Skip to main content

Full text of "The plays and fragments. With critical notes, commentary, and translation in English prose"

See other formats







| ZIZELELO (92 € 


Digitized by the Internet Archive 
in 2008 with funding from 
Microsoft Corporation 


httos://archive.org/details/playsfragmentswi01 sophuoft 








a: » 
SAF, 


t dis 
ie 
oY, y= 


᾿ ᾿ a 
Sid 
ae 





SOP τ θυ δ τὰς 


ΓΗ ον ΕΝ Ὁ PRAGM EN ES: 


PAK lk 
TBE OEDIPUS LY RANNUS, 


ἘΝ 


CAMBRIDGE UNIVERSITY PRESS 
C. F. CLAY, MANAGER 
Dondor: FETTER LANE, E.C. 
EHvinburgh: soo PRINCES STREET 








Berlin; A. ASHER AND CO. 
DLeipsig: F. A. BROCKHAUS 
few Bork: G. P. PUTNAM’S SONS 
Bombay and Calcutta: MACMILLAN AND Co., "τυ. 
Toronto: J. M. DENT AND SONS, Lr. 
Tokyo: THE MARUZEN-KABUSHIKI-KAISHA 


All vights reserved 


° ΄ 


SOPHOCLES 


WITH CRITICAL NOTES, COMMENTARY, AND 
TRANSLATION IN ENGLISH PROSE, 


BY 


Sim RICTIAR DC. TEBE. irr. 


FORMERLY REGIUS PROFESSOR OF GREEK 
AND FELLOW OF TRINITY COLLEGE, CAMBRIDGE. 


Lear ταῦ ee F- 
THE OEDIPUS TYRANNUS, 


Gs 


15: 


CAMBRIDGE: 
AT THE UNIVERSITY PRESS. 


1914 


first Edition 1883. 
Second Edition 1887. Third Edition (stereotyped) 1893. 


Reprinted 1902, 1014 


PH 
HUIS 
2, 

IF /ua 
at, | 
Cop. > 


PREEPACE, 10, ΤῊ SECOND 2 DITION. 


is preparing a second edition of this volume, I have profited 
by several criticisms with which the work has been favoured, 
and by various other contributions to the study of Sophocles 
which have come into my hands since 1883. The modification 
of detail which is chiefly noticeable in the present edition is the 
substitution of English for Latin as the language of the critical 
notes on the text. Without having altered the opinion which I 
formerly expressed, that Latin possesses unequalled merits for 
this purpose, I had been led to feel that a combination of Latin 
critical notes with an English commentary on the same page 
suffered from acertain want of unity and harmony. There seemed 
to be also a practical objection, viz. that some readers were 
harassed by the change of mental attitude involved in turning 
from a Latin to an English note on the same passage. The 
intrinsic superiority of Latin as a vehicle of textual criticism 
could hardly be deemed to outweigh these disadvantages; and 
it is by this consideration that my choice has now been decided. 

The Autotype Facsimile of the Laurentian Ms. of Sophocles, 
published in 1885 by the Society for the Promotion of Hellenic 
Studies, is by far the most important boon ever conferred on 
students of the text. A possessor of this perfectly executed and 
durable photograph commands an aid of indefinitely greater value 
than the most minute and most exact collation; so far, indeed, as 


the purposes of textual criticism are concerned, he has the 


vl PREFACE. 


manuscript itself before him. I have used the facsimile in care- 
fully verifying the report of the Laurentian readings given in my 
first edition, and on a few points have been enabled to supple- 
ment it, or to render it more precise. In this connection I may 
briefly advert to another point of detail which distinguishes the 
present re-issue. Some of my friendly critics in Germany have 
observed that those MSS. which are later than the Laurentian, 
and which are all more or less signally inferior to it, were 
reported in my first edition with a superfluous fulness, which 
somewhat encumbered the critical apparatus, and also tended to 
obscure the leading facts. The view which, for a long period 
of time, has been steadily gaining ground in Germany is that, 
whether the Laurentian MS. is or is not actually the sole source 
of all the other Mss. of Sophocles now extant, at least the cases 
are very rare in which any correction of the Laurentian by 
another MS. is of a higher order than could have been furnished 
by a grammarian’s conjecture. The difficulties in the way of sup- 
posing the Laurentian to be, in fact, the unique source still seem 
to me very considerable. But the experience gradually gained 
in the progress of this work has impressed me, more and more, 
with the truth of the other proposition just noticed,—viz., that 
_ the positive worth of the corrections supplied by the other Mss. 
is no greater than it easily might have been if the Laurentian 
were their common parent. Forty years have passed since 
Cobet first maintained that the Laurentian is the Ms. from 
which all the rest have been immediately or indirectly tran- 
scribed; and, though I cannot share the confidence with which 
that view has since been defended by such scholars as Dindorf 
and Moriz Seyffert, I can now comprehend it, at least, better 


than formerly, Be our view of the genealogical facts what it 


PREFACE. vii 


may, it cannot be questioned that, in critical notes on the text 
of Sophocles, the paramount significance of the Laurentian Ms, 
must be brought into clear and bold relief. Dindorf effects this 
by referring to the later MSS. under the generic name of ‘apo- 
grapha.’ Mekler, in the 6th Teubner edition of Dindorf’s text 
(1885) uses the letter ‘r’ to denote ‘lectio e recentiorum 
librorum consensu aut uno alterove ducta.’ This symbol, ‘r,’ 
has been adopted by me in the critical notes of this edition 
to denote ‘one or more of the MSs, other than the Laurentian’; 
but it is used only in those cases where a more specific 
statement was unnecessary. By thus combining the use of a 
general symbol with occasional recourse to more particular 
statement, I have sought to exhibit the relative importance of 
the documents in a just perspective, without any undue sacrifice 
of precision. 

The commentary, as it is now set forth, will furnish suffi- 
cient evidence of the desire which I have felt to profit by any 
criticism which has convinced my own judgment, and to express 
gratitude for such criticism in the most practical form. Among 
my foreign reviewers, mention is due to Professor Wecklein, 
and to Dr Kaibel, the editor of the Epzgrammata Graeca. 
To the latter I am indebted for calling my attention to 
epigraphic evidence of the 5th and 4th centuries B.c. in regard 
to the Attic orthography of certain words. The Grammatik der 
Attischen Inschriften, by Professor Meisterhans (1885), is an 
excellent hand-book of reference on this subject. Among 
English critics, I owe grateful acknowledgments to the authors 

1 In v. 68 I should have given ηὕρισκον, not εὕρισκον, had I then known the 
evidence collected by Meisterhans from Attic inscriptions of the 5th and 4th centuries 


B.C. for the temporal augment in the historical tenses of verbs beginning with ev. 
Following that evidence, I have given ηὕρηκ᾽ in 546 and ηὑρῆσθαι in 1050. 


viil PREFACE. 


of unsigned reviews in several journals, as well as to some 
eminent scholars whom I am permitted to thank by name,— 
Professor Butcher,—whose examination of this work, in the 
Fortnightly Review, has been to me an exceptionally valuable 
source alike of instruction and of stimulus,—Professor Tyrrell, 
Mr A. Sidgwick, and Mr R. Whitelaw. The criticisms of Mr 
Whitelaw occupy a large space in the Transactions of the 
Cambridge Philological Society for 1886. Although I have not 
always been able to agree with his views, I have been indebted 
to them for amendments on some points, and have never differed 
from them without careful consideration; nor has anything 
given me more pleasure in connection with this book than the 
very kind and generous manner in which he has referred to it. 

I must once again express my best thanks to the Managers 


and staff of the Cambridge University Press. 


THE COLLEGE, GLASGOW, 
November, 1887. 


CONTENTS: 


INTRODUCTION . : ‘ : . page xi—li 


§ 1. General characteristics of the play and of the fable. ~ 
§ 2. References in the Homeric Poems. ὃ 3. Other epic versions, 

§ 4. Pindar. ὃ 5. The logographers. ὃ 6. The dramatists.—/ 
Aeschylus. 

/§ 7. Sophocles. Original features of his plot. ὃ 8. Imagined 
antecedents. ὃ 9. Analysis.“ ὃ 10. Aristotle’s criticisms.” The 
element of improbability. § 11. The characters.“ § 12. Oedipus. 

§ 13. Iocasta. ὃ 14. Teiresias. Creon. ὃ 15. Supposed allusions 
to contemporary events. Alleged defeat of the play. ὃ 16. The 
actor Polus. Significance of a story concerning him. 

§ 17. Other plays on the subject. § 18. The Oedipus of 
Seneca. ὃ 19. His relation to Sophocles. ὃ 20. The Oedipe of 
Corneille. ὃ 21. The Oedifus of Dryden. ὃ 22. The Oedipe of 
Voltaire. § 23. His criticisms. § 24. Essential difference between 
Sophocles and the moderns. § 25. Their references to prophetic 
instinct in Oedipus and Iocasta. ὃ 26. The improbable element ν΄ 
—how managed by the moderns. 

§ 27. Recent revivals of Greek plays. ὃ 28. The Oedipus 
Tyrannus—a crucial experiment. ὃ 29. The result at Harvard. 

§ 30. Ocdipe Rot at the Théatre Frangais.—Conclusion. 


MANUSCRIPTS, EDITIONS, ETC. . . ; ‘ : .  lii—lxi 
METRICAL ANALYSIS . : : : ; : Ixilli—xcv 


ANCIENT ARGUMENTS TO THE PLAY; DRAMATIS PERSONAE; 


STRUCTURE ; ; ; : Ρ : ; 37-9 
395.4 oe : ; : : : Σ ἐ Ἢ ᾿ 2 tO== 200 
APPENDIX . : ; ; : ; : : 20I—234 


INDICES : ᾿ , : : : ; Ἶ 235—251 


fi 








ἘΚ τοΝ; 


§ 1. THE Oedipus Tyrannus is in one sense the masterpiece 
of Attic Tragedy. No other shows an equal degree of art in 
the development of the plot; and this excellence depends on the 
powerful and subtle drawing of the characters. Modern drama, 
where minor parts can be multiplied and scene changed at 
will, can more easily divorce the two kinds of merit. Some 
of Voltaire’s plays, for instance, not first-rate in other ways, are 
models of ingenious construction. The conditions of the Greek 
stage left less room for sucha result. In the Oedipus Tyrannus 
the highest constructive skill is seen to be intimately and 
necessarily allied with the vivid delineation of a few persons. 

Here it is peculiarly interesting to recover, so far as we 
can, the form in which the story of Oedipus came to Sopho- 
cles; to remark what he has altered or added; and to see how 
the same subject has been handled by other dramatists. 

The essence of the myth is the son slaying his unknown 
father, and thereby fulfilling a decree of fate. The subsequent 
marriage, if not an original part of the story, seems to have 
been an early addition. The central ideas are, (1) the irresis- 
tible power of destiny, and (2) the sacredness of the primary 
natural ties, as measured by the horror of an unconscious sin 
against it. The direct and simple form in which these ideas 
are embodied gives the legend an impress of high antiquity. 
This might be illustrated by a comparison with the story of 
Sohrab and Rustum as told in Mr Matthew Arnold’s beautiful 
poem. The slaying of the unknown son by the father is there 
surrounded with a pathos and a chivalrous tenderness which 
have no counterpart in the grim simplicity of the Oedipus myth, 
as it appears in its earliest known shape. 


Homeric 
Poems. 


ΧΙ INTROD CCTION, 


§ 2. The /éad, which knows the war of Polyneices and his 
allies against Thebes (4. 378), once glances at the tale of 
Oedipus—where Mecisteus, father of Euryalus, is said to have 
visited Thebes in order to attend tne funeral games which were 
celebrated after the death of Oedipus (23. 679 ἢ) :— 


ὅς ποτε Θήβασδ᾽ ἦλθε δεδουπότος Οἰδιπόδαο 
ἐς τάφον, -- 
—‘who came to Thebes of yore, when Oedipus had fallen, to his 
burying.’ 


The word δεδουπότος plainly refers to a violent death in 
fight, or at the hand of an assassin; it would not be in accord 
with the tone of epic language to understand it as a figurative 
phrase for a sudden fall from greatness. But more than this the 
Iliad does not tell. The poet of the 23rd book imagines 
Oedipus as having died by violence, and received burial at 
Thebes, in the generation before the Trojan war. 

The Nekyia in the Odyssey gives the earliest sketch of an 
integral story (11. 271 ff.):— 


Μητέρα τ᾽ Οἰὐἰδιπόδαο ἴδον, καλὴν ᾿Επικάστην, 
ἣ μέγα ἔργον ἔρεξεν aidpeinoe νόοιο 

͵7 @ tan ¢ Paes εν | io Pee , 
γημαμένη ᾧ viel’ ὁ δ᾽ ὃν πατέρ᾽ ἐξεναρίξας 

A » ᾿] 9 Z \ 7 3 , 

ynuev' ἄφαρ δ᾽ ἀνάπυστα θεοὶ θέσαν ἀνθρώποισιν. 
ἀλλ᾽ ὁ μὲν ἐν Θήβῃ πολυηράτῳ ἄλγεα πάσχων 
Καδμείων ἤνασσε θεῶν ὀλοὰς διὰ βουλάς" 
ἡ δ᾽ ἔβη εἰς ᾿Αἴδαο πυλάρταο κρατεροῖο, 
ἁψαμένη βρόχον αἰπὺν ad’ ὑψηλοῖο μελάθρου, 
ᾧ ἄχεϊ σχομένη" τῷ δ᾽ ἄλγεα καάλλιπ᾽ ὀπίσσω 
eX Dad ja AL , a4 

\ I~? e/ \ 3 / 3) , 
πολλὰ μάλ᾽, ὅσσα TE μητρὸς ‘Epivves ἐκτελέουσιν. 


‘And I saw the mother of Oedipodes, fair Epicasté, who wrought a 
dread deed with unwitting mind, in that she wedded her son; but he 
had slain his father ere he wedded her; and presently the gods made 
these things known among men. Yet he still ruled over the Cadmeans 
in lovely Thebes, suffering anguish by the dire counsels of the gods; 
but she went to the house of Hades, the strong warder, when she had 
fastened a noose on high from the roof-beam, possessed by her pain; 
and to him she bequeathed sorrows full many, even all that a mother’s 
Avengers bring to pass.’ 


INTRODUCTION. ΧΗ 


With regard to this outline in the Odyssey, it is to be noted 
that it ignores (a) the deliverance of Thebes from the Sphinx— 
though this may be implied in the marriage with Epicaste: 
(0) the self-blinding of Oedipus: (c) the expulsion of Oedipus 
from Thebes—herein agreeing with the indication in the //ad. 
It further seems to exclude the notion of Epicasté having borne 
children to Oedipus, since the discovery followed ‘presently’ 
on the union,—unless, indeed, by ἄφαρ the poet merely meant 
‘suddenly.’ 


8.3. Lost poems of Hesiod may have touched on the story Other epic 
of Oedipus; but in his extant work there is only a passing ““'°”* 
reference to the war at Thebes (between Polyneices and 
Eteocles), in which heroes fell, ‘fighting for the flocks of 
Oedipus. Hesiod knows the Sphinx as the daughter of 
Echidna and as the pest of Thebes’. 

But the story of Oedipus was fully treated in some of those 
lost epics which dealt with the Theban cycle of myths. One of 
these was the ‘ Oedipodeia, Οἰδιπόδεια (ἔπη). According to this, 
the four children of Oedipus were not borne by Iocasta, but by 
a second wife, Euryganeia. Pausanias, who follows this account, 
does not know the author of the poem*. It will be observed 
that this epic agrees with the Odyssey in not making Iocasta 
bear issue to Oedipus. It is by Attic writers, so far as we know, 
that she was first described as doing so. Poets or logographers 
who desired to preserve the favour of Dorians had a reason for 
avoiding that version. There were houses which traced their 
line from the children of Oedipus,—as Theron, tyrant of Acragas, 
claimed descent from Thersandros, son of Polyneices*. To 
represent these children as the offspring of an incestuous 


1 Hes. Op. 162: war slew the heroes, τοὺς μὲν ἐφ᾽ ἑπταπύλῳ Θήβῃ... μαρναμένους 
μήλων ἕνεκ᾽ Οἰδιπόδαο. The Sphinx: Zheog. 326, ἡ δ᾽ (Echidna) ἄρα Bik’ ὀλοὴν τέκε, 
Καδμείοισιν ὄλεθρον. The hill near Thebes on which the Sphinx sat was called Φίκειον 
ὄρος. References in lost Hesiodic poems: schol. on 71. 23. 680. 

2 He speaks merely of ὁ τὰ ἔπη ποιήσας ἃ Οἰδιπόδεια ὀνομάζουσι (9. 5. 11). But the 
inscription known as the ‘marmor Borgianum’ refers it to Cinaethon, a Lacedae- 
monian poet who treated epically the Dorian family legends, and who is said to have 
flourished about 775 B.C. Pausanias, however, who quotes Cinaethon on several 
points of genealogy, certainly did not regard the Oedipodeva as his work. 

8 Pind. O/. 2. 35. 


Pindar. 


XIV INTRODUCTION. 


union would have been to declare the stream polluted at its 
source, 

We learn from Proclus that in the epic called the Cyprian 
Lays (Κύπριαλ), which included the preparations for the Trojan 
war, Nestor related ‘the story of Oedipus’ (τὰ περὶ Οἰδίπουν) 
in the course of a digression (ἐν παρεκβάσει) which comprised 
also the madness of Heracles, as well as the story of Theseus 
and Ariadne. ‘This was probably one of the sources used by 
the Attic dramatists. Another source, doubtless more fertile in 
detail, was the epic entitled the Zhebaid (Θηβαΐς), and now 
usually designated as the ‘Cyclic Thebaid, to distinguish it from 
a later epic of the same name by Antimachus of Colophon, the 
contemporary of Euripides. Only about 20 verses remain from 
it. The chief fragment relates to the curse pronounced by 
Oedipus on his sons. They had broken his strict command by 
setting on his table the wine-cups (ἐκπώματα) used by Laius; 
and he invoked a curse upon them :— 


& \ \ en Ψ > ͵ὔ 9 A 
aia δὲ παισὶν ἑοῖσι μετ᾽ ἀμφοτέροισιν ἐπαρὰς 
/ a an 

apyaréas ἠρᾶτο" θεὸν δ᾽ ov λάνθαν᾽ *Epuviv: 

€ “ ς fiw? 2 / ’ 
ὡς οὗ οἱ πατρωϊ ἐνηείῃ φιλότητος 

, 9 ᾿] ΄ ᾽ Μ Lé ’ / 
δάσσαιιτ᾽, ἀμφοτέροισι δ᾽ ἔοι πόλεμός TE μάχαι TE. 


‘And straightway, while his two sons were by, he uttered dire 
curses,—and the Avenging goddess failed not to hear them,—that they 
should divide their heritage in no kindly spirit, but that war and strife 
should be ever between them.’ 


This 7hebatd—tracing the operation of a curse through the 
whole history of the house—must have had an important share 


‘in moulding the conception of the Aeschylean trilogy. 


§ 4. Pindar touches on the story of Oedipus in OZ 2. 42 ff. 
Destiny has often brought evil fortune after good,— 
ἐξ οὗπερ ἔκτεινε Λᾷον μόριμος vids 
συναντόμενος, ἐν δὲ Πυθῶνι χρησθὲν 
παλαίφατον τέλεσσεν" 
ἰδοῖσα δ᾽ ὀξεῖ" ᾿Ερινὺς 
ἔπεφνέ οἱ σὺν ἀλλαλοφονίᾳ γένος ἀρήιον--- 


1 See the Didot ed. of the Cyclic fragments, p. 587. 


st 


INTRODUCTION. XV 


‘from the day when his doomed son met Laius and killed him, and 
accomplished the word given aforetime at Pytho. But the swift Erinys 
beheld it, and slew his warlike sons, each by the other’s sword.’ 


Here the Fury is represented as destroying the sons in direct 
retribution for the parricide, not in answer to the imprecation of 
Oedipus. A fragment of Pindar alludes to the riddle of the 
Sphinx, and he uses ‘the wisdom of Oedipus’ to denote counsel 
wrapped in dark sayings,—since the skill which solves riddling 
speech can weave it’. 


§ 5. The logographers could not omit the story of Oedipus The logo- 
in a systematic treatment of the Theban myths. Hellanicus of wi al 
Mitylene (circ. 450 B.C.) is mentioned by the Scholiast on the 
Phoenissae (61) as agreeing with Euripides in regard to the self- 
blinding of Oedipus*. The contemporary Pherecydes of Leros 
(usually called ‘Athenian’ since Athens was his home) treated 
the legends of Thebes in the fifth of ten books forming a com- 
prehensive survey of Greek tradition®. According to him, Iocasta 
bore two sons to Oedipus, who were slain by the Minyae: but, 
as in the Oedzpodeia, his second wife Euryganeia bore Eteocles 
and Polyneices, Antigone and Ismene. This seems to be the 
earliest known version which ascribes issue to the marriage of 
Tocasta with Oedipus. 


§ 6. However incomplete this sketch may be relatively to The dra- 
the materials which existed in the early part of the fifth century ™4S‘ 
B.C., it may at least serve to suggest the general conditions under 
which Tragedy entered on the treatment of the subject. The 
story of Oedipus, defined in its main features by a tradition older 
than the Odyssey, had been elaborated in the epics of later poets 
and the prose of chroniclers. There were versions differing in 
detail, and allowing scope for selection. While the great outlines 


1 Pind. fr. 62 αἴνιγμα παρθένου | ἐξ ἀγριᾶν ἡνάθων: Pyth. 4. 263 τὰν Οἰδιπόδα 
σοφίαν. Pindar’s elder contemporary Corinna had sung of Oedipus as delivering 
Thebes not only from the Sphinx but also from τὴν Τευμησσίαν ἀλώπεκα---α fox from 
the Boeotian village of Teumessus: but we hear no more of this less formidable 
pest. (Bergk, Poet. Lyr. p. 949.) 

2 Miiller, Frag. Histor. τ. 85. 

% Miiller, 2d. 1. 48. 


Aeschylus. 


xvi INTRODUCTION. 


were constant, minor circumstances might be adapted to the 
dramatist’s chosen view. 

Aeschylus, Sophocles, and Euripides agree in a trait which 
does not belong to any extant version before theirs. Iocasta, not 
Euryganeia, is the mother of Eteocles and Polyneices, Antigone 
and Ismene. They agree also in connecting the doom of the 
two brothers with a curse pronounced by Oedipus. Neither 
the scanty fragments’ which alone represent the Oedipus of 
Euripides, nor the hints in the Phoenzssae, enable us to de- 
termine the distinctive features of his treatment. With regard 
to Aeschylus, though our knowledge is very meagre, it suffices 
at least to show the broad difference between his plan and that 
of Sophocles. 

Aeschylus treated the story of Oedipus as he treated the story 
of Agamemnon. Oedipus became the foremost figure of a 
trilogy which traced the action of an inherited curse in the house 
of Labdacus, even as the Oresteia traced the action of such a 
curse in the house of Pelops. That trilogy consisted of the 
Laius, the Oedipus, and the extant Seven against Thebes; the 
satyric drama being the Sphinx. From the Laius only a few 


1 Nauck Zur. Fragm. 544—561, to which Unger adds Soph. /*. incert. 663, 
Meineke adesfota 107, 309, others adesp. 6. Almost all the verses are commonplaces. 
From fr. 546, 547 I should conjecture that the Creon of Eur. defended himself 
against a charge of treason in a passage parallel with Soph. O. 7. 583—615. One 
fragment of two lines is curious (545): ἡμεῖς δὲ Πολύβου παῖδ᾽ ἐρείσαντες πέδῳ | ἐξομ- 
ματοῦμεν καὶ διόλλυμεν κόρας. Quoting these, the Schol. on Eur. PA. 61 says: ἐν δὲ 
τῷ Οἰδίποδι of Λαΐου θεράποντες ἐτύφλωσαν αὐτόν. This would seem to mean that, 
after the discovery, the old retainers of Laius blinded Oedipus—for the Schol. is 
commenting on the verse which says that he was blinded by Azmse/f. But the tragic 
force of the incident depends wholly on its being the king’s own frantic act. I incline 
to suspect some error on the Scholiast’s part, which a knowledge of the context might 
possibly have disclosed. 

From the prologue of the Phoentssae it appears that Eur. imagined Oedipus to have 
been found on Cithaeron by the ἱπποβούκολοι of Polybus, and taken by them to the 
latter’s wife. The Iocasta of Eur. herself relates in that play how, when the sons of 
Oed. grew up, they held him a prisoner in the palace at Thebes—that the disgrace 
might be hidden from men’s eyes. It was then that he pronounced a curse upon 
them. When they have fallen, fighting for the throne, Iocasta kills herself over their 
bodies, and Creon then expels Oedipus from Thebes. The mutilated ὑπόθεσις to 
the Phoenissae does not warrant us in supposing that the Oenomaus and Chrysippus 
of Eur.,—the latter containing the curse of Pelops on Laius—formed a trilogy with 
his Oedipus. 


INTRODOCTION. XVil 


words remain ; from the Oedipus, three verses; but some general 
idea of the Oedipus may be gathered from a passage in the 
Seven against Thebes (772—791). Oedipus had been pictured 
by Aeschylus, as he is pictured by Sophocles, at the height of 
fame and power. He who had delivered Thebes from ‘the 
devouring pest’ (τὰν ἁρπαξάνδραν κῆρα) was admired by all 
Thebans as the first of men. ‘But when, hapless one, he came 
to knowledge of his ill-starred marriage, impatient of his pain, 
with frenzied heart he wrought a twofold ill’: he blinded 
himself, and called down on his sons this curse, that one day 
they should divide their heritage with the sword. ‘And now I 
tremble lest the swift Erinnys bring it to pass.’ 

Hence we see that the Oedipus of Aeschylus included the 
imprecation of Oedipus upon his sons. This was essential to 
the poet’s main purpose, which was to exhibit the continuous 
action of the Erinnys inthe house. Similarly the Laius doubtless 
included the curse called down on Laius by Pelops, when bereft 
by him of hisson Chrysippus. The true climax of the Aeschylean 
Oedipus would thus have consisted, not in the discovery alone, 
but in the discovery followed by the curse. And we may safely 
infer that the process of discovery indicated in the Seven against 
Thebes by the words ἐπεὶ δ᾽ ἀρτίφρων | éyéveto...yauwv (778) was 
not comparable with that in the play of Sophocles. It was 
probably much more abrupt, and due to some of those more 
mechanical devices which were ordinarily employed to bring 
about a ‘recognition’ on the stage. The Oedipus of Aeschylus, 
however brilliant, was only a link in a chain which derived its 
essential unity from ‘the mindful Erinnys.y. 


§ 7. The Oedipus Tyrannus of Sophocles was not part of a Sophocles 
trilogy, but a work complete in itself. The proper climax of such! 
a work was the discovery, considered in its immediate effects, oo Cevsth 
in its ulterior consequences. Here the constructive art of the ΙΝ 
dramatist would be successful in proportion as the discovery was 
naturally prepared, approached by a process of rising interest, 
and attended in the moment of fulfilment with the most 


astounding reversal of a previous situation. In regard to the Original 
structure of the plot, this is what Sophocles has achieved. Before tures of 


: his plot. 
> ἜΒΗ. - b 


XVlll LNIRODOCTION. 


giving an analysis of his plot, we must notice two features of it 
which are due to his own invention. 

(1) According to previous accounts, the infant Oedipus, 
when exposed on Mount Cithaeron, had been found by herds- 
men, and reared either in Southern Boeotia, or at Sicyon, a place 
associated with the worship of the Eumenides. Sophocles 
makes the Theban herd of Laius give the babe to the herd 
of Polybus, king of Corinth, who rears it as his own. Thus are 
prepared the two convergent threads of evidence which meet in 
the final discovery. And thus, too, the belief of Oedipus con- 
cerning his own parentage becomes to him a source, first of 
anxiety, then of dread, then of hope—in contrast, at successive 
moments, with that reality which the spectators know. 

(2) The only verses remaining from the Oedzpus of Aeschylus 
show that in that drama Oedipus encountered and slew Laius at 
a meeting of three roads near Potniae, a place in Boeotia, on the 
road leading from Thebes to Plataea. At the ruins of this place 
Pausanias saw ‘a grove of Demeter and Persephone’. It ap- 
pears to have been sacred also to those other and more terrible 
goddesses who shared with these the epithet of aotvtas,—the 
Eumenides (ποτνιάδες θεαί, Eur. Or. 318). For the purpose of 
Aeschylus, no choice of a scene could have been more fitting. 
The father and son, doomed by the curse in their house, are 
brought together at a spot sacred to the Erinnyes :— 

ἐπῆμεν τῆς ὁδοῦ τροχήλατον 
σχιστῆς κελεύθου τρίοδον, ἔνθα συμβολὰς 
τριῶν κελεύθων Ἰ]οτνιάδων ἡμείβομεν", 


‘We were coming in our journey to the spot from which three high- 
roads part, where we must pass by the junction of triple ways at Potniae.’ 


But for Sophocles this local fitness did not exist. For him, 
the supernatural agency which dominates the drama is not that 
of the Furies, but of Apollo. He transfers the scene of the 
encounter from the ‘three roads’ at Potniae to the ‘three roads’ 
near Daulia® in Phocis. The ‘branching ways’ of Potniae can no 


1 ἄλσος Δήμητρος καὶ Κόρης, 9. 8. 1. 
2 Aesch. fr. 173 (Nauck). 
3 Dauliswasthe Homeric form of the name, Daz/ia the post-homeric (Strabo 9. 423). 


INTRODUCTION. X1X 


longer be traced. But in the Phocian pass a visitor can still feel 
how the aspect of nature is in unison with the deed of which 
Sophocles has made it the theatre’. This change of locality has 
something more than the significance of a detail. It symbolises 
the removal of the action from the control of the dark Avenging 
Powers to a region within the influence of that Delphian god who 
is able to disclose and to punish impurity, but who will also give 
final rest to the wanderer, final absolution to ΠΗ; weary mourner 
of unconscious SIN. - rata es fe hee OY pf μι 9 

§ 8. The events which had preceded the action of the Oedzpus Supposed 
Tyrannus are not set forth, after the fashion of Euripides, in a ae Ee the 
formal prologue. They have to be gathered from incidental hints plot. 
in the play itself. It is an indispensable aid to the full compre- 
hension of the drama that we should first connect these hints into 
a brief narrative of its antecedents as imagined by Sophocles. 

Laius, king of Thebes, being childless, asked the oracle of 
Apollo at Delphi whether it was fated that a son should be born 
to him. The answer was, ‘I will give thee a son, but it is doomed 
that thou leave the sunlight by the hands of thy child: for thus 
hath spoken Zeus, son of Cronus, moved by the dread curse of 
Pelops, whose own son (Chrysippus) thou didst snatch from him ; 
and he prayed all this for thee.’ When a son was indeed born 
to Laius of Iocasta his wife, three days after the birth he caused 
it to be exposed in the wilds of Mount Cithaeron. An iron pin 
was driven through the feet of the babe, fastening them together 
—that, if perchance it should live to be found by a stranger, he 
might have the less mind to rear a child so maimed ; from which 
maiming the child was afterwards called Oedipus’. 

The man chosen to expose the babe received it from the 
hands of the mother, Iocasta herself, with the charge to destroy 
it. This man was a slave born in the house of Laius, and so be- 
longing to the class of slaves whom their masters usually treated 
with most confidence. He was employed in tending the flocks 


1 See the note on verse 733. 

2 The incident of the pierced feet was tidently invented to explain the name 
Οἰδίπους (*Swellfoot,’ as Shelley renders it). In v. 397 ὁ μηδὲν εἰδὼς Οἰδίπους suggests 
a play on οἶδα. 


b2 


XX INTIRODOCTION. 


of Laius on Mount Cithaeron, where they were pastured during 
the half-year from March to September. 

In the glens of Cithaeron he had consorted with another 
herdsman, servant to Polybus, king of Corinth. Seized with 
pity for the babe, the Theban gave it to this herdsman of Polybus, 
who took it to Corinth. Polybus and his wife Meropé were 
childless. They reared the child as their own; the Corinthians 
regarded him as heir to the throne; and he grew to a man’s estate 
without doubting that he was the true son of the Corinthian 
king and queen. 

But one day it chanced that at a feast a man heated with 
wine threw out a word which sank into the young prince’s mind; 
he questioned the king and queen, whose resentment of the 
taunt comforted him; yet he felt that a whisper was creeping 
abroad ; and he resolved to ask the truth from Apollo himself at 
Delphi. Apollo gave him no answer to the question touching 
his parentage, but told him these things—that he was doomed to 
slay his father, and to defile his mother’s bed. 

He turned away from Delphi with the resolve never again to 
see his home in Corinth; and took the road which leads east- 
ward through Phocis to Boeotia. 

At that moment Laius was on his way from Thebes to 
Delphi, where he wished to consult the oracle. He was not 
escorted by the usual armed following of a king, but only by 
four attendants. The party of five met Oedipus at a narrow 
place near the ‘ Branching Roads’ in Phocis; a quarrel occurred; 
and Oedipus slew Laius, with three of his four attendants. The 
fourth escaped, and fled to Thebes with the tale that ὦ band of 
robbers had fallen upon their company. This sole survivor was 
the very man who, long years before, had been charged by Laius 
and Iocasta to expose their infant son on Cithaeron. 

The Thebans vainly endeavoured to find some clue to the 
murderer of Laius. But, soon after his death, their attention was. 
distracted by a new trouble. The goddess Hera—hostile to 
Thebes as the city of her rival Semelé—sent the Sphinx to 
afflict it_—a monster with the face of a maiden and the body of a 
winged lion; who sat on a hill near Thebes (the Φίκειον ὄρος), 
and chanted a riddle. ‘What is the creature which is two-footed, 


INTRODUCTION. XX1 


three-footed, and four-footed; and weakest when it has most 
feet?’ Every failure to find the answer cost the Thebans a life. 
Hope was deserting them; even the seer Teiresias had no help 
to give; when the wandering stranger, Oedipus, arrived. He 
solved the enigma by the word man: the Sphinx hurled herself 
from a rock; and the grateful Thebans gave the vacant throne 
to their deliverer as a free gift. At the same time he married 
Iocasta, the widow of Latus, and sister of Creon son of Menoeceus. 

The sole survivor from the slaughter of Laius and his com- 
pany was at Thebes when the young stranger Oedipus ascended 
the throne. The man presently sought an audience of the queen 
Iocasta, knelt to her, and, touching her hand in earnest supplica- 
tion, entreated that he might be sent to his old occupation of 
tending flocks in far-off pastures. It seemed a small thing for so 
old and faithful a servant to ask; and it was readily granted. 

An interval of about sixteen years may be assumed between 
these events and the moment at which the Oedipus Tyrannus 
opens. Iocasta has borne four children to Oedipus: Eteocles, 
Polyneices, Antigone, Ismene. Touches in the closing scene of 
the play forbid us to suppose that the poet imagines the daugh- 
ters as much above the age of thirteen and twelve respectively. 
Oedipus has become thoroughly established as the great king, 
the first of men, to whose wisdom Thebans turn in every trouble. 

And now a great calamity has visited them. A blight is 
upon the fruits of the earth; cattle are perishing in the pastures; 
the increase of the womb is denied; and a fiery pestilence is 
ravaging the town. While the fumes of incense are rising to 
the gods from every altar, and cries of anguish fill the air, a body 
of suppliants—aged priests, youths, and children—present them- 
selves before the wise king. He, if any mortal, can help them. 
It is here that the action opens. 


§ 9. The drama falls into six main divisions or chapters. ee 
The following analysis exhibits in outline the mechanism of the ἣν !% 
plot, which deserves study. 


I. Prologue: 1—150. Oedipus appears as the great prince 
whom the Thebans rank second only to the gods. He pledges 


xxii | INTRODUCTION. 


himself to relieve his afflicted people by seeking the murderer of 
Laius. 

Parodos: 1§51—215. The Chorus bewail the pestilence and 
invoke the gods. 


II. First Episode: 216—462. Ocdipus publicly invokes a 
solemn curse upon the unknown murderer of Laius. At Creon’s 
suggestion he sends for the seer Teiresias, who refuses to speak, 
but finally, stung by taunts, denounces Oedipus himself as the 
slayer. 


2 {. 


First a eerie 463—512. The Chorus forebode that the 
unknown murderer is doomed; they refuse to believe the 
unproved charge brought by the seer. 


III. Second Episode: 513—862. Creon protests against the 
suspicion that he has suborned Teiresias to accuse Oedipus. 
Oedipus is unconvinced. lIocasta stops the quarrel, and Creon 
departs. Oedipus then tells her that he has been charged with 
the murder of Laius. She replies that he need feel no dis- 
quietude. Laius, according to an oracle, was to have been slain 
by his own son; but the babe was exposed on the hills; and 
Laius was actually slain by vodbders, at the meeting of three roads. 

This mention of three roads (v. 716) strikes the first note of 
alarm in the mind of Oedipus. 

He questions her as to (1) the place, (2) the time, (3) the per- 
son and the company of Laius. All confirm his fear that ἦε 
has unwittingly done the deed. 

He tells her his whole story—the taunt at Corinth—the visit 
- to Delphi—the encounter in Phocis. But he has still one hope. 
The attendant of Laius who escaped spoke of vodders, not of one 
robber. 

Let this survivor—now a herdsman—be summoned and 
questioned. 


Second Stasimon: 863—910. The Chorus utter a prayer 
against arrogance—such as the king’s towards ‘Creon; and 
impiety—such as they find i in locasta’s mistrust of oracles. 





IV. Third Be 911—1085. A messenger from Corinth 
announces that Polybus is dead, and that Oedipus is now king 


INTRODUCTION. Xxiil 


designate. Jocasta and Oedipus exult in the refutation of the 
oracle which had destined Oedipus to slay his sire. 

But Oedipus still dreads the other predicted horror—union 
with his mother. 

The messenger, on learning this, discloses that Polybus and 
Meropé were not the parents of Oedipus. The messenger 
himself, when a herdsman in the service of Polybus, had found 
the infant Oedipus on Cithaeron, and had brought him to 
Corinth. Yet no—not found him; had recezved him from another 
herdsman (v. 1040). 

Who was this other herdsman? The Corinthian replies :— 
He was said to be one of the people of Latus. 

Iocasta implores Oedipus to search no further. He answers 
that he cares not how lowly his birth may prove to be—he will 
search to the end. With a cry of despair, Iocasta rushes away. 


tell that Oedipus will prove to be a native of the land—perchance 
of seed divine. 


V. Fourth Episode: 1110—1185. The Theban herdsman 
is brought in’. 

‘There,’ says the Corinthian, ‘is the man who gave me the 
child.” Bit by bit, the whole truth is wrung from the Theban. 
‘The babe was the son of Laius; the wife of Latus gave him to 


me, Oedipus knows all, and with a shriek of misery he rushes 
away. 


Fourth Stasimon: 1186—1222. The Chorus _bewail the. 
great king's | fall. 


VI. Exodos: 1223—1530. A messenger from the house 
announces that Iocasta has hanged herself, and that Oedipus 
has put out his eyes. Presently Oedipus is led forth. With 
passionate lamentation he beseeches the Chorus of Theban 
Elders to banish or slay him. 


1 The original object of sending ior him had been to ask,—‘ Was it the deed of 
several men, or of one?’—a last refuge. But he is not interrogated on that point. 
Voltaire criticised this as inconsistent. It is better than consistent; it is natural. A 
more urgent question has thrust the other out of sight. 


The 
method of 
discovery. 


Aristotle’s 
criticisms. 


XXIV INTRODUCTION. 


Creon comes to lead him into the house. Oedipus obtains 
from him a promise of care for his young daughters; they are 
presently brought to their father, who takes what he intends to 
be a last farewell. For he craves to be sent out of the land; 
but Creon replies that Apollo must pronounce. 

As Creon leads Oedipus within, the Chorus speak the 
closing words: No mortal must be called happy on this side 
death. 


With reference to the general structure of the plot, the first 
point to observe is the skill with which Sophocles has managed 
those two threads of proof which he created by his invention of 
the second herdsman. 


We have :— 


(1) The thread of evidence from the reported statement 
of the Theban herdsman as to the place of the murder, in con- 
nection with Iocasta’s statement as to the time, the person of 
Laius, and the retinue. This tends to show that Oedipus has 
slain Latus—deing presumably in no wise his kinsman. The 
proof of Oedipus having slain Laius is so far completed at 
754 (αἰαῖ, τάδ᾽ ἤδη διαφανῆ) as to leave no longer any moral 
doubt on the mind of Oedipus himself. * 

(2) The thread of evidence from the Corinthian, showing, 
in the first instance, that Oedipus is zo¢ the son of Polybus and 
Meropé, and so relieving him from the fear of parricide and 
incest. Hence the confident tone of Oedipus (1076 ff.), which so 
powerfully contrasts with the despair of Iocasta: se has known 
the worst from v. 1044. ὌΝ 


(3) The convergence of these two threads, when the Theban 
herdsman is confronted with the Corinthian. This immediately 
follows the moment of relief just noticed. It now appears that 
the slayer of Laius has a/so committed parricide and incest. 


§ 10. The frequent references of Aristotle to the Oedipus 
Tyrannus indicate its value for him as a typical masterpiece, 
though the points for which he commends it concern general 
analysis of form, not the essence of its distinctive excellence. 
The points are these :— 


INTRODUCTION. XXV 


AA 

1. The ‘recognition’ (ἀναγνῶρισις) is contrived in the best 
way; Zé, it is coincident with a reversal of fortunes {περὲξ 
πέτεια). ᾿ 

2. This reversal is peculiarly impressive, because the 
Corinthian messenger had come to bring tidings of the honour 
in store for Oedipus. 

3. Ocdipus is the most effective kind of subject for such a 
reversal, because he had been (4) great and glorious, (0) mo¢ 
preeminently virtuous or just, (c) and, again, one whose reverses 
are not due to crime, but only to unconscious error. 

4. ~The story is told in such a manner as to excite pity and 
terror by hearing without seeing (as in regard to the exposure of 
the child, the killing of Laius, the death of Iocasta). 

5. If there is any improbability in the story, this is not in 
the plot itself (ἐν tots πράγμασιν), but in the supposed antece- 
dents (ἔξω τῆς τραγῳδίας). 

In this last comment, Aristotle indicates a trait which Improba- 
5 bility in 
is certainly open to criticism—the ignorance of Oedipus as ine inte- 
to the story of Laius. He knows, indeed, the name of his cedents. 
predecessor—though Creon does not think it unnecessary to 
remind him of the name (103). He also knows that Laius had 
met a violent death: but he does not know whether this had 
befallen at Thebes, or in its neighbourhood, or abroad (100---Ἴ 1 3). 

Nor does he know that Laius was reported to have been slain by 
robbers, and that only one of his followers had escaped (116— 
123): and he asks if no search had been made at the time 
(128, 566). JIocasta, who has now been his wife for many years, 
tells him, as if for the first time, the story of the oracle given to 
Laius, and he tells her the story of his own early fortunes— 
though here we need not press the fact that he even names to 
her his Corinthian parents: that may be regarded as merely 
a formal preface to a connected narrative. It may be conceded 
that the matters of which Oedipus is supposed ignorant were 
themes of which Iocasta, and all the persons about the new king, 
might well have been reluctant to speak. Still it is evident that 
the measure of past reticence imagined, both on their part and 
on his, exceeds the limit of verisimilitude. The true defence of 
this improbability consists in frankly recognising it. Exquisite 


The char- 


acters. 


XXvl INTRODUCTION. 


as was the dramatic art exercised within the scope of the action 
(ἐν Tots πράγμασι), this art was still so far naive as to feel no 
offence at some degree of freedom in the treatment of that 
which did not come within the framework,—of that which, in 
Aristotle's phrase, lay ‘outside the piece,’ ἔξω τῆς τρωγῳδίας. It 
is as if a sculptor neglected to remove some roughness of sup- 
port or environment which, he felt, would not come into account 
against the effect of a highly finished group. 


δ΄ 11. A drama is itself the only adequate commentary on 
its persons. It makes them live for us, or it does not. If we 
submit them to ethical analysis, this may be interesting to ws, 
and instructive to those who have not seen or read the piece. 
But, for a spectator or reader of the play, the men and women 
must be those whom he finds there. When we personally know 
a character in real life, another’s estimate of it is seldom more 
than a key to his point of view—rarely a mental light which we 
feel that we can appropriate. And it may be permitted to 
Say in passing that this is a reason why the reviving taste for 
good drama—a result for which, in this country, so much is due 
to Mr Irving—seems likely to aid in correcting a literary fault 
of the day which is frequently acknowledged—the tendency to 
adopt ready-made critical estimates of books which the adopter, 
at least, has not read. No one who sees a play can help forming 
some impression of 4zs own about the characters. If he reports 
it honestly, that is criticism; not necessarily good, but not 
sham. To any one who reads this play of Sophocles with 
even moderate attention and sympathy, how living is Oedipus! 
Common experience proves so much; but almost every reader 
will probably feel that by no attempt at analysis or description 
could he enable another to see precisely 4zs Oedipus :-no, though 
the effort should bring out ‘a point or two as yet unseized by 
the Germans. The case is somewhat different, however, when 
a particular reading of certain characters in a play is the ground 
for the attribution to it of a tendency; then it is useful to 
inquire whether this reading is right—whether, that is, these 
persons of the drama do indeed speak and act in the tone 
ascribed to them. 


INTRODUCTION. XXVIl 


And certainly one of the most interesting questions in the Is 
Oedipus Tyrannus concerns the intellectual position of Oedipus Saha 
and Iocasta towards that divine power of which the hand is laid unbelief? 
so heavily upon both. Sophocles had found in human nature 
itself the sanction of ‘the unwritten laws, and the seal of faith 
in a beneficence immortal and eternal; but his personal attitude 
towards the ‘sceptical’ currents of thought in his age was never, 
so far as we can judge, that of admonitory protest or dogmatic 
reproof. It was his temperament to look around him for 
elements of conciliation, to evoke gentle and mediating influ- 
ences, rather than to make war on the forces which he regarded 
as sinister :—it might be said of him, as of a person in one of 
his own plays, οὔτοι συνέχθειν ἀλλὰ συμφιλεῖν ἔφυ. But is 
there any reason to think that the Oedipus Tyrannus marks 
a moment when this mind—‘which saw life steadily, and saw 
it whole’—was partly shaken in its self-centred calm by the 
consciousness of a spiritual anarchy around it which seemed 
fraught with ultimate danger to the cohesion of society, and 
that a note of solemn warning, addressed to Athens and to 
Greece, is meant to be heard throughout the drama? Our 
answer must depend upon the sense in which we conceive 
that he places Oedipus or Iocasta at issue with religion. 


§ 12. As regards Oedipus, it might be said that, in this par- Oedipus. 
ticular aspect, he is a modern character, and more especially, 
perhaps, a character of the nineteenth century. The instinct of 
reverence for the gods was originally fundamental in his nature: 
it appears in the first act of his manhood—the journey to 
Delphi. Nor did he for a moment mistrust the gods because the 
doom assigned to him was bitter. Then he achieved a great in- 
tellectual success, reached the most brilliant prosperity, and was 
ranked by his fellow-men as second to the gods alone. He is 
not spoiled by his good fortune. We find him, at the opening 
of the play, neither arrogant nor irreverent; full, rather, of 
tenderness for his people, full of reverence for the word of 
Apollo. Suddenly, however, the prophet of Apollo denounces 
him. Instantly his 8. appeal is to the intellect. If it comes to 
that, what claim has any other human mind to interpose between 





Tocasta. 


XXVI1 INTRODUCTION: 


his mind and Heaven? Is he not Oedipus, who silenced the 
Sphinx? Yes, but presently, gradually, his own mind begins to 
argue on the other side. No one is so acute as he, and of course 
he must be the first to see any facts which tell against himself. 
And now, when he is face to face with the gods, and no prophet 
stands between, the instinct of reverence inborn in his noble 
nature finds voice in the prayer, ‘ Forbid, forbid, ye pure and 
awful gods, that I should see that day!’ After varying hopes 
and fears, his own mind is convinced of the worst. Reason, 
which had been the arbiter of faith, now becomes the inexorable 
judge of sin, the most instant and most rigorous claimant for 
his absolute abasement before the gods. 


*~ §13. Plainly, it would be a misreading to construe the fate 
of Oedipus as a dramatic nemesis of impiety; but the case of 
Iocasta is at first sight less clear. She, at least, is one who 
openly avows scorn for oracles, and urges her lord to share it. 
It miay often be noticed—where the dramatist has known how 
to draw from life—that the true key-note of a dominant mood 
is struck by a short utterance on which no special emphasis is 
thrown, just as, in life itself, the sayings most truly significant 
of character are not always long or marked. For Iocasta, such 
a key-note is given in the passage where she is telling Oedipus 
that a response from the Delphian temple had warned Laius 
that he was destined to be slain by the child whom she bore to 
him. ‘An oracle came to Laius once—/ wll not say from 
Phoebus himself, but from his ministers’ (ν. 712). Tocasta 
thoroughly believes in the power of the gods to effect their 
will (724),—to punish or to save (921). But she does not be- 
lieve that any mortal—be he priest or prophet—is permitted by 
them to read the future. Had not the Delphian priests doomed 
her to sacrifice her first-born child,—and this, without saving 
the life of her husband, Latus? The iron which years ago had 
entered into the soul of the wife and mother has wrought in 
her a result similar to that which pride of intellect has produced 
in Oedipus. Like Oedipus, she still believes in the wise omni- 
potence of the gods; like him also, she is no longer prepared to 
accept any mortal interpreter of their decrees. Thus are the 


INTRODUCTION. XXIX 


two foremost persons of this tragedy separated from the offices 
of human intercession, and directly confronted in spirit—one by 
his self-reliance, the other by her remembered anguish—with 
the inscrutable powers which control their fate. It is as a study 
of the human heart, true for every age, not as a protest against 
tendencies of the poet’s own, that the Oedipus Tyrannus illustrates 
the relation of faith to reason. 


§ 14. The central figure of the drama is brought into clearer Teiresias. 
relief by the characters of Teiresias and Creon. Teiresias exists “°°™ 
only for the god whom he serves. Through him Apollo speaks. 
As opposed to Oedipus, he zs the divine knowledge of Apollo, 
opposed to human ignorance and blindness. While ‘the servant 
of Loxias’ thus stands above the king of Thebes, Creon stands 
below him, on the humbler but safer ground of ordinary 
humanity. Creon is shrewd, cautious, practical, not sentimental 
or demonstrative, yet of a fervid self-respect, and with a strong 
and manly kindliness which comes out in the hour of need’. It 
might be said that the Creon of the Oedipus Tyrannus embodies 
a good type of Scottish character, as the Creon of the Aztigone 
—an earlier sketch—is rather of the Prussian .type, as it is 
popularly idealised by some of its neighbours. Teiresias is the 
gauge of human insight matched against divine; Creon, of 
fortune’s heights and depths, compared with the less brilliant 
but more stable lot of commoner men. ‘Crave not to be master 
in all things; for the mastery which thou didst win hath not 
followed thee through life’—are his words to Oedipus at the 
end; and his own position at the moment exemplifies the 
sense in which ‘the god ever gives the mastery to the middle 
State. 


§ 15. There is no external evidence for the time at which Supposed 


ὶ ° references 
the Oedipus Tyrannus was first acted. Internal evidence warrants {6 ¢on- 


temporary 
events. 
1 Lest it should be thought that in the note on p. 77 the harsher aspect 


of Creon’s character is unduly prominent, I may observe that this note relates 
to vv. 512—862, and deals with Creon only as he appears there. The scene which 
begins at v. 1422—and more especially vv. 1476 f—must of course be taken into 
account when we offer, as here, a more general estimate of the character. 

2 παντὶ μέσῳ τὸ κράτος θεὸς ὥπασεν, Aesch. Zum. 528. 


Alleged 
defeat of 
the play. 


XXX INTRODUCTION. 


the belief that it was composed after the Aztigone, and before 
the Oedipus Coloneus. The probable limits thus indicated might 
be roughly given as about 439—412 B.c. More than this we 
cannot say. Modern ingenuity has recognised Pericles in 
Oedipus,—the stain of Alcmaeonid lineage in his guilt as the 
slayer of Lafus,—the ‘Dorian war, and a pestilence therewith’ 
in the afflictions of Thebes. This allegorical hypothesis need 
not detain us. But it may be well briefly to remark the differ- 
ence, for drama, between association of ideas and direct allusion. 
If Sophocles had set himself to describe the plague at Athens as 
he had known it, it might have been held that, in an artistic 
sense, his fault was graver than that of Phrynichus, when, by 
representing the capture of Miletus, he ‘reminded the Athenians 
of their own misfortunes.’ If, however, writing at a time sub- 
sequent to the pestilence which he had survived, he wished to 
give an ideal picture of a plague-stricken town, it would have 
been natural and fitting that he should borrow some touches 
from his own experience. But the sketch in the play is far too 
slight to warrant us in saying that he even did this; perhaps 
the reference to the victims of pestilence tazuting the air (@avat- 
adopa v. 180) is the only trait that might suggest it. Thucydides 
(1. 50), in describing the plague of 430 B.C., notices the number 
of the unburied dead. The remarks just made apply equally to 
the supposed allusion in vv. 883 ff. to the mutilation of the 
Hermae (see the note on 886). 

A tradition, dating at least from the 2nd century B.C’, 
affirmed that, when Sophocles produced the Oedipus Tyrannus, 
he was defeated for the first prize by Philocles—a poet of 
whose work we know nothing. Philocles was a nephew of 
Aeschylus, and, as Aristeides observes’, achieved an honour which 


‘l The words in the prose ὑπόθεσις (given on p. 4) are simply, ἡττηθέντα ὑπὸ 
Φιλοκλέους, ὥς φησι Δικαίαρχος. The Dicaearchus who wrote ὑποθέσεις τῶν Evpurldov 
καὶ Σοφοκλέους μύθων has been generally identified with Dicaearchus of Messana, the 
Peripatetic, a pupil of Aristotle and a friend of Theophrastus. We might place 
his ‘floruit,’ then, somewhere about 310 B.c.; there are indications that he survived 
296 B.c. If, on the other hand, the ὑποθέσεις were ascribed to the grammarian 
Dicaearchus of Lacedaemon, a pupil of Aristarchus, this would bring us to about 
140 B.C, 

2 11. 256. 


INTRODUCTION. ΧΧΧῚ 


had been denied to his uncle. The surprise which has been 
expressed by some modern writers appears unnecessary ; the 
composition of Philocles was probably good, and it has never 
been held that the judges of such prizes were infallible. 

§ 16. The name of an actor, once famous in the chief part of τς actor 
this play, is of interest also on more general grounds. Polus,a ©” 
native of Aegina, is said to have been the pupil of another tragic 
actor, Archias of Thurii4. He flourished, then, in the middle or 
latter part of the 4th century B.c.—only some 50 or 60 years 
after the death of Sophocles. Physically well-gifted, and of ver- 
satile grace, he was equally successful as Oedipus the King, and 
in the very different but not less difficult part of Oedipus at Co- 
lonus*. Like the poet whose masterpieces he interpreted, he 
enjoyed a vigorous old age; and it is recorded that, at seventy, 
he acted ‘eight tragedies in four days’. In the Electra of 
Sophocles, an urn, supposed to contain the ashes of Orestes, is 
placed in the hands of his sister, who makes a lament over it. 
Polus once acted Electra not long after the death of his son. 
An urn, containing the youth’s ashes, was brought from the 
tomb; the actor received it, and, on the scene, suffered a natural 
grief to have vehement course *. 


1 Plut. Dem. 28 τοῦτον δὲ [Archias] Θούριον ὄντα τῷ γένει λόγος ἔχει τραγῳδίας 
ὑποκρίνεσθαί ποτε, καὶ τὸν Αἰγινήτην Πῶλον, τὸν ὑπερβαλόντα τῇ τέχνῃ 
πάντας, ἐκείνου γενέσθαι μαθητὴν ioropodow.—Schaefer (Dem. u. 5. Zeit, 1. 219 1.) 
and A. Miiller (Gr. Biihnenalterthiimer, p. 186, n. 3) distinguish this Polus from 
an elder, whom they place in the time of Socrates. They seem mistaken. In Plut. 
περὶ φιλίας, fr. 16 (p. 833 ed. Wyttenbach), Socrates is quoted, and then Polus is 
mentioned; but not as contemporary with Socrates. As to Lucian calling Polus 
ὁ Σουνιεύς, see below, note 4. 

, 2 Stobaeus Mori/. p. 522 (XCVII. 28), in an extract from the mporperrixal 
ὁμιλίαι of Arrian: ἢ οὐχ ὁρᾷς ὅτι οὐκ εὐφωνότερον οὐδὲ ἥδιον ὁ Πῶλος τὸν τύραννον 
Οἰδίποδα ὑπεκρίνετο ἢ τὸν ἐπὶ Κολωνῴ ἀλήτην καὶ πτωχόν ; (οὐδὲ ἥδιον is Gaisford’s 
emendation of οὐδὲν dv’ ὧν.) 

3 Plut. Mor. 785 α Πῶλον δὲ τὸν τραγῳδὸν ᾿Ερατοσθένης καὶ Φιλόχορος ἱστοροῦσιν 
ἑβδομήκοντα ἔτη “γεγενημένον ὀκτὼ τραγῳδίας ἐν τέτταρσιν ἡμέραις διαγωνίσασθαι μικρὸν 
ἔμπροσθεν τῆς τελευτῆς. 

4 Aulus Gellius 7. 5 Histrio in terra Graecia fuit fama celebri qui gestus et 
vocis claritudine ceteris antestabat....Polus lugubri habitu Electrae indutus ossa 
atque urnam a sepulcro tulit filii, et quasi Orestis amplexus opplevit omnia non 
simulacris neque imitamentis sed luctu atque lamentis veris et spirantibus. 

Lucian Jupp. Tragoed. ὃ 3 οὐχ ὁρῶ... ἐφ᾽ ὅτῳ Tddos ἢ ᾿᾿Αριστόδημος ἀντὶ Διὸς 
᾿ ἡμῖν ἀναπέφηνας. Id. Menippus § 16 (on the contrast between the life of actors 


Signific- 
ance of 
the story. 


ΧΈΧΙΙ INTRODUCTION, 


Little as such an incident may accord with modern feeling or 
taste, it is at least of very clear significance in relation to the 
tone of the Attic stage as it existed for a generation whose 
grandfathers were contemporary with Sophocles. Whether the 
story was true or not, it must have been conceived as possible. 
And, this being so, nothing could better show the error of sup- 
posing that the old Greek acting of tragedy was statuesque in 
a cold or rigid sense,—in a sense excluding declamation and 
movement suitable to the passions which the words expressed. 
Play of feature, indeed, was excluded by the use of masks; but 
this very fact would have increased the need for appropriate 
gesture. The simple grouping—as recent revivals have helped 
us to feel—must have constantly had a plastic beauty rarely 
seen on our more crowded stage’; but it is inconceivable, and 
the story just noticed affords some direct ground for denying, 
that this result was obtained at any sacrifice of life and truth in 
the portrayal of emotion. Demosthenes tells us that some of 
the inferior tragedians of his time were called ‘ranters’* It 
might be said, of course, that this indicates a popular preference 
for an undemonstrative style. But it might with more force be 
replied that ‘ranting’ is not a fault which a coldly ‘statuesque’ 
tradition would have generated. 
on and off the stage) ἤδη δὲ πέρας ἔχοντος τοῦ δράματος, ἀποδυσάμενος ἕκαστος αὐτῶν 
τὴν χρυσόπαστον ἐκείνην ἐσθῆτα καὶ τὸ προσωπεῖον ἀποθέμενος καὶ καταβὰς ἀπὸ 
τῶν ἐμβατών πένης καὶ ταπεινὸς περιέρχεται, οὐκέτ᾽ ᾿Αγαμέμνων ὁ ᾿Ατρέως οὐδὲ 
Κρέων ὁ Μενοικέως, ἀλλὰ Πῶλος Χαρικλέους Σουνιεὺς ὀνομαζόμενος ἢ 
Σάτυρος Θεογείτονος Μαραθώνιος. [‘Polus, son of Charicles, of Sunium,’ is not 
inconsistent with τὸν Αἰγινήτην in Plut. Dem. 28, for the great actor may have 


been a native of Aegina who was afterwards enrolled in the Attic deme of Sunium.] - 
Id. De mercede conduct. § 5 τοῖς τραγικοῖς ὑποκριταῖς...οἱ ἐπὶ μὲν τῆς σκηνῆς ᾿Αγα- 
μέμνων ἕκαστος αὐτῶν ἢ Κρέων ἢ αὐτὸς Ἡρακλῆς εἰσιν, ἔξω δὲ ἸΠώλος ἢ ᾿Αριστόδημος, 
ἀποθέμενοι τὰ προσωπεῖα, γίγνονται. 

The Aristodemus coupled by Lucian with Polus is the actor mentioned by 
Aeschines and Demosthenes; the latter specially notices that he and Theodorus had 
both often acted the Antigone of Sophocles (or. 19. § 246): Satyrus is the comic actor 
mentioned by the same orators (Aeschin. 2. § 156, Dem. or. 19. § 193). Thus we 
see how, in later Greek literature, Polus had become one of a small group of names 
typical of the best histrionic art of the classical age. 

1 On the sense in which a ‘plastic’ character is common to Greek Sculpture, 
Tragedy, and Oratory, cp. my Aétic Orators, vol. 1. pp. xcviii—ciii. 

2 Dem. or. 18. § 262 μισθώσας αὑτὸν τοῖς βαρυστόνοις ἐπικαλουμένοις ἐκείνοις 
ὑποκριταῖς, Σιμύλῳ καὶ Σωκράτει, ἐτριταγωνίστεις. 


INTRODUCTION. XXX11l 


δ 17. The story of Oedipus was one of a few subjects which Other 


the Greek dramatists never tired of handling. Some eight or ΕἸΣῚ: on 


nine tragedies, entitled Oedzpus, are known by the names of subject. 
their authors, and by nothing else’. Plato, the poet of the Old 
Comedy, wrote a Laius, which was perhaps a parody of the 
Aeschylean play; and the Middle Comedy was indebted to 
Eubulus for an Oedipus from which a few verses are left—a 
travesty of the curse pronounced upon the unknown criminal’. 
Julius Caesar, like the younger Pitt, was a precocious dramatist, 
and Oedipus was his theme*®. The self-blinded Oedipus was a 
part which Nero loved to act*, and the last public recitation 
which he ever gave, we are told, was in this character. The 
Greek verse at which he stopped is on record: whose it was, we 
know not®. Of all the Greek versions, not one remains by which 
to gauge the excellence of Sophocles. But the literatures of 
other languages make some amends. 

Nothing can better illustrate the distinctive qualities of the 
Sophoclean Oedipus than to compare it with the treatment of 
the same theme by Seneca, Corneille, Dryden and Voltaire. So 
far as the last three are concerned, the comparison has a larger 


‘ An Οἰδίπους by the Carcinus whom Aristophanes ridicules is quoted by Arist. 
het. 5. τό. 11. Xenocles is said to have been victorious, with a series of plays 
including an Οἰδίπους, against Euripides, one of whose pieces on that occasion was 
the Zvroades, probably in 415 B.c. An Οἰδίπους is also ascribed to Achaeus (Nauck 
Trag. fr. p. 584), Theodectes (p. 623), and, more doubtfully, to Diogenes of Sinope 
(p. 627); also by Suidas to Philocles, and to each of two poets named Nicomachus 
(one of Athens, the other of the Troad). 

2 Meineke Com. Frag. pp. 231 (Plato), Eubulus (451). Of the latter’s five 
verses, the last three are—éoris δ᾽ ἐπὶ δεῖπνον ἢ φίλον τιν᾽ ἢ ξένον | καλέσας ἔπειτα 
συμβολὰς ἐπράξατο, | φυγὰς γένοιτο μηδὲν οἴκοθεν λαβών. It seems quite possible, 
as has been suggested, that Eubulus was parodying verses from the Οαϊζφης of 
Euripides. 

3 Sueton. /u/. Caes. 56 Feruntur et a puero et ab adulescentulo quaedam scripta, 
ut laudes Herculis, tragoedia Oedipus. : 

4 Sueton. Vero 21 Tragoedias quoque cantavit personatus. Inter cetera cantavit 
Canacen parturientem, Orestem matricidam, Oedipodem excaecatum, Herculem 
insanum. 

5 7b. 46 Observatum etiam fuerat novissimam fabulam cantasse eum [Neronem] 
publice Oedipum exsulem, atque in hoc desisse versu, οἰκτρῶς θανεῖν μ᾽ ἄνωγε 
σύγγαμος πατήρ. Dio Cassius (63. 28) also quotes the verse as one on which Nero’s 
mind dwelt: τὸ ἔπος ἐκεῖνο συνεχῶς ἐνενόει. : 


15:1" : oe 


The 
Oedipus 
of Seneca. 


XXXIV TNERODUIEC TION. 


value. The differences between the spirit of the best Greek 
Tragedy and that of modern drama are not easily expressed in 
formulas, but can be made clearer by a particular example. 
Perhaps the literature of drarna hardly affords any example so 
apposite for this purpose as the story of Oedipus. 

§ 18. Seneca has followed, and sometimes paraphrased, 
Sophocles with sufficient fidelity to heighten the contrast be- 
tween the original and the rhetorical transcript. For the com- 
parative student of drama, however, the Roman piece is by no 
means devoid of instruction or of interest. Seneca’s plot diverges 
from that of Sophocles in three main points. (i) Teiresias does 
not intuitively know the murderer of Latus. When his aid is 
invoked by Oedipus, he has recourse to the arts of divination. 
Manto, the daughter of the blind seer, reports the signs to 
him, and he declares that neither voice of birds nor inspection of 
victims can reveal the name. Laius himself must be called up 
from the shades. In a grove near Thebes, Teiresias performs 
the awful rites which evoke the dead; the ghastly shape of 
Laius rises— | 


Stetit per artus sanguine effuso horridus— 


and denounces his son. This scene is related to Oedipus by 
Creon in a long and highly-wrought speech (530—658). Here, 
as in the earlier scene with Manto (303—402), copious use is 
made of detail from Roman augural lore, as well as of the 
Nekyia in the eleventh book of the Odyssey—suggesting a 
contrast with the lightness of touch which marks that passage of 
the Sophoclean Aztigone (998—1011) where Teiresias describes 
the failure of his appeal to augury. There, the technical signs 
are briefly but vividly indicated; in Seneca, the erudition is 
heavy and obtrusive. 

(ii) After the discovery of the parricide and the incest, and 
when Oedipus has now blinded himself, Iocasta meets and thus 
accosts him :— 


Quid te vocem ? 
Natumne? dubitas? natus es, natum pudet, 
Invite, loquere, nate: quo avertis caput 
Vacuosque vultus? 


INTIRODOCCTION. XXXV 


Oed. Quis frui et tenebris vetat? 
Quis reddit oculos? matris, heu, matris sonus. 
Perdidimus operam. Congredi fas amplius 
Haud est. Nefandos dividat vastum mare... 


Iocasta presently kills herself on the stage. Here, at least, 
Seneca has the advantage of Euripides, whose Iocasta speaks 
the prologue of the Phoentssae, and coldly recites the horrors of 
her past life-—adding that Oedipus has been imprisoned by his 
sons, ‘in order that his fate might be forgotten—for it needs 
much art to hide it’. The Iocasta of Sophocles rushes from the 
scene, not to re-appear, at the moment when she finds Oedipus 
resolved to unbare that truth of which she herself is already cer- 
tuin, and leaves the terrible cry thrilling in our ears— 

ἰού, ἰού, SvaTnve’ τοῦτο yap σ᾽ ἔχω 

μόνον προσειπεῖν, ἄλλο δ᾽ οὔποθ᾽ ἵστερον. 

In the truth and power of this touch, Sophocles is alone. 
Neither Seneca, nor any later dramatist, has managed this 
situation so as to express with a similar union of delicacy and 
strength the desperate anguish of a woman whom fate has 
condemned to unconscious crime. 

(iii) Seneca had no ‘Oedipus at Colonus’ in view. He was 
free to disregard that part of the legend according to which 
Oedipus was expelled from Thebes by Eteocles and Polyncices, 
and can therefore close his play by making Oedipus go forth 
into voluntary exile :— 


Mortifera mecum vitia terrarum extraho. 
Violenta fata et horridus morbi tremor 
Maciesque et atra pestis et tabidus dolor 
Mecum ite, mecum: ducibus his uti libet. 


§ 19. Thecloseness with which Seneca has studied Sophocles Seneca’s 


: pas 5 . relation to 
can be judged from several passages*. It is instructive to notice Sophocles. 


that, while Seneca has invented rhetorical ornament (as in the 


1 Eur. Phoen. 64 ἵν᾽ ἀμνήμων τύχη | γένοιτο, πολλών δεομένη σοφισμάτων. 

2. Such are, the scene in which Oedipus upbraids Creon (Sen. 678—708, cp. Soph. 
532—630); the questioning of Iocasta by Oedipus (Sen. 773—783, cp. Soph. 740— 
755); the scene with the messenger from Corinth, and the final discovery (Sen. 783— 
881. Cp. Soph. 955—1185). 


C2. 


The 
Ocdipe of 
Corneille. 


XXXVI INTRODOGCTION: 


opening dialogue, 1—105, and the Nekyia, 5 30—568), he has not 
known how to vary the natural development of the action. He has 
compressed the incidents of Sophocles into the smallest compass; 
and hence, notwithstanding the rhetorical episodes, the whole 
play consists only of 1060 lines, and would not have occupied 
more than an hour and a half in representation. Seneca is 
thus a negative witness to the mastery shown by the artist who 
could construct such a drama as the Oedipus Tyrannus with 
such materials. The modern dramatists, as we shall see, teach 
the same lesson in a more positive form. Walter Scott’s estimate 
of Seneca’s Oedi~us needs modification, but is just in the main. 
‘Though devoid of fancy and of genius,’ he says, it ‘displays the 
masculine eloquence and high moral sentiment of its author; 
and if it does not interest us in the scene of fiction, it often 
compels us to turn our thoughts inward, and to study our own 
hearts.’ Seneca’s fault, however, so far as the plot is concerned, 
seems less that he fails to interest, than that, by introducing the 
necromantic machinery, and by obliterating the finer moral traits 
of his Greek original, he has rendered the interest rather ‘ sensa~ 
tional’ than properly dramatic’. 

§ 20. The Ocedipe of Corneille was produced at Paris in 1657. 
After an interval which followed the unfavourable reception of his 
Pertharite in 1653, it was with the Oedzpe that Corneille returned 
to the theatre, at the instance of his patron, Nicolas Fouquet, to 
whom it is dedicated. It is immaterial for our purpose that this 
play is far from exhibiting Corneille at his best; nor need we 
here inquire what precise rank is to be assigned to it among his 
less successful works. For the student of Sophocles, it has the 
permanent interest of showing how the subject of the Oedipus 
Tyrannus was adapted to the modern stage by a typical artist of 
the French classical school. The severely simple theme of Sopho- 
cles, with its natural elements of pity and terror, is found too 
meagre by the modern dramatist. He cannot trust to that 


1 A small trait may be noticed as amusingly characteristic of the Roman poet of 
the Empire. The Laius of Sophocles goes to Delphi Basés—with only four at- 
tendants (752). Seneca makes Laius set out with the proper retinue of a king ;—but 
most of them lose their way. lures fefellit error ancipitis viae: Paucos fidelis- 
curribus tunxit labor. 


INTRODUCTION: XXXVI 


alone ; he feels that he needs some further source of variety and 
relief. To supply this, he interweaves an underplot of secondary 
persons—‘ the happy episode of the loves of Theseus and Dirce.’ 
Theseus is the king of Athens; Dircé is a daughter of the 
deceased Latus. 

The drama opens with a love-scene, in which Theseus is 
urging Dircé not to banish him from her presence at 
Thebes :— 


N’écoutez plus, madame, une pitié cruelle, 
Qui d’un fidéle amant vous feroit un rebelle... 


To the end, the fortunes of this pair divide our attention 
with those of Oedipus and Iocasta. Corneille does not bring 
Teiresias on the scene; but Nérine, ‘lady of honour to IJocasta,’ 
relates how the seer has called forth the shade of Laius. The 
ghost does not (as with Seneca) denounce Oedipus, but declares 
that the woes of Thebes shall cease only ‘when the blood of 
Laius shall have done its duty.’ The discovery is brought about 
nearly as in. Sophocles, though the management of the process is 
inferior in a marked degree. The herdsman of Latus—whom 
Corneille, like Dryden and Voltaire, names Phorbas, after 
Seneca’s example—kills himself on the stage ; Iocasta, snatching 
the poniard from him, plunges it in her own breast. Ocdipus 
blinds himself. No sooner have the gory drops flowed from his 
eyes, than the pest which is ravaging Thebes ceases: the mes- 
sage of the spirit is fulfilled :—‘the blood of Laius has done its 
duty.’ Theseus and Dircé, we understand, are made happy. 

The chief character, as drawn by Corneille, shows how an 
artificial stoicism can destroy tragic pathos. The Oedipus of 
Corneille is an idealised French king of the seventeenth century 
—one of those monarchs concerning whom Dircé says, 


Le peuple est trop heureux quand il meurt pour ses rois; 


he learns the worst with a lofty serenity ; and his first thought is 
to administer a stately rebuke to the persons whose misdirected 
forethought had saved him from perishing in infancy :— 


Voyez οὶ m’a plongé votre fausse prudence. 


Dircé admires his impassive fortitude :— 


XXXVIil INTRODUCTION. 


La surprenante horreur de cet accablement 
Ne οοὔϊε ἃ sa grande 4me aucun égarement. 


Contrast with this the life-like and terrible power of the 
delineation in Sophocles, from the moment when the cry 
of despair bursts from the lips of Oedipus (1182), to the 
end. 

one a § 21. Twenty-two years after Corneille, Dryden essayed the 

Dryden. same theme. His view was that his French predecessor had 
failed through not rendering the character of Oedipus more 
noble and attractive. On the other hand, he follows Corneille 
in the essential point of introducing an underplot. Dryden’s 
Eurydicé answers to Corneille’s Dircé, being, like her, the 
daughter of Laius. Corneille’s Theseus is replaced by Adrastus, 
king of Argos,—a personage less likely, in Dryden’s opinion, to 
eclipse Oedipus. When the play opens, Oedipus is absent from 
Thebes, and engaged in war with Argos. Meanwhile plots are 
being laid against his throne by Creon—a hunch-backed villain 
who makes love to Eurydice, and is rejected by her much as 
Shakspeare’s Richard, Duke of Gloster—who has obviously 
suggested some traits—is repulsed by the Lady Ann. Pre- 
sently Oedipus returns, bringing the captive Adrastus, whom 
he chivalrously sets free to woo Eurydicé. From this point, the 
piece follows the general lines of Sophocles, so far as the dis- 
covery is concerned. Oedipus is denounced, however, not by 
Teiresias, but, as in Seneca, by the ghost,—which Dryden, unlike 
Seneca, brings on the stage. 

It is singular that Dryden should have committed the same 
mistake which he perceived so clearly in Corneille. Eurydice 
and Adrastus are less tiresome than Dircé and Theseus, but 
their effect is the same. The underplot spoils the main plot. 
The tragic climax is the death of Eurydicé, who is stabbed by 
Creon. Creon and Adrastus next kill each other; then Iocasta 
slays herself and her children ; and finally Oedipus throws him- 
self from an upper window of the palace. ‘Sophocles,’ says 
Dryden, ‘is admirable everywhere; and therefore we have fol- 
lowed him as close as we possibly could.’ In a limited verbal 
sense, this is true. There are several scenes, or parts of scenes, in 


INTRODUCTION, XXXIX 


which Dryden has almost transcribed Sophocles’. But the dif- 
ference of general result is complete. The Oedipus of Sophocles 
does perfectly that which Tragedy, according to Aristotle, ought 
to do. It effects, by pity and terror, the ‘purgation’ of such 
feelings; that is, it separates them from the alloy of mean acci- 
dent, and exercises them, in their pure essence, on great objects 
—here, on the primary instincts of natural affection. In relation 
to pity and terror, Tragedy should be as the purgatorial fire,— 


exemit labem, purumque reliquit 
Aetherium sensum atque aurai simplicis ignem. 


Now, Dryden’s play first divides our sympathy between 
the fate of Eurydice and that of Oedipus; next, it involves it 
with feelings of a different order,—loathing for the villainy of 
Creon, and disgust at the wholesale butchery of the end. In- 
stead of ‘purging’ pity and terror, it stupefies them; and the 
contrast is the more instructive because the textual debt of 
Dryden to Sophocles has been so large. 

It is right to add that, while the best parts of the play—the 
first and third acts—are wholly Dryden’s, in the rest he was 
assisted by an inferior hand%» And, among the places where 
Dryden’s genius flashes through, it is interesting to remark one 
in which he has invented a really Greek touch,—not in the 
manner of Sophocles, certainly, yet such as might occur in 
Euripides. Oedipus is pronouncing the curse on the unknown 
murderer :— 


But for the murderer’s self, unfound by man, 

Find him, ye powers celestial and infernal ! 

And the same fate, or worse than Laius met, 

Let be his lot: his children be accurst ; 

His wife and kindred, all of his, be cursed! 
Both Priests. Confirm it, heaven! 


1 As in the scene with the suppliants (Act 1. Sc. i.); that between Oedipus and 
Iocasta (Act III. Sc. i.); and that between Oedipus and Aegeon (the messenger from 
Corinth, Act Iv. Sc. i.). 

2 «What Sophocles could undertake alone, Our poets found a work for more than 
one’ (Epilogue). Lee must be held accountable for the worst rant of Acts Iv. and 
v.; but we are not concerned here with the details of execution, either in its merits or 
in its defects. 


The 
Ocedipe of 
Voltaire. 


x] INTRODOCTION. 


Enter Jocasta, attended by Women. 


Joc. At your devotions? Heaven succeed your wishes ; 
And bring the effect of these your pious prayers 
On you, and me, and all. 
Pr. Avert this omen, heaven ! 
Ocdti~. O fatal sound! unfortunate Jocasta ! 
What hast thou said? an il! hour hast thou chosen 
For these foreboding words! why, we were cursing! 
Joc. Then may that curse fall only where you laid it. 
Ocdip. Speak no more! 
For all thou say’st is ominous: we were cursing ; 
And that dire imprecation hast thou fasten’d 
On Thebes, and thee, and me, and all of us. 


§ 22. More than either Dryden or Corneille, Voltaire has 
treated this subject in the spirit of the antique. His Oedipe was 
composed when he was only nineteen. It was produced in 1718 
(when he was twenty-four), and played forty-six times consecu- 
tively—a proof, for those days, of marked success. In 1729, the 
piece having kept its place on the stage meanwhile, a new 
edition was published. It is not merely a remarkable work for 
so young a man; its intrinsic merit, notwithstanding obvious 
defects, is, I venture to think, much greater than has usually 
been recognised. The distinctive ‘note’ of the modern versions 
—the underplot—is there, no doubt; but, unlike Corneille and 
Dryden, Voltaire has not allowed it to overshadow the main 
action. 

The hero Philoctetes revisits Thebes, after a long absence, 
to find Oedipus reigning in the seat of Laius. The Thebans 
are vexed by pestilence, and are fain to find a victim for the 
angry god; Philoctetes was known to have been the foe of 
the late king, and is now accused of his murder. Iocasta had 
been betrothed to Philoctetes in youth, and loves him still. She 
urges him to fly, but he resolves to remain and confront the false 
charge. At this moment, the seer Teiresias denounces Oedipus 
as the criminal. Philoctetes generously protests his belief in the 


king’s innocence; and from this point (the end of the third Act) 
appears no more, 


INTRODUCTION. xl 


Thenceforth, the plot is mainly that of Sophocles. The first 
scene of the fourth Act, in which locasta and Oedipus inform 
each other of the past, is modelled on Oed. Tyr. 698—862, with 
some characteristic differences. Thus, in Sophocles, the first 
doubt of Oedipus as to his parentage springs from a taunt 
uttered at a feast (779). Here is Voltaire’s substitute for that 
incident (the scene, of course, being Corinth) :— 


Un jour, ce jour affreux, présent ἃ ma pensée, 
Jette encor la terreur dans mon 4me glacée; 
Pour la premiére fois, par un don solennel, 
Mes mains, jeunes encore, enrichissaient |’autel: 
Du temple tout-a-coup les combles s’entr’ouvrirent ; 
De traits affreux de sang les marbres se couvrirent; 
De l’autel, ébranlé par de longs tremblemens, 
Une invisible main repoussait mes présens ; 
Et les vents, au milieu de la foudre éclatante, 
Portérent jusqu’é moi cette voix effrayante : 

“Ne viens plus des lieux saints souiller la pureté; 

‘*Du nombre des vivans les dieux t’ont rejeté; 

“Tis ne recoivent point tes offrandes impies ; 

“Va porter tes présens aux autels des Furies; 

“Conjure leurs serpens préts a te déchirer ; 

“Va, ce sont la les dieux que tu dois implorer.” 


This is powerful in its way. But where Voltaire has introduced 
a prodigy—the supernatural voice heard amid lightnings— 
Sophocles was content to draw from common life, and to mark 
how a random word could sink into the mind with an effect 
as terrible as that of any portent. Voltaire has managed the 
final situation on Corneille’s plan, but with infinitely better 
effect. The High Priest announces that Oedipus has blinded 
himself, thereby appeasing the gods; and the play closes with 
the death of Iocasta: 


IOCASTE. 


O mon fils! hélas! dirai-je mon époux? 
O des noms les plus chers assemblage effroyable! 
Il est donc mort? 


, Voltaire’s 
criticisms. 


xlil INTRODUCTION. 


LE GRAND PRETRE. 


I] vit, et le sort qui laccable 
Des morts et des vivans semble le séparer’; 
Il s’est privé du jour avant que d’expirer. 
Je lai vu dans ses yeux enfoncer cette épée, 
Qui du sang de son pére avait été trempée; 
Il a rempli son sort, et ce moment fatal 
Du salut des Thébains est le premier signal. 
Tel est ordre du ciel, dont la fureur se lasse; 
Comme il veut, aux mortels 1] fait justice ou grace; 
Ses traits sont épuisés sur ce malheureux fils: 
Vivez, il vous pardonne. 


LOCASTE. 


Et moi je me punis. (2 72 se frappe.) 
Par un pouvoir affreux réservée ἃ linceste, 
La mort est le seul bien, le seul dieu qui me reste. 
Laius, regois mon sang, je te suis chez les morts: 
Jai vécu vertueuse, et je meurs sans remords. 


LE CHOEUR. 
O malheureuse reine! 6 destin que j’abhorre! 


IOCASTE. 


Ne plaignez que mon fils, puisqu’il respire encore. 
Prétres, et vous Thébains qui ffites mes sujets, 
Honorez mon bficher, et songez ἃ jamais 

Qu’au milieu des horreurs du destin qui m’opprime 
J’ai fait rougir les dieux qui m’ont forcée au crime. 


§ 23. Voltaire was conscious of the objections to his own 
episode of Philoctetes; no one, indeed, could have criticised it 
with more wit or force. ‘Philoctetes seems to have visited 
Thebes only for the purpose of being accused’: not a word is 
said of him after the third Act, and the catastrophe is absolutely 


1 Voltaire borrowed this verse from Corneille,—‘parce qu’ayant précisément la 
méme chose ἃ dire,...i] m’était impossible de l’exprimer mieux’; and Corneille was 
himself translating Seneca’s ‘mec uivis mixtus, nec sepultis.’ Voltaire was perhaps 
unconscious that the ground which he assigns here was exactly that on which the 
repetition of passages in the Greek orators was defended—viz. that τὸ καλῶς εἰπεῖν 
ἅπαξ περιγίγνεται, dis δὲ οὐκ ἐνδέχεται (Theon, προγυμνάσματα 1: see my Aftic 
Orators, vol. 1. p. 1xxii). 


INTRODUCTION. xl 


independent of him. Ina letter to the Jesuit Porée, with whom 
he had read the classics, Voltaire apologises for Philoctetes by 
saying that the Parisian actors would not hear of an Oedipus 
with no love in it; ‘I spoiled my piece, he says, ‘to please 
them.’ 

But it is certain, from what he says more than once else- 
where, that he regarded some underplot as a necessity. His 
remarks on this point are worth noting, because they touch an 
essential difference between the old Greek view of drama and 
that which has prevailed on our stage. ‘The subject (Oedipus) 
did not, in itself, furnish me with matter for the first three Acts; 
indeed, it scarcely gave me enough for the last two. Those who 
know the theatre—that is, who are as much alive to the difficulties 
as to the defects of composition—will agree with what I say.’ 
‘In strictness, the play of Oedipus ought to end with the first 
Act.’ Oedipus is one of those ancient subjects ‘which afford 
only one scene each, or two at most—not an entire tragedy.’ 
In short, to demand a modern drama on the szmple story of 
Oedipus was like setting one to make bricks without straw. 
Corneille found himself constrained to add the episode of 
Theseus and Dircé; Dryden introduced Adrastus and Eurydicé’. 


1 «All we could gather out of Corneille,’ says Dryden, ‘was that an episode must 
be, but not his way.’ Dryden seems to have felt, however, that it was demanded 
rather by convention than by artistic necessity. The following passage is interest- 
ing as an indication that his instinct was better than his practice:—‘The Athenian 
theatre (whether more perfect than ours, is not now disputed), had a perfection 
differing from ours. You see there in every act a single scene, (or two at most), 
which manage the business of the play; and after that succeeds the chorus, which 
commonly takes up more time in singing, than there has been employed in speaking. 
The principal person appears almost constantly through the play; but the inferior 
parts seldom above once in the whole tragedy.. The conduct of our stage is much 
more difficult, where we are obliged never to lose any considerable character, which 
we have once presented.’ [Voltaire’s Philoctetes broke this rule.] ‘Custom likewise 
has obtained, that we must form an underplot of second persons, which must be 
depending on the first; and their bye-walks must be like those in a labyrinth, which 
all of them lead into the great parterre; or like so many several lodging chambers, 
which have their outlets into the same gallery. Perhaps, after all, if we could think 
so, the ancient method, as it is the easiest, is also the most natural and the best. For 
variety, as it is managed, is too often subject to breed distraction; and while we 
would please too many ways, for want of art in the conduct, we please in none.’ 
(Preface to Oedipus.) . 


Essential 
difference 
between 
Sophocles 
and the 
moderns. 


xliv INTRODUCTION: 


§ 24. Now, why could Sophocles dispense with any such ad- 
dition, and yet produce a drama incomparably more powerful? 
The masterly art of Sophocles in the structure and development 
of the plot has already been examined, and is properly the first 
attribute of his work which claims attention. But this is not the 
only, or the principal, source to which the Oedipus Tyrannus 
owes its greatness; the deeper cause is, that Sophocles, in the 
spirit of Greek Tragedy, has known how to make the story of 
Oedipus an ideal study of character and passion. Corneille, 
Dryden, Voltaire—each in his own way—were thinking, ‘How 
am I to keep the audience amused? Will they not find this 
horrible story of Oedipus rather too painful and monotonous? 
Will they not desire something lighter and pleasanter—some 
love-making, for instance, or some intrigue?’ ‘What an insipid 
part would Iocasta have played,’ exclaims Voltaire, ‘had she not 
retained at least the memory of a lawful attachment, and trembled 
for the existence of a man whom she had once loved!’ There is 
the secret frankly told. 

Sophocles, on the other hand, concentrates the attention of the 
audience on the destiny of Oedipus and Iocasta. The spectators 
are enchained by the feelings which this destiny moves at each 
step in its course. They are made to see into the depths of two 
human souls. It is no more possible for them to crave minor 
distractions than it would be for our eyes or thoughts to wander, 
if we were watching, without the power of arresting, a man who 
was moving blindfold towards a precipice. The interest by 
which Sophocles holds us is continuous and intense; but it is 
not monotonous, because alternations of fear lead up to the 
worst ; the exciting causes of pity and terror are not unworthy 
or merely repulsive, for the spectacle offered is that of a noble 
and innocent nature, a victim to unknown and terrible forces 
which must be counted among the permanent conditions of life, 
since the best of mankind can never be sure of escaping them. 
When the worst has befallen, trex Sophocles knows how to 
relieve the strain; but it is a relief of another order from that 
which Corneille affords by the prospect of Theseus being made 
happy with Dircé. It is drawn from the natural sources of the 
tragedy itself; the blind king hears the voices of his children. 


INTRODUCTION. xlv 


§ 25. A comparison may fitly close with a glance at two References 
points in which the modern dramas illustrate Sophocles, and SRE © 
which have more than the meaning of details. Dryden has instinct. 
represented Oedipus and Iocasta as haunted, from the first, by 
a mysterious instinct of their true relationship. Thus she says 
to him :— 

When you chid, methought 
A mother’s love start’ up in your defence, 
And bade me not be angry. Be not you; 
For I love Laius still, as wives should love, 
But you more tenderly, as part of me’. 


Voltaire has the same thought (Act II. Sc. ii.), where Iocasta 
is speaking of her marriage with Oedipus  : 


je sentis dans mon Ame étonnée 
Des transports inconnus que je ne concus pas: 
Avec horreur enfin je me vis dans ses bras. 


There is a similar touch in Corneille. Oedipus is watching 
Dirce—whom he believes to be his step-daughter, but who is in 
fact his sister—with her lover Theseus (Act III. Sc. iv.): 


Je ne sais quelle horreur me trouble ἃ leur aspect; 
Ma raison la repousse, et ne m’en peut défendre. 


Such blind warnings of nature are indeed fitted to make the 
spectator shudder; but they increase the difficulty of explaining 
why the truth was not divined sooner; and they also tend to 
lessen the shock of the discovery. In other words, they may be 
poetical,—they may be even, in the abstract, tragic,—but they 
are not, for this situation, dramatic; and it is due to the art of 
Sophocles to observe that he has nowhere admitted any hint of 
this kind. 


§ 26. Next, it should be noticed that no one of the later The im- 
dramatists has been able to avoid leaving a certain element of im- oe a 
probability in the story. We saw above that Aristotle alludes to how ma- 
the presence of such an element, not in the plot : itself, but in the ene 

moderns. 


1 =‘started,’ as again in this scene: ‘Nature herself start back when thou wert 
born.’ 
2 Act 1. Sc. i.: cp. what Oedipus says in Act II. Sc. i. 


xlvi INTRODUCTION. 


supposed antecedents. It consists in the presumed ignorance of 
Oedipus and locasta regarding facts with which they ought to 
have been familiar. Sophocles tacitly accepts this condition, 
and, by doing so, minimizes its prominence; so much so, that it 
may be doubted whether many readers or spectators of the 
Ocdipus Tyrannus would think of it, if their attention had not 
been drawn to it previously. Seneca has not attempted to im- 
prove on that example. But the moderns have sought various 
ways of evading a critical censure which they foresaw; and it is 
instructive to consider the result. The Oedipus of Corneille 
knows that Lafus was said to have been killed by robbers; he 
also knows the place and the date. Further, he distinctly re- 
members that, at the same place and at the same date, he himself 
had slain three wayfarers. Strange to say, however, it never 
occurs to him that these wayfarers could possibly have been 
Laius and his attendants. He mildly suggests to locasta that 
they may have been the robbers (Act I. ὅς. i.); though, as appears 
from the circumstances which he himself afterwards relates 
(Act Iv. Sc. iv.), he had not the slightest ground for such a sup- 
position. This device cannot be deemed an improvement on 
Sophocles. Dryden's expedient is simpler :— 


Tell me, Thebans, 
How Laius fell; for a confused report 
Pass’d through my ears, when first I took the crown; 
But full of hurry, like a morning dream, 
It vanish’d in the business of the day. 


That only serves to show us that the dramatist has an uneasy 
conscience. Voltaire’s method is subtler. Oedipus thus excuses 
himself for having to question Iocasta concerning the death 
of Laius :— 

Madame, jusqu’ici, respectant vos douleurs, 

Je n’ai point rappelé le sujet de vos pleurs; 

Et de vos seuls périls chaque jour alarmée 

Mon ame ἃ d’autres soins semblait étre fermée. 


But, as the author admits, the king ought not to have been 
so long deterred, by the fear of displeasing his wife, from inform- 
ing himself as to the death of his predecessor: ‘this is to have 


INTRODUCTION. xl vii 


too much discretion and too little curiosity... Sophocles, accord- 
ing to Voltaire, ought to have suggested some explanation of 
the circumstance that Oedipus, on hearing how Laius perished, 
does not at once recollect his own adventure in the narrow pass. 
The French poet seeks to explain it by hinting at a miraculous 
suspension of memory in Oedipus :— 


Et je ne concois pas par quel enchantement 
Joubliais jusqu’ici ce grand événement ; 

La main des dieux sur moi si long-temps suspendue 
Semble éter le bandeau quwils mettaient sur ma vue. 


But this touch, though bold and not unhappy, must be classed 
with the transparent artifices of the stage. The true answer to 
the criticisms on this score which Voltaire directs against Sopho- 
cles, Corneille, and himself is contained in a remark of his own, 
that a certain amount of improbability is inherent in the story 
of Oedipus’. If that improbability is excluded at one point, 
it will appear at another. This being so, it is not difficult to 
choose between the frank treatment of the material by Sophocles, 
and the ingenious but ineffectual compromises of later art. 


§ 27. The recent revivals of Greek plays have‘had their great Revivals 
reward in proving how powerfully the best Greek Tragedy can litte 
appeal to modern audiences. Those who are furthest from being 
surprised by the result will be among the first to allow that the 
demonstration was needed. The tendency of modern study had 
been too much to fix attention on external contrasts between the 
old Greek theatre and our own. Nor was an adequate corrective 
of this tendency supplied by the manner in which the plays have 
usually been studied; a manner more favourable to a minute 
appreciation of the text than to apprehension of the play as 
a work of art. The form had been understood better than the 
spirit. A vague feeling might sometimes be perceived that the 
effectiveness of the old Greek dramas, as such, had depended 
-essentially on the manners and beliefs of the people for whom 


1 In the fifth letter to M. de Genonville:—‘I] est vrai qu’il y a des sujets de 
-tragédie ot l’on est tellement géné par la bizarrerie des événemens, qu’il est pres- 
-qwimpossible de réduire l’exposition de sa piéce ἃ ce point de sagesse et de vrai- 
-semblance. Je crois, pour mon bonheur, que le sujet d’CEdipe est de ce genre.’ 


The 
Ocedipus 
Tyrannus 
—a crucial 
experi- 
ment. 


The result 
at : 
Harvard. 


xlvill LNERODOCTION. 


they were written, and that a successful Sophocles presupposed 
a Periclean Athens. Some wonderment appeared to greet the 
discovery that a masterpiece of Aeschylus, when acted, could 
move the men and women of to-day. Now that this truth has 
been so profoundly impressed on the most cultivated audiences 
which England or America could furnish,—in Germany and 
France it had been less unfamiliar,—it is not too much to say 
that a new life has been breathed into the modern study of the 


Greek drama. > 


§ 28. Recent representations of the Oedipus Tyrannus have 
a peculiar significance, which claims notice here. The incestuous 
relationship—the entrance of Oedipus with bleeding eyes—these 
are incidents than which none could be imagined more fitted to 
revolt a modern audience. Neither Corneille nor Voltaire had 
the courage to bring the self-blinded king on the stage; his deed 
is related by others. Voltaire, indeed, suggested’ that the spec- 
tacle might be rendered supportable by a skilful disposition of 
lights,—Oedipus, with his gore-stained face, being kept in the 
dim back-ground, and his passion being expressed by action 
rather than declamation, while the scene should resound with the 
cries of Iocasta and the laments of the Thebans. Dryden dared 
what the others declined; but his play was soon pronounced 
impossible for the theatre. Scott quotes a contemporary witness 
to the effect that, when Dryden’s Oedipus was revived about the 
year 1790, ‘the audience were unable to support it to an end; 
the boxes being all emptied before the third act was concluded.’ 


§ 29. In May, 1881, after seven months of preparation, the 
Oedipus Tyrannus was acted in the original Greek by members 
of Harvard University. Archaeology, scholarship, and art had 
conspired to make the presentation perfect in every detail; and 
the admirable record of the performance which has been published 
has a permanent value for every student of Sophocles®. Refer- 


1 In one of his notes on Corneille’s Preface to the Oedif~e (Oeuvres de Corneille, 
vol. VII. p. 262, ed. 1817). 

2 An Account of the Harvard Greek Play. By Henry Norman. Boston: 
James R. Osgood and Co., 1882. The account is illustrated by 15 photographs of 
characters and groups, and is dedicated by the Author (who acted the part of Creon) 
to Professor J. W. White. See Appendix, p. 201. 


INTRODUCTION. xfix 


ences to it will be found in the following commentary. But it is 
the impression which the whole work made on the spectators of 
which we would speak here. Nothing of the original was altered 
or omitted; and at the last Oedipus was brought on the scene, 
‘his pale face marred with bloody stains.’ The performances 
were seen by about six thousand persons,—the Harvard theatre 
holding about a thousand at a time. As an English version was 
provided for those who needed it, it cannot be said that the lan- 
guage veiled what might else have offended. From first to last, 
these great audiences, thoroughly representative of the most 
cultivated and critical judgment, were held spell-bound. ‘The 
ethical situation was so overwhelming, that they listened with 
bated breath, and separated in silence.’ ‘The play is over. 
There is a moment’s silence, and then the theatre rings with 
applause. It seems inappropriate, however, and ceases almost 
as suddenly as it began. The play has left such a solemn 
impression that the usual customs seem unfitting, and the 
audience disperses quietly’. There is the nineteenth century’s 
practical interpretation of Aristotle. This is Tragedy, ‘effect- 
ing, by means of pity and terror, the purgatzon of such feelings.’ 


§ 30. A few months later in the same year (1881), the Oedife Roi 
Oedipus Tyrannus was revived in a fairly close French transla- oe 
tion at the Theatre Francais. When the version of Jules Frangais. 
Lacroix was played there in 1858, the part of Oedipus was 
filled by Geoffroy; but on this occasion an artist was available 
whose powers were even more congenial. Probably no actor 
of modern times has excelled M. Mounet-Sully in the union 
of all the qualities required for a living impersonation of the 
Sophoclean Oedipus in the entire series of moods and range 
of passions which the part comprises; as the great king, at 
once mighty and tender; the earnest and zealous champion of 
the State in the search for hidden guilt; the proud man startled 
by a charge which he indignantly repels, and embittered by the 
supposed treason of a friend; tortured by slowly increasing 
fears, alternating with moments of reassurance; stung to frenzy 
by the proof of his unspeakable wretchedness; subdued to a 


1 Account of the Harvard Greek Play, pp. 36, 103. 
129. a 


᾿ INTRODUCTION. 


calmer despair ; finally softened by the meeting with his young 
daughters. The scene between Oedipus and Iocasta (vv. 700 
—862) should be especially noticed as one in which the 
genius of Sophocles received the fullest justice from that of 
M. Mounet-Sully. In the words of a critic who has finely 
described the performance’:— 


‘Every trait of the tragedian’s countenance is now a witness to the 
inward dread, always increasing upon him, as he relates his own adven- 
ture, and questions her for more minute details of the death of Laius. 
His voice sometimes sinks to a trembling gasp of apprehension, as the 
identity of the two events becomes more and more evident. He seems 
to be battling with fate.’ 


With a modern audience, the moment at which the self- 
blinded Oedipus comes forth is that which tests the power of the 
ancient dramatist; if, at that sight, repugnance overpowers 
compassion, the spell has been imperfect; if all other feelings 
are absorbed in the profound pathos of the situation, then 
Sophocles has triumphed. We have seen the issue of the ordeal 
in the case of the representation at Harvard. On the Paris 
stage, the traditions of the French classical drama (represented 
on this point by Corneille and Voltaire) were apt to make the 
test peculiarly severe. It is the more significant that the moment 
is thus described in the excellent account which we have cited 


above :— 


‘Oedipus enters, and in the aspect of the man, his whole history is 
told. It is not the adjunct of the bleeding eyes which now most deeply 
stirs the spectators. It is the intensity of woe which is revealed in every 
movement of the altered features and of the tottering figure whose 
bearing had been so majestic, and the tone of the voice,—hoarse, yet 
articulate. The inward struggle is recognised in its necessary outward 
signs. The strain on the audience might now become too great but for 
the relief of tenderness which almost immediately succeeds in the part- 
ing of Oedipus from his children. Often as pathetic farewells of a 
similar kind have been presented on the stage, seldom has any made an 
appeal so forcible.’ 


1 Saturday Review, Nov. 19, 1881. 


INTRODUCTION. li 


In the presence of such testimonies, it can no longer be Conclu- 
deemed that the Tragedy of ancient Greece has lost its virtue ”°” 
for the modern world. And, speaking merely as a student of 
Sophocles, I can bear witness that the representation of the 
Ajax at Cambridge (1882) was to me a new revelation of 
meaning and power. Of that performance, remarkable in so 
many aspects, I hope to say something in a later part of this 
edition. Here it must suffice to record a conviction that such 
revivals, apart from their literary and artistic interest, have also 
an educational value of the very highest order. 


ΝΜ ΞΘ ΡΞ ΘΙ ΤΟΝ ΝΙ) 
COMMENTARIES: 


MSS. used. § 1. The manuscripts of the Oedipus Tyrannus which have been 
chiefly used in this edition are the following’. 


In the Biblioteca Mediceo-Laurenziana, Florence. 


L, cod. ΧΧΧΙΙ. 9, commonly known as the Laurentian Ms., first half 
of 11th century. 


In the Bibliothéque Nationale, Paris. 


; COGc2712,.1 3tn- century. 

, cod. 2787, ascribed to the 15th cent. (Catal. 11. 553). 
, cod, 2884, ascribed to the 13th cent. (? 2d. Il. 565). 
προ ΣΎ ΙΝ cent, 


- Ξ 9.» 


In the Biblioteca Marciana, Venice. 


V, cod. 468, late 13th century or early 14th. 
V’, cod. 616, probably of the 14th cent. 

ΝΜ cod: 467; 14th: cent, 

V=;-€00s..472,. 14th cent, 


In the Bodleian Library, Oxford. 


Cod. Laud. Misc. 99 (now Auct. F. 3. 25), late 14th century. 
Cod. Laud. 54, early 15th cent: 
Cod. Barocc. 66, 15th cent. 


In the Library of Trinity College, Cambridge. 


Cod. R. 3. 31, mainly of the late 14th century, in parts perhaps of " 
the early 15th. 
These mss. I have myself collated. 


The following are known to me in some cases by slighter personal 


1 There isno doubt that L belongs to the first half of the 11th century, and none 
(I believe) that A is of the 13th. These are the two most important dates. In the 
case of several minor MSs., the tendency has probably been to regard them as some- 
what older than they really are. The dates indicated above for such Mss. are given 
on the best authority that I could find, but I do not pretend to vouch for their preci- 
sion. This is, in fact, of comparatively small moment, so long as we know the 
general limits of age. Excluding L and A, we may say broadly that almost all other 
known Mss. of Sophocles belong to the period 1300—1600 A.D. 


MANUSCRIPTS hii 


inspection, but more largely from previous collations, especially from 
those of Prof. L. Campbell (2nd ed., 1879) :—Pal. = Palat. 40, Heidel- 
berg: Vat. a=cod. 40 in the Vatican, 13th cent. (ascribed by some to 
the:z2th): Vat.b,.cod) Urbin. 14 0,-20., παι cent: Vat. ¢, cod. Urbin, 
140, 20., 14th cent.: M, cod. G. 43 sup., in the Biblioteca Ambrosiana, 
Milan, 13th or early 14th cent.: Μ΄, cod. L. 39 sup., 22. early 14th 
cent.: L’, cod. 31. ro (14th cent.) in the Bibliot. Med.-Lor., Florence ; 
[cod Abbat. 152, date.1eth,-70.: A, cod. Abbat. 41, 14th cent.,20;3 
Rice. cod. 34, in the Biblioteca Riccardiana, Florence, sometimes 
ascribed to the 14th cent. but really of the 16th (see P. N. Papa- 
georgius, ‘cod. Laurent. von Soph.,’ εἴς.) p. 406, Leipzig, Teubner, 1883). 

In making a first selection of Mss. to be collated, I was guided 
chiefly by what I already knew of their character and of their relations 
to each other, as these might be inferred from the previous reports ; 
and this list was afterwards modified by such light as I gradually 
gained from my own experience. L stands first and alone. A is 
perhaps next—though at a long interval—in general value. The 
selection of 14th and 15th century mss. could have been enlarged ; 
but, so far as I can judge, the list which has been given is fairly 
representative. In the present state of our knowledge, even after 
all that has been done in recent years, it would, I think, be generally 
allowed that the greatest reserve must still be exercised in regard 
to any theory of the connections existing, whether by descent or 
by contamination, between our mss. of Sophocles. We have not here 
to do with well-marked families, in the sense in which this can 
be said of the manuscript authorities for some other ancient texts ; the 
data are often exceedingly complex, and such that the facts could be 
equally well explained by any one of two, or sometimes more, different 
suppositions. This is a subject with which I hope to deal more fully on 
a future occasion; even a slight treatment of it would carry me far 
beyond the limits which must be kept here. Meanwhile, it may be 
useful to give a few notes regarding some of the mss. mentioned above, 
and to add some general remarks. 

8.2. L, no. XxxI1. 9 in the Laurentian Library at Florence, is a vellum The Lau- 
MS., written in the first half of the eleventh century. It forms a volume ἴα ἴδῃ ΜΒ. 
measuring 124 by 84 inches, and containing 264 leaves (= 528 pages), 
of which Sophocles fills 118 leaves (= 236 pp.). It contains the seven 
plays of Sophocles, the seven plays of Aeschylus (with a few defects), 
and the Arvgonautica of Apollonius Rhodius. Marginal and interlinear 
scholia accompany the texts. i 

Since the first edition of this volume appeared, an autotype fac- 


The first 
hand. 


The first 
corrector. 


Later cor- 
rectors of 


Unique 
value of L. 


liv MANUSCRIPTS. 


simile of the text of Sophocles in L has been published by the 
London Society for the Promotion of Hellenic Studies (1885). In 
an Introduction issued with the facsimile, the palaeographical character 
of the ms. has been described by Mr E. M. Thompson, Keeper of 
Manuscripts and Egerton Librarian in the British Museum. The s. 
was produced in a regular workshop or scriptorium at Byzantium, 
The scribe wrote a clear and flexible hand; the characters are minus- 
cule, in that more cursive style which distinguishes other classical mss. 
of the same period from the biblical and liturgical, As the form of 
the ruling shows, the scribe prepared the Ms. to receive scholia; but 
his own work was confined to writing the text. The scholia were 
copied into the ms. by another person, under whose supervision the 
scribe appears to have worked. This person is usually designated as 
the ‘diorthotes,’ because he was the first corrector; or as ‘S,’ because 
he wrote the scholia. In some cases he himself corrected the errors 
of the first hand; in some others, where the first hand has corrected 
itself, this was probably done under his guidance; and he usually 
reserved to himself the part of supplying in the margin any verse 
which the first hand had omitted. In writing the scholia, the corrector 
used a mixture of minuscule and uncial (‘half-uncial’): but, in correct- 
ing or supplementing the text, he often used a more minuscule style, as 
if for the sake of greater uniformity with the first hand. Hence there 
is sometimes a doubt between the two hands, though, as a rule, they 
are easily distinguished. 

In the 12th and 13th centuries, at least three different hands added 
some notes. Hands of the r4th, 15th, or 16th century have been 
recognised in some other notes, both marginal and superscript. These 
later hands can usually be distinguished from that of the first corrector 
(the ‘diorthotes,’ or S), but very often cannot be certainly distinguished 
from each other. The attempt to do so is of the less moment since 
the additions which they made are seldom of any value. For much 
else that is of palaeographical interest in regard to L, readers may be 
referred to Mr Thompson’s Introduction: the facts noticed here are 
those which primarily concern a student of Sophocles. 

§ 3. Lis not only the oldest, but also immeasurably the best, ms. 
of Sophocles which we possess. In 1847 Cobet expressed the opinion 
that L is the source from which all our other mss. are ultimately 
derived. This view has been supported by Dindorf in the preface to 
his 3rd edition (Oxon. 1860), and by Moriz Seyffert in the preface 
to his Philoctetes (1867). The contrary view—that some of our mss, 
come from a source independent of L—has also found able supporters, 


MANUSCRIPTS. lv 


among whom have been Anton Seyffert (Quaestiones criticae de Codicibus 
recte aestimandts, Halle, 1863); Prof. N. Wecklein (Ars Sophocltis emen- 
dandt, pp. 2 ff., 1869), and Prof. L. Campbell (Sophocles, vol. 1. pp. 
xxiv ἢ 1879). I learn, however, that Prof. Wecklein has since 
become disposed to retract his opinion. In the second part of the 
Introduction to the Facsimile of L (pp. 15 ff.), I have shortly stated 
some of the objections to regarding L as the unique source. ‘Two of 
them are furnished by this play: viz. (i) verse 800, omitted in the text 
of L, and inserted in the margin by a hand certainly later than several 
of the mss. which have the verse in the text: (11) the words πονεῖν 
ἢ τοῖς θεοῖς written at v. 896 in the text of L,—these being corrupted 
from a gloss, πανηγυρίζειν τοῖς θεοῖς, which exists in full in the Trinity 
Ms., and elsewhere’. The chief argument for L being the unique 
source is briefly this, that, though other Mss. sometimes correct L on 
small points, no one of them supplies any correction which was clearly 
beyond the reach of a fairly intelligent scribe or grammarian. The 
question is one which does not seem to admit of demonstrative proof 
either way: we must be content with the probabilities, which will be 
differently estimated by different minds. Apart, however, from this 
obscure question, all scholars can agree in recognising the paramount 
importance of L as the basis of our text. The sense of L’s incom- 
parable value is one which steadily grows upon the student as he 
proceeds with the labour of textual criticism. Wecklein’s words are 
not too strong, when properly understood: ‘A critic will hardly go 
wrong if he treats every letter, every stroke in L as worthy of particular 
attention, while he regards the readings of other ss. rather in the light 
of conjectures,’—that is, where these Mss. diverge from L otherwise 
than by correcting its trivial errors. Instances in which they correct L 
may be seen in this play at vv. 43, 182, 221, 296, 332, 347, 657, 730, 
967, 1260, 1387, 1474, etc. But, notwithstanding all such small cor- 
rections, it remains true that, with L safe, the loss of our other Mss. 
would have been a comparatively light misfortune. As instances in 
which a true reading has been preserved in a citation of Sophocles by 
an ancient author, but neither in L nor in any other Ms., we may notice 
vv. 466, 528, 1170. 

§ 4. Of the other Florentine ss., L? cod. xxxI. 10 (14th cent.) con- Other Mss, 
tains all the seven plays, while Τ' (cod. Abbat. 152), of the late 13th 
cent., has only 42, ZZ, O. Z:, Phil.; and A (cod. Abbat. 41), of the 
14th cent., only 4z., £7, O. 7. 


1 A valuable discussion of this point is given by Prof. Campbell, vol. 1. pp. xxv— 
xli. 


lvi MANUSCRIPTS. 


A, no. 2712 in the National Library of Paris, is a parchment of the 
13th century’. It is a volume of 324 pages, each about 114 inches by 
g in size, and contains (1) Eur. Hee, Or., Phoen., Androm., Med., 
Hipp. : (2) p. 117—214, the seven plays of Soph.: (3) Ar. Plut., Mub., 
Ran., δῷ, Av., Acharn., Eccl. (imperfect). The text of each page is in 
three columns; the writing goes continuously from left to right along 
all three, so that, ¢.g., vv. 1, 2, 3 of a play are respectively the first lines 
of columns 1, 2, 3, and vy. 4 is the second line of col. 1. The contrac- 
tions are naturally very numerous, since the average breadth of each 
column (#.e. of each verse) is only about 2 inches; but they are regular, 
and the Ms. is not difficult to read. 

B, no. 2787, in the same Library, written on thick paper, contains 
(ἢ AESCN τ Vey cL 00 reise (2) OPN. τ τ (rach; Patt... Os Ὁ. 
Codex E, no. 2884, written on paper, contains (1) the same three plays 
of Aesch., (2) Soph. 4z., #2, O. Z., (3) Theocr. /dyl/. 1—14. Both 
these mss. have short interlinear notes and scholia. In E the writing 
is not good, and the rather frequent omissions show the scribe to have 
been somewhat careless. Though the Catalogue assigns E to the 13th 
cent., the highest date due to it seems to be the middle or late 14th. 
T, no. 2711, on thick paper, a Ms. of the 15th cent., exhibits the seven 
plays of Sophocles in the recension of Demetrius Triclinius, the gram- 
marian of the 14th cent. The single-column pages, measuring about 
114 by 74, contain copious marginal scholia, which are mainly Tri- 
clinian. ‘The general features of the Triclinian recension are well- 
known. He occasionally gives, or suggests, improved readings, but 
his ignorance of classical metre was equalled by his rashness, and 
especially in the lyrics he has often made havoc, 

Of the Venetian mss., V, no. 468, a paper folio of the late 13th or 
early 14th cent., contains (1) Oppian; (2) Aesch. P. V., Zheb., Pers., 
Agam. (imperfect): (3) Soph., the 7 plays (but Zvach. only to 18, O. C. 
only from 1338). V*, no. 616, a parchment in small folio, probably of 
the 14th cent., contains (1) Soph., the 7 plays: (2) Aesch., 5 plays (Cho. 
and Sufpé. wanting). Μ΄, no. 467, a paper 8vo. of the 14th cent., has 
the 7 plays of Sophocles. Μ΄, no. 472, a paper 8vo. of the 14th cent., 
has (1) Ar. Plut., αὐ δι, Ran. ; (2) Soph. Az., £7, Ant. (imperfect), ΟΣ Z., 
with marginal scholia. 

Of the Bodleian mss., Laud. Misc. 99 (Auct. F. 3. 25), late 14th 
cent., contains Soph. O. Z:, £2, Az: Laud. 54 (early 15th cent.) the 
same three: Barocc. 66, 15th cent., the same three, with Eur. Phoex. 


1 It contains the entry, ‘Codex optimae notae. Codex Memmianus. Anno D, 
1731 Feb. 16 Die.’ In 1740 it had not yet been collated (Catal. 11. 542). 


MANUSCRIPTS. lvil 


The ms. of Trin. Coll. Camb. (late 14th—early 15th) has £7, Az, 
0:2: 

§ 5. In relation to a text, the report of manuscript readings may be Scope of 
valuable in either, or both, of two senses, the palaeographical and the Saas 
critical. For example, in O. Z. 15 L reads “προσηϊμεθα, and in 17 tion. 
στένοντες. These facts have a palaeographical interest, as indicating 
the kind of mistakes that may be expected in ss. of this age and class. 
But they are of no critical interest, since neither προσήμεθα nor στένον- 
tes is a possible variant: they in no way affect the certainty that we 
must read προσήμεθα and σθένοντες. In a discussion on the character- 
istics and tendencies of a particular MS., such facts have a proper (and 
it may happen to be, an important) place, as illustrating how, for 
instance, « may have been wrongly added, or @ wrongly altered, else- 
where. The editor of a text has to consider how far he will report facts 
of which the direct interest is palaeographical only. 

The general rule which I have followed is to report only those read- 
ings of mss. which have a direct critical interest, that is, which affect a 
question of reading or of orthography; except in the instances, not 
numerous in this play, where a manuscript error, as such, appeared 
specially significant. Had I endeavoured to exhibit all, or even a con- 
siderable part, of the mere mis-spellings, errors of accentuation, and the 
like, which I have found in the mss. which I have collated, the critical 
notes must have grown to an enormous bulk, without any correspond- 
ing benefit, unless to the palaeographical student of the particular codex 
and its kindred. On the other hand, I have devoted much time, care, 
and thought to the endeavour not to omit in my critical notes any point 
where the evidence of the mss. known to me seemed to have a direct 
bearing on the text. 

§ 6. The use of conjecture is a question on which an editor must be The use of 
prepared to meet with large differences of opinion, and must be content Ὁ Ὁ) ἴθ. 
if the credit is conceded to him of having steadily acted to the best of 
his judgment. All students of Sophocles would probably agree at least 
in this, that his text is one in which conjectural emendation should 
be admitted only with the utmost caution. His style is not seldom 
analogous to that of Vergil in this respect, that, when his instinct felt a 
phrase to be truly and finely expressive, he left the logical analysis of it 
to the discretion of grammarians then unborn. I might instance νῦν 
πᾶσι χαίρω (O. 7. 596). Such a style may easily provoke the heavy 
hand of prosaic correction ; and, if it requires sympathy to interpret and 
defend it, it also requires, when it has once been marred, a very tender 
and very temperate touch in any attempt to restore it. Then in the lyric 


Our text— 
bow trans- 
mitted. 


Its general 
condition. 


Iviii LE SCOPE. OF (COMNJECTORE, 


parts of his plays Sophocles is characterised by tones of feeling and 
passion which change with the most rapid sensibility—by boldness and 
sometimes confusion of metaphor—and by occasional indistinctness of 
imagery, as if the figurative notion was suddenly crossed in his mind by 
the literal. 

87. Now consider by what manner of process the seven extant plays 
of this most bold and subtle artist have come down to us through about 
23 centuries. Already within some 70 years after the death of Sophocles, 
the Athenian actors had tampered in such wise with the texts of the 
three great dramatists that the orator Lycurgus caused a standard copy 
to be deposited in the public archives of Athens, and a regulation to be 
made that an authorised person should follow in a written text the 
performances given on the stage, with a view to controlling unwarranted 
change’. Our oldest manuscript dates from 1400 to 1500 years after 
the time of Lycurgus. The most ancient sources which existed for the 
writers of our MSs. were already, it cannot be doubted, seriously 
corrupted. And with regard to these writers themselves, it must not be 
forgotten what their ordinary qualifications were. They were usually 
men who spoke and wrote the Greek of their age (say from the r1th to 
the 16th century) as it was commonly spoken and written by men of 
fair education. On the other hand, as we can see, they were usually 
very far from being good scholars in old classical Greek ; of classical 
metres they knew almost nothing; and in respect of literary taste or 
poetical feeling they were, as a rule, no less poorly equipped. In the 
texts of the dramatists they were constantly meeting with things which 
they did not understand, and in such cases they either simply transmitted 
a fault of the archetype, or tried to make sense by some expedient of 
their own. On the whole, the text of Sophocles has fared better in the 
mss. than that of either Aeschylus or Euripides. This needs no 
explanation in the case of Aeschylus. The style of Euripides, ap- 
parently so near to common life, and here analogous to that of Lysias, 
is, like the orator’s, full of hidden snares and pitfalls for a transcriber : 
λείη μὲν yap ἰδεῖν, as the old epigram says of it, εἰ δέ τις αὐτὴν | εἰσ- 
Baivor, χαλεποῦ τρηχυτέρη σκόλοπος. Where, however, our mss. of 
Sophocles do fail, the corruption is often serious and universal. His 
manuscript text resembles a country with generally good roads, but an 
occasional deficiency of bridges. 

Is there reason to hope that, in such places, more light will yet be 
obtained from the manuscripts or scholia now known to exist? It 


1 [Plut.] Vit. Lycurg. § 11. 


THE SCOPE OF CONJECTURE, lix 


appears hardly doubtful that this question must be answered in the 
negative. The utmost which it seems prudent to expect is a slightly 
increased certitude of minor detail where the text is already, in the 
main, uncorrupted. I need scarcely add that the contingency of a new 
Ms. being discovered does not here come into account. 

§ 8. Such, then, are the general conditions under which an editor of Textual 

Sophocles is required to consider the treatment of conjectural emendation. τ ΠΝ 
It would seem as if a conservative ¢emdency were sometimes held to be have no 
desirable in the editor of a text. When a text has been edited, we vias 
might properly speak of the vesw/¢ as ‘conservative’ or the contrary. 
But an editor has no more right to set out with a conservative tendency 
than with a tendency of the opposite kind. His task is simply to give, 
as nearly as he can ascertain it, what the author wrote. Each particular 
point affecting the text must be considered on its own merits. Instances 
have not been wanting in which, as I venture to think, editors of Sopho- 
cles have inclined too much to the side of unnecessary or even disastrous 
alteration. On the other hand, it is also a serious fault to place our 
manuscripts above the genius of the ancient language and of the author, 
and to defend the indefensible by ‘construing,’ as the phrase is, ‘through 
thick and thin.’ Who, then, shall be the judge of the golden mean? 
The general sense, it must be replied, of competent and sympathetic 
readers. ‘This is the only tribunal to which in such a case an editor 
can go, and in the hands of this court he must be content to leave the 
decision. 

§ 9. The following table exhibits the places where the reading Conjec- 
adopted in my text is found in no Ms., but is due to conjecture. The ae 
reading placed first is one in which L agrees with some other MS. or critics, 
MSS., except where it is differently specified. After each conjecture is ler in 
placed the name of the critic who (to the best of my knowledge) first 
proposed it: where the priority is unknown to me, two or more names 
are given. 

198 τέλει] τελεῖν Hermann. 200 A long syllable wanting. <rév> 
Hermann. 214 —vo wanting. «σύμμαχον» Wolff. 248 ἀμοιρον)] 
ἄμορον Porson. 351 προσεῖπας] προεῖπας Brunck. 360 λέγειν] λέγων 
Hartung. 376 με...γε σοῦ] oe...y ἐμοῦ Brunck. 478 πέτρας ὡς 
ταῦρος (πετραῖος ὁ ταῦρος first hand of L)] πέτρας ἰσόταυρος J. F. 
Martin and E. L. Lushington. 537 ἐν ἐμοὶ] ἔν μοι Reisig. 538 γνω- 
ρίσοιμι]ὔ γνωριοῖμι Elmsley. 539 κοὐκ] ἢ οὐκ A. Spengel. 657 σ᾽ inserted 
by Hermann after λόγῳ 666 καὶ τάδ᾽ τὰ δ᾽ Kennedy (τάδ᾽ Herm.). 

672 ἐλεεινὸν] ἐλεινὸν Porson. 693 εἴ σε νοσφίζομαι] εἴ σ᾽ ἐνοσφιζόμαν 
Hermann, Hartung, Badham. 696 εἰ δύναιο γενοῦ (δύνᾳ first hand in L)] 


Ix Lut SCOPE OF CONJECTORE: 


ἂν γένοιο Blaydes. 741 τίνα δ᾽] τίνος Nauck. 763 ὁ δέ γ᾽ (6 γ᾽ L)] οἵ 
Hermann. 790 προὐφάνη] προὔφηνεν Hermann. 815 τίς τοῦδέ γ᾽ 
ἀνδρὸς νῦν ἔστ᾽ ἀθλιώτερος (others τίς τοῦδέ γ᾽ ἀνδρός ἐστιν ἀθλιώτερος) 
τίς τοῦδε νῦν ἔστ᾽ ἀνδρὸς αθλιώτερος; I had supposed this obvious 
remedy to be my own, but find that P. N. Papageorgius (Betrage p. 26, 
1883) ascribes it to Dindorf in the Poet. Scen.: this then must be some 
former edit., for it is not in that of 1869 (the 5th), and in the Oxford 
ed. of 1860 Dind. ejected the verse altogether: see my crit. note on 
the place. 817 ᾧ...τινα] ὅν...τινι Wunder. 825 μήτ᾽ (μήστ᾽ first hand 
in L)] μήδ᾽ Dindorf. 876 ἀκροτάταν εἰσαναβᾶσ᾽] ἀκρότατα γεῖσ᾽ ἀναβᾶσ᾽ 
Wolff. 877 ἀπότομον] ἀποτμοτάταν Schnelle. . 801 ἕξεται (έξεται, sic, 
L)] θίξεται Blaydes. 893 θυμῶι (others θυμῶ or θυμοῦ)] θεῶν Hermann. 
906 —v-v or v-vg wanting. παλαίφατα Linwood. 943 f. ἦ τέθνηκε 
Πόλυβος ; εἰ δὲ μὴ | λέγω γ᾽ ἐγὼ τἀληθὲς] Triclinius conjectured 4 τέθνηκέ 
που Πόλυβος, γέρον; | εἰ μὴ λέγω τἀληθὲς, which Erfurdt improved by 
substituting Πόλυβος, ὦ γέρον for που Πόλυβος γέρων. 987 μέγας] μέγας 
γ᾽ Porson. 993 ἢ οὐ θεμιτὸν] ἢ οὐχὶ θεμιτὸν Brunck. 1002 ἔγωγ᾽ οὐ 
(ἔγωγ᾽ οὐχὶ A)] ἐγὼ οὐχὶ Porson. 1025 τεκὼν] τυχὼν Bothe, Foertsch. 
1062 οὐκ ἂν ἐκ τρίτης] οὐδ᾽ ἐὰν τρίτης Hermann. 1099 τῶν] τᾶν Nauck. 
1100 προσπελασθεῖς πατρὸς πελασθεῖσ᾽ Lachmann. 1101 ἢ σέ γε 
θυγάτηρ] ἢ σέ γ᾽ εὐνάτειρά τις Arndt. 1109 ᾿Ἑλικωνιάδων] ᾿Ἑλικωνίδων 
Porson. 1137 ἐμμήνους (ἐκμήνους cod. Trin.)] éxuyvovs Porson. 1193 
τὸ σόν τοι] τὸν σόν τοι Joachim Camerarius. 1196 οὐδένα] οὐδὲν 
Hermann. 1205 τίς ἐν πόνοις, τίς ἄταις ἀγρίαις] τίς atas ἀγρίαις, τίς 
ἐν πόνοις Hermann. 1216 A long syllable wanting. «ὧ» Erfurdt. 
1218 ὀδύρομαι] Svpoua Seidler. 1244 ἐπιρρήξασ᾽] ἐπιρράξασ᾽ Dobree. 
1245 κάλει] καλεῖ Erfurdt. 1264 πλεκταῖς ἐώραις ἐμπεπλεγμένην (L 
ἐμπεπληγμένην)" ὁ δὲ | ὅπως δ᾽ (A omits δ). πλεκταῖσιν αἰώραισιν ἐμπε- 
πλεγμένην ὁ δὲ | ὅπως δ᾽ also occurs.] πλεκταῖσιν αἰώραισιν ἐμπεπλεγμέ- 
νην. | ὁ δ᾽ ὡς Campbell. 1279 αἵματος (others αἵματός τ᾽] αἱματοῦς 
Heath. 1310 διαπέταται] διαπωτᾶται Musgrave, Seidler. 1315 ἀδάμασ- 
τον] ἀδάματον Hermann. 70. A syllable “ wanting. «ὄν» Hermann. 
1341 τὸν ὀλέθριον μέγαν (others péya)] τὸν μέγ᾽ ὀλέθριον Erfurdt. 1348 
pnd ἀναγνῶναί ποτ᾽ ἄν (or ποτε)] μηδέ γ᾽ ἂν γνῶναί ποτε Hermann. 1350 
νομάδος] νομαάδ᾽ Elmsley. 1360 ἀθλιος] ἄθεος Erfurdt. 1365 ἔφυ] ἔτι 
Hermann. 1401 μέμνησθ᾽ ὅτι] μέμνησθέ τι Elmsley. 1494 f. τοῖς 
ἐμοῖς | γονεῦσιν] ταῖς ἐμαῖς γοναῖσιν Kennedy. 1505 μή σφε παρίδῃς] μή 
σφε περιίδῃς Dawes. 1513 ἀεὶ] ἐᾷ Dindorf. 1517 εἰμι] εἶμι Brunck. 
1521 νῦν...νῦν] νυν «νῦν Brunck. 1526 ὅστις... καὶ τύχαις ἐπιβλέπων] 
οὗ τίς...ταῖς τύχαις ἐπέβλεπεν Hartung, partly after Martin and 
Ellendt. 


EDITIONS AND COMMENTARIES. 1x1 


§ 10. The following emendations, adopted in the text, are due to Con- 
the present editor. The grounds on which they rest are in each case eae 
stated in the commentary :— editor. 

227 ὑπεξελὼν | αὐτὸς] ὑπεξελεῖν αὐτὸν. 
624 ὅταν] ὡς av. 
640 δρᾶσαι... δυοῖν] δυοῖν... δρᾶν. 

τορι Οἰδίπου] Οἰδίπουν. 

1218 ὡς περίαλλα ἰαχέων (vv. Ul. περίαλα, ἀχέων)] ὥσπερ ἰάλεμον χέων. 

1405 ταὐτὸν] ταὐτοῦ. 

One conjectural supplement is also the editor’s: 

493 «βασανίζων». 

In a few other places, where I believe the text to be corrupt, I have 
remedies to suggest. But these are cases in which the degree of proba- 
bility for each mind must depend more on an ἄλογος αἴσθησις. Here, 
then, the principles of editing which I have sought to observe would 
not permit me to place the conjectures in the text. In the commentary 
they are submitted to the consideration of scholars, with a statement of 
their grounds in each case. 1090 οὐκ ἔσει τὰν αὔριον] τὰν ἐπιοῦσαν ἔσει. 

ΙΙΟῚΙ ἢ σέ γε θυγάτηρ | Λοξίου" ;] ἢ σέ γ᾽ ἔφυσε πατὴρ | Λοξίας"; 1315 
δυσούριστον ¥] δυσούριστ᾽ ἰόν, 1350 νομάδ᾽] μονάδ᾽. 

δ΄11. In my text, a conjecture is denoted by an asterisk, ἔτελεῖν for Notation. 
τέλει in v. 198: except in those cases where a slight correction, which at 
the same time appears certain, has been so generally adopted as to have 
become part of the received text; as ἄμορον for ἄμοιρον in 248. In 
such cases, however, no less than in others, the fact that the reading is 
due to conjecture is stated in the critical note. A -word conjecturally 
inserted to fill a lacuna is enclosed in brackets, as «τᾶν; in ν. 200. 

The marks { 7 signify that the word or words between them are be- 
lieved by the editor to be unsound, but that no conjecture seemed to him 
to possess a probability so strong as to warrant its insertion in the text. 

§ 12. Editions.—The following is an alphabetical list of the Editions. 
principal editions of Sophocles, with their dates. Separate editions of 
this play are marked with an asterisk.—Aidus (Venice, 1502: the ed. 
princeps).—Bergk (1858).—Blaydes (1859).—Bothe (1806).—Brunck 
(1786).—Burton (Soph. O. Z7:, O. C., Ané., with Eur. Phoen., and Aesch. 
Theb.: 2nd ed., with additions by T. Burgess, 1779).—Camerarius, 
Joachim (15 34).—L. Campbell (2nd ed., 1879).—Canter (1579).—Dindorf 
(3rd Oxford ed., 1860: 6th Leipsic ed., revised by S. Mekler, 1885).— 
Elmsley (1825).—Erfurdt and G. Hermann (1809-1825 : new ed., 1830 


1 See Appendix on verse 1190. 


Subsidia. 


Ix EDITIONS AND COMMENTARIES. 


-1866. Hermann’s first recension of the Oed. Zyr., in the above edition, 
appeared in 1811; the second, in 1823; the third, in 1833).—Hartung 
(1851).—*Herwerden (1851).—T. Johnson (t745).—Junta (Florence, 
and ed., 1547).—*Kennedy (1882).—*Kennedy, with notes by T. H. 
Steel (1885).—Linwood (4th ed., 1877).—J. F. Martin (1822).—Matthiae 
(1825).—Musgrave(1800).— Neue (1831).—*Fr. Ritter(1870).—Schaefer 
(1810: new ed., 1873).—M. Schmidt (1871).—Schneider (2nd ed., 
1844).—Schneidewin, revised by Nauck (new ed., 1886).—H. Stephanus 
(H. Estienne, 1568).—Tournier (2nd ed., 1877).—Turnebus (Paris, 
1552-3).—Vauvilliers (1781).—Wecklein (1876).—*White, J. H. (new 
ed., 1879).—* Wolff-Bellermann (2nd ed., 1876).—Wunder (new English 
ΕΠ} 185.5). 

§ 13. Subsidia.—The scope of the following list is limited to in- 
dicating some of the principal writings consulted for this edition.— 
Arndt (Quaestiones criticae, &vc., 1844: Kritische u. exegetisthe Bemer- 
kungen, &¢., 1854: Bettrage 2. Krittk des Soph. Textes, &¢., 1862).— 
Badham (A/iscellanea, 1855).—Butcher (in Fortnightly Review, June, 
1884).—Cobet (Var. Lectiones, 2nd ed., 1873).—Dobree (Adversaria, 
1831).—Doederlein (Minutiae Sophocleae, 1842-47).—Ellendt (Lexicon 
Sophocleum, 1872).—Emperius, Ad. (Axalecta critica, 1842).—Gleditsch, 
Hugo (Die Sophokletschen Strophen metrisch erklart, 1867-8).—Heath 
(Wotae sive Lectiones, &c., 1762).—Heimsoeth (K7ritische Studien, 1865 : 
Commentatio critica on textual emendation, continued in several parts, 
1866-1874).—K vicala, Joh. (Beitrage 2. Kritik, &c. des Soph., part 1v., 
1869).—Otto, Clem. (Quaestiones Soph. Criticae, 1868-1876).—Papa- 
georgius, P. N. (Betrage 2. Erklarung, &c. des Sophokles, 1883).— 
Porson (Adversaria, 1812).—Purgold, L. (Odbss. Crit. in Soph., &¢., 
1802).—Reiske (Animadverstones ad Sophoclem, 1743?).—Schmidt, F. W. 
(Kritische Studien, 1886: also several earlier tracts).—Seyffert, M. 
(Kritische Bemerkungen zu Soph. Oecd. Tyr., 1863).—Wecklein (Ars 
Sophoclis emendandi, 1869).—Whitelaw, R. (lVotes on the Oed. Rex, in 
Transactions of the Cambridge Philological Society, vol. 111., part 1., 
1886. ‘The same part of the vol. contains Grammatical Annotations 
upon the Oed. Rex, by J. P. Postgate: and Vole on Oed. Rex, 43 599., 
by C. A. M. Fennell).—Occasional reference has also been made 
to many other scholars who have discussed particular points or 
passages of this play. A useful clue to many of these is given by 
H. Genthe’s /udex Commentt. Sophoclearum from 1836 to 1874 (the 
date of issue), in which §§ 541—616 (pp. 66—73) relate to the Oedipus 
Tyrannus. 


METRICAL ANALY ΞΕ 


In my text, I have exhibited the lyric parts with the received 
division of verses, for convenience of reference to other editions, and 
have facilitated the metrical comparison of strophe with antistrophe by 
prefixing a small numeral to each verse. 

Here, in proceeding to analyse the metres systematically, I must 
occasionally depart from that received division of verses—namely, 
wherever it differs from that which (in my belief) has been proved to be 
scientifically correct. These cases are not very numerous, however, and 
will in no instance cause difficulty. 

The researches of Dr J. H. Heinrich Schmidt into the Rhythmic 
and Metric of the classical languages have thrown a new light on the 
lyric parts of Greek Tragedy’. A thorough analysis of their structure 
shows how inventive and how delicate was the instinct of poetical and 
musical fitness which presided over every part of it. For the criticism 
of lyric texts, the gain is hardly less important. Conjectural emend- 
ation can now in many cases be controlled by more sensitive tests 
than were formerly in use. To take one example from this play, we 
shall see further on how in v. 1214 the δικάζει τὸν of the MSS. is cor- 
roborated, as against Hermann’s plausible conjecture δικάζει τ, The 
work of Dr Schmidt might be thus described in general terms. Setting 
out from the results of Rossbach and Westphal, he has verified, cor- 


1 Dr Schmidt’s work, ‘Die Kunstformen der Griechischen Poesie und ihre Be- 
deutung,’ comprises four volumes, viz. (1) ‘Die Eurhythmie in den Chorgesangen der 
Griechen,’ &c. Leipzig, F. C. Vogel, 1868. (2) ‘Die antike Compositionslehre,’ &c. 
16. 1869. (3) ‘Die Monodien und Wechselgesinge der attischen Tragiédie,’ ἄς. 20. 
1871. (4) ‘Griechische Metrik,’ 7. 1872. | 


Prelimin- 


ary 
remarks. 


lxiv METRICAL ANALY SIS. 


rected, and developed these by an exhaustive study of the Greek 
metrical texts themselves. ‘The essential strength of his position con- 
sists in this, that his principles are in the smallest possible measure 
hypothetical. ‘They are based primarily on internal evidence afforded 
by Pindar, Aeschylus, Sophocles, Euripides and Aristophanes. To 
Dr J. W. White, Assistant Professor of Greek at Harvard University, 
is due the credit of having introduced Dr Schmidt’s system to English 
readers’. 

With regard to the lyric parts of this play, were I to give merely 
a skeleton scheme of them, the application of it to the Greek text 
might prove a little difficult for those who are not already acquainted 
with the results indicated above. For the sake, therefore, of greater 
clearness, I give the Greek text itself, with the scheme applied to it. 
Such notes as appeared requisite are added. 

A few explanatory remarks must be premised. 

A syllable of speech, like a note of music, has three conditions of 
utterance: (1) length of tone, (2) strength of tone, (3) height of tone. 

(1) Length of tone—according as the voice dwells a longer or 
shorter time on the syllable—is the affair of Quantity. A ‘short’ 
syllable, as distinguished from a ‘long,’ is one which is pronounced 
in a shorter time. (2) Strength of tone—according to the stronger or 
weaker ‘beat,’ z¢fus, which the voice gives to the syllable—is the affair 
of Rhythm. ‘Rhythm’ is measured movement. The unity of a 
rhythmical sentence depends on the fact that one syllable in it has a 
stronger ictus than any other. (3) Height of tone—according as the 
voice has a higher or lower pitch—is the affair of Accent. 

In modern poetry, Accent is the basis of Rhythm. In old Greek 
poetry, Quantity is the basis of Rhythm, and Accent has no influence 
which we can perceive. The facts which we have now to notice fall, 
then, under two heads: I. Quantity, as expressed in Mere: and II. 
Rhythm. 


1 By his excellent translation, made conjointly with Prof. Dr Riemenschneider, 
and revised by Dr Schmidt, of the ‘Leitfaden in der Rhythmik und Metrik der 
Classischen Sprachen’ (Leipzig, 1869)—an epitome, for schools, of the principles 
established in the ‘ Kunstformen.’ The ‘Introduction to the Rhythmic and Metric of 
the Classical Languages’ was published at Boston, by Ginn and Heath, 1878; and in 
Prof. White’s edition of this play (2d. 1879) the lyrics are constituted in conformity 
with it. Here, I have felt it necessary to assume that few of my English readers 
would be familiar with Dr Schmidt’s results, and have therefore deemed it expedient 
to give fuller explanations than would otherwise have been necessary. 


METRICAL ANALYSIS. Ixv 


I. Metre. §1. In Greek verse, the short syllable, denoted by v, Metre. 
is the unit of measure, and is called ‘a time’ (Lat. mora): a long 
syllable, --, has twice the value of a short; so that πο is a foot of 


‘three times.’ The short syllable has the musical value of a quaver eh 


or ἃ note (1.6. eight of which make zz). The long syllable has there- 
fore the value of J or a i note. 

§ 2. As in music ak signifies that the + note has been made one- 
half as long again (#.e. 44+ ἢ -- 8), so in Greek verse the long syllable 
could be prolonged by a pause, and made equal to ¢hree short syllables. 
When it has this value, instead of — we write -. 

ὃ 3. Ina metrical foot, there is always one syllable on which the 
chief strength of tone, or ictus, falls. This syllable is called the arszs 
of the foot. The rest of the foot is called the ¢heszs’. When a long 
syllalle forms the avszs of a measure, it can have the value of even 
more than three short syllables. When it becomes equivalent to four 


[Ξ =I a 4 note), it is written thus, 4. When to five (= oe 5 note), 


thus): 

§ 4. When the long syllable (written “) is made equal to ¢hree 
short, it can be used, alone, as a metrical substitute for a whole foot of 
three short ‘times,’ viz. for — v (trochee), ὦ -- (ambus), or vv (tribrach). 
So, when (written J) it has the value of four short, it can represent a 
whole foot in $ ($) measure, viz. -~ wu (dactyl), ὦ ὦ -- (anapaest), or 
——(spondee). And so w can replace any δ measure, as -u-, —vuv, 
vuv- (paeons), v--, —-v (bacchii). This representation of a whole 
foot by one prolonged syllable is called symcoge, and the foot itself is ‘a 
syncopated trochee,’ &c. 

§ 5. When two short syllables are used, by ‘resolution,’ for a long 
one ( Bah, for o) this is denoted by *. Conversely the sign σὺ 
means that one long syllable is used, by ‘contraction,’ for two short 
ones. | 

§ 6. An ‘irrational syllable’ (συλλαβη ἄλογος) is one which has a 
metrical value to which its actual ¢2e-value does not properly entitle it. 


1 This is the reverse of the old Greek usage, in which θέσις meant ‘ putting down 
the foot’ (and so the syllable which has the ictus), ἄρσις, the ‘lifting’ of it. Roman 
and modern writers applied avszs to ‘the raising of the vozce,’ ἐλεεῖς, to the lowering of 
it. Dr Schmidt has reverted to the Greek use, which is intrinsically preferable, 
since the modern use of the term ‘arsis’ tends to confuse ¢c¢us with accent. But 
the modern use has become so general that, in practice, it appears more convenient to 
retain it; and I have done so. 


Soe e 


Rhythm. 


Ixvi METRICAL ANALYSIS. 


The most frequent case is when a long stands for a short in the thesis of 
a foot, which is then ‘an irrational foot.’ The irrational syllable is 
marked >. Thus in the trochaic verse (O. Z: 1524), ὦ warp | as 


6nB\ns, the syllable θη is irrational, and as θηβ is an irrational 
trochee. ‘The converse use of an irrational short syllable instead of a 
long is much rarer, occurring chiefly where — uv is replaced by an 
apparent συν (written ὦ}, or —— by an apparent —v (written 
— =). Ina metrical scheme 2 means that a long syllable is admitted as 
an irrational substitute for a short one. 

§ 7. When a dactyl takes the place of a trochee, it is called 8. 


cyclic dactyl, and written ~v. The true dactyl (- υὐ}Ξ J 42: the 
cyclic = Γ > |: 2.6. the long syllable loses 4 of its value, and the first 


I 
short loses 4, so that we have τῷ ἐπε τῖ-ϑ8. 


3 So the cyclic anapaest, 


wv, can replace an iambus. 


ὃ 8. A measure can be introduced by a syllable external to it, and 
having no ictus. This syllable is called the anacrusis (ἀνάκρουσις, 
‘upward beat’). It can never be longer than the thesis of the measure, 
and is seldom less. Thus, before -v, the anacrusis would properly 
be ὦ (for which an irrational syllable>can stand). Before —uy, it 
would be vv or—. The anacrusis is divided from the verse by three 
vertical dots :. 

§ 9. It will be seen that in the Parodos, 2nd strophe, 1st period, 
3rd verse, the Greek letter ὦ is printed over the syllables στόλος which 
form the anacrusis. This means that they have not the full value 
of uv or two ἃ notes ( J), but only of two τς notes ( 5). 

§ 10. auses. The final measure of a series, especially of a verse, 
might always be incomplete. Then a pause represented the thesis of 
the unfinished foot. Thus the verse viv δ᾽ ἔπικεκλομένἃ vv is in- 
complete. The lacking syllables vu are represented by a pause. The 
signs for the pause, according to its length, are as follows :— 

A pause equal to v is denoted by a, musically ἡ for a 


99 2) αἰεὶ 2) 2) A ’ ” [Ὁν» a 
aS | 

3) ” = ” ” A> 3) [> 92 we 
| 

” ” “τὺ ” 2) A> ” wm 99 


II. Rhythm. § 11. Metre having supplied feet determined by 
quantity, Rhythm combines these into groups or ‘sentences’ determined 
by ictus. Thus in verse 151, ὦ Διὸς adverés dati, || τίς ποτε τᾶς 


METRICAL ANALYSTS. Ixvil 


πολυχρύσου, there are two rhythmical sentences. The first owes its 
rhythmical unity to the chief ictus on ὦ, the second to the chief ictus 
on τίς. Such a rhythmical κῶλον or sentence almost always consists of 
feet equal to each other. The end of a sentence is denoted by the sign |]. 

§ 12. Rhythmical seztences are again combined in the higher unity 
of the rhythmical Zeriod. Here the test of unity is no longer the 
presence of a chief ictus on one syllable, but the accurate correspond- 
ence with each other of the sentences which the period comprises. The 
period is seen to be such by the fact that it is neither less nor more than 
an artistic and symmetrical whole. 

§ 13. In the choric type of lyrics, which Tragedy uses, we find, as 
in other Greek lyric types, the rhythmical sentence and period. ‘Their 
correspondence is subordinate to that of strophe and antistrophe. 
Each strophe contains usually (though not necessarily) more than one 
rhythmical period. Each period of the strophe has its rhythmical 
counterpart in a period of the antistrophe. And, within each period, 
the rhythmical ‘sentences’ (κῶλα) accurately correspond with each other. 

§ 14. In the choric dance which accompanied the choric song, the 
antistrophe brought the dancer back to the position from which, at the 
beginning of the stvophe, he set out. Hence the necessity for strict 
metrical correspondence, z.e. for equal duration in time. When any 
part of a choric song is non-antistrophic, this means that, while that part 
was being sung, the dancers stood still. A non-antistrophic element 
could be admitted in any one of three forms: viz. (1) as a verse 
prefixed to the first strophe—a ‘prodde’ or prelude, τὸ προῳδικόν, ἡ 
προῳδός, denoted by zp.: (2) as a verse inserted between strophe and 
antistrophe—a ‘mesode’ or znterlude, τὸ μεσῳδικόν, ἡ μεσῳδός : (3) as a 
verse following the last antistrophe—an ‘ epode’ or postlude, τὸ ἐπῳδικόν, 
ἢ ἐπῳδός". 

During the pause at the end of a verse in a choric ode of Tragedy, 
the dance and song momentarily ceased ; but instrumental music pro- 
bably filled the brief interval. Such pauses correspond no less exactly 
than the other rhythmical divisions. 

We will now see how these principles are exemplified in the lyrics 
of the Oedipus Tyrannus. Under each line of a strophe I give in 
smaller type the corresponding line of the antistrophe, since the 
comparison is often instructive, especially with regard to irrational 
syllables. 


1 Distinguish the masc. ὁ ἐπῳδός, a retrain, esp. the epodic distichon as used by 
Archilochus and Horace. 


e2- 


LG 


ik 


Ixvill METRICAL ANALYSTS. 


I. Parodos, vv. 151—21I5. 


First STROPHE. 


(I., II., denote the First and Second Rhythmical Pertods. The 


sign || marks the end of a Rhythmical Sentence; ἢ marks that of a 


Period.) 
Sr aed, pees DA ah) vv are VY “ea Δ ποι τ 
1. w διος | adver | ες date || τις ποτε | τας πολυ | χρυσου || 
πρωτα ce | κεκλομεν | os Ovyar || ep dios | apBporad| ava | 
Low Livy L_ . - 
2. mv : θωνος | αγλα | ac eB | ac A || 
yor : aox |ovrad| edde | a Ι 
πο ἀν. “τῷ! ὦ τον ΣΝ — VY FS a 
3. OnBas | exterap | at φοβερ || av φρενα | δειματι | παλλων || 
αρτεμιν | α κυκλο | evrayop || as θρονον | evxrea | θασσει || 
= vy πων UW - 
4. e δ me | dare | rac] av ἡ] 
και :potBov ex | aBorov| « | w 7 
= - Po ΠΑ Ξ Pw δ ΩΣ = WS en ad Nd — GO bee vv 
I. api cor αζομεν | os τι μοι | 7 veov || ἡ περι] τελλομεν | ats wp | ats παλιν |} 
τρισσοι a | λεξιμορ] οἱ προφαν | ητε μοι || εἰποτε | και προτερ | ag ar | as ὑπερ |i 
eI Ne. — wv ἐπ Nd, re Δ Vv Vv — Δ, τον VY ap 
2. εξανυσ | evs xpeos | εἰπε μοι] w χρυσε || as τεκνον | ελπιδος | ap pore | Papo] 
ορνυμεν | as rode | ἡνυσατ | ex rome || av φλογα | πηματος] ελθετε | και νυν} 
l.. first Pertod = 4: verses... Metre, dactyic:. Verse 1. The 
comma after — in the 3rd foot denotes caesura. Verse 2. The 


dots : after zv show that it is the amacrusis: see ὃ 8. The sign 
- means that the long syllable here has the time-value of —v or a 
2 note, so that @wvos=a dactyl, -vv: see § 2. This verse forms a 
rhythmical sentence of 3 dactyls, a dactylic tripody. It is known as a 
‘Doric sentence,’ because characteristic of Doric melodies ¢ Pind. Οἱ. 
8. 27 κίονα | dapovi| av QA ||: 2. 40 εἷς δ᾽ ἐσόρ | ουσε Bo | aoas ||. 
The sign ( marks a pause equal to vu: 566 ὃ το. Verse 3. σὺ shows. 


~ LI 

that as represents, by contraction, vw. Verse 4. παι has the time- 
value of a whole dactyl —vv, or £ measure: this is therefore a case of 
syncope, see ὃ 4. When syncope occurs thus in the penultimate measure 


METRICAL ANALYSIS. Ixix 


of a rhythmical sentence or of a verse, it imparts to it a melancholy 
cadence: and such is called a ‘fa//img’ sentence or verse. 

Now count the sentences marked off by |i. In v. 1, we have 2 
sentences of 3 feet each; 3, 3. In v. 2 one sentence of 4 feet; 4. 
In v. 8, the same as inv. 1. Inv. 4, the same as in v. 2. The series 
thus is 3 3. 4. 33.4. This determines the form of the entire Rhythmical 
Period, which is expressed thus :— 


Here the curve on the ἐξ means that one whole 
group (verses 1, 2) corresponds with the other whole 
group (verses 3, 4). The curves on the righ# mean 
4 that the rst sew¢ence of the 1st group corresponds to 
the 1st of the 2nd, the 2nd of the 1st to the 2nd of 
the 2nd, the grd of the 1st to the 3rd of the 2nd. 


.“ W 


9 . 

The vertical dots mean that the figure or figures be- 
ὃ tween any two of them relate to a single verse. 
4 This is called the palinodic period: meaning that 


a group of rhythmical sentences recurs once, in the 
same order. 


II. Second Period: 2 verses. Metre, still dactylic. Verse 1. The 


last foot, acs EFS is a true dactyl (not a ‘cyclic,’ see § 7); it is not 
contracted into ——; and it closes a rhythmical sentence. Now, when 
this happens, it is a rule that the immediately preceding foot should be 
also an uncontracted dactyl. Why do not as wp, as at, break this rule? 
Because, in singing, two + notes, oe’ instead of one } note, o were 


given to the syllable wp, and likewise to ar. This is expressed by 
GS ww 
writing wp, and not merely wp. 

In ν. 1 we have two rhythmical sentences of 4 feet each: 4, 4. In 

v. 2, the same. The series, then, is 44. 44., and the form of the 


Rhythmical Period is again palinodic :-— 


1 


Ιχχ METRICAL ANALYSIS. 


SECOND STROPHE, 


Ὡ» ἐλ ἘΦ ΤΟΣ NSF τ), δε = = 


. I. ὦ : πόποι αν | αριθμα | yap dep | w A || 


wy : mods av | αριθμος | ολλυ | ται 
> a A --- Vv am Vv ag 

2. πη : ματα vor | ede | por mpo | ras A |} 
vn : Nea de | γενεθλα | προς wed | w 


ω ΨΥ τὼ ΠΩ. ὦ ι- - 
3. στολος : ovd ενι [φροντιδος | eyx | os A J 
Oavar : agopa | κειται αν | oxr | ws 
el Δ aaah καὶ, 4 on it Vv Sa NS. 
I. w tis a | λεξεται | ovre yap | exyova || 


ενδ adox | οἱ πολι | air eme | ματερες 


ΞΞ — wv —= vw VW —- vy vw —_—- 


2. κλυτ : as χθονος | αὐξεται | ovre tox | οισιν || 


axr :av mapa | βωμιον | αλλοθεν | αλλαι 
> - OS — vv Se EYES Lj - 
3. 6 ἢ t | wv kapar | wv avex || ovor yuv | ax | es A || 
λυγρ : wy πὸν | wy uxt | npes er || ἐστεναχ | ovo | w 
- --ὖ Le Ss he Oe ae th aL wate KOPROD eG 
4. αλλ: ovd av | αλλ | w προσιδ || ots απερ | evrrepov | opvw || 
mat : av de |λαμπ|ει στονο || εσσα τε | γῆήρυς om | avdos 
ὅπ VW Vv πὶ ale NGS, = vv πε πω 


5. κρεισσον a | μαιμακετ | ov πυρος | ορμενον || 
wy ὑπερ | w xpvce | a Ovyar | ep διος 
πολ ee oe ere) ge 
6. axt : av προς | ἐσπερ | ov | Oeov A J 


ev :w πα | πεμψον | αλκ] av 


Ln lapstePeviod = 3:verses «The metrical basis of the rhythm is the 
choree (or ‘trochee,’ -- υ), for which the cyclie dactyl (~ ὦ, see § 7) and 
tribrach (uv vu) can be substituted. The rhythm itself is ogacedic’. When 


1 The name λογαοιδικός, ‘ prose-verse,’ meant simply that, owing to the apparently 
lawless interchange of measures (τον, VY, — >, for τ ὦ) in this rhythm, the old 
metrists looked upon it as something intermediate between prose and verse. It should 
be borne in mind that the essential difference between choreic and logaoedic rhythm 
is that of zc¢us, as stated above. The admission of the cyclic dactyl is also a specially 
logaoedic trait, yet not excluszvely such, for it is found occasionally in pure choreics 
also. The question, ‘Is this rhythm choreic or logaoedic?’ can often be answered 
only by appeal to the whole poetical and musical character of the lyric composition,— 


METRICAL ANALYSTS. Ixx1 


chorees are arranged in ordinary chorezc rhythm, the ictus of arsis is to 
that of thesis as 3 to 1 aa when in logaoedic, as 3 to 2 (ad)): The 
latter has a lighter and livelier effect. Verse 1. The anacrusis w is 
marked >, since it is an ‘irrational’ syllable (§ 6),—a long serving for a 
short. The anacrusis can here be no more than ὦ, since it can never 
be longer than the thesis (§ 8), which is here ὦ, since σὺ ὦ represents 
—v. Verse 3. ὦ written over στόλος means that the two short syllables 
here have only the time-value of v, or SS, not of ω ὦ or ro: see § 9. 
oe o 


ουδενι and φροντιδος are cyclic dactyls (~ υ =—v), not true ones (—vv), 
see 87. The second syllable of eyxés is marked Jong, because the last 
syllable of a verse (syllaba anceps, συλλαβὴ ἀδιάφορος) always can 
be so, and here os is the first of a choree, —u, which the pause A 
completes. 

Verses I, 2, 3 contain each one rhythmical sentence of 4 feet; the 
series is therefore. 4.4. 4., and the form of the period is :— 


4 When ¢wo rhythmical sentences of equal length correspond to 
each other, they form a ‘stichic’ period (στίχος, a line or verse); 

when, as here, more than two, they form a repeated stichic 
Ὶ period. 


Il. Second Period: 6 verses. Metre, dactylic. Verse 2. The 
anacrusis κλυτ is marked = since it is a really short syllable serving 
‘irrationally’ (δ 6) as a long: for, the measure being — vv, the anacrusis 
should properly be uv or — (as axz in the antistr. actually is). Verse 8. 


ee (§ 4). This syncope (§ 4) in the penult. measure makes a 
‘falling’ verse: see on Str. 1. Per. LD v. 4. A =a pause equal to vu 


(§ το). 


the logaoedic ictus being always more vivacious than the choreic. See, on this subject, 
Griech. Metrik § 19. 3. Students will remember that ‘logaoedic verse’ is a generic term. 


Sat SA) ἘΞ 
Three kinds of it have special names: (1) the logaoedic digodia, as καμπυλον | αρμα ἢ, 
. ᾽ . . xd = ay = πυ Ἷ . 
is an ᾿Αδώνιον μέτρον : (2) the ¢ripodia, βυρσοτον | ov κυκὰ | wua ||, a Pepexpdrecov: 
᾿ : a . ae Vv — Vv ona er, 
(3) the ze¢rvapodia, which is very common, νυν yap eu | οἱ wed | εἰ χορ | ευσαι ||, is the 
‘glyconic,’ Τλυκώνειον. (2) and (3) can vary the place of the cyclic dactyl, and can 


be catalectic. The logaoedic (5) pemtapodia and (6) hexapodia, both of which occur 
in tragedy, are not commonly designated by special names. 


ΙΧΧΙΙ METRICAL ANALYSTS. 


Verse 1 contains 1 rhythmical sentence of 4 feet: v. 2, the same: 


Vv. 3, two sentences each of 3 feet: v. 4, the same: vv. 5, 6, the same 
ΔΘ “SClIeS? 4, 33-3134. 4, and the form of period 1s :-— 


Tes: 


( ᾿ The curves on the ἐγ show the corre- 
| 4 spondence of whole rhythmical groups; 
those on the χίρλέ, that of rhythmical sen- 

3 tences. 
3 If the second group οὗ. 3 3. had followed 
4 the second of .4.4., this would have been 
3 a simple palinodic period, like the 1st of 
λ Strophe 1. But as the groups are repeated 


4— in reversed order, it is called a palinodic 


iN antithetic period. 
ἰ 4 
THIRD STROPHE. 


ap neh ea te τον | μαλερον | os || νυν a | χαλκος | cous | wy A || 

λυκ : εἰ av | a& | ta Te ca | χρὺσ || ocrpop | ων απ | αγκυλ | αν 
“sv συ ~v -v ue = 

precy : εἰ pe | περιβο] aros | αντι[ af | wv A | 


BeX i€a0eX| οιμαν | adauar | evdar | aod | αἱ 


Vv πὰρ ae τ ΡΣ ee, - ΕΞ 
mar : ισσυτ] ov δραμ | nua | νωτισ | αἱ πατρ | ας A || 

ap : wya | προσταθ | evra | τας τε | πυρῴορ | ous 

ἘΠ 1 oe ep en ce ee ee ἊΣ αν 

ex : ovpov | εἰτ | ες pey | αν Ἰ θαλαμον | ee | sper | as A JJ 
apr : eucdos | avy | as ξυν | acs || λυκι op | ἢ δι | goo | εἰ 


a =~ Vv συ ι-- ι- we = “ -- — 

eit : ες τον απ ofevov | opp | ον || θρῃκι | ον KAvd | wv | a A || 
Tov : χρυσομίιτρ | av te Kc| κλησκ | w || tacder | ωὠνυμ | ov | yas 

Ξ a an Nad, ΡΥ ἂν [ΞΞΞ Fa ΤΑΝ ἀπε δὰ τς = 

TeX : εἰν ἘΣ | εἰ τι | νυξ αφ | 7 || τουτ ex | ἡμαρ! ee os at A || 
ov : wma | Baxxov | eve | ov || μαιναδ | wv ou | οστολ | ov 

eh, Sg: ον μετ ον ae ee os 


tov : w | tav | πυρφορ | wr || aatpar | av ΤῈΣ | n ven | wy A || 
πελ : ασθ] nv | αἱ φλεγ | ovr  αγλὰ | wre | συμμαχ ον 


METKICAT, ANALYSIS. Ixxill 


> Se NS SINS = Vv fete NS) ι- ΞΞ 
4. wo : fev rat | ep vt0 | cw φθισ | ov κερ | avy | w A |] 


meux : a me | rovaro | timov | ev Be | os | θεον 


I. First Period: 4 verses. The choree —u is again the fundamental 
measure, as in Str. 11. Per. 1., but the choreic rhythm here expresses 
greater excitement. Verse 1. The place of the sywcope (L--, § 4) at τον 
and os, each following a tribrach, makes a ‘7éstmg’ rhythmical sentence, 


in contrast with the ‘ fad/ing’ sentence (see Str. 1. Per. 1. v. 4), such as 
> 


verse 4. This helps to mark the strong agitation. Verse 4. em means 
that the proper anacrusis, ὦ, can be represented by an ‘irrational’ 
syllable (as apr in the antistr.). 

Verse 1 has 2 sentences of 4 feet each: 2, 1 of 6% 3; ‘the same: 
4, the same as 1. Series: .44.6.6.44. Form of period :— 


4 
ty 

6 A palinodic antithetic period, like the 
᾿ last. 


II. Second Period: 4 verses. Metre, still choretc. Note the weighty 
effect given by syncope () in the ‘falling’ sentences of v. 1, and in 
v. 3. Inv.1; er is marked > (‘ irrational’), because the following dactyl 
is only cyclic (equal to —v), and the thesis being vu, the anacrusis cannot 
be more: cp. v. 4. 

Verses 1, 2, 3, having each 2 sentences of 4 feet each. Verse 4 
forms 1 sentence of 6 feet, to which nothing corresponds: ze. it is an 
epode (δ 14), during the singing of which the dancers stood still. (This 
was dramatically suitable, since Oedipus came on the scene as the last 
period began, and his address immediately follows its conclusion.) 

Series :—4 4.44.44. 6-- ἐπῳδικόν. Form of period :— 


ἰχχὶν METRICAL ANALYSIS. 


SS The period is generically palinodic, since a group 
recurs, with the sentences in the same order. But 
4 the group recurs more than once. This is therefore 
- called a repeated palinodic period, with ‘epode’ or 
postlude. 
4 
‘i 
6 = ἔπ. 
II. First Stasimon, vv. 463—512. 
FIRST STROPHE. 
v OW L_ ~"UY στὸν « = = NS ι- “Ὄπ 


? 
I, 1. tus : ovrw | a | Oeomer | eta || Seddhis | εἰπε | wetp | a A || 


εἰ λαμψε | yap | rov vido | εντος  αρτι [ὡς φαν | εἰσ [α 
a =<) i= τ Vv axe τ ϑλὺ, =i eh ι- — 

2. appyt | appyt | wv teA€ | σαντα || φοινι | aur [χερσ ιν A J 
gaya | rapvacc | ovtova | Sydov. || avdpa | wavr xv | εὖ | ew 


= το ἘΞ iN. - 
11. 1. wp : ανινα ‘| edad [ων A || 
. gor : ayapur | ayp [αν 
> -ῳ v = 
2. wma : wv obevap | wrep | ov A | 
υλ : ἂν avat | avrpa | και 
vy ON, lL 
2. pry : @ moda | vom | av A ἢ 
metp :as to | Tavup | os 
ω = v “π- ὴὴω v ae = 
III. 1. evorA : os yap ex | avrov ex | ενθρωσκ | εἰ A || 


pete : os pere | w modi | xnpev | wy 


METRICAL ANALYSIS. xxv 


ω πο Vv wy b= Vov ὩΣ 
8. πυρι : και στεροπ | αἷς ο δι | os yever | as A | 
Taper: ομῴαλα | -yasamro | νοσφιζ | wy 
> ~ - the — 0 VvVviy tL. 
3. Sev : αἱ ὃ αμεπ᾿ | ονται | κηρες | uvardax | nr {or A ἢ 


μαντ : earad | ae | ζωντα | meperor | ar | a 


I. First Period: 2 verses. Rhythm, /ogacedic, based on the choree, 
πὸ: see Parodos Str. 2. Period 1. Each verse has 2 sentences of 4 
feet each. Series: .44.44. Form of period :— 


A palinodic period, like the 1st of Parod. Str. 1. 


Il. Second Period: 3 verses. Rhythm, the same, but in shorter, 
more rapid sentences. Each verse has 1 sentence of 3 feet. Series: 
- 3-3-3. Form of period :— 


A repeated stichic period: see Parod. Str. πὶ Per. 1. 
3 


III. Zhird Period: 3 verses. Rhythm, the same: remark the 
weighty hexapody of v. 8, expressing how the hand of the avenging god 
will be heavy on the criminal. In v. 2, w-written over yever (see § 9) 
means that the time-value of the two syllables was here 43: 1.2. OS yeveT 


was not a true cyclic dactyl, = ose’ but = Js: In the antistr., the 
corresponding νοσφιξ is — > for —v. | 

Verses 1 and 2 have each 1 sentence of 4 feet: v. 3 has 1 of 6 feet, 
an ἐπῳδικόν, during which the dance ceased. Series: .4.4.6.= ἐπ. 
Form of period :— 


ΤΊ. 


Ιχχνὶ METRICAL ANALY S/S: 


4 
A stichic period (see Parod. Str. 11. Per. 1.), with postlude. 


4 
6 -Ξ- ἐπ. 
SECOND STROPHE. 
ee Nd: Vw = παν ee τ. Vw Ww = πο ΜΝ λον 


. I. dewa μεν ουν | deva tapace || εἰ σοφος οι | ὠνοθετας || 


αλλ o μὲν ow | evs or ἀπολὰλ || wy Evveror | και ra βροτων 
eae Na ANS, ne hee ee Le ee Ferree oh i rd oy SONS oe 
2. οὔτε doxouvt | ovt ἀποφασκ || ovr ore AcE | ὦ ὃ aropw J] 
εἰδοτες avdp | wv δ ore parr || ts πλεον ἢ | yw φερεται 
vy —{~- vu saa ae OE) ~—~|~ vv Ld | 
I. wetopm : avd ελπισιν | ovt evGadop || wy ovt οπισ | w A | 
κρισις : οὐκ ἐστιν ad| 7Ons code || ᾳ ὃ αν σοφι | αν 
ν -- vw U 
2. τι γαρ ;: ἡ λαβδακιδ | ats A || 
παρα : μειψειεν av | np 
Saree en πο ον πὸ πον, ὑπ A ET MN ed Porc ἀπ δα 
3. ἡ τω πολυβ᾽ ov νεικος Ex εἰτ OvTE Tap ]  οιθεν ποτεγ ὠγουτετα νυνπω A || 
αλλ οὑποτ ey | wyav πριν 16 | on ορθον em || os μεμῴομεν | wy αν κατα [ φαιὴν 
vu U 
4. ἐμαθ : ov προς ot | ov dn Bacar || wv Bacay [ὦ A || 


Sued ay od, χοῦ ph eg eS Lee Sanh a NS Ld 


gavep : a yap er | avrw mrepo || eco ηλθε Kop [α΄ 
VV LJ VY ay 
5. ἐπι : ταν ere | dapov A || 
ποτε. : Kat cogos | whbn 
“Ἕπ-τ-οἂου -, - vv Lue -Ί ὺ π ge oi UI 
6. ἜΧΟΝ. : εἰμ, ovdi7r0d α AaBdaxid | ats ἐπι || κουρος a | δηλων θανατ[ wv A J] 


Bacay : ῳφθ αδυπολ | ις τω απ eulasdpevos|| ouror opX| noe και | αν 


I. First Period: 2 verses. Metre, choriambic (-uv-). This 
measure suits passionate despair or indignation: here it expresses the 
feeling with which the Chorus hear the charge against their king. 
Choriambics do not admit of anacrusis. 

Each verse has 2 sentences of 2 feet each. Series:.22.22. Form 
of period :— 


METRICAL ANALYSTS. Ixxvil 


A palinodic period. 


II. Second Period: 6 verses. Metre, zonzc (-—wv), an animated, 
but less excited, measure than the preceding choriambic. Note that 
one verse (8) has 20 anacrusts. Such an ionic verse is most nearly akin 
to a choriambic, in which anacrusis 15 never allowed. Here we see the 
consummate skill of Sophocles in harmonising the character of the two 
periods. Verse 1. o=——(§ 4): A =a pause equal to uv (δ 10): the 
whole is thus -—vuv. 

Verse 1 has 2 sentences of 2 feet each: v. 2, 1 of 2 feet: v. 3, 2 
of 3 feet: v. 4, same as 1; v. 5, same as 2; v. 6, same as 3, Series: 
.22.2.33.22.2.33. Form of period :— 


A palinodic period. 


τ 
Ἐν 


φΦωοσ)ὼ) - 


ΠΕ 


ITI. 


τὺ» 


Ixxvill METRICAL ANALYSIS. 


III. First Kommos, vv. 649—6971. 


πιθ : ov OeX| no | as dpov | ys || as ταν | αξ  λισσομ [αἱ A 7 
yu > arte | μελλ | εἰς kom | ef [εἰν dou | wy | rovd eg [ὦ 
[Here follows an iambic dimeter.] 
tov : ovre|mpiv| νηπι | ov ||vuy τ εν opk || w wey | av an αιδεσ [αι A ἢ 
δοκ : ησις | αγν | wsroy | wy'|| ηλθε | δαπτὶ εἰ δε | και το | μὴ νδικ] ον 
[Here follows an iambic trimeter. ] 
Vv ND ND ea ears ENS ὁ τὸ aN, 
I. τον : evayy "ΠΝ | ov μη || ποτ ev αι τι [ᾷ a All 
an : ws ἐμοιγ αλ | ts yas || προπονουμεν | as 
ANI made A, el, See —, 
2. συν : adaver doy | wea || ιμον Bar |ev A J 
paw : εταιενθε | Anger || avrov μεν | ew 


[Here follow two iambic trimeters. ] 


en gt eee ar ne το: Ἢ 
I. οὐ : tov [παντί Ὁ ἐν lion panies fay Kl 

ων : af | em | ον μεν | συχα | mak μον | ov 

ar A Ce ee VvVYV wwe Vv SISA NS ν᾽ 


2. αλι | ov επει | αθεος | αφιλος | οτι πυμ] a tov A || 
wht | Se rapa| φρονιμον | amopov | επι dpov | tua 
Vv a a ad gece oot ἔπαρε cad NS ἀν ς ες 

3. oA : οιμαν Ὧν» | now ει | tavd exo || 


me : gavOa wav [εἰσ evocd | tfouar 


1 The received constitution of this xoupss—which, for convenience of reference to 
other editions, I have indicated in my text of the play—is as follows: (1) 1s¢ strophe, 
649—659, (2) 20d strophe, 660—668; (3) τοί antistr., 678—688, (4) 2nd antistr., 
689—697. The division exhibited above is, however, in stricter accord with scientific 
method. Here, Periods I. II. III. correspond to the 1st strophe and ist antistrophe 
of the traditional arrangement: Period IV. corresponds to the 2nd strophe and 2nd 
antistrophe. Thus the whole κομμός, so far as it is lyric, might be conceived as forming 
a single strophe and antistrophe. These terms, however, are not applicable to the 
κομμοί, nor to the μονῳδίαι (lyrics sung by individual actors, μέλη ἀπὸ σκηνῆς), in the 
same accurate sense as to the odes sung by the Chorus, since here there was no 
regular dance accompanying the song. Consequently there was no need for the same 
rigour in the division of the composition. The principles which governed the 
structure of the κομμοί and povwdia have been fully explained by Dr Schmidt in vol. 
111. of his Kumstformen, ‘ Die Monodien und Wechselgesdnge der Attischen Tragidie.’ 


METRICAL ANALYSTS. Ixxix 


—-_ - vua_- -- vr - 


4. αλλ 3 a μοι δυσ | popw ya | φθινουσα || 


oor : εμαν γαν | φιλαν ev | πονοισιν 
Σ Sa =< No 7 — Vv -- 


5. τρυχ : εἰ] ψυχ] αν tad | ει Kak | ots κακ | a || 
ay iv | ovo | av κατ] ορθον | ovpio | as 
aS Cl Oe Lak ἀν ΤΩΣ = 
6. προσ : ay | a | τοις tad | αι τα | προς | cfov A J 


τα : νυν [οὐ] πομπὸς |avyer| οἱ | o 


I. Sirst Period: 1 verse, choreic. Two sentences of 4 feet each, 
forming :— 


4 aoe: 
A stich d. 
ὴ ) sticnic perio 


II. Second Period: τ verse, choreic. The rhythmical sentence of 2 
feet νυν τ ev opx || has nothing corresponding with it, but stands between 
2 sentences of 4 feet each: Ze. itis a μεσῳδός or tnterlude. The form 
of the period is thus :— 


2 A stichic period. 


11. Third Period: 2 verses. Rhythm, dochmiac. When an inter- 
change of measures occurs in Greek verse, it is nearly always between 
measures of equal length: as when the ionic, -—wvy, in 2 time, 15 
interchanged with the dichoree, -U—v, in § time. The peculiarity of 
the dochmius (ποῦς δόχμιος, ‘oblique’ foot) is that it is an interchange 
of measures zof equal to each other,—viz. the bacchius ὦ -- -- or -- τ ἧ΄Ί 
(with anacrusis), and shortened choree,— A. The fundamental form is 
vi——-v]|- ||. The varieties are due to resolution of long syllables, 
or to the use of ‘irrational’ instead of short syllables. Seidler reckoned 
32 forms ; but, as Schmidt has shown, only 1g actually occur, and some 
of these very rarely. With resolution, the commonest form is that seen 
here,v : vwu—wv | —A ||. Each verse contains two dochmiac sentences: 
2.6. we have | 


Bock METRICAL ANALYSIS. 


ἝΝ 


Doch. 
A palinodic period. 
age 


Doch.. 


IV. Fourth Period: 6 verses. In 1, 2, 5, 6, the metre is choreic 
(—v). In 8, 4, the metrical basis is the paeon, here in its primary form, 
the ‘amphimacer’ or ‘cretic,’ -- --, combined with another measure 
of the same time-value (3), the bacchius () —-- or -- -- υ)", 

Verse.1has' 1-sentence of 6 feet; Ὑ2.2, the sames v. 2, 10 3 feet; 
Vo qethe same sve Ὁ, 0 thersame-<das.1, 2. series: 05.023 .35.0.. 6.) he, 


6 
6 Here we have no repetition of whole groups, 
pi but only of single sentences. The period is not 
ἘΠ therefore palinodic. And the single senténces 
Pee. 3 correspond in an inverted order. This is called 
simply an antithetic period. 
6 
(i 


1 In v. 4, if Dindorf’s conjecture φθινὰς for φθίνουσα is received, we should write : 


Ξε I τ πο ad 9,5: 
αλλα μοι | δυσμορῳ | ya φθινας || 
oor ἐμὰν | yay φιλαν | εν πονοιξ. 

The ear will show anyone that this is rhythmically better than what I obtain 
with the MS. φθίνουσα and πόνοισιν, and the conjecture φθινὰς is entitled to all the 
additional weight which this consideration affords. On other grounds—those of 
language and of diplomatic evidence—no less distinct a preference seems due to 
φθίνουσα. 





METRICAL ANAL YSIS, Ixxxi 


IV. Second Stasimon, vv. 863—g10. 


First STROPHE 


> an ANS Lu - a, OSs = es -“-Ξ-.ὦ - --» 
Tr εἰ : μοι ξυν | εἰ | 1 dep | οντι || μοιρα | ταν ev || σεπτον | ayve | 
υβρ : ws gut | ev | εἰ τὺρ | avvor|| vps | εἰ moAA|| wy um | ερπλησθ | 


ete ee 
av Noy | wv A ἢ 
n mar | av 
ιν er ie 1, = Vv ι- = 
II. 1. epy : ων te | παντων | wv vo | οἱ προ | κειντ | ae A || 
a ipynm'| Kapa | pode | cumgep | ovr | a 
VVYV il Nd = 
2. vl : urodes | ovpave| av A || 
axp : otata | yer ava! Bac 
3. δὲ : αἰθερα | τεκνωθ | evres | ὧν o | Avr | os Λ΄] 
a : motuorar| avwp | ovcev | ecsav| ayKx | av 
SN, = = > “VV VU -ὶν - 
III. 1. πα : τήρμονος | ovde | νιν Ova | τα φυσις | avep | wy A || 
v0 : οὐ ποδι | χρῆσι | pw χρὴ | Ta το καλ | ws dex | wy 
ν -͵υ - υ wy = we εὑ- ι- - 
2. € : τικτεν | οὐδε | μη ποτε | λαθ || a κατα | κοιμ [ασ [{ῃ A || 
πολ : εἰ mar | αἰσμα [ μὴ ποτε | Ave || αἱ θεον | ar | ov | μαι 
ω “ἀπ του -, ὖ ι- ε΄ - 
4. μεγας : εν Tout | os Geos | οὐδε | γηρ[ ασκ [εἰ ΔΛ] 


θεον : ov ληξ | w ποτε | mpocrar | av | wx [ὡν 


I. First Period: 1 verse. Rhythm, logacedic. 


Two sentences, of 4 feet each, are separated by a mesode or inter- 
lude, consisting of the sentence of 2 feet μοιρα | ταν ev: 2.6, 


4 
2 A stichic mesodic period. 
4 





Ixxxil METRICAL ANALYSIS. 


II. Second Period: 3 verses. Rhythm the same’. 


Verse 1 has 1 sentence of 6 feet: v. 2 is a mesode of 3 feet: v. 3, 
the same as I: 2.6. 


3 A stichic mesodic period. 


III. TZhird Period: 3 verses. Rhythm the same. For the mark 
w over peyas and θεὸν in 3, see § 9, and Parod. Str. m1. Per. 1. v. 3. 


Verses 1, 3 have each 1 sentence of 6 feet: v. 2, 2 of 4 each: 2.2: 


6 


4 
A +) An antithetic period. (See First Kommos, Per. tv.) 


1 The conjectural reading οὐρανίᾳ | αἰθέρι, adopted by Prof. White and by Dr 
Schmidt, would give in v. 3 
ad : epe τεκν | wO | eves | wy o | AvuT | os A || 
In the antistrophe, Prof. White reads simply ἀκρότατον εἰσαναβᾶσ | ἀπότομον - 
ὥρουσεν els ἀνάγκαν, which similarly would give 


am : οτομον | wp | ovcey | as av | ayx | αν A || 


Now, there is no apparent reason for doubting the genuineness of the reading on 
which the Mss. agree, οὐρανίαν | δι᾽ αἰθέρα: while in the antistr. the most probable 
reading seems to be ἀκρότατα γεῖσ᾽ ἀναβᾶσ᾽ | ἀποτμοτάταν x.7.A. (See crit. ni and 
comment. on 876 f.) 


2. 


4. 


5. 


6. 


Se Se a Se τ -v tle - 
3. ἡ itwva| θικτων | Oger | at par | af | wy A J 
ge ταν τεΪ cava | Oavarov| aev | apx| αν 
- Ww v ι- ee Ce CI en eS = 
III. 1. τις : ere ποτ | ev | τοισὃ av | np Ge | wv BeA| y A || 
gow : ovta |yap| λαὲ |ovmadr| agar la 
πα ΡΥ τὴν ΠΥ ee 
evget | αἱ ψυχ | as ap | ννειν || 
θεσῴφατ | ekaip | ovow | dn 
= = = Sat cen gioaee ee 
3. εἰ bis | ae τοι | ade | πραξεις | τιμι ey at A || 
κουδαμ | ov τιμ | asa | πολλων | εμφαν | ns 
ἘΠ σ. Vv baat 
4. τι : dev με χορ | every ἢ 


METRICAL ANALY STS. 


SECOND STROPHE, 


NA VW VY WV Sad Vv μος το Vv 


ele | τις ὑπερ | οπτα | χερσιν || 


oveer | « Tov a | θικτον | etme 


ae ae nee ee 
ἢ Aoy | ὦ πορ doa evet | ae A || 
yas er | oudad | ov ceB | wy 
eres hoe Εν Ἢ 

δικ : as ages | yros | ov | de A || 
ovd : es τον αβ | ae | va | ov 


-π- vu - - -- 


δαιμον | wv τὸ | 7 ΝΣ | wy A || 


ovde | τὰν o | λυμπι | αν 
a ws, Vv pe L = 


kak : avwed| oto | pop | a A || 
ec : pn ταδε | χειρο | der | a 
“ὦ “τ = 


Svororp | ov xap | w xd | as A || 


macw | apuoo | εἰ Bpor | os 

> - vu -> = -v lL - 

εἰ : μη TO κερδος | copay | εἰ δικ | ac | ws A || 
ahd : wkpat| ver | emep | op? ax | ov | es 

= - Vv - > =u = 


. καὶ : τῶν a | σέπτων | ερξετ | αἱ A || 


fev i mavray| accwy | μηλαθ | οι 


epp : εἰ de τα | θεια 


]xxxiil 


ΙΧΧΧὶν METRICAL ANALYSIS. 


I. Lirst Period: 6 verses. Rhythm, logacedic. 
Each verse contains 1 sentence of 4 feet: and the six verses fall into 
3 groups: 2.¢. 


; = A repeated palinodic period. 
\ 


II. Second Period: 3 verses. Rhythm, the same. In v. 8 σὺ over 
θιξ means that in the antistrophe hover represents, by resolution, a long 
syllable, see § 5. 

Verses 1 and 3 have each one sentence of 6 feet: v. 2 is a mesode 
of 4 feet: 2.2: 


A stichic mesodic period. 


SPOON ei etsier ONES 


III. Third Period: 4 verses. Rhythm, the same. In v. 4, the 
last syllable of xopevew is marked short, because, being the last of a 
verse, it can be either long or short; and here it is the second of a 
choree, —v. 

Verses 1 and 3 have each 1 sentence of 6 feet: v. 2 is a mesode of 
4 feet: v. 4 is an epode of 2 feet. Thus, in this period, the dancers 
stood still during the alternate verses, 2 and 4. The form is :— 


6 


4 
ὃ Ὶ A stichic mesodic period, with postlude. 


METRICAL ANALYSIS. Ixxxv 


V. Third Stasimon (properly a Hyporcheme’), vv. 1086—1109. 


11 


Ὡν 


ὙΠ τ 


Ὁ 


~~ VY — = - 9 ἘΣ Lat Nal Ἐω -» == 
evrep ey | w | partes | εἰμι || και κατ an a γνωμ | av dp | us A || 
τις ce TeKy [ον | Tis σε | TiKTE || των μακρ | ac wy | wy ap la 

-ω Vv —~ Vv i de ι- ἘΞ 


ov τον o λυμπον α  πειρων | w κιθ | ap | w A || 


mavosop | ecotBar | ama | rpos red | ασθ | ao 
- Vv ae = L_ we τ be ea! ata ns τ 
. οὐκ ex | εἰ ταν | avpe | ov || πανσελ | yvov | μηου ce | ye A J 
*n cey | ewar | epa | ms || dog | ov Tw | yap πλακ | ες 
-~v τ om ar) = 


και πατρι | ὦ ταν | οιδιπ | ovv A || 


aypovou | o mac | a pir | ac 


- --ὄ > -“ ὦ — vu 


. και ao | ov και | ματερ | αὐξειν || 


e6 o | xkvdd\av | asav | acowy 


ey er = Vv = =, συ συ Φ == 


. καὶ χορ | everO | αἱ προς | ἡμων || ws exe | ἤρα dep in οντα || τοις a | 


e8 o  βακχει | os Oe | os ναι || ὧν eraxp | wy ope | wvevp|| nua | 
Sere oe 
ous τυρ | avy | os A || 
detar | ex | Tov 
> τυ -- ὦ Ll = 
ι ξ: nee | φοιβε] σοι | de A || 
vuud : αν ελικ | wd | wy | as 


Gee, pele sien τ 
tavtap |eor| ε [η A J 


πλειστα | συμ | mag | εἰ 


1 ὑπόρχημα, ‘a dance-song,’ merely denotes a melody of livelier movement than 


the ordinary στάσιμα of the tragic Chorus, and is here expressive of delight. Thus 
Athenaeus says (630 E) ἡ δ᾽ ὑπορχηματικὴ (ὄρχησις) TH κωμικῇ οἰκειοῦται, ἥτις καλεῖται 
κόρδαξ᾽ παιγνιώδεις δ᾽ εἰσὶν ἀμφότεραι : ‘the hyporchematic dance is akin to the comic 


dance called ‘‘cordax, 


” and both are sportive.’ Fragments of ὑπορχήματα, which 


were used from an early age in the worship of Apollo, have been left by several 
lyric poets,—among whom are Pratinas (who is said to have first adapted them to 
the Dionysiac cult),—Bacchylides, and Pindar. | 


Ixxxvl METRICAL ANALYSIS. 


I. First Period: 3 verses. Rhythm, J/ogaoedic. If in the first 
sentence οὖν. 3 we adopt for the antistrophe Arndt’s conjecture, ἢ σέ γ᾽ 
εὐνατειρά τις (which is somewhat far from the mss.), then verses 1 and 3 
have each 2 sentences of 4 feet, and verse 2 has 1 of 6 feet; ze. 


6 A palinodic period, with mesode. 


If, on the other hand, we should hold that ἢ σέ γέ τις θυγάτηρ represents 
the true metre (being corrupted from ἢ σέ γ᾽ ἔφυσε πατὴρ) and that οὐκ 
ἔσῃ τὰν αὔριον should be amended to τὰν ἐπιοῦσαν ἔσῃ, the rhythmical 
correspondence of sentences would be different. The rhythmical divi- 
sion of verses 2 and 3 would then be :— 


πους NET 6 δ Ὡς gE ae hs Aire Aas ape Se 

2. ov Tovo | λυμπον a | rep | wv || wo κιθ | ap | wy [ταν A | 
mavosop | ecoBar | a | ma || τρος weAX| acd | εἰσ | ἢ . 
ee ec a Bee eae 

3. emu : ουσαν eo | εἰ [πανσελ | ἡνον | wn ov ce | ye A 
σεγΎε : φυσε πα [τὴρ] λοξι |astw| γαρπλακ | ες 


and ν. 3 would be an epode, the form being :— 


4 

4 

4 A palinodic period, with postlude. 
4 

6 


II. Second Pertod: 5 verses. Rhythm, the same. Verses 1, 2, 4, 5 
have each one sentence of 4 feet: v. 3 has 3 sentences, the first and 
third of 4 feet each, the second of 3 (the words ws ἐπὶ ἦρα φέροντα). 
Senies 124). ade ἢ Ae Ate 


σσσν 


μι 


εξ 


METRICAL ANALYSTS. IXXxvil 


᾿ Here, single sentences correspond in an Ζ2:- 
4 verted order, while the middle sentence of v. 3 
has nothing corresponding to it, but forms a 
4 ἫΝ mesode or interlude. This is therefore a mesodic 
ΝΣ period. We need not add ‘antithetic,’ because, 
where more than two séngle sentences (and not 
4 groups) are arranged about a mesode, their 
arrangement is zorma/ly inverted. 
4 
VI. Fourth Stasimon, vv. 1186—1222. 
FIRST STROPHE 
(forming a single period). 
= πο ἐν χὰ = . = 
u | w-yeve | at Bpor | wv A || 
og | τις kaO ur | ep Bor | av 
a - τ ν᾿ = ἘΞ ι- = 2.-. ὑπο CY ROY ι- - 
. os vy | ας ισα | και το | μη || dev Coo | as evap | ιθμ} ὦ A || 
τοξευσ | as expar | noe | Tov || παντ ev | δαίμονος | ολβ | ov 
--- - a a -: υ - 
. τις | yap τις av | np wAe | ov A || 
w | gev κατα | μεν φθισ | as 
πο ὩΣ ἘΔ ΥΣ oy ἔστ 
τας ev | δαιμονι | as pep | εἰ A || 
ταν γαμψ | wruxa | παρθεν | ov 
Se Stee Re eG " 
. ἢ too | ovrov oo | ov dox | ew A |i 
xpnopwd | ov θανατ | wien | a 
ea) ANS AP cae ems 
και dof | avt απο | κλιν | a A || 
xwpa | wupyosav| εστ | a 
=> lL wy ue = ee 
. tov ? cov | τοι παρα | δειγμ ex | wv A | 


ov | και βασιλ | evs Kad | εἰ 


IXxxvill METRICAL AANALYSIS. 


of 
9. 


Ὁ -- τ τ Sh το ι- -ς τ τ ae τ = 
tov : σον | δαιμονα | tov cov | w|| τλαμον | οιδιποδ | a Bpor | wy A 
eu : os || Καύτα wey | ἴστε |rimi| ans | ταῖς μεγάλ aow | ev 


ee one ele 


. ov | dev paxop | | w A || 


θη | βαισιν αν | aco | wy 


Rhythm, /ogacedic. Verse 1 contains 1 sentence of 4 feet: v. 2, 2 
4 feet each: v. 3, 1 of 4 feet; to which answer respectively vv. 7, 8, 
Verses 4, 5, 6 also contain each 1 sentence of 4 feet, v. 4 answering 


to v. 6, and v. 5 forming a mesode. The series .4.44.4-,4-4-4-, 


re 


44.4.thus forms the period :— 





Since the whole group, consisting of 
vv. 1, 2, 8, recurs once, the period is 
palinodic ; since the sentences formed 
by vv. 4 and 6 are grouped about the 
interlude formed by v. 5, it is also 
MeSOALC. 





SECOND STROPHE. 


Vv =- v be =u πο -υ ΕΣ 

τα : νυν Sax ου Ϊ εἰν τις | αθλι | wrep [ος A |] 
ep : εὑρεσ | α | κονθο | πανθ op | wy χρον | os 

Vv Le be ~vU ey sao ents = 

τις : at | aus | aype | as τις | εν πον | os A 


δικ : af | εἰ [τὸν ayau | ον yau | ov mad | αἱ 


METRICAL ANALYSIS. ἸΧΥΧΙΣ 


--  .ὼ _— -- wv — 


Vv VY 
3. €vv : oxos | adAay | a Bi | ov A J 
τεκν : ovvra | και τεκν | οὐμεν | ov 
A 


17 το τη ὦ | κλεινὸν | ovdur | ov Kap | ae || 


NA, NS, — tw — 


e|[@ | Aat | evo» | wrexv | ov 
at the ch aes 
2. w μεν | asrin | nv A || 
εἰθεσ | εἰθε [σε 
ye ον χες eo ee 
3. αὐτὸς | ἡρκεσ | ev A | 
μήποτ | edou | av 
πος “ὦ τ COPS) ~— ww ao yy “- 
4. παιδι και πα | τρι θαλαμ | ἡπολ | w πεσ [εἰν A ἢ 
δυρο | wacyap | womepe | aren | ov xe | wy 
<7 VV Ξ - aad VY = ἐξν “τω 4 VY aa Vv as 
III. 1. πως ποτε | πως ποθ | ac ratp | w || ato αλοκ | ες pep | ev Tad | as A|| 


ex orouat | wvt0d | ορθον | em || evaverv | evoat | εκ σεθ | εν 


Sone uae oe τοῦ ee 
σιγ εδυν | α | θησαν | ες Too | ov | de ἄς ἢ 


-- vu -- ὦ 


2 


καὶ κατε | κοιβμ | σα | τουμον | ομμ | a 


I. first Period: 3 verses. Rhythm, choreic. Verses 1 and 2 have 
each 1 sentence of 6 feet: v. 8 forms an epode or postlude of 4 
feels 2:6, 

6 
6 A stichic period, with postlude. 


4 = ἐπ. 


II. Second Period: 4 verses. Rhythm, the same. In v. 4 Τῆι 
θαλᾶμ is an apparent tribrach, representing a cyclic dactyl, ~v, and 
having the time-value of ees ἘΝ (5εε ὃ 7). This denoted by writing = vv, 


because the ‘irrational’ character, though in strictness shared by the 
first and second short syllables, is more evident in the first. 

Verses 1, 4 have each 1 sentence of 6 feet, vv. 2, 3 each 1 of 
cee 


ΧΟ METRICAL ANALYSIS. 


: 2 An antithetic period: see First Kommos, Per. Iv. 
3 


III. Third Period: 2 verses. Rhythm, the same. Verse 1 has 
2 sentences, each of 4 feet: v. 2 has 1 of 6 feet, and forms an epode or 
postlude: ze. 


᾿ 
4 A stichic period, with postlude: see Parod. 
DU Ly Pern dy Olas. ΘΠ dy bel, ἫΝ 


VII. Second Kommos’, vv. 1297—1368. 


(After the anapaests of the Chorus, 1297—1306, and of Oedipus, 
1307—1311, followed by one iambic trimeter of the Chorus, 1312, the 
strophic system of lyrics begins at 1313.) 


FirsT STROPHE 
(forming a single period). 


v Us - 
I. t : woot | ov A || 
t : wor | os 


« A i A ae RO vVvY VV Vv vvVV WV vw 
2. ved : os ἐμὸν απο | τροπον em || erAopevov a | φατον A || 


συ : μεν εμος επι | πόλος ετ || εἰ μονιμος ετ | εἰ yap 


1 At v. 1336, and in the corresponding 1356, an iambic dimeter is given to the 
Chorus (Period I1I., v. 3). With this exception, the Chorus speaks only iambic: 
trimeters, which follow a lyric strophe or antistrophe assigned to Oedipus. Since, 
then, the lyrics belong all but exclusively to Oedipus, the passage might be regarded 
as his μονῳδία, interrupted by occasional utterances, in the tone of dialogue, by the 
Chorus. If, however, regard is had to the character and matter of the whole com- 
position, it will be felt that it may be properly designated as a κομμός, the essence of 
which was the alternate lament. On a similar ground, I should certainly consider it 
as beginning at 1297, though the properly lyric form is assumed only at 1313. 


ee ee 


METRICAL ANALYSIS. 


Vv NA se WS Se Vv σας - ον τι Ὁ ἝΞ 
a: : δαματον τε | Kau ἘΣ || ουριστον | ov A J 
vr : opmevers μὲ | Tov τυφλὰ || ov xy dev | wy 


[Here follow four iambic trimeters.] 


ΧΕΙ 


Rhythm, dochmiac: see First Kommos, Period 111. It will be 
seen that every dochmiac metre here is a variation of the ground- 
form ὦ : - πὸ | — A ||, by substitution either of ὦ ὦ for —, or of > (an 
irrational syllable, apparently long) for ὦ, as in v. 8, κηδεύων. Verse 1 
is a dochmiac used as a prelude (προῳδικόν), ὦ being prolonged to the 


time-value of -—. Vv. 2, 3 have each 2 dochmiac sentences: Ze. 


Dach Ξε τρ. 
ae 
Doch. A palinodic period, with prelude. 
ae 


Doch. 


SECOND STROPHE. 


ww ae ὌΝ Zs 
Το ia ΞΡ δὲν fab | nv a || Nap nx ἴοι οι A || 
oN : 08 oars | νος || αγριας red | as 
Vv Vv WI AS ONS — Ww Vv V VV VV VV 


2. 0 : κακα kaka TeX | wv ev || a Tad eua wad | ca A J 
vou : ad επιποδι | aoe || Avo aro re | govov 
Re eee eye ge is ee Ae - 

1 € : παισε ὃ | αὐτο | χειρ νιν a ουτις || αλλ ey | w [|τλαμ | wv A ἢ 

epp : vuto | κανεσ | woe mw | οὐδεν! es xap | ιν | πρασσ | wy 
So og ae pee ΡΞ 


III. 1. τι : yap ede p op oa av A || 


ToT : εγαραν Bay | wr 
= 
τ ἢ re a, hoe 2 
2: ιν προ ork 
οὐκ : ny gir | οισιν | ουδ εμ [οἱτοσ | ονδαχ | os 


PY. 


xcil METRICAL ANALYSIS. 


Ξε in) Ὁ πὶ. -» πο eee 


3. nv : ταυθ or | worep | και ov | dys A || 


OeX :  ovte | Kapoe | τοῦ av | nv 
Soe πνο, we ee - 


4. te: Oyrew | oe | oer "τὴ Ι στερκτον aa n ΓΞ | » yop | ov A || 
οὐκ : ow ma | rposy | av gov |evs || ηλθον | οὐδὲ | vuude | os 
- - v Le Ee -ὸ -τΟοι αὶ = 

5. er i ect ax | ov | ew |adov|aqgir| a A J 


Bpor : owe [κληθ] mv | wre | Puvamr|o 
ἜΣ ῳ᾿ Ὁ ——  ὰν τε τὺ - 


I. απ : ayet ex tom | tov or [[ εἰ ταχιστα | με A || 


νυνδ : abeospev | εἰμαν || οσιων de | mars 


Vv ee gh ρ , ς > “πω - 


2. ar : ayer ὦ διὰ ἢ οἱ Tov || wey ολεθρι | ov A || 


om : oyevns dap | wy aur || oo epuy Tar | as 
es SI NSD ee SO Sa Ν δ, WwW WV ae Vv a, 


3. τὸν : Katapato | τατον er || ει δε και Ge | ors A || 
εἰ : δετιπρεσβυ | τερον er || & κακου κακ | ov 
> Dah = 

4. εχθρ : οτατον nce | wy A J 


Tour : ελαχ οιδιπ | ous 


[Here follow two iambic trimeters.] 


I. First Period: 2 verses. Rhythm, dochmiac. In verse 1 (anti- 
strophe), we have aypias: observe that if we read am’ ἀγρίας the 
dochmiac would have one ὦ too much, and see my note onv. 1350. In 
v. 2, the Ms. reading νομάδος is zmpossible, as the metre shows. ovov, 
by resolution for —, as in the strophe, since the last syllable of a verse 
can be either long or short: see on Parod. Str. 11. Per. 1. v. 1, and cp. 
Xopevety, Stas. 11. Str. 11. Per. 11. v. 4. Metre would admit ἔλαβέ pw’ or 
ἔλαβεν, but not, of course, ἔλυσέ μ᾽ or ἔλυσεν. 

Each verse has two dochmiac sentences, 2.¢. 


ἀν 
Doch. 

ς A palinodic period. 
ee P P 
Doch. 


METRICAL: ANALYSIS, xclll 


Il. Second Period: 1 verse. Rhythm, chorvetc. Two sentences, 
each of 4 feet: 26. 


Ν᾽ A stichic period. 
4 


Ill. Third Period: 5 verses. Rhythm, choretc, except in verse 1, 
which is a dochmiac, serving as prelude (προῳδικόν). 

Verse 2 has 1 sentence of 6 feet: v. 3, 1 of 4 feet: v. 4, 2 of 4 feet 
each: v. 5, 1 of 6 feet. The first of the 2 sentences in v. 4 forms a 
mesode; which can either (as here) begin a verse, or close it, or stand 
within it, or form a separate verse. Series: .6.4.4.4.6.: form :— 


Doch.=anp. 
6 


4 A mesodic period, with prelude. See Stas. 11. 


; Pere ΤΠ: 
oy, 


4 ee 
δ΄ 
IV. Fourth Period: 4 verses. Rhythm, dochmiac. Verses 1, 2, 3 
have each two dochmiac sentences: v. 4 has one, which forms an 
epode: Z.2. 
‘igs 
Doch. 


f Doch. 


A repeated palinodic period, with post- 
Doch. 


lude. 


( Doch. 
Deck: 


Doch. = ἐπ, 


XC1V METRICAL ANALYSIS, 


RELATIONS OF LYRIC FORM AND MATTER. 


In the lyric parts of Tragedy, the poet was a composer, setting 
words to music. Words, music, and dance were together the expression 
of the successive feelings which the course of the drama excited in the 
Chorus, or typical spectator. It is obvious, then, that the choice of 
lyric rhythms necessarily had an ethical meaning, relative to the mood 
which in each case sought utterance. It is everywhere characteristic of 
Sophocles that he has been finely sensitive to this relation. So much, 
at least, moderns can see, however far they may be from adequately 
appreciating the more exquisite secrets of his skill. Without attempt- 
ing minute detail, we may glance here at some of the chief traits in 
which this skill is exemplified by the lyrics of the Oedipus Tyrannus. 


I, Paropos. /irst Strophe. ‘The Theban Elders are reverentially 
awaiting the message from Delphi, and solemnly entreating the gods for 
deliverance from their woes. With this mood the dactylc rhythm is in 
unison. The Greek dactylic measure was slow and solemn, the fitting 
utterance of lofty and earnest warning—as when oracles spoke—or, as 
here, of exalted faith in Heaven. 

Second Strophe. Period 1. The chorees, in Zogacedic rhythm, express 
the lively sense of personal suffering (ἀνάριθμα yap φέρω | πήματα). 
Per. 11. Dactyls, somewhat less stately than those of the opening, 
again express trust in the gods who will banish the pest. 

Third Strophe. Choreic rhythms of the strongest and most excited 
kind embody the fervid prayer that the Destroyer may be quelled by 
the Powers of light and health. 


II. First Srasimon. The doom has gone forth against the unknown 
criminal ; and the prophet has said that this criminal is Oedipus. /zrst¢ 
Strophe. While the rhythm is Jogaoedic throughout, the fuller measures 
of Period 1. are suited to the terrible decree of Delphi; those of Per. 1. 
to the flight of the outlaw; those of 11. to the rapid pursuit, and, 
finally, to the crushing might, of the Avenger. 

Second Strophe. Period1. The choriambic rhythm—the most pas- 
sionate of all, adapted to vehement indignation or despair—interprets 
the intensity of emotion with which the Theban nobles have heard the 
charge against their glorious king. Period 1. Passing to their reasons 
for discrediting that charge, the Chorus pass at the same time from the 
choriambic rhythm to the kindred but less tumultuous zovz¢, which is 
here (as we have seen) most skilfully linked on to the former. 


METRICAL ANALYSIS. XCV 


III. The First Kommos, in its 3rd and 4th Periods, shows how 
dochmiac measures, and paeonic combined with choreic, can suit varying 
tones of piteous entreaty or anxious agitation; an effect which, as 
regards dochmiacs, the SecoNnD Kommos (VII) also exhibits in a still 
more impressive manner. 


IV. Inthe SECOND STASIMON, Jogacedics are the vehicle of personal 
reflection and devotion ; the lively measures of the Hyporcheme which 
holds the place of TH1RD 5ΤΑΒΙΜΟΝ (V) speak for themselves. 


VI. Inthe FourTH Stasimon we have a highly-wrought example of 
lyric art comparable with the First Stasimon, and with the Parodos. The 
utter ruin of Oedipus has just been disclosed. First Strophe. It was 
a general rule that, when a verse was opened with a syncope, anacrusis 
must precede. By the aisvegard of this rule here, an extraordinary 
weight and solemnity are imparted to the first accent of the lament: 


Cy ig. tae tee 


u| w yeve | ac Bpor | wv A ||. (See the musical rendering of this, Appen- 
dix, § 10, p. 205.) So, again, in the profoundly sorrowful conclusion 
ΠΩ ΔΕ ἃ τς ed τῶν 


drawn from the instance of Oedipus, ovd | ev μακαρ | iw A||. And, since 
his unhappy fate is here contemplated in its entirety, the whole strophe 
forms a single rhythmical period. 

The Second Strophe—reflecting on particular aspects of the king’s 
destiny—is appropriately broken up into three short periods; and the 
choreic rhythm is here so managed as to present a telling contrast with 
the logaoedic rhythm of the first strophe. The weightiest verses are 
those which form the conclusion. | 

I have but briefly indicated relations of which the reader’s own ear 
and feeling will give him a far more vivid apprehension. There are no 
metrical texts in which it is more essential than in those of ancient 
Greece never to consider the measures from a merely mechanical point 
of view, but always to remember waft the poet is saying. Noone who 
cultivates this simple habit can fail to attain a quicker perception of the 
delicate sympathies which everywhere exist between the matter and the 
form οἵ Greek lyrics. 


bh 





ΟΦ ΘΙ ΧΕ OY = 


OO One ΞΡ ΝΟΣ 


1.5.1.» 


ZOO ΣΙ ΡΕΟΥΣ 
ΟΠ oy cee ele NINN OC 


[. 


APISTOSANOYS TPAMMATIKOY ὙΠΟΘΕΣΙΣ. 


Λιπὼν Κόρινθον Οἰδίπους, πατρὸς νόθος 
‘ A ε ’ὔ 4 ἡ. 
πρὸς τῶν ἁπάντων λοιδορούμενος ἕένος, 
ἦλθεν πυθέσθαι ἸΤυθικῶν θεσπισμάτων 
ζητῶν ἑαυτὸν καὶ γένους φυτοσπόρον. 
ε \ Ν , 3 A ε A 
εὑρὼν δὲ τλήμων ἐν στεναῖς ἁμαξιτοῖς : 5 
” ” fee , 
ἄκων ἔπεφνε Aatov γεννήτορα. 
Σφιγγὸς δὲ δεινῆς θανάσιμον λύσας μέλος 
ἤσχυνε μητρὸς ἀγνοουμένης λέχος. 
λοιμὸς δὲ Θήβας εἷλε καὶ νόσος μακρα. 
Κρέων δὲ πεμφθεὶς Δελφικὴν πρὸς ἑστίαν, 1ο 
ὅπως πύθηται τοῦ κακοῦ παυστήριον, 
ἤκουσε φωνῆς μαντικῆς θεοῦ πάρα, 
τὸν Λαΐειον ἐκδικηθῆναι φόνον. 
ὅθεν μαθὼν ἑαυτὸν Οἰδίπους τάλας 
δισσάς τε χερσὶν ἐξανάλωσεν κύρας, 15 


αὐτὴ δὲ μήτηρ ἀγχόναις διώλετο. 


ἈΡΙΣΤΟΦΆΝΟΥΣ ΥΠΟΘΕΣΙΣ] ᾿Αριστοφάνους ἐπίγραμμα εἰς τὸν τύραννον 
οἰδίπουν A. The word ἐπίγραμμα, which could denote the ‘title’ of a book, is not a 
correct substitute for ὑπόθεσις. 3 θεσπισμάτων] νόμων θέλει A, which indicates 
that ἐλθὼν was a v.1. for ἦλθεν in this verse. 11 πύθηται MSS., vivid for πύθοιτο, 
which Brunck unnecessarily conjectured. 15 δισσαῖς MSS., δισσάς Elmsley. 
πόρπαισι δισσὰς Brunck. 16 αὐτὴ δὲ] αὐτή τε Elmsley. But the composer may 
have imitated the irregular sequence τε---δέ which sometimes occurs (as Z/. 1099, 
Ai, 836). 





1... 2. 


Io 


15 


A ΣΟΦΟΚΛΕΟΥΣ 


ΑΡΙΣΤΟΦΑΝΟΥ͂Σ TPAMMATIKOTY] The first of the three prose ὑποθέσεις to 
the 4x¢iyone is also ascribed in the Mss. to Aristophanes of Byzantium (flor. 200 B.C.). 
His name is likewise given in the Mss. to the metrical ὑποθέσεις prefixed to all the 
extant comedies of his namesake except the Z7hesmophoriazusae. All these ascrip- 
tions are now generally held to be false. There is no reason to think that the 
fashion of metrical arguments existed in the Alexandrian age: and the language 
in every case points more or less clearly to a lower date. The verses above 
form no exception to the rule, though they are much more correct than the comic 
ὑποθέσεις. See Nauck’s fragments of the Byzantine Aristophanes, p. 256: Dindorf 
agrees with him, Schol. Soph. vol. 11. p. xxii. 


ΤΙ: 
AIA TI TYPANNOS ETNTETPAITAL 


O TYPANNOS ΟἸΔΙΠΟΥΣ ἐπὶ διακρίσει θατέρου ἐπιγέγραπται." 
χαριέντως δὲ ΤΥΡΑΝΝΟΝ ἅπαντες αὐτὸν ἐπιγράφουσιν, ὡς ἐξέχοντα πάσης 
τῆς Σοφοκλέους ποιήσεως, καίπερ ἡττηθέντα ὑπὸ Φιλοκλέους, ws φησι 
Δικαίαρχος. εἰσὶ δὲ καὶ οἱ ΠΡΟΤΕΡΟΝ, οὐ ΤΎΡΑΝΝΟΝ, αὐτὸν ἐπιγράφ- 
οντες, διὰ τοὺς χρόνους τών διδασκαλιών καὶ διὰ τὰ πράγματα: ἀλήτην 
γὰρ καὶ πηρὸν Οἰδίποδα τὸν ἐπὶ Κολωνῷ εἰς τὰς ᾿Αθήνας ἀφικνεῖσθαι. 
ἴδιον δέ τι πεπόνθασιν οἱ μεθ᾽ Ὅμηρον ποιηταὶ τοὺς πρὸ τῶν Τρωϊκῶν 
βασιλεῖς ΤΎΡΑΝΝΟΥΣ προσαγορεύοντες, ὀψέ ποτε τοῦδε τοῦ ὀνόματος 
εἰς τοὺς Ἕλληνας διαδοθέντος, κατὰ τοὺς ᾿Αρχιλόχου χρόνους, καθάπερ 
Ἱππίας ὁ σοφιστής φησι. Ὅμηρος γοῦν τὸν πάντων παρανομώτατον 


> 
Ἔχετον βασιλέα φησὶ Kat ov τύραννον" 


Εἰς Ἔχετον βασιλῆα, βροτῶν δηλήμονα. 
προσαγορευθῆναι δέ φασι τὸν τύραννον ἀπὸ τῶν Τυρρηνῶν: χαλεποὺς γάρ, 
τινας περὶ λῃστείαν τούτους γενέσθαι. ὅτι δὲ νεώτερον τὸ τοῦ τυράννου 
ὄνομα δῆλον. οὔτε γὰρ Ὅμηρος ovte Ἡσίοδος οὔτε ἄλλος οὐδεὶς τῶν 
παλαιῶν τύραννον ἐν τοῖς ποιήμασιν ὀνομάζει. ὁ δὲ ᾿Αριστοτέλης ἐν Κυμ- 
αίων πολιτείᾳ τοὺς τυράννους φησὶ τὸ πρότερον αἰσυμνήτας προσαγορ- 
εύεσθαι. εὐφημότερον γὰρ ἐκεῖνο τοὔνομα. 

2 ἐπιγράφουσιν] So Dindorf with L: vulg. ἐπέγραφον. 4 IIPOTEPON, οὐ 
ΤΎΡΑΝΝΟΝ, αὐτὸν] L, Dind.: vulg. IPOTEPON αὐτὸν, οὐ TTYPANNON, 





2 τύραννον...ἐπιγράφουσιν] The distinguishing title was suggested by v. 514 of the 
play, τὸν τύραννον Οἰδίπουν, v. 925 τὰ τοῦ τυράννου.. Οἰδίπου. Sophocles doubtless 
called it simply Οἰδίπου. ο κατὰ τοὺς ᾿Αρχιλόχου χρόνους] circ. 670 B.c. Itis about 
679 8.6. that Orthagoras is said to have founded his dynasty at Sicyon, and ‘the despots. 
of Sikyon are the earliest of whom we have any distinct mention,’ Grote 111. 43. 


OIAITOY2 ΤΥΡΑΝΝΟΣ ς 


152 Ἔχετον] Od. 18. 85. 15 οὔτε γὰρ Ὅμηρος] For the writer of this ὑπόθεσις, then 
(unless he made an oversight), ‘Homer’ was not the author of the ‘Homeric hymn’ 
to Ares, 8. 5, ἀντιβίοισι τύραννε, δικαιοτάτων ἀγὲ φωτών. The earliest occurrences 
of the word τύραννος which can be approximately dated are (1) Alcaeus fr. 37 
Bergk, circ. 606 B.C., refering to Pittacus; see below on 17: (2) Pind. Py¢h. 3. 85, 
where it is convertible with βασιλεύς, 26. 70 (Hiero of Syracuse), date perh. 474 B.C. 
(see Fennell’s introd.): and (3) Aesch. P. V. 736 ὁ τῶν θεών τύραννος (Zeus), date 
circ. 472—469 B.C. On the question as to the origin of τύραννος, scholars will read 
with interest the opinion of the author of Greek and Latin Etymology. Mr Peile has 
kindly communicated to me the following note:—‘‘ There seems no reason to doubt 
the usual connection of τύραννος with ,/¢tur, a by-form of ,/TAR. It does not occur, 
I think, in Greek, but it is used in Vedic,—as is also the common epithet /z7-a, 
‘strong,’ applied chiefly to Indra, but also to other gods. Jarer cognates are /urvazz, 
= ‘victory,’ and ¢rvaz¢=‘ victorious,’ also of Indra. The primary meaning of the 
root was ‘to bore’—then ‘to get to the end’ of a thing—then ‘to get the better of? it. 
There is another family of words, like in form, with the general sense of ‘haste’; 
e.g. turvanya, a verb-stem in Vedic=‘to be eager,’ and ¢uranyu an adjective. 
These, I think, are distinct in origin. In form they come nearer to τύραννος. But I 
think that they are /ate Vedic forms, and therefore cannot be pressed into the service. 
The form in Greek is difficult to explain in either case. If there were an Indo-Eur. 
turvan (whence the Sanskrit word), the Greek might have formed a secondary 
turan-yo: but one would expect this to have taken the form τυραινο. Taking into 
account the entire absence of all cognates in Greek, I think that it is probably a 
borrowed word, and that from being an adjective (?=‘ mighty’), it became with the 
Greeks a title.” 16 ἐν Κυμαίων πολιτείᾳ] Cp. schol. in Eur. Zed. το (Dind. vol. 
Iv. p. 8) αἰσυμνᾷ' ἡγεῖται καὶ ἄρχει" ἰδίως δέ φησιν ’ApiororéAns ὑπὸ Κυμαίων αἰσυμνήτην 
τὸν ἄρχοντα λέγεσθαι. “αἰσυμνῆται δὲ κριτοὶ ἐννέα πάντες ἀνέσταν᾽ [Od. 8. 258] 
τοὺς ἄρχοντας τῶν ἀγώνων (sc. ὁ ποιητὴς λέγει). 17. The αἰσυμνητεία resembled 
the τυραννίς in being absolute, but differed from it in being e/ective; hence it is called 
by Arist. αἱρετὴ τυραννίς, Pol. 3. 14. Alluding to the choice of Pittacus as αἰσυμνήτης 
by the Mityleneans, Alcaeus said ἐστάσαντο τύραννον, 10.: but this was ad znvidiam. 


111: 


AAAQS. 


‘O Tvpavvos Οἰδίπους πρὸς ἀντιδιαστολὴν τοῦ ἐν τῷ Κολωνῷ ἐπι- 
γέγραπται. τὸ κεφάλαιον δὲ τοῦ δράματος γνῶσις τῶν ἰδίων κακῶν Οἰδίποδος, 
πήρωσίς τε τῶν ὀφθαλμών, καὶ δι᾿ ἀγχόνης θάνατος ᾿Ιοκάστης. 

‘Haec in fine fabulae habet L, om. A, qui de sequentibus nihil habet praeter 
aenigma Sphingis,’ Dind. Schol. 11. 13. 


XPHSMOS O ΔΟΘΕῚΣ AATOQ: TO. ΘΗΒΑΙΩ͂ι. 
Adie Λαβδακίδη, παίδων γένος ὄλβιον αἰτεῖς. 
δώσω τοι φίλον υἱόν: ἀτὰρ πεπρωμένον ἐστὶν 
παιδὸς ἑοῦ χείρεσσι λιπεῖν φάος. ὡς γὰρ ἔνευσε 


6 ZOPOKAEOY2 


Ζεὺς Κρονίδης, Πέλοπος στυγεραῖς ἀραῖσι πιθήσας, 


,’ 
οὗ φίλον ἥρπασας υἱόν: ὁ δ᾽ ηὔξατό σοι τάδε πάντα. 


ΧΡΗΣΜΟΣ... ΘΗΒΑΙΩΙ.] So L: vulg. χρησμὸς δοθεὶς Λαΐῳ. 2 δώσω... ἐστὶν] 
Another reading was τέξεις μὲν φίλον υἱόν" ἀτὰρ τόδε σοι μόρος ἔσται" cp. Valckenaer, 
Eur. Phoen. p. xvi. . 3 παιδὸς ἑοῦ] Valck. ἀξ. cites this reading from the cod. 
Augustanus, and it is probably right, ἑοῦ here meaning ‘thine,’ in which sense Zeno- 
dotus rightly wished to substitute it for éfjos in //7. 1. 393, 15. 138, 24. 422, 550. The 
pron. é6s (=ofés) properly meant merely ‘own,’ and (like the pron. stem szva, ‘self’) 
was applicable to the rst and 2nd persons, sing. or plur., no less than ‘to the 3rd. 
Vulg. σοῦ παιδὸς. 


ΠΟ AINIDMAy ΤΗΣ ΣΦΙΓΘΣ: 


2 , CPN A Ν 4 x , , 
Ἔστι δίπουν ἐπὶ γῆς Kat τετράπον, od pia φωνή, 

Ν / 3 / Ν Ν ’ὔ “ φῶ et Faas A 
καὶ τρίπον: ἀλλάσσει δὲ φυὴν μόνον ὅσσ᾽ ἐπὶ γαῖαν 

Ν A δ ΟΣ, > , Ν \ , 

ἑρπετὰ κινεῖται ava T αἰθέρα καὶ κατὰ πόντον. 
Ω 3 Ἑ , 4 » XN 
ἀλλ᾽ ὁπόταν πλείστοισιν ἐρειδόμενον ποσὶ βαίνῃ, 


” , , 3 , , 2 A 
ἔνθα TAXOS γυιοισιν αφαυρότατον πέλει αυτου. 


4 φυὴν] φύσιν Athen. 4568, βοὴν L, A. 3 κινεῖται] γίνηται 1,. 4 ἐρειδό- 
μενον a specious but unsound reading. The contrast is not between haste and slow- 
ness, but between the number of the feet, and the weakness of the support which they 
afford. 





Athenaeus 456 8B introduces his quotation of the riddle thus: Kal τὸ τῆς Σ φιγγὸς 
δὲ αἴνιγμα ᾿Ασκληπιάδης ἐν τοῖς Τραγῳδουμένοις τοιοῦτον εἷναι φησίν. Asclepiades 
of Tragilus in Thrace, a pupil of Isocrates, wrote (circ. 340 B.C.) a work called 
Τραγῳδούμενα (‘Subjects of Tragedy’) in six books, dealing with the legendary 
material used by the tragic poets, and their methods of treatment. The Αἴνιγμα, 
in this form, is thus carried back to at least the earlier part of the fourth century B.c. 


ΑΥΣΙΣ TOY AINITMATOS. 


Κλῦθι καὶ οὐκ ἐθέλουσα, κακόπτερε Μοῦσα θανόντων, 
φωνῆς ἡμετέρης σὸν τέλος ἀμπλακίης. 

” / a £4 a 3 ΄ 

ἄνθρωπον κατέλεξας, ὃς ἡνίκα γαῖαν ἐφέρτπει, 
πρῶτον ἔφυ τετράπους νήπιος ἐκ λαγόνων" 

/ ἂς / , / , > / 
5 γηραλέος δὲ πέλων τρίτατον πόδα βάκτρον ἐρείδει, 

αὐχένα φορτίζων, γήραϊ καμπτόμενος. 


5 ἐρείδει Gale: ἔχει or ἐπάγει MSS. 





The Avovs is not in the Mss. of Sophocles, but is given by the schol. on Eur. 
Phoen. 50 (αἴνιγμ᾽ ἐμὸς παῖς Οἰδίπους Σφιγγὸς μαθών)...τὴν δὲ λύσιν τοῦ αἰνίγματος 
οὕτω τινές φασιν" “Κλῦθι᾽ κιτιλ, Valckenaer, Schol. Phoen. p. 28, gives it as above 
from a collation of three Mss, 


ΟΙΔΙΠΟῪΣ ΤΎΡΑΝΝΟΣ 7 


TA TOT APAMATOS IIPOZOITIA, 


OIAITIOYS. IOKASTH. 
IEPEYS. ATTEAOS. 
KPEON. @EPATION Λαΐου. 
ΧΟΡΟΣ γερόντων Θηβαίων, ΞΑΤΓΤΈΛΟΣ. 
TEIPESIAS. 





The ἱκέται in the opening scene (I—150) are a body of κωφὰ πρόσωπα 
like the citizens whom Eteocles addresses in Aesch. 7%. 1—38, or the 
Areiopagites in Lum, 566 ff. They would probably come within the 
meaning of the term παραχορήγημα, which denoted anything furnished by 
the choregus zm supplement to the ordinary requirements of a drama. 
Some, however, deny this, holding that it was an ordinary duty of the 
choregus to provide all ‘mute persons,’ however numerous (A. Miiller, 
Gr. Bihnenalterth., p. 179). The distribution of the parts among the 
three actors would be as follows :— 


OEDIPUS, πρωταγωνιστής. 


ΙΟΘΑΒΤΑ, 

PRIEST OF ZEUS, SarresesarieHe 
MESSENGER from the house (ἐξάγγελος), Y es 
SERVANT OF Laius, 

CREON, 

TEIRESIAS, | τριταγωνιστής. 


MESSENGER from Corinth (ayyeAos), } 


8 2ZOPOKAEOY2 


STRUCTURE OF THE PLAY. 


I. πρόλογος, verses I—I50. 


ty 
e 


πάροδος, 15I—215. 





3. ἐπεισόδιον πρῶτον, 216—462. 


4. στάσιμον πρῶτον, 463—512. 





5. ἐπεισόδιον δεύτερον, 513—862, with κομμός, 649—697. 


6. στάσιμον δεύτερον, 863—9I0. 





7. ἐπεισόδιον τρίτον, g1 I—1085. 


8. στάσιμον τρίτον, 1o86—I1109. 





9. ἐπεισόδιον τέταρτον, 11 10O—I1185. 


10. στάσιμον τέταρτον, 1186--- 1222. 





11. ἔξοδος, 1223—15 30. 


In reference to a Greek tragedy, we cannot properly speak οὗ ‘ Acts’; 
but the πάροδος and the στάσιμα mark the conclusion of chapters in the 
action. The Oedipus Tyrannus falls into six such chapters. 

The parts named above are thus defined by Aristotle (Poet. 12) :— 

I. πρόλογος = μέρὸς ὅλον τραγῳδίας τὸ πρὸ χοροῦ παρόδου, ‘all that 
part of a tragedy which precedes the parodos’ (or ‘entrance’ of the 
Chorus into the orchestra). 

2. πάροδος -- ἡ πρώτη λέξις ὅλου χοροῦ, ‘the first utterance of the 
- whole Chorus.’ 

3. ἐπεισόδιον = μέρος ὅλον τραγῳδίας τὸ μεταξὺ ὅλων xopiKdv μελῶν, 
‘all that part of a tragedy which comes between whole choric songs.’ 

4. στάσιμον -- μέλος χοροῦ τὸ ἄνευ ἀναπαίστου Kal Tpoxaiov, ‘a song 
of the Chorus without anapaests or trochaics.’ στάσιμον is ‘stationary’: 
στάσιμον μέλος, a song by the Chorus at its s¢a¢ion— after it has taken up 
its place in the orchestra—as distinguished from the πάροδος or entrance- 
song. [I do not now think that the notion of ‘unbroken’—by anapaests 
or dialogue—can be included in the term. ] 

Aristotle’s definition needs a few words of explanation. (1) The 
anapaestic was especially a marching measure. Hence the πάροδος of 


OIAITOY2 ΤΥΡΆΝΝΟΣ 9 


the older type often began with anapaests (eg. Aesch. Agam. 40—103, 
Eum. 307—320), though, in the extant plays of Soph., this is so with 
the Ajax alone (134—171). Buta στάσιμον never degins with anapaests. 
Further, the antistrophic arrangement of a στάσιμον is never znterrupted 
by anapaests. Yet, after an antistrophic στάσιμον, the choral utterance 
may evd with anapaests: thus the third στάσιμον of the Antigone is 
antistrophic from 781 to 800, after which come immediately the choral 
anapaests 801—8o05: and we should naturally speak of 781—8o05 as 
the third stasimon, though, according to Arist., it strictly consists only 
of 781—800. (2) By tpoyaiov Arist. plainly means the trochaic ¢efra- 
meter: 1.6. ἃ στάσιμον must not be interrupted by dialogue (such as 
that which the Chorus holds in trochaic tetrameters with Aegisthus and 
Clytaemnestra, Aesch. Ag. ad jin.) Measures into which trochaic 
rhythms enter are, of course, frequent in στάσιμα.- 

5. ἔξοδος = μέρος ὅλον τραγῳδίας μεθ᾽ ὃ οὐκ ἔστι χοροῦ μέλος, ‘all that 
part of a tragedy after which there is no song of the Chorus.’ 

Verses 649—697 of the second ἐπεισόδιον form a short Koppés. The 
Chorus are pleading with Oedipus, lyric measures being mingled with 
iambic trimeters. Arist. (7e¢. 12) defines the κομμός as θρῆνος κοινὸς 
χοροῦ καὶ ἀπὸ σκηνῆς, 1.6. a lamentation in which the Chorus (in the 
orchestra) took part with the actor on the stage. An example of the 
κομμός on a larger scale is Soph. £2. 121—250. 


10 ZOPOKAEOY2 


OIAITIOTS. 


Ὦ TEKNA, Καδμου τοῦ πάλαι νέα τροφή, 

τίνας ποθ᾽ ee τάσδε μοι θοάζετε 

ἱκτηρίοις κλάδοισιν ἐξεστεμμένοι ; 

πόλις δ᾽ ὁμοῦ μὲν θυμιαμάτων γέμει, 

ὁμοῦ δὲ παιάνων τε καὶ στεναγμάτων' 5 
ἀγὼ δικαιῶν μὴ παρ᾽ ἀγγέλων, τέκνα, 

ἄλλων ἀκούειν αὐτὸς ὧδ᾽ ἐλήλυθα, 

O πᾶσι κλεινὸς Οἰδίπους καλούμενος. 


ἀλλ᾽, ὦ γεραιέ, φράξ, 


L=cod. Laur. 32. 9 (first half of eleventh century). 


later MSS.: see Introd. on the text. 


ἐπεὶ πρέπων ἔφυς 
πρὸ τῶνδε φωνεῖν, τίνι τρόπῳ καθέστατε, 


Io 


r=one or more of the 


This symbol is used where a more particular 





Scene :— Before the palace of Oedipus 
at Thebes. In front of the large central 
doors (βασίλειος θύρα) there ts an altar; 
a smaller altar stands also near each of 
the two side-doors: see verse 16. Sup- 
pliants—old men, youths, and young 
children—are seated on the steps of the 
aliars. They are dressed in white tunics 
and cloaks,—their hair bound with white 
fillets. On the altars they have laid down 
olive-branches wreathed with fillets of 
wool. The PRIEST OF ZEUS, a venerable 
man, ts alone standing, facing the central 
doors of the palace. These are now thrown 
open: followed by two attendants (πρόσπο- 
“ λοι), who place themselves on either side 
of the doors, OEDIPUS enters, in the robes 
of a king: for a moment he gazes silently 
on the groups at the altars, and then 
speaks. See Appendix, Note 1, § 1. 

1—77 Oedipus asks why they are 
suppliants. The Priest of Zeus, speak- 
ing for the rest, prays him to save them, 
with the gods’ help, from the blight and 
the plague. Oedipus answers that he 
has already sent Creon to consult Apollo 
at Delphi, and will do whatever the god 
shall bid. 

1 véa, last-born (not ‘young,’ for réxva 
includes the old men, v. 17), added for 
contrast with τοῦ πάλαι, Oedipus,—who 
believes himself a Corinthian (774),— 


marks his respect for the ancient glories 
of the Theban house to whose throne he 
has been called: see esp. 258 f. 80 the 
Thebans are στρατὸς Καδμογενής “Aesch. 
Theb. 303, Καδμογενὴς γέννα Eur.Phoen. 
808, or Καδμεῖοι. τροφή = Ofgmuara 


(abstract for concrete); Eur. Cycl. 189 
ἀρνῶν tpopal=dpves ἐκτεθραμμέναι. Cad- 


mus, as guardian genius of Thebes, is 


. Still τροφεύς of all who are reared in the 


δῶμα Καδμεῖον (v. 29). Campbell under- 
stands, ‘my last-born care derived from 
ancient Cadmus,’—as though the τροφεύς 
were Oedipus. But could Κάδμου τροφή 
mean ‘[72)'] nurslings [derived from] Cad- 
mus’? It is by the word τέκνα that 
Oedipus expresses his own fatherly care. 
2 ἕδρας. The word &pa=‘posture,’ 
here, as usu., sitting: when kneeling is 
meant, some qualification is added, as 
Eur. Ph. 293 γονυπετεῖς ἕδρας προσ- 
πίτνω σ᾽, ‘I supplicate thee on my 
knees.’ The suppliants are sitting on 
the steps (βάθρα) of the altars, on which 
they have laid the κλάδοι: see 142: cp. 
15 προσήμεθα, 20 θακεῖ: Aesch. um. 40 
(Orestes a suppliant in the Delphian 
temple) ἐπ’ ὀμφαλῴ (on the omphalos) 
ἕδραν ἔχοντα προστρόπαιον... «ἐλαίας θ᾽ 
ὑψυγέννητον κλάδον. θοάζετε prob. = θάσ- 
cere, ‘sit,’ ἕδρας being cognate acc. In 
Eur. θοάζω (@0ds) always=‘to hasten’ 


OIAITOY2 TYPANNOZ II 


OEDIPUS. 


My children, latest-born to Cadmus who was of old, why 
are ye set before me thus with wreathed branches of suppliants, 
while the city reeks with incense, rings with prayers for health 


and cries of woe? 


I deemed it unmeet, my children, to hear 


these things at the mouth of others, and have come hither myself, 


I, Oedipus renowned of all. 


Tell me, then, thou venerable man—since it is thy natural 
part to speak for these—in what mood are ye placed here, 


statement is unnecessary. 
known to the editor. 


‘mss.,’ after a reading, means that it is in all the mss. 





(transitive or intrans.). But Empedocles 
and Aesch. clearly use θοάξω as = θάσσω, 
the sound and form perh. suggesting the 
epic θαάσσω, θόωκος. See Appendix. 

8 ixrnplois κλάδοισιν. The suppliant 
carried a branch of olive or laurel (ixe- 
τηρία), round which were twined festoons 
of wool (στέφη, oréupara,—which words 
can stand for the ixernpla itself, zzfra 
913, Z/. 1. 14): Plut. Zhes. 18 jv 6é [ἢ 
ixernpla] κλάδος ἀπὸ τῆς ἱερᾶς ἐλαίας, ἐρίῳ 
λευκῷ κατεστεμμένος. He laid his branch 
on the altar (Eur. Her. 124 βωμὸν κατα- 
στέψαντες), and left it there, if unsuc- 
cessful in his petition (Eur. Swfp/. 259) ; 
if successful, he took it away (23. 359, 
infra 143). tr. KA. ἐξεστεμμένοι -- ἱκτη- 
ρίους κλάδους ἐξεστεμμένους ἔχοντες: Xen. 
Anab. 4. 3. 28 διηγκυλωμένους τοὺς 
ἀκοντιστὰς καὶ ἐπιβεβλημένους τοὺς 
τοξότας, ‘the javelin-throwers w2¢/ javelins 
grasped by the thong (ἀγκύλη), and the 
archers with arrows /it/ed to the string.’ 
So 18 ἐξεστεμμένον absol.,=provided 
with στέφη (z.c. with ἱκετηρίαι: see last 
note). Triclinius supposes that the sup- 
pliants, besides carrying boughs, wore 
garlands (ἐστεφανωμένοι), and the priests 
may have done so: but ἐξεστεμμ. does not 
refer to this. 

4 : μοῦ μὲν... ὁμοῦ δὲ. The verbal con- 
trast is merely between the fmes of in- 
cense burnt on the altars as a propitiato 
offering (Z/. 8. 48 τέμενος βωμός τε dukes, 
and the sownds—whether of invocations 
to the Healer, or of despair. 

7 ἄλλων. Redundant, but serving to 
contrast ἀγγέλων and αὐτός, as if one 


said, ‘from messengers, —at second hand.’ 
Blaydes cp. Xen. Cyr. τ. 6. 2 ὅπως μὴ 
δι’ ἄλλων ἑρμηνέων τὰς τῶν θεῶν συμ- 
βουλίας συνείης, ἀλλ᾽ αὐτὸς... γιγνώσκοις. 
ὧδε-Ξ- δεῦρο, as in vv. 144, 298, and often 
in Soph.: even with βλέπειν, ὁρᾶν, as in 
Trach. 402 βλέφ᾽ ὧδε-- βλέπε δεῦρο. 

8 ὁ πάσι κλεινὸς... καλούμενος. πᾶσι 
with κλεινός (cp. 40 πᾶσι κράτιστον), not 
with καλούμενος : ‘called Oedipus famous 
in the sight of all,’ not ‘called famous 
Oed. by all.’ Cp. πασίγνωστος, πασί- 
δηλος, πασιμέλουσα, πασίφιλος. The tone 
is Homeric (Od. 9. 19 εἴμ᾽ ᾿Οδυσεύς... 
kal μευ κλέος οὐρανὸν ἵκει, imitated by 
Verg. Aen. I. 378 sum pius Aeneas...fama 
super aethera notus): Oedipus is a type, 
for the frank heroic age, of Arist.’s μεγα- 
λόψυχος----ὁ μεγάλων αὑτὸν ἀξιῶν, ἄξιος 
ὧν (Eth. N. 4. 3). 

9 ἔφυς, which is more than εἶ, refers, 
not to appearance (φυή), but to the na- 
tural claim (φύσις) of age and office com- 
bined. 

10 πρὸ τῶνδε, ‘in front of,’ and so 
‘on behalf of,’ ‘for’ these. Ellendt: 
‘Non est ἀντὲ τῶνδε, nec ὑπὲρ τῶνδε, sed 
μᾶλλον 5. μάλιστα τώνδε, prae ceteris 
dignus propter auctoritatem et aetatem.’ 
Rather ἀντὶ révde=‘as their deputy’: 
ὑπὲρ twrde=‘as their champion’: mpd 
τῶνδετε “45 their spokesman.’ So O.C. 
811 ἐρῶ yap καὶ πρὸ τῶνδε. τίνι τρόπῳ 
with καθέστατε only: δείσαντες ἢ στέρ- 
ἕαντες Ξ- εἴτε ἐδείσατέ τι, εἴτε ἐστέρξατε (not 
πότερον δείσαντες; ἢ στέρξαντες ;), ‘in what 
mood are ye set here, whether it be one 
of fear or of desire ?” 


12 ΣΟΦΘΚΛΕΘΥΣ 


δείσαντες ἢ ἢ στέρξαντες ; ὡς θέλοντος ἃ ἂν 
ἐμοῦ προσαρκεῖν may" δυσάλγητος yap av 
εἴην τοιάνδε μὴ οὐ κατοικτίρων ἕδραν. 


ΤΟΡΕΣ, 


ἀλλ᾽, ὦ κρατύνων Οἰδίπους χώρας ἐμῆς, 

ὁρᾷς μὲν ἡμᾶς ἡλίκοι προσήμεθα ; 15 
βωμοῖσι Tots σοῖς, οἱ μὲν οὐδέπω μακρὰν 
πτέσθαι σθένοντες, οἱ δὲ σὺν γήρᾳ βαρεῖς, 


ἱερῆς, ie μὲν Ζηνός, οἵδε τ᾽ ἠθέων 
ὁ δ᾽ ἄλλο φῦλον ἐξεστεμμένον 


λεκτοί:" 


11 στέρξαντες L rst hand, changed by a later hand into στέξαντες : 


gloss, ἤδη πεπονθότες. 


πο: 


marginal 


The reading στέξαντες, found in r, was intended to mean, 


‘having endured,’ and may have been suggested by the glosses παθόντες, ὑπομείναντες, 


explaining στέρξαντες. 


Μ55.: ἱερῆς Brunck: ἱερεὺς Bentley: 


18 μὴ οὐ κατοικτείρων 1,: μὴ κατοικτείρων r. 
ἱερεὺς ἔγωγε Nauck.—oi δέ 7’ ἠϊθέων L: 


18 ἱερεῖς 
the 7’ 





11 στέρξαντες, Shaving formed a de- 
sire’: the aor. part., as Ad. 212 ἐπεί 
e... | στέρξας ἀνέχει ‘is constant to the 
love which he hath formed for thee.’ £7. 
1100 καὶ τί βουληθεὶς πάρει; At. 1052 
αὐτὸν é\micavres.. (ἄγειν. Cpr. ὩΣ Ὁ: 
1093 καὶ τὸν ἀγρευτὰν ᾿Απόλλω | καὶ κα- 
σιγνήταν.. .] στέργω διπλᾶς ἀρωγὰς | μο- 
recy, ‘I desire’: where, in such an invo- 
cation (iw...Zev,...répos, «.7..), στέργω 
surely cannot mean, ‘I am content.’ Oed. 
asks: ‘Does this supplication mean that 
some new dread has seized you (δείσαντες) ? 
Or that ye have set your hearts (στέρξαντες) 
on some particular boon which I can 
grant?’—Others render στέρξαντες ‘hav- 
ing acquiesced.’ This admits of two views. 
(i) ‘Are ye afraid of suffering? Or have ye 
‘ already /earned to bear suffering?’ To this 
point the glosses ὑπομείναντες, παθόντες. 
But this seems unmeaning. He snows 
that the suffering has come, and he does 
not suppose that they are resigned to it 
(cp. v. 58). (ii) Prof. Kennedy connects 
ἢ στέρξαντες ὡς θέλοντος ἂν | ἐμοῦ προσ- 
αρκεῖν πᾶν; 1.6. are ye come in vague 
terror, or in contentment, as believing 
that I would be willing to help you? 
This is ingenious and attractive. But 
(2) it appears hardly consonant with the 
kingly courtesy of this opening speech for 
Oedipus to assume that their belief in his 
good-will would reconcile them to their 
present miseries. (4) We seem to re- 
quire some direct and express intimation 


of the king’s willingness to help, such 
as the words ws θέλοντος...πᾶν give only 
when referred to φράζε. (c) The rhythm 
seems to favour the question at orép- 
ἕαντες.---στέξαντες, explained as ‘hav- 
ing endured,’ may be rejected, because 
(1) the sense is against it—see on (i) 
above: (2) στέγειν in classical Greek =‘ to 
be proof against,’ not ‘to suffer’: (3) 
στέξω, ἔστεξα are unknown to Attic, 
which has only the pres. and the imperf. 
ὡς θέλοντος ἂν (to be connected with 
pave) implies the apodosis of a conditional 
sentence. Grammatically, this might be 
either (a) εἰ δυναίμην, θέλοιμι ἄν, or (ὁ) εἰ 
ἠδυνάμην, ἤθελον av: here, the sense 
fixes it to (a). ὡς, thus added to the 
gen. absol., expresses the supposition on 
which the agent acts. Xen. Mem. 2. 6. 
32 ws οὐ mpocolcovros (ἐμοῦ) τὰς χεῖρας, .... 
δίδασκε: ‘as (you may be sure) I will not 
lay hands on you, teach me.’ 

13 κατοικτίρων. οἰκτίρω, not οἰκτείρω, 
is the spelling attested by Attic inscrip- 
tions of circ. 550—350 B.C.: see Meister- 
hans, Grammatik der Attischen Inschrif- 
ten, p. 89. μὴ οὐ κατοικτίρων. An infini- 
tive or participle, which for any-s:ason 
would regularly take μή, usually takes μὴ 
ov if the principal verb of the sentence is 
negative. Here, δυσάλγητος Ξεοὐκ εὐάλ- 
yntos: Dem. Fals. Legat. § 123 (πόλεις) 
χαλεπαὶ λαβεῖν...μὴ οὐ χρόνῳ Kal πολιορ- 
kia (sc. λαμβάνοντὴ), where xaderai=ovd 
ῥᾷδιαι: ‘cities mot easy to take, unless 


OlAHTOyY = .TyYPANNOS 13 


with what dread or what desire? Be sure that I would gladly 
give all aid; hard of heart were I, did I not pity such sup- 
pliants as these. 


PRIEST OF ZEUS. 


Nay, Oedipus, ruler of my land, thou seest of what years we 
are who beset thy altars,—some, nestlings still too tender for far 
flights,—some, bowed with age, priests, as I of Zeus,—and these, 
the chosen youth; while the rest of the folk sit with wreathed 


does not seem to have ever been π᾿, but may have been made from τε. οἱ δ᾽ ἠϊθέων r. 
—Dobree conj. οἱ δέ γ᾽ or οἵδε δ᾽ : Elmsley, οἱ δ᾽ ἔτ᾽: Wecklein οἱ δ᾽ ἑξῆς θεών (‘ceteri 
ex ordine lecti deorum sacerdotes’). Dindorf edits οἱ δ᾽ ἐπ᾽ ἠθέων (which Diibner 
believes to have been written by the rst hand in L): and this had been conjectured by 
Wunder, who afterwards edited οἱ δ᾽ ἰηθέων, relying on a corrupt reading, of δέ 7’ 





by a protracted siege.’ The participial wv, unmarried youths: //. 18. 593 ἠΐθεοι 


clause, μὴ οὐ κατοικτίρων, is equivalent 
to a protasis, ef μὴ κατοικτίροιμι. Prof. 
Kennedy holds that the protasis is εἰ μὴ 
θέλοιμι understood, and that μὴ οὐ κα- 
τοικτίρων is epexegetic of it:—‘ Yes (γάρ) 
I should be unfeeling, zf 7 did not wish 
(to help you): that is, if I refused to pity 
such a supplication as this.’ But the 
double negative μὴ οὐ could not be ex- 
plained by a negative in the rotaszs 
(εἰ μὴ θέλοιμι) : it implies a negative in 
the apodosis (δυσάλγητος ἂν εἴην). Since, 
then, the resolution into οὐκ εὐάλγητος av 
εἴην is necessary, nothing seems to be 
gained by supposing a suppressed protasis, 
εἰ μὴ θέλοιμι. 

16 βωμοῖσι τοῖς σοῖς. The altars of 
the προστατήριοι θεοί in front of the 
palace, including that of Apollo Λύκειος 
(919). μακρὰν πτέσθαι. So Andromache 
to he: child—veoocds ὡσεὶ πτέρυγας ἐσ- 
πίτνων ἐμάς Eur. 7γυ. 746. The proper 
Attic form for the aor. of méroua was 
ἐπτόμην, which alone was used in prose 
and Comedy. Though forms from ἐπ- 
τάμην sometimes occur in Tragedy, as 
in the Homeric poems, Elms. had no 
cause to wish for πτάσθαι here. 

17 σὺν γήρᾳ βαρεῖς-- βαρεῖς ws γήρᾳ 
συνόντες. O.C. 1663 σὺν νόσοις | ἀλγει- 
νός. 

18 ἐγὼ μὲν. The answering clause, οἱ 
δὲ ἄλλων θεῶν, must be supplied mental- 
ly: cp. Z/. 5. 893 τὴν μὲν ἐγὼ σπουδῇ 
δάμνησ᾽ ἐπέεσσι (sc. Tas δὲ ἄλλας ῥᾳδίως). 
It is slightly different when μέν, used 
alone, emphasizes the personal pronoun, 
as in ἐγὼ μὲν οὐκ olda Xen. Cyr. 1. 4.12 
οἵδε τ. The conjecture οἱ δ᾽ ἐπ᾽ (‘chosen 
to represent the youth’) involves a ques- 
tionable use of ἐπέ: cp. Ant. 787n. ἢθέ- 


kal παρθένοι: Eur. Phoen. 944 Αἵμονος... 
γάμοι | σφαγὰς ἀπείργουσ᾽" οὐ γάρ ἐστιν 
ἤθεος: Plut. Zhes. 15 ἠθέους ἑπτὰ καὶ παρ- 
θένους. 

19 ἐξεστεμμένον : see on 3. 20 ἀγο- 
ραῖσι, local dative, like οἰκεῖν οὐρανῷ 
Pind. Mem. 10. 58. Thebes was divided 
from N. to S. into two parts by the 
torrent called Strophia. The W. part, 
between the Strophia and the Dircé, was 
the upper town or Cadmeia: the E. part, 
between the Strophia and the Ismenus, 
was ἡ κάτω πόλις. The name Καδμεία 
was given especially to the S. eminence 
of the upper town, the acropolis. (1) 
One of the dyopal meant here was on a 
hill to the north of the acropolis, and was 
the ἀγορὰ Καδμείας. See Paus. 9. 12. 3. 
(2) The other was in the lower town. 
Xen. Hellen. 5. 2. 29 refers to this—} 
βουλὴ ἐκάθητο ἐν τῇ ἐν ἀγορᾷ στοᾷ, διὰ τὸ 
τὰς γυναῖκας ἐν τῇ Καδμείᾳ θεσμοφοριάζειν: 
unless Καδμεία has the narrower sense of 
‘acropolis.’ Cp. Arist. Pol. 4 (7). 12. 2 
on the Thessalian custom of having two 
d-yopai—one, ἐλευθέρα, from which every- 
thing βάναυσον was excluded. πρός τε 
Παλλάδος... ναοῖς. Not ‘doth at the two 
temples,’ &c. asif this explained ἀγοραῖσι, 
but ‘and,’ &c.: for the dyopat would have 
their own altars of the ἀγοραῖοι θεοί, as 
of Artemis (161). One of the διπλοῖ ναοί 
may be that of Παλλὰς “Oyxa, near the 
’Oyxala πύλη on the W. side of Thebes 
(πύλας ["Ογκας ᾿Αθάνας Aesch. Thed. 487, 
Ὄγκα Παλλάς 26. 501), whose statue and 
altar ἐν ὑπαίθρῳ Paus. mentions (9. 12. 2). 
The other temple may be that of Athene 
Καδμεία or of Athena Iounvia—both 
mentioned by the schol., but not by Paus. 
Athena Zwornpia, too, had statues at 


14 ZOPOKAEOYS 


ἀγοραῖσι θακεῖ, πρός τε Παλλάδος διπλοῖς 20 
ναοῖς, ἐπ᾽ Ἰσμηνοῦ Te μαντείᾳ σποδῷ. 

πόλις γάρ, ὥσπερ καὐτὸς εἰσορᾷς, ἄγαν 

non σαλεύει ᾿κανακουφίσαι κάρα 

βυθῶν er οὐχ οἵα τε φοινίου “σάλου, ¥ 
φθίνουσα μὲν κάλυξιν ἐγκάρποις i bones: x 25 
φθίνουσα δ᾽ ἀγέλαις. βουνόμοις τόκοισί τε 


ἀγόνοις γυναικῶν" 


ἐν δ᾽ ὁ πυρφόρος θεὸς 


σκήψας ἐλαύνει, λοιμὸς ἔχθιστος, πόλιν, 
ὑφ᾽ οὗ κενοῦται δῶμα Kadpetov: μέλας δ᾽ 


Αιδης στεναγμοῖς καὶ γόοις πλουτίζεται. 


ἰηθέων in Suidas 5. ν. λεκτός. 


21 μαντείᾳ L, made from μαντείασ: 


30 


the upper part 





Thebes (Paus. 9. 17. 3). The schol. 
mentions also’AXaAkomevia, but her shrine 
was at the village of Alalcomenae near 
Haliartus (Paus. 23 ΣΕ). Lt owas 
enough for Soph. that his Athenian 
hearers would think of the Erechtheum 
and the Parthenon—the shrines of the 
Polias and the Parthenos—above them 
on the acropolis. 

21 ἐπ᾽ Ἴσμ. p. σποδῷ. ‘The oracular 
ashes of Teaeny =the altar in the temple 
of Apollo Ἰσμήνιος, where divination by 
burnt offerings (ἡ δι’ ἐμπύρων μαντεία) was 
practised. So the schol., quoting Philo- 
chorus (in his περὶ μαντικῆς, circ. 290B.C.). 
σποδῷ: the embers dying down when 
the μαντεῖον has now been taken from the 
burnt offering: cp. Amt. 1007. Soph. 
may have thought of ᾿Απόλλων Σπόδιος, 
whose altar (ἐκ τέφρας τῶν ἱερείων) Paus. 
saw to the left of the Electrae gates at 
Thebes: 9. 11. 7. ᾿Ισμηνοῦ, because 
the temple was by the river Ismenus: 
‘ Paus. g. 10. 2 ἔστι δὲ λόφος ἐν δεξιᾷ τῶν 
πυλῶν (on the right of the ᾿Ηλέκτραι πύλαι 
on the S. of Thebes, within the walls) 
ἱερὸς ᾿Απόλλωνος " καλεῖται δὲ ὅ τε λόφος 
καὶ ὁ θεὸς ᾿Ισμήνιος, παραρρέοντος τοῦ ποτα- 
μοῦ ταύτῃ τοῦ ᾿Ισμηνοῦ. Ismenus (which 
name Curtius, tym. 617, connects with 
rt és, to wish, as=‘ desired’) was described 
in the Theban myths as the son of 
Asopus and Metope, or of Amphion and 
Niobe. The son of Apollo by Melia (the 
fountain of the Ismenus) was called Is- 
menius. Cp. Her. 8. 134 (the envoy of 
Mardonius in the winter of 480—79) τῷ 
Ἰσμηνίῳ ᾿Απόλλωνι ἐχρήσατο" ἔστι δὲ 
κατάπερ ἐν ᾿Ολυμπίῃ ἱροῖσι χρηστηριάζε- 
σθαι: Pind. Olymp. 8. init. Οὐλυμπία | 


..lva μάντιες ἄνδρες | ἐμπύροις τεκμαιρό- 
μενοι παραπειρῶνται Διός. Ιῃ Pind. 
Pyth. 11. 4 the Theban heroines are 
asked to come πὰρ MeXlav (because she 
shared Apollo’s temple) ‘to the holy 
treasure-house of golden tripods, which 
Loxias hath honoured exceedingly, and 
hath named it /smenian, a truthful seat 
of oracles’ (MSS. μαντείων, not μαντίων, 
Fennell): for the tripod dedicated by the 
δαφναφόρος, or priest of Ismenian Apollo, 
see Paus. g. 10. 4. Her. saw offerings 
dedicated by Croesus to Amphiaraus ἐν 
τῷ νηῷ τοῦ Ἰσμηνίου ᾿Απόλλωνος (1. 52), 
and notices inscriptions there (5. 59). The 
Ἰσμήνιον, the temple at Abae in Phocis, 
and that on the hill Πτῶον to the E. of 
Lake Copais, were, after Delphi, the chief 
shrines of Apollo in N. Greece. 

24 βυθῶν, ‘from the depths,’ 2.6. out 
of the trough of the waves which rise 
around. CP: Ant. 337 περιβρυχίοισιν | 
περῶν ὑπ᾽ οἴδμασιν, under swelling waves 
which threaten to engulf him. Arat. 426 
ὑπόβρυχα ναυτίλλονται. dowlov here 
merely poet. for θανασίμου, as 77. 770 
φοινίας | ἐχθρᾶς ἐχίδνης ἰός: O.C. 1689 
φόνιος ᾿Αἴδας. But in Az. 351 φοινία ζάλη 
=the madness which drove Ajax to 
bloodshed. ἔτ' οὐχ οἵα re: for position of 
ἔτι, cp. Trach. 161 ὡς ἔτ᾽ οὐκ ὦν, Phil. 
1217 ἔτ᾽ οὐδέν elu. With οἷός τε the- 
verb is often omitted, as 1415, O.C. 
1136, Zr. 742, Ar. Eg. 343 

25 2. φθίνουσα μὲν.. “φθίνουσα δέ, 
rhetorical iteration (ἐπαναφορά) ; cp. 259, 
370, O.C. 5, 610, etc. The anger of 
heaven is shown (1) by a dlight (φθίνουσα) 
on the fruits of the ground, on flocks and 
on child-birth: (2) by a pestilence (λοιμός) 


OIAITOY2 ΤΥΡΆΝΝΌΣ 15 


branches in the market-places, and before the two shrines of 
Pallas, and where Ismenus gives answer by fire. } 

’ For the city, as thou thyself seest, is now too sorely vexed, 
and can no more lift her head from beneath the angry waves 
of death; a blight is on her in the fruitful blossoms of the land, 
in the herds among the pastures, in the barren pangs of women; 
and withal the flaming god, the malign plague, hath swooped on 
us, and ravages the town; by whom the house of Cadmus is 


made waste, but dark Hades rich in groans and tears. 


of the o can be traced. μαντεῖα or μαντεία r. 


29 καδμεῖον L. καδμείων τ. 


Cp. 





which ravages the town. Cp. 171 ff. 
For the threefold blight, Her. 6. 139 
ἀποκτείνασι δὲ τοῖσι Πελασγοῖσι τοὺς σφε- 
τέρους παῖδάς τε καὶ γυναῖκας οὔτε γῆ 
καρπὸν ἔφερε οὔτε γυναῖκές τε καὶ ποῖμναι 
ὁμοίως ἔτικτον καὶ πρὸ τοῦ: Aeschin. 2722 
Ctes. $111 μήτε γῆν καρποὺς φέρειν μήτε 
. γυναῖκας τέκνα τίκτειν γονεῦσιν ἐοικότα, 
ἀλλὰ τέρατα, μήτε βοσκήματα κατὰ φύσιν 
γονὰς ποιεῖσθαι. Schneid. and Blaydes 
cp. Philostratus Vit. Apoll. 3. 20, p. 51. 
21 ἡ γῆ ov ξυνεχώρει αὐτοῖς ἵστασθαι" τήν 
τε γὰρ σπορὰν ἣν ἐς αὐτὴν ἐποιοῦντο, πρὶν 
ἐς κάλυκα ἥκειν, ἔφθειρε, τούς τε τῶν γυ- 
ναικῶν τόκους ἀτελεῖς ἐποίει, καὶ τὰς ἀγέ- 
λας πονηρῶς ἔβοσκεν.---κάλυξιν ἐγκάρ- 
ποις. The datives mark the points or 
parts in which the land φθίνει. κάλυξ 
ἔγκαρπος is the shell or case which en- 
closes immature fruit,—whether the 
blossom of fruit-trees, or the ear of 
wheat or barley: Theophr. Ast. Plant. 
8. 2. 4 (of κριθή and πυρός) πρὶν ἂν προαύ- 
ξηθεὶς (ὁ στάχυς) ἐν TH κάλυκι γένηται. 

26 ἀγέλαι βουνόμοι (ρατοχγί.) Ξ- ἀγέλαι 
βοῶν νεμομένων : but ἀκτὴ βούνομος, pro- 
paroxyt., a shore on which oxen are 
pastured, Z7. 181. Cp. Z/. 861 χαλαρ- 
γοῖς ἐν ἁμίλλαις Ξε ἁμίλλαις ἀργῶν χηλῶν: 
Pind. Pyth. 5. 28 ἀρισθάρματον...γέρας = 
γέρας ἀρίστου ἅρματος. The epithet 
marks that the blight on the flocks is 
closely connected with that on the 
pastures: cp. Dionys. Hal. 1. 23 (de- 
scribing a similar blight) οὔτε πόα κτήνε- 
σιν ἐφύετο διαρκής. τόκοισι, the labours 
of child-bed: Eur. Med. 1031 στερρὰς 
ἐνεγκοῦσ᾽ ἐν τόκοις ἀλγηδόνας: Tph. T. 
1466 γυναῖκες ἐν τόκοις ψυχορραγεῖς. 
Dionys. Hal. 1. 23 ἀδελφὰ δὲ τούτοις (.6. 
to the blight on fruits and crops) ἐγίνετο 
περί τε προβάτων καὶ γυναικῶν γονάς" ἣ 
γὰρ ἐξημβλοῦτο τὰ ἔμβρυα, ἣ κατὰ τοὺς 
τόκους διεφθείρετο ἔστιν ἃ καὶ τὰς φερούσας 
συνδιαλυμηνάμενα. 


27 ἀγόνοις, abortive, or resulting in a 
still birth. ἐν δ᾽, adv., ‘and among our 
other woes,’ ‘and withal’: so 181, 77. 
206, Az.675. Not in ‘tmesis’ with σκή- 
yas, though Soph. has such tmesis else- 
where, Ant. 420 ἐν δ᾽ ἐμεστώθη, 10. 1274 
ἐν δ᾽ ἔσεισεν. For the simple ox has, 
cp. Aesch. Ag. 308 εἶτ᾽ ἔσκηψεν, ‘then it 
swooped.’ So fers. 715 λοιμοῦ τις ἦλθε 
σκηπτός. ὁ πυρφόρος θεὸς, the bringer of 
the plague which spreads and rages /i/e 
fire (176. κρεῖσσον ἀμαιμακέτου πυρός, τοῖ 
φλέγει με): but also with reference to 
Sever, πυρετός. Hippocrates 4. 140 ὁκόσοισι 
δὲ τῶν ἀνθρώπων πῦρ (= πυρετὸς) ἐμπίπτῃ: 
fl. 22. 31 καί τε φέρει (Seirius) πολλὸν 
πυρετὸν δειλοῖσι βροτοῖσι (the only place 
where πυρετός occurs in 724. or Od.). In 
O.C. 55 ἐν δ᾽ ὁ πυρφόρος θεὸς | Τιτὰν 
Προμηθεύς refers to the representation of 
Prometheus with the narthex, or a torch, 
in his right hand (Eur. Phoen. 1121 δεξιᾷ 
δὲ λαμπάδα | Τιτὰν ἹΤρομηθεὺς ἔφερεν ὡς). 
Cp. Aesch. Zheb. 432 ἄνδρα πυρφόρον, | 
φλέγει δὲ λαμπάς, κιτιλ. Here also the 
Destroyer is imagined as armed with a 
deadly brand,—against which the Cho- 
rus presently invoke the holy fires of 
Artemis (206) and the ‘blithe torch’ of 
Dionysus (214). For θεός said of λοιμός, 
cp. Simonid. Amorg. fr. 7. 101 οὐδ᾽ αἶψα 
λιμὸν οἰκίης ἀπώσεται, | ἐχθρὸν συνοικη- 
τῆρα, δυσμενέα θεόν. Soph. fr. 837 ἀλλ᾽ 
ἡ φρόνησις ἁγαθὴ θεὸς μέγας. 

29 μέλας δ᾽: elision at end of v. is 
peculiar in Trag. to Soph., who is said 
to have adopted it from a poet Callias 
(Athen. ro p. 453 E): hence it was called 
εἶδος Σοφόκλειον. Examples: δ᾽ 785, 791, 
22} Οἷς ΤΣ ΑΝ, 1041} ἘΣ τ 7: 
τ᾿ below, 1184: ταῦτ᾽ 332. [In O.C. 1164 
μολόντ᾽ should prob. be μόνον.} In Comedy: 
δ᾽ Ar. Av. 1716, Zecl. 351: μ᾽ Ran. 298. 

30 πλουτίζεται with allusion to Πλού- 
τῶν, as Hades was called by an euphem- 


16 ZOPOKAEOYS 


θεοῖσι μέν νυν οὐκ ἰσούμενόν σ᾽ ἐγὼ 
οὐδ᾽ οἵδε παῖδες ἐζόμεσθ' ἐφέστιοι, 
ἀνδρῶν δὲ πρῶτον ἔν τε συμφοραῖς βίον 


κρίνοντες ἔν ΤΕ 


αιμό νων συναλλαγαῖς" 


ὅς γ᾽ ἐξέλυσας, ἄστυ Καδμεῖον μολών, 25 
σκληρᾶς ἀοιδοῦ δασμὸν ὃ ὃν παρείχομεν" 

καὶ ταῦθ᾽ ὑφ᾽ ἡμῶν οὐδὲν ἐξειδὼς πλέον 

οὐδ᾽ ἐκδιδαχθείς, ἀλλὰ προσθήκῃ θεοῦ 

λέγει “νομίζει θ᾽ ἡμὶν ὀρθῶσαι βίον: 


νῦν τ᾽, ὦ κράτιστον πᾶσιν Οἰδίπου κάρα, 


40 


ἱκετεύομέν σε πάντες οἶδε πρόστροποι 
ἀλκήν τιν᾽ εὑρεῖν ἡμίν, εἴτε του θεῶν 
φήμην ἀκούσας εἴτ᾽ ἀπ᾽ ἀνδρὸς οἶσθά που" 
ὡς τοῖσιν ἐμπείροισι καὶ τὰς ξυμφορὰς 


ν. 35. Θ1 οὐκ ἰσούμενον. 
Μ55.: 


The « in L has been made from x or xl. 
és τ᾽ Elmsley, for correspondence with viv 7’ in v. 40.--- καδμεῖον L: καδμείων r. 


35 ὅς γ᾽ 





ism (ὑποκοριστικῶς, schol. Ar. Plut. 727), 
ὅτι ἐκ τῆς κάτωθεν ἀνίεται ὁ πλοῦτος (crops 
and metals), as Platosays, (γαί. 4034. Cp. 
Soph. fr. 251 (Nauck’) (from the satyric 
drama /xachus) Πλούτωνος (="Acdov) ἥδ᾽ 
ἐπείσοδος: Lucian Zimon 21 (Πλοῦτος 
speaks), ὁ Πλούτων (Hades) ἀποστέλλει 
pe παρ᾽ αὐτοὺς ἅτε πλουτοδότης καὶ μεγαλό- 
δωρος καὶ αὐτὸς ὦν" δηλοῖ γοῦν καὶ τῷ 
ὀνόματι. Schneid. cp. Statius Zhe. 2. 48 
pallentes devius umbras Trames agit nt- 
grique [ovis vacua atria ditat Mortibus. 
31 μέν νυν as in 77. 441.—O0vK ἰσού- 
μενόν σ᾽, governed by κρίνοντες in 34. 
But he begins as if instead of ἑζόμεσθ᾽ 
ἐφέστιοι, ἱκετεύομεν were to follow: hence 
ἰσούμενον instead of toov. It is needless 
to take ἰσούμενον (1) as accus. absol., or 
(2) as governed by ἑζόμεσθ᾽ ἐφέστιοι in 
the sense of ixerevouev,—like φθορὰς... 
ψήφους ἔθεντο Aesch. Ag. 814, or γένος... 
véwoov aivov Suppl. 533. Musgrave conj. 
ἰσούμενοι as=‘deeming equal,’ but the 
midd. would mean ‘making owrselves 
equal,’ like ἀντισουμένου Thuc. 3. II. 
Plato has ἰσούμενον as passive in Phaedr. 
238E, and ἰσοῦσθαι as passive in Parm. 
156 B: cp. 581 ἰσοῦμαι. 
84 δαιμόνων συναλλαγαῖς = ‘conjunc- 
tures’ caused by gods (subjective gen.), 
special visitations, as opposed to the or- 
dinary chances of life (συμφοραῖς βίου). 


Such συναλλαγαί were the visit of the 
pea ie and of the πυρφόρος θεός 
(27). Eee νόσου συναλλαγῇ, a visita- 
tion in τ orm of disease (defining gen.). 
Here, the sense might indeed be, ‘deal- 
ings (of men) with gods,’=6érav ἄνθρωποι 
συναλλάσσωνται δαίμοσιν: but the abso- 
lute use of συναλλαγή for ‘a conjuncture 
of events’ in O.C. 410(n.) favours the 
other view. In 77. 845 ὀλεθρίαισι συναλ- 
λαγαῖς = ‘at the fatal meeting’ of Deia- 
neira with Nessus. But in Ant. 157 θεῶν 
ouwvrvxiat=fortunes sent dy gods. The 
common prose sense of συναλλαγή is 
‘reconciliation,’ which Soph. has in 42, 
792: 

85 ὅς y. The γε of the Mss. suits 
the immediately preceding verses better 
than the conjectural te, since the judg- 
ment (κρίνοντες) rests solely on what Oed. 
has done, not partly on what he is ex- 
pected to do. Owing to the length of 
the first clause (35—39) τ᾽ could easily 
be added to νῦν in 40 as if another τε 
had preceded. ἐξέλυσας.. δασμὸν. The 
notion is not, ‘paid it in full,’ but ‘loosed 
it,’—the thought of the tribute suggesting. 
that of the riddle which Oed. solved. 
Till he came, the δασμός was as a 
knotted cord in which Thebes was 
bound. Cp. Zrach. 653 "ΑρηΞ...ἐξέλυσ᾽ | 
ἐπίπονον ἁμέραν, ‘has burst the bondage 


OIAITOY2 TYPANNOS 17 


It is not as deeming thee ranked with gods that I and these 
children are suppliants at thy hearth, but as deeming thee first 
of men, both in life’s common chances, and when mortals have 
_ to do with more than man: seeing that thou camest to the town 
of Cadmus, and didst quit us of the tax that we rendered to the 
hard songstress ; and this, though thou knewest nothing from us 
that could avail thee, nor hadst been schooled ; no, by a god’s 
aid, 5 said and believed, didst thou uplift our life. 

And now, Oedipus, king glorious in all eyes, we beseech thee, 
all we suppliants, to find for us some succour, whether by the 
whisper of a god thou knowest it, or haply as in the power of 
man; for I see that, when men have been proved in deeds past, 


40 viv δ᾽ Blaydes. 


43 του L, with πον written over it by a late hand. 


που Yr. 





of the troublous day.’ Eur. Phoen. 695 
ποδῶν σῶν μόχθον ἐκλύει παρών, ‘ his pre- 
sence dispenses with (solves the need for) 
the toil of thy feet.’ This is better than 
(1) ‘freed the city from the songstress, in 
respect of the tribute,’ or (2) ‘freed the 
city from the tribute (δασμόν by attrac- 
tion for δασμοῦ), to the songstress.’ 

36 σκληρᾶς, ‘hard,’ stubborn, relent- 
less. Eur. Andr. 261 σκληρὸν θράσος. 
In 391 κύων expresses a Similar idea. 

87 καὶ ταῦθ᾽, ‘and that too’: Azz. 322 
(ἐποίησας τὸ ἔργον) καὶ ταῦτ᾽ ἐπ᾽ dp- 
γυρῷ γε τὴν ψυχὴν προδούς: Li. 614. 
οὐδὲν πλέον, nothing more than anyone 
else knew; nothing that could help thee. 
Plat. Crat. 387 A πλέον τι ἡμῖν ἔσται, 
we shall gain something. Syzzpos. 217 Ὁ 
οὐδὲν γάρ μοι πλέον ἦν, it did not help 
me. ἐξειδὼς--ἐκδιδαχθείς: not having 
heard (incidentally)—much less having 
been thoroughly schooled. 

38 προσθήκῃ θεοῦ, ‘by the aid of a 
god.’ [Dem.] lx Aristog. 1. § 24 ἡ εὐ- 
Takia τῇ τῶν νόμων προσθήκῃ τῶν αἰσχρῶν 
περίεστι, ‘discipline, with the support of 
the laws, prevails against villainy.’ Dionys. 
Hal. v. 67 προσθήκης μοῖραν ἐπεῖχον οὗτοι 
τοῖς ἐν φάλαγγι τεταγμένοις, ‘these served 
as supports to the main body of the troops.’ 
προστίθεσθαί τινι, to take his side: Thuc. 
6. 80 τοῖς ἀδικουμένοις...προσθεμένους: 50 
Soph. O.C. 1332 οἷς ἂν σὺ προσθῇ. (The 
noun προσθήκη does not occur as= ‘ man- 
date,’ though Her. 3. 62 has τό τοι προσέ- 
θηκα πρῆγμα.) The word is appropriate, 
since the achievement of Oed. is viewed as 
essentially a triumph of human wit : a di- 
vine agency prompted him, but remained 
in the background. 


iS. ik 


40 νῦν τ᾽: it is unnecessary to read 
γῦν δ᾽: see on 35. πᾶσιν, ethical dat. 
masc. (cp. 8), ‘in the eyes of all men.’ 77... 
1071 πολλοῖσιν οἰκτρόν. 

42 εἴτε οἷσθα ἀλκήν, ἀκούσας φήμην. 
θεῶν του (dy having heard a voice from. 
some god), εἴτε οἶσθα ἀλκὴν ἀπ᾽ ἀνδρός 
που. We might take ἀπ᾽ ἀνδρὸς with 
ἀλκήν, but it is perh. simpler to take it 
with olga: cp. 398 am’ οἰωνῶν μαθών, 
Thuc. 1. 125 ἐπειδὴ, ἀφ᾽ ἁπάντων ἤκουσαν 
τὴν γνώμην : though παρά (or πρός) τινος 
is more frequent. 

43 φήμην, any message (as ina dream, 
φήμη ὀνείρου, Her. 1. 43), any rumour, 
or speech casually heard, which might be 
taken as a hint from the god. Od. 20. 
98 Ζεῦ πάτερ... | φήμην τίς μοι φάσθω... 
(Odysseus prays), ‘Let some one, I pray, 
show me @ word of omen. Then a 
woman, grinding corn within, is heard 
speaking of the suitors, ‘may they now 
sup their last’: χαῖρεν δὲ κλεηδόνι δῖος 
᾿Οδυσσεύς, ‘rejoiced in the sign of the 
voice.’ ὀμῴφή was esp. the voice of an 
oracle; κληδών comprised inarticulate 
sounds (x). δυσκρίτους, Aesch. P.V. 486). 

44 f. ὡς τοῖσιν... βουλευμάτων. I take 
these two verses with the whole context 
from v. 35, and not merely as a comment 
on the immediately preceding words εἴτ᾽ 
am’ ἀνδρὸς οἶσθά mov. Ocdipus has had 
practical experience (ἐμπειρία) of great 
troubles; when the Sphinx came, his 
wisdom stood the trial. Men who have 
become thus ἔμπειροι are apt to be also. 
(καί) prudent in regard to the future. 
Past facts enlighten the counsels which 
they offer on things still uncertain; and 
we observe that the issues of their coun- 


το 


18 ZOPOKAEOYS 


ζώσας ὁρῶ μάλιστα τῶν βουλευμάτων. 

ἴθ᾽, ὦ βροτῶν ἀριστ᾽, ἀνόρθωσον πόλιν" 
ἴθ. evraBnOn ὡς σὲ νῦν μὲν ἥδε γῆ 
σωτῆρα κληήζει τῆς πάρος προθυμίας: 
ἀρχῆς δὲ τῆς σῆς μηδαμῶς μεμνώμεθα Red rot 
στάντες τ᾽ ἐς ὀρθὸν καὶ πεσόντες ὕστερον, 
ἀλλ᾽ ἀσφαλείᾳ τήνδ᾽ ἀνόρθωσον πόλιν. 
ὄρνιθι γὰρ, καὶ τὴν τότ᾽ αἰσίῳ τύχην 
παρέσχες ἡμῖν, καὶ τανῦν ἴσος γενοῦ. 

ὡς εἴπερ ἄρξεις τῆσδε γῆ; ὥσπερ κρατεῖς, 
ξὺν ἀνδράσιν κάλλιον ἢ κενῆς κρατεῖν" 

ὡς οὐδέν ἐστιν οὔτε πύργος οὔτε ναῦς 
ἔρημος ἀνδρῶν μὴ ξυνοικούντων ἔσω. 


48 πάρος L. The rst hand wrote πάλαι, and then poo over λαι. 


50 


Bo 


The corrector de- 


leted Aa, and wrote pogo in the text. 


49 μεμνώμεθα MSS. : 


μεμνῴμεθα Eustathius. 





sels are not usually futile or dead, but 
effectual. Well may we believe, then, 
that he who saved us from the Sphinx 
can tell us how to escape from the 
plague. Note these points. (1) The 
words ἐμπείροισι and βουλευμάτων serve 
to suggest the antithesis between ast 
and future. (2) τὰς ξυμφορὰς τῶν "Εν 
λευμάτων = literally, the occurrences con- 
nected with (resulting from) the counsels. 
The phrase, ‘issues of counsels,’ concisely 
expresses this. The objection which has 
been made to th’s version, that ξυμφορά is 
not τελευτή, rests on a grammatical fallacy, 
viz., that, in ξυμφορὰ βουλεύματος, the 
genitive must be of the same kind as in 
τελευτὴ βουλεύματος. τύχη is not τελευ- 
τή, yet in O.C. 1506 it stands with a 
gen. of connection, just as ξυμῴφορά does 
here: (θεῶν) τύχην τις ἐσθλὴν τῆσδ᾽ ἔθηκε 
τῆς ὁδοῦ (a good fortune connected with 
this coming). Cp. Thuc. τ. 140 ἐνδέ- 
xeTa yap Tas ξυμφορὰς τῶν πραγμά- 
των οὐχ ἧσσον ἀμαθῶς χωρῆσαι ἢ Kal Tas 
διανοίας τοῦ ἀνθρώπου: the issues of hu- 
man affairs can be as incomprehensible 
in their course as the thoughts of man 
(where, again, the ‘occurrences connect- 
ed with human affairs’ would be more 
literal) : ib. πρὸς τὰς ξυμφορὰς καὶ τὰς 
γνώμας τρεπομένους, altering their views 
according to the events. 3. 87 τῆς Evp- 
φορᾶς TH ἀποβάντι, by the zsswe which 
has resulted. (3) ζώσας is not ‘success- 
ful,’ but ‘ operative,’—effectual for the 


purpose of the βουλεύματα: as v. 482 
ζώντα is said of the oracles which re- 
main operative against ane guilty, and 
Ant. 457 $n ταῦτα of laws which are 
ever in force. Conversely - λόγοι Ov7- 
σκοντες μάτην (Aesch. Cho. 845) are 
threats which come to nothing. The 
scholium in L gives the sense correctly: 
—év τοῖς συνετοῖς Tas συντυχίας Kal 
Tas ἀποβάσεις τῶν βουλευμάτων 
ὁρῶ ζώσας καὶ οὐκ ἀπολλυμένας. See 
Appendix. 

47 εὐλαβήθητι, have a care for thy 
vepute—as the next clause explains. Oed. 
is supposed to be above personal risk; 
it is only the degree of his future glory 
(55) which is in question; a fine touch, 
in view of the destined sequel. 

48 τῆς πάρος τροθυμίας, causal genit.: 
Plat. Crito 43 Β πολλάκις μὲν δή σε... 
εὐδαιμόνισα τοῦ τρόπου. 

49 μεμνώμεθα. This subjunctive oc- 
curs also in Od. 14. 168 πῖνε καὶ ἄλλα 
παρὲξ μεμνώμεθα, Plat. Politicus 285 C 
φυλάττωμεν ... Kal... μεμνώμεθα,  Pheleb, 
21ᾺῚᾺ μεμνώμεθα δὴ καὶ ταῦτα περὶ ἀμφοῖν. 
Eustathius (1303. 46, 1332. 18) cites the 
word here as μεμνῴμεθα (optative). We 
find, indeed, μεμνῷο Xen. “αὖ. τ. 7. 5 
(υ. 2. wep fo), μεμνεῷτο /l. 23. 361, με- 
μνῷτο Xen. Cyr. τ. 6. 3, but these are 
rare exceptions. On the other hand, pe- 
μνήμην Ll. 24. 745, μεμνῇτο Ar. ΖΦ κέ. 
gor, Plat. Rep. 518 A. If Soph. had 
meant the optative he would have written 


ΟἸΔΊΠΟΥΣ. TYPANNOS 19 


the issues of their counsels, too, most often have effect. 


On, best of mortals, again uplift our State! 


‘ On, guard thy 


fame,—since now this land calls thee saviour for thy former 
zeal ;'and never be it our memory of thy reign that we were first 
restored and afterward cast down: nay, lift up this State in such 


wise that it fall no more! 


With good omen didst thou give us that past happiness ; 


now also show thyself the same. 


For if thou art to rule this 


land, even as thou art now its lord, ’tis better to be lord of men 
than of a waste: since neither walled town nor ship is anything, 
if it is void and no men dwell with thee therein. 


69 στάντες τ The rst hand in L omitted 7’, which was added by the corrector. 





μεμνήμεθα : cp. Philoct. 119 ἂν... κεκλῇο. 
See Curtius Greck Verd 11. 226 (Eng. tr. 
p. 423). The personal appeal, too, here 
requires the subjunct., not optat.: cp. O. 
ς. 174 μὴ δῆτ᾽ ἘΣ. Trach. 802 μηδ᾽ 
αὐτοῦ θάνω. 

580 στάντες T K.T.AX. ee partic. with 
μέμνημαι cp. Xen. Cyr. 3. I. 31 ἐμέμνητο 
γὰρ εἰπών : Pind. Mem. 11. 15 θνατὰ pe- 
μνάσθω περιστέλλων μέλη: for Te...Kal, 
Ant. 1112 αὐτός τ᾽ ἔδησα καὶ παρὼν ἐκλύ- 
goat, as I bound, so will I loose. 

51 ἀσφαλείᾳ, ‘in steadfastness’: a 
dative of manner, equivalent to ἀσφαλῶς 
in the proleptic sense of ὥστε ἀσφαλῆ 
twat. Cp, O.C.1318 κατασκαφῇ laste 
δῃώσειν, ἢ. Thuc. 3. 56 of μὴ τὰ ξύμ- 
gopa πρὸς τὴν ἔφοδον αὑτοῖς ἀσφαλείᾳ 
πράσσοντες, those who securely made terms 
on their own account which were not for 
the common good in view of the inva- 
sion. 2. 82 ἀσφαλείᾳ δὲ τὸ ἐπιβουλεύ- 
σασθαι (where ἀσφάλεια is a false read- 
ing), to form designs 77 security, opp. 
to τὸ ἐμπλήκτως ὀξύ, fickle impetuosity. 
The primary notion of ἀσφαλής (‘not 
slipping’) is brought out by πεσόντες 
and ἀνόρθωσον. 

52 ὄρνιθι... αἰσίῳ, like secunda alite 
or fausta avi for bono omine. A bird of 
“omen was properly οἰωνός: Od. 15. 531 
οὔ τοι ἄνευ θεοῦ ἔπτατο δεξιὸς ὄρν ι:" | 
ἔγνων γάρ μιν ἐσάντα ἰδὼν οἰωνὸν ἐόντα: 
Xen. Cyr. 3. 3. 22 οἰωνοῖς χρησάμενος 
αἰσίοις. But cp. Eur. 7, A. 607 ὄρνιθα 
μὲν Tove’ αἴσιον ποιούμεθα : Her. 730 dp- 
vidos οὕνεκα : Ar. Av. 720 φήμη γ᾽ ὑμῖν 
ὄρνις ἐστί, πταρμόν τ᾽ ὄρνιθα καλεῖτε, | 
ξύμβολον ὄρνιν, φωνὴν ὄρνιν, θεράποντ᾽ 
ὄρνιν, ὄνον ὄρνιν. For dat., Schneid. ς 
Hippénax fr. 63 (Bergk) δεξιῷ᾽. ἐλθὼν 
pwd (heron). In Bergk Poet. Lyr. p- 


rh, 


1049 fr. incerti 27 δεξιῇ σίττῃ (woodpecker) 
is a conject. for δεξιὴ σίττη. καὶ is better 
taken as=‘also’ than as ‘both’ (answer- 
ing to καὶ τανῦν in 53) 

54 ἄρξεις.. κρατεῖς... «κρατεῖν. κρατεῖν 
τινός, merely to hold in one’s power; 
ἄρχειν implies a constitutional rule. Cp. 
Plat. Rep. 338 D οὐκοῦν τοῦτο κρατεῖ ἐν 
ἑκάστῃ πόλει, τὸ ἄρχον; Her. 2. 1 ἄλ- 
λους τε παραλαβὼν τῶν ἤρχε καὶ δὴ καὶ 


Ἑλλήνων τῶν ἐπεκράτεε, 1.6. the Asiatics 


who were his lawful subjects, and the 
Greeks over whom he could exert force. 
But here the poet intends no stress on a 
verbal contrast : it is as if he had written, 
εἴπερ ἄρξεις, ὥσπερ ἄρχεις. Cp. 7γαελ. 
457 κεὶ μὲν δέδοικας, οὐ καλῶς ταρβεῖς : 

below 973 προὔλεγον... | ηὔδας. 

55 ξὺν ἀνδράσιν, not ‘ with the help 
of men,’ but ‘with men in the land,’ = ἄν- 
Opas ἐχούσης γῆς. Cp. 207 ξὺν als=as 
ἔχουσα. El. 191 ἀεικεῖ σὺν στολᾷ' Ai. 
30 σὺν νεορράντῳ ξίφει. Ant. 116 ξύν θ᾽ 
ἱπποκόμοις κορύθεσσι. 

56 ὡς οὐδέν ἐστιν «.t.A. Thuc. 7. aT 
ἄνδρες γὰρ πόλις, καὶ οὐ τείχη aN νῆες 
ἀνδρῶν κεναί. Dio Cass. 56. 6 ἄνθρωποι 
γάρ πον πόλις ἐστίν, οὐκ οἰκίαι, κ.τ.λ. 
Her. 8. 61 (Themistocles, taunted by 
Adeimantus after the Persian occupation 
of Athens in 480 B.c. with being ἄπολις, 
retorted) ἑωυτοῖσι.. ως εἴη καὶ πόλις καὶ 
γῆ μέζων ἦπερ κείνοισι, ἔστ᾽ ἂν διηκόσιαι 
νῆές opt ἔωσι πεπληρωμέναι.---πτύργος 
= the city wall with its towers: the sing. 
as below, 1378: Ant. 953 οὐ πύργος, οὐχ 
ἁλίκτυποι | ...vaes: Eur. Hee. 1209 πέριξ 
δὲ πύργος εἶχ᾽ ἔτι πτόλιν. 

57 Lit., ‘void of men, when the do 
not dwell with thee in the city’: ἂν ρῶν 
depends on ἔρημος, of which μὴ ξυνοι- 
κούντων ἔσω is epexegetic. Rhythm and 


2—2 


20 ΣΟΦΟΚΛΕΘῪΣ 


ΟΙ. ὦ παῖδες οἰκτροί, γνωτὰ κοὺκ ἀγνωτά μοι 


προσήλθεθ' ἱμείροντες" 


εὖ γὰρ οἷδ᾽ ὅτι 
νοσεῖτε πάντες, καὶ νοσοῦντες, ὡς ἐγὼ 


60 


οὐκ ἔστιν ὑμών. ὅστις ἐξ ἴ ἴσου νοσεῖ. 

τὸ μὲν γὰρ ὑμῶν ἄλγος εἰς ev ἔρχεται 

μόνον καθ᾽ αὑτόν, κοὐδέν᾽ ἄλλον" n δ᾽ ἐμὴ 

ψυχὴ πόλιν τε κἀμὲ Kal oO ὁμοῦ στένει. 

ὥστ᾽ οὐχ ὕπνῳ γ᾽ evdovTa μ᾽ ἐξεγείρετε, 65 
ἀλλ᾽ ἴστε πολλὰ μέν με δακρύσαντα δή, 

πολλὰς Oo ὁδοὺς ἐλθόντα φροντίδος πλάνοις. 


ΕἾ 
ἣν δ᾽ εὖ σκοπῶν ηὕρισκον ιασιν μόνην, 


ταύτην “ἔπραξα. παῖδα γὰρ Μενοικέως 


Κ ρέοντ᾽, 


ἐμαυτοῦ γαμβρόν, ἐς τὰ Πυθικὰ 70 


ἔπεμψα Φοίβου δώμαθ', ὡς πύθοιθ' ὅ τι 
δρῶν ἢ τί φωνών ἘΣ ῥυσαίμην πόλιν. 


στάντες γ᾽ Triclinius. 


67 πλάνοισ L, but altered from πλάναισ: above is written, 





Sophoclean usage make this better than 
to take ἀνδρῶν μὴ ξυνοικ. @ as a gen. 
absol. Cp. Az. 464 γυμνὸν φανέντα τών 
ἀριστείων ἄτερ: Phil. 31 κενὴν οἴκησιν ἀν- 
θρώπων δίχα : Lucret. 5. 841 muta sine 
ore etiam, sine voltu caeca. 

58 γνωτὰ κοὐκ dyvwta. The empha- 
sis of this formula sometimes appears to 
deprecate an opposite impression in the 
mind of the hearer: ‘known, and not (as 
you perhaps think) unknown. ZZ. 3. 59 
ἐπεί με κατ᾽ αἷσαν ἐνείκεσας οὐδ᾽ ὑπὲρ 
αἷσαν, duly, and not,—as you perhaps 
expect me to say,—unduly. Pier 13.25 
ἐμμανήςτε ἐὼν καὶ οὐ ppevijpns—being nad, 
—for it must be granted that no man in 
his right mind would have acted thus. 
Ο. Ο. 397 βαιοῦ κοὐχὶ μυρίου χρόνου, soon, 
and not after such delay as thy impatience 
might fear. 

60 νοσοῦντες...νοσεῖ. We expected 
καὶ νοσοῦντες οὐ νοσεῖτε, ws ἐγώ. But at 
the words ὡς ἐγώ the speaker’s conscious- 
ness of his own exceeding pain turns him 
abruptly to the strongest form of expres- 
sion that he can find—ovx« ἔστιν ὑμῶν ὅστις 
νοσεῖ, there ἐξ not one of you whose pain is 
as mine. In Plat. Philed. 19 B (quoted 
by Schneid.) the source of the anaco- 
louthon i is the same: μὴ γὰρ δυνάμενοι 
τοῦτο κατὰ παντὸς ἑνὸς καὶ ὁμοίου καὶ ταὐ- 
τοῦ δρᾶν καὶ τοῦ ἐναντίου, ὡς ὁ παρελθὼν 
λόγος ἐμήνυσεν, οὐδεὶς εἰς οὐδὲν οὐ- 


δενὸς ἂν ἡμῶν οὐδέποτε γένοιτο ἄξιος, --- 
instead of the tamer οὐκ ἂν γενοίμεθα. 

62 εἰς ἕνα.. “μόνον καθ᾽ αὑτόν. καθ᾽ 
αὑτόν, ‘by himself’ (0.C. 966), is strictly 
only an emphatic repetition of μόνον : but 
the whole phrase els ἕνα μόνον καθ᾽ αὑτόν 
is virtually equivalent to εἰς ἕνα ἕκαστον 
καθ᾽ αὑτόν, each several one apart from 
the rest. 

64 πόλιν τε κἀμὲ καὶ σ᾽. The king’s 
soul grieves for the whole State,—for 
himself, charged with the care of it,—and 
for each several man (σέ. As the first 
contrast is between public and private 
care, κἀμέ stands between πόλιν and σέ. 
For the elision of σέ, though accented, 
CP. 329 τἄμ᾽, ws av εἴπω μὴ τὰ σ᾽: 404 καὶ 
τὰ ser, 1499 τὰ γοῦν σ᾽: Phil. 339 
οἴμοι μὲν ἀρκεῖν σοί γε καὶ τὰ σ᾽: Eur, 
Hipp. 323 ἔα μ᾽ ἁμαρτεῖν" ob yap és σ᾽ 
ἁμαρτάνω. 

65 The modal dat. ὕπνῳ, more forci- 
ble than a cogn. acc. ὕπνον, nearly = 
‘soundly.? Cp. Ant. 427 γόοισιν ἐξῴ- 

μωξεν: Trach. 176 φόβῳ, φίλαι, ταρβοῦ- 
σαν: [Eur.] fr. 1132 (Nauck?) 40 ὀργῇ 
χολωθείς (where Nauck, rashly, I think, 
conjectures &pyet). Verg. Aen. 1. 680 
sopitum somno. εὕδειν, καθεύδειν (Xen. 
An. τ. 3. 11) oft.=‘to be at ease’ (cp. 
ἔνθ᾽ οὐκ ἂν βρίζοντα ἴδοις, of Agam., Z/. 4. 
223): the addition of ὕπνῳ raises and in- 
vigorates a trite metaphor. 


OIAITOY2 TYPANNOS 51 


ΟΕ. Oh my piteous children, known, well known to me are 
the desires wherewith ye have come: well wot I that ye suffer 
all; yet, sufferers as ye are, there is not one of you whose suffer- 
ing is as mine. Your pain comes on each one of you for himself 
alone, and for no other; but my soul mourns at once for the 
city, and for myself, and for thee. 

So that ye rouse me not, truly, as one sunk in sleep: no, be sure 
that I have wept full many tears, gone many ways in wander- 
ings of thought. And the sole remedy which, well pondering, 
I could find, this I have put into act. I have sent the son of Me- 
noeceus, Creon, mine own wife’s brother, to the Pythian house of 
Phoebus, to learn by what deed or word I might deliver this town. 


ἀντὶ τοῦ πλάναις θηλυκῶς. 


πλάναις τ, but with exceptions: thus T has πλάνοις (with 





67 πλάνοις has excellent manuscript 
authority here; and Soph. uses πλάνου 
Ο. Ο. 1114, πλάνοις Phil, 758, but πλάνη 
nowhere. Aesch. has πλάνη only: Eur. 
πλάνος only, unless the fragment of the 
Rhadamanthus be genuine (659 Nauck?, 
v. 8, οὕτω βίοτος ἀνθρώπων πλάνη). Aris 
stoph. has πλάνος once ( 7725. 872), πλάνη 
never. Plato uses both πλάνη and πλάνος, 
the former oftenest: Isocrates has πλάνος, 
not πλάνη. 

68 ηὕρισκον, ‘could find’ (impf.). 
Attic inscriptions of the 5th or early 4th 
cent. B.C. support the temporal augment 
in the historical tenses of εὑρίσκω (Meis- 
terhans, Gram. Att. Inschr., p. 78). 
Our best Ms. of Soph. (L), however, pre- 
serves no trace of it, except in Ant. 406 
(see cr. n. there). Curtius (Verd. I. 139, 
Eng. tr. 93) thinks that, while the omis- 
sion of the syllabic augment was an ar- 
chaic and poetical license, that of the 
temporal was ‘a sacrifice to convenience 
of articulation, and was more or less 
common to all periods’: so that elxafov 
could exist in Attic by the side of ἥκαζον, 
εὕρισκον by the side of ηὕρισκον. ὴ 

69 ταύτην ἔπραξα, a terse equivalent 
for ταύτῃ ἔργῳ ἐχρησάμην. 

71 £.8 τι δρῶν...τί φωνῶν. Cp. Plat. 
Rep. 414 D οὐκ οἷδα ὁποίᾳ τόλμῃ ἢ 
ποίοις λόγοις χρώμενος épS. These are 
exceptions to the rule that, where an in- 
terrogative pronoun (as ris) and a relative 
(as ée¢rts) are both used in an indirect 
question, the former stands first: cp. Plat. 
Crito 48 A οὐκ ἄρα.. φροντιστέον, τί ἐροῦ- 
διν οἱ πολλοὶ ἡμᾶς, ἀλλ᾽ ὅ τι ὁ ἐπαΐων, 
k.T..: Gorg. 448 E οὐδεὶς ἐρωτᾷ ποία τις 
εἴη ἡ Τοργίου τέχνη, ἀλλὰ τίς, καὶ ὅντινα 


δέοι καλεῖν τὸν Topyiav: 2b. 500 A ἐκλέξ- 
ασθαι ποῖα ἀγαθὰ καὶ ὁποῖα κακά: Phileb. 
17 Β (ἴσμεν) πόσα τέ ἐστι καὶ ὁποῖα.--- 
δρῶν ἢ φωνῶν : there is no definite contrast 
between doing and bidding others to do: 
rather ‘deed’ and ‘word’ represent the 
two chief forms of agency, the phrase 
being equivalent to ‘in what possible 
way.’ Cp. Aesch. P. V. 659 θεοπρόπους 
ἴαλλεν, ws μάθοι τί xph | SpGvr’ ἢ λέ- 
γοντα δαίμοσιν πράσσειν φίλα.---ἠἰῥυσαί- 
μὴν (L’s reading)*is right: ῥυσοίμην is 
grammatically possible, but less fitting. 
The direct deliberative form is τί dpav 
ῥύσωμαι; the indirect, πυνθάνομαι 8 
τι (or τί) δρῶν piowua, ἐπυθόμην 6 
τι (or τί) δρῶν ῥυσαίμην. This indirect 
deliberative occurs, not only with verbs 
of ‘doubting’ (Xen. A. 7. 4. 39 ἠπόρει 
ὅ τι χρήσαιτο τῷ πράγματι), but also with 
verbsof ‘asking’: Thuc. 1. 25 τὸν θεὸν ἐπή- 
ροντο, εἰ παραδοῖεν... τὴν πόλιν (oblique of 
παραδῶμεν τὴν πόλιν). Kennedy wrongly 
says that ῥυσαίμην here could be only the 
oblique of ἐρρυσάμην (as if, in Thuc. /.c., 
παραδοῖεν could be only the oblique of 
παρέδοσαν); and that, for the sense, it 
would require dy. This would also be 
right, but in a different constr., viz., as 
oblique of ri δρῶν ῥυσαίμην dv; Cp. 77>. 
991 ov yao éxw πώς ἂν | στέρξαιμι, and 
Ant. 270 ff. n. In £7. 33 ws μάθοιμ᾽, ὅτῳ 
τρόπῳ πατρὶ | δίκας ἀροίμην, the opt. is 
that of ἠρόμην, being oblique for ἄρωμαι, 
rather than of ἀροῦμαι.---ῥυσοίμην would 
be oblique of τί δρῶν picoua; ῥυσοίμην 
(oblique for ῥύσομαι) would imply that he 
was confident of a successful γέρο, and 
doubtful only concerning the means; it 
is therefore less suitable. 


22 ΣΟΦΘΟΚΑΕΟΥΣ 


καί μ᾽, ἦμαρ ἤδη ξυμμετρούμενον χρόνῳ 
λυπεὶ Tl πράσσει" τοῦ γὰρ εἰκότος “πέρα 
ἄπεστι πλείω τοῦ καθήκοντος χρόνου. 75 
ὅταν δ᾽ ἵκηται, τηνικαῦτ᾽ ἐγὼ κακὸς 
μὴ δρῶν ἂν εἴην πάνθ' ὅσ᾽ ἂν δηλοῖ θεός. 
IE. ἀλλ᾽ εἰς καλὸν σύ τ᾽ εἶπας, οἶδε τ᾽ ἀρτίως 
Κρέοντα προσστείχοντα σημαίνουσί μοι. 
Ol. ὦναξ ἼΛπολλον, εἰ γὰρ ἐν τύχῃ γέ τῳ 80 
σωτῆρι βαΐη λαμπρὸς ὥσπερ ὄμματι. 


IE. ἀλλ᾽ εἰκάσαι μέν, ἡδύς. 


οὐ γὰρ ἂν κάρα 


πολυστεφὴς ὧδ᾽ εἷρπε παγκάρπου “δάφνης. 


Ol. τάχ᾽ εἰσό peo Oa. 


ὕμμετρος γὰρ ὡς κλύειν. 


ἄναξ, ἐμὸν κήδευμα, παῖ Μενοικέως, 85 
τίν ἡμὶν ἥκεις τοῦ θεοῦ φήμην φέρων ; 
ΚΡΕΩΝ. 


ἐσθλήν" λέγω γὰρ καὶ τὰ δύσφορ᾽, εἰ τύχοι 


ΧΩ) κατ᾽ ὀρθὸν ἐξελθόντα, πάντ᾽ ἂν εὐτυχεῖν. 


ais written above), a marginal schol. quoting τοὺς φυγαδικοὺς πλάνους. 
Porson conj. περᾷ, proposing to omit v. 75: 


74 répa L. 


see note. 79 προστείχοντα MSS., 


meaning, however, doubtless, the compound with πρός, not with πρό: cp. on O.C. 


986. - προσστείχοντα Erfurdt. 


87 τὰ δύσθρο᾽ is Heimsoeth’s conj. suggested by the 





73 καί μ᾽ ἦμαρ...χρόνῳ. Lit., ‘and 
already the πὰ compared with the lapse 
of time [since his departure], makes me 
anxious what he doth’: 2.6. when I think 
what day this is, and how many days ago 
he started, I feel anxious. ἤδη, showing 
that to- day is meant, sufficiently defines 
ἦμαρ. χρόνῳ is not for τῷ χρόνῳ, the time 
since he left,—though this is implied,— 
but is abstract, —time in its course. The 
absence of the art. is against our taking 
χρόνῳ as ‘the time which I had allowed 
for his journey.’ ξυμμετρούμενον : cp. 
Her. 4. 158 συμμετρησάμενοι τὴν ὥρην τῆς 
ἡμέρης, νυκτὸς παρῆγον, ‘having calculated 
the time, they led them past the place by 
night’: lit., ‘having compared the season 
of the day (with the distance to be tra- 
versed).’ Eur. Or. 1214. καὶ δὴ πέλας νιν 
δωμάτων εἶναι δοκῶ" ] τοῦ γὰρ χρόνου τὸ 
μῆκος αὐτὸ συντρέχει ‘for the length of 
time (since her departure) just tallies 
(with the time required for the journey).’ 

74 λυπεῖ τί πράσσει: Ai. 794 ὥστε μ᾽ 
ὠδίνειν τί φής. τοῦ γὰρ εἰκότος πέρα. 
τὸ εἰκός is a reasonable estimate of the time 


required for the journey. Thuc. 2. 73 
nuepas...év als εἰκὸς ἣν κομισθῆναι (αὐτούς), 
the number of days which might reason- 
ably be allowed for their journey (from 
Plataea to Athens and back). Porson 
conjectured τοῦ γὰρ εἰκότος περᾷ, as= ‘for 
he overstays the due jimit’—thinking 
Vv. 75, ἄπεστι...χρόνου, to be a spurious 
interpolation. The same idea had oc- 
curred to Bentley. But (1) περᾶν with 
the genitive in this sense is strange (in 
674 θυμοῦ περᾶν is different), and would 
not be readily understood as referring to 
time; (2) it is Sophoclean to explain and 
define τοῦ εἰκότος πέρα by πλείω τοῦ mate 
κοντος χρόνου. 

78 εἰς καλὸν, to fit purpose, ‘ oppor- 
tunely’: Plat. Symp. 174 E els καλὸν 
ἥκεις. Az. 1168 καὶ μὴν ἐς αὐτὸν καιρὸν 

. | πάρεισιν. Cp. Ar. Ach. 686 els τά- 
xos = ταχέως, Av. 805 els εὐτέλειαν Ξε εὐ- 
τελώς. οἵδε: some of those suppliants 
who are nearer to the stage entrance on 
the spectators’ left—the conventional one 
for an arrival from the country—have 
made signs to the Priest. Creon enters, 


OIAITOY2 TYPANNOS 23 


And already, when the lapse of days is reckoned, it troubles 
me what he doth; for he tarries strangely, beyond the fitting 


space. 
not all that the god shows. 
PR. 


But when he comes, then shall I be no true man if I do 


Nay, in season hast thou spoken; at this moment these 


sign to me that Creon draws near. 
Or. O king Apollo, may he come to us in the brightness of 
saving fortune, even as his face is bright ! 


Pir 


Nay, to all seeming, he brings comfort; else would he 


not be coming crowned thus thickly with berry-laden bay. 


OE. 


We shall know soon: he is at range to hear.—Prince, 


my kinsman, son of Menoeceus, what news hast thou brought us 


from the god? 


CREON. 


Good news: I tell thee that even troubles hard to bear,—if 
haply they find the right issue,—will end in perfect peace. 


schol., λέγω γὰρ πάντα ἂν εὐτυχεῖν τὴν πόλιν, εἰ Kal τὰ δύσ φημα τύχοι [ἂν] κατ᾽ ὀρθὸν 


ἐξελθόντα. 


But the schol. uses that word only to illustrate his own comment on 


ἐσθλήν : ἀπὸ yap τῶν εὐφήμων ἄρξασθαι θέλει, and clearly read δύσῴφορ᾽, which is in the 


lemma of another schol. 


88 ἐξελθόντα MSS. ἐξιόντα Suidas and Zonaras 5.ν. 





wearing a wreath of bay leaves bright 
with berries, in token of a favourable 
answer. See Appendix, Note 1, § 2. 

80 ξ. ἐν TUX Y...dppate: may his radiant 
look prove the herald of good news. 
λαμπρὸς with ἐν τύχῃ «.7.d.,—being ap- 
plicable at once to dril/iant fortune and 
(in the sense of φαιδρός) to a beaming 
countenance. ἐν τύχῃ, nearly = μετὰ 
τύχης, ‘invested with,’ ‘attended by’: 
cp. 1112 ἔν τε γὰρ μακρῷ | γήρᾳ ξυνάδει: 
Ai. 488 σθένοντος ἐν πλούτῳ. τύχη σωτὴρ 
(Aesch. Ag. 664), like χεὶρ πράκτωρ (26. 
111), θέλκτωρ πειθώ (Aesch. Suppl. 1040), 
καρανιστῆρες δίκαι (Zu. 186). 

82 εἰκάσαι μέν, ἡδύς (sc. βαίνει). Cp. 
El. 410 ἐκ δείματός του νυκτέρου, δοκεῖν 
ἐμοί. Ο. Ο. 151 δυσαίων ] μακραίων 7’, 
ἐπεικάσαι. ἡδύς, not ‘joyous,’ but 
‘pleasant to us,’ ‘bringing good news’: 
as 510 ἡδύπολις, pleasant to the city: £7. 
929 ἡδὺς οὐδὲ μητρὶ δυσχερής, a guest 
welcome, not grievous, to her. In 7rach. 
869 where ἀηδὴς καὶ συνωφρυωμένη is said 
of one who approaches with bad news, 
ἀηδής is not ‘unwelcome,’ but rather 
‘sullen,’ ‘gloomy.’ 

88 πολυστεφὴς.. δάφνης. The use 
of the gen. after words denoting fulness 
is extended to the notions of encompas- 
sing or overshadowing: e.g. περιστεφῆ]ἁ 


. avdéwy θήκην (El. 895), στέγην... «ἧς [v. 
δ. ἢ] κατηρεφεῖς δόμοι (Eur. Hipp. 468). 
But the dat. would also stand: cp. Od. 
9. 183 σπέος.. .δάφνῃσι κατηρεφές : Hes. Of. 
513 λάχνῃ δέρμα κατάσκιον. παγκάρπου, 
covered with berries: cp. O.C. 676. 
Plin. 15. 30 maximis baccis atque e viridi 
rubentibus (of the Delphic laurel). The 
wreath announces good news, 77. 179: 
so in Eur. App. 806 Theseus, returning 
from the oracle at Delphi to find Phaedra 
dead, cries ti δῆτα τοῖσδ᾽ ἀνέστεμμαι 
κάρα | πλεκτοῖσι φύλλοις, δυστυχὴς θεωρὸς 
ὧν; So Fabius Pictor returned from 
Delphi to Rome coronatus laurea corona 
(Liv. 23. 11). 

84 ξύμμετρος γὰρ ὡς κλύειν. He is 
at a just distance for hearing: ξύμμετρος 
=commensurate (in respect of his dis- 
tance) wth the range of our voices (im- 
plied in xkAvew). 

85 κήδευμα, ‘kinsman’ (by marriage), 
Ξεκηδεστής, here=yauBpds (70). Ant. 
756 γυναικὸς ὧν δούλευμα μὴ KWTLArE με. 
Eur. Or. 928 τἄνδον οἰκουρήματα-ε τὰς 
ἔνδον οἰκουρούσαξ. 

87 f. λέγω γὰρ... εὐτυχεῖν. Creon, 
unwilling to speak plainly before the 
Chorus, hints to Oedipus that he brings 
a clue to the means by which the anger 
of heaven may be appeased. ἐξελθόντα, 


24 ΣΟΦΟΚΛΕΟΥΣ 


ΟΙ. ἔστιν δὲ ποῖον τοὔπος ; οὔτε γὰρ θρασὺς 
οὔτ᾽ οὖν προδείσας εἰμὶ τῷ γε νῦν λόγῳ. 9O 
KP. εἰ τῶνδε χρήζεις πλησιαζόντων κλύειν, 
ἕτοιμος εἰπεῖν, εἴτε καὶ στείχειν ἔσω. 
Ol. ἐς πάντας avda. τῶνδε γὰρ πλέον φέρω 
τὸ πένθος ie καὶ τῆς ἐμῆς ψυχῆς πέρι. 
ΚΡ. λέγοιμ᾽ ἂν ot ἤκουσα τοῦ θεοῦ πάρα. 95 
ἄνωγεν ἡμᾶς Φοῖβος ἐμφανῶς ἄναξ 
μίασμα χώρας, ὡς τεθραμμένον χθονὶ 
ἐν τῇδ᾽, ἐλαύνειν, μηδ᾽ ἀνήκεστον τρέφειν. 
ΟΙ. ποίῳ abana: τίς ὁ τρόπος τῆς ξυμφοράς; 
ΚΡ. ἀνδρηλατοῦντας, ἢ φόνῳ φόνον πάλιν ΙΟΟ 
λύοντας, ὡς τόδ᾽ αἷμα χειμάζον πόλιν. 
ΟΙ. ποίου γὰρ ἀνδρὸς τήνδε μηνύει τύχην; 
KP ἦν ἡμίν, ὦναξ, Λάϊός ποθ᾽ ἡγεμὼν 


γῆς τῆσδε, πρὶν σὲ τήνδ᾽ ἀπευθύνειν πόλιν. 


δύσφορα, probably by a mere error. 


99 τρόπος] πόρος conj. F. W. Schmidt. 
101 χειμάζον L, with εἰ written over ov. 


The εἰ may be from the rst hand, as 





of the event, ‘having issued’; cp. ΙΟΤῚ 
μή μοι Φοῖβος ἐξέλθῃ σαφής; so 1182 ἐξή- 
κοι. The word is chosen by Creon with 
veiled reference to the duty of danishing 
the defiling presence (98 ἐλαύνειν). πάν- 
τα predicative with εὐτυχεῖν, ‘will all of 
them (= altogether) be well.’ eyo εὐ- 
τυχεῖν ἀν-Ξ λέγω ὅ ὅτι εὐτυχοίη ἄν. 

89 ξ. τοὔπος, the actual oracle (τοὔπος 
τὸ θεοπρόπον, 77. 822): λόγῳ (go), Creon’s 
own saying (λέγω, 87). προδείσας, a- 
larmed beforehand. Cp. Her. 7+ 50 κρέσ- 
σον δὲ πάντα θαρσέοντα ἢ ἥμισυ τῶν δέιν ὧν 
_ πάσχειν μᾶλλον ἢ πᾶν χρῆμα προδειμαΐί- 
νοντα μηδαμὰ μηδὲν παθεῖν. No other 
part of προδείδω occurs: προταρβεῖν, προ- 
φοβεῖσθαι -Ξ- ‘to fear beforehand,’ but 
ὑπερδέδοικά σου, I fear for thee, Ant. 82. 
In compos. with a verb of caring for, 
however, πρό sometimes=vrép, ¢.g. προ- 
κήδομαι Ant. 741. 

91 f. πλησιαζόντων ΠεΙε-ε- πλησίον 
ὄντων : usu. the verb=either (1) to ap- 
proach, or (2) to consort with (dat.), as 
below, 1136. εἴτε---καὶ στείχειν ἔσω 
(xpnges), (ἕτοιμός εἰμι τοῦτο δρᾶν). So 
Eur. Jom 1120 (quoted by Elms., etc.) 
πεπυσμέναι γάρ, el θανεῖν ἡμᾶς χρεών, 
| ἥδιον dv θάνοιμεν, εἴθ᾽ ὁρᾶν φάος: 2.6. 
εἴτε ὁρᾶν φάος (χρή), (ἥδιον ἄν ὁρῷμεν 


αὐτό). εἰ....εἴτε, as Aesch. Zum. 468 σὺ 
δ᾽, εἰ δικαίως εἴτε μή, κρῖνον δίκην. 

98 f. ἐς πάντας. Her. 8. 26 οὔτε 
ἠνέσχετο σιγῶν εἶπέ τε ἐς πάντας τάδε: 
Thuc. 1. 72 ἐς τὸ πλῆθος εἰπεῖν (before the 
assembly). πλέον adverbial, as in Az, 
1101, etc.: schol. περὶ τούτων πλέον 
ἀγωνίζομαι ἢ περὶ τῆς ἐμαυτοῦ ψυχῆς. 
—tevde, object. gen. with τὸ πένθος 
(not with περί): cp. £7, 1097 τᾷ Ζηνὸς 
εὐσεβείᾳ..----ἢ καὶ, ‘than even.’ This must 
not be confounded with the occasional 
use of ἢ καί in megative sentences con- 
taining a comparison: ¢.¢. Az. 1103 οὐκ 
ἔσθ᾽ ὅπου σοὶ τόνδε κοσμῆσαι πλέον | ἀρχῆς 
ἔκειτο θεσμὸς ἢ καὶ τῷδε σέ: El. 1145 
οὔτε γάρ ποτε | μητρὸς σύ γ᾽ ἦσθα μᾶλλον 
ἢ κἀμοῦ φίλος: Antiphon de caed. Her. 
§ 23 ésnretro οὐδέν τι μᾶλλον ὑπὸ τῶν 
ἄλλων ἢ καὶ ὑπ᾽ ἐμοῦ (where καί. ἴα. re- 
dundant, = ‘on my part’). 

95 λέγοιμ᾽ ἂν, a deferential form, 
having regard to the permission just 
given. Cp. Phil. 674 χωροῖς ἂν εἴσω: 
£i. 637 κλύοις ἂν ἤδη. 

97 ὡς marks that the partic. τεθραμ- 
μένον expresses the view held by the 
subject of the leading verb (ἄνωγεν): 2.¢., 
‘as having been harboured’=‘ which (he 
says) has been harboured.’ Cp. Xen. 


OIAITOY2 TYPANNOS 25 


OE. 
neither bold nor yet afraid. 
CR. 
to speak ; or else to go within. 
OE. 
more than for mine own life. 


CR. 


But what is the oracle? 


So far, thy words make me 


If thou wouldest hear while these are nigh, I am ready 
Speak before all: the sorrow which I bear is for these 


With thy leave, I will tell what I heard from the god. 


Phoebus our lord bids us plainly to drive out a defiling thing, 
which (he saith) hath been harboured in this land, and not to 
harbour it, so that it cannot be healed. 


OE. 


manner of the misfortune ? 


By what rite shall we cleanse us? 


What is the 


Cr. By banishing a man, or by bloodshed in quittance of 
bloodshed, since it is that blood which brings the tempest on our 


city. 
OE. 


And who is the man whose fate he thus reveals! 


Cr. Laius, king, was lord of our land before thou wast pilot 


of this State. 


Diibner thinks: but there is room for doubting whether it was not due to the dcopAw- 


τής or first corrector (S). 


A, and other of the later Mss., have χειμάζον : and χειμάζει, 





Am. τ. 2. 1 ἔλεγε θαρρεῖν ws καταστησο- 
μένων τούτων eis τὸ δέον: he said, ‘Take 
courage, 77 the assurance that’ &c. 

98 ἐλαύνειν for ἐξελαύνειν was regular 
in this context: Thuc. 1. 126 τὸ ἄγος 
ἐλαύνειν τῆς θεοῦ (1.6. to banish the Alc- 
maeonidae): and so I. 127, 128, 135, 
2. 13.—pyd’ ἀνήκεστον τρέφειν. The 
μίασμα is ἀνήκεστον in the sense that it 
cannot be healed dy anything else than 
the death or banishment of the blood- 
guilty. But it can still be healed if that 
expiation is made. Thus ἀνήκεστον is a 
proleptic predicate: cp. Plat. Rep. 565 
τοῦτον τρέφειν τε καὶ αὔξειν μέγαν : O.C. 
g27n. See Antiphon 727. T. y. § 7 
ἀντὶ τοῦ παθόντος (in the cause of the 
dead) ἐπισκήπτομεν ὑμῖν τῷ τούτου φόνῳ 
τὸ μήνιμα τῶν ἀλιτηρίων ἀκεσαμέ- 
vous πᾶσαν τὴν πόλιν καθαρὰν Tod μι- 
doparos καταστῆσαι, ‘to heal with this 
man’s blood the deed which angers the 
avenging spirits, and so to purge the 
whole city of the defilement.’ 

99 ποίῳ.. ξυμφορᾶς. By what puri- 
fying rite (does he command us ἐλαύνειν 
τὸ μίασμα) What is the manner of our 
misfortune (1.6. our defilement)? Eur. 
Phoen. 390 tis ὁ τρόπος αὐτοῦ; τί φυ- 
γάσιν τὸ δυσχερές; ‘what is the manner 
thereof? (sc. τοῦ κακοῦ, exile). ξυμφο- 
pas, euphemistic for guilt, as Plat. Legg. 


9348 λωφῆσαι πολλὰ μέρη THs τοιαύτης 
ξυμφορᾶς, to be healed in great measure 
of such a malady (viz., of evil-doing): 
ib. 8564 Ὁ ἐν τῷ προσώπῳ Kal ταῖς χερσὶ 
γραφεὶς τὴν ξυμφοράν, ‘with his mzsfortune 
[the crime of sacrilege] branded on his 
face and hands.’ Her. 1. 35 συμφορῇ 
ἐχόμενος Ξεἐναγής, under a ban. Prof. 
Kennedy understands: ‘what is the mode 
of compliance (with the oracle)?’ He 
compares O.C. 641 τῇδε yap ξυνοίσομαι 
(‘for with that choice I will comply’). 
But elsewhere, at least, συμφορά does not 
occur in a sense parallel with συμφέ- 
ρεσθαι, ‘to agree with.’ 

100 f£. ἀνδρηλατοῦντας. As if, in- 
stead of ποίῳ καθαρμῷ, the question had 
been τί ποιοῦντας ;—os τόδ᾽ αἷμα χει- 
μάζον πόλιν, since it is this blood [τόδε, 
viz. that implied in φόνον] which brings 
the storm on Thebes. χειμάζον, acc. 
absol. os presents the fact as the ground 
of belief on which the Thebans are com- 
manded to act: ‘Do thus, assured that it 
is this blood,’ etc. Cp. O.C. 380: Xen. 
Hellen. 2. 4. 1 of δὲ τριάκοντα, ws ἐξὸν 
ἤδη αὐτοῖς τυραννεῖν ἀδεῶς, προεῖπον, Κιτ.λ. 
Cp. Eur. Suppl. 268 πόλις δὲ πρὸς πόλιν | 
ἔπτηξε χειμασθεῖσα, ‘city with city seeks 
shelter, when vexed by storms.’ 

104 ἀπευθύνειν, to steer in a right 
course. Theinfin. is of the imperf., = πρό- 


26 ZOPO KAEOY2 


ov \ 
Ol. ἔξοιδ᾽ ἀκούων: ov yap εἰσεῖδόν γέ πω.- 105 
GP. τούτου θανόντος νῦν ἐπιστέλλει σαφῶς 
τοὺς αὐτοέντας χειρὶ τιμωρεῖν τινας. Χ 
Ol; οἱ δ᾽ εἰσὶ ποῦ γῆς; ποῦ τόδ᾽ εὑρεθήσεται 
ἴχνος παλαιᾶς δυστέκμαρτον αἰτίας ; 
ΚΡ. ἐν τῇδ᾽ ἔφασκε γῇ. τὸ δὲ ζητούμενον 110 
ἁλωτόν, ἐκφεύγει δὲ τἀμελούμενον. 
OI. πότερα δ᾽ ἐν οἴκοις ἢ ᾽ν ἀγροῖς ὁ Λαΐος 
ἢ γῆς ἐπ᾽ ἄλλης τῷδε συμπίπτει φόνῳ; 
ΚΡ. θεωρός, ὡς ἔφασκεν, ἐκδημῶν πάλιν 


\ > ΓᾺ δὴ Ψ > ε 9 , 
προς οἶκον οὐκέθ᾽ ἵκεθ᾽, ὡς ἀπεστάλη. 


115 


OI. οὐδ᾽ ἀγγελός τις οὐδὲ συμπράκτωρ ὁδοῦ 


KATELO , 


KE. 


OI.’ τὸ ποῖον; 


ἕν γὰρ πόλλ᾽ ἂν ἐξεύροι μαθεῖν, 


ν 3 Ν > ΄, br ead 
ὅτου τις ἐκμαθὼν ἐχρήσατ᾽ αν; 
θνήσκουσι γάρ, πλὴν εἷς τις, ὃς φόβῳ φυγὼν 


ὧν εἶδε πλὴν € ἕν οὐδὲν εἶ 


᾿ εἰδὼς φράσαι. 


ἀρχὴν βραχεῖαν εἰ λάβοιμεν ἐλπίδος. 


found in a few later MSS., seems to have been merely a conjecture. 
The scribe placed a dot over σ, to indicate that it should be deleted; 


without accent. 


107 τινασ L, 


but this dot was afterwards almost erased, whether by his own hand or by another. 


τινασ Or τινὰσ YX. 


The reading τινά seems to occur in no MS., but only in the Milan 





τερον ἢ ἀπηύθυνες, before you were steer- 
ing (began to steer). Oedipus took the 
State out of angry waters into smooth : 
cp. 696 ἐμὰν γᾶν φίλαν | ἐν πόνοις ἀλύου- 
σαν κατ᾽ ὀρθὸν οὔρισας: fr. 151 πλήκτροις 
ἀπευθύνουσιν οὐρίαν τρόπιν, ‘with the helm 
(πλῆκτρα, the blades of the πηδάλια) they 
steer their bark before the breeze.’ 

105 ov γὰρ εἰσεῖδόν yé mw. As Oecd. 
knows that Laius is dead, the tone of un- 
‘ concern given by this colloquial use of 
οὔπω (instead of οὔποτε) is a skilful touch. 
Cp. £1. 402 XP. σὺ δ᾽ οὐχὶ πείσει...; EA. 
ov δῆτα᾽ μήπω νοῦ τοσόνδ᾽ εἴην κενή: Eur. 
Hec. 1278 μήπω pavein Tuvdapls τοσόνδε 
mais: 71. 12. 270 ἀλλ᾽ οὔπω πάντες ὁμοῖοι] 
ἀνέρες ἐν πολέμῳ: cp. our (ironical) ‘I 
have yet to learn.’ 

107 τοὺς αὐτοέντας...τινας. τούς im- 
plies that the death 4ad human authors; 
τινας, that they are unknown. So in 
O.C. 290 ὅταν δ᾽ ὁ κύριος. | παρῇ τις, 
‘the master—whoever he be.’ τιμωρεῖν, 

‘punish.’ The act., no less than the 
midd., is thus used even in prose: Lysias 
In Agor.§ 42 τιμωρεῖν ὑπὲρ αὑτοῦ ws φονέα 
fa. to punish (Agoratus), on his own 


account, as his murderer. χειρὶ τιμω- 
ρεῖν, here, either ‘to slay’ or ‘to expel by 
force,’ as distinguished from merely fining 
or dis franchising : in 140 τοιαύτῃ χειρὶ 
τιμωρεῖν 15 explained by κτανὼν in £39. 

108 f. ποῦ 768’...airlas; τόδε ἴχνος 
αἰτίας =ixvos τῆσδε αἰτίας, cp. τοὐμὸν φρε- 
νῶν ὄνειρον El. 1390. αἰτίας, ‘crime’: 
Ai. 28 τήνδ᾽ οὖν ἐκείνῳ πᾶς τις αἰτίαν 
νέμει. For δυστέκμαρτον, hard to track, 
cp. Aesch. Zum. 244 (the Furies hunting 
Orestes) elev’ τόδ᾽ ἐστὶ τἀνδρὸς éxpaves 
τέκμαρ. The poet hints a reason for 
what might else have seemed strange— 
the previous inaction of Oedipus. Cp. 219. 

110 ἔφασκε, sc. ὁ θεὸς (εὑρεθήσεσθαι 
τὸ ἴχνος). τὸ δὲ ζητούμενον: δὲ -has a 
sententious force,=‘now.’ The γνώμη, 
though uttered in an oracular tone, is not 
part of the god’s message. Cp. Eur. fr. 435 
αὐτός τι viv δρῶν εἶτα δαίμονας κάλει" | τῷ 
γὰρ πονοῦντι καὶ θεὸς συλλαμβάνει. 

118 συμπίπτει. The vivid historic 
present suits the alertness of a mind 
roused to close inquiry: so below, 118, 
716, 1025: 77. 748: Zl. 679.—Cp. Ai. 


429 Kako-s τοιοῖσδε συμπεπτωκότα. 


ΘΟΙΑΙΤΟΥΣ ΤΥΡΆΝΝΟΣ 27, 


OR: 
(ΕΣ 


He was slain; 


I know it well—by hearsay, for I saw him never. 
and the god now bids us plainly to 


wreak vengeance on his murderers—whosoever they be. 


OE. 


And where are they upon the earth? 


Where shall the 


dim track of this old crime be found? 


CR. 


In this land,—said the god. What is sought for can be 


caught; only that which is not watched escapes. 
Or. And was it in the house, or in the field, or on strange 
soil that Laius met this bloody end? 


GR. 


*Twas on a visit to Delphi, as he said, that he had left 


our land; and he came home no more, after he had once set forth. 


OE. 


And was there none to tell ? 


Was there no comrade 


of his journey who saw the deed, from whom tidings might have 


been gained, and used ? 


Cr. All perished, save one who fled in fear, and could tell 
for certain but one thing of all that he saw. 


ΘῈ. And what was that? 


One thing might show the clue 


to many, could we get but a small beginning for hope. 


ed. of Suidas (ed. Demetrius Chalcondylas, 1498 A.D.), the other editions of Suidas 


giving τινάς (s. v. ἐπιστέλλει). 


117 The rst hand in L wrote émov, which has 
been altered to ὅτου, perhaps by the first corrector. 


[I had doubted this; but in the 





114 θεωρός: Laius was going to 
Delphi in order to ask Apollo whether 
the child (Oedipus), formerly exposed 
by the god’s command, had indeed 
perished: Eur. Phoen. 36 τὸν ἐκτεθέντα 
παῖδα μαστεύων μαθεῖν | εἰ μηκέτ' εἴη. ὡς 
ἔφασκεν, as Laius told the Thebans at 
the time when he was leaving Thebes. 
ἐκδημῶν, not goizg abroad, but deing 
[=having gone] abroad: cp. Plat. Lege. 
864E οἰκείτω τὸν ἐνιαυτὸν ἐκδημῶν. ὡς 
Ξξ ἐπεί : Xen. Cyr. 1. 3. 2 ὡς δὲ ἀφίκετο 
τάχιστα... ἠσπάζετο. Cic. Brut. 5 ut zllos 
libros edidisti, nihil a te postea accepimus. 

116 οὐδ᾽ ἄγγελος... ἐχρήσατ᾽ av; The 
sentence begins as if ἄγγελός τις were to 
be followed by ἦλθε: but the second 
alternative, συμπράκτωρ ὁδοῦ, suggests 
κατεῖδε [had seex, though he did not 
speak]; and this, by a kind of zeugma, 
stands as verb to ἄγγελος also. Cp. Her. 
4. 106 ἐσθῆτα δὲ φορέουσι τῇ Σκυθικῇ 
ὁμοίην, γλῶσσαν δὲ ἰδίην. οὐδ᾽ ἄγγελος : 
Jl. 12. 73 οὐκέτ᾽ ἔπειτ᾽ ὀΐω οὐδ᾽ ἄγγελον 
ἀπονέεσθαι. ὅτου, gen. masc.: from 
whom having gained knowledge one 
might have used it. 

117 expady=a protasis, εἰ ἐξέμαθεν, 
ἐχρήσατ᾽ ἄν, sc. τούτοις ἃ ἐξέμαθεν. Plat. 
Gorg. 465 E ἐὰν μὲν οὖν καὶ ἐγὼ σοῦ ἀπο- 
κρινομένου μὴ ἔχω 6 τι χρήσωμαι, if, when 


.Μ8. fluctuates. 


you answer, I also do not know what use 
to make [of your answer, sc. τούτοις ἃ ἂν 
dmoxplvy),—where shortly before we have 
οὐδὲ χρῆσθαι TH ἀποκρίσει ἣν σοι ἀπεκρι- 
νάμην οὐδὲν οἷός 7’ ἦσθα. 

118 2. θνήσκουσι. Thee subscript in 
the pres. stem of this verb is attested by 
Attic inscriptions (Meisterhans, Gram. p. 
86). The practice of the Laurentian 
It gives the ¢ subscript 
here, in. 023; 1157: 0: C. 611; Ant. 847: 
761; Zl. 1022. It omits the ¢ subscript 
in. 27. 63; 153, 640; 14865: 77.707, 708; 
Ph. 1085. Cp. Ztym. M. 482, 29, θνῇ- 
σκω, μιμνῃσκω. Δίδυμος [czrc. 30 B.C.] 
χωρὶς τοῦ 7...7) μέντοι παράδοσις ἔχει τὸ L.— 
φόβῳ φυγὼν, ‘having fled in fear’: φόβῳ, 
modal dative; cp. Thuc. 4. 88 dia te τὸ 
ἐπαγωγὰ εἰπεῖν τὸν Βρασίδαν καὶ περὶ τοῦ 
καρποῦ φόβῳ ἔγνωσαν : 5. 70 ἐντόνως καὶ 
ὀργῇ χωροῦντες.---εἰδὼς, with sure know- 
ledge (and not merely from confused 
recollection, dcapys δόξα): so 1151 λέγει 
γὰρ εἰδὼς οὐδὲν ἀλλ᾽ ἄλλως πονεῖ: 5. 
41 ὅπως ἂν εἰδὼς ἡμὶν ἀγγείλῃς σαφῆ. 
Ιοσαβία says (849), in reference to this 
same point in the man’s testimony, κοὐκ 
ἔστιν αὐτῷ τοῦτό ἼΑ oe πάλιν. 


120 τὸ ποῖον; Cp. 201: El. 670 
πρᾶγμα πορσύνων μέγα. | KA. τὸ ποῖον, 


ὦ ξέν᾽; εἶπέ. Ar. Pax 696 εὐδαιμονεῖ. 


28 ΣΟΦΟΚΛΕΘῪΣ 


λῃστὰς ἔφασκε συντυχόντας οὐ μιᾷ 


ῥώμῃ κτανεῖν νιν, ἀλλὰ σὺν πλήθει χερών. 


ἐπράσσετ᾽ ἐνθένδ᾽ ; 


πῶς οὖν ὁ λῃστής, εἴ τι μὴ 
ἐς τόδ᾽ ἂν τόλμης ἔβη; 
δοκοῦντα ταῦτ᾽ ἦν: Λαΐου δ᾽ ὀλωλότος 


ξὺν ἀργύρῳ 


3 Ν 3 \ > a Sia 
οὐδεὶς αρωγος €V κακοις ἐγίγνετο. 


Ol. 


Ν \ A Ε] \ / 
κακὸν δὲ ποῖον ἐμποδὼν τυραννίδος 


Ψ ΄ Ss ΨᾺΣ 1.8). τε 5 ΄, 
οὕτω πεσούσης εἶργε τοῦτ᾽ ἐξειδέναι ; 


KP. 


ἡ ποικιλῳδὸς Σφὶγξ τὸ πρὸς ποσὶ σκοπεῖν 


120 


μεθέντας ἡμᾶς Tapavy, «προσήγετο. 


ΟΙ. 


ἀλλ᾽ ἐξ ὑπαρχῆς Aes auT ἐγὼ pave. 
ἐπαξίως yap Φοῖβος, ἀξίως δὲ σὺ 

πρὸ τοῦ θανόντος τήνδ᾽ ἔθεσθ' ἐπιστροφήν: 1 
ὥστ᾽ ἐνδίκως ὄψεσθε κἀμὲ σύμμαχον, 

γῇ τῇδε τιμωροῦντα τῷ θεῴ 


wt 


ἅμα. 


ὑπὲρ γὰρ οὐχὶ τῶν ἀπωτέρω φίλων 


autotype facsimile of L the original 7 is clear. ] 


ὅτου τ. 134 πρὸ Toi L. The τϑί 


hand had written πρὸ στοῦ, separating the σ (as he often does) from the syllable to 


which it belonged, and forming στ in one character; 


the corrector erased the o. 





πάσχει δὲ θαυμαστόν. ‘EPM. τὸ τί; ἐξεύ- 
po. μαθεῖν. One thing would find out 
how to learn many things, z.e. would 
prove aclue to them. The infin. μαθεῖν 
as after a verb of teaching or devising: 
Her. 1. 196 ἄλλο δέ τι ἐξευρήκασι νεωστὶ 
γενέσθαι. Plat. Rep. 5108 ἐν ὅλῃ τῇ 
πόλει τοῦτο μηχανᾶται ἐγγενέσθαι. 

122 f. ἔφασκε σε. ὁ φυγών (118). οὐ 
μιᾷ ῥώμῃ:Ξ οὐχ ἑνὸς ῥώμῃ, in the strength 
not of one man. Cp. Her. 1. 174 πολλῇ 
χειρὶ ἐργαζομένων τῶν Κνιδίων. Ant. 14 
. διπλῇ xept=by the hands of twain. So 
perh. χερὶ διδύμᾳ Pind. Pyth. 2. 9.---αὖν 
πλήθει: cp. on 5 58: 

124f. εἴ τι μὴ κιτιλ., if some intrigue, 
aided by (ξὺν) money, had not been 
working from Thebes. tt is subject to 
ἔπράσσετο: distinguish the adverbial τί 
(=‘perchance’) which is often joined to 
el μή in diffident expressions, as 969 εἴ τι 
μὴ τὠμῷ πόθῳ | κατέφθιτ᾽, ‘unless 247- 
chance’: so O.C. 1450, 77. 586 etc. 
Schneid. cp. Thuc. 1. 121 καί τι αὐτῷ καὶ 
ἐπράσσετο és Tas πόλεις ταύτας προδοσίας 
πέρι: and 5. 82 ὑπῆρχε δέ τι αὐτοῖς καὶ ἐκ 
τοῦ "Αργους αὐτόθεν πρασσόμενον .---ἔπράσ- 
σετο.. ἔβη: the imperf. refers here to a 
continued act in past time, the aor. to an 


act done at a definite past moment. Cp. 
402 ἐδόκεις---ἔγνως : 432 ἱκόμην---ἐκάλεις. 

126 δοκοῦντα...ἦν expresses the vivid 
presence of the δόξα more strongly than 
ταῦτα ἐδόκει would have done (cp. 274 
τάδ᾽ ἔστ᾽ ἀρέσκονθ᾽): Her. 1. 146 ταῦτα δὲ 
ἦν γινόμενα ἐν Μιλήτῳ. 

128 ἐμποδὼν sc. ὄν, with κακὸν, not 
with εἶργε, ‘what trouble (being) i in your 
path?’ Cp. 445 παρὼν... ἐμποδὼν | ὀχλεῖς. 
τυραννίδος. Soph. conceives the Theban 
throne as having been vacant from the 
death of Laius—who left no heir—till the 
election of Oed. The abstract τυραννίδος 
suits the train of thought on which Oed. 
has already entered,—viz. that the crime 
was the work of a Theban faction (124) 
who wished to destroy, not the king 
merely, but the kingship. Cp. Aesch. 
Cho. 973 ἴδεσθε χώρας τὴν διπλῆν τυραν- 
vida (Clytaemnestra and Aegisthus). 

130 ποικιλῳδὸς, singing ποικίλα, sud- 
tleties, αἰνίγματα: cp. Plat. Symp. 1824 
ὁ περὶ τὸν ἔρωτα νόμος ἐν μὲν ταῖς ἄλλαις 
πόλεσι νοῆσαι ῥάδιος" ἁπλῶς γὰρ w- 
ρισται" ὁ δὲ ἐνθάδε καὶ ἐν Λακεδαίμονι 
ποικίλος. Her. 7. 111 πρόμαντις δὲ ἡ 
χρέουσα, κατάπερ ἐν Δελφοῖσι, καὶ οὐδὲν 
ποικιλώτερον, ‘the chief prophetess is she 


ΟἸΙΔΙΠΟῪΣ TYPANNOS 29 


Cr. He said that robbers met and fell on them, not in one 
man’s might, but with full many hands, 

Or. How, then, unless there was some trafficking in bribes 
from here, should the robber have dared thus far ? 

Cr. Such things were surmised; but, Lafus once slain, amid 
our troubles no avenger arose. 

Or. But, when royalty had fallen thus, what trouble in your 
path can have hindered a full search? 

Cr. The riddling Sphinx had made us let dark things go, 
and was inviting us to think of what lay at our doors. 

OE. Nay, I will start afresh, and once more make dark things 
plain. Right worthily hath Phoebus, and worthily hast thou, be- 
stowed this care on the cause of the dead; and so, as is meet, 
ye shall find me too leagued with you in seeking vengeance for 
this land, and for the god besides. On behalf of no far-off friend, 


Among the later mss., A and a few more have πρὸ (sometimes with the gloss ὑπὲρ): 
others have πρός.---τήνδ᾽ ἔθεσθ᾽ ἐπιστροφήν] A variant recorded in the margin of L, 
τήνδε θεσπίζει γραφήν, is instructive, as indicating the lengths to which arbitrary 





who gives the oracles, as at Delphi, and 
in no wise of darker speech.’ 

131 The constr. is προσήγετο ἡμᾶς, 
μεθέντας τὰ ἀφανῆ, σκοπεῖν TO πρὸς ποσί. 
προσήγετο, was drawing us (by her dread 
song), said with a certain irony, since 

- προσάγεσθαι with infin. usually implies 
a gentle constraint (though, as a milit. 
term, ἀνάγκῃ προσηγάγοντο, reduced by 
force, Her. 6. 25): cp. Eur. 7071 659 χρόνῳ 
δὲ καιρὸν λαμβάνων προσάξομαι | δάμαρτ᾽ 
ἐᾶν σε σκῆπτρα τἄμ᾽ ἔχειν χθονός. τὸ πρὸς 
ποσὶ (cp. ἐμποδὼν 128), the zzstant, 
pressing trouble, opp. to τὰ ἀφανῆ, ob- 
scure questions (as to the death of Laius) 
of no present or practical interest. Pind. 
Isthm. 7. 12 δεῖμα μὲν waporxdpmevor | 
καρτερὰν ἔπαυσε pépiyuvav' τὸ δὲ πρὸς 
ποδὸς ἄρειον ἀεὶ σκοπεῖν | χρῆμα πᾶν. 
Ant. 1327 τἂν ποσὶν κακά. 

182 ἐξ ὑπαρχῆς, 1.6. taking up anew the 
search into the death οὗ Laius. Arist. de 
Anim. 2. 1 πάλιν δ᾽ ὥσπερ ἐξ ὑπαρχῆς 
ἐπανίωμεν : SO πάλιν οὖν οἷον ἐξ ὑπαρχῆς 
Rhet. 1. 1. 14: [Dem.] or. 40 § 16 πάλιν 
ἐξ ὑπαρχῆς λαγχάνουσί μοι δίκας. The 
phrase ἐν τῇ τῆς ἐπιστήμης ὑπαρχῇ occurs 
in the paraphrase by Themistius of Arist. 
περὶ φυσικῆς ἀκροάσεως 8. 3 (Berlin ed. 
vol. I. 247 ὁ 29): elsewhere the word 
occurs only in ἐξ ὑπαρχῆς. Cp. Zl. 725 
ὑποστροφῆςτεὑποστραφέντες : Her. 5. 116 
ἐκ νέης: Thuc. 3. 92 ἐκ καινῆς. αὖθις, as 


. 


he had done in the case of the Sphinx’s 
riddle: αὐτά Ξ- τὰ ἀφανῆ. 

188 ἐπαξίως (which would usually 
have a genitive) implies the standard— 
worthily of his own godhead, or of the 
occaston—and is slightly stronger than 
ἀξίως. Cp. Eur. ec. 168 ἀπωλέσατ᾽, 
ὠλέσατ᾽: Or. 181 διοιχόμεθ᾽, οἰχόμεθ᾽ : 
Alc. 400 ὑπάκουσον, ἄκουσον. 

134 πρὸ, on behalf of, cp. πρὸ τῶνδε το, 
O.C. 811: Xen. Cyr. 8.8.4 el tis...dta- 
κινδυνεύσειε πρὸ βασιλέως: 1. 6. 42 ἀξιώ- 
σουσι σὲ πρὸ ἑαυτῶν βουλεύεσθαι. Campb. 
reads πρὸς τοῦ θανόντος, which here could 
mean only “αὐ the instance of the dead.’ 
πρός never=‘on behalf of,’ ‘for the sake 
of,’ but sometimes ‘oz the szde of’: e.g. 
Her. 1. 124 ἀποστάντες ἀπ᾽ ἐκείνου καὶ 
γενόμενοι πρὸς σέο, ‘ranged themselves on 
your side’: 1. 75 ἐλπίσας πρὸς ἑωυτοῦ τὸν 
χρησμὸν εἶναι, that the oracle was on his 
side: below, 1434, πρὸς σοῦ..«φράσω, I 
will speak on your side,—in your in- 
terest: Zrach. 479 καὶ τὸ πρὸς xelvov 
λέγειν, to state his side of the case also. 
- ἐπιστροφήν, a turning round (0.C. 
1045), hence, attention, regard: ἐπιστρο- 
φὴν τίθεσθαι (like σπουδήν, πρόνοιαν τίθ., 
Ai. 13, 536)-: ἐπιστρέφεσθαί (τινος), Phil. 
599. Dem. 2022 Aristocr. 8 136 οὐκ 
ἐπεστράφη ‘heeded not’=ovdév ἐφρόντισε 
tb. § 135. 

137 ὑπὲρ γὰρ οὐχὶ K.T.A., 2.4. not 


30 ΣΟΦΟΚΛΕΟΥΣ 


5 +) 5 ἣν e nw “a 3 5 ~ 4 
ἀλλ᾽ αὐτὸς αὐτοῦ τοῦτ᾽ ἀποσκεδῶ μύσος. 
9 δὴ 3 A ε \ (eae eS 
ὅστις γὰρ ἣν ἐκεῖνον ὁ κτανὼν τάχ᾽ ἂν 


Koy av τοιαύτῃ χειρὶ τιμωρεῖν θέλοι. 


140 


κείνῳ προσαρκῶν οὖν ἐμαυτὸν opera. : Rona 
ἀλλ᾽ ὡς τάχιστα, παῖδες, ὑμεῖς μὲν βάθρων 
ἵστασθε, τούσδ᾽ ἄραντες ἱκτῆρας κλάδους, 


ἄλλος δὲ Κάδμου λαὸν ὧδ᾽ ἀθροιζέτω, 


ὡς πᾶν ἐμοῦ δράσοντος" ἢ γὰρ εὐτυχεῖς 


145 


σὺν τῷ θεῷ φανούμεθ᾽, ἢ πεπτωκότες. 


IE. ὦ παῖδες, ἱστώμεσθα. 


τῶνδε γὰρ χάριν 


καὶ δεῦρ᾽ ἔβημεν ὧν ὅδ᾽ ἐξαγγέλλεται. 
Φοῖβος δ᾽ ὃ πέμψας τάσδε μαντείας ἅμα 


σωτήρ θ᾽ ἵκοιτο καὶ νόσου παυστήριος. 


150 


XOPOS: 


στρ. α΄. 
5 Πυθῶνος ἀγλαὰς ἔβας 


conjecture was sometimes carried. Cp. on 1520. 


ὦ Διὸς ἀδυεπὲς ati, Tis ποτε τᾶς πολυχρύσου 


138 αὐτοῦ L: αὑτοῦ τ. 





merely in the cause of Laius, whose widow 
he has married. The arrangement of the 
words is designed to help a second mean- 
ing of which the speaker is unconscious: 
‘in the cause of a friend who is 72:02 far 
off’ (his own father). The reference to 
Laius is confirmed by κείνῳ προσαρκῶν 
In I4l. 

138 αὑτοῦ-- ἐμαυτοῦ. The reflexive 
αὑτοῦ, etc., is a pron. of the rst pers. in 
O.C. 966, Zi. 285, Ad. 1132: of the 
2nd pers., in O.C. 853, 930, 1356, 77. 
451. ἀποσκεδῶ, dispel, as a taint in 
- the air: cp. Od. 8. 149 σκέδασον δ᾽ ἄπο 
κήδεα θυμοῦ: Plat. Phaed. 77 Ὁ μὴ...ὁ 
ἄνεμος αὐτὴν (τὴν ψυχὴν) ἐκβαίνουσαν ἐκ 
τοῦ σώματος διαφυσᾷ καὶ διασκεδάννυσιν. 

189 f£. ἐκεῖνον ὁ κτανὼν. 
emphasis: cp. 820.---τοιαύτῃ, referring to 
κτανὼν, implies φονίᾳ: on τιμωρεῖν see 
107. The spectator thinks of the time 
when Oed. shall be blinded by his own 
hand.—For the double ἂν cp. 339s 862, 
1438. 

142 παῖδες. The king here, as the 
priest in 147, addresses a// the suppliants. 
ἄλλος (144) is one of the king’s attend- 
Δηί5.--- βάθρων | ἵστασθε κιτιλ. Cp. Ant. 
417 χθονὸς... ἀείρας: Fhil. 630 νεὼς ἄγον- 
τα. Prose would require a compound 


ἐκεῖνον has - 


verb: Xen. Symp. 4. 31 tbravicravra.... 
θάκων. ἄραντες. Aesch. Suppl. 481 κλά- 


dous γε τούτους aly’ ἐν ἀγκάλαις λαβὼν! 
βωμοὺς ἐπ᾿ ἄλλους δαιμόνων ἐγχωρίων | θές. 

145 πᾶν... δράσοντος, to do every- 
thing=to leave nothing untried: for ὡς 
cp. 97. Plat. Afol. 39 A ἐάν τις τολμᾷ 
πᾶν ποιεῖν καὶ λέγειν. Xen. //ellen. 7. 4. 
21 πάντα ἐποίει ὅπως, εἰ δύναιτο, ἀπαγά- 
you. εὐτυχεῖς... πεπτωκότες : ‘fortunate,’ 
if they succeed in their search for the 
murderer, who, as they now know, is in . 
their land (110): ‘ruined,’ if they fail, 
since they will then rest under the ἀνή- 
κεστον μίασμα (98). The unconscious 
speaker, in his last word, strikes the 
key-note of the destined περιπέτεια. 

147 ff. ὦ παῖδες: see on 142.---καὶ 
δεῦρ᾽ ἔβημεν, we een came here: 2.é. this 
was the motive of our coming in the-first 
instance. Phil. 380 ἐπειδὴ καὶ λέγεις 
θρασυστομῶν: Lys. Jn Lratosth. § 29 
παρὰ τοῦ more καὶ λήψεσθε δίκην; ἐξ- 
αγγέλλεται, proclaims on his own part 
(midd.), of himself: 2.5. promises un- 
asked, wliro pollicetur. Cp. Ai. 1376 
ἀγγέλλομαι.. «εἶναι φίλος, ‘I offer friend- 
ship.’ Eur. has thus used égayy. even 
where metre permitted the more usual 
ἐπαγγέλλομαι: Heracl. 531 κἀξαγγέλ- 


ΟἸΔΊΠΟΥΣ. TYPANNOS 31 
no, but in mine own cause, shall I dispel this taint. For who- 
ever was the slayer of Latus might wish to take vengeance on me 
also with a hand as fierce. Therefore, in doing right to Laitus, 
I serve myself. 

Come, haste ye, my children, rise from the altar-steps, and 
lift these suppliant boughs; and let some other summon hither 
the folk of Cadmus, warned that I mean to leave nought un- 
tried ; for our health (with the god’s help) shall be made certain 
—or our ruin. 

Pr. My children, let us rise; we came at first to seek what 
this man promises of himself. And may Phoebus, who sent 
these oracles, come to us therewith, our saviour and deliverer 
from the pest. 


CHORUS. 
O sweetly-speaking message of Zeus, in what spirit 
hast thou come from golden Pytho unto — glorious 


189 ἐκεῖνον has been made from ἐκεῖνοσ in L. The false reading ἐκεῖνος occurs in 


some of the later MSS. 





λομαι | θνήσκειν, I offer to die.—dpa: 
Ζ.6. may the god, who has summoned us 
to put away our pollution, a¢ the same time 
come among us as a healing presence. 

151—215 The Chorus consists of 
Theban elders—men of noble birth, ‘the 
foremost in honour of the land’ (1223) 
—who represent the Κάδμου λαός just 
summoned by Oedipus (144). Ocedipus 
having now retired into the palace, and 
the suppliants having left the stage, the 
Chorus make their entrance (πάροδος) 
into the hitherto vacant ὀρχήστρα. For 
the metres see the Analysis which follows 
the Introduction. 

ist strophe (t51—158). Is the god’s 
message indeed a harbinger of health? 
Or has Apollo some further pain in store 
for us? 

ist antistrophe (159—166). May 
Athene, Artemis, and Apollo succour us! 

and strophe (167—178). The fruits of 
the earth and the womb perish. 

and antistrophe (t7g—189). The un- 
buried dead taint the air: wives and 
mothers are wailing at the altars. 

3rd strophe (1g0—202). May Ares, the 
god of death, be driven hence: may thy 
lightnings, O Zeus, destroy him. 

37d antistrophe (203—215). May the 
Lycean Apollo, and Artemis, and Diony- 
sus fight for us against the evil god. 

151 dar, of a god’s utterance or oracle 


(1440), a poet. equivalent for φήμη: cp. 
310 dm’ οἰωνῶν φάτιν. Διὸς, because 
Zeus speaks by the mouth of his son; 
Aesch. Aum. 19 Διὸς προφήτης δ᾽ ἐστὶ 
Λοξίας πατρός. adverts, merely a general 
propitiatory epitnet : the Chorus have not 
yet heard whether the response is com- 
forting or not. It is presently told to 
them by Oed. (242). Cp. Zl. 480 ἁδυ- 
Tv Lwy...dvepatwv, dreams breathing com- 
fort (from the gods). τίς ποτε.. ἔβας; 
What art thou that hast come? ze. in 
what spirit hast thou come? bringing us 
health or despair? 

152 IIvOevos, from Pytho (Delphi): 
for the gen. see on 142 βάθρων | ἵστασθε. 
Tas πολυχρύσου, ‘rich in gold,’ with 
allusion to the costly ἀναθήματα dedicated 
at Delphi, and esp. to the treasury of the 
temple, in which gold and silver could be 
deposited, as in a bank, until required for 
use. Jliad g. 404 οὐδ᾽ ὅσα...λάϊνος οὐδὸς 
ἀφήτορος ἐντὸς ἐέργει | Φοίβου ᾿Απόλλωνος, 
Πυθοῖ ἐνὶ πετρηέσσῃ. Thuc. 1. 121 vav- 
τικόν τε ἀπὸ τῆς ὑπαρχούσης τε οὐσίας 
ἐξαρτυσόμεθα, καὶ ἀπὸ τῶν ἐν Δελφοῖς καὶ 
᾿Ολυμπίᾳ χρημάτων. Athen. 233 Ε τῷ 
μὲν οὖν ἐν Δελφοῖς ᾿Απόλλωνι τὸν πρότερον 
ἐν τῇ Λακεδαίμονι χρυσὸν καὶ ἄργυρον 
[πρότερον Ξε Ὀεΐίοτα the time of Lysander] 
ἱστοροῦσιν ἀνατεθῆναι. Eur. Andr. 1093 
θεοῦ | χρυσοῦ γέμοντα γύαλα (recesses), 
θησαυροὺς βροτῶν. Jon 54 Δελφοί 


rst 


strophe. 
Xp - 


32 TOPOKAEOYS 


3OnBas; ἐκτέταμαι, φοβερὰν φρένα δείματι πάλλων, 


Γ 4 ἰήιε Δάλιε Παιάν, 


aay Ν Ν cys , Ἃ / 
ὅ ἀμφὶ σοὶ αζόμενος τί μοι ἢ νέον 


155 


6% περιτελλομέναις ὥραις πάλιν ἐξανύσεις χρέος. 
1 εἰπέ μοι, ὦ χρυσέας τέκνον ᾿Ελπίδος, ἄμβροτε Φάμα. 


> , 
QvT. a. 


 γαιάοχόν τ᾽ ἀδελφεὰν 


πρῶτά σε κεκλόμενος, θύγατερ Διός, auBpor ᾿Αθάνα, 


160 


3” Apteuw, ἃ κυκλόεντ᾽ ἀγορᾶς θρόνον εὐκλέα θάσσει, 


159 κεκλόμενος L, with w written over os by a late hand. A few of the later ss. 





σφ᾽ ἔθεντο (the young Ion) χρυσοφύ- 
λακα τοῦ θεοῦ, | ταμίαν τε πάντων. Pind. 
Pyth. 6. 8 ἐν πολυχρύσῳ ᾿Απολλωνίᾳ... 
νάπᾳ (1.6. ἐν ἸΤυθοϊ). 

158 The bold use of ἐκτέταμαι is in- 
terpreted by φοβερὰν φρένα δείματι πάλ- 
λων, which is to be taken in close con- 
nection with it. ἐκτείνεσθαι is not found 
elsewhere of mental tension (though 
Dionys. De Comp. Verb. c. 15 ad fin. has 
ἡ τῆς διανοίας ἔκτασις καὶ τὸ τοῦ δείματος 
ἀπροσδόκητον... Cp. Xen. Cyr. τς 3. 11 
ἕως παρατείναιμι τοῦτον, ὥσπερ οὗτος 
ἐμὲ παρατείνει ἀπὸ σοῦ κωλύων, ----“ γαεξ,᾽ 
‘torture’ him. But παρατείνεσθαι, when 
used figuratively, usually meant ‘to be 
worn out,’ ‘fatigued to death’: eg. Plato 
Lysis 204 C παραταθήσεται ὑπὸ σοῦ ἀκούων 
θαμὰ λέγοντος, ἐ716ςαὐϊέτ7,, he will be tired 
to death of hearing it. So Xen. Mem. 3. 
13. 6 παρατέταμαι μακρὰν ὁδὸν πορευ- 
θείς. Triclinius explains here, ‘I am 
prostrated by dread’ (ἐκπέπληγμαι, map’ 
ὅσον οἱ ἐκπλαγέντες ἔκτασιν σώματος καὶ 
ἀκινησίαν πάσχουσιν: cp. Eur. Med. 585 
ἕν yap ἐκτενεῖ σ᾽ ἔπος): so Ph. 858 ἐκτέ- 
ταται νύχιος (of ἃ sleeper). But the con- 
text favours the other view.—dAdov, 
transitive, governing φρένα, making my 
heart to shake; not intransitive, for παλ- 
Aduevos, with φρένα as accus. of the part 
affected. An intransitive use of πάλλω 
in this figurative sense is not warranted 
by such instances as Ar. Lys. 1304 κοῦφα 
πάλλων, ‘lightly leaping in the dance’: 
Eur. Zl. 435 ἔπαλλε δελφίς (=éoxipra), 
‘the dolphin leaped’: 24. 477 ἵπποι ἔπαλ- 
λον ‘quivered’ (in death). Cp. Aesch. 
P. V. 881 κραδία φόβῳ φρένα λακτίζει: 
so, when the speaker is identified with 
the troubled spirit within him, we can 
say φρένα mad\\w,—where φρένα has a less 
distinctly physical sense than in Aesch. 


Z.c., yet has physical associations which 
help to make the phrase less harsh. 

154 Addte. The Delphian Apollo is. 
also Delian—having passed, according to 
the Ionic legend, from his native Delos, 
through Attica, to Delphi (Aesch.- Zum. 
9). A Boeotian legend claimed Tegyra 
as the birthplace of Apollo: Plut. Pe/op. 
16 ἐνταῦθα μυθολογοῦσι τὸν θεὸν γενέσθαι, 
καὶ τὸ μὲν πλησίον ὄρος Δῆλος καλεῖται. 
We can scarcely say, however, with 
Schneidewin that Δάλιε here ‘bewrays 
the Athenian,’ when we remember that 
the Theban Pindar hails the Delphian 
Apollo as Λύκιε καὶ Δάλου ἀνάσσων Φοῖβε 
(Pyth. τ. 30).---ήϊε (again in 1096), in- 
voked with the cry (7: cp. 77. 221 ἰὼ ἰὼ 
Παιάν. Soph. has the form παιών, 
παιήων as==‘a healer’ (not with ref. to 
Apollo), Phil. 168, 832. 

155 ἁζόμενος (rt. dy, whence ἅγιος) im- 
plies a religzous fear: cp. Od.9.478 σχέτλι᾽, 
ἐπεὶ ξείνους οὐχ ἅζεο σῷ ἐνὶ οἴκῳ | ἐσθέμε- 
ναι. τί μοι... χρέος: ‘what thing thou 
wilt accomplish for me’: z.¢., what expia- 
tion thou wilt prescribe, as the price of 
deliverance from the plague. Will the 
expiation be of a new kind (véov)? Or 
will some ancient mode of atonement be 
called into use once more (πάλιν) πά- 


“λιν recalls Aesch. Ag. 154 μίμνει yap 


φοβερὰ παλίνορτος | οἰκονόμος δολία 
μνάμων μῆνις τεκνόποινος. νέον, adjective 
with χρέος : πάλιν, adverb with ἐξανύσεις. 
τί μοι νέον χρέος ἐξανύσεις; ἢ τί χρέος 
πάλιν ἐξανύσεις; The.doubling of ἤ harshly 
co-ordinates νέον and πάλιν, as if one said 
τίνας ἢ μαχομένους ἢ ἀμαχεὶ ἐνίκησαν ; 
χρέος here=xpjua, ‘matter’ (implying 
importance): cp. Aesch. Supl. 374 (of a 
king) χρέος | πᾶν ἐπικραίνεις : Eur. H. 7. 
530 τί καινὸν ἦλθε τοῖσδε δώμασιν χρέος; 
Others take it 485: ‘obligation’ (cp. O. C. 


OIAITOY2 TYPANNOS 


Thebes Ὁ 


33 


I am on the rack, terror shakes my soul, O thou 


Delian Healer to whom wild cries rise, in holy fear of thee, what 
thing thou wilt work for me, perchance unknown before, per- 
chance renewed with the revolving years : tell me, thou immortal 


Voice, born of Golden Hope! 


First call I on thee, daughter of Zeus, divine Athena, 


and on thy sister, guardian of our land, Artemis, 


who 


sits on her throne of fame, above the circle of our Agora, 


have κεκλομένῳ or Κεκλομένω.---κέκλομαι, ὦ Blaydes.—duBpor’] ἄντομ᾽ Wecklein. 





235), but against this is ἐξανύσεις, which 
could not mean either to ‘impose’ or to 
‘exact’ it. Whitelaw renders, ‘ what re- 
quirement thou wilt enact (by oracular 
voice),’ finding this use of ἀνύω in O. C. 
454, Ant. 1178; but there (as below, 720) 
it has its normal sense, ‘fulfil.’ 

156 περιτελλομ. ὥραις, an epic phrase 
which Ar. Av. 697 also has. Od. 14. 293 
ἀλλ᾽ ὅτε δὴ μῆνές τε καὶ ἡμέραι ἐξετε- 
λεῦντο | ay περιτελλομένου ἔτεος, καὶ ἐπή- 
λυθον ὧραι. 

157 χρυσέας κιτιλ. The answer (not 
yet known to them) sent by Apollo is 
personified as Pdpa, a divine Voice,— 
‘the daughter of golden hope,’ because— 
whether favourable or not—it is the zsszze 
of that hepe with which they had awaited 
the god’s response. 

159 κεκλόμενος is followed in 164 by 
προφήνητέ μοι instead of εὔχομαι προ- 
gav? at Cp. Plat. Legg. 686 Ὁ ἀπο- 
βλέψας yap πρὸς τοῦτον τὸν στόλον οὗ 
πέρι διαλεγόμεθα ἔδοξέ μοι πάγκαλος... 
εἶναι. Antiphon Jer. B. β. ὃ το ἀπο- 
λυόμενος δὲ ὑπό τε τῆς ἀληθείας τῶν 
πραχθέντων ὑπό τε τοῦ νόμου καθ᾽ ὃν διώ- 
κεται, οὐδὲ τῶν ἐπιτηδευμάτων εἵνεκα δί- 
καιοι τοιούτων κακῶν ἀξιοῦσθαί ἐσμεν. 
Xen. Cyr. 8. 8. το ἦν δὲ αὐτοῖς νόμιμον 
.. voulfovres. The repetition of ἄμ- 
Bpor’ has provoked some weak and need- 
less conjectures: see on 517. 

160 yatdoxov, holding or guarding 
our land; so Aesch. Suppl. 816 γαιάοχε 
παγκρατὲς Zed. In O. C. 1072 it is the 
Homeric epithet of Poseidon, ‘girdling 
the earth,’ τὸν πόντιον γαιάοχον. Cp. 
Ταλλὰς πολιοῦχος Ar. Hg. 581 (πολιάοχος 
Pind. Οἱ. 5. 10), πολισσοῦχοι θεοί Aesch. 
Thed. 69. 

161. κυκλόεντ᾽ ἀγορᾶς Opdvoy=xv- 
κλοέσσης ἀγορᾶς θρόνον: cp. Ant. 793 
νεῖκος ἀνδρῶν ξύναιμον, Trach. 9093 ὦ 


1.5.1} 


Κηναία κρηπὶς βωμῶν. ‘Round throne of 
the marketplace’ means simply (I now 
think) ‘throne consisting of the round 
marketplace.” The sitting statue of 
Artemis is in the middle of the agora; 
hence the agora itself is poetically called 
her throne. The word κύκλος in con- 
nection with the Athenian agora, of 
which it perhaps denoted a special part; 
schol. Ar. Zg. 137 ὁ δὲ κύκλος ᾿Αθήνησίν 
ἐστι καθάπερ μάκελλος, ἐκ τῆς κατασκευῆς 
(form) τὴν προσηγορίαν λαβών. ἔνθα δὴ 
πιπράσκεται χωρὶς κρεών τὰ ἄλλα ὥὦνια, καὶ 
ἐξαιρέτως δὲ οἱ ἰχθύες. Cp. Eur. Or. 919 
ὀλιγάκις ἄστυ κἀγορᾶς χραίνων κύκλον, ‘the 
circle of the agora,’ zie. ‘its bounds’: cp. 
Thuc. 3. 74 τὰς οἰκίας τὰς ἐν κύκλῳ τῆς 
ἀγορᾶς, ‘all round’ the agora. In //. 18. 
504, cited by Casaubon on Theophr. 
Char. 2. 4, ἱερῷ ἐνὶ κύκλῳ refers merely to 
the yépovres in council. This is better 
than (1) ‘her round seat in the agora’— 
κυκλόεντα meaning that the pedestal of the 
statue was circular; (2) ‘her throne in 
the agora, round which κύκλιοι χοροί 
range themselves.’ This last is im- 
possible, 

εὐκλέα, alluding to Artemis Εὔκλεια, 
the virgin goddess of Fair Fame, wor- 
shipped esp. by Locrians and Boeotians: 
Plut. Arist. 20 βωμὸς yap αὐτῇ καὶ ἄγαλμα 
παρὰ πᾶσαν ἀγορὰν ἵδρυται, καὶ προθύουσιν 
αἵ τε γαμούμεναι καὶ οἱ γαμοῦντες : also at 
Corinth, Xen. Hellen. 4. 4. 2. 
saw a temple of “Apreyis Εὔκλεια, with 
a statue by Scopas, near the Προιτίδες 
πύλαι on the N.E. side of Thebes. Near 
it were statues of Apollo Boedromios and 
Hermes Agoraios. The latter suggeststhat 
the Agora of the Lower Town (which 
was deserted when Pausanias visited 
Thebes) may have been near. In men- 
tioning the ἀγορά, Soph. may have been 
further influenced by the fact that. Artemis 


3 . 


Ist anti- 
strophe. 


Pausanias 


34 ΣΟΦΟΚΛΕΟῪΣ 


4 καὶ Φοῖβον ἑκαβόλον, ἰὼ 


ὅ τρισσοὶ ἀλεξίμοροι προφάνητέ μοι, 
6 εἰ ποτε καὶ προτέρας ἄτας ὕ ὕπερ ὀρνυμένας πόλει 165 
7 ἠνύσατ᾽ ἐκτοπίαν φλόγα πήματος, ἔλθετε καὶ νῦν. 


στρ. β΄. ὦ πόποι, ἀνάριθμα γὰρ φέρω 
2 πήματα" νοσεῖ δέ μοι πρόπας στόλος, οὐδ᾽ ἔνι φροντίδος 


ἔγχος 
4 δ, ἢ 
8 ᾧ τις ἀλέξεται. 


NO οι Ὁ. 


8 ἀκτὰν πρὸς ἑσπέρου 


οὔτε γὰρ ἔκγονα 

lal ‘ » » ’ 

κλυτᾶς χθονὸς αὔξεται, οὔτε τόκοισιν 

ἰηίων καμάτων ἀνέχουσι γυναῖκες" 

ἄλλον δ᾽ dv ἄλλῳ προσίδοις ἅπερ εὔπτερον ὄρνιν 
κρεῖσσον ἀμαιμακέτου πυρὸς ὄρμενον 

εοῦ 


171 
174 


ἀντ. B. ὧν πόλις ἀνάριθμος ὄλλυται: 
2 νηλέα. δὲ γένεθλα πρὸς πέδῳ θαναταφόρα Ketrar 


ἀνοίκτως" 


8 ἐν δ᾽ ἄλοχοι πολιαΐ τ᾽ ἐπὶ ματέρες 


4 ἀκτὰν παρὰ βώμιον ἄλλοθεν ἄλλαι 


162 ἰὼ ἰὼ L: ἰὼ r, and Heath. 


182 


180 The rst hand in L seems to have 


written θαναταφόρω (sic), which a later hand altered to θαναταφόρα (or θανατάφορα, 





was worshipped as ’Ayopala: thus in the 
altis at Olympia there was an ’Apreyuldos 
᾿Αγοραίας βωμός near that of Ζεὺς ᾽Αγο- 
ραῖος (Paus. Rios, 4)- 

165 ἄτας ὕπερ, ‘on account of ruin’ 
(z.e. ‘to avert it’): cp. Amt. 932 κλαύ- 
pa’ ὑπάρξει βραδυτῆτος ὕπερ. So Aesch. 
Theb. 111 ἴδετε παρθένων ἱκέσιον λόχον 
δουλοσύνας ὕπερ, ‘to avert slavery.’ Cp. 
187. épvupévas πόλει: the dat. Geet} 
as after verbs of attacking, e.g. ἐπιέναι, 
ἐπιτίθεσθαι. Musgrave’s conj. ὑπερορνυ- 
μένας πόλει (the compound nowhere oc- 
curs) has been adopted by some editors. 

166 ἠνύσατ᾽ ἐκτοπίαν, made ἐκτοπίαν, 
= ἐξωρίσατε, arare use of ἀνύω like ποιεῖν, 
καθιστάναι, ἀποδεικνύναι : for the ordi- 
nary use, cp. 720 ἐκεῖνον ἤνυσεν | φονέα 
γενέσθαι, effected that he should become. 
In Ant. 1178 τοὔπος ws ἄρ᾽ ὀρθὸν ἤνυσας, 
the sense is not ‘made right,’ but ‘ drought 
duly 4o pass.’ ἔλθετε kal νῦν, an echo of 
προφάνητέ μοι, προτέρας having sug- 
gested καὶ viv: as in 338 ἀλλ᾽ ἐμὲ ψέγεις 
repeats ὀργὴν ἐμέμψω τὴν ἐμήν. 

167 ὦ πόποι is merely a cry like 
παπαῖ: Trach. 853 κέχυται νόσος, ὦ πόποι, 


οἷον, κιτ.λ. 

170 στόλος, like στρατός (Pind. Pyth. 
2. 46, etc.), Ξελαός.---ἔνι ΞΞ ἔνεστι, is avail- 
able.—povrlSos ἔγχος, not, a weapon 
consisting in a device, but a weapon 
discovered by human wit, ἔγχος ᾧ Tis 
ἀλέξεται being a bold equivalent for μη- 
xavh ἀλεξητηρία. 

171 This future has the support of the 
best Mss. in Xen. An. 7. 7. 3 οὐκ ἐπιτρέ- 
ψομεν.. «ὡς πολεμίους ἀλεξόμεθα: and of 
grammarians, Bekk. Anecd. 5 415: the 
aorist ἀλέξαι, ἀλέξασθαι also occurs. 
These forms are prob. not from the stem 
ἀλεξ (whence present ἀλέξω, cp. ἀέξω, 
ὁδάξω) but from a stem ἀλκ with un- 
consciously developed e, making ἀλεκ (cp. 
ἄλ-αλκον) : see Curtius, Verd, 11. 258, 
Eng. tr. 445. Homer has the fut. ἀλεξή- 
ow, and Her. ἀλεξήσομαι.---Ορ. 539. 

173 τόκοισιν, dy births. Women are 
released from travail, not by the birth 
of living children, but either by death 
before delivery, or by still births. See on 
26, and cp. Hes. Op. 244 οὐδὲ γυναῖκες 
τίκτουσιν. If rékow=‘in child-bed? 
(and so the schol., ἐν τοῖς réxos), the 


OIAITOYS ΤΥΡΆΝΝΟΣ 35 


and on Phoebus the far-darter: O shine forth on me, my three- 
fold help against death! If ever aforetime, in arrest of ruin 
hurrying on the city, ye drove a fiery pest beyond our borders, 
come now also! 


Woe is me, countless are the sorrows that I bear; a plague is 
on all our host, and thought can find no weapon for defence. 
The fruits of the glorious earth grow not; by no birth of children 
do women surmount the pangs in which they shriek ; and life on 
life mayest thou see sped, like bird on nimble wing, aye, swifter 
than resistless fire, to the shore of the western god. 


By such deaths, past numbering, the city perishes: unpitied, 
her children lie on the ground, spreading pestilence, with none to 
mourn:and meanwhile young wives, and grey-haired mothers with 
them, uplift a wail at the steps of the altars, some here, some there, 
for there are traces of an accent over the a). Some of the later Mss. (including A) 


have the dative, others the nomin. 182 ἀκτὰν] αὐδὰν Hartung, ἀχὰν Nauck.— 
παραβώμιον. L, with most of the later Mss. (including A); some others have παρὰ 





meaning would be simply, ‘women die 
in child-bed,—not necessarily ‘before 
child-birth’; but the point here is the 
blight on the fruits of earth and womb,— 
not merely the mortality among women. 

175 ἄλλον δ᾽... ἄλλῳ, ‘one after an- 
other.’ The dative here seems to depend 
mainly on the notion of adding implied 
by the iteration itself; though it is pro- 
bable that the neighbourhood of zpos in 
προσίδοις may have beeri felt as softening 
the boldness. That προσορᾶν could be 
used as=‘to see im addition’ is incon- 
ceivable; nor could such use be justified 
by that of ἐνορᾶν τινι as=dpav ἔν τινι. 
And no one, I think, would be disposed 
to plead lyric license for ἄλλῳ πρὸς ἴδοις 
on the strength of ἀκτὰν πρὸς ἑσπέρον 
θεοῦ in 177. Clearly there was a ten- 
dency (at least in poetry) to use the dative 
thus, though the verd of the context 
generally either (a) helps the sense of 
‘adding,’ or (4) leaves an alternative. 
Under (a) I should put Z/. 235 τίκτειν 
ἄταν ἄταις: Eur. Helen. 195 δάκρυα δά- 
κρυσί μοι φέρων. Under (4), Eur. Or. 1257 
πήματα πήμασιν ἐξεύρῃ: Phoen. 1496 
φόνῳ φόνος | Οἰδιπόδα δόμον ὥλεσε: where 
the datives mzght be instrumental. On 
the whole, I forbear to recommend ἄλλον 
δ᾽ ἂν ἄλλᾳ προσίδοις, though easy and 
tempting; cp. Thuc. 2. 4 ἄλλοι δὲ ἄλλῃ 
THs πόλεως σποράδην ἀπώλλυντο. 

177 ὄρμενον, aor. part. (72 11. 571 
Sodpa...doueva πρόσσω), ‘sped,’ ‘hurried,’ 


since the life is quickly gone. κρεῖσσον 
οὐ πυρὸς, because the πυρφόρος λοιμός 
drives all before it. 

178 ἀκτὰν πρὸς for πρὸς ἀκτάν, since 
the attributive gen. ἑσπέρου θεοῦ is equiv. 
to an adj. agreeing with ἀκτάν : cp. O.C. 
84 ἕδρας | πρώτων ἐφ" ὑμῶν, 26. 126 ἄλσος 
és...xopav: El. 14 τοσόνδ᾽ ἐς ἥβης: so 
Aesch. P. V. 653, Theb. 185: Eur. Or. 
94. ἑσπέρου θεοῦ: as the Homeric 
Erebos is in the region of sunset and 
gloom (Od. 12. 81), and Hades is ἐννυ- 
χίων ἄναξ O. C. 1559. 

179 ὧν.. ἀνάριθμος. ὧν, masc., re- 
ferring to G\Aov...a\\y,—‘ to such (deaths) 
knowing no limit’: cp. ἀνάριθμος θρήνων 
El. 232, μηνῶν  ἀνήριθμος Ai. 602. An 
adj. formed with a privative, whether 
from noun or from verb, constantly takes 
a gen. in poetry: see on 190 (dxaX«os), 
885 (apdByrTO0s). 

180 γένεθλα (πόλεως), Sher sons’: cp. 
1424 τὰ θνητῶν γένεθλα, the sons of men. 
νηλέα, unpitied ; ἀνοίκτως, without οἶκτος, 
lament, made for them: they receive 
neither ταφή nor θρῆνος. Cp. Thuc. 2. 
50 πολλών ἀτάφων γιγνομένων (in the 


plague, 430 B.C.). 


181 ἐν δ᾽, cp.on 27. ἐπὶ, adv.: Her. 
ἡ. 65 τόξα δὲ καλάμινα εἶχον, ...ἐπὶ δέ, 
σίδηρον (v. ὦ. -os) ἦν. But ἔπιΞεἔπεστι, 
χε Vs BER. 

182 ἀκτὰν παρὰ βώμιον, ‘at the steps 
of the altars’: Aesch. Cho. 722 ἀκτὴ 
χώματος, the edge of the mound: Eur. 


32° 


and 
strophe. 


2nd anti- 
strophe. 


36 ΣΟΦΟΚΛΕΟΥΣ 


5 λυγρῶν πόνων ἱκτῆρες ἐπιστενάχουσιν. 
θ παιὰν δὲ λάμπει στονόεσσά τε γῆρυς ὅμαυλος" 


185 


7 ὧν ὕπερ, ὦ χρυσέα θύγατερ Διός, 


8 εὐῷπα “πέμψον ἀλκάν" 


στρ. γ. 


2 φλέγειιε περιβόατος ἀντιάζων, 


ν ΄ Ν ’ a A » 3 ΄ 
Αρεά τε τὸν μαλερόν, ὃς νῦν ἄχαλκος ἀσπίδων 


ΙΟΙ 


8 παλίσσυτον δράμημα νωτίσαι πάτρας 


4 ἔπουρον εἴτ᾽ ἐς μέγαν 
, 3 ’, 
5 θάλαμον ᾿Αμφιτρίτας 


195 


yes. 1.9 X ee Y 
6 €lT ἐς TOV amo €evov ορμον 


7 ΠΡ. ον κλύδωνα" 


8 *rehely γάρ, εἴ τι νὺξ ἀφῇ, 


βώμιον.---ἄλλαι MSS.: ἄλλαν Dindorf. 


185 ἐπιστονάχουσι L: ἐπιστενάχουσι r. 


191 περιβόατοΞ] περιβόατον Dindorf, placing a comma after it, and reading ἀντιάζω 


with Hermann. 


194 érovpoy, the true reading, was written by the ist hand in L, 


but altered by a later hand into ἄπουρον, over which is the gloss μακράν (the prep., 





Herc. F. 984 ἀμφὶ βωμίαν | ἔπτηξε κρηπῖδ᾽, 
at the base of the altar, ἄλλοθεν ἄλλαι 
(with ἐπιστενάχουσι), because the sounds 
are heard from various quarters. 

185 ἱκτῆρες with Avypav πόνων, en- 
treating on account of (for release from) 
their woes, causal gen.: cp. ἀλγεῖν τύχης, 
Aesch. Ag. 571. 


186 λάμπει: 473 ἔλαμψε... φάμα: 
Aesch. Theb. 104 κτύπον δέδορκα. ὅμαυ- 


λος, 2.5. heard at the same time, though 
not σύμφωνος with it. 

188 ΣΦ. ὧν ὕπερ: see on 165.—evora 
ἀλκάν : cp. ἀγανὴ σαίνουσ᾽ | ἐλπίς, Aesch. 
Ag. 101 (where Weil προφανεῖσ᾽), ἱλαρὸν 
φέγγος Ar. Ran. 455. 

190 “Apea te x.7.X. The acc. and 
infin. “Apea...vwtioat depend on δός or 
the like, ay aah by the preceding 
words. Cp. //. 7 Ζεῦ πάτερ, ἢ Αἴαντα 
λαχεῖν ἢ Τυδέος ΤΣ eee that). Aesch. 
Theb. 253 θεοὶ πολῖται, μή με δουλείας 
τυχεῖν. μαλερόν, raging: cp. μαλεροῦ πυ- 
pos 7]. 9. 242: μαλερών..«λεόντων Aesch. 
Ag. 141. Ares is for Soph. not merely the 
war-god, but generally Bporohovyos, the 
Destroyer: cp. At. 706. Here he is iden- 
tified with the fiery plague. ἄχαλκος 
ἀσπίδων (cp. 21. 36 ἄσκευον ἀσπίδων : 
Eur. Phoen. 324 ἄπεπλος φαρέων): Ares 
cqmes not, indeed, as the god of war 
(ὁ χαλκοβόας “Apns, O.C. 1046), yet 
shrieks of the dying surround him with 
a cry (βοή) as of battle. 


191 περιβόατος could not mean ‘cry- 
ing loudly’: the prose use (‘famous’ 
or ‘notorious,’ Thuc. 6. 31) confirms the 
pass. sense here. ἀντιάζων, attacking : 
Her. 4. 80 ἠντίασάν μιν (acc.) οἱ Θρήϊκες. 
Aesch. has the word once only, as=‘to 
meet’ (not in a hostile sense), 4g. 1557 
πατέρ᾽ ἀντιάσασα: Eur. always as=‘to 
entreat’; and so Soph. Z/. 100g. Din- 
dorf reads φλέγει we περιβόατον (the 
accus. on his own conject.), ἀντιάζω (sug- 
gested by Herm.), ‘I fray that’ etc. 
But the received text gives a more vivid 
picture. 

192 νωτίσαι, to turn the back in flight 
(Eur. Andr. 1141 πρὸς φυγὴν ἐνώτισαν), 
a poet. word used by Aesch. with acc. 
πόντον, to skim (Ag. 286), by Eur. Ph. 
651 (Dionysus) κισσὸς ὃν...ἐνώτισεν as 
=‘to cover the back of.’ δράμημα, cog- 
nate acc.: πάτρας, gen. after verb o 
parting from: see on βάθρων, 142. 

194 ἔπουρον -Ξ- ἐπουριζόμενον (ironical), 
Lidd. and Scott s.v. refer to.» Clemens 
Alexandr. Paed. 130 τῷ τῆς ἀληθείας 
πνεύματι ἔπουρος apbels, ‘lifted on a prose 
pering gale by the spirit of Truth.’ So 
Trach. 815 οὖρος ὀφθαλμῶν ἐμῶν | αὐτῇ 
γένοιτ᾽ ἄπωθεν ἑρπούσῃ καλώς: tb. 467 
ἀλλὰ ταῦτα μὲν | pelrw κατ᾽ οὖρον. Active 
in Trach. 954 ἔπουρος ἑστιῶτις αὔρα (schol, 
ἄνεμος οὔριος ἐπὶ τῆς οἰκίας), ‘wafting.’ 
The v.l. ἄπουρον would go with πάτρας, 
‘away from the dorders of my country’— 


OIAITNOYS TYPANNOZ a7 


entreating for their weary woes. The prayer to the Healer 
rings clear, and, blent therewith, the voice of lamentation: for 
these things, golden daughter of Zeus, send us the bright face 
of comfort. 


And grant that the fierce god of death, who now with no 3rd 
brazen shields, yet amid cries as of battle, wraps me in the sttophe 
flame of his onset, may turn his back in speedy flight 
from our land, borne by a fair wind to the great deep of 
Amphitrité, or to those waters in which none find haven, 
even to the Thracian wave; for if night leave aught undone, 


meaning that πάτρας ἄπουρον ΞΞ ‘far from our country’). 
The wrong one, ἄπουρον, prevailed in the later Mss. 
198 τέλει MSS. (τέλη in Bodl. Barocc. 66, 15th cent., is doubtless a 


ings. 
Doderlein. 


The schol. knew both read- 
196 ὅρμον] ὅρμων 





from Ionic οὖρος Ξε ὅρος, like ὅμουρος (Her. 
I. 57), πρόσουρος (Phil. 691), ξύνουρος 
(Aesch. Ag. 495), τηλουρός. Pollux 6. 
198 gives ἔξορος, ἐξόριος, but we nowhere 
find an Ionic ἄπουρος: while for Attic 
writers ἄφορος (from ὅρος) would have 
been awkward, since ddopos ‘sterile’ was 
in use. 

μέγαν | θάλαμον ᾿Αμφιτρίτας, the At- 
lantic. θάλαμος ’Audurplrns alone would 
be merely ‘the sea’ (Od. 3. gt ἐν πελάγει 
μετὰ κύμασιν ’Apudirplrns), but μέγαν helps 
to localise it, since the Atlantic (ἡ ἔξω 
στηλέων θάλασσα ἡ ᾿Ατλαντὶς καλεομένη, 
Her. 1. 202) was esp. ἡ μεγάλη θάλασσα. 
Thus Polyb. 3. 37 calls the AZediterrancan 
τὴν καθ᾽ Huds,—the Atlantic, τὴν ἔξω καὶ 
μεγάλην προσαγορευομένην. In Plat. 
Phaedo 109 B the limits of the known 
habitable world are described by the 
phrase, τοὺς μέχρι τῶν Ἡρακλείων στηλῶν 
ἀπὸ Φάσιδος (which flows into the Euxine 
on the E.), Eur. 47122. 3 ὅσοι τε πόντου 
(the Euxine) τερμόνων τ᾽ ᾿Ατλαντικῶν 
| ναίουσιν εἴσω: Herc. F. 234 ὥστ᾽ ᾿Ατ- 
λαντικῶν πέρα | φεύγειν ὅρων ἄν. 

196 ἀπόξενον. Aesch. has the word 
as=‘estranged from’ (γῆς, Ag. 1282), 
cp. ἀποξενοῦσθαι. Here it means ‘away 
Jrom strangers,’ in the sense of ‘keeping 
them at a distance.’ Such compounds 
are usu. fassive in sense: cp. ἀπόδειπνος 
(Hesych., =ddecmvos), ἀπόθεος, ἀπόμισθος, 
ἀπόσιτος, ἀπότιμος (215), dwoxphuaros.— 
ἀπόξενος ὅρμος, the Euxine: an oxy- 
moron,=6puos ἄνορμος, as in Phil. 217 
ναὸς ἄξενον ὅρμον. Strabo 7. 298 ἄπλουν 
γὰρ εἶναι τότε τὴν θάλατταν ταύτην καὶ 
καλεῖσθαι ΓΑξἕενον διὰ τὸ δυσχείμερον 
‘Kal τὴν ἀγριότητα τῶν περιοικούν- 


των ἐθνῶν καὶ μάλιστα τῶν Σκυθικῶν, 
ξενοθυτούντων, κιτιλ. The epithet 
Θρήκιον here suggests the savage folk 
to whom Ares is ἀγχίπτολις on the W. 
coast of the Euxine (Ant. 969). Ovid 
Trist. 4. 4.55 Frigida me cohibent Euxint 
litora Ponti: Dictus ab antiquis Axenus 
tlle fuit. 

198 τελεῖν γὰρ... ἔρχεται. Reading re- 
λεῖν, as Herm. suggested, instead of τέλει» 
I construe thus :—ei τὶ νὺξ ἀφῇ, ἦμαρ ἐπέρ- 
χεται τελεῖν τοῦτο, ‘If night omit anything 
(in the work of destruction), day comes 
after it to accomplish this.’ τελεῖν is 
the infin. expressing purpose, as often 
after a verb of going or sending, where 
the fut. participle might have been used: 
cp. Her. 7. 208 ἔπεμπε... κατάσκοπον 
ἱππέα, ἰδέσθαι [-Ξεὀψόμενον] ὁκόσοι τέ 
εἰσι, κιτιλ.: Thuc. 6. 50 δέκα δὲ τῶν νεῶν 
προὔπεμψαν ἐς τὸν μέγαν λιμένα πλεῦσαί 
τε καὶ κατασκέψασθαι... καὶ κηρῦξαι. 
Here the gres. inf. is right, because the 
act is not single but repeated. Observe 
how strongly τελεῖν is supported by the 
position of the word (‘To accomplish, — 
if night omit aught,—day follows’). No 
version of τέλει explains this. The 
most tolerable is:—‘ 7» fudness—if night 
omit aught—day attacks (ἐπέρχεται) this’ : 
but I do not think that such a rendering 
can stand. See Appendix.—el...agq. Cp. 
874 εἰ ὑπερπλησθῇ (lyric): O. C. 1443 
el στερηθῶ (dialogue): Amt. 710 κεῖ τις 
ἦ (do.). In using ef with subjunct., the 
Attic poets were influenced by the epic 
usage, on which see Monro, Homeric Gram- 
mar § 292. The instances in classical 
prose are usu. doubtful, but in Thue. 
6. 21 el ξυστῶσιν has good authority. 


38 ΣΟΦΟΚΛΕΟΥΣ 


ΕΘ See ὙΦ » Ν 
9 τοῦτ᾽ ἐπ᾿ ἦμαρ ἔρχεται 
10 τόν, ὦ « τᾶν πυρφόρων 
11 ἀστραπᾶν κράτη νέμων, 


200 


12 ὦ Ζεῦ πάτερ, ὑπὸ σῷ φθίσον κεραυνῷ. 


ς ; , > ¥ ΄, ᾿ , 9. 9. 9 ΝΥ 
ἀντ. γ. Λύκει᾽ ἀναξ, τά TE σὰ χρυσοστρόφων ἀπ᾽ ἀγκυλᾶν 


2 βέλεα θέλοιμ᾽ dv ἀδάματ᾽ ἐνδατεῖσθαι 


205 


8 ἀρωγὰ προσταθέντα, τάς τε πυρφόρους 
4 ᾿Αρτέμιδος αἴγλας, ξὺν αἷς 


ὦ» . 
ὅ AUKU ὄρεα διάσσει 
θ τὸν 
1 τᾶσδ᾽ ἐπώνυμον ‘yas, 
na ie » 
8 οἰνῶπα Βάκχον εὔιον, 
᾽ὔ ε , 
9 Μαινάδων ὁμόστολον 
κ᾿ 3 
10 πελασθῆναι φλέγοντ 


mere slip). See note. 
(=v. 213 πελασθῆναι φλέγοντ᾽). 


200 τὸν ὦ πυρφόρων MSS. 
Hermann inserts τᾶν after ὦ: Wolff οὖν after τόν. 
Lachmann proposed τόν, ὦ Ζεῦ (omitting Zed in v. 202). 


υσομίτραν TE κικλήσκω, 


210 


A long syllable is wanting 


In La late hand has written o 


over ὦ in πυρφῴρων, and A has εἰ written over 7 in κράτη. These are traces of the reading 





199 én’...épxerat: for the adverbial 
ἐπί separated from ἔρχεται, cp. O. C.1777 
μηδ᾽ ἐπὶ πλείω | θρῆνον éyelpere. This is 
‘tmesis’ in the larger sense: tmesis proper 
is when the prep. is essential to the sense 
of the verb: //. 8. 108 οὕς ποτ᾽ ἀπ’ Αἰνείαν 
ἑλόμην =ovs ἀφειλόμην Αἰνείαν : cp. Monro 
H. G. § 176. 

200 τόν--Ξδν, sc. “Apea (190). 
1379 0. 

203 Avxee, Apollo, properly the god 
of light (Aux), whose image, like that of 
Artemis, was sometimes placed before 
houses (Z/. 637 Φοῖβε προστατήριε, Aesch. 
Theb. 449 προστατηρίας | ᾿Αρτέμιδος), so 
that the face should catch the first rays 
of the morning sun (δαίμονες.. ἀντήλιοι 
Agam. 519): then, through Avxecos being 
explained as λυκοκτόνος (Soph. Zi. 7), 
Apollo the Destroyer of foes: Aesch. 
Theb. 145 Λύκει᾽ ἄναξ, Λύκειος γενοῦ] 
στρατῷ δαΐῳ. Cp. below, 919. 

204 aykvAav. ἀγκύλη, ἃ cord brought 
round on itself, a noose or loop, here=the 
νευρά of the dent bow. ἀγκύλων, the 
reading of L and A, was taken by Eu- 
stath. 33. 3 of the dow (ἄγκυλα τόξα). 

205 ἐνδατεῖσθαι, pass., to be distri- 
buted, z.¢. showered abroad on the hostile 
forces. The order of words, and the 
omission of σέ, are against making évdar. 


Cp. 


midd., though elsewhere the pass. occurs 
only in δέδασμαι: Appian, however, has 
γῆς διαδατουμένης 1. 1. It is possible that 
Soph. may have had in mind //. 18. 263 
ἐν πεδίῳ, ὅθι περ Τρῶες καὶ ᾿Αχαιοὶ | ἐν 
μέσῳ ἀμφότεροι μένος “Apnos δατέονται, 
‘share the rage of war,’ give and take 
blows. Others understand, ‘I would fain 
celebrate? a sense of ἐνδατεῖσθαι derived 
from that of distributing words (λόγους 
ὀνειδιστῆρας ἐνδατούμενος, Eur. Herc. 2. 
218). The bad sense occurs in 7rach, 
791 τὸ δυσπάρευνον λέκτρον ἐνδατούμενος : 
the good, only in Aesch. fr. 340 ὁ δ᾽ ἐν- 
δατεῖται τὰς ἑὰς εὐπαιδίας, ‘celebrates his 
happy race of children.’ 

206 προσταθέντα from προΐστημι, not 
προστείνω. Cp. At. 803 πρόστητ᾽ avay- 
kalas τύχης. Ll. 637 Φοῖβε προστατήριε. 
Ο.7. 881 θεὸν οὐ λήξω προστάταν ἴσχων. 
For 1st aor. pass. part., cp. κατασταθείς 
Lys. or. 24.9, συσταθείς Plato Legg. 685 C. 
Theconject.rpooradévra (as= ‘ launch- 
ed’) is improbable (1) because it would 
mean rather ‘having set out on a journey’; 
cp. O.C. 20: (2) on account of the meta- 
phor in dpwyd. προσταθέντα from προσ- ° 
telvw (a verb which does not occur) would 
scarcely mean ‘directed against the ene- 
my,’ but rather ‘strained against the bow- 
string.’ προσταχθέντα, found in one 


ΘΙΔΙΠΟΥΣ. ΤΥΡΆΝΝΟΣ 


day follows to accomplish this. 


39 
O thou who wieldest the 


powers of the fire-fraught lightning, O Zeus our father, slay him 


beneath thy thunder-bolt. 


Lycean King, fain were I that thy shafts also, from thy bent 3rd anti- 
bow’s string of woven gold, should go abroad in their might, our 
champions in the face of the foe; yea, and the flashing fires 
of Artemis wherewith she glances through the Lycian hills. 
And I call him whose locks are bound with gold, who is named 
with the name of this land, ruddy Bacchus to whom Bacchants 
cry, the comrade of the Maenads, to draw near with the blaze 


(found in E) ὦ πυρφόρον | ἀστραπὰν κράτει νέμων. 


Erfurdt. 


205 ἀδάμαστ᾽ MSS.: ἀδάματ᾽ 


206 προσταθέντα L, with gloss προϊστάμενα. Dindorf’s conjecture, προσ- 


ταχθέντα, stands in at least one late Ms. (B, 15th cent.), but the rest agree with L. 





MS., would make dpwyd prosaic, while 
προσταθέντα- ἰΐ not strictly suitable—is 
at least poetical: the difference is like 
that between speaking of ‘auxiliary forces’ 
. vand of ‘champions.’ 

207 ᾿Αρτέμιδος αἴγλας, the torches 
with which Artemis was represented,— 
holding one in each hand (Ar. Ran. 1362 
διπύρους ἀνέχουσα λαμπάδας, Trach. 214 
“Apreuw dudlrvpov),—in her character 
of Διύλύκη, σελασφόρος, φωσφόρος, ἀνθή- 
Xuos,—names marking her connection 
with Selene; cp. Aesch. fr. 164 ἀστερω- 
mov ὄμμα Λητῴας κόρης. 

208 Ait ὄρεα διάσσει as ἐλαφη- 
βόλος, ἀγροτέρα, huntress: Od. 6. 102 
οἵη δ᾽ “Aprems εἶσι κατ᾽ οὔρεος ἰοχέαιρα, | 
«τερπομένη κάπροισι καὶ ὠκείῃς ἐλάφοι- 
ow? | τῇδέ θ᾽ ἅμα νύμφαι. Δύκια: the 
Lycian hills are named here in order to 
associate Artemis more closely with her 
brother under his like-sounding name of 
Λύκειος. At Troezen there was even a 
temple of Ἄρτεμις Avxela: Paus. says 
(2. 31. 4) that he could not learn why 
she was so called (és δὲ τὴν ἐπίκλησιν 
οὐδὲν εἶχον πυθέσθαι παρὰ τῶν ἐξηγητῶν), 
and suggests that this may have been her 
‘title among the Amazons—a guess which 
touches the true point, viz. that the Av- 
κεία was a feminine counterpart of the 
Avxetos. 

209 τὸν χρυσομίτραν. μίτρα, asnood : 
-Eur. Bacch. 831 Al. κόμην μὲν ἐπὶ σῷ 
Kpart tavadv ἐκτενῶς TENOETZ. τὸ 
δεύτερον δὲ σχῆμα τοῦ κόσμου τί wo; ΔΙ. 
πέπλοι ποδήρεις᾽ ἐπὶ κάρᾳ δ᾽ ἔσται μίτρα. 

210 τᾶἄσδ᾽ ἐπώνυμον γᾶς. As he is 
-Baxxos, sois Thebescalled Βακχεία(7 rach. 
510), while he, on the other hand, was 
Kaductas νύμφας ἄγαλμα (1115). The 


mutual relation of the names is intended 
here by ἐπώνυμον. The word usually 
means called after (Tivos), But ἄρχων 
ἐπώνυμος, ἥρωες ἐπώνυμοι were those who 
gave names to the year, the tribes: and 
so Soph. Az. 574 (σάκος) ἐπώνυμον, the 
shield which gave its name to Eurysaces. 
Cp. Eur. 2072 1555 where Athena says, 
ἐπώνυμος δὲ σῆς ἀφικόμην χθονός, giving 
my name to thy land. 

211 οἰνῶπα.. εὔιον, ‘ruddy’—‘to whom 
Bacchants cry evot.? Note how in this 
passionate ode all bright colours (χρυ- 
σέας, εὐῶπα, χρυσοστρόφων, αἴγλας, xpu- 
σομίτραν, οἰνῶπα, ἀγλαώπιρ), and glad 
sounds (ἰήιε Παιάν, εὔιον), are contrasted 
with the baleful fires of pestilence and 
the shrieks of the dying. 

212 Μαινάδων ὁμόστολον Ξε στελλό- 
μενον ἅμα ταῖς Μαινάσιν, setting forth, 
roaming with the Maenads: Apoll. Rhod. 
2. 802 ὁμόστολος ὑμὶν ἕπεσθαι. The 
nymphs attendant on Dionysus, who 
nursed the infant god in Nysa, and after- 
wards escorted him in his wanderings, 
are called Μαινάδες, Θυιάδες, Βάκχαι. ἢ. 
6. 132 μαινομένοιο Διωνύσοιο τιθήνας | cede 
κατ᾽ ἠγάθεον Νυσήιον" αἱ δ᾽ ἅμα πᾶσαι | 
θύσθλα (2.6. thyrsi and torches) χαμαὲ 
κατέχευαν. Aesch. fr. 307 πάτερ Θέοινε, 
Μαινάδων ζευκτήριε, who bringest the 
Maenads under thy spell. 24. 22. 460 
μεγάροιο διέσσυτο, μαινάδι ἴση, | waddo- 
μένη κραδίην. Catullus 63. 23 capita 
Maenades vi iactunt hederigerae: as Pind. 
fr. 224 ῥιψαύχενι σὺν κλόνῳ. Lucian may 
have had our passage in mind, when he 
mentions the μίτρα and the Maenads 
together: Dial. D. 18 θῆλυς οὕτω,... μίτρᾳ 


“μὲν ἀναδεδεμένος τὴν κόμην, τὰ πολλὰ δὲ 


μαινομέναις ταῖς γυναιξὶ συνών. 


strophe 


40 TOPOKAEOYE 


11 ἀγλαῶπι « σύμμαχον > 


, SPTN \ $f nop > A , 
12 πεύκᾳ “ml τὸν ἀπότιμον ἐν θεοῖς θεόν. 


ΟἹ. 


205 


> A Δ 3 3 = ¥ 3 Ὁ ΄ » 
αἰτεῖς" ἃ δ᾽ αἰτεῖς, tay ἐὰν θέλῃς ἔπη 


, » “A ’ ε A 
κλύων δέχεσθαι τῇ νόσῳ θ᾽ ὑπηρετεῖν, 
/ δὺς oS 

ἀλκὴν λάβοις ὄν κανακούφισιν κακών 


c Ν ᾽’ὔ ἈΝ “A 4 vO 5 ἴω 
ἁγὼ ξένος μὲν τοῦ λόγου τοῦδ᾽ ἐξερώ, 
ἕένος δὲ τοῦ πραχθέντος" οὐ γὰρ ἂν μακρὰν 
Ν 3 


non 


ες  iyvevov αὐτός, μὴ οὐκ ἔχων τι σύμβολον. 
νῦν δ᾽, ὕστερος γὰρ ἀστὸς εἰς ἀστοὺς τελώ, 
ὑμῖν προφωνῶ πᾶσι Καδμείοις τάδε' 
ὅστις ποθ᾽ ὑμῶν Λάϊον τὸν Λαβδάκου 


κάτοιδεν ἀνδρὸς ἐκ τίνος διώλετο, 


414 ἀγλαῶπι πεύκᾳ MSS. 


225 


The metrical defect (cp. v. 201) is supplied by Wolff 





214 ἀγλαῶπι. A cretic has been lost. 
G. Wolff's σύμμαχον is simple and ap- 
propriate. Arndt’s conjecture, daig (‘de- 
stroying, consuming,’ prob. from rt. daf, 
to kindle, Curt. Atym. § 258), is sup- 
ported by the possibility of a corruption 
AAIAI having been rejected as a gloss 
on πεύκᾳ. Cp. Jl. 9. 347 δήϊον πῦρ, 
Aesch. 7heb. 222 πυρὶ dat. But in con- 
nection with the ‘blithe torch’ of Dio- 
nysus such an epithet is unsuitable. 

215 τὸν ἀπότιμον. See on ἀπόξενον 
196. Ares is ‘without honour’ among 
the gentler gods: cp. //. 5. 31 (Apollo 
speaks), “Apes, “Apes βροτολοιγέ, μιαιφόνε, 
τειχεσιπλῆτα: and 10. 890 where Zeus 
says to Ares, ἔχθιστός τέ μοι ἔσσι θεῶν, 
κιτιλ. So the Erinyes are στύγη θεῶν 
(Zum. 644); and the house of Hades is 
hateful even to the gods (//. 20. 65). 
—edv, one syll., by synizesis: cp. 1519. 

216- 462 First ἐπεισόδιον. Oedipus 
re-enters from the palace. He solemnly 
denounces a curse on the unknown mur- 
derer of Laius. The prophet Teiresias 
declares that the murderer is Oedipus. 

216 αἰτεῖς: Oedipus had entered in 
time to hear the closing strains of that 
prayer for aid against the pestilence which 
the Chorus had been addressing to the 
gods. ἃ δ᾽ αἰτεῖς. The place of λάβοις 
is against taking ἀλκὴν καἀνακούφισιν 
κακῶν as in apposition with &: rather 
the construction changes, and ἃ is left 
as an accus. of general reference. 

217 κλύων not strictly Ξε πειθαρχών, 
‘obediently’ (in which sense κλύειν takes 
gen., τῶν ἐν τέλει, At. 1352), but simply, 


‘on hearing them’: δέχεσθαι, as Phil, 
1321 κοῦτε σύμβουλον δέχει. Tap ems 
phatic by place: ‘you pray (to the gods): 
hear me and (with their help) you shall 
have your wish.’ τῇ νόσῳ ὑπηρετεῖν, = 
θεραπεύειν τὴν νόσον, to do that which 
the disease requires (for its cure), like 
ὑπηρετοίην τῷ παρόντι δαίμονι Zl. 1306. 
In Eur. fr. 84, 7 οὐδ᾽ αὖ πένεσθαι καξυ- 
πηρετεῖν τύχαις | οἷοί τε, Nauck now 
gives with Athenaeus 413 C καὶ ξυνηρετ- 
μεῖν. Acc. to the commoner use of the 
word, the phrase would mean 20 humour 
the disease, 2.6. obey morbid impulses: 
cp. Lysias Jn Eratosth. § 23 τῇ ἑαυτοῦ 
παρανομίᾳ προθύμως ἐξυπηρετῶν, eager- 
ly indulging the excess of his own law- 
lessness. 

218 ἀλκὴν, as well as ἀνακούφισιν, 
with κακῶν: Hes. Of. 199 κακοῦ δ᾽ οὐκ 
ἔσσεται ἀλκή: Eur. Aled. 1322 ἔρυμα πο- 
λεμίας χερός: below 1200 θανάτων... πύρ- 
γος. 

219-- 228 ἁγὼ ξένος μὲν... τάδε. Οε- 
dipus has just learned from Creon that 
Laius was believed to have been mur- 
dered by robbers on his way to Delphi, 
but that, owing to the troubles caused 
by the Sphinx, no effective search had 
been made at the time (114—131). He 
has at once resolved to take up the mat- 
ter—both because Apollo enjoins it; and 
as a duty to the Theban throne (255). 
But the murder occurred before he had 
come to Thebes. He must therefore ap- 
peal for some clue—ovpBorov—to those 
who were at Thebes when the rumour 
was fresh. 


OIAITOYS ΤΎΡΑΝΝΟΣ 41 


of his blithe torch, our ally against the god unhonoured among 
gods. 


ΟΕ. Thou prayest: and in answer to thy prayer,—if thou 
wilt give a loyal welcome to my words and minister to thine 
own disease,—thou mayest hope to find succour and relief from 
woes. These words will I speak publicly, as one who has been 
a stranger to this report, a stranger to the deed ; for I should not 
be far on the track, if I were tracing it alone, without a clue. 
But as it is,;—since it was only after the time of the deed 
that I was numbered a Theban among Thebans,—to you, the 


Cadmeans all, I do thus proclaim. 
Whosoever of you knows by whom Laius son of Labdacus 


with σύμμαχον. 


was slain, 


221 αὐτὸ L: αὐτὸς r (including A). 





219 ξένος, ‘a stranger’ to the affair, is 
tinged with the notion, ‘unconne&ted 
with Thebes’: and this is brought out by 
ἀστὸς in 222. For other explanations of 
the passage, see Appendix. 

220 τοῦ πραχθέντος, the murder. 
Not, ‘what was dove at the time by way of 
search’: for (a) τὸ πραχθέν, as opp. to ὁ 
λόγος, must mean the ἔργον to which the 
λόγος is related: (ὁ) Oed. has lately ex- 
pressed his surprise that nothing effective 
was done (128), and could not, therefore, 
refer with such emphasis to τὸ πραχθέν in 
this sense. 

220f. ov γὰρ dv μακρὰν txvevov. In 
his Syntax of the Moods and Tenses of the 
Greek Verb (1889), § 511, Prof. Goodwin 
deals with this passage. His view agrees 
with that given in my second ed., so far 
as concerns two points, viz.: (1) that the 
chief protasis is not contained in μὴ οὐκ 
ἔχων: and (2) that μὴ οὐκ ἔχων is still neces- 
sarily conditional. But his analysis of the 
whole is simpler; it is as follows. 

The chief protasis is contained in the 
word αὐτός, ‘unaided,’ which is equiva- 
lent to, ef μόνος ἔχνευον, if I were at- 
tempting to trace it alone. [I had said 
that αὐτός ‘implies the protasis’; but had 
taken the protasis itself to be, ef μὴ ἐξεῖ- 
mov, supplied from ἐξερῶ: tf J had not 
thus spoken,—appealing to you for help.] 
Then, μὴ οὐκ ἔχων is equivalent to εἰ μὴ 
εἶχον. Now, the difficulty here seemed 
to be that εἰ μὴ εἶχον would imply, ‘but 
I have a clue’: whereas, in fact, he has 
none. [I met this by suggesting that 
μὴ οὐκ ἔχων expresses the fact (of his 
having no clue), not simply as a fact, but 
as a condition,—‘ zz a case where I had no 


clue’; being equivalent, not to εἰ μὴ εἶχον, 
but rather to ὅτε μὴ εἶχον. Goodwin’s 
answer is that the conditional sentence, 
written in full, would stand thus,—(1r) and 
(2) denoting respectively the chzef prota- 


. sis, and the sudordinate protasis: (1) εἰ 


μόνος txvevov, οὐκ ἂν μακρὰν ἴχνευον, (2) 
εἰ μὴ εἶχόν τι σύμβολον. Now (1) is an 
unreal supposition (he is mot tracking 
alone); and that makes the whole suppo- 
sition unreal. εἰ μὴ εἶχον is here a part 
of that unreal supposition; and therefore 
it can have that form, although, as a fact, 
he has noclue. (Suppose it to be said of 
a man too old for work : ‘ // he were young, 
he would not be doing well, tf he did not 
work’: εἰ νέος ἦν, οὐκ ἂν εὖ ἐποίει, εἰ μὴ 
ἐπόνει. The chief protasis, εἰ νέος ἦν, being 
unreal, makes all the rest unreal. The 
fact is, οὐ πονεῖ: and εἰ μὴ ἐπόνει does not 
imply, πονεῖ. Compressed, this would be, 
οὐκ ἂν εὖ ἐποίει νέος ὦν, μὴ οὐ πονῶν.) 

αὐτός, unaided: cp. //. 13. 729 ἀλλ᾽ 
οὔπως ἅμα πάντα δυνήσεαι αὐτὸς ἑλέσθαι. 

222 νῦν δ᾽, ‘but asit is’: z¢., ‘since 
it would be vain to attempt the search 
alone—since I came to Thebes only after 
the event.’ ὕστερος, sc. τοῦ πραχθέντος : 
for the adj. instead of δὴ adv., cp. Az. 
217 νύκτερος...ἀπελωβήθη: Ll. τ. 424 χθι- 
fds ἔβη: Xen. An. 1. 4. 12 τοῖς προτέροις 
(=mpébrepov) μετὰ Κύρου ἀναβᾶσι. εἰς 
ἀστοὺς τελῶ, inter cives censeor: ἃ 
metaphor from being rated (for taxation) 
in a certain class: Her. 6. 108 εἰς Boww- 
τοὺς rekéew: Eur. Bacch. 822 ἐς γυναῖκας 
ἐξ ἀνδρὸς TENG. ἀστὸς εἰς ἀστοὺς, like 
At. 267 κοινὸς ἐν κοινοῖσι: 16. 467 ξυμπε- 
σὼν μόνος μόνοις: Ph. 135 ἐν ξένᾳ ξένον: 
tb. 633 ἴσος ὧν ἴσοις ἀνήρ. 


42 ΣΟΦΟΚΛΕΟΥΣ 


~ ’ , 
τοῦτον κελεύω πάντα σημαίνειν ἐμοί: 


Kei μὲν φοβεῖται, τοὐπίκλημ᾽ 


* 


® ὑπεξελεῖν 


> | A > «ε “~ 4 A » ΑἉ 
αὐτὸν καθ αὑτοῦ" πείσεται γὰρ ἄλλο μὲν 


ἀστεργὲς οὐδέν, γῆς δ᾽ ἄπεισιν ἀβλαβής" 


εἰ δ᾽ αὖ τις ἄλλον οἷδεν ἐξ ἄλλης χθονὸς 
τὸν αὐτόχειρα, μὴ σιωπάτω" τὸ γὰρ 
κέρδος τελῶ ᾽γὼ χὴ χάρις προσκείσεται.- 


230 


adh 


> Δ 
εἰ δ᾽ αὖ σιωπήσεσθε, καί τις ἢ φίλου 
, ¥ “ἡ “ 
δείσας ἀπώσει τοὔπος ἢ χαύὐτοῦ τόδε, 


ac τῶνδε δράσω, ταῦτα χρὴ κλύειν ἐμοῦ. 


235 


τὸν ἀνδρ᾽ ἀπαυδῶ τοῦτον, ὅστις ἐστί, γῆς 
τῆσδ᾽, ἧς ἐγὼ κράτη τε καὶ θρόνους νέμω, 
μήτ᾽ ἐσδέχεσθαι μήτε προσφωνεῖν τινα, 
μήτ᾽ ἐν θεῶν εὐχαῖσι μήτε θύμασιν 


\ 
κοινὸν ποεῖσθαι, 


227 τ. ὑπεξελὼν αὐτὸς MSS. 
Blaydes) αὐτόν. 


μήτε χέρνιβος νέμειν" 


I read ὑπεξελεῖν (already proposed by K. Halm and 
229 ἀσφαλής L, with yp. ἀβλαβής in margin. 


240 


Most of the later 


MSS. (including A) have ἀβλαβής, which is the reading of the Aldine, Brunck, Her- 
mann, Elmsley, Linwood, Wunder, Blaydes, Kennedy: while among the editors who 
prefer ἀσφαλής are Schneidewin, Nauck, Dindorf (with the admission, ‘hic tamen aptius 





227 £. Kel μὲν φοβεῖται τοὐπίκλημ᾽ 
ὑπεξελὼν | αὐτὸς καθ᾽ αὑτοῦ is the read- 
ing of all the Mss.: for the ὑπεξελθὼν of 
the first hand in one Milan ms. of the 
early 14th cent. (Ambros. L 39 sup., 
Campbell’s M?) is a mere slip. I read 
ὑπεξελεῖν | αὐτὸν καθ᾽ αὑτοῦ, the change 
of αὐτὸν into αὐτὸς having necessarily 
followed that of ὑπεξελεῖν into ὑπεξελὼν 
due to an interpretation which took the 
latter with φοβεῖται. Cp. Thuc. 4. 83 
(Arrhibaeus, the enemy of Perdiccas, 
makes overtures to Brasidas, and the 
Chalcidians exhort Brasidas to listen): 
ἐδίδασκον αὐτὸν μὴ ὑπεξελεῖν τῷ Περ- 
δίκκᾳ τὰ δεινά, ‘they impressed upon 
him that he must not remove the dangers 
Jrom the path of Perdiccas’—by repulsing 
the rival power of Arrhibaeus. ὑπεξε- 
λεῖν τὰ Sewa=to take them away (ἐκ) 
Jrom under (ὑπό) the feet,—from the path 
immediately before him: τῴ Περδίκκᾳ 
being a dat. commodi. Similarly Her. 7. 
8 τούτων... ὑπεξαραιρημένων, ‘when these 
have been taken out of the way.’ So 
here: kel μὲν φοβεῖται, and if heis afraid 
(as knowing Aimse/f to be the culprit), 
then 71 bid him (κελεύω continued from 
226) ὑπεξελεῖν τὸ ἐπίκλημα fo take the 
peril of the charge out of his path, αὐτὸν 


καθ᾽ αὑτοῦ (σημαίνοντα) dy speaking 
against himself. If the culprit is de- 
nounced by another person, he will be 
liable to the extreme penalty. If he 
denounces himself, he will merely be 
banished. By denouncing himself, he 
forestalls the danger of being denounced 
by another. For other explanations, see 
Appendix. 

229 ἀβλαβής, the reading of A and 
most MSs., ‘without damage,’ ἀζήμιος, is 
far more suitable than ἀσφαλής to this 
context: and Soph. has the word as a 
cretic in Z/. 650 ζῶσαν ἀβλαβεῖ βίῳ. 
Although in L ἀσφαλής appears as the 
older reading, so common a word was 
very likely to be intruded; while it would 
be difficult to explain how the compara- 
tively rare ἀβλαβής could have supplanted 
it. A metrical doubt may have. first 
brought ἀσφαλής in. Dindorf, reading 
ἀσφαλής, recognises the superior fitness of 
ἀβλαβής here, and thinks that it may be 
the true reading, even though its ap- 
pearance in the margin of L was due to 
conjecture. 

230 ἀλλον...ἐξ ἄλλης χθονὸς, ‘another 
[i.c. other than one of yourselves, the 
Thebans] from a strange land’: an alien, 
whether resident at Thebes, or not: cp, 


ΟΙΔΙΠΟῪΣ TYPANNOS 43 


I bid him to declare all to me. And if he is afraid, I tell 
him to remove the danger of the charge from his path by 
denouncing himself; for he shall suffer nothing else unlovely, 
but only leave the land, unhurt. Or if any one knows an alien, 
from another land, as the assassin, let him not keep silence ; for 
I will pay his guerdon, and my thanks shall rest with him besides. 

But if ye keep silence—if any one, through fear, shall seek 
to screen friend or self from my behest—hear ye what I then 
shall do. I charge you that no one of this land, whereof 
I hold the empire and the throne, give shelter or speak word 
unto that murderer, whosoever he be,—make him partner 
of his prayer or sacrifice, or serve him with the lustral rite; 


videtur ἀβλαβής), Wecklein, Wolff, Tournier, Campbell, White. 230 ἐξ ἄλλης 
χθονὸς] For ἐξ, Vauvilliers conj. ἢ ᾽ξ: Seyffert, ἐξ ἀμῆς : but see note. 289 μήτε 
θύμασιν] μηδὲ θύμασιν Elmsley. 240 χέρνιβοσ was written by the rst hand in 
L (and occurs in at least one later MS., L?, cod. Laur. 31. 10), but was changed by 





451 οὗτός ἐστιν ἐνθάδε, | ξένος λόγῳ μέτ- 
οἰκος. The cases contemplated in the 
proclamation (223—235) are (1) a Theban 
denouncing another Theban, (2) a Theban 
denouncing himself, (3) a Theban de- 
nouncing an alien. 

231 τὸ κέρδος, the (expected) gain, τὰ 
μήνυτρα. Trach. 191 ὅπως | πρὸς σοῦ τι 
κερδάναιμι καὶ κτῴμην χάριν. 

292 προσκείσεται, will be stored up 
besides (cp. Eur. Alc. 1039 ἄλγος ἄλγει... 
προσκείμενον, added). χάρις κεῖται is 
perf. pass. of χάριν τίθεμαι or κατατίθεμαι 
(τινί or παρὰ twl),—a metaphor from de- 
posits of money: τὰ χρήματα... κείσθω 
map ols τισιν ἂν ὑμῖν δοκῇ [Plat.] Zpist. 
346 C. 

238 f. φίλου, αὑτοῦ, with ἀπώσει only 
(71. 15. 503 ἀπώσασθαι κακὰ vnov).—Bet- 
cas φίλου 45-- δείσας ὑπὲρ φίλου (like κή- 
δομαι, φροντίζειν) would be too harsh, and 
rhythm is against it. τοὔπος...τόδε, this 

command to give up the guilty. 

_ 286-- 240 ἀπαυδῶ (ἀπ-, because the 
first clauses are negative), I command, 
(uh) τινα γῆς τῆσδε that no one belong- 
ing to this land, μήτ᾽ ἐσδέχεσθαι μήτε 
προσφωνεῖν shall either entertain or 
accost, τὸν ἄνδρα τοῦτον, ὅστις ἐστί. 
For the gen. γῆς, cp. Plat. Prot. 316 B 
Ἱπποκράτης ὅδε ἐστὶ μὲν τῶν ἐπιχωρίων, 
᾿Απολλοδώρου υἱός, οἰκίας μεγάλης καὶ 
εὐδαίμονος. Since μήτε... μήτε in 238 
connect ἐσδέχεσθαι and προσφωνεῖν, we 
require either (2) separate verbs for εὖ- 
χαῖσι and θύμασιν, or (ὁ) as Elms. pro- 
posed, μηδὲ instead of μήτε before θύμα- 
ow. Cp. O. C. 1297, where in a similar, 


though simpler, sentence I receive Her- 
mann’s οὐδ᾽ for οὔτ΄. Here, however, I 
hesitate to alter, because the very fact 
that μήτε has already been thrice used 
might so easily have prompted its use 
(instead of μηδέ) before θύμασιν. As the 
MS. text stands, we must suppose a μήτε 
suppressed before evxatow,: the constr. 
being μήτε κοινὸν ποιεῖσθαι [μήτε] ἐν... 
εὐχαῖσι μήτε θύμασιν. Cp. Aesch. Ag. 
532 Πάρις γὰρ οὔτε συντελὴς πόλις: Cho. 
204 δέχεσθαι δ᾽ οὔτε συλλύειν τινά. 

240 κοινὸν Πειε-ε κοινωνόν, cp. Ai. 
267 ἦ κοινὸς ἐν κοινοῖσι λυπεῖσθαι ξυνών. 
Plat. Legg. 868 E (the slayer) ξυνέστιος 
αὐτοῖς μηδέποτε γιγνέσθω μηδὲ κοινωνὸς 
ἱερῶν. χέρνιβος (partitive gen.) is more 
suitable than χέρνιβας to the idea of ex- 
clusion from all fellowship in ordinary 
worship: χέρνιβας νέμειν would rather 
suggest a special κάθαρσις of the homi- 
cide. When sacrifice was offered by the 
members of a household (κοινωνὸν εἶναι 
χερνίβων...κτησίον βωμοῦ πέλας Aesch. 
Ag. 1037) or of a clan (χέρνιψ φρατέρων 
Lum. 656), a brand taken from the altar 
was dipped in water, and with the water 
thus consecrated (χέρνιψ) the company 
and the altar were sprinkled: then holy 
silence was enjoined (εὐφημία ἔστω) : and 
the rite began by the strewing of barley 
meal (οὐλοχύται!) on altar and victim. 
(Athenaeus 409: Eur. 4. F. 922 ff.) 
Acc. to Dem. Adv. Left. ὃ 158 a law of 
Draco prescribed χέρνιβος [so the best 
MSS.: Ὁ. ὦ. χερνίβων] εἴργεσθαι τὸν ἀνδρο- 
φόνον, σπονδῶν, κρατήρων, ἱερῶν, ἀγορᾶς. 
This was a sentence of excommunication 


44 ZOPOKAEOYS 


3 A > ree J ¥ ’ ε , 

ὠθεῖν δ᾽ am οἴκων πάντας, ὡς μιάσματος 
Ags Εν ¥ ε Ν \ A“ 

τοῦδ᾽ ἡμὶν ὄντος, ὡς τὸ Πυθικὸν θεοῦ 

μαντεῖον ἐξέφηνεν ἀρτίως ἐμοί. 

ἐγὼ μὲν οὖν τοιόσδε τῷ τε δαίμονι 


τῷ T ἀνδρὶ τῷ θανόντι. σύμμαχος πέλω" 


245 


κατεύχομαι δὲ τὸν δεδρακότ᾽, εἴτε τις 
εἷς ὧν λέληθεν εἴτε πλειόνων μέτα, 
κακὸν κακῶς νιν ἄμορον ἐκτρῖψαι βίον. 


ἐπεύχομαι ΝῊ 


οἴκοισιν εἰ ξυνέστιος 
ἐν τοῖς ἐμοῖς γένοιτ᾽ ἐμοῦ συνειδότος, — - 


250 


παθεῖν ἅπερ τοῖσδ᾽ ἀρτίως ἠρασάμην. 
ὑμῖν δὲ ταῦτα πάντ᾽ ἐπισκήπτω τελεῖν 
ὑπέρ T ἐμαυτοῦ τοῦ θεοῦ τε τῆσδέ τε 
γῆς ὧδ᾽ ἀκάρπως καἀθέως ἐφθαρμένης. 


οὐδ᾽ εἰ γὰρ ἣν τὸ πρᾶγμα μὴ θεήλατον, 


259 


ἀκάθαρτον ὑμᾶς εἰκὸς ἦν οὕτως ἐάν, 
ἀνδρός γ᾽ ἀρίστου βασιλέως T ὀλωλότος, 
ἀλλ᾽ ἐξερευνᾶν" νῦν δ᾽, ἐπεὶ κυρῶ T eyo 
ἔχων μὲν ἀρχὰς as ἐκεῖνος εἶχε πρίν, 


an early hand to χέρνιβασ, which is in almost all the later Mss. 


248 κακὸν κακῶσ 


νιν κἄμοιρον ἐκτρίψαι βίον 1, st hand: the «x before ἄμοιρον was afterwards erased. 


One of the later Mss. (B) has κἄμοιρον, and all seem to have ἄμοιρον. 


ἄμορον Porson. 


257 βασιλέως 7’) The 1st hand in L had joined the or in one character (cp. on v. 





(1) from the life of the family and the 
clan, (2) from the worship common to all 
Hellenes, who, as opposed to βάρβαροι, are 
(Ar. Lys. 1129) of μιᾶς ἐκ χέρνιβος | βωμοὺς 
περιρραίνοντες, ὥσπερ ξυγγενεῖς, | ᾽Ολυμ- 
πίασιν, ἐν Πύλαις, Πυθοῖ. The mere pre- 
sence of the guilty could render sacrifice 
‘inauspicious: Antiph. De Caed. Her. § 82 
ἱεροῖς παραστάντες πολλοὶ δὴ καταφανεῖς 
ἐγένοντο οὐχ ὅσιοι ὄντες καὶ διακωλύοντες 
τὰ ἱερὰ μὴ γίγνεσθαι (bene succedere) τὰ 
νομιζόμενα. 

241 ὠθεῖν δὲ, sc. αὐδῶ, understood from 
the negative ἀπαυδῶ: cp. Her. 7. 104 οὐκ 
ἐών φεύγειν... ἀλλὰ ἐπικρατέειν. 

246. 251 ‘These six verses are placed 
by some editors between 272 and 273. 
See Appendix. | 

246 κατεύχομαι. Suidas κατεύχεσ- 
θαι" τὸ καταρᾶσθαι. οὕτω Πλάτων. καὶ 
Σοφοκλῆς, κατεύχομαι δὲ τὸν δεδρακότα 
τάδε. Phot. Lex. p. 148. * κατεύχεσθαι 
τῶν ᾿Αχαιών' ἀντὶ τοῦ κατὰ τῶν ᾿Αχαιών 
εὔχεσθαι. οὕτως Σοφοκλῆς. HEH re the ref. 


is to Plato Rep. 393 E τὸν δὲ (the Homeric 
Chryses, priest of Apollo)...caretvxeoOau 
τῶν ᾿Αχαιῶν πρὸς θεόν. But Photius pre- 
fixes the words, κατεύχεσθαι" τὸ καταρᾶσ- 
θαι. οὕτως Πλάτων. It is clear, then, 
that in Photius οὕτως Σοφοκλῆς and οὕτως 
Πλάτων have changed places. The ‘Soph. 
fr. 894,” quoted by Lidd. and Scott under 
κατεύχομαι as=imprecari, thus vanishes 
(Nauck Fragm. 7rag.? p. 357). Cp. Aesch 
Theb. 632 πόλει | οἵας ἀρᾶται καὶ κατεύ- 
χεται τύχας. But where, as here, κατεύ- 
Xopat is used without gen. (or dat.), it is 
rather fo pray solemnly: often, however, 
in a context which ¢mlies imprecation: 
e.g. Plat. Legg. 935 A κατεύχεσθαι ἀλλή- 
λοις ἐπαρωμένους: Rep. 394 A κατεύχετο 
τῖσαι τοὺς ᾿Αχαιοὺς τὰ ἃ δάκρυα. elre Tis: 
whether the unknown man (tts) who has 
escaped discovery is εἷς, alone in the 
crime, or one of several. tus, because 
the person is indefinite: cp. 107. 

248 viv dpopov: Porson (prac/. Hee. 
p- ix.) defends the redundant νιν by 


OIAITOYS TYPANNOS 45 


but that all ban him their homes, knowing that ¢hzs is our de- 
filing thing, as the oracle of the Pythian god hath newly shown 
me. I then am on this wise the ally of the god and of the slain. 
And I pray solemnly that the slayer, whoso he be, whether his 
hidden guilt is lonely or hath partners, evilly, as he is evil, may 
wear out his unblest life. And for myself I pray that if, with 
my privity, he should become an inmate of my house, I may 
suffer the same things which even now I called down upon 


others. 


And on you [ lay it to make all these words good, for 


my sake, and for the sake of the god, and for our land’s, thus 
blasted with barrenness by angry heaven. 

For even if the matter had not been urged on us by a god, it 
was not meet that ye should leave the guilt thus unpurged, 
when one so noble, and he your king, had perished ; rather were 


ye bound to search it out. 


And now, since ’tis I who hold the 


powers which once he held, 


134). An early hand (perhaps that of the first corrector) afterwards erased the 7’, and 


then wrote it separately from the o. Some later Mss. omit the 7’. 


258 κυρώ τ᾽ MSS.: 


κυρῶ γ᾽ T. F. Benedict (Odservationes in Soph., Lips., 1820: cp. Blaydes ad Joc.). 





Trach. 287 αὐτὸν δ᾽ ἐκεῖνον, εὖτ᾽ ἂν 
ἁγνὰ θύματα | ῥέξῃ πατρῴῳ Ζηνὶ τῆς ἁλώ- 
σεως, | φρόνει νιν ὡς ἥξοντα. The form 
ἄμορος occurs in Eur. Med. 1395 (where 
ἄμοιρος is a Ὁ. 1); ἄμμορος in Hee. 421, 
Soph. Pil. 182. κακὸν κακώς: Prz/. 
1369 ἔα κακῶς αὐτοὺς ἀπόλλυσθαι κακούς. 
Ar. Plut. 65 ἀπό σ᾽ ὀλῶ κακὸν κακῶς. 

249 ἐπεύχομαι, imprecate on mysel/: 
Plato Critias 120 B ταῦτα ἐπευξάμενος 
ἕκαστος αὐτῶν αὑτῷ καὶ τῷ ἀφ᾽ αὑτοῦ 
γένει. οἴκοισιν.. ξυνέστιος : not tautolo- 
gical, since ξυνέστιος is more than ἔνοικος, 
implying admission to the family worship 
at the ἑστία and to the σπονδαί at meals. 
Plat. Lege. 868 E ἱερῶν μὴ κοινωνείτω 
pnde...Evvéotios αὐτοῖς μηδέποτε γιγ- 
νέσθω μηδὲ κοινωνὸς ἱερῶν. Plat. Zuthy- 
phro 4 8 καὶ εἰ μὲν ἐν δίκῃ [ἔκτεινεν], ἐᾶν, 
if he slew the man justly, forbear; εἰ δὲ 
μή, ἐπεξιέναι (prosecute the slayer), ἐάν- 
περ ὁ κτείνας συνέστιός cot Kal ὁμο- 
τράπεζος ἧ. ἴσον γὰρ τὸ μίασμα γίγνεται, 
ἐὰν ξυνῇς τῷ τοιούτῳ ξυνειδὼς καὶ 
μὴ ἀφοσιοῖς σεαυτόν τε καὶ ἐκεῖνον τῇ δίκῃ 
ἐπεξιών. 

251 τοῖσδ᾽, the slayer or slayers (247): 
see on 246. 

254 ἀκάρπως καἀθέως: Zi, 1181 ὦ 
σῶμ᾽ ἀτίμως κἀθέως ἐφθαρμένον : below 
661 ἄθεος, ἄφιλος, forsaken by gods and 
men. 

256 εἰκὸς ἦν. The imperfect indic. of 


‘ but which is not so.—ovtws, 


a verb denoting obligation (ἔδει, χρῆν, 
προσῆκεν, εἰκὸς ἣν), when joined without 
av to an infinitive, often implies a condi- 
tional sentence with imperfect indic. in 
protasis and apoddsis: ¢.g. οὐκ εἰκὸς ἦν 
ἐᾶν =ovx ἂν εἰᾶτε (εἰ τὰ δέοντα ἐποιεῖτε), 
you would not (now) be neglecting it (if 
you did your duty): Xen. Mem. 2.7. 10 
εἰ μὲν τοίνυν αἰσχρόν τι ἔμελλον ἐργάσεσθαι 
ΠΕῚ were now intending—as I am not], 
θάνατον ἀντ᾽ αὐτοῦ προαιρετέον ἦν, τε 
προῃρούμην av (εἰ τὰ δέοντα ἐποίουν). 
Thuc. 6. 78 καὶ μάλιστα εἰκὸς ἦν ὑμᾶς... 
προορᾶσθαι, Ξεπροεωρᾶτε ἂν εἰ τὰ εἰκότα 
ἐποιεῖτε. So ἐβουλόμην, ἠξίουν, without 
ἄν, of that which one wishes were true, 
in this 
(careless) manner: cp. O. C. 1278 ws μή 
μ᾽ ἄτιμον... | οὕτως apy με: Ant. 315, Ph. 
1067. 

257 βασιλέως τ᾽: τε is to be retained 
after. βασιλέως, because (1) there is a 
climax, which is destroyed if βασιλέως 
stands merely in apposition with ἀνδρὸς 
ἀρίστου : (2) ἀνδρὸς ἀρίστου represents the 
claim of birth and personal merit, as βασι- 
λέως represents the special claim of a king 
on his people. Cp. Phil. 1302 ἄνδρα πολέ- 
puov | ἐχθρόν τε. 

258 κυρῶ τ᾽ ἐγὼὠ-Ξ ἐγώ τε κυρῷ, an- 
swered by κοινῶν τε, Κιτιλ, For τε so 
placed cp. 212. 249 ἔρροι τ᾽ ἃν αἰδὼς | ἁπάν- 
των F εὐσέβεια θνατῶν. 


46 


ἔχων͵ δὲ λέκτρα καὶ γυναῖχ᾽ ὁμόσπορον, 


ΣΟΦΟΚΛΕΟῪΣ 


κοινῶν τε παίδων κοίν᾽ ἄν, εἰ κείνῳ ee wee 
μὴ ᾿δυστύχησεν, ἦν ἀν ἐκπεφυκότα, 

νῦν δ᾽ ἐς τὸ κείνου κρᾶτ' ἐνήλαθ' ἡ τύχη" 

ἀνθ᾽ ὧν ἐγὼ τάδ᾽, ὡσπερεὶ τοὐμοῦ πατρός, 


ὑπερμαχοῦμαι, κἀπὶ πάντ᾽ ἀφίξομαι 


ζητῶν τὸν αὐτόχειρα τοῦ φόνου λαβεῖν 

τῷ Λαβδακείῳ παιδὶ Πολυδώρου τε καὶ 

τοῦ πρόσθε Κάδμου τοῦ πάλαι T ᾿Αγήνορος. 
καὶ ταῦτα τοῖς μὴ δρῶσιν εὔχομαι θεοὺς 


Xv 


μήτ᾽ ἄροτον αὐτοῖς γῆς ἀνιέναι τινὰ 


270 


μήτ᾽ οὖν γυναικῶν παῖδας, ἀλλὰ τῷ πότμῳ 
τῷ νῦν φθερεῖσθαι κἄτι τοῦδ᾽ ἐχθίονι" 


260 ἔχων δὲ] ἔχω δὲ L ist hand; an early hand added », 





260 ὁμόσπορον = ὁμοίως σπειρομένην, 
tt. ἣν καὶ ἐκεῖνος ἔσπειρε: but in 460 πα- 
Tpos | ὁμόσπορος = ὁμοίως (τὴν αὐτὴν) σπεί- 
ρων. ὁμογενής in 1361 is not similar. 

τ 261 κοινῶν παίδων κοινὰ ἦν ἂν ἐκπε- 
φυκότα, common things of (-Ξ ἐΐές con- 
sisting 22) kindred children would have 
been generated: =xkowGy παίδων κοινὴ φύσις 
ἐγένετο ἄν, a brood, common to Laius 
and Oedipus, of children akin to each 
other (as having the same mother, Io- 
casta) would have issued: ‘children born 
of one mother would have made ties be- 
tween him and me.’ For ἄν doubled 
CP- 139, 339: κοινῶν = ἀδελφῶν, ὁμαίμων 
(Ant. τ ὦ κοινὸν αὐταδελῴφον᾽ Ισμήνης κάρα). 
The language of this passage is carefully 
framed so as to bear a second meaning, 
of which the speaker is unconscious, but 
which the spectators can feel: locasta 
‘ has actually borne children to her own 
son Oedipus: thus in κοινῶν παίδων 
κοινὰ... ἐκπεφυκότα, the obvious sense of 
κοινά, ‘common to Laius and Oedipus,’ 
has behind it a second sense, in which it 
hints at a brood who are brothers and 
sisters of their own sire: see below 1403 f. 
This subtle emphasis—so ghastly, ξυνε- 
toiow—of the iteration in κοινῶν κοινά 
must not be obliterated by ees 
κοίν᾽ ἄν into κύματ᾽ (Nauck) or σπέρματ 
(Blaydes). Similarly, εἰ κείνῳ γένος | μηὴ 
᾽δυστύχησεν, is susceptible of the sense— 
‘if his son (Oed. himself) had not been 
ill-fated.’ κείνῳ γένος ἐδυστύχησε (his 
hope of issue was disappointed) is here a 
bold phrase for κεῖνος ἐδυστύχησε τὰ περὶ 


γένος : for Oed. is not xow supposed to 
know the story of the exposed babe (see 
Vi OE C . Eur. Andr. 418 πᾶσι δ᾽ 
ἀνθρώποις ap ἦν | ψυχὴ τέκν᾽" ὅστις δ᾽ αὔτ᾽ 
ἄπειρος ὧν ψέγει, | ἧσσον μὲν ἀλγεῖ, δυσ- 
τυχών δ᾽ εὐδαιμονεῖ: εὖ. 7τἰι ἥ στεῖρος 
οὖσα μόσχος οὐκ ἀνέξεται | τίκτοντας ἄλ- 
λους, οὐκ ἔχουσ᾽ αὐτὴ τέκνα" [ ἀλλ᾽ εἰ τὸ 
κείνης δυστυχεῖ παίδων πέρι, κ.τ.λ.: 
Suppl. 66 εὐτεκνία opp. to δυστυχία. 

263 νῦν δ᾽, ‘but as it is,’ with aor. 
equivalent to a "perf, 85; ()..|0.: 84, 371. 
Cp. below 948 καὶ viv ὅδε ] πρὸς τῆς 
τύχης ὄλωλε. So with Azstoric pres., Lys. 
in Erat. ὃ 36 εἰ μὲν οὖν ἐν τῷ δικαστηρίῳ 
ἐκρίνοντο, ῥᾳδίως ἂν ἐσῴζοντο"... viv δ᾽ εἰς 
τὴν βουλὴν εἰσάγουσιν. -ἐνήλατο: ἔ. 6. he 
was cut off by a timeless fate, leaving no 
issue, Cp. 1300: Ant. 1345 ἐπὶ κρατί μοι} 
πότμος.. «εἰσήλατο : so the Erinyes say, 
μάλα γὰρ οὖν ἁλομένα | ἀνέκαθεν βαρυ- 
πεσῆ | καταφέρω ποδὸς ἀκμάν Aesch. 
Eum. 369, Ag. 1175, δαίμων ὑπερβαρὴς 
ἐμπίτνων : Pers. 515 ὦ δυσπόνητε δαῖμον, 
ὡς ἄγαν βαρὺς Ἱ ποδοῖν ἐνήλλου παντὶ 
Περσικῷ γένει. The classical constr. with 
ἐνάλλομαι, as with ἐνθρῴσκω and ἐμπηδάω, 
is usually the dat., though εἰς with accus. 
occurs in later Greek ; a point urged by 
Deventer in his objections to this verse, 
which is, however, clearly sound. 

264 ἀνθ᾽ dy, properly wherefore (0.C. 
1295): here, cherefore. The protasis ἐπεὶ 
κυρώ (258) required, an apodosis intro- 
duced by ἀντὶ τούτων : but the parenthesis 
νῦν δ᾽ és τὸ κείνου k.T-X. (263) has led to 
ὧν being irregularly substituted for rov- 


OIAITOYS ΤΎΡΑΝΝΟΣ 47 


who possess his bed and the wife who bare seed to him; and 
since, had his hope of issue not been frustrate, children born of 
one mother would have made ties betwixt him and me—but, 
as it was, fate swooped upon his head ; by reason of these things 
will I uphold this cause, even as the cause of mine own sire, 
and will leave nought untried in seeking to find him whose 
hand shed that blood, for the honour of the son of Labdacus 
and of Polydorus and elder Cadmus and Agenor who was of old. 

And for those who obey me not, I pray that the gods send 
them neither harvest of the earth nor fruit of the womb, but that 
they be wasted by their lot that now is, or by one yet more dire. 


261 κοινῶν τε] καὶ νῷν τὰ M. Schmidt. 


270 γῆν L: γῆς Vauvilliers. 





των. Cp. 1466: Antiphon De Caed. 
Herod. § 11 δέον σε διομόσασθαι κ.τ.λ... ἃ 
σὺ παρελθών, where the length of the 
protasis has similarly caused α΄ to be 
substituted for ταῦτα. Distinguish from 
this the use of ἀνθ᾽ ὧν, by ordinary attrac- 
tion, for ἀντὶ τούτων ἃ or ὅτι, =decause, 
Ant. 1068.—r1d5’, cogn. acc. to ὑπερ- 
μαχοῦμαι as “411. 1346 σὺ ταῦτ᾽ ᾿Οδυσσεῦ 
τοῦδ᾽ ὑπερμαχεῖς ἐμοί; Cp. 71. 5. 185 οὐχ 
ὅ γ᾽ ἄνευθε θεοῦ τάδε μαίνεται. Brunck, 
Nauck and Blaydes adopt Mudge’s conj. 
τοῦδ᾽. But the Mss. agree in the harder 
and more elegant reading. 

265 ὑπερμαχοῦμαι only here: in Azz. 
194, Az. 1346 Soph. uses ὑπερμαχεῖν. 
But we need not therefore, with Elms. 
and Blaydes, read ὑπὲρ μαχοῦμαι. The 
derivative form ὑπερμαχέω, to be a 
champion, implies ὑπέρμαχος, as συμ- 


paxéw is from σύμμαχος, προμαχέω from: 


πρόμαχος: ὑπερμάχομαι is a simple com- 
pound, like συμμάχομαι (Plat., Xen.), 
mpopaxoua (Ziad, Diod., Plut.).—Kaml 
πάντ᾽ ἀφίξομαι with ζητῶν, will leave 
nothing untried in seeking: a poetical 
variation of ἐπὶ πᾶν ἐλθεῖν (Xen. Anad. 
3. I. 18 dp’ οὐκ ἂν ἐπὶ πᾶν ἔλθοι... ὡς 
φόβον παράσχοι), asin Eur. Hipp. 284 εἰς 
πάντ᾽ ἀφῖγμαι, ‘I have tried all means.’ 
In prose ἀφικνεῖσθαι εἴς τι usu.=to be 
brought to a situation, as Her. 8. 110 és 
πᾶσαν βάσανον ἀπικνεομένοισι, though put 
to any torment; Plat. Zuthyd. 292 E εἰς 
πολλήν γε ἀπορίαν ἀφίκεσθε. 

267 τῷ AaPSakelw παιδὶ, a dat. fol- 
lowing ζητῶν x.7.X. as -- τιμωρούμενος. For 
«Λαβϑακείῳ---ἸΤολυδώρου τε cp. Eur. Med. 
404 τοῖς Σισυφείοις τοῖς τ᾽ ᾿Ιάσονος γάμοις : 
for the adj., Od. 3. 190 Φιλοκτήτην Ποιάν- 
τιον [= Iloiavros] ἀγλαὸν υἱόν : Her. 7. 105 
τοῖς Μασκαμείοισι ἐκγόνοισι: Ph. 1131: 


Tr.1219. Her. (5. 59) saw in thetemple of 
the Ismenian Apollo at Thebes an inscrip- 
tion which he assigns to the age of Laius: 
ταῦτα ἡλικίην ἂν εἴη κατὰ Λάϊον τὸν Λαβ- 
δάκου τοῦ Πολυδώρου τοῦ Κάδμου. Cadmus, 
in the myth, is the son of Agenor king of 
Phoenicia, whence Carthage is ‘ Agenor’s 
city’ (Verg. Ae. 1. 338): Polydorus, son 
of Cadmus and Harmonia, was king of 
Thebes. 

269 f. construe: καὶ εὔχομαι τοῖς 
ταῦτα μὴ δρῶσιν [for them, PZ. 1019 καί 
σοι πολλάκις τόδ᾽ ηὐξάμην] θεοὺς ἀνιέναι 
αὐτοῖς μήτ᾽ ἀροτόν τινα γῆς, μήτ᾽ οὖν 
γυναικῶν παῖδας. The acc. θεοὺς as 
subject to ἀνιέναι is better than a dat. 
θεοῖς with εὔχομαι would be: Xen. Anaé. 
6. τ. 26 εὔχομαι δοῦναί μοι τοὺς θεοὺς 
αἴτιόν τινος ὑμῖν ἀγαθοῦ γενέσθαι: Ar. 
Thesm. 350 ταῖς δ᾽ ἄλλαισιν ὑμῖν τοὺς 
θεοὺς | εὔχεσθε πάσαις πολλὰ δοῦναι 
κἀγαθα. 

271 μήτ᾽ οὖν: ‘20, nor.’ Aesch. Ag. 
474 par εἴην πτολιπόρθης, | μήτ᾽ οὖν αὐτὸς 
ἁλούς, κιτιλ, Soph. Phil. 345 εἴτ᾽ ἀληθὲς 
εἴτ᾽ ἄρ᾽ οὖν μάτην: cp. above v. go. But 
οὖν with the frst clause, below, 1049: 
ΕἸ. 199, 560: see on 25. 

272 φθερεῖσθαι, a fut. found also in 
Eur. Andr. 708 (φθερεῖ 2 sing.) : Thuc. 7. 
48 φθερεῖσθαι: Ionic φθαρέομαι: Her. 9. 
42,8. 108 (φθαρήσομαι in Hippocr., Arist., 
Plut.). The schol. says, φθαρῆναι δεῖ 
γράφειν, od φθερεῖσθαι, distinguishing 
εὔχομαι with fut. infin., ‘I vow’ (to do), 
from εὔχομαι with pres. or aor. infin., ‘I 
pray.” But verbs of wishing or praying 
sometimes take a fut. infin. instead of 
pres. or aor.: Thuc. 6.57 ἐβούλοντο...προ- 
τιμωρήσεσθαι: 6. 6 ἐφιέμενοι pev...THs 
πάσης ἄρξειν: 1. 27 ἐδεήθησαν... ξυμπρο- 
πέμψειν : ἡ. 56 διενοοῦντο κλήσειν. See 


48 ZOPOKAEOYS 


ὑμῖν δὲ τοῖς ἄλλοισι Καδμείοις, ὅσοις 
τάδ᾽ ἔστ᾽ ἀρέσκονθ', n τε σύμμαχος Δίκη 


χοὶ πάντες εὖ ξυνεῖεν εἰσαεὶ θεοί. 
ὥσπερ ve ἀραῖον ἔλαβες, ὧδ᾽, ἄναξ, ἐρώ. 


ΧΟ. 


ΕἾ 
OUT EKTAVOV yap OUTE TOV KTQVOVT. ἔχω 


δεῖξαι. 


Φοίβου τόδ᾽ εἰπεῖν, ὅστις εἴργασταί ΠΟΤΕ. 


Ol. 


δίκαι: ἔλεξας: ἀλλ᾽ ἀναγκάσαι θεοὺς 


a \ , 209 ἃ ae ΄ διν 3 
av μὴ θέλωσιν οὐδ᾽ ἂν εἷς δύναιτ᾽ ἀνήρ. 


XO. 
Ol. εἰ καὶ τρίτ' ἐστί, μ 
XO. 


Ν , oe ed a ϑ τ , > ε \ A 
Ta δεύτερ᾽ ἐκ τῶνδ᾽ ἂν λέγοιμ, aol δοκεῖ 


ἄνακτ᾽ ἄνακτι ταῦθ᾽ ὁρῶντ᾽ ἐπίσταμαι 
a 
μάλιστα Φοίβῳ Τειρεσίαν, παρ᾽ ov τις av 


275 
τὸ δὲ ζήτημα τοῦ πέμψαντος ἣν 
280 
μὴ «παρῇς τὸ μὴ οὐ φράσαι. 
285 


a 99 > 3 , , 
σκοπῶν τάδ᾽, ὠναξ, ἐκμάθοι σαφέστατα. 
> > > 3 3 A ION oat > ’ὔ 
ΟΙ. ἀλλ᾽ οὐκ ἐν ἀργοῖς οὐδὲ τοῦτ᾽ ἐπραξάμην. 
ἔπεμψα γὰρ Κρέοντος εἰπόντος διπλοῦς 
πομπούς: πάλαι δὲ μὴ παρὼν θαυμάζεται. 


273 τοῖς 7 ἄλλοισι Jernstedt: τοῖς ἄλλοισι Καδμείοις θ᾽ F. W. Schmidt. 





Goodwin, Joods and Tenses § 113 (new 
ed.). 

273 2. τοῖς ἄλλοισι. The loyal, as 
Opp. to οἱ μὴ ταῦτα δρῶντες (:69).---ἔσ τ᾽ 
ἀρέσκοντ᾽, cp. 126. ἥ τε σύμμαχος Δίκη, 
Justice who ever helps the righteous 
cause; Blaydes needlessly writes ἡ Δίκη 
Te gUuuaxos. O. C. 1012 ἐλθεῖν ἀρωγοὺς 
συμμάχους τε (τὰς θεάς). 

475 Ξ. εὖ: cf. 7γαεἦ. 220 ἀλλ᾽ εὖ μὲν 
ἵγμεθ᾽ , εὖ δὲ προσφωνούμεθα.---ὥΩὝόψσΜ-πτερ μ᾽ 
ἀραῖον κιτ.λ. As you have brought me 
into your power under a curse [if I speak 
not the truth], so (ὧδε, 2.6. &opxos) I will 
speak. Aeschin. /z Ctes. 8.90 play 
ἐλπίδα λοιπὴν κατεῖδε σωτηρίας, ἔνορκον 
λαβεῖν τὸν ᾿Αθηναίων δῆμον... βοηθήσειν, 
to bind them by an oath that they would 
help. λαβεῖν here has nearly the same 
force as in λαβεῖν αἰχμάλωτον. etc.: Lys. 
or. 4 § 5 ὑποχείριον λαβὼν τὸ σῶμα, having 
got his person into my power. —dpatoy = 
τῇ ἀρᾷ ἔνοχον, cp. Spxios...Aéyw Ant. 305. 
The paraphrase of Eustath. 1809. 14 ὡσ- 
περ με εἷλες διὰ τῆς ἀρᾶς is substantially 
right. The use of καταλαβεῖν is not really 
similar (Her. 9. 106 πίστι τε καταλα- 
Bovres καὶ ὁρκίοισι, Thuc. 4. 85 Spkous... 
καταλαβὼν τὰ τέλη), since the κατά in 
comp. gives the sense of overtaking, and 
500 binding. Nor can we compare Ο. 


C. 284 ὥσπερ ἔλαβες τὸν ἱκέτην ἐχέγ- 
νον, where the sense is, ‘As thou hast 
received the (self- surrendered) suppliant 
under thy pledge.’ 

277 γὰρ after ἔκτανον merely prefaces 
the statement: Plat. Prot. 320 C δοκεῖ 
τοίνυν... μῦθον ὑμῖν λέγειν. ἦν γάρ ποτε 
Κ-ΤιᾺ.- 

278 δεῖξαι, ‘point to.’ Note the em- 
phatic place of the word: the speaker 
knows not that he is face to face with 
the slayer. τὸ ζήτημα, acc. of general 
reference. The simpler form would have 
been, ἦν τοῦ πέμψαντος τὸ ζήτημα καὶ 
λῦσαι: but, instead of a verb which 
could govern ζήτημα, τόδ᾽ εἰπεῖν is 
substituted, because it conveniently in- 
troduces the clause ὅστις εἴργασται, εχ- 
plaining what the {jrnua itself was. τὸ 
ζήτημα is then left much as ἃ αἰτεῖς is left 
in 216 when the insertion of ἀλκὴν κιτ.λ, 
has modified the construction. 

281 dv μὴ θέλωσιν «.7.r. Cp. Phil. 
1368 κἄμ᾽ ἀναγκάζεις τόδε. Gv as 580, 
7492.0. Ὁ: 13, Ant. 1057, Phil. 1276, 
Ai. 1085. οὐδ᾽ ἂν εἷς: Ant. 884 οὐδ᾽ ἂν di 
els παύσαιτ᾽ dv: O. C. 1656 οὐδ᾽ ἂν els | 
θνητῶν ppacee. In this emphatic form 
even a prep. could be inserted (Xen, 
Hellen. 5. 4. τ οὐδ᾽ ὑφ᾽ ἑνός, Cyr. 4. τ. 
14 μηδὲ πρὸς μίαν), and in prose οὐδὲ 


OIAIMOYS ΤΎΡΑΝΝΟΣ 49 


But for all you, the loyal folk of Cadmus to whom these things 
seem good, may Justice, our ally, and all the gods be with you 


graciously for ever. 


Cu. As thou hast put me on my oath, on my oath, O king, 


I will speak. 


I am not the slayer, nor can I point to him who 


slew. As for the question, it was for Phoebus, who sent it, to 
tell us this thing—who can have wrought the deed. 
Or. Justly said; but no man on the earth can force the 


gods to what they will not. 
Ca, 
OE. 
(Ἢ: 


I would fain say what seems to me next best after this. 
If there is yet a third course, spare not to show it. 
I know that our lord Teiresias is the seer most like to 


our lord Phoebus; from whom, O king, a searcher of these things 


might learn them most clearly. 


ΟΕ. Not even this have I left out of my cares. On the hint 
of Creon, I have twice sent a man to bring him; and this long 
while I marvel why he is not here. 


281 ἂν Brunck; the MSs. have ἀν (as L), or ἂν. 





εἷς stood without elision: in Ar. Ran. 
927 etc., where the MSs. have οὐδὲ ἕν 
(Dind. writes οὐδεὲν), οὐδ᾽ ἂν ἕν is a 
possible τ. ἢ. 

282 ἐκ τῶνδε-- μετὰ τάδε: Dem. or. 
18 8 313 λόγον ἐκ λόγου λέγων.---ΕῸΥ 
δεύτερα, second-dest, cp. the proverb dev- 
repos πλοῦς: Plat. Legg. 943 C τὴν τῶν 
ἀριστείων κρίσιν... καὶ τὴν τῶν δευτέρων Kal 
τρίτων .--- ἂν λέγοιμι: see ON 95. 

288 τὸ μὴ οὐ, not τὸ μή, because the 
sentence is negative: below, 1232: Azz. 
544 μή μ᾽ ἀτιμάσῃς τὸ μὴ οὐ | θανεῖν. But 
even in such a negative sentence the 
simple τὸ μή occurs: below, 1388: Azz. 

43: 
ἢ 284 ἀνακτ᾽: Od. 11. 151 Τειρεσίαο 
ἄνακτος.---ταὐτὰ ὁρῶντα, ποί -- ταὐτὰ φρο- 
νοῦντα or γιγνώσκοντα, ‘taking the same 
views, but seeimg in the same manner, 
7.6. with equal clearness: ὁρῶντα absol., 
as O.C. 74 ὅσ᾽ av λέγοιμι, πάνθ᾽ ὁρῶντα 
λέξομαι: ταὐτὰ adverbial=xara ταὐτά: 
the dat. ἄνακτι as O.C. 1358 ἐν πόνῳ] 
ταὐτῷ BeBynxws...€uol. Her. 4. 119 τωὐτὸ 
av ὑμῖν ἐπρήσσομεν. 

287 οὐκ ἐν ἀργοῖς τοῦτο κατέλιπον 
would have meant, ‘I did not leave this 
among things neglected.’ Soph. fuses 
the negative form with the positive, and 
instead of κατέλιπον writes ἔπραξάμην : 
‘I saw to this (midd.) in such a manner 

at it also should not be among things 
neglected.’ πράσσεσθαι (midd.) else- 


age 


where usu. = ‘to exact’ (Thuc. 4. 65 etc.) : 
here=dtarpdooecOar, effect for oneself. 
Cp. Az. 45 ἐξεπράξατο (effected his pur- 
pose). 6. Wolff, sharing Kviéala’s ob- 
jections to the phrase ἐν ἀργοῖς πράσσεσ- 
θαι, places a point after τοῦτ᾽ (‘but neither 
is this among things neglected:—I did 
it’). The extreme harshness of the asyn- 
deton condemns this; and the suggested 
ἔπραξα μήν is no remedy. For ἐν cp. 
οὐκ ἐν ἐλαφρῷ ἐποιεύμην (Her. 1. 118), 
ἐν εὐχερεῖ | ἔθου (ταῦτα) Phil. 875, ταῦτ᾽ 
οὖν ἐν αἰσχρῷ θέμενος Eur. Hee. 806. 
ἀργοῖς, not things wdone, but things at 
which the work is sluggish or tardy; 
O. C. 1605 κοὐκ ἦν ἔτ᾽ οὐδὲν ἀργὸν ὧν 
ἐφίετο: Eur. Phoen. 776 ὃν δ᾽ ἐστὶν ἡμῖν 
ἀργόν, εἴ τι θέσφατον | οἰωνόμαντις Τειρε- 
σίας ἔχει φράσαι, 2.6. ‘in one thing our 
zeal has lagged,—the quest whether’ etc. : 
Theognis however (583 Bergk 3rd ed.) 
has τὰ μὲν προβέβηκεν ἀμήχανόν ἐστι γε- 
νέσθαι | ἀργά, = ἀποίητα, infecta. 

288 διπλοῦς | πομπούς : he had sent 
two successive messages—one messenger 
with each. moumdés=one who is sent to 
escort (πέμπειν) or fetch a person (0.C. 
70). The words could mean (as Ellendt 
takes them) ‘two sets of messengers’: 
but the other view is simpler, and con- 
sists equally well with οἵδε in 297. 

289 μὴ παρὼν. θαυμάζεται-- θαυμάζω 
εἰ μὴ πάρεστι; but with οὐ, Ξε θαυμάζω ὅτι 
οὐ πάρεστι: differing nearly as ‘I wonder 


4. 


50 ZOPOKAEOY2 


XO. καὶ μὴν τά ay ἄλλα κωφὰ καὶ Tahal’ ἔπη. 


290 


295 


OI. ta mova ταῦτα: πάντα γὰρ σκοπῶ λόγον. 

XO. θανεῖν ἐλέχθη πρός τινων ὁδοιπόρων. 

Ol. ἤκουσα κἀγώ’ τὸν δ᾽ ἰδόντ᾽ οὐδεὶς ὁρᾷ. 

ΧΟ. ἀλλ᾽ εἴ TL μὲν δὴ δείματός as ἔχει μέρος, 
τὰς σὰς ἀκούων οὐ μενεῖ τοιάσδ᾽ ἀράς. 

Ol. @ ᾽στι δρῶντι τάρβος, οὐδ᾽ ἔπος φοβεῖ. 

ΧΟ ἀλλ᾽ οὐξελέγξων αὐτὸν ἔστιν" οἵδε γὰρ 
τὸν θεῖον ἤδη μάντιν ὧδ᾽ ἄγουσιν, ᾧ 
τἀληθὲς ἐμπέφυκεν ἀνθρώπων μόνῳ. 

Ol. ὦ πάντα νωμῶν Τειρεσία, διδακτά τε 


200 


π᾿ ἄρρητά τ᾽, οὐράνιά τε καὶ χθονοστιβῆ, 
πόλιν μέν, εἰ καὶ μὴ βλέπεις, φρονεῖς δ᾽ ὅμως 


290 τὰ τά. Ὲ (including A, where the 1st hand had begun to write τὰ δ᾽). 


293 τὸν δ᾽ ἰδόντ᾽ MSS. 


τὸν δὲ δρῶντ᾽ is an anonymous conjecture cited by Burton. 


294 The rst hand in L wrote δείματοστ᾽, (there is no trace of an accent on ο,) joining 
or in one character; the corrector afterwards wrote τ᾿ separately, as in 134, 257. 
(The facsimile shows ees this τ᾿ was not made from γ᾽.) δείματός τ᾽ was the reading of 


almost all the later MSS. 


: indeed, it does not appear certain that any one of them has 





poe and ‘I wonder ¢hat.’ Xen. Anad. 

- 4. 15 (he spoke of) τὰ μὴ ὄντα ὡς οὐκ 
ae 4.2: εἴ τι μὴ ἦν, ἔλεγεν ὅτι οὐκ ἦν. 

290 τά γ᾽ ἀλλα.. ἔπη: the rumours 
which were current—afart from the 
knowledge which the seer may have to 
give us. Not ‘the other rumours.’ Cp. 
Plat. Phaed. 110 E καὶ λίθοις καὶ γῇ καὶ 
τοῖς ἄλλοις wos τε καὶ φυτοῖς. κωφὰ: 
the rumour has died down; it no longer 
gives a clear sound. Cp. fr. 604 λήθην 
Te τὴν ἅπαντ᾽ ἀπεστερημένην, | κωφήν, 
ἄναυδον. Ai. gti ὁ πάντα κωφός, ὁ πάντ᾽ 
ἄϊδρις, reft of all sense and wit. 

291 τὰ ποῖα, cp. 120. 

292 ὁδοιπόρων: the survivor had 
spoken of λῃσταί, 122. The word now 
used comes nearer to the truth (cp. 801 
ὁδοιπορῶν); but, as the next v. shows, 
Oed. does not regard this rumour as a 
different one from that which Creon had 
mentioned. 

293 τὸν δ᾽ ἰδόντ᾽ : the surviving eye- 
witness: cp. 119 ὧν εἶδε, πλὴν ἕν k.7.X. 
Oed. has not yet learned that this wit- 
ness could be produced: cp. vv. 754 ff. 
ἰδύντα is better than the conj. δρῶντα 
(1) as expressing, not merely that the 
culprit is unknown, but that no eye- 
witness of the deed is now at hand: 
(2) because, with ὁρᾷ, it has a certain 
ironical point, —expressing the king’s in- 


credulity. as to anything being made of 
this clue. Cp. 105, 108. 

294 The subject to tye is the mur- 
derer, who is foremost in the thoughts of 
the Chorus,—not the eye-witness (ὁ ἰδών, 
293). The reversion from plural (ὁδοιπό- 
ρων, 292) to singular is unconscious, just as 
in 124 we have ὁ λῃστής, after λῃστάς in 
122.—Selpardés γ᾽. δεῖμα, prop. ‘an object 


of fear,’ is used by Her. and the poets 
as=déos: Her. 6. 74 Knreouévea...detua 


ἔλαβε Σπαρτιητέων : Aesch. Suppl. 566 
χλωρῷ δείματι θυμὸν  πάλλοντ᾽ : Eur. 
Suppl. 599 ws μοι ὑφ᾽ ἥπατι δεῖμα χλοερὸν 
ταράσσει: id. El. 767 ἐκ δείματος, from 
fear. Cp. above, 153. The ye gives 
emphasis: the ἀραί of Oed. were enough 
to scare the boldest. Hartung conjec- 
tures δειμάτων ἔχει μέρος. The plur. 
δείματα means either (a) objects of fear, 
or (4) much more rarely, fears, with re- 
ference to some particular objects already 
specified: as in Z/. 636 δειμάτων ἃ νῦν 
ἔχω, ‘the terrors which I now suffer,’ 
alluding to the dveams. Here we seem 
to need the sing., ‘fear.’ 

295 ff. tds σὰς... ἀράς, thy curses: 
τοιάσδε, being such as they are.—ov€e- 
λέγξων. The present οὐξελέγχων would 
mean, ‘there is one who convicts him’: 
i.e. the supposed criminal, whom threats 
scare not, is already detected; for the 


ΘΟΙΙΠΟΥΣ TEPANNOS 51 


ΓΗ. 
ΘῈ: 


Indeed (his skill apart) the rumours are but faint and old. 
What rumours are they? I look to every story. 


(ΓΗ. Certain wayfarers were said to have killed him. 
Or. I, too, have heard it, but none sees him who saw it. 
CH, Nay, if he knows what fear is, he will not stay when he 


hears thy curses, so dire as they are. 

OE. When a man shrinks not from a deed, neither is he 
scared by a word. 

CH. But there is one to convict him. For here they bring at 
last the godlike prophet, in whom alone of men doth live the truth. 


Enter TEIRESIAS, led by a Boy. 
ΟΕ. Teiresias, whose soul grasps all things, the lore that 
may be told and the unspeakable, the secrets of heaven and the 
low things of earth,—thou feelest, though thou canst not see, 


᾽.-Όειμάτων ἔχει Hartung. 297 The ist hand in 1, wrote οὐξελλέγχων : the 
first \ has been erased, and -ξων written above, either by the 1st hand itself (as 


Diibner thinks), 
οὑξελέγξων and οὐξελέγχων : 


or by the first corrector. 


The later MSS. are divided between 


A supports the former, which, on the whole, has the ad- 





prophet has come. Cp. Isocr. or. 8. 
§ 139 ὥστ᾽ οὐκ ἀπορήσομεν μεθ᾽ ὧν κω- 
λύσομεν τοὺς ἐξαμαρτάνοντας, ἀλλὰ πολ- 
λοὺς ἕξομεν τοὺς ἑτοίμως καὶ προθύμως 
συναγωνιζομένους ἡμῖν: where, how- 
ever, the present part. συναγωνιζομένους 
is relative to the future ἕξομεν. To this 
it may be objected: (1) the present parti- 
ciple with ἔστιν would not be suitable 
unless the conviction were in act of 
taking place: (2) the fut. partic. not 
only suits the context better—‘one éo 
convict him’ [supposing he is here]—but 
also agrees ‘with the regular idiom: ¢.g. 
Phil, 1242 τίς ἔσται μ᾽ οὑπικωλύσων ἘΠΕ: 
£i. 1197 οὐδ᾽ οὑπαρήξων οὐδ᾽ ὁ κωλύσων 
wapa; (cp. Ant. 261:) Aesch. P. V. 27 
ὁ λωφήσων γὰρ οὐ πέφυκέ πω: Xen. 
An. 2. 4. 5 ὁ ἡγησόμενος οὐδεὶς ἔσται. 

298 ᾧ: this pron. ends av. O. C. 14, 
77. 810. £1. 878. 

299 ἐμπέφυκεν, a divine gift of pro- 
phecy: Her. 9. 94 (of the seer Evenius) 
καὶ μετὰ ταῦτα αὐτίκα ἔμφυτον μαντικὴν 
εἶχε.--- ἀνθρώπων μόνῳ, above all other 
men: cp. Ο. C. 261 μόνας... | σώζειν οἵας 
τε κιτιλ., Athens, above all other cities, 
can save: Isocr. or. 14 8. 57 ὀφείλετε δὲ 
μόνοι τῶν Ἑλλήνων τοῦτον τὸν ἔρανον, 
unice (though others owe it also). 

800 ὦ πάντα νωμῶν: νωμάω (ven) 
means (1) to distribute, (2) to dispose, 
and so to wield, ply, (3) figuratively, 
to ponder, azzmo versare: ἐνὶ φρεσὶ κέρδε᾽ 
ἐνώμας Od. 18. 216: ἐν ὠσὶ νωμῶν καὶ 


φρεσὶν πυρὸς δίχα | χρηστηρίους ὄρνιθας 
ἀψευδεῖ τέχνῃ Aesch. Theb. 25 (of Tei- 
resias): (4) then, absolutely, to observe: 
Her. 4. 128 vwuwrres...cira ἀναιρεομένους, 
observing the moment when they were 
cutting forage. Similarly here,—with 
the idea of mental grasp unaided by eye- 
sight. Plato (Crat. 411 Ὁ) fancifully con- 
nects γνώμη with νώμησις,---τὸ yap νωμᾶν 
καὶ τὸ σκοπεῖν ταὐτόν.---διδακτά τε--ἄρ- 
ρητά τε, cp. the colloquial ῥητὸν ἄρρητόν 
τ᾿ ἔπος (O.C. 1001 dicenda tacenda): ἄρ- 
ρηταΞτεἀπόρρητα : Her. 6. 135 ἄρρητα ἱρὰ 
ἐκφήνασαν. 

801 οὐράνιά τε καὶ χθονοστιβῆ: 
not in apposition with ἄρρητα and δι- 
δακτά respectively, but both referring to 
each, lore that may or that may not be 
told, whether of the sky or of the earth. 
Dindorf cp. Nicephorus Gregoras 7152. 
Byz. 695 Ὁ ἄκτιστα γενέσθαι πάντα τά τ᾽’ 
οὐράνια τά τε χθονοστιβῆ καὶ ὑδραῖα γένη: 
where, however, χθονοστιβῆῇ has its literal 
sense,—‘ walking the earth’: here it is 
poet. for ἐπίγεια, ‘the lowly things of 
earth.’ Cp. Hom. hymn. 29. 2 ἀθανά- 
τῶν τε θεῶν χαμαὶ ἐρχομένων τ᾽ ἀνθρώ- 
πων. 

802 μέν is not balanced by φρονεῖς δ᾽ 
(as if we had οὐ βλέπεις μέν), but by the 
thought of the expected healer (310). 
The δὲ after φρονεῖς introduces the 
apodosis after a concessive protasis, as 
Her. 8. 22 ef δὲ ὑμῖν ἐστι τοῦτο μὴ 
δυνατὸν ποιῆσαι, dudes δὲ (then) ἔτι καὶ 


4-2 


52 ΣΟΦΟΚΛΕΟΥΣ 


οἵᾳ νόσῳ σύνεστιν 2 


ἡ ς σε προστάτην 


σωτὴρά τ᾽, ὦναξ, μοῦνον ἐξευρίσκομεν. 

Φοῖβος γάρ, εἰ καὶ μὴ κλύεις τῶν ἀγγέλων, 205 
πέμψασιν ἡμῖν ἀντέπεμψεν, ἔκλυσιν 

μόνην ἂν ἐλθεῖν τοῦδε τοῦ νοσήματος, 

εἰ τοὺς κτανόντας Λάϊον μαθόντες εὖ 


κτείναιμεν, ἢ γῆς φυγάδας ἐκπεμψαίμεθα. 


’ὔ’ 
συ ἍΔΕ 


θονήσας μήτ᾽ ἀπ᾽ οἰωνῶν φάτιν 


210 


μήτ᾽ εἴ TW ἄλλην μαντικῆς ἔχεις ὁδόν, 
ῥῦσαι σεαυτὸν καὶ πόλιν, ῥῦσαι δ᾽ ἐμέ, 
ῥῦσαι δὲ πᾶν μίασμα τοῦ τεθνηκότος. 

ἐν σοὶ γὰρ ἐσμέν: ἄνδρα δ᾽ ὠφελεῖν ἀφ᾽ ὧν 


ἔχοι τε καὶ δύναιτο κάλλιστος πόνων. 


315 


TEIPESIAS. 
φεῦ φεῦ, φρονεῖν ws δεινὸν ἔνθα μὴ τέλη 


λύῃ φρονοῦντι. 


vantage in authority, and is also recommended by Greek usage: see comm. 
εἰ μὴ καὶ F. V. Fritzsch. 

310 σύ νυν] The rst hand in L seems to 
have written od viv, which a later hand changed to σὺ δ᾽ οὖν. 


εἴ τι μὴ L. Stephani: 
808 εὖ] ἢ Meineke. 


καὶ μὴ Μ585.: 
Blaydes. 


ταῦτα γὰρ καλῶς ἐγὼ 


8905 εἰ 
ΘΟ τοῦδε] τήνδε 


(1 formerly thought 





νῦν ἐκ τοῦ μέσου ἡμῖν ἕζεσθε. Xen. Cyr. 
5. 5. 21 ἀλλ᾽ εἰ μηδὲ τοῦτο... βούλει ἀπο- 


κρίνασθαι, σὺ δὲ τοὐτεῦνθεν λέγε: 

8038 ἧς sc. νόσου. προστάτην νόσου. a 
protector from ἃ plague: strictly, one who 
stands in front of, shze/ds, the city’s dis- 
tempered state. Cp. Az. 803 πρόστητ᾽ 
ἀναγκαίας τύχης, shelter my hard fate. In 
Eur. Andr. 220 xelpov’ ἀρσένων νόσον | ταύ- 
την νοσοῦμεν, ἀλλὰ προὔστημεν Kaus, 
_ ‘we suffer this distemper more cruelly 

than men, but ever rule it well,’ the idea 
is that of administering (not protecting), 
as in προΐστασθαι τῆς ἡλικίας, to regulate 
one’s own early years, Isocr. or. 15 ὃ 200. 
Cp. 882. 

804 μοῦνον: this Tonic form (like 
κοῦρος, δουρί, ξεῖνος, γούνατα) is used in 
dialogue by Soph.: Aesch. has not μοῦνος, 
though in) FP: .V. 804 Τὸν re bowara 
στρατόν. In [Eur. ] Rhes. 31 μόναρχοι is 
now restored for μούναρχοι. 

805 εἰ καὶ μὴ κλύεις, ‘if indeed... 
implying that he probably has heard it. 
Ai. 1127 δεινόν γ᾽ εἶπας, εἰ καὶ (7s 
θανών. On εἰ καί and καὶ εἰ see Ap- 
pendix. Others would render, ‘if you 
have not heard from the messengers a/so,’ 


supposing it to be a hyperbaton for εἰ μὴ 
κλύεις καὶ τῶν ἀγγέλων. This is impossi- 
ble. Prof. Campbell compares Thuc. 5. 
45 καὶ ἣν ἐς τὸν δῆμον ταῦτα λέγωσιν, ἃ5 
if put for ἢν καὶ ἐς τὸν δῆμον : but there 
the passage runs thus; (Spartan envoys 
had been pleading with effect before the 
Athenian Βουλή :)---τὸν ᾿Αλκιβιάδην ἐφό- 
βουν μὴ καί, ἣν ἐς τὸν δῆμον ταὐτὰ λέγω- 
σιν, ἐπαγάγωνται τὸ πλῆθος καὶ ἀπωσ- 
θῇ ἡ ᾿Αργείων συμμαχία: where the καῇ 
before ἤν goes with ἐπαγάγωνται. Some 
adopt the conj. ef τε μή, ‘Sunless Zer- 
chance’: for re so used, see below 969, 
O. C. 1450, Zr. 586, 712: but no change 
is required.-For the pres. κλύεις, cp. 
Ph. 261. 

808 μαθόντες εὖ. εὖτ’ with care,’ ‘a- 
right’: cp. Ai. 18 ἐπέγνως εὖ: 10. 528 
ἐὰν τὸ ταχθὲν εὖ τολμᾷ τελεῖν. Meineke’s. 
conj. ἢ, adopted by Nauck, is weak, and 
against the rhythm. 

310 ΣΦ. dm’ οἰωνῶν φάτιν: ἴοτ ἀπό, 
see 43: φάτιν, 151.—dAAnv ὁδόν, as di- 
vination by fire (see on 21), to which 
Teiresias resorts (At. 1005) when the 
voice of birds fails him. 

812 ff. ῥῦσαι σεαυτὸν κιτ.λ. ῥύεσθαί 


ΟἸΔΊΠΟΥΣ, TYPANNOZ 53 
what a plague doth haunt our State,—from which, great prophet, 
we find in thee our protector and only saviour. Now, Phoebus— 
if indeed thou knowest it not from the messengers—sent answer 
to our question that the only riddance from this pest which 
could come was if we should learn aright the slayers of Latus, 
and slay them, or send them into exile from our land. Do 
thou, then, grudge neither voice of birds nor any other way 
of seer-lore that thou hast, but rescue thyself and the State, 
rescue me, rescue all that is defiled by the dead. For we are 
in thy hand; and man’s noblest task is to help others by his 
best means and powers. 


TEIRESIAS. 


Alas, how dreadful to have wisdom where it profits not the 


that the rst hand had written σὺ οὖν, omitting δ᾽.) 
ἔχει τ.---πόνοσ L, with wy written above oo by the first corrector (S). 


wise! Aye, I knew this well, 


σὺ δ᾽ οὖν τ. 915 ἔχοι 1,: 


Several of 


the later Mss. (including A) have πόνων, though πόνος continued to be current as a 


variant. 


317 λύηι L: λύει Or λύη Υ. 





τι is to draw a thing to oneself, and so to 
protect it, ῥῦσαι μίασμα here =literally, 
‘take the defilement under thy care’; 1.6. 
‘make it thy caré to remove the defile- 
ment.’ Cp. πρόστητ᾽ ἀναγκαίας τύχης (Az, 
803), shelter my hard fate, (instead of, 
‘shelter me from it.’)—wav μίασμα, the 
whole defilement, as affecting not only 
human life but also the herds and flocks 
and the fruits of the earth: cp. 253.—rod 
τεθνηκότος, gen. of the source from which 
the μίασμα springs,—more pathetic than 
τοῦ φόνου, as reminding the hearer that 
vengeance is due for innocent blood. 
Both πᾶν and the usual sense of μίασμα 
forbid us to understand, ‘avenge the un- 
cleanness [z.e. the unpunished murder] of 
the dead man.’ For ῥῦσαι δὲ Blaydes 
conj. λῦσον δὲ, comparing Eur. Or. 598 
μίασμα λῦσαι. But the triple ῥῦσαι is 
essential to the force. 

814 ἐν col=fenes te: O.C. 248 ἐν 
ὑμῖν ws θεῷ | κείμεθα TAduoves: Eur. Ale, 
278 ἐν σοὶ δ᾽ ἐσμὲν καὶ ζῆν καὶ μή.---ἄνδρα, 
accus. before, not after, ὠφελεῖν, asin Azz. 
710 ἀλλ᾽ ἄνδρα, κεἴ Tis ἦ σοφός, τὸ μανθά- 
νειν | πόλλ᾽ αἰσχρὸν οὐδέν. In both places 
ἄνδρα has a certain stress—‘for mortal 
man.’ But in Az. 1344 ἄνδρα δ᾽ οὐ δίκαιον, 
el θάνοι, | βλάπτειν τὸν ἐσθλόν, ἄνδρα is 
the object, agreeing with τὸν ἐσθλόν. 

ad ὧν ἔχοι te καὶ δύναιτο, by means 
of all his resources and faculties. The 


optat. is thus used in universal state- 
ments, and therefore especially in γνῶμαι: 
cp. 979: Ant. 666 ἀλλ᾽ ὃν πόλις στήσειε, 
τοῦδε χρὴ κλύειν: Xen. Cyr. 1. 6. 19 
ἀλλὰ τοῦ μὲν αὐτὸν λέγειν, ἃ μὴ σαφῶς 
εἰδείη, φείδεσθαι δεῖ. So here we supply 
ἐστί (not ay εἴη) with κάλλιστος. The 
diiference between ἀφ᾽ ὧν ἂν ἔχῃ (‘may 
have’), and ἔχοι (‘might have’), is that 
the latter form treats the ‘having’ as an 
abstract hypothesis (εἴ τι ἔχοι). 

317 Avy: for subjunct. without ἄν, 
cf, O. C. 395 ὃς νέος πέσῃ: At. 1074 ἔνθα 
μὴ καθεστήκῃ δέος: 77. 1008 6 τι καὶ 
μύσῃ. The subjunct., ἔνθα μὴ λύῃ,ΞΞ “ἴῃ ἃ 
case where it may not profit’: the indic., 
ἔνθα μὴ λύει, =‘in a case where it does 
not profit.” The use of μή, whether with 
subjunct. or with indic., generalises the 
statement. Cp. O. C. 839 μὴ ᾿πίτασσ᾽ a 
μὴ κρατεῖς: 1b. 1442 μὴ πεῖθ᾽ ἃ μὴ δεῖ. 
But L has λύηι, and some other Mss. have 
Avy: and it is much more likely that this 
should have become λύει than wice versa. 
τέλη λύῃ -Ξ λυσιτελῇ, only here: cp. Eur. 
Alc. 627 φημὶ τοιούτους γάμους | λύειν 
βροτοῖς.---ταῦτα γὰρ (I have to bewail 
this now), for, though I once knew it, 
I had forgotten it. Teiresias, twice sum- 
moned (288), had come reluctantly. 
Only now, in the presence of Oedipus, 
does he realise the full horror of the se- 
cret which he holds. 


54 ZOPOKAEOY2 


3 \ 4 3 >) \ x “~ >) δ 4 
εἰδὼς διώλεσ᾽- οὐ yap ἂν δεῦρ ἱκόμην. 


4 21. 
τί δ᾽ ἔστιν; 
» 3 9 » 
ἄφες μ᾽ ἐς οἰκους" 


ὡς ἄθυμος εἰσελήλυθας. 
ῥᾷστα γὰρ τὸ σόν τε σὺ 


5 Ν Ψ > , “Δ 9 ‘\ UA 
καγω διοίσω τουμον, ἫΝ EOL πίθῃ. 
ψ 5 ον > > ¥ a , 
OL ὐτ ἔννομ εἰπας οὔτε Tpor pry πόλει 
τῇδ᾽, ἥ o ἔθρεψε, τήνδ᾽ “ἀποστερῶν φάτιν. 


πρὸς καιρόν" 


ὁρῶ γὰρ οὐδὲ σοὶ τὸ σὸν φώνημ᾽ ἰὸν 
ὡς οὖν μηδ᾽ ἐγὼ ταὐτὸν πάθω. 3.28 
μὴ πρὸς θεῶν φρονῶν γ᾽ ἀποστραφῇς, ἐπεὶ 


πάντες Oe προσκυνοῦμεν οἵδ᾽ ἱκτήριοι. 


ἐγὼ δ᾽ οὐ μή ποτε 
ἐκφήνω κακά. 


τί φής; ξυνειδὼς οὐ BS ἀλλ᾽ ἐννοεῖς 330 


TE. πάντες γὰρ οὐ φρονεῖτ΄. 
ε » Ν 
TOL ὡς ἂν εἴπω μὴ τὰ σ᾽ 
ΟἹ: 
ἡμᾶς προδοῦναι καὶ καταφθεῖραι πόλιν; 
TE. ἐγὼ οὔτ᾽ ἐμαυτὸν οὔτε δ᾽ ahyuva. 


τί ταῦτ᾽ 


ἄλλως ἐλέγχεις; οὐ γὰρ ἂν πύθοιδ μου. 


322 ἔννο μ᾽ L, with δῇ erasure between o and μ᾽. 
the correction may be due either to the 1st hand itself, or 
L has προσφιλῆ, with es written above, by S (I think), rather 

Many later Mss. (including A) combine ἔννομ᾽ with προσφιλὲς, 


(found in some later MSs.) ; 
to the διορθωτής (S). 
than by the rst hand. 


though the latter error was prob. generated by ἔννομον. 


The rst hand had written ἔννομον 


925 μηδ᾽ ἐγὼ] μὴ λέγων 





318 διώλεσ᾽ =let slip out of my me- 
mory; cp. σῴζεσθαι to remember, El. 
993, 1257, 77. 682: Plat. Theaet. 153 B 


κτᾶταί τε μαθήματα καὶ σῴζεται: Hep. 
455 Β ἃ ἔμαθε, σῴζεται. So Terent. 
Phormio 2 5: 39 perit hercle: nomen 


perdidi, ‘have forgotten. —Some explain, 
‘suppressed the thought.’ 

319 τί δ᾽ ἔστιν; Zl. 920 φεῦ τῆς 
,ἀνοίας.. ΧΡΥΣ. τί δ᾽ ἔστιν; and so often 
in Soph. (as 1144, 77. 339, Zl. g21): δέ 
marking that the attention is turned to a 
new point, as in τί δ᾽; gutd vero? (941), 
or to a new person: Isaeus or. 8 § 24 od 
δὲ ris ef; 

321. Solow, bear to the end: Eur. 
Hipp. 1143 δάκρυσι διοίσω | πότμον ἄποτ- 
μον, live out joyless days: Thuc. I. 11 εἰ 
ξυνεχῶς τὸν πόλεμον διέφερον. διαφέρειν 
could not mean ‘to bear apart? (from 
each other), though that is implied.— 
πίθῃ, 2.6. obey me by letting me go home. 

322 οὔτ᾽ ἔννομ᾽ x.7.A.: not in con- 
formity with usage, which entitled the 
State to benefit by the wisdom of its 
μάντις. The king’s first remonstrances 
are gentle. 


323 ἀποστερῶν, ‘withholding’: Arist. 
Rhet. 2. 6. 3 ἀποστερῆσαι παρακαταθήκην, 
depositum non reditere.— φάτιν, of adivine 
message, I5I. 

324 ὁρῶ γὰρ «.7.d.: (7 do mot speak), 
for I see that e/ther dost thou speak op- 
portunely: (I am silent) therefore, lest I 
too should speak unseasonably. 

325 πρὸς καιρόν = καιρίως, as At. 38, 
Ph. 1279, Tr. 59.—@s οὖν x.7.A.: “(1 do 
not speak), then, in order that mezther 
(μηδέ) may I share -your mishap (of 
speaking amiss).’? If he speaks not, med- 
ther will he speak wrongly. Cp. Thue. 
2. 63 εἰκὸς...μὴ φεύγειν τοὺς πόνους, ἢ 

μηδὲ τὰς τιμὰς διώκειν. I now prefer 
this view to taking μηδ᾽ ἐγώ as irregular 
for μὴ καὶ ἐγώ (‘lest I too...’),—resolving 
μηδέ into μή not, δέ on the other hand; 
though the place of éyw suggests this. 
Kviéala’s μὴ λέγων is ingenious and at- 
tractive; it may, indeed, be right; but 
seems hardly necessary. 

326 μὴ πρὸς θεῶν κιτ.λ. The attri- 
bution of hice two verses to the Chorus 
in some MSS. is probably due to the plur. 
in 327 having misled those who did not 


ΟΙΔΙΠΟῪΣ ΤΥΡΑΝΝΟΣ 55 


but let it slip out of mind; else would I never have come 


here. 
OE. 
ghee 


What now ? 


How sad thou hast come in! 
Let me go home; most easily wilt thou bear thine own 


burden to the end, and I mine, if thou wilt consent. 

ΟΕ. Thy words are strange, nor kindly to this State which 
nurtured thee, when thou withholdest this response. 

TE. Nay, I see that thou, on thy part, openest not thy lips 
in season: therefore I speak not, that neither may I have thy 


mishap. 
OE. 


For the love of the gods, turn not away, if thou hast 


knowledge: all we suppliants implore thee on our knees. 
TE. Aye, for ye are all without knowledge; but never will 
I reveal my griefs—that I say not thine. 


OE. How sayest thou ? 


Thou knowest the secret, and wilt 


not tell it, but art minded to betray us and to destroy the State ? 


TE 
these things ? 


Kviéala. 


I will pain neither myself nor thee. 
Thou wilt not learn them from me. 


326 f. L rightly assigns these two verses to Oedipus. 


Why vainly ask 


Several later 


MSS. give them to the Chorus, probably because v. 327 was thought less suitable to 
the person of the king. But there is no fitting place for the interposition of the 


Chorus before v. 404. 


332 ἐγώ τ᾽ L (with οὔτε written over ἐμαυτόν) : ἐγὼ οὔτε r. 





see that the king speaks for all Thebes. 
—dpovev γ᾽, if thou hast understanding 
(of this matter): cp. 569 ἐφ᾽ ols γὰρ μὴ 
φρονῶ σιγᾶν φιλῶ: not, ‘if thou art sane.’ 
But in 328 οὐ @povetre=‘are without un- 
derstanding,’ ‘are senseless.’ 

328 f. ἐγὼ δ᾽ οὐ μή ποτε ἐκφήνω τὰ ἐμὰ 
(ὡς ἂν μὴ εἴπω τὰ σὰ) κακά: I will never 
reveal my (not to call them ἐλ) griefs. τὰ 
ἐμὰ κακά, = those secrets touching Oedipus 
which lie heavy on the prophet’s soul: τὰ 
σὰ κακά, those same secrets in their im- 
port for Oedipus. We might render ws 
ἂν εἴπω μὴ τὰ σ᾽ either (i) as above, or 
(ii) ‘in order that I may not utter thy 
griefs.’ But (i) is preferable for these 
reasons:—(1) The subjunct. εἴπω with 
μή was familiar in such phrases. Plat. 
Rep. 487 Ὁ τοὺς μὲν πλείστους καὶ πάνν 
ἀλλοκότους γιγνομένους, ἵνα μὴ παμπο- 
νήρως εἴπωμεν, ‘becoming very strange 
persons,—not to use a more unqualified 
epithet’: Rep. 507 Ὁ οὐδ᾽ ἄλλαις πολλαῖς, 
ἵνα μὴ εἴπω ὅτι οὐδεμιᾷ, τοιούτου προσ- 
δεῖ οὐδενός, 1.6. few,—not to say none: 
Hippias minor 372 Ὁ τοιοῦτός εἰμι olds 
πέρ εἰμι, ἵνα μηδὲν ἐμαυτὸν μεῖζον 
elmw,—to say nothing more of myself. 
The substitution of ὡς dv for the com- 


moner ἵνα in no way alters the meaning. 
For ὡς dv pr, cp. Ar. Av. 1508 τουτὶ... 
τὸ σκιάδειον ὑπέρεχε ἄνωθεν, ws ἂν μή μ᾽ 
ἴδωσιν οἱ θεοί. For ὡς av εἴπω μὴ instead 
of ὡς ἂν μὴ εἴπω, cp. 255, Phil. 66 εἰ δ᾽ 
ἐργάσει | μὴ ταῦτας Ο. Ο. 1365 εἰ δ᾽ 
ἐξέφυσα τάσδε μὴ ᾿μαυτῷ τροφούς. Her. 
7. 214 εἰδείη γὰρ ἂν καὶ ἐὼν μὴ Μηλιεὺς... 
τὴν ἀτραπόν. (2) The emphatic position 
of τἄμ᾽ suits this version. (3) ἐκφήνω is 
more forcible than εὔπω. If the meaning 
were, ‘I will not veveal my griefs, in 
order that I may not mention (εἴπω) thy 
griefs, the clauses would be ill-balanccd. 
See Appendix, n. on vv. 328 f. 

330 ξυνειδὼς, because ἐκφήνω implied 
that he knew. Cp. 704 αὐτὸς ξυνειδώς, 7 
μαθὼν ἄλλου mapa; 1.6. of his own know- 
ledge, or on hearsay? Not, ‘being an 
accomplice’ (as Ant. 266 Evwedévac | τὸ 
πρᾶγμα βουλεύσαντι): Oed. can still con- 
trol his rising anger. 

882 ἐγὼ οὔτ᾽, synizesis. The rugged 
verse is perh. designed to express agi- 
tation. Cp. 1002 ἐγὼ οὐχί: O.C. 939 
ἐγὼ οὔτ᾽ ἄνανδρον, 0998 ἐγὼ οὐδέ, 1436 
τελεῖτ᾽, ἐπεὶ οὔ μοι: Ant. 458 ἐγὼ οὐκ 
ἔμελλον: Ph. 1390 ἐγὼ οὐκ ᾿Ατρείδα-.--- 
ταῦτ᾽, 20 ἢ. 


ae 


339 


56 ZO DPOKAEOY 2 

Ol. οὐκ, ὦ κακῶν κάκιστε, καὶ γὰρ ἂν πέτρου 
φύσιν σύ ca ὀργάνειας, ἐξερεῖς ποτέ, 
ἀλλ᾽ ὧδ᾽ ἄτεγκτος κατελεύτητος φανεῖ; 

TE. op) ὴν ἐμέμψω τὴν ἐμήν, τὴν σὴν δ᾽ ὁμοῦ 


ναίουσαν οὐ κατεῖδες, ἀλλ᾽ ἐμὲ ψέγεις. 
ΟΙ. τίς γὰρ τοιαῦτ᾽ ἂν οὐκ ἂν ὀργίζοιτ᾽ ἔπη 

κλύων, ἃ νύν σὺ τὴνδ᾽ ἀτιμάζεις πόλιν ; 240 
TE. ἥξει γὰρ αὐτά, κἂν ἐγὼ σιγῇ στέγω. 


ΟἹ. 
[οὐ Κ᾿ av πέρα φράσαιμι. 


οὐκοῦν ᾶ τ ἡ ζει καὶ σὲ χρὴ λέγειν ἐμοί. 


πρὸς τάδ᾽, εἰ θέλεις, 


θυμοῦ δι᾿ ὀργῆς ἥτις ἀγριωτάτη. 


OI. καὶ μὴν παρήσω γ᾽ οὐδέν, ὡς ὀργῆς ἔχω, 
ia Ot “γὰρ 


καὶ ξυμφυτεῦσαι τούργον, εἰργάσθαι θ᾽, 


ἅπερ Evvinp . 


336 κἀπαραίτητος Sehrwald. 


345 


οκών ἐμοὶ 
Ψ 
οσον 


837 ὁρμὴν List hand. Ὑ has been written over 


“ by an early hand (prob. 5), which has also sought to make yu into y in the text. 





334 πέτρου | φύσιν: Eur. Med. 1279 
ὦ τάλαιν᾽, ws ἄρ᾽ ἦσθα πέτρος ἢ olda'pos. 
For the periphrasis cp. Plat. Phaedr. 251 
B ἡ τοῦ πτεροῦ φύσις, Ξετὸ πτερόν, πεφυκὸς 
ὥσπερ πέφυκε, being constituted as it is: 
Timae. 45 B τὴν τῶν βλεφάρων φύσιν: 74 
Ὁ τὴν τῶν νεύρων φύσιν : 84 C ἡ TOU μυελοῦ 
φύσις: Legg. 145 Ὁ τὴν ὕδατος φύσιν. 
And so often in -Arist., δι. ἡ τοῦ πνεύ- 
ματος φύσις Meteor. 2. 8: h τῶν νεύρων 
φύσις Hest. Anim. 3. 5. 

8335 ποτέ, tandem aliguando: Phil. 
816 μέθες ποτέ: 2b. 1041 τίσασθ᾽ ἀλλὰ τῷ 
χρόνῳ ποτέ. 

336 ἀτελεύτητος, not brought to an 
end: //. 4. 175 ἀτελευτήτῳ ἐπὶ ἔργῳ. 


.Plut. 2707. 114 F τὸ γὰρ δὴ ἀτελεύτητον 


νομίζειν τὸ πένθος ἀνοίας ἐστὶν ἐσχάτης. 
Here, a man ‘with whom one cannot 
make an end,’—who cannot be brought 
to the desired issue. In freely render- 
ing, ‘Wilt thou never make an end?’ we 
remember, of course, that the adj. 
could not literally mean ‘not finishing.’ 
Possibly it is borrowed from the col- 
loquial vocabulary of the day: the tone is 
like that of the Latin odzosus. 

337 ἐμέμψω, aor. referring to the 
moment just past: so oft. ἐπήνεσα, ξυνῆκα, 


noOnv: ἕπτηξα (O. C. 1466): ἔφριξα (AZ. 
693): ἐδεξάμην (Εἰ, 668): ἀπέπτυσα 
(Eur. Avec. 1276). ὁμοῦ | ναίουσαν, 


while (or though) it dwells close to 


thee, —possesses and sways thee. So 
One: 1134 κηλὶς κακῶν ξύνοικος: Ll. 784 
βλάβη | ξύνοικος: At. 639 συντρόφοις] 
ὀργαῖς. But (as Eustathius saw, 755. 14) 
the words have a second meaning: ‘thou 
seest not that thine own [τὴν σήν, thy 
kinswoman, thy mother] i is dwelling with 
thee [as thy wife].’ The ambiguity of 
τὴν σὴν, the choice of the phrase ὁμοῦ 
ναίουσαν, and the choice of karetdes, 
leave no doubt of this. Cp. 261. 

338 GAN ἐμὲ ψέγεις : the thought of 
ὀργὴν ἐμέμψω τὴν ἐμήν returns upon itself, 
as if from a sense that the contrast be- 
tween ἐμέμψω and κατεῖδες would be 
imperfectly felt without such an iteration. 
This is peculiarly Sophoclean; cp. above 
166 (ἔλθετε καὶ νῦν): Schneidewin cp. 
also Az. 1111 οὐ. τῆς σῆς οὕνεκ᾽... | ἀλλ᾽ 
οὕνεχ᾽ ὅρκων... | σοῦ δ᾽ οὐδέν : and ‘similar- 
ly Ant. 465 ff., Trach. 431 ff., Zl. 361 ff. 

339 The emphasis on τοιαῦτα as well 
as on ovK warrants the repeated dv:.cp. 
139: Ant. 69 f.: Eur. Andr. 934 οὐκ ἂν 
ἔν γ᾽ ἐμοῖς δόμοις  βλέπουσ᾽ ἂν αὐγὰς τἄμ᾽ 
ἑκαρποῦτ᾽ ἂν λέχη. 

340 ἃ.. ἀτιμάζεις πόλιν: ἅ cogn. 
accus.: Az. 1107 τὰ σέμν᾽ ἔπη | κόλαζ᾽ 
ἐκείνους: Ant. 550 τί ταῦτ᾽ amas μ᾽; 
ἀτιμάζεις, by rejecting the request that he 
would speak : Ant. 544. 

841 ἥξει γὰρ αὐτά. The subject to 

ἥξει is Ἡρεβ μα μὲ left indeterminate: 


OIAITOYS TYPANNOS 57 


ΟΕ. What, basest of the base,—for thou wouldest anger a 


very stone,—wilt thou never speak out? 

Wilt thou never make an end? 
Thou blamest my temper, but seest not that to which 

no, thou findest fault with me. 

And who would not be angry to hear the words with 


thee? 
ἘΠ ΕΣ 

thou thyself art wedded : 
OE: 


Can nothing touch 


which thou now dost slight this city? 


Boe 
silence. 
OE. 
shouldst tell me thereof. 
es. 


I will speak no further ; 


The future will come of itself, though I shroud it in 


Then, seeing that i¢-must come, thow ὉΠ ἘΠ part 


rage, then, if thou wilt, with 


the fiercest wrath thy heart doth know. 
OE. Aye, verily, I will not spare—so wroth I am—to speak 


all my thought. 


Know that thou seemest to me een to have 


helped in plotting the deed, and to have done it, short of 


ὀργὴν τ.---τὴν σὴν δ᾽ L, and so almost all the later Mss. 
347 εἰργάσθαι δ᾽ L ist hand, but the δ᾽ has been 


σοὶ δ᾽, which Dindorf adopts. 


But one at least (V4) has τὴν 





‘(the things of which I wot) will come 
of themselves.’ The seer is communing 
with his own thought, which dwells 
darkly on the κακά of v. 329. αὐτά-Ξ 
αὐτόματα: 71. 17. 252 ἀργαλέον δέ μοί 
ἐστι διασκοπιᾶσθαι ἕκαστον... ἀλλά τις 
αὐτὸς ἴτω. Cp. the phrase αὐτὸ δείξει, 
ves ipsa arguct, the result will show: Soph. 
fr. 355 ταχὺ δ᾽ αὐτὸ δείξει τοὔργον. 

942 οὐκοῦν ἅ γ᾽ ἥξει. Elmsley, 
Nauck and Hartung ral οὐκ odv...€uol; 
but the positive χρὴ is stronger without 
the query. ‘Then, seeing that they will 
come, thou on thy part (καὶ σὲ) shouldest 
tell them to me.’ ‘The stress of καὶ falls 
primarily on σὲ, but serves at the ‘same 
time to contrast λέγειν with ἥξει. In ἅ 
y ἥξευ the causal force of the relative is 
brought out by ye: guippe guae ventura 
sint. 

848 ΣΦ. οὐκ dv πέρα φράσαιμι. The 
courteous formula (95, 282), just because 
it is such, here expresses fixed resolve.— 
ἥτις ἀγριωτάτη: 71. .17. 61 ὅτε τίς τε 
λέων... βοῦν ἁρπάσῃ ἥτις ἀρίστη : Plat. 
Apol. 23 A πολλαὶ ἀπέχθειαι...καὶ οἷαι 
χαλεπώταται: Dem. or. 2 ὃ 18 εἰ μὲν 
γάρ τις ἀνήρ ἐστιν ἐν αὐτοῖς οἷος ἔμπειρος 
πολέμου καὶ ἀγώνων [:ε. ἐστί], τούτους, 
K.T.A. 

345 kal μὴν with ye, ‘aye verily’: cp. 
El. 554, where ἢν ἐφῇς μοι is answered 
(556) by καὶ μὴν ἐφίημ᾽. (For ἃ slightly 
different καὶ μήν...γε, see O. Ο. 396.)— 
ὡς ὀργῆς ἔχω -Ξ ἔχων ὀργῆς ὡς ἔχω, being 


so wrothasIam. Thuc. 1. 22 ὡς ἑκατέ- 
ρων τις εὐνοίας ἢ μνήμης ἔχοι: Eur. Helen. 
313 πῶς δ᾽ εὐμενείας τοισίδ᾽ ἐν δόμοις ἔχεις ; 
παρήσω... οὐδὲν (τούτων) ἅπερ ξυνίημ᾽, 
I will leave unsaid nothing (of those 
things) which I comprehend, ze. I will 
reveal my whole insight into the plot. 
ξυνίημι suits the ‘intellectual pride of 


Oedipus: he does not say ‘think’ or 
‘suspect’: cp. 628. For γὰρ after ἴσθι 
cp. 277- 


347 Kal ξυμφυτεῦσαι.. -eipyao Bat θ᾽. 
καί, τε could no more stand for ‘amd’ 

..‘doth’ than et...gue could. καί here 
(adeo) implies, ‘no mere sywpathiser, but 
actually the plotter.’ Cp. O.C. 1394 καὶ 
(een) πᾶσι Καδμείοισι τοῖς σαυτοῦ θ᾽ ἅμα. 
ξυμφυτεῦσαι: Pind. /s¢h. 5 (6). 12 σύν τέ 
οἱ δαίμων φυτεύει δόξαν: ‘di. 953 Παλλὰς 
φυτεύει πῆμα: El, 198 δεινὰν δεινῶς 
προφυτεύσαντες | μορφάν (of crime). Her- 
mann preferred δ᾽ to τ᾽ after εἰργάσθαι, as 
meaning, ‘du¢ hast done it (only) by an- 
other’s hands’ (2.6. ‘though thou hast not 
executed it thyself’): this, however, be- 
sides being forced, destroys the climax.— 
ὅσον (εἶχες εἰργάσθαι) μὴ καίνων, so far 
as you could be the author of the deed 
without slaying: Thuc. 4. 19 φυλάσσειν 
δὲ καὶ τὴν νῆσον ᾿Αθηναίους μηδὲν ἧσσον, 
ὅσα μὴ ἀποβαίνοντας: ene πῆς γῆς 
ἐκράτουν ὅσα μὴ προϊόντες πολὺ ἐκ τῶν 
ὅπλων: Tr. 1214 | ὅσον γ᾽ av (sc. δρῴην 
τοῦτο) αὐτὸς μὴ ποτιψαύων χεροῖν. 


58 TO OKAEOYS 


μὴ χερσὶ Kalen" ei O ,ἐτύγχανες βλέπων, 
καὶ τοὔργον ἃ ἂν σοῦ τοῦτ᾽ ἔφην εἶναι μόνου. 


TE. ἄληθες ; ἐννέπω σὲ τῷ κηρύγματι 350 
ᾧπερ προεῖπας ἐμμένειν, Kad ἡμέρας 
τῆς νῦν προσαυδᾶν μήτε τούσδε μήτ᾽ ἐμέ, 
ὡς ὄντι γῆς τῆσδ᾽ ἀνοσίῳ μιάστορι. 
ΟΥ οὕτως ἀναιδῶς ἐξεκίνησας τόδε 
TO ῥημᾶ; καὶ ποῦ τοῦτο φεύξεσθαι δοκεῖς ; 355 
TE. πέφευγα' τἀληθὲς γὰρ ἰσχῦον τρέφω. 
Ol. πρὸς τοῦ διδαχθείς ; : οὐ γὰρ ἔκ γε τῆς τέχνης. 
TE. πρὸς σοῦ: σὺ γάρ μ᾿ ἄκοντα προὐτρέψω λέγειν. 
ΟἹ: ποῖον λόγον ; λέγ᾽ αὖθις, ὡς μᾶλλον μάθω. 
TE. οὐχὶ ξυνῆκας πρόσθεν ; ἢ ᾿κπειρᾷ “λέγων ; 260 
ΟΙ. οὐχ ὥστε γ᾽ εἰπεῖν γνωστόν: ἀλλ᾽ αὖθις φράσον. 


ἣν 


re-touched, to make θ᾽. εἰργάσθαι 6’ r 349 εἶναι was omitted by the rst hand 
in L, but has been written in very pale and faint ink above the line, between ἔφην and 
μόνον, by a hand of perh. the r2th cent. The later Mss. have εἶναι. Kirchhoff conj. 
τοῦτ᾽ ἔφην ἅπαν μόνου. 8360 L has ἢ ᾽κπειρᾶι λέγειν, with o written under the 
accent on Ney, and a mark of abbreviation, Δ, over εἰν. Diibner thinks that the rst 
hand wrote λέγ, denoting ew by the mark aforesaid, and indicating by o a reading 
λόγων, to which a marginal gloss by a later hand refers, εἰ πεῖραν λόγων Kuweis: then 





849 καὶ τοὔργον... τοῦτο, the doing - 


of this thing also, αὐτὴν τὴν πρᾶξιν, as 
dist. from the plotting and the direction 
of the act. 

850 ἄληθες: κιτιλ. The same word 
marks the climax of Creon’s anger in 
Ant.758:cp.Ar. Av. 393 ἐτεόν ; ete. ἐννέ- 
πω σὲ.. ἐμμένειν, 1 command that thou 
abide: so Phzl. τοι λέγω σε... λαβεῖν. 

351 ᾧπερ προεῖπας (sc. ἐμμένειν), by 
which thou didst proclaim that (all) 
should abide: this is better than taking 
᾿ᾧπερ as by attraction for ὅπερ, since mpo- 
εἴπον could take an acc. of the thing pro- 
claimed (e.g. ξενίαν, πόλεμον, θάνατον), 
but not of the edict itself (as κήρυγμα). 

353 ds dvtt...pideropt, an anaco- 
louthon for ws évra.. “μιάστορα, as if ἐν- 
νέπω σοί had preceded. ἐμέ just before 
made this necessary. In Eur. Med. 57 
most MSS. give ὥσθ᾽ ἵμερός μ᾽ ὑπῆλθε γῇ 
τε κοὐρανῷ | λέξαι μολούσῃ δεῦρο δε- 
σποίνης τύχας, where Porson, reading 
μολοῦσαν, admits that the dat. stands in 
Philemon’s parody (Athenaeus 288 D), 
ws ἵμερός μ᾽ ὑπῆλθε γῇ τε κοὐρανῷ | λέξαι 
μολόντι τοῦψον ὡς ἐσκεύασα. Elms. 
cp. Eur. Z. A. 491 ἄλλως τέ μ᾽ ἔλεος τῆς 


ταλαιπώρου κόρης | εἰσῆλθε συγγένειαν ἐν- 
νοουμένῳ. Conversely Thuc. 6. 85 § 2 
(rots ἐκεῖ ξυμμάχοις followed by Χίους, etc., 
in appos.). 

354 ἐξεκίνησας. ἐκκινεῖν is used of 
Starting game, Z/. 567 é&exlynoev πο- 
dow | ...€\apov: of rousing one from 
rest, 7; ry. 1242, and fig. of exciting pain 
which had been lulled, 26. 979. Here 
the notion is that of a startling utterance. 
Cp. the use of κινεῖν in the sense of 
mooting subjects which should not have 
been touched: Eur. Z/. 302 ἐπεὶ δὲ κινεῖς 
μῦθον, 1.6. since thou ast broached this 
theme: cp. O. C. 15264 δ᾽ ἐξάγιστα μηδὲ 
κινεῖται λόγῳ. In Eur. Med. 1317 τί 
τάσδε κινεῖς κἀναμοχλεύεις πύλας ; Porson, 
with the author of the Christus Patiens, 
reads λόγους, thinking that Ar. ud. 
1399 © καινῶν ἐπῶν. κινητὰ καὶ μοχλευτά 
alluded to that place. So ἀκίνητα (ἔπη) 
=dmoppyta O. C. 624, Ant. 1060 ὄρσεις 
με τἀκίνητα διὰ φρενῶν ppdaoa. | κίν ει, 
K.T.A. 

355 καὶ ποῦ «.7.\. And on what 
ground dost thou think to escape (punish- 
ment for) this thing? For ποῦ cp. 390: 
At. 1100 ποῦ σὺ στρατηγεῖς τοῦδε; Dis- 


OIAITOY2 ΤΥΡΆΝΝΟΣ 59 
slaying with thy hands. Hadst thou eye-sight, I would have 
said that the doing, also, of this thing was thine alone. 

TE. In sooth ?—I charge thee that thou abide by the decree 
of thine own mouth, and from this day speak neither to these nor 
to me: ¢ou art the accursed defiler of this land. 

ΟΕ. So brazen with thy blustering taunt? 
dost thou trust to escape thy due? 


And wherein 


TE. I have escaped: in my truth is my strength. 

ΟΕ. Who taught thee this? It was.not, at least, thine art. 

TE. Thou: for thou didst spur me into speech against my 
will. 

OE. What speech? Speak again that I may learn it better. 

Tre. Didst thou not take my sense before? Or art thou 


tempting me in talk? 
Or. No,I took it not so that I can call it known :—speak 
again. 


another hand wrote εἰν in full. Campbell holds that the 1st hand wrote λέγοι. All 

the later mss. have λέγειν; and I believe, with Diibner, that this was what the rst | 
hand in L meant to give. The superscript 0, however, is not (I think) from the first 
hand, but from a later one, prob. the same that wrote the marg. gloss. The ew may 
be from the first corrector (S). —Hartung reads ἢ πειρᾷ λέγων ; Campbell, ἢ ᾿κπειρᾷ 
λόγῳ; Wecklein and Bellermann, ἢ ἐκπειρᾷ λόγων ; Blaydes proposes οὐχὶ ξυνῆκας; 
πρὸς τί μου ᾿κπειρᾷ λέγειν; Mekler, ἢ πέτρᾳ ’deyov; F. W. Schmidt, ἢ ἑτέρᾳ λέγω; 





tinguish καί (1) prefixed to interrogative 
particles, when it expresses an objection : 
Aesch. Ag. 280 καὶ tis τόδ᾽ ἐξίκοιτ᾽ av 
ἀγγέλων τάχος; Dem. or. 19 § 257 (with 
Shilleto’s note), and καὶ πώς; passim: 
(2) suffxed, where, granting a fact, it 
asks for further information: Agam. 
278 ποίου χρόνου δὲ καὶ πεπόρθηται πό- 
Aus; (assuming it to be taken, when was 
it taken?) Eur. Alc. 834 ποῦ καί σῴε 
θάπτει; τοῦτο φεύγειν Πετε-Ξ τούτου τὴν 
δίκην ἐκφεύγειν: Eur. Med. 795 παίδων 
φόνον | φεύγουσα, fleeing from (the penal- 
ties of) the murder: Cic. Pro C/uent. 59 
§ 163 calumniam (=crimen calumniae) 
non effugiet. Butin Lys. 75, Erat. ὃ 34 
τοῦτο...οὐ φεύγωτε “1 do not avoid this 
point.’ 

356 ΖΦ. ἰσχῦον expresses the living 
strength of the divine instinct within 
him: cp. ζῶντα 482.—rpépw: see on 
ἐμπέφυκεν 299.—téxvys, slightly con- 
temptuous; cp. 288, 562, 709. 

858 προὐτρέψω: the midd., as 1446: 
but the act., Ant. 270, El. 1193. 

860 ἢ ᾿κπειρᾷ λέγων; or (while you 
do understand my meaning already) are 
you merely trying by your talk (λέγων) 


to provoke a still fuller statement of it? 
Her. 3. 135 δείσας μή εὑ ἐκπειρῷτο Aa- 
petos, was making trial of him: Ar. ΞΖ. 
1234 καί σου τοσοῦτο πρῶτον ἐκπειράσο- 
μαι, ‘thus far make trial of thee’ (test 
thee by one question). The notion of 
ἐκ in the compound is that of drawing 
forth something from the person tested. 
λέγων here implies za/e talk, cp. 1151 
λέγει yap εἰδὼς οὐδέν: Phil. 55 τὴν Φι- 
λοκτήτου σε δεῖ | ψυχὴν ὅπως λόγοισιν ἐκ- 
κλέψεις λέγων: where, as here, the 
partic. denotes the process. If we read 
λέγειν, we must supply ὥστε : ‘tempt- 
ing me so that I should speak’: a weak 
sense. λόγῳ could only mean, ‘by 
thy talk’: whereas it would naturally 
mean ‘in word’ (only, and not ἔργῳ). 
Musgrave conj. λοχών (laying a snare 
for me); Arndt μ᾽ ἑλεῖν ; (to catch me): 
Madvig ἐκ πείρας λέγεις; But, with λέ- 
γῶν, all i is, ᾿ think, sound. 

361 οὐχ ὥστε Ὗ K.T.A. οὐ (ξυνῆκα) 
οὕτω γ ἀκριβῶς ὥστε εἰπεῖν: cp. 1131- 
γνωστόν: ‘known.’ So the MSs.: but 
γνωτὰ 58, γνωτὸν 806. In fr. 262 ἐκ 
κάρτα βαιῶν γνωτὸς ἂν “γένοιτ᾽ ἀνήρ, yvw- 
τός = ‘well-known,’ γνώριμος : but Soph. 


60 ZOPOKAEOY2 


ΠῈ: 
ΟΙ. ae οὐ TL χαίρων 


φονέα σε φημὶ τἀνδρὸς οὗ ζητεῖς κυρεῖν. 
δίς γε πημονὰς ἐρεῖς. 


ΤΕ. εἴπω τι δῆτα καλλ᾽, ἵν ὀργίζῃ πλέον ; 
Ol. ὅσον γε χρήζεις" ὡς μάτην εἰρήσεται. 365 
TE. λεληθέναι σε φημὶ σὺν τοῖς φιλτάτοις 


αἴσχισθ' ὁμιλοῦντ᾽, οὐδ᾽ ὁρᾶν ἵν εἶ κακοῦ. 
Ol. ἢ καὶ γεγηθὼς ταῦτ᾽ ἀεὶ λέξειν δοκεῖς ; 


ἐν Ὁ 


εἴπερ τί γ᾽ ἐστὶ τῆς ἀληθείας σθέυος. 
OI. ad eo, πλὴν aol: σοὶ δὲ τοῦτ᾽ οὐκ ἔστ᾽, ἐπεὶ 


Ὁ 


τυφλὸς τά τ᾽ ὦτα τόν τε νοῦν τά τ ὄμματ᾽ εἶ. 


TE. 


σὺ δ᾽ ἀθλιός γε ταῦτ᾽ ὀνειδίζων, ἃ σοὶ 


οὐδεὶς ὃς οὐχὶ τῶνδ᾽ ὀνειδιεῖ τάχα. 


Ol. 


μιᾶς τρέφει πρὸς νυκτός, ὥστε μήτ᾽ ἐμὲ 


μήτ᾽ ἄλλον, ὅστις φώς ὁρᾷ, βλάψαι TOT av. 375 


TE. 


οὐ γάρ σε μοῖρα πρός γ᾽ ἐμοῦ πεσεῖν, ἐπεὶ 


ἱκανὸς ᾿Απόλλων, ᾧ τάδ᾽ ἐκπρᾶξαι eee 
OI. Κρέοντος ἢ σοῦ ταῦτα τἀξευρήματα; 


ΤΕ. 
Ol. 


Κρέων δέ σοι THM οὐδέν, ἀλλ᾽ αὐτὸς σὺ σοί, 
ὦ πλοῦτε καὶ τυραννὶ καὶ τέχνη τέχνης 


380 


ὑπερφέρουσα τῷ πολυζήλῳ Pia, 


374 judas] μαίας G. Wolff. 


376 με μοῖρα πρός γε σοῦ L (and so the later Mss., 





used γνωστός in the same sense in the 
Hermione (Antiatticista 87. 25). It has 
been held that, where a sigmatic form 
of the verbal (as Ὑνωστός) existed along 
with the non-sigmatic (as γνωτόξ), Attic 
usage distinguished γνωστός as = ‘what 
can be know ῃ from words as= ‘what zs 
known.’ But there is no ground for as- 
suming that such a distinction was ob- 

served. See Appendix, ἢ. on v. 361. 
‘862 οὗ {nreis κιτ.λ. φημί ce φονέα 
κυρεῖν (ὄντα) τοῦ ἀνδρὸς οὗ (τὸν φονέα) 
ζητεῖς. 

868 ἀλλ’ οὔ τι χαίρων: cp. Ph. 1299 
(η.). πημονὰς: 2.4. such charges are 
downright calamities, infamies. There 
is something of a colloquial tone in 
the phrase: cp. 42. 68 μηδὲ “συμφορὰν 
δέχου Ι τὸν ἄνδρα: El. 301 ὁ πάντ᾽ ἄναλ- 
Kis οὗτος, ἡ πᾶσα βλάβη. Cp. 336 ἀτε- 
λεύτητος. 

364 εἴπω, delib. subjunct. : Eur. Jon 
758 εἴπωμεν, ἢ σιγώμεν, ἢ τί δράσομεν; 

866 σὺν τοῖς φιλτάτοις κ.τιλ.-- σὺν 
τῇ φιλτάτῃ (Locasta): since ὁμιλοῦντ᾽ im- 
plies wedlock, and not merely the com- 


panionship denoted by ξυνών in 457: for 
the allusive plural, cp. 77. 335 οὕστινας 
(meaning Iolé): £7. 652 φίλοισι (Ae- 
gisthus). 

367 ἵν᾽ εἴ κακοῦ: 


ΗΡ' 413) 1442. 
Tr. 375 ποῦ ποτ᾽ εἰμὶ πράγματος; 
868 ἢ καὶ: ‘dost thou indeed? 


Aesch. Zum. 402 ἢ καὶ τοιαύτας τῷδ᾽ 
ἐπιρροιζεῖς φυγάς ; 

370 πλὴν col σοὶ δὲ κιτ.λ. Note 
in these two vv. (1) the rhetorical itera- 
tion (ἐπαναφορα) of the pers. pron., as 
in O.C. 250 πρός σ᾽ ὅτι σοι φίλον ἐκ σέθεν : 
7b. 787 οὐκ ἔστι σοι ταῦτ᾽, ἀλλά σοι ταῦτ᾽ 
ἔστ᾽: Phil. 1054 πλὴν els σέ' σοὶ δέ: 
Isocr. or. 15 § 41 κιδυνεύων τὰ μὲν ὑφ᾽ 
ὑμών τὰ δὲ μεθ᾽ ὑμών τὰ δὲ δι᾽ ὑμᾶς τὰ 8 
ὑπὲρ ὑ ὑμών. (2) the ninefold τ (παρήχησι5) 
1Π 721} ΟΡ. 425: ΟΣ Ὁ. 1847: At. £36 
ἐὰν τὸ ταχθὲν εὖ τολμᾷ τελεῖν. Similarly 
π, Zl. 210, Ai. 1112: σ, Eur. Med. 476 
ἔσωσά σ᾽" ὡς ἴσασιν Ελλήνων ὅσοι, κ.τ.λ. : 
Ennius “47172.1. 151 Ο Tite tute Tati tibi 
tanta tyranne tulisti: Cic. Pro Cluent. 
35 § 96 non fuit igitur tllud iudicium 
tudictt simile, tudices. 


ΟΥΑΙ ΠΟΥΣ ΤΥΡΆΝΝΟΣ 61 


ae 
thou seekest. 
OE: 
so dire. 
peo, 
more wroth ? 
Or. 
TE. 


I say that thou art the slayer of the man whose slayer 
Now thou shalt rue that thou hast twice said words 
Wouldst thou have me say more, that thou mayest be 


What thou wilt ; it will be said in vain. 
I say that thou hast been living in unguessed shame 


with thy nearest kin, and seest not to what woe thou hast come. 
ΟΕ. Dost thou indeed think that thou shalt always speak 


thus without smarting ? 
TE. 


Yes, if there is any strength in truth. 


ΟΕ. Nay, there is,—for all save thee ; for thee that strength 
is not, since thou art maimed in ear, and in wit, and in eye. 

TE. Aye, and thou art a poor wretch to utter taunts which 
every man here will soon hurl at thee. 

ΟΕ. Night, endless night hath thee in her keeping, so that 
thou canst never hurt me, or any man who sees the sun. 


EE: 


No, thy doom is not to fall by me- Apollo is enough, 
whose care it is to work that out. 


ΟΕ. Are these Creon’s devices, or thine ? 
TE. Nay, Creon is no plague to thee; thou art thine own. 
OE. Ὁ wealth, and empire, and skill surpassing skill in 


except that A has ge...ye σοῦ): ce μοῖρα πρός γ᾽ ἐμοῦ Brunck. 


life’s keen rivalries, 


379 Κρέων δέ 





872 ἀθλιος, of wretched jolly. Cp. 
the use of dvoABos, Az. 1156, Ant. 1025 
(joined with ἄβουλος), μέλεος (Az. 621), 
κακοδαίμων, K.T.r. 

373 οὐδεὶς (ἔστιν) ὃς οὐχὶ-τε πᾶς τις: 
[Plat.] Adc. 1. 103 Β οὐδεὶς ὃς οὐχ ὑπερ- 
βληθεὶς... πέφευγε. At. 725 ἤρασσον... 
οὔτις ἔσθ᾽ ὃς ov. More properly οὐδεὶς 
ὅστις οὐ, declined (by attraction) in both 
parts, as Plat. Phaedo 117 Ὁ οὐδένα ὅντινα 
οὐ κατέκλασε τῶν παρόντων. 

874 μιᾶς τρέφει πρὸς νυκτός, thou 
art cherished by (thy life is passed in) 
one unbroken night: the pass. form of 
pla νύξ oe τρέφει. Cp. fr. 524 (N.?), 
τερπνῶς yap del πάντας ἁνοία τρέφει, 
folly ever gives a joyous /ife: fr. 532. 4 
βόσκει δὲ τοὺς μὲν μοῖρα ducapepias, | τοὺς 
δ᾽ ὄλβος ἡμῶν: Eur. “7122. 367 ὦ πόνοι 
τρέφοντες βροτούς, cares that make up the 
life of men. μιᾶς might be simply μόνης, 
but, in its emphatic place here, rather= 
‘unbroken,’ unvaried by day: cp. Ar. 
LRhet. 3. 9. τ (λέξιν) εἰρομένην καὶ τῷ συν- 


δεσμῷ μίαν, forming one continuous 
chain. Theingenious conj. walas (nurse) 
seems to me far less forcible. 

376 (οὐκ ἐγώ ce βλάψω), οὐ γὰρ μοῖρα 
σε πεσεῖν κ.τ.λ. 

877 ἐκπράξαι, ‘to accomplish’ (not 
to ‘exact’); τάδε has a mysterious vague- 
ness (cp. 341), but includes τὸ πεσεῖν σε, 
as in 1158 τόδ᾽ refers to ὀλέσθαι. 

379 Κρέων δὲ-- May, Creon,’—in- 
troducing an objection, as 77. 729 τοι- 
adra δ᾽ ἂν λέξειεν κιτ.λ.: O.C. 395 γέροντα 
δ᾽ ὀρθοῦν φλαῦρον: and 72d. 1443. 

881 to πολυζήλῳ βίῳ, locative 
dative, defining the sphere of ὕὑπερ- 
φέρουσα, like ἔτι μέγας οὐρανῷ | Ζεύς 
ΕἸ. 174. twodv{yAw=full of emulation 
(ζῆλος). Others understand, ‘in the much- 
admired life’ (of princes). This is the 
sense of πολύζηλον (πόσιν) in 77. 185. But 
(1) βίῳ seems to denote life generally, 
rather than a particular station: (2) the 
phrase, following πλοῦτε καὶ τυραννΐ, 
would be a weak addition. τέχνη τέχ- 


62 ΣΟΦΟΚΛΕΘΥΣ 


ὅσος παρ᾽ ὑμῖν ὃ φθόνος φυλάσσεται, 
εἰ τῆσδέ ig ἀρχῆς οὕνεχ᾽, ἣν ἐμοὶ πόλις 
δωρητόν, οὐκ αἰτητόν, ,εἰσεχείρισεν, 


ταύτης Κρέων ὁ πιστός, ous ἀρχῆς φίλος, 


385 


λάθρᾳ μ᾽ ὑπελθὼν ἐκβαλεῖν ἱμείρεται, 
ὑφεὶς μάγον τοιόνδε μηχανορράφον, 
δόλιον ἀγύρτην, ὅστις ἐν τοῖς κέρ εσιν 


μόνον δέδορκε, τὴν τέχνην δ᾽ ἔφυ τυφλός. 


ἐπεὶ φέρ᾽ εἰπέ, “ποῦ σὺ μάντις εἶ “σαφής; 
ἡ ῥαψῳδὸς ἐνθάδ᾽ ἣν κύων, 


ἰχὰ 


πῶς οὔχ, O 


390 


ηὔδας τι τ ἀστοῖσιν ἐκλυτήριον ; 
καΐτοι τό γ᾽ αἰνιγμ᾽ οὐχὶ τοὐπιόντος ἣν 
ἀνδρὸς διειπεῖν, ἀλλὰ “μαντείας ἔδει: 


> 
ἣν οὔτ 


ἀπ᾽ οἰωνῶν σὺ προὐφάνης ἔχων 


395 


οὔτ᾽ ἐκ θεῶν του γνωτόν" ἀλλ᾽ ἐγὼ μολών, 
ὁ μηδὲν εἰδὼς Οἰδίπους, ἔπαυσά νιν, 
γνώμῃ κυρήσας οὐδ᾽ ἀπ᾽ οἰωνῶν ΠΩΣ: 


: Κρέων γε Brunck. 


996 τοῦ L, τοὺ Τ. 





vys | ὑπερφέρουσα refers to the view that 
the art of ruling is the highest of arts: 
cp. Phd, 138 τέχνα γὰρ τέχνας ἑτέρας 
προὔχει | καὶ γνώμα, map’ ὅτῳ τὸ θεῖον | 
Διὸς σκῆπτρον ἀνάσσεται: for skill and 
wit (γνώμη), surpassing those of other 
men, belong to him by whom is swayed 
the godlike sceptre which Zeus gives. 
Xen. Mem. 4. 2. 11 μεγίστης ἐφίεσαι 
τέχνης" ἔστι yap τῶν βασιλέων αὕτη, Kal 
καλεῖται βασιλική. But there is also an 
allusion to the skill shown in solving the 
riddle, by which QOed. surpassed the 
μαντικὴ τέχνη of Teiresias (cp. 357). 
382 tap ὑμῖν.. φυλάσσεται, is guard- 
ed, stored, in your keeping: 7z.¢. how 
much envy do ye tend to excite against 
those who receive your gifts. φυλάσ- 
σεται, stronger than τρέφεται, represents 
envy as the zzseparable attendant on 
success: cp. O.C. 1213 σκαιοσύναν φυ- 
λάσσων, stubborn in folly: Eur. Jon 735 
ἄξι᾽ ἀξίων γεννητόρων | ἤθη φυλάσσεις. 
384 δωρητόν, οὐκ αἰτητόν, feminine. 
The adjectives might be neuter: ‘a thing 
given, not asked.’ but this use of the 
neuter adj., when the subject is regarded 
in its most general aspect, is far most 
common in simple predications, as //. 2. 
204 οὐκ ἀγαθὸν πολυκοιρανίη: Eur. Hipp. 


109 τερπνὸν ἐκ kuvaylas | τράπεζα πλήρης. 
And γνωτόν in 396—which must agree 
with 7»—favours the view that here also 
the adjectives are fem. Cp. 2). 2. 742 
κλυτὸς Ἱπποδάμεια: Thuc. 2. 41 γῆν 
éoBarév: 7. 87 ὀσμαὶ οὐκ dvexrol: Plat. 
Rep. 573 Β μανίας... ἐπακτοῦ: [Plat.] 
Lryxias 398 Ὁ ἀρετὴ διδακτός: O.C. 
1460 πτερωτὸς βροντή: 77. 446 εἰ...μεμπ- 
τός εἰμι (Deianeira). 

985 pile redundant, for emphasis: 
Ken. Cyr.8. 7. 9 τὸ δὲ προβουλεύειν 
καὶ τὸ ἡγεῖσθαι, ἐφ᾽ ὅτι a” καιρὸς δοκῇ 
εἶναι, τοῦτο προστάττω. 

887 ὑφεὶς, having secretly sent as his 
agent, ‘having suborned.’ [Plat.] A xio- 
chus 368 E προέδρους éyxabérous b Pév TES, 
‘having privily brought in suborned presi- 
dents.’ The word μάγος expresses con- 
tempt for the rights of divination practised 
by Teiresias: ἀγύρτης taunts him as a 
mercenary impostor. So Plut. Mor. 
165 F joins ἀγύρτας καὶ γόητας, Zosimus 
I. Ir μάγοις τε καὶ ἀγύρταις. The pas- 
sage shows how Asiatic superstitions had 
already spread among the vulgar, and 
were scorned by the educated, in Greece. 
The Persian μάγος (as conceived by the 
Greeks) was one who claimed to com- 
mand the aid of beneficent deities (daé- 


ΟἸΔΙΠΟῪΣ ΤΥΡΑΝΝΟΣ 63 


how great is the envy that cleaves to you, if for the sake, yea, 
of this power which the city hath put into my hands, a gift 
unsought, Creon the trusty, Creon mine old friend, hath crept 
on me by stealth, yearning to thrust me out of it, and hath 
suborned such a scheming juggler as this, a tricky quack, who 





hath eyes only for his gains, but in his art is blind! 


Come, 
seer ? 


was need of a seer’s skill; 


now, tell me, where hast thou proved thyself a 
Why, when the Watcher was here who wove dark 
song, didst thou say nothing that could free this folk ? 
the riddle, at least, was not for the first comer to read; 


Yet 
there 


and none such thou wast found 


to have, either by help of birds, or as known from any god: 
no, I came, I, Oedipus the ignorant, and made her mute, 
when I had seized the answer by my wit, untaught of birds. 





ἕονες ἀγαθοεργοί), while the γόης was 
properly one who could call up the dead 
us 1. 490: cp. Plut. De Defect. Orac. 
c. 10). So Eur. O7. 1496 (Helen has 
been spirited away), ἢ φαρμάκοισιν (by 
charms), ἢ ἢ μάγων | τέχναισιν, ἢ θεῶν κλο- 
παῖς. 

888 ἀγύρτην (ἀγείρω), a priest, esp. 
of Cybele (μητραγύρτης, or, when she 
had the lunar attributes, μηναγύρτης), 
who sought money from house to house 
(ἐπὶ τὰς τῶν πλουσίων θύρας ἰόντες, Plat. 
Rep. 364 B), or in public places, for pre- 
dictions or expiatory rites: Maximus Ty- 
rius 19. 3 τῶν ἐν τοῖς κύκλοις ἀγειρόν- 
TWV..., of δυοῖν ὀβολοῖν τῷ προστυχόντι 
ἀποθεσπίζουσιν. ---ἐν τοῖς κέρδεσιν, in the 
case of gains: cp. “42.1315 ἐν ἐμοὶ θρασύς ; 
rather than, ‘on opportunities for gain’ 
(=6rav ἢ κερδαίνειν) as Ellendt takes it. 
Cicero’s videbat in litteris (Tusc. 5. 38. 
112, quoted by Schneid.) seems not 
strictly similar, meaning rather ‘in the 
region of letters’ (like 272 tenebris). 

390 ἐπεὶ -- “(οτ᾽ (if this is of true): 
Εἰ. 351 οὐ ταῦτα...δειλίαν ἔχει; | ἐπεὶ 
δίδαξον, «.7.A.; so O. C. ούο.---ποῦ; 
where? 2.6. in what sense? Tur. 7071 
528 ποῦ δέ μοι πατὴρ σύ ;--εἶ σαφής = πέ- 
φηνας ὦν: Cp. 358: 
᾿ς 991 κύων, esp. because the Sphinx 
» was the watchful agent of Hera’s wrath: 
cp. 36. Ar. Ran. 1287 has a line from 
the Σῴίγξ of Aesch., Σφίγγα δυσαμε- 
pidv [vulg. δυσαμερίαν] πρύτανιν κύνα 
πέμπει, “the watcher who presides over 
evil days’ (for Thebes).—pawSds, chant- 
ing her riddle (in hexameter verse), as 
the public reciters chanted epic poems. 
The word is used with irony: the baneful 


lay of the Sphinx was not such as the 
servant of Apollo chants. Cp. 130. 
393 f. τό γ᾽ alvtyp is nominative: 


_the riddle did not belong to (was not for) 


the first comer, that he should solve it. 
O. C. 751 οὐ γάμων | ἔμπειρος, ἀλλὰ τοὐ- 
πιόντος ἁρπάσαι. Thuc. 6. 22 πολλὴ γὰρ 
οὖσα [ἡ στρατιὰ] οὐ πάσης ἔσται πόλεως 
ὑποδέξασθαι. ὁ ἐπιών, any one who 
comes up; cp. Plat. Rep. 372 D ws νῦν 
ὁ τυχὼν Kal οὐδὲν προσήκων ἔρχεται ἐπ᾽ 
αὐτό.---διειπεῖν, ‘to declare,’ ‘to solve’: 
cp. 854. διά implies the drawing of 
clear distinctions; cp. O. C. 295 διειδέναι, 
ditudicare, Ὡς 

395 f. ἣν ott ἀπ᾽ οἰωνῶν ἔχων οὔτ᾽ ἐκ 
θεῶν του γνωτὸν (ἔχων) προὐφάνης : and 
thou wast not publicly seen to have this 
art, either from (ἀπ᾽) birds, or as known 
through the agency of (ék) any god. 
προὐφάνης. when brought to ἃ public 
test. For ἀπό cp. 43: ἐκ with θεῶν του, 
of the primary or remoter agent (Xen. 
Flellen. 3. 1.6 ἐκ βασιλέως ἐδόθη), mean- 
ing by a φήμη (43) or other sign. yvwrov: 
cp. on 384.—podov: he was a mere 
stranger who chanced to arrive then. 

397 ὁ μηδὲν εἰδὼς -- ὅστις μηδὲν ἤδη, 
‘I, ὦ man who knew nothing,’ the generic 
μή, here with concessive force,—‘though 
I knew nothing, I silenced her’ (qui 
nihil sezvem, vici tamen). So in Dem. or. 
19 ὃ 31 the generic μή has a causal force: 
ἡ βουλὴ δέ, ἡ μὴ κωλυθεῖσα ἀκοῦσαι τἀληθῆ 
παρ᾽ ἐμοῦ, οὔτ᾽ ἐπήνεσε τούτους, κιτ.λ. | 
(‘the senate, a dody which had not been 
prevented,’ etc.). See Whitelawin Zrans. 
Camb. Phil. Soc., 1886, Ὁ. 17. Cp. °38, 
875, Iotg. “8 


64 


ΣΟΦΟΚΛΕΟΥΣ 


ὃν δὴ σὺ πειρᾷς ἐκβαλεῖν, δοκῶν θρόνοις 


παραστατήσειν τοῖς Κρεοντείοις πέλας. 


400 


κλαίων δοκεῖς μοι καὶ σὺ χώ συνθεὶς τάδε 
ἀγηλατήσειν᾽ εἰ δὲ μὴ ᾿δόκεις γέρων 
εἶναι, παθὼν ἔγνως ἂν οἷά περ φρονεῖς. 


rs 
© 


ἡμῖν. μὲν εἰκάζουσι καὶ τὰ τοῦδ᾽ ἔπη 
ὀργῇ λελέχθαι καὶ τὰ σ᾽, Οἰδίπου, δοκεῖ. 


405 


eo οὐ τοιούτων, ἀλλ᾽ ὅπως τὰ τοῦ θεοῦ 
μαντεῖ" ἄριστα λύσομεν, τόδε σκοπεῖν. 


TE. 


εἰ καὶ τυραννεῖς, ἐξισωτέον τὸ “γοῦν 


ἴσ᾽ ἀντιλέξαι" τοῦδε. γὰρ κἀγὼ κρατώ. 


οὐ γάρ τι σοὶ ζῶ δοῦλος, ἀλλὰ Λοξίᾳ: 


ATO. τ 


ὥστ᾽ οὐ Κρέοντος προστάτου γεγράψομαι. 


λέγω δ᾽, 


ἐπειδὴ καὶ τυφλόν μ᾽ ὠνείδισας: 


σὺ καὶ δέδορκας κοὺ βλέπεις ἵ ἵν εἶ κακοῦ, 
οὐδ᾽ ἔνθα ναίεις, οὐδ᾽ ὅτων οἰκεῖς μέτα. 


ap’ otc θ᾽ ἀφ᾽ ὧν ei * 


Kat λέληθας ἐχθρὸς ὧν 


415 


τοῖς σοῖσιν αὐτοῦ ἘΣ κἀπὶ γῆς ἄνω, 


405 Οἰδίπου. 


L and the other Mss. support this form of the voc. here, and in O. C. 
557, 13463 but Οἰδίπους (voc.) in twelve other places. 


Elmsley and Reisig, whom 





400 πέλας, adv., so Aesch. Zhed. 669 
παραστατεῖν πέλας. 

401 κλαίων: cp. 368, 1152: Ant. 754 
κλαίων φρενώσεις.---ὁ συνθεὶς, Creon, as 
whose agent (387) Teir. is regarded: so 
in Thuc. 8. 68 ὁ τὴν γνώμην εἰπών is 
contrasted with ὁ τὸ πρᾶγμα ξυνθείς. 

402 ἀγηλατεῖν -- τὸ ἄγος ἐλαύνειν (see 
on 98), in this case ἀνδρηλατεῖν (100), to 
expel the μιάστωρ. Her. 5. 72 Κλεομένης 

«.«ἀγηλατέει ἑπτακόσια ἐπίστια (house- 
holds) ᾿Αθηναίων. The smooth breathing 
is supported by Hesychius, by the gram- 
marians in Bekker’s Anecd. 1. 328. 32, 
and by most Mss. of Soph.; while the 
aspirate is given by L here, by Eusta- 
thius (1704, 5), and by Suidas, who quotes 
this verse. Curtius distinguishes (1) dy-, 
dy-os, guilt, object of awe, whence éva- 
vis: Skt. dg-as, vexation, offence: Etym. 
§ 116: (2) root ay, afo-wat reverence, 
dy-to-s holy, a@y-vé-s pure: Skt. jag (7d g- 


᾿ς @-mi), reverence, consecrate: Etym. § 118. 


In Aesch. Cho. 154 and Soph. Ant. 775 
he would with Herm. write ayos as = ‘con- 
secrated offering.’ In both places, how- 
ever, ἄγος (=fiaculum) satisfies the sense 


=) 


. (see ng on “22. 775); 


and for ayos there 
is no other evidence. But this, at least, 
seems clear: the compound synonym for 
τὸ ἄγος ἐλαύνειν (Thuc. 1. 126) should be 
written ἀγηλατεῖν. 

᾽δόκεις is the scornful phrase of an 
angry man; I know little concerning 
thee, but from thine aspect I should 
judge thee to be old: cp. 562 where Oed. 
asks, τότ᾽ οὖν ὁ μάντις οὗτος ἦν ἐν τῇ 
τέχνῃ; Not (1) ‘seemed,’ as opposed to 
really being; nor (2) ‘wast felt by me’ 
to be old: a sense which the word surely 
could not yield. 

408 παθὼν, by bodily pain, and not 
merely μαθών, by reproof: cp. 641,—old 
περ φρονεῖς : see on 624 οἷόν ἐστι τὸ φθο- 
νεῖν. 

405 ὀργῇ, modal dat., cP 
θυμῷ.--- καὶ τὰ σ᾽ κιτ.λ., the eli 
329: see on 64. 

407 τόδε emptigtical 
λύσομεν, this we must consid 
ταύτης: 50 77. 458 τὸ un πυθέσθα. 
τό μ᾽ ἀλγύνειεν dv: Ph. 913. : 

408 εἰ καὶ κτλ. For εἰ καὶ see on 
205.---ἐξισωτέον κ.τ.λ. Ξ δεῖ ἐξισοῦν τὸ γοῦν 














ὃ. 6 
as in 


OIAITTOYS TYPANNOS 65 


And it is I whom thou art trying to oust, thinking to stand 
close to Creon’s throne. Methinks thou and the plotter of 
these things will rue your zeal to purge the land. Nay, didst 
thou not seem to be an old man, thou shouldst have learned 
to thy cost how bold thou art. 

CH. To our thinking, both this man’s words and thine, 
Oedipus, have been said in anger. Not for such words is our 
need, but to seek how we shall best discharge the mandates of 
the god. 

TE. King though thou art, the right of reply, at least, 
must be deemed the same for both; of that I too am lord. 
Not to thee do I live servant, but to Loxias; and so I 
shall not stand enrolled under Creon for my patron. And 
I tell thee—since thou hast taunted me even with blindness 
—that thou hast sight, yet seest not in what misery thou art, 
nor where thou dwellest, nor with whom. Dost thou know 


of what stock thou ae 
foe to thine own kin, 


Dindorf follows, hold Οἰδίπους to be alone correct. 
It is more probable that both forms were admissible. 


mends: Οἰδίπου. 


And thou hast been an unwitting 
in the shades, and on the earth above; 


Here, at least, euphony recom- 
413 δέδορκασ 





ἴσα ἀντιλέξαι, one must equalize the right 
at least of like reply; 2. 6. you must make 
Foam CLR TMCS as to grant me the 
right of replying at the same length. 
The phrase is a pleonastic fusion of (1) 
ἐξισωτέον τὸ ἀντιλέξαι with (2) σνγχώρη 
τέον τὸ ἴσα ἀντιλέξαι. 

410 £. Λοξίᾳ: see note to 853.---ὥστ᾽ 
ov Κρέοντος κιτιλ. ‘You charge me 
with being the tool of Creon’s treason. 
I have a right to plead my own cause 
when I am thus accused. I am not like 
a resident alien, who can plead before 
a civic tribunal only by the mouth of that 
patron under whom he has been regis- 
tered.” Every μέτοικος at Athens was 
required ἐπιγράφεσθαι προστάτην, ἢ. ὃς to 
have the name of a citizen, as patron, 
inscribed over his own. In default, he 
was liable to an ἀπροστασίου γραφή. Ar. 







or ἣν ὃ ἐν ᾿Ωρωπῷ μετοί- 
paying the alien’s tax) 
een, scorns, will 


cp. Ar. Z£g. 1370 οὐδεὶς 
Ovi εγγραφήσεται, ἀλλ᾽ ὡσ- 
wep ἦν Τὸ πρῶτον ἐγγεγράψεται: 
* Theocr. 18. 47 γράμματα δ᾽ ἐν prog γε- 
Ls. iE 
9 


γράψεται, remain written. aie: the gen. 
Κρέοντος cp. Ar. £y. 714 τὸν δῆμον σε- 
αυτοῦ νενόμικας. 

412 λέγω δ᾽, a solemn exordium, be- 
Pee attention : Cp. 449. - τυφλόν μ᾽ 

νείδισας. As ὠνείδισας could not stand 
for ἀπεκάλεσας, ‘called me reproachfully,’ 
τυφλόν must stand for ws τυφλὸν ὄντα. 
For the ellipse οἵ ὄντα, cp. 2. 899 ὡς δ᾽ 
ἐν γαλήνῃ πάντ᾽ ἐδερκόμην τόπον : for that 
of ὡς, Ο. C. 142 μή μ᾽, ἱκετεύω, προσίδητ᾽ 
ἄνομον. 

418 σὺ καὶ δέδορκας. ‘Thou doth 
hast sight azd@ dost not see,’ ζ. 6. thou hast 
sight, and at the same time dost not see. 
The conject. of Reiske and Brunck, σύ, 
kal dedopxws (though having sight), οὐ 
βλέπεις, spoils the direct contrast with 
τυφλόν. 

414 ἔνθα ναίεις might mean, ‘in what 
a situation thou art’: but, as distinguished 
from the preceding and following clauses, 
is best taken literally: ‘where thou 
dwellest,’ —viz. in thy murdered father’s 
house. 

415 dp οἶσθα κιτ.λ. Thy parents 
are unknown to thee. Yea, and (καὶ) 
thou knowest not how thou hast sinned 
against them,—the dead and the living. 


5. 


66 ZOPOKAEOYS 


καί σ᾽ ἀμφιπλὴξ μη ee TE καὶ τοῦ σοῦ πατρὸς 


ἐλᾷ ποτ᾽ ἐκ vas τῆσδε 
βλέποντα νῦν μὲν ὄρθ᾽, 


δεινόπους apd, 
ἔπειτα δὲ σκότον. 
βοῆς δὲ τῆς σῆς ποῖος οὐκ ἔσται ,λιμήν, 


420 


ποῖος Κιθαιρὼν οὐχὶ σύμφωνος τάχα, 
ὅταν καταίσθῃ τὸν ὑμέναιον, ὃν δόμοις 
V ἄνορμον εἰσέπλευσας, εὐπλοίας τυχών; 


ἄλλω 


ων δὲ πλῆθος οὐκ ἐπαισθάνει κακῶν, 
ἃ σ᾽ ἐξισώσει σοΐ τε καὶ τοῖς σοῖς τέκνοις. 


425 


πρὸς ταῦτα καὶ Κρέοντα καὶ “τοὐμὸν στόμα 
προπηλάκιζε' σοῦ γὰρ οὐκ ἔστιν βροτῶν 
κάκιον ὅστις ἐκτριβήσεταί ποτε. 


ΟἹ: 


οὐκ εἰς ὄλεθρον ; 


ἢ ταῦτα Ont ἀνεκτὰ πρὸς τούτου κλύειν ; 
οὐχὶ θς OOOvD ; 


οὐ Nas 


430 


-ἄψορρος οἴκων τῶνδ᾽ ἀποστραφεὶς ἄπει ; 
TE. οὐδ᾽ ἱκόμην ἔγωγ᾽ av, εἰ σὺ μὴ κάλεις. 


ΟΙ. οὐ γάρ τί σ᾽ ἤδη μῶρα φωνήσοντ᾽, 


ἐπεὶ 


Ὁ σχολῇ σ᾽ ἂν οἴκους τοὺς ἐμοὺς ἐστειλάμην. 


καὶ L. δεδορκὼς κοὺ r. 


420 λιμὴν] μυχὸς Wecklein. 


434 σχολῇ σ᾽ MSS.: 





417 ἀμφιπλὴξ: as in 77. 930 ἀμφι- 
πλῆγι φασγάνῳ-εα sword which smites 
with both edges, so here ἀμφιπλὴξ 
apd is properly a curse which smites on 
both stdes,—on the mother’s and on the 
father’s part. The pursuing ᾿Αρά must 
be conceived as bearing a whip with 
double lash (διπλῇ μάστιξ, Az. 242). Cp. 
ἀμφίπυρος, carrying two torches (7; Te 
214). The genitives μητρός, πατρός 
might be causal, with ἀμφιπλήξ, ,‘smiting 
twice—for mother and for sire,’ but are 
better taken with dpa, which here= 
Ἐρινύς: cp. Aesch. Zheb. 70 'Αρά τ᾽, 
᾿Ερινὺς πατρὸς -ἡ μεγασθενής. 

418 δεινόπους, with dread, untiring 
chase: so the Fury, who chases guilt ‘as 
a hound tracks a wounded fawn’ (Aesch. 
Eum. 246), is χαλκόπους (£7. 401), τανύ- 
mous (Az. 837), καμψίπους (' fleet,’ Aesch. 
Theb. 791). 

419 βλέποντα x.7.d., 2.6. τότε σκότον 
βλέποντα, εἰ καὶ νῦν ὀρθὰ βλέπεις. The 
Greek love of direct antithesis often co- 
ordinates clauses where we must subordi- 
nate one to the other: cp. below, 673: 
Isocr. or. 6 § 54 πῶς οὐκ αἰσχρόν, ...τὴν 
μὲν Εὐρώπην καὶ τὴν᾿ Aclav μεστὴν πεποιη- 
κέναι τροπαίων, .. «ὑπὲρ δὲ τῆς πατρίδος... 


μηδὲ μίαν μάχην φαίνεσθαι μεμαχημένους ; 
βλέπειν σκότον, like ἐν σκότῳ...] ὀψοίατο 
(1273), Eur. Bacch. 510 σκότιον εἰσορᾷ 
κνέφας. 

420 βοῆς δὲ κιτιλ. Of thy cry what 
haven shall there not be (z.e. to what 
place shall it not be borne),—what part 
of Cithaeron shall not be resonant with 
it (σύμφωνος ἔσται sc. αὐτῇ), re-echo it? 
If we took σύμφωνος ἔσται (and not ἔσται 
alone) with λιμήν as well as with Κιθαι- 
ρών, the figurative force of λιμήν would 
be weakened. We must not understand: 
What haven of the sea or what mountain 
(as if Cithaeron stood for ὄρος) shall not 
resound? λιμήν, poet. in the sense of 
ὑποδοχή, for that in which anything is 
received: Aesch. Pers. 250 ὦ Περσὶς ala 
καὶ μέγας πλούτου λιμήν (imitated by Eur, 
Or. 1077): the augural seat of Teiresias 
is παντὸς οἰωνοῦ λιμήν, Ant. τ : the 
place of the dead is “Acdou Nhe, 
cp. below, 1208. 

421 f. ποῖος Κιθαιρὼν, vigorous fe 
ποῖον μέρος Κιθαιρῶνος.---τὸν Deyo Lov y by, 
εἰσέπλευσας, the marriag dima which t 
didst sail: δόμοις, i in the house, ἡ 
(381): the marriage (ὑμέναιος, here=ya- 
os) was the haven into which he sailed, 









ΟἸΔΙΠΟῪΣ ΤΥΡΑΝΝΟΣ 67 


and the double lash of thy mother’s and thy father’s curse 
shall one day drive thee from this land in dreadful haste, with 
darkness then on the eyes that now see true. 

* And what place shall not be harbour to thy shriek, what of 
all Cithaeron shall not ring with it soon, when thou hast learnt 
the meaning of the nuptials in which, within that house, thou 
didst find a fatal haven, after a voyage so fair? And a throng 
of other ills thou guessest not, which shall make thee level with 


thy true self and with thine own brood. 
Therefore heap thy scorns on Creon and on my message: for 
no one among men shall ever be crushed more miserably than thou. 


OE. 
ruin take thee! 
thee from these doors! 

ΤΕ: 

OE. 


Are these taunts to be indeed borne from 4zm?— Hence, 
Hence, this instant! 


Back !—away !—avaunt 


I had never come, not I, hadst thou not called me. 
I knew not that thou wast about to speak folly, or it 


had been long ere I had sent for thee to my house. 


σχολῇ γ᾽ Suidas, and so Porson, inserting σ᾽ after ἐμούς. 





—a haven which seemed secure, but 
which, in reality, was for him a ὅρμος 
ἄνορμος.---εὐπλοίας τυχών, because Oed. 
seemed to have found ὄλβος, and also be- 
cause the gale of fortune had borne him 
swiftly on: cp. οὔθ᾽ ὁρῶν οὔθ᾽ ἱστορῶν, 
1484.—The ὑμέναιος was the song sung 
while the bride and bridegroom were 
escorted to their home, 74. 18. 492 νύμ- 
gas δ᾽ ἐκ θαλάμων δαΐδων ὑπὸ λαμπομε- 
νάων | ἠγίνεον ἀνὰ ἄστυ, πολὺς δ᾽ ὑμέναιος 
ὀρώρει, as distinguished from the ἐπι- 
θαλάμιον afterwards sung before the 
bridal chamber: Azz. 813 οὔθ᾽ ὑμεναίων] 
ἔγκληρον, οὔτ᾽ ἐπινύμφειός | πώ μέ τις 
ὕμνος ὕμνησεν. 

424 ἄλλων δὲ κιτιλ. Verses 422—425 
correspond with the actual process of the 
drama. The words καταίσθῃ τὸν ὑμέναιον 
refer to the first discovery made by Oed., 
—that his wife was the widow of one 
whom he had himself slain: cp. 821. 
The ἄλλων πλῆθος κακῶν denotes the 
further discovery that this wife was his 
mother, with all the horrors involved 
(1405). 

425 ἅ σ᾽ ἐξισώσει, which shall make 
thee level with thy (true) self,—by show- 
ing thee to be the son of Laius, not of 
Polybus ;—and level with chine own 
children, t.e. like them, the child of 
Iocasta, and thus at once ἀδελφὸς καὶ 
πατήρ (458). For ἅ σ᾽ Markland conject. 
ὅσ᾽, which shall de made equal for thee 


and for thy children: and so Porson in- 
terpreted, conjecturing doo ’ from Agathon 
fr. 5 ἀγένητα ποιεῖν doo’ ἂν ἦ “πεπραγμένα. 
Nauck ingeniously conj. ἅ σ᾽ ἐξισώσει σῷ 
τοκεῖ καὶ σοῖς τέκνοις. But the vulgate is 
sound: for the παρήχησις cp. 371. 

426 ff. τοὐμὸν στόμα: 72.2. it is 
Apollo who speaks by my mouth, which 
is not, as thou deemest, the ὑπόβλητον 
στόμα (O. C. 794) of Creon. --προπη- 
λάκιζε: acc. to Arist. Top. 6.. 6 προπη- 
λακισμός was defined as ὕβρις μετὰ χλευα- 
σίας, insult expressed by scoffing: so in 
Eth. 5. 2. 13 κακηγορία, προπηλακισμός = 
libellous language, gross abuse: and in 
Ar. ZThesm. 386 προπηλακιζομένας is ex- 
plained by πολλὰ καὶ παντοῖ᾽ ἀκουούσας 
κακά. Dem. or. 21 ὃ 72 has ἀήθεις... 
τοῦ προπηλακίζεσθαι as=‘unused to gross 
contumely’ (generally, but with imme- 
diate ref. to a Ὀ]ον).---ἐκτριβήσεται, 
rooted out. Eur. Hipp. 683 Ζεύς σε γεν- 
νήτωρ ἐμὸς | πρόρριζον ἐκτρίψειεν. 

430 οὐκ εἰς ὄλεθρον; cp. 1146: Ar. 
Plut. 394 οὐκ és κόρακας; Zr. 1183 οὐ 
θᾶσσον οἴσεις ; Cratinus Νόμοι fr. 6 
(Meineke p. 27) οὐκ ἀπερρήσεις σὺ θᾶττον ; 
Aesch. Zheb. 252 οὐκ ἐς φθόρον σιγῶσ᾽ 
ἀνασχήσει τάδε ;---πάλιν ἄψορρος, like Z7. 

53 ἄψορρον eee πάλι : the gen. οἴκων 
aye" with ἀποστραφείς. 

432 ἱκόμην... «ἐκάλεις : Cp. 125, 402. 

484 σχολῇ σ᾽ av. The simple σχολῇ 
is stronger than σχολῇ ye would be: 


orien 


435 


440 


445 


68 ZOPOKAEOYS 

TE. ἡμεῖς τοιοίδ᾽ ἔφυμεν, ὥς μὲν σοὶ δοκεῖ, 
μώροι, γονεῦσι ὃν οἵ o ἔφυσαν, ἔμφρονες. 

Ol. ποίοισι; μεῖνον. τίς δέ μ ἐκφύει βροτῶν ; 

TE. ἥδ᾽ ἡμέρα φύσει σε καὶ δ: 

OL. ws πάντ᾽ ἄγαν αἰνικτὰ Karan λέγεις. 

TE οὔκουν σὺ ταῦτ᾽ ἄριστος εὑρίσκειν. ἔφυς; 

ΟΙ. τοιαῦτ᾽ ὀνείδιζ᾽ οἷς ep εὑρήσεις μέγαν. 

TE. av7y γε μέντοι σ᾽ ἡ τύχη διώλεσεν. 

ΟἹ. DN εἰ πόλιν τήνδ᾽ ἐξέσωσ᾽, οὔ μοι μέλει. 

TE. ἄπειμι τοίνυν" καὶ σύ, παῖ, κόμιζέ με. 

ΟΙ. κομιζέτω ont: ὡς παρὼν σύ γ᾽ ἐμποδὼν 
ὀχλεῖς, συθείς oe ἂν οὐκ ἂν ἀλγύνοις πλέον. 

ΤΕ. εἰπὼν ἄπειμ᾽ ὧν οὕνεκ᾽ ἦλθον, οὐ τὸ σὸν 


δείσας πρόσωπον" οὐ γὰρ ἔσθ' ὅπου “ ὀλεῖς. 
λέγω δέ σοι τὸν ἄνδρα τοῦτον, ὃν πάλαι 


438 7δ᾽ ἡμέρα φύσει σε] τῇδ᾽ ἡμέρᾳ πεύσει ope Nauck. 
445 σύ γ᾽ ἐμποδὼν] 1, has σύγ᾽ in an erasure. 


the 7’ has been erased. 


489 dyarr’ L ist hand: 
The rst 





Ant. 390 σχολῇ ποθ᾽ ἥξειν (where σχολῇ 
7’ ἄν isan inferior v. 7) Plat. Soph. 233 B 
σχολῇ ποτ᾽. .«.«ἤθελεν ἄν, Prot. 330 Ε σχολῇ 
μέντ᾽ ἂν ἄλλο τι ὅσιον εἴη and οἥθρη. --- 
οἴκους: O. C. 643 δόμους στείχειν ἐμούς. 
-ἐστειλάμην = μετεστειλάμην, μετεπεμ- 
ψάμην. Distinguish στέλλεσθαι, to sum- 
mon 40 oneself, from στέλλειν said (1) of 
the messenger, below 860 πέμψον τινὰ 
_ στελοῦντα: (2) of him who sends word 
by a messenger, Piz/. 60 οἵ σ᾽ ἐν λιταῖς 
στείλαντες ἐξ οἴκου μολεῖν : having urged 
thee with prayers to come: Amt. 164 ὑμᾶς 
... Toptotow...  ἔστειλ᾽ ἱκέσθαι, sent you 
word to come. 

435 f. τοιοίδ᾽ refers back to the taunt 
implied in μῶρα φωνήσοντ᾽, and is then 
made explicit by μώροι.. ιἔμφρονες : cp. 
Fhil, 1271 τοιοῦτος ἦσθα (referring to 
what precedes—thou wast such as thou 
now art) τοῖς λόγοισι χὧῶτε μου τὰ τόξ᾽ 
ἔκλεπτες, πιστός, ἀτηρὸς λάθρα. In 
fr. 700 (quoted by Nauck), καὶ τὸν θεὸν 
τοιοῦτον ἐξεπίσταμαι, σοφοῖς μὲν αἰνικ- 
τῆρα,... | σκαιοῖς δὲ φαῦλον, we have not 
the preceding words, but doubtless τοιοῦ- 
tov referred to them.—as μὲν σοὶ δοκεῖ. 
col must be accented; else the contrast 
would be, not partly ‘between σοὶ and 
γονεῦσι, but solely between δοκεῖ and 
some other verbal notion. σοὶ does 


not, however, cohere so closely with δο- 
κεῖ as to form a virtua] cretic. It is need- 
less, then, to read (as Elms. proposed) ws 
μέν σοι or ws σοὶ μὲν. Cp. O. C. 1543 
ὥσπερ σφὼ πατρί: Eur. Herach 641 
σωτὴρ νῷν βλάβης. As neither σφὼ nor 
νῷν adheres to the following rather than 
to the preceding word, it seems unneces- 
sary to read with Porson ws πρὶν σφὼ or 
νῷν σωτήρ. Here we have ὡς μὲν σοὶ in- 
stead of ὡς σοὶ μὲν, because, besides the 
contrast of persons, there is also a con- 
trast between semblance (ὡς δοκεῖ) and 
fact.—yovevou, ‘for’ them, z.e. in their 
judgment: Ant. 904 καίτοι σ᾽ ἐγὼ ᾽τίμησα, 
τοῖς φρονοῦσιν, ed. Ar. Av. 445 πᾶσι 
νικᾶν τοῖς κριταῖς. 

437 ἐκφύει (i). The pres. is not histo- 
ric (for ἐξέφυσε), but denotes a permanent 
character: ‘is my sire.’ Eur. Jon 1560 
noe τίκτει σ᾽, is thy mother: so perh. 
Feracl. 208 πατὴρ δ᾽ ἐκ τῆσδὲ γεννᾶται 
σέθεν. Xen. Cyr. 8. 2. 27 ὁ δὲ μὴ νικῶν 
(he who was not victorious) τοῖς μὲν 
νικῶσιν ἐφθόνει: and so φεύγε!Ψν Ξε φυγὰς 
εἶναι passim. Shilleto thus takes οἱ ἐπα- 
γόμενοι in Thuc. 2. 2, οἱ προδιδόντες 10. 
5, οἱ διαβάλλοντες 3. 4; which, however, 
I should rather take simply as imperfect 
participles, = ol ἐπήγοντο, προὐδίδοσαν, διέ- 
βαλλον. He well compares Verg. Aen. 


OIAITOYS TYPANNOZ 69 


ES 
parents who begat thee, sane. 
What parents ? 


Such am JI,—as thou thinkest, a fool; but for the 


: Stay...and who of men is my sire? 
ΓΕ. This day shall show thy birth and shall bring thy ruin. 





What riddles, what dark words thou always speakest ! 
Nay, art not thou most skilled to unravel dark speech ? 
Make that my reproach in which thou shalt find me 


Yet twas just that fortune that undid thee. 

Nay, if I delivered this town, I care not. 

Then I will go: so do thou, boy, take me hence. 

Aye, let him take thee: while here, thou art a hin- 


drance, thou, a trouble: when thou hast vanished, thou wilt 


ΠΟΙ ΝΟΣ me “mote. 


TE 


I will go when I have done mine errand, fear- 
less of thy frown: for thou canst never destroy me. 


And 


I tell thee—the man of whom thou hast this long while 


hand seems to have written ταῦτ᾽ : an early corrector (S?) wrote yp. σύ ye in the 


margin, and altered the word in the text. 


(B) τά γ᾽. 


One later Ms. (Vat. a) has σύμ᾽ ; another 
446 ἀλγύναισ L: ἀλγύνοις Elmsley. 





9. 266 guem dat Sidonia Dido (is the 
giver): in Persius 4. 2 sorbztio tollit quem 
dira cicutae, I find rather a harsh historic 
res. 

᾿ 440 f. οὔκουν κιτ.λ. Well (οὖν--- I 
do speak riddles), art not thou most 
skilled to read them?—ro.atr ὀνείδιζέ 
(μοι), make those things my reproach, in 
which [ots, dat. of circumstance] thou 
_wilt find me great: z.e. mock my skill in 
reading riddles if thou wilt; but thou 
wilt find (on looking deeper) that it has 
brought me true honour.—ro.atra...ots, 
as O. C. 1353 (n.), Ant. 691, etc. 

442 2. αὕτη ye μέντοι. It was just 
(ye) that fortune, however (μέντοι), that 
ruined thee. γε emphasises the preceding 
word: so 778, 1292: P£zl. 93 πεμφθείς γε 
μέντοι (since I have been sen), 1052 
νικᾶν γε μέντοι: Ant. 233 τέλος γε μέντοι, 
ib. 405 μισῶ γε μέντοι.---τύχη implies 
some abatement of the king’s boast, γνώμῃ 
κυρήσαξ, 308.---ἐξέσωσ᾽, ist pers., not 3rd. 

445 Kopilérw δῆθ᾽. δῆτα in assent, 
as Aesch. Suppl. 206 Ζεὺς δὲ γεννήτωρ 
ἴδοι. ΔΑΝ. ἔδοιτο δῆτα. ---ἐμπτοδὼν with 
παρὼν, ---᾿χεβεηΐ where thy presence irks: 
cp.128. σύ γε here gives a scornful force: 
the use of ov ye in r1o1 (n.) is different. 
The reading τά γ᾽ ἐμποδὼν (found in B) 
is explained by Brunck and Erfurdt (with 
Thomas Magister) ‘thou hinderest the 


business before us,’ comparing Eur. 
Phoen, 706 ἃ δ᾽ ἐμποδὼν μάλιστα (‘most 
urgent’) ταῦθ᾽ ἥκω φράσων. 

446 ἀλγύνοις suits the continuing 
action better than ἀλγύναις. The aor. 
occurs 77. 458 (ἀλγύνειεν) and Eur. 7. A. 
326 (ἀλγῦναι) : but ats and a, as optative 
endings, are not elsewhere found in 
Soph. 

448 πρόσωπον, ‘thy face,’—thy angry 
presence: the blind man speaks as though 
he saw the ‘vultus instantis tyranni.’ 
Not, ‘thy erson’ (z.e. thy royal quality): 
πρόσωπον is not classical in this sense, 
for which cp. the Hellenistic προσωποληπ- 
τεῖν, ‘to be a respecter of persons,’ and 
the spurious Phocylidea 10 (Bergk Poet. 
Lyr. Ὁ. 361) μὴ ῥίψῃς πενίην ἀδίκως" μὴ 
κρῖνε πρόσωπον.---οὐκ ἔσθ᾽ ὅπου, there is 
no case in which...: cp. 355, 300. 

449 λέγω δέ σοι, cp. 412.---τὸν ἄνδρα 
τοῦτον.. οὗτός ἐστιν κιτ.λ. The ante- 
cedent, attracted into the case of the 
relative, is often thus prefixed to the 
relative clause, to mark with greater 
emphasis the subject of a coming state- 
ment: 7%. 283 τάσδε δ᾽ ἅσπερ εἰσο- 
pas | ...xwpotor: 7]. το. 416 φυλακὰς δ᾽ 
ἃς εἴρεαι, ἥρως, | οὔτις κἐκριμένη ῥύεται 
στρατόν : Hom. hymn. Cer. 66 κούρην τὴν 
ἔτεκον... | τῆς ἀδινὴν dm’ ἄκουσα: Ar. 
Plut.200 τὴν δύναμιν ἣν ὑμεῖς φατὲ | ἔχειν 


70 


στρ. α΄. 


ΣΟΦΟΚΛΕΟΥΣ 


ζητεῖς ἀπειλῶν κἀνακηρύσσων φόνον 

τὸν Λαΐειον, οὗτός ἐστιν ἐνθάδε, 

ἕένος λόγῳ μέτοικος, εἶτα δ᾽ ἐγγενὴς 
φανήσεται Θηβαῖος, οὐδ᾽ ἡσθήσεται 5 
τῇ ξυμφορᾷ' τυφλὸς γὰρ ἐκ δεδορκότος 
καὶ πτωχὸς ἀντὶ πλουσίον ξένην ἔ ἔπι 
σκήπτρῳ προδεικνὺς γαῖαν ἐμπορεύσεται. 
φανήσεται δὲ παισὶ τοῖς αὑτοῦ ξυνὼν 
ἀδελφὸς αὐτὸς καὶ πατήρ, κἀξ ἣ ἧς ἔφυ 
γυναικὸς υἱὸς καὶ πόσις, καὶ τοῦ πατρὸς 
ὁμόσπορός τε καὶ φονεύς. καὶ ταῦτ᾽ ἰὼν 
εἴσω λογίζου. κἂν λάβῃς ἐψευσμένον, 


φάσκειν ἔμ᾽ ἤδη μαντικῇ μηδὲν φρονεῖν. 


ν 3 ε > 
XO. tis ὄντιν᾽ a θεσπιέπεια Δελφὶς εἶπε πέτρα 


450 


455 


460 


461 λάβῃσ ἐψευσμένον L: λάβῃς pw ἐψευσμένον r, which Brunck and Hermann 
preferred. Blaydes suggests that, with λάβῃς μ᾽, ἔμ᾽ ἤδη might be changed to τότ᾽ 


ἤδη. 


Wilamowitz conj. λάβῃς ἐψευσμένα. 468 εἶπε L. The letters εἰ 


(written 7) are in an erasure, which would have been unnecessary if the word first 





με, ταύτης δεσπότης γενήσομαι. 
Trinum. 985 lllum quem ementitu’s, ts 
ego sum ipse Charmides. 


Plaut. to, ze. feeling, ψηλαφῶν, the ground 
before him: so of a boxer, χερσὶ mpo- 


δεικνύς, sparring, Theocr. 22. 102. Cp. 


450 ἀνακηρύσσων φόνον, proclaiming 
(a search into) the murder: cp. Xen. 
Mem. 2. 10. 2 σῶστρα τούτου ἀνακηρύτ- 

wy: Andoc. or. 1 ὃ 40 ¢nrnrds τε ἤδη 
ἡρημένους...καὶ μήνυτρα κεκηρνυγμένα 
ἑκατὸν μνᾶς. 

451 £. τὸν Aateov: cp. 267.---ξένος 
μέτοικος, a foreign sojourner: ξένος, be- 
cause Oed. was reputed a Corinthian. 
In poetry μέτοικος is simply one who comes 
to dwell with others: it has not the full 
technical sense which belonged to it at 
Athens, a resident alien: hence the 
addition of ξένος was necessary. Cp. 
Ὁ Ὁ. 934 μέτοικος τῆσδε γῆς: Ant. 868 
πρὸς ovs (to the dead) ἅδ᾽ ἐγὼ μέτοικοβ 
ἔρχομαι.---εἶτα δὲ opp. to νῦν μέν, im- 
plied in év@dSe.—éyyevys, ‘native,’ as 
γεννητός is opp. to ποιητός (adoptivus). 

454 τῇ Evphopa: the (seemingly 
happy) event: cp. 47. 1230 κἀπὶ συμφο- 
ραϊσὶ μοι | γεγηθὸς ἕρπει δάκρυον .---ἐἰς 
δεδορκότος : Xen. Cyr. 3. 1.17 ἐξ ἄφρονος 
σώφρων γεγένηται. 

455 f. ξένην ἔπι, sc. γῆν Ο. C. 184 
ξεῖνος ἐπὶ ξένης: Ph. 135 ἐν ξένᾳ ξένον. 
—yatav with προδεικνὺς only: pointing 


Lucian Hercules 1 τὸ τόξον ἐντεταμένον 
ἡ ἀριστερὰ προδείκνυσι, ζ.6. holds in 
front of him: id. Hermotimus 68 θαλλῷ 
προδειχθέντι ἀκολουθεῖν, ὥσπερ τὰ πρό- 
Bara. Seneca Oed. 656 repet incertus 
viae, | Baculo senili triste praetentans iter. 
The order of words is against taking ξένην 
with γαῖαν (when we should write ἐπὶ), 
and supplying τὴν ὁδόν with προδεικνύς. 

457 f. ξυνὼν: the idea of daily 
converse under the same roof heightens 
the horror. Cp. Andoc. or. 1 § 49 
οἷς... ἐχρῷ καὶ ols συνῆσθα, your friends 
and associates.—aSehpos αὑτὸς. If ἀδελ- 
gos stood alone, then αὐτὸς would be 
right: A¢tmse/f the brother of Azs own 
children: but with ἀδελφὸς καὶ πατὴρ 
we should read αὗτός αὐ once sire and 
brother of his own children, Cp. Phil. 
11g σοφός τ᾽ ἂν αὑτὸς κἀγαθὸς κεκλῃ ἅμα: 
Eur. Alc. 143 καὶ πῶς ἂν αὑτὸς κατθάνοι 
τε καὶ βλέποι; : 

460 ὁμόσποροφ: Βατε δοῖ., κετὴν αὐτὴν 
σπείρων: but passive above, 260. Acc. 
to the general rule, verbal derivatives 
with a short penult. are paroxytone when 
active in meaning (see on βουνόμοις, v. 


OIAITOY2 TYPANNOS 71 
been in quest, uttering threats, and proclaiming a search into the 
murder of Laitus—that man is here,—in seeming, an alien so- 
journer, but anon he shall be found a native Theban, and shall 
not be glad of his fortune. A blind man, he who now hath 
sight, a beggar, who now is rich, he shall make his way to a 
strange land, feeling the ground before him with his staff. And 
he shall be found at once brother and father of the children 
with whom he consorts; son and husband of the woman who 
bore him ; heir to his father’s bed, shedder of his father’s blood. 


So go thou in and think on that; and if thou find that I have 
been at fault, say thenceforth that I have no wit in prophecy. 


CHORUS. 
Who is he of whom the divine voice from the Delphian rock hath 


written had been εἶδε: it seems to have been ἥδε. 
Ist hand wrote εἶδε, which has been corrected to εἶπε. 


In one of the later mss. (I) the 
The Scholiast knew both 


readings: but it is hardly doubtful that εἶδε was a conjecture or a corruption. 





26). But those compounded with a 
preposition (or with a privativum) are 
excepted: hence διάβολος, not διαβόλος. 
So ὁμόσπορος here, no less than in 260. 
_ On the other hand πρωτοσπόρος = ‘sowing 
first,’ πρωτόσπορος =‘ first sown.’ 

461 λάβῃς ἐψ., without we: cp. Pi. 
768 (ἀλλ᾽ ἐᾶν etc.), 801 (ἔμπρησον). 

462 φάσκειν, inf. for imperat., ‘say,’ 
z.e. ‘deem,’ as in Ph. 1411, Zl. 9. Cp. 
Her. 3. 35 ἢν δὲ ἁμάρτω, φάναι Πέρσας 
τε λέγειν ἀληθέα καί με μὴ σωφρονέειν.---- 
μαντικῇ : 27 respect to seer-craft: for dat., 
cp. Eur. Z A. 338 τῷ δοκεῖν μὲν οὐχὶ 
χρήζων, τῷ δὲ βούλεσθαι θέλων. 

463—512 First στάσιμον. Teiresias 
has just denounced Oedipus. Why do 
not the Chorus at once express their 
horror? This ode is the first since v. 
215, and therefore, in accordance with 
the conception of the Chorus as per- 
sonified reflection, it must comment on 
all that has been most stirring in the 
interval. Hence it has two leading 
themes: (1) ‘ Who can be the murderer?’: 
1st strophe and antistrophe, referring to 
vv. 216—315. (2) ‘I will not believe 
that it is Oedipus’: 2nd strophe and an- 
tistrophe, referring to vv. 316—462. 

st strophe (463—472). Who is the 
murderer at whom the Delphic oracle 
hints? He should fly: Apollo and the 
Fates are upon him. 

1st antistrophe (473—482). The word 
has gone forth to search for him. Doubt- 


less he is hiding in waste places, but he 
cannot flee his doom. 

and strophe (483—497). Teiresias 
troubles me with his charge against 
Oedipus: but I know nothing that con- 
firms it. 

and antistrophe (498—512). Only gods 
are infallible; a mortal, though a seer, 
may be wrong. Oedipus has given proof 
of worth. Without proof, I will not 
believe him guilty. 

463 θεσπιέπεια, giving divine oracles 
(ἔπη), fem. as if from θεσπιεπής (not 
found): cp. ἀρτιέπεια, ἡδυέπεια. Since 
0é-om-t-s already involves the stem cer 
(Curt. Z. § 632), the termination, from fer 
(2b. 620), is pleonastic.—AeAdls πέτρα. 
The town and temple of Delphi stood in 
a recess like an amphitheatre, on a high 
platform of rock which slopes out from 
the south face of the cliff: Strabo 9. 418 
ol Δελφοί, πετρῶδες χωρίον, θεατροει- 
δές, κατὰ κορυφὴν (1.6. at the upper part 
of the rocky platform, nearest the cliff) 
ἔχον τὸ μαντεῖον καὶ τὴν πόλιν, σταδίων 
ἑκκαίδεκα κύκλον πληροῦσαν: 2.6. the 
whole sweep of the curve extends nearly 
two miles. Hom. hymn. Apoll. 1. 283 
ὕπερθεν | πέτρη ἐπικρέμαται (the rocky 
platform overhangs the Crisaean plain) 
κοίλη δ᾽ ὑποδέδρομε βῆσσα (the valley of 
the Pleistus).—elwe τελέσαντα (for εἶπε 
τελέσαι) is somewhat rare, but is not ‘a 
solecism’ (as Kennedy calls it): cp. Ο. C. 
1580 λέξας Οἰδίπουν ὀλωλότα: [Eur.] 


1st 
strophe. 


72 


¥ > > 4 / / ’ 
2appynT ἀρρήτων τελέσαντᾳ φοινίαισι χερσίν; 
9 


3 wpa νιν ἀελλάδων 
4 ἵππων σθεναρώτερον 
δ φυγᾷ πόδα νωμάᾶν. 


ZOPOKAEOYS 


465 


» A ἀρ. a Ων 3 ’ 
6 EVOTAOS Y2p €7 QvUTOV erevO pda Ket 


Ν A “A c \ 
7 πυρὶ καὶ στεροπαῖς ὁ Διὸς γενέτας" 
9 vee 


8 δειναὶ δ᾽ ἅμ᾽ ἔπονται 
9 Κῆρες ἀναπλάκητοι. 


> ΄ 
QvT. a. 


470 


¥ N a ΄, > ΄ A 
ἔλαμψε yap Tov νιφόεντος ἀρτίως φανεῖσα 


2 φάμα Παρνασοῦ, τὸν ἄδηλον ἄνδρα πάντ᾽ ἰχνεύειν. 475 


΄“ x € 3 > 4 
3 φοιτᾷ yap um ἀγρίαν 
4v\av ava τ ἄντρα καὶ 
5 πέτρας Ἰἰσόταυρος, 


466 ἀελλοπόδων MSS.; ἀελλάδων Hesychius. 


472 κῆρεσ has been made from 


χεῖρεσ in 1.,.---ἀὀναπλάκητοι L, with » written above the second a. The false reading 


ἀναμπλάκητοι is found in most (but not all) later Mss. 


In T there is a Triclinian 


note, ἀναπλάκητοι yap γράφειν (on metrical σοι 5)... εὕρηται γὰρ καὶ ἔν τινι τῶν 


παλαιοτάτων βιβλίων. 


478 L now has πέτρα σ wo ταῦροσ, with an erasure 





Rhes. 755 αὐδᾷ ξυμμάχους ὀλωλότας : Plat. 
Gorg. 481 C πότερόν σε φῶμεν νυνὶ σπου- 
δάζοντα ἢ παίζοντα; 

465 dppyt ἀρρήτων: Blaydes cp. 
O. C. 1237 πρόπαντα | κακὰ κακῶν, Phil. 
65 ἔσχατ᾽ ἐσχάτων, Aesch. Pers. 681 ὦ 
πιστὰ πιστῶν ἡλικές τ᾽ ἥβης ἐμῆς, | Πέρσαι 
γέροντες. Cp. also 1301 μείζονα τῶν μα- 
κίστων. (But Zl. 849 δειλαία δειλαίων 
[κυρεῖς], cited by Blaydes, and by Jelf 
§ 139, is not in point.) 

466 ἀελλάδων: O. C. 1081 ἀελλαία 
ταχύρρωστος πελειάς: fr. 621 ἀελλάδες 
φωναί. Not, ‘daughters of the storm,’ as 
if alluding to the mares impregnated by 
Boreas, //. 20. 221. For the form, cp. 
θυστάδας λιτάς Ant. τοι. 

467 ἵππων, instead οὗ ἵππων ποδός: 
Her. 2. 134 πυραμίδα δὲ καὶ οὗτος ἀπ- 
ελίπετο πολλὸν ἐλάσσω τοῦ πατρός: 
Xen. Cyr. 3. 3. 4ι χώραν ἔχετε οὐδὲν 
ἧττον ἔντιμον τῶν πρωτοστατῶν. 

470 στεροπαῖς. The oracular Apollo 
is Διὸς προφήτης. As punisher of the 
crime which the oracle denounced, he is 
here armed with his father’s lightnings, 
not merely with his own arrows (205).— 
yevéras, one concerned with γένος, either 
passively, = ‘son,’ as here (cp. γηγενέτᾳ 
Eur. Phoen. 128), or actively, =‘ father.’ 
Eur. has both senses. Cp. γαμβρός, son- 


in-law, brother-in-law, or father-in-law: 
and so κηδεστής or wevOepds could have 
any one of these three senses. 

472 Kies: avenging spirits, identified 
with the Furies in Aesch. Z7heb. 1055 
Kijpes Ἐρινύες, αἵ τ᾽ Οἰδιπόδα [γένος 
ὠλέσατε. Hesiod Zheog. 217 (Νὺξ) καὶ 
Μοίρας καὶ Kijpas ἐγείνατο νηλεοποί- 
vous... | αἵ τ᾽ ἀνδρῶν τε θεῶν τε παραι- 
βασίας ἐφέπουσαι οὐδέποτε λήγουσι θεαὶ 
δεινοῖο χόλοιο, [ πρίν γ᾽ ἀπὸ τῷ δώωσι 
κακὴν ὄπιν, ὅστις ἁμάρτῃ. The Μοῖραι 
decree, the ῆρες execute. In 77. 133 
κῆρες = calamities. — ἀναπλάκητοι, not 
erring or failing in pursuit: cp. 77. 120 
ἀλλά τις θεῶν αἰὲν ἀναμπλάκητον 
“Avia ope δόμων ἐρύκει, some god szf- 
Jers not Heracles to fail, but keeps him 
from death. Metre requires here the 
form without μ. ἀμπλακεῖν is prob. a 
cognate of πλάξω (from stem πλαγΎ for 
max, Curtius Etym. § 367), strength- 
ened with an inserted μ᾽; cp. ἄβροτος, 
ἄμβροτος. 

478 ἔλαμψε: see on 186.---τοῦ νιφόεν- 
τος: the message flashed forth like a 
beacon from that snow-crowned range 
which the Thebans see to the west. I 
have elsewhere noted some features of 
the view from the Dryoscephalae pass 
over Mount Cithaeron :—‘At a turn of 


ΟἸΔΊΠΟΥΣ, ΤΥΡΆΝΝΟΣ 


73 


spoken, as having wrought with red hands horrors that no 


tongue can tell? 


It is time that he ply in flight a foot stronger than the feet 
of storm-swift steeds: for the son of Zeus is springing on him, 
all armed with fiery lightnings, and with him come the dread, 


unerring Fates. 


Yea, newly given from snowy Parnassus, the message hath 


flashed forth to make all search for the unknown man. 


Into the 


wild wood’s covert, among caves and rocks he is roaming, fierce 


between a and oa, and traces of correction at wo τ. 
πετραῖοσ 6 Tavpoo: the correction is old, perh. by the first corrector (S). 


as a bull, 


The rst hand had written 
Most of the 


later Mss. have πέτρας ws ταῦρος: one or two, werpatos ws ταῦρος.---ἶ. F. Martin, and 


(later, but independently) E. 


L. Lushington, conjectured πέτρας ἰσόταυρος: M. 


Schmidt, πέτρας ἴσα ravpos: Dorville, πέτρας ἅτε ταῦρος : Campbell, πέτραισιν ἔναυ- 





the road the whole plain of Boeotia bursts 
upon the sight, stretched out far below 
us. There to the north-west soars up 
Helicon, and beyond it, Parnassus; and 
though this ts the middle of May, their 
higher cliffs are still crowned with dazzling 
snow. Just opposite, nearly due north, is 
Thebes, on a low eminence with a range 
of hills behind it, and the waters of Lake 
Copais to the north-west, gleaming in 
the afternoon sun.’ (Modern Greece, p. 
" .) 

495 Join τὸν ἄδηλον ἄνδρα, and take 
πάντα as neut. plur., ‘by all means.’ The 
adverbial πάντα is very freq. in Soph., 
esp. with adj., as Az, ο11 ὁ πάντα κωφός, 
ὁ πάντ᾽ ἄϊδρις : but also occurs with verb, 
as Zr. 338 τούτων ἔχω γὰρ πάντ᾽ ἐπι- 
στήμην ἐγώ. Here, the emphasis on 
πάντα would partly warrant us in taking 
it as acc. sing. masc., subject to lyvevew. 
But, though the masc. nominative πᾶς 
sometimes=7ds τις, it may be doubted 
whether Soph. would have thus used the 
ambiguous πάντα alone for the acc. sing. 
masc. Ellendt compares 226, but there 
πάντα is acc. plur. neut. 

478 πέτρας ἰσόταυρος is J. Ε΄. Martin’s 
and E. L. Lushington’s brilliant emenda- 
tion of πετραῖος ὁ ταῦρος, the reading of 
the first hand in L. It is at once closer 
to the letters, and more poetical, than 
πέτρας ἅτε ταῦρος (Dorville,—where the 
use of ἅτε is un-Attic), πέτρας ἴσα ταύροις 
(M. Schmidt), or πέτρας ὡς ταῦρος, which 
last looks like a prosaic correction. I 
suppose the corruption to have arisen 
thus. A transcriber who had before him 


IIETPAZIZOTATPOZ took the first O 
for the art., and then amended IIETPA- 
ΣῚΣ into the familiar word METPAIO2. 
With a minuscule Ms. this would have been 
still easier, since in πετρασισοταυροσ the 
first o might have been taken for o (not a 
rare mistake), and then a simple transpo- 
sition of « and the supposed o would have 
given metpacoo. It is true that such 
compounds with ἐσο- usu. mean, not 
merely ‘like,’ but ‘as good as’ or ‘no 
better than’: eg. ἰσοδαίμων, ἰσόθεος, 
ἰσόνεκυς, ἰσόνειρος, ἰσόπαις, ἰσόπρεσβυς. 
Here, however, ἰσόταυρος can well mean 
‘wild’ or ‘fierce of heart’ as a bull. And 
we know that in the lost Κρέουσα Soph. 
used ἰσοθάνατος in a way which seemed 
too bold to Pollux (6. 174 οὐ πάνυ dvex- 
tév),—probably in the sense of ‘dread as 
death’ (cp. Az. 215 θανάτῳ yap ἴσον πάθος 
ἐκπεύσει.. The bul] is the type of a 
savage wanderer who avoids his fellows. 
Soph. in a lost play spoke of a bull ‘that 
shuns the herd,’ Bekk. Amecd. 459. 31 ἀτι- 
μαγέλης᾽ ὁ ἀποστάτης τῆς ἀγέλης 
ταῦρος" οὕτω Σοφοκλῆς. Verg. Geo. 3. 
225 (taurus) Victus abit, longeque ignotis 
exulat oris. Theocr. 14. 43 alvds Onv 
λέγεταί tis, ἔβα καὶ ταῦρος dv’ ὕλαν" a 
proverb ἐπὶ τῶν μὴ ἀναστρεφόντων 
(schol.). The image also suggests the 
fierce despair of the wretched outlaw: 
Aesch. Cho. 275 ἀποχρημάτοισι ζημίαις 
ταυρούμενον, ‘stung to fury by the 
wrongs that keep me from my heritage’: 
Eur. Med. 92 ὄμμα ταυρουμένην: Ar. 
Ran. 804 ἔβλεψε γοῦν ταυρηδὸν ἐγκύψας 
κάτω: Plat. Phaed. 117 B ταυρηδὸν 


Ist anti- 
strophe. 


74 


A \ 
8 μαντεῖα" τὰ δ᾽ Gael 
9 ζώντα περιποτᾶται. 


ΣΟΦΟΚΛΕΟῪΣ 


6 μέλεος μελέῳ ποδὲ χηρεύων, 
1 τὰ μεσόμφαλα yas ἀπονοσφίζων ἜΤ 


480 


Sel ν Φ ὃ \ , ‘ 5 , 
στρ β΄. dewa μὲν οὖν, δεινὰ ταράσσει σοφὸς οἰωνοθέτας, 483 
¥ a ES OOD , Neri , x > a 
2 οὔτε δοκοῦντ᾽ OVT ἀποφάσκονθ'" ὅ τι λέξω δ᾽ ἀπορῶ. 485 
8 πέτομαι δ᾽ ἐλπίσιν, ovr ἐνθάδ᾽ ὁρῶν οὔτ᾽ ὀπίσω. 


4 τί γὰρ ἢ Λαβδακίδαις 


ΕἾ κι 
[οὔτε τανυν TW 


x A λύ A ¥ > ” , θέ > »¥ 3, 
57 Tw Ilo vou VELKOS EKELT , OVTE TAPOLVEV TOT Eywy 


9 


6 ἔμαθον, πρὸς ὅτου δὴ «βασανίζων» βασάνῳ 
τ t 
ἅμο 


SN ἊΝ 3 4 
= 7 ETL TAV ἐπίὸ 


ν φάτιν εἶμ᾽ Οἰδιπόδα, Λαβδακίδαις 495 


8 ἐπίκουρος ἀδήλων θανάτων. 


λος. 


483 δεινὰ μὲν οὖν] δεινά με viv Bergk: δεινά με νοῦν Nauck. 


493 There 


is a defect in the text as given by L and the other Mss., the antistrophic verse (508) 


being φανερὰ γὰρ ἐπὶ αὐτῷ πτερόεσσ᾽ ἦλθε κόρα. 


(See Metrical Analysis.) The 


alternatives are, (1) to supply ~~-— after ἔμαθον, or after ὅτου δὴ: (2) to supply 





ὑποβλέψας πρὸς τὸν ἄνθρωπον. With 
regard to the reading πετραῖος ὁ ταῦρος, 
see Appendix. 

479 xnpevov, solitary, as one who is 
ἀφρήτωρ, ἀθέμιστος, ἀνέστιος (71. 9. 63): 
he knows the doom which cuts him off 
from allhuman fellowship (236f.). Aesch. 
Eum.656 ποία δὲ χέρνιψ φρατέρων προσ- 
δέξεται; 

480 τὰ μεσόμφαλα yas μαντεῖα -- τὰ 
ἀπὸ μέσου ὀμφαλοῦ γᾶς: 2. 1386 δωμάτων 
ὑπόστεγοιτεὑπὸ στέγῃ δωμάτων: Eur. 
Phoen. 1351 λευκοπήχεις κτύπους χεροῖν. 
The ὀμφαλός in the Delphian temple 
(Aesch. Zum. 40), a large white stone in 
the form of a half globe, was held to 
mark the spot at which the eagles from 
east and west had met: hence Pindar 
calls Delphi itself μέγαν ὀμφαλὸν εὐρυκόλ- 
που | ...x9ovbs (Vem. 7. 33): Liv. 38. 48 
Delphos, umbilicum orbis terrarum.—amo- 
γοσφίζων, trying to put away (from him- 
self): the midd. (cp. 691) would be more 
usual, but poetry admits the active: 894 
ψυχᾶς ἀμύνειν: Eur. Or. 294 ἀνακάλυπτε 
κάρα: Pind. Pyth. 4. 106 κομίζων ΞΞ 
κομιζόμενος (seeking to recover): O. C. 6 
φέρονταΞε φερόμενον. In PAzl. 979 ἀπονο- 
σφίζειν τινά τινος Ξεῖο rob one of a thing: 
but here we cannot render ‘ frustrating.’ 

482 ζῶντα, ‘living,’ 2.5. operative, 
effectual ; see on 45 ζώσας.---περιποτᾶται: 
the doom pronounced by Apollo hovers 
around the murderer as the οἷστρος around 
some tormented animal: he cannot shake 


off its pursuit. The haunting thoughts of 
guilt are objectively imaged as terrible 
words ever sounding in the wanderer’s 
ears. 

483 f. The Chorus have described 
the unknown murderer as they imagine 
him—a fugitive in remote places. They 
now touch on the charge laid against 
Oedipus,—but only to say that it lacks 
all evidence. δεινὰ μὲν οὖν. οὖν marks 
the turning to a new topic, with some- 
thing of concessive force: “12 zs true that 
the murderer is said to be here’: μὲν is 
answered by δὲ after λέξω. For μὲν οὖν 
with this distributed force, cp. 0.C. 664, 
Ant. 65: for the composite μὲν οὖν 
(=‘nay rather’), below, 705.—Sewd is 
adverbial: for (1) ταράσσει could not 
mean κινεῖ, stirs up, raises, dread ques- 
tions: (2) δοκοῦντα, ἀποφάσκοντα are 
acc. sing. masc., referring to we under- 
stood. The schol., οὔτε πιστὰ οὔτε ἄπι- 
στα, has favoured the attempt to take the 
participles as acc. neut. plur., ἀποφά- 
σκοντα being explained as ‘negative’ in 
the sense of ‘admitting of negation,’ ἀπό- 
φασιν καὶ ἀπιστίαν δεχόμενα (Triclinius). 
This is fruitless torture of language. 
Nor will the conj. ἀπαρέσκοντ᾽ (Blaydes) 
serve: for, even if the Chorus found the 
charge credible, they would not find it 
pleasing. Soxovvta is not ‘believing,’ 
but ‘approving. Cp. Ant. 1102 καὶ 
ταῦτ᾽ ἐπαινεῖς καὶ δοκεῖς mapexadeiv; ‘and 
you recommend this course, and approve 


OIAITIOYS TYPANNOS 75 


wretched and forlorn on his joyless path, still seeking to put 
from him the doom spoken at Earth’s central shrine: but that 
doom ever lives, ever flits around him. 


Dreadly, in sooth, dreadly doth the wise augur move me, who 
approve not, nor am able to deny. How to speak, I know not; 
I am fluttered with forebodings; neither in the present have 1 
clear vision, nor of the future. Never in past days, nor in these, 
have I heard how the house of Labdacus or the son of Polybus 
had, either against other, any grief that I could bring as proof 
in assailing the public fame of Oedipus, and seeking to avenge 
the line of Labdacus for the undiscovered murder. 


—~~- after βασάνῳ. It may be noticed that in L the words πρὸσ ὅτου δὴ stand in a 
line by themselves, the large space left after them suggesting the loss of something 
there. See comment.—One later Ms. (Bodl. Laud. 54) has παρ᾽ ὅτου, with the gloss 


map οὗ, ἤγουν τοῦ νείκους. 





of yielding?” The pregnant force of δο- 
κοῦντα is here brought out by the direct 
contrast with ἀποφάσκοντα. In gauging 
the rarer uses of particular words by an 
artist in language so subtle and so bold as 
Soph. we must never neglect the context. 

485 f. λέξω, probably deliberative aor. 

subj.: though it might be fut. indic. (cp. 
1419, and n. on O.C. 310).—évOd8e, the 
actual situation, implies the known facts 
of the past; ὀπίσω refers to the seer’s 
hint of the future (v. 453 φανήσεται x.7.X.) : 
cp. Od. τι. 482 σεῖο δ᾽, ᾿Αχιλλεῦ, | οὔτις 
ἀνὴρ προπάροιθε μακάρτατος, οὔτ᾽ ἄρ᾽ 
ὀπίσσω (nor will be hereafter). 

487 £. ἢ Λαβδακίδαις ἡ τῷ Πολύ- 
βου. A quarrel might have originated 
with either house. This is what the dis- 
junctive statement marks: since ἔκειτο, 
‘had been made,’ implies ‘had been pro- 
voked.’ But we see the same Greek ten- 
dency as in the use of re καί where καί 
alone would be more natural: Aesch. P. 
V. 927 τό τ’ ἄρχειν καὶ τὸ δουλεύειν δίχα : 
cp. Hor. Zp. 1. 2. 12 2γι167 Hectora Pria- 
miden animosum atque inter Achillen. 

493 πρὸς ὅτου. In the antistr., 509, 
the words γὰρ ἐπ᾽ αὐτῷ are undoubtedly 
sound: here then we need to supply 

I incline to believe 
that the loss has been that of a participle 
going with βασάνῳ. Had this 5 

σανίζων, the iteration would help to ac- 
count for the loss. Reading πρὸς ὅτου 
δὴ βασανίζων βασάνῳ, I should take πρὸς 
with βασάνῳ : ‘testing o# the touchstone 
whereof ’—‘using which (νεῖκος) as a 
test.” [Receiving my βασανίζων, Kennedy 
(ed. 1885) replaces the word βασάνῳ by 


~~ ee ee or - -»»ν.- 


een βα-᾿ 


πιθανῶς.17 To Brunck’s βασάνῳ χρησά- 
μενος (Plat. Legs. 946 C βασάνοις χρώ- 
μενοι) the objections are (1) the aorist 
part. where we need the pres., (2) the 
tame and prosaic phrase. Wolff writes, 
πρὸς ὅτου δή, βασάνῳ «πίστιν ἔχων: : 
Wecklein and Mekler (in his recension 
of Dindorf’s ed., Teubner, 1885) indicate a 
lacuna, -~~-, after βασάνῳ. Two other 
courses of emendation are possible: (i) 
To supply after ἔμαθον something to ex- 
press the informant, as twos ἀστῶν or, 
προφέροντος, when πρὸς ὅτου would mean 
‘at whose suggestion.” This remedy 
seems to me improbable. (ii) Tosupply 
σύν and an adj. for βασάνῳ, as σὺν 
ἀληθεῖ B., or B. σὺν pavepg. As the 
mutilated verse stands in the MSS., it can- 
not, I think, be translated without some 
violence to Greek idiom. The most toler- 
able version would be this:—‘setting out 
from which (πρὸς ὅτου neut., referring to 
νεῖκος), I can with good warrant (βα- 
σάνῳ) assail the public fame of Oecd.’ 
Then βασάνῳ would be an instrumental 
dative equivalent to βάσανον ἔχων : and 
πρὸς ὅτου would be like 1236 πρὸς τίνος 
mor αἰτίας; Ant. 51 πρὸς αὐτοφώρων 
ἀμπλακημάτων: πρός denoting the source 
back to which the act can be traced. 

4965 ἐπὶ φάτιν εἶμι, a phrase from war: 
it is unnecessary to suppose tmesis: Her. 
1.157 στρατὸν ἐπ᾽ ἑωυτὸν ἰόντα : Eur. 7. A. 


and 
strophe. 


349 ταῦτα μέν σε πρῶτ᾽ ἐπῆλθον, ἵνα σε ἃ 


πρῶθ᾽ ηὗρον κακόν, censured thee: Andr. 
688 ταῦτ᾽ εὖ φρονῶν σ᾽ ἐπῆλθον, οὐκ ὀργῆς 
χάριν. 

497 The gen. θανάτων after ἐπίκου- 
pos is not objective, ‘against’ (as Xen. 


ΤῸΝ 


" argument: 


76 
ἄντ. β΄. 


ΣΟΦΟΚΛΕΟΥΣ 


βροτῶν 


ἀλλ᾽ ὁ μὲν οὖν Ζεὺς oT ᾿Απόλλων ξυνετοὶ καὶ τὰ 


2 εἰδότες" ἀνδρῶν δ᾽ ὅτι μάντις πλέον ἢ 7) "γὼ φέρεται, 500 
κρίσις οὐκ ἔστιν ἀληθής: σοφίᾳ δ᾽ ἂν σοφίαν 


ϑ 
4 παραμείψειεν ἁνή Ρ' 
5 


ἀλλ᾽ οὔποτ᾽ ἔγωγ᾽ ἄν, πρὶν ἴδοιμ᾽ ὀρθὸν ἔπος, μεμφομένων . 


ἂν καταφαίην. 


“Ip Ὁ 


IP: 


φανερὰ γὰρ ἐπ᾿ αὐτῷ πτερόεσσ᾽ ἦλθε κόρα 
ποτέ, καὶ “σοφὸς ὥφθη βασάνῳ θ᾽ ἀδύπολις" τῷ ἀπ᾽ ἐμᾶς 
8 φρενὸς οὐποτ᾽ ὀφλήσει κακίαν. 


512 


ἄνδρες πολῖται, δείν᾽ ἔπη πεπυσμένος 


κατηγορεῖν μου τὸν τύραννον Οἰδίπουν 


πάρειμ᾽ ἀτλητῶν. 


εἰ γὰρ ἐν ταῖς ξυμφοραῖς 


915 


ταῖς νῦν νομίζει πρός γ᾽ ἐμοῦ πεπονθέναι 


508 φανερὰ γὰρ ἐπ᾿ αὐτῷ] Hermann, thinking v. 493 (ἔμαθον x.7.d.) to be com- 


plete as it stands in the Mss., 


omitted the words γὰρ ἐπ᾽ αὐτῷ in his first ed. 
(though he afterwards replaced them); and Dindorf did likewise. 


Triclinius 


omitted ἐπ᾽ αὐτῷ, merely on the ground that he thought them unsuitable, but 





Mem. 4. 3.  πῦϑρ...ἐπίκουρον.. -«Ψύχου-), 
but causal, ‘on account of’; being soften- 
ed by the approximation of ἐπίκουρος to 
the sense of τιμωρός: Eur. 2 . 135 ἔλθοις 
τῶνδε πόνων ἐμοὶ τᾷ μελέᾳ λυτήρ, | ..-«πατρί 
θ᾽ αἱμάτων | ἐχθίστων ἐπίκουρος (=‘aven- 
ger’). The allusive plur. θανάτων is like 
αἱμάτων there, and δεσποτῶν θανάτοισι 
Aesch. Ch. 52: cp. above 366, τοῖς 
φιλτάτοις. 

498 It is true (οὖν, cp. 483) that gods 
indeed (μέν) have perfect knowledge. 
But there is no way of deciding in a strict 
sense (ἀληθής) that any mor¢a/ who essays 
to read the future attains to more than I 
do—i.e. to more than conjecture: though 
I admit that one man may excel another 
in the art of interpreting omens accord- 
ing to the general rules of augural lore 
(σοφίᾳ: cp. σοφὸς olwvobéras 484). The 
disquieted speaker clings to the negative 
‘ Teiresias is more likely to be 

right than a common man: still it is not 
certain that he is right.’ 

500 πλέον φέρεται, achieves a better 
result,— deserves to be ranked above me: 
Her. 1. 31 δοκέων πάγχυ δευτερεῖα γῶν 

αοἴσεσθαι, ‘thinking that he was sure of 
the second place at least.’ 

504 παραμείψειεν : Eur. 1. A. 145 μή 
τίς σε λάθῃ | τροχαλοῖσιν ὄχοις παραμει- 
ψαμένη ].. ἀπήνη. 


506 πρὶν ἴδοιμ, After an optative 


‘winged body of a lion: 


of wish or hypothesis in the principal 
clause, πρίν regularly takes optat.: Ph. 
961 ὄλοιο μήπω πρὶν μάθοιμ᾽ εἰ καὶ πάλιν] 
γνώμην μετοίσεις. So after ὅπως, ὅστις, 
ἵνα, εἴς. : Aesch. Zum. 297 ἔλθοι... | ὅπως 
γένοιτο: Eur. Helen. 435 τίς ἂν.. «μόλοι 
ὅστις διαγγείλειε... ;—6p0ov: the notion is 
not ‘upright,’ established, but ‘straight,’ 

—justified by proof, as by the application 
of a rule: cp. Ar. Av. 1004 ὀρθῷ μετρήσω 
κανόνι προστιθείς: so below, 853, Ant. 
1178 τοὔπος ws dp’ ὀρθὸν ἤνυσας. Hartung 
(whom Wolff follows) places the comma 
ofter ὀρθόν, not after ἔπος: ‘until I see 
(it) established, I will not approve the 
word of censurers’: but the acc. ἔπος 
could not be governed by καταφαίην in 
this sense. 

507 Kkarada(ny: Arist. AJetaphys. 3. 
6 ἀδύνατον ἅμα καταφάναι καὶ ἀποφάναι 
ἀληθῶς. Defin. Plat. 413 C ἀλήθεια ἕξις 
ἐν καταφάσει καὶ ἀποφάσει. 

508 ἐπ᾽ αὐτῷ, against him: cp. O. C. 
1472.---πττερόεσσα... κόρα: the Sphinx 
having the face of a maiden, and the 
Eur. Phoen. 
1042 ἁ πτεροῦσσα παρθένος. See Ap- 
pendix, ἢ. on v. 508. 

510 βασάνῳ with ἁδύπολις only, 
which, as a dat. of manner, it qualifies 
with nearly adverbial force : commending 
himself to the city under a practical test, 
—i.¢. ἔργῳ καὶ οὐ Abyy. Pind. Pyth. το. 


OIAITOYS TYPANNOS 77 


Nay, Zeus indeed and Apollo are keen of thought, and know 2nd anti: 
the things of earth; but that mortal seer wins knowledge above ‘tfoPhe 
mine, of this there can be no sure test; though man may surpass 
man in lore. Yet, until I see the word made good, never will I 
assent when men blame Oedipus. Before all eyes, the winged 
maiden came against him of old, and he was seen to be wise; he 
bore the test, in welcome service to our State; never, therefore, 
by the verdict of my heart shall he be adjudged guilty of crime. 


CREON. 


Fellow-citizens, having learned that Oedipus the king lays 
dire charges against me, I am here, indignant. If, in the 
present troubles, he thinks that he has suffered from me, 


retained γάρ. 510 ἡδύπολις MSS.: ἁδύπολις Erfurdt and Dindorf. 516 πρόσ 
τ᾽ ἐμοῦ L, with traces of erasure at τ᾿ and é. The rst hand had written πρόστεμοῦ (or 
possibly mpéoyeuod), joining σ, as so often, to the following letter: the corrector 
erased the 7 (or y), and wrote 7’ separately (cp. 134, 257, 294).—mpés γ᾽ ἐμοῦ τ, and 


Suidas (s.v. βάξιν).----πρός re μου Hartung. 


This was an old conjecture: τί is written 





67 πειρῶντι δὲ καὶ χρυσὸς ἐν βασάνῳ mpé- 
πει] καὶ νόος ὀρθός: ‘an upright mind, 
like gold, is shown by the touchstone, 
when one assays it’: as base metal τρίβῳ 
τε καὶ προσβολαῖς | μελαμπαγὴς πέλει | 
δικαιωθείς Aesch. Ag. 301.---ἀδύπολις, in 
the sense of ἁνδάνων τῇ πόλει (cp. Pind. 
Nem. 8. 38 ἀστοῖς ἀδών) : boldly formed 
on the analogy of compounds in which 
the adj. represents a verb governing the 
accus., as φιλόπολις Ξε φιλῶν τὴν πόλιν, 
ὀρθόπολις (epithet of a good dynasty) = 
ὀρθῶν τὴν πόλιν (Pind. Olymp. 2.7). In 
Ant. 370 ὑψίπολις is analogous, though 
not exactly similar, if it means ὑψηλὸς ἐν 
πόλει, and not ὑψηλὴν πόλιν ἔχων (like 
δικαιόπολις = δικαίας πόλεις ἔχουσα, of 
Aegina, Pind. Pyth. 8. 22). 

“= 511 τῷ, ‘therefore,’ as //. 1. 418 etc.; 
joined with wv, 7. 7. 352 etc.: Plat. 
Theaet. 179 D τῷ τοι, ὦ φίλε Θεόδωρε, 
μᾶλλον σκεπτέον ἐξ ἀρχῆς.---ἀπ᾽, on the 
part of: 77. 471 Kam’ ἐμοῦ κτήσει χάριν. 
The hiatus after τῷ is an epic trait, 
occasionally allowed in tragic lyrics, as in 
the case of interjections (cp. PA. 832 n.), 
Here the stress on τῷ, and the caesura, 
both excuse it. Cp. 4z. 194 ἀλλ᾽ ἄνα ἐξ 
ἑδράνων: El. 148 ἃ Ἴτυν: 2b. 157 ola 
Χρυσόθεμις ζώει καὶ ᾿Ιφιάνασσα (cp. /7. 9. 
145). Neither πρὸς (Elmsley) nor παρ᾽ 
(Wolff) is desirable. 

513—862 ἐπεισόδιον δεύτερον, with 
κομμός (649—697). Ocdipus upbraids 
Creon with having suborned Teiresias. 


The quarrel is allayed by Iocasta. As 
she and Oedipus converse, he is led to 
fear that he may unwittingly have slain 
Laius. It is resolved to send for the 
surviving eye-witness of the deed. 

Oedipus had directly charged Creon 
with plotting to usurp the throne (385). 
Creon’s defence serves to bring out the 
character of Oedipus by a new contrast. 
Creon is a man of somewhat rigid nature, 
and essentially matter-of-fact. In his 
reasonable indignation, he bases his ar- 
gument on a calculation of interest (583), 
insisting on the substance in contrast with 
the show of power, as in the Antigone his 
vindication of the written law ignores the 
unwritten. His blunt anger at a positive 
wrong is softened by no power of imagin- 
ing the mental condition in which it was 
done. He cannot allow for the tumult 
which the seer’s terrible charge excited 
in the mind of Oedipus, any more than 
for the conflict of duties in the mind of 
Antigone. 

515 ἀτλητῶν. The verb ἀτλητέω, 
found only here, implies an active sense 
of ἄτλητος, tmpatiens: as μεμπτός, pass. 
in O. C. 1036, is active in 77. 446. So 
from the act. sense of the verbal adj. 
come ἀλαστέω, ἀναισθητέω, ἀναισχυντέω, 
ἀνελπιστέω, ἀπρακτέω. 

516 πρός γ᾽ ἐμοῦ: 77. 738 τί δ᾽ ἐστίν, 
ὦ παῖ, πρός γ᾽ ἐμοῦ στυγούμενον; The 
conj. πρός τί μου was prompted by the 
absence of re with φέρον: but cp. Aesch. 


78 ZOO KAEOYS 


λόγοισιν εἴτ᾽ ἔργοισιν εἰς βλάβην φέρον, 
nw / 


» ’ὔ 
οὔτοι βίου μοι τοῦ μακραίωνος πόθος, 


φέροντι τήνδε βάξιν. 


5 Ν 5 ε n 
ov yap εἰς amour 


ἡ ζημία μοι τοῦ λόγον τούτου φέρει, 520 
ἀλλ᾽ ἐς μέγιστον, εἰ κακὸς μὲν ἐν πόλει, 
κακὸς δὲ πρὸς σοῦ καὶ φίλων κεκλήσομαι. 

ΧΟ. ἀλλ᾽ ἦλθε μὲν δὴ τοῦτο τοὔνειδος τάχ᾽ ἂν 
ὀργῇ Brac Gey ἄλλον ἢ γνώμῃ φρενών. 

ΚΡ. τοὔπος δ᾽ ἐφάνθη ταῖς ἐμαῖς γνώμαις ὅτι nos 
πεισθεὶς ὁ μάντις τοὺς λόγους ψευδεῖς λέγοι ; 

ΧΟ. ηὐδᾶτο μὲν τάδ᾽, οἶδα δ᾽ οὐ γνώμῃ τίνι. 

KP. ἐξ ὀμμάτων δ᾽ ὀρθῶν τε καξ ὀρθῆς φρενὸς 
κατηγορεῖτο τοὐπίκλημα τοῦτό μου; 

ΧΟ. οὐκ οἷδ᾽- ἃ γὰρ δρώσ᾽ ot κρατοῦντες οὐχ ὁρῶ. 530 


35 ὮΝ δυο νον ἵν , » a 
αὐτὸς δ᾽ ὅδ᾽ ἤδη δωμάτων ἔξω περᾷ. 
e ΄ κ αν 5 ἜΣ Σ ‘7 99 ¥ 
OI. οὗτος σύ, πῶς δεῦρ᾽ ἦλθες ; ἢ τοσόνδ᾽ ἔχεις 
τόλμης πρόσωπον ὦστε τὰς ἐμὰς στέγας 


above the line in L, and in several of the later mss. 


than cause, of the false reading πρός 7’. 


It may have been a result, rather 
517 ἔργοισί τι βλάβην φέρον Kennedy. 


525 τοῦ πρόσ δ᾽ 1,.. Of the later Mss. some (as B) have τοῦ πρὸς δ᾽ : others (as A) 
πρὸς τοῦδ᾽ (not τοῦ 6’): others (as I and 1,2) τοῦπος or τοῦπος.---τοὔπος is read by most 





Ag. 261 σὺ δ᾽ εἴτε (uv. 2. εἴ τι) κεδνὸν εἴτε 
μὴ πεπυσμένη : Plat. Soph. 237 C χαλεπὸν 
ἤρου: Meno 97 Ἑ τῶν ἐκείνου ποιημάτων 
λελυμένον μὲν ἐκτῆσθαι οὐ πολλῆς τινος 
ἄξιόν ἐστι τιμῆς. 

517 For the single etre, cp. 77. 236: 
Plat. Legg. go7 Ὁ ἐάν τις ἀσεβῇ λόγοις εἴτ᾽ 
ἔργοις : Pind. Pyth. 4. 78 ξεῖνος alr’ ὧν 
ἀστός.---φέρον: 519 φέροντι: 520 φέρει: 
such repetitions are not rare in the best 
Greek and Latin writers. Cp. 158, 159 
(duS8por’), 1276, 1278 (ὁμοῦ), Lucr. 2. 54— 
59 tenebris—tenebris—tenebris—tenebras, 
See on O. C. 554, Ant. 76. 

618 βίου τοῦ paxp.: Az. 473 τοῦ 
μακροῦ χρήζειν βίου: O. C. 1214 al 
μακραὶ | ἁμέραι, where the art. refers to 
the normal span of human life. For Blos 
μακραίων cp. 77. 791 δυσπάρευνον λέκ- 
Tpov. 

M519 εἰς ἁπλοῦν. The charge does not 
hurt him in a szmgle aspect only,—z.e. 
merely in his relation to his family and 
friends (ἰδίᾳ). It touches him also in 
relation to the State («ow ἢ), since treachery 
to his kinsman would be treason to his 
king. Hence it ‘tends to the largest 
result’ (φέρει és μέγιστον), bearing on the 


sum of his relations as man and citizen. 
The thought is, ἡ ζημία οὐχ ἁπλῇ ἐστιν 
ἀλλὰ πολυειδής (cp. Plat. Phaedr. 270 Ὁ 
ἁπλοῦν ἢ πολυειδές ἐστιν): but the proper 
antithesis to ἁπλῇ is merged in the com- 
prehensive μέγιστον. 

523 ἀλλὰ... μὲν δὴ : cp. 77. 627.— 
ἦλθε. τάχ᾽ dv, ‘might perhaps have 
come.’ ἤλθεν ἂν is a potential indicative, 
denoting for past time what ἔλθοι ἂν 
denotes for future time. That is, as 
ἔλθοι dv can mean, ‘it might come,’ so 
ἦλθεν ἂν can mean, ‘it might have come.’ 
ἦλθεν ἂν does not necessarily imply that 
the suggested possibility is contrary to 
fact; 2.6.7 it does not necessarily imply, 
ἀλλ᾽ οὐκ ἦλθεν. Cp. Dem. or. 37 ὃ 57 
πῶς dv ὁ μὴ παρὼν...ἐγώ τί σε ἠδίκησα; 
‘how was I likely to do you any wrong?’ 

[This was the view taken in my first 
edition. Goodwin, in the new ed. of his 
Moods and Tenses (1889), has illustrated 
the ‘ potential’ indicative with dy (§ 244), 
and has also shown at length that ἦλθεν 
dv does not necessarily imply the un- 
reality of the supposition (§ 412). This 
answers the objection which led me, in a 
second edition, to suggest that τάχ᾽ ἄν 


OIAITOY2 ΤΥΡΑΝΝΟΣ 79 


by word or deed, aught that tends to harm, in truth I crave not 
my full term of years, when I must bear such blame as this. 
The wrong of this rumour touches me not in one point alone, 
but has the largest scope, if I am to be called a traitor in the 


city, a traitor too = thee and by my friends. 


CH.” -Nayibu 


is taunt came under stress, perchance, of 


anger, rather than from the purpose of the heart. 


CR: 


seer to utter his falsehoods ? 


And the saying was uttered, that my counsels won the 


CH. Such things were said—I know not with what meaning. 
Cr. And was this charge laid against me with steady eyes 


and steady mind? 
CH: 


I know not; I see not what my masters do: but here 


comes our lord forth from the house. 
OEDIPUS. 


Sirrah, how camest thou here? 


Hast? thow “a.front- so 


bold that thou hast come to my house, 


of the recent edd. : see comment. 
made from τε by a later hand). 
ὀμμάτων ὀρθῶν τε. 


528 ἐξ ὀμμάτων ὀρθῶν δὲ L (the δὲ having been 
Most of the later Mss. have either this, or (as A) ἐξ 
The reading which seems preferable, ἐξ ὀμμάτων δ᾽ ὀρθῶν τε, is 





was here no more than τάχα, and that 
the usage arose from an ellipse (ἦλθε, 
τάχα δ᾽ ἂν ἔλθοι). In O. C. 964f. also I 
should now take ἦν. τάχ᾽ ἂν as=‘per- 
chance it may have been.’] 

525 I formerly kept τοῦ πρὸς δ᾽, with 
L. But the anastrophe of πρός seems to 
be confined to instances in which it is 
immediately followed by an attributive 
genitive, equiv. to an epithet: see on 178. 
For πρὸς τοῦ δ᾽ we could indeed cite 
Aesch. Zum. 593 πρὸς τοῦ δ᾽ ἐπείσθης καὶ 
τίνος βουλεύμασιν; But I now prefer τοὔ- 
mos δ᾽, because (1) Creon seems to ask 
the Chorus for a confirmation of the al- 
most incredible report that Oed. had 
brought such a charge: he would naturally 
be less concerned to know whether any 
one had uttered it defore Oed. (2) Verse 
527 favours τοὔπος.---(ρΡ. 848 ἀλλ᾽ ws 
φανέν γε Todos. 

527 ηὐδᾶτο: these things were said 
(by Oedipus); but I do not know how 
much the words meant; z.e. whether he 
spoke at random, or from information 
which had convinced his judgment. 

528 The reading ἐξ ὀμμάτων δ᾽ ὀρθῶν 
τε gives a fuller emphasis than ἐξ ὀμμά- 
τῶν ὀρθῶν δὲ: when δ᾽ had been omitted, 
τε was naturally changed to 8€ The 


place of te (as to which both verse and 
prose allowed some latitude) is warranted, 


“since ὀμμάτων-ὀρθῶν opposed to ὀρθῆς- 


φρενός forms asingle notion. é&=‘ with’: 
El. 455 ἐξ ὑπερτέρας χερός: 77. 875 ἐξ 
ἀκινήτου ποδός. ὀμμάτων ὀρθῶν: cp. 
1385: Az. 447 κεὶ μὴ τόδ᾽ ὄμμα καὶ φρένες 
διάστροφοι γνώμης ἀπῇξαν τῆς ἐμῆς : Eur. 
47. F. 931 (when the frenzy comes on 
Heracles) ὁ δ᾽ οὐκέθ᾽ αὑτὸς ἦν, | ἀλλ᾽ ἐν 
στροφαῖσιν ὀμμάτων ἐφθαρμένος, κ.τ.λΔ. 
In Hor. Carm.1. 3.18 Bentley gave sectis 
oculis for stccts. 

530 οὐκ οἶδ᾽, Creon has asked: ‘Did 
any trace of madness show itself in the 
bearing or in the speech of Oedipus?’ 
The Chorus reply: ‘Our part is only to 
hear, not to criticise.’ These nobles~of 
Thebes (1223) have no eyes for indiscre 
tion in their sovereign master. 

582 2. Join οὗτος σύ: cp. 1121: 
Eur. Hee. 1280 οὗτος σύ, μαίνει καὶ κακῷ 
ἐρᾷς τυχεῖν; where οὗτος, σὺ μαίνει is if. 
possible.—réApmys, gen. of quality (or 
material); cp. Ant. 114 χιόνος πτέρυγι: 
El. 19 ἄστρων εὐφρόνη.---τοσόνδε τόλ- 
μης-πρόσωπον, like τοὐμὸν φρενῶν-ὄνειρον 
(ZZ. 1390), γνεῖκος-ἀνδρῶν᾽ ξύναιμον (Ant. 
793): 


80 ΣΟΦΟΚΛΕΟῪΣ 


ἵκου, φονεὺς ὧν τοῦδε τἀνδρὸς ἐμφανῶς 


λῃστής Τὶ ἐναργὴς τῆς ἐμῆς τυραννίδος ; : 


599 


φέρ᾽ εἰπὲ πρὸς θεῶν, δειλίαν ἢ μωρίαν 
ἰδών τιν ἔν μοι ταῦτ᾽ ἐβουλεύσω ποεῖν ; 
ἢ τοὔργον ὡς οὐ γνωριοῖμί σον τόδε 


j δόλῳ προσέρπον 


“ἢ οὐκ ἀλεξοίμην μαθών ; 
ἄρ᾽ οὐχὶ μώρόν ἐστι τοὐγχείρημαά σου, 


540 


ἄνευ τε πλήθους καὶ φίλων τυραννίδα 
θηράν, ὃ πλήθει χρήμασίν P ἁλίσκεται ; 


ΚΡ. 


οἶσθ᾽ ὡς πάησον; 


ἀντὶ τῶν εἰρημένων 


io ἀντάκουσον, κάτα Kpw αὐτὸς μαθών. 


ΟΙ. 


λέγειν σὺ δεινός, “μανθάνειν δ᾽ ἐγὼ κακὸς 


545 


σοῦ" δυσμενῆ γὰρ καὶ βαρύν σ᾽ ηὕρηκ᾽ ἐμοί. 


KY. 


given by Suidas and a few later ss. (Ρ, Δ, Trin.). 
γνωριοῖμι Elmsley. 
‘The conjecture πλούτου, first made by an anony- 


Reisig. 
κοὐκ MSS. 


538 γνωρίσοιμι MSS.: 
541 πλήθους MSS. 


TOUT αὐτὸ νῦν μου πρῶτ᾽ ἄκουσον ὡς ἐρώ. 


597 ἐν ἐμοὶ MSS.: ἔν μοι 
539 7 οὐκ A. Spengel:’ 





535 τῆς ἐμῆς closely follows τοῦδε 
τἀνδρός, as O. C. 1329: so Az. 865 μυθή- 
σομαι immediately follows Alas θροεῖ. If 
a Greek speaker rhetorically refers to 
himself in the third person, he usu. reverts 
as soon as possible to the first. 

537 ἔν μοι. The MSS. have ἐν ἐμοί, 
making a verse like 77. 4, ἐγὼ | δὲ τὸν 
-éuldv, kal πρὶν εἰς Acdov μολεῖν. But such 
a verse is rare, and unpleasing. When a 
tribrach holds the second place in a tragic 
senarius, we usually find that (a) the tri- 
brach is a single word, as PA. 1314 ἥσθην] 
πατέρα | Tov ἀμὸν εὐλογοῦντά σε: or (4) 
there is a caesura between the first and 
the second foot, as Ο. C. 26 ἀλλ᾽ ὅσ᾽τις ὁ 
τόπ]ος: Ph. 1232 παρ᾽ οὗπερ ἔλαβον : Eur. 
Tro. 496 τρυχηρὶὰ περὶ] τρυχηρὸν εἱμένην 
χρόα: Eur. Phoen. 511 ἐλθόντ]α σὺν ὅπλοις 
τόνδε καὶ πορθοῦντα γῆν,---ἰ there we 
should not read ἐλθόντ᾽ ἐν ὅπλοις. On 
such a point as ἐμοὶ versus μοι the au- 
thority of our MSs. is not weighty. And 
the enclitic μου suffices: for in this verse 
the stress is on the verbal notion (ἰδών), -- 
Creon’s supposed zzszght: the reference 
to Oedipus is drawn out in the next two 
verses by the verbs in the rst person, -yrw- 
ριοῖμι----ἀλεξοίμην .---ἰδών..«ἐν : prose would 
say geod either with or without ἐν 
(Thuc. 1. 95: ὅπερ καὶ ἐν τῷ Παυσανίᾳ 
ἐνεῖδον : is 30 δ... τοῖς πολεμίοις ἐνορῶνὴ : 
cp. Her. 1. 37 οὔτε τινὰ δειλίην παριδών 


μοι (remarked in me) οὔτε ἀθυμίην. 

ποεῖν ; Attic inscrr. of ¢. 450—300 B.C. 
omit the « before ε or ἡ (not before o or w), 
as L usu. does, when the 1st syll. is short: 
Ph. 120 n. 

538 ἢ τοὔργον κ-τ.λ. Supply νομίσας 
or the like from ἰδών : ‘thinking that 
either I would’ not see,...o7 would not 
ward it off’: an example of what Greek 
rhetoric called χιασμός (from the form of 
X), since the first clause corresponds 
with μωρία, and the second with δειλία. 
--οἮνωριοῖμι. ‘Futures in -low are not 
common in the good Attic period: but 
we have no trustworthy collections on 
this point’: Curtius, Verd 11. 312, Eng. 
tr. 481. On the other hand, as he says, 
more than 20 futures in -.@ can be quoted 
from Attic literature. And though some 
ancient grammarians call the form 
‘ Attic,’ it is not exclusively so: instances 
occur both in Homer (as //. το. 331 ἀγλα- 
ϊεῖσθαι, cp. Monro, Hom. Gram. § 63) 
and in Herodotus (as 8. 68 ἀτρεμιεῖν. be- 
sides about ten other examples in Her.). 
Thus the evidence for γνωριοῖμι outweighs 
the preference of our Mss. for γνωρίσοιμι. 

539 ἢ οὐκ. The κοὐκ of the Mss. can- 
not be defended here—where stress is 
laid on the dilemma of δειλία or μωρία---- 
by instances of ἤ...τε carelessly put for 
%—7 in cases where there is no such 
sharp distinction of alternatives: as //. 2. 


OIAITOYS TYPANNOZ 81 


who art the proved assassin of its master,—the palpable robber 
of my crown? Come, tell me, in the name of the gods, was it 
cowardice or folly that thou sawest in me, that thou didst plot 
to do this thing? Didst thou think that I would not note this 
deed of thine creeping on me by stealth, or, aware, would not 
ward it off? Now is not thine attempt foolish,—to seek, with- 
out followers or friends, a throne,—a prize which followers and 


wealth must win? 


CR. 


Mark me now,—in answer to thy words, hear a fair 


reply, and then judge for thyself on knowledge. 

Or. Thou art apt in speech, but I have a poor wit for thy 
lessons, since I have found thee my malignant foe. 

Cr. Now first hear how I will explain this very thing— 


mous German translator of the play in 1803, has been adopted by Nauck and others. 


546 ηὕρηκ᾽] εὕρηκ᾽ L. See comment. 


Cpritost. 





289 ἢ παῖδες veapol χῆραί τε γυναῖκες: 
Aesch. Lum. 524 ἢ πόλις βροτός θ᾽ 
ὁμοίως.---ἀλεξοίμην : see on 171. 

541 πλήθους refers to the rank and 
file of the aspirant’s following,—his popu- 
lar partisans or the troops in his pay; t- 
λων, to his powerful connections,—the 
men whose wealth and influence support 
him. Thus (542) χρήμασιν is substituted 
for φίλων. Soph. is thinking of the his- 
torical Greek τύραννος, who commonly 
began his career as a clemagogue, or else 
‘arose out of the bosom of the oligarchies’ 
(Grote, vol. 3 p. 25). 

542 ὃ, a thing which, marking the 
general category in which the τυραννίς is 
to be placed: cp. Xen. Mem. 3. 9. 8 φθό- 
νον δὲ σκοπῶν 6 τι εἴη. So the neut. adj. 
is used, Eur. App. 109 τερπνὸν... | τρά- 
meta πλήρης: Eur. Hel. 1687 γνώμης, ὃ 
πολλαῖς ἐν γυναιξὶν οὐκ ἔνι. 

548 οἶσθ᾽ ὡς πόησον; In more than 
twelve places of the tragic or comic poets 
we have this or a like form where a per- 
son is eagerly bespeaking attention to a 
command or request. Instead of οἷσθ᾽ ὡς 
δεῖ σε ποιῆσαι; or οἷσθ᾽ ὥς ce κελεύω ποιῆ- 
σαι; the anxious haste of the speaker 
substitutes an abrupt imperative: οἷσθ᾽ ὡς 
ποίησον; That the imperative was here 
felt as equivalent to ‘you are to do,’ ap- 
pears clearly from the substitutes which 
sometimes replace it. Thus we find (1) 
fut. indic.; Eur. Cycl. 131 οἷσθ᾽ οὖν ὃ 
δράσεις ; Med. 600 οἷσθ᾽ ws μετεύξει καὶ 
σοφωτέρα φανεῖ; where the conjectures 
δρᾶσον (Canter) and μέτευξαι (Elmsley) 


5 ie Bi 


are arbitrary: so with the rst pers., 7. 7. 
759 ἀλλ᾽ οἷσθ᾽ ὃ δράσω; (2) a periphrasis: 
Eur. Suppl. 932 ἀλλ᾽ οἷσθ᾽ ὃ δρᾶν σε Bov- 
λομαι τούτων πέρι; Only a sense that. 
the imperat. had this force could explain: 
the still bolder form of the phrase with 
3rd pers.: Eur. 7. 7. 1203 οἷσθα vu & 
μοι γενέσθωΞεἃ δεῖ γενέσθαι μοι: Ar. Ach. 
1064 οἷσθ᾽ ὡς ποιείτω τ- ὡς δεῖ ποιεῖν αὐτήν, 
where ποιεῖτε is a conjecture. There is 
no reason, in logic or in grammar, against 
this ‘subordinate imperative,’ which the 
flexible Greek idiom allowed. Few 
would now be satisfied with the old 
theory that οἷσθ᾽ ὡς ποίησον stood, by 
transposition, for ποίησον, οἶσθ᾽ ws; 

5451. For κακὸς with inf, cp. Thue. 
6. 38 § 2 ἡμεῖς δὲ κακοὶ...προφυλάξασθαι. 

σοῦ, emphatic by place and pause: cp. 
El. 1505 χρῆν δ᾽ εὐθὺς εἶναι τήνδε τοῖς πᾶ- 
σιν δίκην [ ὅστις πέρα πράσσειν γε τῶν νό- 
μων θέλει, | κτείνειν" τὸ γὰρ πανοῦργον 
οὐκ ἂν ἦν πολύ.---Σηὕρηκ᾽ : as to the aug- 
ment, cp. 68 n. 

547 £, τοῦτ᾽ αὐτὸ κιτ.λ. Oedipus flings 
back Creon’s phrases, as the Antigone of 
Aeschylus bitterly echoes those of the 
κῆρυξ (αὐδῶ — αὐδῶ — τραχύς ---- τράχυν᾽, 
Theb. 1042 f.). An accent of rising 
passion is similarly given to the dialogue 
between Menelaus and Teucer (Az. 1142 
ἤδη ποτ᾽ εἶδον ἄνδρ᾽ éyw—1150 ἐγὼ δέ γ᾽ 
ἄνδρ᾽ ὄπωπα). Aristophanes parodies this 
style, Ach. 1097 AAMAXOZ. παῖ, παῖ, 
φέρ᾽ ἔξω δεῦρο τὸν γύλιον ἐμοί. ΔΙΚΑΙΟ- 
ΠΟΛΙΣ. παῖ, παῖ, φέρ' ἔξω δεῦρο τὴν 
κίστην ἐμοί.----ὡς ἐρῶ, how I will state this 


6. 


92 ΣΟΦΟΚΛΕΟΥΣ 


Ol. τοῦτ᾽ αὐτὸ μή μοι. φραξζ, ὅπως οὐκ εἶ κακός. 
KP. εἴ τοι νομίζεις κτῆμα τὴν αὐθαδίαν 
εἶναί τι τοῦ νοῦ χωρίς, οὐκ ὀρθώς φρονεῖς. 550 
OI. εἴ τοι νομίζεις ἄνδρα συγγενῆ κακῶς 
δρῶν οὐχ ὑφέξειν. τὴν δίκην, οὐκ εὖ φρονεῖς. 
ΚΡ. ξύμφημί σοι ταῦτ᾽ ἔνδικ᾽ εἰρῆσθαι. τὸ δὲ 
πάθημ᾽ ὁποῖον φὴς παθεῖν δίδασκέ με. 
ΟΙ. ἔπειθες, ἢ οὐκ ἔπειθες, ὡς χρείη μ᾽ ἐπὶ 555 
τὸν σεμνόμαντιν ἄνδρα πέμψασθαί τινα; 
KP. καὶ νῦν ἔθ᾽ αὐτός εἰμι τῷ βουλεύματι. 
Ol. πόσον tw ἤδη nf ὁ Λάϊος χρόνον 
ΚΡ. d€dpake ποῖον ἔργον ; οὐ γὰρ ἐννοῶ. 
ΟΙ. dgavros ἔρρει θανασίμῳ χειρώματι ; 560 
KP. μακροὶ παλαιοί T ἂν μετρηθεῖεν χρόνοι. 
OI. τὸτ᾽ οὖν» ὁ «μάντις οὗτος ἣν ἐν τῇ τέχνῃ; 
KP: σοφός γ᾽ ὁμοίως κἀξ ὦ ἰσου τιμώμενος. 
ΟἿ. ἐμνήσατ' οὖν ἐμοῦ τι τῷ τότ᾽ ἐν χρόνῳ; 
KP. οὔκουν ἐμοῦ γ᾽ ἑστῶτος οὐδαμοῦ πέλας. 565 
OI. ἀλλ᾽ οὐκ ἔρευναν τοῦ θανόντος € ἔσχετε; 
KP. Taper x oper, πῶς δ᾽ οὐχί; κοὐκ ἠκούσαμεν. 
Ol. πῶς οὖν τόθ᾽ οὗτος 0 σοφὸς οὐκ ηὔδα τάδε; 
KP. οὐκ οἷδ᾽. ἐφ᾽ οἷς γὰρ μὴ φρονῶ σιγᾶν ΕΠ 


555 χρείη Dawes. 
and the” 


L has xpe?’ 7, but the accentuation is due to the first corrector, 
over 7 has been re-touched by a later hand. The 1st hand may have in- 


tended Xen OF χρείη, though the space between εἰ and ἡ is rather unduly wide. 
xpet’ ἢ is in almost all the later Mss. (χρεῖ ἦν T'; χρείμ᾽ Bodl. Barocc. 66, with a 





very matter (my supposed hostility to 
you): 2.6. in what a light I will place 
it, by showing that I had no motive 
for it. 

549 f. κτήμα: cp. Ant, 1050 ὅσῳ 
κράτιστον κτημάτων εὐβουλία. - αὐθαδίαν, 
poet. for αὐθάδειαν (Aesch. P. V. 70, 
εἰς.).--τοῦ νοῦ χωρίς; for αὐθάδεια is 
not necessarily devoid of intelligence: as 
Heracles says (Eur. 27. /. 1243) αὔθαδες 
ὁ θεός" πρὸς δὲ τοὺς θεοὺς ἔγώ. 

555 ἢ οὐκ: Aesch. 7heb. 100 ἀκούετ᾽ ἢ 
οὐκ ἀκούετ᾽ ἀσπίδων κτύπον ; Od. 4. 682 ἢ 
εἰπέμεναι δμωῇσιν ᾿Οδυσσῆος θείοιο. Such 
‘synizesis’ points to the rapidity and ease 
of ancient Greek pronunciation: see J. 
H. H. Schmidt, ἀγέλης und Metrik 
§ 3 (p. 9 of Eng. tr. by Prof. J. W. 
White). 


556 While such words as ἀριστόμαντις, 
ὀρθόμαντις are seriously used in a good 
sense, σεμνόμαντις refers ironically to a 
solemn manner: cp. σεμνολογεῖν, σεμνο- 
προσωπεῖν, σεμνοπανοῦργος, σεμνοπαρά- 
GtTO$, Etc. 

557 αὑτός: ‘I am the same man in 
regard to my opinion’ (dat. of respect): 
not, ‘am identical with my former 
opinion’ (when the dat. would be like 
Φοίβῳ in 285). Thuc. can dispense with 
a dative, 2. 61 καὶ ἐγὼ μὲν ὁ αὐτός εἰμι 
καὶ οὐκ ἐξίσταμαι : though he adds it in 3. 
38 ἐγὼ μὲν οὖν ὁ αὐτός εἰμι TH Ὑν ὦ μῃ. 

559 δέδρακε. Creon has heard only 
what Oedipus said of him: he does 
not yet know what Teiresias said of 
Oedipus (cp. 574). Hence he is startled 
at the mention of Laius.—ovd yap ἐννοῶ : 


OIAITOYS ΤὙΥὙΒΆΝΝΟΣ 83 


OE. 
CR, 


good gift, thou art not wise. 
OE. 


Explain me not one thing—that thou art not false. 
If thou deemest that stubbornness without sense is a 


If thou deemest that thou canst wrong a kinsman 


and escape the penalty, thou art not sane. 


GR: 


Justly said, I grant thee: but tell me what is the 


wrong that thou sayest thou hast suffered from me. 


OE. 
send for that reverend seer ? 
CR 


Didst thou advise, or didst thou not, that I should 


And now I am still of the same mind. 


And how was it that this sage did not tell his story 


I know not; where I lack light, ’tis my wont to be silent. 


561 ἀναμετρηθεῖεν A, a reading-which no other Ms. 
Cp. 1348, where av γνῶναι has been changed to ἀναγνῶναι in all 


OE. How long is it, then, since Latus— 
Cre csince Palussa? a take not thy-dritts.. 
OE. —was swept from men’s sight by a deadly violence? 
Cr. The count of years would run far into the past. 
Or. Was this seer, then, of the craft in those days? 
Cr. Yea, skilled as now, and in equal honour. 
Or. Made he, then, any mention of me at that time? 
Cr. Never, certainly, when I was within hearing. 
Or. But held ye nota search touching the murder? 
Cr. Due search we held, of course—and learned nothing. 
OE. 
then ? 
CR. 
superscript). Cp. v. 791. 
seems to have. 
the MSS. 566 θανόντος] κτανόντος Meineke: θενόντος M. Schmidt. 


567 κοὐκ 


ἠκούσαμεν] κοὐκ ἰχνεύσαμεν Mekler: κοὐδὲν ἤνομεν Nauck. 





2.6. ‘I do not understand what Laius has 
to do with this matter.’ 

560 χειρώματι, deed of a (violent) 
hand: Aesch. Zhed. 1022 τυμβόχοα χειρώ- 
para=service of the hands in raising a 
mound. In the one other place where 
Aesch. has the word, it means ‘ prey’ 
(Ag. 1326 δούλης θανούσης εὐμαροῦς xet- 
ρώματοΞ) : Soph. uses it only here (though 
he has dvoxelpwua Ant. 126): Eur. 
never. 

561 μακροὶ «.7.\.: long and ancient 
times would be measured; 2.4. the reckon- 
ing of years from the present time would 
go far back. into the past; μακροὶ de- 
noting the course, and παλαιοί the point 
to which it is retraced. Some sixteen 
years may be supposed to have elapsed 
‘since the death of Laius. 

562 ἐν τῇ τέχνῃ: slightly con- 
temptuous. ἐν of a pursuit or calling: 
Her. 2. 82 τῶν Ἑλλήνων of ἐν ποιήσει 


γενόμενοι: Thuc. 3. 28 οἱ ἐν τοῖς mpay- 
μασι: Isocr. or. 2 ὃ 18 of ἐν ταῖς ὀλιγαρ- 
χίαις καὶ ταῖς δημοκρατίαις (meaning, the 
administrators thereof): Plat. Phaed. 
59 A ws ἐν φιλοσοφίᾳ ἡμῶν ὄντων: Legg. 
762 A τῶν ἐν ταῖς yewpylas: Protag. 
317 C (Protagoras of himself as a σοφισ- 
τής) πολλά γε ἔτη ἤδη εἰμὶ ἐν τῇ τέχνῃ. 

565 οὐδαμοῦ with ἑστῶτος πέλας, 
‘when I was standing anywhere near’; 
but equivalent in force to, ‘on any oc- 
casion when I was standing near’: cp. 
Ai. 1281 ὃν οὐδαμοῦ φὴς οὐδὲ συμβῆναι 
ποδί. 

567 παρέσχομεν, we held it, asin duty 
bound: παρέχειν, as distinct from 
ἔχειν, expressing that it was something 
to be expected on their part. Cp. Ο. C. 
1498 δικαίαν χάριν παρασχεῖν παθών. 
For παρέσχομεν after ἔσχομεν cp. 133 
ἐπαξίως... ἀξίως : 575 μαθεῖν... : 576 ἐκ- 
μάνθαν᾽. 


ΟΞ 


84 ΣΟΦΟΚΛΕΟῪΣ 


ΟΙ. τοσόνδε γ οἶσθα καὶ λέγοις ἂν εὖ φρονών. 570 
ΚΡ. ποῖον τόδ᾽ ; εἰ γὰρ οἶδά γ᾽, οὐκ ἀρνήσομαι. 
ΟΙ. ὁθούνεκ᾽, εἰ μὴ σοὶ ξυνῆλθε, τὰς ἐμὰς 
οὐκ av ποτ᾽ εἶπε Λαΐου διαφθοράς. 
KP. εἰ μὲν λέγει τάδ᾽, αὐτὸς οἶσθ᾽" ἐγὼ δὲ σοῦ 
μαθεῖν δικαιῶ ταὐθ᾽ ἅπερ κἀμοῦ σὺ νῦν. 575 
OI. ἐκμάνθαν᾽. οὐ γὰρ δὴ φονεὺς “ἁλώσομαι. 
KP: τί Ont ἀδελφὴν Ty ἐμὴν γήμας ἔχεις ; 
ΟΙ. ἄρνησις οὐκ ἔνεστιν ὧν ἀνιστορεῖς. 
ΚΡ ἄρχεις δ᾽ ἐκείνῃ ταὐτὰ γῆς, ἴσον νέμων ; 
ΟΙ. av ἡ θέλουσα πάντ᾽ ἐμοῦ κομίζεται. 580 
KP. οὔκουν ἰσοῦμαι σφῷν ἐγὼ δυοῖν τρίτος ; 
Ol. ἐνταῦθα “γὰρ δὴ καὶ κακὸς φαίνει φίλος. 
KP. οὔκ, εἰ διδοίης γ᾽ ὡς ἐγὼ σαυτῷ λόγον. 
σκέψαι δὲ τοῦτο πρῶτον, εἴ TW ἂν δοκεῖς 
ἄρχειν ἑλέσθαι ξὺν φόβοισι μᾶλλον ἢ 585 
ἄτρεστον εὕδοντ᾽, εἰ τά ᾽ αὖθ᾽ ἕξει κράτη. 
ἐγὼ μὲν οὖν οὔτ᾽ αὐτὸς ἱμείρων ἔφυν 
τύραννος εἶναι μάλλον ἢ τύραννα δρᾶν, 
οὔτ᾽ ἄλλος. ὅστις σωφρονεῖν. ἐπίσταται. 
νῦν μὲν γὰρ ἐκ σοῦ πάντ᾽ ἄνευ φόβου dépw, 590 


570 τοσόνδε Ὑ] τὸ σὸν δέ 1, ist hand: 
indicate the reading τοσόνδε. 


the corrector changed σὸν to σόν, as if to 
τοσόνδε is in a few of the later Mss. (as B, with gl. 


τοσοῦτον): τὸ σὸν δέ in A and others.—ré σὸν δέ γ᾽ is read by Brunck, and others: 


τοσόνδε γ᾽ by Porson (Eur. Med. 461), Elmsley, and others. 


The reading τόσον δέ γ᾽, 


already known to Triclinius, and also suggested by Reisig, is preferred by Wunder 





570 τοσόνδε ¥. If we read τὸ σὸν 

ἐγ, the coarse and blunt τὸ σὸν would 
destroy the edge of the sarcasm. Nor 
would τὸ σὸν consist so well with the 
calm tone of Creon’s inquiry in 571. 
τοσόνδε does not need δέ after it, since 
οἶσθα is a mocking echo of olda. Cp. 
Eur. 7. 7. 554 OP. παῦσαί νυν ἤδη, μηδ᾽ 
ἐρωτήσῃς πέρα. ID. τοσόνδε γ᾽, εἰ ζῇ τοῦ 
ταλαιπώρου δάμαρ. Against the conject. 
τόσον δέ γ᾽ it is to be noted that Soph. 
has τόσος only in 42. 185 (lyric, τόσ- 
gov), 277 (δὶς ré0’), and 77. 53 φράσαι 
TO Gov. 

572 ‘The simple answer would have 
been :—‘ that you prompted him to make 
his present charge’: but this becomes :— 
‘that, if you had not prompted him, he 
would never have made it.’ ξυνῆλθε: 
Ar. Eg. 1300 φασὶν ἀλλήλαις συνελθεῖν 


τὰς τριήρεις ἐς λόγον, ‘the triremes laid 
their heads together’: 26. 467 ἰδίᾳ δ᾽ ἐκεῖ 
τοῖς Λακεδαιμονίοις ξυγγίγνεται. ---τὰς ἐμὰς : 
the conject. τάσδ᾽ ἐμὰς mars the passage: 
‘he would never have described this slay- 
ing of L. as mine.’—ovk dy εἶπε τὰς ἐμὰς 
«Λαΐου διαφθοράς--οὐκ ἂν εἶπεν ὅτι ἐγὼ 
Λάϊον διέφθειρα, but with a certain bitter 
force added;—‘we should never have 
heard a word of this slaying of Laius by 
me.’ Soph. has purposely chosen a turn 
of phrase which the audience can re- 
cognise as suiting the fact that Oed. had 
slain Laius. For διαφθοράς instead of a 
clause with διαφθείρειν, cp. Thue. 1. 137 
γράψας τὴν ἐκ Σαλαμῖνος mpodyyekow τῆς 
ἀναχωρήσεως καὶ τὴν τῶν γεφυρῶν... οὐ" 
διάλυσιν. ᾿ 
574 2. To write σοῦ instead of cov 
is not indeed necessary; but we thus ob- 


ὌΠ ΟΥΣ aie NINOS 


ὃς 


ΟΝ. Thus much, at least, thou knowest, and couldst de- 


clare with light enough. 
Cr. What is that? 
OE. 


If I know it, I will not deny. 
That, if he had not conferred with thee, he would 


never have named my slaying of Laius. 


CR. 


If so he speaks, thou best knowest; but I claim to 


learn from thee as much as thou hast now from me. 


I shall never be found guilty of the 


wouldst reason with thine own 


Opn eam thy fill: 
blood. 
Cr. Say, then—thou hast married my sister ? 
ΟΕ. The question allows not of denial. 
Cr. And thou rulest the land as she doth, with like sway ? 
Ok. She obtains from me all her desire. 
Cr. And rank not I as a third peer of you twain? 
OE. Aye, tis just therein that thou art seen a false friend. 
Cr. Not so, if thou 
heart as I with mine. 


And first weigh this,—whether thou 


thinkest that any one would choose to rule amid terrors 
rather than in unruffled peace,—granting that he is to have 
the same powers. Now I, for one, have no yearning in 
my nature to be a king rather than to do kingly deeds, 
no, nor hath any man who knows how to keep a sober 
mind. For now I win all boons from thee without fear; 


and others. 572 τὰς MSS.: τάσδ᾽ Doderlein. 575 ταῦθ᾽ MSS.: ταῦθ᾽ Brunck. 
679 Wecklein writes τῆς τιμῆς instead of γῆς ἴσον : Heimsoeth conjectures τοῦ 
κράτους for ταὐτὰ γῆς: F. W. Schmidt, ἀρχῆς δ᾽ ἐκείνῃ ταῦτ᾽ ἔχεις ἴσον νέμων. 
583 ἐγὼ] ἔχω is Heimsoeth’s conjecture, who might point to v. 1061, where ἐγὼ is 





tain a better balance to kapod.—pabety 
ταὔθ᾽, to question in like manner and 
measure. ταῦθ᾽ (MSS.) might refer to the 
events since the death of Laius, but has 
less point. 

576 ov γὰρ δηὴ rejects an alternative: 
here, without ye, as Azz. 46: more often 
with it, as O. C. rro (n.). 

577 γήμας ἔχεις: simply, I think, 
Ξε γεγάμηκας, though the special use of 
ἔχειν (Od. 4. 569 ἔχεις Ἑλένην καί opw 
γαμβρὸς Διός ἐσσι) might warrant the 
version, ‘hast married, and hast to wife.’ 

579 γῆς with ἄρχεις: ἴσον νέμων ex- 
plains tavrd,—‘with equal sway’ (cp. 
201 κράτη νέμων, and 237): γῆς ἴσον 
νέμων would mean, ‘assigning an equal 
share of land.’ The special sense of vé- 
pov is sufficiently indicated by the con- 
text; cp. Pind. P. 3. 70 ὃς Συρακόσσαισι 
νέμει βασιλεύς (rules at S.). 

580 f. ἢ θέλουσα : cp. 126, 274, 747. 
—tplros: marking the completion of the 


Incky snumiber;-as- Ὁ Ὁ 8... νς anya, 
Aesch. Lumen. 759 (τρίτου | Zwrijpos): 
Menander Sent. 231 θάλασσα καὶ πῦρ καὶ 
γυνὴ τρίτον κακόν. 

For the gen. ἐμοῦ, cp. 1163 (του). 

582 ἐνταῦθα yap: (yes indeed:) for 
otherwise your guilt would be less glaring ; 
it is just this fact that deprives it of excuse. 

583 διδοίης λόγον : Her. 3. 25 λόγον 

ἑωυτῷ δοὺς ὅτι... ἔμελλε x.7.rA. fon re- 
Jflecting that,’ εἴς. : [Dem.] or. 45 § 7 (the 
speech prob. belongs to the time of 
Dem.) λόγον δ᾽ ἐμαυτῷ διδοὺς εὑρίσκω 
κιτιλ. Distinguish the 22:7. in Plato’s 
ποικίλῃ ποικίλους ψυχῇ...διδοὺς λόγους, 
applying speeches (Phaedr. 277 C). 

587 ovr αὐτὸς would have been 
naturally followed by οὔτ᾽ ἄλλῳ παραι- 
νοῖμ᾽ ἄν, but the form of the sentence 
changes to οὔτ᾽ ἄλλος (ἱμείρει). 

590 ἐκ σοῦ: ἐκ is here a correct sub- 
stitute for παρά, since the king is the 
ultimate source of benefits: Xen. Hedlen. 


86 ZLTOPOKAEOY2 


εἰ δ᾽ αὐτὸς ἦρχον, πολλὰ κἂν ἄκων ἔδρων. 
TOS Ont ἐμοὶ τυραννὶς ἡδίων ἔχειν 

ἀρχῆς ἀλύπου καὶ δυναστείας ἔφυ ; 

οὕπω τοσοῦτον ἡπατημένος. κυρῶ 


ὥστ᾽ ἄλλα χρἤζειν ἢ τὰ σὺν κέρδει καλά. 


905 


νῦν πᾶσι χαίρω, νῦν με πᾶς ἀσπάζεται, 
vuv οἱ σέθεν χρήζοντες ἐκκαλοῦσί με' 

τὸ γὰρ τυχεῖν αὐτοῖσι πᾶν ἐνταῦθ᾽ ev. 
πῶς δῆτ᾽ ἐγὼ κεῖν ἂν ha Bou’ ἀφεὶς τάδε; 


οὐκ ἂν γένοιτο νοῦς κακὸς kahos φρονῶν. 


600 


ἀλλ᾽ οὔτ᾽ ἐραστὴς τῆσδε τῆς “γνώμης ἔφυν 
οὔτ᾽ ἂν pet ἄλλου δρώντος ἃ ἂν τς ποτέ. 
καὶ τῶνδ᾽ ἔλεγχον τοῦτο μὲν Πυθώδ᾽ i 

πεύθου τὰ χρησθέντ᾽, εἰ σαφῶς ον: σοι" 


right, and the Mss. give ἔχω. 
above. 


597 ἐκκαλοῦσι L, with a gloss προκαλοῦσιν written 
There is no trace of a variant in the later Mss., for in Εἰ καλοῦσι is a mere 


blunder, and the rapa written in the margin of L and A was meant to explain ἐκ, not 


to suggest av. ὦ. παρακαλοῦσι. 


That ἐκκαλοῦσι was rightly understood, appears from 


such glosses as μεσίτην} ποιοῦσι (B), εἰς βοήθειαν μεσοῦντα (Ε)).---οαἰκάλλουσι Musgrave. 
598 τὸ γὰρ τυχεῖν αὐτοῖσ ἅπαν ἐνταῦθ᾽ ἔνι 1.. The accent on αὐτοῖσ has been either 
made οὐ re-touched by the first corrector (5); Diibner and Campbell think that the 





3. 1. 6 ἐκείνῳ δ᾽ αὕτη ἡ χώρα δῶρον ἐκ 
βασιλέως ἐδόθη.--- φέρω -Ξ- φέρομαι, as 1100; 
Ὁ: Ὁ: Ὁ εἴς. 

591 κἀν ἄκων: he would do much of 
his own good pleasure, but much also 
(kal) against it, under pressure of public 
duty. 

594 f. οὔπω, ironical: see on 105.— 
τὰ σὺν κέρδει καλά: honours which bring 
substantial advantage (real power and 
personal comfort), as opp. to honours in 
which outward splendour is joined to 
heavier care. £/. 61 δοκῶ μέν, οὐδὲν ῥῆμα 
σὺν κέρδει κακόν : 1.6. the sound matters 
not, if there i is κέρδος, solid good. 

596 πᾶσι χαίρω, ‘all men wish me 
joy’: lit. ‘I rejoice with the consent of 
all men’: all are content that I should 
rejoice. ‘Cp. O. C. 1446 ἀνάξιαι γὰρ 
πᾶσίν ἐστε δυστυχεῖν, all deem you unde- 
serving of misfortune: Ar. Av. 445 πᾶσι 
νικᾶν Tots κριταῖς | καὶ τοῖς θεαταῖς πᾶσι. 
The phrase has been suggested by χαῖρέ 
pot, but refers to the meaning rather than 
to the form of the greeting: 2.6. πᾶσι 
χαίρω is not to be regarded as if it meant 
literally, ‘I have the word χαῖρε said to 
me by all.’ This is one of the boldly 


subtle phrases in which the art of Soph. 
recalls that of Vergil. Others under- 
stand: (1) ‘I rejoice in all,’—instead of 
suspecting some, as the τύραννος does, who 
φθονέει... τοῖσι ἀρίστοισι... χαίρει δὲ τοῖσι 
κακίστοισι τῶν ἀστῶν Her. 3. 80: (2) “1 
rejoice in relation to all’—de. am on 
good terms with all: (3) ‘I rejoice in the 
sight of all’: ἐ.6. enjoy a happiness which 
is the greater because men see it: (4) ‘I 
rejoice in all things.” This last is im- 
possible. Of the others, (1) is best, but 
not in accord with the supposed position 
of Oedipus 6 πᾶσι κλεινός. 

597 ἐκκαλοῦσι. Those who have a 
boon to ask of Oed. come to the palace 
(or to Creon’s own house, see on 637) 
and send in a message, praying Creon to 
speak with them. Seneca’s Creon says 
(Oed. 687) Solutus onere regio, regnt bonis 
Fruor, domusque civium coetu viget. In 
Greek tragedy the king or some great 
person is often thus called forth. Cp. 
Aesch. Cho. 663: Orestes summons an 
οἰκέτης by knocking at the épxela πύλη, 
and, describing himself as a messenger, 
says—¢Eed dere τις δωμάτων τελεσφόρος | 
γυνὴ T6mapxos,—when Clytaemnestra her- 


ΟΙΔΙΠΟῪΣ ΤΥΡΑΝΝΟΣ 87 


but, were I ruler myself, I-should be doing much e’en against 
mine own pleasure. 

How, then, could royalty be sweeter for me to have than 
painless rule and influence? Not yet am I so misguided as to 
desire other honours than those which profit. Now, all wish 
me joy; now, every man has a greeting for me; now, those who 
have a suit to thee crave speech with me, since therein is all 
their hope of success. Then why should I resign these things, 
and take those? No mind will become false, while it is wise. 
Nay, I am no lover of such policy, and, if another put it into 
deed, never could I bear to act with him. 

And, in proof of this, first, go to Pytho, and ask if I brought 

thee true word of the oracle; 


rst hand wrote αὐτοὺσ. This is possible, but seems hardly certain. They also find 
traces of τ, written by an early hand after ἅπαν, but now erased. Of the later Mss., 
a few have ἅπαν, the majority (as A) ἅπαντ᾽, but two (Τ' and 1,2) the probably true 
reading, πᾶν .---πάντ᾽ is read by Bothe and Burges.—Wecklein brackets the verse as 
spurious. 602 δρῶντος] δρῶν τάσ᾽ Bellermann; δρῶν τόδ᾽ Forster. 604 πεύθου 
L, the letters πεν in an erasure; the rst hand perh. wrote ἐπύθου, as Diibner thinks. 
πεύθου prevails in the later Mss., but Τ' has πύθου, and Pal. πυθοῦ. Nauck prefers 





self appears. So in Eur. Bacch. 170 
Teiresias says—ris ἐν πύλαισι Κάδμον 
ἐκκαλεῖ δόμων; ‘where is there a servant 
at the doors to call forth Cadmus from 
the house ?’—Irw τις, εἰσάγγελλε Τειρεσίας 
ὅτι | ζητεῖ vw: then Cadmus comes forth. 
The active ἐκκαλεῖν is properly said (as 
there) of him who takes in the message, 
the middle ἐκκαλεῖσθαι of him who sends 
it in (Her. 8. 19): but in PA. 1264 éxxa- 
Meio Ge (n.) is anexception. The Lat. evo- 
care=éxxandeloOa in Cic. De Orat. 2. 86. 
Musgrave’s αἰκάλλουσι is not a word 
which a man could complacently use to de- 
scribe the treatment of himself by others. 
αἴκαλος." κόλαξ Hesych. (for ἀκ-ίαλος, 
from the same rt., with the notion of sooth- 
ing or stilling, as ἀκεῖσθαι, ἧκα, ἀκέων, 
ἄκασκα, ἁκασκαῖος): Ar. ἔφ. 47 ὑποπεσὼν 
τὸν δεσπότην | ἤκαλλ᾽, ἐθωπευ᾽, ἐκολάκευ᾽, 
‘fawned, wheedled, flattered’: in tragedy 
only once, Eur. Andr. 630 φίλημ᾽ ἐδέξω, 
προδότιν αἰκάλλων κύνα. 

598 τὸ.. τυχεῖν sc. ὧν χρήζουσιν. The 
reading ἅπαντ᾽, whether taken as accus, 
after τυχεῖν (‘to gain all things’), or as 
accus. of respect (‘to succeed in all’) not 
only mars the rhythm but enfeebles the 
sense. When αὐτοῖσι was corrupted into 
αὐτοῖς, πᾶν was changed into ἅπαν, as it 
isin L. évravOa=év τῷ ἐκκαλεῖν με, in 
gaining my ear: cp. O. C. 585 ἐνταῦθα 
γάρ μοι κεῖνα συγκομίζεται, in ¢hzs boon I 


find ¢hose comprised. 

599 πῶς Shr. Cp. Her. 5. 106 
(Histiaeus to Dareius) βασιλεῦ, κοῖον ἐφ- 
θέγξαο ἔπος ; ἐμὲ βουλεῦσαι πρῆγμα ἐκ τοῦ 
σοί τι ἢ μέγα ἢ σμικρὸν ἔμελλε λυπηρὸν 
ἀνασχήσειν; τί δ᾽ ἂν ἐπιδιξζήμενος ποιέοιμι 
ταῦτα; τεῦ δὲ ἐνδεὴς ἐών, τῷ πάρα μὲν 
πάντα ὅσαπερ σοί, πάντων δὲ πρὸς σέο 
βουλευμάτων ἐπακούειν ἀξιεῦμαι; 

600 οὐκ ἀν γένοιτο κιτ.λ. Creon has 
been arguing that Ae has no motive for 
treason. He nowstates a general maxim, 
‘No mind would ever tur to treason, 
while it was sound.’ As a logical in- 
ference, this holds good only of those 
who are in Creon’s fortunate case. If, 
on the other hand, καλῶς φρονῶν means 
‘alive to its own highest good,’ and not 
merely to such self-interest as that of 
which Creon has spoken, then the state- 
ment has no strict connection with what 
precedes: it becomes a new argument of 
a different order, which might be illus- 
trated from Plato’s κακὸς ἑκὼν οὐδείς. It 
would be forcing the words to render: 
‘A base mind could not approve itself 
wise,’ z.e. ‘such treason as you ascribe to 
me would be silly.’ 

603 ἔλεγχον, accus. in apposition 
with the sentence: Eur. 27. 7. 57 ἡ δυσ- 
πραξία | ἧς μήποθ᾽, ὅστις καὶ μέσως εὔνους 
ἐμοί, | τύχοι, φίλων ἔλεγχον ἀψευδέστα- 


TOV. 


88 ΣΟΦΟΚΛΕΘῪΣ 


τοῦτ᾽ ἀλλ᾽, 


ἐάν με τῷ τερασκόπῳ λάβῃς 


605 


κοινῇ τι βουλεύσαντα, μή μ ἁπλῇ ᾿κτάνῃς 
ψήφῳ, διπλῇ δέ, τῇ T ἐμῇ καὶ σῇ, λαβών. 


γνώμῃ 
οὐ γὰρ 


ἀδήλῳ μή με χωρὶς αἰτιώ. 
ἴκαιον οὔτε τοὺς κακοὺς μάτην 
χρηστοὺς νομίζειν οὔτε τοὺς χρηστοὺς κακούς. 


610 


φίλον, γὰρ ἐσθλὸν ἐκβαλεῖν. ἴσον λέγω 

καὶ τὸν παρ᾽ αὐτῷ βίοτον, ὃν πλεῖστον φιλεῖ. 
ἀλλ᾽ ἐν χρόνῳ γνώσει τάδ᾽ ἀσφαλῶς, ἐπεὶ 
χρόνος δίκαιον ἄνδρα δείκνυσιν μόνος, 


κακὸν δὲ κἂν ἐν ἡμέρᾳ γνοίης μιᾷ, 
,. καλῶς ἔλεξεν εὐλαβουμένῳ πεσεῖν, 


615 


ἀναξ' φρονεῖν γὰρ οἱ ταχεῖς οὐκ ἀσφαλεῖς. 


ὅταν ταχύς τις οὐπιβουλεύων λάθρᾳ 


χωρῇ, ταχὺν δεῖ κἀμὲ βουλεύειν πάλιν. 


εἰ δ᾽ ἡσυχάζων προσμενώ, τὰ τοῦδε μὲν 
πεπραγμέν᾽ ἔσται, τἀμὰ δ᾽ ἡμαρτημένα. 
ἢ με γῆς ἔξω βαλεῖν; 


ἥκιστα" θνήσκειν, οὐ , φυγεῖν σε βούλομαι, 


τί δῆτα χρήζεις; ἡ 


620 


“ὡς ἂν προδείξῃς οἷόν ἐστι τὸ φθονεῖν. 


ὡς 4 % 


KP. ov 


πυθοῦ, as Dindorf did in Poet. Scen. 


conject. γνώμης δὲ δήλου. 


ὡς οὐχ ὑπείξων οὐδὲ πιστεύσων λέγεις ; 


γὰρ φρονοῦντά σ᾽ εὖ ἀλέα 
ἀλλ᾽ ἐξ ἴσου δεῖ κἀμόν. 


ed. 5 (1869). 
623 θνήισκειν L. 


625 


a2 


ΩΣ 


OI. τὸ γοῦν ἐμόν. 
ΟΙ. ἀλλ᾽ ἔφυς κακός. 


608 Bellermann 
See comment. on 118. 





605 τοῦτ᾽ ἀλλο--τοῦτο δέ. Soph. has 
τοῦτο μέν irregularly followed by τοῦτ᾽ 
αὖθις (Ant. 165), by εἴτα (PA. 1345), by 
δέ (Az. 670, O. C. 440).—T@ τερασκόπῳ. 
This title (given to Apollo, Aesch. Zum. 
62) has sometimes a shade of scorn, as 
when it is applied by the mocking 
Pentheus to Teiresias (Eur. Bacch. 248), 
and by Clytaemnestra to Cassandra 
(Aesch. Ag. 1440). 

608 χωρὶς, ‘apart’: z.e. solely on the 
strength of your own guess (γνώμη ἄδη- 
dos), without any evidence that I falsified 
the oracle or plotted with the seer. 

612 τὸν παρ᾽ αὑτῷ βίοτον κ.τ.λ. : the 
life is hospes comesque corporis, dearest 
guest and closest companion : cp. Plat. 
Gorg. 479 B μὴ ὑγιεῖ ψυχῇ συνοικεῖν: 
and the address of Archilochus to his 


own θυμός as his trusty ally (Bergk fr. 
66),—Oupé, θύμ᾽ ἀμηχάνοισι κήδεσιν κυκώ- 
μενε, | ἐνάδευ, δυσμενῶν δ᾽ ἀλέξευ προσβα- 
λὼν ἐναντίον | στέρνον .---φιλεῖ sc. τις, Sup- 
plied from αὑτῷ: Hes. Op. 12 τὴν μέν 
κεν ἐπαινήσειε νοήσας | ἡ δ᾽ ἐπιμωμητή. 
614 f. χρόνος: cp. Pind. fr. 132 
ἀνδρῶν δικαίων χρόνος σωτὴρ ἄριστος: 
Olymp. τι. 53 ὅ 7’ ἐξελέγχων μόνος | ἀλά- 
θειαν ἐτήτυμον | χρόνος.---κακὸν δὲ: the 
sterling worth of the upright man is not 
fully appreciated until it has been long 
tried: but a knave is likely (by come 
slip) to afford an early glimpse of his real 
character. The Greek love of antithesis 
has prompted this addition, which is 
relevant to Creon’s point only as imply- 
ing, ‘if I Aad been a traitor, you would 
probably have seen some symptom of it 


ΘΙ ΤΟΥΣ. ΤΥΡΑΝΝΟΣ 89 


then next, if thou find that I have planned aught in concert 
with the soothsayer, take and slay me, by the sentence not of 
one mouth, but of twain—by mine own, no less than thine. 
But make me not guilty in a corner, on unproved surmise. It 
is not right to adjudge bad men good at random, or good men 
bad. I count it a like thing for a man to cast off a true friend 
as to cast away the life in his own bosom, which most he loves. 
Nay, thou wilt learn these things with sureness in time, for time 
alone shows a just man; but thou couldst discern a knave even 


in one day. 


Cu. Well hath he spoken, O king, for one who giveth heed 
not to fall: the quick in counsel are not sure. 
Or. When the stealthy plotter is moving on me in quick 


sort, I, too, must be quick with my counterplot. 


If I await him 


in repose, his ends will have been gained, and mirfe missed. 


Cr. What wouldst thou, then? 


Cast me out of the land? 


Or. Not so: I desire thy death—not thy banishment— 
that thou mayest show forth what manner of thing is envy. 


Cr. Thou speakest as resolved not to yield or to believe ? 
[Oz. No; for thou persuadest me not that thou art worthy of belief.] 


Cr. No, for I find thee not sane. 


mine own interest. 


Cr. Nay, thou shouldst be so in mine also. 


thou art false. 


OE. 


Sane, at least, in 


On Nav: 


624 f. ws ἂν is my conjecture for ὅταν. The Mss. give v. 624 to Creon, and v. 625 





ere now.’ Cp. Pind. Pyth. 2. go (speak- 
ing of the φθονεροί): στάθμας δέ τινος 
ἑλκόμενοι | περισσᾶς ἐνέπαξαν ἕλκος ὀδυνα- 
ρὸν ἑᾷ πρόσθε καρδίᾳ, | πρὶν ὅσα φροντίδι 
μητίονται τυχεῖν. Ant. 493 φιλεῖ δ᾽ ὁ 
θυμὸς πρόσθεν ἡρῆσθαι κλοπεὺς | τῶν μηδὲν 
ὀρθῶς ἐν σκότῳ τεχνωμένων. 

617 The infin. φρονεῖν is like an 
accus. of respect (¢.g. βουλήν) construed 
with both adjectives: ‘in counsel, the 
quick are not sure.’ Cp. Thuc. 1. 70 ἐπι- 
νοῆσαι ὀξεῖς. 

618 ταχύς τις χωρῇ, advances in 
quick fashion; nearly=raxéws πως. AZ. 
1266 φεῦ, τοῦ θανόντος ws Taxed τις 
βροτοῖς | χάρις διαρρεῖ, 7% what quick sort 
does it vanish. 

622-- 626 τί δῆτα χρήζεις ;...τὸ γοῦν 
ἐμόν. (1) Verse 624, ὅταν προδείξῃς x.7.X., 
which the Mss. give to Creon, belongs to 
Oedipus: and for ὅταν we should (I 
think) read ὡς dv. The argument that 
the stichomuthia should not be broken 
shows inattention to the practice of Soph. 
He not seldom breaks a stichomuthia, 


when a weighty utterance (as here, the 
king’s threat) claims the emphasis of two 
verses. See (¢.g.) 356—369, broken by 
366 f. (the seer’s denunciation): Azz. 
40—48, broken by 45 f. (Antigone’s re- 
solve): O. C. 579—606, broken by 583 f. 
(where Theseus marks the singularity in 
the proposal of Oed.). (2) Verse 625 ws 
οὐχ ὑπείξων x.T.A., Which the MSS. give to 
Oedipus, belongs to Creon. (3) Between 
625 and 626 a verse spoken by Oedipus 
has dropped out, to such effect as οὐ 
yap με πείθεις οὕνεκ᾽ οὐκ ἄπιστος el. 
The fact of the next verse, our 626, also 
beginning with οὐ γὰρ may have led to 
the loss by causing the copyist’s eye to 
wander. The echoed οὐ γὰρ would suit 
angry dialogue: cp. 547, 548 KP. τοῦτ᾽ 
αὐτὸ νῦν μου πρῶτ᾽ ἄκουσον ws ἐρῶ. OL. 
τοῦτ᾽ αὐτὸ μή μοι φράζ᾽. (See also on 
Ph. τ252.) The traditional interpretations 
fail to justify (1) οἷόν ἐστι τὸ φθονεῖν, as 
said by Creon: (2) πιστεύσων, as said by 
Oed. See Appendix. 


Γ᾿ 


ee" 


go ZOPOKAEOYS 


KP, εἰ δὲ ξυνίης μηδέν: Ol. 

ΚΡ, οὗτοι. κακῶς γ᾽ ἄρχοντος. 

ΚΡ. καμοὶ πόλεως μέτεστιν, οὐχὶ σοὶ μόνῳ. 
XO. παύσασθ᾽, ἄνακτες: 


ἀρκτέον γ᾽ ὅμως. 
ΟΙ. ὦ πόλις πόλις. 
630 


καιρίαν δ᾽ ὑμῖν ὁρῶ 


τήνδ᾽ ἐκ δ 4: στείχουσαν ᾿Ιοκάστην, μεθ᾽ ἧς 
τὸ νῦν παρεστὸς νεῖκος εὖ θέσθαι χρεών. 


IOKASTH. 


, \ » ἊΣ / , 
τί τὴν ἄβουλον, ὦ ταλαίπωροι, στάσιν 


᾿ γλώσσης ἐπήρασθ᾽ ; 


οὐδ᾽ ἐπαισχύνεσθε, γῆς 
οὕτω νοσούσης, ἴδια κινοῦντες κακά; 

οὐκ εἶ σύ T οἴκους σύ τε, Κρέον, κατὰ στέγας, 
καὶ μὴ τὸ μηδὲν ἄλγος εἰς μέγ᾽ οἴσετε; 


635 


640 


KP. ὅμαιμε, δεινά μ᾽ Οἰδίπους ὁ σὸς πόσις 
δυοῖν δικαιοῖ “δρᾶν ἀποκρίνας κακοῖν, 
ἴω ἴω “ἡ fa) Α 
ἢ γῆς ἀπῶσαι πατρίδος, ἢ κτεῖναι λαβών. 
to Oedipus. After v. 625 a verse seems to be lost. 


629 apxovros L, made 


from ἄρχοντεσ either by the first hand or by the first corrector (S).—apxovras 


Musgrave. 
which the second was εἰ 
καιρίαν. 


631 καιρίαν] κυρίαν L, the uv in an erasure of two letters, of 
in the margin, yp. καιρίαν. 
684 τὴν] Déoderlein conj. τήν δ᾽, 


Most of the later mss. have 
635 The rst hand in L wrote 


ἐπήρασθ᾽, but an early corrector changed this to érjpar’, as most of the later Mss. 





628 ἀρκτέον -- δεῖ ἄρχειν, one must 
rule: cp. Amt. 677 duré ἐστὶ τοῖς 
κοσμουμένοις. Isocr. or. 14 § 10 οὐ τῶν 
ἄλλων αὐτοῖς ἀρκτέον (they ought not to 
rule over others) ἀλλὰ πολὺ μᾶλλον ’Opxo- 
μενίοις φόρον olaréov. In Plat. 7272. 48 B 
ἀρκτέον = det ἄρχεσθαι, one must begin; 
in Ai. 853 ἀρκτέον τὸ mpayua=must be 
begun, Some understand—‘ one must be 
ruled,’ and οὔτοι κακῶς γ᾽ ἄρχοντος, ‘No, 
not by one who rules ill’: but (a) though 
ἀρκτέα πόλις might mean, ‘the city is to be 
ruled,’ an absolute passive use of ἀρκτέον 
is certainly not warranted by such an 
isolated example as οὐ καταπληκτέον 
ἐστίν (‘we must not be unnerved’) in 
Dein. Jn Dem. § 108: (ὁ) ἄρχομαί τινος, 
‘I am ruled by one’ (instead of ἐκ or 
ὑπό), could only plead the analogy of 
ἀκούω τινός, and lacks evidence. 

629 ἄρχοντος, when one rules. ἀρκ- 
τέον being abstract, ‘it is right to rule,’ 
there is no harshness in the gen. absol. 
with τινός understood (cp. 612), which is 
equivalent to ἐάν τις ἄρχῃ: cp. Dem. or. 
6 ὃ 20 λέγοντος ἄν Twos πιστεῦσαι οἴεσθε; 
‘think you that, if any one had said it, 


they would have believed ?’ = οἴεσθε, εἴ τις 
ἔλεγε, πιστεῦσαι av (αὐτούς) ;---ὦ πόλις 
πόλις : here, an appeal: in Attic comedy, 
an exclamation like o /empora, 0 mores: 
Blaydes cp. Eupolis af..Athen. 424 B ὦ 
πόλις, πόλις | ws εὐτυχὴς ef μᾶλλον ἢ 
καλῶς φρονεῖς : and so Ar. Ach, 27. 

630 πόλεως. Most of the Mss. have 
μέτεστι τῆσδ᾽ οὐχί. Had they μέτεστι 
τῆσδ᾽ οὐ (which appears only in a few in- 


- ferior Mss.) we should hardly be war- 


ranted in ejecting τῆσδ᾽ : but, having the 
choice, we may safely prefer μέτεστιν. 
οὐχὶ to μέτεστι τῆσδ᾽ οὐ. ‘I have some 
right in Thebes, as well as you.’ Creon 
speaks not as a brother of Iocasta, but as a 
Theban citizen who denies that ‘the city 
belongs to one man’ (Ant. 737). Plat. 
Legg. 768 B δεῖ δὲ δὴ καὶ τῶν ἰδίων δικῶν κοι- 
νωνεῖν κατὰ δύναμιν ἅπαν τα ς" ὁ γὰρ ἀκοινώ- 
νητος ὧν ἐξουσίας τοῦ συνδικάξειν ἡγεῖται τὸ 
παράπαν τῆς πόλεως ᾿ μέτοχος εἶναι. 
637 οὐκ εἶ... καὶ «οἴσετε; cp. AZ. 
75 n.—olkovs (the eile ’s palace), acc. 
after εἶ (cp. 5 ae κατὰ with στέγας only, 
referring to the house of Creon, who is 
not supposed to be an inmate of the 


OIAITTOYS ΤΥΡΑΝΝΟΣ ΟΙ 


CR. 
rule. 
Cr. Not 16 thou rule il 
CR: 
(Η. ‘Cease, princes ; 


But if thou understandest nought? 


OE. 
Thebes is for me also—not for thee alone. 
and in good time for you I see Iocasta 


Or. Yet must I 


Hear him, O Thebes! 


coming yonder from the house, with whose help ye should com- 


pose your present feud. 


IOCASTA. 
Misguided men, why have ye raised such foolish strife of 


tongues? 
stir up troubles of your own? 


Are ye not ashamed, while the land is thus sick, to 
Come, go thou into the house, 


—and thou, Creon, to thy home,—and forbear to make much 


of a petty grief. 


Cr. Kinswoman, Oedipus thy lord claims to do dread 
things unto me, even one or other of two ills,—to thrust me 
from the land of my fathers, or to slay me amain. 


read, though one or two (as V, V4) have ἐπήρασθ᾽. 


ov τ᾽ and οἴκουσ. 
nearly all the later Mss. 
by correction from κρέων : 


637 L has an erasure between 


The rst hand seems to have intended σύ τ᾽ ἐσ οἴκουσ.---κρέων L, and 
In 1459 L again has κρέων as voc., but in Amt. 211 Kpéov 
but E has Κρέον, and so Elmsley. 


δικαιοῖ δυοῖν ἀποκρίνας κακοῖν Μϑ85.---δυοῖν. 


640 δρᾶσαι 


. δρᾶν is my conjecture: see comment. 





palace: see 515, 533. 

638 τὸ μηδὲν Agee: the generic use 
of μή (‘a grief such as to be naught,’— 
‘ quod nihili 522), here giving a causal 
force (‘seeing that it is naught’): cp. 397, 
1019; Zl. 1166 δέξαι... | THY μηδὲν és 
τὸ μηδέν : εἰς μέγα φέρειν, make into a 
great matter: cp. (2.11. 259) νόσος | ἀεὶ 
τέθηλε κἀπὶ μεῖζον ἔρχεται. 

640 διυοῖν.. ἀποκρίνας κακοῖν. The 
traditional reading, δρᾶσαι... δυοῖν, is, the 
only extant example of δνοῖν scanned as 
one syllable, though in the tragic poets 
alone the word occurs more than 50 
times. Synizesis of uv is rare in extant 


Greek poetry: Pind. Pyth. 4. 225 γενύων: 
Anthol. 11. 413 (epigram by Ammianus, 
Ist century A.D.) ὥκιμον, ἡδύοσμον, πήγα- 
γον, ἀσπάραγος. Eur. 7. 7. 970 ὅσαι δ᾽ 
Ἐρινύων οὐκ ἐπείσθησαν νόμῳ, and 2d. 1456 
οἴστροις ᾿Ερινύων, where most editors 
write Ἐρινῦν, as 2b. 299’ Ἐρινῦς (acc. plur.). 
Hes. Scut. 3 Ἠλεκτρύωνος. It might be 
rash to say that Soph. could not have 
used δυοῖν as a monosyllable; for he has 
used the ordinary synizesis in a peculiarly 
bold way, Az. 1129 μή νυν ἀτίμα. θεοὺς 
θεοῖς σεσωμένος : but at least it moves the 
strongest suspicion. 


ἀποκρίνας, on the other hand, seems 
genuine. ἀποκρίνειν is properly secernere, 
to set apart: e.g. γῆν (Plat. Rep. 303 Ὁ): 
or to select: id. Leggy. 946 A πλήθει τῶν 
ψήφων ἀποκρίναντας, having selected (the 
men) according to the number of votes 
for each. Here, ‘having set apart (for 
me) one of two ills’ is a phrase suitable 
to the arbitrary rigour of doom which 
left a choice only between death and 
exile. 

For δυοῖν Elms. proposed τοῖνδ᾽ or 
τοῖνδέ γ᾽ : Herm., τοῖνδ᾽ ἕν : A. Spengel, 
δείν᾽, 1 should rather believe that δρᾶν 
was altered into δρᾶσαι by a grammarian 
who looked to ἀπῶσαι, κτεῖναι, and 
perh. also sought a simpler order. But 
for pres. infin. combined with aor. infin. 
cp. 623 OvynoKketv...pvyetv: Ant. 204 
μήτε κτερίζειν μήτε kwkioat See 
also O. C. 732 ἥκω γὰρ οὐχ ὡς δρᾶν τι 
βουληθείς, where in prose we should have 
expected δρᾶσαι. The quantity of ἀπο- 
κρίνας is supported by Aesch. P. V. 24 
ἀποκρύψει: ἀποτροπή and its cognates in 
Aesch. and Eur. : ἐπικρύπτειν Eur. Suppl. 
296: ἐπικράνων 7. 7. 51. Blaydes conj. 
δοὺς δυοῖν κρῖναι κακοῖν (1.6. ‘giving me 
my choice of twoiills’; cp. Ο. Ο. δφοτούτων 

«οὐδίδωμί σοι | κρίναντι χρῆσθαι): Dindorf, 


κομμός. 
tA 
O7p. a. 


στρ. β΄. 


92 TO bOKAEOYS 


Ol. ξύμφημι: δρῶντα γάρ νιν, ὦ γύναι, κακῶς 
εἴληφα τοὐμὸν σώμα σὺν τέχνῃ κακῇ. 
KP. μή νυν ὀναίμην, ἀλλ᾽ ἀραῖος, εἴ σέ τι 


, > > / - > “ lal 
δέδρακ᾽, ὀλοίμην, ὧν ἐπαιτιᾷ pe Spar. 
10. ὦ πρὸς θεῶν πίστευσον, Οἰδίπους, τάδε, 
2 Ν 4 > ν 3 Ν “ 
μάλιστα μὲν τόνδ᾽ ὅρκον αἰδεσθεὶς Dear, 
ἔπειτα κἀμὲ τούσδε θ᾽ ot πάρεισί σοι. 


645 


ΧΟ. 1 πιθοῦ θελήσας φρονήσας 7, ἀναξ, λίσσομαι. 649 
OI. ἃ τί σοι θέλεις OT εἰκάθω; τ 

ΧΟ. 
OI. 4 οἶσθ᾽ οὖν ἃ χρήζεις; ΧΟ. οἶδα. ΟἹ. φράζε δὴ τί φής. 


XO. 5 τὸν ἐναγῆ φίλον μήποτ᾽ ἐν αἰτίᾳ 656 
6 σὺν ἀφανεῖ λόγῳ σ᾽ ἄτιμον βαλεῖν. 
Ol. τεὖ νυν ἐπίστω, ταῦθ᾽ ὅταν ζητῇς, ἐμοὶ 
“ » “Δ νὴ > “A A 
8 ζητῶν ὄλεθρον ἢ φυγὴν ἐκ τῆσδε γῆς. 
ΧΟ. 1 οὐ τὸν πάντων θεῶν θεὸν πρόμον 660 


The word cuvifnots, written over δυοῖν in T, seems to show a consciousness 
of the singularity. 648 πάρεισί σοι made in L from πάρεισ᾽ ἴσοι. Cp. Zi. 1201. 
656 f. L has τὸν ἐναγῆι φίλον μήποτ᾽ ἐν αἰτίαι | σὺν ἀφανεῖ λόγον ἄτιμον ἐκβαλεῖν. 
Over λόγον an early hand has written yw, indicating λόγῳ, which is found in most of 
the later mss. (including A); a few others (as V) have λόγων. Hermann inserted 
σ᾽ after λόγῳ. The false reading ἐκβαλεῖν is in almost all the later Mss.; but T agrees 





\ » \ , ΄“ bs} ν , ͵΄ : 
8 TOV οὔτε πρὶν νήπιον νῦν T ἐν ὅρκῳ μέγαν καταίδεσαι. [MPI 


θάτερον δυοῖν κακοῖν (where I should 
at least prefer κακόν): ‘but since, with 


either of these supposed readings, the 


construction would have been perfectly 
clear, it is hard to see how ἀποκρίνας---ἃα 
far-sought word—could have crept in as 
an explanatory gloss. That, however, 
is Whitelaw’s view, who suggests that 
the original may have been something 
like φαῦλον αἵρεσίν γ᾽ ἐμοί. Wolff would 
compress vv. 640 f. into one, thus: δρᾶσαι 
δικαιοῖ, δείν᾽, ἀποκτεῖναι λαβών. 

642 δρῶντα κακῶς τοὐμὸν σῶμα would 
properly describe bodily outrage: here it 
is a heated way of saying that Creon’s 
supposed plot touched the ferson of the 
king (who was to be dethroned), and not 
merely the νόμοι πόλεως. 

644 ἀραῖος-- ὥσπερ αὐτὸς ἐπαρῶμαι. 

647 ὅρκον θεῶν (object. gen.), an oath 
by the gods (since one said ὀμνύναι θεούς) : 
Od. 2. 377 θεῶν μέγαν ὅρκον ἀπώμνυ: το. 
299 μακάρων μέγαν ὅρκον ὁμόσσαι: Eur. 
Hipp. 657 ὅρκοις θεῶν. But in O. C, 
1767 Διὸς Ὅρκος is personified. 

649—697 The κομμός (see p. 9) has 


a composite strophic arrangement: (1) 
1st strophe, 649—659, (2) 2d strophe, 
660—668; answering respectively to (3) 
ist antistr., 678—688, (4) 2nd antistr., 
689—697. 

649 θελήσας, having consented (πισ- 
tevew). O. C. 757 κρύψον (hide thy 
woes), θελήσας ἄστυ καὶ δόμους μολεῖν. 
1586. or. 8 § 11 ταῦτα ποιῆσαι μὴ θελήσας. 
Plut. Mor. τ40 F συνδειπνεῖν μὴ θελήσαν- 
τος.---φρονήσας, having come to a sound 
mind. Isocr. or. 8 § 141 καλόν ἐστιν ἐν 


Tais τῶν ἄλλων ἀδικίαις καὶ μανίαις πρώτους 


εὖ φρονήσαντας προστῆναι τῆς τῶν ᾿ Ελ- 
λήνων ἐλευθερίας. ' 
651 εἰκάθω : the aor. subj. is certainly 
most suitable here: Phz/. 761 βούλει λά- 
βωμαι; El. 80 θέλεις | μείνωμεν; In 
such phrases the gres. subj. (implying a 
continued or repeated act) is naturally 
much rarer: βούλει ἐπισκοπῶμεν Xen, 
Mem. 3. 5. τ. As regards the form of 
εἰκάθω, Curtius ( Verb 11. 345, Eng. tr. 505), 
discussing presents in -@w-and past tenses 
in -θον from vowel stems, warns us a- 
gainst ‘looking for anything particularly 


OIAITOYS ΤΥΡΆΝΝΟΙΣ 93 


Or. Yea; for I have caught him, lady, working evil, by ill 
arts, against my person. 

Cr. Now may I see no good, but perish accursed, if I have 
done aught to thee of that wherewith thou chargest me! 

Io. O, for the gods’ love, believe it, Oedipus—first, for the 
awful sake of this oath unto the gods,—then for my sake and 
for theirs who stand before thee? 


Kommos. 
CH. Consent, reflect, hearken, O my king, I pray thee! Ist ᾿ 
Or. What grace, then, wouldest thou have me grant thee? °"°P3® 


(ΓΗ. Respect him who aforetime was not foolish, and who 


now is strong in his oath. 


ΟΕ. Now dost thou know what thou cravest ? 


ΘΗ: Wea; 


ΟΕ. Declare, then, what thou meanest. 
CH. That thou shouldest never use an unproved rumour to 
cast a dishonouring charge on the friend who has bound himself 


with a curse. 


ΟΕ. Then be very sure that, when thou seekest this, for me 
thou art seeking destruction, or exile from this land. 


Cu. No, by him who stands in the front of all the heavenly host, 2nd 


with Suidas (s.v. évayjs) in Badetv.—For évay Musgrave conjectured ἀναγῆ: for σὺν, 


Seidler ov γ᾽, reading λόγων (which Musgrave, too, preferred). 
by the rst hand in L, has been changed to φυγὴν by an early corrector. 
In 1, θεὸν is partially effaced, and in most of the latet Mss. it is omitted; 


θεὸν. 


659 φυγεῖν, written 
660 θεῶν 


thus in A it has been completely erased, a space of four letters being left between 





aoristic in the θ᾽ of these verbs. In 
Greek usage, he holds, ‘a decidedly 
aoristic force’ for such forms as σχεθεῖν 
and εἰκαθεῖν ‘never established itself’: 
and he justly cites Z/. 1orqg as a place 
where εἰκαθεῖν is in no way aoristic. He 
would therefore keep the traditional 
accent, and write σχέθειν, εἰκάθειν, with 
Buttmann. Now, while believing with 
Curtius that these forms were prob. in 
origin presents, I also think that in the 
usage of the classical age they were often 
aorists: as ¢.g. σχεθεῖν in Aesch. Zhed. 
429 distinctly is.- 

652 οὔτε mplv...vov τε: cp. O. Cy 
1397 f.—péyay, ‘great,’ 2.6. strong, worthy 
of reverence, ἐν ὅρκῳ, by means of, in 
virtue of, his oath: Eur. 770. 669 ξυν έ- 
σει γένει πλούτῳ τε κἀνδρείᾳ μέγαν : for 
ἐν, cp. Phil. 185 ἔν τ᾽ ὀδύναις ὁμοῦ | λιμῷ 
τ᾽ οἰκτρός. 

656 ‘That thou shouldest never lay 
under an accusation (ἐν αἰτίᾳ βαλεῖν), so 
as to dishonour him (ἄτιμον), with the 


help of an unproved story (σὺν ἀφανεῖ 
λόγῳ), the friend who is liable to a curse 
(€vayy)’: 24. who has just said (644) 
ἀραῖος ὀλοίμαν x.7.A. Aeschin. Jz Ciles. 
§ 110 γέγραπται yap οὕτως ἐν τῇ apa: εἴ 
τις τάδε, φησί, παραβαίνοι, ...ἐν αὙὝ ἡ ς, φη- 
σίν, ἔστω τοῦ ᾿Απόλλωνος, ‘let him 
rest under the ban of Apollo’: as Creon 
would rest under the ban of the gods by 
whom he had sworn. Her. 6. 56 ἐν τῷ 
ἄγεϊ ἐνέχεσθαι, to be liable to the curse. 
ἐν αἰτίᾳ βαλεῖν : [Plat.] Zpist. 7. 341 A 
ws μηδέποτε βαλεῖν ἐν αἰτίᾳ τὸν δεικνύντα 
ἀλλ᾽ αὐτὸν αὑτόν, ‘so that he may never 
blame his teacher, but only himself,’ 
equiv. to ἐμβαλεῖν αἰτίᾳ: cp. the prose 
phrases ἐμβάλλειν els συμφοράς, γραφάς, 
ἔχθραν, κιτ.λ. Eur. 770. 305 εἰς ἔμ᾽ αἰτίαν 
βάλῃ. Seidler’s σύ γ᾽ ἀφανεῖ λόγων, which 
Wolff adopts, is specious. 

660 ov τὸν -- οὐ μὰ τὸν, as not seldom ; 
usu. followed by a second negative (as if 
here we had οὐκ ἔχω τάνδε φρόνησω) : 
1088, Ant. 758, εἰο.---πρόμον, standing 


ἄντ. a. 


94 


1 2 "AXuov: 


3 Ν ¥ ¥ 4 , 
ἐπεὶ ἄθεος ἄφιλος ὁ τι πύματον 


ΣΟΦΟΚΛΕΟΥΣ 


ι" 


> / (A > U > » 
8 ὀλοίμαν, φρόνησιν εἰ τάνδ ἔχω. 


4 ἀλλά μοι δυσμόρῳ γᾶ φθίνουσα 


665 


5 τρύχει ψυχάν, τὰ δ᾽ εἰ κακοῖς κακὰ 
ὃ προσάψει τοῖς πάλαι τὰ πρὸς σφῷν. 


0 δ᾽ οὖν ἴτω, Kel χρή με παντελῶς θανεῖν, 
ἢ γῆς ἄτιμον τῆσδ᾽ ἀπωσθῆναι βίᾳ. 


669 


τὸ γὰρ σόν, οὐ τὸ τοῦδ᾽, ἐποικτίρω στόμα 


ἐλεινόν᾽ οὗτος Θὲ 
υμοῦ περάσῃς. 


ΟἹ, 


ἔνθ᾽ ἂν ἢ στυγήσεται. 
στυγνὸς μὲν εἴκων δῆλος εἶ, βαρὺς δ᾽, 
αἱ δὲ τοιαῦται φύσεις 
αὑταῖς δικαίως εἰσὶν ἄλγισται φέρειν. 

οὔκουν μ᾽ ἐάσεις KAKTOS εἶ ; 


ὅταν 


ΚΡ. πορεύσομαι, 


“A \ \ > ~ 5 δὲ lal 3 » 
σου μεν τυχὼν ayvwiTos, εν εἰ ΤΟῖσ' loos. 


XO. 1 γύναι, τί μέλλεις κομίζειν δόμων τόνδ᾽ ἔσω ; 


θεῶν and πρόμον. 


A few, however, (as V,) keep θεὸν and omit θεῶν. 
665 φθίνουσα] φθινὰς Dindorf: cp. v. 694 
τάδ᾽ Hermann, omitting καί, which the metre (cp. v. 695) condemns. 


678 


T keeps both. 
666 τὰ δ᾽ Kennedy: καὶ τάδ᾽ Mss.: 
668 προσ- 





foremost in the heavenly ranks, most 
conspicuous to the eyes of men: the god 
‘who sees all things and hears all things’ 
(Zl. 3. 277 ὃς πάντ᾽ ἐφορᾷς καὶ πάντ᾽ ἐπα- 
Kovets): invoked Zyrach. 102 as ὦ κρατι- 
στεύων kar’ ὄμμα. 

668 ὅ τι πύματόν (ἐστι), (τοῦτο) 
ὁλοίμαν : schol. φθαρείην ὅπερ ἔσχατον, 
ἤγουν ἀπώλειαν ἥτις ἐσχάτη. 

666 ξ. τὰ δ᾽.- σφῴν: and, on the 
other hand, if the ills arising from you 
two are to be added to the former ills. 
Prof. Kennedy gives td 8’, rightly, I 
think: for ya φθίνουσα refers to the 
blight and plague (25): τάδ᾽ would ob- 
scure the contrast between ¢hose troubles 


and the new trouble of the quarrel.—rpoo-. 


awe. intrans., as perh. only here and in 
fr. 348 καί μοι τρίτον ῥίπτοντι... | ἀγχοῦ 
προσῆψεν, ‘he came near to me.’ Eur. 
Hipp. 188 τὸ μέν ἐστιν ἁπλοῦν" τῷ δὲ 
συνάπτει | Urn τε φρενῶν χερσίν τε πό- 
vos, ‘is joined.’ It is possible, but harsh, 
to make mpocawe act. with “γῆ as subject. 
Since in 695 ἀλύουσαν κατ᾽ ὀρθὸν odpicas 
is clearly sound, Herm. rightly struck out 
καὶ before τὰ δ᾽ here. See on 696. 

669 6 δ᾽ οὖν: then /e¢ him go: Ai. 
114 σὺ δ᾽ οὖν... | χρῶ χειρί. 


672 ἐλεινόν: tertiary predicate: ‘I 


compassionate thy words, piteous as they 
are.” Where a possessive pron. with art. 
has preceded the subst., Soph. sometimes 
thus subjoins an adj., which really has 
the predicative force to which its position 
entitles it, though for us it would be 
more natural to translate it as a mere 
attributive: Ant. 881 τὸν δ᾽ ἐμὸν πότμον 
ἀδάκρυτον | οὐδεὶς... στενάζει: Phil. 1456 
τοὐμὸν ἐτέγχθη I κρᾶτ᾽ ἐνδόμυχον: Li. 
1143 τῆς ἐμῆς πάλαι τροφῆς | ἀνωφελήτου. 
In 1199 (where see note) τὰν γαμψ,. παρθ. 
χρησμῳδόν is not a similar case. Prof. 
Kennedy, placing a comma after ἐποίκ- 
telpw, but none after τοῦδ᾽, construes: τὸ 
σὸν στόμα ἐλεινόν (ἐστι), οὐκ ἐποικτείρω 
τὸ τοῦδε.---στυγήσεται, pass. Other ex- 
amples in Soph. are 1500 ὀνειδιεῖσθε : 
O. C. 581 δηλώσεται, 1186 λέξεται : Ant. 
210 τιμήσεται, 637 ἀξιώσεται:..Ε],. 971 
καλεῖ: Phil. 48 φυλάξεται: among many 
found in prose as well as in verse are ἀδι- 
κήσομαι, ἁλώσομαι, ἐάσομαι, ζξημιώσομαι, 
τιμήσομαι, ὠφελήσομα. The middle 
forms of the aorist were alone peculiar to 
that voice; the so-called ‘future middle,’ 
like the rest, was either middle or pas- 
sive. 

673 £. στυγνὸς... περάσῃς: 
seen to be sullen when t 


‘thou art 
ou yieldest, 


νὸν 


ove 


OIAITOY2 ΤΎΡΑΝΝΟΣ 95 


no, by the Sun! Unblest, unfriended, may I die by the utter- 
most doom, if I have that thought! But my unhappy soul 
is worn by the withering of the land, and again by the thought 
that our old sorrows should be crowned by sorrows springing 
from you twain. 


ΟΕ. Then let him go, though I am surely doomed to death, 
or to be thrust dishonoured from the land. Thy lips, not his, 
move my compassion by their plaint; but he, where’er he be, 


shall be hated. 


Cr. Sullen in yielding art thou seen, even as vehement in 
the excesses of thy wrath; but such natures are justly sorest 


for themselves to bear. 
OE. 
CR. 


but in the sight of these I am just. 


Then wilt thou not leave me in peace, and get thee gone? 
I will go my way; I have found thee undiscerning, 


(Exit. 


Cu. Lady, why dost thou delay to take yon man into the 


house? 


dyer] Nauck conj. προσάξεις.---τὰ προσφῶιν L, 1.2. τὰ πρὸς σφῷν, which is the 


only reading known to the later Mss. 
γενοῦ in 696). 


672 ἐλεεινὸν MSS.: ἐλεινὸν Porson. 


Nauck gives τὰ πρόσφατα (reading εἰ δύνᾳ, 


679 δόμον L: δόμων r. 





but fierce when thou hast gone far in 
wrath’: z.¢., as thou art fierce in passion, 
so art thou sullen in yielding. Greek 
idiom co-ordinates the clauses, though 
the emphasis is on στυγνὸς μὲν εἴκων, 
which the other merely enforces by con- 
trast: see on 419.-- βαρὺς, bearing heavily 
on the object of anger, and so, ‘vehe- 
ment,’ ‘fierce’: Az. 1017 δύσοργος, ἐν 
γήρᾳ βαρύς, 16. 656 μῆνιν βαρεῖαν: Phil. 
1045 βαρύς τε καὶ βαρεῖαν ὁ ξένος φάτιν 
τήνδ᾽ εἶπε: Ant. 767 νοῦς δ᾽ ἐστὶ τηλικοῦ- 
τος ἀλγήσας βαρύς.---περάσῃς absol.,= 
πρόσω ἔλθῃς: O. C. 154 περᾷς, (you go 
too far), 26. 885 πέραν  περῶσ᾽ οἵδε 57.— 
θυμοῦ, partitive gen.: cp. J/. 2. 785 
διέπρησσον πεδίοιο: Her. 3. 105 προλαμ- 
βάνειν...τῆς 6000: sometimes helped by a 
prep. or adverbial phrase, as Xen. Afol. 
30 προβήσεσθαι πόρρω μοχθηρίας : 2 Hpist. 
Tim. 2. τό ἐπὶ πλεῖον γὰρ προκόψουσιν 
doeBelas.—Others render: ‘resentful [or 
‘remorseful ’] even when thou hast passed 
out of wrath’: but (a) περάσῃς with a 
simple gen. could not bear this sense: 
(Ὁ) the antithesis pointed by μὲν and δὲ is 
thus destroyed. 

677 ἀγνῶτος, active, as in 681, 1133: 
but passive, ‘unknown,’ P%. 1008, Anz. 
1ooi. Ellendt is not quite accurate in 
saying that Soph. was the first who used 


ἀγνώς in an active sense, for it is clearly 
active in Pind. Pyth. g. 58 (478 B.C.) οὔτε 
παγκάρπων φυτῶν νήποινον οὔτ᾽ ἀγνῶτα 
θηρῶν (χθονὸς αἶσαν), “ἃ portion of land 
not failing in tribute of plants bearing all 
manner of fruit, nor @ stranger to beasts 
of chase.” The passive use was, however, 
probably older than the active: compare 
Od. 5. 79 ἀγνῶτες... «ἀλλήλοισι (pass.) with 
Thuc. 3. 53 dyvwres ἀλλήλων (act.).—cv 
δὲ τοῖσδ᾽ ἴσος : ἐν of the tribunal or com- 
pany by whom one is judged: Ant. 459 
ἐν θεοῖσι τὴν δίκην δώσειν : Eur. Hipp. 
988 οἱ γὰρ ἐν σοφοῖς | φαῦλοι παρ᾽ ὄχλῳ 
μουσικώτεροι λέγειν : and so, more boldly, 
O. C. 1213 σκαιοσύναν φυλάσσων ἐν ἐμοὶ 
(me zudice) κατάδηλος ἔσται.---ἴσος, aeguus, 
just: Plat. Legg. 975 C τὸν μέλλοντα 
δικαστὴν ἴσον ἔσεσθαι. [Dem.] or. 7 ὃ 35 
(by a contemporary of Dem.) ἴσῳ καὶ κοινῷ 
δικαστηρίῳ. So Ph. 685 ἴσος ὧν ἴσοις 
ἀνήρ. The Scholiast explains, παρὰ δὲ 
τούτοις τῆς ὁμοίας δόξης ἣν καὶ πρώην εἶχον 
περὶ ἐμέ, 1.6. ‘of the same repute as before.’ 
To me such a version of isos appears 
most strange. 

678 Creon leaves the scene. The 
Chorus wish Iocasta to withdraw Oedipus 
also, that he may be soothed in the house: 
but she wishes first to learn how the dispute 
began. 


Ist aniti- 
strophe. 


gine 
, ἈΕῚ ᾿ 


di 


680 


685, 


οὗ ΣΟΦΟΚΛΕΟῪΣ 
IO. 2 padovod y aris ἢ τύχη. 
XO. 8 δόκησις ἀγνὼς λόγων ἦλθε, δάπτει δὲ καὶ τὸ μὴ ᾽νδικον. 
ΙΟ. 4 ἀμφοῖν ἀπ᾽ αὐτοῖν; ΧΟ. ναίχι. 10. καὶ τίς ἣν λόγος; 
ΧΟ. 5 ἅλις ἔμοιγ᾽, ἅλις, γᾶς προπονουμένας, 

6 φαίνεται, ἔνθ᾽ ἔληξεν, αὐτοῦ μένειν. 
Ol. τόορᾷς W ἥκεις, ἀγαθὸς ὧν γνώμην ἀνήρ, 

8 τοὐμὸν παριεὶς καὶ καταμβλύνων κέαρ ; 


πεφάνθαι p av, εἴ o 


ao oO fF ὦ DH μὰ 


684 λόγος L: ὁ λόγος r. 


placing a note of interrogation (;) after ἥκεις. 
In L and A there is a marg. gloss ἐκλύων 
693 εἴ σε νοσφίζομαι MSS. εἴ σ᾽ ἐνοσφιζόμαν Hermann, Hartung 

694 ὅς 7’ MSS.: ὅς γ᾽ Turnebus, and so Wecklein.—révoas 
πόνοισιν Bergk, which obviates the metrical necessity of altering φθίνουσα to 


Cobet). 
on παριείς. 

(-ην), Badham. 
MSS. 


All Mss. give the participles. 


Ls > ον > Ψ , 
ὦναξ, εἶπον μὲν οὐχ ἀπαξ μόνον, 
» \ 4 ¥ ba ld 
ἴσθι δὲ παραφρόνιμον, ἄπορον ἐπὶ φρόνιμα 
> 


* 


689 


ἐνοσφιζόμαν, 


ν 3 > Ν ~ 4 5 4 
ὃς T ἐμαν γᾶν φίλαν ἐν πόνοισιν 
3 > > \ ¥ 
ἀλύουσαν Kat ὀρθὸν ovpicas, 

A ΕῚ » x 
τανῦν T εὐπομπος ἂν 


695 


γένοιο. 


688 Hartung conjectures παρίης καὶ καταμβλύνεις, 


So Wecklein (writing παριεῖς with 





680 μαθοῦσά γ᾽: sc. κομιῶ: cp. 77. 
335 (n.). ") 

681 Sdxyors...Adywv, a suspicioz rest- 
ing on mere assertions (those made by 
Oedipus), and not supported by facts (ép- 
ya): hence ἀγνὼς, unknowing, guided by 
no real knowledge. Thuc. 1. 4 οὐ λόγων 
κόμπος τάδε μᾶλλον ἢ ἔργων ἐστὶν 
ἀλήθεια : 3. 43 τῆς οὐ βεβαίου δοκήσεως-.--- 
δάπτει δὲ: Oedipus was incensed against 
Creon, without proof; on the other hand 
(δὲ) Creon also (kal) was incensed by the 
unjust accusation. —8dare. might be 
historic pres., but need not be so taken: 
Creon is still pained. Aesch. P. V. 437 
συννοίᾳ δὲ δάπτομαι κέαρ. The version, 
‘and even injustice wounds,’ would make 
the words a reflection;—‘An accusation 
galls, even when unfounded’: but this is 
unsuitable. 

683 f. ἀμφοῖν dm’ αὐτοῖν sc. ἦλθε τὸ 
νεῖκος; Thus far, locasta only knew 
that Oedipus charged Creon with treason. 
The words of the Chorus now hint that 
Oedipus himself was partly to blame. 
“80 then,’ Iocasta asks, ‘provocation had 
been given on both sides?’—dédyos, the 
story (of the alleged treason): for the 
words of Oed. (642 δρῶντα κακῶς, τέχνη 
κακή) had been vague. 


685 προπονουμένας, ‘a/ready troubled,” 
not, ‘troubled exceedingly.’ προπονεῖν 
always=to suffer defore, or for: Lucian 
Llupp. Trag. ὃ 40 ᾿Αθηνᾷ "Αρην καταγωνί- 
ferat, ἅτε καὶ προπεπονηκότα olua ἐκ, 
τοῦ τραύματος, already disabled. 

687 The evasive answer of the Chorus 
has nettled Oedipus by implying that the 
blame was divided, and that both parties 
ought to be glad to forget it- He could 
never forget it (672).—op@s tv’ ἥκεις con- 
veys indignant reproach: a grave charge 
has been laid against your king; instead 
of meeting it with denial, you are led, by 
your sympathy with Creon, to imply that 
it cannot be directly met, and must be 
hushed up. Ant. 735 ὁρᾷς τάδ᾽ ws εἴρηκας 
ws ἄγαν νέος: El. 628 ὁρᾷς ; πρὸς ὀργὴν 
ἐκφέρει. 

688 παριεὶς with τοὐμὸν κέαρ, seek- 
ing to relax, enervate, my resentment: ἃ 
sense which the close connection with 
καταμβλύνων interprets, though the more 
ordinary meaning for παριεὶς, had it 
stood alone here, would be ‘neglecting,’ 
‘slighting ’ (πόθος παρεῖτο, El. 545): cp. 
Ar. Za. 436 τοῦ ποδὸς παρίει, slack away 
(some of) the sheet: Eur. Cyel. 591 ὕπνῳ 
παρειμένος : Or. 210 τῷ λίαν παρειμένῳ, 
(neut.) by too great languor. Schneidewin 


fe 


OIAITOYE TYPANNOS 97 


lo. I will do so, when I have learned what hath chanced. 

Cu. Blind suspicion, bred of talk, arose; and, on the other 
part, injustice wounds. 

Io. It was on both sides? 

CH Ave 

Io. And what was the story? 

CH. Enough, methinks, enough—when our land is already 
vexed—that the matter should rest where it ceased. 

ΟΕ. Seest thou to what thou hast come, for all thy honest 
purpose, in seeking to slack and blunt my zeal? 


Cu. King, I have said it not once alone—be sure that I 
should have been shown a madman, bankrupt in sane counsel, 
if I put thee away—thee, who gavest a true course to my 
beloved country when distraught by troubles—thee, who now 
also art like to prove our prospering guide. 


~Owadsin 665. Blaydes suggests πόνοις τότ᾽. 695 ἀλύουσαν] σαλεύουσαν Dobree. 
696 τὰ νῦν δ᾽ 1, ist hand: but δ᾽ has been changed to 7’ by an early | corrector, 
perh. the first. A has 7’, but δ᾽ prevailed in the later Mss.—el δύναιο γενοῦ L. The 
Ist hand wrote εἰ δύναι γενοῦ. The o was added to δύναι (as Diibner thinks) by the 
first corrector, S. Over the letters αὐ something has been erased,—two accents, 





understands, ‘n neglecting my interest, and 
blunting (your) feeling’: but τοὐμὸν must 
surely agree with κέαρ. 

692 ἐπὶ φρόνιμα : [Dem.] or. 25 § 31 
ἐπὶ μὲν καλὸν ἢ χρηστὸν ἢ τῆς πόλεως 
ἄξιον πρᾶγμα οὐδὲν οὗτός ἐστι χρήσιμος. 

698 πεφάνθαι dv, oblique of πεφασμέ- 
vos ἄν ἦν : for the tense cp. Isocr. or. 5 
§ 56 λοιπὸν ἂν ἦν...εἰ μὴ ἐπεποίητο. 
Whitelaw, taking πεφάνθαι μ᾽ ἄν as oblique 
of πεφασμένος av εἴην, defends the εἴ ce 
voogifoua of the mss. by Plat. Phaedr. 
228 Ae ἐγὼ Φαῖδρον ἀγνοῶ, καὶ ἐμαυτοῦ 
ἐπιλέλησμαι, and Apol. 25 B πολλὴ ἄν τις 
εὐδαιμονία εἴη περὶ τοὺς νέους, εἰ εἷς μὲν 
μόνος αὐτοὺς διαφθείρει, κιτ.λ. But the 
playful or ironical tone which εἰ with the 
pres. indic. gives to those passages seems 
hardly in place here. The change of one 
letter restores the required ἐνοσφιζόμαν. 

694 ὅς τε is not for ὅς, though i in £7. 
151 d7’=7, and 77. 824 67’=6: rather 
Te goes with οὔρισας: cP: El. 249 ἔρροι 
τ᾽ dv αἰδὼς | ἁπάντων τ᾽ εὐσέβεια θνατῶν. 

695 ἀλύουσαν, of one maddened by 
suffering, Ph. 1194 ἀλύοντα χειμερίῳ 
Avg. The conj. σαλεύουσαν is tame. 

696 ἂν γένοιο. The mss. have εἰ δύ- 
vavo γενοῦ : for δύναιο, the rst hand of 1, 
had written diva, 2.5. diva. Now εἰ 
δύνᾳ γενοῦ is satisfactory in itself, since 


᾿Εν ΟΕ ἐν 


δύνᾳ for δύνασαι has good authority in 
Attic, as Eur. Hee. 253 δρᾷς δ᾽ οὐδὲν ἡμᾶς 
εὖ, κακῶς δ᾽ ὅσον diva. But then we 
must correct the strophe, 667,—as by 
writing there τὰ πρὸς σῴφῴν τοῖς πάλαι 
προσάψετον, which I should prefer to 
Nauck’s ingenious προσάψει Tots πάλαι τὰ 
πρόσφατα. Verse 667, however, seems 
right as it stands: it gives a better 
rhythm for the closing cadence than we 
should obtain by adding a syllable. And 
if so, εἰ δύναιο (or δύνᾳ) γενοῦ here must 
be reduced to~—~=. (1) If with Hermann 
we simply omit γενοῦ, the elliptical εἰ 
Sivato—understanding ἴσθι or γενοῦ---ἰϑ 
intolerably harsh; to me it does not seem 
even Greek. (2) εἰ γένοιο, ‘mayest thou 
become!’ is read by Bergk and Dindorf; 
cp. 863 εἴ μοι ξυνείη. (3) To this I much 
prefer dv γένοιο, which Blaydes adopts; 
but I do so for a reason which he does 
not give. I suspect that εἰ δύναιο was a 
marginal gloss intended to define the 
sense of dv γένοιο, : and that dv γένοιο was 
corrupted to γενοῦ when εἰ δύναιο had 
crept into the text. (4) Prof. Kennedy 
conjectures εἶ τό γ᾽ ἔν σοι: ‘now also 
with thy best shill thou ably waftest. 
Since the metre of 667 is not certainly 
sound, no treatment of our verse can be 
confident. 


7 


and anti- 
strophe. 


98 ΣΟΦΟΚΛΕΟΥΣ 

10: πρὸς θεῶν δίδαξον. Kap , ἄναξ, ὅτου ποτὲ 
μῆνιν τοσήνδε πράγματος. στήσας ἔχεις. 

ΟἹ: ἐρῶ" σὲ γὰρ τῶνδ᾽ ἐς πλέον, γύναι, σέβω" 700 
Κρέοντος, οἷά μοι βεβουχέυκὼς ἔχει. 

[Ο. λέγ᾽, εἰ σαφῶς τὸ νεῖκος ἐγκαλῶν ἐρεῖς. 

Ol. φονέα με φησὶ Λαΐου καθεστάναι. 

10. αὐτὸς ξυνειδώς, n μαθὼν ἄλλου πάρα; 

Ol. μάντιν μὲν οὖν κακοῦργον εἰσπέμψας, ἐπεὶ 705 
τό γ᾽ εἰς ἑαυτὸν πᾶν ἐλευθεροῖ στόμα. 

IO. σύ νυν ἀφεὶς σεαυτὸν ὧν λέγεις πέρι" 
ἐμοῦ ᾿πάκουσον, καὶ pa οὕνεκ᾽ ἐστί σοι 

«--- βρότειον οὐδὲν μαντικῆς ἔχον τέχνης. 
φανῶ δέ σοι σημεῖα τῶνδε σύντομα. 710 


χρησμὸς γὰρ ἦλθε Λαΐῳ ποτ᾽, οὐκ ἐρῶ 
Φοίβου. γ᾽ ἀπ᾽ αὐτοῦ, τών δ᾽ ὑπηρετῶν ἄπο, 
ὡς αὐτὸν ἤξοι μοῖρα πρὸς παιδὸς θανεῖν, 


according to Diibner; Campbell suggests σύ. ---εἰ δύναιο γενοῦ (εἰ δύναι ὁ γενοῦ Bodl. 


Barocc. 66) is also the reading of the later Mss. 
M. Seyffert. 


κυρεῖς Ἐρσοτί.---ἐγκαλεῖν ἔχεις 


See comment. 702 ἐρεῖς] 
709 ἔχον] τυχὸν Hartung ; λαχὸν 





697 f. «dp: these men know it: 
allow me also to know it.—étov...mpdy- 
ματος, causal gen.; Ant. 1177 πατρὶ 
μηνίσας φόνου. «π-στήσας ἔχεις, hast set 
up, z.¢. conceived as an abiding senti- 
ment, referring to 672 and 689. Cp. 


Eur. Z A. 785 ἐλπὶς... | olay... | o77- 
σασαι τάδ᾽' és ἀλλήλας | μυθεύσουσι 
(Fritzsch). 


700 f. τῶνδ᾽ é és πλέον = πλέον ἢ τούσδε, 
not πλέον ἢ olde. The Chorus having 
hinted that Oedipus was partly to blame, 
he deigned no reply to their protests of 
loyalty (689 f.). But he respects Iocasta’s 
judgment more, and will answer her.— 
Κρέοντος, sc. στήσας ἔχω τὴν μῆνιν: 
causal gen. answering to ὅτου πράγματος. 
“-ἰεβουλευκὼς: in this periphrasis, the 
perf. part. is rarer than the aor. part.: 
Ph, 600 n. 

702 λέγ᾽ : speak, if you can make a 
clear statement (εἰ σαφώς épets) in im- 
puting the blame of the feud: 2.6. if you 
are prepared to explain the vague ola 
(701) by defining the provocation.—éyxa- 
λεῖν νεῖκός (7w1)=to charge one with 
(beginning) a quarrel: as Phzl. 328 χόλον 
(τινὸς) kar’ αὐτῶν ἐγκαλῶν, charging them 
with having provoked your anger at a deed. 

704 2. αὐτὸς ξυνειδώς: 2.4. does he 


speak as from his jown knowledge (of 
your guilt)?—pév οὖν, ‘nay.’ 151. 1503. 

Ar. £q. 13 NI. λέγε σύ. ΔΗ. σὺ μὲν οὖν 
λέγε. Distinguish μὲν οὖν in 483, where 
each word has a separate force. 

706 τό γ᾽ εἰς ἑαυτὸν, in what concerns 
himself: Eur. 7. 7. 691 τὸ μὲν γὰρ εἰς ἔμ᾽ 
οὐ κακῶς ἔχει. ---πτᾶν ἐλευθεροῖ, sets wholly 
free (from the discredit of having brought 
such a charge): Amt. 445 ἔξω βαρείας 
αἰτίας ἐλεύθερον : Plat. Legg. 756 Ὁ ἐλεύ- 
θερον ἀφεῖσθαι τῆς ζημίας. 

707 ἀφεὶς σεαυτόν, an appropriate 
phrase, since ἀφιέναι was the regular 
term when the natural avenger of a slain 
man voluntarily released the slayer from 
the penalties: Dem. or. 38 § 59 ἂν ὁ 
παθὼν αὐτὸς ἀφῇ τοῦ φόνου τὸν Spleen: 
Antiph. or. 2 § 2 οὐ τὸν αἴτιον ἀφέντες τὸν 
ἀναίτιον διώκομεν. ὟΝ 

708 pad’ κ-τ.λ. : learn that thou canst 
find no mortal creature sharing in the art 
of divination.—oov ethic dat.:. ἐστὶν 
gxov=éxee (Eur. Suppl. 427 τί τούτων 
ἐστὶν ob καλῶς ἔχον :): τέχνης, partitive 
gen. The gods have prescience (498); 
but they impart it to no man,—not even 
to such ministers as the Delphian priests. 
Iocasta reveres the gods (Gan): it is to 
them, and first to Apollo, that she turns 


06 


ΟἸΔΙΠΟῪΣ TYPANNOS 99 


10; 


In the name of the gods, tell me also, O king, on what 


account thou hast conceived this steadfast wrath. 


OE. 


That will I; for I honour thee, lady, above yonder 


men :—the cause is Creon, and the plots that he hath laid 


against me. 


Io. Speak on—if thou canst tell clearly how the feud 


began. 


OE. 


another ? 


He says that I stand guilty of the blood of Laius. 
Io. As on his own knowledge? 


Or on hearsay from 


OE. Nay, he hath made a rascal seer his mouth-piece; as 
for himself, he keeps his lips wholly pure. 

Io. Then absolve thyself of the things whereof thou speak- 
est; hearken to me, and learn for thy comfort that nought 


of mortal birth is a sharer in the science of the seer. 


give thee pithy proof of that. 


I will 


An oracle came to Laius once—I will not say from Phoebus 
himself, but from his ministers—that the doom should overtake 
him to die by the hand of his child, 


Heimsoeth. 


713 ἤξοι L rst hand, changed by an early hand to ἥξει. 


Most of 


the later Mss. have ἥξει, but one or two (V, 1.2) 7&o..—Canter conject. ἕξει: K. Halm, 





in trouble (911). But the shock which 
had befallen her own life,—when at the 
bidding of Delphi her first-born was 
sacrificed without saving her husband 
Laius—has left a deep and bitter con- 
viction that no mortal, be he priest or 
seer, shares the divine foreknowledge. 
In the Greek view the μάντις might be 
(1) first, the god himself, speaking 
through a divinely frenzied being in 
whom the human reason was temporarily 
superseded (hence the popular derivation 
of μαντική from μανία) : Plat. Zim. 71 
E μαντικὴν ἀφροσύνῃ θεὸς ἀνθρωπίνῃ δέ- 
δωκεν" οὐδεὶς γὰρ ἔννους ἐφάπτεται μαν- 
τικῆς ἐνθέου καὶ ἀληθοῦς: this was much 
the same as the Egyptian belief, Her. 2. 
83 μαντικὴ δὲ αὐτοῖσι ὧδε διακέεται. ἀν- 
θρώπων μὲν οὐδενὶ προσκέεται 7 τέχνη, 
τῶν δὲ θεῶν μετεξετέροισι. (2) Secondly, 
the μάντις might be a man who reads 
signs from birds, fire, etc., by rule of 
mystic science: it was against this τέχνη 
that scepticism most readily turned: Eur. 
El. 399 Λοξίου yap ἔμπεδοι | χρησ- 
μοί, βροτῶν δὲ μαντικὴν χαίρειν 
λέγω. Iocasta means: ‘I will not say 
that the message came through the lips 
of a truly god-possessed interpreter; but 
at any rate it came from the priests; it 


was an effort of human μαντική.͵ So in 
946, 953 θεῶν μαντεύματα are oracles 
which grofessed to come from the gods. 
Others render:—‘Nothing in mortal 
affairs ts connected with the mantic art’: 
7.2. is affected by it, comes within its ken. 
Then ἐστὶν ἔχον will not stand for ἔχεται 
(which it could not do), but for ἔχει, as 
meaning ‘zs of,’ ‘belongs to.’ Her. has 
ἔχειν as=elvar with expressions equivalent 
to an adverb, as 2. QI ἀγῶνα γυμνικὸν διὰ 
πάσης ἀγωνίης ἔχοντα, ‘consisting in 
every sort of contest,’ as he might have 
said πολυτρόπως ἔχοντα: 50 3. 128 περὶ 
πολλῶν ἔχοντα πρηγμάτων (=ToA- 
Aax@s): 6. 42 κατὰ χώρην (Ξεἐμπέδως) 
ἔχοντες: 7. 220 ἐν ἔπεσι ἑξαμέτροισι 
ἔχοντα. But such instances are wholly 
different from the supposed use of ἔχειν 
alone ἃ5 Ξε εἶναι with a partitive genitive. 

711 οὐκ ἐρῶ «.7.A. The exculpation 
of Apollo Azmself here is obviously not 
inconsistent with 720, which does not 
ascribe the prediction to him. And in 
853 (ὅν γε Λοξίας | διεῖπε) the name of 
the god merely stands for that of his 
Delphian priesthood. 

713 ἥξοι is better than the conject. 
ἕξοι (‘constrain’), as expressing the sud- 
denness with which the doom should 


7—2. 


«<3 


100 


ZOPOKAEOYS 


ὅστις γένοιτ᾽ ἐμοῦ ΤΕ κἀκείνου πάρα. 


καὶ τὸν μέν, ὥσπερ γ᾽ ἡ φάτις, ἕένοι ποτὲ 
x λῃσταὶ φονεύουσ᾽ ἐν τριπλαῖς ἁμαξιτοῖς" “" 


1.19 


a 





παιδὸς δὲ βλάστας οὐ διέσχον ἡμέραι 
τρεῖς, καί νιν ἄρθρα κεῖνος ἐνζεύξας ποδοῖν 
ἔρριψεν ἄλλων χερσὶν εἰς ἄβατον ὄρος. 


κἀνταῦθ᾽ ᾿Απόλλων οὔτ᾽ ἐκεῖνον ἤνυσεν 


720 


φονέα γενέσθαι πατρός, οὔτε Λαΐον, 

Ν \ e A Ν Ν A 
τὸ δεινὸν οὐφοβεῖτο, πρὸς παιδὸς θανεῖν. 
τοιαῦτα φῆμαι μαντικαὶ διώρισαν, 
ὧν ἐντρέπου σὺ μηδέν" ὧν γὰρ av θεὸς 


χρείαν ἐρευνᾷ ῥᾳδίως αὐτὸς φανεῖ. 
οἷόν μ᾽ ἀκούσαντ᾽ “ἀρτίως ἔχει, γύναι, 


725 


ψυχῆς πλάνημα κανακίνησις φρενών. 


ἔδοξ᾽ ἀκοῦσαι σοῦ τόδ᾽, 


κατασφαγείη πρὸς anaes ἁμαξιτοῖς. 
ηὐδᾶτο “γὰρ ταῦτ᾽, οὐδέ πω λήξαντ' ἔχει. 
ὥρος οὗτος οὗ τόδ᾽ ἦν πάθος ; 


Φωκὶς μὲν n Yn κλήζεται, σχιστὴ δ᾽ ὁδὸς 


καὶ ποῦ oF ὁ 


ποίας μερίμνης τοῦθ᾽ ὑποστραφεὶς λέγεις ; 


ὡς ὁ Adtos 


730 


ἐς ταὐτὸ Δελφῶν κἀπὸ Δαυλίας ἄγει. 


ἕξοι. 719 εἰς ἄβατον ὄρος MSS.: 


ἄβατον εἰς ὄρος Musgrave. 
L yp. παθεῖν has been written above by a late hand: A has the same gloss. 


722 θανεῖν MSs. In 
728 ὑπο- 





overtake him. El. 489 ἥξει... Ἐρινύς. 
The simple acc. αὐτὸν, since ἥξοι-Ξ κατα- 
λήψοιτο: cp. Her. 9. 26 φαμὲν ἡμέας 
ἱκνέεσθαι ἡγεμονεύειν, instead of ἐς ἡμέας 
(2. 29). 

714 ὅστις γένοιτ᾽ is oblique for ὅστις 
ἂν γένηται (whoever may be born), not 
for ὅστις ἐγένετο (who has been born): 
Laius received the oracle before the birth 
of the child. 

715 ξένοι: 
his own blood. 

26: See on 733. 

17 διέσχον. ‘Three days had not 
separated the child’s birth from us’: 
three days had not passed since its birth. 
Plut. 776. Gracch. ὃ 18 κελεύσαντος éxel- 
vou διασχεῖν τὸ πλῆθος, to keep the crowd 
off.—BAderas cannot be acc. of respect 
(‘as to the birth’), because διέσχον could 
not mean ‘had elapsed’: when διέχειν is 
intrans. it means (4) to be distant, Thuc. 


not Thebans, much less of 


Ἑλλὰς ὠνόμαζεν Οἰδίπουν. 


8. 79 διέχει δὲ ὀλίγον ταύτῃ ἡ Σάμος τῆς 


ἠπείρου: or (6) to extend, Her. 4. 42 
διώρυχα... διέχουσαν és τὸν ᾿Αράβιον 
κόλπον. 


718 καί--ὅτε (parataxis instead of 
hypotaxis): Thuc. 1. 50 ἤδη δὲ ἦν dye... 
kal οἱ ἹΚορίνθιοι ἐξαπίνης πρύμναν ἀκρού. 
οντο.---“ἄρθρα ποδοῖν--τὰ σφυρά: ἐνζεύ- 
fas, fastened together by driving a pin 
through them, so as to maim the child 
and thus lessen the chance of its being 
reared if it survived exposure:-Eur. Pi. 
22 (Iocasta speaks) ἔσπειρεν ἡμῖν παῖδα, 
kal σπείρας βρέφος, | γνοὺς τἀμπλάκημα 
τοῦ θεοῦ τε τὴν φάτιν, | λειμῶν᾽ ἐς Ἥρας 
καὶ Κιθαιρῶνος λέπας | δίδωσι βουκόλοισιν 
ἐκθεῖναι βρέφος, | σφυρῶν σιδηρᾷ κέντρα 
διαπείρας μέσον (better μέσων), | ὅθεν νιν 
Seneca Oed. 
812 Forata ferro gesseras vestigia, Tumore 
nactus nomen ac vitio pedum. 

719 εἰς ἄβατον ὄρος: the tribrach con- 


ΟἸΔΊΠΟΥΣ TYPANNO2 


ΙΟΙ 


who should spring from him and me. 
Now Laius,—as, at least, the rumour saith,—was murdered 


one day by foreign robbers at a place where three highways 


meet. 


And the child’s birth was not three days past, when 


Laius pinned its ankles together, and had it thrown, by others’ 


hands, on a trackless mountain. 


So, in that case, Apollo brought it not to pass that the babe 
should become the slayer of his sire, or that Laius should die— 
the dread thing which he feared—by his child’s hand. Thus 


did the messages of seer-craft map out the future. 
Whatsoever needful things the god 


them, thou, not at all. 


Regard 


seeks, he himself will easily bring to light. 
OE. What restlessness of soul, lady, what tumult of the 
mind hath just come upon me since I heard thee speak ! 
Io. What anxiety hath startled thee, that thou sayest this? 
ΟΕ. Methought I heard this from thee,—that Laius was 


slain where three highways meet. 


Io. Yea, that was the story; nor hath it ceased yet. 

ΟΕ. And where is the place where this befell ? 

Io. The land is called Phocis; and branching roads lead to 
the same spot from Delphi and from Daulia. 


στραφεὶσ L: ὕπο στραφεὶς r, which Dindorf and others prefer. 


730 διπλαῖσ L: τριπλαῖς r. 


ἐπιστραφεὶς Blaydes. 





tained in one word gives a ruggedness 
which is certainly intentional here, as in 
1496 τὸν πατέρα πατήρ, Az. 459 πεδία 
τάδε. A tribrach in the sth place, always 
rare, usually occurs either when the pen- 
ultimate word of the verse is a pacon 
primus (-~~~), as Zl. 326 ἐντάφια 
χεροῖν, or when the last word is a paeon 
guartus (-~~—), as Phil. 1302 ἄνδρα πο- 
λέμιον. Verse 967 below is exceptional. 

720 κἀνταῦθ᾽: cp. 582. 

722 It is more likely that, as our Mss. 
suggest, παθεῖν should have been a com- 
mentator’s conjecture than that θανεῖν 
should have been a copyist’s error (from 
v. 713). No objection can be drawn 
from the occurrence of πρὸς παιδὸς θα- 
νεῖν so soon after 713: see on 519. 

723 τοιαῦτα... διώρισαν, 1.5. made 
predictions at once so definite and so 
false: φῆμαι, a solemn word used scorn- 
fully: cp. 86. The sense of διώρισαν ἴῃ ὃ 
1083 is slightly different: here we might 
compare Dem. or. 20 ἃ 158 ὁ Δράκων... 
καθαρὸν διώρισεν εἶναι, Shas laid down 
that the man is pure.’ ' 

725 ὧν χρείαν ἐρευνᾷ : a bold phrase 


blended, as it were, from ὧν dv χρείαν 
ἔχῃ and ἃ av χρήσιμα (ὄντα) ἐρευνᾷ: cp. 
Phil. 327 τίνος... | χόλον... ἐγκαλῶν, in- 
stead of τίνος χόλον ἔχων or τί ἐγκαλῶν. 

726—754 The mention of ‘three 
roads’ (716) has startled Oedipus. He 
now asks concerning (1) the place, (2) the 
time, (3) the person. The agreement of 
(1) with (2) dismays him; that of both 
with (3) flashes conviction to his mind. 

727 πλάνημα denotes the fearful 
‘wandering’ of his thought back to other 
days and scenes; as ἔδοξ᾽ (729) is the 
word of one who has been in a troubled 
dream. 

728 ποίας pep. ὕποστρ., having turned 
round on account of (=startled by) what 
care,—like a man whom a sound at his 
back causes to turn in alarm:—far more 
expressive than ἐπιστραφείς, which would 
merely denote attention. For the gen., 
cp. Az. 1116 Tod δὲ σοῦ ψόφου | οὐκ ἂν 
στραφείην. : 

731 λήξαντ᾽ : the breath of rumour is 
as a breeze which has not yet fallen: cp. 
AZ. 258 νότος ws λήγει, and O. C. 517. 

733 σχιστὴ δ᾽ ὁδός. In going from 


ΣΟΦΟΚΛΕΟΥΣ 


795 


102 

OI. καὶ τίς χρόνος τοῖσδ᾽ ἐστὶν οὐξεληλυθώς ; 

10. σχεδόν τι πρόσθεν ἢ σὺ τῆσδ᾽ ἔχων χθονὸς 
ἀρχὴν ἐφαίνου τοῦτ᾽ ἐκηρύχθη πόλει. 

ΟΙ. ὦ Ζεῦ, τί μου δρᾶσαι βεβούλευσαι πέρι; 


740 


745 


ΙΟ. τί δ᾽ ἐστί σοι τοῦτ᾽, Οἰδίπους, ἐνθύμιον ; 
’ὔ 3 > 4 Ν οὗ foe » 
Ol. μήπω μ᾽ ἐρώτα" τὸν δὲ Λάϊον φύσιν 
ΣΕ Soe ΄ ὃς 7 > \ Ψ ¥ 
TW εἶχε φράζε, “τίνος ἀκμὴν nBys ἔχων. 
IO. μέγας, χνοάζων ἄρτι λευκανθὲς κάρα, 
μορφῆς δὲ τῆς σῆς οὐκ ἀπεστάτει πολύ. 
ΟΙ. οἴμοι τάλας" ἔοικ᾽ ἐμαυτὸν εἰς ἀρὰς 
δεινὰς προβάλλων ἀρτίως οὐκ εἰδέναι. 
10. πῶς φής; ὀκνῶ τοι πρὸς σ᾽ ἀποσκοποῦσ᾽, ἀναξ. 
ΟἹ. δεινῶς ἀθυμῶ μὴ βλέπων ὁ μάντις 7. 
δείξεις δὲ μᾶλλον, ἢν ἕν ἐξείπῃς ἔτι. 
IO. καὶ μὴν ὀκνῶ μέν, av δ᾽ ἔρῃ μαθοῦσ᾽ ἐρῶ. 


740 φύσιν | τίν᾽ εἶχε φράζε" τίνα δ᾽ ἀκμὴν ἥβης ἔχων. L. The only variation in the 


later MSS. is ἔσχε for εἶχε (A). 


I adopt a former conjecture of Nauck’s, τίνος for τίνα 


δ᾽. Wecklein changes ἥβης ἔχων to ἔχων ἔβη: Meineke changes ἥβης to τότ᾽ ἦλθ᾽: 
Wolff gives, τίν᾽ εἶχε, φράζ᾽ ἔτ᾽" ἦν δ᾽ ἀκμὴν ἥβης ἔχων; Others seek a substitute 
either (1) for ἔχων, as Brunck τότε, Kennedy ἔτι: or (2) for εἶχε, as Dindorf ἦλθε, 


Hartung ἔτυχε, Schneidewin and Blaydes elpze. 


742 μέγασ L. A few later 


mss. (A, Pal., and V as corrected) have μέλας, which Wecklein adopts.— xvod (wy 





Thebes to Delphi, the traveller passes by 
these ‘Branching Roads,’—still known 
as the rploda, but better as the στενό: 
from Daulia it is a leisurely ride of about 
an hour and a half along the side of Par- 
nassus. The following is from my notes 
taken on the spot:—‘ A bare isolated hil- 
lock of grey stone stands at the point 
where our path from Daulia meets the 
road to Delphi, and a third road that 
stretches to the south. There, in front, 
we are looking up the road down which 
Oedipus came [from Delphi]; we are 
moving in the steps of the man whom he 
met and slew; the road runs up a wild 
and frowning pass between Parnassus 
on the right hand and on the left the 
spurs of the Helicon range, which here 
approach it. Away to the south a wild 
and lonely valley opens, running up 
among the waste places of Helicon, a 
vista of naked cliffs or slopes clothed with 
scanty herbage, ’a scene of inexpressible 
grandeur and desolation’ (Modern Greece 
p- 79). At this σχιστὴ ὁδός Pausanias 


] 


᾿ 


saw τὰ τοῦ Λαΐου μνήματα καὶ οἰκέτου 
τοῦ ἑπομένου : the legend was that Dama- 
sistratus king of Thebes had found the 
bodies and buried them (10. 5 ὃ 4). The 
spot has a modern monument which 
appeals with scarcely less force to the 
imagination of a visitor,—the tomb of a 
redoubtable brigand who was killed in 
the neighbourhood many years ago. 

734 ταὐτὸ, but in 325 ταὐτὸν : cp. 
Tr. 325 n. ἀπὸ with both genitives: cp. 
761, 1205. 

735 τοῖσδ᾽. For the dat. cp. Her. 2. 
145 Διονύσῳ μέν vuv,..xata ἑξακόσια 
ἔτεα καὶ χίλια μάλιστά ἐστι ἐς ἐμέ: Ἥ ρα- 
κλέϊ δὲ.. κατὰ εἰνακόσια ἔτεα’ Πανὶ δὲ 
κατὰ τὰ ὀκτακόσια μάλιστα ἐς ἐμέ. Then 
from persons the idiom is transferred to 
things: Thuc. 3. 29 ἡμέραι μάλιστα ἦσαν 
TH Μυτιλήνᾳ ἑαλωκυίᾳ ἑπτά. 

736 σχεδόν τι πρόσθεν. The interval 
supposed between the death of Laius and 
the accession of Oedipus must be long 
enough to contain the process by which 
the Sphinx had gradually brought Thebes 


ΟἸΔΊΠΟΥΣ. ΤΥΡΑΝΝΌΣ 103 

ΟΕ. And what is the time that hath passed since these 
things were? 

Io. The news was published to the town shortly before thou 
wast first seen in power over this land. 

Or. O Zeus, what hast thou decreed to do unto me? 

Io. And wherefore, Oedipus, doth this thing weigh upon 
thy soul? 

Or. Ask me not yet; but say what was the stature of 
Laius, and how ripe his manhood. 

Io. He was tall,—the silver just lightly strewn among his 
hair; and his form was not greatly unlike to thine. 

Or. Unhappy that I am! Methinks I have been laying 
myself even now under a dread curse, and knew it not. 

Io. How sayest thou? I tremble when I look on thee, 
my king. 

Or. Dread misgivings have I that the seer can see. 
thou wilt show better if thou wilt tell me one thing more. 

Io. Indeed—though I tremble—I will answer all thou ask- 
est, when 1 hear it. 


But 


L, not altered from xvodfov: nor is the latter (so far as I know) in any MS.—dev- 
κανθὲς L, which is the usual reading in the later Mss.; only one or two have 
λευκανθεὶς (I') or λευκανθὲν (A). Hartung reads χνοάζον.. λευκανθεὶς κάρα. 743 In 
L ἀπεστάτει has been made from ἀποστάτει by an early hand. 749 ἃ δ᾽ ἀν L, 
and so nearly all the later Mss. (but ἂν δ᾽ Dresd. a, ἄν δ᾽ Bodl. Laud. 54). On 
such a point as ὦ δ᾽ dv versus ἃν δ᾽, the authority of our Mss, is not decisive. In 
O. C. 13 dv δ᾽ seems clearly preferable to ἃ δ᾽ ἂν (1, there has ay, omitting δ᾽; and 





to despair: but Soph. probably had no 
very definite conception of it: see on 758. 

738 ὦ Zev. A slow, halting verse, 
expressing the weight on his soul: the 
neglect of caesura has this purpose. 

739 ἐνθύμιον: Thuc. 7. 50 ἡ σελήνη 
ἐκλείπει... καὶ οἱ ᾿Αθηναῖοι... ἐπισχεῖν ἐκέ- 
λευον τοὺς στρατηγούς, ἐνθύμιον ποιού- 
μενοι. 

740 I do not believe that Soph., or 
any Greek, could have written vcvw | 
τίν᾽ εἶχε, φράζε, τίνα δ᾽ ἀκμὴν ἥβης 
ἔχων, which Herm. was inclined to defend 
as if rlva φύσιν εἴχεξετίς ἦν φύσι. Now 
τίνος would easily pass into τίνα δ᾽ with a 
scribe who did not follow the construc- 
tion; and to restore τίνος seems by far 
the most probable as well as the simplest 
remedy. No exception can be taken to 
the phrase τίνος ἀκμὴν ἥβης as= ‘the ripe- 
ness of what period of vigorous life.’ 

742 χνοάζων λευκανθὲς κάρα-- ἔχων 
χνοάζον λευκαῖς κάρα: Ar. MVub. 978 


χνοῦς ὥσπερ μήλοισιν ἐπήνθει (the down on 
his chin was as the bloom on apples): 
here the verb marks the /igh¢ strewing of 
silver in dark hair. Cp. 27. 43 ἦνθισ- 
μένον.Ό As Aesch. has μελανθὲς γένος, 
‘swarthy’ (Suppl. 154), so in Anthol. 
12. 165 (Jacobs 11. 502) λευκανθής -- ‘of 
fair complexion’ as opp. to meAlypous. 

744 τάλας, as being for τάλανς: Ar. 
Av. 1494 οἴμοι τάλας, ὁ Ζεὺς ὅπως μή μ᾽ 
ὄψεται. In Anthol. 9. 378 (Jac. 11. 132) 
kal κοιμῶ μεταβάς, ὦ τάλας, ἀλλαχόθι, 
τάλαν is an easy remedy: but not so in 
Theocr. 2. 4 ἀφ᾽ ὦ τάλας οὐδέποθ᾽ ἥκει, 
where πέλας has been conjectured.—ouxa 
00K εἰδέναι ΞΞ ἔοικεν ὅτι οὐκ ἤδη: cp. 
236 f. 

749 καὶ μὴν, ‘indeed’ I fear (as you 
do): Ant. 221, Z/. 556.—dv δ᾽ is certainly 
preferable to ὦ δ᾽ ἂν in a poet whose ver- 
sification is not characterised by any love 
of unnecessary διάλυσις. Even in prose we 
find ὃς av δέ instead of os δὲ dv, Her. 7. 8. 


104 ΣΟΦΟΚΛΕΟῪΣ 
Ol. πότερον ἐχώρει βαιός, ἦ πολλοὺς ἔχων 750 
ἄνδρας λοχίτας, οἵ ἀνὴρ ἀρχηγέτης; 
IO. πέντ᾽ ἦσαν ou ξύμπαντες, ἐν δ᾽ αὐτοῖσιν ἣν 
κῆρυξ' ἀπήνη δ᾽ ἦγε Λάϊον μία. 
OI. αἰαῖ, τάδ᾽ ἤδη διαφανῆ. τίς ἦν ποτὲ 
ὁ τούσδε λέξας τοὺς λόγους ὑμῖν, γύναι ; 755 
10. οἰκεύς τις, ὅσπερ ἵκετ᾽ ἐκσωθεὶς μόνος. 
Ol. ἦ Kav δόμοισι τυγχάνει τανῦν παρών; 
10: ov δῆτ᾽ . ad’ ov yap κεῖθεν ἦλθε καὶ κράτη 
σέ T εἶδ᾽ ἔχοντα Λάϊόν T ὀλωλότα, 
ἐξικέτευσε τῆς ἐμῆς χειρὸς θιγὼν 760 
ἀγρούς σφε πέμψαι κἀπὶ ποιμνίων νομάς, 
ὡς πλεῖστον εἴη τοῦδ᾽ arom TOS ἄστεως. 
κάπεμψ᾽ ἐγώ νιν" ἄξιος γὰρ ot ἀνὴρ 
δοῦλος φέρειν ἣν τῆσδε καὶ μείζω χάριν. 
OI. πῶς ἂν μόλοι δηθ' ἡμὶν ἐν “τάχει πάλιν ; 765 
10. πάρεστιν" ἀλλὰ πρὸς τί τοῦτ᾽ ἐφίεσαι; 
ΟΙ. δέδοικ᾽ ἐμαυτόν, ὦ γύναι, μὴ πόλλ᾽ ἄγαν 


εἰρημέν᾽ ἦ μοι, Ov a νιν εἰσιδεῖν θέλω. 


here, too, it gives a more Sophoclean rhythm. 


the first corrector (S): 


the rst hand seems to have written ὥσπερ. 


756 ὅσπερ L, as re-touched by 
763 of Her- 





750 βαιός identifies the chief with his 
retinue,—the adjective, when so used, 
suggesting a collective force like that of 
a stream, full or thin: so πολὺς pet, πολὺς 
πνεῖ of vehement speech, etc.; Eur. Or. 
1200 ἣν πολὺς παρῇ, if he come in his 
might: συχνὸν πολίχνιον, a populous 
town (Plat. Rep. 370 D). 

751 doxlras: cp. Aesch. Cho. 766 
XO. πώς οὖν κελεύει viv μολεῖν ἐσταλμέ- 
vov; | ...9 ξὺν λοχίταις εἴτε καὶ μονοστιβῆ; 
ΤΡ. ἄγειν κελεύει δορυφόρους ὀπάονας (said 
of Aegisthus). 

753 κῆηρνξ, as the meet attendant of a 
king on the peaceful and sacred mission 
of a θεωρός (114). The herald’s presence 
would add solemnity to the sacrifice and 
libation at Delphi: Athen. 660 A ἔδρων 
(Ξ: ἔθυον) δὲ of κήρυκες ἄχρι πολλοῦ, Bov- 
θυτοῦντες... καὶ σκευάζοντες καὶ μιστύλλον- 
τες, ἔτι δὲ οἰνοχοοῦντεςς ἀπήνη ἦγε μία -- 
μία ἦν ἀπήνη, ἣ ἦγε: Pind. Nem. 9. 41 
ἔν θ᾽ ᾿Αρέας πόρον ἄνθρωποι καλέοισι-- ἔνθα 
πόρος ἐστὶν ὃν ’A. καλοῦσιν. The ἀπήνη, 
properly a mule-car (Pind. Pyth. 4. 94) 
but here drawn by colts (802), and in the 


Odyssey synonymous with ἅμαξα (6. 37, 
57), was a four-wheeled carriage used for 
travelling, as dist. from the two-wheeled 
war-chariot (ἅρμα) : its Homeric epithet 
ὑψηλή indicates that it stood higher on its 
wheels than the ἅρμα: it could be fitted 
with a frame or basket for luggage (ὑπερ- 
tepln Od. 6. 70, πείρινς Jl. 24. 190). 

756: cp. 118. οἰκεύς -- οἰκέτης, as in 
the Odyssey and in a véduos Σόλωνος in 
Lysias or. 10 § 19, who explains it by 
θεράπων. The Jlad has the word only 
twice, both times in plur., of ‘inmates’ 
(slave or free: 5. 413: 6. 366). 

757 ἦ καὶ marks re interest: 51. 
314 ἡ κἂν ἐγὼ θαρσοῦσα μᾶλλον ἐς λόγους] 
τοὺς σοὺς ἱκοίμην ; | 

758 The poet has neglected clearness 
on aminor point. The olxeds—sole sur- 
vivor of the four attendants—had fled 
back to Thebes with the news that Laius 
had been slain by robbers (118—123). 
This news came before the trouble with 
the Sphinx began: 126—131. And the 
play supposes an interval of at least 
several days between the death of Laius 


OIAITTOYS TYPANNOS 105 

ΟΕ. Went he in small force, or with many armed followers, 
like a chieftain ? 

Io. Five they were in all,—a herald one of them; and there 
was one Carriage, which bore Laius. 

ΟΕ. Alas! ’Tis now clear indeed.—Who was he who gave 
you these tidings, lady? 

Io. A servant—the sole survivor who came home. 

ΟΕ. Is he haply at hand in the house now? 

Io. No, truly ; so soon as he came thence, and found thee 
reigning in the stead of Laius, he supplicated me, with hand 
laid on mine, that I would send him to the fields, to the pastures 
of the flocks, that he might be far from the sight of this town. 
And I sent him; he was worthy, for a slave, to win e’en a larger 


boon than that. 


ΟΕ. Would, then, that he could return to us without delay ! 


LO: 
OE. 


It is easy: but wherefore dost thou enjoin this? 
I fear, lady, that mine own lips have been unguarded ; 


and therefore am I fain to behold him. 


mann: ὥς γ᾽ Campbell (who cites ὡς from K,=Flor. Abb. 66). 
768 δι᾽ a] δι᾿ 6 Turner. 


ὁ δ᾽, or wd’, 1. 


dy’ 1,: ὁ δέ γ᾽, ὅδ᾽, 





and the election of Oedipus: see on 736. 
Hence κεῖθεν ἦλθε καὶ... εἶδε cannot mean 
that the οἰκεύς, on reaching Thebes, found 
Oedipus already reigning. Nor can we 
suggest that he may have fled from the 
scene of the slaughter before he was 
sure that Laius had been killed: that is 
excluded by 123 and 737. Therefore 
we must understand:—‘when he had 
come thence, and [afterwards] found that 
not only was Laius dead, but you were his 
successor.’ (For the parataxis σέ τε... 
Λάϊόν re see on 673.) I incline to sus- 
pect, however, that Sophocles was here 
thinking of the man as coming back to 
find Oedipus already on the throne, and 
had overlooked the inconsistency. The 
conjecture Λαΐου re δώματα for Adidy τ᾽ 
ὀλωλότα (Wolff) would remove the diffi- 
culty, but seems very improbable. 

760 χειρὸς θιγὼν, marking that the 
ixereia was formal; as when the suppliant 
clasped the knees (ἅπτεσθαι γονάτων). 
Eur. Hee. 850 τύχας σέθεν, | Ἑκάβη, δι᾽ 
οἴκτου χεῖρά θ᾽ ἱκεσίαν ἔχω. 

761 ἀγρούς might be acc. of motion 
to (O. C. 1769 Θήβας δ᾽ ἡμᾶς |...πέμψον); 
but it is better here governed by ἐπὶ: for 
the position of the prep. cp. 734, 1205, 
El. 780 οὔτε νυκτὸς οὔτ᾽ ἐξ Huépas.—vopas: 
on Cithaeron, or near it, 1127. The man 
had formerly served as a shepherd (1039), 


and had then been taken into personal 
attendance on Laius (olkevs). 

762 τοῦδ᾽ ἄποπτος ἄστεως, ‘far from 
the sight of this town’: that is, far from 
the power of seeing it: whereas in 42. 
1487 κτανὼν mpddes |...ἄποπτον ἡμῶν ΞΞ 
‘far from our eyes’: the gen. as after 
words of ‘distance from.’ See Appendix. 

768 οἷ᾽ : the 6 γ᾽ of L (clumsily amend- 
ed to ὁ δέ γ᾽ in other Mss.) prob. came 
from of’, rather than from ὡς or ὡς γ᾽. 
Phil. 583 ot’ ἀνὴρ πένης, ‘for a poor man’: 
Eur. Or. 32 κἀγὼ μετέσχον, οἷα δὴ γυνή, 
φόνου, ‘so far as a woman might.’ ὡς, 
however, is commoner in this limiting 
sense (1118); ofa more often='‘like’ 
(751). Here ota qualifies ἄξιος, imply- 
ing that in strictness the faithful service 
of a slave could not be said to create 
merit, 

764 φέρειν : cp. 590. 

766 πάρεστιν: ‘it is easily done.’ 
Eur. Bacch. 843 ΠΕ. ἐλθών γ᾽ és οἴκους av 
δοκῇ βουλεύσομαι. | AI. ἔξεστι" πάντῃ τό 
y’ ἐμὸν εὐτρεπὲς πάρα. Not, ‘he is here’ 
(nor, ‘he is as good as here,’ as the schol. 
explains): in 769 téerar=‘he will come 
Srom the pastures.’ 

768 δι Gd. The sense is: ‘I fear that 
I have spoken too many words; and on 
account of those words I wish to see him’: 
cp. 744, 324. Not: ‘I fear that my 


106 
IO. ἀλλ᾽ ἵξεται. μέν: 
ΟῚ, 


ἐμοῦ βεβώτος. 


ΣΟΦΟΚΛΕΟΥΣ 


ἀξία δέ που μαθεῖν 
κἀγὼ τά γ᾽ ἐν σοὶ δυσφόρως ἔ ἔχοντ᾽, ἄναξ, 
κοὺ “μὴ στερηθῇς Y: ἐς τοσοῦτον ἐλπίδων 
τῷ γὰρ ἂν καὶ μείζονι 


770 


λέξαιμ' ἂν ἢ σοί, ‘Sud τύχης τοιᾶσδ᾽ ἰών; 
ἐμοὶ πατὴρ μὲν Πολύβος ἣν Κορίνθιος, 


μήτηρ 


δὲ Μερόπη Δωρίς. 


ἠἡγόμην δ᾽ ἀνὴρ 778 


ἀστῶν μέγιστος τῶν ἐκεῖ, πρίν" μοι τύχη 
τοιάδ᾽ ἐπέστη, "θαυμάσαι μὲν ἀξία, 

. σπουδῆς γε μέντοι τῆς ἐμῆς οὐκ ἀξία. 
avnp γὰρ ἐν δείπνοις be ὑπερπλησθεὶς μέθῃ 





καλεῖ παρ᾽ οἴνῳ, πλαστὸς ὡς εἴην πατρί. 


780 


κἀγὼ βαρυνθεὶς τὴν μὲν οὖσαν ἡμέραν 

μόλις κατέσχον, Oarépa δ᾽ ἰὼν πέλας 
μητρὸς πατρός τ᾽ ἤλεγχον" οἱ δὲ δυσφόρως 

τοὔνειδος ἤγον τῷ βεθέντι τὸν λόγον. 


καγὼ τὰ μὲν κείνοιν ἐτερπόμην, ὅμως δ᾽ 


ὑφεῖρπε γὰρ πολύ. 


3 ΦΙΩΝ ay 
ἔκνιζέ μ᾽ αεὶ τοῦθ᾽. 


785 


λάθρᾳ δὲ μητρὸς καὶ πατρὸς πορεύομαι 
Πυθώδε, καί μ᾽ ὁ Φοῖβος ὧν μὲν ἱκόμην 


779 μέθηι L 1st hand, changed by an early hand to μέθης. 


The latter prevails in 





words have given me only too much 
cause to desire his presence.’ A comma 
after μοι is here conducive to clearness. 
770 κἀγὼ and ποὺ express the wife’s 
sense that he should speak to her as to 
a second self.—éy ool=within thee, in 
thy mind (not ‘in thy case’): cp. ἐν with 
the reflexive pronouns, Plat. 7heaet. 192 Ὁ 
ἐν ἐμαυτῷ μεμνημένος : Crat. 384 A προσ- 
ποιούμενός τι αὐτὸς ἐν ἑαυτῷ διανοεῖσθαι. 
771 ἐς τοσοῦτον ἐλπίδων : Isocr. or. 
8 § 31 εἰς τοῦτο γάρ τινες ἀνοίας ἐληλύ- 
θασιν: Ar. Nub. 832 σὺ δ᾽ ἐς τοσοῦτον 
τῶν μανιῶν ἐλήλυθας. ‘The plural οἵ ἐλπίς 
is rare as=anxious forebodings: but cp. 
87. 
Ἶ 772 μείζονι : strictly, ‘more important’: 
cp. Dem. or. 19 ὃ 248 ἀντὶ... τῆς πόλεως 
τὴν Φιλίππου ξενίαν καὶ φιλίαν πολλῷ μεί- 
ἕζονα ἡγήσατο αὐτῷ καὶ λυσιτελεστέραν 
(alluding to Ant. 182 καὶ μείζον᾽ ὅστις 
ἀντὶ τῆς αὑτοῦ πάτρας] φίλον νομίζει) : 
Ant. 637 ovdels...yapos | μεέξων φέρεσ- 
θαι σοῦ καλῶς ἡγουμένου, no marriage can 
be a greater prize than your good guid- 


ance. The καὶ with λέξαιμ᾽ ἂν :—could 
I speak? Lysias or. 12 ὃ 29 παρὰ τοῦ 
ποτε καὶ λήψεσθε δίκην ; from whom wi// 
you ever exact satisfaction? 

773 ἰών, present, not future, part. : 
Ant. 742 διὰ δίκης ἰὼν πατρί. Xen. An. 
3. 2. 8 διὰ φιλίας ἰέναι. 

775 The epithet ‘Dorian’ carries ho- 
nour: Meropé was of the ancient stock, 
claiming descent from Dorus son of Hellen, 
who settled in the region between Oeta 
and Parnassus. The Scholiast’s comment, 
ΠΠελοποννησιακή, forgets that the Theban 
story is laid in times before the Dorian 
conquest. 

776 πρίν pov.. «ἐπέστη (1) πρίν with 
infin. =our ‘before,’ whether the sentence 
is affirmative or negative: ἦλθε πρὶν κλη- 
θῆναι, οὐκ ἦλθε πρὶν κληθῆναι. (2) πρίν 
with a /fimite mood (indic., subj., or opt.) 
=our ‘until’ in megative sentences. Thus 
οὐκ ἦλθε πρὶν “ἐκλήθη differs from οὐκ 
ἦλθε πρὶν κληθῆναι by implying that at 
last he was called, and then came. Here, 
the form of the sentence is affirmative 


ΟΙΔΙΠΟΥ ΣῪ TYPANNOZ 107 

Io. Nay, he shall come. But I too, methinks, have a claim 
to learn what lies heavy on thy heart, my king. 

Or. Yea, and it shall not be kept from thee, now that my 
forebodings have advanced so far. Who, indeed, is more to me 
than thou, to whom I should speak in passing through such a 
fortune as this? 

My father was Polybus of Corinth,—my mother, the Dorian 
Meropé; and I was held the first of all the folk in that town, 
until a chance befell me, worthy, indeed, of wonder, though not 
worthy of mine own heat concerning it. At a banquet, a man 
full of wine cast it at me in his cups that I was not the true 
son of my sire. And I, vexed, restrained myself for that day 
as best I might; but on the next I went to my mother and 
father, and questioned them; and they were wroth for the taunt 
with him who had let that word fly. So on their part I had 
comfort; yet was this thing ever rankling in my heart; for it 
still crept abroad with strong rumour. And, unknown to mother 
or father, I went to Delphi; and Phoebus sent me forth 


the later Mss. (but μέθη I). 





(ἠγόμην), and ἕως would therefore be more 
strictly correct. But the thought is nega- 
tive (‘nothing happened to disturb me’) ; 
hence πρίν. So Thuc. 3. 29 rods...’ A0n- 
vatous λανθάνουσι (-ε οὐχ ὁρῶνται ὑπὸ τῶν 
᾽Α.) πρὶν δὴ τῇ Δήλῳ ἔσχον. Cp. White- 
law in Zrans. Cam. Phil. Soc. 1886,p. 26. 
-- ἐπέστη: a verb often used of enemies 
suddenly coming upon one: Isocr. or. 
9 ὃ 58 μικροῦ δεῖν ἔλαθεν αὐτὸν ἐπὶ τὸ 
βασίλειον ἐπιστάς: Her. 4. 203 ἐπὶ τῇ 
Κυρηναίων πόλι ἐπέστησαν. 

779 ὑπερπλησθεὶς μέθῃ, lit., intoxi- 
cated by drinking (caus. dat.): μέθη al- 
ways=‘drinking’ (not ‘strong wine’): 
cp. Her. 5. 20 καλῶς ἔχοντας... μέθης 
(‘having had enough of drinking’). For 
the dat. cp. Aesch. Fers. 132 λέκτρα... 
πίμπλαται δακρύωασιν. 

780 παρ᾽ οἴνῳ: Plut. Mor. 143 C τοὺς 
τῇ λύρᾳ χρωμένους παρ᾽ οἶνον. Thuc.6.28 
μετὰ παιδιᾶς καὶ οἴνου.----πλαστὸς ὡς εἴην 
instead οἵ πλαστόν, as if preceded by 
ὀνειδίζει μοι instead of καλεῖ pe. Some- 
what similarly ὀνομάξω-ε λέγω, as Plat. 
Prot. 311 E cogicriv...dvoudgouct...7dv 
ἄνδρα elva. πλαστὸς, ‘feigned (in 
speech),’ ‘falsely called a son,’ πατρί, 
‘for my father,’ 2.6. to deceive him. Eur. 
Alc. 639 μαστῷ γυναικὸς offs ὑπεβλήθην 
λάθρᾳ. whence ὑποβολιμαῖος = νόθος. 

782 κατέσχον, sc. ἐμαυτόν. In clas- 


sical Attic this use occurs only here: in 
later Greek it recurs, as Plut. Artaxerxes 
§ 15 εἶπεν οὖν μὴ κατασχών. ὑμεῖς μέν 
κιτιλ. Cp. ἔχε, σχές, ἐπίσχες (‘stop’), 
in Plat., Dem., etc. ° 

784 τῷ μεθέντι : the reproach was like 
a random missile: Menander fr. 88 οὔτ᾽ 
ἐκ χερὸς μεθέντα καρτερὸν λίθον | ῥᾷον 
κατασχεῖν, οὔτ᾽ ἀπὸ γλώσσης λόγον. The 
dat., because δυσφόρως τοὔνειδος ἦγον = 
ὠργίζοντο ἕνεκα τοῦ ὀνείδους. 

785 ὅμως δ᾽ : cp. 791, and n. on 29. 

786 ὑφεῖρπε γὰρ πολύ: so ὑφέρπειν of 
malicious rumour, Aesch. Ag. 450 φθο- 
νερὸν δ᾽ ὑπ᾽ ἄλγος ἕρπει | προδίκοις ᾿Ατρεί- 
δαις. Libanius 784 A (quoted by Mus- 
grave) πολὺς τοιοῦτος ὑφεῖρπε λόγος (per- 
haps suggested by this passage). Pind. 
Isthm. 3. 58 τοῦτο yap ἀθάνατον φωνᾶεν 
ἕρπει, | εἴ τις εὖ εἴπῃ τι. Cp. Ant. 700 
τοιάδ᾽ ἐρεμνὴ σῖγ᾽ ἐπέρχεται φάτις. For 
πολύ cp. Ο. C. 517 τὸ πολύ τοι καὶ μη- 
δαμὰ λῆγον, that strong rumour which 
is in no wise failing: 26. 305 πολὺ...τὸ 
σὸν ὄνομα | διήκει πάντας. This version 
also agrees best with 775, which implies 
that the incident had altered his popular 
repute. We might render: ‘it was ever 
recurring to my mind with force’: but 
this (a) is a repetition: (4) is less suited 
to πολύ, which implies diffusion. 

788 ὧν ἱκόμην ἄτιμον -- ἄτιμον τούτων 


τοῦ 


Ἂ \ ‘\ 4 
AGE δεινὰ Poe δύστηνα 


ΣΟΦΟΚΛΕΟΥΣ 


ἄτιμον ἐξέπεμψεν, ἄλλα δ᾽ ἄθλια 


ὡς μητρὶ μὲν χρείη με μιχθῆναι, γένος δ᾽ 


ἄτλητον ἀνθρώποισι δηλώσοιμ᾽ ὁρᾶν, 
φονεὺς δ᾽ ἐσοίμην τοῦ φυτεύσαντος πατρός. 
κἀγὼ 'πακούσας ταῦτα τὴν Κορινθίαν 


--- 


ἄστροις τὸ λοιπὸν ἐκμετρούμενος χθόνα 


ἔφευγον, ἔνθα μήποτ᾽ ὀψοίμην κακῶν 

χρησμῶν ὀνείδη τῶν ἐμῶν τελούμενα. 

στείχων δ᾽ ἱκνοῦμαι τούσδε τοὺς χώρους ἐν οἷς 
σὺ τὸν τύραννον τοῦτον ὄλλυσθαι λέγεις. 


καί σοι, γύναι, τἀληθὲς ἐξερώ. 


ὅτ᾽ ἢ κελεύθου τῆσδ᾽ ὁδοιπορῶν πέλας, 
ἐνταῦθά μοι κῆρυξ τε κἀπὶ πωλικῆς 


ἀνὴρ ἀπήνης ἐμβεβώς, οἷον σὺ φής, 


789 ἄλλα θ᾽ ἄθλια L: 
would read ἄλλα δ᾽ ἀθλίῳ. 


the rst hand had written ἀθλίω, ἄλλα δ᾽ ἄθλια τ. 
790 προὐφάνη MSS. 
gloss προέδειξε in E may be a reminiscence of such a reading. 
too, that προὐφάνην is cited by Campbell from M?,= 
ne L, the ¢ after ἡ almost erased. Cp. on 555. 


* a poudnvev λέγων, 790 
795 

τριπλῆς 800 
eihanve Hens (ihe 


It may be remarked, 
= Ambros. L. 39.) 791 χρεῖ 
797 τελούμενα. In L there has 


been an erasure at and after a, and there are traces of an accent above the second e. 





a ἱκόμην, not graced in respect of those 
things (responses) for which I had come: 
Eur. Andr. 1014 ἄτιμον ὀργάναν χέρα τεκ- 
τοσύνας, not rewarded for its skill. Ford 
ἱκόμην (cogn. accus. denoting the errand, 
like ἔρχομαι ἀγγελίαν) Cp. 1005 τοῦτ᾽ 
ἀφικόμην: O. C. 1291 ἃ δ᾽ ἦλθον... θέλω 
λέξαι: Ar. Pl. 966 ὅ τι μάλιστ᾽ ἐλήλυθας : 
Plat. Prot. 310 E ἀλλ᾽ αὐτὰ ταῦτα καὶ νῦν 
ἥκω παρὰ σέ (where the acc. is cogn. to 
ἥκω, not object to the following διαλεχ- 
Θῃ9). 

790 προὔφηνεν, suggested by Herm., 
has been adopted by several recent edi- 
tors. Cp. Herod. 1. 210 τῷ δὲ ὁ δαίμων 
προέφαινε, and so 3. 65, 7-37: Plut. Dem. 
§ 19 ἐν οἷς ἥ τε Πυθία δεινὰ προὔφαινε μαν- 
τεύματα καὶ ὁ χρησμὸς ἤδετο: Camill. ὃ 4 
(a man who pretended to μαντική) λόγια 
προὔφαινεν ἀπόρρητα: Dem. or. 21 § 54 
τοῖς ἐφ᾽ ἑκάστης μαντείας προφαινομένοις 
θεοῖς, the gods announced (as claiming 
sacrifice) on each reference to the oracle. 
Yet the fact that rpogalvecy was thus a 
vox sollennis for oracular utterance would 
not suffice to warrant the adoption of 
προὔφηνεν, if the προὐφάνη of the Mss. 
seemed defensible. προὐφάνη λέγων 


would mean, ‘came into view, , telling’: cp. 
above, 395, and £7. 1285 νῦν δ᾽ ἔχω σε" 
προὐφάνης δὲ | φιλτάταν ἔχων πρόσοψιν. 
It might apply to the sudden appearance of 
a beacon (cp. ὁ φρυκτὸς ἀγγέλλων πρέπει, 
Aesch. Ag. 30): but, in reference to 
the god speaking through the oracle, it 
could only mean, by a strained metaphor, 
‘flashed on me with the message,’ 2.6. 
announced it with startling suddenness 
and clearness. The difficulty of conceiv- 
ing Sophocles to have written thus is to 
me so great that the sfeczal appropriate- 
ness of προὔφηνεν turns the scale. 

791 f. γένος 8’: see on 29.—épav with 
ἄτλητον, which, thus defined, is in con- 
trast with δηλώσοιμ᾽ : : he was to show 
men what they could not bear to look 
upon. 

794 ff. ἐπακούσας (708), ‘having 
given ear ’—with the attention of silent 
horror.—tiv Κορινθίαν : ‘ Henceforth 
measuring from afar (ἐκμετρούμενος) by 
the stars the region of Corinth, I went 
my way into exile, to some place where 
I should not see fulfilled the dishonours 
of [=foretold by] my evil oracles.’ do- 


τροις ékpetpotpevos: 2.6. visiting it no 


ΟἸΔΊΠΟΥΣ. ΤΥΡΆΝΝΟΣ 109 
disappointed of that knowledge for which I came, but in his 
response set forth other things, full of sorrow and terror and 
woe; even that I was fated to defile my mother’s bed; and 
that I should show unto men a brood which they could not 
endure to behold; and that I should be the slayer of the sire 
who begat me. 

And I, when 1 had listened to this, turned to flight from the 
land of Corinth, thenceforth wotting*of its region by the stars 
alone, to some spot where I should never see fulfilment of the 
infamies foretold in mine evil doom. And on my way I came 
to the regions in which thou sayest that this prince perished. 
Now, lady, I will tell thee the truth. When in my journey I 
was near to those three roads, there met me a herald, and a 
man seated in a carriage drawn by colts, as thou hast described ; 


The 1st hand had written τελουμένων, which the first corrector (S) altered.—Some 
later Mss. (B, V, ν 3, V4) add γ᾽ to χρησμῶν. 800 This verse does not stand 
in the text of L, but has been added in the margin by a later hand. With regard to 
the age of the hand, Mr E. M. Thompson observes :—‘ This writing is of the style 
which appears in the latter part of the thirteenth century, and continues with little 





more, but only thinking of it as a dis- 
tant land that lies beneath the stars in 
this or that quarter of the heavens. 
Schneidewin cp. Aelian Azst. Anim. 
(περὶ ζῴων ἰδιότητος) 7. 48 ἧκε δ᾽ οὖν 
(᾿Ανδροκλῆθ) ἐς τὴν Λιβύην καὶ τὰς μὲν 
πόλεις ἀπελίμπανε καὶ τοῦτο δὴ τὸ λε- 
γόμενον ἄστροις αὐτὰς ἐσημαίνετο, 
προΐει δὲ ἐς τὴν ἐρήμην : ‘proceeded to 
leave the cities, and, as the saying ts, 
knew their places only by the stars, and 
went on into the desert.” Wunder quotes 
Medea’s words in Valer. Flacc. 7. 478 
quando hic aberis, dic, quaeso, profundi 
Quod caeli spectabo latus? ἔφευγον might 
share with éxperp. the government of τὴν 
Kop. χθόνα, but is best taken absolutely. 
Sense, not grammar, forbids the version :— 
‘I went into exile from the Corinthian 
land (τὴν Κορινθίαν), thenceforth mea- 
suring my way on earth (χθόνα) dy the 
stars. Phrases like ὕπαστρον... μῆχαρ 
ὁρίζομαι γάμου δύσφρονος | φυγᾷ (Aesch. 
Suppl. 395), ἄστροις τεκμαίρεσθαι ὁδόν (Lu- 
cian Jcaromenippus ὃ 1), are borrowed 
from voyages in which the sailor has no 
guides but the stars. Such phrases could 
be used figuratively only of a journey 
through deserts: as Hesych. explains the 
proverb ἄστροις σημειοῦσθαι: μακρὰν καὶ 
ἐρήμην ὁδὸν βαδίζειν" ἡ δὲ μεταφορὰ 
ἀπὸ τῶν πλεόντων. 

796 ἔνθα-- ἐκεῖσε ἔνθα, as in Ph. 1466. 


φεύγω ἔνθα μὴ ὄψομαιτε “1 fly to such a 
place that I shall not see’; the relative 
clause expresses purpose, and μή gives a 
generic force: cp. 1412: Az. 659: £7. 380, 
436: Trach. 800. ‘Here, the secondary 
tense ἔφευγον permits ὀψοίμην. Remark, 
however, that in such relative clauses (of 
purpose or result) the fut. indic. is usually 
retained, even where the optat. is admis- 
sible. A rare exception is Plat. Rep. 
416 C φαίη ἄν τις...δεῖν.. οὐσίαν τοιαύτην 
αὐτοῖς παρεσκευάσθαι, ἥτις μήτε...παύσοι 
κιτιλ.: where παύσοι (if sound) is pro- 
bably due to φαίη ἄν (see on O.C. 778) 
rather than to δεῖν as Ξξ ὅτι ἔδει. 

800 kal σοι...τριπλῆς. The hand 
which added this verse in the margin of 
L seems to be ‘as early as the beginning 
of the fourteenth century’ (Mr E. M. 
Thompson, /ztrod. to Facsimile of Laur. 
MS.). The verse is in A (13th cent.) and 
all our other Mss. To eject the verse, 
as Dindorf and Nauck have done, is 
utterly unwarrantable. It has a fine 
dramatic force. Oedipus is now at the 
critical point: he will hide nothing of 
the truth from her who is nearest to 
him. It is part of his character that 
his earnest desire to know the ¢rwth never 
flinches: cp. 1170. 

8O2 κηρυξ τε, not κῆρύξ τε: see 
Chandler, Accentuation § 971. 

808 ἀπήνης: see on 753.—olov ad- 


ZOPOKAEOYS 


ξυνηντίαζον: κἀξ ὁδοῦ μ᾽ ὃ θ᾽ ἡγεμὼν 


αὐτός θ᾽ ὁ πρέσβυς πρὸς βίαν ἠλαυνέτην. 
Ν 


805 


A κἀγὼ TOV ἐκτρέποντα, TOV τροχηλάτην, 

τ: ,ὕ 32 , > ε , ε Ca 
παίω dv ὀργῆς: καί μ᾽ ὁ πρέσβυς ὡς ὁρᾷ, 
ὄχου, παραστείχοντα τηρήσας, μέσον 
κάρα διπλοῖς κέντροισί μου καθίκετο. 


> Ν ¥ > » > \ ‘4 
ov μὴν tonv y έἔτεισεν, alka συντόμως 


SIO 


σκήπτρῳ τυπεὶς EK τῆσδε χειρὸς ὕπτιος 
μέσης ἀπήνης εὐθὺς ἐκκυλίνδεται" 


’ Ν A 4 
κτείνω δὲ τοὺς ξύμπαντας. 


εἰ δὲ τῷ ξένῳ 


, Z oh , 
τούτῳ προσῆκει Aaiw TL συγγενές, 


τίς τοῦδε “νῦν ἔστ᾽ ἀνδρὸς ἀθλιώτερος ; 
> / ” xa ’, 3 τ 
ἐχθροδαίμων μᾶλλον ἂν γένοιτ᾽ ἀνήρ; 
ὃν μὴ ἕένων ἔξεστι μηδ᾽ ἀστῶν ὅτινι 


4 
τὶς 
ἧς ἃ 


815 


δόμοις δέχεσθαι, μηδὲ προσφωνεῖν τινα, 


variation for some fifty years or more. 


The line may therefore, without much 


hesitation, be placed as early as the beginning of the fourteenth century.’ (Intro- 
duction to the Facsimile of the Laur. Ms. of Sophocles, p. 11.) All the later mss. 


have this verse in the text. 


808 ὄχου MSS.: ὄχον Schaefer: ὄχους Déderlein. 


814 Λαίῳ Mss.: Λαΐου Bothe. Blaydes suggests, εἰ δέ τι ξένῳ [ τούτῳ προσήκει 
Λαίῳ τε συγγενές: Heimsoeth, εἰ δὲ τῷ ξένῳ | τούτῳ προσῆν καὶ Λαΐῳ τι συγγενές. 
815 τίς τοῦδέ γ᾽ ἀνδρὸσ νῦν ἔστ᾽ ἀθλιώτεροσ L. The νῦν is almost erased, and over 


it a late hand has written ἄλλωσ, probably meant for ἄλλοσ. 


The later mss. either 





verbial neut.=ds, referring to Iocasta’s 
whole description; not acc. masc., re- 
ferring to the person of Laius as described 
by her. 

804—812 The κῆρυξ is, I think, 
identical with the ἡγεμών, and distinct 
from the τροχηλάτης. I understand the 
scene thus. Oedipus was coming down 
the steep narrow road when he met the 
herald (to be known for such by his stave, 
κηρύκειον) walking in front of the carriage 
(ἡγεμών). The herald rudely bade him 
stand aside; and Laius, from the car- 
riage, gave a like command. (With the 
imperfect ἠλαυνέτην, ‘were for driving,’ 
πρὸς βίαν need not mean more than a 
threat or gesture.) The driver (τροχη- 
λάτης), who was walking at his horses’ 
heads up the hill, then did his lord’s 
bidding by actually jostling the wayfarer 
(ἐκτρέποντα). Oedipus, who had forborne 
to strike the sacred herald, now struck the 
driver ; in another moment, while passing 
the carriage, he was himself struck on 
the head by Laius. He dashed Laius 
from the carriage; the herald, turning 


back, came to the rescue; and Oedipus 
slew Laius, herald, driver, and one of two 
servants who had been walking by or 
behind the carriage; the other servant 
(unperceived by Oedipus) escaped to 
Thebes with the news. 

808 ὄχου: ‘from the chariot—having 
watched for the moment when I was 
passing—he came down on me, full on my 
head (μέσον κάρα acc. of part affected), 
with the double goad.’ The gen. ὄχου 
marks the point from which the action 
sets out, and is essentially like τᾶς πολυ- 
χρύσου  ἸΤυθῶνος...ἔβας v. 151: cp. Od. 
21. 142 ὄρνυσθε... | ἀρξάμενοι τοῦ χώρου 
ὅθεν τέ περ οἰνοχοεύει, from the place. In 
prose we should have had ἀπ᾽ éxov. As 
the verb here involves motion, we cannot 
compare such a gen. as ἷζεν... τοίχου τοῦ 
ἑτέρου (Z/. 9. 219), where, if any prep. 
were supplied, it would be πρός.---τηρή- 
σας: [Dem.] or. 53 § 17 (contemporary 
with Dem.) τηρήσας με ἀνιόντα ἐκ Πει- 
ραιῶς ὀψὲ... «ἁρπάζει. 

809 καθίκετο governs pov, which 
μέσον Kapa defines: Plut. Anion. § 12 


ΟἸΔΙΠΟΥΣ. TYPANNOZ ἘΠῚ 


and he who was in front, and the old man himself, were for 
thrusting me rudely from the path. Then, in anger, I struck 
him who pushed me aside—the driver; and the old man, seeing 
it, watched the moment when I was passing, and, from the 
carriage, brought his goad with two teeth down full upon my 
head. Yet was he paid with interest; by one swift blow from 
the staff in this hand he was rolled right out of the carriage, on 


his back; and I slew every man of them. 
But if this stranger had any tie of kinship with Laius, who 


is now more wretched than the man before thee? 
could prove more hated of heaven? 


What mortal 
Whom no stranger, no 


citizen, is allowed to receive in his house; whom it is unlawful 


agree with L, or give rls τοῦδέ γ᾽ ἀνδρός ἐστιν ἀθλιώτερος (as A). 
the latter, and so Campbell (with τἀνδρὸς for γ᾽ dvdpés). 


that any one accost; 


Kennedy adopts 
But νῦν seems forcible 


here. Dindorf proposed νῦν ἔτ᾽ (which Wecklein receives); he afterwards wrote tis 


τοῦδ᾽ ἀκούειν ἀνδρὸς ἀθλιώτερος : but now rejects the verse. 
I would merely transpose ἀνδρὸς and omit γ᾽, which might easily 


(to go with γένοιτ᾽). 


Bellermann writes viv ἂν 


have been intruded, for metre’s sake, when the proper order of words had been de- 


ranged. 


817 ¢...7rwa L. Schaefer wrote ὃν. τινὰ (so that ἔξεστι should be abso- 





σκύτεσι λασίοις...καθικνούμενοι τῶν ἐν- 
τυγχανόντων : Lucian Symp. ὃ 16 τάχα 
δ᾽ ἄν τινος καθίκετο τῇ βακτηρίᾳ: Lcaro- 
menippus § 24 σφόδρα ἡμῶν ὁ πέρυσι 
χειμὼν καθίκετο. This verb takes accus. 
only as=to reach, lit. or fig. (as 71. 14. 
104 μάλα πώς με καθίκεο θυμόν) .---διπλοῖς 
κέντροισι: a stick armed at the end with 
two points, used in driving. Cp. 21. 23. 
387 (horses)...dvev κέντροιο θέοντες. The 
τροχηλάτης had left it in the carriage when 
he got out to walk up the hill. 

810 ov μὴν ἴσην y: not merely an 
even penalty (cp. τὴν ὁμοίαν ἀποδιδόναι, 
par pari referre): Thuc. 1. 35 οὐχ ὁμοία 
ἡ ἀλλοτρίωσις, the renunciation of such 
an alliance is more serious.—traceyv. 
'«τείσω, ἔτεισα, ἐτείσθην (not τίσω, etc.) 
were the Attic spellings of the poet’s age: 
see the epigraphic evidence in Meister- 
hans, Gramm. Ὁ. 88.-- συντόμως, in a 
way which made short work: cp. Thuc. 
ἡ. 42 ἠπείγετο ἐπιθέσθαι τῇ πείρᾳ καί οἱ 
ξυντομωτάτην ἡγεῖτο διαπολέμησιν, the 
quickest way of deciding the war: Her. 
5. 17 ἔστι δὲ σύντομος κάρτα (sc. ὁδός), 
there is a short cut. The conject. συν- 
τόνως (77. 923 συντόνῳ χερί) would 
efface the grim irony. 

812 μέσης implies that a moment be- 
fore he had seemed firmly seated: ‘right 
out of the carriage.’ Eur. Cycl. 7 ἱτέαν 
μέσην θενών, striking Κεἰ on the shield: 


1 7. 1385 νηὸς δ᾽ ἐκ μέσης ἐφθέγξατο | 
βοή τις, from within the ship itself: 22. 
965 ἄρκυν els μέσην, right into the net. 

814 εἰ συγγενές τι τῷ Λαΐῳ if any tie 
with Laius προσήκει τούτῳ τῷ ξένῳ ὁεέ- 
longs to this stranger. συγγενής can take 
either dat. (akin to) or gen. (kin of): and 
here several editors give Aatov. But the 
dat. Aatw, making it verbally possible 
to identify the ξένος with Laius, suits the 
complex suggestiveness with which the 
language of this drama is often contrived : 
cp. τῶν in 1167. Again, τῷ ξένῳ τούτῳ 
might apply to Oedipus himself (452). 
Had we tt without συγγενές, Λαΐου (part. 
gen.) would then be zecessary. The con- 
structions of προσήκειν are (1) προσήκω 
τινί, Lam related to: (2) προσήκει μοί τινος, 
I have a right in, or tie with: (3) προσήκει 
μοί τι, it belongs tome. Here it is (3). 

817 ὃν...τινι. The MS. ᾧ...τινα must 
be rendered, with Hermann: ‘to whom it 
is not allowed that any one should receive 
(him)’: but the words would naturally 
mean: ‘to whom it is not allowed to re- 
ceive any one.’ In 376, where σε...γ᾽ 
ἐμοῦ is certain, all our Mss. have με...γε 
σοῦ: much more might the cases have 
been shifted here. 

818 ΣΦ. μηδὲ... τινα, sc. ἔξεστι, abso- 
lutely: nor zs 7z¢ ἠατυγμί that anyone 
should speak to him.——o@etv 8’: the posi- 
tive δεῖ must be evolved from the negative 


fone 


>) 9 » 
ὠθεῖν ὃ ἀπ᾽ οἴκων. 
xX 


ZOPOKAEOYS 


καὶ τάδ᾽ οὔτις ἄλλος ἣν 
ἢ ᾽γὼ or ἐμαυτῷ τάσδ᾽ ἀρὰς O προστιθείς. 


820 


λέχη δὲ τοῦ «θανόντος ἐν χεροῖν ἐμαῖν 


ραίνω, δι ὧνπερ WAET . 


ap ἔφυν κακός; 


dp οὐχὶ πᾶς ἄναγνος ; εἴ με χρὴ φυγεῖν, 
καί μοι φυγόντι μῆστι τοὺς ἐμοὺς ἰδεῖν, 


* μηδ᾽ ἐμβατεύειν πατρίδος, ἢ γάμοις με δεῖ 


825 


μητρὸς ζυγῆναι καὶ πατέρα κατακτανεῖν 


Πόλυβον, ὃς ἐξέφυσε. κἀξέθρεψέ με. 


ap’ OUK ἀπ᾽ ὠμοῦ ταῦτα δαίμονός τις ἂν 
κρίνων ἐπ᾽ ἀνδρὶ τῷδ᾽ ἂν ὀρθοίη λόγον ; 


μὴ δῆτα, μὴ δῆτ᾽, ὦ θεῶν ἁγνὸν σέβας, 


ἴδοιμι ταύτην ἡμέραν, ἀλλ᾽ ἐκ βροτῶν 
βαίην ἄφαντος πρόσθεν ἢ rowdy? ἰδεῖν 


τὸν ἄνδρα τὸν 


: κηλῖδ᾽ ἐμαυτῷ συμφορᾶς ἀφιγμένην. 
: ἡμῖν μέν, ὦναξ, ταῦτ᾽ ὀκνήρ᾽" 
πρὸς τοῦ παρόντος ἐκμάθῃς, ἔχ᾽ ἐλπίδα. 
καὶ μὴν τοσοῦτόν γ᾽ ἐστί μοι τῆς ἐλπίδος, 
Bornpa προσμεῖναι μόνον. 
πεφασμένου δὲ τίς oF ἢ προθυμία; 


a > 
ἕως αν ουν 


Ol. ἐγὼ διδάξω δ ἢν γὰρ εὑρεθῇ λέγων 


σοὶ ταῦτ᾽, ἔγωγ᾽ ἂν ἐκπεφευγοίην πάθος. 


840 


10. ποῖον δέ μου περισσὸν ἤκουσας λόγον ; 


lute): Dindorf, ὃν... τινι, 
of τινα). 824 μῆστι. 


ἴο μήτε. 


Nauck proposes ef μὴ ξένων... τινι} 
The rst hand in L wrote μήστι, which an early hand changed 
The latter is in most of the later Mss. (with yp. μή ᾽στι in some, as T). 


.. προσφωνεῖν ἐμέ (instead 


825 μηδ᾽ ἐμβατεύειν) L has μήτ᾽, made by an early hand from μῆστ, as Campbell 


thinks, and as seems most probable; or, as Diibner thinks, from μή μ᾽. 


Dindorf’s 





οὐκ ἔξεστι: cp. El. 71 καὶ μή μ᾽ ἄτιμον 
τῆσδ᾽ ἀποστείλητε γῆς | ἀλλ᾽ ἀρχέπλουτον 
(sc. καταστήσατε). See above, 241.---καὶ 
τάδ᾽. And these things—these curses— 
none but I laid on myself. And as the 
thought proceeds, the speaker repeats 
τάδε in a more precise and emphatic 
form: cp. Plat. Rep. 606 B ἐκεῖνο κερδαί- 
νειν ἡγεῖται, THY Noovip. 

821 ἐν χεροῖν, not, ‘in their embrace,’ 
but, ‘by their agency’: //. 22. 426 ws 
ὄφελεν θανέειν ἐν , χερσὶν ἐμῇσιν. 

822 f. ap —dp οὐχὶ. Where ἄρα is 
equivalent in sense to ap’ ov, this is be- 
cause it means, ‘are you Satisfied that it 
is so?’ 2.6. ‘is it not abundantly clear?’ 


(51. 614). Here, the transition from dpa 
to dp’ οὐχὶ is from bitter irony to > despair 
ing earnest. 

827 Πόλυβον. Wunder fee others 
think this verse spurious. But it is, in 
fact, of essential moment to the develop- 
ment of the plot. Ocedipus fears that he 
has slain Laius, but does not yet dream 
that Laius was his father. This verse 
accentuates the point at which his belief 
now stands, and so prepares us for the 
next stage of discovery. A few Mss. give 
ἐξέθρεψε κἀξέφυσε: but the Homeric 
πρότερον ὕστερον (Od. 12. 134 θρέψασα 
τεκοῦσά Te) seems out of place here just 
because it throws a less mafura/ emphasis 


OIAITOYS ΤΥΡΑΝΝΌΣ 113 
whom all must repel from their homes! And this—this curse 
—was laid on me by no mouth but mine own! And I pollute 
the bed of the slain man with the hands by which he perished. 
Say, am I vile?) Oh, am I not utterly unclean ?—seeing that 
I must be banished, and in banishment see not mine own 
people, nor set foot in mine own land, or else be joined in 
wedlock to my mother, and slay my sire, even Polybus, who 


begat and reared me. 


Then would not he speak aright of Oedipus, who judged these 


things sent by some cruel power above man? 
ye pure and awful gods, that I should see that day! 


Forbid, forbid, 
No, may 


I be swept from among men, ere I behold myself visited with 


the brand of such a doom! 


CH. 


To us, indeed, these things, O king, are fraught with 


fear; yet have hope, until at least thou hast gained full know- 
ledge from him who saw the deed. 


OE. 


Hope, in truth, rests with me thus far alone; I can 


await the man summoned from the pastures. 
Io. And when he has appeared—what wouldst thou have 


of him? 


ΟΕ. I will tell thee. 


If his story be found to tally with 


thine, I, at least, shall stand clear of disaster. 
Io. And what of special note didst thou hear from me? 


μηδ᾽ is clearly right. 


The alternatives would be to read μῆστι τοὺς ἐμοὺς ἰδεῖν, | μῆστ᾽ 


ἐμβατεύειν, which does not seem Sophoclean, or μήτε... μήτ᾽, supplying ἔξεστι (as Elmsley 


suggested), which is much worse. 


verse.—eiépuce κἀξέθρεψε 1,: ἐξέθρεψε κἀξέφυσε r. 


827 Wunder, Dindorf, and Nauck reject this 


840 πάθος MSS. : ἄγος has been 





on ἐξέφυσε. 

829 ἐπ᾽ ἀνδρὶ τῷδε with ὀρθοίη λόγον, 
speak truly in my case. Isaeus or. 8 
§ 1 ἐπὶ τοῖς τοιούτοις, ὦ ἄνδρες, ἀνάγκη 
ἐστὶ χαλεπῶς φέρειν, in such cases. //. 


19. 181 σὺ δ᾽ ἔπειτα δικαιότερος καὶ ἐπ᾿ 


ἄλλῳ | ἔσσεαι, in another’s case. 

832 f. τοιάνδε, not τοιᾶσδε : cp. 533. 
π- κηλῖδα: cp. ἄγος 1426: O. C. 1133 
κηλὶς κακῶν. 

884 δ᾽ οὖν. So where the desponding 
φύλαξ hopes for the best, Aesch. 4g. 34, 
γένοιτο δ᾽ οὖν K.T.r. 

835 τοῦ παρόντος, imperf. part.,= 
ἐκείνου ds παρῆν : Dem. or. 19 ὃ 129 οἱ 
συμπρεσβεύοντες καὶ παρόντες κατα- 
μαρτυρήσουσιν, ζ. 6. of συνεπρέσβενον καὶ 
παρῆσαν. 

886 τῆς ἐλπίδος. The art. is due to 
the mention of ἐλπίδα just b-fore, but its 
force is not precisely, ‘the hope of which 
you speak.’ Rather ἐλπίδα is ‘some hope,’ 
τῆς ἐλπίδος is ‘hope’ in the abstract: 


ἘΘΘῚ 


For συμφορᾶς, see on 90. ~ 


cp. Dem. or. 19 § 88 ἡλίκα πᾶσιν ἀνθρώ- 
ποις ἀγαθὰ ἐκ τῆς εἰρήνης γίγνεται, 2.6. 
‘from peace,’ not ‘¢he peace.’ 

838 πεφασμένου, sc. αὐτοῦ : gen.absol. 
Fil. 1344 τελουμένων εἴποιμ᾽ dv, when (our 
plans) are being accomplished. 

830 πάθος, a calamity,—viz. that of 
being proved blood-guilty. The conjec- 
ture ἄγος is specious. But πάθος shows 
a finer touch; it is the euphemism of a 
shrinking mind (like the phrase ἤν τι 
τὺ for θάνω). For perf. with ἄν cp. 

93- 
841 περισσόν, more than ordinary, 
worthy of special note: Her. 2. 32 τοὺς 
ἄλλα τε μηχανᾶσθαι... περισσά, 1. 6. among 
other remarkable enterprises: Eur. Supp/. 
790 τὸ μὲν γὰρ οὐκ ἤλπιζον ἂν πεπονθέναι 
| πάθος περισσόν, εἰ γάμων ἀπεζύγην, I 
had not deemed it a more than common 
woe. locasta is unconscious of any pont 
peculiar to her version, on which a hope 
could depend: she had reported the story 


8. 


\toux π 


ϑωχ 


114 


ZOPOKAEOYS 


Ole λῃστὰς ἔφασκες αὐτὸν ἄνδρας ἐννέπειν 


ὡς νιν κατακτείνειαν. 


εἰ μὲν οὖν ἔτι 


λέξει τὸν αὐτὸν ἀριθμόν, οὐκ ἐγὼ 'κτανον' 


οὐ γὰρ γένοιτ᾽ ἂν εἷς γε τοῖς πολλοῖς ἴσος" 


845 


ἼΔΕ ἄνδρ᾽ ἕν᾽ οἰόζωνον αὐδήσει, σαφώς 
τοῦτ᾽ ἐστὶν ἤδη τοὔργον εἰς ἐμὲ ῥέπον. 


10: 


ἀλλ᾽ ὡς φανέν γε τοὔπος ὧδ᾽ ἐπίστασο, 


κοὐκ ἔστιν αὐτῷ τοῦτό "ἢ ἐκβαλεῖν πάλιν" 


πόλις γὰρ ἤκουσ᾽, 


οὐκ ἐγὼ “μόνη, τάδε. 


850 


εἰ δ᾽ οὖν τι κἀκτρέποιτο τοῦ πρόσθεν λόγου, 
οὔτοι ποτ᾽, ὦναξ, τόν γε Λαΐου φόνον 

φανεῖ δικά ϊως ὀρθόν, ov γε Λοξίας 

διεῖπε χρῆναι παιδὸς ἐξ ἐμοῦ θανεῖν. 


καΐτοι νιν οὐ κεῖνός γ᾽ ὁ δύστηνός ποτε 
’ὔ > 3 3 ΩΝ ’ὕ »ν 

κατέκταν,, ἀλλ αὐτὸς πάροιθεν ὠλετο. 

ὥστ᾽ οὐχὶ μαντείας x, av οὔτε τῇδ᾽ ἐγὼ 


855 


ΕἾ βλέψαιμ᾽ ἂν οὕνεκ᾽ οὔτε τῇδ᾽ ἂν ὕστερον. 


_w@onjectured by Arndt, Blaydes, and M. Schmidt. 
. the letters ac are in an erasure, having been made by an early corrector. 
“ that the ist hand wrote κατακτείνοιεν. 


843 L has κατακτείναιεν, but 
Wolff thinks 


As the last ε is certainly from the rst hand, the 


ist hand must have written either that or κατακτείνειεν, which is in at least one later MS. 


(Pal.), others having κατακτείναιεν (as A), or κατακτείνειαν. 


Most of the recent edd. 





of the slaughter in the fewest words, 715 
—716. 

nae f. τὸν αὐτὸν ἀριθμόν, z. 4. πλείους 
and not ἕνα: or, in the phrase of gram- 
marians, τὸν πληθυντικὸν and not τὸν 
ἑνικὸν ἀριθμόν.---ἴσος : ‘one cannot be 
made to tally with (cannot be identified 
with) those many’: τοῖς πολλοῖς, refer- 
ring to the plur. Anords (842). 

846 οἰόζωνον, journeying alone. The 
peculiarity of the idiom is that the second 
part of the compound is equivalent to a 
separate epithet for the noun: ζ. 6. olé6- 
{evos, ‘with solitary girdle,’ signifies, 
‘alone, and girt up.’ O. C. 717 τῶν 
ἑκατομπόδων Νηρήδων, not, ‘with a 
hundred feet each,’ but, countless, and 
dancing: 20. 17 πυκνόπτεροι anddves, 
not, thickly-feathered, but, many and 
winged: 7b. 1055 dcarédous ἀδελφάς, not, 
separately-journeying sisters, but, two 
sisters, journeying: Az. 390 δισσάρχας 
βασιλῆς, not, diversely-reigning kings, but, 
two reigning kings: Eur. A/c. gos κόρος 
povémats, not, a youth with one child, 
but, a youth, his only child: Phoen. 683 


διώνυμοι θεαί, not, goddesses with con- 
trasted names, but, several goddesses, each 
of whom is invoked. So I understand 
Eur. Ov. 1004 μονόπωλον ᾿Αῶ, ‘Eos 
who drives her steeds alone’ (when moon 
and stars have disappeared from the sky). 

847 εἰς ἐμὲ ῥέπον: as if he were stand- 
ing beneath the scale in which the evi- 
dence against him lies; that scale proves 
the heavier of the two, and thus descends 
towards him. 

848 ἐπίστασο φανὲν τοὔπος ὧδε, know 
that the tale was thus set forth: ἐπίστασο 
as φανὲν τοὔπος ὧδε, know that you may 
take the story to have been thus set forth: 
where ὡς merely points to the mental 
attitude which the subject of ἐπίστασο 
isto assume. hil. 567 ὡς ταῦτ᾽ ἐπίστω 
δρώμεν᾽, οὐ μέλλοντ᾽ ἔτι, know that you 
may assume these things to be a-doing, 
not delayed: and 26. 253, 415: below 
956. So with the gen. abs,: Az. 281 
ὡς ὧδ᾽ ἐχόντων τῶν δ᾽ ἐπίστασθαί σε χρή, 
these things being so, you must view them 
in that belief. 

849 ἐκβαλεῖν, repudiate: Plat. Crito 


ΟἸΔΊΠΟΥΣ. TYPANNOS 115 


ΟΕ. Thou wast saying that he spoke of Laius as slain by 
robbers. If, then, he still speaks, as before, of several, I was 
not the slayer: a solitary man could not be held the same with 
that band. But if he names one lonely wayfarer, then beyond 
doubt this guilt leans to me. 

Io. Nay, be assured that thus, at least, the tale was first 
told; he cannot revoke that, for the city heard it, not I alone. 
But even if he should diverge somewhat from his former story, 
never, king, can he show that the murder of Laius, at least, is 
truly square to prophecy; of whom Loxias plainly said that he 
must die by the hand of my child. Howbeit that poor innocent 
never slew him, but perished first itself. So henceforth, for what 
touches divination, I would not look to my right hand or my left. 


give κατακτείνειαν. It is perhaps safest to do so, in the absence of better evidence for -acey 
(or -ovev) than we have in this passage. Yet cp. the inscription in Kaibel’s Zpzgram- 
mata (24. 2), ἐχθροὶ στήσαιεν Ζηνὶ τρόπαιον ἕδος (date, circ. 400—350 B.C.); to which 
Meisterhans (Gramm. der Attischen Inschriften, p. 75) refers in proof that ‘the poets 
of the 4th cent. B.c. could use, without metrical necessity, the un-Attic forms of the 


aorist optative.’ 


851 κἀκτρέποιτο L: καὶ τρέποιτο r. 


852 τόν ye L: τόνδετ: 





46 B τοὺς δὲ λόγους ods ἐν τῷ ἔμπροσθεν 
ἔλεγον οὐ δύναμαι νῦν ἐκβαλεῖν. 

851 εἰ κἀκτρέποιτο, if he should turn 
aside: see on 772 καὶ.. λέξαιμ᾽ ἄν. 

852 τόν ye Aatov φόνον.  Iocasta 
argues: ‘Even if he should admit that 
the “eed was done by ove man (a circum- 
stance which would confirm our fears that 
the deed was yours), at any rate the death 
of Laius cannot be shown to have hap- 
pened as the oracle foretold; for Laius was 
to have been killed by my son, who died 
ininfancy. The oracular art having failed 
in this instance, I refuse to heed Tei- 
resias when he says that you will yet be 
found guilty of slaying your father 
Polybus.’ Iocasta, bent on cheering 
Oedipus, merely alludes to the possi- 
bility of his being indeed the slayer of 
Laius (851), and turns to the comforting 
aspect of the case—viz., the undoubted 
failure of the oracle, om any supposition. 
This fine and subtle passage is (to my 
apprehension) utterly defaced by the con- 
jecture σόν γε Λαΐου φόνον (Bothe), ‘it 
cannot be shown that your slaying of 
Laius fulfils the oracle. Herm. reads 
τόνδε, ‘this slaying’ (of which you think 
yourself guilty): but the ye is needed. 

853 δικαίως ὀρθόν, in a just sense 
correct, z.e. properly fulfilled: for ὀρθόν 
see 506.—Aoflas: a surname of the ora- 
cular Apollo, popularly connected with 
λοξός, ‘oblique’ (akin to λέχ-ριος, obliguus, 


luxus, ‘sprained’), as=the giver of Ζ7:- 
direct, ambiguous responses (λοξὰ καὶ 
ἐπαμῴφοτερίζοντα, Lucian Dial, Deor. 
16): Cornutus 32 λοξῶν δὲ καὶ περι- 
σκελῶν ὄντων τῶν χρησμῶν ods δίδωσι 
Λοξίας ὠνόμασται, and so Lycophron 14. 
1467: to this Pacuvius alludes, Flexa xox 
Salsa autumare dictio Delphis solet. The 
association of Apollo with Helios sug- 
gested to the Stoics that the idea con- 
necting λοξός with Λοξίας might be that 


of the ecliptic: to which it might be re-. 


plied that the name Λοξίας was older 
than the knowledge of the fact. It is 
not etymologically possible to refer Λοξίας 
to Aux, dwx. But phonetic correspon- 
dence would justify the connection, sug- 
gested by Dr Fennell, with ἀ-λεξ (Skt. 
vrak-sh). Λοξίας and his sister Λοξώ (Cal- 
lim. Del. 292) would then be other forms 
of Phoebus and Artemis ἀλεξητήριοι, 
ἀλεξίμοροι (above, 164), ‘defenders.’ Io- 
casta’s utterance here is not really incon- 
sistent wi'a her reservation in 712: see 
note there. 

854 διεῖπε: expressly said: cp. δια- 
δείκνυμι, to show clearly (Her.), διαδηλόω, 
διαρρήδην, ‘in express terms’: so above, 
394 αἴνιγμα... διειπεῖν = ‘to declare’ (solve) 
a riddle. 

857 2. οὔτε τῇϑε-- οὔτε THSe=ob7’ ἐπὶ 
τάδε οὔτ᾽ ἐπὶ θάτερα, neither to this side 
nor to that: Phil. 204 ἤ που τῇδ᾽ ἢ τῇδε 
τόπων: 1, 12. 237 (Hector to Polyda- 


8—2_ 


~ 


στρ.α. 


116 


OI. καλώς νομίζεις. 


IO. πέμψω ταχύνασ᾽" 


ΣΟΦΟΚΛΕΟΥΣ 


ἀλλ᾽ ὅμως τὸν ἐργάτην 
πέμψον τινὰ στελοῦντα, μηδὲ τοῦτ᾽ ἀφῇς. 
ἀλλ᾽ ἴωμεν ἐς δόμους: 


860 


οὐδὲν yap av mpdkay dv ὧν ov σοὶ φίλον. 


A 
je 


εἴ μοι ξυνείη φέροντι τὐζίς 


2 μοῖρα τὰν εὔσεπτον ἁγνείαν λόγων 


8 ἔργων LAS πάντων, ὧν νόμοι πρόκεινται 


4 ὑψίποδες, οὐρανίαν 


δός 


ὅ δι αἰθέρα τεκνῳθέντες, ὧν Ὄλυμπος 


mas): τύνη δ᾽ οἰωνοῖσι τανυπτερύγεσσι 
κελεύεις | πείθεσθαι" τῶν οὔτι μετατρέπομ᾽ 


οὔτ᾽ ἀλεγίζω, | εἴτ᾽ ἐπὶ δεξί᾽ ἴωσι πρὸς ἠῶ 
τ᾽ ἠέλιόν τε, | εἴτ᾽ ἐπ᾽ ἀριστερὰ τοί γε 
ποτὶ ζόφον ἠερόεντα. ---μαντείας y ...00- 
veka, 50 far as it is concerned: O. C. 22 
χρόνου μὲν οὕνεκ᾽, n. 

859 f. καλῶς νομίζεις : he assents, al- 
most mechanically—but his thoughts are 
intent on sending fer the herdsman.— 
στελοῦντα, ‘to summon’: στέλλειν =‘ to 
cause to set out’ (by a mandate), hence 
‘to summon’: O. C. 297 σκοπὸς δέ vw | 
ὃς κἀμὲ δεῦρ᾽ ἔπεμπεν οἴχεται στελῶν.---μη- 
δὲ τοῦτ᾽ ἀφῇς, ‘and do not neglect this.’ 
With a point after στελοῦντα we could 
render: ‘neglect 7102 even this’: but Oed. 
does not feel, nor feign, indifference. 

862 γάρ, since ἴωμεν x.7.A. implies 
consultation. The doubled ἂν gives em- 
phasis: cp. 139. —dv οὐ σοὶ φίλον :-Ξ τού- 
των ἃ πρᾶξαι οὐ σοὶ φίλον ἐστί. Pihil. 
1227 ἔπραξας ἔργον ποῖον ὧν οὔ σοι πρέ- 
Tov ; 

863—910 Second στάσιμον. The 
second ἐπεισόδιον (512—862) has been 
marked by the overbearing harshness of 
Oedipus towards Creon; by the rise of a 
dreadful suspicion that Oedipus is avay- 
vos—blood-guilty for Laius; and by the 
avowed contempt of Iocasta, not, indeed, 
for Apollo himself, but for the μαντική of 
his ministers. These traits furnish the 
two interwoven themes of the second 
stasimon: (1) the prayer for purity in 
word as in deed: (2) the deprecation of 
that pride which goes before a fall; 
—whether it be the insolence of. the τύ- 
pavvos, or such intellectual arrogance as 
Iocasta’s speech bewrays (λόγῳ, v. 884). 
The tone of warning reproof towards 
Oedipus, while only allusive, is yet in 
contrast with the firm though anxious 
sympathy of the former ode, and serves 





to attune the feeling of the spectators for 
the approach of the catastrophe. 

ist strophe (863—-872). May I ever be 
pure in word and deed, loyal to the un- 
written and eternal laws. 

ist antistrophe (873—882). <A tyrant’s 
selfish insolence hurls him to ruin. But 

may the gods prosper all emulous effort 
for the good of the State. 

and strophe (883—8096). Irreverence 
in word or deed shall not escape: the 
wrath of the gods shall find it out. 

and antistrophe (897—910). — Surely 
the oracles concerning Laius will yet be 
justified: O Zeus, suffer not Apollo’s 
worship to fail. 

868 εἴ μοι Evveln μοῖρα φέροντι is 
equivalent to εἴθε διατελοῖμι φέρων, the 
part. implying that the speaker is al; eady 
mindful of ἁγνεία, and prays that he may 
continue to be so: whereas εἴ μοι ξυνείη 
μοῖρα φέρειν would have been equivalent 
to εἴθε μοι γένοιτο φέρειν, an aspiration 
towards ἁγνεία as not yet attained. 
Though μοῖρα is not expressly personified 
(cp. Pind. Pyth. 3. 84 τὶν δὲ μοῖρ᾽ εὐδαι- 
μονίας ἕπεται), the conception of it is so 
far personal that ξυνείη (‘be with’) is 
tinged with the associations of ξυνειδείη 
(‘be witness to’), and thus softens any 
boldness in the use of the participle; a 
use which, in principle, is identical with 
the use after such verbs as διατελῶ, τυγ- 
χάνω, λανθάνω. dépovt. (= φερομένῳ, 
see On 500)... ἁγνείαν, winning purity, 
regarded as a precious κτῆμα (Ant. 150): 
cp. 1190 πλέον τᾶς εὐδαιμονίας φέρει: 31. 
968 εὐσέβειαν. ..οἴσ ει (will win the praise 
of piety): Eur. Or. 158 ὕπνου.. ᾧ ερο- 
μένῳ xapav.—Others take φέροντι as= 

‘bearing about with me’ (or ‘within me’). 
Cp. Ant. 1090 τὸν νοῦν τ᾽ ἀμείνω τῶν φρε- 
νών ἢ νῦν φέρει (where it=7pépew in 
1089): Zr. 108 εὔμναστον δεῖμα φέρουσαν 


OIAITOYS ΤΥΡΆΝΝΟΣ 


ΟΕ. Thou judgest well. 


ΤΕ 


But nevertheless send some one 


to fetch the peasant, and neglect not this matter. 


ΤΟ, 


I will send without delay. But let us come into the 


house: nothing will I do save at thy good pleasure. 


CH. May destiny still find me winning the praise of rever- 1st 


ent purity in all words and deeds sanctioned by those laws of StPhe- 


range sublime, called into life throughout the high clear heaven, 


(where Casaubon τρέφουσαν, as Blaydes 
τρέφοντι here). ‘This may be right: but 
the use here, at least, would be bold; and 
I still incline to the former view. 

864 εὔσεπτον, active, ‘reverent,’ only 
here: so 890 τῶν ἀσέπτων, also act., ‘irre- 
verent deeds,’ as in Eur. Helen. 542 Upw- 
τέως ἀσέπτου παιδός, impious, unholy: 
see On 515. 

865 ὧν νόμοι πρόκεινται ὑψίπ., ‘for 
which (enjoining which) laws have been 
set forth, moving on high,’—having their 
sphere and range in the world of eternal 
truths: ὑψίποδες being equiv. to ὑψηλοὶ 
καὶ ὑψοῦ πατοῦντες : see on οἰόζωνον 846, 
and contrast χθονοστιβῇ 301. The meta- 
phor in νόμοι was less trite for a Greek 
of the age of Sophocles than for us: cp. 
Plat. Legg. 793 A τὰ καλούμενα ὑπὸ 
τῶν πολλών ἄγραφα νόμιμα--οὔτε 
νόμους δεῖ προσαγορεύειν αὐτὰ οὔτε ἄρ- 
ρητα ἐᾶν.----τρόκεινται (Thuc. 3. 45 ἐν οὖν 
ταῖς πόλεσι πολλῶν θανάτου ζημία πρόκει- 
ται) strengthens the metaphor: Xen. 
Mem. 4. 4. 21 δίκην γέ τοι διδόασιν οἱ 
παραβαίνοντες τοὺς ὑπὸ τῶν θεῶν κει- 
μένους νόμους, ἣν οὐδενὶ τρόπῳ δυνατὸν 
ἀνθρώπῳ διαφυγεῖν, ὥσπερ τοὺς ὑπ᾽ ἀν- 
θρώπων κειμένους νόμους ἔνιοι δια- 
φεύγουσι τὸ δίκην διδόναι: where Socrates 
speaks of the ἄγραφοι νόμοι which are ἐν 
πάσῃ χώρᾳ κατὰ ταὐτὰ voucsduevor,—as to 
revere the gods and honour parents. Arist. 
Rhet. τ. 13. 2: “1 consider law (νόμον) 
as particular (ἴδιον) or universal (k oc- 
v 6v), the particular law being that which 
each community defines in respect to 
itself,—a law partly written, partly un- 
written [as consisting in local custom]; 
the universal law being that of nature 
(τὸν κατὰ φύσιν). For there is a cer- 
tain natural and universal right and wrong 
which all men divine (μαντεύονται), even if 
they have no intercourse or covenant with 
each other; as the Antigone of Sophocles 
is found saying that, notwithstanding the 
interdict, it is right to bury Polyneices’ 


whose father is Olympus 


(Ant. 454, where she appeals to the ἄ- 
γὙραπτα κἀσφαλῆ θεών νόμιμα). Cp. 
Cope’s Introd. to Arist. Rhet. p. 239. 

866 οὐρανίαν δι᾽ αἰθέρα τεκνωθέντες, 
called into a life that permeates the hea- 
venly ether (the highest heaven): the 
metaphor of texvw@évres being qualified 
by its meaning in this particular applica- 
tion to νόμοι, viz. that they are revealed 
as operative; which allows the poet to 
indicate the sphere throughout which they 
operate by δι᾽ αἰθέρα, instead of the ver- 
bally appropriate ἐν αἰθέρι: much as if 
he had said δι᾽ αἰθέρα évepyol ἀναφανέντες. 
So, again, when he calls Olympus, not 
Zeus, their πατήρ, the metaphor is half- 
fused with the direct notion of ‘source.’ 
Cp. Arist. Rhet. 1. 13. 2 quoted on 865, 
which continues (illustrating τὸ φύσει 
δίκαιον): καὶ ws ᾿Εμπεδοκλῆς λέγει περὶ 
τοῦ μὴ κτείνειν τὸ ἔμψυχον᾽ τοῦτο γὰρ οὐ 
τισὶ μὲν δίκαιον τισὶ δ᾽ οὐ δίκαιον, ᾿Αλλὰ 
τὸ μὲν πάντων νόμιμον διά τ᾽ εὐρυ- 
μέδοντος αἰθέρος ἠνεκέως τέταται 
διά τ᾽ ἀπλέτου αὖ γῆς (so Scaliger 
rightly amended αὐγῆς: Emped. 438): 
where the special reference of Empedo- 
cles is to a principle of life common to 
gods, men, and irrational animals (πνεῦμα 
τὸ διὰ παντὸς τοῦ κόσμου διῆκον ψυχῆς Tpd- 
πον, Sextus Emp. Adv. Math. 9. 127 : cp. 
Cope ad loc.).—atOépa: //. 16. 364 ὡς δ᾽ 
ὅτ᾽ dm’ Οὐλύμπου νέφος ἔρχεται οὐρανὸν 
εἴσω | αἰθέρος ἐκ δίης : where, Olympus 
being the mountain, the οὐρανός is above 
the αἰθήρ, since ἐξ αἰθέρος could not=éf 
alOpas, after clear weather: and so 77. 2. 
458 δι᾽ αἰθέρος οὐρανὸν ἵκει: 71. 19. 351 
οὐρανοῦ ἐκκατέπαλτο δι᾽ αἰθέρος: cp. Ant. 
420. Here οὐρανίαν αἰθέρα Ξ πε highest 
heaven. : - 

867 ᾿"Ολυμπος : not the mountain, as 
in the J/iad, but, as in the Odyssey (6. 
42), the bright supernal abode of the 
gods: and so=the sky itself: Ὁ. C. 1654 
γῆν τε προσκυνοῦνθ᾽ ὁμοῦ | kal τὸν θεών 
"Ὄλυμπον. 


118 


6 πατὴρ μόνος, οὐδέ νιν 
7 θνατὰ φύσις ἀνέρων 


8 ἔτικτεν, οὐδὲ μή ποτε λαθα κατακοιμάσῃ" 


ΣΟΦΟΚΛΕΟΥΣ 


870 


9 μέγας ἐν τούτοις θεός, οὐδὲ γηράσκει. 


> ’ 
αντ. a. 


ν Zz , 
ὕβρις φυτεύει τύραννον" 


873 


2 ὕβρις, εἰ πολλῶν ὑπερπλησθῇ μάταν, 


8 ἃ μὴ 'πίκαιρα. μηδὲ συμφέροντα, 
γεῖσ' ἀναβᾶσ᾽ 
ἀποτμοτάταν ὥρουσεν εἰς ἀνάγκαν, 


4 ἀκρότατα 
Ε ἘΣ 


er 


6 ἔνθ᾽ οὐ ποδὶ χρησίμῳ 
7 χρῆται. τὸ καλῶς 


875 


ἔχον. 
8 πόλει πάλαισμα μήποτε λῦσαι θεὸν αἰτοῦμαι. 


880 


9 θεὸν ov λήξω ποτὲ προστάταν ἴσχων. 


σόν γε Bothe. 
(as E). 


Elmsley has been followed by a majority of edd. in giving μήποτε. 


870 οὐδὲ μήν ποτε λάθραι (the p almost erased) κατακοιμάσηι L. 
Most of the later Mss. (as A) have λάθα, and κατακοιμάσει: 


some have μήν, others μή 
««κατακοι- 





870 ἔτικτεν, ‘was their parent,’ some- 
times used instead of érexe where the 
stress is not so much on the fact of the 
birth as on the parentage, 1099, O. C. 982, 
fr. sor: Pind. P. 9. 15 dy rore=Nails... 
ἔτικτεν. (It would be prosaic to render, 
‘brought forth successively,’—developed.) 

οὐδὲ μή ποτε κατακοιμάσῃ. I formerly 
gave οὐδὲ μάν ποτε κατακοιμάσει,----τερατά- 
ing L’s μήν as more significant than its κα- 
τακοιμάσηι. But I now think that the pro- 
babilities are stronger for μήν having come 
from μή. In point of fitness, the readings 
are here equal. οὐ μή expresses conviction : 
Plat. Phaedo 105 Ὁ οὐκοῦν ἡ ψυχὴ τὸ ἐναντίον 
ᾧ αὐτὴ ἐπιφέρει ἀεὶ οὐ μή ποτε δέξηται, ὡς 
ἐκ τῶν πρόσθεν ὡμολόγηται; 

871 μέγας ἐν τούτοις θεός : the divine 
virtue inherent in them is strong and un- 
failing. θεός without art., as 880: Ο. C. 
1694 τὸ φέρον ἐκ θεοῦ. For this use of 
the word, to express an indwelling power, 
cp. Eur. fr. inc. 1007 ὁ νοῦς yap juw 
ἐστιν ἐν ἑκάστῳ θεός. 

873 ὕβρις. The tone of Oedipus to- 
wards Creon (esp. 618—672) suggests the 
strain of warning rebuke. Aeschylus, 
with more elaborate imagery, makes 
ὕβρις the daughter of δυσσεβία and the 
parent of a νέα ὕβρις which in turn begets 
κόρος and θράσος (Ag. 764).---τύραννον, 


here not ‘a prince,—nor even, in the 
normal Greek sense, an unconstitutionally 
absolute ruler (bad or good),—but, in our 
sense, ‘a tyrant’: cp. Plat. Fol. 301 Ὁ 
ὅταν μήτε κατὰ νόμους μήτε κατὰ ἔθη 
πράττῃ τις εἷς ἄρχων, προσποιῆται δὲ 
ὥσπερ ὁ ἐπιστήμων ὡς ἄρα παρὰ τὰ γε- 
γραμμένα τό γε βέλτιστον ποιητέον, ἡ δέ 
τις ἐπιθυμία καὶ ἄγνοια τούτον τοῦ 
μιμήματος- ἡγουμένη, μῶν οὐ τότε τὸν 
τοιοῦτον ἕκαστον τύραννον κλητέον; Rep. 
573 Β ἄρ᾽ οὖν... καὶ τὸ πάλαι διὰ τὸ τοιοῦτον 
τύραννος ὁ Ἔρως λέγεται ; 

874 f. εἰ.. ὑπερπλησθῇ: Plat. Rep. 
ἘΠ) 0 τυραννικὸς δὲ.. ἀνὴρ ἀκριβῶς. γίγ- 
νεται, ὅταν ἢ φύσει ἢ ἐπιτηδεύμασιν ἢ ἀμ- 
φοτέροις μεθυστικός τε καὶ ἐρωτικὸς 
καὶ μελαγχολικὸς γένηται. For εἰ 
with subj., see on 198.---ὁ μή : the generic 
μή (such wealth as is not meet): cp. 397 n 

876 The reading of all the Mss., ἀκ. 
ροτάταν εἰσαναβάσ᾽, is accounted for by 
Wolffs emendation, which I have now 
received, ἀκρότατα γεῖσ᾽ ἀναβᾶσ᾽. The 
change of y into v was very easy for cur- 
sive minuscule; while on the other hand 
the presence of ἀνάγκαν in the next verse 
is not enough to explain the change of 
an original ἀκρότατον into the unmetrical 
axpordrav.—yetora, the coping of a wall: 
cp. Eur. Phoen. 1180 (of Capaneus) ἤδη 


ΟΠ ΘΥΣΤΎΡΑΝΝΟΣ [19 
alone; their parent was no race of mortal men, no, nor shall 
oblivion-eéver lay them to sleep; the god is mighty in them, 
and he grows not old. 


Insolence breeds the tyrant; Insolence, once vainly surfeited 
on wealth that is not meet nor good for it, when it hath scaled 
the topmost ramparts, is hurled to a dire doom, wherein no 
service of the feet can serve. But I pray that the god never 
quell such rivalry as benefits the State; the god will I ever hold 
for our protector. 


μήσῃ. 876 f. ἀκροτάταν εἰσαναβᾶσ᾽ ἀπότομον  ὥρουσεν εἰσ ἀνάγκαν L. All Mss. 
have ἀκροτάταν. Instead of ἀπότομον, A has ἄποτμον, with o written above.—dxpérara 





δ᾽ ὑπερβαίνοντα γεῖσα τειχέων | βάλλει safe landing-place. For the paronomasia 


κεραυνῷ Ζεύς vw (as Ant. 131, of the same, 
βαλβίδων | ἐπ᾽ ἄκρων ἤδη | νίκην ὁρμῶντ᾽ 
ἀλαλάξαι). So here the ὕβρις is hurled 
down, Capaneus-like, at the crowning 
moment of wicked triumph. In Eur. 
Suppl. 728 there is a similar image of in- 
solent ambition hurled down, as from the 
topmost round of a scaling-ladder: ὑβρισ- 
τὴν λαόν, ὃς πράσσων καλῶς | els ἄκρα 
βῆναι κλιμάκων ἐνήλατα | ζητῶν ἀπώλεσ᾽ 
ὄλβον. 
. 877 With the MS. ἀπότομον wpov- 
σεν els ἀνάγκαν, there is a defect of ~~ 
or—. Reading ἀκρότατον in 876, Arndt 
supplies αἷπος before ἀπότομον, as I for- 
merly supplied ἄκρον in the same place: 
E. L. Lushington thought of ὄρος to follow 
ἀπότομον : Campbell reads ἐξώρουσεν. But 
none of these remedies, nor any other of 
a like kind, is satisfactory, or very pro- 
‘bable. I now agree with Wecklein in 
preferring Schnelle’s ἀποτμοτάταν for 
ἀπότομον. This is metrically exact (=867 
δι’ αἰθέρα rexv-), and removes the neces- 
sity for any conjectural supplement. (The 
superlative of ἄποτμος occurs Od. 2. 219.) 
—dpovoev, snomicaor. (cp. O.C. 1215 κατ- 
ἐθεντο).---ἠἀνάγκαν, a constraining doom 
from the gods: Eur. PA. 1000 els ἀνάγκην 
. δαιμόνων ἀφιγμένοι. Cp. Plat. Legg. 716A 
ὁ δέ res ἐξαρθεὶς ὑπὸ μεγαλαυχίας ἠνχρήμα- 
σιν ἐπαιρόμενος ἢ τιμαῖς ἢ καὶ σώματος 
εὐμορφίᾳ, ἅμα νεότητι καὶ ἀνοίᾳ φλέγεται 
τὴν ψυχὴν μεθ᾽ ὕβρεως... μετὰ δὲ χρόνον οὐ 
πολὺν ὑποσχὼν τιμωρίαν τῇ δίκῃ ἑαυτόν τε 
καὶ οἶκον καὶ πόλιν ἄρδην ἀνάστατον ἐποίησε. 
878 χρησίμῳ... χρῆται : where it does 
not use the foot to any purpose: 2.4. the 
leap is to headlong destruction ; it is not 
one in which the feet can anywhere find a 


cp. Pind. P. 2. 78 κερδοῖ δὲ τί μάλα τοῦτο 
κερδαλέον τελέθει; ‘but for the creature 
named of gain,’ (the fox) ‘what so gainful 
is there here?” 

879 τὸ καλῶς δ᾽ ἔχον : but I ask that 
the god never do away with, abolish, 
that struggle which is advantageous for 
the city,—z.e. the contest in which citizen 
vies with citizen who shall most serve the 
State. The words imply a recognition 
of the προθυμία which Oed. had so long 
shown in the service of Thebes: cp. 48, 
93: 247. 

880 πάλαισμα: cp. Isocr. 22. 7 87 
τοῖς καλῶς τὰς πόλεις τὰς αὑτῶν διοικοῦσιν 
ἁμιλλητέον καὶ πειρατέον διενεγκεῖν av- 
τών. Plut. Mor. 820 C ὥσπερ οὐκ ἀργυ- 
ρίτην οὐδὲ δωρίτοην ἀγῶνα πολιτείας 
ἀγωνιζομένοις (the emulous service of 
the State), ἀλλὰ ἱερὸν ws ἀληθώς καὶ στε- 
φανίτην (like the contests in the great 
games). 

882 f. προστάταν : defender, cham- 
pion: not in the semi-technical sense of 
‘patron,’ as in 411.—tmépotra, adverbial 
neut. of ὑπέροπτος [not ὑπερόπτα, epic 
nom. for ὑπερόπτης, like ἱππότα]: cp. 
O. C. 1695 οὔτοι κατάμεμπτ᾽ ἔβητον, ye 
have fared not amiss. //. 17. 75 ἀκίχητα 
διώκων | ἵππους : Eur. Suppl. 770 ἄκραντ᾽ 
ὀδύρει: Ph. 1739 ἄπειμι... ἀπαρθένευτ᾽ ddw- 
μένα: Lon 255 ἀνερεύνητα δυσθυμεῖ (hast 
griefs which I may not explore).—xepolv, 
in contrast with Ἀόγῳ, merely ΞΞ ἔργοις, 
not ‘ deeds of violence’: cp. Eur. PA. 312 
πῶς... | καὶ χερσὶ Kal λόγοισι... | περι- 
χορεύουσα τέρψιν... λάβω, find joy in deed 
and word of circling dance, z.é. in linking 
of the hands and in song: cp. 864. 


Ist anti- 
strophe. 


120 


στρ. β΄. 
2 Δίκας ἀφόβητος, οὐδὲ 
3 δαιμόνων ἕδη σέβων, 
4 κακά νιν ἕλοιτο μοῖρα, 


5 δυσπότμου χάριν χλιδάς, 


ΣΟΦΟΚΛΕΘΥΣ 


εἰ δέ τις ὑπέροπτα χερσὶν ἢ λόγῳ πορεύεται, 885 


88ς 


6 εἰ μὴ τὸ κέρδος κερδανεῖ δικαίως 


3 
Υ καὶ τῶν ἀσέπτων ἔρξεται, 


890 


8 ἢ τῶν ἀθίκτων * θίξεται ματάζων. 
9 τίς ἔτι TOT ἐν τοῖσδ᾽ ἀνὴρ Ἐβεῶν βέλη 


10 *ev€erar ψυχᾶς ἀμύνειν ; 


γεῖσ᾽ ἀναβᾶσ᾽ Wolff; ἀποτμοτάταν (for ἀπότομον) Schnelle. 


See comment. 890 ép- 


ξεται L. The scribe had begun to write x as the third letter, but corrected it to &. 


The later Mss. have the same word, with variations of breathing. 
In L the breathing has been added (or retouched) by the first corrector. 


Blaydes. 


891 ἕξεται MSS. 
θίξεται 


(The mode of writing ἕξεται in L, where the first € is large, suggests the ease 





885 Δίκας ἀφόβητος, not fearing Jus- 
tice: cp. 969 ἄψαυστος éyxous, not touch- 
ing aspear. The act. sense is preferable 
only because class. Greek says φοβηθεὶς 
τὴν δίκην, not φοβηθεὶς ὑπὸ τῆς δίκης : the 
form of the adj. would warrant a pass. 
sense: cp. 77. 685 ἀκτῖνος... ἄθικτον. 
With ἄφοβος (Az. 366) ἀφόβητος cp. arap- 
βής (77. 23) ἀτάρβητος (Az. 197). 

886 ἕδη, zmages of gods, whether sit- 
ting or standing; but always with the 
added notion that they are placed in a 
temple or holy place as objects of wor- 
ship. Timaeus p. 93 &dos" τὸ ἄγαλμα 
καὶ ὁ τόπος ἐν ᾧ ἵδρυται: where τόπος 
prob. denotes the small shrine in which 
an image might stand. Dionys. Hal. 1. 

47 uses ἕδη to render fenates. Liddell 
Τὴ Scott s.v. cite the following as places 
in which ἕδος ‘may be a ¢emfle’: but in 
all of them it must mean zwage. Isocr. 
or. 15 § 2 Φειδίαν τὸν τὸ τῆς ᾿Αθηνᾶς 
ἕδος ἐργασάμενον, 2.6. the chryselephan- 
tine Athena Parthenos; cp. Plut. Per, 

13 ὁ δὲ Φειδίας εἰργάζετο μὲν τῆς θεοῦ τὸ 
χρυσοῦν Edos: Xen. Hellen. το 4. 12 
Πλυντήρια ἦγεν. ἡ πόλις, τοῦ ἕδους κατα- 
κεκαλυμμένου τῆς ᾿Αθηνᾶς : 1.6. the ἀρχαῖον 
βρέτας of Athena Polias in the Erech- 
theum was veiled in sign of mourning 
(the death of Aglauros being commemo- 
rated at the festival of the Plunteria). 
Paus. 8. 46. 2 φαίνεται δὲ οὐκ ἄρξας ὁ Αὔ- 
γουστος ἀναθήματα καὶ ἕδη θεῶν ἀπά- 
γεσθαι παρὰ τῶν κρατηθέντων (ἰ.6. carry 
off to Italy): where ἀναθήματα are dedi- 


cated objects generally, ἔδη images wor- 
shipped in temples. Is Sophocles glancing 
here at the mutilators of the Hermae in 
415 B.C., and especially at Alcibiades? 
We can hardly say more than this :—(r) 
There is no positive probability as to the 
date of the play which can be set against 
such a view. (2) The language suits it,— 
nay, might well suggest it; nor does it 
matter that the ‘Epuai, though ἀναθήματα 
(Andoc. De Myst. § 34), were not properly 
ἕδη. (3) It cannot be assumed that the 
dramatic art of Sophocles would exclude 
such a reference. Direct contemporary 
allusion is, indeed, uncongenial to it. 
But a light touch like this—especially in 
a choral ode—might fitly strike a chord 
of contemporary feeling in unison with 
the emotion stirred by the drama itself. 
I do not see how to affirm or to deny 
that such a suggestion was meant here. 
(ΟΡ Ὁ Οὐ 1537 2.) ᾿ 

888 δυσπότμου, miserably perverse : 
Ant. 1025 οὐκέτ᾽’ ἔστ᾽... | ἄβουλος οὔτ᾽ 
ἄνολβος. 

890 τῶν ἀσέπτων : see on 864.---ἔρξε- 
ται, Κορ himself from: O. C. 836 εἴργου, 
‘stand back’: Her. 7. 197 ws κατὰ τὸ 
ἄλσος ἐγένετο, αὐτός τε ἔργετο αὐτοῦ καὶ 
τῇ στρατιῇ πάσῃ παρήγγειλε. Plat. Legg. 
838 A ws εὖ τε καὶ ἀκριβώς elpyovra τῆς 
τών καλῶν ξυνουσίας. As to the form, Her. 
has ἔργω or éépyw: in Attic the Mss. give 
Aesch. Zum. 566 κατεργαθοῦ : Soph. Az, 
593 ξυνέρξετε: Thuc. 5. 11 περιέρξαντες 
(so the best Mss., and Classen): Plat. 


. OIAITOYS ΤΥΡΆΝΝΟΙΣ 


r2T 


But if any man walks haughtily in deed or word, with no 
fear of Justice, no reverence for the images of gods, may an evil 
doom seize him for his ill-starred pride, if he will not win his 
vantage fairly, nor keep him from unholy deeds, but must lay 


profaning hands on sanctities. 


Where such things are, what mortal shall boast any more 
that he can ward the arrows of the gods from his life? 


with which θίξ might have become ξξ.)---ματἄιζων L, ματάζων r. 


892 f. τίσ ἔτί 


(sic) ποτ᾽ ἐν τοῖσδ᾽ ἀνὴρ θυμῶι βέλη ἐρξεται (sec) | ψυχᾶσ ἀμύνειν L. The later Mss. 
have in some cases θυμῶ or θυμοῦ: a few have ἐν τούτοις (as E), or αὐτοῖς (B), for 
ἐν τοῖσδ᾽.---ΕΟΥ θυμῶι, Hermann restored θεῶν: for ἔρξεται, Musgrave εὔξεται. 





Gorg. 461 Ὁ καθέρξῃς (so Stallb. and 
Herm., with mss.): Aes. 461 B ξυνέρξαν- 
tos: Fol. 285 B ἕρξας. So far as the 
MSS. warrant a conclusion, Attic seems 
to have admitted ép- instead of elp- 7 the 
forms with §. The smooth breathing is 
right here, even if we admit a normal 
distinction between eipyw ‘to shut out’ 
and elpyw ‘to shut in.’ 

891 θίξεται. This conjecture of Blaydes 
seems to me certain. The form occurs 
Eur. Aippol. 1086 κλαίων τις αὐτῶν ap’ 
ἐμοῦ γε θίξεται: Her. 652 εἰ δὲ τῶνδε 
προσθίξει χερί. Hesych. has θίξεσθαι. 
L has έξεται with no breathing. Soph. 
could not conceivably have used such a 
phrase as ἔχεσθαι τῶν ἀθίκτων, fo cling to 
things which should not even be touched. 
He himself shows the proper use of 
ἔχεσθαι in fr. 327 τοῦ ye κερδαίνειν ὅμως 
ἀπρὶξ ἔχονται, ‘still they cling tooth 
and nail to gain’: fr. 26 τὰ μὲν | δίκαι᾽ 
ἐπαίνει τοῦ δὲ κερδαίνειν ἔχου. Some 
explain ἕξεται as ‘abstain’: Od. 4. 422 
σχέσθαι τε Bins λῦσαί Te γέροντα: Her. 6. 
85 ἔσχοντο τῆς ἀγωγῆς. To this there 
are two objections, both insuperable: 
(1) the disjunctive 7,—with which the 
sense ought to be, ‘unless he gain &c.... 
or else abstain’: (2) ματᾷάζων, which could 
not be added to éfera: as if this were 
παύσεται.---ματάζων, acting with rash 
folly: Her. 2. 162 ἀπεματάϊσε, behaved 
in an unseemly manner: Aesch. 4g. 995 
σπλάγχνα δ᾽ οὔτι ματάζει, my heart does 
not vainly forebode. The reason for 
writing ματάζων, not ματάζων, is that the 
form ματαΐζω is well attested (Her., Jo- 
sephus, Hesych., Herodian): while there 
is no similar evidence for ματάζω, though 
the latter form might have existed, being 
related toa stem ματα (μάτη) as δικαζ-ω 
to δικα (δίκη). 

892 τίς ἔτι ποτ᾽... ἀμύνειν; Amid 


such things (if such deeds prevail), who 
shall any longer vaunt that he wards off 
from his life the shafts of the gods? The 
pres. ἀμύνειν, not fut. duvveiy, because 
the shafts are imagined as already as- 
sailing him. ἐν τοῖσδ᾽: 1319: Azz. 38 
el τάδ᾽ ἐν τούτοις. 

893 θεῶν βέλη. The mss. have θυ- 
μῶι, θυμοῦ or θυμῶ: in A over θυμῶι 
βέλη is written τὴν θείαν δίκην. This 
points to the true sense, though it does 
not necessarily presuppose the true read- 
ing. The phrase θυμοῦ βέλη, ‘arrows of 
anger,’ could mean, ‘taunts hurled by an 
angry man’; but, a/one, could ot mean, 
‘the arrows of the: divine wrath.’ The 
readings of the Mss. might have arisen 
either through the ν of θεῶν being written, 
as it often is, in a form resembling μ, 
and w having then been transposed (so 
that θυμῶ would have arisen before 6u- 
wax); or from a gloss θυμοῦ on ψυχᾶς. 
For βέλη cp. Plat. Lege. 873 E πλὴν ὅσα 
κεραυνὸς ἢ TL παρὰ θεοῦ τοιοῦτον βέλος 
ἰόν. 

894 εὔξεται. This conject. of Mus- 
grave (which Blaydes adopts) involves 
only the change of one letter from épée- 
ται; and nothing would have been more 
likely than a change of εὔξεται into ἔρξεται 
if the scribe’s eye or thought had wandered 
to ἔρξεται in 890, especially since the lat- 
teris not obviously unsuited to the general 
sense. But ἔρξεται here is impossible. 
For (1) we cannot render: ‘will keep off 
the shafts from himself, so as to ward 
them from his life’: this would be in- 
tolerable. Nor (2), with Elmsley: ‘who 
will abstain from warding off the shafts 
of the soul (the stings of conscience, 
ψυχᾶς βέλη) from his mind (θυμοῦ)ν᾽ 7.2. 
who will not become reckless? This 
most assuredly is not Greek. εὔξεται, 
on the other hand, gives just the right 


and 
strophe. 


ZOPOKAEOYS 


122 


895 


> \ «ε id ἊΝ La 
ll εἰ yap at τοιαίδε πράξεις τίμιαι, 
12 τί δεῖ με χορεύειν ; 


ἀντ. β. οὐκέτι τὸν ἄθικτον εἶμι γᾶς ἐπ᾽ ὀμφαλὸν σέβων, 

2 οὐδ᾽ ἐς τὸν ᾿Αβαῖσι ναόν, 

8 οὐδὲ τὰν ᾿Ολυμπίαν, 

4 εἰ “μὴ τάδε “χειρόδεικτα 

ὅ πᾶσιν “ἁρμόσει βροτοῖς. 

θ ἀλλ᾽, ὦ κρατύνων, εἴπερ op? ἀκούεις, ᾿ 
7 Zev, πάντ᾽ ἀνάσσων, μὴ λάθοι 
8 
ὃ 


900 


σὲ τάν TE σὰν ἀθάνατον αἰὲν ἀρχάν. 
φθίνοντα γὰρ Λαΐου « παλαίΐίφατα > 
10 θέσφατ᾽ ἐξαιροῦσιν ἤδη, 
11 κοὐδαμοῦ τιμαῖς ᾿Απόλλων ἐμφανής" 
12 ἔρρει δὲ τὰ θεῖα. 


905 


910 


896 After χορεύειν, L has in the same verse πονεῖν ἢ τοῖσ θεοῖσ. These words are 
found in at least four other Mss.,—Pal., M (as corrected), M?, M®: being a corruption 
of a gloss, πανηγυρίζειν τοῖς θεοῖς, found in the Trin. and other Mss. (Campbell, 1. 
xxvii). Dr E. M. Thompson points out that this corruption, hardly possible in 
uncial writing, would have been comparatively easy in minuscule, and regards it as 
indicating that the archetype of L was a minuscule Ms. (Introd. to Facsimile, p. 8.) 
899 ’APaior] Erfurdt wrote ἔλβαισι, on the authority of Arcadius (104. 11). Eusta- 
thius knew both modes of writing it (on //. I. 536, p. 279. 1). 903 ὀρθὸν L, ὄρθ᾽ r. 





sense: ‘If justice and religion are tram- Bacch. 181 δεῖ... Διόνυσον... ὅσον καθ᾽ ἡμᾶς : 


pled under foot, can any man dare to 
boast that he will escape the divine 
wrath?’ 

896 χορεύειν. The words πονεῖν ἢ τοῖς 
θεοῖς added in a few Mss. (including L) 
have plainly arisen from a contracted 
writing of πανηγυρίζειν τοῖς θεοῖς which 
occurs in a few others. This gloss cor- 
rectly represents the general notion of 
χορεύειν, as referring to the χοροί con- 
nected with the cult of Dionysus, Apollo 
and other gods. The χορός was an ele- 
ment so essential and characteristic that, 
in a Greek mouth, the question τί de? με 
χορεύειν ; would import, ‘why maintain 
the solemn rites of public worship?’ Cp. 
Polybius 4. 20 (speaking of the youth of 
Arcadia) μετὰ δὲ ταῦτα τοὺς Φιλοξένου 
καὶ Τιμοθέου νόμους μανθάνοντες (learning 
the music of those masters) πολλῇ φιλο- 
τιμίᾳ χορεύουσι κατ᾽ ἐνιαυτὸν τοῖς Διο- 
νυσιακοῖς αὐληταῖς ἐν τοῖς θεάτροις, οἱ μὲν 
παῖδες τοὺς παιδικοὺς ἀγῶνας, οἱ δὲ νεα- 
νίσκοι τοὺς τῶν ἀνδρῶν λεγομένους. Eur. 


δυνατὸν αὔξεσθαι μέγαν" | ποῖ δεῖ χορεύειν, 
ποῖ καθιστάναι πόδα, | καὶ κρᾶτα σεῖσαι 
πολιόν ; ἐξηγοῦ σύ μοι] γέρων γέροντι, 
Teipecta. The Theban elders need not, 
then, be regarded as momentarily for- 
getting their dramatic part. Cp. 1095 
xXopever Bar. 

897 ἄθικτον: cp. the story of the 
Persian attack on Delphi in 480 B.c. 
being repulsed by the god, who Would 
not suffer his priests to remove the trea- 
sures, φὰς αὐτὸς ἱκανὸς εἶναι τῶν ἑωυτοῦ 
προκατῆσθαι, Her. 8. 36. -πὀμφαλόν: see 
on 480. 

899 τὸν ᾿Αβαῖσι ναόν. The site of 
Abae, not far N. of the modern village 
of Exarcho, was on a hill in the north- 
west of Phocis, between Lake Copais 
and Elateia, and near the frontier of the 
Opuntian Locrians. Her. 8. 33 ἔνθα ἦν 
ἱερὸν ᾿Απόλλωνος πλούσιον, θησαυροῖσί τε 
καὶ ἀναθήμασι πολλοῖσι κατεσκευασμένον" 
ἦν δὲ καὶ τότε καὶ νῦν ἐστὶ χρηστήριον αὖ- 
τόθι' καὶ τοῦτο τὸ ἱερὸν συλήσαντες ἐνέπρη- 


ΟΙΔΙΠΟῪΣ TYPANNOS 


123 


Nay, if such deeds are in honour, wherefore should we join in 


the sacred dance? 


No more will I go reverently to earth’s central and inviolate 
shrine, no more to Abae’s temple or Olympia, if these oracles 
fit not the issue, so that all men shall point at them with the 


finger. 


Nay, king,—if thou art rightly called—Zeus all-ruling, 


may it not escape thee and thine ever-deathless power! 
The old prophecies concerning Laius are fading; already 
men are setting them at nought, and nowhere is Apollo glorified 


with honours ; 


904 πάντ᾽ ἀνάσσων] πάντα λεύσσων B. Arnold.—Addoc L: λάθη τ: 
θέσφατ᾽ L: the three dots meaning that παλαιὰ (written in 


906 φθίνοντα γὰρ λαΐου + 


the margin by a later hand) was to be inserted there. 
a few place παλαιὰ before λαΐου or after θέσφατα.) 


φθίνοντα yap λαΐου παλαιὰ θέσφατ᾽: 


the worship of the gods is perishing. 


λάθῃ Brunck. 


(Most of the later Mss. have 


--παλαίφατα i is the conjecture of Arndt, and of Linwood (who prefixes τὰ to Λαΐου, 


reading ὧν τοιόσδ᾽ for ἐν τοῖσδ᾽ in 892). 


Schneidewin supplied Πυθόχρηστα before 





σαν (the Persians in 480 B.c.). Hadrian 
built a small temple beside the ancient 
ἱερόν, Paus. 10. 35. 3. 

900 τὰν ᾿Ολυμπίαν, called by Pindar 
δέσποιν᾽ ἀλαθείας (Οἱ. 8. 2), because divi- 
nation by burnt offerings (μαντικὴ δι᾽ éu- 
πύρων) was there practised on the altar 
of Zeus by the Iamidae, hereditary μάν- 
τεις (Her. 9. 33): Pind. O/. 6. 7o Ζηνὸς 
ἐπ᾿ ἀκροτάτῳ βωμῷ.. «χρηστήριον θέσθαι 
κέλευσεν (Apollo): | ἐξ οὗ πολύκλειτον καθ᾽ 
Ἕλλανας γένος Ἰαμιδᾶν. 

901 εἰ μὴ τάδε ἁρμόσει, if these things 
(the prophecy that Laius should be slain 
by his son, and its fulfilment) do not come 
_ right (fit each other), χειρόδεικτα πᾶσιν 
βροτοῖς, so as to be signal examples for 
all men. Cp. Azz. 1318 740’ οὐκ ἐπ’ 
ἄλλον βροτῶν ἐμᾶς ἁρμόσει ποτ᾽ ἐξ αἱ- 
τίας, can never be adjusted to another,— 
be vightly charged on him. Prof. Camp- 
bell cites Plat. Soph. 262 C πρὶν ἄν τις 
τοῖς ὀνόμασι τὰ ῥήματα κεράσῃ. τότε δ᾽ 
ἥρμοσέ τε, κιτιλ., where I should suppose 
ἥρμοσε to be transitive : ἥρμοσέ Tis Tots 
ὀνόμασι Ta ῥήματα: if so, it is not paral- 
lel. χειρόδ. only here. 

903 ἀκούεις, audis, alluding chiefly 
to the title Ζεὺς βασιλεύς, Xen. Anad. 3. 
1. 12; under which, after the victory at 
Leuctra in 371 B.C., he was honoured 
with a special festival at Lebadeia in 
Boeotia, Diod. 15. 53. 

904 The subject to λάθοι is not defi- 
nitely τάδε (go2), but rather a motion to 
be inferred from the whole preceding 


sentence, —‘the vindication of thy word,’ 
Elms. cp. Eur. Med. 332 Zed, μὴ λάθοι 
σε τῶνδ᾽ ὃς αἴτιος κακῶν. 

906 After φθίνοντα γὰρ Aatov we 
require a metrical equivalent for θεῶν 
βέλη in 893. ‘The παλαιά in the marg. 
of L and in the text of other Mss. favours 
παλαίφατα, proposed by Linwood and 
Arndt, which suits φθίνοντα : cp. 56r. 
Schneidewin conj. Πυθόχρηστα Λαΐου. 
Aatov, object. gen.: cp. Thuc. 1. 140 
τὸ τῶν Μεγαρέων ψήφισμα (adore them). 

908 ἐξαιροῦσιν, are putting out of ac- 
count. This bold use comes, I think, not 
from the sense of destroying (Xen. Hellen. 
2.2. 19 μὴ σπένδεσθαι ᾿Αθηναίοις ἀλλ᾽ ἐξαι- 
ρεῖν), but from that of setting aside, exs 
cluding from consideration: Plat. Soph. 
249 B τούτῳ τῷ λόγῳ ταὐτὸν τοῦτο ἐκ τῶν 
ὄντων ἐξαιρήσομεν, ‘by this reasoning we 
shall strike this same thing out of the 
list of things which exist.’ Cp. Zheaet. 
162 Ὁ θεοὺς... οὕς ἐγὼ ὁ ἔκ τε τοῦ λέγειν καὶ 
τοῦ γράφειν περὶ αὐτῶν, ὡς εἰσὶν ἢ ὡς οὐκ 
εἰσίν, ἐξαιρῷ The absence of ἃ gen. 
like λόγου for ἐξαιρουσιν is softened by 
φθίνοντα, which suggests ‘fading from 
men’s thoughts.’ 

909 τιμαῖς... ἐμφανής, manifest ἐ7 
honours (modal dat.): 7.¢. his divinity 
is not asserted by the rendering of such 
worship as is due to him. Aesch. Vs 
171 (of Zeus) σκῆπτρον τιμάς τ᾽ ἀποσυ- 
λᾶται. 

910 τὰ θεῖα, ‘religion,’ both faith and 
observance: cp. O. C. 1537. 


2nd anti- 
strophe. 


124 


ZOPOKAEOY2 


ΤΟ: χώρας ἄνακτες, δόξα μοι παρεστάθη 
ναοὺς ἱκέσθαι δαιμόνων, τάδ᾽ ἐν χεροῖν 
στέφη λαβούσῃ καπιθυμιάματα. 
ὑψοῦ γὰρ αἴρει θυμὸν Οἰδίπους ἄγαν 


λύπαισι παντοίαισιν' 


ἔννους τὰ καινὰ τοῖς πάλαι τεκμαίρεται, 
ἀλλ᾽ ἔστι τοῦ λέγοντος, ἢν φόβους λέγῃ. 
ὅτ᾽ οὖν παραινοῦσ' οὐδὲν ἐς πλέον ποιῶ, 
πρὸς σ᾽, ὦ AvKe ἽΛπολλον, ἄγχιστος γὰρ él, 


ἱκέτις ἀφῖγμαι τοῖσδε σὺν “κατεύγμασιν, 


ὅπως λύσιν τιν᾽ ἡμὶν εὐαγῆ mops: 
ὡς νῦν ὀκνοῦμεν πάντες ἐκπεπληγμένον 
κεῖνον βλέποντες ὡς κυβερνήτην νεώς. 


3-9, IK > ε Lo) s i“ / 3 ν 
ap ἂν παρ᾽ ὑμῶν, ὦ Edvor, μάθοιμ᾽ ὅπου 


οὐδ᾽ ὁποῖ ἀνὴρ O15 
920 

ATTEAOS. 
“55 


\ ἴω 4 ’ὕ 9 Σὰ Ν >) ’ 
τὰ τοῦ τυράννου δώματ᾽ ἐστὶν Οἰδίπου ; 
» > Ὗ » 9 
μάλιστα δ᾽ αὐτὸν εἴπατ᾽, 


Aatov.—For Λαΐου, Mekler writes Δαλίου, Nauck Λοξίου. 
ἣν is in erasure, having been corrected (doubtless from ec) either by 


λέγη (not λέγηι. 


3 ΄, Ba 
ει κάτισθ O7TOUV. 


917 Lnowhas ἦν φόβουσ 


the 1st hand itself, or by the first corrector: 7 is written in the form H. There is an 


erasure above ἣν (possibly of 3jv itself, which had been noted as a variant on εὖ. 
H of λέγη is above the line, οἱ having been erased below it. 


The 
Most of the later Mss. have 





911—1085 ἐπεισόδιον τρίτον. A 
messenger from Corinth, bringing the 
news that Polybus is dead, discloses that 
Oedipus was not that king’s son, but a 
Theban foundling, whom the messenger 
had received from a servant of Laius. 
Iocasta, failing to arrest the inquiries of 
Oedipus, rushes from the scene with a 
cry. 

911—928 JIocasta comes forth, bear- 
ing a branch (ἱκετηρία), wreathed with 
festoons of wool (στέφη), which, as a 
suppliant, she is about to lay on the altar 
of the household god, Apollo Avxecos, in 
front of the palace. The state of Oedi- 
pus frightens her. His mind has been 
growing more and more excited. It is 
not that she herself has much fear for the 
future. What alarms her is to see ‘the 
pilot of the ship’ (923) thus unnerved. 
Though she can believe no longer in 
human pavrixn, she has never ceased to 
revere the gods (708); and to them she 


turns for help in her need. 

912 ναοὺς δαιμόνων can only mean 
the public temples of Thebes, as the two 
temples of Pallas and the ᾿Ισμήνιον (20). 
The thought had come to Iocasta that 
she should supplicate the gods; and in 
effect she does so by hastening to the 
altar which she can most quickly reach 
(919). 

613 στέφη : see on 3.--ἐπιθυμιάματα, 
offerings of incense: cp. 4. In 4/. 634, 
where Clytaemnestra comes forth to the 
altar of Apollo προστατήριος, an attendant 
carries θύματα πάγκαρπα, offerings of 
fruits of the earth. AaBovtoy. λαβοῦσαν 
would have excluded a possible ambi- 
guity, by showing that the δόξα had come 
before and not after the wreaths were 
taken up: and for this reason the accus, 
often stands in such a sentence: Xen. 
An. 3. 2. 1 ἔδοξεν αὐτοῖς προφυλακὰς 
καταστήσαντας συγκαλεῖν τοὺς στρα- 
τιώτας. 


OIAITOYS TYPANNO2 125 

Io. Princes of the land, the thought has come to me to 
visit the shrines of the gods, with this wreathed branch in my 
hands, and these gifts of incense. For Oedipus excites his soul 
overmuch with all manner of alarms, nor, like a man of sense, 
judges the new things by the old, but is at the will of the 
speaker, if he speak terrors. 

Since, then, by counsel I can do no good, to thee, Lycean 
Apollo, for thou art nearest, I have come, a suppliant with these 
symbols of prayer, that thou mayest find us some riddance from 
uncleanness. For now we are all afraid, seeing Az affrighted, 
even as they who sce fear in the helmsman of their ship. 


MESSENGER. 


Might I learn from you, strangers, where is the house of the 
king Oedipus? Or, better still, tell me where he himself is—if 
ye know. 


qv... Aéyn (λέγοι T). 920 κατεύγμασιν MSS.: κατάργμασιν Wunder. 926 κά- 
τοισθ᾽ L, with most of the later Mss.: κάτισθ᾽ A. L’s reading may, as Dindorf remarks, 
have prompted the statement of a grammarian in Bachmann’s Avxecdota (vol. 2, 
p- 358. 20), who says that Sophocles used τὸ οἷσθε ἀπὸ τοῦ οἴδατε κατὰ συγκοπήν. 





916 τὰ καινὰ, the prophecies of Tei- 
resias, Tots πάλαι, by the miscarriage of 
the oracle from Delphi: 710f. 

917 τοῦ λέγοντος: Plat. Gorg. 508 D 
εἰμὶ δὲ ἐπὶ τῷ βουλομένῳ, ὥσπερ οἱ ἄτιμοι 
τοῦ ἐθέλοντος, ἄν τε τύπτειν βούληται, κ-τ.λ. 
—as outlaws are at the mercy of the first 
comer: O. C. 752 τοὐπιόντος ἁρπάσαι. 
ἣν φόβους λέγῃ has better Ms. authority 
than εἰ λέγοι, and is also simpler: the 
latter would be an opt. like Az. 520 ἀνδρί 
To. χρεὼν (Ξεχεὴ) | μνήμην προσεῖναι, 
τερπνὸν εἴ τί που πάθοι: cp. 16. 1344: 
Ant. 666. But the statement of abstract 
possibility is unsuitable here. εἰ... λέγῃ 
has still less to commend it. 

918 ὅτε, seeing [(Παί,-- ἐπειδή: Azt. 
170: £7. 38: Dem. or. 1 § 1 ὅτε Toivyy 
οὕτως ἔχει: so ὁπότε Thuc. 2. 60. 

919 Λύκει᾽ "Απολλον: see on Λύκειε 
203. 

920 κατεύγμασιν, the prayers sym- 
bolised by the ixernpia and offerings of 
incense. The word could not mean ‘vo- 
tive offerings.’ Wunder’s conject. katdp- 
ypaciv, though ingenious, is neither need- 
ful nor really apposite. That word is 
used of (a) offerings of first-fruits, pre- 
sented along with the εἰρεσιώνη or harvest- 
wreath, Plut. 7hes. 22: (4) the οὐλοχύται 
or barley sprinkled on the altar and victim 


at the degrzning of a sacrifice: Eur. 7. 7. 
244 χέρνιβάς Te kal κατάργματα. 

921 λύσιν... εὐαγῆ, a solution without 
defilement: 1.6. some end to our anxieties, 
other than such an end as would be put 
to them by the fulfilment of the oracles 
dooming Oedipus to incur a fearful ἄγος. 
For εὐαγὴς λύσις as=one which will 
leave us εὐαγεῖς, cp. Pind. Olymp. 1. 26 
καθαροῦ λέβητος, the vessel of cleansing. 

923 ὡς κυβερνήτην νεώς, not ὡς (ὄντα) 
κυβερν. v., because he is our pilot, but ὡς 
(ὀκνοῖμεν dv) βλέποντες κυβερν. v. ἐκπε- 
πληγμένον : Aesch. Thed. 2 ὅστις φυλάσσει 
πρᾶγος ἐν πρύμνῃ πόλεως | οἴακα νωμῶν, 
βλέφαρα μὴ κοιμῶν ὕπνῳ. 

924 When the messenger arrives, Io- 
casta’s prayer seems to have been im- 
mediately answered by a λύσις εὐαγής 
(921), as regards part at least of the 
threatened doom, though at the cost of 
the oracle’s credit. 

926 μάλιστα denotes what stands 
jirst among one’s wishes: cp. 1466: 
Trach. 799 μάλιστα μέν με θὲς | ἐνταῦθ᾽ 
ὅπου με μή τις ὄψεται βροτῶν᾽ | εἰ δ᾽ οἶκτον 
ἴσχεις, κιτιλ. : Phil. 617 οἴοιτο μὲν μά- 
λισθ᾽ ἑκούσιον λαβών, | εἰ μὴ θέλοι δ᾽, 
ἄκοντα : Ant. 327. ἀλλ᾽ εὑρεθείη μὲν μά- 
λιστ᾽" ἐὰν δέτοι | ληφθῇ τε καὶ μὴ κ.τ.λ. 


ZOPOKAEOY2 


XO. στέγαι μὲν aide, καὐτὸς ἔνδον, ὦ Eve: 
γυνὴ δὲ μήτηρ ἥδε τῶν κείνου τέκνων. 
ΑΓ, ἀλλ ὀλβία τε καὶ ξὺν ὀλβίοις ἀεὶ 
γένοιτ᾽, ἐκείνου y οὖσα παντελὴς δάμαρ. 930 
IO, αὕτως δὲ καὶ σύ γ᾽, ὦ fv ἄξιος γὰρ εἶ 
τῆς εὐεπείας οὕνεκ᾽. ἀλλὰ φράζ᾽ ὅτου 
χρήζων ἀφῖξαι yo τι σημῆναι θέλων. 
eb ἀγαθὰ δόμοις τε καὶ πόσει τῷ σῷ, γύναι. 
10. τὰ ποῖα ταῦτα; παρὰ τίνος δ᾽ ἀφιγμένος: 935 
AI. ἐκ τῆς Κορίνθου. τὸ δ᾽ ἔπος οὑξερῶ τάχα, 
ἤἥδοιο μέν, TOS δ᾽ οὐκ av; ἀσχάλλοις δ᾽ ἴσως. 
10. ri 3 ἐστί; ποίαν δύναμιν ὧδ᾽ ἔχει διπλὴν ; 
AT. τύραννον αὐτὸν οὐπιχώριοι. χθονὸς 
τῆς ᾿Ισθμίας στήσουσιν, ὡς ηὐδάτ᾽ ἐκεῖ. 940 
IO. τί Be οὐχ ὁ πρέσβυς Πόλυβος ἐγκρατὴς ἔτι: 
Ady iit, ἐπεί νιν θάνατος ἐν τάφοις “ἔχει. 
IO. TOS εἶπας ; ἢ τέθνηκε Πόλυβος, « ὦ γέρον ;> 
AI. εἰ μὴ λέγω τἀληθές, ἀξιῶ θανεῖν. 


990 γένοιτ᾽) γένοι᾽ Wecklein. 
hand in L, and then altered to x’ ὦ τι. 
variants. 


9338 x’ worl seems to have been written by the rst 
x@s τι (V, Pal.) and καὶ τί (ΓΤ) were known as 
935 The rst hand in L wrote παρὰ, which an early hand changed to 


πρὸς, the common reading of the late Mss. (but παρὰ i and Pal.).—The δ᾽ after τίνοσ 


in L was added by an early hand, 


943 f£. πῶσ elrac’ ἦ τέθνηκε πόλυβοσ; | εἰ δὲ 





928 γυνὴ δέ. Here, and in 930, 950, 
the language is so chosen as to empha- 
sise the conjugal relation of Iocasta with 
Oedipus. 

930 παντελής, because the wife’s es- 
tate is crowned and perfected by the birth 
of children (928). The choice af the 
word has been influenced by the associa- 
tions of τέλος, τέλειος with marriage. 
Aesch. Zum. 835 θύη πρὸ παίδων καὶ 
γαμηλίου τέλους (the marriage rite): 20. 
214 Ἥρας τελείας καὶ Διὸς πιστώματα: 
schol. on Ar. 7hesm. 973 ἐτιμῶντο ἐν 
τοῖς γάμοις ὡς πρυτάνεις ὄντες τῶν γάμων" 
τέλος δὲ ὁ γάμος: Pindar Mem. 10. 18 
τελεία μήτηρ-Ξ- Ἥρα, who (Ar. Zh. 976) 
κλῇδας γάμου φυλάττει. In Aesch. Ag. 
972 ἀνὴρ τέλειος -- οἰκοδεσπότης : as δόμος 
ἡμιτελὴς (7. 2. 700) refers to a house left 
without its lord: cp. Lucian Dial. Mort. 
§ 19 ἡμιτελῆ μὲν τὸν δόμον καταλιπών, 
χήραν δὲ τὴν νεόγαμον γυναῖκα. 

931 αὔτως (77. 1040 ὧδ᾽ αὔτως ὡς μ᾽ 
ὥλεσε) can be nothing but adverb from 


αὐτός (with Aeolic accent), Ξε ‘in that very 
way’: hence, according to the context, 
(a) simply ‘likewise,’ or (6) in a depre- 
clatory sense, ‘only thus,’—z.e. ‘ineffi- 
ciently,’ ‘vainly.’ The custom of the 
grammarians, to write αὕτως except when 
the sense is ‘vainly,’ seems to have come 
from associating the word with οὗτος, or 
possibly even with airés. For Soph., as 
for Aesch. and Eur., our Mss. on the whole 
favour αὕτως: but their authority cannot 
be presumed to represent a tradition 
older than, or independent of, the gram- 
marians. It is, indeed, possible that 
αὕτως was an instance of old aspiration on 
false analogy,—as the Attic ἡμεῖς (Aeolic 
ἄμμες for doués) was wrongly aspirated 
on the analogy of ὑμεῖς (see Peile, Greek 
and Latin Etymology p. 302, who agrees 
on this with “Curtius). in the absence 
of evidence, however, that αὕτως was a 
like instance, it appears most reasonable 
to write αὔτως. 

932 εὐεπείας, gracious words, = εὐφη- 


ΟΙΙΤΟΥΣ ΟτΤΥΡΆΝΝΟΣ 127 

Cu. This is his dwelling, and he himself, stranger, is within; 
and this lady is the mother of his children. 

ΜΕ. Then may she be ever happy in a happy home, since 
she is his heaven-blest queen. 

Io. Happiness to thee also, stranger! ’tis the due of thy 
fair greeting —But say what thou hast come to seek or to 
tell. 

ME. 
band. 

Io. What are they? And from whom hast thou come? 

ME. From Corinth: and at the message which I will speak 
anon thou wilt rejoice—doubtless; yet haply grieve. 

Io. And what is it? How hath it thus a double potency ? 

ΜῈ. The people will make him king of the Isthmian land, 
as twas said there. 

Io. Howthen? Is the aged Polybus no more in power? 

ΜΕ. No, verily: for death holds him in the tomb. 

Io. How sayest thou? Is Polybus dead, old man ? 

Mer. If I speak not the truth, I am content to die. 


Good tidings, lady, for thy house and for thy hus- 


μὴ | λέγω γ᾽ ἐγὼ τἀληθὲσ, ἀξιῶ θανεῖν LL. The words εἰ δὲ μὴ are in a line by them- 
selves. After πόλυβοσ, and before εἰ, are marks like =. Triclinius conjecturally 
added γέρων after Πόλυβος, and some late Mss. have γέρον, but none (it seems) ὦ γέρον, 
Bothe’s reading. Nauck proposed (1856) πώς εἶπας; 7 τέθνηκεν Οἰδίπου πατήρ; | τέθνηκε 


Πόλυβος" εἰ δὲ μή, ἀξιώ θανεῖν. 


The correction of the first verse is specious; not so 





plas, in this sense only here: elsewhere= 
elegance of diction: Isocrates τὴν evé- 
πειαν ἐκ παντὸς διώκει καὶ τοῦ γλαφυρῶς 
λέγειν στοχάζεται μᾶλλον ἢ τοῦ ἀφελῶς 
(Dionys. Jsocr. 538). 

935 παρὰ τίνος. The change of παρά 
into πρός by an early hand in L is remark- 
able. I formerly received πρός, support- 
ing the phrase by Od. 8. 28 ξεῖνος ὅδ᾽, οὐκ 
015’ ὅστις, ἀλώμενος ἵκετ᾽ ἐμὸν δώ | ἠὲ 
πρὸς ἠοίων ἢ ἐσπερίων ἀνθρώπων. There, 
however, πρός is more natural, as vir- 
tually denoting the geographica: regions 
(cp. Od. 21. 347 πρὸς Ἤλιδος, ‘on the 
side of Elis’), And πρὸς θεῶν wpun- 
μένος (E72. 70) would be parallel only if 
here we had éoraduévos. Questioning, 
then, whether ἀφικνεῖσθαι πρός τινος is 
defensible, I now read παρά, with most 
edd. 

936 τὸ δ᾽ ἔπος, ‘at the word,’ accus. 
of the object which the feeling concerns: 
Eur. £7. 831 τί χρῆμ᾽ ἀθυμεῖς ; 

937 ἀσχάλλοις, from root cex, prop. 


‘not to hold oneself,’ ‘to be impatient,’ 
the opposite of the notion expressed by 
σχο-λή (Curt. Ztym. § 170): the word 
occurs in Her., Xen., Dem.; and in Od. 
2. 193 replaces the epic ἀσχαλάαν. Cp. 
Aesch. Ag. 1049 πείθοι av, εἰ πείθοι", 
ἀπειθοίης δ᾽ tows. 

941 ἐγκρατὴς --εἐν κράτει: cp. ἔναρχος 
=év ἀρχῇ, in office, Appian Bel/, Civ. 
1. I4. 

943 A defective verse, πῶς εἶπας; ἦ 
τέθνηκε Πόλυβος; has been patched up 
in our best Mss. by a clumsy expansion 
of the next verse (see crit. note). The 
γέρων supplied by Triclinius (whence 
some late Mss. have yépov) was plainly a 
mere guess. Nauck’s conj. ἦ τέθνηκεν 
Οἰδίπουν πατήρ ; is recommended (1) by 
the high probability of a gloss Πόλυβος 
on those words: (2) by the greater force 
which this form gives to the repetition of 
the question asked in 941: (3) by the 
dramatic efiect for the spectators. 


128 ZOPOKAEOYS 

10. ὦ πρόσπολ᾽, οὐχὶ δεσπότῃ τάδ᾽ ὡς τάχος 945 
μολοῦσα λέξεις | ὦ θεῶν μαντεύματα, 

fe. ἵν᾽ ἐστέ: τοῦτον Οἰδίπους πάλαι τρέμων 
τὸν ἄνδρ᾽ ἔφευγε μὴ κτάνοι: καὶ νῦν ὅδε 
ΤΣ τῆς τύχης ὄλωλεν οὐδὲ τοῦδ᾽ ὕπο. 

Ol. ἵλτατον γυναικὸς Ἰοκάστης κάρα, 950 
τί μ᾽ ἐξεπέμψω δεῦρο τῶνδε δωμάτων ; 

10. ἄκουε τἀνδρὸς τοῦδε, καὶ σκόπει κλύων 
τὰ σέμν᾽ ἵν᾽ ἥκει τοῦ θεοῦ μαντεύματα. 

Ol. οὗτος δὲ Tis ποτ᾽ ἐστὶ καὶ τί μοι λέγει; 

ΤῸ} Yen τῆς Κορώνθου, πατέρα τὸν σὸν ἀγγελῶν 955 
ὡς οὐκέτ᾽ ὄντα Πόλυβον, ἀλλ᾽ ὀλωλότα. 

ΟἹ" τί φής, ξέν᾽; αὐτός μοι σὺ σημάντωρ γενοῦ. 

AP: εἰ τοῦτο πρῶτον δεῖ μ᾽ ἀπαγγεῖλαι σαφώς, 
εὖ ἴσθ᾽ ἐκεῖνον θανάσιμον βεβηκότα. 

ΟΙ. πότερα δόλοισιν, ἢ νόσου ξυναλλαγῇ ; : 960 

AT. σμικρὰ παλαιὰ σώματ᾽ εὐνάζει ῥοπή. 

ΟἹ: νόσοις ὁ τλήμων, ὡς ἔοικεν, ἔφθιτο. 

AI. Kat τῷ μακρῷ ve “συμμετρούμενος χρόνῳ. 

Ol. φεῦ φεῦ, τί Ont ἄν, ὦ γύναι, σκοποῖτό τις 
τὴν Πυθόμαντιν ἑστίαν, ἢ τοὺς ἄνω 965 
κλάζοντας ὄρνεις, ὧν ὑφηγητῶν ἐγὼ 

that of 944. Mekler rejects both vv. 950 Two of the later mss. (M, A) 


have nilarns for ‘loxdorns,—either a mere error, or a conjecture. 
a corrector has changed this to σημάντωρ. 


Ist hand in L wrote onunvac: 


957 The 





946 ὦ θεῶν μαντεύματα. LTocasta’s 
scorn is pointed, not at the gods them- 
selves, but at the μάντεις who profess to 
speak in their name. The gods are wise, 
but they grant no πρόνοια to men (978). 
Cp.. 712: 

947 ἵν᾽ ἐστέ: tva=br ἐνταῦθα, ‘to 
think that ye have come to this !?: cp. 
1311.---τοῦτον τὸν ἄνδρα... .τρέμων ἔφευγε, 
he feared and avoided this man, μὴ κτάνοι 
(αὐτόν). 

949 πρὸς τῆς τύχης, 2.4. in the course 
of nature, and not by the special death 
which the oracle had foretold. Cp. 977. 

951 ἐξεπέμψω, the midd. as in ἐκκα- 
λεῖσθαι (see on 597), μεταπέμπεσθαι, etc., 
the act. being properly used of the sum- 
moner or escort: see on στελοῦντα (860). 

954 τί μοι λέγει; ‘ what does he tell 
(of interest) for me?’ (not ‘what does he 


say to me?’: nor ‘what, pray, does he 
say?’). 

956 ws: see on 848. 

957 σημάντωρ is, I think, unquestion- 
ably right. A is among the Mss. which 
have it, and in several it is explained by 
the gloss μηνυτής. That the word was 
not unfamiliar to poetical language i in the 
sense (‘indicator,’ ‘informant’) which it 
has here, may be inferred from Axnthol. 
6. 62 (Jacobs 1. 205) κυκλοτερῆ μόλιβον, 
σελίδων σημάντορα πλευρῆς, the pencil 
which makes notes in the margin of pages: 
Nonnus 37. 55! σημάντορι φωνῇ. On the 
other hand, σημήνας γενοῦ could mean 
nothing but ‘ place yourself in the position 
of having told me,’ and could only be ex- 
plained as a way of saying, ‘tell me at 
once.’ But such a use of γενέσθαι with 
aor. partic. would be unexampled. The 


OIAITOYS TYPANNOZ 


129 


Io. O handmaid, away with all speed, and tell this to thy 


master ! 


O ye oracles of the gods, where stand ye now! 


denis 


is the man whom Oedipus long feared and shunned, lest he 
should slay him; and now this man hath died in the course of 


destiny, not by his hand. 
OE. 
forth from these doors ? 


[Exter OEDIPUS. 


Iocasta, dearest wife, why hast thou summoned me 


Io. Hear this man, and judge, as thou listenest, to what the 
awful oracles of the gods have come. - 

ΟΕ. And he—who may he be, and what news hath he for me? 

Io. He is from Corinth, to tell that thy father Polybus 
lives no longer, but hath perished. 


OE. How, stranger? 
ME. 
that he is dead and gone. 


Let me have it from thine own mouth. 
If I must first make these tidings plain, know indeed 


OE. By treachery, or by visit of disease ? 
Me. A light thing in the scale brings the aged to their rest. 


OE. 


Ah, he died, it seems, of Bicktiess ? 


ME. Yea, and of the long years that he had told. 


OE. Alas, alas! 


Why, indeed, my wife, should one look to 


the hearth of the Pythian seer, or to the birds that scream above 


our heads, on whose showing I 


The first corrector (5) had written in the margin, yp. onudvrwp. The later Mss. also 


have onudvrwp (but σημήνας T). 
Hartung: ἔξισθ᾽ Meineke. 


959 εὖ ἴσθ᾽ MSS.: 
966 ὄρνις MSS. The Attic form ὄρνεις (L. Dindorf, 7 hes. 


σάφ᾽ ἴᾳθ᾽ Porson: κάτισθ᾽ 





only proper use of it is made clear by such 
passages as these: Az. 588 μὴ προδοὺς 
ἡμᾶς γένῃ, do not make yourself guilty of 
having betrayed us: PAz/. 772 μὴ σαυτόν 
θ᾽ ἅμα | kdue...xrelvas γένῃ, do not make 
yourself guilty of having slain both your- 
self and me. 

959 εὖ ἴσθ᾽. Dionys. Hal. τ. 41 thus 
quotes a verse from the Προμηθεὺς Λυό- 
μενος οὗ Aesch. (Nauck fr. 193. 2) ἔνθ᾽ οὐ 
μάχης εὖ οἶδα καὶ θοῦρός περ ὦν, where 
Strabo p. 183 gives σάφ᾽ οἷδα: and so 
Pors. here would write σάφ᾽ ἴσθι. Butthe 
immediately preceding σαφῶς i is decisive 
against this. Soph. had epic precedent, 
Jl. 1. 385 εὖ εἰδὼς ἀγόρευε, etc. Cp. 1071, 
lod ἰού. --θανάσιμον βεβηκότα: Az. 516 
μοῖρα... | καθεῖλεν “Αἰιδου θανασίμους οἰκή- 
Topas: Phil. 424 Oavav...ppoddos. 

960 ξυναλλαγῇ: see ON 34. 

961 σμικρὰ ῥοπή, /eve momentum: 
the life is conceived as resting in one 
scale of a nicely poised balance: diminish 
the weight in the other scale ever so little, 
and the inclination (ῥοπή), though due to a 


rE ΤΡ 


slight cause (σμικρα), brings the life to the 
ground (εὐνάζει). Plat. Rep. 556 E ὥσπερ 
σῶμα νοσῶδες μικρᾶς ῥοπῆς ἔξωθεν δεῖται 
προσλαβέσθαι πρὸς τὸ κάμνειν,.. οὕτω δὴ 
καὶ ἡ κατὰ ταὐτὰ ἐκείνῳ διακειμένη πόλις 
ἀπὸ σμικρᾶς προφάσεως... νοσεῖ, 

968 Yes, he died of infirmities (νόσοις 
ἔφθιτο), and of the long years (τῷ μακρῷ 
χρόνῳ, causal dat.), in accordance with 
their term (cupperpodpevos, sc. αὐτοῖς, lit. 
‘commensurably with them’): the part. 
being nearly equiv. to συμμέτρως, and ex- 
pressing that, if his years are reckoned, 
his death cannot appear premature. Cp. 
1113, and Ant. 387 ποίᾳ ξύμμετρος mpov- 
βην τύχῃ; ‘seasonably for what hap?’ 

964 f. σκοποῖτο, midd. as 77. 296.— 
τὴν IL. ἑστίαν -- τὴν Πυθοῖ μαντικὴν ἑστίαν, 
as Apollo himself is Πυθόμαντις, 2.4. ὁ 
Πυθοῖ μάντις, Aesch. Cho. 1030: cp. Πυ- 
θόκραντος, Πυθόχρηστος, Πυθόνικος. ἑστίαν, 

as Ο. C. 413 Δελφικῆς ἀφ᾽ ἑστίας : Eur. 
πὰ 461 Φοιβήιος...γᾶς | μεσόμφαλος ἑστία. 

966 κλάζοντας, the word used by Tei- 
resias of the birds when their voice (φθόγ- 


9° 


130 


( 


ΣΟΦΟΚΛΕΟΥΣ 


A » le \ > 4 ε Ἂ \ 
κτενεῖν ἔμελλον πατέρα Tov ἐμόν; ὁ δὲ θανὼν 
’ 4 XN ~ > \ > 7Q> > 4 
κεύθει κάτω δὴ γῆς" ἐγὼ δ᾽ ὅδ᾽ ἐνθάδε 
ἄψαυστος ἔγχους' εἴ τι μὴ τὠμῷ πόθῳ 


κατέφθιθ᾽. οὕτω δ᾽ ἂν θανὼν εἴη ᾿ἕ 


9 A 
ἐμου. 


970 


Ν 2 Σ , A ΄ 
τὰ δ᾽ οὖν παρόντα συλλαβὼν θεσπίσματα 
A td 4 » > > 4 
κεῖται παρ᾽ “Avon Πόλυβος a&v’ οὐδενός. 


IO. 
OI. 
IO. 
Ol. 
IO. 


οὔκουν ἐγώ σοι ταῦτα προὔλεγον πάλαι; 
ηὔδας" ἐγὼ δὲ τῷ φόβῳ παρηγόμην. 

μή νυν ἐτ᾽ αὐτῶν μηδὲν ἐς θυμὸν βάλῃς. 

καὶ πῶς τὸ μητρὸς λέκτρον οὐκ ὀκνεῖν με δεῖ; 
s 3 “ἡ lovee) ¥ κκ \ “A 4 

τί δ᾽ ἂν φοβοῖτ᾽ ἀνθρωπος, ᾧ τὰ τῆς τύχης 


075 


“ 4 > > \ > Ν 4 
-KpaTel, T POVOLa ὃ εστιν οὐδενὸς σαφής; 


5. 2224) is supported by the Ravenna Ms. in Ar. Av. 717, 1250, 1610: and in Eur. 


Hipp. 1059 by M (cod. Ven. Marc. 471) and the 1st hand in V. 


967 κτανεῖν L, 


and almost all the later Mss. : it may, indeed, be an accident that one, at least, of them 


(V°) has κτενεῖν, which Elmsley required. 
omitted 67, but added it above the line. 


968 After κάτω, the rst handin L had 


No suspicion of δή is warranted by the fact 
that one or two of the later Mss. (Trin., I) omit it. 


Dindorf, who once conjectured 





vos) had ceased to be clear to him, Azz. 
IOOI κακῷ | κλάζοντας οἴστρῳ καὶ BeBap- 
βαρωμένῳ.---ὧν ὑφηγητῶν sc. ὄντων, guibus 
indicibus: 1260 ὡς ὑφηγητοῦ τινος: Ο. C. 
1588 ὑφηγητῆρος οὐδενὸς φίλων. In these 
instances the absence of the part. is soft- 
ened by the noun which suggests the 
verb; but not so in O. C. 83 ws ἐμοῦ μόνης 
πέλας. 

967 κτενεῖν. κτανεῖν, which the MSS. 
give, cannot be pronounced positively 
wrong; but it can hardly be doubted that 
Soph. here wrote κτενεῖν. If κτανεῖν is 
right, it is the only aor. infin. after μέλλω 
in Soph., who has the fut. infin. 9 times 
(ZZ. 359, 379, 538: 42. 925, 1027, 1287: 
Ant, 458: Phil. 483, 1084): and the 
pres. infin. g times (Z/. 305, 1486: Az. 
443: O. T. 678, 1385: O. C. 1773: 77. 
79, 750: Phil. 409). Aeschylus certainly 
has the aor. in P. V. 625 μήτοι με κρύψῃς 
τοῦθ᾽ ὅπερ μέλλω παθεῖν. Excluding the 
Laconic ἰδῆν in Ar. Lys. 117, there are 
but two instances in Comedy, Av. 366 τί 
μέλλετ᾽ ---ἀπολέσαι, and Ach. 1159 μέλ- 
λοντος λαβεῖν. Cp. W. G. Rutherford, 
New Phrynichus pp. 420—425, and 
Goodwin, Greek Moods and Tenses § 23. 
2. The concurrence of tribrachs in the 
4th and sth places gives a semi-lyric 
character which suits the speaker’s agi- 
tation. 


968 κεύθει, is hidden. Az. 635"Adg 
κεύθων. In Zr. 989 σιγῇ κεύθειν may be 
regarded as transitive with a suppressed 
acc., ‘to shroud (thy thought) in silence.’ 
Elsewhere κεύθω is always trans., and 
only the perf. xéxev@a intransitive. —8y 
here nearly=76n: cp. Ant. 170 ὅτ᾽ οὖν 
@AovTo... | ἐγὼ κράτη δὴ...ἔχω. 

969 ἀψαυστος ---οὐ ψαύσας: cp. ἀφό- 
βητος 885(η.): Her.8.124 ἄκριτος, without 
deciding: id. 9. 98 ἄπιστος, mistrustful ; 
O. C. 1031 πιστός, trusting (n.): Phil. 687 
ἀμφίπληκτα ῥόθια, billows beating around: 
Tr. 446 μεμπτός, blaming: Eur. Mec. 1117 
ὕποπτος, suspecting. Cp. note on ἀτλητῶν 
515.—el τι μὴ, an abrupt afterthought :— 
unless perchance: see on 124.--τὠμῷ 
πόθῳ: cp. 797: Od. 11. 202 σὸς...πόθος, 
longing for thee. 

970 εἴη ᾽ξ: cp. 1075: Phil. 467 πλεῖν 
μὴ ᾿ξ ἀπόπτου. ἐξ, as dist. from ὑπό, is 
strictly in place here, as denoting the 
ultimate, not the proximate, agency. 

971 τὰ δ᾽ οὖν παρόντα: but the ora- 
cles as they stand, at any rate (δ᾽ οὖν, 
669, 834), Polybus has carried off with 
him, proving them worthless (d§’ ov- 
Sevds, tertiary predicate), and is hidden 
with Hades.—ra παρόντα, with empha- 
sis: even supposing that they have been 
fulfilled in some indirect and figurative 
sense, they certainly have not been ful- 


OIAITOYS ΤΥΡΆΝΝΟΣ 131 
was doomed to slay my sire? But he is dead, and hid already 
beneath the earth; and here am I, who have not put hand to 
spear.— Unless, perchance, he was killed by longing for me: 
thus, indeed, I should be the cause of his death. But the oracles 
as they stand, at least, Polybus hath swept ‘with him to his rest 
in Hades: they are worth nought. 

Io. Nay, did I not so foretell to thee long since? 

ΟΕ. Thou didst: but I was misled by my fear. 

Io. Now no more lay aught of those things to heart. 

OE. But surely I must needs fear my mother’s bed ? 

Io. Nay, what should mortal fear, for whom the decrees of 
Fortune are supreme, and who hath clear foresight of nothing? 


κάτωθεν, has replaced κάτω δή. Nauck proposes κεύθει κάτω γῆς. Οἰδίπους (instead 
of ἐγὼ) δ΄. Cobet and Blaydes, κάτω κέκευθε γῆς. 970 οὕτω δ᾽] οὕτω γ᾽ Wecklein. 
976 καὶ πῶσ τὸ μρσ λέχοσ οὐκ ὀκνεῖν με δεῖ 1.. The first corrector has written λέκτρον 
over λέχοσ. A and others have λέκτρον in the text. Dindorf would place λέχος after 
ὀκνεῖν (or after δεῖ, Bergk reads λέχος - ἔτ᾽ -- οὐκ ὀκνεῖν με δεῖ, and so Wecklein. 
I prefer to read λέκτρον, with Blaydes, Wolff, Campbell, Kennedy, and others. 





filled to the letter. The oracle spoke of 
bloodshed (φονεύς, 794), and is not satis- 
fied by κατέφθιτο ἐξ ἐμοῦ in the sense just 
explained.—ovAdAaBoy is a contemptuous 
phrase from the language of common life: 
its use is seen in Aristophanes Pluz. 1079 
νῦν δ᾽ ἄπιθι χαίρων συλλαβὼν τὴν μεί- 
ρᾶακα, now be off—with our blessing and 
the girl: Av. 1469 ἀπίωμεν ἡμεῖς συλ- 
λαβόντες τὰ πτερά, let us pack up our 
feathers and be off: Soph. has it twice 
in utterances of angry scorn, O. C. 1383 
σὺ δ᾽ épp ἀπόπτυστός τε κἀπάτωρ ἐμοῦ | 
κακῶν κάκιστε, τάσδε συλλαβὼν ἀράς, 
begone...and take these curses with thee: 
Phil. 577 ἔκπλει σεαυτὸν συλλαβὼν ἐκ 
τῆσδε γῆς, ‘hence in thy ship—pack from 
this land!’ 

974 ηὔδας instead of προὔλεγες : see 
on 54. 

975 νυν, enforcing the argument in- 
troduced by οὔκουν (973), is clearly better 
than the weak viv.—és θυμὸν βάλῃς : Her. 
7. 51 és θυμὸν βαλεῦ καὶ τὸ παλαιὸν ἔπος : 
8. 68 καὶ τόδε ἐς θυμὸν βαλεῦ, ws κ.τ.λ.: 
1. 84 ἰδὼν. «τῶν τινα Λυδῶν καταβάντα... 
ἐφράσθη καὶ ἐς θυμὸν ἐβάλετο. Theactive 
in the Βίος ‘Opnpou ὃ 30 ἐς θυμὸν ἔβαλε 
τὸ ῥηθέν. In ΞΔ, 1347 οὐδέ γ᾽ ἐς θυμὸν 
φέρω is not really similar. 

977 ᾧ, ‘for whom,’ in relation to 
whom: not, ‘in whose opinion.’—rd τῆς 
τύχης is here somewhat more than a 
mere periphrasis for ἡ τύχη, since the 
plur. suggests successive incidents. τύχη 


does not here involve denial of a divine 
order in the government of the world, 
but only of man’s power to comprehend 
or foresee its course. Cp. Thuc. 5. 104 
πιστεύομεν TH μὲν τύχῃ ἐκ τοῦ θείου μὴ 
ἐλασσώσεσθαι. Lysias or. 24 ὃ 22 οὗ 
μόνου μεταλαβεῖν ἡ τύχη μοι ἔδωκεν ἐν τῇ 
πατρίδι, the only privilege which Fortune 
(4.6. my destiny) has permitted me to 
enjoy in my country. 

978 πρόνοια. Bentley on Phalaris 
(xvul, Dyce ii. 115) quotes Favorinus in 
Laertius Plat. § 24 as saying that Plato 
πρῶτος ἐν φιλοσοφίᾳ... ὠνόμασε... θεοῦ πρό- 
voay. Bentley takes this to mean that 
Plato was the first to use πρόνοια of divine 
providence (not merely of human fore- 
thought), and cites it in proof that Pha- 
laris Ep. 3 (=40 Lennep) ἕως ἂν ἡ διοι- 
κοῦσα πρόνοια THY αὐτὴν ἁρμονίαν τοῦ κόσ- 
μου φυλάττῃ is later than Plato. Lennep, 
in his edition of Phalaris (p. 158), puts 
the case more exactly. The Stoics, not 
Plato, first used πρόνοια, wethout further 
qualification, of a divine providence. 
When Plato says τὴν τοῦ θεοῦ... πρόνοιαν 
(Zim. 30 6), προνοίας θεῶν (44 6), the 
phrase is no more than Herodotus had 
used before him, 3. 108 τοῦ θείου ἡ mpo- 
νοίη. The meaning of Favorinus was 
that Plato first established in phivosophy 
the conception of a divine providence, 
though popular language had known such 
a phrase before. Note that in O. C. 
1180 πρόνοια τοῦ Oeof=‘ reverence for 


9--2. 


[32 


ZOPOKAEOY2 


eas , “ ν δύ , 
εἰκῇ ΚΑΤ ΤΟΥ ζῆν, fe UVVQALTO TUS. 


σὺ δ᾽ εἰς Ta “μητρὸς μὴ φοβοῦ νυμφεύματα:' 980 
πολλοὶ γὰρ ἤδη κἀν ὀνείρασιν βροτῶν 
μητρὶ ξυνευνάσθησαν. ἀλλὰ ταῦθ᾽ ὅτῳ 
' παρ᾽ οὐδέν ἐστι, ῥᾷστα τὸν βίον φέρει. 

ΟΙ. καλῶς ἅπαντα ταῦτ᾽ ἂν ἐξείρητό. σοι, 
εἰ μὴ ᾿κύρει ζῶσ᾽ ἡ τεκοῦσα' νῦν δ᾽, ἐπεὶ 985 
ζῇ, πᾶσ᾽ ἀνάγκη, κεὶ καλῶς λέγεις, ὀκνεῖν. 

ΙΟ. καὶ μὴν μέγας γ᾽ ὀφθαλμὸς οἱ πατρὸς τάφοι. 

ΟΙ. μέγας, ξυνίημ᾽" ἀλλὰ τῆς ζώσης φόβος. 

AT. ποΐας δὲ καὶ γυναικὸς ἐκφοβεῖσθ' ὕπερ; : 

ΟἹ. Μερόπης, γεραιέ, Πόλυβος ἧς ᾧκει μέτα. 990 

AT. ti ὁ ἐσ ἐκείνης ὑμὶν ἐς φόβον φέρον ; 

ΟΙ. θεήλατον μάντευμα δεινόν, ὦ Eve. | “ι 

AT. Ἢ pyrov; 0 οὐχὶ θεμιτὸν ἄλλον εἰδέναι; 

ΟΙ. μάλιστά γ᾽" εἶπε γάρ με Λοξίας ποτὲ 
χρῆναι μυγῆναι μητρὶ τἠμαυτοῦ, τό τε 995 


πατρῷον αἷμα χερσὶ ταῖς ἐμαῖς ἑλεῖν. 
ὧν οὕνεχ᾽ ἡ Κόρινθος ἐξ ἐμοῦ πάλαι 


987 μέγας γ}] γ᾽ was restored by Porson (Eur. Phoen. 1638) : 


‘Ita postulat metrum.. 


idemque coniecit nescio quis in editione Londinensi a. 1746, sed neglexit Brunckius.’ 
The loss of γ᾽ in the Mss. may have arisen from μέγας having been written short, wey4 
(as it is in A), when γ᾽, following it, might easily have been mistaken for a dittographia 





the god’: in Eur. Phoen. 637 a man acts 
θείᾳ προνοίᾳ =‘ τ inspired foresight’ : 
in Xen. Mem. 1. 4. 6 Tpovorrrixais = not, 
sifovidentially: but simply, ‘with fore- 
thought.’ 

979 εἰκῆ: cP. Plat. Gorg. 503 E οὐκ 
εεἰκῆ ἐρεῖ, ἀλλ᾽ ἀποβλέπων πρός τι (with 
“some definite object in view). π--κράτιστον 

ὅπως δύναιτο. Cp. Ant. 666 ἀλλ᾽ ὃν 
πόλις στήσειε τοῦδε χρὴ κλύειν: where χρὴ 
κλύειν = δικαίως ἂν κλύοι. So here, though 
ἐστί (not ἦν) must be supplied with κρά- 
τιστον, the whole ρῆγαδβε Ξ εἰκῇ κράτιστον 
ἄν τις ζῴη. Xen. Cyr. 1.6.19 τοῦ.. αὐτὸν 
λέγειν ἃ μὴ σαφῶς εἰδείη φείδεσθαι Sei= 
ὀρθῶς av φείδοιτο. 

980 φοβοῦ. φοβεῖσθαι εἴς τιΞεῖο have 
fears regarding it: 77. 1211 εἰ φοβεῖ πρὸς 
τοῦτο: O. C. 1119 μὴ θαύμαζε πρὸς τὸ λι- 
παρές. 

981 κἀν ὀνείρασιν, in dreams a/so 
(as well as in this oracle) ; and, as such 
dreams have proved vain, so may this 
oracle. Soph. was prob. thinking of the 


story in Her. 6. 107 that Hippias had 
such a dream on the eve of the battle of 
Marathon, and interpreted it as an omen 
of his restoration to Athens. Cp. the 
story of a like dream coming to Julius 
Caesar on the night before he crossed 

the Rubicon (Plut. Caes. 32, Suet. 7). 

983 παρ᾽ οὐδέν: Ant. 34 τὸ πρᾶγμ᾽ 
ἄγειν οὐχ ὡς παρ᾽ οὐδέν. 

984 ἐξείρητο: the ἐξ- glances at her 
blunt expression of disbelief, not her frank 
reference to a horrible subject. 

987 ὀφθαλμὸς : the idea is that of a 
bright, sudden comfort: so Tr. 203 De- 
ianeira calls on her household to rejoice, 
ὡς ἄελπτον ὄμμ᾽ ἐμοὶ [ φήμης ἀνασχὸν 
τῆσδε νῦν καρπούμεθα (the unexpected 
news that Heracles has returned). More 
often this image denotes the ‘darling’ of 
a family (Aesch. Cho. 934 ὀφθαλμὸς οἴκων), 
or a dynasty that is ‘the light’ of a land 
(Σικελίας δ᾽ ἔσαν | ὀφθαλμός, Pind. OV. 
2. 9: ὁ Βάττου παλαιὸς ὄλβος...πύργος 
ἄστεος, ὄμμα τε φαεννότατον | ἕξένοισι, 


ΟἸΔΊΠΟΥΣ: ΤΥΡΆΝΝΟΣ 133 
"Tis best to live at random, as one may. But fear not thou 
touching wedlock with thy mother. Many men ere now have 
so fared in dreams also: but he to whom these things are as 
nought bears his life most easily. 

Or. All these bold words of thine would have been well, 
were not my mother living; but as it is, since she lives, I must 
needs fear—though thou sayest well. 

Io. Howbeit thy father’s death is a great sign to cheer us. 


Or. Great, I know; but my fear is of her who lives. 

ΜΕ. And who is the woman about whom ye fear? 

ΟΕ. Merope, old man, the consort of Polybus. 

ΜΕ. And what is it in her that moves your fear ? 

OE. A heaven-sent oracle of dread import, stranger. 

ΜΕ. Lawful, or unlawful, for another to know ? 

OE. Lawful, surely. Loxias once said that I was doomed 


to espouse mine own mother, and to shed with mine own hands 
my father’s blood. Wherefore my home in Corinth was long kept 


by a copyist inattentive to metre. 


998 ἢ οὐ θεμιτὸν MSS. 
οὐχὶ θεμιτὸν : Johnson, 7 οὐ θεμιστὸν : see comment. 
Laud. 54) has ἄλλοις for ἄλλον, but prob. by a mere error. 


Brunck conjectured ἢ 
One of the later mss. (Bodl. 
Blaydes conjectured ἢ οὐκ 





Pyth. 5. 51). Not merely (though this 
notion comes in) ‘a great help to seeing’ 
that oracles are idle (δήλωσις ὡς τὰ μαν- 
τεύματα κακῶς ἔχει, schol.). A certain 
hardness of feeling appears in the phrase: 
Tocasta was softened by fear for Oedipus 
and the State: she is now elated. 

989 καὶ with ἐκφοβεῖσθε; 772, 851. 

991 ἐκείνης, what is there belonging 
to her, iz her (attributive gen.): Eur. 7. 
A. 28 οὐκ ἄγαμαι ταῦτ᾽ ἀνδρὸς ἀριστέως-.---- 
ἐς φόβον φέρον, tending to fear: cp. 510. 

992 θεήλατον, sent upon us by the 
gods: cp. 255.- 

993 The mss. having οὐ θεμιτὸν, the 
question is between οὐχὶ θεμυτὸν and οὐ 
θεμιστὸν. The former is much more 
probable, since θεμιτός is the usual form, 
found in Attic prose, in Eur. (as Or. 97 
σοὶ δ᾽ οὐχὶ θεμιτόν), and in Soph. O. C. 
1758 ἀλλ᾽ οὐ θεμιτὸν κεῖσε μολεῖν. On the 
other hand θεμιστός is a rare poet. form, 
found once in Pindar (who has also θε- 
purés), and twice in the lyrics of Aesch. 
Had we ἄλλῳ, the subject of θεμιτὸν would 
be μάντευμα: the accus. ov shows 
θεμιτὸν to be impersonal, as in Eur. Or. 
97, Pind. Pyth. 9. 42 οὐ θεμιτὸν ψεύδει 
θιγεῖν. 

996 τὸ πατρῷον αἷμα ἑλεῖν is strictly 
“ἴο achieve (the shedding of) my father’s 


blood.’ Classical Greek had no such 
phrase as αἷμα χεῖν or ἐκχεῖν in the sense 
of ‘to slay.’ αἱρεῖν is to make a prey of, 
meaning ‘to slay,’ or ‘to take,’ accord- 
ing to the context (77. 353 Εὔρυτόν θ᾽ 
ἕλοι | τὴν θ᾽ ὑψίπυργον Οἰχαλίαν). Cp. 
fr. 731 ἀνδρὸς αἷμα συγγενὲς | κτείνας, 
which is even bolder than this, but simi- 
lar, since here we might have had simply 
τὸν πατέρα ἑλεῖν, ‘to slay my father’: 
Eur. Or. 284 εἴργασται δ᾽ ἐμοὶ | μητρῷον 
αἷμα, 1 have wrought the murder of a 
mother. 

997 The simplest view of ἡ Κόρινθος 
ἐξ ἐμοῦ ἀπῳκεῖτο is, as Whitelaw says, 
that it means literally, ‘Corinth was 
lived-away'-from by me,’—being the pas- 
sive of ἐγὼ ἀπῴκουν τῆς Κορίνθου. It 
is thus merely one of those instances in 
which a passive verb takes as subject 
that which would stand in gen. or dat. 
as object to the active verb: cp. the 
passive καταγελώμαι, καταφρονοῦμαι, κα- 
ταψηφίζομαι, ἐπιβουλεύομαι, etc. [I for- 
merly took it to be passive of ἐγὼ ἀπῴ- 
κουν τὴν Κόρινθον, ‘I inhabited C. only 
at a distance,’—a paradoxical phrase like 
ἐν σκότῳ ὁρᾶν (1273).] ἀποικεῖν is a com- 
paratively rare word. Eur. has it twice 
(H. F. 557: 7. A. 680: in both with 
gen., ‘to dwell far from’): Thuc. once 


134 


ZOPOKAEOYS 


μακρὰν ἀπῳκεῖτ᾽ εὐτυχῶς μέν, ἀλλ᾽ ὅμως 
τὰ τῶν τεκόντων. ὄμμαθ᾽ ἥδιστον βλέπειν. 


ΑΓ. ἢ γὰρ τάδ᾽ ὀκνῶν κεῖθεν ἦσθ' ἀπόπτολις ; ; 1000 
OF. πατρός τε χρήζων μὴ φονεὺς εἶναι, γέρον. 
APS srt Ont ἐγὼ οὐχὶ τοῦδε τοῦ nega ao, ava€, 
ἐπείπερ εὔνους ἦλθον, ἐξελυσάμην; 
Ol. καὶ μὴν χάριν γ᾽ ἂν ἀξίαν λάβοις ἐμοῦ. 
AT. καὶ μὴν μάλιστα τοῦτ᾽ ἀφικόμην, ὅπως 1005 
σοῦ πρὸς δόμους ἐλθόντος εὖ πράξαιμί Tt. 
OI. aA ovmor εἶμι τοῖς φυτεύσασίν Ὗ ὁμοῦ. 
AI. ὦ “παῖ, καλῶς εἶ δῆλος οὐκ εἰδὼς τί δρᾷς. 
Ol. TOS, ὦ γεραιέ; πρὸς θεῶν δίδασκέ με. 
AT. εἰ τῶνδε φεύγεις οὕνεκ᾽ εἰς οἴκους μολεῖν. ΙΟΙΟ 
OL. ταρβῶν γε μή μοι DoiBos ἐξέλθῃ σαφής. ἢ 
ΑΓ, ἡ μὴ μίασμα τῶν υτευσάντων λάβῃς; 
ΟΙ. τοῦτ᾽ αὐτό, πρέσβυ, τοῦτό μ᾽ εἰσαεὶ φοβεῖ. 
AT. ap’ οἶσθα δῆτα πρὸς δίκης οὐδὲν τρέμων ; 
OI. πῶς δ᾽ οὐχί, παῖς γ᾽ εἰ τῶνδε γεννητῶν ἔφυν; 1015 
AT. ὁθούνεκ᾽ ἣν σοι Πόλυβος οὐδὲν ἐν γένει. 
ΟΙ. πῶς εἶπας; οὐ γὰρ Πόλυβος ἐξέφυσέ με; 
AT. οὐ μᾶλλον οὐδὲν τοῦδε τἀνδρός, ἀλλ᾽ ἴσον. 
ἄλλοισι θεμιτὸν εἰδέναι, which had also occurred to the present ed. 1001 πατρός 


Te MSS. 
by Elmsley and Blaydes. 


Hermann proposed, but afterwards recalled, marpés ye, a conjecture adopted 
1002 ἐγὼ for ἔγωγ᾽ Porson. 


The 1st hand in L wrote 


ἔγωγ᾽ οὐχὶ, but the x! has been partly erased. The later Mss. have either ἔγωγ᾽ οὐχὶ 





with μακρὰν (3. 55) aps Xen. once (Oecon. 
μὴ 6),—both absol., as=‘to dwell Gar’: 
as prob. Theocr. 15. a (reading ὦ μέλ᾽ 
ἀποικεῖς with Meineke): Plato once thus 
(Legg. 753 A), and twice as=to emigrate 
(ἐκ Τόρτυνος, Legg. 708 A, és Θουρίους, 
Euthyd. 271 6): in which sense Isocr. 
also has it twice (or. 4 § 122, or. 6 § 84) ς 
Pindar once (with accus. of motion to 
a place), Pyth. 4. 258 Καλλίσταν ἀπῴ- 
κησαν, they went and settled at Callista. 
998 f. εὐτυχῶς, because of his high 
fortunes at Thebes.—rav τεκόντων ΞΞ- τῶν 
γονέων: Eur. Hipp. 1081 τοὺς τεκόντας 
ὅσια δρᾶν, sand oft. : cp. H. F. 975 βοᾷ 
δὲ μήτηρ, ὦ τεκών [ΞΞ ὦ πάτερ], τί δρᾷς ; 
1000 ἀπόπτολιξβ, exile, as O. Ο. 
208. 
1001 πατρός te. Sothe MSS., rightly. 
It is the fear of Oed. regarding his 
mother by which the messenger’s atten- 


tion has been fixed. In explaining this, 
Oed. has indeed mentioned the other 
fear as to his father; but in v. 1000, 7 
yap τάδ᾽ ὀκνῶν, the messenger means: 
*So this, then, was the fear about her 
which kept you away?’—alluding to his 
own question in 991. As the speaker’s 
tone seems to make light of the cause, 
Oed. answers, ‘and that further dread 
about my father which I mentioned.’ 
πατρός ‘ye is unsuitable, since it would 
imply that this was his so/e fear. 

1002 ἐγὼ οὐχὶ : synizesis: see on 332 
ἐγὼ οὔτ᾽. 

1008 ἐξελυσάμην : the aor. implies, 
‘why have I not done it already?’ ζΖ.6. 
‘why do I not do it at once? Aesch, 
P. V. 447 τί δὴτ᾽ ἐμοὶ ζῆν κέρδος, ἀλλ᾽ 
οὐκ ἐν τάχει | ἔρριψ᾽ ἐμαυτὴν τῆσδ᾽ ἀπὸ 
στύφλου πέτρας; 

1004 καὶ μὴν, properly ‘however’; 


OIAITOYS ΤΥΡΆΝΝΟΣ 


135 


by me afar; with happy event, indeed,—yet still ‘tis sweet to 


see the face of parents. 


ΜΕ. Was it indeed for fear of this that thou wast an exile 


from that city? 
OE. 
my sire. 


And because I wished not, old man, to be the slayer of 


Mr. Then why have I not freed thee, king, from this fear, 
seeing that I came with friendly purpose ? 


OE. 
ME. 


Indeed thou shouldst have guerdon due from me. 
Indeed ’twas chiefly for this that I came—that, on thy 


return home, I might reap some good. 
ΟΕ. Nay, I will never go near my parents. 
ΜΕ. Ah my son, ’tis plain enough that thou knowest not 


what thou doest. 
OE. How, old man? 
ME. 


For the gods’ love, tell me. 
If for these reasons thou shrinkest from going home. 


OE. Aye, I dread lest Phoebus prove himself true for me. 
ME. Thou dreadest to be stained with guilt through thy 


parents? 


OE. Even so, old man—this it is that ever affrights me. 


ME. 


Dost thou know, then, that thy fears are wholly vain? 


OE. How so, if I was born of those parents? 
ME. Because Polybus was nothing to thee in blood. 


OE. What sayest thou? 


Was Polybus not my sire? 


ΜΕ. No more than he who speaks to thee, but just so much. 


(as A), or ἔγωγ᾽ οὐ, which Brunck retained. 


If that, however, had been genuine, οὐ 


could hardly have been corrupted into οὐχί, whereas the opposite corruption would 


easily have caused the change of ἐγὼ into ἔγωγ᾽. 


1011 ταρβῶ L: ταρβῶν τ and 





here, like our ‘well indeed’ (if you would 
do so). The echoing καὶ μὴν of 1005 
expresses eager assent. Cp. Azz. 221. 

1005 τοῦτ᾽ ἀφικόμην : see on 788. 

1008 καλῶς, pulchre, belle, tho- 
roughly, a colloquialism, perh. meant 
here to be a trait of homely speech: cp. 
Alciphron Z/. 1. 36 πεινήσω τὸ καλόν 
(‘I shall be fine and hungry’): Aelian 
Ep. 2 ἐπέκοψε τὸ σκέλος πάνυ χρηστῶς 
(‘in good style’). 

1011 With Erfurdt I think that ταρ- 
Bov is right; not that ταρβῶ could not 
stand, but Greek idiom distinctly favours 
the participle. Amz. 403 KP. 7 καὶ Evins 
καὶ λέγεις ὀρθῶς ἃ φής; PT. ταύτην γ᾽ 
ἰδὼν θάπτουοαν. 10. 517 ΑΝ... ἀδελφὸς 
ὥλετο. ΚΡ. πορθῶν γε τήνδε γῆν. Plat. 
Symp. 164 E εἶπον οὖν ὅτι.. ἥκοιμι.---κα- 
λῶς (v. 1. καλῶς y’), ἔφη, ποιῶν. Cp. 1130 


ξυναλλάξας. ---ἐξέλθῃ; cp. 1182 ἐξήκοι 
σαφῆ, come true. 

1018 Cp. 77. 408 τοῦτ᾽ αὔτ᾽ ἔχρῃζον, 
τοῦτό σου μαθεῖν. 

1014 πρὸς δίκης, as justice would 
prompt, ‘justly.’ πρὸς prop. =‘ from the 
quarter of,’ then ‘on the side of’: Thuc. 
3. 59 οὐ πρὸς τῆς ὑμετέρας δόξης...τάδε, 
not in the interest of ‘your reputation: 
Plat. Gorg. 459 C ἐάν τι ἡμῖν πρὸς λόγου 
ἢ, ‘if it is in the interest of our dis- 
cussion.” ep. 470 C ovdév...dmd τρόπου 
Aéyes* Spa δὴ καὶ ef τόδε πρὸς τρόπου 
λέγω, ‘correctly.” Theophr. Char. 30 
(=26 in my ist ed. p. 156) πρὸς τρόπου 
πωλεῖν, to sell on reasonable terms. 

1016 ἐν γένει: [Dem.] or. 47 § 70 οὐκ 
ἔστιν ἐν γένει got ἡ ἄνθρωπος, compared 
with § 72 éuol δὲ οὔτε γένει προσῆκεν. 


136 ZLTOPOKAEOYS 

Ol. καὶ πῶς ὁ φύσας ἐξ ἴσου τῷ μηδενί; 

AT. ἀλλ᾽ οὐ σ᾽ ἐγείνατ᾽ οὔτ᾽ ἐκεῖνος οὐτ᾽ ἐγώ. 1020 
OI. ἀλλ᾽ ἀντὶ τοῦ δὴ παῖδά μ᾽ ὠνομάζετο: 

AT. δώρόν ποτ᾽, ἴσθι, τῶν ἐμῶν χειρῶν λαβών. 

OI. καθ᾽ ὧδ᾽ am ἄλλης χειρὸς ἔστερξεν μέγα; 

AI. ἡ γὰρ πρὶν αὐτὸν ἐζξέπεισ᾽ ἀπαιδία. 

OI. σὺ δ᾽ ἐμπολήσας ἢ τυχών μ᾽ αὐτῷ δίδως ; 1025 
AT. εὑρὼν ναπαίαις ἐν Κιθαιρῶνος πτυχαῖς. 

OI. ὠὡδοιπόρεις δὲ πρὸς τί τούσδε τοὺς τόπους : 


AP. 
Ol. 
"Ὁ δὲ 
ΟΙ. 
1 
ΟἹ. 
AY: 
Ol. 


Erfurdt. 


1025 τυχών Bothe: τεκών MSS. 


ἐνταῦθ᾽ ὀρείοις ποιμνίοις ἐπεστάτουν. 
ποιμὴν γὰρ ἦσθα κἀπὶ θητείᾳ πλάνης; 
σοῦ δ᾽, ὦ τέκνον, σωτήρ γε τῷ τότ᾽ ἐν χρόνῳ, 1030 
τί δ᾽ ἄλγος ἴσχοντ᾽ " ἀγκάλαισι λαμβάνεις ; 

ποδῶν ἂν ἄρθρα μαρτυρήσειεν τὰ σά. 
οἴμοι, τί τοῦτ᾽ ἀρχαῖον ἐννέπεις κακόν; 
λύω σ᾽ ἔχοντα διατόρους ποδοῖν ἀκμάς. 
δεινόν γ᾽ ὄνειδος σπαργάνων ἀνειλόμην. 


pulps 
xe 1035 


(Hermann, however, cites that cor- 


rection as made by C. Foertsch, Obss. crit. in Lysiae orationes, p. 12 sq.)—n κιχών μέ 


που δίδως Heimsoeth. 


Elmsley, with one later ms. (I). 
σοῦ γ΄. See comment. 


1028 ἐπεστάτουν. 
t. Wecklein conj. ἐπιστατῶν (Ars Soph. emend. p. 12). 


In L the second e has been made from 
1030 σοῦ y L. σοῦ δ᾽ 


Hermann once proposed σοῦ 7’, but reverted to 
1031 τί δ᾽ ἄλγοσ ἴσχοντ᾽ ἐν καιροῖσ λαμβάνεισ L. 


ἴσχοντ᾽ 


has been corrected from ἴσχων, and the rst hand has also written ἔσχοντ᾽ in the left 





1019 τῷ μηδενί, dat. of ὁ μηδείς, one 
who is szch as to be of account (in respect 
of consanguinity with me),—the generic 
use of μή (cp. 397, 638). 

1023 ἔστερξεν, came to love me (in- 
gressive aor.): cp. If n.—am’ ἄλλης 
XErpos sc. λαβών. 

1025 ἐμπολήσας...ἢ τυχών: 2.4. ‘Did 
you buy me, or did you light upon me 
in the neighbourhood of Corinth?’ Oed. 
is not prepared for the Corinthian’s reply 
that he had found the babe on C7thaeron. 
ἐμπολήσας: cp. the story of Eumaeus 
(Od. 15. 403—483) who, when a babe, 
was carried off by Phoenician merchants 
from the wealthy house of his father in 
the isle Syria, and sold to Laertes in 
Ithaca: the Phoenician nurse says to the 
merchants, τόν κεν ἄγοιμ᾽ ἐπὶ νηός, ὁ δ᾽ 
ὑμῖν μυρίον ὦνον | ἄλφοι, ὅπῃ περάσητε 
kar’ ἀλλοθρόους ἀνθρώπους. τυχών is 
answered by εὑρών (1026) as in 973 
προὔλεγον by ηὔδας. Cp. 1039. The 
τεκών of the Mss. is absurd after vv. 1016 
—1020. The man has just said, ‘Poly- 


bus was no more your father than I am’; 
Oed. is anxiously listening to every word. 
He could not ask, a moment later, ‘ Had 
you bought me, or were you my father?’ 

1026 The fitness of the phrase varratats 
πτυχαῖς becomes vivid to anyone who 
traverses Cithaeron by the road ascending 
from Eleusis and winding upwards to the 
pass of Dryoscephalae, whence it descends 
into the plain of Thebes. 

1029 ἐπὶ θητείᾳ, like ἐπὶ μισθῷ Her. 
5. 65 etc. θητεία, labour for wages, 
opp. to δουλεία : Isocr. or. 14 ὃ 48 πολ- 
Aovs μὲν.. δουλεύοντας, ἄλλους δ᾽ ἐπὶ θη- 
τείαν ἰόντας. πλάνης, roving in search of 
any employment that he can find (not 
merely changing summer for winter pas- 
tures, 1137). The word falls lightly from 
him who is so soon to be ὁ πλανήτης Οἰδί- 
mous (O. C. 3). 

1080 σοῦ δ᾽. With the σοῦ γ᾽ of 
most Μ88. : ‘Yes, and thy preserver’ (the 
first γε belonging to the sentence, the 
second to σωτήρ) Cp. Her. 1. 187 μὴ 
μέντοι γε μὴ σπανίσας γε ἄλλως ἀνοίξῃ: 


eS δ 


OIAITOYS TYPANNOS 


OE. 
nought to me? 


137 


And how can my sire be level with him who is as 


ME. Nay, he begat thee not, any more than I. 
OE. Nay, wherefore, then, called he me his son? 


ME. 
hands of yore. 
OE. 


from another’s hand? 
ME. 


Know that he had received thee as a gift from my 
And yet he learned to love me so dearly, who came 


Yea, his former childlessness won him thereto. 


ΟΕ. And thou—hadst thou bought me or found me by 
chance, when thou gavest me to him? 


ME. Found thee in Cithaeron’s winding glens. 

ΟΕ. And wherefore wast thou roaming in those regions? 
ΜΕ. I was there in charge of mountain flocks. 

ΟΕ. What, thou wast a shepherd—a vagrant hireling ? 

ΜῈ. But thy preserver, my son, in that hour. 

Or. And what pain was mine when thou didst take me in 


thine arms ? 


ΜΕ. The ankles of thy feet might witness. 


Or. Ah me, why dost thou speak of that old trouble ? 
ΜΕ. I freed thee when thou hadst thine ankles pinned 
together. 


OE. Aye, ‘twas a dread brand of shame that I took from 


my cradle. 


margin. The later Mss. have ἐν καιροῖς we λαμβάνεις (Pal.), or ἐν κακοῖς με λαμβάνεις 
(as A), or ἐν κακοῖς λαμβάνεις (as M).—For ἐν καιροῖς Theodor Kock conjectures 
ἀγκάλαις με: Verrall, ἴσχον τἀγκάλισμα: Wunder, ἐν carp με (Weil ἐν καλῴ σὺ): 
Blaydes, ἢ κακόν με: W. W. Walker, ἐν χεροῖν με: Dindorf, ἐν νάπαις με: Nauck, ἐν 
σκάφαισι (‘in cunis’): Wecklein, ἐν δέοντι: F. W. Schmidt, τί δ᾽; ἐσχάτοις ὄντ᾽ ἐν 
κακοῖς με λαμβάνεις ;--Ἰ had thought of éyxupwy, ‘when you lighted on me’ (a verb 





where the second ye belongs to σπανί- 
gas. There is no certain example of a 
double ye in Soph. which is really similar. 
With σοῦ δ᾽: ‘ But thy preserver’: the ye 
still belonging to σωτήρ, and δὲ opposing 
this thought to that of v. 1029. For δέ 
yecp. Aesch. 4g. 938 AT. φήμη γε μέντοι 
δημόθρους μέγα σθένει. KA. ὁ δ᾽ ἀφθόνη- 
ros γ᾽ οὐκ ἐπίζηλος πέλει. ‘True, but....’ 
The gentle reproof conveyed by δέ ye is 
not unfitting in the old man’s mouth: 
and a double ye, though admissible, is 
awkward here. 

1081 τί δ᾽ ἄλγος x.7r.A. And in what 
sense wast thou my owrijp? The ἐν ka- 
κοῖς of the later Mss. is intolerably weak: 
‘what pain was I suffering when you 
found me 7m trouble?’ The ἐν καιροῖσ' 
of L (found also, with the addition of 


με, in one later Ms., Pal.) seems most 
unlikely to have been a corruption of ἐν 
κακοῖς. Among the conjectures, ἀγκάλαις 
με (Kock), or, better, ἀγκάλαισι, is perh. 
most probable; being slightly nearer the 
letters than Verrall’s ingenious ἴσχον ray- 
κάλισμα. (For the dat. ἀγκάλαις without 
ἐν, cp. Eur. J, 7. 289, etc.) Such con- 
jectures as ἐν δέοντι (Wecklein), ἐν καλῷ 
(Wunder), presuppose that ἐν καιροῖς was 
a gloss: but it is more probable that it 
was a corruption. 

1035 δεινόν ye in comment, as Ph. 
1225, Al. 341, Az. 1127.—omapydvev, 
‘ from my swaddling clothes’: 2.4. ‘from 
the earliest days of infancy’ (cp. Ovid 
Heroid. 9.22 £t tener in cunis tam Love 
dignus eras). The babe was exposed a 
few days after birth (717). £¢. 1139 


138 ΣΟΦΟΚΛΕΟῪΣ 
AT. wor ὠνομάσθης ἐκ τύχης ταύτης ὃς εἶ. 
ΟΙ. ὦ πρὸς θεών, πρὸς μητρός, ἢ πατρός ; φράσον. 
JIE οὐκ oto’ Se δὲ ταῦτ᾽ ἐμοῦ λῷον φρονεῖ. 
Ol. 7 yep παρ᾽ ἄλλου i ἔλαβες οὐδ᾽ αὐτὸς τυχών ; 
AT οὔκ, ἀλλὰ ποιμὴν ἄλλος ἐκδίδωσί μοι. 1040 
OI. tis οὗτος ; ἢ κάτοισθα δηλῶσαι λόγῳ; 
A TOV Λαΐου δήπου τις ὠνομάζετο. 
ΟΙ. ἢ τοῦ τυράννου τῆσδε γῆς πάλαι ποτέ; 
AT. μάλιστα" τούτου τἀνδρὸς οὗτος ἦν Bornp. 
Ol. ἡ καστ᾽ ἔτι ζών οὗτος, ὥστ᾽. ἰδεῖν ἐμέ; 1045 
AE: ὑμεῖς γ᾽ ἄριστ᾽ εἰδεῖτ᾽ av οὐπιχώριοι. 
ΟΙ. ἔστιν τις ὑμῶν τῶν παρεστώτων πέλας 
ὅστις κάτοιδε τὸν Botnp ὃν ἐννέπει, 
εἴτ᾽ οὖν ἐπ᾽ dy pav εἴτε κἀνθάδ᾽ εἰσιδών ; 
σημήναθ᾽, ὡς ὁ καιρὸς ηὑρῆσθαι τάδε. 1050 
XO. οἶμαι μὲν οὐδέν᾽ ἄλλον ἢ τὸν ἐξ ἀγρῶν, 
ὃν κἀμάτευες πρόσθεν εἰσιδεῖν: ἀτὰρ 
ἥδ᾽ ἂν τάδ᾽ οὐχ ἥκιστ᾽ ἂν Ἰοκάστη λέγοι. 
ΟΙ. γύναι, νοεῖς ἐκεῖνον OVTW ἀρτίως 
μολεῖν ἐφιέμεσθα; τόνδ᾽ οὗτος λέγει; 1055 


used in £7. 863; cp. 1025, 1039 τυχών). 
on 68. 


1050 ηὑρῆσθαι εὑρῆσθαι 1,.. See comment. 
1055 μολεῖν ἐφιέμεσθα: τὸν θ᾽ οὗτος λέγει; L. Most of the later Mss. have τόν θ᾽, 





οὔτε.. «πυρὸς | ἀνειλόμην ... ἄθλιον βάρος. 
Some understand, ‘I was furnished with 
cruelly dishonouring tokens of my birth,’ 
δεινῶς ἐπονείδιστα σπάργανα, alluding to 
a custom of tying round the necks of 
children, when they were exposed, little 
tokens or ornaments, which might after- 
wards serve as means of recognition (cre- 
pundia, monumenia): see esp. Plautus 
Rudens 4. 4. 111—126, Hpidicus 5. 1.34: 
and Rich s. v. Crepundia, where a wood- 
cut shows a statue of a child with a string 
of crepundia hung over the right shoulder. 
Plut. Zhes. 4 calls such tokens γνωρίσματα. 
In Ar. Ach. 431 the σπάργανα of Tele- 
phus have been explained as the tokens 
by which (in the play of Eur.) he was re- 
cognised ; in his case, these were paxwpara 
(431). But here we must surely take 
σπαργάνων with ἀνειλόμην. 

1036 ὥστε assents and continues: 
‘(yes,) and so...’—6s εἶ, 2.4. Οἰδίπους : 
see on 718. 

1037 πρὸς μητρές, ἢ πατρός; sc. 


ὄνειδος ἀνειλόμην (1035): ‘was it at the 
hands of mother or father (rather than at 
those of strangers) that I received such 
a brand?’ The agitated speaker follows 
the train of his own thoughts, scarcely 
heeding the interposed remark. He is 
not thinking so much of his parents’ pos- 
sible cruelty, as of a fresh clue to their 
identity. Not: ‘was I so named by 
mother or father?’ The zame—even if it 
could be conceived as given before the 
exposure—is not the sting; and on the 
other hand it would be forced to take 
‘ named’ as meaning ‘doomed to bear the 
name.’ 

1044 βοτήρ: cp. 837, 761. 

1046 ἐεἰδεῖτ᾽ =eldelnre, only here, it 
seems: but cp. elre=elyre Od. 21. 195 
(doubtful in Amz. 215). εἰδεῖμεν and el- 
μεν occur in Plato (Rep. 581 E, Theaet. 
147 A) as well as in verse. In Dem. or, 
14 § 27 καταθεῖτε is not certain (κατά- 
Gore Baiter and Sauppe): in or. 18 § 324 
he has ἐνθείητε. Speaking generally, we 


ΟΙΔΙΠΟῪΣ ΤΥΡΑΝΝΟΣ 


130 


ME. Such, that from that fortune thou wast called by the 


name which still is thine. 
OE. 

father’s ? 
ME. 

that-than 1 


Speak ! 


ΟΕ. What, thou hadst me from another? 


light on me thyself ? 
ME. 
OE. 
ME. 
OE. 
ME. 

herd. 
OE. 
ME. 
ORF; 


Who was he? 


Oh, for the gods’ love—was the deed my mother’s or 


I know not; he who gave thee to me wots better of 


Thou didst not 


No: another shepherd gave thee up to me. 

Art thou in case to tell clearly? 

I think he was called one of the household of Latus. 
The king who ruled this country long ago? 

The same: ’twas in his service that the man was a 


Is he still. alive, that I might see him? 
Nay, ye folk of the country should know best. 
Is there any of you here present that knows the herd 


of whom he speaks—that hath seen him in the pastures or the 


town? Answer! 
be finally revealed. 


The hour hath come that these things should 


CH. Methinks he speaks of no other than the peasant whom 
thou wast already fain to see; but our lady Iocasta might best 


tell that. 


Or. Lady, wottest thou of him whom we lately summoned? 
Is it of him that this man speaks ? ᾿ 


which was taken as=6v θ᾽ (thus in B there is a gl. ὅντινα, and in Bodl. Laud. 54 ὅν). 





may say that the contracted termination 
-elev for -einoay is common to poetry and 
prose; while the corresponding contrac- 
tions, -e@uev for -elnuev and -cire for -είητε, 
are rare except in poetry. 

1049 οὖν with the first ere, as £7. 
199, 560: it stands with the second 
above, go, 271, PA. 345.—ém’ ἀγρῶν: 
Od. 22. 47 πολλὰ μὲν ἐν μεγάροισιν...πολ- 
a δ᾽ ἐπ᾽ ἀγροῦ: (cp. Ο. C. 184 ἐπὶ ξένης, 
El. 1136 κἀπὶ γῆς ἄλλης ) the usual Attic 
phrase was ἐν ἀγρῷ or κατ᾽ ἀγρούς. 

1050 ὁ καιρὸς : for the art., cp. [Plat.] 
Axiochus 364 B viv ὁ καιρὸς ἐνδείξασθαι 
τὴν ἀεὶ θρυλουμένην πρὸς σοῦ codiay.— 
ηὑρῆσθαι: Bellermann (objecting to the 
tense) reads εὑρέσθαι, citing Az. 1023 
(where, as usual, the aor. midd.=‘to 
gain’): but the perf. is right, and for- 
cible, here; it means, ‘to be discovered 
once for all.’ For the form, cp. 546n. 
Isocr. or. 15 ὃ 295 τῶν δυναμένων λέγειν ἢ 
παιδεύειν ἡ πόλις ἡμῶν δοκεῖ γεγενῆσθαι 


διδάσκαλος, to be the established teacher. 

1051 Supply ἐννέπειν (αὐτόν), not 
ἐννέπει. The form οἶμαι, though often 
parenthetic (as 77. 536), is not less com- 
mon with infin. (Plat. Gorg. 474 Α οἷον 
ἐγὼ οἶμαι δεῖν εἶν αι), and Soph. often so 
has it, as Z/, 1446. 

1058 av...dv: see on 862. 

1054 νοεῖς Ξε γοιι wot of,’ the man— 
i.¢. you understand to whom I refer. We 
need not, then, write εἰ κεῖνον for ἐκεῖνον 
with A. Spengel, or νοεῖς; ἐκεῖνον with 
Blaydes, who in 1055, reading τόνδ᾽, has 
a comma at ἐφιέμεσθα. Cp. 859. 

1055 τόνδ᾽ is certainly right: τόν θ᾽ 
arose, when the right punctuation had 
been lost, from a desire to connect λέγει 
with ἐφιέμεσθα. Dindorf, however, would 
keep τόν θ᾽: ‘know ye him whom we 
summoned and him of whom this man 
speaks?’ 2.5. ‘Can you say whether the 
persons are identical or distinct?’ But 
the language will not bear this. 


ΣΟΦΟΚΛΕΟΥΣ 
ae 


τί, δ᾽ ὅντιν᾽ εἶπε; μηδὲν ἐντραπῇς. τὰ δὲ 
ῥηθέντα βούλου μηδὲ μεμνῆσθαι μάτην. 
οὐκ av γένοιτο Toul, ὅπως ἐγὼ λαβὼν 
σημεῖα τοιαῦτ᾽ οὐ φανῶ τοὐμὸν γένος. 


IO. μὴ πρὸς θεῶν, εἴπερ τι τοῦ σαυτοῦ βίου 1060 
κήδει, ματεύσῃς τοῦθ᾽" ἅλις νοσοῦσ᾽ ἐγώ. 

ΟἹ: θάρσει" σὺ μὲν γὰρ οὐδ᾽ ὃ ἐὰν τρίτης ἐγὼ 
μητρὸς φανῶ τρίδουλος ἐκφανεῖ κακή. 

10: ὅμως πιθοῦ μοι, λίσσομαι." μὴ δρᾶ τάδε. 

OI. οὐκ ἀν πιθοίμην μὴ οὐ τάδ᾽ ἐκμαθεῖν σαφῶς. 1065 

10. καὶ μὴν φρονοῦσά γ᾽ εὖ τὰ λῷστά σοι λέγω. 

ΟῚ. τὰ λῷστα τοίνυν ταῦτά ΠῚ ἀλγύνει πάλαι. 
ye’ “10. ὦ δύσποτμ᾽, εἴθε μήποτε γνοίης ὃς εἶ. 

Ol. ἄξει τις ἐλθὼν δεῦρο τὸν βοτηρά μοι; ἐφ 5 

ταύτην δ᾽ ἐάτε πλουσίῳ χαίρειν γένει. 1070 


IO. tov iov, δύστηνε' τοῦτο γάρ o ἔχω 
μόνον προσειπεῖν, ἄλλο δ᾽ οὔποθ' ὕστερον, 
ΧΟ. τί ποτε βέβηκεν, Οἰδίπους, ὑπ᾽ ἀγρίας 


ἀξασα λύπης ἡ γυνή; 


But a few, at least, have τόνδ᾽ (M, M? ist hand, A). 
1062 θάρσει Brunck: θάρρει L.—oud’ ἂν ἐκ τρίτης 
In 1, @ has its accent from the rst hand, but its breathing from another. 


νοσοῦσ᾽ ἐγώ schol. (on 1056). 
ἐγὼ MSS. 


δέδοιχ᾽ ὅπως 


1061 νοσοῦσ᾽ ἔχω Μϑ85.: 


Hermann restored οὐδ᾽ ἐὰν τρίτης éyw (in which Tournier suggests ἀπὸ for ἐγὼ) : but 





1056 τί δ᾽ 6 ὅντιν᾽ εἶπε: Aesch. P. V. 
765, θέορτον ἢ βρότειον [γάμον γαμεῖ]; εἰ 
ῥητόν, φράσον. ΠΡ. τί δ᾽ ὅντιν᾽; Ar. Av. 

97 σὺ δ᾽ εἶ τίς ἀνδρῶν ; Μ. ὅστις εἴμ᾽ ἐγώ; 


έτων. Plat. EHuthyphr. 2 Β τίνα ypa- 
φήν σε γέγραπται; ΣΏ. ἥντινα ; οὐκ 
ἀγεννῆ. 


1058 Since οὐκ ἔστιν ὅπως, οὐκ ἂν 
γένοιτο ὅπως mean ‘there is, there could 
be found, πὸ way in which,’ τοῦθ᾽ is 
abnormal; yet it is not incorrect : ‘this 
thing could not be attained, namely, a 
mode in which, etc. Cp. the mixed 
constr. in Az. 378 οὐ γὰρ γένοιτ᾽ ἂν ταῦθ᾽ 
ὅπως οὐχ ὧδ᾽ ἔχειν (instead of ἕξει). 

1060 Since the answer at 1042, Io- 
casta has known the worst. But she is 
still fain to spare Oedipus the misery of 
that knowledge. Meanwhile he thinks 
that she is afraid lest he should prove 
to be too humbly born. The tragic power 
here is masterly. 

1061 ἅλις (εἰμὶ) νοσοῦσ᾽ ἐγώ instead 


οἵ ἅλις ἐστὶ τὸ νοσεῖν ἐμέ: cp. 1368: AZ. 
76 ἔνδον ἀρκείτω μένων : 16. 635 κρείσσων 
γὰρ “Αιδᾳ κεύθων (n.): Her. 1. 37 ἀμείνω 
ἐστὶ ταῦτα οὕτω ποιεύμενα : Dem. or. 4 ὃ 34 
οἴκοι μένων, βελτίων : 1586. or. 2 § 7 ἱκανὸς 
γὰρ αὐτὸς ἔφη ἀτυχῶν εἶναι: Athen. 435 D 
χρὴ πίνειν, ᾿Αντίπατρος γὰρ ἱκανός ἐστι 
νήφων 

1062 For the genitive τρίτης μητρὸς 
without ἐκ, cp. El. 341 οὖσαν πατρός, 
366 καλοῦ | τῆς μητρός. τρίτης μητρὸς 
τρίδουλος, thrice a slave, sprung from the 
third (servile) mother: ¢.¢. froma mother, 
herself a slave, whose mother and grand- 
mother had also been slaves. No com- 
mentator, so far as I know, has quoted 
the passage which best illustrates this: 
Theopompus fr. 277 (ed. Miiller 1. 325) 
Πυθονίκην...ἣ Baxxldos μὲν ἦν δούλη τῆς 
αὐλητρίδος, ἐκείνη δὲ Σινώπης τῆς Op¢r- 
TNS). «ὥστε γίνεσθαι μὴ μόνον τρίδουλον 
ἀλλὰ καὶ τρίπορνον αὐτήν. [Dem.] or. 
58 8 17 εἰ γὰρ ὀφείλοντος αὐτῷ τοῦ πάπ- 


OIAITTOY2 TYPANNOZ 


Io. Why ask of whom he spoke? 
not a thought on what he said.. 


OE. 


[41 


Regard it not...waste 


. twere idle. 
It must not be that, with such clues in my grasp, I 


should fail to bring my birth to light. 
Io. For the gods’ sake, if thou hast any care for thine own 


life, forbear this search! 


OE. Be of good courage; 


My anguish is enough. 
though I be found the son of 


servile mother,—aye, a slave by three descents,—¢hou wilt not 


be proved base-born. 


Io. Yet hear me, I implore thee: do not thus. 


OE. 


I must not hear of not discovering the whole truth. 


Io. Yet I wish thee well—I counsel thee for the best. 
Or. These best counsels, then, vex my patience. 


Io. Ill-fated one! 
thou art! 


Mayst thou never come to know who 


OE. Go, some one, fetch me the herdsman hither,—and 
leave yon woman to glory in her princely stock. 

Io. Alas, alas, miserable !—that word alone can I say unto. 
thee, and no other word henceforth for ever. 


[Ske rushes into the palace. 


CH. Why hath the lady gone, Oedipus, in a transport of 


afterwards preferred οὐδ᾽ ἂν el ’x τρίτης ἐγώ, 


reads. Dindorf, οὐδ᾽ ἐὰν ἐγὼ ’« τρίτης. 


changed it to δρᾶν by writing v above the line, also adding ane subscript. 


wild grief? I misdoubt, 


which (with the omission of ’«) Campbell 
1064 μὴ δρᾶ List hand; a late hand has. 


1070 xai- 


pew] χλιδᾶν Nauck, from schol. τρυφᾶν, ἐναβρύνεσθαι: which words, however, manifestly: 





που πάλαι... διὰ τοῦτ᾽ οἰήσεται δεῖν ἀπο- 
φεύγειν ὅτι πονηρὸς ἐκ τριγονίας ἐστίν 
..y ‘if, his grandfather having formerly 
been a debtor,...he shall fancy himself 
entitled to acquittal because he is a rascal 
of the third generation.” Eustathius Od. 
1542. 50 quotes from Hippénax ᾿Αφέω 
τοῦτον τὸν ἑπτάδουλον (Bergk fr. 75), 2.6. 
‘seven times a slave.’ For the force of 
τρι-, cp. also τριγίγας, τρίπρατος (thrice- 
sold,—of a slave), τριπέδων (a slave who 
has been thrice in fetters). Note how 
the reference to the female line of servile 
descent is contrived to heighten the con- 
trast with the real situation. 

1068 kak = δυσγενής, like δειλός, opp. 
to ἀγαθός, ἐσθλός : Od. 4. 63 ἀλλ᾽ ἀνδρῶν 
γένος ἐστὲ διοτρεφέων βασιλήων | σκηπ- 
τούχων᾽ ἐπεὶ οὔ κε κακοὶ τοιούσδε τέκοιεν. 

1067 τὰ «ταῦτα: cp. Ant. 
96 τὸ δεινὸν τοῦτο (ὦ é. of which you 
speak). 


1068 ὃς--ὅστις : O. C. 1171 ἔξοιδ᾽ 
ἀκούων τῶν δ᾽ ὅς ἐσθ᾽ ὁ προστάτης (n.). 

1072 Iocasta rushes from the scene—. 
to appear no more. Cp. the sudden exit. 

of Haemon (Axt. 766), of Eurydicé (20. 
1245), and of Deianeira (77. 813). In 
each of the two latter cases, the exit 
silently follows a speech dy another person, 
and the Chorus comments on the de- 
parting one’s sz/ence. Locasta, like Hae- 
mon, has spoken passionate words zm- 
mediately before going: and here σιωπῆς. 
(1075) i is more strictly ‘reticence’ than 
‘silence.’ 

1074 δέδοικα has here the construc-. 
tion proper to a verb of taking thought 
(or the like), as προμηθοῦμαι ὅπως μὴ. 
yevhoerar,—implying a desire to avert, 
if possible, the thing feared. Plat. Zu- 
thyphr. 4 E οὐ φοβεῖ δικαζόμενος τῷ πατρί, 
ὅπως μὴ αὖ σὺ ἀνόσιον πρᾶγμα τυγχάνῃς: 
πράττων; 


142 


μὴ ᾽κ τῆς σιωπῆς τῆσδ᾽ ἀναρρήξει κακά. 
ὁποῖα xp cer ῥηγνύτω. τοὐμὸν δ᾽ ἐγώ, 


ΣΟΦΟΚΛΕΟΥΣ 


Ἢ γυυς 


IO75 


Kel σμικρόν ἐστι, σπέρμ ἰδεῖν βουλήσομαι. ; 


αὕτη δ᾽ ἴσως, φρονεῖ γὰρ ὡς γυνὴ μέγα,»  - 
τὴν δυσγένειαν τὴν ἐμὴν αἰσχύνεται. 


ἐγὼ δ᾽ ἐμαυτὸν παῖδα τῆς Τύχης νέμων 


1080 


τῆς εὖ διδούσης, οὐκ ἀτιμασθήσομαι. 
τῆς γὰρ πέφυκα μητρός" οἱ δὲ συγγενεῖς 


μὴῆνές με μικρὸν καὶ μέγαν 


ιώρισαν. 


τοιόσδε δ᾽ ἐκφὺς οὐκ ἂν ἐξέλθοιμ᾽ ἔτι 


3 » 4 re “A 3 \ ’ 
ποτ᾽ ἄλλος, ὥστε μὴ ᾿κμαθεῖν τοὐμὸν γένος. 


suit χαίρειν here. 1075 ἀναρρήξη L. 


ἀναρρήξει is in V, Bodl. Laud. 54 E (from -n), Trin. (ἀναρήξει). 
hand in L wrote τοιόσδ᾽ ἐκφὺς wo οὐκ ἂν ἐξέλθοιμ᾽ ἔτι. 


1085 


Most of the later Mss. agree with L, but 
1084 The Ist 
A later hand wrote de over 


τοιόσδ (2.e. τοιόσδε δ᾽), and indicated by dots over.wo that it was to be deleted. The 





1075 The subject to ἀναρρήξει is 
κακά, not ἡ γυνή: for (t) ἡ γυνὴ ἀναρρή- 
fe κακά would mean, ‘the woman will 
burst forth into reproaches,’ cp. Ar. Eq. 
626 ὁ δ᾽ ἄρ᾽ ἔνδον ἐλασίβροντ᾽ ἀναρρηγνὺς 
ἔπη: Pind. fr. 172 μὴ πρὸς ἅπαντας ἀναρ- 
ρῆξαι τὸν ἀχρεῖον λόγον : (2) the image is 
that of a storm bursting forth from a 
great stillness, and requires that the mys- 
terious κακά should be the subject: cp. 
Ai. 775 ἐκρήξει μάχη: Arist. Meteor. 2. 8 
ἐκρήξας.. «ἄνεμος. 

1076 f. χρήζει scornfully personifies 
the κακά.---βουλήσομαι, ‘I shall wish’: 
i.e. my wish will remain unaltered until 
it has peu a ae Cp. 1446 προσ- 
τρέψομαι: 681 ὠφελεῖν βουλήσομαι, 
it shall πως be my aim: Eur, 
Med. 259 τοσοῦτον οὖν σου τυγχάνειν βου- 
λήσομαι, I shall wish (shall be content) 
to receive from you only thus much 
(cp. Az. 825 αἰτήσομαι δέ σ᾽ οὐ μακρὸν 
γέρας λαχεῖν). Ο. C. 1289 καὶ ταῦτ᾽ ἀφ᾽ 
ὑμῶν... βουλήσομαι | ..«κυρεῖν ἐμοί: Pind. 
Olymp. 7. 20 ἐθελήσω...διορθῶσαι λόγον, 
I shall have good will to tell the tale 
aright. That these futures are normal, 
and do not arise from any confusion of 
present w7sh with future act, may be 
seen clearly from _ Plat. Phaedo ΟἹ A καὶ 
ἐγώ μοι δοκῶ ἐν τῷ παρόντι τοσοῦτον ᾿ μόνον 
ἐκείνων. διοίσειν" οὐ γὰρ ὅπως τοῖς πα- 
ροῦσιν ἃ ἐγὼ λέγω δόξει ἀληθῆ προθυμη- 
θήσομαι: and 26. 191 C. 

1078 ὡς γυνὴ, for a woman: though, 


as it is, her ‘proud spirit’ only reaches 
the point of being sensitive as to a lowly 
origin. She is proud of her lineage; 
Oedipus, of what he is. Whitelaw well 
compares Tennyson: ‘Her pride is yet 
no mate for mine, Too proud to care 
from whence I came.’ Cp. Eur. feracl. 
978 πρὸς ταῦτα τὴν θρασεῖαν ὅστις ἃ ἂν θέλῃ] 
καὶ τὴν φρονοῦσαν μεῖζον ἣ γυναῖκα 
χρὴ | λέξει: Lipp. 640 μὴ γὰρ ἔν γ᾽ ἐμοῖς 
δόμοις εἴη φρονοῦσα πλεῖον ἢ γυναῖκα χρή. 
ὡς is restrictive; cp. 1118: Thuc. 4. 84 
ἣν" δὲ οὐδὲ Asivaree. ws Λακεδαιμόνιος, el- 
πεῖν (not a bad speaker, for a Lacedae- 
monian): imitated by Dionys. το. 31 (of 
L. Icilius) ws Ῥωμαῖος, εἰπεῖν οὐκ ἀδύ- 
varos. See on 763. 

1081 Whatever may have been his 
human parentage, Oed. is the ‘son of 
Fortune’ (said in a very different tone 
from ‘ Fortunae filius’ in Hor. Sat. 2. 6. 
49): Fortune brings forth the months with 
their varying events; these months, then, 
are his brothers, who ere now have known 
him depressed as well as exalted. He has 
faith in this Mother, and will not shrink 
from the path on which she seems to 
beckon him ; he will not be false to his 
sonship. We might recall Schiller’s epi- 
gram on the Wolfians; whatever may be 
the human paternity of the Lliad, ‘hat es 
doch Eine Mutter nur, Und die Ziige der 
Mutter, Deine unsterblichen Ziige, Natur.’ 
--τῆς εὖ διδούσης, the beneficent: here 
absol., usu. with dat.,as σφῷν δ᾽ εὖ διδοίη 


OIAITOY2 TYPANNOZ 143 
a storm of sorrow will break forth from this silence. 

OE. Break forth what will! Be my race never so lowly, I 
must crave to learn it. Yon woman, perchance—for she is 
proud with more than a woman’s pride—thinks shame of my 
base source. But I, who hold myself son of Fortune that gives 
good, will not be dishonoured. She is the mother from whom I 
spring; and the months, my kinsmen, have marked me some- 
times lowly, sometimes great. Such being my lineage, never 
more can I prove false to it, or spare to search out the secret of 
my birth. 


origin of the corruption plainly was that, δ᾽ having dropped out after τοιόσδε, some one 
unskilled in metre thought to complete the verse with ws (as=‘be sure that,’ cp. AZ. 
9).—Blaydes conj. τοιόσδε δὴ p’s.—Dindorf, who once conjectured οὐκ ἂν ἐξέλθοιν ποτὲ 


ἀλλοῖος, now rejects both verses (1084 f.). 


1085 or’ ἄλλος] ἄτιμος Nauck.—wore 





Ζεύς, O. C. 1435. Not gen. abs., ‘ while 
she prospers me,’ since the poet. τῆς for 
αὐτῆς could stand only at the beginning 
of a sentence or clause, as 1082. 

1082 συγγενεῖς, as being also sons of 
Τύχη: the word further expresses that 
their lapse is the measure of his life: cp. 
963: ἀλκᾷ ξύμφυτος αἰών (Ag. 107), years 
with which bodily strength keeps pace. 
Pind. Mem. 5. 40 πότμος συγγενής, the 
destiny born with one. 

1083 διώρισαν: not: ‘have determined 
that I should be sometimes lowly, some- 
times great’; to do this was the part of 
controlling Τύχη. Rather: ‘have distin- 
guished me as lowly or great’: 2.¢., his life 
has had chapters of adversity alternating 
with chapters of prosperity; and the 
months have marked these off (cp. 723). 
The metaphor of the months as sympa- 
thetic brothers is partly merged in the 
view of them as divisions of time: see on 
866, 1300. 

1084 ‘Having sprung of such parent- 
age (ἐκφὺς, whereas dvs would be merely 
‘having been born such’) I will never after- 
wards prove (ἐξέλθοιμι, evadam, cp. 1011) 
another man’ (ἄλλος, z.¢. false to my own 
nature). The text issound. The license 
of ποτ᾽ at the beginning of 1085 is to be 
explained on essentially the same prin- 
ciple as μέλας δ᾽ | , etc. (29, cp. 785, 791) 
at the end of a verse; viz. that, where the 
movement of the thought is rapid, one 
verse can be treated as virtually continuous 
with the next: hence, too, 4z. 986 οὐχ 
ὅσον τάχος | δῆτ᾽ αὐτὸν ἄξεις δεῦρο: Ph. 
66 εἰ δ᾽ ἐργάσει | μὴ ταῦτα. So here Soph. 
has allowed himself to retain ἔτι | in 
their natural connexion instead of writing 


ἔτι | ἄλλος ποτ΄. The genuineness of ποτ᾽ 
is confirmed by the numerous instances 
in which Soph. has combined it with ἔτι, 
as above, 892, below, 1412: Az. 98, 687: 
Tr. 830, 922. 

1086—1109 This short ode holds 
the place of the third στάσιμον. But it 
has the character of a ‘dance-song’ or 
ὑπόρχημα, a melody of livelier move- 
ment, expressing joyous excitement. The 
process of discovery now approaches its 
final phase. The substitution of a hypor- 
cheme for a regular stasimon has here a 
twofold dramatic convenience. It short- 
ens the interval of suspense; and it pre- 
pares a more forcible contrast. For the 
sake of thus heightening the contrast, 
Soph. has made a slight sacrifice of pro- 
bability. The sudden exit of Iocasta has 
just affected the Chorus with a dark pre- 
sentiment of evil (1075). We are now 
required to suppose that the spirited 
words of Oedipus (1076—1085) have 
completely effaced this impression, leav- 
ing only delight in the prospect that he 
will prove to be a native of the land. 
A hyporcheme is substituted for a stasi- 
mon with similar effect in the 47ax, where 
the short and joyous invocation of Pan im- 
mediately precedes the catastrophe (693— 
717): and inthe Antig., 1115—1154. The 
stasimon in the 7rachiniae 633—662 may 
also be compared, in so far as its glad anti- 
cipations usher in the beginning of the end. 

Strophe(1086—1097). Our joyous songs 
will soon be celebrating Cithaeron as na- 
tive to Oedipus. 

Antistrophe (1098—1109). Is he a son 
of some god,—of Pan or Apollo, of 
Hermes or Dionysus? 


144 
XO. 


στρ. 


ΣΟΦΟΚΛΕΟΥΣ 


» oN , “ΓΝ Ν Ν , ¥ 
SEER εγώ eagle ειμυ και κατα γνωμαν ἴδρις, 


’ 
2 ov TOV Ὄλυμπον ἀ εἰρων, 


8 ὦ Κιθαιρών, οὐκ ἔσει τὰν αὔριον 


ΙΟ90 


4 πανσέληνον, μὴ οὐ σέ γε καὶ πατριώταν * Oidimour 
5 καὶ τροφὸν καὶ ἜΑΡΟΣ αὔξειν,--- 


6 καὶ χορεύεσθαι" 
ἐμοῖς τυράννοις. 


ρὸς ἡμῶν, ὡς ἐπὶ ἦρα φέροντα τοῖς 


ee an Ν a > > > » 
1 inte Φοῖβε, σοὶ δὲ ταῦτ᾽ apéoT εἴη. 


> 
αντ. 


2 Πανὸς ὀρεσσιβάτα ἢ 


μὴ ᾿κμαθεῖν) ὥστε μὴ οὐ μαθεῖν Blaydes. 
ἔσει τὰν αὖρι Nauck: 
comment., and cp. 1101. 


, , , 2» 
TLS σέ, TEKVOV, TLS O ETLKTE 
πτα- 


οὐκ ἔσει τὰν ἦρι Wecklein: 
1091 Οἰδίπου Mss. 


*rav μακραιώνων apa 1098 


IIOO 
οὐκ 


οὐκέτι τὰν ἑτέραν Dindorf. See 
I write Οἰδίπουν. 1097 σοὶ δὲ 


1090 οὐκ ἔσει τὰν αὔριον MSS.: 





MSS.: σοὶ δ᾽ οὖν Kennedy. 1099 τῶν MSS.: τἂν Heimsoeth.—dpa L: dpa Heath. 
1086 μάντις: as El. 472 εἰ un γὼ Hermann Anz. 11. ὃ 59. Wolff remarks 


παράφρων μάντις ἔφυν καὶ γνώμας | λειπο- 
μένα σοφᾶς: cp. Ο. C. 1080, 41". 1160, 
At. 1419: and μαντεύομαιΞε ‘to presage. 

1087 κατὰ with an accus. of respect 
is somewhat rare ( Tr. 102 κρατιστεύων. 
Kar’ dupa: 1b. 379 7 κάρτα λαμπρὰ καὶ 
κατ᾽ ὄμμα Kal φύσωυν), except in such 
phrases as κατὰ πάντα, Kar’ οὐδέν, κατὰ 
τοῦτο. Cp. Metrical Analysis. 

1088 οὐ--οὐ μὰ: see on 660.— 


ἀπείρων = ἄπειρος : Hesych. 1. 433 ἀπεί- 


povas’ ἀπειράτους. Σοφοκλῆς Θυέστῃ. 
Ellendt thinks that ἀπειράτους here meant 
ἀπεράντους (‘limitless ’ yeabue elsewhere 
ἀπείρατος ἅΙγαγδεε " untried’ or ‘inex- 
perienced.’ Conversely Soph. used ἄπει- 
pos in the commoner sense of ἀπείρων, 
‘vast,’ fr. 481 χιτὼν ἄπειρος ἐνδυτήριος 
κακῶν. περά-ω, to go through, πεῖρα 
(repia), a going-through (ferttus, peri- 
culum), are closely akin to πέρα, beyond, 
πέρας, πεῖραρ a limit (Curt. Ztym. §§ 356, 
357): in poetical usage, then, their deriva- 
tives might easily pass into each other’s 
meanings. 

1090 τὰν αὔριον πανσέληνον, ‘the 
full-moon of to-morrow,’ acc. of ἡ αὔριον 
πανσέληνος (there is no adj. adpios), as 
Eur. Alc. 784 τὴν αὔριον μέλλουσαν, acc. 
of ἡ αὔριον μέλλουσα, Hipp. 1117 τὸν 
αὔριον χρόνον. At Athens the great 
Dionysia were immediately followed by 
the ia, a festival held at full-moon 
in the middle of the month Elaphebolion_ 
(at the beginning of April): cp. A. 
Mommsen feortol, p. 389, and (Ὁ. F. 


that, if this play was produced on the 
last day of the Dionysia, the poet would 
have known that arrangement long be- 
forehand, and may have intended an 
allusion to the Πάνδια which his Athenian 
hearers would quickly seize. This would 
explain why precisely ‘to-morrow’s full- 
moon’ is named. —Nauck reads αὖρι (as 
=Taxéws, § the coming’ full- moon) : 
Wecklein, ἦρι (dat. of 7p), ‘the vernal 
full-moon’—that, namely, in Elaphe- 
bolion. τ-- πανσέληνον (sc. ὥραν) : Her. 2. 
47 ἐν τῇ αὐτῇ πανσελήνῳ. For the accus., 
cp. on 1138 χειμώνα. The meaning is: 
‘At the next full-moon we will hold a 
joyous παννυχίς, visiting the temples with 
χοροί (Ant. 153), in honour of the dis- 
covery that Oedipus is of Theban birth; 
and thou, Cithaeron, shalt be a theme of 
our song.’ Cp. Eur. 7072 1078, where, in 
sympathy with the nocturnal worship of 
the gods, ἀστερωπὸς | ἀνεχόρευσεν αἰθήρ, | 
χορεύει δὲ Σελάνα. The rites of the The- 
ban Dionysus were νύκτωρ τὰ πολλά (Eur. 
Bacch. 486). 

1091 πατριώταν, since Cithaeron 
partly belongs to Boeotia; so Plutarch 
of Chaeroneia calls the Theban Dionysus 
his πατριώτην θεόν, Mor. 671 C.—I read 
Οἰδίπουν instead of Οἰδίπου. With the 
genitive, the subject to αὔξειν must be 
either (1) ἡμᾶς understood, which is im- 
possibly harsh; or (2) τὰν.. «πανσέληνον. 
Such a phrase as ἡ πανσέληνος αὔξει σε, 
z.é., ‘sees thee honoured,’ is possible ; cp. 
438 ἥδ᾽ ἡμέρα φύσει σε καὶ διαφθερεῖ: but. 


OIAITOYS TYPANNOZ 


CH. 


145 


If I am a seer or wise of heart, O Cithaeron, thou Strophe. 


shalt not fail—by yon heaven, thou shalt not!—to know at to- 
morrow’s full moon that Oedipus honours thee as native to him, 
as his nurse, and his mother, and that thou art celebrated in 
our dance and song, because thou art well-pleasing to our prince. 
O Phoebus to whom we cry, may these things find favour in thy 


sight! 


Who was it, my son, who of the race whose years are many Anti- 
that bore thee in wedlock with Pan, the mountain-roaming StoPhe- 


Blaydes conject. κορᾶν. 
πελασθεῖσα, without elision.) 


1100 πανὸσ ὀρεσσιβάτα προσπελασθεῖσ᾽ MSS. 
To supply the want of a syllable after ὀρεσσιβάτα, Her- 


(L has προσ- 


mann inserted τις, Heath που: Wunder and others wrote ὀρεσσιβάταο: Dindorf con- 


jectured Νύμφα ὀρεσσιβάτᾳ που Mavi πλαθεῖσα. 


Lachmann restored πατρὸς πελασθεῖσ᾽. 





it is somewhat forced; and the order of 
the words is against it. The addition of 
one letter, giving Οἰδίπουν, at once 
yields a clear construction and a pointed 
sense. ‘Thou shalt not fail to know 
that Oedipus honours thee both as native 
to him, and as his nurse and mother (z.¢., 
not merely as belonging to his Theban 
fatherland, but as the very spot which 
sheltered his infancy); and that thou art 
celebrated in choral song by ws (πρὸς 
ἡμῶν), seeing that thou art well-pleasing 
to him.’ μὴ οὐ with αὔξειν, because οὐκ 
ἀπείρων ἔσει =a verb of hindrance or denial 
with a negative. αὔξειν, not merely by 
praises, but by the fact of his birth in the 
neighbourhood: as Pindar says of a victor 
in the games, Olymip. 5. 4 τὰν σὰν πόλιν 
αὔξων, Pyth. 8. 38 αὔξων πάτραν. The 
acc. φέροντα, instead of φέρων, may be 
explained by supposing that σέ ye is 
carried on as subject to xopeverGar: cp. 
Tr. 706n. Another defence of the acc. 
would be to take καὶ χορ. πρὸς ἡμῶν as a 
parenthesis (cp. Amz. 1279 n.): so Tyrrell 
in Class. Rev. 11. 141. : 

1092 τροφὸν, as having sheltered him 
when exposed: τί pw’ édéxov; 1391. 
ματέρ᾽, as the place from which his life 
rose anew, though it had been destined 
to be his τάφος, 1452. 

1094 χορεύεσθαι, to be celebrated 
with choral song: At. 1153 πάννυ- 
xo | χορεύουσι τὸν ταμίαν Ἴακχον. (Not 
‘danced over,’ like ἀείδετο τέμενος, Pind. 
- Ol. 11. 76.) 

1095 ἐπὶ ἦρα φέροντα: see Merry’s 
note on Od. 3. 164 αὖτις ἐπ᾽ ᾿Ατρείδῃ 
᾿Αγαμέμνονι ἦρα PépovTes. ἦρα was 
probably acc. sing. from a nom. 7p, from 


ΠΑ: Ἂμ 


root dp (to fit), as=‘pleasant service. 
After the phrase ἦρα φέρειν had arisen, 
ἐπὶ was joined adverbially with φέρειν, 
ἐπὶ ἦρα φέρειν being equivalent to ἦρα 
ἐπιφέρειν. Aristarchus, who according to 
Herodian first wrote ἐπίηρα, must have 
supposed an impossible tmesis of a com- 
pound adj. in the passage of the Od. just 
quoted, also in 16. 375, 18. 56.—rots 
ἐμοῖς Tup., 2.4. to Oedipus: for the plur., 
see on θανάτων, 497. 

1096 ἰήϊε, esp. as'the Healer: see on 
154. 
1097 σοὶ δὲ: Z/. 150 NidBa, σὲ δ᾽ 
ἔγωγε νέμω θεόν .---ἀρέστ᾽ : 2.4. consistent 
with those oracles which still await a 
λύσις εὐαγής (921). 

1098 ἔτικτε: see on 870. 

1099 τάν μακραιώνων : here not god- 
desses (Aesch. 77%. 524 δαροβίοισι θεοῖ- 
ow), but the Nymphs, who, though not 
immortal, live beyond the human span; 
fom. Hymn. 4.260 αἵ ῥ᾽ οὔτε θνητοῖς οὔτ᾽ 
ἀθανάτοισιν ἕπονται" | δηρὸν μὲν ζώουσι καὶ 
ἄμβροτον εἶδαρ ἔδουσιν. They consort with 
Pan, ὅς 7’ ἀνὰ πίση | δενδρήεντ᾽ ἄμυδις 
φοιτᾷ χοροήθεσι Νύμφαις, Hymn. 19. 2. 

1100 In Ilavds ὀρεσσιβάτα προσπε- 
λασθεῖσ᾽, the reading of the MssS., we 
note (1) the loss after ὀρεσσιβάτα of one 
syllable, answering to the last of ἀπείρων 
in 1087: (2) the somewhat weak com- 
pound προσπελασθεῖσ᾽: (3) the gen., 
where, for this sense, the dat. is more 
usual, as Aesch. P. V. 896 μηδὲ πλαθείην 


γαμετῇ. 1, has κοίτῃ written over dpec- 
o.Bara. I had thought of λέκτροις 
πελασθεῖσ. But the gen. is quite ad- 


missible: and on other grounds Lach- 
mann’s πατρὸς πελασθεῖσ᾽ is far better, 


10." 


146 
8 τρὸς πελασθεῖσ᾽; ἢ σέ γ᾽ 
4 Λοξίου ; 
ὅ «0 ὁ Κυλλάνας ἀνάσσων, 
6 


δέξατ᾽ ἔκ του 


ΣΟΦΟΚΛΕΟΥΣ 


Sees Στ ΄, 
ευνατειρα TU 


A \ , > μὰ a 4 
τῷ yap πλάκες ἀγρόνομοι πᾶσαι Pidau: 


1104 


et ὁ Βακχεῖος θεὸς ναίων ἐπ᾽ ἄκρων ὀρέων εὕρημα 


7 Νυμφᾶν ᾿Ελικωνίδων, αἷς πλεῖστα συμπαίζει. 


OF. 


εἰ χρή TL κἀμὲ μὴ συναλλάξαντά πω, 


1151 


πρέσβεις, σταθμᾶσθαι, τὸν βοτῆρ᾽ ὁρᾶν δοκώ, 


ὅνπερ πάλαι ζητοῦμεν. 


ἔν τε γὰρ μακρῷ 


γήρᾳ ξυνάδει τῷδε τἀνδρὶ σύμμετρος, 
ἄλλως TE TOUS ἄγοντας ὥσπερ οἰκέτας 


1101 ἢ σέ γε θυγάτηρ λοξίου L. Most: of the later MSS. insert τίς before θυγάτηρ, 


while a few agree with L. Arndt conjectures ἢ σέ γ᾽ εὐνάτειρά TLS. 


οὔρειος κόρα. 


with almost all the later mss. 


Hartung, ἢ σέ γ᾽ 


1107 εὕρημα] σ᾽ εὕρημα Dindorf: ἄγρευμα M. Schmidt: γέννημα or 
λόχευμα Wecklein: δώρημα Gleditsch: σε θρέμμα Wolff. 


1109 ἑλικωνιάδων L, 


(A has ἑλικωνιάδων by correction from ἑλικωνίδος. = 





since πατρὸς, written προσ, would explain 
the whole corruption. 

1101 If in 1090 we keep οὐκ ἔσει 
τὰν αὔριον, it is best to read here with 
Arndt, ἢ σέ γ᾽ εὐνάτειρά, τις. On the 
view that in 1090 τὰν ἐπιοῦσαν ἔσει was a 
probable emendation (see Appendix on 
that verse), I proposed to read here, ἢ σέ 
γ᾽ ἔφυσε πατὴρ | Λοξίας ; If the ce of 
ἔφυσε had once been lost (through a 
confusion with the preceding σέ), ΓῈ- 
ΦΥΠΑΤῊΡ might easily have become 
TEOYTTATHP: the τις (which is not in 
L) would have been inserted for metre’s 
sake, and the change of Λοξίας to Λοξίου 
would have followed. (It cannot be ob- 
jected that a mention of the mother is 
required here, since, as the context shows, 
the foremost thought is, ‘what god was 
thy sire?’?) It would be a very forced 
way of taking ἢ σέ γέ τις θυγάτηρ to make 
θυγάτηρ depend on μακραιώνων, and Λοξίου 
on πελασθεῖσ᾽ (z.e., ‘some daughter of the 
Nymphs wedded to Pan, or haply to 
Loxias’). Nor does it seem easy'to take 
θυγάτηρ with τᾶν μακραιώνων in both 
clauses (‘some daughter of the Nymphs, 
wedded to Pan, or perhaps to Loxias’). 
On the whole, I now prefer Arndt’s cor- 
rection.— For σέ ye in the second alter- 
native, cp. PA. 1116 πότμος σε δαιμόνων 
τάδ᾽, | οὐδὲ σέ ye δόλος ἔσχεν. Her. 7. 
to (ad fin.) διαφορεύμενον ἤ κου ἐν γῇ τῇ 
᾿Αθηναίων ἢ σέ γε ἐν τῇ Λακεδαιμονίων. 


1103 πλάκες ἀγρόνομοι--πλ. ἀγροῦ 
νεμομένου, highlands affording open pas- 
turage: so dypov. αὐλαῖς, Ant. 785. 
Apollo as a pastoral god had the title of 
Νόμιος (Theocr. 25. 21), which was esp. 
connected with the legend of his serving 
as shepherd to Laomedon on Ida (71. 21. 
448) and to Admetus in Thessaly (Z/. 2. 
766: Eur. Adc. 572 pndovduas). Macro- 
bius 1. 17. 43 (Apollinis) aedes ut ovium 
pastoris sunt apud Camirenses [in Rhodes] 
ἐπιμηλίου, apud Naxios ποιμνίου, 
itemque deus ἀρνοκόμης colitur, et apud 
Lesbios ναπαῖος [cp. above, 1026], εἴ 
multa sunt cognomina per diversas civt- 
tates ad det pastoris offictum tendentia. 
Callim. Hymn. Apoll. 47 οὐδέ κεν atyes | 
δεύοιντο βρεφέων ἐπιμηλίδες, How ᾿Απόλ- 
λων | βοσκομένῃς ὀφθαλμὸν ἐπήγαγεν. 

1104 ὁ Κυλλάνας ἀνάσσων, Hermes: 
Hom. Hymn. 3. 1 ἙἭἍἨρμῆν ὕμνει, Μοῦσα, 
Διὸς καὶ Μαιάδος υἱόν, | Κυλλήνης μεδέ- 
οντα καὶ ᾿Αρκαδίης πολυμήλου : Verg. Aen. 
8. 138 guem candida Maia| Cyllenes 
gelido conceptum vertice fudit. The peak 
of Cyllene (now Ziria), about 7300 ft. 
high, in N. E. Arcadia, is visible from 
the Boeotian plain near Leuctra, where 
Cithaeron is on the south and Helicon to 
the west, with a glimpse of Parnassus 
behind it: see my Modern Greece, p. 77. 

1105 ὁ Baxxetos θεὸς, not ‘the god 
Bdxxos’ (though in O. C. 1494 the Mss. 
give Ποσειδαωνίῳ θεῷ Ξε- Ποσειδῶνι), but 


OIAITOYS TYPANNOZ 147 
father? τῶ was, it a@. bride-ol. Loxias: that ‘bore thee? = Por 
dear to him are all the upland pastures. Or perchance ’twas 
Cyllene’s lord, or the Bacchants’ god, dweller on the hill-tops, 
that received thee, a new-born joy, from one of the Nymphs 
of Helicon, with whom he most doth sport. 


ΟΕ. Elders, if ’tis for me to guess, who have never met with 
him, I think I see the herdsman of whom we have long been in 
quest; for in his venerable age he tallies with yon stranger’s years, 
and withal I know those who bring him, methinks, as servants 


᾿Ελικωνίδων Porson. ἑἐλικωπίδων Wilamowitz. 1111 πρέσβει L. A letter (evi- 
dently o) has been erased after «. A very late hand has written vy over εἰ. The other 
MSS. have mpéoBe (A), πρέσβυ (received by Blaydes and Campbell), or πρέσβυν (Elmsley 
and Hartung). Dindorf cp. Aesch. Pers. 840 (where the chorus is addressed), ὑμεῖς 
δέ, πρέσβεις, χαίρετ᾽. 1114 ἄλλως τε] Nauck gives δμώάς τε, and further conjec- 





‘the god of the Βάκχοι,᾽ the god of Bac- 
chic frenzy; Hom. Hymn. 19. 46 ὁ Βάκ- 
χειος Διόνυσος : O. C. 678 ὁ Baxxwwras... 
Διόνυσος. Some would always write Bak- 
xetos (like ‘Ounpetos, Αἰάντειος, etc.): on 
the other hand, Βακχεῖος is said to have 
been Attic (cp. Kadmetos): see Chandler, 
Greek Accentuation, § 381, 2nd ed. 

1107 εὕρημα expresses the sudden 
delight of the god when he receives the 
babe from the mother,—as Hermes re- 
ceives his new-bornson Pan from the Νύμφη 
ἐὐπλόκαμος, Hom. Hymn. 19. 40 Tov δ᾽ aly’ 
Ἑρμείης ἐριούνιος és χέρα θῆκεν | δεξάμενος" 
χαῖρεν δὲ νόῳ περιώσια δαίμων. The word 
commonly=a lucky ‘find,’ like ἕρμαιον, 
ora happy thought. In Eur. Jom 1349 it 
is not ‘a foundling,’ but the box contain- 
ing σπάργανα found by Ion. 

1109 συμπαίζει: Anacreon fr. 2 (Bergk 
Ῥ. 775) to Dionysus: ὦναξ, @ δαμάλης 
(subduing) "Epws | καὶ Νύμφαι κνανώ- 
aides | roppupén τ᾽ ᾿Αφροδίτη | συμπαί- 
ζουσιν' ἐπιστρέφεαι δ᾽ | ὑψηλῶν κορυφὰς 
ὀρέων. ᾿“Ἑϊλικωνίδων is Porson’s correc- 
tion of ᾿Ελικωνιάδων (MSS.), ad Eur. Or. 
614. Since αἷς answers to δέ in 1097, 
Nauck conjectured ᾿Ελικῶνος αἷσι. But 
this is unnecessary, as the metrical place 
allows this syllable to be either short or 
long: so in £2. 486 αἰσχίσταις answers 
to 502 νυκτὸς εὖ. 

1110—1185 ἐπεισόδιον τέταρτον. 
‘The herdsman of Laius is confronted with 
the messenger from Corinth. It is dis- 
covered that Oedipus is the son of Laius. 

1110—1116 The οἰκεύς, who alone 
escaped from the slaughter of Laius and 
his following, had at his own request been 


sent away from Thebes to do the work 
of a herdsman (761). Oedipus had sum- 
moned him in order to see whether he 
would speak of λῃσταί, or of one λῃστής 
(842). But meanwhile a further question 
has arisen. Is he identical with that 
herdsman of Laius (1040) who had given 
up the infant Oedipus to the Corinthian 
shepherd? He is now seen approaching. 
With his coming, the two threads of dis- 
covery are brought together. 

1110 κἀμὲ, as well as you, who per- 
haps know better (1115).--- μὴ συναλ- 
Adgavrd πω, though I have never come 
into intercourse with him, have never 
met him: see on 34, and cp. 1130. 

1112 ἐν... γήρᾳ: ἐν describes the con- 
dition zz which he is, as Pz. 185 ἔν τ᾽ 
ὀδύναις ὁμοῦ | λιμῷ τ᾽ οἰκτρός: Az. 1017 
ἐν γήρᾳ βαρύς. 

1118 ξυνάδει with τῷδε τἀνδρὶ: σύμ- 
μετρος merely strengthens and defines it: 
he agrees with this man in the tale of his 
years. 

1114 ἄλλως te, and moreover: cp. 
Her. 8. 142 ἄλλως τε τούτων ἁπάντων 
αἰτίους γενέσθαι δουλοσύνης τοῖσι “Ἑλλησι 
᾿Αθηναίους οὐδαμῶς ἀνασχετόν (‘and Jde- 
sides,’ introducing anadditional argument). 
Soph. has ἄλλως τε καί τε ‘especially,’ £7. 
1324. ‘I know them as servants’ would 
be ἔγνωκα ὄντας οἰκέτας. The ὥσπερ 
can be explained only by δὴ ellipse: ὥσπερ 
ἂν γνοίην οἰκέτας ἐμαυτοῦ (cp. 923). Here 
it merely serves to mark hes first impres- 
ston as they come in sight: ‘I know those 
who bring him as (mettinks) servants of 
mine own.’ 


ΤΟ 


148 ΣΟΦΟΚΛΕΟΥΣ 
ἔγνωκ᾽ ἐμαυτοῦ" ™m ἐπιστήμῃ σύ μου I115 
προὔχοις τάχ᾽ av που, τὸν οτῆρ᾽ ἰδὼν πάρος. 
ΧΟ. ἔγνωκα γάρ, σάφ᾽ ἴσθι: Λαΐου γὰρ ἦν 
εἴπερ τις ἄλλος, πιστὸς ὡς νομεὺς ἀνήρ. 
ΟΙ. σὲ πρῶτ᾽ ἐρωτῶ, τὸν Κορίνθιον ἕένον, 
7 TOVOE φράζεις ; AI’. τοῦτον, ovmep εἰσορᾷς. 1120 
Ol. οὗτος σύ, πρέσβυ, δεῦρό μοι φώνει βλέπων 
ὅσ᾽ ἄν σ᾽ ἐρωτῶ Λαΐου ποτ᾽ ἦσθα σύ; 
ΘΕΡΑΠΩΝ, 
ἦ, δοῦλος οὐκ ὠνητός, ἀλλ᾽ οἴκοι τραφείς. 
ΟΙ. ἔργον μεριμνῶν ποῖον ἢ βίον τίνα ; 
OE. ποίμναις τὰ πλεῖστα τοῦ βίου συνειπόμην. 1125 
OI. χώροις μάλιστα πρὸς τίσι ξύναυλος ὦν; 
OE. ἦν μὲν Κιθαιρών, ἣν δὲ πρόσχωρος τόπος. 
ΟΙ. τὸν ἄνδρα τόνδ᾽ οὖν οἶσθα τῇδέ που μαθών ; 
ΘΕ. τί χρῆμα δρῶντα; ποῖον ἄνδρα καὶ λέγεις ; 
OI. τόνδ᾽ ὃς πάρεστιν: ἢ ξυναλλάξας τί πω; ΤΥ το 


tures ὄντας for ὥσπερ. See comment. 


1130 71, 1st hand, corrected to 7 by a later 


hand.—éuvaddatac L, the first ἃ made from ν, as if the scribe had begun to write 


ξυναντήσας. 


The later MSS, are divided between the alternative readings, ἢ ξυναλλάξας 
(as E, Bodl. Laud. 54, Vat. a, c), and ἢ ξυνήλλαξας (as A, T, V, A). 


The change of 





1117 ydp, in assent (‘you are right, 
170... Cte.) 731: Ph. 756: Ant. 639, etc.— 
Λαΐου γὰρ ἦν.. νομεὺς : a comma at ἦν 
is admissible (cp. 1122), but would not 
strictly represent the construction here, 
in which the idea—Aatou ἦν πιστὸς νομεύς, 
εἴπερ τις a\Xos—has been modified by 
the restrictive ὡς before vouetsx—os only 
means that the sense in which a νομεύς 
can show πίστις is narrowly limited by 
the sphere of his work. See on 763: cp. 
1078. 

1119 τὸν Κορίνθ. ξένον, with σὲ, in- 
stead of a vocative, gives a peremptory 
tone: Ant. 441 σὲ δή, σὲ τὴν νεύουσαν els 
πέδον κάρα, | φὴς ἢ καταρνεῖ x.7.r., where 
the equivalent of ἐρωτῶ here is under- 
stood. Cp. Az. 71 οὗτος, σὲ τὸν τὰς K.T.X. 
So in the nomin. Xen. Cyr. 4. 5. 22 od 
δ᾽, ἔφη, ὁ τῶν Ὑρκανίων ἄρχων, ὑπόμεινον. 
Blaydes thinks that τῷ Κορινθίῳ ξένῳ in 
Ar. Zh. 404 comeshence. Surely rather 
from the Sthenoboea of Eur. apf. Athen. 
427 E πεσὸν δέ νιν λέληθεν οὐδὲν ἐκ χερός, 
| ἀλλ᾽ εὐθὺς αὐδᾷ, τῷ Κορινθίῳ ξένῳ. 


1121 Cp. Tr. 402 οὗτος, βλέφ᾽ ὧδε. 

1128 ἦ, the old Attic form of the 1st 
pers., from éa (//. 4. 321, Her. 2. 19): 
so the best Mss. in Plat. Phaed. 61 B, etc. 
That Soph. used 7 here and in the Niobe 
(fr. 409) 7 yap φίλη ᾽γὼ τῶνδε τοῦ προ- 
φερτέρου, is stated by the schol. on 712. 
5. 533 and on Od. 8. 186. L has ἦν 
here and always, except in O. C. 973, 
1366, where it gives q- In Eur. Z7o. 
474 ἦ μὲν τύραννος κεἰς τύρανν᾽ ἐγημάμην 
is Elmsley’s corr. of ἦμεν τύραννοι x.r. λ. 
On the other hand Eur., wat least, has ἦν 
in several places where ἢ is impossible : 
Hipp. 1012 μάταιος ap’ ἣν, οὐδαμοῦ μὲν 
οὖν φρενῶν: 47. F. 1416 ὡς ἐς τὸ λῆμα 
παντὸς ἦν ἥσσων ἀνήρ: Alc. 655 mais δ' 
ἣν ἐγώ σοι τῶνδε διάδοχος δόμων: Jon 280 
βρέφος νεογνὸν μητρὸς ἣν ἐν ἀγκάλαις. τος 
οἴκοι τραφείς, and so more in the con- 
fidence of the master: cp. schol. Ar. ἔφ. 
2 (on Παφλάγονα τὸν νεώνητον), πεφύ- 
καμεν γὰρ καὶ τῶν οἰκετῶν ον πισ- 
τεύειν τοῖς οἴκοι γεννηθεῖσι καὶ τραφεῖσιν 
ἢ οἷς av κτησώμεθα πριάμενοι. Such vernae 


OVATION Ss ΥΡΑΝΝΟΣ 


of mine own. 


149 


But perchance thou mayest have the advantage 


of me in knowledge, if thou hast seen the herdsman before. 


CH. 


Aye, I know him, be sure; he was in the service of 


Laius—trusty as any man, in his shepherd’s place. 


OE. 
thou meanest ? 
OE. 


| Zhe herdsman ts brought in. 


I ask thee first, Corinthian stranger, is this he whom 
ΜΕ. This man whom thou beholdest. 
Ho thou, old man—I would have thee look this way, 


and answer all that I ask thee-—VThou wast once in the service 


Or liaise. 


HERDSMAN. 


I was—a slave not bought, but reared in his house. 


OE. 
ΗΕ. 


Employed in what labour, or what way of life? 
For the best part of my life I tended flocks. 


ΟΕ. And what the regions that thou didst chiefly haunt ? 


HE. 
bouring ground. 
OE. 


parts— 
HE. 


Sometimes it was Cithaeron, sometimes the neigh- 
Then wottest thou of having noted yon man in these 


Doing what ?...What man dost thou mean ?... 


OE. This man here—or of having ever met him before? 


ἢ into 7 probably induced the change of the aor. participle into the aor. indic.—7w] 


In L the w has been made from o or a after erasure of at least two other letters. 
word was never 7wo or που: Diibner suggests πούσ, Campbell ποτέ. 


The 
The last letter 


seems to have been o, and the word may perhaps have been πάροσ.---πὼσ r: ov 





were called οἰκογενεῖς (Plat. AZen. 82 B: 
Dio Chrys. 15. 25 τοὺς παρὰ σφίσι γεν- 
νηθέντας ovs οἰκογενεῖς καλοῦσι), olkorpa- 
φεῖς (Pollux 3. 78), ἐνδογενεῖς (oft. in 
inscriptions, as C. 7. G. 1. 828), or οἰκό- 
TpiBes [Dem.] or. 13 § 24, Hesych. 2. 
766. 

1124 μεριμνῶν. In classical Greek 
μεριμνᾶν is usu. ‘to give one’s thought 
to a question’ (as of philosophy, Xen. 
Mem. 4. 7.6 τὸν ταῦτα μεριμνῶντα) ; here 
merely= ‘to be occupied with’: cp. Cyr. 
8. 7. 12 τὸ πολλὰ μεριμνᾶν : and so in the 
NV. 7., 1 Cor. 7. 33 μεριμνᾷ τὰ τοῦ κόσ- 
μου. 

1126 ξύναυλος, prop. ‘dwelling with’ 
(μανίᾳ ξύναυλος Az. O11): here, after πρὸς, 
merely: ‘having thy haunts’: an instance 
of that redundant government which 
Soph. often admits: below 1205 ἐν πό- 
νοις | ξύνοικος : Az. 464 γυμνὸν...τῶν ἀρισ- 
τείων ἄτερ: Ph. 31 κενὴν οἴκησιν ἀνθρώ- 
πων δίχα: Ant. ο10 ἔρημος πρὸς φίλων : 
445 ἔξω βαρείας αἰτίας ἐλεύθερον. 


1127 ἦν μὲν, as if replying to χῶροι 
τίνες ἦσαν πρὸς ols ξυν. ἦσθα; 

1128 οἶσθα with μαθών, are you aware 
of having observed this man here? Cp. 
1142 οἶσθα... δούς; We could not render, 
‘do you £zow this man, through having 
observed him?’ εἰδέναι, implying intui- 
tive apprehension, is said of knowing 
facts and propositions: in regard to per- 
sons, it is not used in the mere sense of 
‘being acquainted with one’ (γνωρίζω), 
but only in that of ‘knowing one’s cha- 
racter,’ as Eur. Med. 39 ἐγῷδα τήνδε. 
So scire, wissen, savotr, Ital. sapere. On 
the other hand, γιγνώσκω, implying a 
process of examination, applies to all 
mediate knowledge, through the senses, 
of external objects: so moscere, kennen, 
connattre, Ital. conoscere. Cp. Cope in 
Journ. of Philology 1. 79. 

1129 καὶ λέγεις: see on 772. 

1130 The constr. is οἶσθα μαθών...ἢ 
ξυναλλάξας ; Oed. takes no more notice 
of the herdsman’s nervous interruption 


150 


ΣΟΦΘΟΚΛΕΘΥῪΣ 


= 3 9 > > ω > / ΄ Ψ 
OK. οὐχ woTe y εἰπειν ἐν τάχει μνήμης ὑπο. 


4d be 


Kovoev ye θαῦμα, δέσποτ᾽" ἀλλ᾽ ἐγὼ σαφῶς 


> Cot er > / Ss \ ἴὸ᾽ ν 

ἀγνῶτ᾽ ἀναμνήσω νιν. εὖ yap oid ὅτι 
3 Ss “ 

κάτοιδεν ἦμος τὸν Κιθαιρῶνος τόπον 


ε Ν a , > Ν δ᾽ CEN τ, 
ὁ μὲν διπλοῖσι ποιμνίοις, ἐγὼ δ᾽ ἑνὶ ΓΞ 
ἐπλησίαζον τῷδε τἀνδρὶ τρεῖς ὅλους 
ἐξ ἦρος εἰς ἀρκτοῦρον ἑκμήνους χρόνους" 
ἴω 5 » > 4 > > » > > Ν 
χειμῶνα δ᾽ ἤδη Taya T εἰς ἔπαυλ᾽ ἐγὼ 
ἤλαυνον οὗτός T εἰς τὰ Λαΐου σταθμα. 
λέγω τι τούτων, ἢ οὐ λέγω πεπραγμένον ; 1140 


ΘΕ. λέγεις ἀληθῆ, καίπερ ἐκ μακροῦ χρόνου, 


Blaydes. 1181 ὕπο] ἄπο Reiske. 
ποιμνίοις, ἔγὼ δ᾽ ἑνί, | ἐπλησίαζε. 


1185 ξ. Heimsoeth conject. νέμων διπλοῖσι 
1137 ἐμμήνουσ L, with almost all the later 
Mss.: but the Trin. Ms. has ἐκμήνους, whence Porson restored ἑκμήνους. 


1138 χει- 





than is necessary for the purpose of stern- 
ly keeping him to the point. ἢ συνήλ- 
Aatas...; ‘have you ever met him?’ mars 
the force of the passage. The testimony 
of L to συναλλάξας has the more weight 
since this is the less obvious reading. Cp. 
verse 1037, which continues after an in- 
terruption the construction of verse 1035. 

1131 οὐχ ὥστε γ᾽ εἰπεῖν: cp. 361.— 
μνήμης ὕπο, at the prompting of memory, 
—vwmé having a like force as in compound 
verbs meaning to ‘suggest,’ etc.: Plut. 
Mor. 813 E λογισμοὺς οὖς ὁ Περικλῆς av- 
τὸν ὑπεμίμνησκεν, recalled to his mind: 
so ὑποβολεύς (ib.), ‘a prompter.’ The 
phrase is more poetical and elegant than 
μνήμης ato, the conjecture of Reiske. 
Blaydes, reading ἄπο, compares ἀπὸ τῆς 
γλώσσης (O. C. 936). 

1132 Ε΄ Kovdév ye: cp. PA. 38 n. 
dyvar =od γιγνώσκοντα, not recognising 
me: 677 n. 

1134 Soph. has the epic ἦμος in two 
other places of dialogue, 77. 531 (an- 
swered by τῆμος) and 155; also once in 
lyrics Az. 935; Eur. once in lyrics (Hee. 
915); Aesch. and Comedy, never.—tov 
Κιθαιρῶνος τόπον. The sentence be- 
gins as if it were meant to proceed thus: 
τὸν K. τόπον ὁ μὲν διπλοῖς ποιμνίοις ἔνε- 
μεν, ἐγὼ δ᾽ ἑνὶ (ἔνεμον), πλησιάζων αὐτῷ : 
but, the verb ἔνεμε having been post- 
poned, the participle πλησιάζων is irregu- 
larly combined with the notion of ἔνεμον 
and turned into a finite verb, érAnolafov : 
thus leaving τὸν K. τόπον without any 


proper government. (In the above ex- 
planation, the act. voice of véuw has 
been used, since this was specially said 
of shepherds: cp. Xen. Cyr. 3. 2. 20 
ἐπεὶ ὄρη ἀγαθὰ ἔχετε, ἐθέλοιτ᾽ ἂν ἐᾶν νέμειν 
ταῦτα τοὺς ᾿Αρμενίους ; The midd. would 
also be correct, as=‘to range over.’) For 
the irregular but very common change of 
participle into finite verb cp. Z/. 190 
οἰκονομῶ.. ὧδε μὲν ἀεικεῖ σὺν στολᾷ | κε- 
vais δ᾽ ἀμφίσταμαι τραπέζαις (instead of 
ἀμφισταμένη) : so Ant. 810 (ὕμνος ὕμνη- 
σεν instead of ὕμνῳ ὑμνηθεῖσαν): Tr. 676 
ἠφάνισται, διάβορον πρὸς οὐδενὸς τῶν ἔν- 
δον, ἀλλ᾽ ἐδεστὸν ἐξ αὑτοῦ φθίνει. Thuc. 
4. 100 προσέβαλον τῷ τειχίσματι, ἄλλῳ 
τε τρόπῳ πειράσαντες καὶ μηχανὴν προσή- 
yayov. Though we can have δώμα πε- 
λάζει (Eur. Andr. 1167), ‘is carried to- 
wards the house,’ the. dat. τῷδε τἀνδρὶ 
after ἐπλησίαζον here is proof in itself 
that the verb does not govern τόπον : 
further the sense required is not ‘ap- 
proached,’ but ‘occupied.’ Brunck, ta- 
king τῷδε τἀνδρὶ as= ἐμοί, was for chang- 
ing ἐπλησίαζον to ἐπλησίαζε : which only 
adds the new complication of an irregular 
μέν and δέ. The text is probably sound. 
Heimsoeth’s conjecture, νέμων for ὁ μέν, 
with ἔπλησίαζε, is attractive, but the pa- 
renthetic éyw 6’ évi is then very awkward, 
Nauck proposes ἐν Κιθαιρῶνος νάπαις | 
(this with Blaydes) voueds διπλοῖσι ποιμνί- 
ous ἐπιστατῶν | ἐπλησίαζε: but this is to 
re-write, not to correct. 

1137 ἐξ ἦρος εἰς ἀρκτοῦρον : from 


ORAIIOV 20 ΤΥΡΆΑΝΝΟΣ 151 


He. Not so that I could speak at once from memory. 

Mer. And no wonder, master. But I will bring clear recol- 
lection to his ignorance. I am sure that he well wots of the 
time when we abode in the region of Cithaeron,—he with two 
flocks, I, his comrade, with one,—three full half-years, from 
spring to Arcturus; and then for the winter I used to drive my 
flock to mine own fold, and he took his to the fold of Laius. 


Did aught of this happen as I tell, or did it not? 


FE: 


udva Li χειμῶνι rv. 


Thou speakest the truth—though ’tis long ago. 


As the accus. was changed into the easier dat., so the dat. in 
turn became the gen. in some copies (I has χειμῶνος, with yp. χειμώνη. 


In A there 


is an erasure over the vu of χειμῶνι, but no trace (I think) of a, 





March to $ In March the 
herd of Polybus drove his flock up to 
Cithaeron from Corinth, and met the 
herd of Laius, who had brought up his 
flock from the plain of Thebes. For six 
months they used to consort in the upland 
glens of Cithaeron; then, in September, 
when Arcturus began to be visible a 
little before dawn, they parted, taking 
their flocks for the winter into home- 
steads near Corinth and Thebes.—dpx- 
τοῦρον, (the star a of the constellation 
Bootes,) first so called in Hes. Of. 566 
where (610) his appearance as a morning 
star is the signal for the vintage. Hippo- 
crates, Apident. 1. 2. 4, has περὶ ἀρκτοῦρον 
as=‘a little before the autumnal equi- 
nox’: and Thuc. 2. 78 uses περὶ ἀρκτού- 
ρου ἐπιτολάς to denote the same season. 
See Appendix. 

éxpyvous. Plato (Lege. 916 8) ἐντὸς 
ἑκμήνου, sc. χρόνου : the statement in Lidd. 
and Scott’s Lexicon (6th ed.) that it is 
Jeminine was due to a misunderstanding 
of the words πλὴν τῆς ἱερᾶς (sc. νόσου) just 
afterwards. Aristotle also has this form. 
Cp. ἕκπλεθρος (Eur.), ἕκπους, ἕκπλευρος. 
The form ἑξμέδιμνον in Ar. Pax 631 is an 
Atticism: cp. ἕξπουν Plat. Comicus fr. 
36, where Meineke quotes Philemon (a 
grammarian who wrote on the Attic dia- 
lect): ᾿Αττικῶς μὲν ἕξπουν καὶ ἕξκλινον λέ- 
γεται, ὥσπερ καὶ παρὰ Σοφοκλεῖ ἑξπηχυστί: 
adding Steph. Byz. 345 “Εξγυιος, πόλις 
Σικελίας, γραφὴν ᾿Αττικὴν ἔχουσα. Be- 
sides ἕκμηνος, Aristotle uses the form 
ἑξάμηνος (which occurs in a perhaps in- 
terpolated place of Xen., ellen. 2. 3. 9); 
as he has also ἑξάπους. The Attic dialect 
similarly preferred πεντέπους to πεντά- 
mous, ὀκτώπους to ὀκτάπους, but always 


said πενταπλοῦς, ἑξαπλοῦς, ὀκταπλοῦς. 
1138 The fact that L has χειμῶνα 
without notice of a variant, while some 
other Mss. notice it asa variant on their 
χειμῶνι, is in favour of the accus., the 
harder reading. It may be rendered ‘for 
the winter,’ since it involves the notion 
of the time during which the flock was to 
remain in the ἔπαυλα. It is, however, 
one of those temporal accusatives which 
are almost adverbial, the idea of duration 
being merged in that of season, so that 
they can even be used concurrently with 
a temporal genitive: Her. 3. 117 Τὸν 
μὲν yap χειμῶνα ὕει σῴφι ὁ θεός.. τοῦ 
δὲ θέρεος σπείροντες. . χρηΐσκοντο τῷ 
ὕδατι. 2. O85 τῆς μὲν ἡμέρης ἰχθῦς ἀ- 
γρεύει, τὴν δὲ νύκτα τάδε αὐτῷ χρᾶται. 
2.2 τὴν ὥρην ἐπαγινέειν σφι αἴγαξ, ‘at 
the due season.’ 7. 151 τὸν αὐτὸν τοῦτον 
χρόνον πέμψαντας... ἀγγέλους. Cp. above, 
1090 τὰν αὔριον πανσέληνον. The ten- 
dency to such a use of the accus. may 
have been an old trait of the popular 
language (cp. dwpiay ἥκοντες Ar. Ach. 
23, καιρὸν ἐφήκεις Soph. Az. 34). Modern 
Greek regularly uses the accus. for the 
old temporal dat.: e.g. τὴν τρίτην ἡμέραν 
for τῇ τρίτῃ ἡμέρᾳ. Classical prose would 
here use the genit.: Thuc. 1. 30 χειμῶνος 
ἤδη a ἀνεχώρησαν. The division of the year 
implied is into ἔαρ, θέρος (including ὀπώ- 
pa), and χειμών (including φθινόπωρον). 

1140 πεπραγμένον, predicate : = πέ- 
πρακταί τι τούτων ἃ λέγω; 

1141 ἐκ, properly ‘at the interval of’; 
cp. Xen. Am. 1. 10. 11 ἐκ πλέονος ἢ τὸ 
πρόσθεν ἔφευγον, at a greater distance: so 
ἐκ τόξου ῥύματος,. αἱ the interval of a bow- 
shot, 26. 3. 3. 15. 


152 ΣΟΦΟΚΛΕΟῪΣ 


AI. Φερ. εἰπὲ νῦν, τότ᾽ οἷσθα παῖδά μοί τινα 


δούς, ὡς ἐμαυτῷ θρέμμα θρεψαίμην ἐγώ: 


ΘΕ. τί δ᾽ ἔστι; 


ΘΕ. οὐκ εἰς ὄλεθρον ; 


OI. a, μὴ κόλαζε, πρέσβυ, τόνδ᾽, 


“πρὸς τί τοῦτο τοὔπος ἱστορεῖς ; 
AL: ὅδ᾽ ἐστίν, ὦ τᾶν, κεῖνος ὃς TOT ἦν νέος. 


δε τα κολαστοῦ μᾶλλον n τὰ τοῦδ᾽ ἔπη. 
ΘΕ. τί δ᾽, ὦ φέριστε δεσποτῶν, ἁμαρτάνω; 


\ 
OI. οὐκ ἐννέπων TOV Tato ὃν οὗτος ἱστορεῖ. 


OE. λέγει γὰρ εἰδὼς οὐδέν, ἀλλ᾽ ἄλλως πονεῖ. 
ΟΙ. σὺ πρὸς χάριν μὲν οὐκ ἐρεῖς, κλαίων δ᾽ ἐρεῖς. 
ΘΕ. μὴ δῆτα, πρὸς θεῶν, τὸν γέροντά μ᾽ αἰκίσῃ. 


OI. οὐχ ὡς τάχος TUS τοῦδ᾽ ἀποστρέψει χέρας ; 
ΘΕ. δύστηνος, αὐτὶ TOU; τί «προσχρήζων μαθεῖν ; 


OI. τὸν wats ἔδωκας τῷδ᾽ ὃν οὗτος ἱστορεῖ; 
ΘΕ. ἔδωκ᾽: ὀλέσθαι δ᾽ ὥφελον TO ἡμέρᾳ. 

ΟΙ. ἀλλ᾽ εἰς τόδ᾽ ἥξεις μὴ λέγων γε τούὔνδικον. 
ΘΕ. πολλῷ γε μᾶλλον, ἢν φράσω, διόλλυμαι. 


ΟΙ. ἀνὴρ ὅδ᾽, ὡς ἔοικεν, ἐς τριβὰς ἐλᾷ. 


ΘΕ. ov δητ᾽ ἔγωγ᾽, ἀλλ᾽ εἶπον ὡς δοίην πάλαι. 
On πόθεν λαβών; οἰκεῖον, ἢ ᾿ξ ἄλλου τινός: 
ΘΕ. ἐμὸν μὲν οὐκ ἔγωγ᾽, ἐδεξάμην δέ του. 

ΘΙ. τίνος πολιτών τῶνδε κἀκ ποίας στέγης ; 


ΘΕ. μὴ πρὸς θεῶν, μή, δέσποθ', ἱστόρει πλέον. 


ΓΕΔ 5 
οὐ σιωπήσας ἔσει; 
ἐπεὶ τὰ σὰ 
[150 
101 5 5 
1160 
1165 


OF ohwhas, εἴ σε ταῦτ᾽ ἐρήσομαι πάλιν. 
ΘΕ. τῶν Λαΐου τοίνυν τις ἦν γεννημάτων. 


1145 νέος] βρέφος Wecklein. 





1144 τ 8 égort;=‘what is the 
matter?’ ‘what do you mean?’ Cp. 
319 (η.).--- πρὸς τί cannot be connected 
as a relative clause with τί δ᾽ ἔστι, since 
τίς in classical Greek can replace ὅστις 
only where there is an indirect question ; 
e.g. εἰπὲ τί σοὶ φίλον; Cp. El. 316: 7}: 
339. Hellenistic Greek did not always 
observe this rule: Mark xiv. 36 ov τί ἐγὼ 
θέλω, ἀλλὰ Ti σύ. 

1145 ὦ τάν, triumphantly, ‘my good 
friend.’ It is not meant to be a trait of 
rustic speech: in PA. 1387 Neoptolemus 
uses it to Philoctetes; in Eur. Her. 321 
Iolaus to Demophon, and 2d. 688 the 
θεράπων to Iolaus; in Bacch. 802 Diony- 
sus to Pentheus. 


1146 οὐκ εἰς ὄλεθρον ; see on 430.— 
οὐ σιωπήσας ἔσει; =a fut. ρετίεοιί, --τα 
once, or once for all; Dem. or. 4 § 50 τὰ 
δέοντα ἐσόμεθα ἐγνωκότες καὶ λόγων μα- 
ταίων ἀπηλλαγμένοι. So Ant. 1067 ἀντι- 
δοὺς ἔσει, Ο. C. 816 λυπηθεὶς ἔσει. The 
situation shows that this is not an ‘aside.’ 
The θεράπων, while really terrified, could ᾿ 
affect to resent the assertion that his 
master had been a foundling. 

1147 κόλαζε: of words, Ai. 1107 
τὰ σέμν᾽ ἔπη | κόλαζ᾽ ἐκείνου. On the 
Harvard stage, the Theban at 1146 was 
about to s¢rzke the Corinthian (see § g of 
the first note in the Appendix). 

1149 ὦ φέριστε: in tragedy only here 
and Aesch. 7%. 39 (Ἐτεόκλεες, φέριστε 


ΟἸΔΊΠΟΥΣ ΤΥΡΆΝΝΟΣ 153 
ME. Come, tell me now—wottest thou of having given me 
a boy in those days, to be reared as mine own foster-son ? 
HE. What now? Why dost thou ask the question ? 
ΜΕ. Yonder man, my friend, is he who then was young. 
HE. Plague seize thee—be silent once for all! 
ΟΕ. Ha! chide him not, old man—thy words need chiding 
more than his. 


He. And wherein, most noble master, do I offend? 

ΟΕ. In not telling of the boy concerning whom he asks. 

ΗΕ. Hespeaks without knowledge—he is busy to no purpose. 

ΟΕ. Thou wilt not speak with a good grace, but thou shalt 
on pain. 

HE. Nay, for the gods’ love, misuse not an old man! 

OE. Ho, some onc—pinion him this instant! 

ΗΕ. Alas, wherefore? what more wouldst thou learn ? 

OE. Didst thou give this man the child of whom he asks? 

ΗΕ. I did,—and would I had perished that day! 

OE. Well, thou wilt come to that, unless thou tell the honest 
truth. 

Hr. Nay, much more am I lost, if I speak. 

OE. The fellow is bent, methinks, on more delays... 

HE. No, no!—I said before that I gave it to him. 

ΟΕ. Whence hadst thou got it? In thine own house, or 
from another ? 

HE. Mine own it was not—I had received it from a man. 

ΟΕ. From whom of the citizens here? from what home? 

ΗΕ. Forbear, for the gods’ love, master, forbear to ask more! 

Or. Thou art lost if I have to question thee again. 

HE. It was a child, then, of the house of Laius. 





Καδμείων ἄναξ) ; ironical in Plat. Phaedr. 
238 D. 

1152 πρὸς χάριν, so as to oblige: 
Dem. or. 8 § 1 μήτε πρὸς ἔχθραν ποιεῖσθαι 
λόγον μηδένα μήτε πρὸς χάριν: Ph. 594 
πρὸς ἰσχύος κράτος, by main force.—kAat- 
ων: see on 401. 

1154 Cp. Az. 72 τὸν τὰς αἰχμαλωτίδας 
χέρας | δεσμοῖς ἀπευθύνοντα (preparatory to 
flogging): Od. 22. 189 σὺν δὲ πόδας χεῖράς 
τε δέον θυμαλγέϊ δεσμῷ | εὖ μάλ᾽ ἀποστρέ- 
ψαντε (of Melanthius the goat-herd) ; then 
κίον ἀν᾽ ὑψηλὴν ἔρυσαν πέλασάν τε δοκοῖ- 
ow: and so left him hanging. 

1155 δύστηνος sc. ἐγώ. This agrees 
best with Soph.’s usage: see 77. 377 ὦ 
δύστηνος (n.): though the adj. could also 
refer to Oed. (cp. 1071). 


1158 εἰς τόδ᾽ -- εἰς τὸ ὀλέσθαι: Ai. 
1365 αὐτὸς ἐνθάδ᾽ ἵξομαι, 1.6. εἰς τὸ θάπ- 
τεσθαι. 

1160 és τριβὰς ἐλᾷ, will push (the 
matter) to delays (Amt. 577 μὴ τριβὰς 
ért),—is bent on protracting his delay: 
ἐλαύνειν as in Her. 2. 124 és πᾶσαν κακό- 
τητα ἐλάσαι, they said that he went ail 
lengths in wickedness: Tyrtaeus 11. 10 
ἀμφοτέρων δ᾽ els κόρον ἠλάσατε, ye had 
taken your fill of both. For the fut., ex- 
pressing resolve, cp. Ar. Av. 759 αἷρε 
πλῆκτρον, εἰ μαχεῖ. 

1161 οὐ δῆτ᾽ ἔγωγε, as Ph. 735, 
Zr. 1208. Remark πάλαι referring to 
1157: so dudum can refer to a recent 
moment. 

1167 The words could mean either: 


154 


OL 


OE. οἴμοι. 


OE. 


ZOPOKAEOYS 


7, δοῦλος, ἢ κείνου τις ἐγγενὴς γεγώς: 
πρὸς αὐτῷ γ᾽ εἰμὶ τῷ 
ΟἹ καγωγ᾽ ἀκούειν" ἀλλ᾽ ὅμως ἀκουστέον. 

κείνου γέ τοι δὴ παῖς exh lel: ἡ δ᾽ ἔσω 


εινῷ λέγειν. 
1170 


κάλλιστ᾽ ἂν εἴποι σὴ γυνὴ τάδ᾽ ὡς ἔχει. 


ΟἹ. ἢ γὰρ δίδωσιν ἦδε σοι; 
ΟΙ. ὡς πρὸς τί χρείας ; 


ΘΕ. μάλιστ᾽, ἀναξ. 


ΘΕ. ὡς ἀναλώσαιμί νιν. 


ΟἹ. τεκοῦσα τλήμων ; OE. θεσφάτων γ ὄκνῳ, κακῶν. 1175 


ΟἹ. ποίων ; : 


ΘΕ. κτενεῖν νιν τοὺς τεκόντας ἣν λόγος. 


ΟΙ. πῶς δῆτ᾽ ἀφῆκας τῷ γέροντι τῷδε σύ; 


OE. 


κατοικτίσας, ὦ δέσποϑ᾽, ὡς ἄλλην a 
δοκῶν ἀποίσειν, αὐτὸς ees ἣν" ὁ δὲ 
κάκ᾽ ἐς μέγιστ᾽ ἔσωσεν. 


ὅν φησιν οὗτος, ἴσθι δύσποτμος γεγώς. 
OI. ἰοὺ ἰού: τὰ πάντ᾽ ἂν ἐξήκοι σαφῆή. 
ὦ φῶς, τελευταῖόν σε προσβλέψαιμι νῦν, 


εἰ γὰρ οὗτος εἶ 1180 
ὅστις πέφασμαι φύς τ᾽ ἀφ᾽ ὧν οὐ χρὴν, ξὺν ols 7 
1185 


οὐ χρῆν ὁμιλῶν, οὖς τέ μ᾽ οὐκ ἔδει κτανών. 


ΧΟ. 


OTp. a. 


ἰὼ γενεαὶ βροτών, 


ἢ ὡς ὑμᾶς ἴσα καὶ τὸ μηδὲν ζώσας ἐναριθμῶ. 


1170 ἀκούων L, with most of the later Μ55., including A. But in some (as V, V®, 


ν᾽, V4) ἀκούων has been made from ἀκούειν. 


The schol. 


reads ἀκούειν (Mor. 522 C, 1093 B). 


Plutarch, who twice quotes this verse, 
in L, κἀγὼ ὡσαύτως εἰμὶ τῷ νῦν 


ἀκούειν, cannot be taken, however, as proving that he read the infin. 5 since τῷ νῦν 





(1) ‘he was one of the children of Laius’ ; 
or (2) ‘he was one of the children of the 
household of Laius,’ τῶν Λαΐου being gen. 
of of Aatov. The ambiguity is brought 
out by 1168. See on 814. 

1168 κείνου τις ἐγγενὴς γεγώς, some 
one belonging by birth to his race, the 
genit. depending on the notion of γένος 
in the adj., like δωμάτων ὑπόστεγοι, Ei. 
1386. 

1169 1 am close on the horror,—close 
on uttering it: (ὥστε) λέγειν being added 
to explain the particular sense in which 
he is πρὸς τῷ δεινῷ, as ἀκούειν defines 
that in which Oedipus is so. Cp. £/. 
542 τῶν ἐμών.. ἵμερον τέκνων...ἔσχε dal- 
σασθαι: Plat. Crito 52 Β οὐδ᾽ ἐπιθυμία 
σε ἄλλης πόλεως οὐδ᾽ ἄλλων νόμων ἔλαβεν 
εἰδέναι. 

1171 While γέ τοι, γε μέντοι, γε μὲν 
δή are comparatively frequent, γέ του δή 


is rarer: we find it in Ar. Mud. 372, 
Plato Phaedr. 264 A, Rep. 476 E, 504 A, 
7:10. 246: 

1174 «s=‘in her intention’: see on 
848.—mpos τί χρείας nearly=zpds ποίαν 
χρείαν, with a view to what kind of need 
a desire, z.¢. with what aim: cp. 1443: 

174 ἐπὶ παντί τῳ χρείας ἱσταμένῳ: 
eae 1229 ἐν τῷ (-ετίνι) ξυμφορᾶς, in 
what manner of plight. 

1176 τοὺς τεκόντας, not, as- usually, 
‘his parents’ (999), but ‘his father’: the 
plur. as τυράννοις, 1095. 

1178 ‘I gave up the child through 
pity,’ ὡς.. δοκῶν, ‘as thinking’ etc.: ze, 
as one might fitly give it up, who 50 
thought. his virtually elliptic use of 
ὡς is distinct from that at 848, which 
would here be represented by ws ἀποί- 
σοντι.---ἄλλην χθόνα ἀποίσειν (αὐτόν): 
cp. Ὅς C. 1769 ‘Oy Bas δ᾽ ἡμᾶς | ras ὠγυ- 


OIAITOY2 ΤΥΡΆΝΝΟΣ 155 


And I of hearing 


A slave? or one born of his own race? 

Ah me—I am on the dreaded brink of speech. 

; yet must I hear. 

Thou must naw, then, that twas said to be his own 


child-—but thy lady within could best say how these things are. 


OE. How? She gave it to thee? 
OE. 
ΟΕ. Her own child, the wretch? 


evil prophecies. 
OE. What were they? 
slay his sire. 
OE. 
ἜΤΕῚ 


ΗΕ Yea, O-king, 


For what end? HE. That I should make away P with it. 


HE. Aye, from fear of 


He. The tale ran that he must 


Why, then, didst thou give him up to this old man? 
Through pity, master, as deeming that he would bear 


him away to another land, whence he himself came; but he 


saved him for the direst woe. 


For if thou art what this man 


saith, know that thou wast born to misery. 


OE. 


Oh, oh! All brought to pass—all true! Thou light, 


may I now look my last on thee—I who have been found 
accursed in birth, accursed in wedlock, accursed in the shedding 


of blood! 


[He rushes into the palace. 


CH. Alas, ye generations of men, how mere a shadow do I 


ἀκούειν might be an instrum. dat. paraphrasing ἀκούων. 
1185 οὐ χρῆν ὁμιλῶν L: 
1186 ἰὼ] The rst hand in L wrote ὥ (found also in later 


conject. μάλιστ᾽. 
edd. Cp. 461. 


count your life! 


1172 καλλιστ᾽] Nauck 
οὐ χρῆν μ᾽ ὁμιλῶν r, and the older 


MsSs.); another has corrected it to ἰώ, rightly, since ἰώ answers to ὅστις in 1197. 
1188 ἐναριθμῶ] ἐναριθμῶι (2.2. ἐν ἀριθμῷ) L rst hand: the final ¢ has been almost 





γίους πέμψον. 

1180 κάκ᾽: a disyllabic subst. or adj. 
with short penult. is rarely elided unless, 
as here, it is (a) frs¢ in the verse, and 
also (6) emphatic: so O.C. 48, 796: see 
, & W. Verrall i in Journ. Phil. X11. 140. 

1182 ἂν ἐξήκοιυ, must have come true 
(cp. ΙΟΙ it), the opt. as Plat. Gorg. 502 Ὁ 
οὐκοῦν ἡ ῥητορικὴ δημηγορία ἂν etn: Her. 
Ἐν ἃ εἴησαν δ᾽ ἂν οὗτοι Κρῆτες: id. 8. 136 
τάχα δ᾽ ἂν καὶ τὰ χρηστήρια ταῦτά οἱ 
προλέγοι. 

1184 ἀφ᾽ ὧν οὐ χρῆν (φῦναι), since he 
was foredoomed to the acts which the two 
following clauses express. 

1186—1222 στάσιμον τέταρτον. See 
§ 10 of the first note in the Appendix. 

1st strophe (1186—1195). How vain 
is mortal life! *Tis well seen in Oedipus: 

1st antistrophe (1196—1203): who 
saved Thebes, and became its king: 

and strophe (1204-1212): but now 
what misery is like to his? 


and antistrophe (1213—1222). Time 
hath found thee out and hath judged. 
Would that I had never known thee! 
Thou wast our deliverer once; and now 
by thy ruin we are undone. 

1187 ὡς with ἐναριθμῶ: τὸ μηδὲν ad- 
verbially with ζώσας: 2.5. how absolutely 
do I count you as living a life which is 
no life. ζώσας should not be taken as= 
‘while you live,’ or ‘though you live.’ 
We find οὐδέν εἰμι, ‘I am no more,’ and 
also, with the art., τὸ μηδέν εἰμι, ‘I amas 
if I were not’: 77. 1107 κἂν τὸ μηδὲν ὦ: 
At. 1275 τὸ μηδὲν ὄντας. Here ζώσας is 
a more forcible substitute for οὔσας, 
bringing out the contrast between the 
semblance of vigour and the real feeble- 
ness.—loa καὶ-εἴσα (or ἴσον) ὥσπερ, a 
phrase used by Thuc. 3. 14 (ἴσα καὶ ἱκέται 
ἐσμέν), and Eur. £7. 994 (ceBifw σ᾽ ἴσα 
kal μάκαρας), which reappears in late 
Greek, as Aristid. 1. 269 (Dind. ).—-€va- 
ριθμῶ only here, and (midd.) in Eur. Or. 


Ist 
strophe. 


156 ΣΟΦΟΚΛΕΟΥΣ 
’ 4 ’ 3 Ν , 

3 τίς yap, τὶς ἀνὴρ πλέον 

4 τᾶς εὐδαιμονίας φέρει 

ὅ ἢ τοσοῦτον ὅσον δοκεῖν 
Ν / > > la 

6 καὶ δόξαντ ἀποκλιναι; 

7 τὸν σόν TOL παράδειγμ᾽ ἔχων, 

8τὸν σὸν δαίμονα, τὸν σόν, 

βροτῶν 
9 οὐδὲν μακαρίζω" 


1190 


ὦ τλᾶμον Οἰδιπόδα, 
Pros 


ὅστις καθ᾽ ,“ὑπερβολὰν 

2 τοξεύσας ἐκράτησε τοῦ πάντ᾽ εὐδαίμονος ὄλβου, 
8 ὦ Ζεῦ, κατὰ μὲν φθίσας 

4 τὰν γαμψώνυχα παρθένον 

ὄ χρησμῳδόν, θανάτων δ᾽ ἐμᾷ 

6 χώρᾳ πύργος ἀνέστα' 

7 ἐξ οὗ καὶ βασιλεὺς καλεῖ 

8 ἐμὸς καὶ τὰ μέγιστ᾽ ἐτιμάθης, ταῖς μεγάλαισιν ἐν 
9 Θήβαισιν ἀνάσσων. 


, 
αντ.α. 


1200 


τανῦν δ᾽ ἀκούειν τίς ἀθλιώτερος ; 1204 


otp. B. 


erased, A gloss ἐντάττω is written above. 1193 τὸ σόν τοι Mss. L has a 
comma after τὸ (added as if to guard against the words being read τόσον), and the 
marg. schol., τὸν σὸν βίον παράδειγμα ἔχων οὐδένα μακαρίζω καὶ εὐδαιμονίζω. As βίον 
would be a natural equivalent for δαίμονα here, the Scholiast may have read τὸν cov 
τοι: though it is also possible that he took τὸ σόν as=‘thy lot. "—rov σόν τοι 
Camerarius, and so most of the recent edd. 1196 οὐδένα MSS.: οὐδὲν Hermann. 


1197 ἐκράτησε Hermann, with some later Mss. 


(ἐκράτησε M?, Ζκράτησὲν Vat. a): 





623 εἰ τοὐμὸν ἔχθος ἐναριθμεῖ κῆδός τ᾽ 
ἐμόν τεὲν ἀριθμῷ ποιεῖ, if you make of 
account. 

1190 déper= φέρεται, cp. 590. 

1191 δοκεῖν ‘to seem,’ sc. εὐδαιμονεῖν : 
not absol., ‘to have reputation,’ a sense 
which οἱ δοκοῦντες, τὰ δοκοῦντα can some- 
times bear in direct antithests to οἱ ἀδο- 
ξοῦντες or the like (Eur. ec. 291 etc.), 
Cp. Eur. Her. 865 τὸν εὐτυχεῖν δοκοῦντα 
μὴ ζηλοῦν πρὶν dv | θανόντ᾽ ἴδῃ Tis: Ad. 
125 ὁρῶ yap ἡμᾶς οὐδὲν ὄντας ἄλλο πλὴν | 
εἴδωλ᾽ ὅσοιπερ ζῶμεν ἢ κούφην σκιάν. 

1192 ἀποκλῖναι, a metaphor from the 
heavenly bodies; cp. ἀποκλινομένης τῆς 
ἡμέρης (Her. 3. 104): and so κλίνει ἡ 
ἡμέρα, ὁ ἥλιος in later Greek: Dem. or. 1 
§ 13 οὐκ ἐπὶ τὸ ῥᾳθυμεῖν ἀπέκλινεν. Xen, 
Mem. 3. 5. 13 Ἢ. πόλιδ. πὸ Τὸ ΧΕΙ͂ΡΟΝ 


ἔκλινεν. 

1193 τὸν σόν τοι κιτιλ. The ap- 
parently long syllable τὸν (=é in 1202) 
is ‘irrational,’ having the time-value only 
of ~: see Metrical Analysis. The τὸ σόν 
τοι of the MSS. involves a most awkward 
construction:—‘having thy example; 
having thy fate, I say, (as an example)’: 
for we could not well render ‘haying, thy 
case (τὸ σόν) as an example.’ Against 
τὸν σόν, which is decidedly more forcible, 
nothing can be objected except the three- 
fold repetition; but this is certainly no 
reason for rejecting it in a lyric utterance 
of passionate feeling. 

1195 οὐδὲν βροτῶν, nothing (2.6. no 
being) among men, a stronger phrase 
than οὐδένα: Nauck compares fr. 652 οἱ 
δὲ τῇ γλώσσῃ θρασεῖς | φεύγοντες ἄτας 


OIAITOYS TYPANNOZ 157 
Where, where is the mortal who wins more of happiness than 
just the seeming, and, after the semblance, a falling away? 
Thine is a fate that warns me,—thine, thine, unhappy Oedipus 
—to call no earthly creature blest. 


For he, O Zeus, sped his shaft with peerless skill, and won 
the prize of an all-prosperous fortune; he slew the maiden with 
crooked talons who sang darkly; he arose for our land as a 
tower against death. And from that time, Oedipus, thou hast 
been called our king, and hast been honoured supremely, bear- 
ing sway in great Thebes. 


But now whose story is more grievous in men’s ears? 


ἐκράτησασ L. Blaydes writes ἐκράτησας és (for τοῦ) πάντ᾽, a former conject. of 
Hermann’s. 1200 ἀνέστα L ist hand: a much later hand has added σι Most of 
the later mss. have ἀνέστας, but L? has dvéora. Hermann preferred ἀνέστας. 
1202 f. καλεῖ ἐμὸς] To avoid the hiatus, Elmsley proposed ἐμὸς | καλεῖ, Blaydes 
καλεῖ τ᾽ | ἐμός, Heimsoeth κλύεις | ἐμός. But, as Wunder said, the hiatus is allowed 
here. Cp. 1190 φέρει | 7, Anz. 11g στόμα | €8a.—For ἐμός, Hermann and Blaydes 
give duds, in order that this verse, like the corresponding one in the strophe (1195), 
may begin with a long syllable; but this is unnecessary, since the anacrusis is com- 





ἐκτός εἰσι τῶν κακῶν" |*Apns yap οὐδὲν 
τῶν κακῶν λωτίζεται, ‘no dastard life’: 
fom. Hymn. 4. 34 οὔπερ τι πεφνγμένον 
ἔστ᾽ ᾿Αφροδίτην | οὔτε θεῶν μακάρων οὔτε 
θνητῶν ἀνθρώπων. Add Phil. 446 (with 
reference to Thersites being still alive) 
ἔμελλ᾽" ἐπεὶ οὐδέν πω κακόν γ᾽ ἀπώλετο, | 
ἀλλ᾽ εὖ περιστέλλουσιν αὐτὰ δαίμονες" [ καί 
πως τὰ μὲν πανοῦργα καὶ παλιντριβῆ 
χαίρουσ᾽ ἀναστρέφοντες ἐξ “Αιδου, τὰ δὲ 
δίκαια καὶ τὰ χρήστ᾽ ἀποστέλλουσ᾽ ἀεί. 
The οὐδένα of the Mss. involves the reso- 
lution of a long syllable (the second of ov- 
δὲν) which has an ictus; this is inadmis- 
sible, as the ear will show any one who 
considers the antistrophic verse, 1203, 
Θήβαισιν ἀνάσσων. 

1197 καθ᾽ ὑπερβολὰν τοξεύσας, having 
hit the answer to the riddle of the Sphinx, 
when Teiresias and all others had failed: 
cp. 398: Aesch. Ag. 628 ἔκυρσας ὥστε 
τοξότης ἄκρος σκοποῦ.---ἐκφράτησε. At 
1193 the Chorus addressed Oedipus: at 
1197 (ὅστις x.7.X.) they turn to invoke 
Zeus as the witness of his achievements ; 
and so in 1200 L, which here has the 
corrupt ἐκράτησας, rightly gives dvéora. 
Then at 1201 (ἐξ οὗ κ.τ.λ.} they resume 
the direct address to Oedipus, which is 
thenceforth maintained to the end of the 
ode. To read ἐκράτησας and ἀνέστας 
would be to eftace a fine trait, marking 


the passion of grief which turns from 
earth to heaven, and then again to earth. 
-τοῦ πάντ᾽ εὐδαίμονος : for the adverbial 
πάντα see on 475; also 823, 1425. 

1198 φθίσας, because the Sphinx, 
when her riddle was solved, threw her- 
self from a rock (Apollod. 3. 5): cp. 397 
ἔπαυσά νιν. 

1199 τὰν γαμψώνυχα κιτ.λ. In poetry, 
when a subs. has two epithets, the first 
may stand, with the art., beforeit, and the 
second after it. This is the ‘divided attri- 


_ bute’: see 2.11. 392 ἢ. τὸν μέγαν ΤΠ ἀκτω- 


λον εὔχρυσον : Ο. C.1234.76 Te κατάμεμπτον 
... [γῆρας ἄφιλον : 2]. 133 τὸν ἐμὸν... «πατέρ᾽ 
ἄθλιον. So Pind. Pyth. I. 95, 5. 00᾽ etc. 
This is not like τὸ σὸν στόμα...ἔλεινόν in 
672 (η.).---παρθένον : see on κόρα, 508. 

1200 θανάτων πύργος: see on 218. 

1204 ἀκούειν, to hear of, defining 
ἀθλιώτερος : Eur. AzZp. 1202 φρικώδη 
κλύειν. Whose woes are more impressive 
to others, or more cruel for himself? Cp. 
O. C. 306 modv...7d σὸν | ὄνομα διήκει 
πάντας. The constr. is τίς ἀθλιώτερος 
ἀκούειν, τίς (ἀθλιώτερος) ξύνοικος ἐν ἄταις 
x.T.\., who is more wretched to hear of 
(whose story is more tragic), who is more 
wretched as dwelling amid woes (whose 
present miseries are sharper)? It is not 
possible to supply μᾶλλον with ξύνοικος 
from ἀθλιώτερος. 


1st anti- 
strophe. 


and 
strophe. 


ZOPOKAEOYS 


, » 3 νι 4 3 / 
2 τις αταις αγρίιαις, TLS EV TOVOLS 205 
“A ’ 
ξύνοικος ἀλλαγᾷ βίου; 
SEN Ν > / ’ 
ἰὼ κλεινὸν Οἰδίπου κάρα, 


8 
sie : 7 Vbo. 
6 αὐτο paneled 
τ πὲς 
8 
9 


1208 
παιδὶ καὶ πατρὶ θαλαμηπόλῳ πεσεῖν, 1210 
TOS ποτε πῶς ποθ᾽ αἱ πατρῴαΐ σ᾽ ἄλοκες φέρειν, τάλας, 
σῖγ᾽ ἐδυνάθησαν ἐς τοσόνδε; 


avr. 8: 


ἐφεῦρέ σ᾽ ἀκονθ᾽ ὃ πάνθ᾽ ὁρῶν χρόνος" 


2 δικάζει τὸν ἄγαμον «γάμον πάλαι 


3 TEKVOUVT και τεκνούμενον. 


4 ἰὼ Λαΐειον « ὦ Σ» τέκνον, 


ὅ εἴθε σ᾽ εἴθε σε 
6 μήποτ᾽ εἰδόμαν. 
7 δύρομαι γὰρ 


mon. 
MSS. 


Cp. Metrical Analysis, p. Ixxxviii. 


1215 


ae ν 3 aN , 
ὥσπερ ἰάλεμον χέων 


1205 τίς ἐν πόνοις, τίς ἀταις ἀγρίαις 


τίς ἄταις ἀγρίαις, τίς ἐν πόνοις Hermann : who, however, in his 3rd ed. (1833) 


preferred τίς ὧδ᾽ ἐν dras, τίς ἐν ἀγρίοις πόνοις, inserting Αἰκὰ before δικάζει in 


1214. 
1214 δικάζει τ᾽ ἄγαμον γάμον : 


λιμήν. 


Hartung: πέλειν Heimsoeth. 


Hartung writes here τίς ἄταις ἀγρίαις πλέον (omitting tls ἐν πόνοις), and in 
and so Heimsoeth, but with τόσαις. for πλέον. 
μέγας λιμὴν] Heimsoeth conject. πῶς γάμου Wee, 


1208 ᾧ 
Mekler 7 στέγας (i.e. στέγη) 


1209 πατρὶ] πόσει Blaydes, as Wunder suggested. rece] ᾿μπεσεῖν 
1214 δικάζει τὸν MSS.: 


δικάζει 7’ Hermann, 


for the sake of metrical correspondence with 1205 τίς dras ἀγρίαις κιτ.λ. Gleditsch, 


keeping τόν here, would insert ἐν before ἀγρίαις in 1205. 


But neither change is 





1205 In 1214 the δικάζει τὸν of the 
Mss. should be kept (see Metrical Analy- 
sis): here the simple transposition of tls 
ἐν πόνοις is far the most probable cure 
for the metre. ἐν with drats as well as 
πόνοις: see on 734: for the redundant 
év...€0v-, 1126. 

1206 The dat. ἀλλαγᾷ might be in- 
strumental, but is rather circumstantial, 
ΞΞ- τοῦ βίου ἠλλαγμένου. 

1208 λιμὴν: schol. ὅτε μήτηρ ἦν καὶ 
γυνὴ ἡ Ἰοκάστη, ἣν, λέγει λιμένα. Cp. 
420 ff. 

1210 πεσεῖν Πεῖε-Ξ- ἐμπεσεῖν (which 
Hartung would read, but unnecessarily). 
Ar. Th. 1122 πεσεῖν és εὐνὰς καὶ γαμήλιον 
λέχος. The bold use is assisted by θαλα- 
μηπόλῳ (bridegroom) which goes closely 
with πεσεῖν. 

1211 ddoxes: cp. 
Aesch. 7h. 753. 

1212 σῖγ᾽: cp. Aesch. Ag. 37 οἶκος 


1256, Ant. 569, 


δ᾽ αὐτός, εἰ φθογγὴν λάβοι, | σαφέστατ᾽ 
ἂν λέξειεν. 

1218 ἀκονθ᾽, not as if he had been a 
criminal who sought to hide conscious 
guilt; but because he had not foreseen 
the disclosure which was to result from 
his inquiry into the murder of Laius.— 
χρόνος, which φύει ἄδηλα (411. 647): fr. 
280 πρὸς ταῦτα κρύπτε μηδέν, ὡς ὁ πάνθ᾽ 
ὁρῶν | καὶ πάντ᾽ ἀκούων (cp. note on 660) 
πάντ᾽ ἀναπτύσσει χρόνος : see on 614. 
Time is here invested with the attributes 
of the divine omniscience and justice. 

1214 δικάζει (see on 1205), prop. 

‘tries,’ as a judge tries a cause (δίκην 
δικάζει): here, ‘brings to justice,’ pun- 
ishes: a perhaps unique poetical use, for 
in Pind. Olymp. 2. 59, which Mitchell 
quotes, ἀλιτρὰ... δικάζει τις = simply ‘tries.’ 
Aesch. has another poet. use, Ag. 1412 
δικάζεις.. «φυγὴν ἐμοί = καταδικάζεις φυγὴν 
ἐμοῦ.---’γάμον πάλαι τεκνοῦντα καὶ τεκ- 


ΟΥΔΙ ΠΟΥ Σ  ΤΥΕΆΝΝΟΣ 159 
Who is a more wretched captive to fierce plagues and troubles, 
with all his life reversed ? 

Alas, renowned Oedipus! The same bounteous place of rest 
sufficed thee, as child and as sire also, that thou shouldst make 
thereon thy nuptial couch. Oh, how can the soil wherein thy 
father sowed, unhappy one, have suffered thee in silence so long? 


Time the all-seeing hath found thee out in thy despite: he 
judgeth the monstrous marriage wherein begetter and begotten 
have long been one. 

Alas, thou child of Laius, would, would that I had never 

seen thee! I wail as one who pours a dirge 


necessary, since the rst syllable of ἀγρίαις can be long: cp. Metrical Analysis, 
p- Ixxxviil. 1216 ἰὼ Λαΐειον τέκνον Mss.: Erfurdt supplied ὦ before 
τέκνον. See comment. 1217 εἴθε σ᾽ εἴθε MSS.: εἴθε σ᾽ εἴθε σε Wunder. 
1218 ὀδύρομαι MSS.: δύρομαι Seidler.—wo περίαλλα | ἰαχέων ἐκ στομάτων L. The 
later MSS. offer no variation, except περίαλα (Bodl. Barocc. 66), and ἀχέων (Ν3). 
—For ἰαχέων, Erfurdt conjectured laxxylwv.—Wecklein has given, δύρομαι γὰρ ὡς 
περίαλλ᾽ ἰαλέμων | ἐκ στομάτων, making ἰαλέμων an adj., and quoting Hesych., 
laréuwy δυστήνων, ἀθλίων: Eur. H. F. τοῦ ἰηλέμων | γόων aodds.—Burges, ὡς 
περίαλλ᾽ lav yéwv.—Neither of the two latter emendations was known to me when 
I conjectured ὥσπερ ἰάλεμον xéwv,—getting ἰάλεμον not, as Wecklein does, from 





vovdpevov: one in which ὁ τεκνούμενος has 
long been identified with ὁ τεκνῶν : Ζ. 6. 
in which the son has become the hus- 
band. The expression is of the same 
order as τά γ᾽ ἔργα μου | πεπονθότ᾽ ἐστὶ 
μᾶλλον ἢ δεδρακύτα, O. C. 266. 

1216 ἰὼ Λαΐειον ὦ τέκνον. Erfurdt’s 
ὦ is the most probable way of supplying 
the required syllable, and Reisig’s objec- 
tion to its place is answered by Az. 395 
ἔρεβος ὦ φαεννότατον. Hermann, how- 
ever, preferred ὦ, as a separate excla- 
mation: ‘Alas, of Laius (oh horror!) the 
son.’ Bothe’s Aatjov could be supported 
by Eur. Z. A. 757 Φοιβήιον δάπεδον: zd. 
fr. 775. 64 ὁσίαν βασιλήιον : but seems 
less likely here. 

1218 ff. The Mss. give δύρομαι yap ὡς 
περίαλλα [sic; in one MS. ὡς περίαλα] 
| taxéwv ἐκ στομάτων. I conjecture δύ- 
popat γὰρ ὥσπερ ἰάλεμον χέων | ἐκ στο- 
μάτων : ‘I lament as one who pours from 
his lips a dirge’: 2.6., Oedipus is to me as 
one who is dead. Cp. Pind. Jsthm. 7. 
58 ἐπὶ θρῆνον...πολύφαμον ἔχεαν, ‘over 
the tomb they poured forth a resounding 
dirge.’ My emendation has been adopted 
by Prof. Kennedy (ed. 1885). 

Every attempt to explain the vulgate 
is unavailing. (1) ὡς περίαλλ᾽ is sup- 
posed to be like ὡς ἐτητύμως, ws μάλιστα, 


‘in measure most abundant.’ Now περί- 
adda could mean only ‘ preeminently, 
‘more than others’: Soph. fr. 225 vd- 
μων | ods Θαμύρας περίαλλα μουσοποιεῖ, 
‘strains which Thamyras weaves with 
art preeminent’: Ar. Th. 1070 Τί tor’ 
᾿Ανδρομέδα | περίαλλα κακῶν μέρος ἐξέ- 
λαχον; ‘why have I, Andromeda, been 
dowered with sorrows above all women?’ 
Pindar Pyth. 11. 5 θησαυρὸν ὃν περίαλλ᾽ 
ἐτίμασε Λοξίας, honoured preeminentily. 
Here, περίαλλα is utterly unsuitable; 
and the added ὡς makes the phrase 
stranger still. 

(2) The ss. have taxéwv. Both idxeiv 
and ἰαχεῖν occur: but the latter should, 
with Dindorf, be written ἰακχέω. Eur. 
Her, 752 ἰακχήσατε: 783 ὀλολύγματα.... 
ἰακχεῖ: Or. 826 Tuvdapls ἰάκχησε τάλαινα: 
965 ἰακχείτω δὲ γᾶ Κυκλωπία. The parti- 
ciple, however, is unendurably weak after 
δύρομαι, and leaves ἐκ στομάτων weaker 
still. 

(3) ἐκ στομάτων can mean only ‘/rom 
my lips’ (the plur. as 77. 938 ἀμφιπίπτων 
στόμασιν, kissing her lips: Eur. Alc. 404 
ποτὶ σοῖσι πίτνων στόμασιν) : it could not 
mean ‘ loudly.’ 

(4) Elmsley, doubtless feeling this, took 
iaxéwv as gen. of a supposed, but most 
questionable, /ayéos, ‘loud,’ formed from 


and anti- 
strophe. 


160 


ZOPOKAEOY2 


9 , \ > 3 δὴ 9 ra 5.) A Sees , 
8 EK στομάτων. 10 Ορ OV ELTTELY, AVETTVEVOAT EK σέθεν 


\ ar ¥ 
9 καὶ κατεκοίμησα τοὐμὸν ομμα. 


10 2.09.2 


ESATTEAOS. 


5 wn lal > 3 
ὦ γῆς μέγιστα τῆσδ᾽ ἀεὶ τιμώμενοι, 


O 


> La) 


ἀρεῖσθε πένθος, εἴπερ eyyevas ἔτι 
4 > la 4 
Λαβδακείων ἐντρέπεσθε δωμάτων. 


~ 


TWV 


ἔργ᾽ ἀκούσεσθ᾽, ota δ᾽ εἰσόψεσθ᾽, ὅσον δ᾽ 


1225 


ΑΝ \ ν᾽. ὃς ν ¥ la x 
οἰμαι yap ovT av lotpoyv ovte Φασιν av 
s A § 
νίψαι καθαρμῷ τήνδε THY στέγην, ὅσα 
/ \ 3 > Y eer} > Ν wn A rN 
" κεύθει, τὰ δ᾽ αὐτίκ᾽ eis TO φῶς φανεῖ κακὰ 


4 5 » 
EKOVTa KOUK ἄκοντα. 


τῶν δὲ πημονῶν 1230 


μάλιστα λυποῦσ᾽ at φανῶσ᾽ avlaipero. .- :, 


ἰαχέων, but from ὡς περίαλλα. 


1231 αἱ L ist hand: ᾽ν added by a later 





lax}. Erfurdt conjectured ἰακχίων, ‘from 
lips wild as a bacchant’s.’? But a Greek 
poet would not have brought lacchos and 
Thanatos so close together; χωρὶς ἡ τιμὴ 
θεῶν. 

(5) ἰάλεμον gives exactly the right force ; 
for them, Oed. is as the dead. ἰάλεμος is 
a wat! for the dead in the four places of 
Eur. where it occurs (Or. 1391, Phoev. 
1033, 770. 600, 1304), in [Eur.] (hes. 
895, and in the one place of Aesch., 
Suppl. 115, which is just to our point: 
the Choius of Danaides say, wa%ea...@peo- 
μένα... | ἰηλέμοισιν ἐμπρεπῆ ζῶσα γόοις με 
τιμῶ, ‘lamenting sorrows meet for funeral 
wails (2.5. the sorrows of those who are 
as dead), while yet living, I chant mine 
own dirge.’ ἐκ στομάτων fits χέων, since 
xetv was not commonly used absolutely 
for ‘to utter’ (as by Pindar, 7. ¢c. above). 

(6) The corruption may have thus arisen 
in a cursive MS.: ἰάλεμον being written 
laxeuo, the last five letters of ὥσπερ- 
ιαλεμοχεων would first generate axewv 
(as in one MS.), or, with the second 
stroke of the μ, taxewv: the attempt to 
find an intelligible word in the imme- 
diately preceding group of letters would 
then quickly produce the familiar περί- 
αλλα (in one MS. περίαλα). The non- 
elision of the final a in the Mss. favours 
this view. As to metre, with πατρὶ in 
1209, ἃ tribrach (-τρὶ θαλαμ) answers to a 
dactyl (ὡς περι-, my ὥσπερ l-), whether 
we keep the traditional text, or adopt 


my conjecture, or that of Wecklein or of 
Burges; though Wecklein, by a strange 
oversight, has noticed this objection as if 
it were peculiar to my conjecture. Wun- 
der’s πόσει for πατρὶ in 1209 would restore 
exact correspondence, and may be right ; 
but I rather prefer, with Heinrich Schmidt 
(Composttionslehre \xiv), to regard the ws 
as an ‘irrational syllable’: see Metrical 
Analysis. 

1221 τὸ δ᾽ ὀρθὸν εἰπεῖν, like ws εἰπεῖν 
ἔπος, prefaces the bold figure of speech: 
I might truly say that by thy means (é« 
σέθεν) I received a new life (when the 
Sphinx had brought us to the brink of 
ruin); and now have again closed my eyes 
in a sleep as of death,—since all our 
weal perishes with thine. The Thebans 
might now be indeed described as στάντες 
τ᾽ és ὀρθὸν καὶ πεσόντες ὕστερον (50).— 
ἀνέπνευσα, ‘revived,’ 2.5. was delivered 
from anguish; cp. //. 11. 382 ἀνέπνευσαν 
κακότητος, had a respite from distress: 
Ai. 274 ἔληξε κἀνέπνευσε τῆς νόσου. 

1222 κατεκοίμησα: cp. Aesch. Ag. 
1293 ὡς ἀσφάδαστος.. ὄμμα συμβάλω 
τόδε: At. 831 καλῶ θ᾽ dua! πομπαῖον 
Ἑρμῆν χθόνιον εὖ με κοιμίσαι. 

1223—1530 ἔξοδος. It is told how 
Tocasta has taken her own life. The self- 
blinded Oedipus comes forth. Creon 
brings to him the children his daughters, 
but will not consent to send him away 
from Thebes until Apollo shall have 
spoken. 


ΟἸΔΙΠΟῪΣ ΤΥΡΆΝΝΟΣ 161 
from his lips; sooth to speak, ’twas thou that gavest me new 
life, and through thee darkness hath fallen upon mine eyes. 


SECOND MESSENGER (from the house). 


2ΜΕ. Ye who are ever most honoured in this land, what 
deeds shall ye hear, what deeds behold, what burden of sorrow 
shall be yours, if, true to your race, ye still care for the house 
of Labdacus! For I ween that not Ister nor Phasis could wash 
this house clean, so many are the ills that it shrouds, or will 
soon bring to light,—ills wrought not unwittingly, but of pur- 


pose. 
own choice. 


hand. Most of the later Mss. have αἱ ᾽ν. 


And those griefs smart most which are seen to be of our 


i 





1223 A messenger comes forth from 
the house. An ἐξάγγελος is one who 
announces τὰ ἔσω γεγονότα τοῖς ἔξω (He- 
sych.), while the ἄγγελος (924) brings 
news from a distance: in Thuc. 8. 51 
(τῷ στρατεύματι ἐξάγγελος γίγνεται ws, 
k.T.A.), one who betrays secrets. 

1224 f. ὅσον δ᾽: see on 20.--ἀρεῖσθε, 
take upon you, z.e. have laid upon you: 
like αἴρεσθαι ἄχθος (so Ant.go7 πόνον, Tr. 
491 νόσον): while in 24. 14. 130 μή πού 
τις ἐφ᾽ ἕλκεϊ ἕλκος ἄρηται is more like 
Ll. 12. 435 μισθὸν ἄρηται, ‘win.’—éyyevos 
Ξξ ὡς ἐγγενεῖς ὄντες, like true men of the 
Cadmean stock to which the house of 
Labdacus belonged (261, 273). 

1227 "Iotpov, the Thracian name for 
the lower course of the river which the 
Kelts called Danuvius (for this rather 
han Danubius is the correct form, Kie- 
pert Anc. Geo. § τοῦ n., Byzantine and 
modern Δούναβι:).---Φᾶσιν (707), di- 


viding Colchis from “Asia Minor and 
flowing into the Euxine. (‘Phasis’ in 
Xen. Az. 4. 6. 4 must mean the Araxes, 
which flows into the Caspian.) Soph. 
names these simply as great rivers, not 
with conscious choice as representatives 
of Europe and Asia. Ovid 2761. 2. 248 
arsit Oronles | Thermodonque citus Gan 
gesque et Phasis et Ister, Commentators 
compare Seneca Hipp. 715 Quis eluct 
me Tanais? aut quae barbaris Maeotis 
undis Pontico incumbens mari? Non 
2256 toto magnus Oceano pater Tantum 
piarit sceleris,and Shaksp. Macbeth 2.2.60 
Will all great Neptune’s ocean wash this 
blood Clean from my hand ?; where, how- 
ever, theagony of personal remorse renders 
the hyperbole somewhat more natural 


} 5.15 


than it is here in the mouth of a messenger. 

1228 καθαρμῷ, modal dat., ‘by way 
of purification,’ so as to purify.—vt- 
Wor: Eur. 7, 7. 1191 ἁγνοῖς καθαρμοῖς 
πρῶτάνιν νίψαι θέλω. The idea of washing 
off a defilement belongs to νίζειν (as to its 
cognates in Sanskrit and Old Irish, Curt. 
Etym. § 439), cp. 21. 11. 830 etc.—doa, 
causal, = ὅτι τοσαῦτα: Her. 1. 31 ἐμακάρι- 
ὧον Thy μητέρα οἵων (= ὅτι τοιούτων) τέκνων 
ἐκύρησε: Aesch. P. V. 908 ἔσται ταπεινός, 
οἷον ἐξαρτύεται | γάμον γαμεῖν: 7. 5. 757 
οὐ νεμεσίζῃ “Ape...  ὁσσάτιόν τε καὶ οἷον 
ἀπώλεσε λαὸν ᾿Αχαιῶν : 7]. 18. 262 οἷος (= 
ἐπεὶ τοῖος) ἐκείνου θυμὸς ὑπέρβιος, οὐκ ἐθε- 
λήσει | μίμνειν ἐν πεδίῳ. Cp. Ο. C. 263 n. 

1229 The construction is ὅσα κακὰ 
(τὰ μὲν) κεύθει, τὰ δὲ αὐτίκα és τὸ φῶς 
φανεῖ: cp. Z/. 1290 πατρῴαν κτῆσιν...] 
ἀντλεῖ, τὰ δ᾽ ἐκχεῖ κατ.λ. The house con- 
ceals (κεύθει) the corpse of Ιοοσαβία; it 
will presently disclose (φανεῖ) the self- 
blinded Oedipus: both these horrors 
were due to conscious acts (ἑκόντα), as 
distinguished from those acts in which 
Oed. and Iocasta had become involved 
without their knowledge (ἄκοντα). ἑκόν- 
Ta...akovTa for ἑκούσια... ἀκούσια, the 
epithet of the agent being transferred to 
the act: see on 1215. 

1231 μάλιστα, because there is not 
the consolation of recognising an inevi- 
table destiny: cp. Az. 260 τὸ yap ἐσλεύσ- 
σειν οἰκεῖα πάθη | μηδενὸς ἄλλου παραπράξ- 
avros | μεγάλας ὀδύνας ὑποτείνει : but here 
λυποῦσι refers rather to the spectators 
than to the sufferers.—at for at ἄν, as oft. 
in poetry (O. C. 395 etc.), rarely in 
prose, Thuc, 4. 17 οὗ μὲν βραχεῖς ἀρκῶσι, 
18 οἵτινες..-νομίσωσι. 


ΤΙ 


162 


ZOPOKAEOY2 


XO. λείπει μὲν οὐδ᾽ ἃ πρόσθεν ἤδειμεν τὸ μὴ οὐ 
βαρύστον᾽ εἶναι" πρὸς δ᾽ ἐκείνοισιν τί φής; 


ΕΞ. 


BE. 


αὐτὴ πρὸς QUTNS. 


ὁ μὲν τάχιστος TOV λόγων εἰπεῖν τε καὶ 
μαθεῖν, τέθνηκε θεῖον ᾿Ιοκάστης κάρα. 
ὦ δυστάλαινα, πρὸς τίνος ποτ᾽ αἰτίας ; 


ἀλγιστ᾽ ἄπεστιν" ἡ γὰρ ὄψις OU πάρα. 


ὅμως ὃς: 


πεύσει τὰ κείνης ἀθλίας παθήματα. 


ὅπως γὰρ ὀργῇ χρωμένη παρῆλθ' ἔσω 
υρῶνος, ler? εὐθὺ πρὸς τὰ νυμφικὰ 
λέχη, κόμην σπῶσ᾽ ἀμφιδεξίοις ἀκμαῖς" 


πύλας δ᾽, ὅ ὅμως εἰσῆλύ', 


καλεῖ τὸν ἤδη Λάϊον πάλαι νεκρόν, 


1235 
τῶν δὲ πρα θέντων τὰ μὲν 
ὅσον γε κἀν ἐμοὶ μνήμης ἔνι, 
1240 
ἐπιρράξασ᾽ ἔσω 
1245 


μνήμην παλαιῶν σπερμάτων ἔχουσ᾽, ὑφ᾽ ὧν 
άνοι μὲν αὐτός, τὴν δὲ τίκτουσαν λίποι 

τοῖς οἷσιν αὐτοῦ δύστεκνον παιδουργίαν. 

-yoato δ᾽ εὐνάς, ἔνθα δύστηνος διπλοῦς 


282 ἤδειμεν MSS. εἴδομεν Wecklein. 


written over ἡ by a later hand. 


1244 ἐπιρρήξασ᾽ MSS. 
ἐπιρράξασ᾽ Dobree. 


In L, a has been 
1245 κάλει MSS.: καλεῖ 





1232 λείπει, fail: Polyb. 2. 14 ἡ τῶν 
“Ad\rewy παρώρεια.. “προκαταλήγουσα λείπει 
τοῦ μὴ συνάπτειν αὐτῷ, the chain of the 
Alps, stopping short, fails of touching 
(the inmost recess of the Adriatic).—py 
ov, because of οὐδὲ with λείπει: the added 
τὸ makes the idea of the infin. stand out 
more independently of λείπευ: cp. 283.— 
ySepev, which the mss. give, should be 
kept. It was altered to ἤδεμεν by Elms. 
on Eur. Bacch. 1345 ὄψ᾽ ἐμάθεθ᾽ ἡμᾶς, ὅτε 
δ᾽ ἐχρῆν, οὐκ ἤδετε: where the εἴδετε of 
the Mss. is possible, but less probable. 
Aeschin. or. 3 ὃ 82 has ἤδειμεν : Dem. or. 
55 § 9 ὕδειτε. See Curtius, Verb 11. 230; 


Eng. tr. 432, who points out that the. 


case of the ¢hird pers. plur. is different: 
for this, the forms in ecay (as ἤδεσαν) 
alone have good authority. 

1235 θεῖον, epic epithet of kings and 
chiefs, as in //. of Achilles, Odysseus, 
Oileus, Thoas, etc., also of heralds, and 
in Od. of minstrels, as δῖος 2b. τό. 1 of 
Eumaeus: Plat. Phaedr. 234 Ὁ ouveBax- 
χευσα μετὰ σοῦ τῆς θείας κεφαλῆς (‘your 
worship’). 

1236 For πρὸς here see note on 493 


ad fin. 

1288 οὐ πάρα--οὐ πάρεστιν ὑμῖν: ye 
have not been eye-witnesses, as I have 
been. 

1239 κἀν ἐμοὶ, fe’en in me, \_ ihe 
your own memory, had you been present, 
would have preserved a more vivid im- 
pression than I can give: cp. [Plat. ] 
Alctb. τ. 127 E ἂν θεὸς ἐθέλῃ εἴ τι δεῖ καὶ 
τῇ ἐμῇ μαντείᾳ πιστεύειν, σύ τε κἀγὼ 
βέλτιον σχήσομεν. ἐν---ἔνι (:ΞΞ ἔνεστι), as 
ἐνεῖναι ἐν Ar. Eg. 1132 etc. 

1241 We are to suppose that, when 
she rushed from the scene in her pas- 
sionate despair (1072), Iocasta passed 
through the central door of the palace 
(βασίλειος θύρα) into the θυρών; "8. short 
passage or hall, opening on the court 
(αὐλή) surrounded by a colonnade (περί- 
στυλον). Across this court she hurried 
to the θάλαμος or bedroom of the master 
and mistress of the house, and shut her- 
self into it. Presently Oedipus burst into 
the court with that cry of which we heard 
the first accents (1182) as he fled from 
the scene (Bowv εἰσέπαισεν, 1252). The 
messenger and others who were in the 


OIAITOY2 TYPANNOS 163 

Cu. Indeed those which we knew before fall not short of 
claiming sore lamentation: besides them, what dost thou an- 
nounce ? 

2 Murs Thisas the shortest tale-tostell and to hear- our 
royal lady Iocasta is dead. 

Cu. Alas, hapless one! From what cause? 

2 ME. By her own hand. The worst pain in what hath 
chanced is not for you, for yours it is not to behold. Never- 
theless, so far as mine own memory serves, ye shall learn that 
unhappy woman’s fate. 

When, frantic, she had passed within the vestibule, she 
rushed straight towards her nuptial couch, clutching her hair 
with the fingers of both hands ; once within the chamber, she 
dashed the doors together at her back ; then called on the name 
of Laius, long since a corpse, mindful of that son, begotten long 
ago, by whom the sire was slain, leaving the mother to breed 
accursed offspring with his own. 


And she bewailed the wedlock wherein, wretched, she had 


borne a twofold brood, 


Erfurdt. (Brunck 'κάλει, Blaydes ἐκάλει.) So in Eur. Ale, 183, Med. 1141 the Mss. 





court watch him in terror as he raves for 
a sword and asks for Iocasta. Then the 
thought strikes him that she is in the 
θάλαμος. He bursts into it (ἐνήλατο 1261). 
They follow. There they find Iocasta 
dead, and see Oedipus blind himself. 
1242 εὐθὺ, ‘straight,’ is obviously 
more forcible here than εὐθύς, ‘without 
delay’; a distinction to which Eur. 27122. 
1197 τὴν evOds” Apyous kam Savplas ὁδόν isan 
exception rare in classical Attic. Nauck, 
with tasteless caprice, writes εὐθὺς és. 
1248 ἀμφιδεξίοις here=not simply 
‘both,’ but ‘belonging to both hands’ (for 
ἀκμαῖς alone would scarcely have been 
used for ‘hands’): soin O.C, 1112 épel- 
σατε πλευρὸν ἀμφιδέξιον can mean, ‘press 
your sides to mine on either hand. ἀμ- 
φιδέξιος usu. means ‘equally deft with 
either hand’ lancredstier|, opp. to duda- 
plorepos, ‘utterly gauche’ (Ar. fr. 432): 
hence ‘ambiguous’ (of an oracle, Her. 5. 
92). The Sophoclean use has at least 
so much warrant from etymology that 
δεξιά, from dex with added o, prop. 
meant merely ‘the catcher’ or ‘re- 
ceiver’: see Curt. Hitym. §§ 11, 266. 
1244 émppdtac’ from ἐπιρράσσω, 
Plut. Alor. 356 C τοὺς δὲ συνόντας ἐπιδρα- 
povras ἐπιρράξαι τὸ πῶμα, hastily put the 
lid on the chest. J/. 24. 452 θύρην δ᾽ 


ἔχε μοῦνος ἐπίβλης | elAarwos, τὸν τρεῖς 
μὲν ἐπιρρήσσεσκον ᾿Αχαιοί, | τρεῖς δ᾽ ἀναοί- 
γεσκον κιτ.λ. (from ἐπιρρήσσω). Hesych, 
ἐπιρρήσσει. ἐπικλείει. Plat. Prot. 314 C 
ἀμφοῖν τοῖν χεροῖν τὴν θύραν... ἐπήραξε 
(from ἐπαράσσω). In O. C. 1503 (χάλαζ) 
ἐπιρράξασα is intrans. 

1245 τὸν ἤδη A. πάλαι vexpov: for 
the order cp. O. C. 1514 αἱ πολλὰ βρονταὶ 
διατελεῖς : Ph. 1316: Z/. 183: Thue. 7. 
23 al mpd τοῦ στόματος νῆες ναυμαχοῦ- 
σαι: Isocr. or. 4 ὃ 179 τήν τε περὶ ἡμᾶς 
ἀτιμίαν γεγενημένην : Dem. or. 18 § 271 
τὴν ἁπάντων... ἀνθρώπων τύχην κοινήν: 
esp. with proper names, as Pind. O/. 13. 
53 τὰν πατρὸς ἀντία Μήδειαν θεμέναν 
γάμον: El, 283. 

1248 παιδουργίαν for παιδουργόν, ζ.6. 
γυναῖκα τεκνοποιόν (Her. 1. 50), abstract 
for concrete: see on 1 (τροφή) : cp. Od. 
3. 49 νεώτερός ἐστιν, ὁμηλικίη δέ μοι αὐτῷ 
(Ξε ὁμῆλιξ). Not acc. in appos. with sen- 
tence, ‘an evil way of begetting chil- 
dren,’ because λίποι [τοῖς οἷσιν αὐτοῦ, 
‘left zo (or for) his own,’ would then be 
very weak. 

1249 yoaro. Cp. Curtius, Verb 1. 
138, Eng. tr. 92: ‘It seems to me best 
on all grounds to suppose that shortly 
before the rise of the Greek Epic the 
[syllabic] augment became occasionally 


11--2 


164 


x ἐξ ἀνδρὸς ἄνδρα καὶ τέκν᾽ ἐκ τέκνων τέκοι. 


ΣΟΦΟΚΛΕΟΥΣ 


1250 


9 Ν 3 ~ > 2 43 5.0.9 3 ’ὕ 
χώπως μὲν ἐκ τῶνδ᾽ οὐκέτ᾽ οἷδ᾽ ἀπόλλυται: 
An ε 4 
. βοῶν γὰρ .εἰσέπαισεν Οἰδίπους, ὑφ᾽ οὗ 
οὐκ ἦν τὸ κείνης ἐκθεάσασθαι κακόν, 
> > > > ~ A > > , 
ἀλλ᾽ εἰς ἐκεῖνον περιπολοῦντ ἐλεύσσομεν. 


φοιτᾷ γὰρ ἡμᾶς ἔγχος ἐξαιτῶν πορεῖν, 
3 3 “ 


Γ2 ΕΒ 


ra ip 3 ν 
γυναῖκά ten OU eee dd Oe ὃ οπου 
υ 


κίχοι διπλῆν ἄρουραν" o 


ε καὶ τέκνων. 


λυσσῶντι δ᾽ αὐτῷ δαιμόνων δείκνυσί τις" 
A A ἴω 
οὐδεὶς γὰρ ἀνδρῶν ot παρήὴμεν ἐγγύθεν. 
ε lal 


δεινὸν δ᾽ dicas, ws ὑφηγητοῦ τινος, 


1260 


πύλαις διπλαῖς ἐνήλατ᾽" ἐκ δὲ πυθμένων 
y ἔκλινε κοῖλα κλῇθρα κἀμπίπτει στέγῃ. 

οὗ δὴ κρεμαστὴν τὴν γυναῖκ ἐσείδομεν, 

πλεκταῖσιν αἰώραισιν ἐμπεπλεγμένην. 


have κύνει for κυνεῖ. 
added o to ἄνδρα. 


1250 ἐξ ἀνδρὸσ ἄνδρα L ist hand; a later hand 
Most of the later Mss. have ἄνδρας (altered in E to ἄνδρα, 


with τὸν Οἰδίποδα written above). The plur. διπλοῦς in 1249 caused the error. 
1260 ὑφ᾽ ἡγητοῦ L (and so the Aldine): ὑφηγητοῦ r (with gloss ὁδηγοῦ in A and 
E). 1264 2.1, has πλεκταῖσ ἐώραισ (corrected from éwpatc) ἐμπεπλεγμένην (from 


ἐμπεπληγμένην)" ὁ δὲ | ὅπως δ᾽ ὁρᾷ νιν. 


Ἱ The poet prob. wrote πλεκταῖσιν αἰώραισιν 
ἐμπεπλεγμένην" | 6 δ᾽ ws ὁρᾷ vw. Then 


(1) alwpacow became alwpais, which is 





exposed to the same tendency towards 
wearing away (Verwitterung) which the 
ἀ of dpa and the é of ἔνερθε could not 
always withstand; that there were, in 
short, pairs of forms then in use, one 
with the augment and one without...The 
omission of the syllabic augment in 
Homer was purely a matter of choice... 
Post-Homeric poetry adopts the power of 
dispensing with the syllabic augment as 
an inheritance from its predecessor, and 
makes the greater use of it in proportion 
as it is removed from the language of or- 
dinary life. Hence it is that, as is shown 
by the careful investigations made by 
Renner (Stud. i. 2. 18 ff.), the omission 
of the syllabic augment is extremely rare 
in iambic, and far more common in ele- 
giac and lyric verse. Hence, as is shown 
(Stud. i. 2. 259) by Gerth, in the dialogue 
of tragedy the range of this license is 
very limited indeed, while the majority 
of instances of it occur in the slighily 
Epic style of the messengers’ speeches, 
or still more commonly in lyric passages.’ 

The tragic ῥήσεις here borrow from a 
practice more marked in epic arrative 
than in epic speeches. In Homer, where 


augmented and unaugmented forms are 
on the whole about equally numerous, 
the proportion of augmented to unaug- 
mented is in the speeches about fo to 3, 
in the narrative about 5 to 7: see Monro, 
Hom. Grammar § 6ρ.---διπλοῦς, acc. 
plur., a twofold progeny, viz. (1) Oedipus 
by Laius (ἐξ ἀνδρὸς ἄνδρα), and (2) her 
four children by Oedipus (τέκνα ἐκ τέκ- 
νων, where the poetical plur. τέκνων is for 
symmetry with τέκνα, as 1176 τοὺς τεκόν- 
Tas=Tdv πατέρα). 

1251 The order (instead οάπόλλυται, 
οὐκέτ᾽ οἶδα) is a bold ‘hyperbaton’: cp. 
O. C. 1427 τίς δὲ τολμήσει κλύων | τὰ 
τοῦδ᾽ ἕπεσθαι τἀνδρός...; and 2b. 135 f. 
Blaydes cp. Eur. Her. 205 σοὶ δ᾽ ὡς 
ἀνάγκη τούσδε βούλομαι φράσαι | σῴζειν, 
where σῴζειν ought to come before βού- 
λομαι. 

1255 φοιτᾷ, moves wildly about. 
Cp. f/. 15. 685 ὡς Alas ἐπὶ πολλὰ θυάων 
ἴκρια νηῶν | φοίτα μακρὰ B.8ds—where he 
has just been likened to a man jumping 
from one horse to another, θρώσκων 
ἄλλοτ᾽ ἐπ’ ἄλλον. So of the sharp, sudden 
visits of the νόσος, Ph. 808 ὀξεῖα φοιτᾷ καὶ 
ταχεῖ ἀπέρχεται. At. 59 φοιτῶντ᾽ ἄνδρα 


OIAITOYS ΤΥΡΆΝΝΟΣ 165 
husband by husband, children by her child. And how there- 
after she perished, is more than I know. For with a shriek 
Oedipus burst in, and suffered us not to watch her woe unto 
the end; on him, as he rushed around, our eyes were set. To 
and fro he went, asking us to give him a sword,—asking where 
he should find the wife who was no wife, but a mother whose 
womb had borne alike himself and his children. And, in his 
frenzy, a power above man was his guide; for twas none of us 
mortals who were nigh. And with a dread shriek, as though 
some one beckoned him on, he sprang at the double doors, 
and from their sockets forced the bending bolts, and rushed 

into the room. 
There beheld we the woman hanging by the neck in a 
twisted noose of swinging cords. 


found in some later Mss. (as B, V): (2) αἰώραις was changed for metre’s sake to 
ἐώραις, as it is in L, A, and others: (3) to complete v. 1264, now too short by 
a toot, the words ὁ δὲ were borrowed from ὁ δ᾽ ws at the beginning of 1265: 
and (4) ws in 1265 became the metrically requisite ὅπως. The δ᾽ after ὅπως in 
L may be a survival from the original ὁ δ᾽ ws. A has ὁ δὲ ὅπως without δ᾽. 
Wecklein reads as I do, but with ὅπως δ᾽ instead of ὁ δ᾽ ws. We seem, however, 
to need the pron, here. The case would thus resemble that cf vv. 943, 944, 
—a gap in the former verse being filled with words borrowed from the latter, 





pavidow νόσοις, ‘raving.’ Curtius (Ztym. (κοῖλα). So Oedipus, within the house, 


§ 417) would refer the word to gu, φοιτάω 
coming from @of-t-ra-w, ‘to be often’ 
(in a place). 

1255 £. πορεῖν is epexegetic of ἐξ- 
αὐτῶν, which governs a double accusa- 
tive.—(éEa:rwv) τε ὅπου κίχοι, optative, 
and not subj., because the pres. φοιτᾷ 
is historic, representing a deliberative 
subjunctive, ποῦ xixw; Cp. n. on 72 ῥυ- 
σαίμην. Xen. Hellen. 7. 4. 39 ἠπόρει τε 
ὅ τι χρήσαιτο τῷ πράγματι: 1.6. his thought 
was, τί χρήσωμαι; 

1257 ἄρουραν: see on [211- 

1259 οὐδεὶς γὰρ ἀνδρῶν : cp. Aesch. 
Ag. 662 ἤτοι τις ἐξέκλεψεν ἢ ᾿ξῃτήσατο | 
θεός τις, οὐκ ἄνθρωπος: At. 243. 

1260 ὡς ὕφηγ.: see on 966. 

1261 πύλαις διπλαῖς, the folding 
doors of the θάλαμος. Od. 2. 344 (the 
θάλαμος of Odysseus) κληισταὶ δ᾽ ἔπεσαν 
σανίδες πυκινῶς ἀραρυῖαι | δικλίδες.--πυθ- 
μένων, prop. ‘bases’: Aesch. 2. V. 1046 
χθόνα δ᾽ ἐκ πυθμένων | αὐταῖς ῥίζαις πνεῦμα 
κραδαίνοι. Here the ‘bases’ of the κλῇ- 
θρα (bolts) are the staples or sockets 
which held them. They were on the 
inner side of the doors, which Iocasta 
had closed behind her(1244). The pres- 
sure of Oedipus on the outer side forces 
the bolts, causing them to bend inwards 


gives the order διοίγειν κλῇθρα, 1287. 
Others understand: ‘forced the doors 
from their hinges or posts’: but this 
gives an unnatural sense to κλῇθρα. 
πυθμένες would then mean the στρόφιγ- 
γες (Theophr. Ast. Pl. 5. 5. 4) or pivots 
(working in sockets called στροφεῖς) 
which served as hinges. 

1264 αἰώραισιν expresses that the 
suspended body was still oscillating, and 
is thus more than dprdvas. aldpa (akin 
to delpw, ἄορ, ἀορτήρ, ἄωρος ‘uplifted,’ 
Od. 12. 89, Curt. EZtym. § 518) meant a 
swing (as in Modern Greek), or swinging 
movement: Plat. Phaed. 111 E ταῦτα δὲ 
πάντα κινεῖν ἄνω τε Kal κάτω ὥσπερ alwpay 
τινὰ ἐνοῦσαν ἐν τῇ Ὑῇ; there is a sort of 
swinging in the earth which moves all 
these things up and down; ..,.alwpetras 
δὴ καὶ κυμαίνει ἄνω καὶ κάτω, so they 
swing and surge: Legg. 789 Ὁ ὅσα τε ὑπὸ 
ἑαυτῶν (κινεῖται) ἢ καὶ ἐν αἰώραις (in 
swings) ἢ καὶ κατὰ θάλατταν ἢ καὶ ἐφ᾽ 
ἵππων ὀχουμένων. Cp. Athen. 618 Ε ἦν 
δὲ καὶ ἐπὶ ταῖς ἐώραις τις, ἐπ᾽ ᾽Ηριγόνῃ, 
ἣν καὶ ἀλῆτιν καλοῦσιν ᾧδήν, “αἱ the 
Feast of Swings there was also a song in 
memory of Erigoné, otherwise called the 
Song of the Wanderer.’ The festival 
was named ἐῶραι (small images, like the 


166 


ὁ δ᾽ ὡς ὁρᾷ νιν, δεινὰ βρυχηθεὶς τάλας 
χαλᾷ κρεμαστὴν ἀρτάνην. 
ἔκειτο “τλήμων, δεινὰ δ᾽ ἦν τἀνθένδ᾽ ὁρᾶν. 


ΣΟΦΘΟΚΛΕΘΥΣ 


1265 
ἐπεὶ δὲ 


ἀποσπάσας γὰρ “εἱμάτων χρυσηλάτους 
τὸ 
περόνας ἀπ᾽ αὐτῆς, αἷσιν ἐξεστέλλετο, 


ἄρας ἔπαισεν ἄρθρα τῶν αὐτοῦ κύκλων, 
ὁθούνεκ᾽ οὐκ ὄψοιντό νιν 


αὐδῶν τοιαῦθ, 


1270 


ov0 of ἔπασχεν. οὐθ᾽ ὁποῖ᾽ , ἔδρα κακά, 
ἀλλ᾽ ἐν σκότῳ τὸ λοιπὸν οὗς μὲν οὐκ ἔδει 


opoial’, 


ods δ᾽ ἔχρῃζεν οὐ γνωσοίατο. 
τοιαῦτ᾽ ἐφυμνῶν πολλάκις τε κοὐχ ἅπαξ 


1275 


ἤρασ σ᾽ ἐπαίρων βλέφαρα: φοίνιαι δ᾽ ὁμοῦ 
γλῆναι γένει᾽ ἔτεγγον, οὐδ᾽ ἀνίεσαν 
τ φόνου μυδώσας CD A ἀλλ᾽ ὁμοῦ μέλας 


ν᾿ ὄμβρος χαλάζης * 


αἱματοῦς ἐτέγγετο. 


which was afterwards expanded.—Nauck conjectures πλεκταῖσιν ἀρτάναισιν alw- 


ρουμένην. 


1279 ὄμβρος χαλάζησ αἵματοσ ἐτέγγετο L. Some later mss. 





oscilla offered to Bacchus, Verg. G. 2. 
389, being hung from trees) because 
Erigone had hanged herself on the tree 
under which she had found the corpse of 
her father Icarius; the name ἀλῆτις al- 
luding to her wanderings in search of him. 
Hesych. s. Ὁ. ἀλῆτις has ἐώρα : the gloss 
of Suidas (ἐώρα: ὕψωσις ἢ μέταρσι) is 
from the schol. here. ἐώρημα for alwpnua 
(the stage μηχανή) occurs in schol. Ar. 
Pax 77. aiwpa, however, is the only 
form for which there is good authority of 
the classical age. [Eustathius on 22. 3. 

108 says: ἠερέθεσθαι δὲ κυρίως μὲν τὸ ἐν 
ἀέρι κρέμασθαι, ἐξ οὗ καὶ ἡ αἰώρα. ὅτι δὲ 
ἡ ῥηθεῖσα alwpa καὶ διὰ τοῦ ε ψιλοῦ ἔχει 
τὴν ἄρχουσαν, ὡς δηλοῖ οὐ μόνον τὸ πλεκ- 
ταῖς ἐώραις ἐμπεπλεγμένην, ἀλλὰ 
καὶ τὸ μετέωρος, ἕτεροι ἐπαγωνιζέσθω- 
σαν. Prof. Kennedy quotes this to prove 
‘the classical use of éwpa.’ But it rather 
indicates that this verse furnished the only 
classical example of é#pa known to Eusta- 
thius; and there is no proof that here he 
was following an older or better Ms. than 
Ι,.1--ἐμπεπληγμένην (see crit. n.) would 
mean ‘having dashed herself into...’: but 
this can hardly be justified by the intrans. 

use of the active, Od. 22. 468 f. ὅταν... 
πέλειαι | ἕρκει ἐνιπλήξωσι: nor is it ap- 
propriate here in reference to the hanging 
corpse. 


1266 γῇ; locative dat.: see on 20: 
cp. 1451 ναίειν ὄρεσιν. 

1267 δεινὰ δ᾽, For δέ introducing 
the apodosis after a temporal protasis 
(even when it is a short one), cp. Od. 7. 
46 ἀλλ᾽ ὅτε δὴ βασιλῆος ἀγακλυτὰ δώμαθ᾽ 
ἵκοντο, | τοῖσι δὲ μύθων ἦρχε θεὰ γλαυκῶ- 
mis ᾿Αθήνη: and 2b. 184 ἐπεὶ σπεῖσάν τ᾽ 
ἔπιόν θ᾽ ὅσον ἤθελε θυμός, | τοῖσιν δ᾽ ᾿Αλ- 
κίνοος ἀγορήσατο. 

1269 περόνας (called πόρπαι by Eur. 
Ph. 62), brooches with long pins which 
could serve as small daggers: one fasten- 
ed lIocasta’s ἱμάτιον on her left shoulder, 
and another her Doric χιτών on the right 
shoulder, which the ἱμάτιον did not cover. 
The Doric χιτών was sleeveless, and 
usually made with a slit at each shoulder, 
requiring the use of brooches. (Cp. 
Guhl and Koner, Life of the Greeks and 
Romans, p. 162 Eng. tr.) In ‘The 
Harvard Greek Play’ (1882), plate-I1. p. 
26 represents Iocasta with the ἱμάτιον 
thus worn. Cp. Her. 5. 87, where the 
Athenian women surround the sole sur- 
vivor of the expedition to Aegina, κεντεύ- 
σας τῇσι περόνῃσι τῶν ἱματίων, and so 
slay him. Thus too in Eur. Hee. 1170 
the women blind Polymestor ; πόρπας 
λαβοῦσαι τὰς ταλαιπώρους κόρας | κεντοῦ- 
ow, αἱμάσσουσιν. 

1270 ἄρθρα can only mean the 


ΟἸΔΊΠΟΥΣ: ΤΎΡΑΝΝΟΣ 167 
But he, when he saw her, with a dread, deep cry of misery, 
loosed the halter whereby she hung. And when the hapless 
woman was stretched upon the ground, then was the sequel 
dread to see. For he tore from her raiment the golden brooches 
wherewith she was decked, and lifted them, and smote full on 
his own eye-balls, uttering words like these: ‘No more shall 
ye behold such horrors as I was suffering and working! long 
enough have ye looked on those whom ye ought never to have 
seen, failed in knowledge of those whom I yearned to know— 
henceforth ye shall be dark !’ 

To such dire refrain, not once alone but oft struck he his 
eyes with lifted hand; and at each blow the ensanguined eye- 
balls bedewed his beard, nor sent forth sluggish drops of gore, 
but all at once a dark shower of blood came down like hail. 


(E, V?) have αἵματός 7’.—aluarods Heath: αἱμάτων Hermann: χάλαξά θ᾽ al- 


ματοῦσσ᾽ Porson. 


For χαλάζης, Herm. once conjectured χαλαΐζῆς (2.6. χαλα ζήει5), 





sockets of the eye-balls (κύκλων). ‘He 
struck his eye-balls in their sockets,’ is a 
way of saying that he struck them full. 
ἄρθρα could not mean κόρας (pupils), as 
the schol. explains it. Eur. has another 
bold use of the word, Cyc. 624 ovyare 
πρὸς θεῶν, θῆρες, ἡσυχάζετε, | συνθέντες 
ἄρθρα στόματος, 1.6. shut your 425 and be 
still. 

1271 οὐκ ὄψοιντο x.7.\. His words 
were :—ovx ὄψεσθέ με οὐθ᾽ ὁποῖ᾽ ἔπασχον 
οὔθ᾽ ὁποῖ ἔδρων κακά, ἀλλ᾽ ἐν σκότῳ τὸ 
λοιπὸν οὗς μὲν οὐκ ἔδει ὄψεσθε, ods δ᾽ 
ἔχρῃζον οὐ γνώσεσθε: Ye shall not see 
the evils which I was (unconsciously) 
suffering and doing [as defiled and de- 
filing], but in darkness henceforth ye 
shall see those whom ye ought never to 
have seen [Iocasta and his children], and 
fail to know those whom I longed to 
know [his parents, Laius and Iocasta].— 
ἔπασχεν... ἔδρα.. ἔδει... ἔχρῃζεν can repre- 
sent nothing but imperfects of the direct 
discourse: had they represented presents, 
they must have been πάσχει, etc., or else 
πάσχοι, etc. ἔπασχεν... ἔδρα mean ‘was 
suffering,’ ‘was doing’ all this time, while 
ye failed to warn me; and express the 
reciprocal, though involuntary, wrong of 
the incestuous relation, with its conse- 
quences to the offspring. (Cp. Amt. 171 
παίσαντές Te καὶ | πληγέντες αὐτόχειρι σὺν 
μιάσματι.) 

1278 £. ἐν σκότῳ... ὀψοίαθ᾽, 2.4. οὐκ 
ὄψονται: see on 997. The other verbs 
being plural (with κύκλοι for subject), the 
subject to ἔχρῃζεν cannot be ἄρθρα κύκλων, 


but only Oed. He had craved to learn 
his true parentage (782 ff.). ὀψοίατο, 
γνωσοίατο, Ionic, as O. C. 44 δεξαίατο, 
921 πυθοίατο, 945 defolaro: E/. 211 ἀπο- 
ναίατο: Aesch. Pers. 369 φευξοίατο, 451 
ἐκσωζοίατο: Eur. H. 2. 547 ἐκτισαίατο: 
Helen. 159 ἀντιδωρησαίατο. So Thuc. 3. 
13 can say ἐφθάραται ᾿Αθηναῖοι... αἱ δ᾽ ἐφ᾽ 
ἡμῖν τετάχαται (and 4. 31, 5. 6, 7. 4). 

1275 ἐφυμνῶν, of imprecation, as 
Ant. 1305 κακὰς | πράξεις ἐφυμνήσασα τῷ 
παιδοκτόνῳ : here the idea of repetition is 
also suggested: cp. Az. 292 Bal’ del δ᾽ 
ὑμνούμενα : 50 Lat. canere, decantare. 

1276 Cp. Ant. 52 ὄψεις ἀράξας αὐτὸς 
αὐτουργῷ χερί. ὁμοῦτεαιί each blow 
(hence Ζριῤεγ7, ἔτεγγον): but in 1278 
duod=all at once, not drop by drop 
(ἀστακτί, and not στάγδην). See on 517 
(φέρον). 

1279 The best choice lies between 
Heath’s ὄμβρος χαλάζης αἱματοῦς and 
Porson’s ὄμβρος χάλαζά θ᾽ αἱματοῦσσ᾽. 
The fact that all the Mss. have χαλάζης 
and that most (including L, A) have αἵμα- 
tos favours Heath’s reading, which is also 
the stronger. Dindorf prefers Porson’s 
on the ground that such forms as aiyua- 
τοῦς, αἱματοῦν are rarer than the feminine 
forms; but this seems an inadequate 
reason. Seneca’s free paraphrase (Qed. 
978 rigat ora foedus imber, et lacerum 
caput Largum revulsis sanguinem venis 
vomit) affords no clue as to his text of 
Sophocles. μέλας ὄμβρος aiparots xa- 
Ad{ns=a shower of dark blood-drops. 
rushing down as fiercely as hail: cp. 


κομμός. 


168 ZOPOKAEOYS 


“a> 9 A » > , 
τάδ εκ δυοῖν asad fae OU Bovey 


Ἂν κάτα, 1280 


ἀλλ᾽ ἀνδρὶ καὶ γυναικὶ συμμιγῆ κακά. 
ὁ πρὶν παλαιὸς δ᾽ ὄλβος ἦν πάροιθε μὲν 
ὄλβος δικαίως" νῦν δὲ τῇδε θημέρᾳ 


στεναγμός, arn, θάνατος, αἰσχύνη, κακῶν 


ὅσ᾽ ἐστὶ πάντων ὀνόματ᾽, οὐδέν ἐστ᾽ ἀπόν. 
νῦν δ᾽ ἔσθ᾽ ὁ τλήμων ἔν τινι σχολῇ κακοῦ; 


XO. 


1285 


BS. βοᾷ διοίγειν κλῇθρα καὶ δηλοῦν τινα 


τοῖς πᾶσι Καδμείοισι τὸν πατροκτόνον, 


τ τὸν μητρός, αὐδῶν ἀνόσι᾽ οὐδὲ ῥητά μοι, 
ὡς ἐκ χθονὸς ῥίψων ἑαυτόν, οὐδ᾽ ἔτι 


[290 


μενών δόμοις ἀραῖος, ὡς ἡράσατο. 

ῥώμης γε μέντοι καὶ προηγητοῦ τινος 
δεῖται" τὸ γὰρ νόσημα μεῖζον ἢ φέρειν. 
δείξει δὲ καὶ σοί" κλῇθρα γὰρ πυλῶν τάδε 


διοίγεται" θέαμα δ᾽ εἰσόψει τάχα 


1295 


τοιοῦτον οἷον καὶ στυγοῦντ᾽ ἐποικτίσαι. 


ΧΟ. ὦ δεινὸν ἰδεῖν πάθος ἀνθρώποις, 


which Blaydes adopts, reading αἱματοῦς. 
κάτα Otto. 


The same emendation had been made by me independently. 


1280 ov μόνου κακὰ MSS. οὐ μόνου 


It is 


received by Wolff and Wecklein.—od μόνῳ κακὰ Schneidewin ; οὐ μόνου πάρα Ken- 
nedy ; οὐ μόνου μόνῳ Lachmann; οὐχ ἑνὸς μόνου Porson; οὐκ ἀνδρὸς μόνου Arndt; οὐ 





O. C. 1502 ὀμβρία | χάλαξ᾽ ἐπιρράξασα. 
Pindar has ἐν πολυφθόρῳ... Διὸς duBpy | 
ἀναρίθμων ἀνδρῶν χαλαζΐζάεντι φόνῳ (/sthm. 
4. 49) of a slaughter in which death- 
blows are rained thick as hail; and so 
χάλαζαν αἵματος (7. 6. 27): so that the 
resemblance is only verbal. 

1280 ΖΦ. Soph. cannot have written 
these two verses as they stand; and the 
fault is doubtless in 1280. Porson’s οὐχ 
ἑνὸς μόνου, though plausible, is in sense 
somewhat weak, and does not serve to 
connect 1280 with 1281. In the conjec- 
ture, οὐ μόνον κάτα, the force of the 
prep. is suitable to the image of a de- 
scending torrent which overwhelms: and 
for its place cp. Az. 969 τί δῆτα τοῦδ᾽ 
ἐπεγγελῷεν av κάτα; 2b. 302 λόγους... 
τοὺς μὲν ᾿Ατρειδῶν κάτα. 

1282 ὁ mplv,=which they had till 
lately: παλαιὸς, because the house of the 
Labdacidae was ἀρχαιόπλουτος ; tracing 
its line to Cadmus and Agenor, 268. 


1288 δικαίως, in a true sense: cp. 
853. 
1284 ΖΦ. Instead of κακὰ πάντα, ὅσα 
ὀνομάζεται, πάρεστιν, we have ὅσα ὀνό- 
ματα πάντων κακῶν ἐστι, (τούτων) οὐδὲν 
ἄπεστιν : ὄνομα κακοῦ standing for κακὸν 
ὀνομαζόμενον. So Aesch, 2. V. 210 Tata, 
πολλῶν ὀνομάτων μορφὴ μίατε μορφὴ μία 
θεᾶς πολλαχῶς ὀνομαζομένης. 

1286 ἔν τινι is right. Even if τίς 
σχολὴ κακοῦ could mean ‘what form of 
respite from misery?’ τίνι would be less 
suitable. The Chorus mean: ‘and is he 
now calmer?’—to which the answer is 
that he is s¢z77 vehemently excited. 

1289 μητέρ᾽ (Schneidewin), suggested 
by Ar. Vesp. 1178, would debase this 
passage. 

1291 δόμοις ἀραῖος, fraught with a 
curse for the house, making it accursed, 
ὡς ἠράσατο, in terms of his own curse 
(238 μήτ᾽ εἰσδέχεσθαι μήτε προσφωνεῖν, 
κιτ.λ.), according to which anyone who 


ΟἸΔΊΠΟΥΣ: TYPANNO2 169 

From the deeds of twain such ills have broken forth, not on 
one alone, but with mingled woe for man and wife. The old 
happiness of their ancestral fortune was aforetime happiness 
indeed ; but to-day—lamentation, ruin, death, shame, all earthly 
ills that can be named—all, all are theirs. 

Cu. And hath the sufferer now any respite from pain? 

2 ΜῈ. He cries for some one to unbar the gates and show 
to all the Cadmeans his father’s slayer, his mother’s—the unholy 
word must not pass my lips,—as purposing to cast himself out 
of the land, and abide no more, to make the house accursed 
under his own curse. Howbeit he lacks strength, and one to 
guide his steps; for the anguish is more than man may bear. 
And he will show this to thee also; for lo, the bars of the gates 
are withdrawn,‘and soon thou shalt behold a sight which even 
he who abhors it must pity. 


OEDIPUS. 
Cu. O dread fate for men to see, 


μονόστολα Winckelmann; οὐ μονοζνγῆ Hermann.—Dindorf rejects vv. 1280, 1281 as 
spurious. 1283 τῇδε θὴἡμέρᾳ] τῆιδέθ᾽ ἡμέραι L. (The final ε, which might easily 
be taken for a comma, is from a later hand.) τῇδ᾽ ἐν ἡμέρᾳ Erfurdt. Cp. Az. 756 
τῇδε Onuépa. 1284 dre L 1st hand, corrected to ἄτη. 1286 ἐν τίνι L. 





was knowingly ξυνέστιος with the crimi- 
nal incurred the like curse as he (270). 
Cp. Eur. Med. 608 καὶ σοῖς dpata γ᾽ οὖσα 
τυγχάνω δόμοις, 1.6. bring a curse on it. 
7. 7. 778 (κόμισαί με)...ἢ σοῖς ἀραία δώ- 
μασιν γενήσομαι. Aesch. Ag. 236 φθόγ- 
γον ἀραῖον οἴκοις. Not μενῶν δόμοις, as 
though the dat. were locative, like γῇ; 
1266. 

12987 φέρειν : Eur. Hec. 1107 κρείσσον᾽ 
ἢ φέρειν κακά: the fuller constr., Her. 3. 
14 μέζω κακὰ ἢ ὥστε ἀνακλαίειν. 

1294 The subject to δείξει is Oedipus. 
Cp. Az. 813 χωρεῖν ἕτοιμος, κοὐ λόγῳ δείξω 
μόνον. O. C. 146 δηλῶ δ᾽ : Sand I prove 
it’ (viz. that I am wretched), like τεκμή- 
ριον δές In Ar. Eccl. 933 δείξει γε καὶ 
σοί" τάχα γὰρ εἶσιν ὡς ἐμέ, a person just 
mentioned is the subject of both verbs, 
as just afterwards we have, 2d. 936, δείξει 
τάχ᾽ αὐτός. On the other hand the verb 
seems really impersonal in Ar. Ran. 1261 
πάνυ γε μέλη θαυμαστά" δείξει δὴ τάχα 
(for the subject cannot well be either μέλῃ 
or Aeschylus): and so in Her. 2. 134 διέ- 
δεξε, it was made clear: as 2. 117 δηλοῖ, 
it is manifest. In 3. 82, however, the 
subject to διέδεξε may be μουναρχίη. Cp. 
Plat. Hipp. mai. 288 B εἰ δ᾽ ἐπιχειρήσας 


ἔσται καταγέλαστος; αὐτὸ δείξει (the event 
will show): cp. Zheaet. 200 E, and see on 
341. The central door of the palace is 
now opened. Oedipus comes forth, lean- 
ing on attendants; the bloody stains are 
still upon his face. 

1296 οἷον ἐποικτίσαι, proper for one 
to pity, kal orvyovvta, even though he 
abhors it. The infin. with οἷος, as with 
other adjectives of ability or fitness (ixa- 
vos, ἐπιτήδειος, etc.): so, too, with ὅσος 
as=sufficient’: Xen. Am. 4. I. 5 ἐλεί- 
πετο τῆς νυκτὸς ὅσον σκοταίους διελθεῖν τὸ 
πεδίον. Cp. 77. 672: fr. 598. 8 φεῦ" κἂν 
ἀνοικτίρμων τις οἰκτίρξιέ νιν. 

1297--1968 A κομμός (see p. 9). 
The Chorus begin with anapaests (1297 
—1306). The first words uttered by 
Oedipus are in the same measure (1307 
—1311). Then, after a single iambic 
trimeter spoken by the Chorus (1312), 
(1) 15¢ strophe 1313—1320=(2) 1st antt- 
strophe 1321— 1328; (3) 22d strophe 1329 
—1348=(4) 22d antistrophe 1349—1368. 
Oedipus here speaks in dochmiac mea- 
sures blended with iambic; the Chorus, 
in iambic trimeters or dimeters only. 
The effect of his passionate despair is 
thus heightened by metrical contrast with 


170 ΣΟΦΟΚΛΕΟΥΣ ; 


ὦ δεινότατον πάντων ὅσ᾽ ἐγὼ 
προσέκυρσ᾽ ἤδη. τίς σ᾽, ὦ τλῆμον, 
προσέβη μανία; τίς ὁ πηδήσας 
μείζονα δαίμων τῶν μακίστων 

πρὸς on δυσδαίμονι μοίρᾳ; 


φεῦ φεῦ, “ δύστην᾽" 

ἀλλ᾽ οὐδ᾽ ἐσιδεῖν δύναμαί σ᾽, ἐθέλων 
πόλλ᾽ ἀνερέσθαι, πολλὰ πυθέσθαι, 
πολλὰ δ᾽ ἀθρῆσαι" 

τοίαν φρίκην παρέχεις μοι. 














1305 


ΟἹ: 


αἰαῖ, φεῦ φεῦ, δύστανος ἐγώ, 
ποῖ yas φέρομαι τλάμων ; πᾷ μοι 
Ἁ la 

φθογγὰ ὃ διαπωτᾶται φοράδην ; 1310 
1299 τλῆμον has been made from τλήμων in L. After this verse, v. 1302 (πρὸς σῇ.. 
μοίρᾳ) had been written by an oversight, but has been partially erased, dots having been 
placed above it: and it is repeated in its proper place. 1301 μακίστων] In L 
the rst hand had written κακίστων, but altered the initial x into uw. Some of the later 
MSS. (as B and V) have κακίστων. 1308 φεῦ φεῦ δύστανοσ L, and so most of 
the later Mss.: but T has φεῦ φεῦ δύσταν᾽, which is preferred by Hermann and 
Bothe. The latter writes δύστην᾽, (and so Elmsley,) because Sophocles did not 
admit Doric forms in choral anapaests. That rule is subject to exceptions (see on 
Ant. 110): but here, at least, the Doric form seems unsuitable; see commentary. 
I formerly read φεῦ δύστανος (the ds could be excused by the pause); but 
now prefer the other reading. Dindorf deletes the words, on the assumption that 





a more level and subdued strain of sor- 
row. Compare Az. 348—429, where the 
κομμός has in this sense a like character, 
Some regard the κομμός as beginning only 
at 1313; less correctly, I think. Its 
essence is the antiphonal lament rather 
than the antistrophic framework. 

1298 804...mpocékupoa: I know no 
other example of an accus. after mpoo- 
κυρεῖν, which usu. takes the dat.: but 
the compound can at least claim the 
privilege of the simple κυρεῖν. The neut. 
plur. accus. of pronouns and adjectives 
can stand after τυγχάνειν and κυρεῖν, not 
as an accus. directly governed by the 
verb, but rather as a species of cognate 
or adverbial accus.: Ph. 509 ἀθλ᾽ ola 
μηδεὶς τῶν ἐμῶν τύχοι φίλων : O. C. 1106 
αἰτεῖς ἃ τεύξει (which need not be ex- 
plained by attraction): Aesch. Cho. 711 
τυγχάνειν τὰ πρόσφορα, 10. 714 κυρούν- 
των..«τὰ πρόσφορα: Eur. Ph. 1666 οὐ γὰρ 
av τύχοις τάδε: cp. Munro on 4g, 1228 ff. 
ola...revierat in Journ. Phil. ΧΙ. 134. 
In Hipp. 746 τέρμονα κύρων is not simi- 


lar, since xtpwy=‘reaching,’ and the 
accus. is like that after ἀφικνεῖσθαι. 
1300 ff. ὁ πηδήσας... μοίρᾳ; ‘who is 
the deity that hath sprung upon thy hap- 
less life with a leap greater than the 
longest leap?’ z.¢.:‘has given thee sorrow 
which almost exceeds the imaginable limit 
of human suffering?’ For μείζονα τῶν 
paklorwvsee on 465 ἄρρητ᾽ ἀρρήτων. The 
idea of a malignant god leaping from 
above on his victim is frequent in Greek 
tragedy: see on 263. But here μακίσ- 
τῶν, as in 311 va, combines the notion 
of swooping from above with that of 
leaping 4o a far point,—as with Pindar 
μακρὰ.. «ἅλματα (Vem. 5. 19) denote sur- 
passing poetical efforts. We should then 
conceive the δυσδαίμων μοῖρα, the ill-fated 
life, as an attacked region, far into which 
the malign god springs. Here we see a 
tendency which may sometimes be ob- 
served in the imagery (lyric especially) of 
Sophocles: the zmage is slightly crossed 
and blurred by the interposing notion 
of the ¢himg: as here he was thinking, 


ΟἸΔΊΠΟΥΣ TYPANNOZ 71 
O most dreadful of all that have met mine eyes! Unhappy one, 
what madness hath come on thee? » Who is the unearthly foe 
that, with a bound of more than mortal range, hath made thine 
ill-starred life his prey? 

Alas, alas, thou hapless one! Nay, I cannot e’en look on 
thee, though there is much that I would fain ask, fain learn, 

‘much that draws my wistful gaze,—with such a shuddering dost 


thou fill me! 
ΟΕ. 


= 


they came in from 1308.—o’ ἐθέλων τ: σε θέλων L. 
rious the words πόλλ᾽ ἀνερέσθαι, πολλὰ πυθέσθαι, πολλὰ δ᾽ ἀθρῆσαι. 
alalat| φεῦ φεῦ" δύστανος ἔγώ" ποῖ yao | etc. 
times (as T), others only twice (as V4, A). 
probably right, in view of the division of the verses. 


πᾶι μοι POoyya | διαπέταται φοράδην |. 


Woe is me! Alas, alas, wretched that I am! Whither, 
whither am I borne in my misery? 
abroad on the wings of the air? 


How is my voice swept 


1304 Nauck rejects as spu- 
1307 Φ. Lhas 
Some of the later mss. have al four 
I now think that the latter is most 
1309 L has φέρομαι TAGuwY* 


The only variants for dvarérarat in the later 


Mss. are the corrupt διέπταται and διαπέπταται, both of which probably arose from 


διαπέταται itself. 


Musgrave and Seidler conjectured διαπωτᾶται, and so Blaydes: 


Kennedy, πέταται: F. Bellermann, διαπεπόταται (Dor. for -πεπότηται), so that the 


verse should be a proceleusmaticus (-4~~Y~—+4~~-4). 


Nauck, following Din- 


dorf’s former view, writes πᾷ μοι φθογγά; without any verb; and then, φοράδην, ὦ 





‘what suffering could have gone further?’ 
See on δι᾽ αἰθέρα τεκνωθέντες, 866. With 
Aeschylus, on the other hand, the ob- 
scurity of imagery seldom or never a- 
rises from indistinctness of outline, but 
more often from an opposite cause,—the 
vividly objective conception of abstract 
notions, 

1302 πρὸς with dat., after a verb of 
throwing or falling, is warranted by epic 
usage: Od. 5. 415 μήπως μ᾽ ἐκβαίνοντα 
βάλῃ λίθακι ποτὶ πέτρῃ | κῦμα μέγ᾽ ap- 
παξαν: 74. 20. 420 λιαζόμενον προτὶ γαίῃ, 
sinking to earth. Az. 95 mpds...crpaTy, 
97 πρὸς ᾿Ατρείδαισιν are different, since 
no motion is strictly implied. Here the 
conjecture ἐπὶ is metrically admissible 
(Ag. 66 κάμακος θήσων Δαναοῖσι, Pers. 48 
φοβερὰν ὄψιν προσιδέσθαι), but needless. 

1808 The Attic δύστην᾽ harmonises 
with of (1302) and φρίκην (1306), while 
δύσταν᾽ would hardly be confirmed by 
μακίστων, since Tragedy used the latter 
form, and not μήκιστος, in dialogue also 
(Aesch. fr. 275: cp. Ag. 289: so Pers. 
698 μάκιστῆρα). Theuse of Attic forms 
by the Chorus helps to bring out the 
more passionate lyric tone which Do- 
ricisms lend to the words of Oedipus 
(1307 f.). Cp. ἢ. on Antz. 804 f. 

1304 The fate of Oedipus is a dark 


and dreadful mystery into which they are 
fain to peer (ἀνερέσθαι, πυθέσθαι : cp. 
the questions at 1299 ff., 1327): in its 
visible presentment it has a fascination 
(ἀθρῆσαι) even for those whom it fills 
with horror. 

1810 διαπέταται (MSS.) is unques- 
tionably corrupt. The view that these 
are anapaests of the ‘freer kind’ (‘ex 
liberioribus,’ Herm.) does not explain 
a verse which is not anapaestic at all. 
διαπωτᾶταν is far the most probable re- 
medy. The epic πωτᾶσθαι, which Pind. 
uses, is admissible in lyrics. When there 
is no caesura after the 2nd foot, there is 
usually one in the 3rd: cp. however 
Aesch. P.V.172 καί μ᾽ οὔ τι μελιγλώσσοις 
πειθοῦς: and Ar. Av. 536, Fax 1002. 
Cp. Ο. C.1771 dtaxwrdtow|uer ἰόντα φόνον. 
The wilder and more rugged effect of such 
a rhythm makes it preferable here to 
φθογγὰ φοράδην διαπωτᾶται, though the 
hiatus before / (in 1311) would be justi- 
fied by the pause. To the conjecture 
πέτεται (or πέταται) it may be objected 
that the notion of dispersed sounds sup- 
ports the compound with διά. Hermann 
simply omitted διαπέταται, dividing thus: 
alat-— | d0cravos—| τλάμων ; πᾶ μοι 
φθογγὰ φοράδην; Bergk, πᾶ μοι | φθογγά; 
διά μοι πέταται φοράδην. Schneidewin 


/ 
στρ: a. 


ys 


> ΄ 
QvT. a. 


172 


ἰὼ δαῖμον, ἵν ἐξήλου. 


ΣΟΦΟΚΛΕΟΥΣ 


΄ 350.» ’ 
ΧΟ. ἐς δεινόν, οὐδ᾽ ἀκουστόν, οὐδ᾽ ἐπόψιμον. 


ΟΣ ’ 
Ol.1i@ σκότου 


, > ‘ - l4 9 (2 ¥ 
νέφος ἐμὸν ἀπότροπον, ἐπιπλόμενον aaron, 


Ψ 
8 ἀδάματόν τε καὶ δυσούριστον « ὄν.» 


» 
4 οιμοι, 


1315 


Ὑ 7\59 > ® 3 “ὃ 3 ν 
δ οἴμοι μάλ᾽ αὖθις" οἷον εἰσέδυ p ἅμα 
“ » Ν ; lal 
6 κέντρων TE τῶνδ᾽ οἴστρημα καὶ μνήμη κακῶν. 
ΧΟ. τ Kai θαῦμά γ᾽ οὐδὲν ἐν τοσοῖσδε πήμασιν 


8 διπλᾶ σε πενθεῖν καὶ διπλᾶ φέρειν κακά. 


ΟΙ. 1 ἰὼ φίλος, 


1320 


\ Ν 5. ΜΌΝ 27 » 4 »Ὰ Ν 
5 σὺ μὲν ἐμὸς ἐπίπολος ἔτι μόνιμος" ETL γὰρ 
8 ὑπομένεις με τὸν τυφλὸν κηδεύων. 


4 φεῦ φεῦ" 


δαῖμον, ἐνήλω. 
Nauck. 


Laud. 54 0 is written over ὦ, with gl. ἐπερχόμενον. 
1315 ἀδάμαστον MSS.: ἀδάματον Hermann.— 


(as B, E, V?, Bodl. Barocc. 66). 


δυσούριστον MSS.: δυσούριστον ὃν Hermann. 


1311 ἰὼ δαῖμον ἵν᾽ ἐξήλου 1, (ἐξήλω τὴ : ἐξήλλου Hermann: ἐνήλω 
1314 ἐπιπλώμενον L. Some of the later Mss. have this reading. In Bodl. 


Others have the true ἐπιπλόμενον 


I conjecture δυσούριστ᾽ ἰόν. 1820 φο- 





(ed. Nauck) πᾶ μοι φθογγά; φοράδην, 
ὦ δαῖμον, ἐνήλω.--- φοράδην = ‘in the man- 
ner of that which is carried’; here corre- 
lative to φέρεσθαι as said of things which 
are swept onward by a tide or current: 
thus, of persons deficient in self-restraint, 
Plat. Zheaet. 144 B arrovres φέρονται 
ὥσπερ τὰ ἀνερμάτιστα πλοῖα, they are hur- 
ried away on currents like boats without 
ballast: Cvat. 411 C ῥεῖν καὶ φέρεσθαι: 
Rep. 496 Ὁ πνεῦμα φερόμενον. He has 
newly lost the power of seeing those to 
whom he speaks. He feels as if his voice 
was borne from him on the air in a direc- 
tion over which he has nocontrol. With 
the use of the adverb here, cp. βάδην, 
δρομάδην, σύδην. Elsewhere φοράδην is 
parallel with φέρεσθαι as=to be carried, 
instead of walking: Eur. Andr. 1166 
φοράδην... δῶμα πελάζει, 1.5. borne in a 
litter: Dem. or. 54 § 20 ὑγιὴς ἐξελθὼν 
φοράδην ἦλθον οἴκαδε. Such adverbs in 
τδην, which were probably accusatives 
cognate to the notion of the verb, are 
always formed from the verbal stem, (a) 
directly, like βά-δην, or (6) with modified 
vowel and inserted a, like φοράδην instead 
of ἔφερδην, σποράδην instead of ἔσπερδην. 


1811 ἐξήλου. Ina paroemiac, the foot 
before the catalectic syllable is usually 
an anapaest, seldom, as here (é&A—), 
a spondee: but cp. Aesch. ers. 
ἵππων 7 ἐλατὴρ Σωσθάνης : Suppl. 7 ψή- 
oy πόλεως γνωσθεῖσαι: tb. 976 Bate 
λαῶν ἐν χώρῳ: Ag. 366 βέλος ἠλίθιον 
σκήψειεν. L and A are of the Mss. 
which give ἐξήλου: and good MS. au- 
thority supports ἐνήλου in Aesch. Pers. 
516, εἰσαλοίμην in Soph. fr. 685, ἥλοντο 
in Xen. Hellen. 4. 4. 11. The evidence, 
so far as it goes, seems to indicate that, 
while ἡλάμην (itself rare in prose) was 
preferred in the indicative, a form ἡλό- 
μὴν was also admitted: see Veitch, Jrreg. 
Verbs, ed. of 1879. Blaydes gives ἐξήλω: 
Elms. gave ἐξάλω, ‘inaudite δωρέξων,᾽ in 
Ellendt’s opinion: but Veitch quotes 
Theocr. 17. 100 ἐξάλατο. The imperf. 
ἐξήλλου, which Dindorf, Campbell and 
others read, was explained by Hermann 
as =dendebas, 1.6. ‘whither wast thou pzr- 
posing toleap?’ To this I feel two ob- 
jections: (1) the unfitness of thus re- 
presenting a swift act: (2) the use of 
ἵνα, which means where. This could 
not be used with the émperfect of a verb 


ΟἸΙΔΙΠΟΥΣ. ΤΥΡΑΝΝΟΣ 


173 


Oh my Fate, how far hast thou sprung! 
CH. To a dread place, dire in men’s ears, dire in their sight. 


ΟΕ. O thou horror of darkness that enfoldest me, visitant 
unspeakable, resistless, sped by a wind too fair! 

Ay me! and once again, ay me! 

How is my soul pierced by the stab of these goads, and 
withal by the memory of sorrows! 

(ΓΗ. Yea, amid woes so many a twofold pain may well be 


thine to mourn and to bear. 


ΟΕ. Ah, friend, thou still art steadfast in thy tendance of 
me,—thou still hast patience to care for the blind man! Ah me! 


petv L, with some of the later Mss.: others (including A) have φέρειν. 
1828 we Erfurdt: ἐμὲ Mss. 


Nauck gives θροεῖν. 


See comment. 
(Instead of ἐμὲ τὸν τυφλόν, T 


has τόν γε τυφλόν, an attempt to restore the metre.) Hermann conjectured ἔτι γὰρ 


ὑπομένεις" τυφλόν Te κήδενε (with δυσούριστον οἴμοι in 1315). 


For κηδεύων, Linwood 





of motion (as iva ἔβαινε, instead of of), 
but only with the Zerfect, as ἵνα βέβηκε 
(1.6. where zs he now) or the aorist 
when equivalent to the perfect: as O. C. 
273 ἱκόμην (I have come) ἵν᾽ ἱκόμην. So, 
here, the aor. alone seems admissible: 
ἵν᾽ ἐξήλου, where ast thou leaped to, 7.2. 
where art thou? cp, 1515 ἵν᾿ ἐξήκεις, and 
see On 947. 

1314 ἀπότροπον -- ὅ τις ἂν ἀποτρέποιτο 
(Hesych.): and so Az. 608 τὸν ἀπότροπον 
ἀΐδηλον “Acdav, such as all would turn 
away from, abhorred. Not, ‘turning 
away from others,’ ‘solitary,’ as Bion 
Idyll. 2. 2 τὸν amérporov..."Epwra,—éte- 
πλόμενον -- ἐπιπελόμενον, pres. part., as 
Od. 7. 261 ἐπιπλόμενον ἔτος ἦλθε. 

1815 δυσούριστον is defective by one 
syllable as compared with 1323 τυφλὸν 
κηδεύων. Now the second syllable of 
κηδεύων is ‘irrational,’ 2.5. it is a long 
syllable doing metrical duty for a 
short one (the third of an antibacchius, 
Hence in this verse also the 
penultimate syllable can be either long 
or short. Hermann’s δυσούριστον ὄν 
is therefore metrically admissible. It is, 
however, somewhat weak, and the sound 
is most unpleasing. I should rather pro- 
pose δυσούριστ᾽ ἰόν: for the adverbial 
neut. plur., cp. ὑπέροπτα.. πορεύεται (883, 
where see note) ; for the part., Plat. Legs, 
873 E παρὰ Geod...Bédos ἰόν. Nauck cone 
jectured δυσοιώνιστον. Blaydes gives 
δυσεξούριστον (not found), in the dubious 


= avy 


sense of ‘hard to escape from.’ 

1318 κέντρων, not literally the pins 
of the brooches, (which we can scarcely 
suppose that he still carried in his hands,) 
but the stabs which they had dealt: as 
piercing pangs are κέντρα, 77. 840. 

1319 ἐν τοσοῖσδε πήμασιν, when thy 
woes are so many: cp. 893 ἐν τοῖσδ᾽. 

1320 πενθεῖν... καὶ φέρειν. The form 
of the sentence, in dependence on θαῦμα 
οὐδέν, seems to exclude the version: ‘It is 
not strange that, ὧς you bear, so you 
should mourn, a double pain’ (parataxis. 
for hypotaxis). Rather the sense is: 
‘that you should mourn (aloud) and (in- 
wardly) suffera double pain ’—7.e., the phy- 
sical pain of the wounds, and the mental 
pain of retrospect. I do not agree with 
Schneidewin in referring διπλᾶ πενθεῖν. 
to the double οἴμοι (1316 f.) as=‘make a 
twofold lament.’ The φέρειν of A must 
be right. φορεῖν can stand for φέρειν 
‘to carry’ when habitual carrying is 
implied (Her. 3. 34, and of bearers in 77. 
965): or fig., of mental habit (ἦθος φορεῖν 
Ant. 705): but φορεῖν κακά could only 


mean ‘to carry ills about with thee’ ;. 


which is not appropriate here. 
1322 μόνιμος, steadfast: Xen. Cyr. 


8. 5. 11 of μονιμώτατοι πρόσθεν ὄντες. 


(said of hoplites), Cp. Az. 348 ff. 
where Ajax addresses the Chorus as μόνοι 


ἐμῶν φίλων, | μόνοι ἐμμένοντες ἔτ᾽ ὀρθῴ. 


νόμῳ. 


Ist 
strophe. 


Ist anti- 
strophe. 


στρ. β΄. 


174 


ὅ οὐ “γάρ με λήθεις, ἀλλὰ γιγνώσκω σαφώς, 


ZOPOKAEOY2 


1325 


6 καίπερ σκοτεινός, τήν YE σὴν αὐδὴν ὅ κως 
ΧΟ. τῷ δεινὰ δράσας, TOS ἔτλης τοιαῦτα σὰς 


ΜΝ an 
8 oWEeLs papavat ; 


4 > > 7 / 
TiS O ETNPE δαιμόνων ; 


ΟΙ. ᾿᾿Απόλλων τάδ᾽ ἦν, ᾿Απόλλων, φίλοι, ἐκ 
ε \ Ν A Sees BNI 19S / 
20 κακὰ κακα τελών ἐμα τάδ ἐμα πάθεα. 1330 
8 ἔπαισε δ᾽ αὐτόχειρ νιν οὔτις, ἀλλ᾽ ἐγὼ τλάμων." 


4 τί γὰρ ἔδει he ὁρᾶν, 


ὅ ὅτῳ γ᾽ ὁρῶντι μηδὲν ἢ ἣν ἰδεῖν γλυκύ; 


ΧΟ. 6 ἦν ταῦθ᾽ ὅπωσπερ καὶ σὺ φής. 


ΟΙ. τί δητ᾽ ἐμοὶ βλεπτόν, ἢ 


8 στερκτόν, n προσήγορον 


9 er ἔστ᾽ ἀκούειν ἡδονᾷ, “φίλοι; 


10 ἀπάγετ᾽ ἐκτόπιον ὅτι τάχιστά με, 
Ε 3" Φ 5 6 
μέγ ολέθριον, 


\ 4 » δὲ ἣν la) 
12 τὸν καταρατότατον, ἔτι δὲ Kal θεοῖς 


5 \ 
ll ἀπάγετ᾽, ὦ φίλοι, τὸν 


18 ἐχθρότατον βροτῶν. 


1340 


1345 


ΧΟ. 14 δείλαιε τοῦ νοῦ τῆς τε συμφορᾶς ἴσον, 


1ιὅ ὡς σ᾽ ἠθέλησα μηδέ γ᾽ 


proposed κηδεμών. 


x a , 
αν γνῶναι ΠΌΤΕ, 


1330 In L the rst hand wrote 6 κακὰ τελῶν τάδ᾽ ἐμὰ πάθεα : 
an early hand added a second κακὰ after ὁ, and a second éua before τάδ᾽, 


Many of the 


later MSS. have κακὰ only once (the second having been taken for a dittographia), 


while they have éua twice (owing to the interposed τάδ᾽). 
1841 τὸν ὀλέθριον μέγαν L: τὸν ὀλέθριον μέγα τ (B, E, T): τὸν 
Turnebus conjectured τὸν ὄλεθρον μέγαν (received by Brunck and 
1348 L has wo (made from ὅσσ᾽ or ὅσ) σ᾽ ἠθέ- 


adova Dindorf. 
μέγ᾽ ὀλέθριον Erfurdt. 
others): Bergk, τὸν ὄλεθρόν με γᾶς. 


1339 ἡδονᾷ MSS.: 





1325 A distinct echo of //. 24. 563 καὶ 
δὲ σὲ γιγνώσκω, IIpiaue, φρεσίν, οὐδέ pe 
λήθεις. Besides λήθω, λήσω, λέληθα, 
Soph. has ἔληθον (5). 1359). Cp. Ο. Ὁ. 
891, where Oed. recognises the voice of 
Theseus. 

1326 σκοτεινός: cp. Af. 
σκοτώσω βλέφαρα καὶ δεδορκότα. 

1329 £. ᾿Απόλλων. The memory of 
Oedipus (cp. 1318) is connecting the 
oracle given to him at Delphi (789) with 
the mandate which afterwards came 
thence (106). Apollo was the author of 
the doom (τελῶν), but the instrument of 
execution (ἔπαισε) was the hand of 
Oedipus. 

1330 ὁ κακὰ κακὰ κιτιλ. The doch- 
miac metre is sound (see Metrical Analy- 
sis): it is vouados in the antistrophe 


85 ἐγὼ 


(1350) which is corrupt. Prof. Camp- 
bell, however, retaining the latter, here 
changes the second κακὰ to κακῶς, and 
the first ἐμὰ to ἐμοί. The iteration of 
τάδε, κακὰ, ἐμὰ is in a style which the 
lyrics of tragedy admitted where vehe- 
ment agitation was expressed. Euripides 
carried it to excess. But here, at least, it 
is in place. 

1331 ar or ὄψεις (adap --οὔτις 
(ἄλλος), ἀλλ - Οὐ. 8. 311 ἀτὰρ οὔ 
τί μοι πον ἄλλο: | ἀλλὰ τοκῆε δύω. 
Schneid. cp. //. 21. 275 ἄλλος δ᾽ οὔτις 
μοι τόσον αἴτιος οὐρανιώνων | ἀλλὰ [instead 
οὗ ὅσον] φίλη μήτηρ. : 

1337 ff. The simple mode of expres- 
sion would have been: ri ἐμοὶ ἡδέως 
βλεπτόν, ἢ στερκτόν, ἢ ἀκουστὸν ἔτ᾽ ἐστίν ; 
what henceforth can be pleasurably seen, 


ΟἸΔΊΠΟΥΣ. ΤΥΡΑΝΝΟΣ 175 
Thy presence is not hid from me—no, dark though I am, yet 
know I thy voice full well. 

Cu. Man of dread deeds, how couldst thou in such wise 
quench thy vision? What more than human power urged thee ? 


ΟΕ. Apollo, friends, Apollo was he that brought these my 
woes to pass, these my sore, sore woes: but the hand that 
struck the eyes was none save mine, wretched that 1 am! Why 
was I to see, when sight could show me nothing sweet ? 

(ΓΗ. These things were even as thou sayest. 

OE. Say, friends, what can I more behold, what can I love, 
what greeting can touch mine ear with joy? MHaste, lead me 
from the land, friends, lead me hence, the utterly lost, the thrice 
accursed, yea, the mortal most abhorred of heaven! 

CH. Wretched alike for thy fortune and for thy sense 
thereof, would that I had never so much as known thee! 


Anoa μὴδ᾽ (sic) ἀναγνῶναί ποτ᾽ ἄν. Instead of ποτ᾽ dv, some later Mss. (including A) have 
more. Asin 561 ἂν μετρηθεῖεν was corrupted to ἀναμετρηθεῖεν, so here ἀναγνῶναι is 
probably a corruption of ἂν γνῶναι. Hermann restored ὥς σ᾽ ἠθέλησα μηδέ γ᾽ av γνῶναί 
ποτε. ‘This is slightly nearer to the mss. than Dindorf’s ὡς ἠθέλησα μηδέ σ᾽ ἂν γνῶναί 
more: and γε suits the emphasis (‘never so much as known thee’).—Dobree proposed 
ws σ᾽ ἠθέλησα μηδαμὰ γνῶναί ποτ᾽ ἄν. (For the short vowel lengthened before γν, cp. 
El. 547 σῆς δίχα γνώμης, Tr. 389 οὐκ ἀπὸ γνώμης.) Wecklein (Ars Soph. em. p. 21) 





or loved, or heard by me? But instead 
of the third clause, we have ἢ προσή- 
yopov | ἔτ᾽ ἔστ᾽ ἀκούειν ἡδονᾷ, ‘or what 
greeting is it longer possible for me to 
hear with pleasure?’ προσήγορον, pas- 
sive in PR. 1353, is here active, as in 
Ant. 1185 Παλλάδος θεᾶς | ὅπως ἱκοίμην 
εὐγμάτων προσήγορος. ἡδονᾷ, modal dat. 
adverbially, as ὀργῇ 405. The form 
ἡδονάν, intermediate between Attic ἡδονήν 
and Doric ἁδονάν, is given by L in £7. 
1277, where Herm. keeps it, but most 
edd. give ddovdy. If right, it was a com- 
promise peculiar to tragedy. The Dori- 
cism of scenic lyrics was not thorough- 
going: here, for instance, we have τλάμων 
(1333) yet προσήγορον (1338). 

1840 éxrémov: cp. 1411 θαλάσσιον, 
and see Appendix on v. 478. 

1841 τὸν μέγ᾽ ὀλέθριον is a certain 
correction of the MS. τὸν ὀλέθριον μέγαν 
(or μέγα), a corruption due to the omis- 
sion and subsequent marginal insertion 
of μέγα. Cp. Z/. τ. 158 ὦ μέγ᾽ ἀναιδές: 
16. 46 μέγα νήπιος: Ph. 419 μέγα | θάλ- 
λοντες. The antistrophic words are αὐτὸς 
ἔφυν τάλας (1363). ὀλέθριον, pass., ‘lost,’ 
as 77. 878 τάλαιν᾽ ὀλεθρία. τίνι τρόπῳ 
θανεῖν ope φής; The objections to the 


conject. ὄλεθρον μέγαν (metrically ad- 
missible as a dochmiac, if the second of 
ὄλεθρον is made short) are: (1) the 
awkward necessity of supplying ὄντα in 
order to defend the position of μέγαν: 
(2) the phrase ὄλεθρον, which belongs to 
the colloquial vocabulary of abuse; Dem. 
or. 18 ὃ 127 περίτριμμα ἀγορᾶς, ὄλεθρος 
γραμματεύς. 

1847 He is to be pitied alike for the 
intrinsic misery of his fate, and for his 
full apprehension (συνέσεως, schol.) of it. 
A clouded mind would suffer less. 

1848 ἀν with ἠθέλησα : ye emphasises 
μηδέ. Ocdipus had been the all-admired 
(8), the ‘saviour of the land’ (48). But 
now the Theban elders wish that they 
had never so much as heard his name or 
looked upon his face. ‘That bitter cry is 
drawn from them by the very strength of 
their sympathy: for his ruin was the re- 
sult of his coming to Thebes. The ob- 
jections to the reading of the MSS., ὥς σ᾽ 
ἠθέλησα μηδ᾽ dvayveval ποτε, are these: 
«r) Eur. Helen. 290 has the Ist aor. pass., 
ἀνεγνώσθημεν ἄν, ‘we should have been 
recognised’: but ἀναγιγνώσκειν occurs 
nowhere else in tragedy; and in Attic its 
regular sense was ‘to read,’ or in the ist 


and 
strophe. 


ἀντ. β΄. 


176 


ZOPOKAEOYS 


ΟΙ. τόλοιθ᾽ ὅστις ἢν ὃς ἀγρίας πέδας 


2 Τνομάδ᾽ ἫΝ ἐπιποδίας ἔλυσ᾽ ἀπό τε φόνου 


1350 


8 EPPUTO | κανέσωσέ μ᾽, οὐδὲν εἰς χάριν πράσσων. 


a 
4 TOTE yap αν ανὼν 


ὅ οὐκ ἣν φίλοισιν οὐδ᾽ ἐμοὶ τοσόνδ᾽ ἄχος. 


1355 


ΧΟ. 6 θέλοντι κἀμοὶ τοῦτ᾽ ἂν ἦν. 
ΟΙ. τούκουν πατρός γ᾽ ἄν φονεὺς 


8 ἦλθον, οὐδὲ νυμφίος 


9 βροτοῖς ἐκλήθην ὧν oe ἄπο. 


10 νῦν δ᾽ ἄθεος μέν εἰ 


ΣΉ. cal 


See 1360 


11 ὁμογενὴς δ᾽ ad’ ὧν αὐτὸς ἔφυν τάλας. 


ὡς σ᾽ ἠθέλησα pn dd’ ἂν γνῶναί ποτε. 


1349 ἀγρίας] am’ ἀγρίας Τ,., Triclinius 


rightly struck out ἀπ᾽, which was probably added to make the construction of the gen. 


clearer. 


Hermann preferred to omit ἦν, reading, ὄλοιθ᾽ ὅστις, ὅς μ᾽ ἀπ᾽ ἀγρίας πέδας. 


1350 νομάδοσ ἐπιποδίασ ἔλυσεν ἀπό τε φόνου | ἔρρυτο κἀνέσωσεν L. ἔλυσεν has been 
made by an early hand from ἔλαβέμ᾽ (Campbell thinks, from ἔλαβέν μ᾽), above which 


had been written Uc. 


The later Mss. have ἔλυσεν (as A), ἔλυσέ μ᾽ (E), ἔλυσ᾽ ew’ ( V4), 





aor. act., ‘to persuade.’ I have not 
found a single example of ἀναγιγνώσκω 
as= ἀναγνωρίζω (‘to recognise’) in Thuc., 
Plato, Xen., or the Orators. (2) But the 
and aor. has that sense in Homer, in 
Pindar (/sthm. 2. 23) and in Herod. 
(2. 91): may not an Attic poet have fol- 
lowed them? Granted. The sense re- 
quired here, however, after μηδέ, is to 
know, not to recognise: the latter would 
be pointless. (3) The ellipse of ἄν with 
the aor. ἠθέλησα would be strangely 
harsh. Such an ellipse with the zwzperf. 
sometimes occurs: as Antiphon or. 5 § 1 
ἐβουλόμην (and so Ar. Ran. 866), 2b. § 86 
ἠξίουν. But if, as seems clear, ἄν is 76- 
quired here, then the probability is 
strengthened that ἀναγνῶναι arose from 
ἂν γνῶναι. Between Dindorf’s ὡς ἠθέ- 
λησα μηδέ σ᾽ ἀν γνῶναι and Hermann’s 
&3 o° ἠθέλησα μηδέ y dv γνῶναι the 
question is: Which is more likely to 
have passed into the reading of the Mss.? 
Now they have ὥς σ᾽, and the loss of 
y through a confusion with the same 
letter in γνῶναι is slightly more probable 
than the double error of omitting σ᾽ be- 
fore av and inserting it after ws. 

1350 The vopddos of the Mss. is cor- 
rupt. It would require an improbable 
alteration in the strophe (see on 1330); 
and it yields no good sense. The Scholi- 
asts hesitated between rendering it (1) 


‘feeding on my flesh’! or (2) ‘in the 
pastures.’ Reading vopdd’, we have a 
dochmiac dimeter, agreeing with 1330: 

see Metrical Analysis. But the use of 
the word is extraordinary. It must mean 
ἐν νομαῖς, ‘in the pastures’—said of the 
babe whom the shepherd had been 
ordered to expose on Cithaeron. Now 
elsewhere νομάς always means ‘roaming,’ 
said (¢.g.) of pastoral tribes, or of animals : 
Tr. 271 ἵππους νομάδας ἐξιχνοσκοπῶν, 
tracking horses that had strayed: fr. 87 
νομὰς δέ τις κεροῦσσ᾽ ἀπ᾽ ὀρθίων πάγων | 
καθεῖρπεν ἔλαφος : of waters wandering 
over the land which they irrigate, O. C. 
686 κρῆναι... | Κηφισοῦ νομάδες ῥεέθρων, 
The idea of ‘wandering movement is in- 
separable from the word. To apply it 
to a babe whose feet were pinned to- 
gether would have been indeed a bold 
use. Prof. Campbell, retaining νομάδος, 
takes πέδας as acc. plur.: ‘that loosed 
the cruel clog upon my feet, when J was 
sent astray.’ But could νομάς, ‘roaming,’ 
be said of the maimed child merely in the 
sense of ‘turned adrift’ by its parents? 
The nomin. νομὰς, referring to the roving 
shepherd (πλάνης 1029) would be intel- 
ligible; but the quadruple -as is against 
it. Now cp. Aesch. Pers. 734 μονάδα 
δὲ Ξέρξην ἔρημον, ‘Xerxes alone and 
forlorn.’ Simpl ra transposing v and μ 1 
conjecture povad’, a word appropriate to 


ΟΙΔΛΙΠΌΥΣΦ ΤΥΡΑΝΝΟΣ 177 

OE. 
pastures from the cruel shackle on my feet, and saved me from 
death, and gave me back to life——a thankless deed! Had I 
died then, to my friends and to mine own soul I had not been 
so sore a grief. 

CH. I also would have had it thus. 

Or. So had I not come to shed my father’s blood, nor been 
called among men the spouse of her from whom I sprang: but 
now am I forsaken of the gods, son of a defiled mother, 
successor to his bed who gave me mine own wretched being: 


Perish the man, whoe’er he was, that freed me in the ἊΝ ae 
5 rop 16. 








or ἔλαβέ μ᾽ (). Some have ἔρρυτο, others ἔρυτο. For νομάδος Elmsley conjectured vo- 
μάδ᾽ : I suggest wovdd’. For κἀνέσωσεν Campbell has given κἀνέσωσέ μ᾽, 1855 ἄχος 
τ, ἄχθος L. Faehsi’s conjecture, ἄγος, is less suitable here. 1360 ἀθλιος MSS. : 
ἄθεος was restored by Erfurdt, and independently (in the same year, 1811) by Seidler, 
De Vers. Dochm. 59. The same emendation was afterwards made by Elmsley, and 


by Reisig (Comect. I. 191). 


1362 ὁμογενὴς MSS. : 


ὁμολεχὴς Meineke: ὁμόγαμος 





the complaint that the babe, sent to the 
lonely mountain, had not been left to 
perish in its solitude. The fact that the 
Corinthian shepherd received the child 
from the Theban is no objection: the 
child was φίλων μεμονωμένος, desolate 
and forlorn. ἔλυσ᾽, which suits the 
dochmiac as well as ἔλαβέ μ᾽, is more 
forcible here. There is a further argu- 
ment for it. The Mss. give dm’ ἀγρίας in 
1349, but the strophe (1329) shows that 
am’ must be omitted, since ᾿Απόλλων, 
φίλοι-- ὃς ἀγρίας πέδας, the first syllable 
of ἀγρίας being short, as in 1205, “11. 
344, 1124. Now πέδας (1.4. rédns) ἔλαβε, 
too’ from the fetter, would be too harsh: 
we could only do as Schneidewin did, 
and refer ἀπό back to πέδας : but though 
Δελφῶν κἀπὸ Δαυλίας (734) admits of such 
treatment, the case is dissimilar here. 
On the other hand πέδας Avo’, loosed 
jrom the fetter, is correct. Thus the 
metrical impossibility of dm confirms 
édvo’. The epithet ἀγρία, ‘cruel,’ is ap- 
plied to πέδη as it is to ὀδύνη in 777. 075. 

1351 ἔρρυτο, a strong aorist of ῥύω, 
formed as if there were a present ῥύμι: 
in //, 18. 515 ῥύατο for ῥύντο is its 3rd 
plur. Cp. 24, §. 23 ἔρυτο σάωσε δέ, where 
the aor. has a like relation to ἐρύω (the 
temporal augment being absent).—els 
χάριν: see on 1152. 

1356 θέλοντι: O. C. 1505 ποθοῦντι 
mpovpdyns: Zr. 18: Thuc. 2. 3 τῷ γὰρ 
πλήθει... οὐ βουλομένῳ ἦν.. ἀφίστασθαι: 
Tac. Agric.18 guibus bellum volentibus erat. 

1357 φονεὺς ἦλθον, have come to be 
the slayer, a compressed phrase for és 


1 S.S 


τοσοῦτον ἦλθον ὥστε φονεὺς εἶναι : cp. 
1510 and Ant. 752 ἢ κἀπαπειλῶν ὧδ᾽ 
ἐπεξέρχει θρασύς; Zr. 1157 ἐξήκεις δ᾽ ἵνα 
φανεῖ. 7. 18. 180 εἴ κέν τι νέκυς yoxup- 
μένος ἔλθῃ, come to be dishonoured (where 
some explain, ‘veach thee dishonoured’) : 
in Xen. Az. 3. 2. 3 ὅμως δὲ δεῖ ἐκ τῶν παρ- 
ὀντων ἄνδρας ἀγαθοὺς ἐλθεῖν (so the MSS.: 
τελέθειν G. Sauppe) καὶ μὴ ὑφίεσθαι, the 
clause ἐκ τῶν παρόντων helps ἐλθεῖν as= 
evadere. In 1433 ἐλθών is not similar. 
No classical use of venire seems really 
parallel: thus in Iuv. 7. 29 ut dignus 
ventas hederis, ventas=‘may come for- 
ward’ (Mayor ad /oc.). 

1359 (τούτων) dd’ dy, 2.6. ταύτης ἀφ᾽ 
qs: plur., as 1095, 1176, 1250. - 

1860 ἄθεος is a necessary correction 
of the Ms. ἄθλιος, the verse being a 
dochmiac dimeter, = 1340 ἀπάγετ᾽ ἐκτόπιον 
ὅτι τάχιστά με. νῦν answers to the short 
first syllable of ἀπάγετ᾽, since the ana- 
crusis can be either long or short: cp. 
Aesch. 7heb. 81, where αἰθερία κόνις is 
metrically parallel to νῦν δ᾽ ἄθεος μέν εἰμ’ 
here. He is ἀνοσίων (.46. dvogias) παῖς 
since through him Iocasta became such. 

1362 f. ὁμογενὴς δ᾽ ad’ dv ἔφυν-Ξ 
κοινὸν γένος ἔχων (τούτοις) ἀφ᾽ ὧν αὐτὸς 
ἔφυν: 1.6. having a common brood (one 
born of the same wife) with those (Laius) 
from whom he sprang. For the plur., 
cp. 366: for (τούτοις) ὧν, Ph. 957 παρέξω 
δαῖθ᾽ ὑφ᾽ ὧν ἐφερβόμην. ὁμογενὴς is usu. 
taken 85 -- ὁμοῦ γεννῶν, z.e. ‘engendering’ 
ὁμοῦ τῇ τεκούσῃ. But ὁμογενής is ἃ com- 
pound from ὁμο- and the stem of γένος, 
and could no more mean γεννῶν ὁμοῦ 


[2 


178 


> / , » σι , 
12 εἰ δέ Tl πρεσβύτερον €TL Κακου Κακον, 


13 τοῦτ᾽ ἔλαχ Οἰδίπους. 


ΣΟΦΘΚΛΕΟΥΣ 


1365 


ΧΟ 1 οὐκ oid” ὅπως σε φῶ βεβουλεῦσθαι καλῶς" 
15 κρείσσων γὰρ ἦσθα μηκέτ᾽ ὧν ἢ ζῶν τυφλός. 


Ol. 
μή μ᾽ ἐκδίδασκε, 
ἐγὼ γὰρ οὐκ οἷ 


ὡς μὲν τάδ᾽ οὐχ ὧδ᾽ ἔστ᾽ ἀριστ᾽ εἰργασμένα, 
μηδὲ συμβούλεν᾽ ἔτι. 
ὄμμασιν ποίοις βλέπων 


1376 


> ἃ 
πατέρα ποτ ἂν προσεῖδον εἰς “Αιδου μολών, 
οὐδ᾽ αὖ τάλαιναν μητέρ᾽, οἷν ἐμοὶ δυοῖν 
— €py ἐστὶ κρείσσον᾽ ἀγχόνης εἰργασμένα. 


ἀλλ᾽ ἡ τέκνων dnt ὄψις ἢν ἐφίμερος, 


13795 


βλαστοῦσ᾽ ὅπως ἔβλαστε, προσλεύσσειν ἐμοί; 
οὐ δῆτα τοῖς γ᾽ ἐμοῖσιν ὀφθαλμοῖς ποτε" 

οὐδ᾽ ἄστυ γ᾽, οὐδὲ πύργος, οὐδὲ δαιμόνων 
ἀγάλμαθ᾽ ἱερά, τῶν ὁ παντλήμων ἐγὼ 


κάλλιστ᾽ ἀνὴρ εἷς ἔν γε ταῖς Θήβαις τραφεὶς 


Musgrave. 


1365 ἔτι Hermann: ἔφυ MSS. 
words ἔτι κακοῦ κακόν answer metrically to ἔτι δὲ καὶ θεοῖς (1345). 


1380 


The correction is necessary, since the 
1368 ἦσθα) ἦσθ᾽ ἂν 


Porson (on 77. 114, ddv. Ρ. 174). Purgold(Ods. Crit. in Soph. etc., 1802) made the same 


conjecture, and Hartung so reads: 


but see comment. 


1376 ἔβλαστεγτ, ἔβλαστεν L. 





than συγγενής could mean γεννῶν σὺν, 
or ἐγγενής, γεννῶν ἐν. In 460 πατρὸς 
ὁμόσπορος as=omelpwy τὴν αὐτὴν ἣν ὁ 
πατήρ is different, since the second part 
of the compound adj. represents a transs 
itive verb. Meineke’s ὁμολεχὴς would 
be better than Musgrave’s ὁμόγαμος: but 
neither is needed. 

1365 πρεσβύτερον, ‘older,’ then, 
‘ranking before’; here, * more serious’ 
ΕΣ  Ἔ: 63 τὰ γὰρ τοῦ θεοῦ τρΑΒὲ τεῆς 
ἐποιεῦντο ἢ τὰ τῶν ἀνδρῶν : Thuc. 4 61 
TOUTO.. εἰπρεσβύτατον.. «κρίνας, τὸ κοινῶς φο- 

βερὸν ἅπαντας εἷ θέσθαι. 

1868 κρείσσων...ἦσθα μηκέτ᾽ dv= 
κρεῖσσον ἦν σε μηκέτ᾽ εἷναι: see on 1061. 
ἄν is omitted, as after ἔδει, εἰκὸς ἦν, etc., 

κρείσσων ἦσθα μὴ ὦν implying the thought, 
οὐκ ἂν ἧσθα, εἰ τὰ βέλτιστα ἔπασχες: see 
on 256. 

1369 ἀριστ᾽ is adverbial, the con- 
struction being οὐχ ὧδε (εἰργασμένα) ἐστὶν 
ἄριστα εἰργασμένα : that, thus done, they 
are not done best. So ἄριστα is adverb 
407, 1046, Az. 160. 

1371 βλέπων --εἰ ἔβλεπον, which is 
more forcible than to take it with ποίοις 


ὄμμασιν. Cp. Ph. 110 πῶς οὖν βλέπων 
τις ταῦτα τολμήσει λαλεῖν ; Her. 1. 37 νῦν 
τε τέοισί με χρὴ ὄμμασι ἔς τε ἀγορὴν καὶ ἐξ, 
ἀγορῆς φοιτέοντα φαίνεσθαι; [Dem.] or. 
25 ὃ 98 (the work of a later rhetorician) 
ποίοις προσώποις ἢ τίσιν ὀφθαλμοῖς πρὸς 
ἕκαστον τούτων ἀντιβλέψετε; Cp. AZ. 
462 καὶ ποῖον ὄμμα πατρὶ δηλώσω φανεὶς | 
Πελαμώνι ; 

1872 εἰς “AvSov. Blind on earth, 
Oed. will be blind in the nether world. 
Cp. Od. 12. 266 καί μοι ἔπος ἔμπεσε 
θυμῷ | μάντηος ἀλαοῦ Θηβαίου Τειρεσίαο, 
where Odysseus is thinking of the blind 
Teiresias as he had found him in Hades. 
Cp. 11. 91, where ἔγνω need not imply 
that the poet of the νέκυια conceived 
Teiresias as having sight. So Achilles 
in Hades is still szwzft-footed (11. 546). 

1878 οἷν.. δυοῖν, a dative of the per- 
sons affected, as, instead of the usual ποιῶ 
ταῦτά oe, we sometimes find ποιῶ ταῦτά 
got: cp. Zr. 808 (dpwo’): Od. 14. 289 
TpwKrns, ὃς δὴ πολλὰ κάκ᾽ “ἀνθρώποισιν 
ἐώργει. Plat. Apol. 30 A ταῦτα καὶ νεω- 
τέρῳ καὶ πρεσβυτέρῳ... ποιήσω, «καὶ ξένῳ 
καὶ ἀστῷ, μᾶλλον δὲ τοῖς ἀστοῖς. Charm. 


OIAITOYS ΤΥΡΆΝΝΟΣ 179 
and if there be yet a woe surpassing woes, it hath become the 
portion of Oedipus. 

Cu. I know not how I can say that thou hast counselled 
well: for thou wert better dead than living and blind. 


OE. Show me not at large that these things are not best 
done thus: give me counsel no more. For, had I sight, I know 
not with what eyes I could eer have looked on my father, 
when I came to the place of the dead, aye, or on my miserable 
mother, since against both I have sinned such sins as strangling 
could not punish. But deem ye that the sight of children, born 
as mine were born, was lovely for me to look upon? No, no, 
not lovely to mine eyes for ever! No, nor was this town with 
its towered walls, nor the sacred statues of the gods, since I, 
thrice wretched that I am,—I, noblest of the sons of Thebes, 


For B\acroio’ Hartung gives βλαστόντ᾽, omitting the comma after ἔβλαστε (‘that I should 
look upon offspring so born’): but see comment. 1379 ἱερὰ L; ἱρὰτ, Dindorf. The 
longer form is the regular one in L (though in O. C. 16 it has ipés). Here, as in 1428, 
the tribrach lends a certain pathos to the rhythm. Nauck unnecessarily writes ἱερά θ᾽ 





ev 


157 C οὐκ ἂν ἔχοιμεν ὅ τι ποιοῖμέν σοι. 
Xen. Hier. 7. 2 τοιαῦτα γὰρ δὴ ποιοῦσι 
τοῖς τυράννοις οἱ ἀρχόμενοι καὶ ἄλλον ὅντιν᾽ 
av ἀεὶ τιμῶντες τυγχάνωσι. Ar. Vesp. 
1350 πολλοῖς γὰρ ἤδη χὰἁτέροις αὔτ᾽ εἰρ- 
χάσω. In Xen. An. 5. 8. 24 Τούτῳ 
τἀναντία ποιήσετε ἢ τοὺς κύνας ποιοῦσι, 
there is warrant for τοῦτον : and in Isocr. 
or. 16 ὃ 49 μηδὲν ἀγαθὸν ποιήσας τῇ 
πόλει, for τὴν πόλιν. 

1974 κρείσσον᾽ ἀγχόνης, not ‘worse 
than hanging’ (such that, rather than do 
them, he would have hanged himself) : 
but ‘too bad for hanging’ (such that 
suicide by hanging would not adequately 
punish their author), Eur. App. 1217 
εἰσορῶσι δὲ [ θέαμα κρεῖσσον δεργμάτων 
ἐφαίνετο, too dreadful to be looked on: 
Aesch. Ag. 1376 ὕψος κρεῖσσον ἐκπηδήμα- 
tos, too high to be leaped over. ay xé- 
vys: cp. Eur. Alc. 229: Ar. Ach. 125 
ταῦτα δῆτ᾽ οὐκ ἀγχόνη; ‘is not this 
enough to make one hang oneself?’ 

1375 f. ἀλλ᾽ introduces (or answers) 
a supposed objection (the ὑποῴφορά of 
technical Rhetoric): Andoc. 1 § 148 τίνα 
yap καὶ ἀναβιβάσομαι δεησόμενον ὑπὲρ 
ἐμαυτοῦ; τὸν πατέρα; ἀλλὰ τέθνηκεν. 
ἀλλὰ τοὺς ἀδελφούς; ἀλλ᾽ οὐκ εἰσίν. ἀλλὰ 
τοὺς παῖδας ; ἀλλ᾽ οὔπω γεγένηνται. ---τέκ- 
νων ὄψις.. βλαστοῦσα -- ὁρώμενα τέκνα 
βλαστόντα: cp. Eur. Ale. 967 Θρήσσαις 
ἐν σανίσιν τὰς |’Opdela κατέγραψεν γῆ- 


pus, which the melodious Orpheus wrote 
down.—6tas ἔβλαστε: Eur. AZed. tort 
ἤγγειλας ol” ἤγγειλας. 

1878 πύργος, the city-wall with its 
towers and its seven gates (already famous 
in the Odyssey, 11. 263 Θήβης ἕδος ἑπτα- 
πύλοιο). Cp. Eur. Bacch. 170 Κάδμον... 
ὃς πόλιν Σιδωνίαν | λιπὼν ἐπύργωσ᾽ ἄστυ 
Θηβαῖον τόδε. Hee. 1209 πέριξ δὲ πύργος 
εἶχ᾽ ἔτι πτόλιν. 

1379 ἀγαλμαθ᾽ ἱερά, the images of 
the gods in their temples: cp. 20.—rTev | 
=av, as Ant. 1086: cp. 1427. Soph. 
has this use in many other places of 
dialogue: see Ὁ: C. 747 n. 

1380 κάλλιστ᾽ ἀνὴρ εἷς.. τραφείς. 
εἷς, in connection with a superlative, is 
strictly correct only where ove is com- 
pared with several:'as Thuc. 8. 40 οἱ 
γὰρ οἰκέται Tots Χίοις πολλοὶ ὄντες καὶ μιᾷ 
γε πόλει πλὴν Λακεδαιμονίων πλεῖστοι γε- 
νόμενοι: Eur. Heracl. 8 πλείστων μετέσχον 
εἷς ἀνὴρ Ηρακλέει. So 77. 460 πλείστας 
ἀνὴρ εἷς...ἔγημε. But here, where the 
question is of degree in nobility, it 
merely strengthens κάλλιστ᾽: cp. Thuc. 
8. 68 πλεῖστα els ἀνήρ, ὅστις ξυμβουλεύ- 
σαιτό τι, δυνάμενος ὠφελεῖν : which, not- 
withstanding πλεῖστα, is really like our 
passage, since we cannot suppose a con- 
trast with the collective wisdom of several 
advisers.—év ye ταῖς Θήβαις: the ye, by 
adding a second limitation, helps, like εἷς 


12—2. 


189 


ZOPOKAEOY2 


5 , > 3 ’ - οὗ ἘΣ , 
ἀπεστέρησ᾽ ἐμαυτόν, αὐτὸς ἐννέπων 

52 9 \ > A \ 9 A 
ὠθεῖν ἅπαντας tov ἀσεβῆ, τὸν ἐκ θεῶν 





, 5 » Ν , “Ὁ “ἢ 
φανέντ αναγνον καὶ γένους του Λαΐου. 
τοιάνδ᾽ ἐγὼ κηλῖδα μηνύσας ἐμὴν. 


ὀρθοῖς ἔμελλον ὄμμασιν τούτους ὁρᾶν ; : 
ἀλλ᾽ εἰ. τῆς ἀκουούσης ἔτ᾽ ἣν 


ἥκιστά ὟΝ 


πηγῆς δι ὦτων φραγμός, οὐκ ἂν ἐσχόμην ΠΡῸΣ 


τὸ μὴ ἀποκλῇσαι τοὐμὸν ἄθλιον 
ἵν᾽ ἢ τυφλός τε καὶ κλύων μηδέν" 


ἐμας, 
τὸ γὰρ 


τὴν φροντίδ᾽ ἔξω τῶν κακῶν οἰκεῖν γλυκύ. 1390 


ἰὼ Κιθαιρών, τί μ᾽ ἐδέχου ; 


Ti pw ov AaBav 


ἔκτεινας εὐθύς, ὡς ἔδειξα μήποτε 
ἐμαυτὸν ἀνθρώποισιν ἔνθεν ἢ “γεγώς; 
ὦ Πόλυβε καὶ Κόρινθε καὶ ve πάτρια 


λόγῳ παλαιὰ δώμαθ', οἷον dpa με 


1395 


κάλλος Kak@v ὕπουλον ἐξεθρέψατε. 


ὧν. 1888 καὶ γένους τοῦ Λαΐου] These words seem sound (see comment.), but have 


been variously amended. 


Blaydes, καὶ γένος τὸν Λαΐου (‘by birth the son of L.’): Har- 


tung, κἂν γένους τοῦ Λαΐου (‘though he be of L.’s race’): Herwerden, καὶ γένους ἀλά- 


στορα: Mekler, καὶ γένους τοὐμοῦ μύσος. 


Benedict (Obs. 271. Soph., 1820) would place 


the full stop after ἄναγνον, and take καὶ γένους τοῦ A. with κηλῖδα (‘a stain on the 


race’); and so Kennedy. 


space between syllables or letters. 


1387 ἀν εσχόμην, L, 1.6. ἀνεσχόμην, as is shown by the 
absence of accent on ἀν and of breathing on e: 


the scribe often thus leaves a small 


Most of the later Mss. have ἀνεσχόμην or ἠνεσχόμην, 





ἀνήρ. to emphasise the superlative. If 
the glories of Thebes can rejoice the sight, 
no 7heban at least had a better right to 
that joy: (and who could have a better 
right than Thebans?) 

1381 ἀπεστέρησ᾽ ἐμαυτόν : a regular 
phrase in reference to separation from 
civic life: Antiphon or. 5 ἃ 78 εἰ δ᾽ ἐν 
Αἴνῳ χωροφιλεῖ, τοῦτο οὐκ ἀποστερῶν γε 
τῶν εἰς τὴν πόλιν ἑαυτὸν οὐδενὸς (not for- 
feiting any of his relations with Athens) 
οὐδ᾽ ἑτέρας πόλεως πολίτης γεγενημένος: 
[Dem.] or. 13 § 22 οὐδενὸς ἔργων τῶν 
τότε ἀπεστέρησαν ἑαυτούς, the Athenians 
of those days did not renounce their 
share in any of the great deeds of the 
Persian Wars. 

1382 τὸν ἀσεβῆ naturally depends 
on ὠθεῖν. But, if so, it would be very 
awkward to take τὸν.. «φανέντα κ.τ.λ. 
with ἀπεστέρησ᾽ ἐμαυτόν. Rather τὸν 
φανέντα x.7.. also depends on ὠθεῖν. 
‘Bidding all to expel the impious one,— 
that man who has [szzce] been shown by 


the gods to be unholy—and of the race 
of Laius.’ His thought passes from the 
unknown person of the edict to himself, 
precisely as in 1440 f. The words καὶ 
γένους τοῦ Λαΐου are a climax, since the 
guilt of bloodshed, which the oracle had 
first denounced, was thus aggravated by 
a double horror. 

1384 κηλῖδα: 
ἐμήν, sc. οὖσαν. 

1385 ὀρθοῖς: see on 528. 

1386 τῆς ἀκουούσης...πηγῆς, the 
source (viz. the orifice of the ear) from 
which sounds flow in upon the sense: 
cp. Plat. Phaedr. 245 C ψυχή... πηγὴ καὶ. 
ἀρχὴ κινήσεως. (Not the stream of sound 
itself.) δι’ ὥτων supplements τῆς ἀκου- 
ovons πηγῆς by suggesting the channel 
through which the sounds pass from the: 
fount. Cp. fr. 773 βραδεῖα μὲν γὰρ ἐν 
λόγοισι προσβολὴ (dies δι’ ὠτὸς ἔρχεται. 
τρυπωμένου. ἡ ἀκούουσα πηγή, instead of 
ἡ πηγὴ τῆς ἀκούσεως, is said with a con- 
sciousness that πηγή means the organ of 


see on 833: μηνύσας. 


OIAITOYS ΤΥΡΆΝΝΟΣ 181 
—have doomed myself to know these no more, by mine own 
command that all should thrust away the impious one,—even 
him whom gods have shown to be unholy—and of the race of 
Laius! 

After baring such a stain upon me, was I to look with steady 
eyes on this folk? No, verily: no, were there yet a way to 
choke the fount of hearing, I had not spared to make a fast 
prison of this wretched frame, that so 1 should have known nor 
sight nor sound; for ’tis sweet that our thought should dwell 
beyond the sphere of griefs. 

Alas, Cithaeron, why hadst thou a shelter for me? When 
I was given to thee, why didst thou not slay me straightway, 
that so I might never have revealed my source to men? Ah, 
Polybus,—ah, Corinth, and thou that wast called the ancient 
house of my fathers, how seeming-fair was I your nursling, and 

what ills were festering beneath! 


but two at least (A, V) give dv ἐσχόμην. 1388 τὸ μὴ ἀποκλεῖσαι MSS.: τὸ μὴ ἀπο- 
κλῇσαι Elmsley. The original form of the verb was κληΐω (being formed from the 
noun-stem KAjft, cp. κονίω, unviw), and κλήω, not κλείω, was the older Attic form, 
still used, doubtless, in the time of Sophocles: thus κληίς occurs in an Attic inscrip- 
tion later than 403 B.c.; though κλείς, κλεῖθρον, etc., occur as early as about 378— 
330 B.C. (Meisterhans, Gramm. Att. Inschr. p. 17.) The spelling of κλείω, etc., 
fluctuates in our Mss.: thus L has κλεῖθρα above in v. 1262, but κλῆιθρα in 1287, 





hearing, just as we might have τὰ ἀκού- 
ovra ὦτα. Seneca paraphrases: utcnam 
quidem rescindere has quirem vias, Mani- 
busque adactis omne gua voces meant 
Aditusgue verbts tramite angusto patet, 
Eruere possem, gnata:...aures ingerunt, 
guicguid mihi Donastis, oculi (Oed. 
226 ff.). 

1387 ἐσχόμην, usu. in this sense with 
gen., as Od. 4. 422 σχέσθαι... βίης. 

1388 To py: cp. 1232. Forthesimple 
μή, where (as here) μὴ οὐ is admissible, 
see Az. 96: Ant. 443: Antiph. Zetral. 
3 B § 4 οὐδεὶς ἡμῖν λόγος ὑπελείπετο μὴ 
φονεῦσιν εἶναι. 

1389 ἵν ἧ. For ἦ (as 1393) see on 
1123. The negative μηδέν here shows 
how in this construction tva is essentially 
final, ‘so that I might have been’; not 
=‘in which case I should have been’— 
for which the negative must have been 
οὐδέν. So ws ἔδειξα μήποτε (1392), that 
I might never have shown. Eur. fr. 442 
φεῦ φεῦ τὸ μὴ τὰ πράγματ᾽ ἀνθρώποις 
ἔχειν | φωνήν, ἵν᾽ ἧσαν μηδὲν οἱ δεινοὶ 
λόγοι. 

1990 ἔξω τῶν κακῶν, 2.6. undisturbed 
by those sights and sounds from the 
outer world which serve to recall past 


ν 


miseries. 

1391 The imperf. ééxov helps the per- 
sonification: ‘wast ready to shelter me.’ 

1392 ὡς ἔδειξα: see on 1389, and cp. 
Aesch. P. V. 776 rl...0dK ἐν τάχει | ἔρριψ᾽ 
ἐμαυτήν.. ὅπως πέδῳ σκήψασα τῶν πάντων 
πόνων ἀπηλλάγην ; 

1994 τὰ πάτρια λόγῳ--τὰ λόγῳ πά- 
τρια, an order the less harsh since πάτρια 
(=of my fathers, not πατρῷα, of my 
father) is supplemented by παλαιά. Cp. 
At. 635 6 νοσῶν μάταν: £1. 792 τοῦ θαν- 
ὄντος ἀρτίως: Aesch. P. V. 1013 τῷ φρο- 
νοῦντι μὴ καλῶς: Eur. Med. 874 τοῖσι 
βουλεύουσιν ed. 

1396 κάλλος κακῶν ὕπουλον, a fair 
surface, with secret ills festering beneath 
it (gen. κακῶν as after words of fulness, 
= κρυπτῶν κακῶν γέμον) : because he had 
seemed most prosperous (775), while the 
doom decreed from his birth was secretly 
maturing itself with his growth.—KdAXos, 
concrete, a fair object, Xen. Cy7. 5. 2.7 
τὴν θυγατέρα, δεινόν τι κάλλος καὶ μέγε- 
θος, πενθικῶς δ᾽ ἔχουσαν.---ὅπουλον, οὗ ἃ 
sore festering beneath an οὐλή or scar 
which looks as if the wound had healed: 
Plat. Gorg. 480 Β ὅπως μὴ ἐγχρονισθὲν τὸ 
νόσημα τῆς ἀδικίας ὕπουλον τὴν ψυχὴν 


ΣΟΦΌΚΛΕΟΥΣ 


νῦν γὰρ κακός τ᾽ ὧν κἀκ κακῶν εὑρίσκομαι. 





ὦ τρεῖς κέλευθοι καὶ κεκρυμμένη νάπη 


δρυμός τι καὶ στενωπὸς ἐν τριπλαῖς ὁδοῖς, 


αἵ τοὐμὸν αἷμα τῶν ἐμῶν χειρῶν aro 


1400 


ἐπίετε πατρός, dpd μου μέμνησθέ τι, 
οἵ ἔργα δράσας ὑμὶν εἶτα δεῦρ᾽ ἰὼν 
ὁποῦ ἔπρασσον αὖθις; ὦ γάμοι γάμοι, 


ἐφύσαθ' ἡμάς, 


ἀνεῖτε * 


τεύσαντες πάλιν 


κα 
ταύτου ΡΝ κἀπεδείξατε 


1405 


πατέρας, ἀδελφούς, παῖδας, αἷμ᾽ «ἐμφύλιον, 
νύμφας γυναῖκας μητέρας τε, χωπόσα 


αἴσχιστ᾽ ἐν ἀνθρώποισιν ἔργα γί 
—= ἀλλ᾽ οὐ “γὰρ αὐδᾶν ἔσθ᾽ ἃ μηδὲ 
ὅπως τάχιστα πρὸς θεών ἔξω μέ που 





εται. 
Ze 
pav καλόν, 
1410 


καλύψατ᾽, ἢ φονεύσατ᾽, ἢ θαλάσσιον 
3 , 2 » ,, 3 3 ’ > Ὁ 
ἐκρίψατ', ἔνθα μήποτ᾽ εἰσόψεσθ᾽ ἔτι. 
UT , ἀξιώσατ᾽ ἀνδρὸς ἀθλίου θιγεῖν' 


πίθεσθε, 


μὴ δείσητε" «τἀμὰ γὰρ κακὰ 





οὐδεὶς οἷός τε πλὴν ἐμοῦ φέρειν βροτῶν. 


1415 





1204. 1401 dpd μου MSS. 


: ap’ ἐμοῦ Brunck, Erfurdt: dpa μὴ Blaydes. 


Linwood 


suggested apd μοι. — μέμνησθ᾽ ὅτι L, with most of the later Mss. (including A); but 


a few have μέμνησθ᾽ ἔτι: 
ταὐτοῦ. Nauck, τοὐμόν. 
all edd. receive. 


μέμνησθε τι Elmsley. 
1414 πείθεσθε MSS.: πίθεσθε Elmsley, which almost 
The pres.=‘be persuaded’: the aor.=‘obey,’ ‘comply with my 


1405 ταὐτὸν Mss. I read 





ποιήσει καὶ ἀνίατον, ‘lest the disease of 
injustice become chronic, and render his 
soul gangrenous and past cure’ (Thomp- 
son). Thuc. 8. 64 ὕπουλον αὐτονομίαν, 
unsound independence opp. to τὴν ἄντι- 
κρυς ἐλευθερίαν. Dem. or. 18 § 307 ἧσυ- 
χίαν ἄγειν ἄδικον καὶ ὕπουλον, unjust and 
insecure peace. Eustath. Od. 1496. 35 
Σοφοκλῆς...λέγεται... ὕπουλον εἰπεῖν τὸν 
δούρειον ἵππον, the wooden horse at Troy, 
as concealing foes. 

1397 κἀκ κακῶν like ἀνοσίων παῖς 
(1360), with reference to the stain in- 
curred by Iocasta. 

1398 f. His memory recalls the 
scene as if he were again approaching 
it on his way from Delphi. First, he 
descries three roads converging in a deep 
glen or ravine (τρεῖς κέλευθοι---κεκρυμμένη 
νάπη): then, descending, he comes to ἃ 
coppice (δρυμός) at a point where his 
own ae narrows (στενωπός) just before 
its junction with the two others (ἐν τρι- 


mais ὁδοῖς). See on 733. The genu- 
ineness of v. 1399 has been groundlessly 
questioned, on the score of supposed tau- 
tology. The language may be compared 
with that of the verses from the Oedipus 
of Aeschylus (fr. 167), quoted in the In- 
troduction. 

1400 τοὐμὸν αἷμα, thus divided from 
πατρός, is more than αἷμα τοὐμοῦ πατρός: 
‘the same blood which flows in my own 
veins—the blood of my father.’ 

1401 For tt, which has a tone of 
bitterness here, see on 124, 969. The ὅτι 
of the Mss. must be explained in one of 
two ways:—(1) as if the construction 
was irregularly changed by ola, ὁποῖα : 
but the immediate succession of οἷα to 
ὅτι makes this intolerably harsh: or (2) 
as if ola, ὁποῖα were exclamatory substi- 
tutes for δεινά or the like: which seems 
inadmissible. 

1405 ἀνεῖτε ταὐτοῦ σπέρμα. By the 
change of one letter, we restore sense to 


OlAITIOYS. ΤΥΡΆΝΝΟΣ 183 
For now I am found evil, and of evil birth. O ye three roads, 
and thou secret glen,—thou coppice, and narrow way where 
three paths met—ye who drank from my hands pea 
blood which was mine own,—remember ye, pérchance, what 
deeds I wrought for you to see,—and then, when I came hither, 
what fresh deeds I went on to do? 

O marriage-rites, ye gave me birth, and when ye had brought 
me forth, again ye bore children to your child, ye created an 
incestuous kinship of fathers, brothers, sons,—brides, wives, 
mothers,—yea, all the foulest shame that is wrought among 
men! Nay, but ’tis unmeet to name what ’tis unmeet to do:— 
haste ye, for the gods’ love, hide me somewhere beyond the 
land, or slay me, or cast me into the sea, where ye shall never 
behold me more! Approach,—deign to lay your hands on a 
wretched man;—hearken, fear not—my plague can rest on no 
mortal beside. 


wish.’ In £/. 1015 and O. C. 520 πείθου is fitting, as in Plat. Crito 44 Β ἔτι καὶ viv 
ἐμοὶ πείθου καὶ σώθητι: on the other hand, in 77. 1227 πιθοῦ is best; and in Aesch. 
P. V. 276 πείθεσθε (d¢s) seems rightly changed to πίθεσθε by Blomfield. Here, as 
in most cases, either pres. or aor. is admissible; but the aor. seems clearly prefer- 





The ταὐτὸν of the MSS. is 
unintelligible. Oedipus was the σπέρμα 
of Laius and Iocasta. When Iocasta weds 
Oedipus, the marriage cannot be said 
ἀνιέναι ταὐτὸν σπέρμα: for it is absurd to 
suppose that the seed sown by Oedipus 


the passage. 


could be identified with Oedipus himself. 


But the marriage can be rightly said 
ἀνιέναι ταὐτοῦ σπέρμα, to yield seed from 
the same man (Oedipus) whom that womb 
had borne. 

1405 ff. The marriage of Iocasta 
with Oedipus constituted (ἀπεδείξατε) 
Oedipus at once father and brother (of 
his children), while he was also soz (of 
his wife),...the closest relation in blood 
(αἷμ᾽ ἐμφύλιον) becoming also the hus- 
band. The marriage made Iocasta the 
bride (vipdas)...aye, and the child-bear- 
ing wife (yvvaikas),—of him to whom 
she was also mother (pytépas). Thus, 
through the birth of children from such a 
marriage, complex horrors of relationship 
arose (ὁπόσα αἴσχιστα ἔργα γίγνεται). 
αἷμ᾽ ἐμφύλιον is in apposition with πα- 
τέρας ἀδελφοὺς tatdas,—‘a blood-kin- 
ship’ standing for ‘a blood-kinsman.’ 
It expresses that the monstrous union 
confounded the closest tie of consan- 
guinity with the closest tie of affinity. 
The phrase ἐμφύλιον αἷμα, like συγγενὲς 
αἷμα, would in Tragedy more often mean 


‘murder of a kinsman.’ But it can, of 
course, mean also ‘kindred blood’ in 
another sense; and here the context 
leaves no ambiguity. Cp. O. C. 1671 (n.) 
ἔμφυτον αἷμα, Eur. Phoen. 246 κοινὸν 
αἷμα, κοινὰ τέκεα | τῆς κερασφόρου πέ:- 
φυκεν ᾿Ιοῦς. 

1410 ff. ἔξω μέ που | καλύψατ᾽ : the 
blind man asks that they will- lead him 
away from Thebes, and Azde him from 
the sight of men in some lonely spot—as 
amid the wilds of Cithaeron (1451). We 
must not transpose καλύψατ᾽ and ἐκρί- 
at’, as is done in Schneidewin’s ed. (as 
revised by Nauck), after Burges. 

1411 f£. θαλάσσιον : cp. Appendix, 
note on v. 478. Cp. Ὁ; CG. 119 n.— 
ἔνθα μή with fut. indic., as dz. 659, Εἰ. 
380, 77. 800. 

1415 No one can share the burden of 
his ills. Other men need not fear to be 
polluted by contact with him, as with 
one guilty of blood. His wunwitting 
crimes and his awful sufferings—alike 
the work of Apollo—place him apart. 
In illustration of the fear which he seeks 
to allay, compare the plea of Orestes that, 
since he has been duly purified from 
bloodshed, contact with him has ceased 
to be dangerous (Aesch. Zum. 285 ὅσοις 
προσῆλθον ἀβλαβεῖ evvovoig).—Contrast 
O. Ο. 1132 ff., where Oed. will not allow 


xO): 


ZOPOKAEOY2 


ἀλλ᾽ ὧν ἐπαιτεῖς ἐς δέον πάρεσθ' ὅδε 


Κρέων τὸ πράσσειν καὶ τὸ (βουλεύειν, ἐπεὶ 
χώρας λέλειπται μοῦνος ἀντὶ σοῦ φύλαξ. 


οἴμοι, τί δῆτα λέξομεν πρὸς τόνδ᾽ ἔπος; 
τίς μοι φανεῖται. πίστις ἔνδικος ; τὰ γὰρ 


πάρος πρὸς αὐτὸν πάντ᾽ ἐφεύρημαι κακός. 


οὐχ ὡς γελαστής, Οἰδίπους, ἐλήλυθα, 


οὐδ᾽ ὡς ὀνειδιῶν τι τῶν πάρος κακῶν. 
ἀλλ᾽ εἰ τὰ θνητῶν μὴ καταισχύνεσθ᾽ ἔτι 


γένεθλα, τὴν γοῦν πάντα βόσκουσαν φλόγα 


1425 


αἰδεῖσ θ᾽ ἄνακτος Ἡλίου, τοιόνδ᾽ ἄγος 
ἀκάλυπτον οὕτω δεικνύναι, τὸ μήτε yn 


μήτ᾽ ὄμβρος ἱ ἱερὸς μήτε ae προσδέξεται. 
ἀλλ᾽ ὡς τάχιστ᾽ ἐς οἶκον ἐσκομίζετε' 


τοῖς ἐν γένει γὰρ 


τἀγγενῆ μάλισθ᾽ ὁρᾶν 


1430 


4 ) 3 id 3 » ͵7 
μόνοις T ἀκούειν evoeBas ἔχει κακά. 


able. 
οὐχ in the margin. 


1422 οὐχ ws] L has οὐ, with a letter erased after it: 


a later hand has written 


The erased letter was probably @ (or 7’), as in the next verse the 


1st hand wrote οὔθ᾽, which a later changed to οὐδ᾽ (A’s reading), while another wrote 


a second ovx in the margin. ovx.. 


.ovd’ seems better here, because simpler, than the 





his benefactor Theseus to touch him. 
There, he feels that he is still formally 
ἄναγνος, and that gratitude forbids him 
to impart a possible taint. Here, he 
thinks only of his unique doom and his 
incommunicable anguish. 

1416 f. ὧν ἐπαιτεῖς és S5¢0ov=season- 
ably in respect of those things which (ὧν = 
τούτων ἅ) youask. For the gen. of rela- 
tion cp. Xen, 27. 6. 2. 9 κεῖσθαι τὴν Kép- 
κυραν ἐν καλῷ μὲν τοῦ Κορινθιακοῦ κόλπου 
καὶ τῶν πόλεων αἱ ἐπὶ τοῦτον καθήκουσιν 
{ conveniently in respect to’), ἐν καλῴ δὲ 
τοῦ τὴν Λακωνικὴν χώραν βλάπτει .---τὸ 
πράσσειν καὶ τὸ βουλεύειν are strictly 
accusatives of respect, ‘as to the doing 
and the planning,’ 2.6. with a view to 
doing and planning. So Azz. 79, El. 1030, 
ΟἿ, 442, PA, 0253, ete; 

1418 μοῦνος: see on 304. Kiihlstadt 
(De Dial. Trag. 104) thinks that Soph. 
never uses μοῦνος for μόνος unless with 
some special emphasis: but, as Ellendt 
remarks, such instances as O. C. 875, 
991, Ant. 705, fr. 434 refute that view. 
Rather it was a simple question of metri- 
cal convenience. The same is true of 
ξεῖνος and ξένος, with this exception, that, 


even where metre admitted ξέν᾽, ἕεῖν" 
occurs as the frst word of an address: 
Eur. 7. 7. 798 ξεῖν᾽, οὐ δικαίως. In OVC. 
928 also, L and A give ξεῖνον map’ ἀστοῖς. 
1420 τίς μοι φανεῖται πίστις ἔνδικος; 
‘what reasonable claim to confidence can 
be produced on my part?’ Oedipus had 
brought a charge against Creon which 
was false, and had repudiated a charge 
against himself which was true. He 
means:—‘How can I expect Creon to 
believe me now, when I represent myself 
as the blind victim of fate,—when I crave 
his sympathy and pity?’ πίστις has two 
main senses, each of which has several 
shades,—(1) faith, and (2) a warrant for " 
faith. Were it is (2) essentially as in 
0. C. 1632 δός μοι χερὸς σῆς πίστιν. Not 
‘a persuasive argument’ in the technical 
sense of Rhetoric, for which πίστεις were 
‘instruments of persuasion,’ whether ἔν- 
Texvot, provided by the Art itself (λογική, 
παθητική, ἠθική), or ἄτεχνοι, external to 
the art, as depositions, documents, etc. 
1421 πάντ᾽ : see on 475. 
1422 Cp. the words of Tennyson’s 
Arthur to Guinevere: ‘Yet think not that 
I come to urge thy crimes.’ 


OIAITOYS TYPANNO2 185 
CH. Nay, here is Creon, in meet season for thy requests, 
crave they act or counsel; for he alone is left to guard the land 
in thy stead. 
Or. Ah me, how indeed shall I accost him? What claim 
to credence can be shown on my part? For in the past I have 
been found wholly false to him. 


CREON. 


I have not come in mockery, Oedipus, nor to reproach thee 
with any bygone fault.—(7Zo the Attendants.) But ye, if ye 
respect the children of men no more, revere at least the all- 
nurturing flame of our lord the Sun,—spare to show thus 
nakedly a pollution such as this,—one which neither earth can 
welcome, nor the holy rain, nor the light. Nay, take him into 
the house as quickly as ye may; for it best accords with piety 


that kinsfolk alone should see and hear a kinsman’s woes. 


more rhetorical οὔθ᾽... ot@’. 


on 1379. 1480 μάλισθ᾽ ὁρᾶν MSS. 


1424. 1491] ἀλλ᾽ εἰ τὰ θνητῶν... .ἔχει κακά. 
transposition of these eight verses, see comment. 
Dobree conjectures μόνοις ὁρᾶν (and so Blaydes, 


On Nauck’s 
1428 ἱερὸς] ipds Dindorf. See 





1424—1431 Nauck gives these verses 
to Oedipus, making them follow 1415. 
He regards τοιόνδ᾽ ἄγος x.7.\. as incon- 
sistent with the profession which Creon 
has just made. Rather may we consider 
them as showing a kinsman’s anxious and 
delicate concern for the honour of Oedipus 
and of the house (1430). Creon, deeply 
moved, deprecates the prolonged indul- 
gence of a painful curiosity (cp. 1304). 
It is again Creon who says ἴθι στέγης ἔσω 
(1515) when Oedipus would fain linger. 
Clearly, then, these verses are rightly 
placed in the Mss. 

1425 βόσκουσαν boldly for τρέφουσαν: 
cp. Aesch. 4g. 633, where the sun is τοῦ 
τρέφοντος...χθονὸς φύσιν. 

1427 f. δεικνύναι depends on αἰδεῖσθε, 
for the constr. of which with (1) acc. of 
persons revered, and (2) infin. of act which 
such reverence forbids, cp. Xen. 4x. 2. 
3. 22 ἠσχύνθημεν καὶ θεοὺς καὶ ἀνθρώπους 
προδοῦναι αὐτόν, ‘respect for gods and for 
men forbade us to betray him.’—7é (=8, 
see on 1379) μήτε, not οὔτε, since τοιόνδ᾽ 
ἄγος indicates a class of ἄγη : not merely 
‘which, but ‘such as, ‘earth will not 
welcome’ (guod Terra non admissura sit) : 
cp. 817, Zl. 654 ὅσων ἐμοὶ | δύσνοια μὴ 
πρόσεστιν. γῆ--ὄμβρος-- φῶς. The pol- 
lution (ἄγος) of Oedipus is such that the 
pure elemental powers—represented by 
earth, the rain from heaven, the “ghti— 


cannot suffer it to remain in their pre- 
sence (προσδέξεται): it must be hidden 
from them. Cp. Aesch. Zum. go4 f., 
where the Erinyes, as Chthonian powers, 
invoke blessings on Attica, γῆθεν---κ τε 
ποντίας δρόσου---ἐξ οὐρανοῦ τε. ὄμβρος 
here is not a synonym but a symbol of 
water generally, as with Empedocles 282 
ὡς τότ᾽ ἔπειτ᾽ ἐδίηνε Kimpis χθόνα δηρὸν ἐν 
ὄμβρῳ | εἴδεα καὶ ποιοῦσα θοῷ πυρὶ δῶκε 
κρατῦναι: cp. Lucr. 1. 714 f. guattuor ex 
rebus posse omnia rentur Ex igni terra 
atque anima procrescere etimbri. In Ant. 
1073 the exposure of the unburied corpse 
is spoken of as a violence to ol ἄνω θεοί 
(βιάζονται). It was a common form of 
oath to pray that, if a man swore falsely, 
neither earth, nor sea, nor air, might 
tolerate the presence of his corpse (Eur. 
Or. 1085, Hipp. 1030). 

1428 The original sense of ἱερός, 
‘strong’ (Curt. Ztym. 8 614), suits a few 
phrases, such as ἱερὸς ἰχθύς (//. 16. 407). 
But in such as ἱερὸν ἦμαρ, κνέφας, ὄμ- 
Bpos, ποταμοί etc. it is more likely that 
the poet had no consciousness of any 
other sense than ‘sacred.’ 

1430 The objection to taking μάλιστα 
with τοῖς ἐν γένει is not that it follows 
these words (see on 1394), but that τἀγ- 
γενῆ intervenes. Rather join it with 
εὐσεβῶς ἔχει. ὁρᾶν μόνοις τ᾽ ἀκούειν = μό- 
νοις ὁρᾶν ἀκούειν τε. 


186 


ΣΟΦΟΚΛΕΟΥΣ 


ΟΙ. πρὸς θεών, ἐπείπερ ἐλπίδος μ᾽ ἀπέσπασας, 
ἄριστος, day πρὸς κάκιστον avop ἐμέ, 
πιθοῦ τί μοι" πρὸς σοῦ yap, οὐδ᾽ ἐμοῦ, φράσω. 


KP: 


Kal TOU με χρείας ὧδε λιπαρεῖς τυχεῖν ; 
ΟΙ. ῥῖψόν με γῆς ἐκ τῆσδ᾽ ὅσον τάχισθ', 


1435 


ὅπου 


θνητῶν φανοῦμαι μηδενὸς προσήγορος. 


ΚΡ, ἔδρασ᾽ ἂν εὖ TOUT lo 


ἄν, εἰ μὴ τοῦ 


εοῦ 


πρώτιστ᾽ ἔχρῃζον ἐκμαθεῖν τί πρακτέον. 


OI. ἀλλ᾽ ἡ γ᾽ ἐκείνου πᾶσ᾽ ἐδηλώθη φάτις, 


1440 


Ν / \ > la > > , 
τὸν πατροφόντην, τὸν ἀσεβῆ μ᾽ ἀπολλύναι. 
Ψ 


KP. οὕτως ἐλέχθη ταῦθ᾽ 


3. δ᾽. 8. ὅν 
ὅμως δ᾽, ἱν ἕσταμεν 


χρείας, “ἄμεινον ἐκμαθεῖν τί δραστέον. 
Ol. οὕτως ap ἀνδρὸς ἀθλίου πεύσεσθ' ὕπερ; 


ΚΡ. 


καὶ γὰρ σὺ νῦν τἂν τῷ θεῷ πίστιν φέροις. 


[445 


OI. καὶ σοί γ᾽ ἐπισκήπτω τε καὶ προστρέψομαι, 
τῆς μὲν κατ᾽ οἴκους αὐτὸς ὃν θέλεις τάφον 
θοῦ: καὶ γὰρ ὀρθῶς τῶν γε σών τελεῖς ὕπερ' 
ἐμοῦ δὲ μήποτ᾽ ἀξιωθήτω τόδε 


πατρῷον ἄστυ ζώντος οἰκητοῦ τυχεῖν, 


1450 


ἀλλ᾽ ἔα με ναίειν ὄρεσιν, ἔνθα κλήζεται 


with μόνοις δ᾽ in 1431): 
Meineke, which Nauck adopts. 


Meineke, μόνοις θ᾽ ὁρᾶν. 


1437 φανοῦμαι] θανοῦμαι 


1445 7’ ἂν L (1.6. τοι av, τὰν), with most of the 





1432 ἐλπίδος μ᾽ ἀπέσπασας, suddenly 
plucked me away from (made me to aban- 
don) my uneasy foreboding: cp. Lat. 
revellere (falsorum persuasionem, Sen. 
Epist. 95)» and our phrase, " a revulsion 
of feeling’: Az. 1382 ws μ᾽ épevoas ἐλ- 
πίδος πολύ. Conversely (22. 809) ἀπο- 
σπάσας... φρενὸς | αἵ μοι μόναι παρῆσαν 
ἐλπίδων. 

1488 ἄριστος ἐλθὼν πρὸς... ἐμέ, having 
come to me in so noble ἃ spirit; cp. 1422 
ἐλήλυθα. This is more natural than to 
render, ‘having , proved thyself most 
noble towards me’ (see on 1357). 

1434 πρὸς σοῦ, in thy interest: Eur. 
Alc, 58 πρὸς τῶν ἐχόντων, Φοῖβε, τὸν νόμον 
τίθης: Zr. 479 δεῖ γὰρ καὶ τὸ πρὸς κείνου 
λέγειν, the argument on his side. 

1435 χρείας, request: O. (. 1754 
προσπίτνομέν σοι. ΘΗ. τίνος, ὦ παῖδες, 
χρείας ἀνύσαι; 

1437 μηδενὸς προσήγορος, accosted 
by no one: for the gen., cp. El. 1214 
οὕτως Grids εἶμι τοῦ τεθνηκότος ; 20. 344 


κείνης διδακτά. With dat. PA. 1353 τῷ 
προσήγορος; see on 1337: for ὅπου μή 
with fut. indic., on 1412. 

14388 For the double dv, Cp. 139. 
τοῦτ᾽ depends on ἴσθι, not édpaca. 

1440 φάτις (151), the message brought 
by Creon from Delphi (86); πᾶσ᾽, ‘in 
full,’ explicitly: Az. 275 κεῖνος... «λύπῃ 
πᾶς ἐλήλαται. The indefinite person of 
the φάτις is identified with Oedipus just 
as in 1382 f. 

1441 ἀπολλύναι could refer either to 
misery in exile (1436), or to death: cp. 
100. Ph. 252 διωλλύμην. 

1442 1. ἵνα.. χρείας, see 367. 

1444 οὕτως with ἀθλίου: Ph. 104 
οὕτως ἔχει τι δεινὸν ἰσχύος θράσος ; ; 

1445 The καὶ belongs to σύ: ‘even 
thou’ who didst not believe ‘Teiresias. 
This is not spoken in mockery, but with 
grave sorrow. The phrase πίστιν φέροις 
as Ξε πιστεύοις (22. 735 τῷ τέλει πίστιν 
φέρων) prob. = ‘vender belief’ (as ἃ tribute 
due), cp. φόρον, δασμόν, χρήματα φέρειν, 


OIAITTOYS ΤΥΡΆΝΝΟΣ 187 

ΟΕ. For the gods’ love—since thou hast done a gentle 
violence to my presage, who hast come in a spirit so noble to 
me, a man most vile—grant me a boon:—for thy good I will 
speak, not for mine own. 

Cr. And what wish art thou so fain to have of me? 

OE. Cast me out of this land with all speed, to a place 
where no mortal shall be found to greet me more. 

Cr. This would I have done, be thou sure, but that I craved 
first to learn all my duty from the god. 

OE. Nay, his behest hath been set forth in full,—to let me 
perish, the parricide, the unholy one, that I am. 

Cr. Such was the purport; yet, seeing to what a pass we 
have come, ’tis better to learn clearly what should be done. 

Or. Will ye, then, seek a response on behalf of such a 
wretch as I am? 

Cr. Aye, for thou thyself wilt now surely put faith in the 

od. 

᾿ On. Yeas and “on thee day 1 Τα σα, to tice «wall 
I make this entreaty:—give to her who is within such 
burial as thou thyself wouldest; for thou wilt meetly render 
the last rites to thine own. But for me—never let this city 
of my sire be condemned to have me dwelling therein, while 
I live: no, suffer me to abide on the hills, where yonder is 


later Mss.; L? and [' have γ᾽ ὧν, which some edd. prefer. 


while ye here would be almost derisive. 


But ro: has a pensive tone, 
1446 προστρέψομαι L: mporpépoua r, 





and the like figure in Pind. Οὐ. 11. 17 
νικῶν | Ἴλᾳ φερέτω χάριν. 

1446 καὶ σοί γ᾽ : yes [I am prepared 
to abide by Apollo’s word], and on ¢hee 
too I lay an injunction, and I will now 
make a prayer to thee; 2.5. as I turn to 
the god for what he alone can give (cp. 
1519 τοῦ θεοῦ μ᾽ αἰτεῖς δόσιν), so I turn 
to ¢hee for that which lies in thine own 
power. The midd. προστρέψομαι as in 
fr. 759 "Epydvny (Athene)...tpoorpémecde: 
the active has the same sense in Az. 831, 
O. C. 50. On the future, see 1077. 
There is no cause to desire ἐπισκήψω: 
each tense has its due force: I now en- 
join, and am going on to ask. Just so in 
Thuc. 2. 44 οὐκ ὀλοφύρομαι μᾶλλον ἢ Tapa- 
μυθήσομαι, where the conjecture é\ogv- 
podua is needless: ‘I do zot bewail them, 
but rather zz¢end ¢o comfort them.’ The 
reading προτρέψομαι must be judged by 
the context. With it, the sense is:— 
yes [7 am sensible of my duty to 
Apollo], and I enjoin on ¢hee, and will 
exhort thee, to do thine. (Cp. 358 mpov- 


tpéyw; Plat. Legg. 711 Β πρὸς ἀρετῆς 
ἐπιτηδεύματα προτρέπεσθαι τοὺς πολίτας.) 
But this strain of lofty admonition seems 
little in accord with the tone of the 
broken man who has just acknowledged 
Creon’s unexpected goodness (1432), and 
is now a suppliant (cp. 1468). In 472. 
831 and O. C. 50, where προστρέπω is 
undoubtedly right, προτρέπω occurs as a 
variant. 

1447 τῆς... κατ᾽ οἴκους: the ame of 
Iocasta has not been uttered since 1235. 
Contrast 950. 

1448 τελεῖς absol., like ἔρδειν, per- 
form rites, 7.5. the ἐντάφια (Isae. or. 8 
§ 38). The special term for offerings to 
the dead was ἐναγίζειν (Isae. or. 3 § 46). 

1449 ἀξιωθήτω, be condemned: Her. 
3-145 ἐμὲ μέν, ὦ κάκιστε ἀνδρῶν, ...ἀδική- 
σαντα οὐδὲν ἄξιον δεσμοῦ γοργύρης ἠξίω- 
σας, doomed me to a dungeon though I 
had done no wrong worthy of bonds. 

1451 ἔα, a monosyllable by synizesis, 
and in Anz. 95 ἀλλ᾽ ἔα με. Cp. Od. 9. 283 
νέα μέν μοι κατέαξε Ποσειδάων ἐνοσίχθων. 


188 


ZOPOKAEOYS 


«ε Ν ΄ \ @ a ld / 
οὑμὸς Κιθαιρὼν οὗτος, ὃν μήτηρ τέ μοι 
πατήρ T ἐθέσθην Covre κύριον τάφον, 
w εξ ἐκείνων, οἵ μ ἀπωλλύτην, θάνω. 


καίτοι τοσοῦτόν γ᾽ οἶδα, μήτε μ᾽ ἂν νόσον 


1455 


μήτ᾽ ἄλλο πέρσαι μηδέν: od yap av ποτε 


’ 3 va \ 5 ᾿ς A ~ 
θνήσκων ἐσώθην, μὴ πί τῷ δεινῷ, κακῷ. 
ἀλλ᾽ ἡ μὲν ἡμῶν “μοῖρ᾽, ὅποιπερ elo, 


¥” 
Τὼ" 


παίδων δὲ τῶν μὲν ἀρσένων μή μοι, Κρέον, 


προσθῇ μέριμναν" ἄν pes. εἰσίν, ὥστε μὴ 


1460 


σπάνιν ποτὲ σχεῖν, ev? ἂν ὦσι, τοῦ βίου' 
τοῖν δ᾽ ἀθλίαιν οἰκτραῖν τε rapbevow ἐμαῖν, 


οἷν οὐποθ᾽ ἡμὴ χωρὶς ἐστάθη βορᾶς 


which some edd. receive: but see comment. 


1453 ζῶντε MSS.: ζῶντι Toup. 


1458 ὅποιπερ L: ὅπηπερ r, which Brunck and others prefer; but Oed. is thinking 
rather of the end to which his destiny may go than of the course by which the end is 


to be reached. 1459 κρέων L: 


κρέον Yr. 


Cp. on 637. 1460 πρόσθῃ (sic) L, 





--ὄρεσιν, locative dative, cp. γῇ, 1266. 

—tv0a κλήζεται «.7.A., lit., ‘where my 
Giteeton yonder is famed,’ = “where yon- 
der is Cithaeron, famed as mine,’—2.e. 
made famous by the recent discovery that 
it is Οἰδίπου τροφὸς καὶ μήτηρ (1092). 
There is an intense bitterness in the 
words; the name of Cithaeron is for ever 
to be linked with his dark story. Statius 
(quoted by Schneidewin) was doubtless 
thinking of this place: habeant te lustra 
tuusque Cithaeron (Theb. 11. 752). κλῇ- 
ἵεται is stronger than καλεῖται, as in 77. 
659 ἔνθα κλύήζεται θυτήρ means, ‘where 
Jame (that brought the tidings of his great 
victory) tells of him as sacrificing.’ For 
the idiom cp. //. 11. 757 ᾿Αλεισίου ἔνθα 
κολώνη | κέκληται. 

1453 The words ἐξ ἐκείνων form the 
decisive argument for the ζῶντε of the 
MSS. against Toup’s specious emendation, 
ζῶντι. His parents in ¢hetr life-time ap- 
pointed Cithaeron to be his grave. Now 
they are dead; but, though he can no 
longer die by their agency, he wishes to 
die ἐξ ἐκείνων, by their doom; 1.6. by self- 
exposure in the same wilds to which they 
had consigned him (cp. 719 ἔρριψεν ἄλλων 
χερσὶν els ἄβατον épos). The thought of 
the dead bringing death upon the living 
is one which Sophocles has also in 41. 
1026 εἶδες ws χρόνῳ | ἔμελλέ σ᾽ “Exrwp καὶ 
θανὼν ἀποφθιεῖν; 77. 1163 (Heracles 
speaking of Nessus) ζῶντά μ᾽ ἔκτεινεν 
θανών: Ant. 871. The reading ζῶντι, 


on the other hand, yields nothing but a 
weak verbal antithesis with τάφον. Had 
his parents meant him to /zve in lonely 
misery on Cithaeron, there would be some 
point in calling it his ‘living grave.’ But 
they meant him to die there forthwith 
(cp. 1174); ζώντι, then, would mean no- 
thing more than that the grave was chosen 
before the babe was dead.—kvptov, ap- 
pointed by their authoritative decision : 

cp. Aesch. Zum. 541 ποινὰ yap éréorat'| 
κύριον μένει τέλος. 

1454 ἀπωλλύτην : for the imperf. of 
intention, cp. Andoc. or. 1 § 41 τὸν πατέρα 
μου ἀπώλλυε (‘sought to ruin’), συνειδότα 
ἀποφαίνων. 

1455 οἶδα μὴ dv πέρσαι-- “1 am con- 
fident that nothing can destroy me.’ μή 
is admissible since olda here = πέποιθα, 
and μὴ av πέρσαι represents a negative 
conception of the mind. So with partic. 
Ὁ: Ὁ: 656 οἶδ᾽ ἐγώ σε μή τινα ] ἐνθένδ᾽ 
ἀπάξοντ᾽. οἷδα οὐκ ἂν πέρσαι would be 
more usual; the difference being that this 
would be the oblique form of οἶδα ὅτι οὐκ 
dv πέρσειε. The ordinary usage is (1) οὐ 
with infin. (Ξε ὅτε with indic.) after verbs 
of saying or thinking, λέγω, φημί, οἴομαι, 
etc.; (2) μή with infin. after verbs of feel- 
ing confident, promising, etc., as πιστεύω, 
πέποιθα, ὑπισχνοῦμαι, ὄμνυμι. Cp. Ph. 
1329. Buta fewexceptions occur both ways, 
when a verb of either class is virtually equi- 
valent to a verbof the other : ¢.g. (1) [Dem. J 
or. 29 § 48 οἴεσθε οὐκ dy αὐτὴν λαβεῖν (= ὅτι 


OPO tre ΤΥΡΆΝΝΟΣ 189 
Cithaeron, famed as mine,—which my mother and sire, while 
they lived, set for my appointed tomb,—that so I may die by 
their decree who sought to slay me. Howbeit of thus much am 
I sure,—that neither sickness nor aught else can destroy me; 
for never had I been snatched from death, but in reserve for 
some strange doom. 

Nay, let my fate go whither it will: but as touching 
my children,—I pray thee, Creon, take no care on thee for 
my sons; they are men, so that, be they where they may, 
they can never lack the means to live. But my two girls, 
poor hapless ones,—who never knew my table spread apart, 


with most of the later Mss. 


109 MSS. give προσθῇ. 
sos OUs 


The ancient grammarians were not agreed on the accen- 
tuation of such forms; cp. Chandler, Greek Accentuation, § 820, 2nd ed. 
Elmsley conjectured προθῇ (V has πρόθη). 

Attic inscriptions of the 5th and 4th cent. B.C. recognise no dual in -a, -aww for 


In Her. 6. 
1462 f. τοῖν 





οὐκ av ἔλαβεν αὐτήν), but Xen. Mem. 1. 
2. 41 οἶμαι μὴ ἂν δικαίως τυχεῖν τούτου τοῦ 
ἐπαίνου τὸν μὴ εἰδότα: (2) Plat. Prot. 
236 Β ὁμολογεῖ μὴ μετεῖναί οἱ μακρολογίας, 
but AZol. 17 A ὁμολογοίην ἂν ἔγωγε οὐ 
κατὰ τούτους εἶναι ῥήτωρ. Cp. Whitelaw 
in Zrans. Cam. Phil. Soc. (1886) p. 34, 
and Gildersleeve in Amer. Fourn. Philol. 
I. 49.—Whitelaw here takes mépoa ἄν 
as=émrepoey ἄν, and reads τῷ (not Tw) 
δεινῷ κακῷ: ‘my parents wished to kill 
me; but nothing could have killed me; I 
was reserved for Ζάζς dread evil.’ Surely, 
however, it is better to connect the 
verses with the wish for death which he 
has just uttered. The poet of Colonus 
gives Oedipus a presentiment that his 
end is not to be as that of other men. 

1457 with μή understand σωθείς, -Ξ εἰ 
μὴ ἐσώθην ἐπὶ κακῷ Tw: cp. Az. 950 οὐκ 
ἂν τάδ᾽ ἔστη τῇδε μὴ θεῶν μέτα, sc. στάν- 
Ta=el μὴ ἔστη. 

1460 προσθῇ μέριμναν, ake care upon 
thee: so often of assuming a meed/ess bur- 
den: Thuc. 1. 78 μὴ... οἰκεῖον πόνον προσ- 
θῆσθε: 10. 144 κινδύνους αὐθαιρέτους μὴ 
προστίθεσθαι: Plat. Prot. 346 Ὁ ἔχθρας 
ἑκουσίας...προστίθεσθαι. Elmsley’s plau- 
sible προθῃ (Z/. 1334 εὐλάβειαν προΐ- 
θέμην) would be weaker.—dvSpes, males 
(though not ἐξηνδρωμένοι) ; cp. 77. 1062 
θῆλυς οὖσα KovK ἀνδρὸς φύσιν. 

1462 ff. τοῖν δ᾽ ἀθλίαιν. Instead of 
supplying πρόσθου μέριμναν, it is better 
to regard οἷν in 1466 as an anacolouthon 
for τούτοιν, arising from the length of the 
preceding clause. Cp. Antiphon or. 5 
§§ 11, 12 δέον ce διομόσασθαι.. ἃ σὺ παρ- 
ελθών, where, after a long parenthetic 


clause, ἅ has been irregularly substituted 
for ταῦτα. 

1468 f£. οἷν for whom ἡ ἐμὴ βορᾶς 
τράπεζα the table at which 7 ate οὔποτε 
χωρὶς ἐστάθη was never placed apart, 
ἄνευ τοῦδ᾽ ἀνδρός (so that they should be) 
without me. Instead of ἄνευ αὐταῖν, we 
have ἄνευ τοῦδ᾽ ἀνδρός, because (οἷν being 
dat. of persons affected) οἷν οὔποτε ἡ ἐμὴ 
τράπεζα χωρὶς ἐστάθη ἄνευ τοῦδ᾽ ἀνδρός 15 
equivalent to ὦ οὔποτε τὴν ἐμὴν τράπεζαν 
χωρὶς σταθεῖσαν εἰδέτην, (ὥστε εἶναι) ἄνευ 
τοῦδ᾽ ἀνδρός. This is simpler than to 
construe: ‘for whom the dinner-table, 
which was (always) mine, was never 
placed apart, or without me’: when ἡμή 
would be a compressed substitute for 7 
ἐμὴ del οὖσα in the sense of ἀλλὰ ἡ ἐμὴ 
del ἦν. We cannot take ἡμὴ βορᾶς 
τράπεζα as merely=‘the table which 1 
provided’: the emphasis on ἡμή would 
alone exclude this. Prof. Kennedy un- 
derstands: ‘apart from whom (οἷν xwpis) 
my dinner-table ne’er was set without my 
bidding,’ 1.5. never except on special oc- 
casions, when I had so directed. dvev 
could certainly mean this (O. C. 926 etc.). 
But can we understand Oedipus as say- 
ing, in effect,—‘who always dined with 
me—except, indeed, when I had directed 
that they should zo¢’??—I am much in- 
clined to receive Arndt’s ἄλλη for ἡμή 
(AA for M),as Wecklein has done.—The 
attributive gen. Bopds is equivalent to an 
adj. of quality like τρόφιμος, as Eur. 
Phoen. 1491 στολὶς τρυφᾶς Ξε στολὶς τρυ- 
φερά: not like ἅμαξαι σίτου (Xen. Cyr. 
2. 4. 18) ‘waggon-/oads of grain.’—éord- 
θη, because a light table is brought in for 


190 


ZOPOKAEOY2 


τράπεζ' ἄνευ τοῦδ᾽ avd ρός, ἀλλ᾽ ὅσων ἐγὼ 


ψαύοιμι, πάντων τῶνδ᾽ ἀεὶ μετειχέτην' 


1465 


οἷν μοι μέλεσθαι: καὶ μάλιστα μὲν χεροῖν 
ψαῦσαί μ᾽ ἔασον κἀποκλαύσασθαι κακά. 


i? ὠναξ, 


νΩ 4 A “A 
10 ω γονῇ γενναιξε. 


τί φημί; 


΄ Ἅ Ν 
χερσί τἂν θιγὼν 

re Sot Y (PT τ 
δοκοῖμ᾽ ἔχειν σφας, ὥσπερ ἡνίκ᾽ ἔβλεπον. 


1470 


οὐ δὴ κλύω που πρὸς θεῶν τοῖν μοι φίλοιν 
δακρυρροούντοιν, καί μ᾽ ἐποικτείρας Κρέων 
» “4 Ν 4 > > ’ 3 ~ 
ἔπεμψέ μοι τὰ φίλτατ᾽ ἐκγόνοιν ἐμοῖν ; 


λέγω τι; 
ie. 


1475 


λέγεις" ἐγὼ γάρ εἰμ᾽ ὁ πορσύνας τάδε, 


γνοὺς τὴν παροῦσαν τέρψιν, ἢ σ᾽ εἶχεν πάλαι. 


Ol. 


ἀλλ᾽ εὐτυχοίης, καί σε τῆσδε τῆς ὁδοῦ 


δαίμων ἄμεινον ἢ μὲ φρουρήσας τύχοι. 


ὦ τέκνα, ποῦ ποτ᾽ ἐστέ; 


δεῦρ᾽ ir, ἔλθετε 1480 


ws τὰς ἀδελφὰς τάσδε Tas ἐμὰς χέρας, 


pronoun-forms in -a, -ἢ. 
terhans, 67. α΄. Att. lnschr. p. 50. 
by Brunck, Erfurdt, and others. 


Thus they give, as fem., τώ, τοῖν, τούτοιν, οἷν. 


See Meis- 


1466 οἷν] Heath’s emendation ταῖν is received 
I found ταῖν in one of the later Mss., V?, and Blaydes 


cites it from cod. Paris. 2820, with gloss τούτων: it was probably an old conjecture, 


intended to smooth the construction. 


See comment. on 1462 ff. 


1470 c¢ac L, 





the meal, and removed after it (cp. 72. 
24. 476, Od. το. 354 etc.).—dvev τοῦδ᾽ 
ἀνδρός, explaining χωρίς, as in PA. 31 
κενὴν οἴκησιν is explained by ἀνθρώπων 
δίχα, Az. 464 γυμνὸν φανέντα by τῶν 
ἀριστείων ἄτερ. avevasin 77. 336 μάθῃς 
ἄνευ τῶν δ᾽, hear apart from these. 

1466 μέλεσθαι, infin. for imper.: cp. 
462. μάλιστα μέν : see on 926. 

1468 ἴθ᾽ ὦναξ. A moment of agitated 
suspense is marked by the bacchius inter- 
rupting the trimeters, as PA. 749 f. (in an 
anxious entreaty, as here) ἴθ᾽, ὦ παῖ. So 
O. C. 1271 τί σιγᾷς; 318 τάλαινα. The 
speech of the agonised Heracles is simi- 
larly broken by short dactylic or chori- 
ambic phrases, 7”. 1081, al, al, ὦ τάλας: 
1085 ὦναξ ᾿Αἴδη δέξαι μ᾽, [ὦ Διὸς ἀκτίς, 
παῖσον. But Soph. has used the license 
most sparingly, and always, it may be 
said, with fine effect. 

1469 γονῇ γενναῖε, noble in the 
grain,—one whose γενναιότης is γνησία, 
inbred, true,—referring to the ἀρετή just 


shown by Creon (1433). γονῇ here is 
not merely intensive of γενναῖε, making 
it=~yevvadrare, (as the sarcastic γένει 
seems to be in Plat. Soph. 231 Β ἡ γένει 
γενναία σοφιστική, ‘the most noble.’) 
Cp. Az. 1094 μηδὲν ὧν γοναῖσιν. 

1470 δοκοῖμ᾽ : for this form, cp. PA. 
895 dpm’ (n.). ἔχειν σφας. σφέας has 
the accent in Homer when it is emphatic, 
as when joined with αὐτούς, being then 
a disyllable: 7/7. 12. 43 σφέας αὐτούς... 
When non-emphatic and enclitic, it is a 
monosyllable: Od. 4. 77 καί σῴεας φωνή- 
σας. The perispomenon σῴφᾶς.. corre- 
sponds to σφέας, as in σφᾶς αὐτούς: the 
enclitic cpas to σῴεας. Thus in Ο. Ci. 
486 we must write ws σῴας καλοῦμεν with 
Herm.; where Elmsley gave ws σφᾶς, 
holding (against the grammarians) that 
this form was never enclitic. Here, as in 
1508, the pronoun is non-emphatic. Ac- 
cording to the rule now generally received, 
a monosyllabic enclitic stands unaccented 
after a paroxytone word, the latter re- 


OIAITOY2 ΤΥΡΆΝΝΟΣ Ig! 
or lacked their father’s presence, but ever in all things shared 
my daily bread,—I pray thee, care for them, and—if thou canst 
—suffer me to touch them with my hands, and to indulge my 
grief. Grant it, prince, grant it, thou noble heart! Ah, could I 
but once touch them with my hands, I should think that they 


were with me, even as when 1 had sight... 
[CREON’S Attendants lead in the children 


Ha? 


ANTIGONE azd ISMENE.] 


O ye gods, can it be my loved ones that I hear 


sobbing,—can Creon have taken pity on me and sent me my 
children—my darlings? Am [I right? 

Cr. Yea: ’tis of my contriving, for I knew thy joy in them 
of old,—the joy that now is thine. 


OE. 


Then blessed be thou, and, for guerdon of this errand, 


may heaven prove to thee a kinder guardian than it hath to 


me! 


My children, where are ye? 


Come hither—hither to 


the hands of him whose mother was your own, 


though the ἃ might easily be taken for ἃ, the accent found in some later mss. 


1474 éyyévow L; ἐκγόνοιν τ (B, V4). 
evidently a prosaic correction. 


1477 ἡ σ᾽ εἶχεν L: ἣν εἶχες τ (including A), 
Wunder, whom Hermann and others follow, adopts 7 
σ᾽ ἔχει from one 14th century Ms. (Laur. 32. 2), taking πάλαι with γνούς. 
σαν Kviéala conjectures πάρος σὴν, Blaydes πάροιθε. 


For παροῦ- 
1481 ὡς Mss.: εἰς Elmsley. 





maining unaffected: we therefore write 
ἔχειν ogas. But, according to Arcadius 
and Herodian, a paroxytone word fol- 
lowed by an enclitic deginning with oo 
took the acute on its last syllable, as 
éxelv ogas: see Chandler, §§ 965, 966, 
2nd ed. ; 

1471 τί φημί; the cry of one startled 
by a sound or sight, as 77. 865: O. Ὁ: 
315 τί φῶ; Aesch. P. V. 561 τίς γῆ; τί 
γένος ; τίνα φῶ λεύσσειν; 

1472 f. τοῖν.. φίλοιν | δακρυρροούν- 
tow. Cp. Ant. 381 οὐ δή πον..; ἴῃ par- 
ticiples belonging to the 3rd declens. the 
masc. form of the dual is often used as 
fem.; indeed the specially fem. forms, 
such as ἐχούσα, are very rare. See O.C., 
append. on 1676, p. 293. Similarly τώ, 
τοῖν, τούτοιν, οἷν were the usual fem. 
forms: cp. 1462 f., 1504, and Azz. 769 n. 
Thus Xen. Cyr. 1. 2. 11 μίαν ἄμφω τούτω 
τὼ ἡμέρα λογίζονται. Plat. Phaedr. 237 
D ἡμῶν ἐν ἑκάστῳ δύο τινέ ἐστον ἰδέα 
ἄρχοντε καὶ ἄγοντε, οἷν ἑπόμεθα. So τὼ 
θεώ, τοῖν θεοῖν (Demeter and Persephone). 

1474 τὰ φίλτατ᾽ ἐκγ. ἐμοῖν, my chief 
treasure, (consisting in) my two daugh- 
ters: cp. on 261 κοινῶν παίδων κοινά: Κ΄. 
682 πρόσχημ᾽ ἀγῶνος, a glory (consisting 
in) a contest, 


1475 λέγω τι; see Plat. Ογαί. 404 A 
κινδυνεύεις τι λέγειν, compared with 
Symp. 205 D κινδυνεύεις ἀληθῇ λέγειν. 
Ar. £9. 333 νῦν δεῖξον ὡς οὐδὲν λέγει τὸ 
σωφρόνως τραφῆναι, ‘what nonsense it is.’ 

1477 yvovs...mdAat: aware of the | 
delight which you now feel,—as you ever 
felt it: 2.6. taught by the past to foresee 
that you would thus rejoice. 

1478 Soph. may have been thinking 
of Aesch. Cho. 1063 ἀλλ᾽ εὐτυχοίης, καί 
σ᾽ ἐποπτεύων πρόφρων | θεὸς φυλάττοι 
καιρίοισι συμφοραῖς. τῆσδε τῆς ὁδοῦ, 
causal gen.: 222. 626 θράσους | τοῦδ᾽ οὐκ 
ἀλύξεις - Eur. Or. 1407 ἔρροι τᾶς ἁσύχου 
προνοίας. 

1479 ἢ ᾽μὲ is required here, since 
with # με the stress would fall wholly on 
φρουρήσας. On the other hand in 1478 
καί oe is right, because, after εὐτυχοίης, 
the Zerson does not need to be at once 
emphasised again. This is not, however, 
like 77. 23. 724 ἢ mw’ ἀνάειρ᾽ ἢ ἐγὼ σέ, 
where we suffices because the sense is, 
‘slay or be slain.’ In El. 383, 1213 μὲ 
and go are justified by the stress on 
ὕστερον and προσήκει respectively. 

1481 ὡς τὰς.. χέρας. As the sense is 
so plainly equivalent to ws ἐμέ, we are 
scarcely justified in changing ὡς to εἰς 


192 


ZOPOKAEOYS 


at τοῦ φυτουργοῦ πατρὸς ὑμὶν ὧδ᾽ ὁρᾶν 
τὰ πρόσθε λαμπρὰ Tpou νη σαν ὄμματα' 


ὃς ὑμίν, ὠ τέκν᾽, 


oul ὁρῶν ov? ἱστορῶν 
πατὴρ ἐφάνθην Bio hiae ἡρόθην. 


1485 


καὶ σφὼ δακρύω" προσβλέπειν γὰρ οὐ σθένω: 
νοούμενος τὰ λοιπὰ τοῦ πικροῦ βίου, 

οἷον βιῶναι σφὼ πρὸς ἀνθρώπων χρεών. 

ποίας γὰρ ἀστῶν ner εἰς ὁμιλίας, 


, > ε ’ » 3 ig 
ποίας δ ἑορτάς, ἔνθεν οὐ κεκλαυμέναι 


[490 


πρὸς οἶκον ἵξεσθ' ἀντὶ τῆς θεωρίας: 
ἀλλ' ἡνίκ᾽ ἂν δὴ πρὸς γάμων ἤκητ᾽ ἀκμάς, 
τίς οὗτος ἔσται, τίς παραρρίψει, τέκνα, 


1487 τὰ λοιπὰ τοῦ πικροῦ] Some of the later Mss. have τὰ πικρὰ τοῦ λοιποῦ, which 


Blaydes prefers, because hitherto their lives had not been bitter. 
been the motive of the change, unless it was a mere oversight: 
is equivalent to τὸν λοιπὸν βίον τὸν πικρόν. 


This may have 
but L’s reading 
1491 ἵξεσθ᾽] ἥξεθ᾽ L ist hand: 





(with Elmsley), or és (with Blaydes). 
Tr. 366 δόμους | ws τούσδε is a slightly 
stronger case for such a change, yet not a 
conclusive one. és is now read for ws in 
Ar. Ach, 242 (ws τὸ πρόσθεν) and in 
Thuc. 8. 36 (ws τὴν Μίλητον), 103 (ws 
τὴν “ABvdov). Soph. has ws ὑμᾶς 77. 
66. 

ὶ 1482 f. Construe: at προὐξένησαν 
ὑμὶν who have effected for you τὰ πρόσθε 
λαμπρὰ τοῦ φυτ. πατρὸς ὄμματα ὧδε 
ὁρᾶν that the once bright eyes of your 
sire should see thus, z.e. should be sight- 
less: cp. his own phrase quoted in 1273 
ἐν σκότῳ τὸ λοιπὸν... ὀψοίατος Ph. 862 ws 
᾿Αἴδᾳ παρακείμενος ὁρᾷ, he sees as the 
dead, 2:5. not at all. Cp. Xen. Aol. 
Socr. ὃ 7 ὁ θεὸς δι᾽ εὐμένειαν προξενεῖ μοι 
οὐ μόνον τὸ ἐν καιρῷ τῆς ἡλικίας καταλῦ- 
σαι τὸν βίον, ἀλλὰ καὶ τὸ ἣ ῥᾷστα, the 
god’s kindly offices grant to me that I 
should close my life etc. προξενεῖν =(1) 
to be a πρόξενος: then (2) fig., to lend 
one’s good offices: either (a) absol., as 
O. C. 465 προξένει, stand my friend: or 
(ὁ) with dat. and acc., or acc. and infin., 
to effect a thing, or result, for one: Xen. 
An, 6. 5. 14 lore...pe...ovd€va πω κίνδυνον 
προξενήσαντα ὑμῖν : Plut. Alex. 22 αὐτῷ... 
τοιαῦτα ὀνείδη προξενῶν (said of one who 
panders to vices): Soph. 77. 726 ἐλπὶς 
ἥτις Kal θράσος τι προξενεῖ. In particular, 
προξενεῖν τινά τινι-ε συνιστάναι, to intro- 


duce one person to another. So Prof. 
Kennedy understands here: ‘which in- 
troduced to you your father’s once 
brilliant eyes, that you should thus 
behold them’—i.e. presented them to 
you in this state. But ὧδ᾽ ὁρᾶν seems 
thus to lose its force: and the ordinary 
usage of προξενεῖν confirms the version 
given above. The conjecture προυσέλη- 
σαν (‘maltreated’) has found some un- 
merited favour. Besides προυσελούμενον 
in Aesch. P. V. 438, we find only πρου- 
σελοῦμεν in Ar. Ran. 730. 

1484 οὔθ᾽ ὁρῶν οὔθ᾽ ἱστορῶν: i.e. | 
neither recognising his mother when he 
saw her, nor possessing any information 
which could lead him to suspect that she 
was such. ἱστορεῖν is (1) to be, or (2) to 
become, ἵστωρ, a knower: 2.¢. (1) to have 
information, or (2) to seek it. Sense (2) 
is more frequent: but Aesch. has (1) in 
Lum. 455 and fers. 454. [In Zr. 382 
οὐδὲν ἱστορῶν ΡΓΟΡ. Ξε ὅτι οὐδὲν ἱστόρει 
(imperf.), ‘did not ask.’] Here (1) is 
best, because it would be almost absurd 
to say that he had wedded Iocasta ‘ with- 
out asking any questions’—as if he could 
have been expected to do so. Cp. 0. C. 
273 νῦν δ᾽ οὐδὲν εἰ δὼς ἱκόμην ἵν᾽ ἡ κεῖ" 

1485 ἠρόθην: cp. 1257, 1210. 

1489 f. ὁμιλίας... ἑορτάς. The poet 
is thinking ae his own Athens, though the 
language is general. ὁμιλίας comprises 


OIAITOY2 TYPANNOS 193 
the hands whose offices have wrought that your sire’s once bright 
eyes should be such orbs as these,—his, who seeing nought, 
knowing nought, became your father by her from whom he 
sprang! For you also do I weep—behold you I cannot—when I 
think of the bitter life in days to come which men will make you 
live. To what company of the citizens will ye go, to what festi- 
val, from which ye shall not return home in tears, instead of shar- 
ing in the holiday? But when ye are now come to years ripe for 
marriage, who shall he be, who shall be the man, my daughters, 


an early corrector (the first, 5, acc. to Diibner) changed this to té6’, writing o 
above the ε, 7.6. ἵξεσθ᾽. Some of the later mss. (B, E, ΝΆ) have ἥξετ᾽, generated, 
doubtless, by ἥξετ᾽ in 1489: as conversely in 1489 T has fier’, prompted by ἵξεσθ' 


here. 


1498 ἔσται, ris] Elmsley conjectured ἐστιν ὃς (one of the later Mss., E, 





all occasions on which Attic women 
could appear in public,—as at the de- 
livery of ἐπιτάφιοι (Thuc. 2. 45): ἑορτάς 
suggests such festivals as the Thesmo- 
phoria, the Panathenaea, or the Dionysia 
(when women were present in the theatre, 
at least at tragedy). To feel the force of 
this passage, we must remember how 
closely the Greek festivals were bound 
up with the life of the famzly. Kinsfolk 
took part in them together: and at such 
moments a domestic disgrace, such as 
that which the sisters inherited, would be 
most keenly felt. In Athenian law-courts 
the fact of association at festivals could 
be cited in evidence of family intimacy: 
Isocr. or. 19 ὃ 10 ἕως μὲν γὰρ παῖδες 
ἦμεν, περὶ πλέονος ἡμᾶς αὐτοὺς ἡγούμεθα ἢ 
τοὺς ἀδελφούς, καὶ οὔτε θυσίαν οὔτε θεω- 
ρίαν (public spectacle) οὔτ᾽ ἄλλην ἑορ- 
τὴν οὐδεμίαν χωρὶς ἀλλήλων ἤγομεν. 
1546. or. 8 § 15 καὶ εἰς Διονύσια εἰς ἀγρὸν 
ἦγεν ἀεὶ ἡμᾶς, καὶ μετ᾽ ἐκείνου τε ἐθεω- 
ροῦμεν (in the theatre) καθήμενοι παρ᾽ 
αὐτόν, καὶ τὰς ἑορτὰς ἤγομεν παρ᾽ ἐκεῖνον 
πάσας. It was the Attic custom for a 
bridegroom Θεσμοφόρια ἑστιᾶν τὰς γυναῖ- 
kas, to provide a banquet at the next 
Thesmophoria for the women of his deme 
(Isae. or. 3 ὃ 80), and also φράτορσι 
γαμηλίαν εἰσφέρειν, to provide a banquet 
for his clansmen when his bride was in- 
troduced into his φρατρία (or. 8 § 18). 
1490 κεκλαυμέναι, only poet.: later 
poets and Plut. have κέκλαυσμαι: the 
poet. δεδακρυμένος also occurs in later 
prose, Plut., Lucian, etc. The festivals 
were religious celebrations, which would 
be polluted by the presence of persons 
resting under an inherited ἄγος (cp. note 


fe eg 


on 240). Some word or act reminds the 
daughters of Oedipus that they are thus 
regarded, and they go home in tears. 
Greek sensitiveness to public notice on 
such occasions might be illustrated by the 
story in Her. of the affront offered to the 
deposed king Demaratus by his successor 
Leotychides at the Spartan festival of the 
γυμνοπαιδίαι (6. 67). Demaratus drew 
his robe over his head, and left the 
theatre: κατακαλυψάμενος ἤϊε ἐκ τοῦ 
θεήτρου és τὰ ἑωυτοῦ οἰκία. Contrast the 
effusive public greeting which Electra 
imagines herself and Chrysothemis as re- 
ceiving ἔν θ᾽ ἑορταῖς ἔν τε πανδήμῳ πόλει 
(1. 982). 

1491 ἀντὶ τῆς θεωρίας, in place of the 
sight-seeing (for which they had looked). 
θεωρία is (1) subjectively, a sight-seeing: 
a objectively, a spectacle. In sense 

1) the article is added here because a 
definite occasion is meant; usually, the 
art. is absent: Thuc. 6. 24 πόθῳ ὄψεως 
kal θεωρίας : Plat. Rep. 556 C ἢ κατὰ θεω- 
plas ἢ κατὰ στρατείας (on travels or cam- 
paigns): Isocr. or. 17 ὃ 4 ἅμα κατ᾽ ἐμπο- 
ρίαν καὶ κατὰ θεωρίαν. In Her. 1. 30 
τῆς θεωρίης ἐκδημήσας... εἵνεκεν, the art. 
is added as in ἡ εἰρήνη (‘peace’) εἰς., 
because ‘seeing the world’ is spoken of 
generically. 

1498 τίς οὗτος ἔσται, tls, κιτ.λ., is 
more animated for τίς οὗτος ἔσται, ὅστις. 
Theocr. 16. 13 τίς τῶν νῦν τοιόσδε; τίς εὖ 
εἰπόντα φιλασεῖ; is compared by Jacobs 
there, and by Schneidewin here, but is 
not really similar, since τοιόσδε there re- 
fers back to v. 5 ἴω, τίς γάρ.. ὑποδέξεται 
(ΚΎΤΟΣ 


ΤΊΣ 


194 


Coie ee 5 , / a 
τοιαῦτ᾽ ὀνείδη λαμβάνων, ἃ 
ὡς A » n~ >] ε “ 4 

γοναῖσιν ἔσται σφῷν θ᾽ ὁμοῦ δηλήματα; 


ΣΟΦΘΚΛΕΟΥΣ 


ὡς ἴω 9 ἴω 
ταις ἐμαις 


τί γὰρ κακῶν ἄπεστι; τὸν πατέρα πατὴρ 
ὑμῶν ἔπεφνε: τὴν τεκοῦσαν ἤροσεν, 

ὅθεν περ αὐτὸς ἐσπάρη, κἀκ τῶν ἴσων 
ἐκτήσαθ' ὑμᾶς ὧνπερ αὐτὸς ἐξέφυ. 


τοιαῦτ᾽ ὀνειδιεῖσθε: 


οὐκ ἔστιν οὐδείς, ὦ τέκν᾽, ἀλλὰ δηλαδὴ 
έρσους φθαρῆναι κἀγάμους ὑμᾶς χρεών. 


@ παῖ Μενοικέως, 


τούτοιν λέλειψαι, νὼ γάρ, ὧ Φυτεύσαμεν, 


ὀλώλαμεν δύ᾽ ὄντε, 


πτωχὰς ἀνάνδρους ἐγγενεῖς ἀλωμένας, 
μηδ᾽ ἐξισώσῃς τάσδε τοῖς ἐμοῖς κακοῖς. 
ἀλλ᾽ οἰκτισόν σφας, ὧδε τηλικάσδ᾽ opav 
πάντων ἐρήμους, πλὴν ὅσον τὸ σὸν μέρος. 


ὕννευσον, ὦ γενναῖε, σῇ ψαύσας χερί. 
δ᾽, ὦ τέκν᾽, εἰ μὲν εἰχέτην ἤδη φρένας, 
--- 


σφῴν 


has ἔσται γ᾽ ὅΞ) : 
σιν MSS. Schenkel conjectures γόνοισιν : 
γοναῖσιν. 


‘at languet hoc,’ as Hermann says. 
Arndt, γαμβροῖσιν : 
Hartung changes ἐμοῖς to γάμοις, and δηλήματα to ᾿κμεμαγμένα (‘re- 


1495 
κἄτα τίς γαμεῖ; I500 
ἀλλ᾽ ἐπεὶ μόνος πατὴρ 
μή σφε ὅ περιίδῃς 1505 

1510 


1494 f. τοῖς ἐμοῖς | γονεῦ- 
Kennedy ταῖς éuais | 


proaches which will cleave to your marriage, on your parents’ account and on your 


own’). 


Heimsoeth would keep γονεῦσιν, and change ἃ τοῖς ἐμοῖς. to ἃ ᾿κ τῆς ἴσης. 


1497 ff. Nauck supposes that Soph. wrote, after ἔπεφνεν, merely οὗπερ αὐτὸς ἐσπάρη, 


κἀκτήσαθ᾽ ὑμᾶς ὧνπερ αὐτὸς ἐξέφυ. 


He now grants that ὅθεν can mean ἐξ ἧς, but 





1494 λαμβάνων instead of the infin. 
with παραρρίψει, as Plat. Legg. 699 A 
οὐδεὶς τότε ἐβοήθησεν οὐδ᾽ ἐκινδύνευσε 
ξυμμαχόμενος. 

1495 γοναῖσιν. The disgraces of the 
polluted house will be ruinous not only 
to the children of Oedipus, but to his 
children’s children (σφῷν, genit., sc. yo- 
vais). Iformerly read γόνοισιν : but Ken- 
nedy justly objects that the plur. of γόνος 
is not used; and his conjecture, ταῖς 
ἐμαῖς yovaiow, gives more point here. 
For γοναί, ‘ offspring,’ cp. O. C. 1192, 
Ant. 641. The γονεῦσιν of the MSs. 
yields no tolerable sense, whether it is 
referred to Laius and Iocasta or to 
Iocasta alone.—dé7Anva is a hurt, bane, 
mischief, in a physical or material sense: 
Od. 12. 286 ἄνεμοι χαλεποί, δηλήματα 
νηῶν : Hom. Hom. Hymn. Apoll. 364 (of 
the dead monster) οὐδὲ σύ ye ζώουσα κα- 
κὸν δήλημα βροτοῖσιν: Aesch. fr. 119 ὁδοι- 


πόρων δήλημα χωρίτης δράκων (the ser- 
pent in the fields, a bane of wayfarers). 
The disgraces are δηλήματα to the sons 
and daughters as involving their ruin in 
life: but could not be called δηλήματα to 
the dead in the remote figurative sense 
of disgracing their memories. Nor would 
there be any fitness in the conjunction 
of harm of another kind to the. living. 
Oedipus here thinks of the living, and 
of the future, alone. The conject. yap- 
βροῖσιν, besides being far from the ss., 
presumes the event which he regards as 
impossible. 
1496 πατέρα: for the tribrach see on 
710. 
1498 τῶν ἴσων is poetically equiva- 
lent to τῶν αὐτῶν, t.e τῆς αὐτῆς: it is 
like saying, ‘from a source which was 
Pie as that whence he sprang,’ instead 
» ‘from the same source whence he 


eee Cp. 845 οὐ γὰρ γένοιτ᾽ ἂν εἷς 


OIAITOYS ΤΥΡΆΝΝΟΣ 195 
that will hazard taking unto him such reproaches as must be 
baneful alike to my offspring and to yours? For what misery 
is wanting? Your sire slew his sire, he 2who 
bare him, and begat you at the sources of his own being ! Such 
are the taunts that will be cast at you; and who then will wed? 
The man lives not, no, it cannot be, my children, but ye must 
wither in barren maidenhood. 

Ah, son of Menoeceus, hear me—since thou art the only 
father left to them, for we, their parents, are lost, both of us,— 
allow them not to wander poor and unwed, who are thy kins- 
women, nor abase them to the level of my woes. Nay, pity 
them, when thou seest them at this tender age so utterly forlorn, 
save for thee. Signify thy promise, generous man, by the touch 
of thy hand! To you, my children, I would have given much 


objects to τῶν ἴσων, and to the marriage being dwelt upon at more length than the 
parricide. 1505 μή σφε παρίδῃς MSS. (παρίδησ L). Dawes conjectured μή ode 
περιίδῃς : Fritzsch, μὴ περί σφ᾽ ἴδῃς: μὴ παρά od’ ἴδῃς Porson: Erfurdt, μή ope δὴ 


(μοι Blaydes) προδῷς, and afterwards μή σφ᾽ ἀτιμάσῃς. 


in L from évyeveic). 


1506 ἐγγενεῖς MSS. (made 


Dindorf conjectures ἐκγενεῖς, comparing ἔκβιος, ἔκτιμος, ἐξούσιος : 
Hermann, doréyous: Schneidewin, ἐκστεγεῖς: Wolff, συγγενής. 


1511 εἰχέτην MSS.: 





γε τοῖς πολλοῖς ἴσος, and note. 

1500 ὀνειδιεῖσθε: see on 672. 

1501 δηλαδή: prosaic, but also in 
Eur. Or. 180) I. A. 1366. 

1508 ἀλλ᾽ after the vocative, like od 
δέ, but stronger, as introducing an ap- 
peal: as O. C. 1405 ὦ τοῦδ᾽ ὅμαιμοι παῖ- 
des, GAN’ ὑμεῖς... μή μ᾽ ἀτιμάσητέ ye: and 
4b, 237. 

1505 δύ᾽ ὄντε, both of us: cp. 77. 539 
δύ᾽ οὖσαι μίμνομεν : Eur. Jon 518 σὺ δ᾽ εὖ 
φρόνει γε καὶ δύ᾽ ὄντ᾽ εὖ πράξομεν.----περιί- 
Sys: on Porson’s objection, see Appendix. 

1506 éyyevets, your kinswomen as 
they are (where in prose we should have 
οὔσας added). The word was full of 
meaning for an Attic audience, who 
would think of Creon as placed by 
Oedipus in the position of ἐπίτροπος 
(guardian) and κύριος (representative be- 
fore the law) of the unmarried girls who 
are here viewed as orphans (1505); their 
brothers not being of age. Cp. Isae. or. 
5 § 10; [Dem.] or. 46 § 18. 

1507 ἐξισώσῃς τάσδε, do not put 
them on the level of my miseries: cp. 
425: for τάσδε instead of τὰ τῶνδε κακά, 
cp. note on 467. 

1508 τηλικάσδ᾽, at their age, Ζ.6. so 
young: Ant. 726 οἱ τηλικοίδε (so old) καὶ 
διδαξόμεσθα δὴ | φρονεῖν πρὸς ἀνδρὸς τηλι- 
κοῦδε (so young) τὴν φύσιν; 


_ of the normal -τον. 


1509 πλὴν ὅσον τὸ σὸν μέρος, ex- 
cept in so far as, on thy part, οὐκ ἔρημοι 
εἰσί." 

1511 εἰχέτην, 2nd pers. dual, with 
the form proper to the 3rd (μετειχέτην, 
1465). Before the Attic period, the 
Greek language had attained to this re- 
gular distinction of active dual forms :— 
(1) primary tenses, 2nd pers. -rov, 3rd 
pers. -rov ; (2) secondary tenses, 2nd pers. 
-Tov, answering to Skt. ¢am: 3rd pers. 
«τὴν, Skt. tam. As regards (2), two 
classes of exceptions occur: (2) Homeric 
3rd pers. in -rov instead of -τὴν ; three 
instances, διώκετον (71. το. 364), érevxe- 
tov (13. 346), λαφύσσετον (18. 583). 
These Curtius refers to ‘the want of 
proper linguistic instinct on the part of 
some late rhapsodist.’ (ὁ) Attic 2nd pers. 
in -τὴν instead of -rov. Our εἰχέτην here 
is the only instance proved by metre: but 
8 others are established. Against these 
fall to be set at least 13 Attic instances 
Curtius regards the 
and pers. in -ryv as due-to a false an- 
alogy. In the ¢hzvd person dual -τὴν 
was distinctive of the secondary tenses. 
Attic speech sometimes extended this 
distinction to the second person also. 
(Curtius, Verb 1. 80, Eng. tr. 53.) Cp. 
n. on O. C. 1378 f. 


13—2, 


ZOPOKAEOY2 


ld > aA 4 A“ A “a 3 ¥ id 
πόλλ᾽ ἂν παρήνουν" νῦν δὲ τοῦτ᾽ εὐχεσθέ pou, 
οὗ καιρὸς *éea ζῆν, τοῦ βίου δὲ λῴονος 
ὑμᾶς κυρῆσαι του φυτεύσαντος πατρός. 


ΚΕ. 
Ol. 
Ol. 


, 3 δὲ τὸ te 
πειστέον, KEL μηδὲν NOV. 
> » ey & > > ὁ 
οἷσ θ᾽ ἐφ᾽ οἷς οὖν εἶμι; 
κλύων. 
Ol. 


οσνν. 


ΟΙ. ἀλλὰ θεοῖς γ᾽ ἔχθιστος ἥκω. 
τάχα. 

ΟΙ. 
μάτην. 

OI. ἀπαγέ νύν μ᾽ ἐντεῦθεν ἤδη. 


δ᾽ ἀφοῦ. 


εἴχετόν γ᾽ Brunck. 


a 7 Y , ¥ 
γῆς μ᾽ OTS πέμψεις ἄποικον. 


1512 εὔχεσθέ μοι MSS. 


te 
ἅλις ἵν᾿ ἐξήκεις δακρύων: ἀλλ᾽ ἴθι στέγης ἔσω. 1515 


ΚΡ, πάντα γὰρ καιρῷ καλά. 


ΕΝ 
ΚΡ. λέξεις, καὶ τότ᾽ εἴσομαι 


ΚΡ. τοῦ θεοῦ μ᾽ αἰτεῖς 


ΚΡ, τοιγαροῦν τεύξει 


φὴς τάδ᾽ οὖν; ΚΡ. ἃ μὴ φρονῶ γὰρ οὐ φιλῶ λέγειν 


1520 
KP. στεῖχέ νυν, τέκνων 


(In L the third ε had been a.)— 


Wunder, εὔχεσθ᾽ ἐμοί: Blaydes, τοῦθ᾽ ἕν εὔχομαι (so Wecklein), suggesting also τοῦτ᾽ 


ἐπεύχομαι: Dindorf, ηὔχθω μόνον. 


Tr. 610 ηὔγμην, midd.: but the imperat. οὗ ηὖγμαι does not occur.) 
The modes of correction tried have been 
(1) Omitting ἕῆν, Elmsley explains thus: εὔχεσθε κυρῆσαι τοῦ βίου 
οὗ καιρὸς det (κυρῆσαί ἐστι), Awovos δὲ τοῦ pur. πατρός. 


καιρὸς ἀεὶ ζῆν τοῦ βίου δὲ λῴονος MSS. 
chiefly three. 


(Plat. Phaedr. 279 C has noxra, pass., and Soph. 


1513 οὗ 


Hermann, also omitting 


ζῆν, makes εὔχεσθε passive (2.6. ‘let that prayer be made for you by me, which is 


fitting at each season’). 


(2) Omitting τοῦ, Hartung writes. οὗ καιρός, αἰεὶ ζῆν. βίου δὲ 





1512 ff. Oedipus now. turns from 
Creon to the children. e few words 
which he addresses to them are spoken 
rather to the older hearers and to him- 
self. τοῦτ᾽ εὔχεσθέ pot, ‘make this 
prayer, as I bid you’ face: ‘ pray on my 
account,’ in which sense Wunder reads 
ἐμοί): the ethic dat. μοι in request, as 
O. C. 1475. In these words Oedipus is 
thinking solely of his children: he has 
now passed away from the thought of 
self (1458). ὑμᾶς in 1514 is no argu- 
ment for understanding μὲ as subject to 
{nv: rather it is added to mark the con- 
trast with πατρός. 

1513 I prefer οὗ καιρὸς ἐᾷ ζῆν, τοῦ 
βίου κιτ.λ. to οὗ καιρὸς ἀεὶ ζῆν, βίου 
k.T.A. on these grounds. 1. τοῦ before 
Blov, though not required, is commend- 
ed, by Greek idiom; it also gives a de- 
cidedly better rhythm; and it is not likely 
to have crept into the text, since the oc- 
currence of del with the a long was not 
so uncommon that it should have sug- 


gested the need of supplementing the 
metre by τοῦ: but, apart from metrical 
motive, there was no other for intruding 
the article. 2. οὗ καιρός, without any 
verb, though a possible phrase, isa harsh 
one. 3. From eat to ae would be an 
easy transition. And καιρὸς ἐᾷ is quite 
a natural expression: cp. Eur. 7. 4. 858 
δοῦλος" οὐχ ἁβρύνομαι τῷδ᾽" ἡ τύχη γὰρ 
οὐκ ἐᾷ. The foreboding of Oedipus is 
that his daughters must become home- 
less exiles (1506) unless Creon shelters 
them at Thebes. ‘To live where occa- 
sion allows’ means in his inner thought, 
‘to live at Thebes, if that may be—if 
not, in the least warms τὲ exile that the 
gods may grant you.’ e monosyllabic 
ἔα (1451, Ant. 95) and ἐᾷ (//. 5. 256 
τρεῖν μ᾽ οὐκ ἐᾷ Παλλὰς ᾿Αθήνη) go far to. 
remove the metrical objection. Meineke’s. 
conjecture, 7, gives a more prosaic phrase, 
and is too far from the ἀεί of the mss. 

1615 ἐξήκεις: see on 1357. 

1516 καιρῷ--ἐν καιρῷς. In Thuc. 4. 


OIAITOYS TYPANNOS 197 
counsel, were your minds mature; but now I would have this 
to be your prayer—that ye live where occasion suffers, and that 
the life which is your portion may be happier than your sire’s. 


Cr. Thy grief hath had large scope enough: nay, pass into 
the house. 

Or. I must obey, though ’tis in no wise sweet. 
for it is in season that all things are good. 

OE. Knowest thou, then, on what conditions I will go? 
Cr. Thou shalt name them; so shall I know them when I hear. 

OE. See that thou send me to dwell beyond this land. 
Cr. Thou askest me for what the god must give. 

ΟΕ. Nay, tothe gods I have become most hateful. Cr. Then 
shalt thou have thy wish anon. 

OE. So thou consentest ? 
idly what I do not mean. 

Or. Then ’tis time to lead me hence. 
but let thy children go. 


Cr. Yea: 


CR: 


*Tis not my wont to speak 


Cr. Come, then,— 


λῴονος. Blaydes and Campbell read thus, but keep del, and place no comma after 
καιρός. (3) Others alter ἀεί. Dindorf gives οὗ καιρὸς ἐᾷ ζῆν, τοῦ βίου δὲ λῴονος. 
This has been the most generally received emendation, and seems the best. Meineke, 
ov καιρὸς ἦ ζῆν : Blaydes, οὗ καιρός, εὖ ζῆν. 1517 εἰμί L: εἶμι Brunck. 1518 πέμ- 
ψεισ L rst hand, corrected to πέμψηισ, and then (by a still later hand) back to 
πέμψεισ. The later Mss, are divided, but most have πέμψεις.---ἀπ᾽ οἴκων L, ov written 
over wy by a late hand. Most of the later Mss. have dm’ οἴκων (over which in A is 
yp. ἄποικον), but V? has ἀποίκων, and B ἄποικον. 1521 νῦν (47s) L, and so Wolff; 
νυν (δ᾽) Brunck, and most edd. T has νῦν. «νυν, but this, at least, can hardly be 





59 most MSS. give εἰ μὴ καιρῷ τύχοιεν 
ἑκάτεροι πράσσοντες: Classen reads ἐν 
καιρῷ on the ground that Thuc. so has 
it in 1. £21, §- 61, 6. 9. 

1517 The words οἶσθ᾽ ἐφ᾽ οἷς οὖν 
εἶμι; were said with some return of his 
former agitation: λέξεις x.7.d. is said by 
Creon with calm, grave courtesy; they 
have nothing in them of such irony as, 
‘I shall know when you are pleased to 
tell me.’ So Aesch. Z7heb. 260 ET. al- 
τουμένῳ μοι κοῦφον εἰ δοίης τέλος: ‘would 
that thou couldst grant me a light boon.’ 
ΧΟ. λέγοις ἂν ws τάχιστα, Kal τάχ᾽ εἴσο- 
μαι (2.¢ and then I shall know if I can 
serve thee). 

1518 ὅπως πέμψεις : sc. dpa: Xen. 
An. τ. 7. 3 ὅπως οὖν ἔσεσθε ἄνδρες, ‘see 
that ye be’: Plat. Rep. 337 A ὅπως μοι, 
ὦ ἄνθρωπε, μὴ ἐρεῖς. Not (εἶμι ἐπὶ rov- 
τοις), ὅπως κιτ.λ. 

1519 ἀλλὰ θεοῖς γ᾽: ze. ‘Nay, the 
gods, who hate me, will not be displeased 
that I should be thrust forth.’ For the 
synizesis in θεοῖς cp. 215.--ἥκω: cp. 


1357, O. C. 1177 ἔχθιστον ἥκει, has come 
to be most hateful. Creon’s reply, τοι- 
γαροῦν τεύξει τάχα, means: ‘if the gods 
do desire thy banishment, thou wilt soon 
have thy wish’—when the oracle at 
Delphi is consulted (1443). According to 
the story which Soph. follows, Oedipus 
was at first detained at Thebes against 
his own wish. But when some time had 
elapsed, and that wish had given place 
to a calmer mood, the Thebans, in their 
turn, demanded his expulsion ; and Creon 
then yielded (0. C. 433 ff.). 

1520 ἃ μὴ φρονῶ. In the O. C. 
(765 ff.) Creon is represented as oppos- 
ing a distinct refusal to this prayer of 
Oedipus. His words here could mean: 
‘No, I do not promise, for I am not 
wont to speak vain words when I lack 
knowledge’ (φρονῶ as in 569): z.¢., ‘Ican- 
not tell how Apollo may decide.’ But I 
now think that, on the whole, it suits the 
context better to take them as expressing 
consent (ὦ μὴ φρονῶ = what I do not mean 
todo). As this consent can be only pro- 


198 ΣΟΦΟΚΛΕΟΥΣ 


OI. μηδαμῶς ταύτας γ᾽ ἕλῃ μου. KP. πάντα μὴ βούλου 
κρατεῖν' 
καὶ γὰρ ἀκράτησας ov σοι τῷ βίῳ ξυνέσπετο. 


: ὦ πάτρας Θήβης ἔνοικοι, λεύσσετ᾽, Οἰδίπους ὅδε, 
ὃς τὰ κλείν᾽ αἰνίγματ᾽ ἊΣ καὶ κράτιστος ἦν avnp, 1525 
* οὗ τίς οὐ ζήλῳ πολιτῶν “Tats τύχαις “ ἐπέβλεπεν, 
εἰς ὅσον κλύδωνα δεινῆς συμφορᾶς ἐλήλυθεν. 
ὥστε θνητὸν 0 ὄντ᾽ ἐκείνην τὴν τελευταίαν ἰδεῖν 
ἡμέραν ἐπισκοποῦντα μηδέν᾽ ὀλβίζειν, πρὶν ἂν 
δ .\ τέρμα τοῦ βίου περάσῃ μηδὲν ἀλγεινὸν παθών. 1530 


right, though νυν .. νῦν would be quite defensible. 1523 τῷ βίῳ] διὰ βίου Nauck. 
1524. 1580 The ss. rightly give these verses to the Chorus. The Scholiast gives 
them to Oedipus, but thinks that the play would end better with v. 1523: τὰ γὰρ 
ἑξῆς ἀνοίκεια, γνωμολογοῦντος τοῦ Οἰδίποδος. This error arose, as Dindorf points out, 
from the fact that in Eur. Phoen. 1758 ff. Oed. speaks similar verses, of which the 
first two are taken almost verbatim from our passage : :—O πάτρας κλεινῆς πολῖται, 
λεύσσετ᾽, Οἰδίπους ὅδε, | ὃς τὰ κλείν᾽ αἰνίγματ᾽ ἔγνω καὶ μέγιστος ἣν dv_ip.—Fr. Ritter 
would delete vv. 1 524—1530: but the close of the play would then be too abrupt. 
1526 ὅστισ οὐ ζήλῳ πολιτῶν Kal τύχαισ ἐπιβλέπων L. In the later Mss. the only 
variations are ἐν for οὐ (V, M, M® 1st hand), and βίῳ for ζήλῳ (M),—mere blunders. 
Musgrave conjectured, ὃν τίς οὐ ζήλῳ πολιτῶν τῆς τύχης ἐπέβλεπεν; (So Blaydes.) 





visional—depending on the approval of 
Apollo—it is not necessarily inconsistent 
with O. C. 765 ff. 

1522 ἕλῃ pov: cp. 1022 χειρῶν λα- 
βών. 

1524--1580 ὅεε critical note. These 
verses are spoken by the Chorus, as Creon 
turns with Oedipus to enter the house. 
The calm close which the tragedy re- 
quires would be wanting if they were 
spoken by the chief sufferer himself. 
Of extant Greek tragedies, the Prome- 
theus and the Agamemnon are the only 
ones which end with words spoken by 
one of the actors; and in each case this 
is justified by the scheme of the trilogy 
to which the play belonged. 

1525 Here, as elsewhere, the MSs. 
fluctuate between Se and ἤδη. The 
Attic ἤδη, as first pers. sing., is con- 
tracted from ἤδεα : in the ¢hird, the 
classical form was not ἤδη but ἤδει, or, 
before a vowel, ἤδειν (as it must be in 
Eur. /on 1187, Ar. Pax 1182 etc.). No 
3rd sing. in ea, from which ἡ could come, 
6 said, or can be supposed, to have ex- 
isted. Aristarchus, indeed, is quoted by 
the schol. on //. 5. 64 in favour of the η- 
But the Doric 3rd sing. ἀπολώλη in 7ad, 
Heracl. 1. 39 is the only such form which 
is beyond question. Curtius (Verd 11. 


237, Eng. tr. 431 ff.) therefore agrees 
with those textual critics who, like La 
Roche, Cobet, and Kontos (Λόγιος ἡ Ἑρμῆς 
p. 61) would always write the 3rd sing. 
ἤδει (or ἤδειν)., ἥδει αἰνίγματα (slur. 
with reference to the hexameter ἔπη 
in which it was chanted) = knew Ζ7:- 
stincti@gly, by the intuition of genius: in 
Eur. Phoen. 1759 the adapter of this 
verse has altered ἤδει (perhaps by a slip 
of memory) to the more natural but less 
forcible ἔγνω, ‘read aright,’ solved. 
1526 οὗ τίς od ζήλῳ.. ταῖς τύχαις 
ἐπέβλ., ‘on whose fortunes what citizen 
did not look with emulous admiration?’ 
(Cp. Xen. Avero 1. 10 πῶς δὲ πάντες 
ἐζήλουν ἂν τοὺς τυράννους") To me it 
appears certain that we should here read 
the interrogative τίς, with ἐπέβλεπεν in- 
stead of ἐπιβλέπων. Cp. O. C. 1133 ᾧ 
tls οὐκ ἔνι κηλὶς κακῶν ξύνοικος; 871 
ὅπου τίς ὄρνις οὐχὶ κλαγγάνει; Zl. τόρ f. 
τί.. οὐκ... 1... «ἀγγελίας: Eur. Phoen. 878 
ἁγὼ τί δρῶν οὐ, ποῖα δ᾽ οὐ λέγων ἔπη, 
εἰς ἔχθος ἦλθον. Dem. or. 18 § 48 ἐλαυ- 
νομένων καὶ ὑβριζομένων καὶ τί κακὸν οὐχὶ 
πασχόντων πᾶσα ἡ οἰκουμένη μεστὴ γέ- 
γονεν. Then the καί of the mss. should 
probably be ταῖς: though it is possible 
(as Whitelaw proposes) to take {Aw καὶ 
τύχαις as ‘his glory and his fortunes’: 


"Ὁ 


Par 


OIAITOY2 TYPANNOS 199 

OE. Nay, take not these from me! CR. Crave not to be 
master in all things: for the mastery which thou didst win hath 
not followed thee through life. 

CH. Dwellers in our native Thebes, behold, this is Oedipus, 
who knew the famed riddle, and was a man most mighty; on 
whose fortunes what citizen did not gaze with envy? Behold 
into what a stormy sea of dread trouble he hath come! 

Therefore, while our eyes wait to see the destined final day, 
we must call no one happy who is of mortal race, until he hath 
crossed life’s border, free from pain. 


Combining ἐπέβλεπεν with two other conjectures (Martin’s οὗ τις, and Ellendt’s ταῖς for 
kai) Hartung restored, ov τίς οὐ ζήλῳ πολιτων Tals τύχαις ἐπέβλεπεν. Nauck now reads, 
οὗ τίς οὐ ζήλῳ πολιτῶν ἦν τύχαις ἐπιβλέπων (ἦν for καὶ with Enger). Campbell con- 
jectures πρῶτος ἐν ζήλῳ πολιτῶν καὶ τύχαις ἐπιφλέγων, citing a gloss ἐπαιρόμενος (on 
ἐπιβλέπων) which occurs in M (not, however, in E, where on p. 110, which contains 
vv. 1518—1530, there is no gloss). 1528 ἐκείνην] κείνην L tst hand: the initial 
eis from the first corrector (S).—idetvy has been suspected: see comment on 1529. 
1629 In L four words (probably belonging to a gloss) have been erased above μηδέν᾽ 
ὀλβίζειν πρὶν ἄν. Inthe margin the first corrector has written yp. πάντα προσδοκᾶν 
ἕως dv: z.¢., some copies had πάντα προσδοκᾶν ἕως (to which the corrector of L has 
wrongly added ἄν) for μηδέν᾽ ὀλβίζειν πρὶν dév,—a conjecture of the same class as that 


noticed on v. 134. 





cp. AZ. 503 οἵας λατρείας ἀνθ᾽ ὅσου ζήλου 
τρέφει. 1 doubt, however, whether ἐπέ- 
βλεπεν, without ζήλῳ, could mean ‘ad- 
mired.’ On the usage of the verb ém- 
βλέπω, see Appendix. 

1529 The use of ἐπισκοποῦντα is 
peculiar. I take the exact sense to be :— 
‘fixing one’s eye on the final day (as on a 


\, point towards which one is moving), ¢hat 


one should see tt, 1.56. ‘until one shall 
have had experience of it.’ Thus ém- 
σκοπεῖν is used in a sense closely akin 
to its common sense of ‘attentively con- 
sidering’ a thing: and the whole phrase 
is virtually equivalent to, ‘wazting medt- 
tatively to see the final day.’ For the 
added infin., cp. Thuc. 3. 2 γεῶν ποίησιν 
ἐπέμενον τελεσθῆναι, καὶ ὅσα ἐκ τοῦ 1Πόν- 
του ἔδει ἀφικέσθαι. Cp. Plin. 7 § 132 
alius de alio iudicat dies, et tamen supre- 
mus de omnibus, zdeogue nullis creden- 
dum est, Hartung proposed to replace 
ἰδεῖν by ye δεῖ (where ye would be in- 
tolerable); Stanley by ἔδει, Seyffert by 
δέον, and Nauck by χρεών. Kennedy, 
keeping ἰδεῖν, changes ἐκείνην into ἄμει- 
νον. But the infin. ὀλβίζειν as a ‘sen- 
tentious’ imperative (see on 462) is ap- 
propriate in this γνώμη. The accus. 
(θνητὸν ὄντ᾽, ἐπισκοποῦντα) stands with 
the infin. when, as: here, the infin. repre- 
sents an imperat. of the ¢hird person; 


cp. 21. 3. 284 εἰ δέ x’ ᾿Αλέξανδρον κτείνῃ 
ξανθὸς Μενέλαος, | Todas ἔπειθ᾽ Ἑλένην 
καὶ κτήματα πάντ᾽ ἀποδοῦναι, with Leaf’s 
note: and Madvig Gr. ὃ 546. When 
the infin.=an imperat. of the second pers., 
the case is regularly the nom. (Od. 11. 
441), rarely the acc. (Hes. Οὐ. 380). 
The view that ὀλβίζειν depends on ὥστε 
requires a shorter pause at ἐλήλυθεν, and 
thus weakens the effect of v. 1527. 

μηδέν᾽ ὀλβίζειν. Eur. Androm. too fi. 
partly reproduces the language of this 
passage: χρὴ δ᾽ οὔποτ᾽ εἰπεῖν οὐδέν᾽ 
ὄλβιον βροτῶν, | πρὶν ἂν θανόντος τὴν 
τελευταίαν ἴδῃς ] ὅπως περάσας ἡμέραν. 
ἥξει κάτω. He has the thought also 
in 770. 510, Heracl. 866, /. A. 161, 
as Soph. in 77 1 and fr. 588. The 
maxim, ‘Call no man happy before death,’ 
first appears in Greek literature as a 
set γνώμη in Aesch. Ag. 928 ὀλβίσαι 
δὲ χρὴ | βίον τελευτήσαντ᾽ ἐν εὐεστοῖ 
φίλῃ: but Aristotle recognises the popular 
tradition which ascribed it to Solon. 
In Her. 1. 32 Solon says that a man 
may be called εὐτυχής 17: life, but ὄλβιος 
only after a life exempt from reverse. 
Cp. Iuv. 10. 274 f. Zt Croesum, quem 
vox tustt facunda Solonis Respicere ad 
longae iussit spatia ultima vitae, where 
Mayor refers to the proverbs Λυδὸς (Croe- 
Sus) ἀποθνήσκει σοφὸς ἀνήρ, and τέλος dpa 


200 ΘΙ ΔΙΠΟΥΣ 


βίου (Paroemiogr. II. 187, I. 315 n.), and 
to notices of the saying in Cic. (De Fin. 
2 § 87, 3 § 76), Diog. Laert. ᾿ § 50 τὰ 
θρυλούμενα), Ovid (Aer. ΓΞ 135), Seneca 
(De Trang. An. 11 § 12), Jose hus Gee 
dud.1.'5: 11=29 § 3), Arrian 8. 106. 7). 

Lucian (Charon 10): cp. Brees 17: 28. 
Does Solon mean, Aristotle asks, (1) that 
a man zs happy when he is dead? Or 
(2) that, after death, he may be said to 
have been happy? If (1), Arist. declines 
to allow that the dead are positively 


ΤΎΡΑΝΝΟΣ 


happy; and popular opinion, he says, 
denies that they are always negatively so, 
t.e. free from unhappiness. If (2), then 
is it not absurd that at the time when he 
zs happy we are not to call him so? The 
fallacy, he concludes, consists in treating 
‘happiness’ as dependent on bright for- 
tunes: ob yap ἐν ταύταις τὸ εὖ ἢ κακῶς, 
ἀλλὰ προσδεῖται τούτων ὁ ἀνθρώπινος βίος, 
καθάπερ εἴπαμεν, κύριαι δ᾽ εἰσὶν αἱ κατ᾽ 
ἀρετὴν ἐνέργειαι τῆς εὐδαιμονίας, αἱ δ᾽ ἐ- 
ναντίαι τοῦ ἐναντίου Γ΄ (ΕἸ. Nic. 1. 11.) 


APPENDIX. 


The Oedipus Tyrannus at Harvard.—Reference has been made in 
the Introduction (§ 29) to the performance of the Oedipus Tyrannus by 
members of Harvard University in May, 1881. The thorough scholar- 
ship, the archeological knowledge and the artistic skill which presided 
over that performance invest the record of it with a permanent value 
for every student of the play. Where the modern imagination most 
needs assistance, this record comes to its aid. Details of stage- 
management and of scenic effect, which a mere reading of the text 
could suggest to few, become clear and vivid. Mr H. Norman’s 
‘Account of the Harvard Greek Play ’—illustrated by excellent photo- 
graphs—is, in fact, a book which must always have a place of its own 
in the literature of the Oedipus Tyrannus. 1 select those passages 
which relate to the principal moments of the action; and, for more 
convenient reference, I arrange them in successive sections. 


§ 1. Opening Scene. ‘Account,’ p. 65. ‘The scene behind the 
long and narrow stage is the palace of Oedipus, king of Thebes,—a 
stately building with its frieze and columns. There is a large central 
door with two broad steps, and two smaller side doors; all three are 
closed. In the centre of the stage in front is a large altar; beside each 
of the smaller doors of the palace is another altar. A flight of steps 
leads from the stage at each side. ‘The sound of the closing doors has 
warned the audience that the long-expected moment is at hand, and an 
immediate silence ensues. Under these circumstances the first notes of 
the orchestra come with great effect, and the entire prelude is unusually 
impressive. As it closes, the spectators are sympathetic and expectant. 

‘Slowly the crimson curtains on the right-hand side below the stage 
are drawn apart, and the Priest of Zeus enters, leaning on a staff, a 
venerable and striking figure....Behind him come two little children. 
They are dressed in soft white tunics and cloaks, their hair is bound 
with white fillets, and they carry in their hands olive branches twined 
with wool,— 

ἐλαίας θ᾽ ὑψιγέννητον κλάδον, 
λήνει μεγίστῳ σωφρόνως ἐστεμμένον. 


202 ALLENIDIX. 


This shows that they come as suppliants. Behind the children come 
boys, then youths, and then old men. All are dressed in white and 
carry suppliant boughs ; in the costumes of the men, the delicate fabric 
of the undergarment, the χιτών, contrasts beautifully with the heavy 
folds of the ἱμάτιον. With grave, attentive faces the procession crosses 
the front of the stage, and mounts the steps; the suppliants lay down 
their branches and seat themselves on the steps of the altars. The 
priest alone remains standing, facing the palace door. 

‘ The first impression upon the spectators was fortunate. The inno- 
cent looks of the children, the handsome figures of the men, the 
simplicity and solemnity of their movements, set off as they were by 
the fine drapery of their garments and the striking groups around the 
altars, had an instant and deep effect. It is safe to say that fears of 
crudeness or failure began rapidly to vanish. The spectacle presented 
at this moment was one of the most impressive of the play. 

‘After a short pause the great doors of the palace are thrown back, 
and the attendants of Oedipus enter and take up their positions on 
each side. They wear thin lavender tunics reaching nearly to the knee. 
Their looks are directed to the interior of the palace, whence, in a 
moment, Oedipus enters. His royal robes gleam now with the purple 
of silk and now with the red of gold; gold embroidery glitters on his 
crimson tunic and on his white sandals; his crown gives him dignity 
and height. 

‘For an instant he surveys the suppliants, and then addresses them.’ 


§ 2. Arrival of Creon from Delphi: verses 78 ff. ‘ Account,’ p. 69. 
‘While Oedipus is speaking, the children on the [spectators’] left of the 
stage have descried some one approaching, and one of them has pointed 
him out to the priest. It is Creon, who enters with rapid strides, 
wearing a wreath of bay leaves sparkling with berries, the symbol of a 
favorable answer. He is dressed in the. short salmon-colored tunic and 
crimson cloak, with hat and staff. A hasty greeting follows; and 
Oedipus, the priest, and the suppliants wait for the answer of the 
oracle.’ 


§ 3. Withdrawal of the Suppliants, and Entrance of the Chorus: 
vv. 143-151, p. 71. ‘With the assurance of speedy aid [for the The- 
bans] he [Oedipus] leads Creon into the palace, and the attendants 
follow and close the doors. Slowly the white-robed suppliants rise ; the 
petition being granted, each one takes his bough, and led by the priest 
they descend the steps and disappear. 

‘As the last figure passes out of sight the notes of the orchestra are 
heard once more, this time with a measured beat which instantly attracts 
attention, and the Chorus of old men of Thebes issues from the same 
entrance. They are men of various ages, dressed in tunics reaching to 
the instep, and full ἱμάτια, of harmonious soft. warm colors. The excel- 
lence of the costumes was marked ; each man seemed to have worn his 
dress for years, and to exhibit his individuality in the folds of it. They 
enter three deep, marching to the solemn beat of the music ; and as the 


APPENDIX. 203 


first rank comes in sight of the audience the strains of the choral ode 
burst from their lips. 


Oise sss ee Fae = er 


Shoulder to shoulder and foot to foot the old men make their way to 
the altar on the floor of the theatre and take up their positions around 
it. This entrance of the Chorus was surpassed in dramatic effect by 
few features of the play: the rhythmical movements, the coloring and 
drapery, the dignity of the faces, the impressive music sung in unison by 
the fifteen trained voices,—all these combined to produce a startling 
effect on the audience.’ 








§ 4. Lntrance of Teiresias, v. 297, p. 75. ‘Atthis moment Teiresias 
enters, a towering venerable figure, with long white hair and beard. He 
is guided to the stage by a boy, whose blue cloak contrasts with the 
snowy draperies of the old man.’ zs exit, v. 462, Ὁ; 79. ‘The two 
men part in deadly anger, Oedipus going within the palace and the boy 
leading Teiresias down the steps [from the stage, see ὃ 1]....Once more 
the music sounds, and the Chorus gives voice to its feelings concerning 
the strange scene which has just been enacted.’ 


§ 5. Lxtrance of Creon, when he comes to repudiate the charge of 
treason brought agatnst him by Oedipus: v. 512, p. 81. ‘As the strains 
of [choral] music die away, Creon is seen hastily ascending the steps [to 
the stage] on the right [of the spectators: cp. § 2]. He is no longer 
dressed as a traveller, but in garments suited to his high rank. His 
tunic is of delicate dark crimson material, with a gold border; his 
ἱμάτιον is of bright crimson cashmere, with a broader gold border ; his 
sandals are of crimson and gold. He strides to the centre of the stage 
and bursts out in indignant denial of the charges that Oedipus has made 
against him.’ 


§ 6. Jocasta enters while high words are passing between Oedipus and 
Creon: v. 631, p. 83. ‘Just as this [altercation] reaches its height the 
doors of the palace are seen to open, and the Chorus bids both angry 
speakers cease, as Jocasta is approaching. ‘The attendants of Jocasta 
enter and place themselves on each side of the door, and a moment 
later the queen herself stands upon the threshold. Oedipus turns to her 
with welcome, and Creon with a gesture of appeal. 

- ‘Her dress consists of a richly trimmed silvery undergarment, and an 
ἱμάτιον of crimped pale yellow silk. She wears a crown, bracelets, and 
necklace, and white sandals embroidered with gold.’ 

It was upon this group—the first complex one in the play—that Mr 
F. D. Millet based his scheme of the costumes, to which he gave long 
study, both from the historical and from the artistic point of view, and 
which he has described in the Century Magazine of Nov., 1881. 
From this article, Mr Norman (p. 83) quotes the following passage :— 


204 APPENDIX. 


‘It was part of the original scheme that in each group the most 
prominent character should, as far as possible, be the focus, not only of 
interest in the text, but from the point of view of costume. Let us see 
how the first complex group fulfilled this condition. On the stage left 
stood Oedipus, in rich but deep-toned red; on the right, Creon, equally 
in red, but of a color entirely different in scale; the attendants of the 
king, in lavender tunics bordered with gold-embroidered white, flanked 
the doorway ; and the two attendants of Jocasta, in delicate blue and 
salmon, brought the eye by a pleasing graduation in intensity of color 
and strength of tone up to the figure of the queen, clothed in lustrous 
and ample drapery.’ 


$7. Arrival of the Messenger from Corinth: v. 924, Ὁ. 89. ‘As 
the Chorus closes, Jocasta enters [v. g11] in a new state of mind. She 
has comforted Oedipus by ridiculing all oracles ; but she is not without 
faith in the power of Gods, and she brings frankincense and garlands, 
and lays them with a prayer upon the altar. 

‘While she is speaking, an old man has entered on the left below the 
stage. He is dressed as a common traveller, in a tunic and short cloak, 
his hat slung over his shoulder, and a stout staff in his hand. It is the 
messenger from Corinth. He looks round as if in search of something, 
and as soon as the queen has finished her prayer he inquires of the 
Chorus where the home of Oedipus, or, better still, the king himself, can 
be found. He is promptly informed that the mansion he sees is the 
palace of Oedipus, and that the lady before it is the queen. Witha 
profound salutation as he ascends to the stage, he declares himself to 
be the bearer of news at once good and bad. Old Polybus, king of 
Corinth, is dead, and the citizens are about to make Oedipus king. 
This is indeed news to Jocasta. Oedipus has long avoided Corinth lest 
he should slay his father, Polybus ; now he can return, as king, all fear 
dispelled. Oedipus enters in response to her summons. His royal 
robes have been exchanged for simpler ones of white and gold. He, 
too, learns the news with triumph.’ 


8 8. Locasta divines the worst:—her final exit; vv. 1040—1072, 
Ρ. 92. ‘But Jocasta? At the other end of the stage the queen is writhing 
in anguish. The deep-red cloak which she wears is twisted about her ; 
now she flings her hands up and seems about to speak, then her hands 
are pressed on her mouth to stop the cries which rise, or on her bosom 
to silence the beating of her heart. She rushes toward the king, but 
stops half-way; her face shows the tortures of her soul. ‘The truth is all 
too clear to her. The spectator feels that this suspense cannot last, and 
relief comes when the Chorus suggests that perhaps Jocasta can tell 
something about the shepherd of Laius. When appealed to by Oedipus, 
she forces the suffering from her face and turns with a smile. But 
Oedipus has gone beyond recall. Her last appealing words are scorned, 
and with the language and the gesture of despair she rushes from the 
stage.’ 


APPENDIX. 205 


89. Zhe Herdsman of Laius is brought in: the whole truth ts ex- 
torted from him: vv. 1110—1185, pp. 94 ff. ‘As the music ceases the 
attendants of Oedipus appear at the entrance on the right, supporting a 
strange figure between them. It is an aged man, with grizzled hair and 
beard, clothed in coarse homespun cloth, and with a rough, untanned 
sheepskin over his shoulders. He supports himself on a sapling staff 
which he has cut in the woods. He mounts the steps with difficulty, and 
faces the king. He is no stranger to the errand on which he has been 
brought, and with the greatest difficulty he is made to speak. The 
contrast between the eagerness of the messenger from Corinth to tell all 
he knows, and the silence of the tender-hearted old shepherd, is very 
striking. The shepherd cannot bear the other’s telltale chatter, and 
with the words, “Confusion seize thee and thine evil tongue!” he swings 
his staff to strike him. Ata gesture from Oedipus the attendant stops 
the blow. The old man must be made to speak. The muscular 
attendants spring forward and seize him. Then the truth is wrung 
from him, word by word. He gave the child to the Corinthian; it 
came from the palace; they said it was the son of Laius; Queen 
Jocasta herself placed it in his hands; they said that an oracle 
had declared that it should kill its father. The truth is out; the 
oracles are not falsified; his father’s murderer, his mother’s husband, 
Oedipus faces his doom. With a fearful, choking cry he pulls his 
robes over his head and face, and bursts into the palace. 

‘This scene...was the dramatic climax of the play. The acting led 
up to it gradually by the excited conversation and the shepherd’s blow. 
When Oedipus burst through the doors of the palace, his attendants 
quickly followed him; the horror-stricken messengers turned with 
despairing gestures and descended the steps, the one to the right, the 
other to the left, and a profound silence fell upon the theatre.’ 


§ 10. Lffect of the fourth stasimon, vv. 1223—1530, p. 98. ‘In the 
opening strains of the last choral ode, which now ring out, the emotions 
of the scene are wonderfully expressed. Each one recognizes the 
solemnity and depth of his own feelings in their pathetic tones.’ 











S11. Zhe Messenger from the House: the entrance of the blinded 
Oedipus, 1223—1296, pp. 98f. ‘As the ode [just mentioned] closes, the 
palace doors are opened violently from within, and the second messenger 
rushes on the stage. He is a servant from the palace, clad, like the at- 
tendants, in a short light tunic. He brings a tale of horror: Oedipus, 
on entering, had called for a sword, and demanded to know where 
Jocasta was. No one would tell him; but at last, seeing the doors of the. 
bedchamber shut, he had broken through them and disclosed the body: 
of the queen hanging by the bed. Tearing down the body, he had. 


206 APPENDIX. 


snatched from the shoulders the golden clasps and had thrust them into 
his eyes.’...‘In a moment Oedipus himself appears, leaning on his at- 
tendants, his pale face marred by bloody stains. The dismayed Chorus 
hide their faces in their robes, and the king’s voice is broken with sobs 
as he cries, αἰαῖ, αἰαῖ, δύστανος ἐγώ. 


8.12. Closing scene, vv. 1416—1530, pp. ror ff. ‘As Oedipus is 
begging to be slain or thrust out of the land, the approach of Creon, who 
has resumed his royal powers, is announced. The memory of all his 
injustice to Creon overwhelms Oedipus, and he cannot bear to meet 
him. But he is blind and unable to flee, so he hides his face and waits 
in silence. Creon enters, crowned, followed by two attendants....His 
first words are reassuring ; the new king does not come with mocking or 
reproach, but directs that a sight so offensive to earth and heaven be 
hidden within the palace. Oedipus asks the boon of banishment, but is 
informed by the cautious Creon that the God must be consulted. Then 
the blind man begs that his wife be buried decently, and reiterates his 
prayer that he may be permitted to leave the city which he has afflicted. 
And one thing more he asks,—that he may embrace his daughters again. 
By asign Creon despatches his own attendants to bring them, and while 
Oedipus is still speaking their voices are heard. 

‘Antigone and Ismene now enter, led by the attendants of Creon, 
and are placed in the arms of Oedipus, who falls on his knees beside 
them, and addresses them with saddest words. The children are too 
young to appreciate the horror of the scene, but they are filled with pity 
for their father’s pain. There is a look of genuine sympathy on the two 
bright faces which watch the kneeling figure. Creon has retired to the 
right of the stage and has wrapped his robe round him, unable to bear 
the sight of the terrible farewell. He is summoned by Oedipus to give 
his hand in token of his promise to care for the helpless girls. ‘The 
children fall back, the blind man waits with outstretched hand, and 
Creon slowly and sadly walks across the stage and gives the sign. Then 
Oedipus turns again to his little ones. The painful scene, however, has 
lasted long enough, and Creon orders Oedipus to leave his children and 
withdraw. It is a dreadful separation, but the king’s order is impera- 
tive. So Oedipus tears himself away, his attendants throw open the 
doors, the attendants of Creon take the children by the hand, and Creon 
himself leads Oedipus up the steps and into the palace....The children 
and the second messenger follow ; the attendants of Oedipus enter last 
and gently close the doors. 

‘The music sounds again in pathetic tones, and the Coryphaeus 
expresses for his fellows the lesson of life.’ 


Verse 2. On the meaning of θοάζετε. The points of the question 
are these. 1. θοάζειν, from θο-ό-ς swift (rt. Ger, Oém; Curt. Ltym. 
5. 313), Occurs ten times in Eur., four times transitively, ‘to impel,’ 
‘urge,’ as Bacch. 66 θοάζω Βρομίῳ, πόνον ἡδύν: six times intransitively, as 
Troad. 349 μαινὰς θοάζουσ΄. If it is the same word here, what would 
Goalev ἕδρας mean? (a) Not, I think, ‘to urge, press your supplication,’ 


APPENDIX. 207 


—referring to the eager gestures or aspect of the suppliants: for rapt 
motion, and not merely eagerness, is implied by doafw. Rather (ὁ) ‘to 
come with eager haste as suppliants’: as Herm. explains Erfurdt’s ‘ cur 
hanc sessionem festinatis?’—‘ cur tanto studio hic sessum venitis ?’ 
Now I can conceive Sophocles saying σπεύδειν or ἐπείγειν or even θοάζειν 
ἱκετείαν : but could he have said Ooalew é5pas? The primary notion of 
a fixed attitude stands out too clearly above the secondary notion of 
a supplication. 

2. For another θοάζειν, ‘to sit,’ only two passages are cited. (1) Em- 
pedocles 52 θάρσει καὶ τότε δὴ σοφίης ἐπ᾽ ἄκροισι θόαζε. This might 
mean ‘hasten on to the heights of wisdom’: though, when ἐπί with dat. 
denotes motion, it usually means ‘against,’ as in Qd. το. 214 οὐδ᾽ οἵ γ᾽ 
ὡρμήθησαν ἐπ᾽ ἀνδράσιν. But the more natural sense would be, ‘sit on 
the heights of wisdom.’ (ii) Aesch. Suppl. 595 ὑπ᾽ ἀρχᾶς [L ἀρχὰς] 
δ᾽ οὔτινος θοάζων | TO 'μεῖον κρεισσόνων κρατύνει" οὔτινος ἄνωθεν ἡμένου 
σέβει κάτω. Hermann renders the first words: ‘hasting at no one’s 
bidding,’ zullius sub tmperio properans. So Mr Paley: ‘ Himself uzged 
to action (θοάζων) by no authority.’ But the Scholiast is right, I believe, 
in rendering θοάζων by καθήμενος. Only ὑπ᾽ ἀρχᾶς οὔτινος θοάζων does 
not mean ‘sitting uw#der no other’s rule,’ but ‘sitting dy no other’s 
mandate.’ (I should prefer ὕπαρχος) For the Aeschylean image of 
Zeus throned on high, cp. Aesch. Agam. 182 δαιμόνων δέ που χάρις } 
βιαίως σέλμα σεμνὸν ἡμένων. 

3. Ancient tradition recognised θοάζειν as=Odooew here. Plut. 
Mor. 22 E says, τῷ θοάζειν ἢ τὸ κινεῖσθαι σημαίνουσιν, ws Εὐριπίδης... 
ἢ τὸ καθέζεσθαι καὶ θαάσσειν, ὡς SopoxAys,—quoting this passage. So 
the Φνηι. Magn. 460. 10 διὰ τί προσθακεῖτε τάσδε τὰς ἕδρας ; τί 
προσχρήζετε ταύταις ταῖς ἕδραις ; If ἢ had stood before τί the last clause 
would have seemed to glance at the other explanation. So the Schol. 
θοάζετε, κατὰ διάλυσιν ἀντὶ τοῦ θάσσετε" but adds, ἢ θοῶς προσκάθησθε. 

4. Buttmann would connect θοάζω ¢o sit with Oe, the stem of τίθημι. 
θοάζω cannot be obtained directly from Oe. It is possible, however, that 
a noun-stem, from which θοάζω fo sit came, may itself have been 
derived from a secondary form of @e. It might be said that θαα-, θοω-, 
suggest a OeF or OaF or θυ akin to Oe: cp. dav (πιφαύσκω) with ¢a, 
στυ (στῦλος) with ora. 

5. To sum up:—Emped., Aesch. and Soph. seem to have used 
θοάζξειν as=Oaccew. We can only say that (i) the sound and form 
of θοάζω may have suggested an affinity with θαάσσω, θόωκος : (ii) as 
a purely poetical word, θοάζω belonged to that region of language in 
which the earlier Attic poets—bold manipulators of old material—used 
a certain license of experiment, not checked by scientific etymology, 
and so liable to be occasionally misled by false or accidental analogies. 


44 f. In discussing these two verses, it is essential that the whole 
context from v. 35 should be kept clearly before the mind :— 


35 ὅς γ᾽ ἐξέλυσας, ἄστυ Καδμεῖον μολών, | 
σκληρᾶς ἀοιδοῦ δασμὸν ὃν παρείχομεν" 


208 APPENDIX. 


καὶ ταῦθ᾽ ὑφ᾽ ἡμῶν οὐδὲν ἐξειδὼς πλέον 
οὐδ᾽ ἐκδιδαχθείς, ἀλλὰ προσθήκῃ θεοῦ 
λέγει νομίζει. θ᾽ ἡμὶν ὀρθῶσαι Biov: 

40 νῦν τ᾽, ὦ κράτιστον πᾶσιν Οἰδίπου κάρα, 
ἱκετεύομέν σε πάντες οἶδε πρόστροποι 
ἀλκήν τιν᾽ εὑρεῖν ἡμῖν, εἴτε τοῦ θεῶν 
φήμην ἀκούσας εἴτ᾽ ἀπ᾽ “ἀνδρὸς οἶσθα που" 
ὡς τοῖσιν ἐμπείροισι καὶ τὰς ξυμφορὰς 

45 ζώσας ὁρῶ μάλιστα τῶν βουλευμάτων. 


The general sense is: ‘Thou didst save us from the Sphinx; and 
now we pray thee to save us from the plague: for, when men are 
experienced, we see that they are also (xa/) most successful in giving 
counsel.’ The last two verses form a comment on the whole preceding 
sentence. The complaint that, thus understood, they involve ‘bathos’ 
is doubly unjust. For, even if the trouble which Oedipus is now asked 
to heal had been precisely similar to the trouble which he had formerly 
healed, yet the general sentiment, ‘ Experience teaches prudence,’ is no 
more ‘bathos’ than is δράσαντι παθεῖν, παθήματα μαθήματα, or many 
other maxims which occur in Greek Tragedy. But in this case the new 
trouble was of a different order from the old; and the definition of the 
old trouble, given in 35 f., naturally suggests a supplementary thought 
which lends a special force to the yvwun. The experience of a great 
national crisis will stand Oedipus in good stead, though the problem 
now presented to him is unlike that which he formerly solved. 

The old scholium on v. 44 in the Laurentian ms. runs thus :—ws 
τοῖσιν ἐμπείροισιν: ἐν τοῖς συνετοῖς τὰς συντυχίας Kal Tas ἀποβάσεις 
τῶν βουλευμάτων ὁρῶ ζώσας καὶ οὐκ ἀπολλυμένας. οὐ σφάλλεται ἀλλὰ 
τὸ ἀποβησόμενον στοχάζεται καλῶς. Prof. Kennedy calls this ‘the poor 
gloss of a medieval scholiast.’ The scribe was medieval; but the gloss? 
The age and origin of the old scholia in L have been discussed by Wunder, 
G. Wolff, O. Pauli, and others, with results of which I have given an out- 
line in the second part of the Introduction to the Facsimile of the Lauren- 
tian MS. (p. 21). These old scholia represent, in the main, the work of 
the Alexandrian scholars, and more especially of two commentators, one 
of whom is unknown, the other being the famous grammarian Didymus, 
who flourished crc. 30 B.c. The other interpreters from whose com- 
ments these scholia were compiled belonged chiefly to the period from 
about 250 B.c. down to the age of Didymus. There is nothing in this 
scholium on v. 44 to suggest a ‘medieval’ rather than an Alexandrian 
origin ; while on the other hand there are definite reasons for believing 
that, like the rest of the old scholia, it represents an explanation which 
had been handed down, through successive generations of Alexandrian 
scholars, from an age when the feeling for classical Greek idiom was 
still fresh. 

The interpretation thus sanctioned by the Greek commentary has 
been accepted by the all but unanimous judgment of modern critics. 
We may here state, and answer, the chief objection which has recently 
been made to it. 


APPENDIX. 209 


It is said that ξυμφορά cannot mean ‘issue’ or ‘outcome’; and that, 
therefore, τὰς ξυμφορὰς τῶν βουλευμάτων cannot mean ‘the issues of their 
counsels.’ The answer is that the phrase, ‘the issues of their counsels,’ 
is only a convenient way of saying, ‘the occurrences connected with 
their counsels’; z.e., in this particular case, ‘the occurrences which 
result from their counsels.’ No one has contended that the word ξυμ- 
φορά, taken by itself, could mean ‘outcome’ or ‘issue. The fallacious 
objection has arisen from the objectors failing to distinguish between 
the use of the English genitive and the much larger and more varied 
use of the Greek genitive. We could not say, ‘the occurrences’ (meaning 
‘consequences ’) ‘of their counsels.’ But our ‘of’ is not an exhaustive 
equivalent for the force of the Greek genitive. ξυμφοραὶ βουλευμάτων, 
‘occurrences connected with, belonging to, counsels,’ could mean, 
according to context, that the occurrences (4) consist of the counsels, 
(6) accompany them, (¢) result from them. It would be just as reason- 
able to object to the phrase Avypwv πόνων ἱκτῆρες at v. 185, because 
‘suppliants of weary woes’ would be unintelligible. The ancient Greek 
commentator has explained the phrase, τὰς ξυμφορὰς τῶν βουλευμάτων, 
with a precision which could not have been happier if he had foreseen 
the objection which we have been noticing; and those who raise that 
objection might have profited by attention to his language. In his 
paraphrase, tas συντυχίας καὶ tas ἀποβάσεις τῶν βουλευμάτων, the first: 
word, συντυχίας, marks that ξυμφοράς bears its ordinary sense: the 
second word, ἀποβάσεις, marks that the relation expressed by the gent- 
tive case is here the relation of cause to effect. It is as if he had 
said: ‘the occurrences connected with—that is (καί), the results of— 
the counsels.’ Similarly in O. C. 1506, καί σοι θεῶν | τύχην τις ἐσθλὴν 
τῆσδ᾽ ἔθηκε τῆς ὁδοῦ, ‘a good fortune connected with this coming,’ means 
‘a good fortune which this coming bestows.’ There, as it happens, we 
can say simply, ‘the good fortune of this coming’: but we might say also, 
‘a happy issue from this coming,’—and that, too, without fear of being 
supposed to think that τύχη means the same thing as τελευτή. In Thuc. 
I. 140 (quoted in my commentary) tas ξυμφορὰς τῶν πραγμάτων is a 
phrase strictly parallel to τὰς ξυμφορὰς τῶν βουλευμάτων. That is, the 
genitive is a genitive of connection; the phrase means literally, ‘the 
occurrences connected with human affairs,’ ze, the ways in which 
human affairs turn out; and therefore we may accurately render, 
‘the issues of human affairs.’ Prof. Kennedy renders it, ‘the course 
of actual events,’ and says that the genitive ‘is attributive or descrip- 
tive, not possessive.’ This is not very clear; but the translation in- 
dicates that he takes the gen. to be descriptive; so that the phrase 
would mean literally, ‘the ξυμφοραί consisting in πράγματα. Such a 
phrase, though oddly expressed, would be intelligible if the course of 
events in real life was being opposed to the course of events in a poem 
or other work of fiction. But it is inadmissible in Thuc. I. 140, 
where the comparison is not between real and imaginary ξυμφοραί, but 
between the incalculable conjunctures of outward circumstances and the 
incalculable caprices of human thought: ἐνδέχεται yap τὰς ξυμφορὰς 
TOV πραγμάτων οὐχ ἧσσον ἀμαθῶς χωρῆσαι ἢ καὶ Tas διανοίας τοῦ ἀνθρώπου. 


ἘΠῊΝ τς Ι4᾽ 


219 APPEN DEX. 


Before leaving this topic, it may be well to say a word on the choice 
of the word ‘issues,’ employed in my translation. In my first edition, 
commenting on tas ξυμφορὰς τῶν βουλευμάτων, I had said, ‘the events, 
issues, of their counsels.’ On this Prof. Kennedy remarks, ‘he seems 
to confuse the words events and zssues, as if they were identical.’ 
A little before, the critic states what he himself regards as the distinction 
between them :— 

‘Etymologically they are much the same, both meaning ozt-come; event from 
evenire, issue from exire. Both can be used in the sense of exding: as ‘the event 
(or the issue) of the battle of Tel-el-Kebir was the defeat of Arabi.’ But we could 
not say, ‘the event of the battle was the surrender of Cairo,’ though we might say 


‘the issue’ &c. In short, event may not be used in the sense of ‘result’ or ‘conse- 
quence’; zsswe may be so used.’ 


The statement that ‘event’ cannot be used in the sense of ‘result or 
consequence’ is surprising. ‘The first two meanings given by Dr 
Johnson to ‘event’ are (1) ‘incident ; anything that happens’: (2) ‘con- 
sequence of an action; conclusion; upshot.’ So Webster defines 
‘event,’ first, as ‘incident,’ secondly as ‘the consequence of any thing ; 
the issue,’ etc. Nor is there the least warrant for saying that ‘event’ 
can denote only an immediate consequence, while ‘issue’ can denote 
also an ulterior consequence. See, σὺ Richard If. 2. 1. 212: 

‘What will ensue hereof, there’s none can tell; 


But by bad courses may be understood 
That their evens can never fall out good.’ 


Ghakeencare would probably have been surprised to learn that he 
ought to have written ‘issues.’ And Tennyson was doubtless unconscious 
of a blunder in the words, 

‘One God, one law, one element, 


And one far-off divine event 
To which the whole creation moves.’ 


‘Event’ and ‘issue,’ both alike, can mean either ‘ending’ (as victory 
15 ἘΠ6 Ὁ event,’ ‘issue,’ of a battle), or ‘consequence.’ The second sense 
belongs to ‘event’ by precisely the same right as to ‘issue’ (exdfus): 
cp. Cicero Juv. 1. 28. 42 eventus est alicuius exitus negotit, in quo quaert 
solet, guid ex quaque re evenerit, eveniat, eventurum sit. The distinction 
in our usage at the present day is simply this. ‘Event’ has become 
familiar in the sense of ‘incident,’ and unfamiliar in the sense of ‘ out- 
come,’ except in certain phrases, such as ‘the event will show,’ etc. 
Hence to say, for instance, ‘the events of human affairs,’ would have 
an awkward sound now; though it is just as correct, and could bear 
exactly the same sense, as ‘the issues of human affairs.’ One cause is 
manifest. We have a verb, ‘to issue,’ but no verb, to ‘evene’; and, 
through saying, ‘the affair issued in that,’ it has become natural to say 
‘the issue’ (rather than ‘the event’) ‘of the affair.’ 

It is this shade of contemporary preference, and no other reason, 
which has guided my use of the words ‘issue’ and ‘event’ in the note 
on wv. 44 f. (p. 18). I have used ‘issue’ in the sense of ‘ outcome,’ 
and ‘event’ only 1 in the sense of ‘occurrence.’ But, when ‘event’ does 
mean ‘outcome,’ then it is synonymous with ‘issue.’ Prof. Kennedy’s 


APPENDIX. 211 


assertion that ‘event’ can mean only (1) ‘occurrence’ or (2) ‘ending,’ 
while ‘issue’ can mean either of these, and also (3) ‘consequence,’ seems 
to have no foundation either in the history of the words or in the usage 
of the best English writers. 

The first modern writer who dissented from the traditional interpre- 
tation was John Young, who held the Chair of Greek at Glasgow from 
1774 to 1821. He rendered ξυμφοράς by collationes, taking the sense 
to be: ‘I see that with men of experience comparisons of counsels also 
are most in use’: ze, such men are not only fitted to be counsellors, 
but are also ready to consult other men. ‘Thus understood, the two 
verses are no longer a comment on the whole preceding sentence ; they 
refer to the latter part of v. 43, εἴτ᾽ am ἀνδρὸς οἶσθά που A view 
identical with Young’s was expressed by Dr Kennedy in 1854, and is 
maintained in his edition. He renders thus :— 

‘ws since τοῖσιν ἐμπείροισιν fo men of experience ὁρῶ 7) see that (not 
only counselling but) καὶ also ras ξυμφορὰς τῶν βουλευμάτων comparisons 
of their counsels μάλιστα ζώσας are in most lively use.’ 

In a note on tas ξυμφορὰς τῶν πραγμάτων (Thuc. 1. 140 ὃ 3) 
Shilleto wrote thus :— 

‘Interpreting here (see ὃ 1) “events, issues, results,” I disagree with 


1 John Young, a very acute and accomplished scholar—known to many by 
his fine criticism on Gray’s Elegy—published nothing on Sophocles. His note on 
O. 7. 44 f. was communicated to Andrew Dalzell, Professor of Greek in the Uni- 
versity of Edinburgh. In 1797 Dalzell published the second volume of his Codlectanea 
Graeca Mavora, containing extracts from poets, as the first volume had contained 
prose extracts. Young’s note does not appear in the edition of 1797, which on v. 44 
gives only Brunck’s note (as below). The book went through several editions. The 
edition of 1822 was revised by Dalzell’s successor in the Greek Chair, George Dunbar, 
who added some comments of his own. There the note on v. 44 stands as follows :— 

“44. ‘Qs τοῖσιν ἐμπείροισι---Ἴ Osu enim peritis video felict quogue eventu consilia 
maximé vigere. BRUNCK. Ita interpretes: sed συμφόραν (sic) pro eventu consilit 
sumi posse non credo; ea enim vox fortuitum aliquid semper innuere videtur: hic 
autem potius in primitivo sensu sumi, locusque adeo totus ita reddi potest: Szcudi 
alicujus deorum vocem audisti, vel etiam ἃ mortalium quocunque quicquam acceperis ; 
video enim apud prudentes expertosque viros etiam collationes constlii maxime in usu 
esse. Ipsius sapientiam supra laudaverat ; iam etiam alios consultdsse posse addit: 
qui sensus vulgato multd melior videtur; otiosum enim alias foret καὶ, neque tota 
sententia loco suo digna. T. Y. Esto ut ξυμφορὰ aliquid fortuiti semper innuit (szc). 
Hoc ipsum est quod quaerimus. Sensus loci esse videtur Safientes Fortuna iuvat. 
_ Cantab, Anon. *Vix credere possum τὰς ξυμφορὰς τῶν βουλευμάτων significare 
collationes consilii. Sensus videtur esse; vzdeo enim apud expertos eventus consiliorum 
maxime vigere, i.e. Ex eventu consiliorum quae prius dederant facilius et rectius de 
futuro iudicare possunt.’ 

The last note, with an asterisk prefixed, is Dunbar’s own. In the initials appended 
to Young’s note, ‘T.’ is a misprint for ‘J.’ (Another obvious misprint, viz. ‘innuit’ 
for ‘innuat,’ closely follows it.) It was very natural that Dr Kennedy should have 
thought this better authority than my statement, and should have continued to speak 
of ‘Dr T. Young.’ (John Young took no degree beyond that of M.A.) But I do not 
know what ground my eminent critic had for saying that Young’s view was ‘accepted 
by Prof. Dalzell.’ The mere printing of Young’s note, along with two others of a 
different tendency, can scarcely be held to prove it. And the fact that Brunck’s note 
is still placed first (as in the ed. of 1797) rather suggests the contrary. Dunbar, it will 
be noticed, records his dissent from Young.—I have to thank my colleagué, the Rev. 
hoe W. P. Dickson, for access to Dunbar’s ed. of Dalzell,—now a somewhat rare 

ook. 


14.:- 


212 APPENDIX. 


such rendering of Soph. Oed. T. 44 ws τοῖσιν ἐμπείροισι καὶ τὰς ξυμφορὰς | 
ζώσας ὁρῶ μάλιστα τῶν βουλευμάτων. 1 have long thought that ‘com- 
parisons of counsels’ was there meant and have compared Aéschyl. Pers. 
528 quoted above on 128, 9. (1 am rejoiced to find that Prof. Kennedy 
and I have independently arrived at the same conclusion. See Journal 
of Philology, Vol. 1. pp. 311, 312.) καὶ seems thus to have more signi- 
ficance. Men of experience may receive suggestions from not only 
gods but from other men (εἴτ᾽ az’ ἀνδρὸς οἶσθά που). Collations also of 
counsels are most effective. It is not improbable that Sophocles had 
in view the adage σύν te δύ᾽ ἐρχομένω καί τε πρὸ ὃ τοῦ ἐνόησεν Hom. 
ΤΠ ΠΧ: 227. 

It will be seen that Mr Shilleto agreed with Professor Kennedy in 
taking ξυμφορᾶς as = ‘comparisons,’ but differed from him (1) in taking 
Cucas—as I do—to mean ‘effective,’ not ‘in vogue’ (an old schol. in L 
has ζώσας, ἀντὶ τοῦ ἐνεργεστέρας): (2) in taking the καὶ (‘also’) to 
imply ‘independently of hints from the gods,’ and not ‘in addition to 
offering counsels.’ 

Mr Whitelaw, too, agrees with Dr Kennedy about ξυμφορας, but not 
about ζώσας, which he takes to mean ‘prospering.’ ‘Conference also of 
counsels prospers for men of experience more than others.’ Remark 
that this version makes tas ξυμφορὰς τῶν βουλευμάτων equivalent to τὸ 
ξυμφέρειν ta βουλεύματα. It is this act that prospers for them. 

Dr Fennell now renders (Zvans. Camb. Phil. Soc., 1886, p. 72), 
‘since I see that with men of experience their co//ections of counsels (ze. 
the counsels which they bring together) are also (as well as a φήμη θεοῦ) 
most of all living.’ Thus ζώσας is virtually the epithet of the counsels, 
since tas & tov B. is taken=rTa ξυμφερόμενα βουλεύματα. By ‘living,’ 
Dr Fennell means ‘effective.’ He remarks, with justice, that his version 
‘embodies a less trite sentiment than that attributed to the poet by 
Professor Kennedy.’ 

One more interpretation of ξυμφορᾶς has lately been given by Sir 
George Young, in a note to his translation of the play. ‘I see that, for 
men of experience, the correspondences of their counsels actually exist? ; 
2.6., ‘the things that actually exist correspond with their counsels.’ In 
other words, their counsels suit the conditions of the crisis. This sense 
must be derived from ξυμφέρεσθαι (to agree, concur), not from ξυμφέρειν 
(to bring together). 

With regard, then, to the advocates of the new interpretation, it is a 
case of ‘quot homines, tot sententiae.’ Dr Kennedy, indeed, exactly 
agrees with John Young; but the rest differ in various points both from 
Dr Kennedy and from each other. The only point on which they are 
unanimous is that ξυμῴοράς must mean something which it never means 
anywhere else. We may first consider this contention. 

I. συμφορά is a word of very frequent occurrence, and yet in the 
extant literature of the classical age it is never found except in one of 
two senses,—(1) an occurrence; (11) an unhappy occurrence,—a mis- 
fortune. That is, usage had restricted this very common noun to 
senses parallel with the intransitive συμφέρειν as meaning ‘to happen’ 
(Thuc. 6. 20 ξυνενέγκοι μὲν ταῦτα ws βουλόμεθα, ita eventant). The limit. 


APPENDIX: 253 


imposed by usage can be illustrated from Lucian. His Lexiphanes is a 
satire on a certain kind of affectation in language. There (§ 6) we 
have the phrase τὸ μὲν δὴ δεῖπνον ἦν ἀπὸ συμφορών, ‘the repast was 
furnished from contributions.’ The point is that the learned speaker 
has employed συμφορά in a sense which derivation warranted, but which 
sounded strangely, as parallel with the transitive συμφέρειν, ‘to bring 
together’; the ordinary phrase would have been ἀπὸ συμβολῶν. To this 
argument Dr Kennedy replies: ‘As to Lucian’s jests (dating in the 
second century of our era), I decline to trouble myself with anything so 
irrelevant to the question.’ The irrelevancy, we gather, depends, first, 
on the fact that Lucian is jesting, and secondly on the fact that he 
flourished about 160 4.D. Now, as to the jests, my point is precisely 
that Lucian did think this use of συμφορά a jest. He cannot have been 
jesting in the sense of pretending to think it ludicrous when he did not 
really think it so. And as to 160 a.D., that date surely did not preclude 
Lucian from treating many points of classical idiom with an authority 
which no modern can claim. Can no illustrations of classical Greek be 
derived from Athenaeus, Arrian, Pausanias, Galen, Hermogenes, or 
Oppian? But Dr Verrall has another way of dealing with Lucian’s 
evidence. Heassumes that Lucian’s satire rested on the fact that some 
earlier writer had actually used συμφορά in the sense of ‘contribution.’ 
This view grants at least the singularity of such a sense, since, if there 
was nothing odd in it, there was no room for ridicule. But does such 
a view suit Lucian’s drift here? His Lexiphanes is especially the 
man who employs words in a sense warranted by etymology but not 
warranted by usage. Thus, a few lines further on, Lexiphanes speaks 
of λάχανα τά τε ὑπόγεια Kal τὰ ὑπερφυῆ, ‘vegetables which grow under 
ground (ze. roots) and above ground.’ His use of ὑπερφυής has just as 
much, and as little, warrant as his use of συμφορά : viz., the etymo- 
logical warrant. If, however, Greek literature had actually recognised 
συμφορά as ‘contribution,’ then the satire would have missed its peculiar 
point. Lexiphanes would merely be using a fine word where a simpler 
one would have served. And is it probable that any classical writer had 
opposed ὑπερφυής to vroyeos? It remains to notice some passages of 
the dramatists in which Dr Verrall has suggested that συμφορά means 
neither ‘occurrence’ nor ‘misfortune.’ In each case his proposed 
version is added in brackets, while the ordinary version immediately 
follows the Greek. 


(1) Aesch. Zum. 897 τῷ yap σέβοντι συμφορὰς ὀρθώσομεν: ‘we will prosper the 
fortunes of our worshippers.’ [‘We will prosper their zsezons,—making them and 
their living possessions fertile.] (2) 26. 101g μετοικίαν δ᾽ ἐμὴν | εὐσεβοῦντες οὔτι 
μέμψεσθε συμφορὰς βίου: ‘while ye revere us as dwellers among you, ye shall not 
complain of the fortunes of your lives.’ [‘Ye shall not complain of the union of our 
life,’—7.e., of our united life.] (3) Soph. Z7. 1179 οἴμοι ταλαίνης apa τῆσδε συμφορᾶς : 
‘Woe is me, then, for this thy wretched plight.’ [‘ For our unhappy meeting.’] (4) 2d. 
1230 ὁρῶμεν, ὦ παῖ, κἀπὶ συμφοραῖσί μοι | yeynOds ἕρπει δάκρυον ὀμμάτων ἄπο: ‘we 
see it, and for thy (happy) fortunes a tear of joy trickles from our eyes.’ [‘For thy 
meeting (with thy brother).’] (5) O. 7. 452 ἐγγενὴς | φανήσεται Θηβαῖος, οὐδ᾽ ἡσθή- 
σεται] τῇ ξυμφορᾷ, ‘and shall not be glad of his fortune.’ .[‘His w#zon with the 
citizen-body.”] (6) [Eur.] Res. 980 ὦ παιδοποιοὶ ξυμφοραί, πόνοι βροτῶν: ‘sorrows 
in the begetting of children, woes for men.’ [‘Child-producing wsioms.’] In these 





214 APPENDIX. 


six places, the unexampled sense of συμφορά is sought from συμφέρεσθαι. In the 
following, it is sought from the active sense of συμφέρειν. (7) Eur. AZed. 552 πολλὰς 
ἐφέλκων ξυμφορὰς ἀμηχάνους : ‘cumbered with many perplexing troubles.’ Jason means 
Medea and his children by her. [‘Much troublesome luggage,’—lit., ‘things carried 
along with me.’] (8) 7. 54 χρηστοῖσι δούλοις ξυμφορὰ τὰ δεσποτῶν | κακῶς πίτνοντα, 
kai φρενῶν ἀνθάπτεται: ‘to good slaves their masters’ ill luck is a misfortune,’ etc. 
[‘Their masters’ ill luck is a durden which they share,—lit. ‘a thing borne jointly’ 
by them.]—The shorter form of the saying in Bacch. 1029, χρηστοῖσι δούλοις ξυμφορὰ τὰ 
δεσποτῶν, may, as Dobree thought, be an interpolation; but in any case ξυμῴφορά can 
mean ‘misfortune,’ since τὰ δεσποτῶν is shown by the context to mean, ‘their 
masters’ troubles.’ 

In each of the above passages the ordinary sense of συμφορά is not 
only perfectly clear, but also perfectly appropriate and _ satisfactory. 
The attempt to invest it with an unexampled meaning is in every 
instance strained; in some of the instances it is extremely so. Is there 
a single one of those passages in which the unusual version would have 
occurred to a critic who was not in search of an argument by which to 
defend the strange version of ξυμφοράς as ‘comparisons’ in O. Z: 44? 
But the process might be carried further. ‘There is hardly any passage 
of Greek literature in which a novel sense for €vudopa, fairly suitable 
to the particular context, might not be devised, if we were free to draw 
upon all the senses both of συμφέρειν and of συμφέρεσθαι. And so at 
last we might prove that συμφορά never meant ‘occurrence’ or ‘ misfor- 
tune.’ 

2. Next, we will suppose that Sophocles intended to hazard an 
exceptional use of the noun, relying on the context to show that 
ξυμφοράς meant ‘comparisons.’ Convenience prescribes the general rule 
that, when a strange use of a word or phrase is risked in reliance on an 
explanatory context, this context should not follow at an interval, but 
should either precede or closely accompany the word or phrase which 
would otherwise be obscure. A rough illustration—the first that occurs 
to me—from our own language will serve to show what I mean. ‘Many 
of the visitors were afterwards present at a collation, and did ample 
justice to the difference of hands in the mss.’ If we heard that read 
aloud, we should be apt to suppose—down to the word ‘to’—that 
‘collation’ meant luncheon; and a certain degree of discomfort would 
attend the mental process of apprehending that it meant a comparison 
of documents. This inconvenience would not arise if the mention 
of the mss. preceded, or closely accompanied, the word ‘collation.’ 
Such an argument applies @ fortiori to συμφορά, since the literary sense 
of the word ‘collation’ is at least thoroughly recognised, while συμφορά 
nowhere else occurs in the sense of ‘comparison.’ Consider now the 
two verses, 

Ws τοῖσιν ἐμπείροισι Kal τὰς ξυμφορας, 
ζώσας ὁρῶ μάλιστα τῶν βουλευμάτων. 


When the first verse was spoken, would any hearer in the theatre doubt 
that ξυμφοράς bore its usual sense, or divine that it was to bear the 
unexampled sense of ‘comparisons’? And the indispensable clue, 
τῶν βουλευμάτων, is postponed to the end of the next line. In the cir- 
cumstances, it is hard to imagine any good writer arranging his words 


a νυ 


APPENDIX, 215 


thus; it is, to me, altogether inconceivable that a skilled writer for the 
stage should so arrange them. If Sophocles had intended to suggest 
ξυμφέρειν βουλεύματα, he would at least have given ξυμφορὰς βουλευ- 
μάτων. In reply to this argument, Dr Kennedy merely says that no 
modern can tell; and that Sophocles has used many words, each of 
which occurs only once in his writings. But he has overlooked the 
distinction between a rare word, and a rare meaning for a common 
word. Suppose that the word συμφορά occurred only in O. 7: 44; 
then his reply would at least be relevant. But the word is exceedingly 
common ; and yet in the entire range of classical Greek literature this 
is the solitary place where any one has even suggested that it means 
‘comparison.’ The argument from the order of words is not, therefore, 
one which can be answered by simply saying that it is an argument 
which no modern is qualified to use. It is an argument which a modern 
writer is here strictly entitled to use. When people hear a familiar 
word, they will take it in its usual sense, unless they are warned to the 
contrary. This, we may presume, was as true in 450 B.C. as it is to-day. 

Now, turning from the phrase τὰς ξυμφορὰς τῶν βουλευμάτων, I wish 
to compare the received version with Dr Kennedy’s in respect of two 
other points: (1) ζώσας : (2) the force of καί Dr Kennedy maintains 
that his version is the only one which suits these words. I grant that 
his version suits them; but I submit that the received version suits 
them equally well. First, as to ζώσας. When Shakespeare says, ‘the 
evil that men do lives after them,’ he is using the verb ‘to live’ as 
Sophocles uses ζῆν here: 7.2, ‘to live’ means ‘to be operative,’ ‘to 
have effect’; as, conversely, ‘dead’ can be used of what has ceased to 
be active. In two other passages of Sophocles (quoted in my note) the 
use of ζῆν is strictly similar. In v. 482 the oracles are ζῶντα, ‘living’ 
—not dead letters—because they remain operative against the criminal ; 
a divine power is active in them, and will not suffer him to escape. In 
Ant. 457 the ‘unwritten and unfailing laws of heaven’ “ve (ζῇ), as 
having an eternal and ever-active validity, which no edict of man can 
extinguish or suspend. Here, the events which flow from the counsels 
of experienced men are said to ‘live,’ because they are effective for their 
purposes,—ecas καὶ οὐκ ἀπολλυμένας, as the old scholium in L has it; 
they do not ‘come to nothing.’ On v. 45 the Scholiast has ζώσας" ἀντὶ 
τοῦ ἐνεργεστέρας: ζ.6.) more ‘operative’ than are the counsels of the 
inexperienced. Dr Kennedy renders, ‘comparisons of counsels are 7 
most lively use. ‘This is quite legitimate; it is as possible to say, τὸ 
ἔθος ζῇ, the custom lives (z.¢., is in lively use), as to say, of νόμοι ζῶσιν, the 
laws live (2.6.5 are in active operation). But Dr Kennedy has not observed 
that, by adding the word ‘/ve/y,’ he has extended the figurative use of 
ζῆν to just those limits which I claim for it, and beyond the limits to 
which he himself seeks to restrict it when he says that, figuratively, it 
can mean only (1) ‘to live qwed/, (2) ‘to survive, to remain alive’ For 
if he rendered ζώσας in real conformity with his second proposed sense, 
he would have to say merely, ‘I see that it is with men of experience 
that comparisons of counsels chiefly survive’ (or ‘remain in use’). That 
is to say, the words would imply that the consulting of other people 


216 APPENDIX: 


was an old-fashioned practice, the survival of which was chiefly due to 
the conservative instincts of experienced persons. Then as to the καί 
Prof. Kennedy takes it to mean: ‘counsellors of experience do also, 
most of any, consult other people.’ I take it to mean: ‘the men of 
experience are also, in most cases, the men whose counsels prove 
effectual.’ To put it more shortly, of ἔμπειροι kat εὐβουλοί εἰσι μάλιστα. 
It is, therefore, incorrect to say that the received version deprives 
καί of its point. It has just as much point in that version as in the 
new one. 

Prof. Kennedy lays peculiar stress on a new canon which he has 
formulated, and which he calls ‘the law of ws, szvce.’ The gist of this 
law is to prove that ws, in O. Z. 44, must necessarily refer to the clause 
εἴτ᾽ az ἀνδρὸς οἶσθά που in 43, and cannot refer to the whole preceding 
sentence from viv 7 in 40 onwards. The law is stated thus:—ws, ‘since,’ 
as used by Sophocles, is invariably ‘referred to words immediately going 
before it.’ This statement lacks something in clearness. On my view 
also ws refers to ‘words immediately going before it,’—only to a greater 
number of them. Nor is it easy to see how ws could do anything else. 
But what Prof. Kennedy evidently means to say is this: —When the sen- 
tence preceding ws, ‘since,’ consists of more than one clause, then Sopho- 
cles always refers ws to the last clause, and never to the whole sentence. 
I venture to hope that some readers will accompany me in an attempt 
to test this canon. Prof. Kennedy begins by referring to seven other 
passages in this play, which will not detain us long. Three of them are 
irrelevant, since the sentence preceding ws is of one clause only: 365 
OI. ὅσον ye χρήζεις: ws etc.: 445 OL. κομιζέτω δῆθ᾽ - ὡς etc.: 1050 OL. 
σημήναθ᾽ - ws etc. Two of them are really apposite for Dr Kennedy’s 
purpose, viz. 47 and 54, in each of which ws refers to the nearest clause 
of the preceding sentence. Two are ambiguous, viz. 922, where ws 
may refer to the whole sentence, from 918 to g21, just as well as to 921 
alone: and 56, where ws may refer to the whole of vv. 54 and 55, 
just as well as to v. 55 alone. ‘The fact is, as might have been ex-. 
pected, that ws (‘since’), when it follows a sentence of more than one 
clause, sometimes refers to the whole sentence, and sometimes to the 
last clause of that sentence. 

Prof. Kennedy proceeds :— 


‘The other places to which I refer are: O. C. 562, 937, 1016, 1028, 1075, 
1229, 1528, 1691 ; Ant. 66, 499, 624, 765, 1337; 77. 385, 391, 453, 488, 592, 596, 
599, 921, 11203 At. 39, 92, 131, 141, 789, 1314; ZV. 17, 21, 324, 369, 470, 633, 821, 
1112, 1319, 1337, 1446, 1489; Ph. 46, 53, 117, 464, 807, 812, 847, 014, 1043, 1442, 
and a few in the fragments. I have examined all, and find the fact to be as I state 
it; and I must confess myself amazed that any scholar can look at this passage care- 
fully without discerning that 44, 45 are in immediate dependence on εἴτ᾽ ἀπ᾽ ἀνδρὸς 
ola 6d, που, even without the clinching proof supplied by this crowd of examples.’ 


The number of passages thus alleged as examples is 50. Prof. 
Kennedy claims them all as proving that ws, in v. 44, must refer to 
εἴτ᾽ az ἀνδρὸς οἶσθά που in v. 43, and could not refer to the whole 
preceding sentence from v 40 to v. 43. 1 have examined all these 50 
passages, and I propose to give here the results of that examination. 


APPENDIX. 217 


I find that Dr Kennedy’s 50 citations can be classified under the 
following heads. 


I. Passages which are irrelevant to O. 7. 40—44, owing to the form of the 
sentence. In each of these, ws refers to a short and compact sentence preceded by 
a full stop. There is no separable clause, like εἴτ᾽ dm’ ἀνδρὸς οἶσθά mov, which could 
appropriate ὡς to itself, and 50 withdraw its significance from the whole sentence. 

(1) O. Ὁ: 937 ΧΟ: ὁρᾷς ἵν᾽ ἥκεις, ὦ ξέν᾽; ws etc. (2) 20. 1016 ΘΗ. ἅλις λόγων, 
ὡς εἴς. (3) 2b. 1028 κοὐκ ἄλλον ἕξεις εἰς τόδ᾽" ws etc. (4) 26. 1074 ἔρδουσ᾽ ἢ μέλλουσιν ; 
ὡς εἴς. (5)26.1689—1691 κατά με φόνιος ᾿Αἴδας ἕλοι | πατρὶ ξυνθαν εῖν γεραιῴ | τάλαιναν' 
wsetc. Similar are (6) A¢.65f. (γ) 10. 400. (δ),7.: [337- (9) 77. 385. (10) 26.391. 
(11) 2b. 453. (12) 2b. 592. (13) Ὁ, SoG. (14) 7 508. {1} 2. 20 ἢ: (10) 700 1120: 
(17) Az, 1313. (18) Al. 15—17. (19) 12.201, (20) 70: 324. (21) 26. 369. (22) 7d. 
470s $(23)'t0. S200 (24) 20. TatSe ἴ28) Ὁ: 1737. (20) 20. 144s) ἢ (27) PA. 104. 
(28) 2b. 807. (29) 2b. 844—847. (30) 26. 914. (31) 16. 1440. 


II. Passages which are irrelevant because in them ws does not mean ‘since,’ but 
either (a) ‘that,’ (6) ‘how,’ (ἡ) ‘how!’ (exclamatory), (4) ‘in order that,’ or 
(e) ‘even as.’ 

a. (32) O. C. 562 ὃς οἷδα καὐτὸς ws ἐπαιδεύθην ξένος. (33) Az. 39 ΑΘ. ws ἔστιν 
ἀνδρὸς τοῦδε τἄργα ταῦτά σοι. (34) Ph. 117 ΟΔ. ὡς τοῦτό γ᾽ ἔρξας δύο φέρει δωρήματα. 
(35) 16. 812 NE. ὡς οὐ θέμις γ᾽ ἐμοὔστι σοῦ μολεῖν ἄτερ. 

ὦ. (36) Az. 789 τοῦδ᾽ εἰσάκουε τἀνδρός, ws ἥκει φέρων etc. 


Ζ. (37) 26. 92 ὦ χαῖρ' ᾿Αθάνα, χαῖρε διογενὲς τέκνον, ὡς εὖ παρέστης. (38) Ei. 
1112 ΗΛ. τί δ᾽ ἔστιν, ὦ ξέν᾽ ; ὡς μ᾽ ὑπέρχεται φόβος. 


a. (39) Ant. 765 (‘I will go’) ws τοῖς θέλουσι τῶν φίλων μαίνῃ συνών. 
e. (40) Az. 141 (following a full stop) ὡς καὶ τῆς νῦν φθιμένης νυκτός etc. 


Thus, of 50 passages cited by Dr Kennedy from plays of Sophocles 
other than the Oed. Zyr., 40 are wholly irrelevant. Of the remaining 
10, one is a wrong reference, viz. Ant. 624. If Ant. 643 (ws...avrapv- 
νωνται) 15 meant, that comes under II. (4) above, and-raises the list of 
40 to 41. The other g illustrate the fact which I stated above; viz., 
that when ws, meaning ‘since,’ follows a sentence of more than one 
clause, it sometimes refers to the whole sentence, and sometimes spe- 
cially to the last clause of that sentence. Dr Kennedy maintains that 
it must always refer to the last clause (as to εἴτ᾽ am ἀνδρὸς οἶσθά που 
here). Among the g passages which now remain to be considered, it 
will be found that there are only three such instances :— 


(1) nie. 45—47 τὸν οὖν παρόντα πέμψον εἰς κατασκοπήν, | μὴ καὶ λάθῃ με προσπε- 
ody’ ὡς μᾶλλον dv | ἕλοιτό μ᾽ ἢ τοὺς πάντας ᾿Αργείους λαβεῖν. Here ws refers to μὴ καὶ 
λάθῃ etc. 

(2) 2b. 50—53 ᾿Αχιλλέως παῖ, δεῖ σ᾽ ἐφ᾽ οἷς ἐλήλυθας | γενναῖον εἶναι, μὴ μόνον 
τῷ σώματι, | ἀλλ᾽ ἤν τι καινὸν ὧν πρὶν οὐκ ἀκήκοας | κλύῃς, ὑπουργεῖν, ὡς ὑπηρέτης 
πάρει. Here the last three words, though they enforce the whole precept, are more 
. particularly a comment on ὑπουργεῖν. 

(3) £7. 632 f. ἐῶ, κελεύω, Ode" μηδ᾽ ἐπαιτιῶ | τοὐμὸν ordu’, ws οὐκ ἂν πέρα λέξαιμ᾽ 
ἔτι. This is the usual punctuation. But we might also place a comma at θῦε, and 
a colon at oréu’, when the passage would be more evidently a case of ws referring to 
the last clause of a sentence. 


In the following passages, on the other hand, ws refers to the whole 
preceding sentence; as I hold that, in O. 7: 44, ws refers to the whole 
sentence from v. 40 onwards :— 


(ip: 77. 484—489 ἐπεί γε μὲν δὴ πάντ᾽ ἐπίστασαι λόγον, κείνου τε καὶ σὴν ἐξ 
ἴσου κοινὴν χάριν καὶ στέργε τὴν γυναῖκα καὶ βούλου λόγους | οὗς εἶπας ἐς τήνδ᾽ ἐμπέδως 


21ὃ APPENDIX. 


εἰρηκέναι" | ws τἄλλ᾽ ἐκεῖνος πάντ᾽ ἀριστεύων χεροῖν | τοῦ τῆσδ᾽ ἔρωτος els ἅπανθ᾽ ἥσσων 
ἔφυ. Here, ws does not refer to the last clause, καὶ βούλου λόγους etc., but to the 
whole sentence from v. 484 to 487. 

(2) Ph. 1040—1044. ὡς in 1443 refers to the whole prayer for vengeance, and 
not merely to the clause εἴ τε κἄμ᾽ οἰκτίρετε in 1042. 

(3) O. C. 1526-—1530. ws in 1528 refers to the whole sentence from 1526. 

(4) Az. 127—133. ws in 131 refers to the whole sentence from 127. 

(5) O. C. 1225—1230. ws in 1229 refers to the whole sentence from μὴ φῦναι 
in 1225. 

(6) £2. 1487—1490. ws in 1489 refers to the whole sentence, and not merely to 
the clause καὶ κτανὼν πρόθες etc. 


We have now examined Prof. Kennedy’s 50 passages, with this 
result :—40 are irrelevant: 3 make for his view: 6 make for mine: and 
1 (Ant. 924) is either irrelevant (being for “πὲ. 643) or undiscoverable. 
It seems, then, permissible to say that the new ‘law of ws’ is as devoid 
of ground in the actual usage of Sophocles as it is contrary to what 
we might have reasonably expected. 

The questions of language raised by the different interpretations 
have now been considered. With regard to the general spirit and tone of 
the speech in which the disputed passage occurs, they appear decidedly 
favourable to the old interpretation, and decidedly adverse to the new. 
The Priest of Zeus salutes Oedipus, not, indeed, as a god, but as unique 
and supreme among mortals. It was by the direct inspiration of a god 
(προσθήκῃ θεοῦ, v. 38), not by any help from man, that Oedipus was 
believed to have solved the riddle of the Sphinx. His success on that 
occasion is the ground assigned for believing that he will succeed now. 
But, according to the new interpretation, the passage expressing this 
belief winds up with a remark to the effect that ‘men of experience are 
just those who are most ready to consult other people.’ In this context, 
such a remark is both illogical and unpoetical. It is illogical, because 
the thought is that, as formerly he found a remedy when Theban 
advice could not aid him (ὑφ᾽ ἡμῶν οὐδὲν ἐξειδὼς πλέον), so he may find 
a remedy now, though the Thebans have no counsels to offer him. It 
is unpoetical, because Oedipus, who has just been exalted far above all 
other men,—to a rank which is only not divine,—is suddenly lowered 
to the ordinary level of shrewd humanity. 

In concluding this Note, I may briefly recapitulate the points which 
it has sought to establish. The old interpretation of verses 44 and 45, 
—that which has come down, presumably, from the Alexandrian age, 
and which modern scholars have been all but unanimous in upholding,— 
suits the general context, employs ξυμῴφορά in its ordinary sense, and 
gives a legitimate meaning both to ζώσας and to καί The new inter- 
pretation gives ξυμφορά a meaning which the word, though extremely 
common, never once bears in the classical literature. Etymology, 
indeed, warrants that meaning; but, as Lucian shows by the example of 
this very word €vudopa, it was possible to observe etymology and yet 
to commit a iudicrous offence against usage. Further, if Sophocles had 
desired to use ξυμφορά in an unexampled sense, it is improbable that he 
would have chosen to arrange his words in such an order as to aggravate 
the obscurity. The contention that ws must refer to the last clause of v. 


APPENDIX. 219 


43, rather than to the whole sentence, is groundless. Lastly, the general 
sense obtained by the new interpretation is not in good harmony either 
with the argument or with the spirit of the context. 

It is among the advantages and the pleasures of classical study that 
it gives scope for such discussions as this passage has evoked. I have 
endeavoured to weigh carefully what can be said on both sides, and to 
give the result,—as it appears to me. If any one prefers a different 
view, κεῖνός τ᾽ ἐκεῖνα στεργέτω, κἀγὼ τάδε. 


198 f. τελεῖν yap, εἴ τι νὺξ ἀφῇ, 


elk pil Yous vast 2 μὴ 
TOUT ἐπ ἡμαρ ἐρχεται. 


Before adopting τελεῖν, I had weighed the various interpretations of 
τέλει, and had for some time been disposed to acquiesce in Elmsley’s 
as the least strained. He renders ‘ommnino,’ ‘ absolute, comparing Eur. 
Bacch. 859 ff. γνώσεται δὲ τὸν Διὸς Διόνυσον ὃς πέφυκεν ἐν τέλει θεὸς 
| δεινότατος, ἀνθρώποισι δ᾽ ἠπιώτατος. On Elmsley’s view, ἐν τέλει there 
means ommnino, ‘in fulness’; and here the sense would be ‘in fulness— 
if night spare aught—day attacks this’: 2.6. so as to make the tale of 
havoc full. Yet I think with Professor Tyrrell that in Bacch. 860 ἐν 
τέλει could not bear the sense which Elmsley gave to it. I should 
prefer there to render it, as Dr Sandys did, ‘in the end’—ze, when 
his wrath has been aroused. I now believe, however, that Munro’s 
brilliant emendation in that place is right,—os πέφυκεν ἐν ἀτελεῖ θεὸς | 
δεινότατος: ‘who is a god most terrible towards the uninitiated’ (Fourn. 
Philol. Vol. x1. p. 280). If, then, τέλει is to mean ‘in fulness’ here, it 
must dispense with even such support as might have been derived from 
the passage in the Bacchae. And, at the best, the sense obtained by 
such a version is hardly satisfactory. Still less would it be so, were 
τέλει joined with ἀφῇ, as=‘spare anything at all’: εἴ τι τέλει ἀφῇ could 
not possibly mean εἰ ὁτιοῦν ἀφῇ. Nor could τέλει go with ἀφῇ as= 
‘remit anything zz regard to completeness’: nor again, as Hermann 
proposed, ‘remit anything Zo the completion’—z.e. fail to complete. 

Others have rendered—‘if night αἱ z¢s close spare anything.’ The 
objections to this are,—(i) the weakness of the sense: (11) the szmple 
dative in this meaning: for ‘at the end’ is ἐπὶ τῷ τέλει (Plat. Polit. 
268 Ὁ), or πρὸς τέλει (Legg. 768 6). The Scholiast who explains τέλει as 
ἐπὶ τῷ ἑαυτῆς τέλει begs the question by his addition of ἐπὶ ro. Of pro- 
posed emendations, the obvious reAety—which Hermann merely sug- 
_ gested, himself preferring the bolder cure mentioned below—is at once 
the simplest and the best. Dindorf spoils it (in my judgment) by taking 
it with ἀφῇ instead of ἐπέρχεται ----- Fortasse igitur scribendum, τελεῖν 
yap εἴ (vel 4) τι νὺξ ἀφῇ, 1.6. nox si (vel ubz) guid malorum perfciendum 
reliquerit, td dies aggreditur et perfictt.’ 

Among other conjectures are: (1) Kayser, τελεῖ γάρ" εἴ τι κιτιλ. ‘for 
Ares will finish his work.’ (2) Hermann, μέλλει yap: εἴ τι νὺξ δ᾽ ἀφῇ 
x.7.\.: ‘Cunctatur enim (sc. Mars): si quid nox autem dimiserit, id 
invadit dies’; μέλλει, ‘delays,’ meaning, I suppose, ‘tarries too long 
among us.’ (3) Arndt would change τέλει into ἀεὶ, and in the 5th ed. 


220 APPENDIX. 


of Schneidewin (revised by Nauck) this is approved, τέλει being pro 
nounced ‘clearly wrong.’ 


ΘΠ: ἀγὼ ἕένος μὲν τοῦ λόγου τοῦδ᾽ ἐξερώ, 
ξένος δὲ τοῦ πραχθέντος: οὐ γὰρ ἂν μακρὰν 
ἴχνευον αὐτός, μὴ οὐκ ἔχων τι σύμβολον. 


Professor Kennedy understands οὐ yap κιτιλ. as referring to a sup- 
pressed clause. ‘On my having been a foreigner at the time of the 
deed, I lay no stress; for had I been no foreigner, but one of the 
citizens, I myself, whatever my native shrewdness, as in guessing the 
riddle of the Sphinx, should not have traced the matter far, seeing that 
I had not (μὴ οὐκ ἔχων) any token (2.6. any clue to guide me).’ 

The difficulties which I feel in regard to the above interpretation 
are these. (a) I do not see how the hearer could be expected to supply 
mentally such a suppressed clause as ‘That, however, matters not; for 
even if I had been a citizen’... (6) The opBodov lacking to Oed. is 
some way of obtaining such a clue. We should not expect him, then, 
to say that, even if he had been a citizen of Thebes at the time, he 
could not have made much progress in the investigation, because he 
would have had no clue. 

According to Professor Campbell, the suppressed clause is εἰ ἴχνευον, 
and the sense is: ‘I have remained a stranger to the matter, for, if I 
had undertaken an inquiry, I could not have followed it far, since I had 
no clue to guide me.’ ‘He offers this excuse for having hitherto 
neglected what he now feels to be an imperative duty.’ But Sophocles 
assumes that Oed. has just heard, for the first time, of the mysterious 
murder (105—-129). On hearing of it, Oed. straightway asked why the 
Thebans themselves had not at the time made a search (128). Here, 
then, we cannot understand him to speak as if he had all along shared 
the knowledge of the Thebans, or as if he were apologising for having 
neglected to act upon it sooner. 

Mr Blaydes understands: ‘For (were it otherwise, had I not been 
thus ignorant), I should not have had to investigate it (αὐτὸ, the foul 
deed) far, without finding (quin haberem) some clue.’ To this the 
objections are that (1) μὴ οὐκ éywv=‘unless I had,’ and could not mean 
‘without finding’: (2) the remark would be suitable only if Oed. had 
already for some time been engaged in a fruitless search, whereas he is 
only about to commence it. 

Schneidewin formerly conjectured 7 [for ov] γὰρ av μακρὰν | ἴχνευον 
αὐτός, οὐκ [for μὴ οὐκ] ἔχων τι σύμβολον : ‘for [if I had mot appealed to 
you] I should have searched long indeed by myself, seeing that I have 
no clue.’ In the sth ed., revised by Nauck, ov is wisely replaced 
instead of 7 (though οὐκ for μὴ οὐκ is kept), and the sense is given 
substantially as I give it. 

Much of the difficulty which this passage has caused seems 
attributable (1) to a prevalent impression that ov ydp...av in such a 
sentence always means, ‘for eé/se,’ etc.: (2) to want of clearness regarding 


ae 


ALLEMDTX. 221 


Now, as to (1), it depends on the context in each case whether ov yap 
av means, ‘for εἶδέ, etc. When it has that force, it has it because there 
is a suppressed protasts. Such is the case in v. 82 ἀλλ᾽ εἰκάσαι μὲν ἡδύς" 
ov γὰρ ἄν. οεεἷρπε: 2.€. εἰ μὴ ἡδὺς nv. Such is also the case in 318 διώλεσ᾽- 
ov yap ἂν ΞΡ ἱκόμην: 1.6. εἰ μὴ διώλεσα. But when the protasis is ot 
suppressed, then, of course, there is no such ellipse as our word ἡ Εἴθε 
implies. Thus Xen. Anad. ἡ. 7. 11 καὶ νῦν. ἄπειμι" οὐδὲ γὰρ ἂν Μήδοκός 
με ὁ βασιλεὺς ἐπαινοίη, εἰ ἐξελαύνοιμι τοὺς εὐεργέτας: ‘and now I 
will go away; for Medocus the king would not commend me, 7f 7 
should drive out our benefactors. ad the protasis εἰ ἐξελαύνοιμι τοὺς 
evepy. been suppressed, then οὐδὲ γὰρ By taser} must have been 
rendered, ‘for e/se he would not commend me’: but, since it is 
given, we do not need “else; So Dem. or. 18 5. 228 ὡμολόγηκε νῦν 
y ἡμᾶς ὑπάρχειν ἐγνωσμένους ἐμὲ μὲν λέγειν ὑπὲρ τῆς πατρίδος, αὐτὸν δ᾽ 
ὑπὲρ Φιλίππου. οὐ γὰρ ἂν μεταπείθειν ὑμᾶς ἐζήτει, μὴ τοιαύτης οὔσης τῆς 
ὑπαρχούσης ὑπολήψεως περὶ ἑκατέρου : ‘he has admitted that, as matters 
stand, we are already pronounced to be speaking, 1, in our country’s 
cause, and he, in Philip’s; for he would not have been seeking to bring 
you over to his view, were not such the extsting impression with regard to 
each.’ Here, μὴ τοιαύτης ovens represents the protasis, εἰ μὴ τοιαύτη ἦν, 
exactly as here in Ὁ. Z: 221 py οὐκ ἔχων represents the protasis εἰ μὴ 
εἶχον: and we do not insert ‘else’ after ‘for.’ 

(2) As regards μὴ ov with the participle, the general principle may, 
I think, be stated thus. Every sense possible for (e¢y.) μὴ ποιῶν is 
possible for μὴ ov ποιῶν when the principal verb of the sentence is 
negative. Take the sentence fadiov ἡμῖν ζῆν μὴ πονοῦσι. The participial 
clause here could represent, according to the sense intended, any one of 
four things, viz. (1 ) εἰ μὴ πονοῦμεν, ‘if,—as is the fact,—we are not 
labouring’: (2) ἐὰν μὴ πονῶμεν, ‘whenever we do not labour,’ ov, ‘if we 
shall not labour’: (3) εἰ μὴ πονοῖμεν, ‘if we should not labour’: (4) εἰ 
μὴ ἐπονοῦμεν, ‘if we had not (then) been labouring, (as in fact we then 
were,)’ ov, ‘if we were not (now) labouring, (as in fact we now are).’ 
So in the negative sentence, ov ῥᾷδιον ἡμῖν ζῆν μὴ ov πονοῦσι, the 
participial clause can equally represent any one of the same four things. 

But from the very fact that μὴ ov can stand only in a negative 
sentence it follows that a participial clause with μὴ ov will, in practice, 
most often express an exception to a negative statement. This must not, 
however, make us forget that μὴ ov with the participle is still equivalent 
to the protasis of a conditional sentence. Thus :— 

Her, 6. 9 πυθόμενοι τὸ πλῆθος τῶν Ἰάδων νεῶν καταρρώδησαν μὴ οὐ 
δυνατοὶ γένωνται “ὑπερβαλέσθαι, καὶ οὕτω οὔτε τὴν Μίλητον οἷοί τε ἔωσι 
ἐξελεῖν μὴ οὐκ ἐόντες ναυκράτορες κιτιλ.: Where μὴ οὐκ ἐόντες -- εἰ μή εἰσι, 
(or ἢν μὴ ἔωσι,) the negative condition. Her. 6. 106 εἰνάτῃ δὲ οὐκ 
ἐξελεύσεσθαι ἔφασαν μὴ οὐ πλήρεος ἐόντος τοῦ κύκλου, 2.6. εἰ μὴ πλήρης 
ἐστὶν ὁ κύκλος, ‘if (as is the case) the moon is not full’ (they are 
speaking on the εἰνάτη itself). Plat. 2γδῖξ 212 Ὁ οὐκ ἄρα ἐστὶ φίλον 
τῷ φιλοῦντι μὴ οὐκ ἀντιφιλοῦν, 2.6. ἐὰν μὴ ἀντιφιλῇ, unless it love in 
return. Soph. Ὁ, C. 359 ἥκεις yap οὐ κενή ye, τοῦτ᾽ ἐγὼ σαφῶς | ἔξοιδα, 
μὴ οὐχὶ δεῖμ᾽ ἐμοὶ φέρουσά τι: ‘thou hast not come empty-handed, 


222 APPENDIX. 


without bringing,’ etc.: where the participial clause, epexegetic of κενή, 
implies εἰ μὴ ἔφερες, (οὐκ ἂν ἧκες,)---- hadst thou not been bringing (as 
thou arf bringing), thou wouldst not have come.’ 

In all the above passages, it is the present participle which stands 
after μὴ ov, as it 1s also τ wh @ aay ae 13, 221. Now compare (1) Dem. 
OF 18 5 34 μὴ κατηγορήσαντος Αἰσχίνου (=ei μὴ κατηγόρησεν Αἰσχίνης) 
Bee ἔξω τῆς γραφῆς οὐδ᾽ ἂν ἐγὼ λόγον οὐδένα ἐποιούμην ἕτερον. (2) or. 
19 § 123 ov yap ἐνῆν μὴ παρακρουσθέντων ὑμῶν (=e μὴ παρεκρού- 
σθητε ὑμεῖς) μεῖναι Φιλίππῳ. Here, though the sentences are nega- 
tive, we vee μή, not μὴ οὐ, with the aorist partic., representing the 
protasis. In (1) the order of clauses affects the question, but not in (2). 
Owing to τς comparative rarity of μὴ ov with the participle, generali- 
sation appears unsafe; but it looks as if prevalent usage had accustomed 
the Greek ear to μὴ οὐ with partic. chiefly in sentences where the pro- 
tasis so represented would have been formed with (1) imperf. indic., or 
(2) pres. subjunct., or (3) pres. optat. In conditional sentences with 
the aor. indicative, even where the negative form admitted μὴ οὐ, 
there may have been a preference for μή. ‘The instances cited seem 
at least to warrant the supposition that, in such a sentence as οὐκ av 
ἀπέθανεν εἰ μὴ ἔπεσε, Demosthenes would have chosen μὴ (rather than 
μὴ ov) πεσών as the participial substitute for the protasis. 


227 f. kel μὲν φοβεῖται, τοὐπίκλημ᾽ ὑπεξελὼν 

αὐτὸς καθ᾽ αὐτοῦ. 

With this, the common reading, it is necessary to suppose some 
ellipse. I believe ὑπεξελὼν and αὐτὸς to be indefensible. If they were 
to be retained, I should then, as the least of evils, translate thus :— 
‘And if he is afraid,—when (by speaking) he will have removed the 
danger of the charge from his own path,—[/et him not fear].’ Such an 
ellipse—though, to my mind, almost impossibly harsh—would at least 
be mitigated by the following πείσεται yap ἄλλο μὲν | dorepyés οὐδέν, 
which we might regard as an irregular substitute for an apodosis in the 
sense of μὴ φοβείσθω, yap being virtually equivalent to ‘I tell him.’ 

Among the interpretations of the received text which have been 
proposed, the following claim notice. 

1. Professor Kennedy renders (the italics are his): ‘and if he fears 
and hides away the charge | against himself, let him speak out.’ Here 
ὑπεξελὼν = ‘having suppressed,’ and μὴ σιωπάτω is mentally supplied 
from v. 231 (three verses further on). 

2. Professor Campbell gives the preference to the following version 
(while noticing two others) :—‘And let the man himself, if he be touched 
with fear, inform against himself, by taking the guilt away with him’: 
2.6. ὑπεξελὼν-- ‘having withdrawn,’ and ‘the words καθ᾽ αὐτοῦ are to be 
construed κατὰ σύνεσιν with v. 226, sc. ποιείτω τάδε, self-banishment 
being in this case equivalent to self-impeachment.’ This is tantamount 
(if I understand rightly) to supplying σημαινέτω from σημαίνειν in 226. 

3. Schneidewin: ‘And if he is afraid, decause he will have revealed 
(ὑπεξελὼν) a charge against himself, —let him not Sear’ (st. μὴ φοβείσθω). 
So Linwood, only supplying σημαινέτω. 


APPENDIX. 223 


4. Elmsley: ‘And if he is afraid, (still let him denounce himself, 
s¢. σημαινέτω,) thus extenuating the guilt (by confession),’—crimen con- 
jfitendo diluens. To say nothing of the sense given to ὑπεξελὼν, the 
aorist part. seems strange on this view. 

5. Matthiae regards the construction as an irregular form of what 
might have been more simply put thus: kei μὲν φοβεῖται, τὸ ἐπίκλημα 
αὐτὸς καθ᾽ αὐτοῦ ὑπεξελὼν (ἀπελθέτω ἐκ τῆς γῆς)" πείσεται yap οὐδὲν 
ἄλλο ἀστεργές: ‘If he is afraid, (let him leave the country,) thus faking 
away the charge against himself.’ He explains ὑπεξελὼν by ‘subripiens, 
2.6. subterfugiens, declinans, ‘evading the danger of being accused.’ 
Neither this nor the ellipse of ἀπελθέτω seems possible. Wunder nearly 
agrees with Matthiae. 

6. Hermann (3rd ed.) translates v. 227 ‘Si metuit, subterfugiens 
accusationem sui ipsius,’ and supposes the apodosis to be γῆς ἄπεισιν 
ἀβλαβής,---αἴμὲν and δὲ having been added because the clause πείσεται 
yep has been put first. Thus he agrees with Matthiae as to urefeAur, 
but takes it with φοβεῖται, not with a supposed ἀπελθέτω. 

7. Dindorf also takes Matthiae’s view of ὑπεξελὼν, but wishes (ed. 
1860) for ὑπεξέλοι in an imperative sense: ‘crimen subterfugiat’: ‘let 
him evade the charge against himself’ (by going into exile). 

Under one or another of the above interpretations those given by 
most other commentators may be ranged. 

Among emendations, the palm for ingenuity seems due to Hartung’s 
kel μὲν φοβεῖται, τοὐπίκλημ᾽ ἐπεξίτω | αὐτὸς καθ᾽ αὐτοῦ : ‘and if he is 
afraid, still let him prosecute the charge against himself.’ This is, how- 
ever, more brilliant than probable. 

Mr Blaydes in his note proposes to read κεὶ μὲν φοβεῖται τοὐπίκλημ᾽ 
ὑπεξελεῖν (to draw forth from the recesses of his own mind), and sup- 
plies, ‘let him feel assured.’ For this view of ὑπεξελεῖν, cp. above, 
no. 3. In his text, however, he gives (on his own conjecture) καὶ μὴ 
φοβείσθω τοὐπίκλημ᾽ ὑπεξελεῖν | αὐτὸς καθ᾽ αὐτοῦ. 


246 ff. Zhe proposed transposition of verses 246—251, κατεύχομαι... 
ἠρασάμην. 

Otto Ribbeck suggested that these six verses should stand imme- 
diately after 272 (é€x@iov.). He thought that their displacement in the 
MSS. arose from a confusion between ὑμῖν δὲ in 252 and the same words 
in 273. Heargued that 251, παθεῖν ἅπερ τοῖσδ᾽ ἀρτίως ἡἠρασάμην, has no 
meaning unless it follows 269—274, καὶ ταῦτα τοῖς μὴ δρῶσι κ.τ.λ. 
Many recent editors adopt the transposition, Against it, and in favour 
of the mss., I would submit these considerations. (1) The transposition 
destroys the natural order of topics. The denunciation of a curse on 
the murderer must stand in the fore-front of the speech, whereas the 
transposition subjoins it, as a kind of after-thought, to the curse on those 
who disobey the edict. It thus loses its proper emphasis. (2) The 
transposition enforces an awkward separation between ταῦτα τοῖς 
μὴ δρῶσιν (269) and τοῖς ἄλλοισι (273). The latter depends for its 
clearness on juxtaposition with the former: but six verses are now in- 
serted between them. (3) In 251 Ribbeck’s objection would fail if we 


224 APPENDIX, 


had τῷδ᾽ instead of τοῖσδ᾽ : but τοῖσδ᾽ is used to include the hypothesis 
of several murderers (247, cp. 122). 


305. εἰ καί and καὶ εἰ.---(1) εἰ καί, in its normal usage, =‘ granting 
that...,, where the speaker admits that a condition exists, but denies 
that it is an obstacle: above, 302: 408, εἰ καὶ τυραννεῖς: El. 547, εἰ καὶ 
ons δίχα γνώμης λέγω. 

(2) In our passage (as in 25). 1127, Zr. 71), the εκὰν has a slightly 
stronger sense,—‘ if z7deed—though I should be surprised to hear it.’ 

(3) Both these uses differ from that in which εἰ καί has the sense 
which properly belongs to καὶ εἰ, ‘even supposing that...,’ where the 
speaker refrains from granting the existence of the alleged condition: 
Tr. 1218 εἰ Kat μακρὰ κάρτ᾽ ἐστίν, ἐργασθήσεται, ‘even if the favour is 
a very large one, it shall be granted.’ 

For the regular distinction between εἰ καί and καὶ εἰ, see 71. 4. 347 
καὶ εἰ δέκα πύργοι ᾿Αχαιῶν | ὑμείων προπάροιθε μαχοίατο, compared with 11. 
5. 410 Τυδείδης, εἰ καὶ μάλα καρτερός ἐστιν. 

The normal use of kal εἰ occurs below, 669, 1077: O. C. 306 κεῖ 
βραδὺς | εὕδει: Ant. 234 Kei τὸ μηδὲν ἐξερῶ: 461 Kei μὴ σὺ προὐκήρυξας: 
El. 617 Kei μὴ δοκῶ σοι. 

Conversely, we have καὶ εἰ for εἰ καί in “412. 536, 692, 962: Ο. C. 661: 
below, 986, 1516. 

(4) All the foregoing uses, in which εἰ καί forms a single expression, 
must be distinguished from those cases in which καί belongs closely to 
the following word, as 283 εἰ καὶ τρίτ᾽ ἐστί: Ant. go εἰ καὶ δυνήσει γ᾽. 

Similarly, for καὶ εἰ, distinguish those cases in which καίτε “Δηά᾽ : 
O. C. 1323 ἐγὼ δὲ σός, Kel μὴ σός, ἀλλὰ τοῦ κακοῦ | πότμου φυτευθείς. 


328 f. οὐ μή ποτε 
4 9 ε x 4 Ν Ν t Jee ! ΄ 
TOL WS ὧν εἰπω μὴ τα oO EKPHIW κακα. 


Prof. Kennedy takes the passage thus:—éeyo δ᾽ οὐ μήποτε εἴπω Tapa, 
7 will never speak my things, ws ἂν (εἴπω), however 7 may call them 
(whatever they may deserve to be called), μὴ ta σ᾽ ἐκφήνω κακά, lest 7 
disclose your things as evil. Or, as he renders it in verse, ‘but mine I 
ne’er will speak, | however named, lest I display thine—evil.’ For ὡς 
av as=‘in whatever way,’ he compares //. 2. 139 ὡς ἀν ἐγὼν εἴπω, 
πειθώμεθα πάντες: Soph. Az. 1369 ws ἂν ποιήσης, πανταχοῦ χρηστός γ᾽ 
ἔσει: Dem. or. 18. 292 [8 192] τὸ...πέρας, ὡς ἂν ὁ δαίμων βουληθῇ, 
πάντων γίγνεται: and adds: ‘We might place commas before and after 
ws av, to indicate the quasi-adverbial character which it acquires. by the 
ellipse [of εἴπω], in reality not more abnormal than that of ἥδοιο in goo 
[937], ἥδοιο μέν, πῶς δ᾽ οὐκ av;’ (Oed. Tyr., pp. 76 f.). 

As Prof. Kennedy has well said elsewhere (Sad. Soph. p. 62), if any 
emendation were to be admitted, the simplest would be εἰπὼν for εἴπω (a 
change which Hermann also once suggested), with a comma after rap’. 
ἐγὼ δ᾽ ov μήποτε (εἴπω) τἀμά, ws ἂν εἰπὼν (dy telling them) py...éxprjve. 
But with him (though our interpretations differ) I believe that the words 
are sound as they stand. 


APPENDIX. 225 


Hardly any passage, however, in Sophocles has given rise to so large 
a number of conjectures. Most of these have been directed to the same 
general object—some such alteration of the words τἄμ᾽ ws ἂν εἴπω as 
shall make it easier to take the second μὴ with ἐκφήνω. The following 
may be mentioned: (1) Wolff, tap’ dav’ εἴπω, ‘my visions, —opavoy 
having that sense in Aesch. Cho. 534. (2) Hartung, τὰ θέσφατ᾽ εἴπω. 
(3) C. F. Hermann, τὰ μάσσον᾽ εἴπω. (4) Campbell, εἴπω τάδ᾽, ὡς ἂν 
μὴ τά σ᾽ ἐκφήνω κακά. (5) Nauck, approved by Bonitz, ἄνωγας εἴπω. 
(6) Campe, Quaest. Soph. 1. 18, ἅγνων ἀνείπω. (7) Arndt, τάλλων 
aveirw. (8) Seytfert, Weismann, Ritter, tay’ ws aveirw. (9) Wecklein, 
τἄμ᾽ ὧδ᾽ ἀνείπω. (το) Pappageorgius, tap’ és σ᾽ aveirw. See his Beitrage 
zur Erklirung und Krittk des Sophokles, p. 22, Tena, 1883. 


361. Zhe forms γνωτός and yvwortos.—yvwrtos is regularly formed 
from the verbal stem yvw with the suffix ro: cp. Skt. g#d-tas, Lat. notus. 
In the form yvworos, the origin of the o is obscure: Curtius remarks 
that we might suppose a stem yvws expanded from yvw, but also a 
present *yvwyw, which might be compared with O. H. G. &ndu. In the 
case of καυστός (Eur.), κλαυστὸς (Soph.), the o is explained by xaFyw 
_ (καίω), κλαβ γω (kAaiw). The existing data do not warrant us in assign- 
ing the forms with or without o to certain periods with such rigour as 
Elmsley’s, for example, when he regarded εὔγνωτος as the only correct 
Attic form. ἄγνωστος occurs in Odyssey, Thucydides, Plato (who has 
also yvworos) ; in Pindar /sthm. 3. 48 ἄγνωστοι is doubtful; Mommsen 
gives ayvwro, and so Fennell, who remarks ad Joc. that in OZ 6. 67 for 
ἄγνωτον (as against ἄγνωστον) Mommsen has the support of two good 
Mss. We have ἄγνωτος in Sophocles and Aristophanes ; εὔγνωστος in 
Sophocles, Euripides, Lysias, etc. 

With regard to the meaning of these verbals, it has been held that, 
where such forms as γνωτός and yvworos existed side by side, Attic 
writers appropriated the fotentzal sense to the szgmatic form, distinguish- 
ing yvworos, as ‘what caz be known,’ from yvwrds, ‘what zs known.’ 
Nothing in the sigmatic form itself could warrant such a distinction. 
However the o be explained, γνωστός; no less than yvwrds, must 
have primarily meant simply ‘known,’ as xavords ‘burnt’ and κλαυ- 
στός ‘wept.’ And we find axAavoros as = ‘unwept’ (not, ‘what can- 
not be wept for’), πολύκλαυστος as=‘much-wept’ (not, ‘worthy of 
many tears’). When the modal idea of ‘may’ or ‘can’ attached itself 
to these verbals, it was merely by the same process as that which in 
Latin brought zzvictus, ‘unconquered,’ to the sense of ‘ unconquerable.’ 
Yet I would suggest, on the other hand, that the special attribution 
of a potential sense to the sigmatic forms may have thus much ground. 
When two forms, such as yvwrds and γνωστός, were both current, regular 
analogies would quicken the sense that yvwrds had a participial nature, 
while γνωστός, in which the o obscured the analogy, would be felt more 
as an ordinary adjective, and would therefore be used with less strict 
regard to the primary participial force. Thus it might be ordinarily 
preferred to yvwros, when ‘knowable’ was to be expressed. At the 
Same time, it would always remain an available synonym for yvwrds 


5 ee ἐν 15 


226 : APPENDIX. 


as=‘known.’ And we have seen in the commentary that Sophocles 
is said to have used yvworos, as well as yvwrds, in the sense of ‘ well- 
known.’ 


478. The reading of the first hand in the Laurentian MS., πετραῖος ὁ 
tatpos.—This reading raises one of those points which cannot be lightly 
or summarily decided by any one who knows the rapid transitions and 
the daring expressions which were possible for the lyrics of Greek 
Tragedy. Hermann—who was somewhat more in sympathy with the 
manner of Aeschylus than with that of Sophocles—characteristically 
adopted the reading,—which he pronounces ‘ multo vulgata fortiorem.’ 
The mere substitution of metaphor for simile is not, indeed, the difficulty. 
Euripides, for instance, has (Jed. 184) ἀτὰρ gees εἰ πείσω | δέσποιναν 
ἐμήν"...«καίτοι τοκάδος δέργμα λεαίνης | ἀποταυροῦται ὃμωσίν. But 
the boldness of λεαίνης so closely followed by δμωσίν is not comparable 
to that which we must assume here, if tov ἄδηλον ἄνδρα were so imme- 
diately followed by πετραῖος ὁ ταῦρος : nor can I persuade myself that 
Sophocles would have so written. 

The further verbal question, whether φοιτᾷ πετραῖος could be said in 
the sense, ‘wanders among rocks,’ is one which must be considered in 
the light of Sophoclean usage. We have below 1340 ,ἀπάγετ᾽ ἐκτόπιον : 
1411 θαλάσσιον | ἐκρίψατ᾽ : Antig. 785 φοιτᾷς δ᾽ ὑπερπόντιος ἔν τ᾽ ἀγρονό- 
pots αὐλαῖς : 221. 419 ἐφέστιον | πῆξαι... σκῆπτρον: Ant. 1301 βωμία... | 
λύει.. (βλέφαρα (she closes her eyes at the altar): and perh. fr. 35 ἐξ 
βωμιαῖον ἐσχάρας λαβών, for Steph. Byz. 191. 8, citing it, says, ro 
τοπικὸν βώμιος καὶ κατὰ παραγωγὴν βωμιαῖος. Given these examples, 
we could scarcely refuse to Sophocles such a phrase (for instance) as 
φοιτᾷ opewos. My own feeling in regard to πετραῖος is that it is 
decidedly bolder—not to say harsher—than any phrase of the kind 
which can be produced; but, on the other hand, I certainly am not 
prepared to say that, in lyrics, Sophocles could not have used it. It is — 
the extreme abruptness of the metaphor in this context, rather than 
the singularity of the phrase, that has decided me against reading 
πετραῖος ὃ ταῦρος. 


508. πτερόεσσα κόρα. The Sphinx.—The Sphinx, with lion’s body 
and human head, has a unique place among the most ancient symbols 
of an irresistible daemonic might, at once physical and mental. The 
Egyptian type was wzngless, and of male sex. The Sphinx of Ghizeh— 
oldest and largest of extant examples—dates from the age of the Fourth 
Dynasty (perhaps from εἶ. 2400 B.C.), as Mariette’s latest results have 
established (Revue archéol., new series 26, 1873, pp. 237 ff.), and was the 
object of a cultus, which ‘does not appear to have been the case with 
any other Egyptian Sphinx. 

The winged type occurs first in the lands of the Euphrates. The 
earliest example which can be approximately dated is afforded by the 
palace of Esharaddon, which belongs to the seventh century B.c. Here 
the winged and crouching Sphinx is female (Milchhoefer, A@¢th. des 
deutschen archaeol. Institutes in Athen, fourth year, 1879, p. 48,—the best 
authority for the present state of knowledge on the subject). Phoenicia 


APPENDIX. 227 


was in this case, as in so many others, the point at which Egyptian and 
Asiatic influences converged. A stelé from Aradus (Alusée Napoléon 
111. xvii. 4) shows a Sphinx with Egyptian head-gear and on a pedestal 
of Egyptian character, but with the Assyrian wings. 

The wingless Sphinx was not unknown to the earlier art of Hellenic 
countries. Such a Sphinx (female, however, and in this respect not 
Egyptian) occurred on the Sacred Way at Miletus (Newton, Zvavels 
Vol. 11. p. 155). At Thebes, singularly enough, was found a terracotta 
figure, about 4 inches long, of a wingless crouching Sphinx (Milchhoefer, 
4 ὦ, p. 54). As is well known, it was maintained by Voss in his 
Mythologische Briefe that the Greek Sphinx, being borrowed from 
Egypt, was wingless until the influence of the Attic dramatists popu- 
larised the winged type. Aeschylus, indeed, like Hesiod, does not 
mention wings in his brief description of the Sphinx on the shield of 
Parthenopaeus (Zed. 541), nor in his only other notice of the monster 
(fr. 232): but the Sphinx of Euripides, like that of Sophocles, is 
winged (Phoen. 1022 ff.). Gerhard argued as far back as 1839 (46- 
hanal. der k. Akad. der Wissensch. 2. Berlin) that the Greek winged 
Sphinx was probably much older than the age of the dramatists, 
and this fact has long been placed beyond discussion. The oldest 
representations of the Sphinx found on the soil of Greece Proper are 
presumably the relievo-figures in gold, ivory, etc., of the graves at Spata 
in the Mesogaia of Attica, and at Mycenae: and these have the wings. 
Three round figures of winged Sphinxes, in Parian marble, have also 
been found in Greece (two in Attica, one in Aegina): a round terracotta 
figure of a winged Sphinx, which possibly served as akroterion of a 
heroon, has been found at Olympia, and a similar figure is reported to 
have been found at Corinth. These Sphinxes are regarded by Milch- 
hoefer as the oldest and most complete Greek examples of polychromy 
applied to round figures. The feathers of the Sphinx’s wings were, in 
two cases at least, painted red and dark-green (or blue?), and in one 
instance a brownish-red colour had been given to three corkscrew ringlets 
which fell on the Sphinx’s breast and shoulders. 

It was not in connection with Thebes and Oedipus that the Sphinx 
was most generally familiar to Greek art. By far her most frequent 
appearance was on sepulchral monuments, as an emblem of the uncon- 
querable and inscrutable power which lays man low,—as the Seiren, 
from another point of view, was similarly applied. But the Oedipus 
myth illustrates in a very striking manner the essential traits both in the 
Asiatic and in the Hellenic conception of the Sphinx. 

(1) Zhe Sphinx oppresses the Thebans. This belongs to the original 
essence of the Sphinx idea, as a manifestation, in mind and body, of 
a force with which mortals may not cope. A grave of the Egyptian 
Thebes shows a bearded Sphinx, with one of its feet on three men 
(Lepsius, Denkm. v. 3. 76 c). An Attic vase shows two Sphinxes, with 
a prostrate man between them. A bowl found at Larnaka represents 
winged griffins and Sphinxes, with a man held captive (Milchhoefer 1. ἃ 
57, 51). The pitiless female Sphinx of Greek mythology belongs to the 
same order of winged pursuers as the Harpies and the Gorgons. 


I5—2 ᾿ 


228 APPENDIX. 


(2) Zhe Sphinx asks a riddle. Were we seem to have a purely 
Hellenic graft on the Egyptian and Asiatic original. To the Greek 
mind, the half-human, half-leonine shape was itself a riddle, and—gzven 
the notion of oppressor—could have suggested the story. The Centaur 
was not characteristically an oppressor of man; in the Chimaera, nothing 
was human ; but in the Sphinx these conditions met, and the crouching 
posture suggested grim expectancy. 

(3) Zhe Sphinx sits on the Φίκειον ὄρος near Thebes. In the Hesiodic 
Theogony the Sphinx is called Φίξ (Φίκ᾽ ὀλοήν, 326). Which was older,— 
the name of the hill, or Φίξ as a name for the monster? If the former, 
then we might well suppose that the localising of the myth had been 
suggested by the accident of a hill with such a name existing near 
a town in which Phoenician and Egyptian influences had long been 
present. 

(4) The Sphinx ts vanquished by Oedipus. ‘This is hyperbole clothed 
in myth. ‘He is so acute that he could baffle the Sphinx.’ For it isa 
distinction of the monumental Sphinx that it never appears as tamed or 
vanquished. The man-headed lions and bulls of Assyria, as Layard 
pointed out, are symbols of hostile forces which have been subdued and 
converted to the service of the conqueror. It is never so with the 
Sphinx of Egyptian, Asiatic, or Hellenic art. 

In conclusion, I may notice the most recent addition—a brilliant 
one—which has been made to the known examples of the Greek winged 
Sphinx. Under the auspices of the Archaeological Institute of America, 
the site of the ancient Assos, opposite Lesbos, on the south coast of the 
Troad, has within the last two years been thoroughly explored by a 
mission of American scholars and archaeologists’. On Oct. 4, 1881, 
was found the fragment of a relief with winged Sphinxes, belonging to 
the Doric temple of Athene, which crowned the Acropolis of Assos. 
The date of the temple may be referred to the early years of the 
5th century B.c. The Assos relief exhibits two Sphinxes crouching 
face to face, and must have decorated the lintel above the central inter- 
columniation of the temple-front—having a heraldic significance, as the 
civic emblem of Assos, like the two crows of the Thessalian Crannon, 
the two axes of the Carian Mylasa, the two heads of Tenedos, and the 
like. Mr J. T. Clarke, in his excellent Report on the investigations at 
Assos, of which he has been the director, (p. 111) writes :— 

‘Of all the sculptures of Assos discovered by the present expedition, 
and in the Louvre’—[those namely given to France in 1838 by 
Mahmoud II., of which the most striking are the bas-reliefs of Centaurs] 
—‘the magnificent Sphinxes are by far the best preserved, they alone 
having been taken from a hard bed of mortar, which had long saved 
them from weathering. The carving of this relief is of a delicacy and 
vigour comparable to the best works of fully developed Greek art. 
Throughout the body the firm muscles and yielding cushions of flesh 
are indicated with an appreciation of natural forms which shows a 
distinct advance beyond the art of Mesopotamia, successful as were its 


1 In the Fortnightly Review (April, 1883) I gave some notes of a tour in the Troad 
(Sept. 1882) which included a visit to Assos. 


APPENDIX. 229 


representations of animals; while the decorative character of the 
composition is maintained by the admirable outline of paws, wings, and 
tail. The heads are of that archaic type familiar in Attic sculptures 
dating near the beginning of the fifth century B.c. The eye, though 
shown nearly in profile, is still too large,—the corners of the mouth 
drawn up to a meaningless smile. The Egyptian derivation of the 
Sphinx is more evident than is elsewhere the case upon Greek works, 
by the closely fitting head-dress, welted upon the forehead and falling 
stiffly behind the ears.’ 


622 ff. KP. τί δῆτα χρήζεις ; ἦ με γῆς ἔξω βαλεῖν ; 
OL. ἥκιστα' θνήσκειν od φυγεῖν σε βούλομαι 
ὡς ἂν προδείξῃς οἷόν ἐστι τὸ φθονεῖν. 
KP. ὡς οὐχ ὑπείξων οὐδὲ πιστεύσων λέγεις ; 
ΟΙ. * Χ * Χ % * 


KP. οὐ yap φρονοῦντά σ᾽ εὖ βλέπωβ. OI. τὸ γοῦν ἐμόν. 


In discussing this passage, I take first the two points which seem 
beyond question. 

I. ν. 624 ὅταν... φθονεῖν, which the Mss. give to Creon, belongs to 
Oedipus. The words προδείξῃς οἷόν ἐστι τὸ φθονεῖν can mean 
nothing but ‘ show forth [by a terrible example] what manner of thing it 
zs to envy, —how dread a doom awaits him who plots to usurp a throne 
(cp. 382). Ant. 1242 δείξας ἐν ἀνθρώποισι τὴν δυσβουλίαν | ὅσῳ μέγισ- 
τον ἀνδρὶ πρόσκειται κακόν. “1. 1382 καὶ δεῖξον ἀνθρώποισι τἀπιτίμια; 
τῆς δυσσεβείας οἷα δωροῦνται θεοί, Thuc. 1. 76 ἄλλους γ᾽ ἂν οὖν οἰόμεθα 
τὰ ἡμέτερα λαβόντας δεῖξαι μάλιστα εἴ τι μετριάζομεν. 6. 77 προθυμότερον 
δεῖξαι αὐτοῖς ὅτι οὐκ “Iwves τάδε εἰσίν. (For the ‘one of the threat, 
cp. also Ant. 308, 325, 77. 1110.) Eur. Heracl. 864 τῇ δὲ νῦν τύχῃ] 
βροτοῖς ἅπασι λαμπρὰ κηρύσσει μαθεῖν, | τὸν εὐτυχεῖν δοκοῦντα μὴ 
ζηλοῦν (said of the captive Eurystheus). It is a mere accident that zpo- 
δείκνυμι does not elsewhere occur as=to show /orth: that sense is as 
natural for it as for προδηλόω, προφαίνω, προκηρύσσω, etc. I do not 
think that ὅταν can be defended by rendering, ‘zen thou shalt first 
have shown,’—a threat of torture before death. This strains the words: 
and death would itself be the essence of the warning example. Read 
ws ἂν, in order that: as Phil. 825 ws ἂν eis ὕπνον πέσῃ. 

2. vV. 625, ws οὐχ ὑπείξων...λέγεις, which the mss. give to Oedipus, 
belongs to Creon. Spoken by Oed., ὑπείξων must mean ‘admit your 
guilt,’ and πιστεύσων ‘obey’ me (by doing so): but the only instance of 
πιστεύειν in this sense is 77. 1228 πείθου" τὸ γάρ τοι μεγάλα πιστεύ- 
σαντ᾽ ἐμοὶ | σμικροῖς ἀπιστεῖν τὴν πάρος συγχεῖ χάριν: with 1251 σοί γε 
πιστεύσας. But there (a) the sense of ‘obeying’ verges on that of faking 
onés word as warranty for the act: and (ὁ) πείθου, ἀπιστεῖν help it out. 
Here, Creon speaking, ὑπείξων means ‘consent to give me a fair hearing,’ 
—under the tests which Creon himself proposed (603 f.),—and πιστεύ- 
σων, ‘believe’ my solemn assurances. 

' 3. Verse 624 having been given to Oedipus, and v. 625 to Creon, 
will the passage have been healed if vv. 625 and 624 change places? I 


230 APPENDIX. 


think not. For v. 624 will then mean: ‘[I will yield, and believe you, 
only| when you have been made an example of envy’: to which Creon 
will reply, ‘Nay, I find you mad’ (ze what you call my envy is but 
remonstrance with your fo//ly). This is too disjointed. I have long 
thought, and still think, that a verse spoken by Oed. has dropped out 
after 625, as 15 explained in the commentary. 


762. azomros.—I believe that ἄποπτος has two distinct uses, and 
that a neglect of the distinction has made some confusion. (1) As a 
verbal adject. of passive sense: seer, though at a distance: Arist. Pol. 2. 
12 ὅπως ἄποπτος ἔσται ἡ Κορινθία ἐκ τοῦ χώματος: (2) in poetry and 
later prose, as an adject. meaning, ‘away from the sight of’: implying 
either (a) ‘seen only afar,’ ‘dimly seen’; or (ὁ) ‘out of sight of’, as 
here: Ze. not seen, or not seeing, according as the ὄψις is that of object 
or subject. Dionys. Hal. 2. 54 ἐν ἀπόπτῳ τίθενται τὸν χάρακα (of an 
ambuscade), ‘zz a place out of sight’ (not, ‘in a place seen afar’). 
ἄποπτος does not occur in the active sense parallel with (1), as = ‘seeing, 
though at a distance’: analogy would, however, warrant it: see on 515. 
Ast strangely gives “τὸ ἄποπτον, specu/a,’ quoting the Platonic Axiochus 
369 A, and Lidd. and Scott, referring to the same passage, give ‘70 
ἄποπτον, a look-out place, watch-tower’: but there ἐξ ἀπόπτου θεώμενος 
= ‘seeing afar off.” In this adverbial phrase (PA. 467 ἐξ ἀπόπτου 
σκοπεῖν, Galen 3. 222 ἐξ ἀπόπτου θεασάμενος) the word has sense (1), 
meaning, ‘so that the place at which you look is ἄποπτος to you.’ 


1090. With Nauck’s αὖρι or Wecklein’s ἦρι we must read Arndt’s ἢ 


σέ γ᾽ εὐνάτειρα (without τις) in v. 1101. αὖρι would be attractive if it - 


had better authority. But Nauck’s note is quite misleading when he 
describes it as ‘et auch von Aschylos (fr. 412, ugl. fr. 274) gebrauchtes 
Adverbium. Aesch. fr. 274, in Nauck’s ed., is simply this word, αὖρι- 
Paras, on which Hesych. 5.0. 1. p. 619 says: Αἰσχύλος τὸ αὔριον ἐπὶ τοῦ 
ταχέως τίθησι: where αὖρι for αὔριον is merely Pauw’s conjecture. And 
Aesch. fr. 412 (Nauck) is merely this conjectured αὖρι quoted from 
Hesychius s.v. αὐριβάτας ! In Bekker Axecd. p. 464. 9 we have aipiBa- 
τον" TO αὖρι τιθέασιν ἐπὶ τοῦ ταχέως καὶ τάχα, οὐκ ἀπὸ τῆς αὔρας, ἀλλὰ 
κατά τινα βαρβαρικὴν λέξιν, τάχα δὲ καὶ ἀπὸ τοῦ αὔριον : but there, too, 
αὖρι is no more than an inference from aipiBarov.—Dindorf changed 
οὐκ ἔσει τὰν αὔριον tO οὐκέτι τὰν ἑτέραν, reading in 1101 ἢ σέ γέ τις γενέ- 
τας. This metre would suit the tone of excitement, as in 77. 96 f.,, 
where Ἅλιον, “Aduov αἰτῶ is followed by τοῦτο καρῦξαι τὸν ᾿Αλκμήνας πόθι 
μοι πόθι παῖς : cp. Zr. 500 οὐδὲ τὸν ἔννυχον “Avdav, followed by ἢ Ποσει- 
δάωνα τινάκτορα γαίας. On this view of the metre, 1 conjectured τὰν ἐπιοῦ- 
σαν ἔσει for οὐκ ἔσει τὰν αὔριον. In Par. A τὴν ἐπιοῦσαν is written over τὰν 
αὔριον : and Par. B has the gloss κατὰ τὴν αὔριον πάνυ λαμπρὰν ἡμέραν. 
Since ἡ ἐπιοῦσα, without ἡμέρα, could mean ‘to-morrow’ (Polyb. 5. 13. 
10), a reader who took τὰν ἐπιοῦσαν here as = ‘the coming day’ might 
have written τὰν αὔριον above it, or in the margin ; and this more familiar 
phrase might have supplanted the other in the text. Then πανσέληνον 


APPENDIX. 231 


would be explained as = πάνυ λαμπράν, and the.whole phrase interpreted 
as in the gloss of Par. B, ‘the all-bright morrow’: οὐκ being added to 
complete the assumed trochaic metre. In 1101, where L has ἢ σέ ye 
θυγάτηρ | Λοξίου, I proposed to read ἢ σέ γ᾽ ἔφυσε πατὴρ | Aogias; but I 
have come to think that the traditional reading, τὰν αὔριον πανσέληνον, 
though undoubtedly strange, may be genuine, and that perhaps the 
safest course is to receive Arndt’s emendation ἢ σέ γ᾽ εὐνάτειρά τις in 
1101. At the same time I wish to leave my conjectures on record, as 
they have been favourably received by some scholars, and may possibly 
have at least a suggestive value. 


1137. ἐξ ἦρος eis ᾿Αρκτοῦρον. The significance of Arcturus in the 
popular Greek calendar. 


= ᾧ Ursa Major 
δχς * ᾿Ξ Ἔ-ὰς δ ω 
Rei Seg gee 
a ea aor 
ae y ἃ 


ὭΣ 
aX Arcturus 


Arcturus is from ἄρκτος and οὖρος, ‘watcher’ (akin to ὁράω, and to 
our ward)—the ‘bear-ward,’ the keeper, or deader, of Ursa Maior. This 
name was also given to the whole constellation Borys (‘ploughman’) of 
which Arcturus is the brightest star: Cic. Avat. 96 Arctophylax, vulzo 
gui dicitur esse Bootes. Greek writers speak of dpxrovpev ἐπιτολή not in 
a geometrical sense, but as meaning ‘earliest visibility’; and this in two 
distinct applications. 

(1) The season when Arcturus first begins to be visible, after sun- 
set, aS an evening star, shortly before the vernal equinox (March zo—2r). 
This is sometimes termed the ‘acronychal’ rising (from ἀκρόνυχος, on the 
verge of night). Hippocrates, who was the contemporary of Sophocles, 
and who illustrates the popular reckoning by Arcturus more clearly than 
any other writer, uses ἀρκτούρου ἐπιτολή in this sense without any quali- 
fying epithet, leaving the context to show what he means: περὶ διαίτης 
3. 68 (vol. vi. p. 598 ed. Littré) μετὰ δὲ ταῦτα [viz. when 44 days have 
elapsed from the winter solstice] ὥρη ἤδη ζέφυρον πνέειν, καὶ μαλακωτέρη 
ἡ ὡρη".. εἶτα δὲ [15 days later] ἀρκτούρου ἐπιτολή, καὶ χελιδόνα ὥρη ἤδη 
φαίνεσθαι, τὸν ἐχόμενον δὲ χρόνον ποικιλῴτερον ἤδη διάγειν μέχρις ἰσημερίης 
[the vernal equinox] ἡμέρας τριάκοντα δύο. 

(2) Far more commonly, ἀρκτούρου ἐπιτολή denotes the season 
when Arcturus begins to be visible as a morning star. This is termed 
the ‘heliacal’ rising (ἡλιακή), because Arcturus is then visible before 
sunrise. In the age of Hippocrates and Sophocles (say in 430 B.C.), 
Arcturus began to be thus visible about a week before the autumnal 
equinox, which falls on Sept. 20o—21; and, in the popular language of 
that age, ‘the rising of Arcturus’ commonly meant, ‘shortly before the 
autumnal equinox.’ Cp. Hippocr. περὶ διαίτης 3. 68 (vi. 594 Littré, before 


232 APPENDIX. 


the passage cited above) τὸν μὲν ἐνιαυτὸν és τέσσαρα μέρεα διαιρέουσιν, 
ἅπερ μάλιστα γινώσκουσιν οἱ πολλοί, χειμῶνα, ἦρ, θέρος, φθινόπωρον. καὶ 
(1) χειμῶνα μὲν ἀπὸ πλειάδων δύσιος ἄχρι ἰσημερίης ἠαρινῆς, (2) ἦρ δὲ ἀπὸ 
ἰσημερίης μέχρι πλειάδων ἐπιτολῆς, (3) θέρος δὲ ἀπὸ πλειάδων μέχρι ἀρκτού- 
ρου ἐπιτολῆς, (4) φθινόπωρον δὲ ἀπὸ ἀρκτούρου μέχρι πλειάδων δύσιος. 
Here he tells us that, according to the reckoning with which the Greeks 
of the 5th century B.c. were most familiar, the year was divided into 
four parts, thus: (1) Winter—from the setting of the Pleiads to the 
vernal equinox: (2) Sprzing—from the vernal equinox to the rising of 
the Pleiads: (3) Swsmer—from the rising of the Pleiads to the rising of 
Arcturus: (4) Autumn—from the rising of Arcturus to the setting of the 
Pleiads. In the sevenfold division of the year (noticed by Hippocrates 
in his περὶ “EBdouadwv), summer was subdivided into θέρος, early sum 
mer, and ὀπώρα, late summer: and the latter ended with the ‘ heliacal’ 
rising of Arcturus, as Galen 5. 347 says: ὅσοι τὸν ἐνιαυτὸν εἰς ἑπτὰ τέμν- 
ουσιν @pas, ἄχρι μὲν ἐπιτολῆς Tod κυνὸς (Sirius) ἐκτείνουσι τὸ θέρος, 
ἐντεῦθεν δὲ μέχρις ἀρκτούρου τὴν ὀπώραν. Hippocrates says that, in 
watching the course of maladies, particular attention should be paid to 
the stars, especially to the rising of Sirius and of Arcturus, and to the 
setting of the Pleiads; for these are the critical seasons at which diseases 
most often mend, cease, or enter on new phases: περὶ ἀέρων, ὑδάτων, 
τόπων τι (vol. 11. p. 52 ed. Littré). The short phrase of Sophocles, εἰς 
ἀρκτοῦρον, can be matched with several of his medical contemporary, 
showing how familiar the sign was: ἐπιδημ. 1. 2. 4 περὶ “ἀρκτοῦρον (=a 
little before the autumnal equinox), 7b. 1. 2. 7 πρὸ ἀρκτούρου ὀλίγον καὶ 
ἐπ᾽ ἀρκτούρου (dcfore, and at, his ‘heliacal rising’): περὶ ἀέρων κιτιλ. τὸ 
μήτε ὑπὸ κύνα μήτε ἐπὶ τῷ ἀρκτούρῳ (neither just before Sirius rises, nor 
just when Arcturus does so). For the Roman writers, though Arcturus 
had no longer the same importance as a mark of the people’s calendar, 
he is especially the symbol of equinoctial storms in September: Plaut. 
Rudens prol. 69 Mam Arcturus signum sum omnium acerrimum: Vehe- 
mens sum exortens: cum occido, vehementior. Cp. Horace Carm, 3.1. 27 
saevus Arcturi cadentis Impetus. Plin. 18.74 (Arcturus rises) vehementissimo 
significatu terra marigque per dies quingue (indicated as Sept. 12—17). 

A passage of curious interest is Plin. 2. 47 usgue ad sidus Arcturi, 
guod exoritur undecim diebus ante aeguinoctium auctumnt. Here Pliny 
ireats the ‘heliacal rising’ of Arcturus as an event of fixed date, 
occurring annually about Sept. 9 or το. But, owing to the precession of 
the equinoxes, this ‘heliacal rising’ becomes progressively later,—as 
will be seen below, about one day later in every 70 years. In Pliny’s 
time (about 70 a.D.) the earliest time at which Arcturus could have 
been seen before sunrise would have been considerably later than 
Sept. 9 or 10. It would seem, then, that Pliny had taken his date 
from a literary source long anterior to his own age. On this point, 
Professor G. H. Darwin has kindly given me the subjoined note :— 

‘A rough calculation gives the following results with respect to the 
rising of Arcturus in the latitude of Athens (38° N.):— 

‘In 430 B.c. the rising of Arcturus (R.A. 185°, decl. 32°) preceded 
that of the sun 


a ee ee 


APPENDIX. jae 


on 7 Sept. (N.S.) by 22 minutes, 
and on 15 Sept. by 61 minutes. 


‘In 70 A.D. the rising of Arcturus (R.A. 191°, decl. 29°) preceded 
that of the sun 
on 15 Sept. by 23 minutes, 
and on 22 Sept. by 62 minutes. 


‘After a star has risen it remains invisible for some time on account 
of mist on the horizon, but if the climate be clear the interval of 
invisibility after geometrical rising is short. It is of course also in- 
visible in the day time and shortly after sunset or before sunrise. If 
therefore a star only rises in the geometrical sense a short time before 
sunrise, it will remain altogether invisible. From the above results 
we see that on Sept. 7, 430 B.c. and on Sept. 15, 70 A.D. Arcturus 
though really above the horizon before sunrise must have been in- 
visible on account of the brightness of the twilight. On the 15 Sept. 
430 B.c. and on the 22 Sept. 7o A.D. it must have been visible after 
geometrical rising, and before there was so much daylight as to ex- 
tinguish stars of the first magnitude. It is likely that Arcturus would 
have thus been first visible as early as 12 Sept. 430 B.c., and as 
20 Sept. 70 a.D. The first visibility of Arcturus took place between 
seven and eight days earlier in the month in 430 B.c. than in 70 A.D. 
In a clear climate like that of Greece the first visibility, after the 
period of invisibility due to the nearness of the sun, would fix the 
time of year within two or three days. At this season the rapid 
decrease of the sun’s declination conspires with the increase of his 
right ascension to produce a rapid increase in the interval by which 
the rise of Arcturus precedes that of the sun. As above stated, this 
interval would increase from 22 to 61 minutes between Sept. 7 and 
15, 430 BC. In a week after Sept. 15 the star would have risen long 
before sunrise, and the appearance of the star in the east and the 
rapidity of its extinction by the rays of the sun would cease to bea 
remarkable phenomenon.’ 


1505. μή ode wepiidys.—Porson on Med. 284 holds that Tragedy 
never admitted περί before a vowel (whether the prep. stood alone or was 
compounded with another word) in senarii, in trochaics, or in a regular 
system of anapaests. In Ar. ZZ. 1070 περίαλλα occurs in an anapaestic 
verse from Eur., but this, says Porson, seems to have belonged to a 
free or irregular system (systema illegitimum). In Soph. fr. 225 περίαλλα 
belongs to lyrics: so περιόργως (not a certain reading) in Aesch. Ag. 216: 
περιώδυνος 2b. 1448: and περιώσια Soph. fr. 611. Where a compound of 
περί occurs elsewhere than in lyrics, Tragedy, Porson says, used tmesis : 
as Eur. Bacch. 619 τῷδε περὶ βρόχους ἔβαλλε : fr. ap. Cornut. De WV. D. 
184 κορυφὴ δὲ θεῶν ὁ περὶ χθόν᾽ ἔχων | φαεινὸς αἰθήρ. Similarly such a 
form as ἠμφιεσμένος (Ar. Lec. 879) belongs to Comedy, not Tragedy. 
Here, then, he would write παρά σφ᾽ ἴδῃς (the Mss. having παρίδῃς): 
Fritzsche, περί of ἴδῃς. But it may be urged: (1) such a tmesis is 
alien from the style of ordinary tragic dialogue: (2) the extant remains 
of Attic Tragedy justify Porson’s remark that compounds of περί were 


234 APPENDIX. 


avoided, but are too small to warrant a rule absolutely excluding them : 
(3) the probability of such a rule, intrinsically slight, is further lessened 
by the περίαλλα of the Euripidean anapaest: (4) one reason why περί 
before a vowel should be usually avoided is evident: a compound with 
ἀμφί would in most cases express the same notion, without resolving 
the foot: 4.9. ἀμπέχω, ἀμφίστημι dispensed with need for περιέχω, 
περιίΐστημι. A single example like our passage goes far to break down 
the assumed universality of the exclusion. 


1526. οὗ ris οὐ ζήλῳ πολιτῶν ταῖς τύχαις éréBAerev.—Lucian once 
uses the verb ἐπιβλέπω with a dative, Astrol. 20 (where he is imitating 
an Tonic style) καί σφισι γιγνομένοισι τῷ μὲν ἡ ᾿Αφροδίτη τῷ δὲ ὁ Ζεὺς 
τῷ δὲ ὁ “Apys ἐπέβλεψαν (looked favourably upon). Plutarch (Caes. 2) 
has τοῖς χρήμασιν ἐποφθαλμιώντος, ‘eyeing the money’ (covetously), 
but that proves nothing for ἐπιβλέπω. ἐπιβλέπω usually takes either 
(2) an accus. with preposition of an object towards whom one looks,— 
eis ἡμᾶς Plato Phacdr. 63 A, ἐπὶ τὴν Θηβαίων πόλιν Deinarch. or. 1 ὃ 72: 
or (ὁ) a simple acc. of a ‘thing which one mentally considers: as λόγους 
Plat. Lege. τυ; ἀτυχίας, cupdopas [socrs or, 3 $$ σὺν 35.) Are we 
warranted, then, in rendering, ‘not looking jealously on the prosperity 
(ζήλῳ, or as Prof. Kennedy translates it, the aspiring hopes) and fortunes 
of the citizens’? 

I take ζήλῳ as a dative of manner with ἐπέβλεπεν. Thebans 
viewed Oedipus, not with jealousy, but with ζῆλος, ze. with a sense 
that he was the type of perfect good fortune, the highest model 
for aspiring effort. ζῆλος is felt by one who 15 impelled to lift himself 
towards the level of a superior; φθόνος, by one who would depress 
that superior to his own; when they are mentioned together, it is 
because baffled ζῆλος often breeds φθόνος : Plat. Alenex. 242 A πρῶτον 
μὲν ζῆλος, ἀπὸ δὲ ζήλου φθόνος. Cf. Eur. Suppl. 176 ff. σοφὸν δὲ 
πενίαν T εἰσορᾶν τὸν ὄλβιον, | πένητά tT ἐς τοὺς πλουσίους ἀπο- 
βλέπειν | ζηλοῦνθ᾽, ἵν᾽ αὐτὸν χρημάτων ἔρως ἔχῃ, Ζ.6. that his ζῆλος of 
the prosperous man may spur him to honourable exertion, The chief 
reason for preferring οὗ... ταῖς τύχαις to ee 5 ὅν. -τῆς τύχης is that 
the latter is so much further from the Mss.: the usage of ἐπιβλέπειν also 
favours the former. The reading of the MSS., ὅστις...καὶ τύχαις ἐπι- 
βλέπων, is nonsense. We cannot supply ἦν with the participle. 

Prof. Kennedy, reading ws τις, renders: ‘mighty man he was, for 
one who never eyed jealously the aspiring hopes and fortunes of the 
citizens’: 2.5. he was as powerful as a τύραννος could be who refrained 
from jealously suppressing all eminence near him. This version raises 
the question noticed above—as to whether ἐπιβλέπων would have been 
used, without any addition, in the sense of zmvzdens. As regards the 
sense, we scarcely seem to need here a clause which qualifies and 
restricts the former mzght of Oedipus, even though this clause at the 
same time implies a tribute to his moral greatness. - 


EN DIGES: 


I. GREEK. 


The number denotes the verse, in the zofe on which the word or matter 15 illustrated. 


When the reference is to the critical note, cr. is added to the number. 


When 


the reference is to a page, p. is prefixed to the number. )( means, ‘as distinguished 


from.’ 


A 

ἀβλαβής as a cretic, 229 
ἀγηλατεῖν, ἄγος, 402 
ἀγκύλη, 204 
ἀγνώς, act. and pass., 677 
ἀγροί, opp. to πόλις, 1049 
ἀγρόνομοι πλάκες, 1103 
ἀγύρτης, 387 
ἀγχόνης κρεῖσσον, 1374 
ἁδύπολις, 510 
ἀελλάδες ἵπποι, 466 
ἅζομαι, 155 
ἀθέως, 254 
ἄθικτος, of Delphi, 898 
ἄθλιος, of folly, 372 
αἰδοῦμαι with (1) accus. of pers., (2) infin. 

of act, 1427 
αἰθήρ γ( οὐρανός, 866 
αἰκάλλειν, 597 
αἷμα αἱρεῖν, 996 
αἷμα ἐμφύλιον, 1406 
αἱματοῦς, 1279 
αἱρεῖν, to ‘take,’ or ‘slay,’ 996 
αἴρεσθαι πένθος, 1225 
αἰσυμνήτης )( τύραννος, p. 5 
αἰώρα, 1264 
ἀκούειν, to be called, 903 ᾿ 
ἀκτὴ (βώμιοΞς), edge of, 182 
ἄκων = ἀκούσιος (of an act), 1229 
ἀλέξομαι as future, 539 


ἄληθες; 350 

ἀλλά, puts and meets an objection, 1375 

GANG... MEV δή, 523 

ἄλλος, ὁ, idiomatic use of, 290 

ἄλλος redundant, 7 

ἄλλος omitted (οὔτις, ἀλλά), 1331 

ἄλλως τε, ‘and moreover,’ 1114 

ἄλοκες, in fig. sense, 1211 

ἀλύειν, 695 

ἀμφιδέξιοι ἀκμαί, 1243 

ἀμφιπλὴξ ἀρά, 417 

᾿Αμφιτρίτης μέγας θάλαμος, 194 

ἁντεἃ ἄν, 281, 749 

ἄν, ellipse of, with imperf. (ἐβουλόμην), 
1348; (ἔδει), 256, 1368 

ἄν omitted after és with subjunct., 1231 

ἄν with infin. or partic., 11 

ἄν with partic. or infin., limit to use of, 
523 | 

dy repeated, 139, 339, 862, 1438 

ἄν before verb corrupted to dva-, 1348 

ἀναγιγνώσκειν not found in Attic prose as 
= ‘to recognise,’ 1348 

ἀνάγκη, a constraining doom, 877 

ἀνακηρύσσειν, 450 

ἄναξ, of a god and of a seer, 284 

ἀναπλάκητος, 472 

ἀναπνεῖν, to revive, 1221 

ἀναρρηγνύναι, intrans., in fig. sense, 1075 

ἄνδρα, accus. defore infin., in a γνώμη, 314 


236 


ἀνδρηλατεῖν, 100 

ἄνευ, senses of, 1463 

ἀνήκεστον, of a μίασμα, 98 

ἀνθ᾽ ὧν -- ἀντὶ τούτων, 264 

ἀντιλαβή, 626 

ἀνύειν with adj., to make such or such, 
166 

ἀξιοῦσθαι, to be condemned (with infin.), 
1449 

ἀπαυδᾶν in commands, 236 

ἀπείρων = Arreipos, 1088 

ἀπευθύνειν, to steer aright, 104 

ἀπήνη, 753 

ἁπλοῦν, εἰς, 519 

ἀπό )( ἐκ, of source, 395 

ἀπό, sense of, in compound adjectives, 196 

ἀπό )( παρά or πρός τινος, 42 

ἀποικεῖσθαι, pass., bold use of, 997 

ἀποκλίνειν, intrans., 1192 

ἀποκρίνειν, 640 

ἀπονοσφίζειν, 480 

am déevos, 196 

ἀπόπτολις, exile, 1000 

ἄποπτος, two senses of, p. 230 

ἄποπτος ἄστεως, 762 

ἀποσπᾶν ἐλπίδος τινά, 1432 

ἀποστερεῖν ἑαυτὸν τῆς πόλεως, 1381 

ἀποστρέφειν χέρας, 1154 

ἀπότομος ἀνάγκη, 877 

ἀπότροπος, 1314 

ἀποφάσκειν, 483 

ἄρα equiv. in sense to ἄρ᾽ οὐ, 822 

ἀρά-Ξ ἐρινύς, 417 

ἀραῖος, bound by an oath, 276 

ἀραῖος δόμοις, sense of, 1291 

ἀραῖος ὁλοίμην, 644 

ἀργός, senses of, 287 

ἄρθρα ποδῶν, 718; κύκλων, 1270 

ἀριθμός, of plural number as opp. to sin- 
gular, 844 

ἄριστα, adv., 1369 

ἀρκτέον, ‘one must rule,’ 628 

ἁρμόζειν, absol., of oracles, to come true, 
902 

ἄρουρα, fig. sense of, 1257 

ἄρρητ᾽ ἀρρήτων, 465 

“Apres ἀμφίπυρος, 207 

ἄρχειν )( κρατεῖν, 54 


INDICES. 


ἄστροις ἐκμετρεῖσθαι γῆν, 795 

ἀσχάλλειν, 937 

ἀτελεύτητος, 336 

ἄτιμος with genit., 788 

ἀτλητεῖν, 515 

αὐθαδία, not necessarily stupid, 550 
αὔξειν, to reflect honour upon, 109g! 
αὔριον always adv., 10go 

αὐτός, ‘unaided,’ 221, 341 

avrés=‘at once’ (ἀδελφὸς καὶ πατήρ), 458 
airés=‘ unaltered in opinion,’ 557 
αὑτοῦ-Ξ ἐμαυτοῦ, 138 

αὕτως, sense and accent of, 931 

ἀφανὴς (Adyos), unproved, 656 

ἀφιέναι ἑαυτόν, to absolve oneself, 707 
ἀφικνεῖσθαι ἐπὶ πάντα, 265 

ἀφόβητος, ‘not fearing,’ with genit., 885 
ἄψαυστος -- οὐ ψαύσας, 969 

ἄψορρος, 431 


Β 


βαιός = with few attendants, 750 
βακχεῖος θεός, 1105 

βάλλειν ἐν αἰτίᾳ, 656 

βάλλειν ἐς θυμόν, 975 

βαρύς, of vehement wrath, 673 
βάσανος, 493 

βασιλεύς, title of Zeus, 903 
βέλη θυμοῦ, θεῶν, 893 

βόσκειν τκετρέφειν, 1425 
βουλήσομαι, 1077 

βούνομος )( βουνόμος, 26 


Ἰ 


yudoxos=guarding the land, 160 

γάρ, merely prefacing statement, 277 

γάρ, in elliptical sentences, 582 

γάρ, in assent, 1117 

γάρ, in negation, 1520 

γε, scornful (σύ γε), 445 

γε...γε, 1030 ' 

γε, added to a repeated pron. (σέ...σέ ye), 
IIoI 

ye μέντοι, 442 

yé rot δή, 1171 

γένεθλα (πόλεως), her ‘sons,’ 180 

yevéras, senses of, 470 

γνωτός and γνωστός, 361, p. 225 


Ὁ (GREEK 


γονῇ γενναῖος, 1469 


Δ 

δάϊος, 214 

δάπτειν, of mental pain, 681 

δαφναφόρος, 21 

δ᾽ at end of verse, 29 

δέ, introducing a γνώμη, 110 

δέ, introducing objection, 379 

δέ, after σέ, etc., in addresses, 1097 

δέ, of apodosis after concessive protasis, 
302 

δέ, when attention is turned to a new 
point, 319 

δέ...γε, 1030 

δὲ οὖν, 669, 834 

δείκνυμι, of a warning example, p. 229 

δεῖμα, δείματα, 294 

δεινά, adv., 483 

δεινόπους ἀρά, 418 

δείξει, δηλοῖ, etc., sometimes impersonal, 
1294 

δεξιά, first sense of, 1243 

δεύτερα, τά, the second-best course, 282 

δή, as nearly = ἤδη, 968 

δηλαδή, 1501 

δήλημα, sense οὗ, 1495 

δην, adverbs in, 1310 

δῆτα, in assent, 445 

δι’ αἰθέρα τεκνωθέντες, 866 

διὰ τύχης ἰέναι, 773 

διαφέρειν, ‘bear to the end,’ 321 

διδακτός, opp. to ἄρρητος, 300 

δίδωμι λόγον ἐμαυτῷ, 583 

διειπεῖν, 394, 854 

διέχειν, trans. and intrans., 717 

δικάζειν, peculiar use of, 1214 

dixalws=‘in a strict sense,’ 853 

Δίκη, 274 

διολλύναι, to forget, 318 

διορίζειν, 723, 1083. 

διπλαῖ πύλαι, 1261 

δοκεῖν, to approve, 483 

δοκεῖν, (1) with infin. understood, (2) ‘to 
have repute,’ 1101 

δυοῖν, never a monosyllable, 640 

δυσούριστον, 1315 

δύσποτμος, of folly, 888 


237 
E 


ε elided after ἡ (εἴη ᾽ξ), 970 

ἔα, ἐᾷ, a monosyllable, 1451, 1513 

ἐγγενῶς, 1225 

ἐγκαλεῖν νεῖκος, sense of, 702 

éyKpaThs =e κράτει, 941 

éyxupwy (conjectured), 1031 

ἔγχος φροντίδος, of a device, 170 

ἐγὼ οὔτ᾽, 332 

ἕδος, sense of, 886 

ἕδρα, of supplication, 2 

εἰ with subjunctive, 198, 874 

el with fut. indic., 702 

el...ecre=elre...elTe, 92 

εἰ καί, 305: distinguished from καὶ el, 
Pp. 224 

εἴ τι μή, in diffident expressions, 124 

eldetre = εἰδείητε, 1046 

εἰδώς, with sure knowledge, 119 

εἰκάθω, 651 

εἰκῆ, sense of, 979 

εἰκός, τό, of a reasonable estimate, 74 

εἰμί understood with an adject., 92 

εἰμί with partic., instead of pres. or im- 
perf., 126 

elpyouat, to abstain from, 890 

els =continuous, 374 

εἷς, with superlat. (κάλλιστ᾽ ἀνὴρ els), 
1380 : 

εἰς ἑαυτόν, τό, in what concerns himself, 706 

εἰς καλόν, 78 

εἰς πάντας (αὐδᾶν), 93 

els τι φοβεῖσθαι, g80 

εἴτε, single instead of double, 517 

εἴτ᾽ οὖν... εἴτε, 1049 

ἐκ in adverbial phrases (ἐξ ὑπαρχῆς), 132 

ἐκ, of a former state (τυφλὸς ἐκ δεδορκό- 
TOS), 454 ᾿ 

ἐκ, of ultimate cause, 590, 1453 

ἐκ (μακροῦ), ‘at a long interval,’ 1141 

éx= ‘since’ (ἐξ οὖ), 1197 

ἐκ τῶνδε-- μετὰ τάδε, 282 

ἐκβάλλειν, to repudiate a statement, 849 

ἐκγενής (conjectured by Dind.), 1506 cr. 

ἐκδημεῖν, to be abroad, 114 


᾿ ἐκκαλεῖν, 597 


ἐκκινεῖν (ῥῆμα), 354 


ἐκλύειν δασμόν, 35 


238 


ἐκμετρεῖσθαι γῆν ἄστροις, 795 

ἕκμηνος, 1137 

ἐκπειρᾶσθαι, 360 

ἐκπέμπομαι, midd., 951 

ἐκτείνομαι, fig. sense of, 153 

ἐκτόπιος ἄγεται (instead of ἐκ τόπων), 1340 

ἐκτρίβειν, 428 

ἑκών Ξε ἑκούσιος (of an act), 1229 

ἐλαύνειν ἄγος, 98 

ἐλαύνειν és τριβάς, 1160 

ἐλευθεροῦν στόμα, sense of, 706 

ἐμπέφυκε, of prophecy, 299 

ἐμπλήσσειν, 1264 

ἐν = ‘in the case of,’ 388 

ἐν, of pursuit or calling (ἐν τῇ τέχνῃ), 562 

ἐν ἀργοῖς (πράσσεσθαι), 287 

ἐν γένει, 1016 

ἐν δέ, adverbial, 27, 181 

ἐν (δικασταῖς), ‘before judges,’ 677 

ἐν ὅρκῳ, 652 

ἐν σοί, penes te, 314 

ἐν σοί, ‘in thy mind,’ 770 

ἔν τινι ὁρᾶν and ἐνορᾶν τινι, 537 

ἐν τύχῃ, γήρᾷ, 80, 1112 

ἐν χεροῖν, dy his hands, 821 

ἐν αγής, ‘liable to a curse,’ 656 

ἐναριθμώ, 1187 

ἐνδατεῖσθαι, 205 

&v0a=éxeice ἔνθα, 796 

ἐνθύμιος, 739 ; 

ἐνταῦθα Ξε" ἴῃ that point,’ 598 

ἐξαγγέλλομαι, 148 

ἐξάγγελος, 1223 

ἐξαιρεῖν, to put out of account, 908 

ἐξελθεῖν, to be fulfilled, 88 

ἐξεστεμμένοι, said of suppliants, 3 

ἐξισοῦν, to bring to a (lower) level, 425, 
1507 

ἐξισωτέον, 408 

éés as= ‘thine,’ p. 6 

ἐπ᾽ ἀγρῶν and like phrases, 1049 

ἐπακούειν, 794 

ἐπεί =‘ for else,’ 390 

ἐπεύχομαι, 249 

ἔπι, adverb, 181 

ἐπὶ ἦρα φέρειν, 1095 

ἐπὶ ἠθέων λεκτοί (conject.), 18 cr. 

«ἐπί with dat. as= ‘against,’ 508 


INDICES. 


ἐπὶ τῷ dvdpl=in his case, 829 

ἐπὶ φρόνιμα ἄπορος, 692 

ἐπιβλέπειν, classical use of, p. 234 

ἐπίκουρος, ‘avenging,’ 497 

ἐπιοῦσα, ἡ, 1090 

ἐπιρράσσω, 1244 

ἐπισκοπεῖν, sense of, 1529 

ἐπιστροφή, 134 

ἐπιτολὴ ἀκρόνυχος and ἡλιακή, p. 230 

ἐπιών, ὁ, the first comer, 393 

ἔπος, of an oracular response, 89 

ἔπουρος, 194 

ἐπῳδός, ἡ, distinguished from ὁ ἐπῳδός, 
Ρ- Ixvii 

ἐπώνυμος, uses of, 210 

Epyw, ἔρξω, ἔρξας, etc., 8go 

ἐρρύμην, aor. of ῥύω, 1351 

ἔρχομαι, to come to be (φονεὺς ἦλθον), 1357 

ἕσπερος θεός ="Atdns, 178 

ἑστία, of Delphi, 965 

εὖ, ‘carefully,’ 308 

εὖ διδόναι, to give good, 1081 

εὖ ἴσθ᾽ with hiatus, 959 

evayns λύσις, 921 

εὐέπεια, senses of, 932 

εὐθύ )( εὐθύς, 1242 

Εὔκλεια, title of Artemis, 161 

εὕρημα, 1107 

εὔσεπτος, act., ‘reverent,’ 864 

εὔχομαι, constr. of, 269 

εὐῶψ, epith. of comfort, 189 

ἐφυμνεῖν, of imprecation, 1275 

ἔφυν, of a natural claim, 9 

ἔχομαι, uses of, 891, 1387 

éxw, with part. of aor., 577, 698: of perf., 
701 

ἔχω, intrans. with adv. (Herod.), 708 

ἐῶραι, ai, the festival, 1264 


Z 
ζῆλος )( φθόνος, p. 234 
ζῆν, to be operative, 45 


H 


i, Ist pers. sing. imperf. of εἰμί, 1123 

%...4, where the first ἤ might be absent, 
487 

ἢ kal=than even, 94 


“>” -,;. ΎΣ 


| ANS ΟΥΑΙ. 239 


7 καί, in question, 368, 757 
ἢ οὐκ as one syllable, 555 
ἤ...τε instead of 4...4, 539 
noe, 3rd sing., 1525 
ἤδειμεν, Here, ἤἥδεσαν, 1232 
ἡδονά, form of, 1337 

ἡδύς = εὐάγγελος, 82 

ἤθεος, 18 

ἡκωε γέγονα, 1519 
ἦλθον = ἐγενόμην, 1357 
ἡλόμην and ἡλάμην, 1311 
ἦμος, in tragic dialogue, 1134 
ηὖγμαι, 1512 cr. 


Θ 


θάλαμος, 1241 

θανάσιμος βεβηκώς, 959 

θεῖα, τά, religion, 910 

θεῖος, epithet of kings, etc., 1235 
θελήσας, 649 

θεμιτός and θεμιστός, 993 

θεός, said of λοιμός, 27 

θεός, without art., 871 
θεσπιέπεια, a really pleonastic form, 463 
θεωρία, uses of, 1401 

θεωρός, to Delphi, 114 

θητεία )( δουλεία, 1029 

θίξομαι, 891 

θοάζειν, as=Odooev, 2, p. 206 
θυρών, 1241 

@w, verbal forms in, 651 


I 


ἰάκχιος, 1218 

ἰάλεμος, 1218 

ἰἀχεῖν, ἰακχεῖν, 1218 

ἱέναι ἐπί (accus.), to attack, 495 

ἱερός, epith. of ὄμβρος, 1428: and pds, 
1379 cr. 

ijos, 154, 1096 

ἴθι, in entreaty, 1468 

ἱκνεῖσθαι εἴς τι, to incur a fate, 1158 

ἱκτήριοι κλάδοι, 3 

iva, ‘where,’ 367 (with genit.), 687 (with 
ἥκειν), 947: limit to its use, 1311 

ἵνα, final, with imperf. and aor. indic., 
1389 

ἵνα μὴ εἴπω, 328 


ἴσα καίΞξεῖσα ὥσπερ, 1187 

ἴσα, τά, poet. for τὰ αὐτά, 1498 

ἴσος, adjectival compounds with, 478 
ἴσος; ‘just,’ 677 

ἰσοῦσθαι, passive, 31 

ἱστάναι ἐλπίδα, 698 

ἱστορεῖν, senses of, 1484 

.@ and -ίσω, futures in, 538 

ἐῶν, pres., not fut., partic., 773 


K 

Kad’ ὑπερβολήν, 1197 

καθικνεῖσθαι, construct. of, 809 

καί, emphasizing verb, 851, 98g, 1129 

kal, ‘e’en,’ where the speaker is diffident 
(κἀν ἐμοί), 1239 

kal=adeo, 347 

kal=6re, 718 

καὶ (δεῦρ᾽ ἔβημεν) = ‘in the first instance,’ 
148 

καί... καίΞε ‘both, and (yed),’ 413 

kal μήν, ‘indeed,’ 749, 1004 

kal μήν γε, 345 

kal σύ, ‘thou on thy part,’ 342 

kal ταῦτα, 37 

καιρός, with art., 1050. 

καιρῷΞεἐν καιρῷ, 1516 

κακός = δυσγενής, 1063 

κάλλος, concrete, a fair thing, 1396 

καλῶς, colloquial use of, 1008 

κατά, with acc. of respect, 1087 

κατά, after its case, 1280 

κατὰ ἑαυτόν, = ‘alone,’ 62 

κατὰ στέγας ἱέναι, 637 

κατακοιμᾶν ὄμμα (of deathlike anguish), 
1222 

κάταργμα, sense of, 920 

κατάφημι )( ἀπόφημι, 507 

κατεύχομαι, 246 

κατέχω, intrans. (to restrain oneself), 782 

κεκλαυμένος, 1400 

κέντρα διπλᾶ, 809 

κέντρα, fig., 1318 

κέρδος, material gain, 595 

κεύθειν, to be hidden, 968 

κήδευμα, of a brother-in-law, 85 

κηλὶς συμφορᾶς, 833: 

Κῆρες )( Μοῖραι, 472 


240 


κλάζειν, of birds, 966 

κλαίων, ‘to thy cost,’ 401 

κλήζομαι )( καλοῦμαι, 1451 

κλῇθρα, door-bolts, 1261 

κοινός = kotvwvds, 240 

κολάζειν, of verbal reproof, 1147 

kp, vowel long before, 640 

κρείσσων εἶ μὴ wv=Kpelacdy ἐστί σε μὴ 
εἶναι, 1368 

κτῆμα, of mental or moral qualities, 549 

κυκλόεις ἀγορᾶς θρόνος, 161 

κύριος, 1506 

κύων, said of the Sphinx, 391 

κωφὰ ἔπη, 290 


A 


λαμβάνειν (dpatov), 276 

λάμπειν, said of sound, 186 

λέγειν, of mere talk, 360 

λέγω δέ, as an exordium, 412 

λέγω τι; 1475 

λείπειν, intrans., to stop short, 1232 
λήγειν, fig., of rumour, 731 

λήθω, parts of, used by Soph., 1325 
λιμήν, poet. for ὑποδοχή, 420, 1208 
λόγων δόκησις, κόμπος, 681 

Λοξίας, 854 

λοχῖται, a king’s body-guard, 751 
Avew, with simple genit., 1350 
λύειν τέλη ελυσιτελεῖν, 317 

Λύκειος, epith. of Apollo, 203 


M 


μάγος, 387 

μακραίωνες, ai, the Nymphs, Tog 
μαλερός, 190 

μάλιστα, of one’s first wish, 926 
μάντις, said of (1) god, (2) man, 708 
μάντις, ‘prescient,’ 1086 

ματάζω, ματάζω, 891 

μέγα, adv. with adj., 1341 

μεγάλη θάλασσα, ἡ, 194 

péyas=in a strong (moral) position, 652 
μεθιέναι λόγον, 784 

μείζονα τῶν μακίστων, 1300 

μείζων, ‘nearer and dearer,’ 772 
μέλλω, fut. or aor. after, 967 
μεμνώμεθα, subjunct., 49 


INDICES, 


μέν, clause with, without expressed an- 
tithesis, 18 

μὲν οὖν, where each word has a separate 
force, 483 

μὲν οὖν, as=‘ nay rather,’ 705 

μεριμνᾶν, uses of, 1124 

μέσης (ἐξ ἀπήνης), ‘right out of,’ 812 

μεσόμφαλος, of Delphic oracle, 480 

μέτεστί μοι πόλεως, sense of, 630 

μέτοικος, sense of, in poetry, 452 

μή, generic, 397, 638, 875, 1019 

μή, where μὴ οὐ could stand, 1388 

μή before the infin., where οὐ could 
stand, 1455 

μή, in a saving clause (with partic. un- 
derstood) = el μή, 1457 

μὴ οὐ, with partic., 13, 221, p. 221 

μὴ ov, τό, with infin., 1232 

μὴ )( οὐ παρὼν θαυμάζεται, 289 

μηδέ, irregularly equiv. to μὴ καί, 325 

μηδείς, ὁ, ‘he who is as nought,’ 1019 

μηδέν, τό, ‘what is as nought,’ 638 

μηδέν, τό, adverbial with ἑῶσας, 1187 

μηδὲν εἰδώς, ὁ (instead of οὐδέν), 397 

μήτε, understood, 239 

μία ῥώμη Ξε ἑνὸς ῥώμη, 122 

poo=‘as I bid you,’ 1512 

μοῖρα, how far personified, 863 

μονάς, 1350 

μόνιμος, 1322 

μόνος, not ‘alone,’ but ‘pre-eminently,’ 
299 

μονῳδίαι, structure of, p. xxviii 

μοῦνος, in dialogue, 304 

μοῦνος, supposed limit to its use by Soph., 
1418 


N 


ναίειν ὁμοῦ (said of feelings, etc.), 337 
νέμω, of sway, 579 

νηλής )( ἄνοικτος, 180 

vigfew, special sense of, 1228 

viv, accus. plur., 1331 

vouds, use of, 1350 

νόμος ἴδιος and κοινός, 865 

νῦν δέ, with aor. equiv. to perf., 263 
ywudw, senses of, 300 

νωτίζειν, 192 


Le? GICLEE, 


=) 


ξεῖνος for ξένος in dialogue, 1418 

ξένη = ξένη γῆ, 455 

ξυμφοράς, τάς, τῶν βουλευμάτων, 44, Ῥ- 
207 


O 


οἷα impossible after ὅτι in 1401 

οἷα (δοῦλος, ‘for a slave’), rarer than ὡς...» 
763 

οἶδα )( γιγνώσκω, 1128 

Οἰδίπους as vocative, 405 cr. 

οἰκεύς -- οἰκέτης, 756 

οἶμαι, only sometimes parenthetic, 1051 

οἰόζωνος, 846 

οἷον (after τοιοῦτον) instead of ὥστε, 1293 

οἷσθ᾽ ws ποίησον; 543 

ὀλέθριος, pass., ‘lost,’ 1341 

ὄλεθρος, colloquial use of, 1341 

Ὄλυμπος, the sky, 867 

ὅμαυλος )( σύμφωνος, 186 

ὄμβρος, symbol of water generally, 1427 

ὁμιλίαι ἀστῶν, sense of, 1489 

ὁμογενής, sense of, 1362 

ὁμόσπορος, 260, 460 

ὁμόστολος, ‘roaming with,’ 212 

ὁμοῦ, senses of, 1276 

ὀμφαλός, the Delphic, 480, 898 

ὄνομα κακοῦ Ξ-- κακὸν ὀνομαζόμενον, 1284 

ὄντες, etc., with a numeral (δύ᾽ ὄν τε), 1505 

ὀπίσω, of the future, 486 

ὅπως μή, after verb of fearing, 1074 

ὅπως πέμψεις, ‘(see) that you send,’ 1518 

ὁρᾶν τὰ αὐτά, sense of, 284 

ὁρᾷς ; in reproach, 687 

ὀρθός, ‘justified,’ 506 

ὅρκος θεῶν, 647 

ὄρμενος, aor. part., ‘sped,’ 177 

ὄρνιθι αἰσίῳ, 52 

ὃς dv δέ instead of ὃς δὲ ἄν (in prose), 749 

ὅσον μή, with partic., 347 

ὅσος with causal force (ξξὅτι τοσοῦτοΞ), 
1228 

ὅστις with superl., εἰμί being understood, 
344, 663 

ov γὰρ av, with protasis suppressed or ex- 
pressed, p. 221 


es tg 


241 


οὐ yap δή, 576 

οὐ (τὸν θεόν) Ξ- οὐ μά, 660 

οὐδ᾽ ἂν εἷς, 281 

οὐδὲ μήν, ‘no, nor,’ 870 

οὐδεὶς ὃς οὐχίΞτε πᾶς τις, 373 

οὐδὲν (instead of οὐδεὶς) βροτῶν, 1195 

οὐκ εἰς ὄλεθρον ; 430 

οὐκ ἴσος, more than equal, 810 

οὕνεκά twos, so far as it is concerned, 
858 

οὔπω instead of οὔποτε, 105 

οὕπω ironically, 594 

οὐρανία αἰθήρ, 866 

ὅτε, ‘seeing that,’ Ξε ἐπειδή, 0918 

οὔτις, ἀλλά, for οὔτις ἄλλος, ἀλλά, 1331 

οὗτος σύ, 532 

οὕτως divided from its adjective, 1444 

ὀφθαλμός, fig. sense of, 987 


II 


πάγκαρπος, epith. of laurel, 83 

πάθος, euphemistic, 840 

παθών, by bodily pain, 403 

Παιάν, of Apollo, 154 

παιδουργία for madoupyés, 1248 

πάλαι, of a recent moment, 1161 

παλαιός, joined with ὁ πρίν (not a pleo- 
nasm), 1282 

πάλαισμα, of civic emulation, 880 

πάλιν, redundant, 430 

πάλλω, trans. and intrans., 153 

πᾶν δρᾶν, etc., 145, 265 

πανσέληνος (wpa), τορο 

πάντα, adv. neut. plur., 475, 1107 

παντελής, of a wife, 930 

παρ᾽ οἴνῳ, 780 

παρ᾽ οὐδέν, 983 

παρά in τὸν παρ᾽ αὑτῷ βίοτον, 612 

παραμείβειν, to outstrip, 504 

παραρρίπτω, with partic., 1494 

παραχορήγημα, P. 7 

πάρεστιν, impers., ‘it can be done,’ 766 

παρέχειν )( ἔχειν, 567 

παρήχησις, rhetorical, 370 

παριέναι κέαρ, 688 

πάροδος of Chorus, 151 

πάτριος )( πατρῷος, 1394 


16 


242 


πατριώτης, said of a place in one’s native 
land, 1091 

πέλας, adv., with παραστατεῖν, 400 

πελασθῆναι, usu. with dat. in conjugal 
sense, 1100 

περᾶν (θυμοῦ), to go far 271, 673 

περί, compounds with, in tragic verse, 
Pp. 233 

περίαλλα, use of, 1218 

περιβόατος, 191 

περισσός, ‘of special note,’ 841 

περιτελλομέναις ὥραις, 156 

περόνη, a brooch, Τ2.0 

πέτομαι, aorist forms of, 16 

merpatos, a doubtful use of, p. 226 

πηγή, ἡ ἀκούουσα, 1386 

πημονή, quasi-colloquial use of, 363 

πίθεσθε )( πείθεσθε, 1414 cr. 

πίπτειν -- ἐμπίπτειν (as on a bed), 1210 

πίστιν φέρειν τινί, 1445 

πίστις, senses Of, 1420 

πλάνης, 1029 

πλάνος, πλάνη, 67 

πλαστός, 780 

πλέον τι, ‘some advantage,’ 37 


πλησιάζειν = πλησίον εἶναι, gi: with dat., 


1134 
Πλούτων, name for Hades, 30 


ποικιλῳδός, chanting 77ddles, 130 

motos Κιθαιρών -- ποῖον μέρος Κιθαιρῶνος, 
421 

πόλις, the, exists where its men are, 56 

πόλις, indignant appeal to, 629 

πόλις, adjectives compounded with, 510 

πολύζηλος, senses of, 381 

πολύς, of strong rumour, 785 

πολὺς pet, etc., of vehement speech, etc., 
750 

πομπός, 288 

πόποι, 167 

mwoté=tandem aliquando, 335 

ποῦ; ‘on what ground?’ 355 

ποῦ; ‘in what sense?’ 390 

πράσσειν, ‘put into act,’ 69 

πράσσειν, of intrigue (pass.), 124 

πράσσεσθαι, midd., senses of, 287 

πρεσβύτερον, ‘more serious,’ 1365 

πρίν, with indic., limit to use of, 776 


INDICLES. 


πρό )( ἀντί, ὑπέρ, πρός with gen., 10, 134 

προδείκνυμι, of a warning example, p. 229 

προδεικνύναι γαῖαν, 456 

προδείσας )( ὑπερδείσας, 80 

πρόμος θεῶν, of the Sun, 660 

πρόνοια, Classical use of, 978 . 

προζενεῖν, senses of, 1482 

προπηλακίζω, 427 

προπονεῖν, senses of, 685 

πρός following its case, 178 

πρός, with dat., after verb of throwing or 
falling, 1302 

πρὸς δίκης, 1014 

πρὸς ποσί, τό, 131 

πρὸς σοῦ, ‘in thy interest,’ 1434 

προς τί, 766, 1027, 1144 

πρός τινος, ‘on one’s side,’ 134 

πρὸς Tivos αἰτίας ; 1236 

πρός τινος )( παρά τινος, 935 

πρὸς (τῷ δεινῷ), close to it, 1169 

πρὸς καιρόν, 325 

πρὸς χάριν, 1152 

προσάγεσθαι, 131 

προσάπτειν, intrans., 666 

προσήγορος, act. and pass., 1337, 1437 

προσήκειν, constructions of, 814 

προσθήκη, aid, 38 

προσκεῖσθαι, 232 

προσκυρεῖν with accus., 1298 

προσταθέντα, said of βέλεα, 206 

προστάτην ἐπιγράφεσθαι, 411 

προστατήριοι θεοί, 203 

προστάτης, champion, 882 

προστάτης νόσου, 303 

προστείχειν for προσστείχειν (MSS.), 79 cr. 

προστίθεσθαι μέριμναν, 1460 

προστρέπεσθαι, 1446 

πρόσωπον, τὸ σόν, ‘thy frown,’ 448 

προτέρον ὕστερον, the so-called figure, 827 

προφαίνειν, said of an oracle, 790 

προφαίνεσθαι, 395 

πυθμένες, sockets of bolts, 1261 

Πυθόμαντις ἑστία, 965 

πύματον (ὅ τι) ὀλοίμαν, 663 

πύργος (city-walls with towers), 56, 1378 

πυρφόρος, of pestilence, 27 

πῶς βλέπων; 1371 

πωτᾶσθαι, 1310 


1 GLEE Ἂς 


P 
ῥαψῳδός, of the Sphinx, 391 


ῥέπειν els τινα, 847 
ῥοπή = momentum, 961 
ῥύεσθαι (μίασμα), 312 


Σ 


σ᾽, elided, though emphatic, 64 

σαφής = ‘proved,’ 390 

σεμνόμαντις, ironical, 556 

σημάντωρ, 957 

σκοτεινός, of blindness, 1326 

σοί, not σοι, required, 435 

σπάργανα, fig. for infancy, 1035 

στάσιμον, Arist.’s definition of, p. 8 

στέγειν, classical use of, 11 

στέλλειν )( στέλλεσθαι, 434, 860 

orépéas, having formed a desire, 11 

στέφητεἱκετηρία, QI 

στόὀλοςΞελαός, 170 

στόμα, of a prophet, 426 

στόματα, said of one mouth, 1218 

συγγενής, with genit. or dat., 814 

συγγενής, said of πότμος, etc., 1082 

συλλαβών, colloquial force of, 971 

σύμμαχος, of gods, 274 

συμμετρεῖσθαι, 73, 963 

σύμμετρος, strengthens ξυνᾷάδειν, 1113 

σύμμετρος ws κλύειν, 84 

συμφορά, classical uses of, p. 212 

συμφορά, euphemistic for guilt, 99 

συμφορά, of a happy event, 454 

συμφυτεύειν, 347 

σύν, ‘by means of,’ 656 

σὺν ἀνδράσιν Ξεἄνδρας ἔχων, 55, 123 

σὺν γήρᾳ βαρύς, 17 

συναλλαγαὶ δαιμόνων, 34 

συνέρχομαι, to conspire with, 572 

συνέστιος, implying a share in family 
worship, 249 

συντιθέναι, to concoct a plot, 401 

συντόμως, 810 

σφας, σφέας, accent of, 1470 

σχιστὴ ὁδός, the, 733, 1398 

σχολῇ, adv., 434 

σῶμα δρᾶν κακῶς, sense of, 642 

σωτήρ, as epithet of τύχη, 80 


243 


zr 

τὰ δέ, answering to τὰ μέν understood 
(after ὅσα), 1229 

τὰ λῴστα ταῦτα (of which you speak), 
1067 

τάλας, last syllable long, 744 

re, irregularly placed, 258, 528, 694 

τε, linking the speaker’s words to those 
of a previous speaker, Ioor 

τε καί where καί alone would suffice, 487 

TEKOVTES, οἱ Ξε οἱ γονεῖς, 009 

τεκόντες, ol=6 πατήρ, 1176 

τέλει, proposed versions for in 198, p. 219 

τελεῖν (absol.), to perform (funeral) rites, 
1448 

τελεῖν els, 222 

τέλειος, τέλος, Of marriage, 930 

τερασκόπος, 605 

τέχνη, human skill, 380 

τῇδε...τῇδε (βλέπειν), to right or to left, 
857 

τηλικόσδε, ‘so young,’ 1508 

τηρήσας, 808 

τι, adv., ‘perchance,’ 969, 1401 

τί δ᾽ ἔστιν; 319, 1144 

τί δ᾽ ὅντιν᾽ εἶπε; 1056 

τί φημί; ἃ startled cry, 1471 

τί χρείας Ξετίς χρεία, 1174 

τιμωρεῖν, ‘to punish,’ 107 

ris and ὅστις combined, 72 

τις, indef., after noun with definite art. 
(ὁ κύριός Tts), 107 

Tis with adv. force (ταχύς τις -- ταχέως 
ws), 618 

τις for ὅστις only in indirect question, 1144 

tls (ἔβας); ‘in what spirit? 151 

τίς οὐτεπᾶς τις, 1526 

τίς οὗτος, τίς... ; for τίς οὗτος, ὅς, 1493 

τοιόσδε, after noun with 6 σός, 295 

τοιόσδε, in appos. with explanatory δά]. 
435 

τόκοι, labours of child-bed, 26 

τόσος, rare in Soph., 570 

τοῦ λέγοντος εἶναι, O17 

τοῦτ᾽ αὐτό, τοῦτο, 1013 

τοῦτο μέν.. τοῦτ᾽ ἄλλο, 605 

τρέφειν, said of the concomitants of one’s 
life, 374 | 


244 


τρίδουλος, 1062 
τρίτος, added, 581 
τυραννίς, of the king as embodying king- 


ship, 128 
τύραννος, earliest occurrences of the word, 
Pp: 5 


τύραννος, probable etymology of, 2d. 
TUpayvos=a ‘tyrant’ in our sense, 873 
τύχη, idea of, 977 


» 


ὕβρις, personified, 873 

ὑμέναιος )( ἐπιθαλάμιον, 422 

ὑπεξαιρεῖν, 227 

ὑπεξελών, proposed versions for in 227, 
Ῥ 222 

ὑπὲρ ἄτας, ‘to avert’ ruin, 165, 188 

ὑπερμάχεσθαι, ὑπερμαχεῖν, 265 

ὑπηρετεῖν νόσῳ, 217 

ὑπὸ μνήμη, 1131 

ὑπόρχημα, p. 1xxxv. 

ὑποστρέφεσθαι μερίμνης, 728 

ὕπουλος, 1396 

ὑποφορά, rhetorical, 1375 

ὑφέρπειν, of rumour, 786 

ὑφιέναι, to suborn, 387 

ὑψίποδες, epith. of νόμοι, 865 


Φ 


φαίνω, to set forth a story, 525 

φάσκειν, =‘be confident,’ 462 

φάτις, of a divine message, 151 

φέρειν )( φορεῖν, 1320 

φέρειν πίστιν τινί, 1445 

φέρεσθαι πλέον, to achieve more, 500 

φέριστε, ὦ, rare in trag., 1149 

φέρω: φέρομαι, 590 

φέρω ἁγνείαν, 863 

φεύγειν τι, to escape the penalty of it, 
359 

φῆμαι μαντικαί, 723 

φήμη )( ὀμφή and κληδών, 43 

φθερεῖσθαι, 272 

φίλοι, powerful friends, 541 

φοβεῖσθαι ἔς τι, οϑο 

φοίνιος, poet. for θανάσιμος, 24 

φοιτᾶν, sense of, 1255 

φοράδην, form and senses of, 1310 


INDICES. 


φρονεῖν, senses of, 326, 1520 
φρονήσας, ‘having become sane,’ 649 
φυλάσσεσθαι παρά τινι, sense of, 382 
φύσις (πέτρον, etc.), 334 


x 


χαίρω πᾶσι, sense of, 596 

χάλαζα, fig. uses of, 1279 

xetv, of song, etc., 1218 

χειρὶ τιμωρεῖν, as Opp. to a fine or to 
ἀτιμία, 107 

χειρόδεικτος, a ἅπαξ λεγόμενον, go2 

χείρωμα, 560 

χέρνιψ, 240 

χερσίν =simply ἔργοις, opp. to λόγῳ, 883 

Xnpevew, 479 

χθονοστιβής, 301 

χιασμός, rhetorical, 538 

χνοάζειν, 742 

χορεύειν, typifying public worship gene- 
rally, 896 

χορεύεσθαι, 1094 

xpela, ‘request,’ 1435 

χρείαν τινὸς ἐρευνᾶν, 725 

χρυσέα, epith. of Hope, 157 

χρυσομίτρας, epith. of Bacchus, 209 

χωρίς =‘ without evidence,’ 608 


2 


és, final, with aor. indic., 1392 

ws, as prep., 1481 

ὡς, marking the mental attitude of the 
subject to the verb, 848, 1174 

ws and ὥσπερ, in comparison, with ellipse 
of a verbal clause, 923, 1114, 1178 

ws, added to a genit. absol., 11, 145 

ws, with accus. absol., ror 

ws ἄν, as=‘in whatever way,’ p. 224 

ws dv μή, 328 

ὡς γυνή, ‘in a woman’s way,’ 1078 

ws (δοῦλος, ‘for a slave’), 763, 1117 

ὡς τεθραμμένον, ‘which (Ae says) has been,’ 
etc., 97 

wore, confirms and continues the last 
speaker’s words, 1036 

ὥστε γε, οὐχ, in reply, 1131 

ὦ τᾶν, 1145 


il, MATTERS. 245 


II. MATTERS. 


A 
Abae, temple at, goo 
abstract for concrete (τροφή θρέμματα), 
I, 1248, 1396 
‘accent’ defined, p. Ixiv. 
», Of Baxxelos, 1105 
>> οἵ κῆρυξ (not κῆρύξ) re, 802 
», Of προσθῇ, 1460 cr. 
»» Of verbal derivatives with short 
penult., 460 
accented forms of pers. pron. preferable, 
435s 5749 1479 
accus. absol., 1o1 
», after κυρεῖν, τυγχάνειν, 1298 
», after notion equiv. to transitive 
verb, 31 
», at beginning of sentence, without 
any regular government, 216, 278, 1134 
», before infin., where dat. could 
stand, 913 
», before infin. with εὔχομαι, 269 
»» cognate, 192, 264, 340, 422 
», cognate, denoting one’s errand 
(ἔρχομαι ἀγγελίαν), 788 
», cogn. to verb of feeling (τὸ ἔπος 
ἥδομαι), 936 
», double, after στέλλεσθαι, 434 
», in appos. with σέ, instead of a 
vocative, 1119 
»» in appos. with whole sentence, 603 
», Of antecedent, prefixed to relative 
clause, 449 
», Of person, after ἥκειν, 713 
9, Of place to which, 1178 
», temporal, almost adverbial in refer- 
ence to a season, 1138 
acting, probable style of old Greek, 
Ὁ. Xxxi. 
adj. agreeing with pers., instead of subst. 
with prep. (ἐκτόπιος ἄγομαι), 1340, p. 226 
», and adv. co-ordinated (ri ἢ νέον 7 
πάλιν Spds;), 155 


adj., comparative, to be carried on to a 


second clause, 1204 

»» compounded with noun of like sense 
with the subst. (βίος μακραίων), 518 

», compound, equiv. to two distinct 
epithets (oldfwvos), 846, 965 

», instead of adv. (ὕστερος), 222 

», instead of proper name in genit. 
(Λαβδάκειος παῖς), 267, 451, 1216 

», ΟΥ̓ pron., as epith. of a compound 
phrase (τοὐμὸν φρενῶν ὄνειρον, not τῶν 
ἐμών), 108 

,9,. second, as epithet, following subst. 
(τὰν γαμψώνυχα παρθένον χρησμῳδόν), 
1100, 1245 

»» simple, instead of adj. with ὦν, 412, 
1506 

», transferred from subst. in the gen. 
to its dependent subst. (τοσόνδε τόλμης- 


πρόσωπον), 532, 832, 1375 


», verbal, in -és, used as fem., 384 


” x,  sigmatic form of, p. 225 
: » With act. sense (ἄψαυστος), 
θοῦ. 


adv., neut. plur., 883 
Aeschylus, apparent reminiscence of, 1478 


+9 Theban trilogy of, p. xvi. _ 


Agenor, 268 

alliteration, rhetorical, 370 

altars on the stage, p. 10 

ambiguity of phrase, intended by the 


dramatist, 137, 261, 572, 814, 1167 


anacolouthon (dat. for accus.), 353 


is (plur. subject, sing. verb), 60 
‘3 through change of construc- 
tion (xexAduevos...mpopdvyré μοι), 159 


‘anacrusis,’ p. lxvi. 
anapaestic paroemiac, spondees in, 1311 
anapaests, excluded by Arist. from ord- 


σιμα, p. 8 


antecedent, attracted into case of relative 


(accus.), 449 


246 


aor. part., of a wish, hope, etc., 11, 649 
with γίγνομαι, 957 
with ἔσομαι, 1146 


” 9 
7 ” 
aor. referring to a moment just past, 337 
Apollo, προφήτης of Zeus, 151 
» with attributes of Zeus, 470 
»» 85 ἃ pastoral god, 1103 
aposiopesis, 1289 
Arcturus, in Greek calendar, 1137, p. 230 
Ares, the Destroyer, 190 
Aristophanes of Byzantium, 
ascribed to, p. 4 
Aristophanes, parodies tragic altercation, 
548 
Aristotle’s criticisms on the Oed. 7}7- 
rannus, p. XXiv. 
a Κυμαίων πολιτεία, pp. 4 f. 


ὑποθέσεις 


“arsis,’ D..1xv- 
Artemis Εὔκλεια and ’Ayopala, 161 
» with a torch in each hand, 207 
art. as relative pron., 200 (lyric): 1379 
(dialogue) 
», With abstract noun (ἡ ἐλπίς, Shope’), 
836 
», with infin. in dependent clause, 1232, 
1388 
»» With καιρός, 1050 
», referring to a previous mention, 845 
article, with interr. pron., in repeated 
question (τὸ 7f;), 120, 291 
Asclepiades of Tragilus, p. 6 
Assos, the American exploration of, p. 228 
Atlantic, the, w. limit of earth, 194 
augment, syllabic, omitted, 1249 
», temporal, omission of, 68 


blight, threefold, 25 

‘Branching Roads,’ the, 733, 1398 
brooches used as daggers, 1269 

bull, the, type of a savage wanderer, 478 


ς 


Cadmeia, the, of ancient Thebes, 20 

caesura, irregular, in anapaests, 1310 

children bought, to be sold as slaves, 
1025 


INDICES. 


choral ode, relation of to preceding ἐπ- 
εισόδιον, 463 

choreic rhythm, p. Ixx. 

choriambic verse, p. 1xxvi. 

chorus almost always close a play, 1524 

Cithaeron, the glens of, 1025 

clauses, rst and 2nd contrasted, and 3rd 
repeating 1st, 338 

colloquial phrases, 336, 363, 971, 1008 

comparison, elliptical form of (οἰκίαν ἔχει 
μείζω τοῦ yelrovos), 467 

condensed expression (ula ἀπήνη iyye= 
μία ἦν, ἣ ἦγε), 753» 1451 

conditional statement of probable fact 
(τάχ᾽ ἂν 7\0e= probably came), 523 

conjectures by the editor, p. 1xi. 

- of former critics, adopted in 

this ed., p. lix. 

construction changed (in answering a 
question which prescribed a different 
form), 1127 

‘contraction,’ metrical, p. lxv. 

co-ordination of clauses, where we should 
subordinate one to the other, 419 

Corneille’s Oedige, p. xxxvi. 

Creon, the, of Sophocles, Ρ. xxix. 

crepundia (Roman), 1035 

Cyllene, mount, rro4 

Cyprian Lays, reference to Oedipus in, 


Pp. Xiv. 
D 


dative after ὁ αὐτός, 284 
4» With βουλομένῳ ἦν, etc., 1356 
» after ὄρνυμαι (as=‘to attack’), 


», alone, in sense of dat. with πρός, 


»» ethic (πᾶσι κλεινός), 8, 40, 596 
.5. 1004]; 20 
»,» Ιοοδῖνε, 381, 422, 1266, 1451 
,, modal (ἀσφαλείᾳ), 51, 909, 1228, 
1556 
i », cognate to idea of verb (ὕπνῳ 
εὕδειν), 65 
Daulia in Phocis, p. xviii., 733 
‘deed and word,’ 72 
‘Delian,’ epith. of Apollo, 154 


ΤΑ ΖΖΤ ΟΝ, 


deliberative subjunct., indirect forms of, 
7,2, 1280 
Delphi, wealth of temple at, 152 
3 topography of, 463 
Dionysus, epithets of, 209 ff. 
dual forms of 2nd pers., 1511: fem., of 
participle, etc., 1472 
E 
echo, of one speaker’s words by another, 
570, 622, 1004 
editions of the play, p. lxi. 
elemental powers, the, profaned by an 
impure presence, 1427 
elision of σέ, etc., though emphatic, 64 
», Of 6 at end of verse, 29 
ellipse of verbal clause after ws, 923 
entrance, stage, for one coming from the 
country, 78 
epanaphora, figure of, 25, 259, 370 
epexegetic clause, after an adject., 57 
‘episode,’ Arist.’s definition of, p. 8 
epithet of agent transferred to act (γάμος 
τεκνῶν καὶ τεκνούμενος), 1214, 1229 
ἢ placed after a subst. which has 
art. and adv. phrase defore it (τὸν ἤδη 
Λάϊον πάλαι νεκρόν), 1245 
‘epode’ in choric songs, p. Ixvii. 
Eubulus, the comic poet, the Oedipus of, 
p- xxxiii. 
Euripides, the Oedipus of, p. xvi. 
a Phoen. 1758 ff., 1524 cr. 
‘exodus,’ Arist.’s definition of, p. 9 
expansion of verses in MSS., 1264 cr. 


Ix 


‘falling’ verse or sentence, p. lxix, 
false characters soon betray themselves, 
615 
festivals, Greek, bound up with family 
life, 1489 
figurative and literal expression half- 
blended, 866, 1300 
Fortune, Oedipus the son of, 1081 
fusion of two modes of expression, 725 
fut. indic. after ἔνθα μή, 1412 
» 99 Of wish, resolve, etc. (βουλή- 
gouat), 1077, 1160, 1446 
9) iN -tow and ιῷ, 538 


247 


fut. interrog., with οὐ, commands, 430, 
[140 
», ‘middle’ as pass., 672 
», optative, 538 f., 792, 796, 1271 ff. 
», partic. with art., 297 
95 perfect, 411, 1146 
a 
Genitive, absol. of subst. without partic., 
966—1260 
»,»  absol., with subject understood 
(ἄρχοντος, when one rules), 629, 838 
», after adj. of active sense, 885 
», after ἄτιμος; 788 
», after compound adj. 
lack (ἄχαλκος ἀσπίδων). 190 
», after ἐπώνυμος, 210 
» after vduo. (laws 
things), 865 
9, after πολυστεφής, 83 
», after mpoordrns, etc., 303 
», after verb of rising or raising, 142 
», after verb of taking (ἕλῃ μου), 
1522 
»,  attributive, forming one notion 
with a subst. which has an epithet 
(τοσόνδε τόλμης πρόσωπον), 532 
(γῆς τις, one of the land), 


denoting 


prescribing 


93 23 


236 

᾽ » (προστάτου γράφε- 
σθαι), 4τι 

9 ae (τί ἔστιν ἐκείνου; in 
him...?), 991 


τι 53 with infin. (οὐ παντός 
ἐστι ποιεῖν), 393, 917 

», causal (τῆς προθυμίας), 48, 697, 
701, 1478 

39 »» Κ(ἰκτὴρ πόνων), 185, 497 

», depending on subst. implied in 
adj. (ὧν ἀνάριθμος), 179, 1168 

»9 ΞΞ8ῃ adj. of quality (στολὶς τρυφᾶς, 
2.5. τρυφερά), 1463 

»» objective (ἀλκὴ κακοῦ), 93, 218, 647 

», defining (τὰ φίλτατ᾽ ἐκγόνοιν), 
1474 

»» Of source (φροντίδος ἔγχος), 170, 
312, 473, ὅδι 

»» Of parent (unrpéds), 1062 


248 


genitive, of place from which an act is 
done (ὄχου), 808 
» Of place whence, 152, 192 
», Of things needed, after els δέον, 
1416 
» _ partitive, 240 
after ἔχειν, 708 
» περᾶν, 673 
in ὡς ὀργῆς ἔχω, 345 
a »» Of point to which (εἰς 
τοῦτ᾽ ἀνοίας), 771 
τς simple, after λύειν, 1350 
goad, driver’s, with two points, 809 
god, an unseen, the agent, 1259 
Greeks, their unity expressed in religious 
rites, 240 


” ” 


happiness, to be predicated of no one 
before death, 1529 

Harvard, Oedipus Tyrannus at, p. xlviii., 
p- 201 

Helicon, nymphs of, rrog 

herald, sacred functions of, 753 

Hermae, supposed reference to mutila- 
tion of, 886 

Hermes, 1104 

Hesiod, reference by, to Oedipus, p. xiii. 

hiatus (εὖ ἴσθ᾽, as if preceded 1), 959: in 
lyrics, 1202 f. 

Hippocrates, references of, to Arcturus, 
Ῥ 221 

Homer, an echo of, 1325 

Homeric poems, notices of Oedipus in, 
ῬΉΎΣΙΝ 

Homeric practice as to syllabic augment, 
1249 

‘honesty the best policy,’ 600 

house of Oedipus, general plan of, 1241 

‘hyperbaton,’ 1251 

‘hyporcheme,’ defined, p. Ixxxv. 

hyporcheme in place of stasimon, 1086 


iambic trimeters interrupted by short 
phrases, 1468 

imperfect, not admissible in 1311 
»» of intention or menace, 805, 1454 
»» οἰ τίκτω, instead of aor., 870 


INDICES. 


imperfect, of willingness (ἐδέχου), 1391 
»» partic. (ὁ παρών -- ὃς παρῆν), 835 
», referring to a result of effort (εὕρισ- 
kov, was able to find), 68 
»» and aor. joined in a condit. sen- 
tence, 125 
», Iindic., of obligation etc. (ἔδει), 256, 
1368 
improbability, element of, in the plot, 
noticed by Aristotle, p. xxv.: how 
treated by the moderns, p. xlv. 
incense in propitiation, 4, 913 
indefin. pronoun (71s) after noun with art., 
107 
indirect discourse turned into direct, 1273 
infin. after ἐξευρίσκειν, 120 
»» after ἐπισκοπεῖν, 1529 
» after λέγω etc. as=ubeo, 350 
», alone, instead of infin. with ὡς (rd 
δ᾽ ὀρθὸν εἰπεῖν), 1221 
»» and accus. in prayer (subaud. δός, 
etc.), 190 
»» defining an adj. (ἄτλητος ὁρᾶν), 792, 
1204 
2» » ἃ phrase, 1169 
» epexegetic (ἐξαιτῶ oe τοῦτο πορεῖν), 


1255 
»» =an accus. of respect (φρονεῖν ra- 
xus), 617 


» for imperat., 462, 1466, 1529 
», Of plup. with ἄν, 693 
» Of purpose, with verb of ‘going,’ 
etc., 198 
»» understood after χρῆν, 1184 
», With art.=an accus. of respect, 1417 
», Without ἄν, representing an optat. 
without ἄν, 1296 
», Without ὥστε (εἰκάσαι), 82 
»» With τὸ μή (οὐ), 1232, 1388 
interrogative (τίς) and relative (ὅστις) pro- 
nouns combined, 71 
Tocasta, the Sophoclean, character of, p. 
XXViii. 
Ionic 3rd plur. (6Wolaro), 1273 
»» verse, p. lxxvii. 
Ionicisms in trag. dialogue, 304 
‘irrational syllable,’ p. Ixv. 
Ismenus, Ismenion, 21 


tie MATTERS: 


Aster, the river, 1227 

iteration of a word, rhetorical, 370, 1330 
J 

Julius Caesar wrote an Oedipus, p. xxxiii. 
K 


king, etc., summoned forth by visitors, 597 
*kommos,’ a, defined, p. 9 


τὸ structure of the rst, p. Ixxviil. 
ΑΝ the 2nd, almost ἃ monody, p. 
XC. 
L 


laurel, worn by θεωροί returning from 
Delphi, 83 

Laurentian MS., general relation of to 
the others, p. liv. 

laws, the ‘unwritten,’ 865 

leaping from above,—fig. of an evil δαί- 
μων, 263, 1300 

life, the, the guest of the body, 612 

logaoedic verse, p. Ixx., n. 

logographers, the, references of, to Oedi- 
pus, p. xv. 

Loxias, 853 

Lycia, haunt of Artemis, 208 

lyrics, relation of the form to the matter 
of, p. Xciv. 

IM 

Maenads, 212 

manuscripts used in this edition, p. lii. 

market-place, statue of Artemis in, 161 

masc. subst. used as fem. adject. (σωτὴρ 
τύχη), 80 
», dual instead of fem., 1472 

mesode in choric songs, p. Ixvii. 

metaphor, a trait of Sophoclean, 866, 
1300 

Pe substituted for simile, p. 226 
‘monodies’ in Tragedy, p. lxxviii. 


N 


Wero fond of acting Oedipus, p. xxxili. 
neut. adj. or pron. referring to masc. or 
fem. noun, 542 
», referring to men (οὐδὲν κακόν for 
οὐδεὶς κακός), 1195 
Nymphs, the, 1099 


249 


° 
Ocdipodeia, the, a lost epic, p. xiil. 
Oedipus—feels his own fate as separating 
him from human kind, r415 
5 the Sophoclean, character of, 
P. XxvVil. 
Olympia, μάντεις at, gor 
Olympus, the sky, 867 
optat., after secondary tense, replacing 
subj. with ἄν, 714 
», in dependent clause, by attraction 
to optat. of wish, etc., s06 
»» instead of subj. with ἄν, after 
primary tense, 315, 979 
», representing a deliberative sub- 
junct. after a secondary tense, 72, 1256 
», simple, where optat. with ἄν is 
more usual, 1296 
», With ἄν, deferential, 95, 282, 343 
», With ἄν, expressing one’s convic- 
tion, 1182 
oratio obliqua, 1271 
order of words, abnormal (τὸν ἤδη Λάϊον 
πάλαι νεκρόν), 1245 
5, (ὅπως, οὐκέτ᾽ οἶδ᾽, ἀπόλλυται), 1251 
», (ὁρᾶν μόνοις τ᾽ ἀκούειν), 1430 
», (Ta πάτρια, λόγῳ, fOr 1..λ. π.)» 1204. 
oscilla (Roman), 1264 
oxymoron, 196 


Ῥ 


paeon, the, in metre, p. 1xxx. 
Pallas, Theban shrines of, 20 
paradoxical phrases such as ἐν σκότῳ ὁρᾶν, 
997, 1482 
Parnassus, snow-crowned, 473 
paronomasia (χρησίμῳ χρῆται), 878 
partic. as tertiary predicate, 1140 
», continuing a question which 
another speaker has interrupted, 1130 
», epithet of agent, transferred to 
his act, 1214 
»» equiv. to protasis of a sentence, 
117 
», imperf. (ὁ παρών = ὃς παρῆν), 835 
»» )( infin., after εἴ μοι ξυνείη μοῖρα, 863 
.» ἴῃ nomin., instead of accus. and 
infin. (ἅλις νοσοῦσ᾽ ἐγώ), 1061, 1368 


250 

Ραχῖϊς,, irregularly replaced by finite verb, 
ELS 
» modal, 
dative, 100 
»» (wv) omitted, 412, 966 
» ΟΥ̓ adj. equiv. to an adv., 963 
» =protasis with εἰ, 1371 
» With ye, instead of finite verb, in 
a reply, 1011 
» With μέμνημαι, so 
»5 With παραρρίπτω, 1494 

parts, cast of the dramatic, p. 7 

pastoral epithets of Apollo, 1103 

patrons of μέτοικοι, 411 

pauses, metrical, p. Ixvi. 

perf. of final result (εὑρῆσθαι, ‘found once 
for all’), 1050 

person, the third, for the first, 535 

Phasis, the river, 1227 

Pherecydes of Leros on Oedipus, p. xv. 

Philocles, traditional defeat of Sophocles 
by pexxx: 

Pindar, reference of, to Oedipus, p. xiv. 

plague at Athens, supposed allusion to, 
ΡΒ χχχ. 

pleonasm, 408 

Pliny, references of, to Arcturus, p. 231 

Plunteria, festival of the, 886 

pluperf. infin. with ἄν, 693 

plural, allusive, for singular, 366, 497, 
IOQI, 1359, 1405 
», neuter as adverb, 883 

pollution, feared from contact with the 
blood-guilty, 1415 

Polus, the tragic actor, p. xxxi. 

position, irregular, of a second epithet, 
1199 

», unusual, of words, giving em- 

phasis, 139, 278, 525 

positive and negative joined (γνωτὰ κοὐκ 
dyvwra), 58 

5» (verb) to be evolved from nega- 

tive, 241 

power, the substance of, better than the 
show, 599 

predicate, adj. as, after subst. with art., 
672, 971 

prep., following its case, 178, 525 


answering to a modal 


INDICES. 


prep., between two nouns, governing 
both, 734 
», needlessly added 
χώροιΞς), 1126 
present infin. after εὔχομαι, 802 
»,  indic. or partic., denoting a per- 
manent character, 437 
»" ΠΙΞΙΟΤΙΟ <Ki2 
proleptic use of adjective, 98 
‘prologue,’ Arist.’s definition of, p. 8 
pronoun in appos. with following subst. 
(τάδε... τάσδ᾽ ἀράς), 819 
»» possessive, for genit. of pers. pron. 
(σὸς πόθος), 969 
5, ΤΙαΪαΐ., as last word of verse, 298 
», relat. instead of demonstrative, 
after a parenthesis, 264 
», with causal force (ὅσα Ξε ὅτι τοσ- 
adra), 1228 
», redundant, 248, 385, 407 
prodde in choric songs, p. lxvii. 
prophecy, Greek view of, 708 


(ξύναυλος πρὸς 


Q 


‘quantity,’ metrical, defined, p. Ixiv. 
», Of vowels before xp, 640 


rain, symbol of water generally, 1427 

recognition of children by tokens, 1035 

redundant expression, 1126, 1463 

repetition (ἀστὸς els ἀστοὺς), 222, 248, 261 
», in euphemism (βλαστοῦσ᾽ ὅπως 
ἔβλαστε), 1375 


rr in lyric lament, 1193, 1330 

ΕΝ of one speaker’s words by an- 
other, 548 

re of the same word, at a short 


interval, 517 
resident-aliens at Athens, and their pa- 
trons, 411 
‘resolution,’ metrical. p. Ixv. 
revivals, recent, of Greek plays, p. xlvii. 
rhetoric, figures of, 370, 538, 1375 
», πίστεις of, 1420 
rhythm defined, p. lxiv. 
rhythmical ‘sentence,’ the, p. Ixvi. 
»» ‘period,’ the, p. Ixviii. - 


Ti MATTERS. 251 


riddle of the Sphinx, pp. 6, 228 
‘rising’ rhythmical sentence, p. lxxiil. 
rivers, representative, 1227 


sacrifices, excommunication from, 240 
seasons, the, Greek reckoning of, by the 
Stars, p. 231 
Seneca’s Oedipus, p. ΧΧΧὶν. 
sentence, structure of, changed as it pro- 
ceeds, 159, 587 
slaves, home-bred, most trusted, 1123 
Solon’s saying, 1529 
Sophocles, and the modern dramatisers of 
the story—essential difference between 
them, p. xliv. 
» general 
style, p. lvii. 
» new traits of the story in- 
vented by, p. xvii. 
Sphinx, death of, 1198 
, Egyptian, Asiatic and Hellenic 
types of, p. 226 
,, relation of, to the Oedipus-myth, 
p. 227 
», riddle of, pp. 6, 228 
,. winged, 508, pp. 227 f. 
stars, the wanderer’s guides, 694 
stasimon, Arist.’s definition of a, p. 8 
State, rivalry in service of the, 880 
subject of verb indefinite, 904 
subjunct. after ὅς without ἄν, 1231 
» deliberative, 364: λέξω doubt- 
ful, 485: usu. aorist, 651 
»» without ἄν, 317 
suppliants, their branches, 3 
a touch the hand, 760 
syllabic augment omitted, 1249 
‘syncope,’ p. lxv. 
synizesis, 555, 1002, 1451, 1518 
»  Oofvrare, 640 
synonym used, instead of repeating the 
same word, 54 


characteristics of his 


= 


table brought in for a meal, 1463 
Teiresias, the, of Sophocles, p. xxix. 


text of Sophocles, general condition of, 
p- lviii. 

Théatre Frangais, the, Oedipe Roi at, p. 
xlix: 

Thebaid, the ‘cyclic,’ fragment of, p. 
xiv. 

Thebes, topography of ancient, 20, 1378 

thesis; -p. 1Xv. 

‘Thracian,’ epith. of Euxine, 196 

time the test of worth, 614, 1213 

title of the Oedipus Tyrannus, p. 4 

tmesis, 27, 199 

tribrach, apparent, for cyclic dactyl, p. 
]xxxix. 
»,  insenarii, usual limits to use of, 
537) 719 

trochaics, in what sense excluded from 
στάσιμα, p. 9 

tunic, women’s Doric, 1269 

tyrannis, the Greek, 541 


Vv 


verb, left to be understood, 683, 1037 
», (or partic.) to be supplied from a 
cognate notion (νομέσας from ἰδών), 538 
» referring to two subjects, though 
appropriate only to one, 116 

verbal adjective, sigmatic form of, p. 
225 

verse, beginning with word which closely 
adheres to preceding verse (ποτ᾽), 1084 
» rhythm of, suited to the thought, 
332) 719, 738, 1310 

vocative of Οἰδίπους, 405 cr. 

Voltaire’s Oedipe, p. xl. 
», CYiticisms, p. xlii. 


ww 
west, the region of the Death-god, 178 
women, position of, 1078 
»» presence of, at festivals, etc. 1489 
δ. 4 
year, popular division of, by the stars, 
p. 231 
Ζ 


zeugma of verb, 116 


2 


Cambringe : 
PRINTED BY JOHN CLAY, M.A. 


AT THE UNIVERSITY PRESS. 














Pee 
εδϑυ 


PA Sophocles 
JAN S The plays and fragments 
Ae 


19118 
ptel 


cop.e3 





University of Toronto Ὁ 
Robarts Library 





fi a 


εὐὐνένς 


se honk τὸ 9 sak 
aieeate 














ghetet4 1 
7 ay ime " ἊΣ ν γι 
Wehbe ). Ar = ἊΨ" ρ ὴ 


pepe pe hs ' 











fet 
Heute 
Had wy 





1 ΠῚ τεὸν 
| ἐν ἐσ Α ἡ he ἀμ μα] 
ἢ Debye ping 

TEER s 
ata sys i λοι 
wae AV ἦν 
ΡΝ Hc 

she ian 
Sree tlE 
ayia ene 





rn mi 
4 ties ἷ ΤΟῪΣ τὸ Ν ᾿ 
boat ta te οὗ νιν hiek 
ik \™ ‘0 sparen test es > 
. ΕΣ , 


i 
ἦν ty oa: Bis tt it 
ten ah eich τς ἥν nth 
ieee ean eres mit ΤΙ ΒΗΝ μὴν 
δ Prevage a 
ma Goh dda 
ἜΡΝΟΣ ΚΡ etehen p 
ADA LR a ον, ἘΜ 
ae a Abeta ae 



























































































































































































































































































wy op ’ = Hats: 
hele: ie 
iy rie Hah 
ὙΠ ᾿ ΝΕ 
Ps one 
ΘΙ, Lita Steet ee dan 
᾿ ste bunk ."- 
ΜᾺ 
᾽ 
. 
ἮΙ 
De φ 5 ; ς 
᾿ ἐδ > ᾿ ᾿ ᾿ oie 
ee ln ὲ γὴν ty εὐ 
νὰ το owe ᾿ Αι δὸς 
phathens 7 ἢ ΤΩΝ 
ihre Bees υς , : hfe τ snails Ab DE De Ue ΠΟ ik) cari ᾿ K τ 
oleh “hey " iy ΤῊΝ Ae Dy ify ἑ " ἢ ᾿ δὺν ; i . 
bie eo: 5 t ᾿ νι: * ΓΤ γ 
ΤΟΙ δ δ is Fay oh 
tl ee Bh if 
τ M4 Selts r¢ oP) ᾿ 
bee ὙΠ 
; pie i Sat ret ἐξ» 
ΜΝ Tey put Nani 
Ῥ 
rt sept? ἘΣ Rie 
89: ᾿ 4 : . 
ἡ i henk Walk 7 “(τ fai τ 
μφὸ κὰ ble ἡ, ni a spite Tait \ Ae ye UR Poy dela hent , ’ t 
ἐσ en ΜΝ, i iat ; 4 Hane : AVS (1h Dp yee ». ᾽ 
ἸΕΚΣΉΝΗΝ pate gh iy Ee ΤῊ tin Ἧι α ἘΝ fee 1 
{ert este avin ib rn ΡΥ Ἵ NiNelinee Le p Reuaeawnd SEPRAL Ea τὸν - 
: ἩΠΕΚΗ ΗΚ Oe ata + WWelg: cents ΕΣ ΓΟ ΗΝ WEE AEE ELEM ie ΡΝ ΤΙ ἐν peta, a be ot [ΤᾺΝ thas tf 
sf pat putes nt a ned ) Der sare rte thal OL it arts a "FA ᾿ ; , 
᾿ y ᾿ ᾽ Vela 
bdo ι aie f Wag wire ‘A 
᾿ ᾿ς 4}. 4 ys ᾿ Bebe ᾿ * 
rat hp bby » ὡς 
44 , ᾿ 
* et ee) Ὁ ᾿ 
Ἢ aia abetes i 
wri'g§ ae Σ᾿ τ δια 
ρον τ τῪ τς ἔνε ΕΔ bee ah ede δ᾽ Ky Py 4 ᾿ Say 4h t 
΄ Η ᾿ Seer tho 
soars ᾿ . een ve Sg) Ὁ} , tha ἐν os ὃ ni mate τὶ " ἣν (bento 
3 oath \ 4 om. ᾽ 
o> ἣ 
᾿ - 
¢ 
- “- 
ἣν wee δῷ ' 
ee Δ shenc an r Ἂ 
ry we . ἃ Α 
ψΨ' Dae ierepb oe! eh δε δεν de ielanens ι . 4 Vga 4 aha 
hb 4 344 sad ΠῚ ἃ fi , 
ry) τὶ σεν 
peated limnne tt .᾿ 
νῷ γὼ Co bi LB Va M Ἷ 
νὶ ft . 
, , - 
bye deel t 7 ᾿ . ‘ 
. ‘ ͵ tone ᾽ ῃ ΄ - ᾿ 
αν Abee ᾿ 
“Ὁ . . 
ahr hie be Sexe 
i » |=) 
. ᾿ cle 
i ΤΟΣ ἫΝ ‘ ἐφ 91. oh ᾿ ete 
Z Φ νὰ ἐἀβ φν ve ae ; ΡΝ Rive vast» τ 
ing citer =a ἐφ « ea het yp τῆι hie +m 
ore silane x She ye ) : ᾿ Weibel ; ἴω 3 ᾿ 
ἐ bole. Petree nee) he r Ὁ POV ht Kans ie | ; ὃ te 
44 eastagtititt Whe, Vari tenn ἘΆΝ DAR AY Sea ene Hae Wi Prat be bb ieeiee eA εἶ Lapa! gS 
bee jhe ἡ 
abe ee ων ' - . 4 + Ψ “ἢν εν ἢ i> 
ng bg vege oa eerer m= > Py, w os me . rae ᾿ " 
bie ot vey ate ’ 
. 
 ymuves e 
δεν. ven ἊΨ ἡ αὐ γενφῦε ὁ ἐγ 
Porn Me . 
- > - 
Ἔα τ γός μ of aire +> ; Iie ne bet, aie te eh briehetese aye p ἷ 
i+w των] σι . srs ei ve 4 
berkot ye ee 4 the oe lies 
ete 
4 the ae baneie re ieee 
pred wap το δογ ther akin - 
sesosesghePinate ΠΣ ΜΝ } ἼΣΗ 
Sere isajregtet sehilien eit steepirtepeten ‘ oh \y + gh Th ao ; λον 
κεν 1h sas, ri “~ bev ᾿ Ne ten “Ὁ PO oem . “ἢν 
picts eters pes Sopeten στ heres ΤΟΥ Lag peg Pye 4 Me jie ᾿ ἘΜ ὍΣ. he 23 ᾿ 
Sr Relates eo. o> " ἡ ταῦ σίγα δῳ ol are - 
Or ted , - Sobehe? 
eerheatense 4s =e 
sischhesatsraovemer wereserarsorncgerestemtinest on iee