PA 3829 D5 1905 MAIN UC-NRLF B M D33 fl^D Zbc THn(v>erslti2 of Cbfcaflo FOUNDED BY JOHN D. ROCKEFELLER ES I THE IDLE ACTOR IN ^SCHYLUS A DISSERTATION [;UBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF ARTS AND LITERATURE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (department of THE GREEK LANGUAGE AND LITERATURE) FRANK W. DIGNAN ;;v^ CHICAGO THE UNIVERSITY OF CHICAGO PRESS 1905 Zbc Tanlverelts of Cbfcago FOUNDED BY JOHN D. ROCKEFELLER THE IDLE ACTOR IN ^SCHYLUS A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF ARTS AND LITERATURE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (department of the greek LANGUAGE AND LITERATURE) BY FRANK VV. DIGNAN CHICAGO THE UNIVERSITY OF CHICAGO PRESS 1905 CoPYItlGHT, 1905 Bv THE University of Chicago IN MEMORIAM CAROLI FRATRIS 159791 MA I ''^ I wish to acknowledge my deep indebtedness to Professor Paul Shorey and to the other members of the Greek Faculty of the University; and in a very special manner to Professor Edward Capps, without whose constant assistance this study would never have seen the light. THE IDLE ACTOR IN ^SCHYLUS The famous scene in the Frogs of Aristophanes, in which ^Eschylus and Euripides ridicule each other's methods, has been made the subject of many dissertations ; but one point — really the central point in the new poet's arraignment of the old — has not yet been investigated with the care which it seems to deserve, ^schylus is charged in general with bombast and in particular with aiming at the statuesque effect of a silent actor. Achilles and Niobe, says Euripides, sit silent through a large part of the play, in order to give an exaggerated effect to their words when they do speak, and these cases are taken as typical of ^schylus's method as con- trasted with that of Euripides. Such a charge seems reasonable enough at first sight in view of iEschylus's elevated and somewhat pompous tone, and accordingly the critics, ancient and modern, have accepted the allega- tion as substantially true. It seems time, however, now that our understanding of the material conditions and the course of development of early tragedy has been greatly enlarged, to reconsider the matter. May it not be that the fault in ^schylus's technique, if it really exists, is due to material limitations and to the restraints of tradition ? This is the question which I shall attempt to answer in the following study. The material at hand is, of course, far from complete, but some Hght should be thrown on the matter by the consideration of the evidence as to the lost plays referred to by Aristophanes, by the examination of the plays still extant, and by a com- parison wath the work of Sophocles and Euripides. It will be well to have before us the text of the passage in the Frogs: EY. TovTov oik Trpwr' iXey^w, oj? ^v dXa^cbv Koi cfiivai, OLOi<i T£ Tov? $€aTa<; 910 iirfTrdra, /Ltwpous Aa/3a)v Tra/aa ^/DWt;(<i) Tfyx(f>evTa<;. irpuiTUTTa fjikv yap eva tlv' av KaOlatv iyKa\viJ/as, A^^tXAea tlv' rj Nto/^T^v, to TrpoauiiTOV ov)(l SeiKvvs, Trp6<j-)^rjfxa ttJs rpaycpbtas, ypv^ovTas ovSe tovtL AI. jxa. Tov l^l' ov 8rjd\ EY. 6 8e xopos y' ijpeiSev 6pfJiu6ov<; av 915 fxeXwv €<^£^s T€TTapa<i ^we^ws av 01 8' eaiywv. AI. iyo} S' t)(aipov rrj (Tiodtjtj, Kat /xe tovt' erepirev oi)( T]TTOv rj vvv oi XaXovvTc?. 7 8 THE IDLE ACTOR IN ^SCHYLUS EY. ^Ai'^ios yap rjcrda, crd(f>' icrdi. AI. KafJLavTio boKw. tl Bk ravr' eSpacr' 6 Sctva; EY. vtt' dXa^ovEtas, Tv' 6 6€arr]<; TrpoaBoKiov Kadfjro, Q20 OTTod' 17 Nto/3r; Tt (fiOey^eTai • to Spa/xa 8' ak Sirjet. AI. w Trafj.iTOvrjpo<i, ol ap' €<f>evaKi^oprjV vir' avTov. Tt' (TKopOLva Koi Sva<l}op(.Z<; ; EY. OTi al'Tov Cs^Aey^oj. KaTTtir' CTTCiS^ ravra Xrjprjaeu Kal to Spapxi 77017 peaoLT). prjpaT' av (Soiui oouStK' etTrev, 925 6(f>px<; €;(Oi'Ta kui Ade^ovs, SetV arra poppopwTrd, dyvwra Tois 6€wp.ivoL<;. A little further on (948 ff.) Euripides contrasts his own method with this: EY. tTreir' ciTro roJi' irpwTwv i-jriLv oi'Skv TraprjK av dpyov, dAA' lAcyev 17 yvi*?; Tt /iot ;(0j SoGAo; ouSei' t;ttov, 950 X*^ he.(JTT6T7i<i \^ TTapdiva ^(7/ ypav<; av. ******* EY. cTTCtTa TovTovdl AaAciv eStSa^a — A IS. ^fjf'-i- xdyw. 955 *"^ TT/jtv OLod$ai y' u!<^cAes /itcros hiappayrjvaL. It is clear that the poet has in mind a general trait in the work of vEschylus as distinguished from that of Euripides; for the passage would be pointless unless the instances mentioned were typical. Stripped of its comic verbiage, what is the essence of the charge ? First, in general, that he imposed upon the audience by bombast, and then, more specifically, that he aroused expectation by the cheap device of a long silence on the part of an actor. That the silence, and not the sitting with veiled head, is the point of the criticism is shown by the contrast with Euripides's method — AoAcTi' eSiSa^a.* Later writers of antiquity seem to have taken the charge seriously, and there are several echoes of this passage which repeat its substance without the comic tone. The Ravennas scholiast on vs. 911 says: * hyiXX€v<; hi Ka6i]piv6<; iaTL Kai ovk aTroKpiv6pivo<; Trap' Alcr^vXw (v opdpaTi (Triypa(f)OptVio ^pvilv ^ 'ExTopos AvTpois. olBiv 8t 6 'A;^tAA£us <f>6€yytTat.' An additional ■ Cf. schol. ad Kan. 048: dpy6f uxrirfp <ri> ttjv yii^rjv Kal Tdv 'Ax'XX^a iTrolrjaas » Thr scnUncc is incomplete, as Beigk saw, Ilcrmcs, X\'III (1S83), p. 483, and the la( una is doubtless to be filled by reference to the Vita ^T\i)v iv apxa^i 6\lya irpbi 'EpfjLTJv ifioipaiay-, so Weeklein, .^sclt. Fab., p. 537. THE IDLE ACTOR IN .^SCHYLUS 9 note is found in the Venetus: "AAXws. eiko? t6v iv toTs ^pvilv 'AxtXAca r] "EktOjoos XvrpOL^ • rj tov iv MvpfJuBocnv, os P-^XP'- '''p'-^v rfp-tpixiv ovSev <j>6iyy€Tai.i Again, in the Vita Aeschyli: ware 8ia to irXtovd^iLv tw /3dpei Twv TTpoarwTTcav Kw/xwSetTai Trapa. 'ApLcrTO(f>dvov<; {'Api(TTO(f>dveL conj. Bcrgk). iv p.€V yap rrj 'Sto/Sr] <( , Nto^Sv; Bothe)' ews Tpirr^s y]jxipa<;'i inLKaO-qfJiivr] TiS Td(f>(o tS)v TraiSwv oiSev (jyOeyyerai, iTriKeaaXvpLfJiivr] • €v re toTs Ekto^os XvTpoi<; 'A;)((.AAevs o/xotws iyK€KaXvp.iJitvo<; ov <f)6iyyeTaL, ttXtjv iv dp;(at? oAtya Trpos 'Ep/x^i/ dixoL/SaZa. Schol. at/ Prom. 436 (Schiitz) : a-Lwirwcri Trapa Trotr/- rats Ta irpoawTra, rj 8t' ar^aStav, ok 'A;(tAAers iv rot? <^/3Utt 2o<^okA£Ovs (At(r;(uAor)* ^ Slol ttjv (Tvp(J30pav, cJs 17 Ntd/3r; Trap' Atcr^uAo); and the later note, a(f 440 (Dind.) : rj cny-q l^et ttoAAols /ne^dSot;? .... w? 17 l^iio/Sr) 8ia T^v vircp^dXXovaav Xvjrrjv iattoTra, kol o'lov to tov A^^tAAews, otc iaTdXrj(Tav Trpos e/celvov 6 TuA^u^ios Kai Eupv/3aTr/?, KuAo{}vTes ets p.d)(r]V, iaiyqcrev. Eustathius orf 0</. 1941, i: Trapa Ato-;(vAci) Kd6r)VTaL irov Trpoauyira cnwrrwvTa i<l>' iKavov Kara (T^^/xa rj TrevOov; r/ OavfiacTfXOv rj Ttvos inpoiov TrdOov; • koI eoiKcv rj TpaywSt'a evrev^ev Xa^ovcra to. TotavTa (T0<l>L^e(T6aL. See also 0(^ //. 1343. 59 ff- In modern times the matter has received little attention from scholars. Among the editors of Aristophanes, Bekker (1829), in commenting on the passage, discusses the place of aTrdTrj in the drama and remarks that Euripides has done the very thing that he blames ^schylus for doing, citing a few instances; Dindorf (1837) has a brief and commonplace note; Fritzsche (1845) discusses at some length the cases referred to by Aristoph- anes, but does not generalize; Kock (1876) quotes the passages from the grammarians, but adds nothing of importance; Blaydes (1889) besides the usual comments has a suggestion that the early prominence of the chorus had something to do with the matter; Van Leeuwen (1896) gives notes on the plays referred to and on Phrynichus, intimating that the silent actor was a natural consequence of the one-actor stage in the development of tragedy; but he does not consider the matter in general. The historians of Greek Hterature have nothing bearing on the point beyond a few words of comment on particular instances when very strik- 3 Hermann {Opusc, III, p. 42): "haec postrema aut hominis sunt indocti, qui quae de Niobe legisset ad Myrmidones transtulit, aut corrupta aliquot verborum omissione." Wecklein, loc. cit., gives the more sweeping opinion: "futilia sunt quae de Achille in Mvrmidonibus velato et taciturno et de legatis ad eum missis scholia recentiora habent," citing the Venetus note ad Ran. 911, and the second note ad Prom., quoted above. 4 The correction rplrov /xipovs, proposed by Victorius, is possibly right, though it was rejected by Hermann. Wecklein adopts it in his recension of the Vita, I, p. 467 of his edition of ^schylus (1885). lO THE IDLE ACTOR IN ^SCHYLUS ing. The same is true, in general, of writers on Greek tragedy. Hermann, in restoring the lost dramas, considers the instances mentioned (Opuscula, Vol. Ill, pp. 37 ff., and Vol. V, pp. 136 fif.); Haigh (Tragic Drama of the Greeks, p. 35) discusses the striking instance in the Suppliants, but does not generalize. The only writer who has even so much as made a collection of notable instances in the extant plays of .-Eschylus is Paul Girard (" L'expression des masques dans les drames d'Eschyle," Revue des eludes grecques, VIII (1895), pp. 118 ff., and pp. 102 ff. of the reprint), and he does it only incidentally to illustrate his remark: "Jamais ils [les poetes] n'ont recule devant les scenes de silence, et Eschyle, en particulier, semble les avoir multipliees de parti pris dans son theatre." He assumes in each case a deUberate use for artistic effect: "Ainsi, le mutisme, un mutisme pathetique, a bien 6t6, comme le lui reproche Euripidedans les Grenouilles, un de sesproced^s" (ibid., p. 109). We must now consider more carefully the accusation of Aristophanes and see what can be made out of the examples mentioned by him. The reference to Phrynichus in vs. 910 is important. s Aristophanes hints that the early history of the drama is in a measure responsible. The dramas of Phrynichus were of the older type, in which the choral element was much more important. Accustomed to this, the audience would be less impatient if an actor in /Eschylus's plays were silent. This, at least, is the meaning of Aristophanes. ^ As to the cases of Achilles and Niobe, not much is now to be had in the way of definite information. ^ The grammarians quoted above seem, in general, to know the plays, Ijut we cannot be certain of this. Nothing is known of the history of these pieces in later antiquity, except that the Myrmidons survived till the time of Accius, who made an imitation of it, probably in the latter part of the second century B. C* 5 Aristophanes nfers to Phrynichus as old-fashioned and as a favorite of the older ,\thenians of his time in Vesp. 220: fiivvpl^oyres m^Xtj ipxaia fie\iffidu>yo(l>pvyi- Xi)paTa (of the dicasts), and 269: x/jwtoj ijfiuv rjytiT' Slv ^Suv i>pvvlxov (of Phiiocleon). Aristophanes clearly approved of him; cf. Av. 750; Thesm. 164. ^ Merry, whom Blaydes quotes apparently with approval, writes this remarkable note: "After being accustomed to the usage of Phrynichus, the audience /f// they were being defrauded by the introduction of a mute person, instead of the actor who supplied the gist of the play, and the inspiration of the chorus." He misses the point entirely. /Eschylus was able to impose upon his audi.-nces, Euripides charges, because thiy were yMpoi, accustomed to nothing better by the leading tragic poets of the day. 7 See Hermann, Opusc, HI, pp. 37 ff., and V, pp. 136 t7.; Welcker, Die griechi- schen Tragddien, I, pp. 33 f. 8 See Ribb<ck, Ndmische Tragodie, pp. 340-55- Wecklein, however, dissents from this view, SitzungsbericlUe der bayrrischen Akademie zu Munchen, 1891, p. 327. THE IDLE ACTOR IN ^SCHYLUS II Apparently the Myrmidons, Nereides, and Phrygians formed a trilogy, in which the first play contained the reluctant consent of Achilles to the use of his troops by Patroclus and the la tier's death; the second, the slaying of Hector; the third, the visit of Priam to Achilles. Did the noteworthy silence of Achilles occur in the first play or in the third ? The Ravennas scholiast on Aristophanes and the earlier commentator on Prom. 440 say it was in the third; so says the writer in the Vita. On the other hand, the late scholiast to the Prometheus can refer only to the Myrmidons, and the Venetus scholiast on Aristophanes is in doubt. Unfortunately neither the fragments of the plays nor a comparison with those of Accius gives a clear indication as to where this scene of silence came. It may well have occurred in either; thus, it might express Achilles's pride and stubbornness when besought by the chorus of Myr- midons and the other Greeks to enter the battle, or it might be due to his grief for his comrade and his hatred of Priam. 9 The most natural explanation of the confusion is to suppose that something of the sort occurred in each drama; though the figure of the veiled Achilles, to which Aristophanes explicitly refers, must be assigned to the Phrygians. The trilogy, it may be added, probably belonged to the middle period of .^schylus, and may be classed with the Persians from the point of view of structure. It is fairly clear that there were but two actors. '° As to Niobe, the nature of her silence is clear. She sits, plunged in grief, at the tomb of her children and refuses to speak or unveil her head. If the corrected reading in the Vita be right, this silence lasts through a third of the piece. Before passing on to a consideration of the extant plays I wish to call attention to certain conditions which influenced the technique of iEschylus. 9 In either case it must have occurred in the first part of the play. This is borne out by the Aristophanes passage. At any rate, Achilles was present while the chorus rendered the parodos in both plays; c/. irsig. Myrtn. 131 (Nauck): rdSe /xiv Xei^crcrets, (paldifj.'' 'Ax'XXeO, 5opL\vfiAvTovs Aavauiv fjidxOovs, .... and a wag in Aristophanes apud Athen., 21 i. (Kock 678) remarks: Toiti ^pvyas oida decjpojv, Sre Tif. Ilpidfj.({! (rvWv(T6fi€voi tov iratS' '^Xdov redveu. ra, TfoXXd .... (j-x'»7/i'tciri(rai'Tas. 10 See Maurice Croiset, Revue des etudes grecques, VII (1894), p. 151, and Paul Girard, ibid., VIII (1895), p. 118, note 4. There are no indications as to the date of the Niobe. 12 THE IDLE ACTOR IN AESCHYLUS The theater at ^schylus's disposal was intended primarily for choral performances and not for the drama; as a result, the chorus was more at home there than the actors. The modem theater comes to a focus at the stage; the /^schylean theater focused at the center of the orchestra. The old circular dancing-place had followed a logical and inevitable course of development. Set on a hillside, it utilized the slope above as an audi- torium, which was prolonged on two sides by means of an embankment. The fourth side could not be so used, and here the orchestra terminated in a retaining-wall. On either side, just below the end of the auditorium, a passageway sloped up to the orchestra. Somewhere within the circle, probably at its center, there must have been an altar. This completed the arrangements for a choral performance. The drama at first required little more. A dressing-booth, somewhere in the vicinity but out of sight, was the first essential. Just where it was placed cannot now be told, but it was probably at the outer end of one of the side passages. With the addition of this dressing-booth, a drama of the earliest sort could be produced. The single actor could appear at intervals and, mounting the steps of the altar, dehver his brief speeches or take part in dialogues with the chorus-leader. This was the theater of Thespis." The earliest existing plays of /Eschylus show us a state of things not essentially different. There are two actors and a primitive stage-setting, but otherwise very little change in material and external conditions. The action still takes place in the orchestra, and an actor must still at every entrance come from the distant booth up through the parodos and over a considerable space in the orchestra before reaching his place. At every exit he must retrace the whole distance. The "setting" does not yet provide a more convenient door. It seems fairly clear that this early setting was an outgrowth of the original altar of Dionysus. At first this altar had been the center of action, but it was soon outgrown. It was a sacred object and could not be transformed into a tomb or enlarged by temporary boarding. More- over, it was i)rohably in the center, and the action gravitated toward the outer edge of the circle, where the actor could face the whole house at once, and where he would be nearest to the parodoi. So it seems to have become customary, if we may be j)erniitted to generalize on the basis of the small number of |)la\s from tliis ])crio(l wliich we possess, to erect a large altar- like structure, (loul)tlcss of boards, on the fartlicr side of the orchestra. " D6ri)fi'l(l-Rii.sth, Das griechischf Thentcr, \^\^. 2^-^b, 193-95, 366-69. THE IDLE ACTOR IN ^SCHYLUS I 3 It appears as an altar in the Suppliants and Septem, as a tomb in the Persians, and as a rock in the Prometheus.^ ^ This helped out the action by furnishing a scene, but the awkwardness of entrance and exit remained. It was not merely that so much time was con- sumed in coming and going; there was the difficulty of finding pretexts for such movements. A modern interior scene is the ideal of easy movement, for a slight pretext will take a character from one room to another; a scene before a palace or temple is almost as good; but an open place with no house in sight involves difficulties which are almost insuperable. If there were any doubt as to the reality of such a period in the history of the theater of Dionysus at Athens, the remains of the theater at Thoricus would be sufficient witness. In spite of its unusual shape, this little build- ing exhibits the essential features of the ^schylean theater. A flat, rec- tangular space on the hillside serves as the orchestra. The slope above has been formed into an elliptical auditorium with retaining-walls at the ends. On either side of the orchestra is a broad passage, the western one being partly occupied by a small temple. Outside the eastern passage are the ruins of another building, probably a storehouse for scenic prop- erties. The remaining side of the orchestra is entirely open and terminates in a retaining- wall. There is no trace of a scene building. '3 It was for such a building as this that the four earlier plays of ^Eschylus were written. Before the time of the Oresteia, however, a great advance had been made. A temporary building was erected along the outer edge of the orchestra and represented a habitation or temple connected with the story. '4 It was now possible for a character who was supposed to live in the palace or serve in the temple to withdraw with little or no expressed motive when his presence was no longer desired. Other characters would still use the side passages, but the management of their exits and entrances I* The most important discussions of the early ^schylean theater (beside the work mentioned in note ii) are: Wilamowitz, "Die Biihne des Aischylos," Hermes, XXI (i886), pp. 597-622; Todt, "Noch einmal die Btihne des Aischylos," Philologiis, XLVIII (1889), pp. 505-41; Bodensteiner, "Szenische Fragen," Jahrbilcher fiir classische Philologie, Suppl. XIX (1893), pp. 639 ff. ; Bethe, Prolegomena zur Geschichte des Theaters im Altertum, chap, v; Robert, "Zur Theaterfrage," Hermes, XXXII (1897), pp. 421 ff. 13 Papers of the American School at Athens, V (18S6-90), pp. 6-26; Dorpfeld- Reisch, op. cit., pp. 109-11. '4 The change must have been made about 465 B. C, for ^schylus's Myrmidons, Nereides, Phrygians, Memnon, Thracians, Lycurgeia, and Philoctetes apparently required a tent or other dwelling in the background, and all of these must have come before the Oresteia; on the other hand, Sophocles's Nausicaa and Polyxena probably required no such background. 14 THE IDLE ACTOR IX .ESCHYLUS also was rendered easier by the increased wealth of incident aflforded by the presence of such a building. I shall attempt to point out how this material fact of a primitive theater and the resulting difficulty in the arranging of a plot are largely responsible for the "idle actor" in ^^schylus. But first another factor should be men- tioned. Every reader of the Attic drama is struck by the strong tendency toward dialogue between two speakers. If three characters are present, one is apt to be neglected for a considerable time.'s The explanation of this convention is to be found, in part at least, in the histor\- of the drama. When the songs of the chorus first began to be diversified by short speeches, the purpose of these interludes was doubtless to give themes for the odes. A messenger announced an event, and then an appropriate song was sung. Dialogue must have begun with the questions of the chorus-leader and the messenger's answers. As the art developed, the speaker's functions were enlarged, especially after the addition of other actors,'^ but the old stiffness of set speeches and regular question and answer was not entirely outgrown. Nor indeed would the freedom of modem ensemble scenes have suited the elevated tone of Greek tragedy or the severity of Athenian taste. '7 To this should be added also the early importance of the chorus. The chorus is at first practically an actor. Though the part it plays is usually a passive one, convention requires that it should figure in all scenes. The old single actor had of course addressed the chorus, and the habit clung even when he might more naturally have conversed with his brother-actor. The inevitable result is that one of the characters is temporarily dropped from notice. '8 The difficulty of constructing scenes under these conditions appears most clearly in the Suppliants. ^^ The only setting is a Koivo/Jw/xia or altar of various gods (vss. 189, 222, etc.) with whose statues it is adorned (vss. 209-20, 463-65). To this altar both Danaus and the chorus, in their char- acter of supphants, frequently betake themselves (see 189, 208, 242, 713, 731, 832, 852). From his elevated position on the steps of the altar Danaus •s C/. Navarre, Dionysos, pp. 219 f. '* C/. .Aristotle, Poetics, 4. ■'C/. Frcytag, Technique of the Drama (English translation, iqoo), chap, ii, sec. iii. '* In e.xamining the plays, I have considered, in general, only the cases in which an actor is idle for twenty lines or more. By "idle" I mean neither speaking nor addressed. '9 The play is now gi^erally conceded to be the oldest e.xtant. See Gilbert, Rheinischea Museum, XW'III (1873), p. 480; Tucker's Introduction; Croiset, His- toire lie la litterature grecque, III', p. 173. THE IDLE ACTOR IN .ESCHYLUS 1 5 sees the king's party approaching, and afterward the ship, and on these same steps the chorus sits.^° The structure must therefore be of con- siderable size, even if the chorus consists of only twelve, and it is evidently not the altar of Dionysus in the center of the orchestra, but a temporary erection at the edge of the circle. ^^ The play opens with the entrance of the chorus through one of the side passages. It would seem that Danaus comes with them; for no reason is given for his remaining behind, whereas elsewhere he never leaves them except for some express purpose. Further, it is carefully arranged that he shall march out with his daughters at the end, and this suggests a similar effect at the beginning. Again, when he begins to speak after the ode, we find that he is upon the "mound" or altar (189) and has perceived the approach of the king. Apparently he enters with the chorus by the parodos and goes at once to the altar, while they take their position in the orchestra for the ode. This leaves him idle during a passage of 175 lines. ^^ After a short scene, in which Danaus calls the chorus to the altar, the king appears, and during the long scene which follows (the most important in the play) Danaus is entirely idle. It is noticeable that he seems to drop from the mind of the poet himself, for the king refers to the suppliants as a band of women (237), and he is mentioned only when the genealogy brings him inevitably to notice (3 18). ^3 When the scene is over, he is addressed by the king, but it is only to get him out of the way for the stasimon. The device is a transparent one, for the pretext on which he is removed is forced (480 ff.). 2° Cf. Capps, "The Stage in the Greek Theatre according to the Extant Dramas," Transactions of the American Philological Association, XXII (1891), pp. 36 and 70!. The references are to the Teubner texts (/^schylus, Weil, 1891; Sophocles, Dindorf, 1889; Euripides, Nauck, 1891). 21 See Bodensteiner, loc. cit., p. 648; Bethe, op. cit., pp. 90 ff., 95 ff. (Bethe, however, beUeves the erection to be the dressing-booth); Dorpfeld-Reisch, op. cit., pp. 195 f. For the old view see Schonborn, Skene der Hellenen, pp. 284 f., and Tucker's Introduction. " Capps, loc. cit., pp. 22 f. Tucker (Introduction) thinks that Danaus appears after the ode. Bodentseiner (p. 709) regards it as uncertain. 23 Maurice Croiset ("Le second acteur chez Eschyle," Memoires presentes 4 r Academic des inscriptions, X, pp. 193 ff.) explains such cases by saying that the part of the deuteragonist is still undeveloped. This is clearly inexact, for here the deuterag- onist speaks and the protagonist is silent. It is noticeable that most of the cases of neglect in ^Eschylus affect the protagonist. Richter {Dramaturgic des Aischylos, p. 122) says with reference to a later scene that if /Eschylus had had three actors he would have made Danaus enter with the king (so Gilbert, loc. cit.), forgetting that in this scene the poet is unable to keep two actors employed. " Auffallendcrweise " is Richter's only comment on this case (p. 112). 1 6 THE IDLE ACTOR IN .ESCHYLUS The king being likewise got out of the way and the chorus brought out into the orchestra (in defiance of all probability), the stasimon is rendered. Danaus then enters and announces the resolution of the people to protect them, and, no further pretext being at hand for removing him, he is left on the scene during the next ode. Here, as at the beginning of the play, he stands upon the altar steps, and immediately after the song he announces the approach of the pursuing ship. His withdrawal to get aid leaves the chorus free for another stasimon. After the scene between king and herald, he is called in to accompany his daughters to the town, and, having exhorted them to prudence, marches out with them while they sing. The exodus consists of 57 lines, and as Danaus has no part in the song, this must be added to the passages in which he is disregarded. ^^ It thus appears that the protagonist, in the character of Danaus, is left idle upon the scene for 175 + 246 + 85+ ? lines, or just about half the piece. Nowhere is any effect intended by the poet.^s The importance of the chorus gives it the lion's share of dialogue as well as of hxic parts; the conventional preference for two-part dialogue excludes Danaus from the chief scene; and, finally, the inconvenient theater makes it impossible to get him out of the way as often as is desired. Had there been a temple in the background, or a palace, the old man might have withdrawn at frequent intervals into that. But the town Hes at a distance, and only an important errand can take him thither. Even more distinct is the influence of the crude theater in the Persians. Had there been a palace in the background, the play would have run as smoothly as the Agamemnon. As it is, the characters (except Darius) must come from and go to a distance; and the moti\-ing is labored. The exact arrangement of the scene has been a matter of much dispute, but it is now generally agreed that there is but one structiu-e visible. This was of course the temporary erection in the background, and it seems to have represented throughout the play the tomb of Darius. Wilamowitz ^^ has argued for an imagined scene-change, the same setting representing suc- cessively the council-house, tomb, and road outside the town. But, to say nothing of the improbabihty of such an imagined shifting with so little to indicate it, we have direct evidence that the poet regarded at least the »4 The lini's have hrrn variously assigned, but Danaus seems clearly not to take part. Capps (p. 15) suggests that Danaus's silence is due to his being at the head of the procession, and hence being the first to disappear; but cf. the exodus of the Persians. »s These notable instances in the Suppliiints are not mentioned by Girard. >6"Dic Perscr des Aischylos," Hermes, XXXII (1897), pp. 382-9S. THE IDLE ACTOR IN ^SCHYLUS 1 7 beginning and middle of the play as having the same scene. On her first departure to procure the libations, the queen says: ij$<a XafSovcra TTcXavov ii oLKiov e/Awv (524); i. e., she is to return to the same spot to make the offering to Darius. If the first and second parts have the same scene, of course the third has also, since there is nothing to prevent Xerxes's meeting the chorus in the vicinity of the tomb. The only real difficulty is in the passage 140 ff., where the elders, after the opening ode, propose to hold a meeting t68' ive^ofxevoL oreyos dpx'nov. But it is easier to suppose either that they intend sitting on the steps of the tomb, or that they are referring to a building supposed to lie just outside the scene, than to believe that the same structure is pointed out to the audience as a senate- house and later is found to be the tomb of Darius. ^ 7 The scene is thus not essentially different from that in the Suppliants. The altar has merely become a tomb. It is true that the queen comes from, and retires to, her palace, but the palace being at a distance, it is hard for the poet to invent reasons for her going. Further, the chorus, though not now an important element in the story, must take part in each conversa- tion. Accordingly, Atossa is several times neglected. ^^ The first instance is at 249-89. The messenger at his first appearance, coming from the parodos, meets the chorus in the orchestra before he approaches the queen. It is easy to understand that the old convention, by which an actor addressed the chorus, would be especially strong at the first appearance of a messenger with important tidings. He accordingly converses with them, and the queen is unnoticed for forty Hnes. We have a regular scheme of two-line speeches by the messenger, interspersed with strophes and antistrophes by the chorus. Atossa would have disturbed this neat balance had she spoken. That the poet felt the awkwardness is shown by Atossa's first words, in which she apologizes for her silence: 27 C/. the hypothesis of the play; Todt, "Noch einmal die Biihne des Aischylos," Philologus, XLVIII (1889), pp. 515 ff.; Bodensteiner, loc. cit., pp. 648 f., 673 ff.; Bethe, op. cit., pp. 92 ff.; Dorpfeld-Reisch, op. cit., pp. 196 f. For the old theory of a palace background see Schonborn, op. cit., pp. igi ff. ; Oemichen, Bilhnenwesen, p. 185 (c/. Bodensteiner, p. 648); A. Miiller, Biihnenalterthiimer, pp. 113 and 116. Miiller later modified his view (see Philologus, Suppl. VI (1891-93), p. 16, note). Richter {Dramaturgic, pp. 103 ff.) is unable to decide. Cf. Jurenka, "Scenisrhes zu ^schylus' Persern," Wiener Studieji, XXIII (1901), pp. 213-25. 28 The essential parts of the story are four: (i) the queen tells her dream and her fears for her son; (2) the messenger reports the calamity at Salamis; (3) the spirit of Darius is summoned to give counsel; (4) Xer.xes appears, and the chorus join him in lamentation. The disturbing element, from the point of view of the dialogue, is the presence of the queen in parts 2 and 3. 1 8 THE IDLE ACTOR IN ^SCHYLUS (Tiyoi TToAat Svarrjvo'; iKTreir\€yfj.€vrj KOKOi; • vTTCp^aAAci yap rjSe crvfi<f>opd, TO p-'J/TC Ac'cui /liT/r' ipoiTTJcrai Tzadrj.'^ During the second stasimon, also, she is present and takes no part, but here a special device does away with the awkwardness: she pours the libations, and the ode is an invocation to the dead. This is the first of several cases in which an actor, present during an ode, is given some employment which prevents his being altogether idle. It is of course possible that in other cases, where no occupation is indicated in the te.xt, the actor could invent "busine.ss," as on the modem stage; for example, Danaus in the Siipplinnts, in the first two periods of silence, may have occupied himself in intent and anxious watching seaward from the steps of the altar. But the simplicity and dignity of an .^schylean tragedy must have made this of little practical use. 3° On the first appearance of Darius we have another case similar to that at the arrival of the messenger. Darius calls upon the chorus to explain why he is summoned ; they are too full of awe to reply, and he turns to Atossa. The first appeal is evidently in deference to the old convention, and the later change to Al^.ssa shows the poet's growing feeling for dialogue between actors. Atossa is thus neglected for twenty-three lines (681-703), though she is nearer the tomb than the chorus and Darius sees her almost at the start (684). 3' Again, after a dialogue with the queen, Darius turns to the chorus, and Atossa is idle for forty-five lines (787-831). This alternation is an attempt to keep all three in play, and it is certainly an improvement on the total neglect of Danaus in the chief scene of the Suppliants. But it is very far from being a sustained conversation among the three. 3» '9 MasquiTay, Theorie des jormes lyriques de la tragedie grecque, pp. 135 f., com- ments on the dramatic cfTcct, but docs not analyze the cause. So Girard, he. cit., p. iiq: "Atossa reste silencieuse, comme ecras^e sous le poids du malhcur." Prickard (note on 290) explains as due to the queen's dignity. 30 The actors of the time, it must be remembered, were little advanced beyond the stage of amateurs, and could not bo expected to furnish much by-play. Acting was not definitely recognized as a profession until the establishment by the state of the tragic actors' contest at the Dionysia in the year 450. 49. See CIA, II, 971, as reconstructed by Capps, Introduction of Comedy into the City Dionysia. And on the exact dale of the first actors' contest see ibid., p. 22, note 62. 31 Jurenka, lac. cit., p. 213, suggests that Darius does not at first see Atossa, because she stands so close to the tomb as to be really beneath him. It is difficult to imagine the poet composing the scene with such considerations in mind. J' It has often been noticed as strange that Atossa is not present in the l.isl scene, and variously explained (see Maurice Croiset, loc. cit., and Wilamowitz, /';> Pcrser THE IDLE ACTOR IN .ESCHYLUS 1 9 In the Septem play, although the setting is as primitive as ever, a number of circumstances make the management of the actors easier, and there is but one case of the idle actor. In the matter of the setting, it is interesting to observe hov^^ strong was the influence of convention. Accus- tomed to an altar as the sole or most important piece of scenery, the poet manages to introduce one here and to make it the resort of suppliants, though such a scene is not essential to the story. There is no evidence for other scenery, for Trvpyots aTreiXet roto-Se (549) and similar references to the fortifications may easily have been uttered without such being in sight. As in the Suppliants, there are images of the gods (94 ff ., 185, 21 1 f., 219 f., 265), and the chorus of maidens flee to them for protection (96 ff.) and are sent back to the orchestra for a stasimon (265 f.). There is no clear indication that an actor mounts the altar steps, but Eteocles's speech to the army (1-38) may have been delivered from that elevation. What is of chief importance here is that there is certainly no palace background. 33 The setting is thus identical with that in the Suppliants, but the nature of the plot gives several advantages for the management of the actors. In the first place, the chorus here is not the virtual protagonist. Its con- nection with the story is loose — so much so that the poet has great difficulty in keeping it occupied. This difficulty appears in the first episode (181- 286), where the only material at hand for the scene is the impatience of Eteocles at the outcry of the girls, and in the long scene between Eteocles and the messenger, in which their part is merely a few words of comment after the sending of each champion. With a chorus so reduced in impor- tance it was of course easier to keep the actors employed. A second advantage lies in the fact that there is but one important character, Eteocles. He can speak to the soldiers, to the chorus, or to the messenger, without danger of being left idle through the participation of other persons. Nor is the minor personage, the messenger, liable to be thus neglected, for messengers regularly depart unnoticed when their message is delivered. A third advantage arises directly from the situation. The scene is the citadel of a besieged city, and the characters are warriors engaged in the des Aischylos, pp. 386 f. and note); but the chief reason (that she would have dis- turbed the balance of one to one) has scarcely been noticed. ^Eschylus was doubtless influenced here by the traditions of the one-actor period, and particularly by the Phcenissa of Phrynichus. 33 See Todt, loc. cit., pp. 518 ff.; Capps, loc. cit., p. 37; Bodensteincr, loc. cit., pp. 649 f.; Bethe, op. cit., p. 94; Dorpfeld-Reisch, op. cit., pp. 197 f. For the old theory of a palace background see Schonborn, op. cit., pp. 125 ff. 20 THE IDLE ACTOR IN .ESCHYLUS defense. In such circumstances, exits and entrances are almost as easily motived as if a building stood in the background. In one case, however, Eteocles is left idle — namely, during the parodos. He has dismissed the soldiers, the messenger has made his report and gone, and Eteocles is praying to the gods, when the chorus rush in. They are full of terror at the prospect of an attack on the city, and their lamentations constitute the parodos — a passage of a hundred Hnes. Eteocles is silent and unnoticed throughout. It has been assumed by several editors 34 that he withdraws without remark just before the ode and returns at the close. If this were so, the case would be unique in the earUer plays. Nowhere else in the plays before the introduction of a back-scene does an important character, or indeed any character, depart and return again without a motive. Let us see what evidence there is for this case. The messenger has urged Eteocles to appoint defenders for the gates as soon as possible (57 f.), and if he goes at this point, it must be for that purpose. But it appears after the ode and his long argument with the chorus that he has not yet accompUshed the task, for he says (282 ff.) eyo) 8' iirdpxov<:35 .... rd^o} fioXwv. The errand is thus used as a pre- te.xt for removing him for the first stasimon. If /Eschylus had meant that the task was begun on the former occasion and interrupted, this would have been indicated here. We have, then, one period of idleness for the protagonist, and this due to the lack of a pretext for removing him. Later he goes to appoint the champions, and again to the battle; more than this the poet could not do without inventing a pretext for his departure that would have been more awkward than his presence without occupation. ^^ 34 As Palcy, Vcrrall, Flagg. Girard docs not notice this instance in the Septem. 35 Canter's conjecture for iir^ Avdpas. 36 The latter part of the play is probably a later addition. It is so inappropriate to the end of a trilogy that Welcker and K. O. Miiller (before the finding of the didas- calia) maintained that this could not have been the final play of a scries. The scene is doubtless an imitation of Sophocles, with whose play in mind the spectators could complete the story for themselves. Weil uses the long silence of the sisters as an argument against authenticity; but we must acknowledge, with Girard, that this objection has little weight in an early play of /Eschylus. See Bergk, Griechische Littcriilurgrsfhichte, III, pp. 303 f. ; Weil, "Traces de remaniement dans Eschyle," Revue dts eludes grecquts, I (188S), pp. 17 fT.; Wilamowitz, " Die Biihne des .Aischylos," Hermes, XXI (1886), p. 606, note 3. The authenticity of the passage is defended by Richter, op. cil., pp. 41 fT., and accepted by Girard, loc. cit. (last article), p. 120. The latest discus-sion of the question is by Wilamowitz, "Drei Schlussscenen griechischer Dramen," SitzungsbericlUe d. k. preuss. Akad. d. Wissenschajtcn, philos-hist. Klasse, XXI (1903), pp. 1-15. He cuts out vss. 861-73 ^"*i 1005 -end, and gives the inter- vening ode to the chorus. His argumentation seems to me conclusive. THE IDLE ACTOR IN .ESCHYLUS 21 Until something like agreement is reached as to the history of the text of the Protnethetis, any consideration of it from a technical standpoint must be purely tentative. The arguments for a revision, while not conclu- sive, have made it necessary to regard that as a distinct possibility. 3? And even if we have the play in its original form, there is still the uncertainty of date. The combination of early and late traits is at first disconcerting, but a careful weighing of the evidence on both sides certainly gives the impression that in the more essential matters the play belongs to ^schylus's earlier style. That so exalted a theme is combined wdth so rambhng a style, so episodical a plot, and so much geographical digression, surely shows the poet of the Septem rather than the poet of the Oresteia. The late characteristics must then be explained away or attributed to revision. The third actor, if employed at all, appears in only one scene, and may be regarded as an experiment which foreshadowed the later usage. As to the monodies by an actor, we have not sufficient evidence to assert that ^schylus might not have introduced them on occasion, even at an early period. The brevity of the choral parts may be due simply to the unimportance of the chorus and the supreme interest of the central figure. 38 The same factors may have influenced the metrical construction of these parts, for the consciousness that he was abbreviating them would naturally lead the poet to disregard many conventions. As to the elaborate machinery, there is not a passage in which its use can be absolutely proved from the text, unless it be the closing scene — and here the revision theory is most tempting. Why should Prometheus sink into the earth, w^hen he is required to be still bound to the rock at the beginning of the next piece ? On the other hand, Dorpfeld's researches in the theater of Dionysus have shown how easily the disappearance of Prometheus with the chorus might have been managed by taking advantage of the elevation of the rear part 37 Westphal, Prolegomena zu Aischylos, pp. 6 and 8; Rossbach-Westphal, Metrik der Griechen, II, p. :!^lviii; Bethe, op. cit., chap. ix. Both Westphal and Bethe lay- stress on the actor-monody, not found elsewhere in ^schylus, and Bethe argues alsO' from inconsistencies in the plot, metrical peculiarities, the use of machinery, and the nature of the conclusion. He seems to me to exaggerate all these difficulties, and especially to forget that the few plays which we have from this period furnish a totally inadequate basis for such generalizing. See also (on plot) KoUsch, "Der Prometheus des Aeschylus," etc., reviewed by Oberdick, Jenaer Litteraturzeitiing, 1876, No. 27, pp. 428 f.; (on metre) Wecklein's Introduction (pp. 25 fl. of English edition); Kramer, Pro?netheum vinctum esse jahulam correctam, pp. 34 ff., 39; Heidler, De compositione metrica Promethei fabiUae Aeschyleae cap. iv; Oberdick, Wochenschrift fiir klassische Philologie, V (1888), cols. i3iof. ; Masqueray, op. cit., pp. 34 if., 165, 270 f. Weil (loc. cit., pp. 21 ff.) thinks the e\'idence insufficient. 38 C}. Masqueray, op. cit., p. 79. 22 THE IDLE ACTOR IN ^SCHYLUS of the orchestra above the level of the ground in front of the temple of Dionysus. It has been well said that the distinct and reiterated descrip- tions of the chorus's wagon and Oceanus's steed as flying may be due to the fact that they </o not lly; consequently the imagination is helped out by verbal description — a well-recognized .T^schylean device. 39 A very simple setting is in reality sufficient on this view of the case. The temporar}' erection at the edge of the orchestra now represents a rock, and nothing more is necessary. All entrances and e.xits are made through the parodoi."" We come, then, to a consideration of the management of the actors. The situation is of course e.xceptional — Prometheus must remain through- out; but in various ways the awkwardness of his presence during choral passages is rendered less noticeable. Thus the parodos takes the form of a kommos, he and the chorus conversing in much the same strain as in the trimeters that follow. The first stasimon is chiefly addressed to Pro- metheus, and after it he apologizes for his silence as Atossa does (436 ff.). The second stasimon begins with general reflections, but the chorus quickly turns to address the central figure. The third stasimon is the only lyric pa.ssage in which Prometheus is entirely neglected, and it is verj' short (887-906). The exodus, like the parodos, is in the form of a kommos. Except for Prometheus, the play has the early episodical character, each scene beginning with the entrance of a new character and ending with his departure, so that none remains through the odes. In the dialogue passages there is shown the same desire to avoid the neglect of an actor, and in general the same success. Leaving out of sight for the moment the opening scene, we find that neither Prometheus nor a subordinate character is left idle at any point. Prometheus converses successively with the chorus, with Oceanus, with the chorus again, with lo, and with Hermes. Generally the chorus is neglected, but in the lo- scene great efforts are made to keep it in play, and we see how difficult for the poet was a three-cornered dialogue. The poet was accustomed to cope with the problem of the immovable chorus; the addition of an immovable actor raised a problem of exceptional difl"iculty. It was 39 Cf. Bodcnstciner, he. cit., pp. 665 f. ; Capps, "Stage in the Greek Theatre," pp. 19 f. *°Cf. C. Fr. Miiller, liymnasialprograni, Stade, Austria, 1S71, rovieweil in Philologisciier Anzeiger, III (1871), pp. 318 ff.; Wecklein's Introductinn; Capps, "Stage in the Greek Theatre," pp. $q f.; Bethc, op. cit., pp. 94 f.; Todt, loc. cit., pp. 520 fl.; Dbrpfeld-Reisch, op. cit., pp. igS f., 216 ff. Reisch's "Felzweg" for the chorus seems unnecessary, for w4Sot 5i pacai (vs. 272) may mean simply "alight," and aTv<t>\ov irirpai (748) may refer to the whole orchestra. THE IDLE ACTOR IN ^SCHYLUS 23 inevitable that the poet should find a partial solution in the reduction of the part of the chorus, both as to the amount of lyric assigned to it and as to its participation in the dialogue. It has seemed worth while to show in some detail the care with which the actors are in general kept occupied, because in this way the pecuhar nature of the first scene is emphasized by contrast. In this scene of eighty- seven lines, during which Prometheus is nailed to the rock, he utters not a word; the conversation is between Hephaestus and Cratos. Was this done deliberately for artistic efifect, as is often said, or was it the result of practical limitations ? The question seems decided by the fact that neither in the scene itself nor in the monologue that follows is there any reference to this silence as a sign of Prometheus 's pride. To plan such an effect and carry it out without calling attention to it by expHcit mention might accord well enough with modern methods, but is absolutely un- yEschylean, not to say un- Greek. There will always be a certain number of critics who will regard it as profanation to seek practical reasons for things where artistic work is concerned. But it is plainly illogical to admit a practical cause in one case and deny it in another merely because the same phenomenon has now an artistic significance. If Danaus, Atossa, and Eteocles are left idle only when the poet could not keep them employed, why not Prometheus also? Just what the difficulty was in this case cannot be told with certainty until the vexed question is decided whether we have here two actors or three. If the former is true, and Prometheus is represented by a lay figure, the reason is not difficult to see. Cratos and Bia are needed to carry the figure in, and the part of Cratos is given to an actor that conver- sation may begin immediately on their appearance. Prometheus of course cannot speak until the scene is ended and the protagonist has taken up his position behind the figure. It is true that the lay-figure theory has been a favorite object of ridicule with the critics who pride themselves on taking a common-sense view of such questions; but the common-sense attitude is too apt to involve the ignoring of conditions. In the great theater of Dionysus a wooden Pro- metheus, nude, fettered, and of superhuman size, may well have been more impressive than a masked and padded actor in the same position. 4» 4' The lay-figure theory was first suggested by Welcker, Trilogie, p. 30, and has been approved by G. Hermann, Opusc, II, p. 146; K. F. Hermann, De distrib., pp. 623, o; Wieseler, Gott. prorect. Program, 1866, p. 5; A. Miiller, Philologus, XXIII (1866), pp. 5195., XXXV, p. 312; Philologischer Anzeiger, III, p. 319; Wecklein, 24 THE IDLE ACTOR IN iESCHYLUS On the other side it is urged (i) that the idea of a lay- figure was first suggested solely because the Prometheus was supposed to belong to the two-actor period, a thing which we have no right to assume; and (2) that the ancients nowhere mention the use of a lay-figure. But if, on the other hand, three actors were employed, the reason for Prometheus's silence is equally easy to detect. The third actor was an experiment, an innovation, and the convention of two-part dialogue could not be overcome. Three actors might appear at once, but a general con- versation among them was against all the traditions of the drama. The writer is inclined, however, to the view that a lay-figure was employed. In any event, the effect upon the audience was the same, and Prometheus must be reckoned as an idle actor in this scene. The Agamemnon and the two following plays are sharply distingmshed from the preceding four by the existence of a back-scene with doors. It is scarcely possible to overestimate the importance of such a change. A mere piece of scenery (altar, tomb, or rock) is of little assistance in the arranging of a plot. It serves to indicate the locality and add \-ividness to the action, but it gives little help in motiving the comings and goings of the characters. But when once the idea is grasped of making the action transpire before a palace, a temple, a tent, or even a cave, new possibilities are opened for the drama. The name (tktjvt^ is good evidence that this back-scene was not developed from the older setting, but from the dressing-booth. The altar or tomb can never have been called o-kt^vt;. That word suggests a real building, slight indeed, but meant as a cover or protection — meant to contain some- thing, and not as a bit of idle show. We are justified in assuming that this new structure on the edge of the orchestra was from the first a retiring- place for the actors — merely the dressing-booth in a new situation. 4^ The Studien zu Aischylos, pp. 31 S., and ed. of i8()6, Introduction, pp. 54 f.; O. Navarre, "De I'hypoth^sc d'un mannequin dans le Promethec enchaine d'Eschyle," Anrwles de la FacuUe des Leilres de Bordeaux, Revue des itudes anciennes, III (1901), 2. The view is rejected by Schoemann, Prometheus, p. 87; Soinmerbrodt, Scaenica, pp. 170 ff.; Girard, loc. cit., p. 123 (of last article), note 5; Croiset, Histoire de la liiteratur grccque III», p. 176; C. Fr. Muller, loc. cit.; Richtcr, op. cit., pp. 50 f.; Bcthc, op. cit., p. 180, note; Bodensteiner, loc. cit., p. 674. The arguments for the view are summed up as follows by Navarre: (i) absence of any sign of movement in Prometheus; (2) his silence under torture; (3) unnecessary brutality, the iron being driven through his l)ody; (4) probability that only two actors are used in the pilay; (5) arrangement of the scene, Cratos remaining behind as if to give the other actor time to take his place behind the figure. 4' Bethe believes that the earlier altar or tomb was likewise the dressing-lxwth. On this view the innovation would be merely a new use of the booth, not a changing of its position. See op. cit., p. iqo. THE IDLE ACTOR IN ^SCHYLUS 25 reason for the change, then, was not a desire for more elaborate scenery, for that would have led to a development of the older setting; it was the need of a more convenient place of withdrawal for the characters. That the device proved satisfactory is shown by the regularity with which it is henceforth employed. Wlien we reach the period of the New Comedy, we find that the ideal of harmony between the arrangements of the theater, on the one hand, and the requirements of the drama, on the other, is at last attained; characters make their exits freely and without motiving when their presence is no longer needed by the poet, as Leo has clearly shown in his Plautinische Forschnngen. The first ctkt/vt; was a simple, temporary structure of wood containing one or more doors, as the piece required. It probably extended on either side as far as the parodoi, so that actors could pass through or behind it unseen. The action was still in the orchestra. The actors changed their costumes in the building, and either entered directly through its doors or passed around to the side and came in through a parodos. They might also on occasion appear upon the roof. 43 The immense advantage of this new arrangement is at once evident. Any character supposed to hve in the building could now come and go with Uttle or no motive or remark. Not being imagined to have gone to a distance, he might reappear as quickly as desired. 44 A comparison of the management of Atossa in the Persians and Clyte- mestra in the Agamemnon will serve to show the difference. The circum- stances of the two plays are very similar; in each we have a queen, a chorus of elders, an absent king, a herald announcing his coming, and the king's appearance. In the older play the queen arrives on a chariot from her distant palace. She must be present through the scene with the herald, and hence the difficulty of arranging that scene. Twice she is explicitly sent off on clumsy pretexts, and finally she does not meet her son when he appears. Clytemestra, on the other hand, after her first scene with the chorus, withdraws without remark. During the herald-scene she is absent, except for a few moments in which she explains why she need not hear the mes- sage. When the king arrives, he is received by the chorus alone, and only then does the queen_ leave her palace to utter her greetings without inter- ference from them. 43 See Wilamowitz, "Die Biihne des Aischylos," Hermes, XXI, pp. 597 ff.; Bodensteiner, loc. cit., p. 645; Dorpfeld-Reisch, op. cit., pp. 199 ff., 370 ff., and Part V. 44 Dorpfeld-Reisch, op. cit., pp. 201, 371 f. 26 THE IDLE ACTOR IX -ESCHYLUS The conditions are thus greatly changed since the time of the Suppli- ants. When the poet now fails to keep his characters employed, the reason must generally be sought, not in the primitive theater, but in other condi- tions, often less tangible and less easy to trace. The movements of the queen are now so easily made that they are but vaguely indicated in the text, and at times it cannot be told with certainty whether she is present or not. The most puzzling case is in the parodos. The anapaestic portion of the ode concludes with an address to the queen: (TV Se, Ti'v8ap€a> dvyartp, /iacrt'Atta KXvTai/xrjcrTpa, Tt \pio^ ; TL viov ; k.t.K. (S3- 103). One naturally expects her to reply forthwith, but instead we have a hnric passage of 160 lines, after which she is again addressed and finally speaks. There is something grotesque in the picture of the queen standing idly by all this time, quite without reason, and the detailed description of the slaughter of Iphigenia is in singularly bad taste if the mother is present. It is more natural to suppose, either that she appears only for a moment (as in the herald-scene), or that she is not present at all until the end, the first address being merely an apostrophe made with lyric freedom. ^s There occurs in this play, however, the most famous case of the idle actor in .^schylus — the long silence of Cassandra. When Agamemnon makes his triumphal entry on the chariot, Cassandra as his captive naturally rides in another chariot following him. She is unnoticed, however, for 168 Hnes; then Agamemnon mentions her, but she is not addressed for 85 lines more; and then she is obstinately silent for another 37 lines. In all she is silent for 290 lines. In this long interval occur the greeting of the chorus, Agamemnon's speech, the scene with Clytemestra, and a stasimon. Much praise has been lavished upon this long silence, as increasing the effectiveness of the mystical prophecy which follows,-"^ but critics have uni- formly lost sight of the fact that the greater part of it was forced upon the poet by the conditions. Cassandra, if introduced at all, must enter in the train of the conqueror. She could not take part in the dialogue until the greetings were over and a stasimon had j)rcparcd the way for a new scene. The skill of the poet is shown, not in inventing the silence as a •♦■; The editors disagree. Werklein, Enger, Vcrrall, and Gilbert think the queen is nuTL-ly aj)ostro|)hi/.((i; Hermann says that she enters only for a moment; Klausen, Karsten, Peile, an<l SidgNsick make her remain through the ode, busy with the offer- ings. So Capps, "Stage in the Greek Theatre," p. 23; Bodensteiner, loc. cit., p. 731 ; DetschetT, Pe tragordiarum graecarum conjormationc, Sardicae (1904), note loi. Arnoldt, Chor im Agamemnon, pp. q f., thinks the queen's silence shows her pride. 46 £. g., Richter, of), cit., p. 165; Girard, loc. cit., p. 124. THE IDLE ACTOR IN ^SCHYLUS 27 dramatic device, but in accepting it as unavoidable and turning it to brilliant effect. 47 Cho'ephoroi. — The Cho'ephoroi, though brought out at the same time with the Agamemnon, is in some respects strikingly different from that play. Whether because of the inferior interest of the subject, or owing to acci- dental circumstances connected with its composition which we cannot now trace, the Cho'ephoroi, regarded as a piece of workmanship, is much less impressive than the Agamemnon. In the close-knit structure and con- stant intensity that give significance to every detail, the play is somewhat lacking. A general air of looseness and remoteness pervades it, not unlike what may be observed in Shakespeare's Macbeth as compared with Othello. This appears particularly in the treatment of two characters, Electra and Pylades. Electra enters with the chorus (f/. 16) and is silent during the parodos. She has come with the company of serving-women to pour libations on her father's grave. At the conclusion of the song she asks their advice as to the prayer she is to offer. This period of silence, like Cassandra's in the Agamemnon, is really forced upon the poet by circumstances. Electra could not well remain behind, since the chorus's entrance would then be unmotived, and, on the other hand, the ode could not be turned into a kommos, for an elaborate kommos with Orestes is to occupy a large part of the piece. 48 The neglect with which Electra is treated appears more distinctly later. In the long kommos, with its succession of prayers by Orestes, Electra, and the chorus, each party is of course left idle in turn, but Electra is neglected more than either of the others. There follows a scene in trimeters, the greater part of which passes between Orestes and the chorus only, Orestes asking and learning of them the reason of the offering. In the scene where Orestes announces to his mother his own supposed death, there is a speech (691-99) which may belong to either Clytemestra or Electra. 49 If Electra is present, she is unnoticed by the other characters during the scene. Henceforth she does not appear at all. 47 Girard, loc. ci/., pp. 1 16 f., describes the ludicrous effect of Cassandra's tragic mask during this long silence when the play was produced at Paris. It is not unlikely that in the original production the mask was concealed by a veil at this point. 48 Richter {op. ciL, p. 215) thinks she is busy with the offerings, but in four vase- paintings, which seem to have been influenced by this scene, Electra sits on the steps of the altar. See Huddilston, Creek Tragedy in the Light of the Vase-Paintings, chap. iii. 45 Bodensteiner {loc. cit., p. 733) thinks Electra is not present. The speech is given by most editors — as Hermann, Dindorf, Paley, Conington, Sidgwick, Wecklein, Verrall — to Clytemestra; Weil, however, and some others, give it to Electra — Robor- tellus to a .servant, Wellauer to one of the chorus. 28 THE IDLE ACTOR IN ^SCrnXUS In Pylades we have a puzzling phenomenon. Apparently he is always at Orestes's side, yet he speaks but three lines in the play. He is repeatedly referred to (20, 561 f., 583 f., 899), but his only words are a reminder to Orestes of Apollo's command (900-902). Is he a regular actor, or a mute who is given a few words to speak ? If the former, why is he so persistently silent; and if the latter, why was the part not adapted to a regular actor, since without him the three actors are used little, if at all ? Whether actor or mute, he must of course produce the effect of a silent actor, since he must wear the costume of a principal character. Apparently the explanation lies in a variety of circumstances. In the first place, as Wilamowitz has shown, ^Eschylus uses Pylades, not because he wanted the character, but because tradition placed him at Orestes's side, and he could not be got rid of. At the same time, he did not choose to encumber his play by the addition of another active personality. It still remains to be explained how, even so, he could be willing to allow this awkwardly silent figure in the play. The reason seems to be this: The sensitiveness on this point often shown in the earlier pieces was due to the conspicuous position which the actors occcupied. The characters were few ; often but a single one appeared in an entire scene. Not many attendants were required, and the scenery was of the simplest description. Under these conditions the attention was strongly focused on the actor. Even when the plot did not require any activity on his part, it would be very noticeable if he were neglected. Hence the poet soon learned to avoid such absurdities as the long silence of Danaus in the Suppliants. For example, the messenger in the Septetn, though a minor character, is never left idle on the scene. Here the condi- tions are different. We have many characters, an ample setting, an elabo- rate plot. The silence of Pylades, though still awkward, would not attract the same attention as at an earlier time.s° Eumenidcs. — No reader of ^schylus can fail to be struck by the differ- ence between the Eiimcnides and the other plays. The remoteness and mysticism are replaced by a matter-of-fact tone, and the drama reads almost like a portrayal of contemporary life. In the trial scene the gods and heroes speak and act like Athenians of the poet's own time. 5' It is easy 50 Vcrrall (Introduction, pp. .xvii ff.) says that Pylades is introduced partly to explain the situation (he being the influential friend who makes Orestes's schemes practicable), and partly as an impressive embodiment of the divine command. There seems .scant evidence in the play for either. S' There may have been a trial scene in the Diiuiiiilcs, hut it was cortaiiiiy not elaborately realistic like this one. See Hermann, Of>iisc., II, pp. 319 IT. ; Nauck, Trui^. Cracc. Frog., /Esch., 44. THE IDLE ACTOR IN ^,SCHYLUS 29 to see that such a development of the drama was inevitable. The poet would be certain sooner or later to pass from mere allusions to whole scenes of practical political import. The significant thing for us in this change is its effect upon the technique. If, for example, the ancient authority of the court of the Areopagus is to be pictured in a trial-scene, evidently the con- ventional method of the drama must be modified. The set dialogues between two characters, or a character and the chorus, will not be sufficient for such a scene. A considerable number of persons must be present, and they cannot all be kept constantly in play. The attention will be fixed on the scene as a whole, rather than on individual characters. The poet will be forced into something very like a modern ensemble scene. This is what we actually find in the Eumenides. In the trial scene the old conventionalities are disregarded to an extent that would be surprising if the reason were not so plain. The drama here comes into close contact with contemporary life: set to imitate a scene in the court, it must adapt itself to the new conditions. 5^ Thus we find several noticeable periods of silence in various characters, for which the trial is mainly responsible. During the epiparodos (244-63) Orestes remains cHnging to the statue of Athene {cj. 258 f.). He keeps this position until the end of the trial, being silent during the first stasimon (307-96), a dialogue of 39 lines between Athene and the chorus (397-435), the second stasimon (490-565), ^3 the opening of the trial (566-84), and the greater part of the trial (614-743). As in the case of Cassandra in the Agamemnon, the silence is turned to account at one point. When his case is about to be decided, Orestes, after a silence of 130 lines, breaks out in the cry : w $oty8' "AttoAAov, ttw? dywv Kpi6rj<T(.Tai ; The dramatic effect of the appeal must have been greatly heightened by the previous silence. Athene is silent during a large part of the trial (585-673, 711-33). As presiding officer, she must listen quietly to the presentation of the case. Apollo also is neglected while Athene establishes the court (681-710). It must be remembered that at the time of the Oresteia the problem of plot-building had been enlarged by the addition of the third actor. It is true that the poet was under no compulsion to have all his actors visible at one time, but circumstances would occasionally make such a scene necessary; and then the difficulty of keeping all the actors employed was of course increased. The cases in the Oresteia are as follows: Agamem- non, 855-974 (Agamemnon, Clytemestra, Cassandra) ; Choephoroi (omitting 52 There are, for example, no songs by actors. 53 Cf. the schohum on 490: ^ m^" 'AdrjvS. atrrjXdev evrpeirlffaL diKaa-rds, 6 di '0/3c'(7T7;s iKereiiwu /J-ivei, ai 5e 'Eptvijes (ppovpomiv OLXirbv. 30 THE IDLE ACTOR IN .ESCH\XUS Pylades), apparently none; Eumenides, 574-753 (Orestes, Athene, Apollo). If these numbers are compared with those in the preceding paragraphs, it will be seen that throughout both passages we have the idle actor except in Eumenides 744-53. These few lines then are the only passage in the extant plays of ^schylus in which three actors are kept employed at once. We are now in a position to collect the instances of the idle actor in the extant plays of /Eschylus and to consider them together. They are as follows : Suppliants: Danaus (protagonist) neglected for 175 + 246 + 85 + (57?) = 563 lines. Persians: Atosssa (protagonist) neglected for 41 + 53 + 23 + 45 = 162 Unes. Septem: Etcocles ^protagonist) neglected for 103 lines. Proniclheus: Prometheus (protagonist) neglected for 30 + 20 = 50 lines. Agamemnon: Cassandra (dcuteragonist) neglected for 253 lines. Choephoroi: Electra (dcuteragonist) neglected for 62 + 26 + 67 = 155 lines. ^^ Orestes (protagonist) neglected for 20 lines. Eumenides: Orestes (protagonist) neglected for 20 + 90 + 39 + 76+19+130 = 374 lines. Athene neglected for 89 + 23 = 1 1 2 lines. Apollo neglected for 30 lines. In all, in the 8,117 lines of the extant plays, there are 1,779 lines (about 22 per cent.) in which at least one actor is neglected. In 113 of these lines, two actors are neglected. 5 5 Of the causes of this phenomenon, the most widely operative is the crude theater. We see its effect in the structure of such a play as the Supplianis.^^ Given the simple scene, the poet has evolved a simple situation — the suppliants at the altar — and it is impossible to remove Danaus as often as dramatic propriety requires. So he stands idly by through parodos and stasimon, epi.sode and exodus. The Persians is but little better: no building being at hand, the queen must wait upon the scene while the chorus converses witli the messenger and with Darius. Similarly, Eteocles in the Septem, for lack of a ])lace of withdrawal, must stand idle through the parodos. The silence of Prometheus also is perhaps due to a material cause — the use of a lay-figure. In the Orcsteia material limitations are less operative, but their effect can still be traced. That Cassandra should enter with Agamemnon and be obliged to wait so long 54 Ii is dilTRult to sclfct the lines, a.s it is not always rU-ar whcthi-r Electra is being addressed or not. 55 The mutes (Bin, Pylades, and Hermes) are not here included. 5* And, through imitation, in later plays, such as the Suppliunls of Euripides. THE IDLE ACTOR IN .ESCHYLUS 3 1 seems a survival of early stiffness, the possibility .of easy movement not yet being fully grasped ; and Electra need not have been so often neglected if the back-scene had been as freely used as it is, for example, in many plays of Euripides. The effect of the preference for dialogue between two speakers is seen in Danaus's idleness during the chief scene in the Suppliants; in that of Atossa in the two scenes just mentioned; possibly in that of Prometheus; certainly in much of the neglect of Electra, and in some of the cases in the Enmenides. The prominence of the chorus must be held accountable also for the chief cases affecting Danaus, Atossa, and Electra. Various incidental causes (mentioned above) complete the explanation, so far as it is possible now to give it. Now it appears in various places that vEschylus was at great pains to avoid the idle actor. In Suppliants, 480 ff., Danaus is sent to the city on a most artificial pretext, merely to have him out of the way for the first stasimon; and in the same play, 968, the king goes before Danaus returns, though he might as well have remained till the end, had the poet been able to employ him. Atossa's apology for her silence (290 ff.) is evidence in the same direction, and so are her poorly motived exits at vss. 523 and 849, just before choral passages. 57 The same thing appears in the care taken to avoid neglect of Prometheus — the turning of the parodos into a kommos, the addressing of most of the odes to him, and his careful expla- nation of his silence at vs. 436. The management of Ciytemestra in the Agamemnon, which by keeping her off the scene avoids all awkwardness on the arrival of the messenger and of Agamemnon, shows clearly that the presence of Atossa in the similar scenes in the Persians would have been avoided if it had been possible, without forcing a motive for her exit, to remove her easily from the scene of action. The final conclusion to be drawn from an examination of the seven extant plays of ^schylus is that, while there are a number of striking cases of the idle actor, not one was introduced as a dramatic device, such as the Euripides of Aristophanes professed to see in the Phrygians and the Niohe. This is not to say, of course, that the poet does not make a virtue of necessity now and then, securing a striking dramatic effect from a situation which would have bafHed a lesser poet. It was, however, a dramatic effect won by somewhat violent means, and as such more or less open to ridicule by the comic critic of a later generation. S7 In the latter case there is, of course, the further reason that she was not to be present in the last scene. 32 THE IDLE ACTOR IN ^SCHYLUS In this survey of the plays, an important matter has been mentioned only incidentally — the so-called mutes. It is not uncommon in Sophocles and Euripides to find characters referred to as present who nevertheless say nothing. Such parts were of course taken, not by actors, but by supernumeraries. In other cases a personage who elsewhere speaks is silent through a scene, and the number of other characters present makes it evident that here the part is temporarily taken by a person who is not a qualified actor. Taking these two sorts of mutes together, we find that in Sophocles and Euripides they average about one to a play. In .^schylus, however, if we omit the doubtful case of Pylades, there are but two instances, and neither very noticeable — those of Bia in the Prometheus and Hermes in the Enmenides (see vs. 90). It is worth noticing that the mere presence of a silent person is not necessarily striking or awkward. If attention is not in .some way called to him, he blends with the attendants and other supernumeraries. In such cases as these, however, the mute must have worn the actor's costume to prevent incongruity, and so the effect must have been practically that of an actor's silence, though not a particularly noticeable one. It will be necessary now, for the sake of comparison, to consider briefly the cases of the idle actor in Sophocles and Euripides. Nothing shows more clearly the superiority of Sophocles's technique than his management of the actors. To this the smooth perfection of such a play as the Anlif^one or the OLdipus Rex is largely due. Brought in and sent out at precisely the right moment, on pretexts so natural that the practical reason is entirely concealed, the characters are seldom felt to be in the way even for an instant. For the producing of such a result the perfected theater was the first essential. In si.x of the seven extant plays the back-scene represents a dwelling — a palace in the Antigone, (Ed i pus Rex, Eleclra, and Trachinia:; a tent in the Ajax; a cave in the Philoctcles. In these plays the movements are so easy and natural, and the whole management of the characters so perfect, that the idleness of an actor, when it does appear, is almost invari- ably fitting and ctTcctivc. In such a case the character is not awkwardly neglected: he is constantly present to the mind of the poet, and is silent because it is proper and dramatically necessary that he should be so. I shall first consider these plays, reserving the Qidipus Coloneus for separate treatment. In the Antigone the heroine stands proudly silent while the messenger reports her disobedience to the king (vss. 384-440). So perfect is the THE IDLE ACTOR IN ^SCHYLUS 33 arrangement of the piece that no other case can be regarded as certain. It may be that Creon remains through two odes (582-625, 944-87). Imme- diately after the former the chorus calls his attention to the approach of Haemon, and at the close of the latter, Tiresias, arriving, addresses him; but there is no great difficulty in supposing that he comes from the palace at both these points, for his movements are as easy as those of Clytemestra in the Agameynnon.^^ In the Ajax, Odysseus is silent through fear while Athene questions the mad Ajax (89-117). Tecmessa, restrained by her husband's rebukes, is quiet while he laments his folly and converses with his followers and his son (371-409, 412-84, 545-77, perhaps 646-84). Teucer is appropriately silent while Ajax and Odysseus discuss his cause (1316-73). ^9 In the (Edipiis Rex, the priest is an interested hstener while Creon reports the oracle (85-134), and Jocasta while the chorus intercedes for Creon (649-77). Brilliantly effective is her silence while (Edipus converses with the messenger, unconsciously revealing the fatal truth to her (989- 1053). Nor is it awkward that the messenger stands idle while (Edipus and Jocasta converse (964-88) and while the shepherd is sent for (1047-1118), nor that Creon is silent while (Edipus talks with his daughters (1480-1502). In the Eleclra, Electra easily remains on the scene during the brief odes, 472-515 and 1058-98, which are partly addressed to her; and also (rebuked into silence) during Clytemestra's prayer (634-59). So, though she is idle 680-787, the messenger's story is really for her ear as well as Clytemestra's. Again, nothing could be more dramatic than the ostensible disregarding of Orestes while Electra, holding the urn, addresses the dead Orestes (1126-73). In the TrachinicB it is natural that the nurse (being a mere servant) should remain upon the scene while Deianeira, following her advice, sends her son after Heracles (61 ff.; see 62). Deianeira is present through the parodos (94-140), but it is partly addressed to her. It is natural that the messenger should wait quietly until Lichas enters the house (200-334), 58 Muff, Chorische Technik des Sophocles, p. loi, thinks, with Nauck, that he is left upon the scene in order that the words of the chorus may be supposed to influence his mind. But the application of the latter song to his case is far from direct, and immediately after each he shows his stubbornness. A better argument for the view that he remains might be drawn from the analogous passages in Euripides. 59 In the first stasimon of the Ajax (596-645) it is possible, but not probable, that Tecmessa remains. See Jebb's note. Welcker, Rheinisches Museum, III (1829), p. 87, thinks that even Ajax remains visible. 34 THE IDLE ACTOR IN iESCHYLUS and that Deianeira should listen in silence while he convicts Lichas of deceit (402-28). In the Philoctetes it is ven* effective that Philoctetes should be allowed to overhear the conversation between Neoptolemus and the supposed merchant (542-77), and should sleep through the ode, 821-66.^° Also that Neoptolemus should listen to the dispute between Odysseus and Philoctetes (982-1065), a struggle going on in his own mind which reveals itself after- ward. Nor is it awkward that he is at first not addressed by Hermes (1409-32), for he is soon included in the admonition of the god. There are, however, three cases in these plays which would certainly be felt by the spectator, as by the reader, to be somewhat stiff and unnatural. These are: Tecmessa in the latter part of the Ajax (i 168-1420); lole in the TrachinuB (225-335); and Pylades in the Electra (1-85, 1098-1375, 1422-36, 1466-1510). Tecmessa in the earlier part of the play has speak- ing parts of considerable length, but lole and Pylades, though occasionally addressed, are silent throughout the piece. The reason is not difficult to discover: they could not be made to speak without the employment of a fourth actor; the parts arc therefore taken by mutes. But the effect of a silent actor is there in any event, particularly when the character is else- where a speaking person. Of course, the plot might have been altered in each case so as to avoid this silent figure, but the loss involved would have been serious. In the Ajax, the poet wanted the scene between Odysseus, Agamemnon, and Teucer, and also the picture of the wife and child beside the body. In the Trachinia:, he must have Deianeira, Lichas, lole, and the messenger present together, or sacrifice much of the effect of the scene. In the Electra, tradition required the presence of Pylades; and Orestes, Electra, and ^gisthus were all needed in the final scene. ^' It appears, then, that in these si.x plays the only awkward cases of the idle character are due to the limitation in the number of actors. Whether this rule had a practical basis or was merely an artistic convention, it has plainly had in these cases an unfortunate effect on Sophocles's work. A striking contrast to the generally successful management in these plays is afforded by the Qidiptis Coloneiis, which involves more long periods of silence than any other extant tragedy. This is doubtless due in part to 6° Neoptolemus also is present, but he takes jiart 839-42, and Bcrgk conjectures a similar four lines lost between 854 and 855. 6' The child parts played by mutes, Eurysaces in the .4/(7.v, Antigone and Ismcnc in the QLdipiis Tyrattnus, need not be considered, since no special awkwardness is involved. THE IDLE ACTOR IN iESCHYLUS 35 the elaborate plot and the poetic tone, but chiefly to the simple fact that the scene is not laid before a building. The background represented a grove (14-18, 38-40, 114, 125, 156 f., 505).^^ There seems to have been an opening in the center of the back-scene, or at least some object (as a clump of trees) behind which one could retire, for (Edipus and Antigone conceal themselves on the first appearance of the chorus (111-16) and reappear suddenly (138 ff.). This could not have been done had they been obHged to use the side entrance. The same means of exit may have been used by Ismene when she goes to perform the sacrifice "beyond the grove" (see 505), and by (Edipus on his final disappearance (iSSS).'^^ Except in these few cases, all entrances and exits must be through the parodoi, and the characters must be supposed to come from or go to a distance. We have thus a reproduction of early conditions, combined with a more elaborate plot than ^Eschylus could have conceived. The result is an extraordinary number of long silences. (Edipus must remain visible through most of the piece, but it is not difficult to keep him occupied for the greater part of the time. He is idle during the quarrel between Theseus and Creon (897-959, 1014-37), during two stasima (1044-95, 12 11-48), and during the conversation between Antigone and Polynices (1405-56). In the remaining lyric passages in which he is present he takes part, or is at least addressed. Antigone fares worse. As her father's constant attendant she is like- wise tied to the spot, and, being a less important character, she cannot be kept well employed. She is thus neglected during the odes 117-69, 510-48, 668-719, and 1211-48; during the dialogue passages 36-80, 84-110, 258-309- 353-420, 421-92, 549-667, 724-827, 1119-80, 1289-1404, 1505-41. The same fate, in a lesser degree, overtakes other characters. Ismene is idle 421-92, 1096-1555; Creon, 887-908; Theseus, 960-1013, 1181-1205. Ismene's long silence (1096-1555) is probably due to the assignment of the part to a mute, three other characters being present. ^4 Euripides began his work with many things in his favor, and the earher extant plays show how easy it had become to manage the actors so that 62 See Schonborn, op. ciL, pp. 272 £f.; Wieseler, Gottingen Nachrichten (1890),, p. 215; Jebb's Introduction, pp. xxxvii f.; Bodensteiner, loc. cit., pp. 652, 776—78; Dorpfeld-Reisch, op. cit., pp. 209 f. 63 With, of course, the reappearance of those who had accompanied him. It seems probable, however, that the exit was used only in the first-named instance (ill— 16, 138), since they were then polluting sacred ground. 64 So Wecklein, ed., p. 8; Teuffel, Rheinisches Museum, N. F. IX (1854), pp. 136 ff. Miiller, Geschichte griechischer Litteratur, I, p. 403, prefers to assume a fourth actor. 36 THE IDLE ACTOR IX -ESCHYLUS all should be kept employed while upon the scene. The Alceslis and Medea are almost entirely free from awkward periods of silence on the part of a character. This is due partly to the improved theater, which made movement easy; partly to the lesser importance of the chorus ;^s and partly to the example of Sophocles, who had shown how perfectly three- actor scenes could be arranged. This smoothness of structure, however, soon began to be more and more broken up by causes which lay partly in circumstances and partly in the nature of Euripides's genius. In the first place, the chorus still remained and must be allowed its part in the performance; nor was it always possible to remove an actor for an ode or to allow him to take part. In the second place, Euripides in his desire for variety introduced novel situations and scenes, which made the improved background practically useless. Finally, and more important than all, as his individual style developed, Euripides showed plainly the lack of Sophocles 's peculiar skill in arranging scenes. In particular, he became fond of a kind of scene which is favorable for displays of rhetoric, but is too stiff and artificial for dramatic effect. A, B, and C appear together; A and B have a long argument in which C is disregarded; then A and C converse, neglecting B. This is of course an attempt to adapt the old two-part dialogue to a three- actor scene and shows plainly the influence of the Athenian law-courts. It is only necessary to compare the three-actor scenes in Sophocles to see how superior he was in this sort of technique. From these various causes, the phenomenon of the neglected actor becomes more and more common in the work of Euripides until the time of the Heracles and the Suppliants, when it reaches a clima.x. It then becomes less frequent, as Euripides's taste for the archaic led him back to simpler plots. The cases are as follows: Alceslis: Admetus, 962-83 (only the latter half of the ode is addressed to him). Medea: Nurse, 144-67 (Medea calls from the house, and the chorus answers); Medea, 410-30, 627-62, 824-45, 1081-1115 (she cannot enter the house of her enemies, and so must remain through the odes). Andromache: Menclaus, 551-78, Andromache, 577-716, 719-47 (three-part scene). Heraclida: Copreus, 69-98, lolaus, 101-117 (the parodos is here a three-part scene); lolaus, 118-80, 250-96, Copreus, 181-249 (three-part scene); lolaus, *5 In Kuripidcs the chorus has comparatively little share in the dialogue. It seldom takes an important part in the conversation when two or more actors are present; hence the cases of the neglect of an actor from this cause are few. THE IDLE ACTOR IN ^SCHYLUS 37 353-80, Alcmene, 748-83 (suppliants at the altar); Demophon, 427-50 (lolaus speaks to children: practically a three-part scene); Demophon, 478-566 (three- part scene); Alcmene, 667-708 (three-part scene); Alcmene, 720-47 (three-part scene); Messenger, 941-60, Eurj'stheus, 961-82 (three-part scene). The temple in the background is but little used; hence much of the awkwardness. Hippolylus: Phjedra, 250-87, Nurse, 364-432 (three-part scene); Phaedra, 525-64 (she listens to what is said within the house); Theseus, 1347-1406, Artemis, 1347-88 (three-part scene). Hecuba: Polyxena, 216-333, Hecuba, 342-71, Odysseus, 404-31 (three-part scene) ; Hecuba, 444-98 (she lies wrapt in her robe during the stasimon) ; Maid, 726-889 (three-part scene; c}. vs. 778); Hecuba, 1056-1121 (she avoids the blinded and furious Polymestor); Hecuba, 1124-86, Agamemnon, 1 197-1232, 1254-79 (three-part scene). Heracles: Megara, 1-59 (while Amphitr}'on speaks the prologue); Megara, 165-274, Lycus, 275-315, Amphitrv'on, 240-77 (three-part scene); Amphitryon, 442-96 (Megara talks to children: practically a three-part scene); Amphitr\'on, 538-61, Megara, 585-625 (three-part scene); Heracles, 1042-87 (kommos while he sleeps); Heracles, 1 163-1202, Amphitryon, 1214-135 7, Theseus, 1358-85 (three- part scene). Megara's periods of neglect in the first part are due to her position as a suppliant at the altar: she cannot enter the house. lati: Ion, 184-218 (during the entrance of the chorus he is busy driving away the birds from the temple);**^ Servant, 859-924 (Creusa's apostrophe); Creusa, 1320-94 (three-part scene); Ion, 1571-1603 (three-part scene). Suppliants: Adrastus, 1-86 (cUnging to the altar, he is idle during the pro- logue and parodos); Adrastus, 87-109, 263-364, ^thra, 110-285 (three-part scene); Adrastus, 365-512, 517-84 (three-part scene); Adrastus, 598-633 (stasi- mon); Adrastus, 634-733 (messenger addresses chorus: practically a three-part scene); Theseus, 772-837 (kommos between Adrastus and Chorus); Adrastus and Theseus, 11 14-64 (kommos between Chorus and children); Adrastus, 1 1 96- 1 23 1 (three -part scene). Through nearly the whole piece one or more characters are neglected. This is due chiefly to the situation: the suppliants must remain at the altar, and the temple in the background is not used for entrances and exits. Trojan Women: Hecuba, 1-97 (she lies wrapt in her robe during the pro- logue); Talthybius, 308-407, Hecuba, 427-57 (three-part scene); Hecuba, 511-67 (stasimon); Hecuba, 709-89, Talthybius, 740-73 (three-part scene); Hecuba, 799-859 (stasimon); Hecuba, 914-65, Menelaus, 971-1028 (three-part scene); Hecuba, 1060-1117 (stasimon). Electra: Peasant, 364-403 (three-part scene); Electra, 598-646, Old Man, 671-93 (three-part scene); Orestes, 907-58 (Electra addresses dead .^gisthus; practically a three-part scene); Electra, 1238-1302 (three-part scene). Helen: Menelaus, 515-45, 1093-1183 (he cannot enter the palace during 66 Detscheff {op. cit., note 105) thinks that he enters the temple. 38 THE IDLE ACTOR IN ^SCHYLUS the odes); Messenger, 623-99, Menelaus, 711-33 (three-part scene); Menelaus, 892-946, Helen, 947-97 (three-part scene); Menelaus, 1 184-1249, Helen, 1250-84 (three-part scene). Iphigenia among the Taurians: Iphigenia, 392-466 (stasimon); Pylades, 494-577, 582-611, 614-46 (three-part scene); Orestes, 744-71, Pylades, 795-901, 924-1055 (three-part scene); Orestes and Pylades, 1056-78 (Iphigenia addresses Chorus: practically a four-part scene); Thoas, 1445-66 (practically a four-part scene). Phcenician Wamen: Eteocles, 469-98, Polynices, 499-567 (three-part scene); Menocceus, 849-969 (three-part scene); Antigone, 1585-1630, Creon, 1595-1619, (Edipus, 1643-75 (three-part scene). Orestes: Orestes, 1-210 (he sleeps through the prologue and parodos); Orestes, 352-79 (three-part scene); Orestes, 470-525, Menelaus, 542-621 (three- part scene); Pylades, 1018-64, Electra, 1065-1176 (three-part scene). Iphigniia at Aulis: Menelaus, 414-35 (three-part scene); Iphigenia, 590-612 (three-part scene); Clytemestra, 640-84 (three-part scene); Achilles, 866-95 (three-part scene); Iphigenia, 1129-1210, Clytemestra, 121 1-75 (three- part scene); Iphigenia, 1345-68, Achilles, 1369-1404, Clytemestra, 1403-33 (three-part scene). Bacchce: Cadmus, 215-50, 255-329, Tiresias, 215-54, 330-57 (three-part scene); Dionysus, 660-786 (three-part scene) ; Agave, 1308-28 (Cadmus addresses dead Pentheus; practically a three-part scene). Some of these cases are natural and dramatically effective, as in Sophocles; such are Hippolyliis, 524-64, 250-87, 364-482, 1347-88; Hecuba, 444-83, io^6~iioS; Heracles, 1042-88, 1 163-1202; /on, 184-236, 1320-94; Orestes, 1-206, 352-79; Iphigenia at Aulis, 590-630. But in general it cannot be supposed that the elTect was desired by the poet. It remains to speak of the cases where important characters are repre- sented, either temjwrarily or throughout the piece, not by actors, but by mutes. Child-parts need not be considered. In Alcestis, 962 fT., Alcestis (brought back from the grave) is persist- ently silent, and this is cxi)lained as due to religious scruples. It is dif^kult to believe that Euripides would have missed such an opportunity for pallios without a practical reason, and the fact that elsewhere only two actors are re(iuircd suggests the true cause: for some reason or other, quite unknown to us, only two were available. The same reason (the paucity of actors) caused the parts of Pylades and Hermione in the last scene of the Orestes to be given to mutes. Pylades and one of the Dio.scuri in the Electro, and one of the Dioscuri in the Helen, were played throughout by mutes. Of these only the case of Pylades is noticeably awkward, and hero the example of .^tischylus and Sophocles had established a convention. THE IDLE ACTOR IN ^SCHYLUS 39 Adding up all the "idle" lines in Euripides and comparing them with the sum of the lines in the plays, we find that they amount to about one- fourth, a slightly larger proportion than in ^^schylus and about the same as in Sophocles. In conclusion I wish to compare the methods of the three dramatists in dealing with a particular phase of the problem; namely, the disposing of the actors during lyric passages — parodos and stasima. It must be remembered that these passages were the earliest to take shape, and that it was about these lyrics as nuclei that the drama grew. In the period of the single actor it must have been comparatively easy to dispose of that actor while the choruses were sung, for his short speeches were mere inci- dents, and he naturally withdrew as soon as they were finished. But with the steady increase in the importance of the dialogue parts and the corre- sponding diminution of the choruses, the removing of the actors at the end of prologue and episode became a matter of no little difficulty. ^^ We have seen how ^Eschylus wrestled with the problem, and partly succeeded, partly failed, in giving verisimilitude to situations in themselves improb- able. In the earher plays he removes characters on unlikely pretexts, or leaves them frankly idle ; in the later ones he often lets them slip unnoticed into the building in the rear. Only once does he turn the parodos into a part-song between actor and chorus. With Sophocles and Euripides the circumstances were changed; the odes were now shorter, the characters more numerous. It was natural that the continuous presence of an actor should be regarded with less aversion, and in fact the occurrence became distinctly more common. At the same time, the parodos was often made a kommos^^ (especially where it was natural that the chorus should question the actor about the state of affairs), and the stasimon was often addressed partly or wholly to the actor. ^^ The parodos and stasima in which an actor is present are the following. I have given in each case the number of fines in which an actor is neglected, 67 The real source of the trouble was, of course, the continuous presence of the chorus, which rendered scene-change practically impossible. 68 The innovation seems to have been slow in coming in, however. The CEdipus Tyrannus and Trachinia, for example, might well have had kommatic parodoi; see Detscheff, op. cit., p. 45. 69 Euripides sometimes gives the first half of an ode to gmeral reflections, the second to an address to the person present. So Alcestis, 962-1005; Medea, 410-45, 824-65, 976-1001; HeraclidcB, 608-29. 40 THE IDLE ACTOR IN -ESCHYLUS i. e., neither speaking nor spoken to, even where it is very small, and italicized the cases where the actor is neglected throughout the ode. iESCHYLUS Suppliants: 1-175 (Danaus, lyj //.); 625-709 (Danaus, S5 II.). Persians: 623-80 CAtossa, 53 11.)- Seplem: 78-180 (Eteocles, loj II.) . Promelheus: 128-92 (Prometheus, o 11.); 397-435 (Prometheus, o 11.); 526-60 (Prometheus, 14 11.); 887-906 {Prometheus, 20 IL). Af^amemnon: 975-1034 (Cassandra, $(^ II.). Choephoroi: 22-83 {Electra, 61 II.). Eumenides: 307-96 {Orestes, go II.); 490-565 {Orestes, 76 //.); 77S-880 (.\thcne, 10 11.). SOPHOCLES Antigone: 582-625 {Creon, 44 II.) (?); 944-87 {Creon, 44 II.) (?). Ajax: 866-973 (Tecmessa, present 891-973, 10 11.); 1 185-1222 {Tecmessa, mute, 3S II.). (Edipus Tyrannus: 1085-1109 (CEdipus, 12 11., messenger, 24 //.). Electra: 121-250 (Electra, o 11.); 472-515 (Electra, 12 11.); 823-70 (Electra, oil.); 1058-97 (Electra, 23 11.). Trachinice: 94-140 (Deianeira, 27 11.); 205-24 (Deianeira, 17 11.). Philoctetes: 135-218 (Neoptolcmus, o 11); 827-64 {Philoctetes, 38 //.). Q^.dipus Coloneus: 117-253 (CEdipus, 40 11., .Vntigone, 93 11.); 510-48 (CEdipus, o 11., Antigone, jg II.); 668-719 (CEdipus, 8 11., Antigone, 52 //.); 1044-95 {(Edi- pus, ^2 II.); 1211-48 {CEdipus, jS //., Antigone, jS II., Ismene, mute, j8 //.). EURIPIDES Alcestis: 962-1005 (Admetus, 22 11.). Medea: 131-213 (Nurse, present 131-203, 24 11.); 410-45 (Medea, 21 11.); 627-62 {Medea, 36 II.); 824-65 (Medea, 22 11.); 976-1001 (Medea 14 11.). Andromache: 117-46 (Andromache, o 11.); 274-308 (.Andromache, o 11.). Heraclidce: 73-110 (Coprcus, 26 11., lolaus, 10 11.); 353-80 {lolaus, 2S II.); 608-29 (lolaus, 10 11.); 748-83 {Alcmene, 36 II.) . Ilippolytus: 525-64 {Pluedra, 3g II.); 1268-82 {Theseus, 75 //.). Hccuha: 98-176 (Hecuba, o 11.); 444-83 {Hecuba, 40 II.). Heracles: 107-37 (Amphiarcus, 30 11., Megara, 28 11.). Ion: 184-236 (Ion, 35 11.). Suppliants: 42-86 {Adrastus, 4^ II., .Ethra, 16 11.); 598-633 {Adrastus ^ 36 II.). Trojan Women: 153-229 (Hecuba, o 11.); 511-67 {Hecuba, 37 H.); 799-859 (Hecuba, 61 //.); 1060-1117 (Hecuba, 55//.). Electra: 167-212 (Electra, o 11.); 859-879 (Electra, o 11.). Helen: 191-251 (Helen, o 11.); 515 27 (.U<7/</(n<.v, / 7 //.); i\o-j-64 (Mcnelaus, 58 II.). THE IDLE ACTOR IN .ESCHYLUS 41 Iphigenia among the Taurians: 123-235 (Iphigenia, 14 11.); 643-56 (Orestes, 5 11., Pylades, 5 H-)- Phomician Women: None. Orestes: 140-207 {Orestes, 68 II. , Electra, o 11.); 316-47 (Orestes, 20 11.). Iphigenia at Aulis: None. Bacch<E: None. The results may be tabulated as follows:'" Number of Plays Odes in Which Actor is Present Average to Play Odes in Which Actor is Entirely Neglected Average to Play Total Lines Neglected Average to Play j^schylus Sophocles Euripides 7 7 17 13 18 33 1.86 2-57 1.94 8 7 14 I. 14 I .00 0.82 746 611 892 107 87 52 The extant tragedies as a whole may be roughly but conveniently divided into three classes — simple, developed, and loose. By a simple tragedy I mean one that shows clearly the influence of the form from which tragedy arose — the alternation of songs with two-part dialogues. In such a tragedy the lines of conventionality are still too closely followed to allow of much verisimilitude in the arrangement of the plot, and the aim of the poet is achieved if the play be neatly constructed, clear, and fairly well motived. Such a play is apt to fall apart on examination into a succession of distinct episodes, not all essential to the story, alternating with choral songs. In this class may be placed the Suppliants, Persians, Septem, and Prometheus of ^Eschylus. A developed tragedy is one in which the old form is less rigidly obvious and now exists as the skeleton which supports the play without impeding it. The divisions are less marked, and the play has become an organic whole. Clever motiving and arrangement bring the characters upon the scene at the right moment, and remove them when they cease to be needed. The various threads of the action are closely interwoven, and there are no loose ends. To this class belong the Agamemnon of ^schylus; the Ajax, Antigone, Electra, Trachinice, (Edipus Tyrannus, and Philodetes of Sophocles; and the Alcestis, Medea, Hippolytus, Ion, Iphigenia among the Taurians, Iphigenia at Aidis, Helen, Electra, and Bacchce of Euripides. 70 Apparently at a still later period more pains were taken to avoid the presence of actors during odes. In Seneca, seven of the eight parodoi and thirty-four of the thirty-six stasima are given without actors present. 42 THE IDLE ACTOR IN -ESCHYLUS By a loose tragedy I mean one in which the old form is not merely concealed, but to some extent disintegrated. It is no longer stiff enough to support the structure, which becomes in consequence somewhat flabby. While the general course of the play may be direct and consistent enough, there is a tendency to lose that distinctness of function of everv' part which was the result of the development of tragedy from a simple original. To this class belong the Cho'ephoroi and Eumenides of .^schylus; the CEdipus Coloneiis of Sophocles; and the Hecuba, Andromache, Heracles, Sup- pliants, Heraclida, Trojan Women, Plvenician Women, and Orestes of Euripides. The idle actor is really a different phenomenon in each of these three classes of plays. In a simple tragedy he appears as a fault in the structure. In spite of all the poet's efforts, occasions arise on which a character can be neither employed nor removed. The characters being few, the neglect is noticeable, and the poet is usually quite conscious of the awkwardness, and .sometimes apologizes for it. In the second class the artist's tools are more elaborate and his mastery of them more complete. The characters are more numerous and move more easily, and if one be dropped from the dialogue, there are various ways of accounting for his silence and making it natural; nor is it in itself so noticeable a thing. The bulk of Sophocles's work is of this sort, and a part of that of Euripides is scarcely inferior, but the latter's scenes are marred by the deliberate adoption of the pecuHar form of three-part scene already described. The plays of the third class are so freely constructed that the idle actor is hardly felt as a defect. Episodes and stasima are often blurred together by the continuous pre.sence of an actor, and the characters are held with so loose a hand that the spectator hardly notices when they come or go. This stage is in a sense the breaking up of the old tragedy as an art-form. I have tried to show that the instances of the idle actor in the e.xtant plays of .^schylus can invariably be referred to practical causes — that when artistically effective they arc so merely because the poet had learned to turn to account a thing that conditions forced upon him; that Sophocles u.sed the idle actor more artistically, but could return to early crudeness when the older conditions were partially reproduced; that Euripides, with all his advantages, was inferior to Sophocles in this sort of technique. It may be objected that I have enlarged the matter far beyond the simple charge of Aristophanes, making it include all instances in which an actor drops from the dialogue or is disregarded during an ode; it may also THE IDLE ACTOR IN iESCHYLUS 43 be said that, since the plays to which Aristophanes refers are lost, we cannot examine what were probably the most striking cases. All this is true; but the point is this: having determined be)'ond a doubt that the classic drama- tists, and especially ^Eschylus, were constantly struggling with the difficulty of keeping the actors employed when present, we are justified in assuming that all instances of the idle actor originated, not in a striving after exagger- ated efifect, but in the obstacles opposed by conditions to the rise of the drama. To return, then, to the criticism of Aristophanes on ^schylus. It remains, of course, a possibiUty that in certain cases the silence of an actor was a thing deliberately sought for, though the evidence of the extant plays is against such a view. It is possible, that is to say, that Achilles and Niobe were shown as silent figures because the poet thought the effect desirable, but the analogy of extant scenes suggests that the thing originated in a practical difficulty. They were idle, says Aris- tophanes, while the chorus sang a long ode; for the reason, we may add, that the poet did not know what else to do with them. Hampered by tradition, by material disadvantages, and by the immaturity of his art, he was forced to do, not what he would, but what he could. A clear under- standing of the difficulties which beset him should increase, rather than diminish, our admiration for the genius of ^Eschylus. ^^0^^f^^\ U.C. BERKELEY LIBR, III CD0b3MQ733