The Classical Papers of A. E. Housman collected and edited. Volume I: 1882-1897

THE CLASSICAL PAPERS OF A. E. HOUSMAN VOLUME I

A. E. H. at the age of 67

THE

CLASSICAL PAPERS OF

A.E.HOUSMAN COLLECTED AND EDITED BY

J. DIGGLE & F.R.D. GOODYEAR Volume I 1882-1897

CAMBRIDGE AT THE UNIVERSITY PRESS 1972

PUBLISHED BYTHE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE The Pitt Building, Trumpington Street, Cambridge, United Kingdom CAMBRIDGE UNIVERSITY PRESS The Edinburgh Building, Cambridge CB2 2RU, UK 40 West 20th Street, New York NY 10011-4211, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia Ruiz de Alarcon 13,28014 Madrid, Spain Dock House, The Waterfront, Cape Town 8001, South Africa http ://www. cambridge.org This collection and edition © Cambridge University Press 1972 This book is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 1972 First paperback edition 2004 A catalogue recordfor this book is available from the British Library Library of Congress Catalogue Card Number: 74-158552 ISBN0 52108243 9Vol.I ISBN0 521 08511 X Vol. II ISBN 0 52108479 2 Vol. Ill ISBN 0 521 08243 9 hardback ISBN 0 521 60694 2 paperback

CONTENTS Portrait of A. E. Housman by Francis Dodd (Reproduced by permission of the National Portrait Gallery)

Frontispiece

Preface

vii

List of editions consulted

xi

Abbreviations I

Horatiana [I]

2

Ibis 539

xv 1

9

On Soph. Electr. 564, and Eur. I.T. 15 and 35 4 On certain corruptions in the Persae of Aeschylus 5 Isocr. Paneg. §40 6 Schol. on Aesch. P. V. 488 [[472] 7 CWOPONH 8 Emendationes Propertianae 9 The Agamemnon of Aeschylus Note on Emendations of Propertius IO II Horatiana [II] 12 Notes on Latin poets [I] J 3 Persius m. 43 14 Conjectural emendations in the Medea Review: T. G. Tucker, The Supplices of Aeschylus i5 16 Review: I. Flagg, Euripides' Iphigenia among the Taurians Horace, Carm. iv. 2, 49 17 18 Notes on Latin poets [II| J 9 Horatiana [III] 20 Emendations in Ovid's Metamorphoses 21 The new fragment of Euripides 22 Adversaria orthographica 2 3 The Oedipus Coloneus of Sophocles 24 Sophoclea Remarks on the Vatican Glossary 3321 25 26 Soph. Oed. Col 527 27 The manuscripts of Propertius [I] 28 The manuscripts of Propertius [II] 29 Review: K. P. Schulze, Catulli Veronensis liber 3

[v]

10

14 22

23 24 29

55 91 92 106

no 112 120

128 131 132 136 162 173 175 181 209 227 231 232 277

305

CONTENTS 30 The manuscripts of Propertius [IIIJ 31 A note on Virgil 32 The manuscripts of Propertius [IV] 33 Review: J. P. Postgate, Sexti Properti carmina 34 Cicero Pro Milone c. 33 §90 35 Ovid's Heroides [I] 36 Ovid's Heroides [II] 37 Ovid's Heroides [III] 38 Ovid's Heroides [IV] 39 Ovid's Heroides [V]

314 348 351 369 378 380 388 396 404 412

PREFACE ' When a scholar of A. E. Housman's eminence has disposed of much of his work in periodicals, the convenience of scholars no less than respect for his memory commonly demands the publication of his Collected Papers.' These are the opening words of A. S. F. Gow's biographical sketch, published in 1936, the year of Housman's death. The reason why the re-publication of Housman's papers has been delayed for over thirty-five years is to be found in a clause of his will: 'I expressly desire and wish my desire to be made as widely known as possible that none of my writings which have appeared in periodical publications shall be collected and reprinted in any shape or form.' For undertaking the publication of these papers in defiance of Housman's express wish we offer no apology, but we offer our reasons, and we offer them with all brevity. It is proper that Housman's friends and scholars nearer to him in time should have bound themselves by this prohibition; but posterity cannot be bound for ever. Housman's reputation is secure; and if he feared that the re-publication of earlier judgements later recanted might impair that reputation, or that the absence of an explicit recantation might be interpreted as a continued endorsement, then his fears were unfounded. And even if such fears had been well founded, littera scripta manet, nescit uox missa reuerti. In fifteen years the original copyright of Housman's papers expires, and it could not be expected that their re-publication would be delayed much beyond that time. It would be sad indeed to see them published in a hasty and slipshod manner, or otherwise than in their entirety. Our edition has, we hope, precluded this danger. We have included all the papers and reviews on classical subjects published by Housman in learned journals; and we have placed in an Appendix the reports of papers read to the Cambridge Philological Society, when these were not published elsewhere in full, and contributions to the papers and books of other scholars when these were expressed in Housman's own words and offered a substantial statement of opinion, and several miscellaneous items which were unsuited to inclusion in the body of the work. We have decided to publish nothing which Housman did not publish or countenance the publication of in his lifetime. We have therefore made no attempt to include the 1911 Cambridge Inaugural Lecture or such other unauthorised pieces as have been printed or mentioned in print from time to time. That the London Inaugural Lecture of 1892 falls outside our scope may perhaps be disputed; but it can conveniently be read elsewhere. Apart from items later published in a fuller form, we have knowingly omitted only one item which may properly be described as a 'classical paper': the transcription of readings from three manuscripts of Ovid [vii]

viii

PREFACE

which Housman appended to a paper by J. P. Postgate in the Journal of Philology 22 (1894), 154-6. Without the bibliography compiled by Mr Gow, and the supplement to it published by Professor G. B. A. Fletcher in the Durham University Journal 38 (1946), 85-93, our task would have been more difficult and the result less complete. We have been able to add only one item to those which they have recorded, and we cannot be certain that there are not other items which we might have included. We have tried to present what Housman wrote with the least possible obtrusion of editorial comment. The papers are presented in chronological sequence by years; within each year they are presented according to the alphabetical sequence of the titles of the periodicals in which they appeared. We have checked every reference. Those which we have found to be wrong (they are not few) we have silently corrected. Where Housman has cited the numeration of an edition which is no longer in general use, we have added in double square-brackets a reference to the numeration, if it differs, of the edition which may now be considered standard. The editions which we have used are listed on pp. xi-xiii. Misprints we have silently corrected, and we have reduced to a standard form certain insubstantial typographical details over which Housman probably had no control or perhaps had no concern. We have followed the practice of the Cambridge University Press in the elision of numerals (e.g. 107-11 instead of 107-111, and 117—18 instead of 117-8 or 117-118), but in all other respects we have maintained the style of reference used in each paper. There are many remaining inconsistencies in presentation which we have thought it unwise or unnecessary to remove. Whatever else we have added, such as the references to the places of original publication, or the cross-references in later papers to the page-numbers in this edition of earlier passages to which Housman has alluded, we have enclosed in brackets. Two vertical strokes in the text indicate the division between pages in the original publication; the original page-numbers are added in brackets at the top of the page. The footnotes have been numbered serially by the pages of this edition. To have retained the original numbers would have been cumbersome and confusing, and the reader can work them out for himself, if he wishes, without undue effort. In the Indexes we have included all passages to whose interpretation or emendation Housman has made some contribution, and all words and topics which are discussed by him. It has often been difficult to adjudge entitlement to inclusion: we have preferred to be liberal rather than parsimonious in our decisions. Where we have added to the text a newer reference in brackets, the Index of Passages incorporates this and not the older reference. We could not have completed this work had we not received at the outset the encouragement and help of various friends, in particular Mr E. J. Kenney, and the consent of Housman's great-nephew, Mr R. E. Symons. Our gratitude is also due

PREFACE

ix

to the Librarian of the University of Cambridge for allowing his staff to undertake the laborious task of producing two xerox prints of every page of the original papers, to the Syndics of the Cambridge University Press for undertaking the publication of this work, and to the staff of their Printing House and Editorial Office for the patience, skill, and expert knowledge which they have brought to its production. We have appealed to Dr J. B. Hall many times for advice over editorial problems; and from him, as from Professor H. D. Jocelyn, we have received generous and welcome help in the correction of the proofs. A friend who has given his encouragement to this work from the beginning, and to whom for many kindnesses we are both indebted, is also the holder of Housman's chair at Cambridge. If we had felt free to prefix a dedication to this edition of Housman's papers, we should have dedicated it to Professor C. O. Brink. Queens College, Cambridge Bedford College, London

J. D. F. R. D. G. 30.9.1971

LIST OF EDITIONS CONSULTED' GREEK

Aeschylus fragmenta Anthologia lyrica Graeca2 Arcadius Bacchylides Callimachus Comicorum Atticorum fragmenta Erotian Euripides fragmenta

Herodotus Hesiod fragmenta Hesychius Hippocrates Libanius Menander Nonnus Orphica Paulus Alexandrinus Pindar

G. Murray A. Nauck H. J. Mette E. Diehl

Oxford (OCT) 1955 Leipzig 1889 Berlin 1959 Leipzig (Teubner) 1936

A. Lentz, Herodiani technici reliquiae vol. i

Leipzig 1867

B. Snell R. Pfeiffer T. Kock

Leipzig (Teubner) 1961 Oxford 1949-53 Leipzig 1880-8

E. Nachmanson G. Murray A. Nauck H. von Arnim, Suppl. Eurip. D. L. Page, Greek lit. pap. vol. i C. Hude F. Solmsen R. Merkelbach and M. L. West K. Latte E. Littre R. Foerster A. Koerte and A. Thierfelder R. Keydell E. Abel E. Boer B. Snell

Goteborg 1918 Oxford (OCT) 1902-9 Leipzig 1889 Bonn 1913 London (Loeb) 1942 Oxford (OCT) 1927 Oxford (OCT) 1970 Oxford 1967 Copenhagen 1953— Paris 1839-61 Leipzig (Teubner) 1903-23 Leipzig (Teubner) 1955-8 Berlin 1959 Leipzig & Prague 1885 Leipzig (Teubner) 1958 Leipzig (Teubner) 1959—64

1 This list comprises editions of those texts in which sometimes or always we have found it necessary to supply in double square-brackets references additional to those given by Housman. For the numerous texts of which no edition is mentioned here it may be assumed that we have checked Housman's references against the latest edition (OCT or Teubner or other) commonly in use and that we have found no discrepancy. 2 From this edition we cite only those poets who are not included by Page in Poetae melici

Graeci.

[xi]

LIST OF EDITIONS CONSULTED Poetae melici Graeci Proclus, chresu Ptolemy, tetr. Sophocles fragmenta Stobaeus Strabo Suidas Theon Smyrnaeus

D. L. Page T. W. Allen, Homed opera vol. v F. E. Robbins A. C. Pearson A. Nauck A. C. Pearson C. Wachsmuth and O. Hense A. Meineke A. Adler E. Hiller

Oxford 1962 Oxford (OCT) 1912 London (Loeb) 1940 Oxford (OCT) 1924 Leipzig 1889 Cambridge 1917 Berlin 1884-1912 Leipzig (Teubner) 1852-3 Leipzig 1928-38 Leipzig (Teubner) 1878

LATIN

Aetna Asconius Ausonius Catullus Charisius Cicero, Q.fr. Cyprianus Gallus, kept. Ennius, arm. trag. Festus Filastrius Firmicus Maternus

F. R. D. Goodyear A. C. Clark R. Peiper R. A. B. Mynors K. Barwick W. S. Watt R. Peiper

J. Vahlen H. D. Jocelyn W. M. Lindsay F. Marx W. Kroll, F. Skutsch and K. Ziegler M. P. J. van den Hout Fronto H. Keil Grammatici Latini H. J. Rose Hyginus,^. P. Brandt Lactantius, inst. epit. C. Bailey Lucretius A. E. Housman Manilius W. M. Lindsay Martial J. P. Waltzing Minucius, Oct. W. M. Lindsay Nonius Oratorum Romanorum H. Malcovati fragmenta

Cambridge 1965 Oxford (OCT) 1907 Leipzig (Teubner) 1886 Oxford (OCT) 1958 Leipzig (Teubner) 1925 Oxford (OCT) 1958 Vienna (CSEL) 1891 Leipzig 1903 Cambridge 1967 Leipzig (Teubner) 1913 Vienna (CSEL) 1898 Stuttgart (Teubner) 1968 Ley den 1954 Leipzig 1855-80 Leyden 1934 Vienna (CSEL) 1890 Oxford (OCT) 1922 Cambridge 1937 Oxford (OCT) 1929 Leipzig (Teubner) 1931 Leipzig (Teubner) 1903 Turin 1967

LIST OF EDITIONS CONSULTED 1

Ovid, her. Ibis met. trist. and ex P. Petronius Phaedrus Poetarum Latinorum fragmenta Poetarum scaenicorum fragmenta2 Probus, in Verg. hue. et georg. Propertius Scholia in Ouidi Ibin Seneca maior Seneca, apoc. nat. Solinus Statius, Ach. silu. Suetonius, Caes. gramm. Tacitus Tibullus Valerius Flaccus

R. Ehwald A. La Penna R. Ehwald S. G. Owen K. Miiller J. P. Postgate W. Morel

Leipzig (Teubner) 1916 Florence 1957 Leipzig (Teubner) 1915 Oxford (OCT) 1915 Munich 1961 Oxford (OCT) 1919 Leipzig (Teubner) 1927

O. Ribbeck

Leipzig (Teubner) 1897-8

G. Thilo and H. Hagen, Seruii in Verg. comm. E. A. Barber A. La Penna H. J. Miiller F. Buecheler and W. Heraeus A. Gercke T. Mommsen H. W. Garrod J. S. Phillimore M. Ihm G. Brugnoli E. Koestermann J. P. Postgate E. Courtney

Leipzig 1880-1902 Oxford (OCT) 1953 Florence 1959 Vienna 1887 Berlin 1922 Leipzig (Teubner) 1907 Berlin 1864 Oxford (OCT) 1906 Oxford (OCT) 1918 Leipzig (Teubner) 1908 Leipzig (Teubner) i960 Leipzig (Teubner) 1960-2 Oxford (OCT) 1914 Leipzig (Teubner) 1970

1 In our references to Ehwald's edition of the heroides we have ignored the re-numbering of 15-21 which he adopts, although we continue to give his line numbers in the individual poems. 2 Except the fragments of Ennius' tragedies, for which we refer instead to Jocelyn.

ABBREVIATIONS AJPh BPh W CPh CQ CR CSEL GLK GLP JPh OCT ORF PC A PCPhS PhW PLF TCPhS

American Journal of Philology Berliner Philologische Wochenschrift Classical Philology Classical Quarterly Classical Review Corpus Scriptorum Ecclesiasticorum Latinorum Grammatici Latini, ed. Keil Greek Literary Papyri Journal of Philology Oxford Classical Text Oratorum Romanorum Fragmenta Proceedings of the Classical Association Proceedings of the Cambridge Philological Society Philologische Wochenschrift Poetarum Latinorum Fragmenta Transactions of the Cambridge Philological Society

[xv]

1 HORATIANA [II* Carm. n 2 1-4 Nullus argento color est auaris abdito terris, inimice lamnae Crispe Sallusti, nisi temperato splendeat usu. Alike Lambinus' 'abditae' and Bentley's the only rational elucidation of the MS reading compel the words * auaris terris' to mean the miser's coffers: now when Horace says carm. ill 3 49 sqq. * aurum inrepertum et sic melius si turn, Cum terra celat, spernere fortior Quam cogere humanos in usus' he is to be sure taking the other side as a poet may, but the parallel does seem to show that 'auaris terris' here must have its natural sense of the mine, ' in her own loins She hutcht the all-worshipt ore' as Comus says. And is not 'inimice lamnae, nisi temperato splendeat usu' or 'auaris abditae terris inimice lamnae' a most dark and helpless way of saying 'open-handed Sallust'? And then how 'inimice' and its train of dependants encumber and overbalance the sentence. If then as seems likely it is in 'inimice' the corruption lies, this is what I would suggest: nullus argento color est auaris abdito terris, minimusque lamnae, Crispe Sallusti, nisi temperato splendeat usu. ' Silver in the mine has no lustre at all, nay even when coined it has next to none, without it is burnished by changing hands.' This at least does away with the obscurity and redresses the || balance of the sentence. It is chiefly I suppose because Horace was at no period unread that the corruptions in his MSS seldom lie on the surface, they present a resemblance at least superficial to sense and metre: if'minimusque' by two common errors became 'inimusce' the further change to 'inimice' was all but inevitable. Carm. 111 5 31-40 Si pugnat extricata densis cerua plagis, erit ille fortis * IJPh 10 (1882), 187-96] 187-8]

GHC 1

1

2

HORATIANA [I]

[188-9

qui perfidis se credidit hostibus, et Marte Poenos proteret altero qui lora restrictis lacertis sensit iners timuitque mortem, hie unde uitam sumeret inscius pacem duello miscuit. o pudor! o magna Carthago probrosis altior Italiae minis. In this the reading of most MSS and well-nigh all editions Bentley justly finds fault with the lame climax 'timuitque mortem', and 'hie* used where the poet should and might have used 'ille': he might too have added, what sort of writer is Horace if 'mortem' and 'uitam' here have nothing to do with one another? But there is this deeper fault in the reading, that it makes Regulus lose the thread of his argument; for what is he debating? not what is done and cannot be undone, the surrender of the army, but its ransom, the matter in hand: his aim is to fence off the pernicies ueniens in aeuum, the flagitio additum damnum, the probrosae Italiae ruinae, and down to v. 36 he is speaking straight to the point; but here with a full stop a t ' mortem' he loses his way and drifts off into mere exclamation about what is past mending and will remain the same whether he gains his cause or loses it. But several good MSS, that of Queen's College Oxford among them, have 'aptius' for 'inscius', and very many more give it for a varia lectio: Bentley then, accepting this, proposed 'timuitque mortem Hi/zc, unde uitam sumeret aptius, Pacem et || duello miscuit', comparing carm. m 11 38 'ne longus tibi somnus unde Non times detur'. This removes at once the faults of the passage and saves Horace's credit as a rhetorician: 'hinc' for 'hie' is the slightest of changes, carm. 1 17 14 and 21 13 the MSS have 'hinc' where 'hie' must be right: but his insertion of 'et' has not much likelihood, as he himself tacitly acknowledges on iv 4 18. But can 'pacem duello miscuit' in Horace mean 'confounded war with peace'? Horace who five times elsewhere uses 'duellum' uses it with a marked restriction, always of some single war, never of war in the abstract: the word's fancied connexion with 'duo' was maybe at the bottom of this: war as opposed to peace is 'bellum' carm. 11 1928 'idem Pads eras mediusque belli9 serm. 11 2 111 i in pace j ut sapiens, aptarit idonea hello9 3 268 'in amore haec sunt mala, helium. Pax rursum': if he wants to use 'duellum' thus he must use the plural epist. 11 1 254 'tuisque Auspiciis to turn confecta duella per orbem, Claustraque custodem pads cohibentia Ianum'. I will suggest then that Horace here too was true to his custom and wrote 'pacem^ue hello miscuit': ' u ' and ' b ' are in his MSS as in others much confused, carm. m 23 19 'mollibit' for 'molliuit',

189-90]

HORATIANA [I]

3

i 20 10 where Munro emends 'uides' for 'bides' or 'bibes', 2520 Aldus' 'Euro' for 'Hebro' is probably right: ' b e ' then might well fall out after ' u e ' , and the senseless 'pacemquello' would be readily altered by the change of one letter to 'pacem duello'. Carm. 11111 15-20

Cessit immanis tibi blandienti ianitor aulae Cerberus, quamuis furiale centum muniant angues caput eius atque spiritus taeter saniesque manet ore trilingui. Perhaps neither 'eius' alone nor 'spiritus manet' alone would be intolerable, but surely the pair of them is more than man can stand: so at least thought Bentley Meineke and || Haupt. Haupt and Meineke however betake themselves to the coward's remedy of declaring the stanza spurious: Bentley perceiving that the alteration of 'eius atque' into a verb for 'spiritus' rids us at one stroke of both inconveniences proposes 'exeatque': he cites instances of'spiritus exit' but candidly adds 'verum hie notandum est, quod in his locis spiritus exit de iis duntaxat dicatur, qui moribundi animam expirant. Quare ad evitandum Ambiguum, utinam Noster scripsisset potius exeatque halitus teter9. I propose then 'effluatque' a word which can well be applied to 'spiritus' or the like, Ovid met. VI 233 'ne qua leuis effluat aura\ Cic. n.d. 11 39 [ioij *aer effluens hue et illuc uentos efficit'. Of all errors ' i ' for '1' is perhaps the most frequent, ' s ' for ' f by no means unusual, and if a copyist read or wrote 'essiu atque' then 'eius atque' was not far off. Carm. 111 26 1—8 Vixi puellis nuper idoneus et militaui non sine gloria: nunc arma defunctumque bello barbiton hie paries habebit, laeuum marinae qui Veneris latus custodit. hie hie ponite lucida funalia et uectes et arcus oppositis foribus minaces. Of all weapons the one which doors and doorkeepers can best afford to laugh at is an 'arcus' in any known sense of the word: Bentley's' securesque' however is not likely, no more is Keller's 'et ascias': indeed it surely is plain enough there is no keeping 'et': you can almost count up the available substantives on your fingers and see that none of them will do. But is it a substantive that is wanted?

4

HORATIANA [I]

[190-2

Theocritus cited by Bentley has irgAeiceis Kai AapnrdSes, and that Horace had this in his head is likely enough; but why then when Theocritus mentions only two sorts of 'arma' should he mention three? Surely hatchets alone or crowbars alone are all that is wanted in addition to the torches, and his 'uectes' do duty for Theocritus' || ireAeKEis. Then as to the symmetry of the sentence: 'funalia' has an epithet to itself, and that 'uectes' should tally with it is at any rate as likely as not. What I am trying to make out is that here we have a corruption such as Bentley detected in 'eius atque', that 'et arcus' represents a single word, probably then an imperative co-ordinate with 'ponite': can it be 'et uectes sacrate Oppositis foribus minaces'? 'sacrate' with the change of one letter is 'et arcus' written backwards: to be sure I know of no quite parallel corruption, but in Propertius (Baehrens) m 5 24 DV give 'integras* for 'et nigras' precisely reversing the first four letters. Carm. iv 4 65-8 Merses profundo, pulchrior euenit; luctere, multa proruet integrum cum laude uictorem, geretque proelia coniugibus loquenda. Many seem to have felt the strangeness of 'merses, euenit' followed by 'luctere, proruet geretque', yet 'exiet' is quite out of the question, and 'proruit' and 'geritque' are not very taking. And then the unexampled use of 'euenire'? The MSS vary between 'merses' 'mersus' and 'mersae': 'mersus' which has most authority is of course impossible and is attributed by Keller to the Mavortian recension: among those which have 'mersae' is Keller's liber archetypus F ( = £pOVTOCS

AocuTrpous 5UV&OTCC$; EUTTPETTOVTOCS ocidepi,

doTgpas, OTOCV 90ivcoaiv, avroA&s T£ TCOV.

/ know the stars and the riders of the seasons, the stars to wit. This of course is one of those sentences which a poet does not write; so most editors with Pauw and Valckenaer bracket v. 7 for spurious. It is a good riddance, that I see, but I do not see on what principles of criticism it can be justified: the Aeschylean archaism TCOV never came from the workshop of an interpolator. Fault has of * \JPh 16 (1888), 244-90] 244-5]

55

56

THE AGAMEMNON O F AESCHYLUS

[245-6

course been found, wore ouyyovov PpoToiai TOV TreaovToc ACCKTICTCCI irAeov, with the initial dactyl, and when the faultfinders have got rid of Cho. 215 [216] KCCI Tiva ovvoiaOd \xo\ KocAoujjievT] ppcrrcov; 984 [986] of the same play f|Aios dvocyvcc litjTpos ipyoc 7r\s £|ifjs, Sept. 640 [653] & OeoiicxvES TH KOCI Oecov laeyoc crrvyos1 and fr. 290, 4, Dind. [300 4 Nauck, 193 5 Mette] fjAios ev fj m/pcoTros EKAdjavyocs XOovi, then they may be heard: not before. But Hermann and others, who suppose themselves to have rescued the verse by trifling with the punctuation, ascribe to commas a cabalistic virtue which did not reside in the seal of Solomon. Mr Margoliouth writes 5-7 as follows: KOCI TOUS cpspovTocs xB^a K0Cl Gcpos ppoTois | Aaijnrpous Suvdcrrocs, ijjrcrpeTrovTOcs aiOepi | doTEpas OTOCV 9O1V00CTIV, CCVTOAOCIS TS TOOV: the Aaprnpoi Suvdorca, he says, are the Pleiades. If a year of sleepless nights has taught the watchman so little astronomy that he singles out this nebulous cluster from the host of heaven to call it Aajjnrpos, he is a signal confutation of his creator's favourite doctrine, 7roc0r)|jiocTcc jJocOfmocTa. Nay Mr Margoliouth's own witnesses turn round and testify against him: oAiyoci KOCI d9eyyees, tinOKEvyacrOca d9ca/pai, * ignis uix tenui longe face fit spectabilis' say Ararus and Auienus; and to set against this damaging evidence Mr Margoliouth can find nothing better than the following citation: 'Cic. Progn. jtffugiet cum lucida uisus Pleias.' Now the employment of lucida, by Cicero or by || any one else, as an epitheton ornans for Pleias in a context which does not pit these stars against their fellows, concerns these verses of Aeschylus not a jot; but what is 'Cic. Progn. 356'? If Mr Margoliouth has access to 356 verses of Cicero's Prognostica, he is more fortunate than the rest of the world who know only 27. Truth to tell, however, the words which he ascribes to Cicero were written in the seventeenth century after Christ by Hugo Grotius. If Mr Margoliouth cares for Cicero's account of the Pleiades, here it is: Phaen. 27 'omnis^orte locatas | parua Vergilias tenui cum luce uidebis', 37 'hae tenuesparuo labentes lumine lucent'. And let the Pleiades be as brilliant as you will, masculine they will never be: even Cic. Progn. 356 does not present us with lucidus Pleias. The passage is I believe to be righted, not by the change of a single letter, but by a simpler remedy; the simplest which can be applied to the text of any poet Greek or Roman. We should have heard no evil of the initial dactyl if the MS gave the verses thus: 6

doTpcov KonroiSa vuKTspcov 6|if|yupiv, Aajjnrpous 5uvdoTas eimpETrovTOCs aiOepi,

5

KOCI TOUS 96povT0cs \Bi\xa KOCI Ospos |3poToTs

dorepocs, OTOCV 90ivcoaiv, oVroAds TE TCOV. The watcher is grown acquainted with the stars, which he likens to a congregation of princes, and chiefly with the down-setting and the uprising of those 1

Pers. 287 |284| and Soph. Aiax 1331 seem to answer Mr Verrall's objections to this verse.

246-7]

THE AGAMEMNON OF AESCHYLUS

57

which bring men winter and summer, the stars of the zodiac. These, by which he reckons the passage of his year's vigil, are singled out from the other stars by KOCI, as in Pers. 751 [[749] the god of the sea is singled out from the other gods whom Xerxes fought against when he bound the Hellespont: OECOV 8E TTCCVTCOV WET', OUK Eu|3ouAia, | Kai TTOCTEIBCOVOS Kpcrrfjcjeiv. It should be said that the transposition must have taken place before the time of Achilles Tatius, who quotes vv. 4-6 in the traditional order.

49-59 TpoTrov cdyurricov, 01V EKTT&yAots dAyEcri TTCCISCOV UTTOCTOI AEXECOV ||

50

O T p O < p o 5 tVOUVTOCl,

TTTepuycov EpETuoicnv Ipeaaouevoi, TTOVOV opTocAixcov OAECTCCVTES * UTTOCTOS 6' dcicov fj TIS 'ATTOAACOV

55

f| TTccv fj ZEUS oicovoOpoov yoov o^upoocv T C O V S E JiETOlKCOV UOTEpOTTOlVOV

TTEUTTEI Trapapaaiv 'Epivuv. The learner of Greek, in quest of probable or even plausible reasons for believing that UTTCCTOI AEXECOV summi cubilium means UTTEp AEXECOV super cubilia, is dismissed by Mr Paley to these references: 'ECTXOVTT] x^ 0V °S Prom. 865 [846], UCTTCCTOU vEobs Suppl. 697 [717], UTTOCTOS X ^ P ^ ^EUS inf. 492 [509]'. Thefirsttwo of these passages, TTOAIS Eaxocrr) x^ovos and OIOKOS UOTOTOU VECOS, prove to him

what he could well believe without proof, that such a phrase as OpiyKos UTTOCTOS TEIXOUS a coping which is the highest part of a wall is Greek; but since vultures on the wing are not the highest part of their eyries the information does not help him. Had he been referred, say, to a passage where a fish following a ship is called UOTOTOS VECOS, then he would have been helped; but Greek literature contains no such passage: such a fish is OcrrEpos VECOS. T O the third reference he turns with keen interest, because it is manifest that Mr Paley's translation of UTTOCTOS X^P 0 ^ must differ widely from the usual rendering. But n o : Mr Paley translates like everyone else 'supreme over the country'; and the learner of Greek returns with a touch of resentment from his fool's errand. I propose TTOCI8COV dTrcrrri AEXOCICOV, because their brood is stolen away. T h e phrase TTOCISGOV AEXOCICOV finds an exact parallel in Sept. 278 [29 ij SpaKovTOcs cos TIS TEKVCOV UTTEPSESOIKEV AEXOCICOV SuCTEUV&Topccs TrdvTpoiios TTEAEI&S, the phrase

TTCCISCOV oorcrrri in Soph. Ant. 630 crrraTocs AEX&OV urapaAycov, wroth that he is

58

THE AGAMEMNON OF AESCHYLUS

[247-9

cheated of his bride. My reading is rather an interpretation than an alteration of the MS text: confusions of dor- and vrr- are to be counted not by scores but by hundreds; and for century on century 01 was identical with TJ in pronunciation, || and £ with ecu In the passage which I have cited from the Septem, Asxocicov had to be restored by Lachmann: the MS there as here gives Asxecov. As for the scansion of Aexocicov as an anapaest, the penultimate cci of TreTpccTos is shortened by Sophocles in lyrics, Ant. 827, of TTOCAOCIOS and 5IKOCIOS by Euripides in senarii, El. 497 and Cycl. 274, of yepocios by Sophocles in lyrics, O. C. 200, and by Euripides in anapaests more than once, of 8eiAccios by Sophocles in lyrics, El. 849, and by Aristophanes in senarii over and over again. In Soph. Ant. 1240 the MS gives w TOC VU^IKOC I T£AT| Aocxcbv SeiAocios ev Ai8ou 66|iois: the conjecture e!v is not to be dreamt of: Sophocles seems to have written Aocxcbv ev wAi8ov SsiAccios 86JJIOIS TeAr| or 86|iois £v wAi8ou 8eiAocios T£AT) Aocxcbv. Finally in the Agamemnon itself, v. 723 [722], is found eu9iA6irai8a ml yepocpois enrixocpTOv: which now is the more prudent, to confer on yepapols an alien and unexampled meaning, unexampled, for Supp. 675 [667] proves nothing at all, or to suppose that here, as in the same word in Eur. Supp. 43 (yepocicov Markland, yepocpcov MSS), a scribe confused two letters which in old uncials can hardly be distinguished, I and P? I take the second alternative: uiris doctis aliter uisum. But another check awaits us in v. 58. The dissension about the meaning of TCOVSE |i£ToiKcov is of long standing. The scholiast refers the words to the nestlings, and renders Crrrsp TCOV neTouaoflEVTCov VEOCTCTCOV, a version which of course is peremptorily forbidden by TCOV8S. Another explanation is given in the scholion on Soph. O. C. 934 and reappears, somewhat curtailed, in Suidas: AlcxXuAos.. .£V 'Ayaii^jivovi.. .jiSToiKOUS.. .ehxs TCOV uv|/r|Acov TOTTCOV TOUS OICOVOUS . . .&VT! TCOV EVOIKCOV. But obviously a poet who writes thus, a poet 6s x ' rrepov jiEV KSUOTI ev! 9pealv aAAo 8e EITTT), cannot hope that his audience will understand him: he might as well call the birds TreAei&Socs OCVTI ociyu-mcov. Mr Paley says that the parent vultures are called HETOIKOI to contrast them with the HEToiKoi of Athens who could obtain redress at law only through upoorocTca; an allusion frigid in itself, and so carefully obscured that even after Mr Paley has told us it is there one scans the Greek for it in vain. || It will be conceded that a copyist who found in his exemplar the letters Tcov5eipi8TOKcov would be likely to make Greek of them by transposing the single letter 1 to the place it holds in the text to-day. Such transpositions, intentional or unintentional, are common enough: in one play I notice these three: Supp. 22 iepoo'i&iriotcn for foioo ibii loicn, 278 [272] A£yoi irpoacos for Aeyois TTpoaco, 961 [950] 1061 |aeV through crOupiev for loiypiev. But if my hypothetical copyist had been so faithful or so dull as to write what he read, criticism would before now have restored, letter for letter, a phrase which seems to me the most appropriate in the world, TCOV OCIVOTOKCOV. The substitutions 6 for a, ei for 1, u for

249-50]

THE AGAMEMNON OF AESCHYLUS

59

v and e for o are so common, not in Aeschylus merely, but all of them in most Greek MSS and some of them in all, that I will not fill with illustrations the pages which might be filled; but take two instances where the cci of crivos by changing to 8si has wrought further mischief: in Soph. O. C 212 T68*; ccivd is restored by Wunder for T66E; 5EIVCC, and in Eur. Med. 640 7rpocr|3dAoi u ' ccivoc by Verrall forTrpoapdAoijJi Seivd.1 The terms aivoT6Kos,aivoT6Keia,aivdT8Kouaaare especially applied to parents rendered wretched by the calamities of their children: //. A414&1J101 TEKVOV k\xovt TI vu <J* ETpecpov aivocTEKoOaa; says Thetis to Achilles; Oppian. Hal. v 526 |jrjTpt Trap' atvoTOKcp, the mother dolphin whose young one the fishermen harpoon; Nonn. Dion. 11 160 [162 Keydell] CCIVOTOKOIO 0Er|nccxov ouvonoc vuucpris, and XLVIII 428 TocvTocAis CUVOTOKEICC, Niobe in both places. But let me ask especial attention to the employment of OCIVOTOKEIOC in Mosch. iv 27 where Megara relates the death of her children at the hands of Heracles: cos 8' opvis 8upr)Tca em oxpETEpoicn VEOCTCTOIS | OAAUHEVOIS, OUOT' aivos 691s ETI vr|TridxovTas |

Oajivois ev TTUKivoTai KOCTEOOIT) * x\ 8E KOCT' OCUTOUS | TrcoTOTai KXajouaa \xaka Aiyu TroTVia liTirrip | . . . cos £Y& aivoTOKEia 91X0V yovov aid^ouaa | piaivopEVoiai Tr65Eaai 56|iov Korra TTOAAOV E901TCOV. The poet who wrote this was imitating first and foremost Iliad B 308 sqq., but he would naturally remember also Aesch. Sept. 278 [[291] and this passage of the Agamemnon: that he did remember this last, I find another indication besides OCIVOTOKEIOC. In v. 2 of the poem you have || EKTrdyAcos ccxsovaoc, in v. 72 EKrrdyXcos 6Ao9upo^ai, in v. 93 8£i|iaivco... EKirdyXcos: in v. 2 is a variant dx&CKn: if EiordyAois OCXEECKJI is to be read, it is neither more nor less than a reproduction of the phrase which Blomfield has in Ag. 49 restored to Aeschylus, EKrrdyAois dAyEcn. The MS reading EKTTCCTIOIS is undoubtedly a word which might have existed, though in face of the resemblance between yA and TI there is nothing approaching proof that exist it did; but in this passage it gives a totally inadequate sense, and EKirdyAois as well as ccivoTOKcov seems to gain support from Moschus.

97-103 TOUTCOV Af)£apa * Kaipos; &pai • mipoi; coprj • TCO Kaipcp; KOC8' copav • KOCTOC Kaipov. The adverbial use of rqv cbpr|V = iusto tempore occurs in Herod. 11. 2: Kaipov itself is used in that sense in Soph. Ai. 34, 1316 and Eur. HeL 479, but Kaipov is perhaps thus employed only with TJKCO or verbs of that meaning. Of course I cannot promise that cbpocv was the very word on which Kaipov is a gloss; but that Kaipov is a gloss, luckily detected by metre, I have no doubt. I render you should have exchanged blows earlier•, in season, when we did this deed. Strictly I suppose TTpiv belongs to TTOCOEIV, cbpocv to Ip£ocvTas. In the same meaning which I here give to

90

THE A G A M E M N O N OF AESCHYLUS

[289-90

mxOetv fp^ocvTocs (smite and be smitten) Euripides Phoen. 480 uses KCCKOV TI Spaaoci Kai TTCCSEIV. In v. 1659 almost all editors now accept Martin's SexoijjeO* dv. But manifestly this of itself is not enough to amend the line. To say s! uoxOcov yevonro dAis the moment after you have said Trnnovfis dAis urr&pxsi is so obviously inconsistent that there is a general consent against the genuineness of dAis. Donaldson proposes and Paley approves OCKOS, which makes good sense. But the verse is to be corrected with much less change than OCKOS, 5e)(oineO' dv. The reading which I propose is really almost identical with that of the MSS: || EI

6s TOI

JJOXOCOV

ysvoiTo TCOVB' ccAr|, crreyoiiJiEO' ocv,

T\ for 1, T for f, y for x- Should there be any way to ward off these ills (civil war), ward them off we should. We moderns know only dAr| error; but the Greeks knew also dXr| tritura akin to dXeco tero and OKT\ defensio akin to OCAECO defendo. The existence of the verbs might support this surmise, even were there no other proof; but it happens that both these lost substantives occur in the Agamemnon, though obscured in one place by the corruption of the copyists and in the other by the mistranslation of the commentators. In v. 204 [194] the winds that blew at Aulis are called ppc-Tcov dAoci, which is supposed to mean causes of wandering to men: a less happy name for winds which prevented the Greeks from sailing and kept their fleet on the shore it would need some ingenuity to devise. The true rendering is suggested by v. 207 [197] dvdos KCCTE£OCIVOV 'ApyEicov Tpi^cp (so I should arrange the words, making no change in the antistrophe but pEiOpois for p££0pois): ppoTciov dAcci are grindings or tribulations of men, winds that wear men away dnrAoia KEvayyEi. In v. 1659 dAr| is akin to OCAECO defendo, a verb preserved, I think, only in Hesych. OCAEE * cpuAccaCTE: I imagine that &AEO|JCXI uito is originally part of the same verb: compare too &AET|, &A£copf|, CVAEOCO, aA^co. The verse

means then el pi6x§cov yEvoiTO TCOVSE