatic chorus and the actors (a commus—κομμός) is defined as a “dirge shared by the χοροῦ καὶ <τῶν> ἀπὸ σκηνῆς.”194 But neither the original meaning of ἐπὶ (ἀπὸ) τῆς σκηνῆς nor this secondary development which brought it into opposition to the thymelic performers and even to the dramatic choreutae presupposes a raised stage for the exclusive use of actors, still less requires that σκηνή should have meant “stage.”
Now οἱ ἐπὶ (ἀπὸ) τῆς σκηνῆς and οἱ ἐπὶ (ἀπὸ) τῆς θυμέλης are exactly equivalent to the more common expressions οἱ σκηνικοί and οἱ θυμελικοί. For example, Euripides is called both ὁ ἐπὶ τῆς σκηνῆς φιλόσοφος and philosophus scaenicus.195 The relationship is an obvious one, but is worth noting because one of Bethe’s pupils has made σκηνικός and θυμελικός the basis of an attempt to prove the existence of a stage in the fourth-century theater at Athens. But since the earlier expressions ἐπὶ (ἀπὸ) τῆς σκηνῆς and ἐπὶ (ἀπὸ) τῆς θυμέλης were used with the same distinctions of meaning but without presupposing a stage, there is obviously no need of one to explain the later expressions. Moreover, Dr. Frei is guilty of an egregious petitio principii: he first accepts Bethe’s hypothesis that the Lycurgus theater had a stage and consequently concludes that the distinction between σκηνικός and θυμελικός must be explained on the basis of difference in the place of performance there, and then uses these conclusions to prove a stage at that period.196 All attempts to forge a pro-stage argument out of any of these expressions must be pronounced a failure. But of course in the Roman era, after most Greek theaters had been provided with a raised stage, the differentiation between ἐπὶ (ἀπὸ) τῆς σκηνῆς and σκηνικός, on the one hand, and ἐπὶ (ἀπὸ) τῆς θυμέλης and θυμελικός, on the other, became doubly appropriate, because the difference in levels now reinforced a distinction which had already existed without it.
Vitruvius, of course, made no philological or archaeological study of the two adjectives but explained them in terms of the theater which was known to him (see pp. 76 f., above). It should be noted, however, that Vitruvius mentions only the tragic and comic actors under the term scaenici and includes under thymelici “the other artists” who perform in the orchestra. Does the dramatic chorus belong among the latter? Or is it simply ignored here? The answer is far from certain. If we were dealing only with new plays, it is conceivable that the choruses were so detached from the histrionic action as to be able to stand ten or twelve feet below the actors. But it is well known that some of the fifth-century tragedies were still popular and frequently acted; and as we have already seen, they were not amenable to any such method of staging. In revivals of early masterpieces, then, did all the performers, actors and chorus alike, appear in the orchestra, as in the old Greek theaters? Or was the chorus so reduced in size, and its manner of performance so altered, that it could stand with the actors on the high and narrow Graeco-Roman stage, as they all certainly did on the low and broad Roman stage? It is impossible to determine. All that can truthfully be said is that Vitruvius does not clearly indicate the place of the dramatic chorus in the Graeco-Roman theater. My own opinion is that he is speaking of two distinct types of performance and is ignoring the dramatic chorus.
The same question arises in connection with Pollux. He catalogues eleven parts of a theater. Of these, only six concern us at present: σκηνή, orchestra, logium, proscenium, parascenia, and hyposcenium (IV, 123). Dörpfeld thinks that Pollux is describing the Greek Hellenistic theater,197 but Pollux was for many years a professor at Athens and dedicated his work to the emperor Commodus (161-92 A.D.). Unless his language prevents it, it is more natural to suppose that he had the Athenian structure of his own day in mind, and this would be the Nero theater. In that case, every term falls into place. For the Nero theater logium could refer to the stage alone; and as there would be no sense in Pollux mentioning two words for stage, and since no other term for scene-building as a whole (including logium, proscenium, and parascenia) appears in his list, σκηνή must still mean scene-building and not stage. Pollux then proceeds to say that “the scene-building belongs to the actors and the orchestra to the chorus,” and a little later that “entering at the orchestra they mount to the scene-building on ladders (steps?).”198 Believing that Pollux is describing the Hellenistic theater, Dörpfeld interprets the first of these passages much as Aristotle’s use of ἐπὶ (ἀπὸ) τῆς σκηνῆς has just been explained. The second passage he considers a reference to some such unusual incident as occurs in Aristophanes’ Clouds, where an actor is bidden to climb (from the orchestra) by means of a ladder to the housetop (i.e., to the top of the scene-building) and destroy the roof.199 There is much merit in this explanation, and it is not necessarily inconsistent with a belief that Pollux is in general dealing with the contemporaneous theater; such learned digressions occur not infrequently in his text. Nevertheless, since stone steps leading from the orchestra to the stage of the scene-building are a part of the Phaedrus theater at Athens, it is not improbable that they belonged also to the Nero stage, if, as Dörpfeld first thought, this was only about six inches higher than the present stage (see p. 74, above). On the other hand, the pro-stage writers boldly cite these passages in support of their views and as if they pertained to the earlier periods of the theater’s history. But though Pollux is probably discussing a theater with a stage, σκηνή does not mean stage in these two sentences any more than in his catalogue of theater parts; and his testimony, however it is to be interpreted, should not be applied to fifth- and fourth-century conditions unless confirmatory evidence for so doing can be produced from these periods. Now the last of these sentences from Pollux concludes a discussion of the conventional significance of the parodi in the ancient theater (see p. 233, below). In my opinion, the Nero stage, though much deeper than the Hellenistic proscenium, was shallow enough so that the parodi still led directly into the orchestra. In that case, when the characters entered by either parodus, as they would when they were thought of as coming from the market place, harbor, or country, they would have to pass through the orchestra first and mount from there upon the stage by means of the steps, exactly as Pollux says. Furthermore, if actors could traverse this route it must have been available also for the chorus. In other words, although at this period the orchestra was the exclusive sphere of the dithyrambic choruses and other thymelic performers and was the normal place for the dramatic chorus, and though the actors regularly stood upon the stage, yet both the actors and the dramatic chorus appeared in either orchestra or stage according to the requirements of the plays. It must be understood, however, that this manner of staging was confined to the Nero theater at Athens; the stage of the Graeco-Roman theaters and the proscenium of the Hellenistic theaters were too high to make it feasible, and in the purely Roman theaters all performers appeared upon the stage. But why is it permissible to accept a low stage for the Nero theater and reject it for the fifth century? In the first place, the stage in Roman times is attested by incontrovertible evidence, both literary and archaeological, but for the fifth century it rests upon pure hypothesis. In the second place, there is no reason to believe that the Athenian chorus in Roman times was brought into actual contact with the tragic actors or had to pass to their place of action so frequently as in fifth-century drama (see p. 88, above).
There is still another sentence in Pollux which needs to be discussed. He declares that “the hyposcenium is adorned with columns and sculptured figures turned toward the audience, and it lies beneath (ὑπό) the logium.”200 There is no doubt as to the general position of the hyposcenium—it is the room201 immediately behind the orchestra and on the same level—but there is a division of opinion as to the type of theater which had one and as to its function. In accordance with his belief that Pollux is describing the Hellenistic theater, Dörpfeld understands it as the first story of the scene-building in a theater of this type.202 The columns and statuary would then refer to the proscenium just in front of it and to the figures which were sometimes placed in the intercolumniations thereof. In Hellenistic theaters Dörpfeld believes the top of the proscenium to have been used by speakers in the public assemblies and for that reason to have been known as a logium (see p. 59, n. 1, above); the hyposcenium, of course, lay on a lower level. Pollux’ statement could not refer to a theater with a stage because the wall beneath the front of the stage was not decorated with columns or statuary (see p. 86, above), the proscenium now being raised one story and appearing at the back of the stage. On the contrary, the pro-stage writers maintain that Pollux refers to the space under a stage. In this instance I agree with them as against Dörpfeld, though I would not look upon Pollux’ statement as applying to the theaters before his own day. Accepting Dörpfeld’s opinion that the Hellenistic theaters had no stage, I think that the first story of their scene-buildings had no special name and that the term “hyposcenium” had not yet come into use; Pollux, however, is referring to the space under the stage in the Nero theater. The front of this was probably adorned with the same frieze as now stands before the Phaedrus stage, and we may not dogmatically assert that no columns stood there as well.203 The Athens theater was inclined to be sui generis at all periods, and these would not be the only particulars in which the Nero theater differed from the Graeco-Roman type.
There remains for discussion a passage in Plutarch. It concerns an episode in the career of Demetrius Poliorcetes (337-283 B.C.) and has been thought to refer to the theater of his day. But a study has been made of Plutarch’s practice in such matters and it has been found that many times he deliberately sought vividness of presentation by modernizing his accounts and picturing his scenes amid the familiar surroundings of contemporaneous life; in other words, the references to the theater in connection with his anecdotes never presuppose any other type of building than the stage-equipped buildings of his own day, and in several instances this method resulted in patent anachronisms. One example will suffice.204 Plutarch declares that Lycurgus, the Spartan lawgiver of about the ninth century B.C., believed that the minds of assemblymen were distracted by “statues and paintings or the proscenia of theaters or the extravagantly wrought roofs of council chambers,” and so caused the Spartans to hold their assemblies in an open space. The author has here modernized his account in two particulars: he speaks as if Lycurgus were familiar with a fully developed theater building and as if it had already come to be used, elsewhere in Greece, as a place of meeting for the popular assembly. Of course, Lycurgus antedated the Greek drama and all but the crudest forms of choral performances by centuries, and this fact was as well known to Plutarch as it is to us.
Now Plutarch says205 that “Demetrius came into the city (Athens) and ordered the entire population to be assembled into the theater and hedged in the scene-building (σκηνήν) on every side with troops and surrounded the stage (λογεῖον) with guards, and himself descending (καταβάς), like the tragic actors, through the upper parodi (διὰ τῶν ἄνω παρόδων) he ended their fears with his very first words.” In my opinion, the word καταβάς (“descending”) clearly shows that λογεῖον means “stage.” The “upper parodi,” then, must be the passages opening upon the logium from the parascenia. As Plutarch visualized the scene and wished his readers to do so, Demetrius came out upon the stage from one of the side entrances but did not address the people from there, as an orator of Plutarch’s own day would have done.206 Instead, in his desire to show the Athenians his good-will he passed on down the central steps, as Plutarch had often seen the actors do in that theater (see p. 99, above), and addressed the assemblage from the orchestra. Since he could have passed through only one side entrance, the plural (παρόδων) must be due to a sort of zeugma, to imply that he came through one upper parodus and one upper entrance, viz., the central steps. The pro-stage writers who seek to apply Plutarch’s words to the Lycurgus theater in which the incident really happened, and who use them as an argument for a stage at that period, are forced to ignore the word καταβάς, for they cannot allow that “tragic actors” regularly descended from the Lycurgus proscenium into the orchestra. If we go back of Plutarch’s words and inquire what Demetrius actually did in the Lycurgus theater, the answer is plain: he simply advanced from the scene-building into the orchestra, and expressions consistent with this must have appeared in the source from which Plutarch derived his account. In fact, in describing a similar scene at Corinth, Plutarch retained words which are vague enough to be applicable to either type of theater.207 He has simply modernized one account and brought over the other unchanged.
The zenith of Attic drama had passed by, entirely for tragedy and almost so for comedy, before the remains of theaters outside of Athens become frequent.208 Nevertheless, these sometimes aid materially in reconstructing or interpreting the Athenian theater, and it will be necessary to dwell briefly upon a few of them. Perhaps the earliest and most primitive is found at Thoricus in southern Attica (Figs. 70 f.). This was built in the fifth or fourth century B.C. and was subsequently enlarged somewhat. The orchestra is oblong rather than circular, being bounded at one side by a temple, at the other side by a greenroom or storage chamber, and at the rear by a retaining wall. There is no reason to believe that a permanent scene-building was ever erected behind the orchestra. It is apparent that this structure has several points of resemblance to the Athenian theater of the period between ca. 499 B.C. and ca. 465 B.C. (see pp. 65 f., above).
The most symmetrical of all the Greek theaters and one of the best preserved is that at Epidaurus (Figs. 46-52 and 72, 2).209 Its architect was the younger Polyclitus, and it was built toward the close of the fourth century B.C. If we are right in believing that the proscenium was not used as a stage, then the Epidaurus theater never had a stage. At any rate, it was not rebuilt and provided with one in Roman times. In the center of the orchestra stands a block of stone with a circular cavity, doubtless the foundation of the thymele. There is not only space for the full circle of the orchestra (in the narrowest sense; see p. 83, n. 2) but the bounding stones are actually continued for the full distance. The stone proscenium, containing half-columns (Fig. 72, 2) of the Ionic order and once eleven feet seven inches or about twelve Roman feet in height, was erected in the second or third century B.C. and replaced a wooden proscenium. The parascenia were rebuilt at the same time and seem originally to have been broader and to have projected farther from the scene-building. In either parodus stood a handsome double gateway (Figs. 49 and 51 f.), one door of which led into the orchestra and the other opened upon a ramp, somewhat sharply inclined, which debouched on the top of the proscenium. Ramps are found also in the Sicyon theater.
THE THEATER AT EPIDAURUS
The theater at Eretria, on the west coast of Euboea, is not only one of the earliest but also presents several unusual features (Figs. 53-55 and 72).210 It falls into three periods. The old scene-building was erected early in the fourth century B.C. A later scene-building was erected in front of the other about 300 B.C. The white marble proscenium belongs to the first century B.C. or later. The precinct of Dionysus at Eretria was situated on level ground, and this fact necessitated different arrangements than were feasible on the usual hillside site. The highest ground in Fig. 55 shows the original level on which the first scene-building, orchestra, and auditorium were erected (Fig. 54). This scene-building was of the common type with projecting parascenia between which the proscenium must have been constructed of wood. The seats at this period apparently were wooden bleachers like the ἴκρια of the primitive orchestra in the old market place at Athens (see pp. 63 f., above); and when they proved unsatisfactory, it seemed easier to excavate the center of the area than to throw up a mound around it. Accordingly, earth to a depth of ten and a half feet was removed to form a new orchestra somewhat north of the old one. In order that the old scene-building might not have to be taken down or lose its serviceability, the earth just in front of it was left standing and was held in place by a retaining wall. Over this space was built a new scene-building, really only an episcenium. Communication between the old level and the new was secured by means of a vaulted passageway and stone steps. Before the retaining wall stood a wooden proscenium, the top of which doubtless continued the floor of the scene-buildings at the original ground level. The boundary of the orchestra (in the narrowest sense) stopped at the semicircumference, but there was sufficient room before the proscenium for the complete circle. A tunnel, six and a half feet high and three feet wide and with stone steps at either end, led from behind the proscenium to the center of the orchestra. Such an arrangement is probably what Pollux referred to as “Charon’s steps”211 and was convenient when an actor was to make an appearance from the earth or, like the ghost of Darius in Aeschylus’ Persians, from some structure which might temporarily be erected in the orchestra. Somewhat similar passages have been found in several other theaters, including Athens, but because of their size or other considerations seem not to have been used by actors. The downward pitch of the parodus, owing to the excavations, is clearly seen in Fig. 55. The marble proscenium is thought to have been about eleven and a half feet high and was supported by rimmed columns (Fig. 72, 1b). The parascenia did not project from this but merely continued the line of the proscenium, as in many of the Asia Minor theaters. Traces of tracks for the wheels of an eccyclema (see pp. 284 ff., below) are said to have been found in this theater on a level with the logium,212 but the stones have now disappeared and their purpose is not free from doubt.
Inscriptions in the island of Delos213 show that contractors received payment for a scene-building and proscenium in 290 B.C. Panels (πίνακες) for the proscenium are mentioned in 282 B.C. Wood for the “logium of the scene-building” was paid for in 279 B.C. Extensive repairs and improvements seem to have been carried through in 274 B.C. Stone was provided for the parascenium in 269 B.C. Wood was used for “panels for the logium” in 180 B.C. These were probably used to close large openings in the episcenium (see the θυρώματα at Oropus on p. 109, below). Most of these entries refer to wooden construction and antedate the extant remains in stone. There is no orchestra in the more restricted sense, but a gutter extends for about two-thirds of a circumference. If prolonged, this would just reach the front wall of the scene-building but would have a large segment subtended by the proscenium. The scene-building is an oblong with three doors in front and one in the rear. It is bounded on all four sides by a portico about nine and a third feet high. The front of this formed the proscenium, and it is clear that what was an ornament and certainly not a stage on the other three sides was primarily an ornament and certainly not a stage also on the fourth side. The oblong pillars, which were left plain on the other three sides of the building, on this side have their front surfaces rounded off into half-columns, and a vertical rim expedited the insertion of panels (Fig. 72, 3). There were no parascenia in the stone theater except as these were provided by the ends of the side porticos. The inscriptions, however, would seem to indicate that the situation had previously been different. From the front corners of the colonnade slanting doorways extended across the parodi. In the orchestra several bases stand in front of the proscenium, probably for the erection of statues or votive offerings.
There are theaters also at Delphi (Fig. 26), Megalopolis (Figs. 27 and 72, 1a), and Sicyon, but it is not possible to discuss every theater on the Greek mainland. We must not, however, pass by the small theater at Oropus in northern Attica (Figs. 56 f. and 72, 4).214 It stood in the precinct of Amphiaraus and dates from the first and second centuries B.C. The auditorium is almost completely destroyed; evidently the seats were always wooden bleachers. Five marble thrones, however, stand within the orchestra, an unusual arrangement which recurs at Priene (see p. 113, below). Another peculiarity is that no orchestra, in the narrowest sense, is marked out, either in whole or in part. But if a circle is drawn through the seats of honor, as has been done in Fig. 56, it falls just outside the proscenium. On the contrary, a circle as determined by the lowest row of seats cuts into the proscenium slightly. The parodi have been banked up so that their outer entrances are on a level with the top of the proscenium. The chief merit of this theater consists in the fact that the superior preservation of its scene-building and the presence of two inscriptions enable us to form a fairly clear picture of how a proscenium and an episcenium looked at this period. The front wall of the scene-building is pierced by one door; the side walls are continued so as to frame the proscenium but themselves turn sharply back along the parodi without forming projecting parascenia. The proscenium consisted of Doric half-columns and was eight and a quarter feet high. Its central intercolumniation was intended to be filled by a door, but the four on either side were so made as to be readily filled in with painted panels (Fig. 72, 4). Across the architrave ran an inscription: “... having been agonothete, dedicated the proscenium and the panels.” Another inscription ran along the top of the episcenium: “... having been priest, dedicated the scene-building and the doors.”215 The last item refers to five (or three) large openings in the front wall of the episcenium. Similar doors are found at Ephesus, and they were doubtless used in connection with the crane (μηχανή, see pp. 67 f., above, and p. 289, below). All in all, Oropus contributes very materially to our knowledge of the ancient theater.
Beginning with the first century B.C. the only kind of Greek theater which was newly built was what Dörpfeld calls the Graeco-Roman type, cf. the theaters at Termessus (Fig. 58)216 and Aspendus in Asia Minor. During this period several Hellenistic theaters (e.g., those at Priene, Magnesia, Tralles, Pergamum [Fig. 28], Athens [?], Syracuse, Pompeii, etc.) were remodeled to the Graeco-Roman type. That this is a Greek and not a Roman form of theater is proved by the fact that its orchestra, though no longer a complete circle, yet exceeded a semicircumference (see p. 77, above). These theaters had a stage varying from eight to ten feet in height and from eleven and a half to twenty in depth. The scene-buildings were of three stories—hyposcenium, logium, and theologium (Fig. 24). The first presented to the spectator an undecorated wall with doors leading into the orchestra; the second was terminated by a proscenium with columns and statues. The proscenium was seldom so simple as in the earlier theaters but was an ornamental façade with projections and recesses (Fig. 59), which added materially to the area of the stage.
Hellenistic theaters could be remodeled either (a) by building a new (undecorated) wall in front of the old proscenium and roofing the two over to form a stage or (b) by moving back the front wall of the scene-building slightly and constructing a stage between this and the old proscenium.217 In either case, a new (decorated) proscenium would be erected at the back of the stage. In the latter case, the columns of the old proscenium would either be removed and a blank surface built in their stead or they would be walled up. As already explained (see p. 86, above) this was done because the floor of the stage was thought of as representing earth or a street. At Priene (Fig. 64) the Hellenistic columns were left standing, but this is the sole instance of a Graeco-Roman hyposcenium having columns.
Method (a) is illustrated at Ephesus (Figs. 24 and 59-62),218 where the first permanent scene-building was built about 300 B.C. (Fig. 60). The dotted lines show the position of the stone proscenium, eight and a half feet high and nine feet ten inches deep, which was erected in the first century B.C. (Fig. 61). There were no parascenia. The seven openings (θυρώματα) in the episcenium furnish an interesting parallel to the five at Oropus (see p. 109, above). In the last half of the first century A.D. this structure was converted into a Graeco-Roman type (Figs. 24 and 62). The new logium was left of the same height as the old proscenium, but was made nearly twenty feet deep; and at certain points this depth received a considerable accession from the recesses of the new proscenium (Fig. 59). These changes were made at the expense of the orchestra, which derived some compensation from the fact that several rows of the lowest seats were removed; as a result the orchestra became a sort of pit (Fig. 24). The hyposcenium was plain and was pierced by three doors leading into the orchestra. The top story of the proscenium in Fig. 59 was not added until the third century A.D.
Method (b) was employed at Priene (Figs. 63 f.).219 This theater enjoys the distinction of being the only one in which an altar was found, and this was not situated in the center of the orchestra, as the foundations at Athens and Epidaurus would seem to indicate was the case there, but on its circumference. Seats of honor were placed in the orchestra, as at Oropus (see p. 108, above); but in Roman times new seats for dignitaries were erected in the center of the fifth row of seats (Fig. 63). The proscenium was of the same age as the scene-building and belongs to the third century B.C. At the Graeco-Roman rebuilding the columns of this proscenium were left standing, but the intercolumniations, except the three which served as doors, were walled up. The front wall of the Hellenistic episcenium was torn down and a new proscenium was built about six and a half feet farther back (see cross-hatched wall in Fig. 63).
The height of the Graeco-Roman stage as compared with the low Roman stage was partly due to convenience in remodeling when it was kept at the same figure as the earlier proscenium, but mostly to the conditions of exhibition.220 The Greeks did not, like the Romans, sit in their orchestras. Choral and musical competitions still were held there, as well as such Roman sports as gladiatorial and animal combats. It was necessary, therefore, that the orchestra should be accessible from the hyposcenium, and the doors could scarcely be lower than six and a half or seven feet. Accordingly, the stage could hardly be less than eight feet high.
But the seats of honor in Greek theaters had always been in the lowest tier (nearest the orchestra), and from there the view of dramatic performances, when presented upon an eight- or ten-foot stage, would be seriously obstructed.221 Usually when such theaters were remodeled, as at Ephesus, Assus, Pergamum, and Delphi, enough tiers were removed so that the lowest seats would be only about five feet below the stage level. The orchestra thus became like a pit and was inclosed with vertical walls (Fig. 24). At Side the space from which seats had been removed was built over with a six-foot wall, which was especially suitable in view of the gladiatorial and animal fights of Roman times. Where the auditorium was not altered, as at Priene and Magnesia, it is supposed that the lowest seats were unoccupied at dramatic performances, but were put to use, as the best places, at orchestral sports and contests.
As to the function of the dramatic chorus in the period of the Graeco-Roman theaters, especially in Asia Minor, we have little information. Nevertheless, it is necessary to consider the question. Already in Hellenistic (New) Comedy the chorus appeared only between acts (see p. 147, below). It is possible that by this time it had disappeared entirely or that it was so detached that, though the comic actors stood on a stage, the entr’actes could be given in the orchestra, or that its numbers were so reduced (see p. 135, below) that it could perform upon a Graeco-Roman stage—in any case, the chorus in contemporaneous comedy is negligible. The number of the tragic choreutae had probably been reduced also (see p. 134, below). But what is still more significant is that, if the fragments of Roman drama are any criterion222, the tragic choruses had abandoned the strophic responsions of the old Greek tragedy, and this means the abandonment of the complicated evolutions which had carried the chorus over the full expanse of the ancient orchestra. It was quite feasible for a small chorus which sang astrophic odes, spoke through its coryphaeus, and danced in a restricted fashion to appear upon a Graeco-Roman stage with the actors, to be closely connected with the plot, and even to participate in the action. As to the reproduction of old plays, the situation was not especially different. Fifth-century comedies were probably never repeated at this period. New Comedy, as we have just seen, would present little difficulty. As to old tragedies, the choral parts could be excised ad libitum or sung on the stage by a reduced chorus without dancing (or at least without evolutions). It will be remembered that I do not accept Dörpfeld’s opinion that the Nero stage at Athens was of the Graeco-Roman type. Accordingly, I believe that different physical conditions and the glory of their traditions kept up a livelier interest in the dramatic chorus at Athens than elsewhere and still retained the Athenian orchestra as the normal place of activity for the dramatic choreutae (see p. 99, above).
The foregoing account shows that there are many points of dispute with regard to the Greek theater and many points concerning which no one can do aught but guess. In closing, let me repeat that we are interested in the Greek theater mainly because of the Greek drama and that the extant pieces belong almost exclusively to the fifth century B.C. Now for that century the irreducible minimum, as shown by the plays themselves, is that there can have been no place, elevated much or little, which was reserved exclusively for the actors.
In the case of the drama the religious origin and the persisting religious meaning are self-evident. Performed at a festival of Dionysus, beside his temple, in the presence of his altar and his priest, tragedy and comedy are the natural response to that Greek demand for the enrichment of worship by art.—Arthur Fairbanks.