The context shows that the intention of the lines was to bring out the surpassing beauty of a certain girl and its value to the chorus as a whole. When the Pleiades rise up the sky, they are followed by a star that far outshines them all: Sirius. In Alcman's image, then, the Pleiades should correspond to the chorus and Sirius to the girl. The point of opdpiaiis that the comparison is not chosen at random, but suggested by something to be seen during the current ceremonies: the Pleiades rise up the sky before dawn when we carry the plough, with Sirius down below them, and they seem like a rival group.

There is no reference to Ortheia, and no rival chorus. We may translate: ‘For the Pleiades range themselves against us, before dawn, as we bear the plough through the ambrosial night, bringing Sirius up with them as they do.’

In his discussion of the ethics of suicide (Phaedo 61 c ff.) Plato alludes to more than one traditional injunction against it:indicates a fairly general acceptance of its wickedness. Cebes has heard the Pythagorean Philolaus, among others, saying that suicide was immoral, but has gathered no satisfactory explanation as to why this should be so. One reason, impressive, but, Socrates admits, difficult is to be found

In two of his dialogues especially, the Sophist and the Parmenides, Plato concerns himself at length with problems presented by the Eleatics. Despite difficulties in the interpretation of individual passages, the Sophist has in general proved the less difficult to understand, and since some of the problems at issue in the two works indicate the same or similar preoccupations in Plato's mind, it is worth considering how far an interpretation of the ‘easier’ dialogue can be used to forward an interpretation of the more difficult. First, therefore, we must identify problems common to the two works; then we must see whether we can understand what Plato understood Parmenides to have done—this may help towards an understanding of what he did in fact do; finally we can apply our findings to the Parmenides itself, particularly to the problem of the unity of the dialogue, in the hope that Plato's intentions may become clearer.

In 191O Wilamowitz suggested that the account of the election of the first Magnesian officials (Laws 751 a–755 b) is a conflation of two originally separate sets of proposals. After long neglect his arguments have been resurrected, with one major modification and in more detail, by Morrow. I intend to argue that both commentators are fundamentally mistaken, and that, properly interpreted, the passage yields limited but valuable information about Plato's plans for coping with the problems of founding a state from scratch. These plans are not simply of theoretical interest: as D. A. Russell has remarked, the Laws is our best guide to the policies and practices of the constitutional advisers sent out by the Academy.

Fuhrmann's work on the manuscripts of Anaximenes', finally made public in his Teubner text (Leipzig, 1966), has left the ground clear for critical operations. A solid start was made by Spengel and Kayser (for references to their contributions, and to those of other scholars, see Fuhrmann, pp. xlvi–xlviii, whose abbreviations are used below); but that there are still serious flaws in the text has recently been shown by R. Kassel (Philologus 1967, 122–6). The main purpose of the following notes is to air difficulties, some afresh, some for the first time.

The second example is apt, the first not, because the author is discussing not outright illegality, which falls within the compass of TO, but two ways of bringing within the compass ofacts that are not expressly illegal. Deletion of the first example must therefore be considered. It should not be thought, however, that there is anything wrong with the asyndetic coupling of two examples that both lead off with(cf. p. 10. 9–11).

A. Agreement of Participle

(1) Masculine takes precedence over feminine: e.g.

In the first two examples (for which see also II. A. 2 below) the participle may be conceived of as agreeing with the nearer of the two subjects, since it is expressed in the masculine singular. Likewise,refers specifically to. But the third example, in which the participle is in the masculine plural, clearly demonstrates the usual preference for masculine.

I should like to draw attention to a metrical phenomenon observable in the hexameters of Callimachus and propound a ‘law’ which so far as I know has not been remarked on before; the accompanying discussion involves some refinements to our understanding of the metrical effect of proclitics of general importance to Greek metrical studies. In analysing the data I have made use of some standard statistical methods which could in my view be used throughout the whole field of Greek metrical studies with great profit (to emphasize this I have presented the evidence as thoroughly as seemed possible). The analysis also provides some detailed evidence useful for a comparison of Callimachus, Apollonius Rhodius, Theocritus, and Aratus, and the earlier writers by whom the Hellenistic poets were influenced.

The text as published runs:

The elegiacs on side (a) of this fragmentary piece of papyrus are identifiable as by Callimachus, probably from the Aetia, and these lines too are undoubtedly by the same author, and almost certainly from the same work. Verse 5 is a surprise, for it was thought until the discovery of this papyrus to be by Euripides; however the only source for this attribution is Stobaeus (Eel. 1. 3. 6), in whom it appears as the first line of a two-line quotation. It is not unusual in Stobaeus for two originally unconnected lines to be mistakenly combined (for further references and comments see John Barns, ‘A new Gnomologium (II)’, CQ N.s. i (1951), 18–19).

In the controversy over the date of Corinna, the following points may be taken as agreed:

1. An edition was made in Boeotia about the end of the third or beginning of the second century B.C.

2. The texts of Corinna current in the late Hellenistic and Roman periods were all descended from that Boeotian edition.

3. Before its dissemination, Corinna was unknown in Greece at large. If she wrote at an earlier period, she must have been remembered only locally.

The difference between Boeotian spelling of the fifth century and that of the fourth is very great: but the difference in this respect between the mid-fourth century and the late third or early second is comparatively slight. It is therefore tenable that whereas there would be a good reason for the re-spelling of fifth-century Boeotian into the later convention of any period, there would be no obvious or adequate reason for re-spelling Boeotian of the fourth century into the orthography of the third, or that of the third into that of the second. Even those features of fourth-century spelling which have ceased to preponderate are by no means unknown or even uncommon at the end of the third century.

Didymus, in modern works of reference, gets rather a good press. It is conceded on all sides that he was not an original researcher and that his remarks often betray a certain want of common sense. But the general estimate is favourable: more recent works do not substantially dissent from Sandys’ verdict (History of Classical Scholarship, i. 143): ‘The age of creative and original scholars was past and the best service that remained to be rendered was the careful preservation of the varied stores of ancient learning; and this service was faithfully and industriously rendered by Didymus.’ His industry is, of course, beyond question. But it is perhaps surprising that his reputation for accuracy did not drop sharply with the publication of the Berlin papyrus of the,the defects of which cast grave doubt on his reliability and general competence.

The literary legacy of Aramaic-speaking Christianity consists predominantly of ecclesiastical works—theological treatises (both original and translations), sermons, hymns, and the like; it is for the most part, one must admit, rather dull stuff. Distinguished from the rest, and of peculiar interest to classical students, are secular works, translated from the Greek, which include, apart from medical and scientific treatises, a handful of writings by Plutarch, Lucian, and Themistius. Baumstark suggests that the translator of these three Greek writers be identified as Sargis (died 536), a learned priest and ⋯ρχιατρός in Theodosiopolis, with a somewhat chequered ecclesiastical career (he changed sides in the christological controversy, starting out as a monophysite and ending up in the Chalcedonian camp), who is known as the translator of a number of philosophical and medical treatises. Sargis has his place in the history of thought, for it was in the first place through his Syriac translations that the Arabs became acquainted with Galen, whose works eventually assumed almost canonical status with them.

It is universally recognized that the Hymn to Ares stands apart from all the other poems in the Homeric collection, and that it was composed centuries later than any of those that can be assigned to a particular period with any degree of confidence. Many older scholars classed it or even printed it with the Orphic Hymns, which are transmitted together with the Homeric Hymns as well as with the hymns of Callimachus and Proclus. But the similarity with the Orphic Hymns is only superficial. It merely consists in the stylistic feature of accumulated epithets, a common characteristic of late hymns. There are three reasons for not lumping the Hymn to Ares with the very homogeneous Orphic series. First, the Orphic Hymns are courteous invocations of different gods to come to a ceremony described as τєλєτή or τєλєτταί, while the Hymn to Ares is an intensely personal prayer for the poet's own soul. Secondly, the Orphic collection already contains a hymn to Ares (65), and duplications are avoided. (There are a number of hymns, to Dionysus, but then he bears a different title in each case, Liknites, Trieterikos, etc.) Thirdly, in the Homeric hymn Ares is addressed as a planetary god. In the Orphic hymn to Ares, this aspect is totally absent, and the same is true of the hymns to Hermes, Aphrodite, Zeus, and Kronos.

The evidence of ancient books points to the surprising conclusion that in texts of drama or prose dialogue changes of speaker were not usually marked by the name of the new speaker. Instead the ancient reader had a colon, sometimes combined with a paragraphus or stroke in the margin, to guide him. The inconvenience of this practice and the muddle it caused need no emphasis. The facts have been assembled for the text of Plato and Lucian by J. Andrieu (Le Dialogue antique, 288 ff.), and for Aristophanes by J. C. B. Lowe (Bull. Inst. Class. Stud, ix (1962), 27–42). As far as prose dialogues are concerned confirmation can be found in an unexpected source: the prologue to the dialogue Eranistes by the fifth-century church father Theodoret (Migne, Patrologia graeca, lxxxiii. 29 b).

The verb λάω is attested in two passages of early epic poetry, (a) Homeric Hymn to Hermes 360, where the infant Hermes is hiding in a dark cave, and (b) τ 229 ff., of a hound seizing a fawn on the brooch of Odysseus. Of the several meanings suggested by the ancient lexicographers for λάω, seeing, gazing, or crying, screeching would suit (a). These senses recur in their explanations of (b), with gripping or devouring as additional possibilities. The most extensive modern treatment of λάω is by Leumann, who explains it as a present falsely formed from the perfect λ⋯ληkα (X 141, of a hawk), and originally intended to describe the cry of a bird of prey. The unfamiliarity of the form led to its being associated later on with the sharp-sightedness of such birds, as well as with the bark of a hound fastening on its quarry.

The following notes on the In Pisonem are largely based on the commentary of R. G. M. Nisbet (Oxford, 1961) (‘N’). The references to the speech are by section and line of his text, and where my note is based on one of his I add a reference to the page of his commentary.

1. 20 (p. 58) voltus …, qui sermo quidam tacitus mentis est: ‘thoughts are usually revealed by the face.’ Add to Otto's, Seyffert-Muller's, and N.'s examples: Curtius 8. 6. 22 ‘voltus haud sane securi animi index’, Juv. 9. 18 ff.

These three Ciceronian references, each used only once, have given rise to a most confusing variety of interpretations. In this article I hope to show, as far as the evidence will allow, who these poets were and what sort of poetry Cicero probably had in mind.

οί νєώτєροι

THE first Epode provides no clear indication of date. We learn only that Maecenas is about to join Octavian on a dangerous expedition and has suggested that Horace should not accompany him, while Horace retorts that he will be unable to enjoy himself in the absence of his patron and would be ready to follow him to the ends of the earth, whatever the danger, in the hope of earning his gratitude.

The Epodes were published about 30 B.C. and, perhaps for that reason, the scholiast Pseudo-Acron confidently assigns the poem to the period immediately before the battle of Actium with the comment: ‘Maecenatem prosequitur euntem ad bellum navale cum Augusto adversum Antonium et Cleopatram.’ It is curious that in referring to the activities of Octavian in 31 the author uses the title Augustus, which was not conferred on him before 27,: but the substance of his comment accords with the known facts. In the spring of 31 Octavian asked all Romans of influence to meet him at Brundisium, thereby to demonstrate their willingness to take part in the war against Antony and Cleopatra, and more than seven hundred senators are known to have offered their services.

I Agree with R. G. Austin, who in his recent paper (CQ, N.S. xviii (1968), 107 ff.) showed that Virgil did not write this proem to the Aeneid, and suggested (p. 115) that it was produced in the first half of the first century, perhaps prompted by the problem mentioned by Servius on A. I. I: ‘multi varie disserunt cur ab armis Vergilius coeperit.’ I wish here to comment briefly on the content of the lines.

(a) gracili

qui... carmen refers to the writing of the Eclogues. As Austin points out (p. 112), Virgil has gracilis only in E. 10. 70 ff., the conclusion of the whole of the Eclogues:

Haec sat erit, divae, vestrum cecinisse poetam,

dum sedet et gracili fiscellam texit hibisco,

Pierides…

It was natural and inevitable that his two Aethiopias, in the eastern and western extremities of the world, should be identified with the countries of the two dark-skinned peoples in the Far East and the Far West of the Ancient World: India and Mauretania.

There was the difficulty that the real Aethiopia was in Africa, neither in the Far East nor in the Far West. Serious writers on geography tried to reconcile Homer and the geographical facts.

I wish to excerpt from Strabo three theories concerning Aethiopia and the Aethiopians which will then be used to elucidate passages in Latin literature.

10. 2. lugentem timentemque custodire solemus, ne solitudine male utatur.

Reynolds does not mention Haupt's conjecture amentemque, which is certainly on the right lines. Bereaved persons may need watching because in the violence of their grief they may do themselves an injury (cf. Ben. 2. 14. 2 ut frigidam aegris negamus et lugentibus ac sibi iratis ferrum), and the same applies to madmen (cf. Macer, Dig. 1. 18. 14 nam custodes furiosis non ad hoc solum adhibentur ne quid perniciosius ipsi in se moliantur, sed sqq.) or to anyone suspected of suicidal inclinations (cf. Sen. Contr. 3. 5 minantem sibi ipsi (sc. filiam) custodio). It does not apply to persons afraid; they may sometimes be glad of company, but do not require surveillance. My only doubt is whether amentem, seldom used of actual insanity, is the right word. Perhaps dementem or furentem, which makes a better match with the participle lugentem.