Bryn Mawr Classical Review 02.01.11


Lloyd-Jones, H. and Wilson, N. G., Sophoclis fabulae. Oxford Classical Texts. Oxford: Oxford University Press, 1990. Pp. xx + 431. ISBN 0-19-814577-2. $24.95.


Lloyd-Jones, H. and Wilson, N. G., Sophoclea: Studies on the Text of Sophocles. Oxford: Oxford University Press, 1990. Pp. 282. ISBN 0-19-814041-X. $59.00.


Reviewed by G.M. Kirkwood, Cornell University.

In the sixty-four years since A. C. Pearson's OCT was published, so much that concerns the text of Sophocles has occurred, and in some matters so long ago, that Pearson's text has been badly out of date for many years. The 1924 text, though generally praiseworthy in its choice of readings (Housman, not an overindulgent critic, called it, among other words of praise in a review in CR 76, "the best critical edition of Sophocles now in existence''), was bound still to the nineteenth century formula of "L-A-rec." in its evaluation of MSS authority. The ensuing years have brought forth de Marco's important recognition of the Roman family of MSS; Turyn's collations and his epoch-making though flawed classification of the MSS tradition; the Budé text of Dain, with its reassertion, against Turyn, of the value of A (Parisinus graecus 2712) as a witness independent of Byzantine criticism, Dawe's further extensive collations and analyses (on which his Teubner text is based) which enabled him to correct much of Turyn's work -- essentially to disassemble Turyn's elaborate stemma of the MSS, especially through his recognition of the importance of "horizontal" transmission in the text tradition of Sophocles; careful editions of individual plays (Webster's Ph., Easterling's Tr., Kells' El., and Dawe's OT); and Kamerbeek's extensive and valuable commentaries on all seven plays. Finally (not to mention the numerous journal articles on a great many points of text), we have the present edition and the separate commentary, Sophoclea, that accompanies it. It is beyond question that we have thus been brought to a far fuller and more secure understanding of Sophoclean text history than was possible before, and have the advantage of an accumulation of learned and skilled analysis in the evaluation of readings and their interpretation.

While the accrued knowledge of the history of the text is the indispensable basis of an edition, editorial tact and judgment determine choices among variant MSS readings and decide which of the myriad of conjectures that over the generations of scholarship have been proposed are the "corrections" their proposers hope them to be, which ones are sufficiently probable to merit mention in the apparatus criticus, and which (the vast majority) can best be disregarded.

Our judgment of the value of a new edition of the text must then rest on these two general criteria: the scholarly knowledge and critical analysis of the history of the MSS tradition shown by the editors and the philological judgment shown in their choice of readings. While the two editors assume joint responsibility for all aspects of their work, references in both the OCT itself, and in the companion volume, Sophoclea, to previous publications indicate that Wilson has contributed primarily to text history, Lloyd-Jones to text criticism.

To begin with the first of these criteria, two questions are appropriate: what does the new edition add to our knowledge of the history of the text of Sophocles and what significance for the text does the editors' grasp of the history have? To the first of these questions the short answer is "not much, but the little that is added is valuable." Principally, MS K (Laurentianus 31.10) has been securely redated by Wilson, in Scrittura e civiltà 7 (1983) 161-76, to the second half of the twelfth century. While this MS has long been recognized as having an affinity with L (Laurentianus 32.9, which remains by far the most important of all Sophoclean MSS) and preserving valuable readings not in or not legible in L, its previous dating in the fourteenth century meant that its readings were under suspicion of being Byzantine conjectures. Now it is seen as the second oldest MS; its supplementations of the frequently defective L and its readings independent of L, which represent an otherwise lost branch of the tradition, become authoritative. While the evaluation of L (with its twin, the palimpsest Leiden BPG 60A) and of the Roman family (G, R, Q, and, providing scholia only, M) are essentially unchanged from the account of them given by Dawe, the present editors have re-examined L to some advantage (El. 769, OT 297). They have used also the recent study, by Barrett, of P.Oxy. 2180, the one papyrus that provides some valuable readings, in OT. One minor but intriguing addition to the history of Sophoclean scholarship is the analysis, in an Appendix to Sophoclea, of the readings, whether conjectures or discoveries in MSS now lost, of the sixteenth-century scholar Livineius. These readings, written in the margins of Livineius's Aldine Sophocles, were published in the Classical Journal of 1813, by a contributor who styled himself "Ctesiphon." The editors judge that the best of these readings are conjectures of Livineius himself; the question of what, if any, MSS he found some of his readings in, as he alleges he did, remains unanswered.

In general, the MSS used in this edition are those used in Dawe's Teubner text and discussed in Dawe's three-volume Studies on the Text of Sophocles (the abbreviations for the MSS are those used by Turyn and Dawe). In addition to the MSS already named, these are: the Z group, MSS used by Demetrius Triclinius, four of which (Zf, Zg, Zn, and Zo) are cited by our editors for the four plays other than the Byzantine triad of Aj., El., and OT; A and its congeners (U, Y, and, for the triad only, D, Xr, Xs, and Zr); a group of MSS written in the late Byzantine period, the five earliest of them, C, N, O, Pa, and Wa, probably written before 1300, containing the triad only, and designated p (for Palaeologan) in this edition; two MSS (T and Ta) that closely reflect the edition of Demetrius Triclinius, made soon after 1300. Some other late MSS are cited occasionally, as representing the earliest witness to a correct reading, only one of which, J (Jena, Bos. q. 7), is listed in the Preface.

While the MSS mentioned in the preceding paragraph do not provide sources previously unused or seriously undervalued, the editors have in a number of cases taken the trouble to reexamine MSS. Overall, there can be no doubt about the editors' scholarly thoroughness in their use of text sources, and I very much doubt that their learning and judgment in these matters can be seriously impugned. There remains one aspect of the investigation of the MSS of Sophocles that cannot be said to be fully dealt with. The editors have used, to a greater or lesser degree, a total of some thirty MSS. According to Turyn's "List of the Manuscripts of Sophocles" (Studies in the Manuscript Tradition of the Tragedies of Sophocles 5-9), there are, to his knowledge, 193 MSS (with another seven that he was unable to consider). I do not know why in the Preface to the new OCT the number of MSS is given as "about one hundred and fifty." Dawe gives the number as 200. While Turyn was probably justified in declaring (Studies p. 201) that his publications "cover almost all the extant mss. of Sophocles (not later than the XVIth century)," it is not clear from Studies that all have been fully collated. Even though these late products are derived from well-examined sources, it is hard to be confident that we know all that can be known about the history of the text until all extant MSS have been fully collated. In the case of Turyn's "Jena" class, for example, though Turyn declares (Studies p. 90) that "the Jena text ... should be completely disregarded by students of the Sophocles text," we find, on page xii of the Preface (not praefatio; this volume breaks the old OCT Latin tradition here -- though not in the apparatus criticus -- for reasons given on p. v, and, though we may lament this further evidence that we are to assume that not all who study Greek texts can read Latin, there is a practical advantage in switching to "the international language of modern times"), the acknowledgment that J, the best known of the Jena group, may preserve readings that are correct and whose presence in this "isolated" source "is very hard to explain." It remains improbable, of course, that any very great new light will be shed by further examination of unexploited sources. But it would be appropriate, in this age of efficiency, for the remote corners to be explored. The search for the text of Sophocles may not yet be ended.

The newly appreciated importance of MS K is a conspicuous, interesting, and valuable feature, authenticating or supporting readings that were previously underrated. At Aj. 679 K alone preserves the superior reading hêmin (êmên cett.). At El. 33, patri (Lac, K) now looks more probable than patros (Lc and most other MSS), which was reasonably chosen by Pearson. At OT 445, where L is not readable, K provides evidence of the L text. At Ant. 211 K alone gives, in the margin, the reading poein, already conjectured by J. F. Martin, an improvement on the vulgate, Kreon or Kreôn. At Ant. 599, K gives hoper, already conjectured by Hermann; it is a valuable reading in this troubled passage, even though the OCT prefers huper. At Ant. 687, which our editors hesitantly delete, K's reading chaterai, also an earlier conjecture by Musgrave and accepted by Kamerbeek, seems to me a very probable reading; the line, which our editors call "feeble," should be kept; its hesitancy is part of Haemon's tact. At Ant. 859 K confirms oiton (conjectured by Brunck), which is clearly superior to the vulgate oikton or oikon. At Tr. 1275-78, K, as well as P.Oxy. 3688, attributes these disputed lines to the Chorus, correctly, Easterling argues, and I think she is right; the OCT editors give them to Hyllus. At OC 457 K's reading moi is preferable to the vulgate mou. At 534 K provides essential guidance, at 946 the OCT editors accept a unique K reading over Lloyd-Jones's 1954 conjecture, and at 1487 K gives a reading different from L's and better.

The new OCT provides a tidy and informative apparatus criticus, using some easily followed sigla (explained on p. xiii) to lessen the number of individual MSS cited: p stands for two or more of the group designated Palaeologan, a for two or more of A and its congeners.

To turn from these matters to the multiple problems of the selection of text from the variety offered by the MSS or by modern criticism is to enter a realm in which disagreements are inevitable. To illustrate the dimensions of this task of criticism it is instructive to make some comparisons between the two most recent editions, the new OCT and Dawe's Teubner (2nd edition). Neither could be said to be a conservative edition in the sense of excessive adherence to the "paradosis," to use the present editors' convenient term, and yet the points and ways of their departures are very different. What is regarded by Dawe as a "correction" may not even gain mention in the OCT app. crit.; thus, of eight readings listed by Dawe as his own corrections, only one is accepted by the OCT, one other is mentioned in the app. crit. (but credited to an earlier critic), the rest not listed in the app. crit., though several are discussed and rejected in Sophoclea.

A similar dissension can be seen in the choice of passages deemed interpolations. In the OCT, forty-one passages, involving about ninety-two lines, are bracketed; in Dawe's Teubner, twenty-eight passages, fifty-one lines. Of these, only nine passages coincide (fourteen lines). Another example, in the field of deletion: the OCT rejects Aj. 1028-39 as "empty bombast," a passage which includes a line (1034) praised by M. L. West as a "fine verse" and four others that the same critic confidently judges genuine. Such disagreements clearly reveal the difficulty that attends evaluation of Sophoclean style. The OCT editors state on page v that Sophocles "took pleasure in experimenting with the syntactical resources of Attic Greek," and in Sophoclea repeatedly warn that "what is unique is not necessarily corrupt" (164, cf. 198, 233), and in the sentence rejecting Aj. 1028-39 speak of the dangers of "a romantic aversion to anything savouring of rhetoric." It is hard to know what is too strange to be Sophoclean and what is just strange enough to be Sophoclean.

The OCT is not greatly given to detecting interpolation. Of three passages most often (and in my opinion wrongly) excised, OT 1524-30, Ant. 904-20, Ant. 1080-85, all are retained as genuine, though a hiatus is assumed, perhaps rightly, after Ant. 1080. Haslam's rejection of El. 1, accepted by Dawe, is resisted, on the valid grounds that the "ceremonious address" of line 1 is appropriate to a Sophoclean prologue. The long note on this line, Sophoclea 42, addressing the question fully and convincingly, contains a minor error: neither Plutarch nor Athenaeus quotes line 3, a fact that, if it has any bearing, weakens the case for excision.

One of the editors' most extensive deletions, Aj. 854-59, a passage that has often been questioned in part or entirely, induces the uneasy feeling that scholars who favor rejection have been too readily repelled by the rhetorical fullness of this peroration to the suicide speech; yet it must be granted that suspicion is justified by the repetitiousness involved in accepting the whole twelve closing lines of the speech. More dubious is the excision of OT 1278-79, the most gruesome part of the Exangelos's description of Oedipus's self-blinding. In support of their deletion, the editors quote M. L. West's view that it is "an obvious interpolation in the interest of goriness." To this judgment I would answer that the full horrifying picture of self-mutilation is thematically appropriate at this terrible moment and there is no reason to think that Sophocles was too delicate to draw it, even as Euripides gives us the full horror of the death of Creon and his daughter in Medea.

Whereas Dawe's text has been criticized, in my opinion justly, for its over-frequent assumptions of lacunae, the new OCT recognizes only seven, apart from the numerous passages where lack of metrical responsion shows that some words are missing in the lyrics. Only one (after Ant. 1080) is new; all are occasioned by doubts about correctness of syntax or continuity of sense; none is of major importance.

The editors are very sparing, also, in transposing lines. Tr. 994-98 are reorganized for an improvement in syntax, OC 198 and 199 are reversed, OC 1028 is placed after 1019. The last of these is the only passage that does not actually require some change. The two passages most often subjected to extensive transpositions, the long report of the Paedagogus in El. and Creon's first speech to Haemon in Ant., are left intact, correctly in my opinion.

The obelus is, by design, used very sparingly, to provide a "text which can be read with few interruptions" (Preface vi). The editors therefore accept readings of which they are frankly uncertain. The practice makes good sense, but it is marred by the editors' willingness to replace acceptable MSS readings by conjectures that they prefer. This comment brings us to two features that remain to be discussed: the relatively restricted but quite crucial problem of attribution of lines, and the fundamental and endlessly disputable matter of conjectural emendation.

Consideration of these questions means discussion of Sophoclea more, even, than of the edition, since, though the readings are in the text and the app. crit., the explanations are in the commentary. Sophoclea is modestly described, on page xvi of the OCT, as a work "in which we explain our view of many difficult passages." Many of the notes in this volume are valuable brief essays on points of style and grammar. The acumen and thoroughness which the editors show in their defense of their text and in the many, varied problems discussed, and the fairness -- for the most part -- of their discussion of differing views are models of scholarship. The exceptions alluded to apply to their clear distaste for the work of two contemporary scholars, Dawe and Kamerbeek. Dawe is, of course, more given to textual alteration than most contemporary critics deem necessary, and his recurrent "corr." in the apparatus is excessive (I count 187 "corr."s in his edition); substantial disagreement is natural. Moreover, our editors fully recognize Dawe's very great contribution to the study and assessment of the MSS. But there is a recurrent note of attack, of irrelevant criticism, that is somewhat jarring. In their attitude toward Kamerbeek the proclivity for making unfavorable general comments is even more in evidence (his "usual nervous horror of emendation," Sophoclea 222, his "nervous horror of the slightest departure from the paradosis," 212), which sort oddly with the several times, especially in the notes to Ph. and OC, that they record agreement with Kamerbeek. Their overall judgment of Kamerbeek's contribution to the study of Sophocles, given in the Introduction to Sophoclea, is similarly ungenerous. Kamerbeek is indeed a conservative text critic, but many of his defenses of the MSS readings deserve respectful attention; his series of commentaries is a major contribution to the study of Sophocles. There is much to be uncertain about in the criticism of the text of Sophocles, and there is much to be said for keeping the paradosis in the many passages where no convincing conjectural change has been proposed. Attractive possibilities have indeed a place; but it is in the app. crit., not the text. The scorn of the more adventurous critic for the conservative's reluctance to accept conjectures, often expressed with exasperation at the latter's faulty comprehension of Sophoclean style, would be more acceptable if the conjecturers were more in agreement with one another.

Before I turn to an examination of readings in the new OCT, a brief mention of what these critics have done with some disputed attributions is appropriate. Probably the most famous and most contested of these is Ant. 572, which the OCT, I am happy to report, leaves with Ismene, in accordance with the paradosis. I do not see how anybody with a sense for Sophocles' style as a dramatist can fail to see that the silent scorn of Antigone, who does not deign to address Creon in this scene (her last words to him are the famous line 523), is more Sophoclean than this timid, pathetic cry. The OCT ascribes to Ismene also both 574 and 576. 574 is Ismene's in the MSS and can reasonably be defended as hers, though it would fit also the Chorus, betokening their first words of resistance to their ruler. (There is some confusion in the note in Sophoclea, contradicting the OCT app. crit., which has it right.) But 576, for which the weight of MSS evidence is for the Chorus, though there is some support for Ismene, is clearly a line for the Chorus, with its opening "It is, then, decreed...," anticipating in its questioning tone their later suggestion of dissent at 770. Moreover, Creon's answer ("Yes, decreed by you and by me") is clearly addressed to the Chorus. Creon is confident that he controls the will of the Chorus; he is not concerned with Ismene's view. The closing lines of OT, rightly accepted as genuine, are ascribed to the Chorus, not Oedipus, and I think the choice is good, though the question is complex. MSS evidence for ascription of the closing lines of Tr., Hyllus or Chorus, is inconclusive. Disagreeing both with Easterling and, in my opinion, with probability, the OCT ascribes to Hyllus.

In the matter of conjectures, no doubt all students of Sophocles have their strong views as to what must be altered and what must not. Here is my score card for what the new edition does with some debatable questions. At OT 873 some critics have changed the MSS Hubris phuteuei turannon to hubrin phuteuei turannis; the OCT editors retain the MSS reading and defend it with an excellent note in Sophoclea. At Ant. 368 it is heart-warming to find them going against the tide for the feeble gerairôn and retaining the much subtler and more meaningful pareirôn; here too the supporting note in Sophoclea is excellent. In the next stasimon of the play, however, they accept the irrelevant but popular kopis in place of the MSS konis (602), inspite of Easterling's strong defense (Dionysiaca 148-49) of konis. Their solution to the difficult problem in 614-15, changing ouden to ouden', which then becomes object of herpei, with biotôi changed to biotos and made subject of herpei, is both more complex and less probable than the usual change of pampolis to pampolu; oudena herpei biotos is not a clause to admire. Their obelus at Tr. 526 (the Chorus as matêr or Zielinski's patêr with neither satisfactory and nothing else to consider) seems appropriate. At OC 1583-84 they do not replace the paradosis with the ingenious and influential conjecture of the eighteenth century critic Zachary Mudge. I do not, however, agree with their interpretation of the passage (in spite of the authority of Hermann, whom they quote) and believe, as Jebb did, that aei in 1584 is corrupt; I have argued the point in Phoenix 41 (1987) 184-88. Lastly in this list, at OC 1752 they accept J. F. Martin's nux apokeitai in place of the reading xun' apokeitai (Reisig), which is virtually the paradosis; their objection to the form xuna ("why should xuna have been preferred to koina?") is hard to understand in view of xunou at Aj. 180, which they accept without question. Their note supporting nux is far from convincing. The meaning, I have argued, TAPA 116 (1986) 99-117, is far better with Reisig's minor change. The introduction of nux weakens the text.

The editors have put in the text, apart from slight changes in spelling and accent, forty-nine of their own conjectures, including several previously published by Lloyd-Jones (designated by "Ll.-J." in the app. crit.; the others by "nos"). This is a modest number (Pearson's text has fifty of his) in view of the many uncertainties in the paradosis; the editors have, however, accepted into their text fully ten times that number of conjectures by others, and while most of these are worthy of notice, many are not probable enough to deserve the honor of a place in the text.

Several of the editors' own conjectures bring substantial improvement to the text. Thus at El. 918, while the paradosis is defensible, the slight change ho (Ll.-J.) for ta, combined with Brunck's hautos in 917, gives a markedly better sense. Slight also, but definitely an improvement, avoiding hiatus, is tôs (Ll.-J.) at OT 510. At Ph. 209, thrênei is a good solution to a metrical problem. At OC 547, where some alteration is needed for sense, atai (Ll.-J.) along with Hermann's alous for allous is better than J. F. Martin's moirai. At OC 1694 the editors have done a good patching job on a faulty text, improving on earlier efforts. Several other conjectures of their own that are taken into the text are interesting contributions but are not entirely persuasive: the rare adverb harmoi, Aj. 245 (Ll.-J.), makes excellent sense, but there are more probable solutions for achieving correspondence between 221 and 245. At Aj. 636, heis (Ll.- J.) is attractive, though it is not certain that ek with the genitive is wrong here. tarchaion (nos), at Aj. 1292, for archaion no doubt improves on the style of the paradosis, but it is not clear that improvement is needed. There are several more changes that have the same kind of interest short of necessity.

At a lower level of probability, in my judgment, are the following. Aj. 775 has enrêxei (nos) for ekrêxei. The argument given is that Ajax must mean that battle will not break "through" where he is, rather than "out"; but this assumes that the ek- prefix must mean literally "out from," whereas ek need not have any marked sense of direction, as in the use of ekdeixeias at El. 348 and ekdeixei at OC 1021. Ajax means that things will not "blow up" where he is stationed; furthermore, the harsh sound of ekrêxei is effective here. At El. 844 the editors have replaced edamê with damar ên, both for meter and because edamê requires Electra to "show almost preternatural acuteness" in realizing that the unexpressed subject of edamê is Eriphyle. But the allusions are no more indirect here than in other places in Sophocles, notably in Stasimon Three of Antigone, and Electra's words at 846, "I know, I know; for an avenger appeared" require edamê. The reading akos at El. 1087, on the theory that the MSS to mê originally was tomên, a gloss for akos, is indeed ingenious, but not very probable. Nor is it by any means certain that to mê kalon kath oplisasa is wrong. There are many ways in which Electra's plan of revenge can be regarded as other than kalon (cf. Jebb's note ad loc. and his Appendix). The change from prosesche to katesche, Ph. 236, is unnecessary; though it has often been suspected here, prosesche has a close parallel to this transitive use in Herodotus 9.99.

The editors' solution to the syntactical problem in Antigone's opening speech to Ismene is to replace the ho- of hopoion with exclamatory a. Apart from the repeated a's that represent Philoctetes' anguished cries of pain, which are surely irrelevant to this proposal, Sophocles uses a twice elsewhere, both times to indicate a spirit of resigned reproach: Oedipus to the shepherd, urging him not to castigate the Theban messenger, OT 1147; Neoptolemus to Philoctetes, Ph. 1300, urging him not to shoot Odysseus. In these opening words Antigone is anything but resigned, and her tone is not one of reproach but of anger, and it is hard to see how this rather quiet exclamation can be appropriate here. A better way of escaping the grammatical awkwardness of the two subordinating conjunctions is to read, as Bothe proposed, oistha ti; I do not understand why this simple solution (listed by Jebb in his Appendix and rejected without comment) has not been more widely accepted. The substitution of esaines for ephaines, El. 1359, is also unpersuasive. The emotional quality of such passages as Ant. 1214 (Creon, of the effect on him of Haemon's voice) and Hippolytus 864 (prossainousi, Theseus, of the effect on him of Phaedra's note) is quite different. Moreover, the parallel at Euripides El. 1234 for the use of ephaines here is excellent; the assertion that Electra is "unlikely to have addressed the Paidagogos in terms quite so hyperbolical" (Sophoclea 72), is contradicted by her opening words in this very speech, ô philtaton phôs.

Many of the conjectures proposed by other critics and incorporated in the OCT text are necessary, and the editors show good judgment in choosing among proposed readings. But there are a great many instances where it would have been possible to retain the paradosis, leaving the attractive but inessential changes in the app. crit. A word about the app. crit. of this edition: there are those who deplore these paginal appendices, bristling with mysterious symbols, but if one accepts the app. crit. as a necessary part of a proper text edition (as I think most scholars do), this is a good one, both succinct and informative. Succinctness has been achieved by a simple and clear reference system, well explained in the Preface. Informativeness has been greatly eased and improved by the use of a not-yet-published repertory of conjectures by Dr. L. van Paassen, of the University of Amsterdam, which was made available to the OCT editors. I add some comments, mostly on passages where the OCT accepts readings that seem to me debatable for a variety of reasons. At El. 84-85 te phêmi replaces t' eph' hêmin, with pherei changed to pherein (Tournier), on the grounds that epi with dative, with a verb of motion, can only express hostility. As the note in Sophoclea shows, the editors are uncertain; in such cases it seems to me preferable to keep the paradosis, with the attractive proposal listed in the app. crit. At El. 278 the editors retain eurous' as being "a perfectly natural expression, as one speaks of 'finding' a birthday on a calendar." But can we think that, in the context of this passage, Clytemnestra has to have her memory of the day of Agamemenon's death jogged? To me, Seyffert's hierous', "keeping sacred," seems almost certainly correct, with exactly the right tone of bitterness in Electra's speech at her mother's wicked "religious" celebration. At El. 459, while the conjecture accepted, melein for melon, does not seem to me essential, the note in Sophoclea is excellent. At El.1423 psegein for legein is a change based solely on an interpretation of the outlook of the Chorus, since the phrase with legein is entirely idiomatic. To those who see a substantial degree of distress on the part of the Chorus at what is going on, this will seem an arbitrary and unwarranted change; the uncertainty expressed by legein is appropriate. The note on OT 1214-15 is excellent, but I think Campbell's solution (given in the OCT app. crit.) is best. Ant. 24 is obelized; Schütz's conjecture, accepted by Dain, deserves mention. Ant. 203: Musgrave's ekkekêruktai for ekkekêruchthai is accepted; but since the paradosis "can be defended" (Sophoclea 122) it should be kept, as the difficilior. While the note on Ant. 1219 is good, it seems to me that exathumou is worth considering. It would make no verbal change from the MSS, but supposes such an adjective (analogous to athumos from athumeô) to exist. At Tr. 7 oknon is more appropriate than otlon and has better MSS support. At Ph. 43 Toup's recherché mastun is accepted, but while it provides a desirable meaning, nostos can mean "journey," not necessarily "return"; two of the scholars whose work is most respected by the editors, Wilamowitz and E. Fraenkel, accept noston; the change is tempting but not needed. Postgate's epathrôn, at Ph. 452, reduces Philoctetes' words to a cliche; as Kamerbeek's note shows, epainôn needs no change. At Ph. 1092 the editors accept Jackson's solution (which they rightly call "brilliant") to a difficult passage. At Ph. 1251 the editors seem not to have noticed that phobon as well as straton repeats a word used by Odysseus; there is no need to change phobon. At Ph. 1401, tethrulêtai is most unlikely to be said by Philoctetes of himself; if this reading is accepted, soi is needed in place of moi. At OC 210, the three 's are indeed "most unusual" (Sophoclea 224), but they are not for that reason wrong; the extreme emphasis is appropriate. The threefold, emphatic negation at Ph. 1300, which the editors do not question, is equally unusual. Moreover, in the Ph. passage the editors are accepting a conjectural reading which introduces the third negative. At OC 487 the editors accept Bake's sôthrious for sôthrion; the paradosis is either a very elementary error, making the adjective agree with whatever is nearby, or a very subtle suggestion of Oedipus's power; I agree with those who interpret it as the second. At OC 516, anaidôs (nos) for anaidê certainly straightens things out, but it sacrifices the interesting oxymoron peponth' anaidê which may well be right. To say concerning OC 883 that Creon "would scarcely boast" (Sophoclea 244) of his hubris may underestimate his arrogance. The note in Sophoclea rejecting Jebb's interpretation of OC 1640 is valid, and it would have been in keeping with the editors' practice of avoiding the obelus to put Maehly's tod' eugenei phreni, which they speak well of, in the text. At OC 1653-55, sun logôi (nos) is not necessary; "shortly, after no long time," is not offensive, cf. Ph. 348-49, while Blaydes's chronôi for logôi at 1655 (which the editors accept presumably to avoid the pointless repetition of logôi) is bad. To say "in the same speech" does not introduce the absurdity that to say "at the same time" inevitably suggests.

Misprints and errors in proofreading are few. The worst, and probably the only potentially misleading one is hêmas for humas at OT 1514. Others that I have noticed: obliterated type at Aj. 20 (a slighter instance also at 9); at Ant. 211 the app. crit. indicates that the text is meant to have poiein rather than poein; since elsewhere the editors accept the spelling poein, which is shared by L and K, it is probably the app. crit. that is wrong. As noted above, the first sentence of the note in Sophoclea on Ant. 574 says the opposite of what is intended, reversing Chorus and Ismene. The note in Sophoclea on Tr. 328 mis-numbers the line as 327. At Ph. 1431, labês is printed for labêis. At OC 1436 the note in Sophoclea has tod' for tad'. The note on OC 1752 in Sophoclea has kona for koina.