Claudian: De Raptu Proserpinae (Cambridge Classical Texts and Commentaries - Volume 11)

CAMBRIDGE CLAS SICAL TEXT S AND C O M M E N T ARIE S EDIT ORS

C . 0 . BRINK

D. W. LUCAS

F. H. SAN DBACH

11 CLAUD IAN D E RAPTU P R O S ERPI N AE

CLAUD IAN DE RAPTU PROSERPINAE

EDITED WITH AN INTRODUCTION AND COMMENTARY BY

J.

B. HALL

CAMB RID GE AT THE UNIVE R S ITY PRE S S !969

Published by the Syndics of the Cambridge University Press Bentley House, 200 Euston Road, London N.W.I American Branch : 32 East 57th Street, New York, N.Y. I0022 © Cambridge University Press 1969

Library of Congress Catalogue Card Number: 69-1 4395 Standard Book Number: 521 07442 8

ISBN:

521074428

Powered by LATINSCAN

Printed in Great Britain at the University Printing House, Cambridge (Brooke Crutchley, University Printer)

C O NTE NT S page

PREFACE

1x

I N TR O D UCTI O N I Catalogue of manuscripts II The manuscript tradition III History of the transmission I V Editions V Date and circumstances of composition V I Sources and style VI I The apparatus criticus

II2

TEXT A N D CRITICAL APPARAT U S

II5

C O M MENTARY

1 85

BIBLIOGRAPHY

244

INDEX NOMINUM

GENERAL INDEX

v

I

3 33 64 76 93 !06

UXOR! CARISSIMAE

PREFACE

The present work began life as a dissertation approved for the Cambridge Ph.D. degree in 1965 . Since then it has been sub­ stantially revised and, I hope, improved, the Introduction in particular being to all intents and purposes a new composition. It is a great pleasure to have this opportunity of acknow­ ledging the help I have received from many quarters in the preparation of this book. The Faculty Board of Classics and the Managers of the Craven Fund in the University of Cambridge provided most welcome financial assistance in the early stages of my researches. A considerable number oflibrarians have put me in their debt by answering my queries, supplying me with films ofMSS and putting their facilities at my disposal ; for these kindnesses I am deeply grateful. Throughout the period during which first the dissertation and then the book was being pro­ duced, I have greatly benefited from the advice and criticism of many Cambridge scholars and friends, among whom I must single out for particular mention Prof. C. 0. Brink, who examined my dissertation and suggested many improvements, Prof. F. R. D. Goodyear, Mr E.]. Kenney, my former research supervisor, Mr A. G. Lee, the late Mr A. Ker, Mr D. W. Lucas, Dr A. H. McDonald, and Prof. F. H. Sandbach, who very kindly read the proofs and made a number of helpful observa­ tions. The Cambridge University Press has handled all stages of the printing and publication with unsurpassable skill and efficiency. Finally, I must pay heartfelt thanks to my wife, Dr J. A. Hall of Westfield College, London, who has been in on the D.R.P. from the beginning and who has criticised every­ thing I have written with wifely candour and fme scholarship ; to her this work, qualecumque est, is fittingly dedicated. ]. B. H.

London June 1969 IX

INTRODUCTION

I. C ATAL O GU E O F M A N U S CR I P T S 1

A I* Antwerp, Museum Plantin-Moretus 85 (Denuce 71), fourteenth century, parchment. Consists of 4 4 fos., of which 1 8 v-37 are devoted to the D.R.P. Each page has one column of thirty-three verses. After m.281 is the spurious verse omnis honos recti vobis sic Jata recedit. The text is fully annotated, and there are corrections by Poelman (Pulmannus ), who used this MS (L in his sigla) for his edition of 1571. Heinsius' ' alter Moreti ' ; V in Birt' s edition. A2* Antwerp, Museum Plantin-Moretus 17. 1 (Denuce 193 ) , fourteenth and sixteenth centuries, parchment and paper. The bulk of Claudian, including the D.R.P., is 1 Included in this catalogue are all existing MSS that I know of except two. The exceptions are Toledo Caj6n 102.17 (Birt, p. clii, note 1 ) and San Daniele 47, both apparently of the fifteenth century; I have no information about either of these MSS because my requests for microfilm have been ignored. Leningrad Caesareus A.O. Sect. CL. N.7 (the Petropolitanus mentioned by Birt on p. cl) was transferred to Warsaw after 1921 and destroyed during the Second World War, along with the bulk of the Polish National Library manuscript collection (communication of the Director of the Biblioteka Narodowa, Warsaw). Toledo Caj6n 102 . 1 8 , mentioned by Birt, p. clii, note I, does not contain the D.R.P. (communication of Sr. L6pez de Toro of the Biblioteca Nacional in Madrid), and Perusinus 1 5349 (my Y2) is wrongly so numbered by Jeep (ed. maior I, li) and Birt (p. cl). It has not proved possible to identify Jeep's 'codex Dresselii', or to locate Birt's Cheltenhamensis 9125 (referred to in the footnote on p. ii). I have myself collated all the MSS listed above, either on the spot or by means of microfilm (MSS known to me only through fJ.lm are denoted by an asterisk after the siglnm). The introduction of a new and systematic set of sigla seemed warranted by the fact that many MSS here mentioned were unknown to the last major editor, Birt, while most of those that he did know were given sigla quite at random, often rather clumsy ones; cross-references to his and also to Jeep's sigla are provided in the case of those MSS that they use in constituting their texts. The present sigla are based as far as possible on d1e current location ofMSS and, where necessary, on their date. Thus,B=ilie single Bern MS; F1 =the oldest MS in Florence; FI6=one of the youngest Florence MSS.

3

1•2

INTRODU CTION

a*

B*

bI

contained o n fos. I--96, which are parchment; fos. 97117V present the carmina minora in a sixteenth-century hand. The D.R.P. (fos. 84-96v) is written with forty­ four verses to the page and is much annotated. Passages omitted are: I.203-I4 (added in margin) ; rn.280-360 (after 279 is the spurious bridge-verse omnis honor recti nobis sic Jata recedit; on fo. 95 there is a note about this omission, very possibly by Poelman) ; 438-48 (added by a later hand) . Pulmannus' C, Heinsius' 'prior Moreti ', W in Birt' s edition. New York, Public Library 96 (gift of J. J. Astor), c. I450, parchment. The D.R.P. is on fos. I-34; there are eighteen verses to the page. 111.332-60 stand after rn.388 (c£ F 6 Y I and Y 2 ) and are followed by the verse omnis honor recti nobis sic Jata recessit. Bern, Burgerbibliothek 398, twelfth century, parch­ ment. The MS has thirty-six verses to the page and presents the D.R.P. on fos. 45-54 v. Due to the loss of a number of folia it lacks the preface to book 1; I.I- 1 15; 25 6-88; the preface to book rr; II.I-44; and I09-237· Other omissions, not confined to this MS alone, are: 1.139-2I4; rn.280-360 (between 279 and 36I is the verse omnis honor recti vobis sicJata recessit) ; and 438-48 (these verses were added later by Pierre Daniel, who once owned the MS). Bologna, Biblioteca Universitaria 222I, fourteenth cen­ tury, parchment. Written with sixty-four verses to the page, the MS consists of I62 fos., of which 86-I52 are devoted to Claudianus maior1 and I52V-I6o to the D.R.P. The following passages are omitted : I.I39-2I4 (I40 however is added in the margin) ; 111.280-360 (after

1 I regularly use this convenient traditional term to denote the ' corpus carminum publicorum cum minorum appendice' (c£ Birt, pp. lxxvii, cxlviii).

4

C A T AL O G U E O F M A N U S C R I P T S

b2

CI

279 is the verse omnis honos recti nobis sic Jata recessit) ; and 43 8-48. The preface to the Panegyric on the Sixth Consulship of Honorius, which a number of MSS have borrowed to introduce book m, is in this case added at the foot of fo. 1 57v, after m.6o. This is Heinsius' 'Bononiensis '. Bologna, Biblioteca Universitaria capsa 401 (entitled Miscellanea Poetarum ), fifteenth/sixteenth century, paper. The text of the D.R.P. stops short on fo. 14 at m.23 ( . . . vite). Cambridge, Corpus Christi College 228, thirteenth century, parchment. The MS consists of 130 fos. and contains both Claudianus maior and the D.R.P. (on fos. n6-28 v); there are thirty-six verses to the page. Passages omitted are: I.I 39-2r4; III.28o-360 (after 279 is inserted omnis honos recti vobis sicfata recedit). The text stops short at m.370 ( . . . cypressi), which concludes fo. 128, one or possibly two folia presumably having been lost. This is B irt s C. Cambridge, Peterhouse 2. r.8 (James 2 15), thirteenth century, parchment. The D.R.P. begins on fo. 72 v and ends on fo. 78 ; each page has two columns of forty-six verses. Passages omitted are : 1. 139 and 146-213 ; m.28o3 60 (after 279 is the curious conflation virginitatis honos nobis sic Jata recessit); and 43 8-48. Cambridge, Peterhouse 2.r.o (James 207 ) , thirteenth/ fourteenth century, parchment. There is no regular foliation ; the leaf on whose verso the D.R.P. ends bears the pencilled number 79. Omitted are II.258-3 17 and m.22r-8o, owing to the excision of two leaves. Written in England. Chiari, Biblioteca Morcelliana 4, finished ' pridie idibus '

C2

C3

c*

5

INTR ODUCTION

Marcij 1470' (note ad fin.). There are twenty-eight verses to the page. Between m.279 and 280 is the verse omnis honos recti vobis sic fama recedit. The MS was written by one Pecino Bigoni. D* Dresden, Sachsische Landesbibliothek (formerly Konig­ liche Bibliothek) De. 157, thirteenth century, parch­ ment. The D.R.P. is presented on fos. 25-50 ; there are twenty-four verses to the page. After m.279 is the verse omnis honos recti vobis sic Jama recedit. According to F. Schnorr von Carolsfeld and L. Schmidt, Katalog der Handschrifien der Kgl. Ojfentlichen Bibliothek zu Dresden, this MS was written in Italy and at one time belonged to Laurentius de Papia. It was badly damaged in the Second World War and is now largely illegible. Metz, Bibliotheque Municipale 647, fifteenth century, d* paper. The D.R.P. occupies fos. I09v-u8 ; there are 1 8-20 verses to the page. Passages omitted are : the preface to book 1; I.I4o-214. The text ends abruptly at ll.79· E 1* El Escorial O.m.25, thirteenth century, parchment. The MS consists of 1 1 8 fos. and contains Claudianus maior in addition to the D.R.P., which takes up fos. 94-108. There are thirty-six verses to the page. Verses omitted are : 1. 139 and 141-214; m.280-36o (between 279 and 3 61 is the verse omnis honor recti nobis sic Jata recedit); and 43 8-48. E 2* El Escorial S.m.29, thirteenth century, parchment. The MS consists of 87 fos., with the D.R.P. preceding Claudianus maior on fos. 1-14 v. There are thirty-nine verses to the page. Passages omitted are : 1.203-14 ; m.28o-360 (between 279 and 361 is the verse omnis honos recti nobis sic Jata recedit); and 43 8-48. 6

C A T A L O GU E O F M A N U S C R I P T S

e 1*

c2*

F1

F2

F3

Erlangen, Universitatsbibliothek 621, :fifteenth century, paper. D.R.P. on fos. 249-72, with twenty-six verses to the page. Between m.279 and 280 is the verse omnis honor vobis recti sic phama recessit. Erlangen, Universitatsbibliothek 626, a. 1469, paper. The D.R.P. is on fos. 1-23 ; there are twenty-seven verses to the page. 1.1 3 9 and 140 have been transposed, and 141 appears only in the lower margin after 148. Florence, Laurentianus plut. xxiv sinistr. 12, twelfth century, parchment. This MS, which formerly belonged to the Library ofS. Croce, consists of 91 fos. and contains Sedulius, Statius' Achilleid and the D.R.P. Written with 25, 26 or 3 2 verses to the page, the D.R.P. occupies fos. 70-91 v (fos. 87 and 88, which contain rrr.279-3 32 and 3 3 3-88 respectively, are misplaced between fo. 86, which ends with m.157, and fo. 89, which begins with III.I 5 8, and should properly follow fo. 90) . There are interlinear glosses and marginalia in profusion. This MS is Lin Jeep's editions, Fin Birt's edition. Florence, Laurentianus no. 250, twelfth/thirteenth century, parchment. The D.R.P. stands on fos. n 8-p v and is preceded by Claudianus maior. There are thirty­ six verses to the page. Passages omitted are : 1. 141-214; m.28o-360 (between 279 and 3 61 is the verse omnis honos recti nobis sic Jata recedit) ; and 43 8-48. The MS once belonged to Niccolo de' Niccoli and later formed part of the Library of S. Mark in Florence. Florence, Laurentianus plut. xxxiii.4, thirteenth century, parchment. The MS contains 157 fos., with Claudianus maior preceding the D.R.P., which is on fos. 141-57. There are thirty-one verses to the page. The following passages are omitted : 1. 139-214; rrr . 280-36o ; and 43 87

I N T RO D U C T I O N

F4

F5

F6

F7

48. Once in the possession of Pietro de' Medici, son of Cosimo, and later collated by N. Heinsius (there is a note to this effect on fo. I ) . It may be that this is Heinsius' ' Mediceus primus '. Florence, Ashburnham L 977, thirteenth/fourteenth century, parchment. The MS consists of I6 fos., with thirty-three verses to the page, and contains only the D.R.P. Passages omitted are: I.I41-214; m.28o-36o; and 441-8. Between rn.279 and 361 appears the verse omnis honos recti nobis sic fata recedit. Florence, Laurentianus plut. xxxiii.8, a. 1404, paper. This MS, written by one Bartolomaeus Iohannis, consists of I 7 fos. and has many marginal and inter­ linear notes, some of them in Tuscan dialect. Between rn.279 and 280 is inserted the verse omnis honos recti vobis sic Jama recedit. Florence, Laurentianus plut. xci sup. 34 (formerly of the Biblioteca Gaddiana) , a. I394, paper. Consists of I3 3 fos., with the D.R.P. on fos. 4I-63 , and has twenty­ six verses to the page. Fo. 61 (containing rn.349-60, the spurious bridge-verse, which appears again between 2 79 and 280, and 3 89-428) and fo . 62 (containing III.322-3 I , 3 6I-88, and 3 32-48 ) are i n the wrong order ; but even if this order were corrected, we should still have III.3326o and the spurious verse out of place behind 3 88, an arrangement which is found also in a Y I and Y 2. The subscription of this MS is worth quoting in full ; it runs : Explicit hoc totum; Ocro, mihi porrige potum. 1394, die Iovis, quarta Junii, Fulginei in maxima paupertate delapsus explevi, et Dominicus Ducii, vocatus Ocro, dedit potum cum cerasis et fecit bene. Florence, Biblioteca Nazionale Centrale n.ix.I32, fif8

C A T AL O G U E O F M A N U S CRI PTS

F8

F9

F 10

FII

teenth century, paper. This MS contains Eutropius, Sextus Rufus, Plinius Secundus and, on fos. 98-II9, the D.R.P. Florence, Laurentianus plut. xxxiii.3 , fifteenth century, parchment. The MS consists of 1 86 fos., with twenty­ six verses to the page, and contains Claudianus maior followed by the D.R.P. (on fos. 166v-1 86) . Passages omitted are : 1.139-214; nr.280-3 60; and 43 8-48. Like F 3 this MS once belonged to Pietro de' Medici. Florence, Biblioteca Riccardiana 718, fifteenth century, paper. The MS consists of 54 fos., with twenty-four verses to the page, and offers the carmina minora as well as the D.R.P., which is on fos. 31-54. Passages omitted are : 1.140/1 ; m.28o-3 6o (after 279 is the verse omnis honos recti nobis sicJata recedit) ; and 43 8-48. On fo. 1 are the entries D. Franciscus Maria utebatur and, in another hand, Deinde D. Iulianus Rillius eiusdem beneficia. Florence, Laurentianus plut. xxxvii. 14, fifteenth century, parchment. The MS consists of 224 fos., with thirty-five verses to the page, and contains Silius Italicus, Cal­ pumius' Bucolics, and, on fos. 207-24, the D.R.P. Between m.279 and 280 is the verse omnis honos recti nobis sic Jama recedit. Florence, Laurentianus plut. xci sup. 4 (formerly of the Biblioteca Gaddiana), fifteenth century, paper. The MS has II4 fos., with twenty-four verses to the page, and contains the Latin Iliad and works of Coluccius, Prudentius and Avienus besides the D.R.P., which is on fos. 89-II3 v. Interesting textual features are the fact that 1.139 and m.28o appear only in the margin, and the following order of verses in book III: 279, then omnis honos recti nobis sicfama recedit, then 3 62-3 (in brackets ) , 9

INTRODUCTION

and finally 281-3 62/3 etc. The M S once belonged to Paolo Morelli de Morellis. F 12 Florence, Laurentianus Leopoldinus cciii (formerly of the Bibliotheca Aedilium Florentinae Ecclesiae), fif­ teenth century, paper. The text of the D.R.P., which is lightly annotated and corrected, is written on fos. 8 1105, with twenty-five verses to the page. Between rn.279 and 280 is the verse omnis honos recti vobis si Jama recedit. At the end of the MS, which contains also the Appendix Vergiliana, Calpurnius, Statius' Achilleid and the Elegia Sapphus, is the entry Georgii Antonij Vespuccii liber. F 13 Florence, Laurentianus plut. xxxiii. 1, fifteenth century, parchment. There are 1 8 8 written folia, with twenty-six verses to each page, and Claudianus maior is presented as well as the D.R.P., which is on fos. 169-88 v. Omitted passages are : 1.141-214 (13 9 and 140 are in reverse order) ; rn.28o-3 6o ; and 43 8-48. Like F 3 , this MS was collated by N. Heinsius. F 14 Florence, Laurentianus plut. xxxiii.2, sixteenth century, paper. Consists of 292 written folia, with nineteen verses to the page, and contains Claudianus maior as well as the D.R.P., which is on fos. 223-54 (then follows the poem on the Phoenix attributed to Lactantius) . Passages omitted are : 1. 141-214 (added at the end of book rn); rn.28o-3 60 (added at the end of book rn). Between rn.279 and 361 is omnis honos recti vobis sic Jata recedit. Subsequently, however, this verse was deleted and a marginal note inserted referring forward to the supple­ ment at the end of book III. Collated by N. Heinsius and known to Burman secundus as the ' incertus Heinsii '. F15 Florence, Fondo Redi 100, fifteenth century, paper. The IO

C A T A L O GU E O F M A N U S C R I P T S

D.R.P. i s on fos. 39-60. In this M S m.279 i s thus re­ shaped : omnis honos recti nobis tantumne relictus. P16 Florence, Fondo Acquisti e Doni 3 58, a. 1477 (so the subscription on fo. 79), paper. Contains : (i) Lactantius; (ii) ]oh. Antonius Gallicanus poeta celeberrimus ; (iii) Inci­ piunt quaedam Phalaridis epistolae scriptae per Franciscum Aretinum; (iv) Valerius Probus; (v) Leonardi Aretini epistolae ; (vi) Claudiani de raptu Proserpinae (fos. 47-70) ; and (vii) Lucij ]unii Moderati Columellae . . . carmen. Between m.279 and 280 of the D.R.P. is the verse omnis honos recti nobis sic Jama recedit. f1* Modena, Biblioteca Estense a.M.9.22 (Lat. 939), fifteenth century, parchment. Contains the D.R.P. on fos. 29v-59v. There are twenty verses to the page. Between m.279 and 280 is the verse Omnis honos recti vobis sic Jama recedit. f2* Modena, Biblioteca Estense aJ.5.I9 (Lat. ro8o), fif­ teenth century, paper. On fos. n 8 v-12or offers only the prefaces to the three books of the D.R.P. f3* Modena, Biblioteca Estense y.R.6.25 (Campori App. 223 ), fifteenth century, paper. Offers on fos. 90r-92v a fragmentary text that breaks off at 1. 123 . There are twenty-three verses to the page. G 1* Berlin, MS lat. 4° 740 (now in the Staatsbibliothek at Marburg) , thirteenth century, parchment. The D.R.P. is on fos. 1-17v; for the most part there are thirty-five verses to the page. Passages omitted are : 1. 139-214 (added by man. rec. at the very begin­ ning of the MS) ; III.28o-3 6o ; and 43 8-48 (added by the same man. rec. that repaired the lacuna in book 1) . !1.256-83 are written twice over; the first version I call ' a ', the second ' b ' (for there are some differences II

INTRODUCTION

between the two). The M S is perhaps of N. Italian ongm. G2* Berlin, MS lat. go I I2 (now at Marburg), :fifteenth century, paper. The D.R.P. is on fos. 1-3 2 v. There are nineteen to twenty verses to the page. G 3 * Berlin, Hamilton 492 (still in Berlin), fifteenth century, parchment. The D.R.P. is on fos. 37v-59v. There are twenty-seven verses to the page. Between m.279 and 280 is the verse omnis honos recti vobis sic Jam a recedit. G4* Berlin, MS lat. fo. 39 (now at Tiibingen), sixteenth century, paper. The D.R.P. is on fos. I98-219v. There are twenty-eight verses to the page. This MS is unique in omitting altogether the prefaces to books 1, n and 111, but its very late date forbids our laying any weight on this peculiarity. H* Worcester, Chapter Library F. 147, fourteenth century, parchment. There are two columns each of forty verses to the page, and a headless text of the D.R.P. begins on fo. 5 at 1.25 (vos . . . ) . The poem ends on fo. I I v. 1.139 and I4I-2I4 alone are omitted. Holkham Hall, Library of the Earl of Leicester 3 3 2, h fifteenth century, paper. The MS consists of 28 fos. and contains only the D.R.P. There are twenty-two verses to the page. After 111.279 is the verse omnis honos recti nobis sic Jama recedit. J I* Lei den, Bibliotheek der Rijks-universiteit 3 8 5 (Bet. 4) , thirteenth century, parchment. The MS contains Claudianus maior and a mutilated version of the D.R.P. (on fos. 82-5 v) . Only m. 19-437 are presented, with the omission however of 280-3 60 (between 279 and 3 61 is the verse omnis honos recti nobis sic Jata recedit) . There are forty-six verses to the page. 12

C A T A L O GU E O F M A N U S C R I P T S

] 2*

]3*

]4*

j*

KI

Leiden, Bibliotheek der Rijks-universiteit 395 (Gronov. 87), thirteenth century, parchment. The MS consists of 55 fos., with thirty or so verses to the page, and a text of theD.R.P.withannotations appears on fos. 3 3-5 1 v. 1.139 is present only in the margin ; m.28o-3 6o are wrongly placed between m.250 and 251 which respectively end and begin folia (it is to be noted that fo. 49, which contains 280-3 60, was written by a second hand roughly contemporary with the first) ; and III . 43 8-48 have been added by a third hand, probably not much later than the other two, which also deleted the original explicit after 43 7· Leiden, Bibliotheek der Rijks-universiteit 294 (Voss. lat. 0. 39), a. 1218, parchment. The MS contains Claudianus maior besides the D.R.P. There are forty verses to the page. 1. 140 is omitted. This is Heinsius' ' Petavianus primus ' ; VI in Jeep's editions. Leiden, Bibliotheek der Rijks-universiteit B.P.L. 105 A (Geel 3 86), fourteenth century, parchment. Written by one Martinus de Vicomercato, the MS consists of 28 fos. and has twenty-four verses to the page. The passage m.28o-3 60 is omitted in its proper place (after 279 is the verse omnis honos recti nobis sic Jata recedit) but is added by the :first hand at the end of the poem. The MS was collated by Burman, who calls it ' Leidensis '. Princeton, NJ., University Library 47 (28oo.86o.3), :fifteenth century, paper. The D.R.P. is on fos. 25-49. There are twenty-five verses to the page. After III.28I is the verse omnis honos recti vobis sic Jata recedit. Milan, Ambrosianus H 73 sup., fourteenth century, parchment. This MS consists of 3 fos. (two of which contain musical notation) + 21, with twenty-nine verses 13

INTRODUCTION

K2

K3

K4

k*

Lr

to the page. Between rn.279 and 280 i s the verse omnis honos recti vobis sic Jama recedit. Bought in Venice by Antonius Olgiatus, first librarian of the Ambrosian. Milan, Ambrosianus Z 73 sup., fifteenth century, parch­ ment. Consists of 2 + 23 fos., with twenty-four verses to the page. Passages omitted are : I. I4I-2I4 ; rn.28o-3 6o ; and 43 8-48. Between rn.279 and 3 6I is the verse omnis honos recti nobis sic Jata recedit. Milan, Ambrosianus I 26 sup., a. I463 , paper. Consists of 64 fos., with twenty verses to the page, and contains Calpurnius' Bucolics in addition to the D.R.P. Like KI, bought in Venice by Olgiatus. Milan, Ambrosianus S 66 sup., fifteenth century, parchment. Consists of I 8o +I fos., with twenty-seven or twenty-eight verses to the page, and contains Claudianus maior in addition to the D.R.P., which is on fos. I57-79· III.28o appears only in the margin ; its place in the text is taken by omnis honos recti sic nobisfama recedit. This is Heinsius' ' Ambrosianus tertius '. Pesaro, Biblioteca Oliveriana I8, sixteenth century (ace. Mazzatinti), parchment. The D.R.P. is on fos. I-22V. There are twenty-five verses to the page. rn.332-6o stand after 3 8 8 (c£ a F 6 Y I Y 2), and are followed by the verse omnis honor recti nobis sic Jata recessit. The folia are out of order and must be read in the following sequence : I-r 6 ; r8 ; I]; 20 ; I9; 21 ; 22. 1.287-8, the preface to book 11, and 11. 1-46 are missing owing to the loss of two folia between fo. 6 and fo. 7. London, Brit. Mus. Egerton 2627, twelfth century, parchment. Consists of I66 fos., with twenty-eight verses to the page, and contains Claudianus maior as well as theD.R.P., which is on fos. 149 v-r66v. Passages omitted I4

C A T A L OGUE O F M AN U S C R I P T S

are : I . I 39-2 1 4; II . II4- 84; m.28o- 360 (between 279 and 36 1 is the verse omnis honor recti vobis sic fata recessit); and 43 8-48. This is Heinsius' ' Petavianus secundus '. L2

London, Brit. Mus. Additional 2 12 13, thirteenth cen­ tury (palimpsest on primary script of eighth century. In the opinion of E. A. Lowe, Cod. Lat. ant. ii no. r69, this primary script was written either in England or in a continental scriptorium with Anglo-Saxon connexions. The entry capitulum ecclesie Q:!.inque ecclesie on the paper fly-leaf in a fourteenth-century hand probably refers, he thinks, to Fiinfkirchen in Hungary), parchment.There are two columns each of fifty-one, fifty-three or fifty-four verses to the page. The D.R.P. follows Maximianus on fo. 2rv (col. r ) to fo. 2 7 (col. r ) . I.I4I-2I4 stand after 1 .241; n1 .43 8-48 are omitted altogether; and m .28o-360, like the verses in book I, are out of place, being preceded by 279 and 3 61 and followed by a second 361. For an interpretation of these phenomena see below, p. 58. There are certain indications that this MS was copied from a difficult exemplar that had no word-division.

L3

London, Brit. Mus. Additional 6o42, thirteenth century, parchment. Consists of I I4 fos., with thirty-nine or forty lines to the page, and contains Claudianus maior as well as the D.R.P. , which is on fos. r-14. Passages omitted are : 1.140 and 203-14; m.28o-360 (after 279 is the verse omnis honos recti nobis sic fata recedit) ; and 43 8-48. The MS once belonged to the Benedictine abbey of S. Augustine (originally of SS. Peter and Paul) at Canterbury (see M. R. James, The ancient libraries of Canterbury and Dover, Cambridge, 19 13, and N. R. Ker, Medieval libraries of Great Britain, London, 1941, p. 27) . This is Birt's D. IS

I N TR O D U C T I O N

L4

London, Brit. Mus. Harley 2753 , thirteenth century, parchment. Consists of 54 fos., with two columns each of forty-four verses to the page, and contains Claudianus maior followed by the D.R.P. (on fo. 49 col. 1 to fo. 54 v col. 2). Omitted passages are : 1. 141-214; 111.280-3 60 (in the margin at 279-361 is a truncated version of the bridge-verse : omnis honos recti nobis; the rest is lost owing to the cutting down of the margin) ; and 43 8-48.

L5

London, Brit. Mus. Royal 15 A vii, thirteenth century, parchment. The D.R.P. is on fos. 56v-76. There are thirty verses to the page. 1.139 stands between 140 and 141, and between 111 . 279 and 280 is the verse omnis honor recti nobis si Jata recedit. This is Heinsius' 'Junii secundus '.

L6

London, Brit. Mus. Additional 12021 (formerly of the Bibliotheca Butleriana), fourteenth century, parchment. Consists of 19 fos., with thirty-five verses to the page, and contains the D.R.P. alone. 1.139 stands between 140 and 141, and between 111.280 and 281 is the verse omnis honos recti sic vobis Jama recedit.

L7

London, Brit. Mus. Harley 5 198, fifteenth century, paper. Consists of 105 fos., with twenty-five verses to the page. The D.R.P. is on fos. 1 5-39, after Persius. On fo. 105 v is the entry Stephani de Bertolinis et amicorum eius.

L8

London, Brit. Mus. Additional 10091, fifteenth century, paper. Consists of seventy-nine written folia, with twenty-nine verses to the page, and offers the D.R.P. on fos. 41-61 v. Between 111.279 and 280 is the verse omnis honos recti vobis sic Jama recedit. 16

C A T AL O GU E O F M A N U S C R I P T S

L IO

London, Phillipps 8776 (now in the keeping of the Phillipps Trust, 16 Pall Mall, London),1 thirteenth/ fourteenth century, parchment. The text of the D.R.P. is accompanied by a full-scale commentary written by one Gaufridus Vitreacensis (this is currently being inves­ tigated by Miss A. K. Clarke of Cambridge). This MS was used by M. A. Delrio in his Notae (he refers to it as ' m.s. liber quem mihi utendum Victor Giselinus singulari vir doctrina concessit'), and by Theodor Poelman for his Plantin edition ( 1571 etc.), in which it is given the siglum P. Readings from this MS, and others, were subsequently entered by Poelman in a copy of the Isengrin edition which was utilised by Barth in his first edition and later examined by N. Heins ius.

M 1* Munich, Bayrische Staatsbibliothek 597, fourteenth century, parchment. The D.R.P. is on fos. 1 v-2o ; there are thirty-two verses to the page. 1. 139 stands between 140 and 141 . The MS is heavily annotated, and there is occasional use of the letters of the alphabet to indicate the construction. M2* Munich, Bayrische Staatsbibliothek 63 1 , fourteenth century, parchment. The D.R.P. is on fos. 17-34 v, with thirty-four verses to the page. Between 111.279 and 280 is the verse omnis honor recti vobis sicJata recessit. There is much annotation. Formerly belonged to Hartman Schedel. M3* Munich, Bayrische Staatsbibliothek 391, fourteenth century, parchment. The D.R.P., preceded by Statius' Achilleid, is on fos. 44-69 v, with twenty-three verses to 1 Since this note was written the MS has been purchased by the Bodleian Library, Oxford, where it is MS Lat. class. c. 12.

2

17

HCD

I N TR O D U C T I O N

the page. 1. 140 is omitted. Also heavily annotated. Formerly belonged to Hartman Schedel. M4* Munich, Bayrische Staatsbibliothek 15740, fifteenth century, paper. The D.R.P. is on fos. 1-26, with twenty­ four or twenty-five verses to the page. III.28o-36o are omitted from their proper place and added at the end of the poem. After m.279 is the verse omnis honos recti vobis sic Jama recedit. The text is interspersed with regular chapter­ headings, as for example (after 1.53 ) O!!_aliter lachesis una parcarum allocuta est plutonem iratum. The MS formerly belonged to the Court Library at Salzburg (see fo. r ) . m* Baltimore, Walters Art Gallery 43 7, fifteenth century, paper. The D.R.P. is on fos. 61-88. There are twenty­ two verses to the page. Between m.279 and 280 is the verse omnis honos recti vobis sic fama recedit. N* Nottingham, University Library Aln. 130/2 (formerly in the possession ofLord Middleton, of Birdsall House, Malton, Yorks), thirteenth century, parchment. The D.R.P., preceded by Statius' Achilleid, is on fos. 1 59-65 and ends abruptly at 11.52. There are twenty-nine verses to the page. n* Naples, Biblioteca Nazionale V. D. 53, fifteenth century, paper. The D.R.P. is on fos. 1 1 1-3 8 (in fact 1 39, for the number 1 32 is used twice). There are twenty-two verses to the page. Or Oxford, Bodleian Library ms. auct. F.5.6 (SC 2195 ) , thirteenth century, parchment. Written in England, this MS consists of 4 + 161 fos. The D.R.P. begins on fo. 5 8 v and ends on fo. 8 r v. There are twenty-three verses to the page. The verses m.28o-3 37 are omitted and 3 3 8-60 are inserted after m.230. After m.279 is the verse omnis honor recti nobis sicjata recedit. 18

C A T A L OGUE O F M AN U S C R I P T S

02

03

04

Oxford, Bodleian Library ms. auct. F.2.16 {SC 2077) , thirteenth century, parchment. Written in England, this MS consists of 104 fos., with forty verses to the page, and contains Claudianus maior as well as theD.R.P., which is on fos. 1-14. Passages omitted are : 1.139-214 (139 and 140, however, have been added in the margin by a second hand, but in reverse order) ; III .280-3 60 (the verse omnis honor recti nobis sicfata recessit has been jotted in the margin by the second hand) ; and 43 8-48. This is Heinsius' ' Oxon. primus '. A in Birt' s edition. Oxford, Bodleian Library ms. auct. F.2.16 (SC 2077), thirteenth century, parchment. Bound in with 02, this MS consists of 169 fos. (numbered 105-273 ) , with forty-four verses to the page, and contains the Alexan­ dreis of Walter of Chatillon, part of Claudian' s Latin Gigantomachia, and the D.R.P., which is on fos. 262-72 v. The text begins at 1.28, and the following passages are omitted: 1.139-214 (140, however, is added in the margin by a second hand) ; III . 280-36o (279 was subsequently deleted by the second hand, which added in the margin hec ait et lectura faces altum nemus intrat, a spurious verse with the same function as omnis honos etc.) ; and 43 8-48. Heinsius' ' Oxon. secundus '. Bin Birt's edition. Oxford, Bodleian Library ms. lat. class. e 47 (formerly in the possession of the late Dr Garrod of Merton College, by whose kind permission I was first able to examine it), thirteenth century, parchment. A frag­ mentary MS, with numbered folia running from 245 to 284. It contains Maximianus (beginning on fo. 245 at El. 1.!5 ), Statius' Achilleid, and the D.R.P. (on fos. 269 v284V=26V-40V in the Bodleian's new numbering). There are thirty-six verses to the page. 1. 1 3 9 stands between 140 and 141 ; m.224-95 are omitted; and III . 2-2 19

I N T R O DU C T I O N

440-8 have been added by a second hand (the leaf containing these verses was cut away and they were therefore supplied by some reader of the MS). Fo. 278 34 is largely cut away so that the ends of n.262-97 and the beginnings of 298-3 3 3 are lost. 05 Oxford, Bodleian Library Rawl. G. r r 6 (SC I484I), fifteenth century, parchment. Written in Italy, the MS consists of I + 68 fos., with nineteen verses to the page. The D.R.P. is on fos. 3 I-64v. At the beginning there is the entry Paulus Franciscus Nuptius inter suos habuit die 3 mensis Marti} 1640. 06 Oxford, Bodleian Library ms. auct. F.2.2 (SC 886o), fifteenth century, paper. Written in Italy, the MS consists of I+ 8I fos., and contains part of Ovid's Fasti and the D.R.P., which is on fos. 59-80. According to Bodl. 8864, fo. 2 v, it came into the possession of N. Heinsius through the agency of Langermannus. It is no. I05 in the Heinsius sale catalogue, tom. ii, 69, and is the ' chartaceus Heinsii ' in Burman's edition. P I* Paris, Bibliotheque Nationale MS lat. 8o8o (formerly Tellerianus), thirteenth century, parchment. It contains Claudianus maior in addition to the D.R.P., which is on fos. 85-96v. There are forty-four verses to the page. Passages omitted are : I .I4I 2I4 (I39 appears only in the margin) ; m.280-36r (subsequently added in the margin by a second hand) ; and 43 8-48 (43 8-40, how­ ever, were later added by a second hand) . This is Heinsius' ' Tholosanus '. P2* Paris, Bibliotheque Nationale MS lat. 8082, thirteenth century, parchment. Contains the D.R.P. (on fos. I-I4) and Claudianus maior. There are forty verses to the page. Passages omitted are : I. I39-2I4 (I40, however, is added =

-

20

C A T A L OGUE OF M A N U S C RI P T S

P 3*

P 4*

P 5*

P6*

in the margin) ; m.28o-3 6o (between 279 and 361 appears the verse omnis honor recti vobis sic Jata rcccdit); and 43 8-48. This MS once belonged to Petrarch (c£ Birt, p. ii) and is Heinsius' ' Regius '. Paris, Bibliotheque Nationale MS lat. 8295 (formerly Colbertinus 4054 and Regius 6165 ) , thirteenth century, parchment. Contains Claudianus maior and, on fos. 1193 5V, the D.R.P. There are thirty-three verses to the page. Passages omitted are : 1. 139 and 203-14 ; m.28o360 (after 279 is the verse omnis hos [sic] recti nobis sic Jata recedit); and 43 8-48. Paris, Bibliotheque Nationale MS lat. 8296, thirteenth century, parchment. Contains Claudianus maior and the D.R.P., which is on fos. 1 10-23 v. There are thirty­ eight verses to the page. Passages omitted are : 1.20314 ( 140 appears only in the margin) ; m.280-3 60 (between 279 and 361 is the verse hec ait et lectura faces intrat nemus altum, with which compare the wording in 0 3 . The other bridge-verse omnis honos recti vobis sicJata recessit appears after III . 3 8 5 ) ; and 43 8-48. This is Heinsius' ' Puteaneus '. Paris, Bibliotheque Nationale MS lat. 15005, thirteenth/ fourteenth century. The D.R.P., heavily annotated, is on fos. 303-19v. This MS once belonged to the abbey of S. Germain-des-Pres. Birt's S. Paris, Bibliotheque Nationale MS lat. 7892, fifteenth century, parchment. The D.R.P. stands on fos. 12341 v. There are thirty-two verses to the page. Due to the disarrangement of some leaves, n.157-284 now figure between 1.115 and 116. After m.279 is the verse omnis honos recti vobis sicJama recedit. On fo. 1 is the entry Claudii Puteanj. 21

INTRO DU CTION

P 7*

Paris, Bibliotheque Nationale MS lat. 8081, fifteenth/ sixteenth century, paper. The D.R.P. is on fos. 103-2ov. There are thirty-four verses to the page. After m.279 is the verse omnis honos recti vobis sic phama recedit. P 8* Paris, Bibliotheque Nationale MS lat. 8297, a. 1 5 1 3 , paper. The D.R.P., heavily annotated, is on fos. 1-29. There are twenty-two to twenty-six verses to the page. P 9* Paris, Bibliotheque Nationale MS lat. I IJ24, fifteenth century, parchment. The D.R.P. is on fos. I-39· There are sixteen verses to the page. After m.279 is the verse omnis honos recti vobis sic Jama recedit. p 1* Padua, Biblioteca Universitaria 1561, 27 January 1460, paper. The D.R.P. is on fos. 1-26. There are twenty­ four verses to the page. After m.279 is the verse omnis honos recti vobis sic Jama recedit. p 2* Padua, Biblioteca Capitolare C.62, fifteenth century. There are thirty-one verses to the page. 1.1 40 appears only in the margin. After m.279 is the verse omnis honor recti vobis sic Jama recedit. Q* Pavia, Biblioteca Universitaria MS Aldini 412 (old catalogue no. C :xxx B 8), 8 April 1467, paper. The D.R.P. is on fos. 1-28. There are twenty-three verses to the page. R I Vatican City, Barberinianus lat. 41, thirteenth century, parchment. The MS contains the Disticha Catonis, Theodulus, Avianus, Gallus (i.e. Maximianus), the D.R.P. (on fos. 32-sov), and Statius' Achilleid. There are thirty-two verses to the page. 1.139 is omitted. R 2 Vatican City, Reginensis lat. 1440, thirteenth century, parchment. The D.R.P. is on fos. 49-64 v, and there are thirty-four verses to the page. Passages omitted are: 1.140 and 203-14 ; m.28o-3 6I (between 279 and 3 62 is 22

CATALO GUE O F MANUS CRIPTS

H. 3

R4

Rs

R6

inserted the verse omnis honos recti sic Jata recessit, thus curtailed) ; and 43 8-48. This is Heinsius' ' Petavianus quartus ' . Vatican City, Reginensis lat. 1540, thirteenth/fourteenth century, parchment. Consists of 90 fos., with forty-one, forty-two or forty-three verses to the page, and contains all Claudian's works. The D.R.P. brings up the rear on fos. 82-90v. Passages omitted are: 1.139-214; II.J27III. I23 ; m.28o-360 (between 279 and 3 61 is the verse omnis honos recti vobis sic fata recessit) ; and 43 8-48. This is Heinsius' ' Petavianus tertius '. Vatican City, Vaticanus lat. 2807, thirteenth/fourteenth century, parchment. Contains Claudianus maior and the D.R.P. (on fos. 68 v-79v) . There are forty-six verses to the page. Passages omitted are : 1.139 and 141-214; m.280-360 (between 279and 3 61 is the verse omnis honos recti vobis sic Jata recedit) ; and 43 8-48. There are a number of annotations. Either this MS or R6 is Heinsius' ' Vaticanus quartus '. Vatican City, Vaticanus lat. 1663, thirteenth/fourteenth century, parchment. Heavily annotated, this MS con­ sists of 77 fos., with twenty-nine verses to the page, and contains the Disticha Catonis, Theodori fiber (i.e. Theo­ dulus), Aviani liber, Liber Laudis et Gloriae Christi, Statii Achilleis, and (on fos. 57-77v ) Claudiani de Raptu Proserpinae. In this MS 1.139 stands between 140and 141; m.279 is omitted ; and a leaf containing m.398-448 is misplaced between leaves ending and beginning with m.173 and 174 respectively. Vatican City, Vaticanus lat. 3290, fourteenth century, parchment. Contains the D.R.P. and carmina minora. There are thirty-four verses to the page. The folios 23

INTRODU CTION

which have the D.R.P. are out of order : the poem begins on fo. r6 at 1.55 (the preface to book 1 and I.I-54 are lost) ; the second book occupies fos. r 8-22 v and I-2 v, and the third fos. 2 v-7v. Passages omitted are : 1. 1 3 9-214; m.28o-3 6o; and 43 8-48. According to Ezra de Clerq van ]ever (apud Burman, p. 936), this is Heinsius' 'Vaticanus quartus ' ; but R4 seems occasion­ ally to be cited as ' quartus ' as well as R 6. The term ' Vat. d.' is used as an alternative to ' Vat. quart.' R7

R8

R9

Vatican City, Reginensis lat. I 556 (formerly Egmunda­ nus, later in the possession of G.]. Voss) , thirteenth/ fourteenth century, parchment. Much annotated, the MS consists of 75 fos., with thirty verses to the page, and contains the Disticha Catonis, Maximianus, the D.R.P. (on fos. 3 7-ss v) , and Statius' Achilleid. I.l39 stands between 140 and I4I ; the spurious bridge-verse appears only in the margin (omnis hotzos recti vobis sic Jata recessit), as does 111.279, which has been added by a second hand ; and 11. I89-248 are omitted due to the loss of one lea£ This is Heinsius' ' Vossianus secundus '. Rome, Biblioteca Corsiniana (Accademia dei Lincei) col. 43 · F. 5, fourteenth/fifteenth century, parchment. Consists of I I I fos., with thirty-five verses to the page, and contains the D.R.P. on fos. 52 v-69v. Between m.279 and 280 is the verse omnis honos recti vobis sicJam a recedit. Vatican City, Vaticanus lat. I66I, fourteenth/fifteenth century, parchment. Consists of 92 fos., with two columns each of thirty-six verses to the page, and contains Sidonius preceded by the D.R.P. (on fos. I -7v) Passages omitted are : I.I3 9-214; m.28o-36o; and 43 8 48. Also lost are 11. 9I-I47· This i s Heinsius' ' Vaticanus tertius '. 24 .

C A T A L OGUE O F M A N U S C R I P T S

R IO

R II

Vatican City, Ottobonianus lat. 2859, fourteenth/ fifteenth century, parchment. Consists of 53 fos., with thirty-three verses to the page, and contains the D.R.P. (on fos. 1-19) and the Ep istles ofHorace. In this MS 1.140 appears only in the margin ; 1.192-202 stand after 1.214; and III.280-3 6o and 43 8-48 are omitted (between 279 and 3 61 is the verse omnis honos recti nobis sic Jata recedit). Vatican City, Ottobonianus lat. 2126, fifteenth century (after 1479 ) , paper. In addition to the D.R.P., which is on fos. 71-87, the MS contains a commentary on Tibullus by Bernardin us V eronensis, Augustini Senensis Elegantiolae, and Cicero's Paradoxa. The order of verses in the D.R.P. is much disturbed : the preface to book 1 comes at the end ofbook III; II.94-III.143 stand between III.3 10 and 3 I I ; and III.144-255 are stationed between 111.3 66 and 367. After III.279 is the verse omnis honos recti vobis sic Jama recedit. Once belonged to Johannes Angelus Dux ab Altaemps.

R 12 Rome, Biblioteca Angelica 2266, fifteenth century, parchment. Consists of 42 fos., with twenty-four verses to the page, and contains Persius and, on fos. 1 7-42, the D.R.P. The verse 1.140 is omitted, and after III.279 is the line omnis honos recti vobis sic Jama recedit. On fo. 42 v is the entry ' mi costa 8 scudi, soldi 10 di moneta di Genova', attributed by Count Manzoni to Bernardino Baldi, abbott of Guastallae and a native of Urbino. R 13

Vatican City, Vaticanus lat. 2808, fifteenth century, paper. Consists of 164 fos., with twenty-eight verses to the page, and contains Claudianus maior in addition to the D.R.P. (on fos. 1 3 3-54 ) . III.332-6o are misplaced after III . 398 ; this fact is noted in marginalia at III.3 3 1 and 398. This is Heinsius' ' Vaticanus quintus '. 25

IN T R O D U C TI O N

R 14 Vatican City, Vaticanus lat. 166o, fifteenth century, parchment. There are 132 fos., and each page contains thirty verses. Claudianus maior precedes the D.R.P., which is on fos. I I3-3 I· Lacking are 1.154-272, which would have filled two whole leaves between fos. I I 5 and I I6. After 111.279 is the verse omnis honos recti vobis sic fama recedit. R 15 Vatican City, Urbinas lat. 657, fifteenth century, parch­ ment. Contains Claudianus maior as well as the D.R.P., which is on fos. 142v-1 6I. Passages omitted are : 1.141214 (13 9 and 140 appear in the text, but in reverse order) ; 111.280-3 60 (between 279 and 3 61 is the verse omnis honos recti nobis sic Jata recedit); and 43 8-48 (after 43 7 a second hand has added 1. 141-213 ) . R 16 Vatican City, Borgianus lat. 343 (formerly of the Collegium de Propaganda Fide), fifteenth century, parchment. The MS consists of 29 fos., with twenty-two verses to the page. 111.2 85-3 60 appear after 398 (compare the misplacement in R 1 3 ) . R17 Rome, Biblioteca Corsiniana 43 · G. 25, a. 1444 (so the note on fo. 23 v : Ego Baldesar Melius explevi hun [sic] Claudianum MCCCC44 in quinto decimo die novembris ora una noctis) , paper. There are twenty-four verses to the page. R 18 Rome, Biblioteca Corsiniana 43 .D.3 6, fifteenth century, paper. Consists of 194 fos., with fourteen verses to the page, and contains the D.R.P. on fos. 1-44. After 111 . 279 is the verse omnis honos recti nobis sic Jama recedit. R19 Vatican City, Reginensis lat. 1369, fifteenth century, parchment. Consists of 42 fos., with twenty-one verses to the page. After m.279 is omnis honos recti vobis sic 26

C A T A L O G U E OF M A N U S C R I P T S

lbo

R2I

R 22

R23

R24

R25

Jama recedit. At the foot o f fo. I is the entry ' Nicolai Heinsii '. Known to Burman as the ' membranaceus Heinsii '. Vatican City, Reginensis lat. I428, fifteenth century, parchment. Consists of 268 fos., with twenty-seven verses to the page, and contains the D.R.P. on fos. I25V­ I48V. I. I39 is omitted, and between rn.28o and 28I is the verse omnis honos recti nobis sic Jata recedit. This is Heinsius' ' Vossianus primus '. Rome, Biblioteca Angelica I345, fifteenth century, parchment. Contains the D.R.P. (on fos. I-2I v) , Statius' Achilleid, and the Satires and Epistles of Horace. After rn.279 is the verse omnis honos vobis si Jama recedit, thus truncated. Vatican City, Barberinianus lat. 68, fifteenth century, parchment. Consists of 57 fos., with twenty-six verses to the page, and contains Mela and the D.R.P. (on fos. 3 5-57) . III. 3 89-44I are omitted. Vatican City, Palatinus lat. I714, fifteenth century, parchment. The D.R.P. is on fos. I 59V-I78. Passages omitted are : 1. I39-2I4; III.280-36o; and 43 8-48. This is Heinsius' ' Palatinus '. Vatican City, Vaticanus lat. 2795, sixteenth century, paper. Consists of 90 fos., with twenty-four verses to the page, and contains Statius' Achilleid, the D.R.P. (on fos. 27-52 ) , and Persius. Fo. 49, which presents III.3 49-96, should follow fo. 50, which presents III . 30I48. After rn.279 is the verse omnis honos recti vobis sic fama recedit. Vatican City, Palatinus lat. I 573 , fourteenth century, parchment. Two MSS bound in one. The first contains Vegetius (fos. I-52 ) , and the second the Disticha Catonis,

27

INTROD U CTION

Avianus, and the D.R.P. (on fos. 99-I I4 v). This second MS, which comprises fos. 53-I I4, once belonged to Gruter (see fa. 55). Claudian, preceded by Statius' Achilleid, is written in single columns of thirty-seven or thirty-eight verses to the page. 1.139 stands between 140 and 141 ; m.28o-3 60 are omitted (between 279 and 3 61 is the verse omnis honos recti sic nobisfata recessit) ; and 448 precedes 447· Used by Barth for his edition, and collated by Heinsius in Rome. R 26 Vatican City, Vaticanus lat. 2809, of various dates, parchment. Contains all the works of Claudian written by three distinct hands : (i) fos. 1-39 (eleventh century according to Jeep, eleventh/twelfth century according to Birt) ; (ii) fos. 4o-71 (fourteenth century in Birt' s view) ; (iii) fos. 72-8ov (fifteenth century), which present the D.R.P. in twin columns each of thirty-five verses to the page. After m.279 is the verse omnis honos recti vobis sic fama recedit. Each episode in the poem is introduced by a prose summary set in the body of the text : before 1.20, for example, stands the heading invo­ catio deorum infernalium, and 1.214 is preceded by the words A iove mittitur in siciliam venus ad rem agendam. The vv.ll. inscribed by Livineius in a copy of the Aldine edition now in Leiden University Library are taken from this MS, which is Heinsius' ' Vaticanus primus '. R27 Vatican City, Vaticanus lat. 7182 and Vaticanus lat. 2864, fifteenth century, paper. The former contains, inter alia, the D.R.P. up to m.304 (fos. 41 8-3 5) ; the latter the rest of the poem up to 448 (fos. 4 5-8). Com­ parison of the scripts and water-marks of the two MSS proves beyond doubt that their Claudian sections were originally united. 2864 was examined by Heinsius and 28

C ATALO GUE O F MANUS CRIP T S

given the siglum ' A ' (c£ Burman, p . 965) . Bound in with 7182 is a copy ofSchurener's edition ofthe D.R.P., printed at Rome in 1475. R28 Vatican City, Vaticanus lat. r r420, fifteenth century, paper. Consists of 8 r fos., with twenty verses to the page, and contains the D.R.P. (on fos. r-26) and poems of Ovid, Horace and Persius. One leaf containing rqo79 is lost after fo. ro; two leaves containing rr.280-3 59 are lost after fo. 15 ; and two containing m. r62-241 after fo. 20. The MS once belonged to Muretus. R29 Rome, Biblioteca Angelica r46r, fourteenth and fifteenth centuries, parchment. Contains : (i) fos. 4-5 Prosperi Aquitanici epigramma ; (ii) fos. 6-3 3 (also numbered r-28) the D.R.P. (both written in the fourteenth century) ; and (iii) the works of Adamus Montaltus (fifteenth century). According to the Narducci cata­ logue (Rome, r 893 ), this MS once belonged to Johannes da Sienna. R3o Vatican City, Chigianus lat. H.vn.23 6, fifteenth century, parchment. On fo. 64, in the midst of Claudianus maior, offers a text of just the preface to book n. r* Ravenna, Biblioteca Classense 120, fifteenth century, paper. The D.R.P. is on fos. 50-65, and there are thirty­ nine or forty verses to the page. After m.279 is the verse omnis honos recti vobis sic fama recedit. T* Venice, Biblioteca Nazionale Marciana class. XII cod. ro (3973 ), thirteenth/fourteenth century, parchment. There are thirty-six verses to the page. One leaf containing III .402-48 is misplaced between leaves ending and beginning with n.223 and 224 respectively. The leaf which contains n.296-367 has largely been torn away. This is Jeep's V. 29

INTRODU CTION

T2* Venice, Biblioteca Nazionale Marciana class. XII cod. I I (4162) , fifteenth century, paper. The D.R.P. is on fos. I -40. There are eighteen verses to the page. T 3* Venice, Biblioteca Nazionale Marciana class. XIV cod. I83 (4628), fifteenth/sixteenth century, paper. The D.R.P. is on fos. I0-3 5 v. There are twenty-three verses to the page. Notable is the order : m.279, 278, omnis honos recti vobis sic phama recedit, 280. T 4* Venice, Biblioteca Nazionale Marciana class. XIV cod. 202 (4294), fifteenth/sixteenth century, paper. The D.R.P. begins on fo. 27. The MS is fragmentary, ending with 11.210 in the middle of fo. 3 5. There are thirty-four verses to the page. Poznan, Miejska Biblioteka Publiczna im. Edwarda t* Raczynskiego MS I79, a. 1 507 (c£ K. Aland, Die Handschriftenbestiinde der polnischen Bibliotheken, p. s r). The D.R.P. is on pages I-52. There are twenty-one or twenty-two verses to the page. Omitted are III . 3 3 2-6o (c£ a F 6 Y r and Y 2) . Between m.279 and 280 is the verse omnis honos recti vobis sic Jama recedit. U* Copenhagen, Kongelige Bibliotek Gl. Kgl. S. I905, twelfth/thirteenth century, parchment. The D.R.P. is on fos. 124V-I3 8, with forty-three verses to the page. The MS is heavily annotated. I.I39 stands between I40 and I4I ; m.293-3 60 are omitted in their proper place and added by the first hand after 448. In the margin at m.279 is the verse omnis honor recti nobis sic Jata recessit, and in the text between m.292 and 36I is a slightly different version of the same verse: omnis honos nobis et mox sic Jata recedit. Urbana, Illinois University Library MCA 4 (formerly u* Cheltenham 3 390), 2 August I469, paper. The MS was 30

CATALO GUE O F MANUS CRIPT S

written by Paulus Rosinus Abbaciensis. The D.R.P. is on fos. 1-3 0v. There are twenty verses to the page. After III.279 is the verse omnis honos recti vobis sic Jama recedit. The D.R.P. is followed by the Greek epigram which was inscribed on the base of the statue of Claudian set up in Trajan's forum (c£ C.I.L. VI. I7IO, and Birt, p. xliii). This epigram is translated by Rosinus both into prose : in uno virgilii mentem & musam homeri I Claudianum Roma & Reges posuerunt, and into verse: Maeonide cui mens Claudi cui musa Maronis I hanc tibi cum Roma Caesar uterque dedit; and this verse translation is printed in the vita Claudiani ofParrhasius' edition, along with another version by Parrhasius himsel£ Vienna, 6sterreichische Nationalbibliothek cod. ser. nov. 9 3 63 , twelfth century, parchment. Consists of three leaves (formerly glued in the covers of Vindob. 1245 Salisburg. 84 and Vindob. 1258 = Salisburg. r63 B), which give a text of m.280-448. Fo. r contains 280-3 I I and 3 12-46 ; fo. 2 , 347-78 and 379-409 ; and fo. 3 , 410-44 and 445-8. V 2* Vienna, 6sterreichische Nationalbibliothek 3087, fif­ teenth century, paper. The D.R.P. is on fos. r-22. There are twenty-seven verses to the page. The leaves have been dislocated and must be read in the following order: r-5, I I-r6, 6-ro, 17-22. The verses 11.134-89 are missing, due no doubt to the loss of one lea£ After III. 279 is omnis honor recti vobis sic Jama recedit. V 3 * Vienna, 6sterreichische Nationalbibliothek 3198, I I Au­ gust 1473 , paper. Written in Florence by Bernardus de S. Pauli de Mathelica. The D.R.P. is on fos. I I4-3 7· There are twenty-six verses to the page. After m.28o is the verse omnis honor recti vobis sicJama recedit. 3! V*

=

INTROD UCTION

W*

w*

Y1

Y2

y*

Wolfenbiittel, Herzog August Bibliothek, Gudianus lat. 228 (453 3 ), thirteenth/fourteenth century, parch­ ment. Consists of 91 fos., with thirty-six verses to the page, and contains the D.R.P. on fos. 75-91. 1.140 stands after 141 , and between 111 . 280 and 28 1 is the verse omnis honor recti vobis sic Jata recessit. Consulted and occasionally cited in his apparatus criticus by Gesner. Jeep's G. Lincoln, Cathedral Library 1 3 2 (C 5.8 ) , thirteenth/ fourteenth century, parchment. The D.R.P. begins on fo. 1 5 1 at 1.93 and ends on fo. 162v. The MS is now in a very poor state and frequently quite illegible. Perugia, Biblioteca Comunale 247 (D.65 ) , fourteenth to sixteenth centuries, paper. Consists of 1 89 fos. and contains the D.R.P., written in 1477 (so fo. 42 ) , on fos. 45-54 and 44-2 (these last three leaves have been inserted upside down and must be read in reverse order !). There are forty-three to forty-five verses to the page. 111 . 332-60 appear after 3 88, and between 3 60 and 3 89 is the verse omnis honor recti nobis sicJata recessit. Perugia, Biblioteca Comunale 63 3 (!.27) , fifteenth century, parchment. Consists of 74 fos., with twenty­ one verses to the page, and contains the D.R.P. on fos. 1-29v. 111.33 2-60 appear after 3 88, and between 360 and 3 89 is inserted the verse omnis honor recti nobis sic Jata recessit (c£ Y1) . This MS is wrongly numbered 1 5349 by Jeep and Birt. Yale, University Library unnumbered, fifteenth century, parchment. The D.R.P. is on fos. 1-2ov. There are twenty-seven verses to the page. The order of verses is very confused : beginning to n.po; 111. 45-170 ; n.p1111 .44; 111. 294 to the end. 111 . 1 71-293 are now lost. At .

32

C A T ALOGUE O F M AN U S C R I P T S

the foot of fo. I is the coat of arms of Fridericus de Schennis. Parma, Biblioteca Palatina 2504, twelfth and fif­ teenth centuries, parchment. Consists of I I 8 fos., of which I-95 are of the twelfth century and contain the D.R.P. (fos. I-I I ) and Claudianus maior, while 96I I 5 are of the fifteenth century and present the carmina minora. In the older part there are thirty-nine verses to the page. The following passages of the D.R.P. are omitted : 1.203-I4 (added in the margin, however, probably by a second hand) ; n1.45-278 (due to the loss of some leaves) ; 280-3 60; and 43 8-48 (these were, however, added later by a second hand). Between m.279 and 36I appears the verse omnis honos recti vobis sic Jata recedit.

Z*

I I. T H E M A NU S C RIP T T R A D ITI O N

Although Heinsius brought to light many MSS of the D.R.P., it was not until the I 87os that Ritschl' s pupil, L. Jeep, undertook the first systematic investigation of the MS tradition. I This was followed in the I89os by the analysis of Th. Birt,2 in which there is scarcely a mention, let alone an assessment, of Jeep's pioneer work. Since neither scholar's theses have ever been examined in detail, it seemed right, and indeed necessary, that the statement of my own views that concludes this chapter ' In Acta Societatis Philologae Lipsiensis I.ii (I87r), 347 ff. (hereafter abbre­ viated as Acta) ; the preface to the I874 ed. of the D.R.P. alone; and the preface to vol. I of the ed. of the whole of Claudian (I 876), xlix ff. Substantially the same views are expressed on all three occasions. Reviews by G. Vitelli in Riv. difil. I (1 873), 3 3o-4; E. Baehrens in ]enaer Literaturzeitung, nr. 8 (I87 5 ), 131-2; L. Mueller, ibid. nr. 18 (I877), 284-5 ; and H. Kostlin in Philologischer Anzeiger VIII (I878), 45-7. z In his edition of Claudian (Mon. Germ. Hist. tom. x) (I892), cxlvii­ clviii. Review by J. P. Postgate in C.R. IX (I895), 162-9. 3

33

HCD

INTRODU CTI ON

(Section 3 ) should be preceded by some discussion of my pre­ decessors' opinions, especially as an opportunity is thereby provided of considering, inter alia, Jeep's mistaken belief in the applicability of stemmatics and the existence of a ' best MS ' (Section I ), and Birt' s implausible contention that the poem passed through ancient recensions (Section 2 ) . I On the basis of their coincidence or non-coincidence in certain omissions, Jeep divides the MSS examined by him (they number sixty-one in Acta) into five classes : Class I, which comprises all those MSS that give the full text and also exhibit before book m the preface to the panegyric on the sixth consulship ofHonorius. This class is subdivided according to whether Jusis or pens is is read at I. 5 3 . Class 1 a, presentingfusis, consists of two MSS (in my sigla, F I and J 3 ), and Class I b, offering pensis, of thirty-seven MSS. Class rr, which is the same as r b, except that it omits m.280-360. There are three MSS belonging to this class (my ]4, M4 and R 2 5 ) , but two of them (J 4 and M 4) have the omitted passage added at the end of the poem. Class m, which is the same as II, but further omits the preface to book m and m.43 8-48. This class has just one member, my R 2. 1 Class rv, which i s a s m , but further omits I. I39-2 I4, while retaining the preface to book m. This class has six members. Class v, which is as IV, but omits the preface to book m. There are nine MSS in this class. The MSS F9 F I4 and J 2 (my sigla) exhibit features characteristic of two of the above classes and therefore cannot be classified.

The following MSS are selected as representatives of their respective classes : ofr a, Laurentianus xxiv sinistr. I2 (my F I) ; 1 Jeep quite ignores the fact that this MS omits 1.203-14. Bonnet (Revue critique, no. 27 (1 875 ), s f£) informed him of the presence of this same omission in two Paris MSS, but he clearly did not realise its significance. Cf. below, p. 58 and n. 2.

34

T H E M A N U S C RI P T TRADITI O N

of I b , Venetus Marcianus xrr x (my T) and Gudianus 228 (my W) ; of n, Palatinus 1573 (my R25) ; of m, Reginensis 1440 (my R2) ; of IV, Laurentianus 250 (my F 2) ; and of v, Mediceus xxxiii.4 (my F 3 ) . To these is added Vossianus 294 (my ] 3 ) which, according to Jeep, represents a stage intermediate between I a and I b. The interrelationships of the five classes are depicted by the following stemma : [•1 --

Ia

-----­

[z) ------

Ib

[x)� I

U

I

rn

--· ----

IV

__,

v

' Hinc sequitur ', concludes Jeep (ed. maior, p. liv) , ut prima classis ad Claudiani raptum Proserpinae emendandum sola sit maioris momenti ; nam ceterae . . . quod ad cognationem cum archetypo attinet, locum tenent secundarium, cum prior primae classis particula [ 1 a] aliis codd. non interpolatis ipsum archetypum repraesentet. Ergo praesertim cum classes rr-v lacunosae sint, severa artis criticae disciplina poscit, ut eas plane praetereamus ; nam scripturae, quas primae classis libris meliores hoc vel illo loco praebcnt, pro librariorum coniecturis habendae sunt, quibus firmissimis nostrae artis fundamentis nisus facile careas. =

The position of F 1 as oldest representative of Class I a calls forth this statement of editorial principle : ' imprimis id agendum erat, ut ubicumque sententia loci concedebat, spretis paulo elegantiorum scripturarum, quas recentiores libri saepis­ sime efferunt, illecebris summa severitate eas lectiones tene­ remus, quas codex Laurentianus praebet. Alius enim liber, qui propius ab archetypo absit, nunc non exstat. Q.!!amquam negari non potest etiam Laurentianum recentioris interpola­ tionis spurcitiis nimis iam esse contactum' (ed. minor, p. xviii). 35

INTR O D U CTION

Jeep's classification and the practical conclusions drawn from it are, in my submission, open to serious objection. (i) The criterion by which Classes I a and I b are distinguished is not valid : in a tradition as thoroughly penetrated by contami­ nation as this one, 1 the incidence of one variant reading is no sure basis for classifying MSS. Faced with the choice between pensis and fusis in an annotated MS of Class I a, how many scribes must by the simple act of choosing the former variant have produced a copy to be assigned to Class I b !2 Moreover, the inference one draws from the classification is that in [z] and all later stages of the tradition, the reading pensis has super­ seded fusis Not so. C r, G r, L4, N and 0 3 (my sigla), none of which was known to Jeep, would have had to be assigned to classes subsequent to [z] , and all offerfusis (ii) But [z] itself is an unnecessary postulate. Bonnet ob­ served that there are occasions where Laurentianus and Gudianus agree in error, while Venetus has the truth, and concluded from this (logically, within the framework ofJeep's scheme) that all three MSS must derive directly from an [a] which contained variant readings. 3 (iii) The subordination of [x] to [z] in the stemma is disputable. Jeep's words are (Acta, p. 369) : ' [x] . . . wegen 1.53 aus einer �elle geflossen sein muB, die wie [z] beschaffen war, so ergibt sich mit logischer Nothwendigkeit, daB die Ueber­ lieferung, die wir durch [x] markirten, der Ueberlieferung [z] als der vollstandigem tmterzuordnen ist '. But the fact that the source of [x] is 'like [z ] ' does not prove that it actually .

.

1 For full evidence, see below, p. 63 .

2 This point was made by Bonnet in his review of Jeep's ed. minor in Revue

critique for 1 875 · Jeep's rejoinder, in the ed. maior, was thatfusis was simply a ' symbolon integrioris lectionis ', but no supporting examples were forth­ coming. Nor can I fmd any. J The same conclusion, that LVG were ' drei selbstandige Reprasentanten des Archetypus ' , was reached also by Baehrens (foe. cit.), but from a different premise.

T H E M A N U S C R I P T T R A D IT I O N

was [z]. An equally possible hypothesis, and one that Jeep makes no attempt to rule out, is that [z] and [x] have a common parent. Combine this possibility with the elimination of stage [z] ( (ii) s up ra ) , and it will be evident that [x] might well be a primary witness to the text of [a]. (iv) But, even if we agreed with Jeep in subordinating [x] to [z] in the stemma, we could hardly concur with him in ignoring it altogether for the constitutio textus. In stemmatic terms, [z] can no more be determined without [x] than [a] can be established without [z]. Yet, throughout his discussion of the tradition, Jeep seems unaware of dus fact. For him, the onlissions characteristic of Classes rr-v disqualify those Classes as witnesses to the text ; and the possibility that in those parts of the text where they have not suffered any loss they nlight represent their source with greater fidelity than less mutilated MSS seems not to have crossed his mind. (v) If Class I a is really as valuable as Jeep makes it out to be, why does he desert it so often ? In the ed. maior, which represents his last thoughts on the text of the D.R.P., I count no less than forty-five cases where he prints the readings of MSS other than Laurentianus and Vossianus, or readings of the edd. vett. ; 1 and that without taking account of mere orthographica, which are numerous, since neither Laur. nor Voss. spells at all well. (vi) In the light of such phrases as ' die jiingsten und schlechtesten Handschriften ' (Acta, p. 347) and ' MSS . . . wegen ihrer Jugend ganz olme Bedeutung ' (ibid. p. 3 5 1) , one is not surprised to discover that Laurentianus, the oldest MS known to Jeep, is regarded as ' codicum . . . Ionge optimus '. But con­ siderations of convenience, no less than his estimate of its intrinsic worth, seem to have played their part in conditioning 1 Most of the readings which Jeep knows only from the edd. vett. are of course found in MSS. Of the forty-five readings in question, only five cannot be traced back beyond the printed editions. 37

INTR O D U CTION

his attitude to Laur., as i s evident from the observation that, unless one cleaves firmly to that MS, ' in talem variarun1 scripturarum labyrinthum innumerabilium fere codicum, eorum quidem plerumque eodem saeculo exaratorum incidis, ut frustra filum qua eras, quod te ad lucem reducat ' (ed. maior I,lvi) . However, both in theory, by admitting that it was ' recen­ tioris interpolationis spurcitiis nimis iam contactus ', and also in practice (see (v) above),Jeep himself went far towards weaken­ ing his case for the pre-eminent authority ofLaur. The view which I consider correct, namely that Laur. is of very average worth, was first advanced by Birt, 1 but by no means as conclusively as might be ; there is room therefore for a fresh look at the matter. The re-examination ofLaurentianus which concludes this criticism ofJeep's analysis of the tradition com­ prises, first, a rebuttal of his arguments in support ofLaur., and secondly, a statement of the case against it. In Acta, pp. 3 7I ff, twenty-two passages are adduced in which, it is alleged, Laur. is alone in preserving or leading to the truth. Were the allegation true, Laurentianus' claim to supremacy would have a good deal in its favour ; but it is not. My exami­ nation of the MSS reveals that other MSS are in agreement with Laur. on eighteen of the twenty-two occasions, and on five of the eighteen the MSS in question are as old as, if not older than, Laur. (I refer to B L I and Z). Further, Laur. seems to me to be right on only five of the twenty-two occasions, with four others where it might possibly be right; on all nine of these occasions other MSS bear it company, B L I and Z being among them in four instances. A brief survey of all the passages may be helpful.2 I. I J 8 ff.3

Laur. has

v.

I39

added in the margin by a second hand

I

Praef. pp. clv, clvii-clviii . Jeep's numbering of verses is idiosyncratic, and I therefore add the traditional numbering in brackets where necessary. 3 For a full consideration ofthis very vexed passage see my commentary ad loc. z

38

THE MANUS CRIPT TRADITIO N

contemporary with the ftrst ; the word ferris, however, is omitted. Subjoined is the gloss Jurtim quia timens raptus. From these pheno­ mena Jeep inferred that the incomplete verse was, like furtim . . . raptus, a gloss concocted by the second hand, and that it was from Laur. that the other MSS offering 1 3 9 got the verse, subsequently completed by the addition of ferris. This preposterous hypothesis was gently demolished by Baehrens, who observed (Fleckeis. Annal. VIII (1 872), 63 4-6) that scribes do not gloss verses they themselves have fabricated, as would be the case if Jeep's idea were right. The truth of the matter is obvious, namely that the second hand has copied 1 3 9 from an earlier MS but omitted terris by oversight. This same notion, that the original glosses of Laur. have been taken over by other MSS, is exemplified in the following five cases. 1I. I 70 ( I 7 1 ) dura text ; cicula ( sicu la) superscript. I favour Heinsius' salida, from the Isengrin's solita, and assume that dura has come from v. r 86 infra. sicula is found also in Z and is therefore unlikely to have originated in Laur. III . 1 3 7 videtur text ; ruenti illi festinanter margin. I agree with Birt (p. clvii) that videtur is ' inanis et quasi sterilis scriptio ' and regard it as a false reading ; ruenti I ftnd in MSS as old as Laur. and consider to be the true reading, added here with the gloss illi Jestin an ter. III . 72 pallescere text ; vel nigrescere superscript. The former reading, found only in Laur., was perhaps inspired by pallet in v. 8 8 ; it is less aptly used of vestes than the latter reading, which, as the word vel intimates, is not a gloss but a variant, and which is found in the text of MSS as old as Laur. m.108 tantum text ; vel saltern superscript. Both readings occur in other MSS, and either is possible. rn. r 3 8 iugales text ; vel dracones superscript. Again, both readings occur in other MSS ; the former, offered also by B and L r, seems superior. =

To these are added : 1.71 gelida. Found in other MSS also, and less attractive than Getica. 1.97 (98 ) vacuas. Found also in Z (amongst others) , and right in my opinion. 39

INTRODU CTION

I.I63 (I65) mittit. Peculiar to Laur., and not to be preferred to nutrit. I. I7I (I74) ojfensus rimata. Likewise uniquely attested in Laur., and unacceptable because (a) meatu can ill do without the qualification provided by offenso, and (b) rimata, which would be exceptionally rare as a passive formed from rimo, gives a poorer sense than rimosa. I.2 I 3 (2 1 5 ) pa11dit. Found elsewhere, and possibly right. I.216(218) pridem est. Other MSS also introduce the copula, almost certainly in error. Claudian's aversion to forms of esse, especially est, is commented on by Birt, p. ccxxiv. I.276 (278) auras. Offered also by Z (amongst others). I accept, but arces is not impossible. n prae£ 52 ab. Might be right, but only if ore is read in place of orbe. Other MSS besides Laur. have ab. II.I I tanti conscia voti. Also found elsewhere. In spite of the earlier appearance of conscia at v. 8, this conformation could be right ; but tanto concita voto is much more forceful. n.42 conscia. Other MSS too have this reading, which makes no sense here. 11.3 56 nullumque. Also in R 5· I believe that the original termination -i has here been assimilated to rogum ; the balance of the sentence certainly demands a nom. form. m.5 8 agmine summa. Unique, and producing doubtful sense and syntax. II1. 127 exudat (interpreted as exsudat) . Found also in L I and weaker than exundat. II1.145 pateat. Other MSS also have this reading, but the lengthen­ ing of -at, though not quite impossible, would be without parallel in Claudian, so far as I know (c£ Birt, p. ccxi), and in any case pateant is found in MSS. II1.2o8 praescierit. Offered also by other MSS. I find the prefix prae- without point in this context, where re- is clearly appropriate. m.396 mons. Read also by B and L I, and obviously right.

To my knowledge, there is not a single instance where Laur. alone provides the true reading, and few indeed where it preserves the truth with a small minority (say five or less) of the vetustiores. I note in fact just nine cases : 1.98 vacuas (with 40

THE MANUS CRIPT TRADITION

M 3 0 4 N Z) ; 1 . 1 43 aetas (with E 2 ] 4 L 3 P 4 R 2) ; rr prae£ r6 opus {with L 6 W) ; n.s6 Naides (with b I R4 R 25) ; n.74 meatu {with D ]4 K r) ; m. ros tua nata (with L 6 R 2) ; m.284 tenui dicto (with L2 R s R 7 U p.c. V) ; III.3 52 !assam (with H ] 3 R r W w) ; and III.4I 5 jerebar (with R2). Where Laur. stands alone against the other MSS, its readings are consistently inferior to those of other MSS. Setting aside orthographical and other trivialities, I count thirty-six passages where Laur. presents otherwise unknown readings ; 1 and in not one of them is it superior to the rest of the tradition. These passages are : adducunt obiceat I I 8 volutans I I9 ducebat( !) 160 henneos 269 suprema 28 1 ingentibus( !) 285 morpheus ll pf. 25 leporem blandi I xpulit ( = expulit) n. 4 animi audax 6 1 alpheus (et) 107 bellis (et) accomoda 1 09 nuntia 1 2 8 flavis 145 inventisque 1 8 0 cum 2 1 2 turpis crassa 232 iram (from 2 1 3 ?) I. 5 3

ll.23 J

74

25 7 297 3 00 3 17 3 19 3 42 3 60 m. 86

88 91 97 1 02 170 1 75 3 49 3 90

tritonia alumpnum (a.c.) etiam Jastigia media eventos( !) [sibi] concrescere fortior inspexere superni tandem cognoscere torva precor ita lugentem ut in umbra

In the light of this evidence I conclude that Laur., far from being ' codex Ionge optimus ' or the ' filum quod te ad lucem 1 Forty, if one includes 1 .165 mittit, 174 offensus rimata, m.sS agmine summa and 72 pallescere, mentioned above.

4!

INTRODU CTION

reducat ', i s in fact a M S of indifferent quality : in the words of Birt (p. clvii), ' non magis ille vincit ceteros quam ab ipsis vincitur '. 2

Birt maintains1 that ' in ipsius antiquitatis exitu' there were produced two ' recensiones ' or ' conformationes ' of the text, the one longer and consisting of 1 , 198 verses, the other shorter and comprising just over I ,ooo verses. The longer recension in fact is identical with Jeep's Class I, and, like it, is divided into two sub-groups, the division, however, being made not on the basis ofJeep'sjusis-pensis, but on what seems to me an equally invalid ground, namely the absence (Class 1 a) or presence (Class I b) of the spurious verse II. I I 8. The shorter recension (Class II ) is the same as Jeep's Classes IV and v combined, the incidence of the preface to book III being discarded, rightly in my opinion, as a ground of separation. To these two Classes is added a third (Class III), with which compare Jeep's Class III. In speaking of ' recensiones ', Birt at once makes it clear how he imagines Classes I and II to have come into existence, but whether he is likely to be right in thus presuming an artificial rather than an accidental origin for Class II remains to be seen. As for Class I, tl1e mere interpolation of a preface borrowed from another work hardly seems a substantial enough change to merit the term ' recensio ', and there is no reason for believing that such augmentation of the text could only have taken place ' at the end of antiquity ' ; its date indeed may well depend on whether or not we hold that tl1ere was an archetype of the MSS.2 Class III, because it appears to stand midway between I and II in textual conformation is tentatively thought ' magis etiam quam altera genuina videri ', but we are nowhere in­ formed of its relationship to the other classes, and it is in any 1

Praef. pp.

2

cxlvili ff.

42

See further p. 49.

THE MANUS CRIPT TRADITION

case composed not of seven MSS, as Birt erroneously main­ tains, but only of one fifteenth-century MS of no distinction (Riccardianus 7 1 8) . I We come now to the origins of the shorter ' recension '. Whereas Jeep's Classes IV and v are thought to have evolved from Class I via two intermediate stages, Birt's Class II is alleged to have been produced by a single ' diorthota ' whose intention was to shorten, almost to excerpt, the poem. As evidence of the activity of this editor, Birt observes (a) that II1.437, the final verse in the shorter version, marks the end of a paragraph, (b) that ' in libro I demptis versibus J40 ( 141)-2r3 optime continuat versus 214 versum 1 3 9 ' , and (c) that there would appear to have been some dissatisfaction felt with III. 28o--J 6o which resulted in the excision of those verses and the substitution for them of the stop-gap omnis honos etc. How cogent is this evidence ? On its own, the fact that the omission at the end of book III coincides with a break in the narrative proves nothing : the final verses of any work are always liable to be lost by accident, 2 and pure chance is no less likely an explanation in this case than intentional removal. The second piece of evidence (b) is vitiated by errors of fact, inas­ much as the omission in Class II MSS invariably includes v. 214, and only one of the fifteen older MSS in this class offers v. 1 3 9 ; and between 1 3 8 and 215 there is n o connexion at all. The third piece of evidence (c) seems to me to be based on a mis' Five of the seven Class m MSS in fact lack I .203-I4 (Mus. Brit. 6042, Antverpiensis m.59, Parisinus 8295 and 8296, and Escorialensis S.m.29), while a sixth (Ottobonianus 2859) inserts those verses between I. I 9 I and I92. Other MSS which do not conform to type and thus are wrongly classified are : (in Class I a) Vaticanus 2808, Monacensis I 5740, Ashbumhamiensis 977 and Laurentianus XCI sup. 4 ; (in Class I b) Laurentianus XCI sup. 34 and Perusinus ; (in Class n) Urbinas 657 and Parisinus 8o8o. On all these misclassified MSS see below, p. 58 and n. 2. 2 My 0 4, for instance, which belongs to Class I , lost m.44o-8 through the excision of one lea£

43

INTRODU CTION

reading of the facts. The assumption that readers found m.28o3 60 objectionable is founded, one suspects, though Birt does not say so, on the remarks of one Martinus de Vicomercato, the scribe of Leidensis 3 86 (quoted by Birt, p. cxlviii, n. 4) , to the effect that ' there are some MSS which have these verses although they were not composed by Claudian, but inserted in the poem after his death ', and ' some people think that Claudian did not compose these verses eo quod non videntur aliquibus tam sacra spiritu alfiata tamque sonora et mellifluo poemate carmina reddimita '. In my view, however, these comments may equally well be understood to mean, not that the passage was thought to be of inferior stuff and therefore was removed, but that in face of the omission in Class II (however it came about) scholars concluded that Class I must have acquired a spurious insertion after m.279. As for the literary quality of m.28o-360, that was almost bound to be depreciated once the authenticity of the passage had been called in question; and there is nothing to show that it was because they were felt to be less ' melodious ' and ' mellifluous ', or for any other literary reason, that the verses came to be omitted. It may further be remarked that the stop-gap omnis honos etc. does not appear in all MSS that lack 280-3 60, as one might expect to be the case were an editor at work here. Of the fifteen Class II MSS three have no bridge­ verse at all, and two others have one only in the margin (yet another, 0 3, deleted 279 in the text and substituted hec ait etc. for it in the margin) . From this I infer that the stop-gap verse was concocted after the loss of 280-360 and not contempora­ neously with it. Apart from the possibility that he might have had the vague feeling that III . 280-3 60 were not up to the standard of the rest of the poem (surely a most improbable causa delendi), what motives might Birt' s ' diorthota ' have had for cutting out the three sections in books I and III ? I must confess I cannot find 44

THE MANUS CRIPT TRADITI ON

any. Had the omission in book I been confined to, say, 16o-78, there would be grounds for thinking that he could not abide excursuses which held up the course of the narrative ; but the removal of everything from I4I to 2I4 is quite unintelligible, the more so as III. I I2 ff , where Ceres is found addressing her mother Cybele in Phrygia, cannot fully be appreciated without I. I 79-2I3 . If, therefore, there really was a ' diorthota ', he was utterly incompetent at his job ; so incompetent in fact that he not only removed parts of the main narrative but also left sentences without end or beginning on two occasions. Such bungling, admittedly, would not be without parallel ; a classic example of senseless excerpting is to be found in the Bern Horace. But that such devastation is to be laid at the door of, not a medieval scribe, but a ' diorthota ' operating ' in ipsius antiquitatis exitu ' I find it impossible to believe. The arguments advanced in support of this early date are (a) that the elimination of the preface to book III and the omission of II. I I 8 are ' features redolent of real antiquity ' ; and (b) that the shorter version comprises about I,ooo lines, 1 a figure which is said to be 'proprius fere librorum antiquorum poeticorum '. These arguments are not convincing. In the first place, not all repre­ sentatives of the shorter ' recension ' omit the preface and n. I I 8 (the preface is present in b i E I F2 L4 0 2 and P 2, the verse II. I I 8 in the margins of G I and L4) ; and in the second place Birt does not prove that the exemplar from which his ' diorthota ' worked must have contained these interpolations. Since they were not in Claudian' s autograph, why may we not believe that the shorter version derives from a medieval exemplar that reflected the original in omitting such intruders ? The figure one thousand, which Birt finds so significant, seems 1 Birt refers to his Das antike Buchwesen, p. 293 . He might also have added pp. J OD-I , where the D.R.P. is mentioned.

45

INTRODU CTION

to me devoid o fall importance. Surely no editor, at any period, would think it desirable to reduce a comparatively short epic of 1 ,200 lines to a round 1 ,ooo for purely theoretical reasons ; nor indeed do the statistics available suggest that the larger figure would have seemed 'paullo amplior debito '. The existence of a ' diorthota ' such as Birt postulates cannot of course be ruled out absolutely, but since indications of a rational plan ofexcision are so totally lacking, it must seem more likely that it was not deliberate policy that produced the shorter conformation of the text but the accidental loss or casual removal of several leaves. Consideration of the number of verses involved in the three omissions (1. 139/140/141-214 = 76/75/74 verses ; III.280-3 6o = 8 1 verses ; and III. 3 61-43 7, which stand between the two lacunae in book III, = 77 verses) might seem at first sight to suggest that there must have been at least one intermediate stage between the full and the abbreviated text, but it is not uncommon for MSS to show a variable number oflines to the page, 1 and not impossible therefore that all three passages should have been lost in one and the same MS. Supposing, though, for the sake of argument, that there was an intermediate stage between the full and the abbreviated text, could there be a reflection of it in Palatinus 1 573 , the repre­ sentative of Jeep's Class II ( = R 25 in my sigla), which omits III.280-3 60 only ? The possibility cannot be ruled out com­ pletely, but it must be noted that Palatinus' testimony is not reconcilable with that of my H, a MS unknown to both Jeep and Birt, which omits 1. 141-214 only. The lines of evolution towards the shorter text that might seem to be warranted by these MSS are in fact mutually exclusive, and their convergence could hardly have had any other outcome than the reciprocal repair of their respective omissions. Therefore, even if the shorter text was preceded by an intermediate stage, it is certain 1

So F 1

L2 and R25,

for example.

T H E MAN U S CRIPT TRADITI O N

that H and R 2 5 cannot both simultaneously reflect such a stage. At the very least one or other of them must be, as it were, an impostor ; and it may well be that both alike are products of a process of supplementary conflation that pre­ sumes the prior existence of the shorter text. 1 In his evaluation of the MSS and the use he makes of them in constituting his text, Birt is more satisfactory than Jeep. He realises that all three classes must be consulted, where they are available, and not just that one in which the full text is present ; and far from clinging to one supposedly ' best ' MS, he gives equal and impartial consideration to representatives chosen from each of the classes. At the root of this new attitude is the important perception that the tradition is affected by contami­ nation ; and evidence is adduced to show how inextricably confused are the interrelationships of the three classes and the MSS selected to represent them.2 Six MSS are cited regularly in the apparatus criticus, and two others occasionally. These are (in my sigla) : F I P 5 (A I) of Class r a ; C I 0 2 (0 3 ) of Class n ; and A 2 L 3 of Class III . The use of fewer MSS than these would, in Birt's opinion, have made for an insufficient and inadequate presentation ' of the history of the text through the twelfth, thirteenth and four­ teenth centuries ' ; the use of more would simply have resulted in the accumulation of ' ineptiae ', there being ' scarcely one reading which ought to be received into the text yet is absent from my selection '. Confronted with more than seventy MSS, Birt of course had some justification for editing on the basis of a representative selection ; but he had no justification for changing his definition of c{ from ' manuscripti ceteri quotquot extant ' (p. clviii) to 1 See further p. 59· I cannot imagine why Birt should then say (p. clvii) : ' Eae autem lectiones

z

paullo maiorem auctoritatem sibi arrogent necesse est, de quibus inter classes binas (vel inter classium binarum singulos libros) convenit atque consensio est.'

47

INTRODU CTION

' codices deteriores ' (p. 3 48) . Even if i t were true that the other MSS offered hardly any true readings not found in the selection, that in itself would be no reason for dubbing them ' deteriores ', for, in point of fact, Birt' s text could be elicited from almost any eight MSS picked at random from the tradition, and his particular selection is no better and no worse than many another would be. But, without realising it, Birt himself gives the lie to his assertion that the c; MSS contain ' scarcely one reading which ought to be received into the text yet is absent from my selection ' in that on a number of occasions he in fact prints c; readings on the mistaken assumption that they are in the text of his eight. Orthographica apart (and they are numerous), the following may be instanced : 1. 19 firmat (not in C) ; 89 Tegeaee (not in A) ; r 72 amnis (not in D) ; 2 so iussis (not in A) ; 11.242 pharetram (not in any of the eight) ; III . 144 decusserit (not in A) ; 14 5 pateant (not in A) ; 1 59 plangitve (not in SW) ; 23 3 albet (not in C) ; 248 et casus (not in D) ; 28 7 an gravis ; 296 actum. At III . 1 7 , however, Naides is openly printed on c; authority ; and at III . 236 pulsu strepituque is adopted from Heinsius' Bononiensis ( = my b r ) . (More frequent still are those cases where a MS reading which I regard as true is either totally unknown to Birt or, if it is known, is assigned to c; or to edd. Once again I omit points of spelling (and also such things as tum for tunc) . Unknown : 1. 162 nullo teritur (yet this order is in Birt's V) ; 11. 162 prolabitur; 3 64 genialia ; III . I03 etiamnum ; 323 }ines. Attributed to c; or edd. : 1. 9 8 aures; 164 molibus ; 11 p£ 2 ebur; 3 1 qui; rr. I 84jractoque ; 2 55 rabido ; 274 sentit; 287 derunt; 111. 44 abdita ; 1 8 9 nunc ubi ; 221 adfatu (yet it is in Birt's BW) ; 262 ultro ; 26 5 avexit; 295 delapsa ; 443 sulcis ; 444 innatat.)

T H E M A N U S C RI P T TRADI T I O N

3

Before embarking on a new classification and evaluation of the extant MSS, it will be worth while considering what can be known about their lost antecedents, and in particular whether there is any evidence for the existence of an archetype1 inter­ mediary between the original and its surviving posterity. Birt, while discussing in some detail his theory of ancient recensions of the poem, omits to explain how these may be supposed to be related to the original, but Jeep argues strongly, and with an apparently overwhelming mass of evidence,2 for the view that behind the extant MSS there stands a single common source or archetype. The present discussion may therefore conveniently start from an examination of the case put forward by Jeep.3 First, it is alleged that there are lacunae after 1.275 and in 11.209 between ille and velut. 4 In the latter passage the elliptical syntax is mmsual, but can be paralleled, and there is therefore no ground for supposing that something has been lost. In the former the absence of any description ofProserpine's encounter with her three divine visitors and the abruptness of the transi­ tion from Proserpine to Pluto may be felt to be stylistic weak­ nesses, but I am not convinced that they necessarily indicate the loss of a passage in which hostess and guests exchanged civili­ ties and Venus laid her trap for the naive Proserpine.s Secondly, 1. 1 7 1-8, 11.3 3 7 and m. 3 1 2 are condemned as spurious, and the preface to book rr , though acknowledged to 1 I use this term in the traditional sense and not as it is employed by A. Dain.

2

Acta, pp. 3 56 ff. For ease of reference I tacitly substitute the standard numbering of verses in cases where Jeep deviates from it. 4 Jeep, rightly in my view, rejects the notion that tl!e preface to book 1 is defective ad .finem. s On botl! tl!ese passages see tl!e commentary ad locc. 3

4

49

HCD

INTRODU CTION

b e by Claudian, is said to b e out o f place in its present home. Not one of these supposed instances of interpolation seems to me at all certain. 11.3 3 7 is indeed superfluous after 3 3 5-6, but one cannot be sure that it is not a variant of Claudian's own composing. I. I7I-8 may strictly be irrelevant to the main course of the narrative, but it seems natural enough that the references to Etna should be followed by a digression on vulcanology, and the resumptive hie in 179 is perfectly intelli­ gible. III.3 12, whose credentials I regard as unimpeachable, needs only to be reinstated as part of Ceres' utterance to elude Jeep's charge that it interrupts the progress of the action. 1 The preface to book II will be dealt with at length in the chapter entitled ' Date and Circumstances of Composition ' ; for the moment I need only say that in my opinion the theory that it reached its present position otherwise than by the author's volition is far-fetched and unnecessary. Further confirmation of the existence of an archetype is found in the supposed disappearance of an original division of the present book m into two books, the division to fall between m.3 3 1 and 3 3 2, where the theme of Ceres' wanderings is taken up. This hypothesis, though not unattractive, is however completely subjective and destitute of supporting evidence. There is no hint of a fourth book in the MSS and the medieval catalogues (including one as old as the eighth century) ; one cannot use 1.26-30, as Jeep does, as though it were a precise table of contents ; and whether Claudian' s composition on the subject of Ceres and Proserpine ever proceeded beyond III . 448 to its ultimate conclusion or not, the mere fact that the poem is now, as everyone admits, incomplete ought to deter us from speculating about the author's intentions concerning the compass of book m. 1 Further discussion of these three passages will be found in the commentary

ad lace.

so

THE MANU S CRIPT TRADITIO N

Finally, pointers to a lost archetype are discovered in a number of allegedly corrupt readings said to occur in all the MSS. 1 Of the seven readings mentioned, however, one is in my opinion true and therefore irrelevant in tlus context (m.23 3 alget) ; and in the remaining six cases Jeep is led astray by his exclusive concentration on ' representative ' MSS, for the readings he lists (1.8 lumina ; 72 flare cupit; 220 peragit; 111.44 addita ; 5 8 -que ; 7 8 quaesivitque), while undoubtedly or probably false, are not in point of fact common to the whole tradition. As it seems to me, two things make it idle to hope for complete success in proving what Jeep has thus failed to prove : the apparent lack of what are perhaps the surest criteria, inter­ polations or lactmae common to all the MSS, and the conse­ quent fact that any conclusion about an archetype must depend on our judgement of the truth or falsehood of individual readings, few of which provide particularly cogent evidence. In two places, 1.123 and m.39, all MSS are at fault in matters of spelling ; but purely orthographical coincidences prove nothing. More weighty is 1.284, where all2 MSS have the nonsensical ethonusque (or similar gibberish), and Jeep's con­ jecture Cthoniusque is very likely right. In all other cases where certainly conjectural readings figure in the text,3 the ousted readings of the MSS are meaningful per se and unsatisfactory only in the light of their context. Least satisfactory of all, and therefore most likely to be generally acknowledged as corrupt, are rr.23 ima parte viget, where C I offers a pointer to 1 Jeep also uses the agreements of his MSS against the vulgate to prove the existence of an archetype. Fourteen out of the fifteen readings cited from his MSS, however, are probably true; and coincidence in truth can prove nothing about an archetype. The fifteenth case (n. 1 6 5 elusos : inclusos) will be discussed below, p. So. 2 Except n, which offers aetonque, presumably a humanist correction. J At n.23, 171, m. 1 3 7, 25 1 , 265 , 3 5 9·

51

INTROD U CTION

the truth in ima viget parte, 111. 1 3 7 sed, and 111 . 3 5 9 petitura,Jeritura etc. If further evidence for the existence of an archetype is lacking, this is probably due to the learned activities both of medieval and renaissance copyists and of early editors, none of whom make it clear whether they are incorporating into the text their own conjectures or readings handed down from older sources. When Parrhasius, for example, offers Coei at 111 . 3 47 in place of the manuscripts' impossible Caci, and when the Isengrin edition gives vomit for movet at 1. 163 and elusos for inclusos at n.r65, are we dealing with conjectural or transmitted truth ?1 Similarly within the MS tradition, in the case of true readings found only in a single MS or mere handful of MSS, one cannot be sure whether one has to do with the deliberate (or accidental) correction of a corrupt archetype or the pre­ carious survival of the truth. In 1. 164 and 220, for example, practically all the MSS have motibus and peragit respectively ; a very small number, however, offer molibus and peragi, which are almost certainly what Claudian wrote. If it could be established beyond all doubt that molibus and peragi were restored by chance or by emendation, motibus and peragit would be revealed as further features of the archetype hitherto so tenuously documented. It is quite on the cards, how­ ever, that molibus and peragi are archetypal, and motibus and peragit simply instances of the widespread diffusion of false readings. If then there was an archetype, and I think on balance there was, its date must be quite conjectural, and the most one can do is to establish the limits within which it must have been written. A terminus ante quem of at latest the eleventh century 1 Other probably or possibly true readings first attested in the early printed edd., without indication of source, are : I.4 concussa, I. I 89 culmi, rr.8 3 Sabaeis, n.84 busto, m.267 iamiamque, m.297 deiecta, m.3 32 Jlavum. The Isengrin seems to come near to the truth at II. I7I solitaque, whence Heinsius' solidaque.

52

THE MANUS CRIPT TRADITIO N

seems suggested by the wealth of divergent readings and other textual phenomena that is already manifest in the earliest extant MSS. 1 The terminus post quem could be very early indeed, and the fact that corruptions common to the whole tradition are apparently neither numerous nor complicated does not preclude the possibility that the archetype could be the first ' published ' edition of the poem, if that were produced from a difficult autograph by someone other than the author, or at least an ancient exemplar of the poem. Finally, some mention must be made of Jeep's theory that the D.R.P. , prior to its casual union with the corpus of pane­ gyrics etc. in some (not all) of the extant MSS, 2 and before the total separation that preceded such union, was transmitted together with the panegyrics etc. in a MS ' in dem uns der Claudian noch als einheitliche Ueberlieferung entgegentritt'. This conclusion is reached from two assumptions, that the borrowed preface to book III was already interpolated in the archetype of the D.R.P. , and that this interpolation could only have been made if and when the panegyrics and the D.R.P. were united in one book. The theory that there was just such a pre-archetypal MS3 enables Jeep to explain not only how the preface to book III came to intrude into the text of the D.R.P., but also hqw the preface to book II, which he (unjustifiably) regards as introductory to a lost panegyric on Stilicho, was 1 I do not of course mean to suggest, on the assumption of a medieval archetype, that the readings offered by the MSS must be either archetypal or post-archetypal. The example of other poets, Juvenal for example (see Housman's ed. p. xl n. 1 ) , is sufficient to show that MSS quite independent of the archetype may have made their contribution to the text through collation with the linear descendants of the archetype ; and many readings true or false could have travelled down by such hidden routes from antiquity. 2 In A 2 b 1 C 1 E 1 F2 F 3 F 8 F 1 3 F 14 ] 1 ] 3 K 4 L 1 L 4 P 1 P 4 R 3 R 4 R 1 3 R 14 R 1 5 R23 R26 (my sigla) the D.R.P. follows Claudianus maior; in E 2 L 3 0 2 P 2 Z it stands first ; and in P 3 it comes in the middle. J Jeep himself uses the term ' Archetypus ' both of this MS and of ' die nachweislich j iingste Q!!elle ' of the D.R.P.

53

INTRODU CTION

divorced from the carmina maiora and reached its present location. 1 The supposed characteristics of this pre-archetype need not detain us ; the crucial question is whether there really was such a MS containing both the D.R.P. and the panegyrics. First of all, it is anything but certain that the archetype of the D.R.P. included the preface to book m. Jeep's distinction between his Classes I a and I b being, as I have argued above (p. 3 6), in­ admissible, any suggestion that the preface was in the archetype must rest on the demonstration, which Jeep nowhere gives, that his [x] derives from [z] . In the absence of such demonstra­ tion, there is nothing to gainsay my view, that the preface was first introduced into my Class a ( =Jeep's Class I) and passed on from there into the other classes by haphazard conflation. 2 Secondly, it is by no means necessary to assume joint trans­ mission of all Claudian' s works to account for the presence of the preface in MSS of our epic, as must be clear from the fact that the first ten verses of the preface appear also in Voss. lat. I I I , a ninth-century MS mainly of Ausonius, and the first two in the eleventh-century Monac. lat. I45 I6 (c£ Anth. Lat. 652 Riese) . Patently the interpolation of the preface could have been made whenever the desire for an introduction to book III made itself felt and a text of the panegyrics was available to the interpolator. I see no reason therefore for believing that the D.R.P. was transmitted together with Claudianus maior in the pre-archetypal stage of its transmission. Unlike Jeep and Birt, I do not regard such random features as the incidence of the spurious verse rr. I I 8 and the interpolated third preface as satisfactory criteria for classification, and my own attempt to distribute the MSS into groups is based solely 1 To this stage also are assigned the hypothetical, and to my mind im­ probable, losses of the conclusion to book IV and one leaf containing fifty-eight verses thought to have stood between 1.275 and 276. See above, pp. so and 49. 2 See below, p. 56.

54

THE MANUS CRIPT TRADITION

on the incidence of the three major lacunae, at I. I4I-2I4,1 III.280-360 and III . 43 8-48. On this basis one may establish : Class a. Those MSS which are free from all the lacunae = A I b 2 C 3 c D e 1 e 2 F 1 F 5 F 7 F I0 F I2 F 1 5 F r6 fi G2 G 3 G4 h ] 3 j Kr K 3 K4 L 5 L 6 L7 L 8 L IO Mr M2 M 3 m n 0 4 05 06 P5 P 6 P 7 P 8 P 9 pi p 2 Q R r R5 R 7 R8 Rn R 12 R 14 R 17 Rr8 R 19 R 2o R21 R22 R24 R26 R 27 R28 R29 r T T2 T 3 u V 2 V 3 W w y. In all, 73 · Class [3 . Those MSS which have all the lacunae = B b r E r F 2 F 3 F 8 F 13 F 14 G r K2 L r L4 0 2 0 3 P 2 R 3 R 4 R 6 R9 R 1 5 R23 (C I , which now lacks m.371 to the end, probably belongs to this class). Including C I, 22. Class y. Those MSS which are defective only in part = A 2 a C 2 E z F4 F 6 F9 F n H ] 2 ] 4 k L2 L 3 M4 O r P r P 3 P 4 R 2 R IO R 1 3 R r6 R25 t U Y r Y2 Z. In all, 29. Fragmentary and therefore unclassifiable are : d (which ends at 11.79, omitting I. 14o-214) ; fz (offering only the three prefaces) ; f3 (ending at 1.123 ) ; J r (presenting only III . 19-279 and 361-43 7 ) ; N (which ends at 11.52 and omits I. I41-214) ; R3o (offering only the second preface) ; T 4 (which ends at n.210) ; and V (which offers only m.28o-448 ) . The ultimate derivation of the Class a MSS from a common ancestor is by no means unquestionable. Nothing is proved about the parentage of Class a by its exemption from the three 1 Few Class i3 MSS have 1 3 9 andjor 140 : the former is present only in F 1 3 K 2 and L 4 , the latter only in E r F 2 F 13 (before 1 3 9) K 2 (before 1 39) L 4 and R4. If these verses were in fact in Class [3's ancestor and formed part of the lacuna, they have obviously been retrieved by contamination in these six cases. But I suspect they are later than the lacuna. See the commentary on 1. 1 3 9 ff.

55

INTRODU CTION

lacunae found in Class !3, for in this respect it simply mirrors the original. Nor does it offer any readings that are found in all its members but nowhere else. The only textual feature in fact that suggests a common source for Class a is the presence in all its members1 of the borrowed preface to book III. Agreed, this preface appears in other MSS as well-in eight of the twenty-two Class !3 MSS and in twenty-four of the twenty­ nine Class y MSS-but it is not a universal characteristic of the other classes, as it is of a. It follows therefore that unless one believes, with Jeep, that the preface figured in the archetype of the MSS (see above, p. 54) but was expelled as an alien intruder in twenty of its surviving progeny (incl. G4), one must suppose, as I think more likely, that the preface originated in Class a and passed horizontally by contamination from Class a into a few of the Class !3 MSS and most of the Class y MSS. If this hypothesis is correct, we shall be justified in regarding Class a as descended from a single, albeit ill-defined, source; otherwise it could be that the MSS assigned to Class a have come down from the archetype along more than one path. 2 The fact that the Class !3 MSS all concur in the three lacunae is proof positive that they derive from a common ancestor. The readings of this ancestor can of course be discovered where all the members of all the classes are in agreement, and also in 1 Except the sixteenth-century G4, which has also, and uniquely, ejected the other two prefaces. 2 Within a three pairs of vetustiores (for definition of this term see below, p. 6o and n. 1 ) are more or less closely interrelated. A 1 and L 10 agree on a number of occasions : see the app. crit. to I.IOI, m . 1 77, 206, 3 1 5 , for example. D seems to be the main source of K 1 : see further, p. 6o. R s and R 7 are so closely linked as to be almost identical (both 01nit 11.3 3 3 , 347, m.59 and 279, and reshape 11.97 and 249 in the same way ; other cases of agreement in the app. crit., passim), though whether as gemelli or as exemplar and copy is not certain. In part they appear to draw on F 1, in their omission of vix at m.91 , for example

(also c£ the app. crit. to m. 1 1 9 and 1 37), but they are definitely not clescripti. Among the recentiores (for defmition of this term see p. 6o and n . I) in this class, R 14 and R 26 frequencly agree : both omit m.299, and both share otherwise unknown readings, for example at 11.189, 246, 271 .

s6

THE MANU S CRIPT TRADITION

perhaps twenty passages where one of two o r more variants is offered by all (or all available) Class 13 MSS, in company with some MSS from the other classes, but while there are many cases where its members disagree amongst themselves and form transitory coalitions with members of the other classes, there is not one case where Class 13 is united against the other classes. The character of l3 's source being thus veiled in obscu­ rity, it is impossible to determine whether that source derived direcdy from the archetype, or indirecdy via an ex exemplar. But no matter, for this question, like most questions concerning the classification, is of no practical importance, given the presence of wholesale contamination. 1 Since the textual conformation of the y MSS is very varied, the siglum under which they are mustered is scarcely more than a symbol of unity in disunity, though it must be added that a majority of them, and perhaps all, were produced by similar procedures and exhibit 13-type texts diversely and more or less thoroughly supplemented from ex. 2 1 See further p. 63 . Within Class (3 the following MSS are closely inter­ related : (i) 0 2 and P 2 (see the app. crit. to n pf. 3 3 , n.no, 280, m.14, 86, 98, 144, for example) ; (ii) F2 and R 1 5 (the latter is, basically, a copy of the former, as Jeep noticed, but its second hand has added material from some other source or sources, e.g. n . I I 8 , which F2 does not know) ; (iii ) the sub-group F 3 F 8 F 1 3 R23, which reveals in miniature how the tradition at large developed. All these MSS contain Claudianus maior in addition to the D.R.P. ; all omit 1.52, and all share a number of unique readings, e.g. n pf. 43 tribuitque busirida, m.267 exhausta. Undoubtedly therefore, F 3 is the main source of the later trio. But (a) F 13 offers 1.139 and 140, whicl1 are not in F 3 ; (b) R23 contains some interesting and otherwise unknown readings, whether of its own making or appropriated from MS (S) now lost (e.g. at 1.23 3 , n.48 and 341 ) ; and (c) all three fifteenth-century MSS occasionally adopt into their texts the variant readings of F 3 and relegate the readings of its text to their margins. 2 The basic exemplar of]4 (c£ M 4 ) might possibly have been of a-type, however. Since the scribe of this MS regarded m.280-360 as spurious, the fact that those verses appear out of place at the end of book m could be due to their having been consciously removed from the body of the text under the influence of Class (3, whose lack of them may have seemed to reflect the original form of text. Alternatively, ] 4's main source was a MS of(3-type whose supplementation from a was confmed to the passages 1. 141 etc. and m.43 8 etc.,

57

INTRODUCTION

Clear indications of such confl.ation are provided by L 2 , for example, in which I . I 4 I-2 I 4 and ru.28o-3 60 appear after 1.241 and m. 3 6 I respectively, I while 111.4 3 8-48 are omitted alto­ gether. From this it must be evident that L2 is a hybrid, the product of the careless and incomplete fusion of an a:-type MS with a MS which had neither I. I4I-2 I 4 nor m.28o-3 60 (nor, one presumes, 111.43 8-48 ) . Of equally mixed origins are other y MSS which in their several ways reveal peculiarities of order or omission that can only be explained on the assumption that a:-type MSS have been used, and carelessly, to mend the losses of !3.2 In some y MSS two of the three lacunae are correctly the verses m.z8o etc. being relegated to the end of the poem because of doubt as to their authenticity. P I might perhaps have been assigned to Class 13. I include it under y for various reasons, among them the fact that it shares with R z the abnormal omission of m.361. Certain Class 13 MSS have also been repaired in part : Pierre Daniel added m.43 8-48 in the Bemensis ; in G I a manus recentior added I.I 3 9-2I4 before book I and m.43 8-48 at the end of the poem ; in R I 5 a second hand appended I.I4I-2I3 [sic] to book m ; and in F 14, I.I4I-2I4 and m.28o-360 have been added after book m. In all these cases the supplements stand outside the body of the text and it is possible to isolate the l3 original. 1 m.36I in fact appears twice : once before 280 and again after 360. 2 A2 E2 L 3 P 3 P 4 R2 and Z omit I.203-I4 (these verses are added in the margin in A2, and by what looks to be a second hand at the foot of the appropriate page in Z), i.e. I.I4I-202 alone have been restored. R IO is a faultily corrected descendant of one of these MSS, or a lost relation of theirs, in that it has the order I. I4I-9 I , 203-14, I92-202. C 2 omits I.146-2 1 3 , i.e. I4I-5 and 2I4 alone have been replaced. t omits m.3 3 2-
s8

THE M AN U S CRI P T TRADITI O N

mended, the third incorrectly o r not at all. Thus, 1.141-214 and III.4 3 8-48 appear regularly in F 6 F n ]4 k M4 0 1 R 13 R 16 t U Y 1 Y 2. In two other MSS, F9 and J 2, only one of the passages, 1. 141-214, appears regularly. In the light of this partial correction of !3 from a it must seem highly probable that H (offering III . 280-3 60 and 43 8-48 but not 1.J41-214) and R25 (offering 1.141-214 and III .43 8-48 but not m.28o-3 6o) do not reflect hypothetical stages in the evolution of the shorter text found in Class !3 1 but are merely additional instances of incomplete supplementation of [3 from a. In all the cases mentioned above, the fusion of Class a with Class !3 has been carried out in so partial or careless a fashion that there can be little or no doubt of the hybrid origins of they MSS. But if such fusion has been botched in these instances, might there not be others in which the restoration has been executed so skilfully as to render the finished product in­ distinguishable from a true member of Class a ? Since the only respect in which a differs from the rest of the tradition is its exemption from the three lacunae, we cannot know for sure; but in one case at any rate, the possibility cannot entirely be excluded. The a MS C 3 offers 1.140 twice (in its regular position and also before 215), and after III . 361 it has the spurious verse hec ait etc. This w1ique arrangement may be explained in one of two ways : either C 3 's scribe worked basically from an a-type exemplar, occasionally consulting a MS of[3-type from which he took the second 140 and the spurious verse;2 or, perhaps less likely, his basic exemplar was a [3-type MS which he mended with almost perfect success from an a-copy.J I

See above, p. 46. In the !3-type MS 140 would precede 2 1 5 . Sheer inadvertence is a possible cause of the reduplication, but perhaps the scribe copied the verse twice because of uncertainty as to its correct position. 3 With 1 .14o-214 inserted between 1 3 9 and 140, and m.28o-361 inserted between 279 and the spurious verse, which was followed by 362. Within 2

59

INT R O D U CTION

The M S tradition falls naturally into two halves, the division coming at its chronological mid-point, about the middle of the fourteenth century. 1 Its older stratum, consisting of fifty-five codices vetustiores, concludes with (amongst others) two MSS, J 4 and K I , which together with the thirteenth-century D have left an appreciable mark on a large part of the younger stratum of seventy-seven codices recentiores. While not agreeing in all respects with D,2 the two later MSS J 4 and K I nevertheless share so many otherwise unknown readings with it that it must be accounted their main source. 3 Certain other vett. attach themselves to this group from time to time, notably F r R2 E 2 (see the app. crit. to 1.172, 254, II pf. I 7, 45, 11.56, I68 ) and W b I M 3 L6 (see the app. crit. to 1.260, rr. 84, I62, 343 , m.ro8, I66, 297, 323 , 3 39, 39I, 41 1 ) , E 2 and L 6 being especially closely related to it. Class y the following MSS are closely interrelated : F4 and P r (see the app. crit. to r. r oo, 121 , rr. r48, r so, 1 54, m.266, 399) ; A2 and P 3 (see the app. crit. to rr. n 3 , 228, 3 3 3 , m. r6s, 248, 367) ; R 13 and R r6 (apart from their dealings with m.28o--3 6o, they share a number of unusual readings in, for example, u.n4 clarescit, 129 hanc [mollis] flos amaracus, m.27 nee conlivescere, I S S atria et amissas) ; the group a F 6 k Y I Y 2 (apart from their handling ofm.280-36o, the following may be instanced : (a) a F 6 k interpolate Stat. Theb. 8 .26 between 1. 5 1 and 52 ; (b) a F 6 k Y 2 reshape m.443 to read omnibus admugit silvis et in equore fulvis; (c) a Y 2 read thamantis at m.347 and F 6 k Y I ramantis) . Slightly related are : }2 and R25 (both transpose m.247 after 241 and read tendat at m.388) ; E2 and R2 (see the app. crit. to 1.236 and m.426, for example). 1 The recc. begin with four MSS of the fourteenth/fifteenth century (R 8 R9/ro/29) and F6, dated 1394. The vett. end with twelve MSS (A I/2 b I H J 4 K I L 6 M I/2/3 R6/25) all written, in my opinion, in the first two or three decades of the fourteenth century. There is thus a gap of some forty years between the latest of the vett. and the earliest of the recc. 2 The following disagreements may be instanced : r.I I attollit K I , extollit D } 4 ; I29 germ ina D, cornua J 4 K I ; m. 88 pudor J 4, rubor D K I. All these readings occur in other vett. also and show in consequence that the texts ofJ 4 and K I are critically eclectic. J See the app. crit. to r.6I , 64, Io6, 1 5 5 (his), 1 56, 1 72, 201 , 2 1 5 , 2 1 8 , 232, 256, 269, 27I, and passim. In many places D is illegible and }4 K1 stand alone, but with the presumption that D, could it be deciphered, would generally be found to agree. So at 1.30, II pf. so, rr.8 3 , 242, 264, for example.

6o

THE MANU S CRIPT TRAD ITI ON

The number of readings that have been passed on from D J 4 K 1 to the recc. is very considerable, and these three MSS, more than any other vett. , though not of course to their exclusion, have moulded the features of the text as it was known in the late fourteenth and :fifteenth centuries. Predictably, it is the a-type recc. whose debt to them is greatest (especially clear cases are : c fI G 2 G 3 j 0 5 P 6 p I p 2 Q R 29 T 2 T 3 y), and the relatively few !3- and y-type recc. which are least affected. The number of recc. that seem not to have been influenced at all by D and its associates is, however, comparatively restricted : less than twenty out of seventy and more. 1 At this juncture certain questions inevitably arise : how far are the recc. independent of the vett. ? are they in fact indepen­ dent at all ? to what extent should their testimony be adduced in establishing the text ? It is surely a measure of the complexity of the tradition that, in spite of the existence of more than fifty vett.-no doubt also because of it-only one rec. may formally be eliminated as in all things derived from an earlier extant MS.2 But it is not necessarily or even reasonably to be expected that the copies made by and for scholars-and that means most of the surviving MSS-will reproduce one older exemplar to the letter;3 on the contrary the amanuensis will feel himself entitled and indeed in duty bound to consult as many sources as circumstances permit and to compose his text eclectically from the store of variants he has at his disposal. He may even be inspired to incorporate his own 'improvements ', perhaps because his exemplars are corrupt or simply because he feels that the text before him is not as elegant or as straightforward as he would wish. This picture of critical conflation and 1 These twenty include F 8 F 13 and Rz3 (whose allegiance, as we have seen, is to F 3 ) , R ro (see above, p. 58 n. 2), and the y sub-group formed by a F 6 k :z. F S, copied from F 3 . R 1 3 R r6 Y r Y 2 (see p. 5 8 n. 2). 3 Servile reproduction i s characteristic o f the ' collector's M S ' , such as F 8 or F 1 3 .

6I

INTR O D U CTION

conjectural alteration i s not in the least fanciful but i s based on hard fact, for there is no other way ofexplaining the appearance in the recc. of more than five hundred new, and often attractive, readings, or of accounting for the fact that even if their primary sources may often be isolated, they hardly ever adhere to them in all respects but regularly form more or less close attachments to other vett. In these circumstances, rigorous eliminatio such as is possible in so-called ' closed ' traditions is quite out of the question, and one must proceed by more rough and ready methods. Do the recc. offer readings found also in the vett. ? In that case con­ venience demands that we assume dependence, even if logic suggests that the recc. might be drawing on lost exemplars. Do they advance readings not known from any older MS ? In that case we may be uncertain whether we have to do with survival from lost exemplars or conjecture by the copyist himself, but at all events we shall not ignore anything that has intrinsic merit. I As the existence of so many vett. naturally leads one to expect, there are very few passages where what is undoubtedly the true reading first emerges in the recc. : two only in fact (ilia for ille at 1.67 and de- for dis-cusserit at III.I44), apart from cases involving orthography (see I. I I , r6, 24, 40, 1 5 1 , 284 ,2 285, II.30, ro6, 1 3 6, 3 3 8, m.3 3 , r 85, 225, 252, 269, 320, 3 62 ) . There are more than a few passages, however, where the recc. might be followed without detriment to sense or style. 3 The likelihood of contamination in the vett. is suggested by the medieval popularity of the poem and the comparatively large number of annotated MSS that have survived from the thirteenth and fourteenth centuries ; its reality is conftrmed by the almost total inscrutability of the manuscripts' interrelation1

In book I alone I cite from the recc. just over one hundred times. Unless G4 is drawing, as it may well be, on Parrhasius' edition. 3 For example, I. IO terrisfremitus, 229 mandata, n.217 foedare, III . I I veneran­ daque, 345 fractos. 2

THE MANUS CRIPT TRADITI ON

ships ; and the problems of handling it are made worse by the apparent lack of any MS that may be considered to have escaped contamination. Evidence to prove the operation of contamination was pro­ duced in adequate measure by Birt, but only for the eight MSS on which he based his text. Additional evidence to show that it is in fact universal in the older, no less than the younger, stratum of the tradition is hardly needed. Only the working of contamination can explain how it comes about (a) that the basic classes discussed above cut across one another and have lost whatever individuality they may once have had, (b) that MSS of one class are constantly found to share false readings, and sometimes even omissions, with members of another class, and (c) that there is frequent, often contemporary, alteration of readings to bring one MS into line with another. Unequivocal proofs of contamination may be found on every page of the app. crit. ; for the present a few examples of (b) may suffice : 1.107 thetin a, L I 0 2 P 2 R4, y ; u2 signa a[3y ; 23 6 omitted by 0 3 and ] 2 ; II p£ 30 om. by P 5 and R 3 ; 43 rubuitque a[3y ; II. I7I siculaque a[3y ; 250 om. by E I F 3 0 3 and P4 ; 284 mirabile] 3 R I, h i F2 G I 0 3 R 6, C 2 ; 344 om. by 0 3 and L 2 ; 3 67 om. by w and Z ; m.24 seges om. by 0 4 and H ; 49 om. by R 5 R 7 and L 2 ; 6 5 om. by L4 and L 2 ; 257 om. by L 4 and 0 I ; 397 haerere J 3 M I w ?, [3y. What are the consequences of contamination as regards the editing of a text in which it has been active ? Since contamina­ tion implies a critical selection from among a store of variants, it is obvious that the currency of a particular reading will indicate only how popular it was and have no bearing on its authenticity; obvious also that the true reading may be found in any part of the tradition, in few MSS or in many, depending on its appeal to the copyist. From this it follows that we can have no other guides in our 63

INTRODU CTION

search for the truth than the requirements o f sense and the linguistic habits of the author. These two general criteria ought to be enough; but in the case of this poem and this writer they are often inadequate. Since its subject is non-technical and its diction often trite and predictable, the very simplicity of the D.R.P. provoked learned copyists to tamper with its expres­ sion ; 1 and because they, like the author himself, had been schooled in the poetry of Virgil, Ovid and Statius, the new readings they concocted are often indistinguishable from the original in style and aptness. Nor is the principle ' lectio difficilior praeferenda' (and its ramifications) of much rele­ vance or value, for there are few readings that are comparatively more difficult than their alternatives. In short, on many occa­ sions one might as well toss a coin to decide which reading to print. III. HI S T ORY O F T H E TRAN S MI S S I O N

Evidence for the history of the poem's transmission, though steadily increasing, is still in rather short supply, and the picture that emerges from such facts as have hitherto been brought to light, notably by Birt and Manitius, 2 is fragmentary and in­ distinct. Our three main sources of information, quotations and reminiscences in the works of later writers, medieval library catalogues, and the surviving MSS themselves, do not become available in combination until the middle ofthe twelfth century, 1 Two hundred years ago J. M. Gesner acutely observed that ' voluisse subinde descriptores veluti ingenii certamen cum poeta inire, & quomodo variari posset oratio, ostendere ' (Prolegomo to ed. p. xii ) . z Birt, praef. pp. lxxvi ff. (' De carminum Latinorum primis fatis ') and ppo cxlvii-cxlviii ; M. Manitius, Philologus XIlX (r 89o), 554ff., ' Beitrage zur Geschichte romischer Dichter im Mittelalter, 20 Claudianus' (not mentioned by Birt), and Geschichte der lat. Litteratur des Mittelalters, Munich, I9I I etc., passim. o o o

H I S T ORY O F THE TRANSMI S S I O N

by which time the text is nearing the point of its widest diffusion and the MS tradition is already too complex to be unravelled. For the period before c. n s o, on the other hand, we have at our disposal only scattered citations or allusions and the testimony of a few library catalogues, from which little can be inferred. Because ofthese limitations to our knowledge, and for other reasons besides, 1 it would be absurd to imagine that this account of the poem's fortunes in its passage down the centuries could be anything other than partial and incomplete, constantly needing to be modified in the light of fresh evidence ; but even so, a resume of what is known at present seemed worth giving. Although there is considerable uncertainty as to the date or dates at which the books of the D.R.P. were written and made available to the reading public, 2 it seems clear that from the start the poem was transmitted separately from Claudianus maior3 and that it quickly achieved a wide circulation through­ out the western half of the empire. It is true that the verbal similarities linking the D.R.P. with poems by Claudian's con­ temporaries Prudentius, Licentius (the friend ofAugustine and, like him, a native of Tagaste), and Paulinus of Nola4 are 1 For example, one cannot draw hard and fast conclusions about the extent to which an author was known and read from the frequency or infrequency with which he is quoted or imitated (he may not be quotable or memorable) ; nor does the availability of MSS automatically guarantee that the text was read. Furthermore, quotation may sometimes be at second hand, and not prove direct knowledge of the work quoted : e.g. Albert of Stade (see below, p. 74) could have got the D.R.P. verses he uses from the Florilegium Gallicum (see below, p. 72). 2 See the chapter entitled ' Date and Circumstances of Composition '. 3 For discussion ofJeep's contention, that all Claudian's poems were trans­ mitted together at an early period, see p. 5 3 · 4 C£ Birt's notes o n D.R.P. 1 . 1 90, u.243 , u.258. I t may be observed here that many of the parallel passages adduced by Birt, both below his text and on pp. lxxviii ff., are far from convincing. That Prudentius and also Rutilius Namatianus were familiar with Claudian's political poems seems certain : see now D. Romano, Claudiano, Palermo, 1958, pp. 143 ff. (appendix entitled ' Linee d'una storia della fortuna di Claudiano '). 5

HCD

INTRODUCTION

tenuous in the extreme and quite inadequate as proof that those writers were familiar with Claudian' s epic, as they may very well have been ; but from the middle of the fifth century on­ wards there can be no doubt of its extensive currency. Dracontius, living under the Vandal kingdom in Africa, frequently borrows from Claudian, 1 and his epyllion De Raptu Helenae shows particularly clear traces of the influence of the D.R.P. To Sidonius, writing in Gaul in the fiftl1 century, Claudian is first and foremost the poet of the D.R.P.2 Others whose works reveal more or less frequent reminiscences of the D.R.P. are the African Corippus, the Spaniard Merobaudes, the Italians Sedulius and Venantius Fortunatus (who emigrated to Gaul at an early age), and the Gallic poets Prosper of Aquitaine, Alcimus Avitus, and Paulinus of Perigucux.3 Doubtless many another writer of the fifth or sixth century might have been added to this list had he chanced to incorpo­ rate in his work an w1mistakable allusion to the D.R.P. 4 The vogue enjoyed by the D.R.P. in the two hundred or so years following Claudian's death was not destined to last, however, and from the seventh to the eleventh century there is hardly a trace of it in the literature that has sur1 C£ C. Rossberg's dissertation entitled ' De Dracontio et Orestis . . . auctore . . . Vergilii Ovidii Lucani Statii Claudiani imitatoribus ', Nordae, 1 8 80, pp. 3o ff. ; also Romano, op. cit. p. 149. 2 In a verse epistle to his friend Felix, Sidonius refers to Claudian in the words non Pelusiaco satus Catwpo (hie tibi legetur) I quiferruginei toros mariti I et Musa canit itiferos superna (Carm. 9.274-6). Sidonius' poems contain numerous reminiscences of the D.R.P. 3 For all these poets c£ Birt, praif. p. lxxix and below his text passim. 4 Birt notes memories of Claudianus maior in the following authors : Boethius, Arator, Ennodius, Rutilius Namatianus, Rusticius Elpidius Domnu­ lus, and Claudius Marius Victor. S. Augustine cites Claud. 7.96 ff. in his De civ. Dei 5.26. F. Raby, Secular Latin Poetry 1 .162, asserts that the Irish monk Columban had read Claudian. If Columban's acquaintance with Cl. extended to the D.R.P., an interesting link would be forged with the oldest surviving library catalogues, esp. that from Bobbio, which was an Irish foundation.

66

H I S T O R Y O F T H E T R A N S MI S S I O N

vived. 1 The seventh-century grammarian Asper and three im­ portant figures at the court of Charles the Great, Alcuin, Angil­ bert and Theodulf, show a slight acquaintance with Claudianus maior, but ofthem only Alcuin has what looks to be a reminiscence of the D.R.P.,2 and an isolated one at that. In the mid-ninth century Florus, deacon of Lyon, recalls the D.R.P. in his use of the clausula Dindyma Callis (Poet. lat. aevi Carol. 11.558, Carm. xxix. r9 = D.R.P. n.269) ; 3 in the tenth century some know­ ledge of Claudianus maior but not the D.R.P. is shown by the anonymous author of the Gesta Berengarii and by Rather and Notker of Luttich, Heriger of Laubach, Wipo, and Gozwin of Mainz ;4 in the eleventh century only Baudri de Bourgueil seems to know the D.R.P. s But although the poem thus appears to have been very largely forgotten by men of letters for near on s oo years, there were certainly copies of it at various times in at least three major centres : the court library of Charlemagne, c. 790 ; 6 and 1 Birt, praef. pp. lxxx-lxx xi . Aldhelm quotes Claud. 7.98 (presumably at second hand from Augustine, De civ. Dei 5.26, or Oros. 7-35 ) and vv. 73 and 8o of the arguably spurious Epithalamium Laurentii (Claud. c.m. app. 5 ) , but apparently shows no knowledge of the D.R.P. : cf. J. D. A. Ogilvy, Books known to Anglo-Latin writers from Aldhelm to Alcuin, Cambridge, Mass., 1936, p. 28. J. de Ghellinck, L'Essor de Ia Litterature Latine au XII• siecle, vol. 2.83, states that there are citations of Claudian in Bede, but Manitius (in Philologus) had found none and neither Plummer (ed. of Baedae Opera Historica, p. 1 and note) nor Ogilvy (op. cit.) mention Claudian in their lists of authors cited by Bede. 2 See Birt's note on n.234. The phrase vestigia lambat (c£ Birt on m.253 ) might have been suggested to the Italian-born scholar Paul the Deacon by Stat. Theb. 10.822. 3 Noted by Manitius, Philologus p . 5 56. 4 For all these authors c£ Manitius, Geschichte, vol. 1.63 3 ; vol. n.52, 223-4,

326, 477· s In Carm. cxcvi, written between 1 0 99 and I I02 according to P. Abrahams, Les CEuvres Poetiques de Baudri de Bourgueil, Paris, 1 926, p. 232, he depicts

Countess Adela supervising the weaving of a tapestry which, like Proserpine's (D.R.P. 1 .247 ff. ) , represents the creation. 6 C£ B. Bischoff, ' Die Hofbibliothek Karls des Grossen', in Karl der Grosse, Lebenswerk und Nachleben, vol. n, Das geistige Leben, Diisseldorf, 1965. The catalogue in question, found in Berol. Diez B. Santen 66, was first published 5 ·2

INTRODUCTION

at Reichenau1 and St. Gallenz in the ninth century. It is also by no means unlikely, though we cannot be certain, that the D.R.P. accounted for at least one of the 'libri Claudiani poete quatuor ' that Bobbio is known to have possessed in the tenth or eleventh century, if not in the ninth. J The rekindling of interest in the D.R.P. and the production of our oldest extant MSS of it occur in the twelfth century, in the context of a general enthusiasm for classical literature. The centres of this new-found interest are England and France, especially N. France. There was a MS of ' Claudianus maior et minor ' in the Cluniac house at Limoges,4 and MSS containing unspecified works of Claudian, very possibly including the D.R.P., were preserved in the Cistercian house at Rievaulxs and in the Benedictine houses at Durham, Peterborough, Bee and St. Amand. 6 The D.R.P. is cited indeed only by Petrus Cantor, bishop of Paris, 7 Bernard Silvester of Tours, 8 and Helinand, the trouvere turned monk, 9 but there can be little by G. Becker, Catalogi bibliothecarum antiqui, Bonn, 1 8 85, 20.6.42, and was thought by B. L. Ullman, Scriptorium VIII (1954), 24 ff., to be that ofCorbie or a related centre. 1 Cf. P. Lehmann, Mittelalterliche Bibliothekskataloge Deutschlands und der z Cf. Becker, op. cit. 1 5.3 3 1 , p. 3 5 . Schweiz, Munich, 1918, 1.266.5. J C f. Becker, op. cit. 32.369--72, p. 69 ; M . Esposito, ]TS xxxii, 3 3 7 ff. 4 C f. L. Delisle, L e cabinet des manuscrits de la Bibliotheque Nationale, tom. n, Paris, 1874, 495 · The description of this MS is identical with that of a MS located at Avignon in the year 1 375 (cf. F. Ehrle, Hist. bibl. roman. pontif. tum Bonifatianae tum Avenionensis 1. 5 I I , 858) ; either therefore the Limoges MS itself was brought to Avignon, or a copy of it was made for Avignon. s For Rievaulx see M. R. James, Catalogue of MSS in the library of Jesus College, Cambridge, 5 5 · 6 For Durham see Becker, op. cit. I I7.3 5 1-2, p . 242 and 126.63 , p. 257; for Peterborough see Becker, op. cit. n6.4o, p. 239 ; for Bee see Becker, op. cit. 86. 9 1 , p. 201 ; and for St. Amand see J. Mangeart, Catalogue des MSS de la bibliotheque de Valenciennes, Paris, x86o, p. 3 3 no. 1 5 . 7 Migne, vol. 205.76 (' Unde poeta : . . opibus dotJatur avaris quidquid i n orbe perit ' = D.R.P. 1.21-2) ; cf. Manitius, op. cit. m.161. s Cf. Manitius, op. cit. m.205 . 9 Migne, vol. 2 12.648 = 75 3 (Ferrea lascivis mollescunt corda sagittis = D.R.P. 1.228). .

68

H I S T O R Y O F T H E T R A N S MI S S I O N

doubt that it was familiar to such celebrated students or teachers ofParis as John of Salisbury, Walter Map, Giraldus Cambrensis, Radulf ofDicetum and Alexander Neckam, all of whom allude to or quote from Claudian' s other works, and Alan of Lille, whose Anti-Claudianus is based on the In Rufinum. 1 Others whose knowledge of Claudianus maior at least is undoubted are Roger of Hoveden, William of Malmesbury, Herbert of Boseham, Joseph of Exeter, Eberhard ofBethune (author of the Graedsmus) , Godefrid of Breteuil and Nigel Wireker.2 Willelmus Cayso est pravus puer I This very human jotting on fo. 152 of N may serve to introduce us to what is without doubt the most interesting feature of the medieval tradition of the D.R.P., the fact that from the twelfth century at latest the epic occupied a place in the schoolboy's reading list. In an article entitled 'De librorum Catonianorum historia atque compositione ' , 3 M. Boas, author of the now standard edition of the Disticha Catonis, traced the development of the school­ reader which in its most expanded form incorporated the D.R.P. In the ninth century this reader comprised only the Distichs and Avianus' Fables, but by degrees other works were added to it : the Latin Iliad, Theodulus, Maximianus, and finally Statius' Achilleid and the D.R.P. , these two last ousting the Iliad. The final conformation, which is not found in any MS older tlun the thirteenth century but is likely to have come into existence in the previous century,4 thus included Cato, Theodulus, Avianus, Maximianus, the Achilleid and the D.R.P. (which sometimes precedes Statius), and the whole miscella1 C£ Birt, praef. p. clxxvii ; Manitius, op. cit. m.267, 624 ff., 639, 797 £, 786 ff. 2 Cf. Manitius, op. cit. m.413 , 467, 614, 652, 75 1 , 779, 8 1 3 . 3 Mnemosyne, n.s. xlii ( 1914 ) , pp. 17ff. C£ also the section on ' Curriculum Authors ' in E. R. Curtius's European Literature and the Latin Middle Ages (Eng. transl.), pp. 48 ff. 4 The eleventh-century Etonensis 1 50 already contains Theodulus, Maxi­ mianus, Stat. Ach., together with Ovid Rem. am. and Her. and Arator.

INTROD UCTION

neous collection seems on occasion at least, to judge from the evidence of L 5, to have carried the titles Libri de moribus or Libri ethicorum. 1 As these titles suggest, it was their value as sources of moral instruction that primarily recommended these works to the grammaticus ; but they were hardly less useful as models of eloquentia. 2 They seem to have maintained their position as elementary readers at least until the fourteenth century.3 Of tl1e extant MSS of the D.R.P. no less than fourteen may be described in Boas' terminology as 'libri Catoniani ', viz. C 2 C 3 H L 2 L 5 M 3 N (unknown to Boas) O I 0 4 (unknown to Boas) R I R 5 R 7 R25 w (unknown to Boas). As one might expect in the case of school books, a few of these MSS are in a very poor state of preservation. 4 The association of the D.R.P. with Statius' Achilleid also occurs independently of the full ' liber Catonianus '-in F I , for example, where the two epics, in a twelfth-century hand, follow a slightly older copy of 1 An index of the contents ofH includes libri ethici, i.e. Cato, Theodulus etc. 2 Interesting in this context are some remarks ofJacques de Vitry (ob. 1 240) in Parisinus 17509, fo. 3 1 , which are thus translated by L. J. Paetow (introd. to ed. of Henri d'Andeli's ' La Bataille des Sept Ars ', p. 16) : ' In spite of the value of the art of eloquence which we derive from the poets, properly called " auctores ", it is better to choose for our instruction those works which contain moral teaching, such as those of Cato, Theodulus, Avianus . . . 3 They figure in the reading-lists prescribed by the schoolmaster Hugo of Trimberg (fl. 1280) and the grammarian Eberhard the German, and are included in a list of the thirteenth or fourteenth century jotted in a thirteenth­ century Ovid at Tours (cf. E. K. Rand, Speculum IV ( 1 929), 265). E. Rickert, Modern Philology XXIX, no. 3 (1932), 257, notes that a ' liber Catonianus ' was bequeathed by William de Ravenstone to St Paul's Almonry School, of which he was master, in 1 3 5 8 (the catalogue transcribed perhaps includes a second ' liber Catonianus ' in item xxvi). 4 H, for example, has lost a number of folia and now contains only Cato, an incomplete text of Theodulus, the D.R.P. beginning at 1 .25, and Statius (followed by other educational works unconnected with the 'liber Cato­ nianus ') ; in N the D.R.P. ends abruptly at 11.52 ; 0 4 begins on fo. 245 and contains only Maxirnian, Statius and Claudian ; and w offers, after Theodulus, Cato . . . Avianus, Maximian and Statius, a text of the D.R.P. which is so badly smudged as to be almost totally illegible. '

70

H I S T O R Y O F THE TRAN S MI S S I O N

another educational text, Sedulius- ; 1 while in 0 3 the D.R.P. is paired with another epic read in the schools, the Alexandreis of Walter of Chatillon. 2 The use of the D.R.P. in schools inevitably created the need for a commentary to assist the master in his exposition and the pupils in their comprehension of the poem. This need was supplied, not later than the twelfth century, 3 by one Gaufridus of Vitry, whose annotations are preserved most fully in L 1o (of the thirteenth/fourteenth century), and in variously abbre­ viated form in at least nine MSS ranging in date from the twelfth/thirteenth century to the fourteenth. Four of these MSS are ' libri Catoniani ' (C 2 M3 0 1 R 1), and one Q4) , like L 1o, contains only the D.R.P. 4 Preceding the notes proper in L 10 is an Introduction (accessus) s comprising inter alia a brief (and ill-informed) biographical sketch of the poet6 and general comments on the subject-matter and purpose ofhis poem, and this Introduction recurs, with more or less substantial re1 Statius accompanies the D.R.P. in A I b 2 D F n F I2 R2I and T as well. 2 The Alexandreis figures in the reading-lists of Hugo and Eberhard (see

p. 70 n. 3 ) . Its great popularity is noted by Henry of Ghent (ob. 1295 ), De scriptoribus ecclesiasticis 20 : in scholis . . . grammaticorum tantae dignitatis est hodie, ut prae ipso veterum poetarum lectio negligatur. J I owe this information about date to Miss A. K. Clarke, who is currently engaged in editing Gaufridus' commentary. 4 The others are : A I Az P 5 U. Doubtless traces of Gaufridus' influence will be found in yet other MSS. s Cf. in general E. A. Q!! ain, ' The Medieval Accessus ad Auctores ', Traditio 3 (I945) , 2I 5-65. 6 To the following effect : ' . . . actor (auctor leg.) si quidem iste filius fuit cuiusdam nobilissimi romani civis claudii nomine a quo et ipse dictus est claudius claudianus qui in tempore florentini romanorum principis valde peritus fuit in rethorica et in a rte poetica unde florentinus ipsum rogavit ut ipsius gesta assumeret describenda claudianus autem se diffidens tante materie describende posse sufficere praesens opus aggressus est . . . ' With this compare the preambles in J 4 (' claudius claudianus florens tempore florentini impera­ toris . . . '), M 3 {' florentinus imperator romanorum rogavit magistrum claudianum ut gesta sua describeret . . . ' ) , R I (' florentinus imperator rogavit claudianum ut sua acta describeret . . . ') , and U (' cum imperator romanorum florentinus rogasset claudianum ut eius facta describeret . . . ') .

71

INTRODUCTION

wording and pnming, in the other MSS that offer an anno­ tated text. What little I know of Gaufridus' work does not suggest either that he was drawing on ancient sources of information or that editions of the D.R.P. will profit much from his exegetical labours, but the student of medieval education and scholarship will assuredly find much to interest him in this Gallic commentator on Claudian. 1 The six authors of the 'liber Catonianus ' also contributed excerpts to the twelfth-century Florilegium Gallicum, 2 a vast collection of prose and verse citations from both classical and medieval writers, and to other more or less closely related anthologies.J The ' flares ' from the D.R.P. that are fow1d in these anthologies are critically without importance, but from the contents of the most frequently quoted passages (rr.294-304, m.28-32, 197, 227, 29o-r) one gets a clear impression of the medieval preoccupation with sentiments of a religious or ethical nature. In the second quarter of the thirteenth century, Henri d' Andeli, author of ' La Bataille des Sept Ars ', laments that ' . . . la gent Gramaire perverse Rant lessie Claudien et Perse, .ij . molt bans livres anciens, Les meillors aus gramairiens.' 4 1 A possibly contemporary commentary on Statius' Achilleid has now been edited by P. M. Clogan in his The Medieval Achilleid of Statius, Leiden, 1968. 2 This survives in five MSS : Paris. lat. 7647 (saec. xii ) , Paris. lat. 17903 (saec. xiii), Escorial. Q.L 14 (saec. xiv) , Berolin. Diez. B. Santen 6o (saec. xiv) , and Atrebat. 6o (saec. xiv) . See Boas, op. cit. ; B. L. Ullman, ' Classical Authors in Medieval Florilegia , C.Ph. xxvii (1932), 1 ff. Birt, pp. clxxiii ff., gives references to the Claudian passages cited in Paris. lat. 7647 and opines that, as least as far as Claudian is concerned, Paris. lat. 17903 and the Escorial MS may be copied from 7647. 3 E.g. Paris. lat. 1 5 1 5 5, written at Orleans at the beginning of saec. xiii in the view of F. Novati, Me1anges Paul Fabre (1902), p. 267. 4 Verses 93-6. See the edition by L. J. Paetow, Berkeley, Calif., 1914, who translates as follows : ' . . . the grammarians perverse f Have for their part for'

72

HIST ORY O F THE TRANSMI S SI O N

This statement seems to need some qualification. Even the most ardent devotee of the ' Artes ' could hardly have avoided a brief confrontation with the D.R.P. at his elementary school, and it cannot be mere accident that whereas only seven twelfth­ century MSS have survived to us, there are no less than thirty­ five of the thirteenth century. In all likelihood, therefore, Henri is simply referring to the conspicuous neglect of the ' Auctores ', the poets, in, for example, Paris, and to their gradual neglect even in and around Orleans, their erstwhile stronghold. 1 That he is not thinking in terms of a universal drift away from the poets is shown by his later observation that the Muse of Poetry has fow1d sanctuary further afield among ' li Breton et li Alemant ' ;2 and such facts as we have tend to confirm that interest in the D.R.P. was now more lively in areas distant from the Loire valley and its immediate environs. Of the five thirteenth-century MSS whose place of origin is fairly certainly established, three are English (C 3 , 0 r and 0 2) , one is Italian (D), and the fifth has a connexion with S.W. Hungary (L2) . Surviving MS catalogues reveal that there were copies of the D.R.P. at Neumiinster bei Wiirzburg, at Rochester, and in the possession of Richard de Fournival of Amiens, 3 while unspecified Claudians (perhaps including the D.R.P.) are attested for Schlettstadt, Pegau, Rome, and the library of Alexander of Jumieges. 4 Citations from the D.R.P. saken Claudian and Persius, I Two very good old books, I The best belonging to the grammarians.' C£ also E. R. Curtius, op. cit. p. 56. 1 C£ La Bataille, vv. 19-20 : ' Logique a les clers en ses mains, I Et Gramaire rest mise au mains ', ' Logic has the students, I Whereas Grammar is reduced in numbers ' (Paetow's transl.). z Ibid. v. 446. It is disputed whether ' Breton ' refers to the inhabitants of Brittany or Britain. J Sec Arch. f Unterfranken, p. 253 (Neumiinster) ; Archaeologia Cantiana m. 54 ff. (Rochester) ; Delisle, Le cabinet des manuscrits n. 524 ff., tab . xi 122 (Richard). 4 See Die Stadtbibliothek zu Schlettstadt, pp. 4 f£ ; Serapeum Intell. 24.53 (Pegau) ; Arch.f Litt.- u. Kirchengesch. d . M.A. 1.36 (Rome) ; Cat. General des

73

INTRODU CTION

are foWld in the Historia Anglorum o f Matthew Parisi and the anonymous Flores Historiarum,2 both written at St. Albans ; Vincent of Beauvais quotes largely from the D.R.P. in his Speculum Doctrinale ;3 Albert of Stade (in Prussia) incorporates three verses from the D.R.P. in his poem Troilus;4 and Emo of Frisia cites four verses from book 1 of the D.R.P.s In the fourteenth century, Geoffrey Chaucer, the chief non­ Italian to show real familiarity with the D.R.P. at that time, commemorated ' Daun Claudian, the sothe to telle, That bar up al the fame of helle, Of Pluto, and of Proserpyne, That quene ys of the derke pyne' ;6 and Petrarch, for whom Chaucer had the greatest admiration, 7 was thoroughly conversant with Claudian's works. 8 There are citations from the D.R.P. in the works of Guglielmo da Pastrengo, who also has a brief biographical and literary note on the poet in his De originibus rerum. 9 One reason for the quite considerable popularity that Claudian seems to have enjoyed Manuscrits des Departements l.xx n. (Alexander). MSS of Claudianus maior are found at Glastonbury Qohn of Glastonbury, Chron. ed. Hearne n.442) and Canterbury (M. R. James, The Ancient Libraries of Canterbury and Do ver, p. 94) . 1 Vol. I.21 8 (Rolls Series), quotes D.R.P. I.23 5-6 (quae for sed, however). 2 Vol. I.597 (Rolls Series), quotes D.R.P. I.23 I-6. 3 Cf. Manitius, Philo/. XLIX.5 5 8 ; Birt, pp. clxxvii clxxviii . 4 D.R.P. I pf. 9-10= Troilus v1. 1 5-16 (illis however replacing ast ubi) ; D.R.P. n.201 = Troilus m.597· s D.R.P. 1.232-5 (nee in 234) : cf. M.G. ss. xxm.496. 6 House ofFame, vv . 1 509-12 ; cf. also v. 449· Chaucer refers to Proserpine's abduction from the ' mede ' of ' Ethna ' in the Merchant's Tale IV (E) 2227-3 3 , adding ' In Claudyan y e may the stories rede '. E. Rickert, Modern Philology XXIX (1932), 257 ff., thinks that Chaucer may have attended St Paul's Almonry School, where there was at least one ' liber Catonianus ' : see above, p. 70 n. 3 . On Chaucer's use of Claudian cf. R. A. Pratt, Speculum XXII (1947), 419-29. 7 See the ' Prologe of the Clerkes Tale ', vv . 26 ff. s Cf. B. L. Ullman, Studies in the Italian Renaissance, p. 1 3 0 ; Romano, op. cit. p. 1 50. The MS Parisinus lat. 8082 (my P 2) belonged to Petrarch. 9 Cf. Sabbadini, Scoperte, 1905 , p. 12. 74

HIST ORY O F THE TRAN SMI S SI O N

with the fourteenth-century humanists is perhaps to b e sought in their strange notion that he was one of them, so to speak, a citizen of Florence. 1 By virtue of this error he even gets a mention in Filippo Villani's Liber de civitatis Florentiae Jamosis civibus, written in I J 8 I-2,2 and when subsequently it was realised that Claudian was Egyptian by birth, Florence still contrived to uphold her claim to him by the timely bestowal of adoptive citizenship. 3 The comparative paucity of the fourteenth-century MSS of the D.R.P. that have survived or are mentioned in contem­ porary catalogues4 is more than made up for by the rapid proliferation of texts, mainly Italian, in the next century, of which more than sixty have been preserved. And not only was the D.R.P. now much read, it was also the subject of lecture courses, as for example at Ferrara, where Magister Lucha Ripa, ' honoratus grammaticus ', expounded it in public, with a good deal of false etymologising and copious use of synonymous glosses. s A puritanical interpreter, to all seeming, was Magister Ripa, for in his considered opinion the poet's intention was to impress upon parents the need to lock up their daughters !6 In face of the obviously considerable demand for texts of the D.R.P., it is hardly surprising that the assistance of the printing press was quickly invoked in reduplicating copies of it. MSS of 1

Gyraldus lays this misconception at Petrarch's door. See Birt, p. ii note.

z Ed. G. C. Galletti, Florence, 1 847. 3 See the epitaph on Claudian concocted by Coluccio Salutati : Birt,

p. ii note. 4 There are seventeen survivors. Dover in 1 3 89 had two copies (c£ M. R. James, op. cit. pp. 489, 494), Avignon one in I J I I (cf. Ehrle, op. cit. 1.91.555) and another, which had previously been at Limoges (?), in 1375 (ibid. 1 5 I I . 858). s Notes taken at his lectures are preserved in Brit. Mus. Harley 53 82. The anonymous student has copied up his notes only as far as 11.66 , with the addition ofjust a few lines on the interpolated third preface. 6 fo. 3 v : ' (Intentio) Publica ut parentes admoneret quod ( ?) custodirent filias suas diligentius quam fecerit ceres.' .

75

INTRODUCTION

course continued to be produced even after the editio princeps had appeared c. 1471, but by the early years of the sixteenth century the triumph of the printed text was complete and the vagaries of the MSS were :finally superseded by the emergence of a vulgate. I V. E D I T I O N S 1

By the year 1482, when the editio princeps of the whole of Claudian (except the carmina minora ) was printed at Vicenza by Jac. Dusensis, the D.R.P. had already passed through six editions, its own editio princeps being tentatively ascribed to the Venetian printer Valdarfer and a date c. 1471 ; and by the end of the incurrable period it had been edited thirteen times, four together with and nine apart from Claudianus maior.2 Biblio­ graphically, these very early editions are of considerable interest, but from the textual standpoint they are practically valueless. A large proportion of their readings, like those of many of the codices recentiores, to which they are obviously related, derive ultimately from the thirteenth/fourteenth century MSS D ] 4 K 1,l and their combined contributions to the restoration of the text amount to nothing more than the 1 A full list of editions up to 1759 was compiled for ]. M. Gesner's edition (pp. xv ff.) by one Nicolaus Niclas ; I discuss only the more important of these. Birt's chapter ' De editionibus nonnullis Claudiani ', pp. clxxxiii ff., supplies a good deal of miscellaneous information. z In approximate order of publication, and with square brackets denoting conjectural ascriptions, obeli editions of the whole of Claudian, and asterisks items that I have not myself been able to examine, these incunables are : [Valdarfer, Venice, c. 1471 ] ; [Ketelaer and Leempt, Utrecht, c. 1473/s ] ; [Schurener, Rome, c. 1475] ; [Ferrara, c. 1480] ; [Matthias von Olmiitz, Naples, c. 1480]* ; [Amdes, Perugia, c. 148 1 ] ; Celsanus, Vicenza, 1482 t ; Marcellinus Verardus, Rome, 1493 ; Ugoletus, Parma, 1493 t ; Ugoletus, Venice, I495 t ; [Kachelofen, Leipzig, c. 1495]* ; Ugoletus, Venice, r soot ; Parrhasius, Milan, 1500. All are described in the Leipzig Gesamtkatalog der Wiegendrucke, vol. VI (1934), nos. 7059-70. J On which see above, p. 6o.

EDITIONS

provision of the correct spelling in four places1 and the removal of a slight corruption in one other.z The edition of Janus Parrhasius3 in fact is the only one that merits prolonged scrutiny, and that primarily because it is the first to be equipped with a commentary. The main purpose of this very diffuse compilation4 is to illustrate the myth and elucidate points of diction, and apart from the occasional recording of variant readings, scant interest is shown in critical matters. Of unique importance and value is the edition of Michael Bentinus, printed at Basle in 1 534 by Michael Isengrin.s Like most other early editions it was founded on one printed exemplar6 and supplemented by material from other sources both printed and manuscript, but there was this fundamental difference, that one at least of the two MSS7 it drew on was in many respects far superior to any now surviving. For the most 1 See the app. crit. to

1.284 ; II.190 ; m.348 ; 390.

z See the app. crit. to m.347·

J To which Camers' edition, Vienna, 1 5 1 0, is largely indebted. Neither Camers however nor the Juntine and Aldine editions (Florence, 1 5 19, and Venice, 1 523, respectively) have anything worthwhile to add to Parrhasius. 4 It encloses the text on three sides and extends to 124 pages. s See Birt, pp. clxxxvi ff. Bentinus died before his work was done, and it was left to one Jo. Honterus to put the finishing touches to the edition. It should be noted that in this and in all subsequent editions that I discuss (except Jeep's ed. minor and the ed. of Paladini), the D.R.P. is accompanied by Claudianus maior. 6 In this case the ed. of Camers. This is proved (a) by the fact that both edd. exhibit Claudian's poems in the same unusual order ; (b) by the appearance in the Isengrin of carm. min. app. I, 2 , 20, 21, which were flrst published by Camers (the Juntine has none of these minor poems, the Aldine only nos. I and 2, with the note ' nos primi ex fragmento quodam uetustissimo in medium protulimus '); and (c) by the textual coincidences between them in the D.R.P. (the Isengrin's text diverges from Birt's c. 360 times ; of these divergent readings c. 300 are found also in Camers, but only twenty in either the Juntine or the Aldine, or both). 7 The number is established by these words from Isengrin's Epistola Nuncupatoria : ' (Bentinus) Vsus fuit exemplaribus ueneranda uetustate com­ mendabilibus : quorum alterum suppeditauit CAPITO.' On Capito see Birt, p. clxxxvii n. 2.

77

INTROD U CTION

part, though not invariably, the readings of these MSS were printed in the margin of the edition, I and from the pattern of their distribution one may with reasonable certainty infer that one of the MSS contained only Claudianus maior, the other only the D.R.P.2 The pre-eminent worth of the former of these two MSS, that one which does not immediately concern us here, is unassail­ able. Not only does it offer a considerable number of true readings such as no Scaliger even could have recovered by conjecture, it has also preserved several whole lines or parts of lines that are largely unknown to the extant MSS.3 But what of the second of Bentinus' MSS, which he utilised for the D.R.P. ? Is it of comparable excellence to the former ? Scaliger and Heinsius evidently thought so, for they accepted a number of readings from this MS, but Birt refused to allow it any value at all. The problem of this MS's trustworthiness is not in fact susceptible of an easy or defmitive solution, for no such un­ equivocal proofs of excellence as the preservation oflost verses are forthcoming in this case, and all we have to go on are a number of new readings of debatable worth, as follows : 1 In the dedicatory epistle Isengrin declares that Bentinus ' tolerabilem . . . utcumque lectionem marginibus assignatam lectoris superciliosi arbitrio reli­ quit integram '. The text itself, however, contains features foreign to Camers and beyond the reach of editorial conjecture, and we must therefore assume that Bentinus occasionally contravened this general principle. 2 In the forty-six pages taken up by the D.R.P. alone there are I 13 marginal singletons, yet in the 379 pages that make up the whole of Claudian there are only three cases of marginal doublets (fo. 46 nunc, vel num ; fo. 47 v mediis. quidam saevi legunt; fo. 49 V descendente, quidam difensata legunt = In Eutrop. 1. 195, 272, 3 86). Had the MSS overlapped, one would have expected more such doublets. There is no adequate reason for supposing, with Birt, that it was Capito's MS rather than the other that offered Claudianus maior. J (a ) Panegyr. de quart. cons. Hon. 3 I 5 (entered in the margins of Voss. 294 and Heinsius' Tholosanus and Petavianus primus, in all three cases by man. rec., and found also in the ed. princeps ; otherwise unknown) ; (b) ibid. 43 2 (other­ wise unknown) ; (c) ibid. 509 (added in the margin of the thirteenth-century Florentinus bibl. nat. VII. 144, probably by the first hand ; otherwise unknown) ; (d) ibid. 636-7 (the supplement jluctus / A/pinos genitor rupit te consule is not

78

EDITIONS

A. Readings of the texti

concussa.* Probably right. culmina.* Possibly right. Accepted by Scaliger and Heinsius. 24 vorticibus. As good as gurgitibus {MSS) and found as a gloss on that word in Parrhasius' ed. Accepted by Scaliger and Heinsius. 66 incestas. Wrong. 92 celeres. Elegant, and accepted by Heinsius. But the speed of the messenger is what requires emphasis, hence my preference for i celer et {MSS). 98 hominum.* Wrong. Presumably intended to clarify aures. II pf. 5 leonum.* Just possible. II . 3 5 6 nullaeque. Accepted by Scaliger and Heinsius, and by no means impossible. m. 75 olim quae. Accepted by Scaliger, but probably suggested by v. 84. 409 Jastaeque. Wrong. I.

4

8

B. Readings in the margin I

exultat. * Very attractive. Accepted by Heinsius. moveri.* Wrong. I63 vomit.* Right. I 86 leves. Wrong. I 8 9 culmi.* Accepted by Heinsius, and probably right. 223 Jalle. Wrong. 226 corpus. Very poor.

pf. I I

I. I 5 I

fotmd before the Isengrin) ; and (e) Panegyr. dictus Prob. et Olybr. coss. 201-4 incl. (not fotmd before the Isengrin) . The authenticity of (a) , (b) and (c) has never been questioned, for these verses aptly fill obvious lactmae ; that of (d) and (e) should never have been questioned, in view of the palaeographical and stylistic considerations that confirm the former (i.e. no interpolator would have repeated te consule or proceeded on the assumption of a ' saut du meme au meme '), and the fact that the latter passage is (i) marred by some minor corruptions which at least rule out the possibility of recent, humanist forgery, and (ii) in any case too much in Claudian's manner to be anything but genuine. 1 An asterisk signifies that the reading in question is discussed in the Commentary.

79

INTRODUCTION

232 256 257

n pf. 14

n.

43 13 83 84 99

canduit. * Favoured by Heinsius, and possibly right. curvantia. Attractive, and possibly right. alnos. Wrong. reparat. Preferred by Heinsius, and possibly right. Cf. reparit 0 1 (a.c. ) . maduit. Inelegant. domitura. Possibly right. Sabaeis.* Accepted by Heinsius, rightly. busto.* Also accepted by Heinsius, rightly. picto. Attractive, but inspired likely enough by pingit in v. 93 ·

165 171 290 m. I I

76

13 5 152 248 262 267 278 297 3 59 425

elusos.* Printed by Scaliger and Heinsius, rightly. solitaque.* A pointer to the true reading. Averni. Q!!ite possible. From it Heinsius conjectured A vernis. et caerula.* Accepted by Heinsius, but not really satisfactory here. stirpe recisam. Attractive. Read by Scaliger and Heinsius. sumat. Not as forceful as mittat (MSS). ex. Barely possible. scrutamur. Wrong. in. Accepted by Heinsius, and possibly right. inhiatque. Syntactically impossible to fit in. cudit. Nonsensical here. deiectaque.* Accepted by Heinsius, and probably right. seu. Makes no sense here. onerata. Wrong.

The forty readings listed above were rejected en bloc by Birt (p. cxciii) as the products of ' librariorum noviciorum ac sciolorum securitas, levitas, procacitas ', partly because in his view the plausibility of some of them was neutralised by the implausibility of the others, partly also, one feels, because his partiality for the MSS on which his own text was based pre­ disposed him against believing that the truth might be found elsewhere. But the fact that bad readings coexist with good is,

8o

EDITI ONS

i f anything, an argument for the Isengrin's essential honesty­ as for the MSS, they are all without exception a mixture of good and bad-and when I find in the vett. a number of readings which Birt had supposed to be peculiar to the Isengrin, 1 I cannot but feel that those listed above should also be tested by the canons applied to the readings of the MSS, and be accepted or rejected on their merits alone. My evaluation of the Isengrin' s new readings comes close to that of Heinsius, whose enthusiastic approval of that edition's contribution to the text stands in marked contrast to Birt' s sweeping disparagement of it. Slightly over half the readings may in my opinion be dismissed at once ; of the remainder seven seem to me certainly or probably true (1.4, 163 , 1 89, n.83, 84, 165, m.297) and nine possibly true (r pf. I I , r.8, 24, 232, 256, n p£ 14, n. r 3 , III.76, 262 ) , and there is one other that seems to point to the truth (II. I7I). If this assessment of the Isengrin is even approximately correct, and it is unlikely that any assessment will ever meet with universal approval, it must be conceded that Bentinus' MS of the D.R.P. was rather more valuable than any now surviving, for while the truth does indeed survive in single MSS on a number of occasions, not one of them preserves it in isolation as often as this lost exemplar. All subsequent editions until the time of Heinsius were dominated directly or indirectly by the Isengrin, and very little progress was made in the criticism of the D.R.P. until the influence of the vulgate which had been handed down from D J 4 K I via the Italian MSS and editions of the fifteenth century to Parrhasius, Camers and the Isengrin was replaced by the weightier and more beneficial influence of the ' sospitator 1 See Birt, p. cxciii. The following should now be removed from his list of ' propriae scripturae ' : I. I S I littora (in T) ; 243 jornace (in F4) ; II pf. I4 reparat (0 I a.c. has reparit) ; m.2 I 3 vetito (in C 2 R2) ; 234 institit altior (in C 2) ; 262 multum (v.l. in U). The phrase patriis procul amandaverit (m.214) is in b I P 4.

6

8!

HCD

INTRODUCTION

poetarum Latinorum ' A handful of new readings, none how­ ever of any particular excellence, were contributed by Gyraldus, Livineius, 1 Pulmannus2 and Delrio,3 and a rather larger number of readings, some of which are almost certainly his own conjectures, were inscribed by the jurist Jacob Cujas in a copy of the lsengrin which is still extant. 4 These Excerpta Gudiana, so called because the lsengrin later came into the possession of Marquard Gude, s are however of disappointingly poor quality, as a cursory glance at them is sufficient to establish. 6 The edition produced by Cujas' pupil Stephanus Claverius7 is problematical to a degree because Claverius hardly ever refers to his various sources in any more precise terms than ' vetus liber ' , ' vetus codex', 'vetus scriptura ', and so forth. .

1 MS annotations in the hands of Gyraldus and Livineius are found in a copy of the Aldine now in Leiden University Library (cat. no. 757 G2). C£ Birt, pp. lxxxv ff. and clviii. Of great importance for Claudianus maior, these annotations arc of negligible value in the D.R.P. I cite from them only three times, at 1.52, 1 8 1 , m.441. 2 In his Plantin edition, Antwerp, 1571. See the app. crit. to II p£ 14, II.321, 3 56, m. 1 3 8 , 3 84. In the margin of this edition are printed readings taken from my A I , A2, L 10, and one other (lost) MS, and also from Parrhasius, the Isengrin and other printed sources. J In his notes on Claudian, Antwerp, 1 572. See my app. crit. to II pf. 3 , 18, n.6, 57, m.97. Delrio consulted my L 10 and one other MS now lost. 4 In the Gottingen University Library, cod. MS philo/. 163 , formerly auct. class. lat. 225 3 . The text of ClaudiatlUS maior is annotated with a collation of Ambrosianus lat. M9 sup. A. C£ ]. Koch's pamphlet ' De codicibus Cuiacianis quibus in edendo Claudiano Claverius usus est ', Marburg, 1 8 89, pp. 8 ff. s From Gude it passed successively to J. A. Fabricius, a Town Councillor of Wolfenbiittel by the name of Weichmann, and J. M. Gesner, who discussed it in his edition, pp. xxviii ff. 6 Lections otherwise unknown are : 1.13 addicta ; 21 animis; 26 vulnere ; 27 laesit ; 3 3 solu' careret; 58 reponis vel resolvis ; 74 occlusit; 83 maerorem ; 125 laetior; 1 3 2 fremunt . . . vota ; 146 conjungi ; 149 projectis, porrectis ; 171 pro­ cellas; 172}luat; 196 raros ; 198 actu ; 224 infera ; 249 hie signabat; II pf. 24 vocibu' capta ; 47 quievit; n. 1 3 laesura ; 36 gaudia ; 40 nectuntur; 78-9 flatu / pubcscant virgulta tuo, ut sic ; 84 renovans ; 85 in terras ; 125 carptura ; 16 1 sulcis serpentibus impedit actum. These and other readings are mentioned in Gesner's apparatus. 7 Paris, 1602.

82

EDITIONS

Were these sources of conspicuous value, such terminological vagueness would be regrettable, but the quality of the new material found in Claverius is in fact so indifferent that it matters little that its provenance is unknown. 1 Readings first fotmd in Claverius are as follows :2 1 The problem o f Claverius' sources was investigated b y Koch in his Marburg pamphlet, the conclusions of which are resumed and modified in Birt's praef. pp. cxcvi-cxcviii. Koch rightly concludes that Claverius' chief debts were (in the D.R.P.) to the Isengrin and (in Claudianus maior) to the Isengrin and Cujas' collation of the thirteenth-century Ambrosianus lat. M9 sup. A, but his thesis concerning Ambros. needs some qualification. Clav. states that he consulted ' duo Claudiani exemplaria, antiqua manu exarata ' belonging to Cujas, together with a copy of the Gryphian ed. of 1 5 3 5 etc. (I agree with Koch that this is what Claverius means by ' codex Gryphianus '. C£ his use of the term ' codex excusus ' on fo. 24 v in his note on D.R.P. 11.363) in which Cujas had copied variants ' aliunde '. From this word ' aliunde ' I infer that the variants did not come from one of the ' two ancient exemplars ', and therefore reject Koch's suggestion that Ambros., which is patently the source of the variants, could at one time have belonged to Cujas and been consulted directly by Clav. ((a) The entry on the involucrum of Ambros. to the effect that ' hunc codicem una cum multis aliis Avenione vehendum curavimus ' was not made by Cujas, as Koch thought likely, but by the first librarian of the Ambrosian, Antonius Olgiatus. C£ tl1e similar entry on the fly-leaf of Ambros. B 71 , a MS containing grammatical texts : ' Hunc codicem . . . Avenione vehendum curavimus . . . Felicibus auspiciis Mmi Card. Federici Borrhomaei . . . Biblio­ thecae nee non scholae Ambrosianae fundatoris Antonius Olgiatus eiusdem primus Bibliothecarius scripsit a. 1605.' The phrase ' Avenione vehendum curavimus ' occurs also in two MSS of Ovid's Metamorphoses, viz. Ambros. N 254 sup. and 0 3 sup. In the light of these entries Birt's note about the provenance of Ambrosian MSS, p. xcvii n. 2, must be corrected. (b) There is little cogency in Koch's contention that whereas those Ambrosian readings in Claverius' ed. that are accompanied by such comments as ' ut . . . notavit Cuiacius ' must come from the annotated Gryphian, the absence of such reference to Cujas can only be explained on the assumption that Clav. is now drawing not on Cujas' collation of Ambros., but on the MS itself). In my view Ambros. was collated by or for Cujas at Avignon and it was this collation that Clav. had at his disposal, not the MS itsel£ Though the Ambrosian variants were copied into the Gottingen Isengrin (see above, p. 82 n. 4) which still surves as well as the now lost ' codex Gryphianus ', Clav. patently never saw the former since it is inconceivable that he could have omitted to cite from the Excerpta Gudiana (above, p. 82 n. 6) had he known them. 2 An asterisk signifies that the reading in question is discussed in the Commentary.

6-2

INTRODU CTION

A . In the text 30 leges (populis) . Accepted by Heinsius, who also expressed himself dissatisfied with et immediately following. But et is unexceptionable if vv. 3o-1 are regarded as theme and variation, and while the notion of Ceres Jrugifera is a commonplace in the poem, that of Ceres legifera appears only here. I consequently reject leges as a learned interpolation. 70 minaci. Wrong. 72 bella (cupit).* Accepted by both Scaliger and Heinsius, but probably conjectural. I I6 Julgidus.* Accepted by Scaliger and Heinsius. I I9 diverso. Wrong. 141 Sicula . . . in ora. Wrong. II pf. 2 seposuisset. Possible. Accepted by Heinsius. 7 eduri. Wrong. II. I I ca llida . * Wrong in my opinion, but accepted by Heinsius. 295 salis. Accepted by Heinsius and not impossible, though Claudian nowhere else employs sal as a synonym of mare. m.23 7 aestus.* Accepted by Scaliger and Heinsius, misguidedly as it seems to me. 23 8 pervenit. A trivialisation. 267 iamiamque.* Right, but probably conjectural. Read by both Scaliger and Heinsius. 44 3 aera. Wrong case, but aere would be possible. Scaliger conjectured ( ?) aethere, whlch was accepted by Heinsius. I.

B. In the notes I.

Not in the Isengrin, as Koch erroneously states apud Birt, p. cxcviii 6.

1 n.

27 ulsit (' vetus liber '),* whence Claverius conjectured ussit, unnecessarily. 3 5 nudasque maritae (maritae in C 1 ) .* Wrong. 1 49 despicit. Probably a correction of respicit (recc.) . 265 penetralia. 1 Much more obvious than sacraria.

EDITIONS m . 1 40

195 3 12

332

liquerit. Feeble. ' hie versus aberat a vet. codice, an oblivione librarij ? ' The answer i s ' Yes '. ' hunc versum non habuit vet. liber.'* The verse was excised by Heinsius, who commented : ' cum Claverius testetur in optima (the epithet is Heinsius' own) co dice non comparuisse, per me licet pro spurio habeatur hie versus.'. At one time I was inclined to agree, but now I consider the verse genuine. Jlavum.* Right, but probably conjectural. Accepted by Heinsius.

There are in this list a number of interesting and even plausible suggestions (e.g. at 1.72, II p£ 2), but I find nothing to suggest that Claverius had access to a MS or MSS of great value. The two true readings that we owe to his edition (m.267 iamiamque and 332 jlavum) 'might be instances of precarious survival like those one finds in many an extant MS, but neither is such that a man of Cujas' learning could not have retrieved it by conjecture. I In the year following the publication of Claverius' edition, an infinitely more gifted pupil of Cujas turned his hand to Claudian. Although the minute edition published by the Plantin establishment at Leiden2 bears no indication of author­ ship, only the vague phrase ' Ex emendatione virorum doc1 Burman II remarks, pre£ to ed. vi ff. , that ' Cujacianas conjecturas pro librorum veterum Mss. lectionibus Claverium passim adripuisse & lectoribus obtrusisse monet saepe Heinsius '. C£ Heinsius' own words, reported by Burman in his note on carm. min. app. 2 . 8 1 . W. Schmid, ' Ein verschollener Kodex des Cuias und seine Bedeutung fiir die Claudiankritik', SIFC xxvii­ xxviii (1956), 498 ff., esp. 505 and n. 2, argues against the idea that Cujas Inight have slipped his own conjectures in amongst MS readings ' um ihnen zu sichererer Geltung zu verhelfen ', but it need not be a question of Cujas' honesty or dishonesty, only of Claverius' failure to distinguish between conj ecture and collation. z Directed by the brothers Raphelengius.

ss

INTRODUCTION

torum', its author beyond all doubt was ] . ] . Scaliger. 1 The most noteworthy and welcome feature of this edition is the appearance in it of a considerable number of Scaliger's con­ jectures, but these are conftned almost exclusively to Claudianus maior, and in the D.R.P. I ftnd only three, all alike unnecessary. 2 As far as the D.R.P. is concerned, therefore, this edition is something of a disappointment, the more so as Scaliger had no recourse to any MS but merely worked over Claverius' edition and the Isengrin vulgate as represented in Pulmannus' anno­ tated text. ' Omnium quotquot vidi optimam ' is how Heinsius described this Raphelengian edition, but its excellence in the D.R.P. is of a negative kind, not to be compared to that of Heinsius' editions. The two editions of Caspar von Barth do little to improve the text of the D.R.P.3 The ftrst edition is nothing more than a reprint of Pulmannus' third edition of 1 596, with some emendations noted in the commentary ; the second is a reprint of the Raphelengian edition, with variant readings taken from R 2 5 (' Codex Manuscriptus Bibliothecae tum Palatinae'), a MS now lost which was purchased in Switzerland, and a copy 1 Barth. Advers. LVIII. 7 , commenting on Panegyr. dictus Prob. et Olybr. 201-4, notes that ' Scaliger, qui Claverianos codices alioquin pressius sequitur, in Raphelengiana minuta editione abs se adornata et emaculata nee tales illos editioni suae indere voluit, sed ut vulgati sunt fecit excudi asterisco tamen in fronte notatos, ne quis exactissimo eius iudicio satisfecisse arbitraretur ', and in his edition of 1650 he remarks : ' Inter editiones laudem praecipuam meretur quae ex Cuiacianis libris et observationibus a St. Claverio Parisiis edita in satis ampla et splendida, paullo post a Io. Scaligero in minutissimam formam solo tamen ut vocamus textu redacta et a Plantinianis est saepius recusa.' Cf. also J. Bernays, Rh. Mus. XV. 163-5. 2 At rr.2, 69, III - 443 · Additional emendations, including a fme correction of III - 359 (see also 1.65, II pf. 39, m. 78-9), were jotted in his copy of the 1 596 reprint ofPulmannus' ed. , which was subsequently consulted by Heinsius and Burman II. Credit for the rediscovery of this volume goes to Jeep : see his ed. maior, vol. II.xxiv-xxvi. 3 The true reading ilia (now found in MSS) is restored at 1.67 ; unconvincing emendations are proposed at II pf. 32 and m.3 5 9 ; and one not implausible reading is adduced from Barth's own MS at 1.68.

86

EDITIONS

o f the Isengrin edition containing Pulmannus' collations of fifteen MSS, among them A r , A2 and L ro. Barth's second edition was followed, later in the same year (r6 s o) , by Heinsius' first edition, publication of which had been held up by a dilatory printer. Malicious gossip was not slow to make capital out of this delay, but the thirty-year-old diplomat's earliest editorial undertaking is so vastly superior in quality to anything that had preceded it that the charge of plagiarism that apparently was levelled against it is quite preposterous. In this epoch-making edition, published by the Elzevir Press in Leiden, 1 Heinsius first revealed those qualities that were to win him the respectful admiration of other scholars : an assiduity in the examination of MSS that was not equalled until the last century ; an unrivalled understanding of Latin poetry, in particular of the imperial period ; and a rare genius for emendation that was refined and sharpened by his literary sensitivity and his familiarity with the ways of scribes. Because of these natural and acquired endowments he was uniquely equipped for the task of editing Claudian. His travels in England, France and Italy prior to r6 s o gave him the opportunity of examining no less than twenty-eight MSS of Claudian, sixteen of them containing the D.R.P. ; a second visit to Italy and the kindness of friends augmented the second edition of r6652 with the readings of a further eleven MSS, six of them offering the D.R.P. ; and a few other MSS are mentioned in the younger Burman's edition, which incorpo­ rates material Heinsius had collected for a projected third edition. Data supplied by the two published editions and the transcript of Heinsius' collations produced for Burman by one 1 In Heinsius' absence the publication was supervised and the proofs corrected by his friend ]. F. Gronovius, whose own annotated copy of Barth's first edition is now in the Rijksuniversiteit Library in Leiden. 2 Printed at Amsterdam, also by the Elzevir Press, with ' Notae Variorum ' negligently compiled by one Cornelius Schrevel.

INTR O D U CTION

Ezra de Clerq van ]ever, 1 make it possible to identify most of these MSS.2 At the time when Heinsius began work on the D.R.P., there was still some scope for the exercise of skill in divination, but much more remained to be achieved by investigating as fully as circumstances permitted the almost completely untapped resources of the MS tradition. Consequently, although Heinsius 1 See Burman's edition, pracf. pp. xxv-xxvii, and, for the actual ' Sylloge variantium lectionum . . . excerptarum e . . . codicibus . . . quos contulit Nicolaus Heinsius ', pp. 739-1026, of which 936-69 cover the D.R.P. The ' Sylloge ' is based on the five editions that Heinsius used for recording his collations, and the ' Schedae Heinsianae ', a huge notebook of over 6oo pages containing yet more collations together with copious notes on the whole of Claudian and rough drafts of the published commentary. The ' Schedae ' are now in the Bibliotheque Nationale in Paris (MS lat. nouv. acq. 5 82) ; of the editions annotated by Heinsius three seem to have disappeared, but Theobaldus Paganus' Lyon edition of 1 5 5 1 (Burman, pp. xxv-xxvi) is now in the Rijks­ universiteit Library in Leiden, shelfmark 757 G 3 , and the Elzevir of 1650 is in the Universiteits Bibliotheek in Amsterdam, shelfmark 6o8 G 22 (communica­ tion of the Keeper of Western MSS, Bibliotheek der Rijksuniversiteit te Leiden). 2 The conclusions reached by Birt, praef pp. cxxvi-cxxviii, are in general correct, but there are some errors to be rectified and several gaps to be filled. Also, Heinsius occasionally uses one and the same label to denote different MSS in different parts of Claudian's works, a fact overlooked here and there by Birt. My identifications of MSS used for the D.R.P. are as follows : ' Vaticanus primus ' = R 26 ; ' Vaticanus alter ' = R 5 ; ' Vaticanus tertius ' = R 9 ; ' Vaticanus quartus ' = R 6 , and sometimes R 4 ; ' Vaticanus quintus ' = R I 3 ; ' Bononiensis ' = b I ; ' Ambrosianus tertius ' = K 4 ; ' Mediceus primus ' = F 3 ; ' Mediceus quartus ' = F 8 or F I 3 ; ' Florentinus S. Marci ' = F 2 ; ' Palatinus ' = R23 ; ' Puteaneus ' = P 4 ; ' Oxoniensis primus ' = 0 2 ; ' Oxoniensis secundus ' = 0 3 ; ' Regius ' = P 2 ; ' Moreti primus ' = A 2 ; ' Moreti secundus ' = A I ; ' Peta­ vianus primus ' =J 3 ; ' Petavianus alter ' = L I ; ' Petavianus tertius ' = R 3 ; ' Petavianus quartus ' = R 2 ; ' Vossianus primus ' = R 2o ; ' Vossianus alter ' = R 7 ; ' Tholosanus ' = P I ; ' Heinsii membranaceus ' = R I9 ; ' Heinsii chartaceus ' = 0 6 ; ' Incertus Heinsii ' = F I4 ; 'Junii secundus ' = L s . The Palatinus used by Barth (R 25) was examined but not collated. Three MSS I have not been able to identify : 'Junii primus ', ' Bartholinianus ', and ' Heinsii tertius litera H notatus ' ; these are presumably now lost. Also consulted by Heinsius were the ' Excerpta Laeti ' and ' Excerpta Schottiana ', the source of neither of which is known, the ' Excerpta Gyraldina ', now in Leiden University Library (see above, p. 82 n. I ) , and the collations made by Pulrnannus (see above, p. 82), including, as Heinsius observed, a very careless one of A2.

88

EDITIONS

has done more than any other scholar for the conjectural restoration of this text, 1 by far his greatest service lay in un­ earthing the readings of at least twenty of the vetustiores, whereby he was able to expose the specious interpolations of the more recent vulgate. In the nineteenth century the critical principles of Heinsius' editions came under heavy fire from Jeep, who, ' auf dem sichern Boden methodischer Kritik stehend ', in his own phrase, charged the older editors, Heinsius included, with having operated ' mit ungemeiner Willkiir und Ungleichheit '.2 In the absence of even one demonstrably reliable and trustworthy MS, however, it is difficult to see how any worthwhile edition can be other than ' capricious ', if indeed caprice consists in seeking out the best readings according to one's lights ; and where there is no ' sure ground ' for the critic to base himself on, it is often impossible to imagine how else he may discriminate between plausible variants except by having recourse to such Heinsian criteria as ' eleganter', ' quod praestat ', ' probe' , ' quod placet ', and so forth. The authority of Heinsius' scholarship and the evident superiority of his work to anything that had gone before elevated his recension to the status of a second vulgate, hardly to be modified or questioned3 during the next 200 years. The 1 Of the fifty and more conjectures he proposed, many appear to be improvements rather than corrections, while not a few are indeed corrections, but of secondary corruptions peculiar to single MSS or small groups of MSS : see the apparatus to 1.8, 21, 30, 6 1, 96, 99 , 1o6 , I I I , I I 9, 121, 146, 1 5 7, 250, 256, II p £ 32, 3 9, II. I I , 23 , 64, 84, 87, 1 3 9 , 163 , 165, 175, 201 , 2 1 8 , 262, 267, 290, 306 (bis), 3 16, 3 37, 342, 367, m. 1 5 , 3 0, 37, so, 5 1, So, 82, 105 , 162, 177, 197, 217, 241 , 255, 273 , 301 , 3 1 2, 347, 418 , 421 . Hardly anything is suggested which could not have been written by Claudian, but of Heinsius' conjectures only three look as if they must have been written by him : 11.171 solida, m.137 2 Acta, pp. 347 , 3 87. sat, 265 premit. 3 Except in the portentous commentary of N. Biffius, Milan, 1684 , which runs to 655 folio pages each with twin columns of minute print. For vacuity coupled with verbosity it would be hard to find the equal of this production.

INTRO D U CTION

edition o f J. M . Gesner, 1 whose text i s essentially that of Heinsius' first edition, aims primarily at providing a candid and straightforward explanation of the author's meaning, with comments designed to help the younger reader to appreciate the qualities of the poem ; 2 that of Burman secundus, 3 likewise founded on Heinsius, sets out to sum up the results of previous investigation, both by offering a more ample ' variorum ' selection than had previously been available, and in particular by making known the contents of Heinsius' unpublished papers and those of the elder Burman, his uncle. These two complementary editions effectively bring to a close the ' aetas Heinsiana ' and usher in a century of comparative inactivity, as far as work on Claudian is concerned. 4 The resumption of critical work on the D.R.P. in the 1 870s was attended by consequences that were little short ofdisastrous, for the two editions of L. Jeep,s far from improving on the Heinsian text, as their author believed they had, in fact im­ ported more falsehood than they removed. Two main causes of this setback may be distinguished : the erroneous assumption that the MS tradition must respond to the newly evolved methods of stemmatic analysis, 6 which resulted in a misplaced confidence in F I and, to a lesser degree, J 3 ; 7 and a recklessness in both formulating and adopting conjectural ' emendations ' ' Leipzig, 1759. Readings are reported from the ' Excerpta Gudiana ' and the Wolfenbiittel MS (W), and two conjectures are proposed, at m. 16o and 361. 2 See his Prolegomena, p. xiv. 3 Amsterdam, 1760. The lion's share of the work had already been executed by the elder Burman. 4 One must however mention the literary studies of B. G. Walch, pub­ lished as a Gottingen doctoral dissertation in 1770, and the elegant ifsuperfluous conjectures of ]. Schrader (see the apparatus to 1.258, II pf. 7, 23, m.263 ) , pub lished by M. Haupt in Hermes v (1871), 326. s Turin, 1874 (ed. minor), and Leipzig, 1 879 (in vol. II of the ed. maior). 6 Full discussion in chapter II. 7 F 1 is preferred to J 3 about five times as often as J 3 to F I . 90

EDITI ONS

that fully equalled the worst excesses of Baehrens. This un­ restrained propensity to divination was doubtless accentuated, if it was not engendered, by the conviction that where F I and J 3 were unsatisfactory, it was proper to resort, not to the other MSS, but either to the edd. vett., which were regarded as so many repositories of conjecture, I or to what modern critics might excogitate.2 In consequence Jeep's texts are loaded with the specious interpolations of F I , which were mistaken for the truth, and with a glut of conjectures which would never have been required had he been prepared to admit the testimony of the forty-eight MSS eliminated in his preliminary analysis. Yet other conjectural alterations, not a few of which were contri­ buted by the reviewers of the ed. minor, Baehrens in particular, 3 seem prompted only by that ' corrigendi & tentandi omnia etiam sanissima prurigo ' which Burman secundus had castigated in his contemporaries. 4 Such superiority as the now standard recension of Theodor BirtS may claim over those of his immediate predecessor is due very largely to his utilisation of a larger and better selection of MSS than Jeep had drawn on. 6 His excessive adherence to the testimony of his six chosen witnesses, however, and his all but constant rejection of readings found elsewhere7 in fact result frequently in a text that is either poor or positively indefen1 These editions, as is now understood, draw mainly on MS sources, and the readings they offer are as likely to be transmitted as those of F I and J 3 · z ' Durch diesen Apparat (i.e. utilisation o f F I and J 3 and very occasionally, in the ed. minor only, W) wird bei Anwendung der nothigen Conjectural­ kritik der ganze Ballast der iibrigen MSS. als iiberfiiissig beseitigt ' (Acta, p. 3 87)3 Not one of Baehrens' suggestions merits serious consideration, though his reconstruction of 1. 139ff. is most ingenious. 4 Praej. p. xix. s Berlin, 1 892. 6 See further p. 47 · ' In keeping with this policy is the fact that comparatively few conjectures are reported in the apparatus, and only one, his own, figures in the text : n.275 deterget. Only two <{ readings are knowingly admitted into the text : m. 17 Naides and 236 pulsu strepituque.

91

INTR O D U CTION

sible. 1 Nor is his critical judgement always to be trusted in cases where the MSS on which his text is founded differ among themselves.2 A less serious but more insidious deficiency of Birt's edition is the poor quality ofhis apparatus criticus, which is not only replete with useless information about totally un­ important corruptions and corrections in the MSS collated, but also riddled with errors of fact : leaving aside mere trivialities, I count over ninety false attributions of readings and other mis-statements, and over fifty cases of omission or misleading implication. 3 Of the editions that have appeared since r 892, little need be said. The Teubner edition produced by Birt' s assistant, J. Koch, 4 diverges from his text and punctuation in twenty-two places, s slightly more often for the better than for the worse, but other­ wise is an exact replica of the larger work, to which it stands avowedly in the relation of an editio minor. M. Platnauer's Loeb edition6 follows Birt no less closely, differing from him, generally for the better, in eighteen places,7 where, in Plat­ nauer's words, his ' conservatism commits him (in my opinion) to untranslatable readings '. The most recent edition, by V. Paladini, 8 has the more ambitious aim of providing a critical revision of the text on the basis of both Birt and Jeep. In this, however, it fails signally, for the general principles on which 1 One may instance the following : 1.67 v ix ille ; 163 movet ; 165 motibus ; rr.165 inclusos ; m. 44 addita ; 1 3 7 sed ; 262 vultu ; 265 advexit. 2 E.g., at 1. 143 magna is preferred to una and aestus to aetas ; rr.62 ademptis read, not aduncis; 132 gramina, not germina ; III. 375 astant, not extant. 3 These errors are tacitly corrected in my own apparatus criticus. 4 Leipzig, 1 893 . It is greatly to be regretted that Koch did not see fit to provide an abbreviated version of Birt's apparatus criticus. s Details of the changes are given in the ' Adnotationes Criticae ', pp. liv-lix . A few readings excerpted from Egerton 2627 (my L 1 ) are mentioned on 6 London, 1922 ; reprinted 1956. With English translation. pp. v-vi. 7 At 1.67, 1 4 3 , 195, 244, 250, rr p£ 42, n.62, So, 84, 1 32, 256, 3 06, m.4 4, 1 5 6, 177, 249, 393 , 443 · On one or two occasions Birt's reading is printed but another translated : c£ the text and translation of 1.72, 243 , rr.148, III .220. 8 Rome, 1952.

92

EDITI O N S

the text is constituted are invalid, 1 the text itself, where it diverges from Birt' s, is twice as often wrong as right, and the apparatus, abbreviated from Birt's, is inadequate and untrust­ worthy.z V. D A T E A N D C I R C U M S T A N C E S OF C OMP O SITION

Whereas the panegyrics and invectives can all be more or less precisely dated because of the references in them to historical events, 3 the lack of such references in the three books of the D.R.P., coupled with the disputed interpretation of its two pre­ faces, has given rise to a number of substantially different dating theories, no one of which has met with universal approval. The evidence which may be admitted in support of any dating theory is severely limited ; more so in fact than those scholars allow who associate the D.R.P. chronologically with other poems on the ground of thematic or verbal similarities,4 or argue from the premise that Claudian could not have written on mythological subjects during the period 395-404, when he was closely involved with the senate and the court. s It is not , 1 One cannot treat Birt's three classes as though they were branches of a tripartite stemma, as Paladini erroneously imagines, p. 9· 2 For example, conjectures appear in the text without any indication that they are conjectures ; critically important dissensions among Birt's six MSS are ignored ; minority readings are printed against Birt without mention of the majority readings preferred by Birt ; many plausible readings are omitted ; and the terminology used in referring to MSS is grossly misleading (Birt's six MSS are emphatically not ' codices antiqui ' [app. to II.289], nor are all his <{ MSS ' recentiores '). 3 See Birt, praef. pp. i ff. , ' De Claudiani vita et scriptis et temporum historia '. 4 For example, D. Romano, Claudiano, pp. 29-30. s Romano, ibid. p. 31 and passim. P. Fabbri, ' Claudiano in Sicilia e il ratto di Proserpina', Raccolta di Scritti in Onore di Felice Ramorino, Milan, 1927, pp. 91-100, anticipates Romano in this respect. It may here be remarked that Fabbri's thesis that the D.R.P. was occasioned by a visit to Sicily at the time of the poet's marriage is absolutely subjective.

93

INTR O D U CTION

for example, legitimate to conclude, as do Fargues and Romano, that because the D.R.P. has features in common with the Gigantomachia, therefore the former work was written or begun shortly after the latter. 1 Not that there is any intrinsic objection to this conclusion (which may or may not be correct) , only the data on which it is based in fact prove nothing more than that after an undefmed period of time (whether shorter or longer remains to be determined), the promptings of his memory or the re-reading of whichever work came first suggested to the poet tl1e idea of varying or repeating earlier phrases or conceits. Similarly, there is nothing inherently improbable in Romano's contention that there was a ' svolgimento della poesia claudianea ' in the sense that historically based composition was preceded by a ' momenta mitico ', 2 but dating on objective grounds must come before any theory of literary evolution. Clearly, these and similar approaches to the problem of dating the D.R.P. must be abandoned, and that once done, the two prefaces alone remain as sources of evidence-.if indeed the preface to book II is an integral part of the D.R.P. and not, as some have main­ tained, a misplaced intruder. Since the thirteenth century there has been a good deal of puzzlement as to the relevance of the second preface to the book that follows it and to the epic as a whole. The view expressed by the annotators of 0 2 P 2 and R 26, that this preface was out of place in the D.R.P. ,3 was cautiously reaffirmed by Heinsius4 and Gesner,s while more recently still L. Jeep took the final step of removing the preface from the epic altogetl1er. 6 1

P. Fargues, Claudien, p. 18 and n. 1 ; Romano, ibid. pp. 29-30.

2 Op. cit. pp. 6 and 30. J

A transcript of these annotations is given in the Commentary. ' Praefatio haec alieno loco inserta videtur. ad rem certe non facit. u t tollenda fortasse sit.' s ' Nihil habet haec praefatio, quod cum argumento libri conjunctum sit.' 6 In this he met with the approval of Fabbri, op. cit. p. 97· 4

94

D ATE AND CIRCUMSTANCES O F COMPO SITION

As s o often, Heinsius confronts u s with just his conclusions and nowhere goes into detail as to why he thought the preface not ' ad rem '. His reasoning can, however, be reconstructed without much difficulty. Like Claverius and Barth before him, and all subsequent editors with the solitary exception of Jeep, he must have accepted the identification of the Florentinus mentioned in the preface (v. so) with the Florentinus who held office as praefectus urbi at Rome from mid-395 to the end of 3 97. 1 Why did Heinsius feel that an address to the city prefect Florentinus was out of place in the D.R.P. ? There can be little doubt, I think, that it was because he laid too great emphasis on the analogy Claudian draws between Hercules and Florentinus. If, he may have reasoned, Claudian dwells with such insistence on the labours of Hercules, he leads us to expect a parallel account of Florentinus' deeds ; but, since such an account is not, and indeed could not be given in a purely mythological epic, it must have formed the substance of a lost encomium of Florentinus, to which the preface was introductory. At the time of his preliminary article on the MS tradition of the D.R.P., this was also the view ofJeep.2 Later, however, in his two editions, he adopted the suggestion made by one Wedekind, a German translator of the D.R.P. , and affirmed that ' Floren tinus ' was not in fact a proper name but an honorific title given to the general Stilicho in recognition of his relief of Florence and defeat of the invader Radagaisus at Fiesole in 406.3 As long as the belief in a lost encomium remained unshaken, there was not a little to be said in favour of this new identifica­ tion. On the one hand, it was difficult, if not impossible, to imagine any circumstances in which Claudian might have seen 1

See 0. Seeck's edition of Symmachus, pp. cxli ff. pp. 3 59-60. J See E. Demougeot, De /'unite a Ia division de /'empire romain 395-41 0, Paris, 1951, pp. 3 5 3 ff. 2

Acta,

95

INTRODUCTION

his way to write a panegyric for the administrator Florentinus who, apart from his tenure of the city prefecture, seems to have lived a withdrawn and politically undistinguished life and who, in contrast to the regular recipients of panegyrics-members of the imperial family, successful generals, and holders of the consulship1-never accomplished any rrp6:�E1S which could form the central theme of an encomium. On the other hand, Stilicho was for long the poet's hero, and there was real point in a long catalogue of Hercules' deeds if an extensive narration of Stilicho's achievements was intended to follow it.2 Con­ firmation that it was Stilicho who was Claudian's addressee might be thought to be provided by Orpheus' song, which has seemed to some scholars to contain allusions to certain of Stilicho's military successes : thus, Diomedes (v. 12) represents Radagaisus,3 Pholoe (v. 44 ) alludes to the moral victory over Alaric on that mountain in 397,4 and Libyci . . sinus (v. 45) recalls the campaign against Gilda in the following year. s Although ingenious and at first sight even plausible, this interpretation of the preface and its function is totally erroneous. Of the supposed allusions to historical events, the first is too tenuous and laboured to carry conviction, while the second and third are no more than instances of a curious coincidence between myth and reality. The basic contention, however, that ' Florentinus ' is an honorific, has not hitherto been conclusive­ ly refuted, for Cremona's assertion that Claudian never uses ' un' espressione antonomastica ' of historical personages is false, 6 .

1

These three categories may of course overlap.

z See Jeep's ed. maior, vol. n.xiv-xv. 3 Fabbri, Zoe. cit. p. 97. 4 Fargues, op. cit. p. 17 n. 6.

s Birt (pp. xvi-xvii ) and Fargues (ibid. n. 3) refer this line to the praefectus urbi; were any reference to the real world in fact intended, it might with more justice be referred to Stilicho : see below, p. 98. 6 V. Cremona, ' La Composizione del " De Raptu Proserpinae " di Claudio Claudiano ', Aevum (1948), p. 23 3 . But Pellaeus= Alexander at 8.374 and Pharius= Hadrianus at c.m. 2 1 .2, 4·

DATE AND CIRCUMSTANCES OF COMPOSITION

and Birt begs the question in stating that ' Stilichonem semper suo nomine vocavit poeta '. 1 The crucial fact, so far ignored, is that such a formation as ' Florentinus ' is neither credible nor possible : honorifics commemorating victory in war seem always to have been formed from the name of the region or place subdued, and never from the name of the place where the fmal victory was won, in cases where the decisive battle was not fought on the enemy's home ground.2 To this one may add, on a more general level, that it is not easy, on Jeep's theory, either to understand why reference to the re­ sounding defeat of the savage and terrible Radagaisus should have been confmed to just one epithet in a poem of fifty-two lines, 3 or to believe that the preface could have survived the loss of an accompanying panegyric and wandered from the tradition of Claudianus maior into the quite separate one of the D.R.P. I therefore conclude that ' Florentinus ' is not Stilicho but the city prefect, and feel that there is no justification for our assuming the loss of a panegyric. The now standard theory about the date of the D.R.P. and the circumstances that occasioned its composition was advanced by Birt (praef. pp. xivf£) and accepted by Vollmer, Platnauer and, with some modification, Fargues. 4,In Birt' s view theD.R.P., with its two prefaces, was written during the period 395-7 as an oblique tribute to Florentinus' alleged services in providing 1 His further observation, p. xvi n. r, that it cannot be proved that ' victorem post pugnam publice Florentinum esse appellatum ', is true but would be irrelevant if the honorific were coined by the poet himsel£ It may also be noted that honorifics such as Macedonius etc. had gone out of official use by the time of Theodosius I's father : cf. Pacat. Panegyr. dictus Theodosio Augusto 5-4· z C£ B. Doer, Die romische Namengebung, Stuttgart, 1937, p. 70. 3 Fabbri, Zoe. cit. p. 97, believes that Claudian celebrated the event ' in modo coperto per eccitar meno l'odio dei potenti avversi a Stilicone '. But Stilicho's popularity was in fact as great as ever after this victory : see Demougeot, op. cit. pp. 3 60-r and notes. 4 Vollmer in R.E., s.v. Claudianus ; Platnauer in the Introduction to his edition, p. xiv ; Fargues, op. cit. p. r6.

7

97

HCD

INTRODUCTION

Rome with corn at the time of Gildo's revolt. The first preface is taken to indicate that the epic was begun after the panegyric for Probinus and Olybrius (recited in January 3 95 ) but before the In Rufinum (early 396 ) ; verses 45-8 of the second, dedica­ tory preface are interpreted as alluding to the Gildonian affair, and the other labours of Hercules as recalling deeds performed by Florentinus c. 3 79, when he is conjectured to have roamed the world like a second Hercules in the capacity of tzotarius ; and 1.30-1 are confidently presumed to look forward to an intended, but never executed, epilogue incorporating an analogy between Ceres' institution of agriculture and Floren­ tinus' arranging for corn to be shipped from Gaul after the loss of the African granary. The unfinished state of the epic Birt attributes to the fact that towards the end of 397 Florentinus was dismissed from office because he was not sufficiently active in mustering troops for Stilicho' s offensive against Gildo : since the epic was dedicated to Florentinus, he contends, its comple­ tion was contingent on Florentinus' remaining in office, for by pressing on with an encomium of a discredited official, Claudian must surely have offended Stilicho, who was just beginning to take note of his potential as a propagandist. The central question posed by this theory is whether the epic really is linked to Florentinus' term as prefect in the way that Birt supposes : are we, in other words, confronted not just with a mythological composition but rather with a subtle combina­ tion of overt story-telling and concealed laudation ? I cannot believe so for the simple reason that such a task as organising imports from the Gallic and German granaries lay far beyond the legal competence of the city prefect and could only be handled by a minister of Stilicho's power and authority. 1 1 On the occasions when he is explicit on this subject, Claudian invariably credits Stilicho with having opened up these alternative supply-routes : see

1 8 .402 ff. , 21.307 ff., 22.392 ff.

D ATE AND CIR C U M S T A N CES O F C O M P O S I T I O N

Florentinus' only claim to fame a t the time of the African rebellion seems in fact to have been that he prevented civil disturbances from arising as a result of the com shortage. 1 There is thus no ground whatsoever for reading into the

D.R.P. an 1.30-1

implicit encomium of Florentinus, or for interpreting

and II pf. 45-8 (let alone references to such creatures as Cerberus and the Stymphalian birds) in other than purely mythological terms ; and the disgrace of Florentinus is totally insufficient as an explanation of the non-completion of the epic, when all reference to the prefect could have been removed by the simple expedient of excising the second preface.

If, as I believe, Birt' s theory about the circumstances in which D.R.P. was composed is wrong, the two prefaces and the

the

epic as a whole can no longer be regarded as necessarily linked to the period

395-7 when Florentinus was city prefect. What is

more, there are serious reasons for holding that even if the first preface and book were to be assigned on whatever grounds to

395-7,

the second preface and second and third books would

have to be dated later : in other words the three books were not all composed at one time. First of all, if the whole of the

D.R.P. is dedicated to Florentinus, as Birt asserts,

why does not

the dedicatory preface precede the first book ? Koch's com­ parison of the two prefaces to the

In Rufinum2

is not relevant,

for the first of these, dealing with Rufinus' assassination, was recited in

396

in Stilicho's absence and therefore does not

address the general directly, while the second was added in

397

on the occasion of a second reading of the invective, this time in his presence, and consequently incorporates an apostrophe to the victor of Pholoe. Secondly, if the first preface of the

1 In the preface to his Symmachus, p. cxliii , Seeck remarks that ' Florentinus, licet inopiae mederi non posset, tamen tanta prudentia in re publica versatus est, ut nulla populi seditio oreretur et administratio eius omnium ore cele­ braretur '. 2 C£ Rh.M. xliv, p. 583 ; Birt, praef. p. xv and n. 2.

99

7-2

INTRODUCTION

D.R.P. was composed during 395 and the second shortly after it, what is the point of II p£ 5 1-2 antraque Musarum Iongo torpentia somno f excutis . . . ? Birt refers the ' long sleep ' to the neglect of epic composition in the centuries between Statius and Claudian, 1 but in view of the close analogy between the poet and Orpheus, who had ' for long neglected and set down his lyre ' (neclectum . . . diu . . . ebur, v. 2 ) , it is hard to resist the conclusion that Claudian is alluding to a particular stage in his own career as a writer. It is possible that the ' long sleep ' refers simply to an intermission of the D.R.P. alone and has no bearing on the composition of other kinds of poetry ( c£ Parrhasius,2 Claverius,3 Barth4 and Gesners) , but I am more attracted to the view that the second preface marks a resumption of work on the epic after a ' long ' period of general literary inactivity. Attempts to delimit this period must depend to some extent on how the first preface, which is obviously anterior to it, is interpreted. When Claudian (book 1 praef.) compares his progress in letters with that of the first mariner advancing from less to more hazardous voyages, and implies that at the time of com­ mencing the D.R.P. he is embarking on his most ambitious enterprise (vv. n-12 ) , to what point in his career is he refer­ ring ? Parrhasius, 6 Delrio, 7 JeepS and Fabbri9 take the implica­ tion to be that the D.R.P. is Claudian's latest work, begun after 1 P. xviii . 2 ' Agit . . . gratias Florentino cuius recenti liberalitate provocatus ad carmen redierat . . . intermissum.' J ' videtur poeta opus intermissum ipsius (Florentini) hortatu persecutus esse.' 4 ' per intervalla scriptos . . . hos libros docet haec praefatio.' s ' indicare, post librum primum de raptu Proserpinae, jacuisse opus, donee excitatus a Florentino suo poeta hunc alterum librum quoque proferret.' 6 ' (scripsit) novissime de raptu Proserpinae libros . . . ' 7 ' novissime scripti, ut ex praefatione illis pr�fixa facile colligitur, aut potius scribi coepti, videantur.' s Ed. maior, vol. r.xxix. 9 Op. cit. pp. 98-9.

100

DATE AND CIRCUMSTANCES OF COMPO SITION

all the historico-political poems were written ; Dempster, on the other hand, believes that we should infer that it is the earliest work of all ; 1 and other critics take up their positions between these two extremes. Because of the vague phrasing of the first preface, its date catmot be determined with certainty, only with greater or less probability. Is it likely, one may ask, that after composing the great cycle of poems for Stilicho, the longest of which, the 1,200 line panegyric on Stilicho's first consulship, had brought him the honour of a statue in Trajan' s Forum, with an inscription couched in the most fulsome terms, 2 Claudian could consider himself only just beginning to cross the open sea of literary creation ? The idea seems preposterous. Well then, is it likely that after writing his first panegyric, for Probinus and Olybrius, he would have expressed himself as he did in the first preface ? Most critics think he must have done so for they believe that the panegyric for the sons of Probus was, if not the first poem Claudian ever wrote in Latin, at any rate the first Latin poem he ever published or intended for publica­ tion,3 and therefore necessarily earlier than the D.R.P. Evidence for this belief is thought to be provided by the Epistle to Probinus (c.m. xli) , vv. 13-14, Romanos bibimus primum te consule Jontes f et Latiae cessit Graia Thalia togae, mistakenly in my opinion, for it seems to me very unlikely that Claudian had not had considerable experience in writing Latin verses before embarking on his first, very accomplished panegyric, or that he had written no Latin verse with tl1e intention of publication prior to 395. Birt maintains that v. 1 3 might be taken at its face value to mean that Claudian came to Rome in the year when Probinus and Olybrius held the consulship, but a similar sense 2 See Birt, praef. pp. xlii-xliii. Apud Burman, p. 732. viii ; Fargues, op. cit. p. 12 and n. 4· Platnauer, p. xiii , makes the astonislllng statement that ' it was not W1til the year 395 that he {Claudian) started to write Latin '. But the panegyric for Probinus and Olybrius was recited in January of that year ! 1

3 So Birt, p.

IOI

INTRODUCTION

can hardly b e extracted from v . 14, even reading accessit as he does in defiance of the poet's metrical practice and with questionable Latinity (also, for what this is worth, against the testimony of the eighth-century Verona MS) . The most likely interpretation of these verses, it seems to me, and the one which makes the best sense of the parallel phrases Romanos Jontes and Latiae togae, is that advanced by Postgate, I Cremona2 and Romano, 3 who take the poet to be referring, not to a change of domicile, let alone a change of language, but to a transition from Greek (mythological) to Roman (historico­ political) composition : in their view Romanos bibimus Jontes means ' I drank from Roman sources of inspiration', 4 and Latiae cessit Graia Thalia togae says in effect ' I ceased writing on Greek themes and began to compose for members of the Roman senate'. s If this interpretation is correct and Graia Thalia in fact implies poetry on Greek subjects, irrespective of language, there is nothing to prevent our dating the first preface and book of the D.R.P. before the panegyric for Probinus and Olybrius. To prove that they should be so dated is probably impossible, but v. 14 of the Epistle, which Birt dates 395/6, not long after the panegyric,6 seems to speak of the ' Greek Muse' as a thing of the past, and it is hard to believe that the first preface could have been written after the autumn of 395, when Claudian began to write his first imperial panegyric, let alone 'post magnae molis carmina . . . quae sunt 1 In his review of Birt's edition, C.R. IX (1895), 163-4. Commenting on Birt's interpretation of the couplet in question, he remarks that ' The metaphor of 13 means that then the poet devoted himself to Roman subjects '. The reading accessit he rejects in favour of cessit which, he maintains, gives ' the excellent sense that the poet's sportive Greek Muse made way or was abandoned for Roman national themes '. J Op. cit. p. 20 n. 39. 2 Loc. cit. pp. 247-8. 4 The use offons to signify sources of poetic inspiration is commonplace : c£ e.g. Prop. 3 . 1.3 ; Hor. Epist. 1 . 3 . 1 0 ; Lucr. 1 .927. s For the use of toga = ' the Roman senate ' Romano compares Claud. 21.330 6 P. ix. and 28.598.

! 02

DATE AND CIR CUM S T A N C E S O F C O M P O SITION

de Rufino . . . '1 I am therefore inclined to agree with Cremona and Romano in thinking that the D.R.P. was begun before 395, possibly while the poet was still living in his native Alexandria, as Cremona opines, but at all events before he had come into contact with members of the Roman senate, to some one of whom otherwise the epic would surely have been dedicated. Whatever the reason, whether temporary loss of interest in the subject or, more probably, the chance of making a lucrative career for himself as a writer on contemporary themes, Claudian wrote only book 1 of the D.R.P. in the first instance. At a later date, still to be determined, he was encouraged by Florentinus to resume work on the epic, and composed two more books before coming to an abrupt and final halt. This second instalment of the epic is introduced by a dedicatory preface at once elaborate in expression and simple in intent. ' Tu Florentine ad haec canenda me excitasti ' is all we expect the preface to convey, says Birt;2 and this, provided only that we do not press too closely the analogy between Hercules and Florentinus, is all that it does in fact convey, the gist of its message being that, just as Hercules provided Orpheus with the incentive to resume his singing, so Florentinus has prompted Claudian to take up composing again. And just as in c.m. iii, in a similar analogy between myth and reality, Aeternalis is called m eus Apollo because, in Birt' s words, ' ab eo . . . se incitari ad lusum poeticum Claudianus fatetur',3 so here it is not because his deeds are in any way comparable to those of Hercules that Florentinus is termed Tirynthius alter, but simply in virtue of his having by his patronage roused the poet's sleeping Muse to renewed activity. This is the only point of contact between Florentinus and Hercules, and attempts to read historical allu­ sions into the mythical incidents recounted in Orpheus' hymn to the hero are totally misguided. 1

Birt, p.

xv.

The In Ru.finum

was

written 395/6.

103

2

P.

xv.

3 P. xiv.

INTR O D U CTION

When, ftnally, was the D.R.P. resumed tmder the auspices of Florentinus ? No defmite answer to this question can be given, and the number of possible alternative answers that may be advanced depends on how we interpret the ftnal couplet of the Florentinus preface and what we think happened to the poet after his last datable work, the panegyric on Honorius' sixth consulship (recited in January 404) If vv. 5 I -2 of the preface are taken to refer back exclusively to book I of the D.R.P. and the poet means us to understand simply that he is now returning to the epic after intermitting it for a year or two, then books II and III may have been composed at any date after the intermission when the poet had time to spare : the extreme limits could be as early as c. 394, if book I was written about 390 (so Cremona), and as late as the period after 404, if Claudian was still alive then. 1 If, on the other hand, the concluding couplet of the second preface implies a general cessation from writing, as I think more .

' Barth and Crt!pin (ed. with French trans!. in the series ' Classiques Garnier ', vol. 2.366 n. 78) believe that the epic was resumed after the death of Stilicho (408). If they are right, one must account for the poet's failure to celebrate Stilicho's second consulship (405) or the annihilation of Radagaisus' army (406) or Honorius' seventh consulship (407) by assuming either that Claudian retired to his native Egypt after 404 or that he fell out of favour with Stilicho and was no longer called upon for the purposes of poetic advertise­ ment. But the balance of probability seems to me to rest v1rith the standard view, according to which the poet died not long after composing his last datable work, on Honorius' sixth consulship. There is much romancing about Claudian's last years and the circumstances in which he died. Vollmer's suggestion, in R.E. m.ii.2655, that he died on his ' Hochzeitsreise ' is pure fiction. The Deprecatio ad Hadrianum (c.m. xxii) , if it has any basis in historical fact and is not just a literary exercise, can hardly be dated after 400/1 (e.g. v. 3 1 nascentia vota could not apply after the poet had been honoured with a public statue), and cannot therefore be used as evidence that the poet was arraigned by Hadrianus in 404 or 405 (so Seeck, Geschichte des Untergangs der antiken Welt, v.56o), or amid the confusion and purges that accompanied Stilicho's assassination in 408 (so Gibbon, Everyman's Library ed. vol. m.21 2-1 3 , followed b y Boissier, L a.fi n du paganisme n.292, and others). Fabbri's ridiculous notion (Athenaeum XVII (1939), 3 8 ) that the De Bello Pollentino was revised by the poet after Alaric's entry into Rome (410), scarcely needed to be refuted by E. Merone (GIF vii (1954), 309-20).

104

D ATE AND CIRCUMSTANCES O F COMPO SITION

likely, one can be somewhat more precise about the possible dates for books II and III. The years after 404 still remain open, with of course the proviso made above, but it must seem im­ probable that an ambitious young poet should have given up all composition in the years before 395 ; on the contrary, one presumes that in order to make a name for himself he would have been writing a good deal at that time, and consequently that the final verses of the Florentinus preface refer to a later date, after 395· Within the period 395-404 Claudian was generally very fully occupied with the writing of his pane­ gyrics, invectives etc., and the only intervals of inactivity protracted enough to be described in retrospect as longus somnus are the fourteen or fifteen months that elapsed between the completion of the In Rufinum and the inception of the pane­ gyric on Honorius' fourth consulship ( ?June 396-autumn 397) and the lengthy gap between the recitation of the panegyric on Stilicho's first consulship and the writing of the De Bello Pollentino Qanuary 40o-springjsummer 403 ) 1 Claverius dates book II to 397, but it is no less possible that it should have been composed at about the same time as the De Bello Pollentino, which is similarly described (Claud. 25.1) as having been written after ' a long sleep ', or at a still later date, if the poet and his patron were alive after 404.z No inference, I think, may legitimately be made from the fact that the epic was never finished, the poet's death being no more likely an explanation than the distraction of some more pressing commitment or a waning of enthusiasm for the project, which could have happened at any time. .

1 The year is disputed, some scholars assigning Pollentia to 40z. C£ e.g. Birt, pp. xlvii ff., who argues for 40z, and J. H. E. Crees, Claudian as an historical authority, Cambridge, 1908, pp. 175 ff. , who argues for 403 . Like Crees, I feel that Claud. z8.1z4-5 is hardly reconcilable with the earlier date. 2 The latest reference to Florentinus comes in a letter of Symmachus (IV.56) which Seeck dates tentatively to 40z.

105

INTRODUCTION

V I. S O U R C E S A N D S T Y L E

For centuries before Claudian' s time the story of Ceres and Proserpine had enjoyed a considerable vogue in Greek and Roman literature. It was handled incidentally in Greek lyric poetry and tragedy, and by many Roman poets, especially of the imperial period ; it was utilised by the compilers of mytho­ logical handbooks, by the euhemerists and the Christian apologists ; it formed the subject of the Homeric Hymn to Demeter, and very likely of an Orphic poem or poems now lost ; and to it, under the influence of an Alexandrian original or originals, are devoted substantial episodes in Ovid's Fasti and Metamorphoses. 1 This long series of compositions was con­ cluded2 by Claudian's D.R.P. which, in spite of its incomplete­ ness, is still by far the most extensive account of the myth that has come down to us. Investigation into the sources on which Claudian may have drawn for his version of the story was initiated by R. Foerster, 3 who distinguished three main streams of influence : Sicilian, exemplified in the choice of locale; Orphic, attested by a number of details, notably Ceres' house ofbronze, Proserpine's weaving, the involvement of the three goddesses Venus, Diana and Minerva, and the participation of Triptolemus ; and Alexandrian, revealed by the use of the metamorphosis-motif (Cyane and the Sirens). Foerster's contention, however, that 1 Nor was the myth less popular with exponents of the visual arts. A comprehensive but (at least as far as Claudian is concerned) not very detailed historical survey of both literary and non-literary representations of the myth is provided by R. Foerster in his monograph Der Raub und die Riickkehr der Persephone, Stuttgart, 1874. 2 Apart from the occasional remarks of scholiasts etc. J In the work mentioned in note I above. Parallel passages had of course been noted by editors from Parrhasius onwards, but with a view rather to elucidating Claudian's linguistic and stylistic practice than to disentangling his sources.

1 06

S O URCES AND S TYLE

Claudian derived this combination of thematic elements from a single lost exemplar of the first or second century of our era, has not been accepted by later scholars, rightly, in my opinion, for the alleged similarity between Claudian's version and a series of imperial sarcophagus-reliefs which is thought to point to a common ancestor is much too superficial to prove such a contention. This first brief discussion was followed a few years later by the supplementary contributions of L. Cerrato 1 and A. Zimmermann, 2 the former of whom set out to document the influence of Orphic literature on Claudian and the latter to compare the various versions of the myth. The hypothesis of a single Vorbild advanced by Foerster was discarded by P. Fargues3 who, sensibly in my view, concluded that ' 1' auteur . . . a eu sous les yeux plusieurs modeles ' (p. 264 ; c£ p. 2 8 1 ) , namely Orphic poems and the Sicilian-Alexandrian tradition which is reflected in the Ovidian accounts. That the Homeric Hymn also influenced Claudian, as Fargues asserts, seems to me questionable, although it is a reasonable presump­ tion that as an Alexandrian born he would have been familiar with it. There is a good deal of ingenuity, but also some perversity, in E. Bernert's article ' Die �ellen Claudians in " De Raptu Proserpinae " '. 4 Slight infelicities and inconsistencies resulting from the conflation of divergent traditions are cleverly laid bare, but it is surely going too far to infer from differences in their disposition of common motifs that ' Claudian kann daher kaum, wie Fargues . . . mit Unrecht annimmt, den Ovid 1 ' De Claudii Claudiani fontibus in poemate De Raptu Proserpinae ', Riv. di .fi l. IX (1 8 8 1 ) , 273-395. 2 De Proserpinae raptu et reditu fabulas varias inter se comparavit A.Z., Lingen progr. (1882). J Claudien, etudes sur sa poesie et son temps, Paris, 1 93 3 . Foerster did not, however, say that his hypothetical exemplar was in Greek or that it did not contain non-Orphic elements. 4 Philologus xcm (193 8-9), 3 52-76.

1 07

INTRODUCTION

benutzt haben ' (p . 3 53 ) . 1 Bernert's final conclusion, that Claudian used r . ein alexandrinisches Gedicht, und 2. ein orphisches Gedicht vom Raube der Kore ', is true insofar as it follows the earlier assumption of both Orphic and Sicilian­ Alexandrian influence, but one should, I think, be chary of specifying the precise number of sources at Claudian's disposal and of particularising their contents : our knowledge of previous handlings of the myth is so fragmentary that there is no means of determining whether the sources from which Claudian's version was conflated were few or many. Although there can be no doubt that Claudian was familiar with a wide range of authors, z in all likelihood far more so than his imitations and adaptations permit us to discern, it is to Virgil, Ovid, Lucan and, above all, Statius that he is especially indebted in the D.R.P. for his ideas and his forms of expression. The influence of these four writers is dearly shown in the parallel passages collected en passant by commentators from Parrhasius onwards, and systematically, though by no means exhaustively, by S. Gramlewicz3 and Cerrato.4 On the basis of '

1 The same conclusion was reached by A. M. Eaton, The influence of Ovid in Claudian, Washington, 1943 , p. u S, on the ground that Claudian's verbal imitations of Ovid's Metamorphoses do not come from the Proserpine episode but from elsewhere in the work. This may be so, but the thematic remini­ scences are unmistakable. z Included in the list of authors said to have been imitated by Claudian which concludes C. Muellner's De imaginibus similitudinibusque, quae in Claudiani carminibus inveniuntur (Vienna, 1 893 ) are : Homer and Apollonius Rhodius ; Lucretius, Catullus, Virgil, Horace, Ovid, Lucan, Valerius Flaccus, Silius, Seneca in his tragedies, and Statius. A partly identical, though much less extensive catalogue of debts is to be found in C. Gunther's progr. De Claudii Claudiani comparationibus, Regensburg, 1 894. Alleged reminiscences of the Roman historians are discussed by E. Stoecker, De Claudiani poetae veterum rerum Romanarum scientia quae sit et unde .fiuxerit, Marburg, 1 8 89. Miss A. K. Clarke, P.C.Ph.S. n.s. I (r9so-r), 4-7, finds in the D.R.P. a know­ ledge of Seneca's Natural Q!!estions, the elder Pliny and Tacitus, but I am not entirely convinced. 3 Of:!aestiones Claudianeae, Breslau, 1 877· Reproduced in part, with additions and corrections, by L. Jeep in his ed. maior, vol. rr (1 879), lxxvi ff. 4 Loc. cit.

!08

S O URCES AND S TYLE

this accumulation of parallel passages Miss A. K. Clarke1 and 0. A. W. Dilkez have attempted to analyse the way in which Claudian used this borrowed material in the D.R.P. Miss Clarke draws attention to the ' complex pattern of allusions ' to be found in the poem and distinguishes three main types : brief traditional formulae, sustained variation on a well-known theme, and less studied, often perhaps unconscious, echoing of earlier passages. This line of investigation is indeed interesting, though inevitably it is fraught with uncertainty : in some of the cases discussed the supposed allusion strikes me as more apparent than real; in others there is the difficulty of deter­ mining whether the echo is intentional or such as might arise unconsciously from a well-stocked memory. Dilke, however, would almost completely eliminate the processes of memory, at least as far as echoes of the Georgics, Lucan and Statius' Achilleid are concerned, for in his submission Claudian did not know these works at all well but was obliged to consult codices containing them, as a scholar turns to his reference books, in the very course of composing his poem. But I find it impossible to believe that so patently learned a poet as Claudian could have operated in this mechanical way, and the 'progression of borrowings ' which Dilke imagines he has detected is based on a good deal of dubious or insubstantial evidence. 3 There is in 1 ' Claudian's methods of borrowing in " De Raptu Proserpinae " ', P.C.Ph.S. n.s. 1 (195 o- r), 4-7. z ' Patterns of borrowing in Claudian's " De Raptu Proserpinae " ', Re vu e beige XLm (1965), no. r , 6o-r . J A considerable number of alleged reminiscences are far from convincing. As examples one might instance the following : D.R.P. 11. 3 3 3 and Luc. 6.78, D.R.P. 1.1 54 and Luc. 8.6zz, D.R.P. rr.r3-14 and Luc. ro.65, D.R.P. 1.129 and S tat. Ach. 1.3 14, D.R.P. II p£ 45 and Stat. Ach. 1.222, D.R.P. rr. r07 and Virg. Georg. 2.447, D.R.P. rr.r28 and Virg. Georg. 2.404. For just what is proved by the appearance in two poets of such identical phrases as nulla sonant (D.R.P. rr. 3 3 3 = Luc. 6.78) and numquam tacitura (D.R.P. r. I 54 = Luc. 8 .622) ? Is it not bound to happen that Latin hexameter writers will from time to time inde­ pendently hit on the same phrase ?

109

INTR O D U CTI O N

fact no reason for denying that Claudian was as familiar with the three works considered by Dilke as with the Aeneid or Thebaid; and there is not a little truth in an earlier remark of Dilke's, that ' Claudian's Rape of Proserpine is to a large extent modelled on the Achilleid'. 1 In respect of form and style the D.R.P. continues the tradi­ tion of the post-Virgilian epic, being composed of a series of loosely connected episodes, with hardly a trace of a more closely integrated structure. 2 Coupled with this lack of a unifying theme is the absence of a central figure : the importance of Ceres does, it is true, become very marked as book III progresses, but for most of the time no one character is pre­ dominant. A large proportion of the poem is given over to speeches and eK
I IO

S O URCES AND S T YLE

From the earliest days of the Renaissance Claudian's poetry was greatly admired : Boccaccio, Buchanan, Muretus, the two Scaligers, Casaubon, Grotius and many others were full of praise for his achievements, I and even in more recent times his talents have not gone entirely w1recognised, though now for the most part in spite of rather than, as then, because of the rhetorical element in him. The excellences of the D.R.P. stem from a vivid and powerful imagination, which is seen to best effect in the brilliant descriptions (e.g. II.7I :ff. and 306:ff. ) and dramatic and impassioned speeches (e.g. 1.89:ff. and m.269 :ff. ) . Its shortcomings o n the other hand arise partly from an excess of exuberance, a tendency to over-elaborate any idea which happens to catch the poet's fancy,2 and partly from what one feels to be the basic unsuitability of the subject-matter, the theme of Proserpine's abduction and Ceres' search not being on the grand and heroic scale which one associates with epic poetry. Indeed, it is not impossible tl1at it was because of a growing realisation that he had made a mistake in his choice of subject that Claudian never brought the poem to completion. infrequency of elision and aphaeresis (the latter because of a decided aversion to the forms of esse), which results in a remarkable smoothness of rhythm. Statistics of comparative frequency are illuminating : D.R.P. I, one elision in 57·6 verses ; n, one in 23·2 ; m, one in 7· 8 . Tl:lls considerable variation between the three books might seem to confrrm the thesis that books I and n were written at different dates (see above, p. 99) ; but in fact there is no consistency in Claudian' s practice, or sign of any development in his attitude to elision, as is shown by the figures for the following works, which are arranged in order of composition : Pan. Prob. et 0/ybr. , one in thirty-one ; In Ruf I, one in 14·3 ; Pan. Q!!art. Cons. Hotz., one in 46· 8 ; Epitha/. Hon. et Mar., one in 56· 8 ; De Bell. Goth., one in I I·3 ; Pan. Sext. Cons. Hon., one in 13·7. 1 C£ the first chapter ( Claudiani carmina omnibus temporibus placuisse narratur ; eruditorum de eo iudicia examinantur ') of B. G. Walch's doctoral dissertation entitled Uberioris cotnmentatiottis de Claudiani carmine De Raptu Proserpinae inscripto specimen, Gottingen, 1770. 2 In the words of Walch, op. cit. p. 21, ' ubicumque sese vertat (sc. Claudia­ nus), luxurians et luxuriandi studio captus in rebus parvis eadem ingenii vi, eadem ornatus ubertate versatur ac in magnis An eloquent summing up of Claudian's literary qualities may be found in Gibbon's Decline and Fall at the end of chapter xxx . '

'.

III

INTRODUCTION

V I I. T H E A P P A R A T U S CRITICU S

The present text is based on all the extant MSS examined, with certain qualifications. Cited throughout are the older MSS (vetustiores), ranging in date from the twelfth to the early fourteenth century. The testimony of the younger MSS (recentiores) on the other hand is recorded only where they offer critically interesting readings unknown to the older stratum of the tradition ; where the recc. agree with the vett. I do not men­ tion the fact. This procedure of suppressing the recc. except where they have something new to say is, admittedly, arbi­ trary ; no less so is that of using all the vett. all the time. On the one hand was the practical need to confme the size of the apparatus criticus within reasonable limits, and the realisation that, as H. Dorrie has said of the tradition of Ovid's Heroides,1 ' Mehrheiten besagen bei dieser Uberlieferung nichts ' ; on the other, the conviction that it was desirable to give a much fuller account of the evidence of the MSS than had hitherto been available. The resultant apparatus criticus, therefore, attempts to strike a balance between excessive brevity and undue prolixity. The apparatus criticus has been composed on the following plan : r . Vett. belonging to one and the same class are cited by their individual sigla if they number five or less than five; if they number more than five they are (except at 1.139-41, m.28o and 43 8 where precise information about lacunae is needed) subsumed under the class sigla a: or � or y . These class sigla may occasionally stand for all or most members of the class in question; except where explicit notice is given to the contrary, 1

Gotting. Nachrichten 1 !2

nr.

5 (1960), 1 8 5.

THE APPARATU S CRITICUS

however, they should only be taken to mean more than five members of a particular class. 2. Five or less than five recc., of whatever class, are cited under their individual sigla ; where a reading occurs in more than five recc. the ' blanket' siglum \j.l is employed, irrespective of class. 3. The apparatus is negative in form (a) where the accepted reading appears in al3y, i.e. more than five vett. of each class ; (b) where 13 is not available (at I. J4I-2I4) and the accepted reading is in ay ; and (c) where l3y are not available (at 111.28o360 and 43 8-48) and the accepted reading is in a. In all other cases the apparatus is positive in form. Thus, to give some typical examples, inrumpit (1 p£ I I) and Maeonius . . . thyrsus (1. 19) are not mentioned because they are found in al3y = more than five vett. of each of the three classes ; at 1. 1 5 8 dextrum laevumque is not mentioned because it is found in ay = more than five vett. of each of the classes a and y ; and at III . 30I fortunam is not mentioned because it is found in a = more than five vett. of class a. Strepitus terris (I. ro) , on the other hand, is mentioned because it is not found in al3y ; likewise nullo teritur (1. 162) because it is not found in ay, and informatura (III . 3 3 1 ) because it is not found in a . 4 · In adducing the testimony of the vett. I list MSS o f each class separately, where necessary distinguishing between the classes by the use of commas. Thus, at 1.3 , prodere is offered by two Class a MSS (J 3 and R I ), three Class 13 MSS (C I F 3 G I) and one Class y MS (R 2) ; at 1.2o, jamulatur is given by more than five MSS each of Classes a and 13 and four MSS of Class y (A 2 C 2 E 2 P 4). Unclassifi.able MSS, where they are available, are listed after classifiable ones : c£ N at 1. I 3 . 5· Trifling corruptions and unimportant orthographica (including of course the constant confusion of e, ae and oe) are normally passed over in silence. Where the correct spelling is 8 II C D 1 13

INTR O D U CTION

assigned without comment to a small number of MSS, it should be understood that the rest of the tradition mis-spells the word in question, e.g. by erroneous insertion or suppression of the aspirate (nothus for notus, plegeton for phlegethon) or replacement of ph- by f- (jlegra for phlegra). So at 1. !2, 16, 24, 88, 92, 1 3 5, 11.248, 255, 3 15, III. 201, 264, 3 3 7· 6. Scholars whose conjectures have subsequently turned up in MSS are of course still mentioned in the apparatus, as equity demands.

I I4

T E XT A ND CRITICA L A P P ARA T U S

S I GLA C O D I C E S VETUS TIORES

AI C3 D FI J3 KI L5 L6 L IO MI M2 M3 04 P5 RI R5 R7 T W w

Classis a: Antverpiensis 8 5 Cantabrigiensis coli. div. Pet. 2. r.o Dresdensis De. 1 5 7 Laurentianus xxiv sinistr. I2 Leidensis 294 Ambrosianus H 73 sup. Londiniensis Mus. Brit. Reg. I S A vii Londiniensis Mus. Brit. Add. I202I Phillippicus 8776 Monacensis 597 Monacensis 63 1 Monacensis 391 Oxoniensis Bodl. class. e 47 Parisinus I 5005 Vaticanus Barberinianus 4I Vaticanus I663 Vaticanus Reginensis I 556 Venetus Marcianus XII x Guelferbytanus Gudianus 228 Lincolniensis I 3 2 Classis 13

B bI CI EI F2 F3 GI LI L4

Bernensis 398 Bononiensis 222I Cantabrigiensis coli. Corp. Christi 228 Escorialensis O.rn.25 Laurentianus 250 Laurentianus xxxiii.4 Berolinensis quart. 740 Londiniensis Mus. Brit. Egerton 2627 Londiniensis Mus. Brit. Harl. 275 3 I I6

SI GLA

02 03 P2 R3 R4 R6

Oxoniensis Bodl. 2077 (pars prior) Oxoniensis Bodl. 2077 (pars altera) Parisinus 8082 Vaticanus Reginensis 1540 Vaticanus 2807 Vaticanus 3290

A2 C2 E2 F4 H ]2 ]4 L2 L3 OI PI P3 P4 R2 R 25

Classis y Antverpiensis 17.1 Cantabrigiensis coli. div. Pet. 2. 1.8 Escorialensis S.m.29 Laurentianus Ashburnhamcnsis L.977 Vigorniensis F 147 Leidensis 395 Leidensis 105 A Londiniensis Mus. Brit. Add. 21213 Londiniensis Mus. Brit. Add. 6042 Oxoniensis Bod!. 2195 Parisinus 8o8o Parisinus 829 5 Parisinus 8296 Vaticanus Reginensis 1440 Vaticanus Palatinus 1 573 Hauniensis Gl. Kgl. S. 1905 Parmensis 2504

u z

JI N v

Incertae Classis Leidensis 3 8 5 Nottinghamiensis Aln. 1 3 0/2 Vindobonensis cod. ser. nov. 9363

CODICES RECENTIORES QUOTQUOT ADHIBENTUR

a c d eI e2

Noveboracensis 96 Clarensis 4 Mettensis 647 Erlangensis 621 Erlangcnsis 626 I I7

SIGLA

F5 F6 F7 F9 F 10 F II F I2 F I3 F I4 F IS F I6 fI G2 G3 G4 h

j

K2 K4 k L7 L8 M4 m

n 05 06 P6 P7 P8 P9 pI p2

Q

R8 R9 R IO

Laurentianus xxxiii.8 Laurentianus xci sup. 3 4 Florentinus bibl. nat. rr.ix. I 3 2 Riccardianus 7 I 8 Laurentianus xxxvii.I4 Laurentianus xci sup. 4 Laurentianus Leopoldinus cciii Laurentianus xxxiii .I Laurentianus xxxiii.2 Laurentianus Fondo Redi IOO Laurentianus Fondo Acq. e Doni 3 5 8 Mutinensis a.M.9.22 Berolinensis octav. I I2 Berolinensis Hamiltonensis 492 Berolinensis fo. 3 9 Holkhamicus 3 3 2 Princetonensis 47 Ambrosianus Z 7 3 sup. Ambrosianus S 66 sup. Pisaurensis I 8 Londiniensis Mus. Brit. Harl. 5 198 Londiniensis Mus. Brit. Add. I009 I Monacensis I5740 Baltimorensis 437 Neapolitanus V.D.53 Oxoniensis Bodl. I484I Oxoniensis Bodl. 8860 Parisinus 7892 Parisinus 8o8 I Parisinus 8297 Parisinus I I 3 24 Patavinus bibl. univ. 1 561 Patavinus bibl. capit. C 62 Papiensis 4I2 Corsinianus col. 43 .F.5 Vaticanus I66 I Vaticanus Ottobonianus 2859

II8

SI GLA

RII R I2 R 13 R 14 R 15 R 16 R 17 R18 R 19 R 2o R21 R 22 R23 R 26 R 27 R 28 R 29 R3o r T2 T3 T4 t

u V2 V3 Y1 Y2 y a

13 y \II

Vaticanus Ottobonianus 2 126 Angelicus 2266 Vaticanus 2808 Vaticanus 1660 Vaticanus Urbinas 657 Vaticanus Borgianus 343 Corsinianus 43 .G.25 Corsinianus 43.D.36 Vaticanus Reginensis 1369 Vaticanus Reginensis 1428 Angelicus 1 3 45 Vaticanus Barberinianus 68 Vaticanus Palatinus 1 7 14 Vaticanus 2809 Vaticanus 7182 et Vaticanus 2864 Vaticanus 1 1420 Angelicus 1461 Vaticanus Chigianus H.vn.236 Ravennas 120 Venetus Marcianus xn. 1 1 Venetus Marcianus :xrv. 183 Venetus Marcianus xrv.202 Posnaniensis 179 Urbanensis MCA 4 Vindobonensis 3087 Vindobonensis 3 198 Perusinus 247 Perusinus 63 3 Yalensis sine numero plus quinque vetustiores classis a plus quinque vetustiores classis 13 plus quinque vetustiores classis y plus quinque recentiores

His etiam compendiis usus sum : a.c. = ante correctionem ; codd. = codices omnes vel plures ; edd. editores ; lac. = lacuna ; m.a. =

I I9

=

SIGLA

manus altera ; m . tert. = manus tertia ; p.c. = post correctionem ; ras. = rasura ; recc. = codices recentiores ; s.c. = sed correxit, sed correctum ; sscr. = superscriptum ; vett. = codices vetustiores ; v.l. = varia lectio. Abbrevianturnomina hunc in modum : Burm(annus) ; Claver(ius) ; Gyrald(us) ; Heins (ius) ; Parrhas(ius) . Gud. = excerpta Gudiana, et Isengr. = editio lsengriniana.

120

NVENT

PRAEFATIO

A primus secuit qui nave profundum et rudibus remis sollicitavit aquas, qui dubiis ausus committere flatibus alnum quas natura negat praebuit arte vias, tranquillis primum trepidus se credidit undis litora securo tramite summa legens ; mox longos temptare sinus et linquere terras et leni coepit pandere vela Noto ;

I

Inscriptiones quotquot in vett. libris adsunt : claudii claudiani de raptu proserpine prirnus liber incipit A I : Praefatio . . . De raptu Proserpine lib. I A 2 (manu Pulmanni ?) : incipit i praefatio claudii claudiani C I : incipit liber claudiani C 3 : liber primus de raptu proserpine F 2 : Prologus in librum claudii claudiani de raptu proserpine G I : Claudianus de Raptu Proserpinae ]2 (man. recentiss.) 0 I : incipit liber K I : incipit liber claudii eloquentissirni hominis de raptu proserpine L I : incipit claudianus M 3 (fo. 42 v), L 2 : prologus claudiani in libro de raptu proserpine L 3 : claudianus L 5 R 5 (m.a.) : Claud. Claudianus de raptu Proserpinae L IO (add. Pulmannus) : incipit prologus de raptu proserpine P 3 : De raptu Proserpinae P 4 (man. recentiss.) : claudius claudianus P 5 : Claudii Claudiani Praefatio in raptum Proserpine R 7 (man. recentiss.) : Claudianus de Raptu Proserpine W (man. recentiss.) Inscriptiones recentiorum notabiles : Publij Claudiani canopei de raptu proser­ pine primus liber foeliciter incipit a : flavii claudiani poete de r.p. liber incipit F 9 (item in F I 3 , om. tamen verbo poetae, et in R I S , clarissirni post poetae addito) : claudianus (sic) poete florentini de r.p. liber prirnus incipit F I S : Claudi Claudiani Proserpina incipit n : Claudiani poete Alexandrini de r.p. liber primus / incipit foeliciter f I (prohoernium pro foeliciter) p 2 u ( . . . feliciter Prohernium) : Claudii Claudiani de r.p. prologus incipit ad florentinum R I 3 : Prohoernium Claudiani poete Alexandrini R I9 : claudii claudiani de r.p. tragedia heroica incipit feliciter R 2o : Claudiani poete optimi egyptiaci de rapftu proserpine praefatio in primum librum incipit r : claudiani canopei de r.p. primus liber incipit Y r . Raptus Proserpinae inscribi iussit Heins. sed vide comment. I p.s.q. C I G I L I/4 : s.q.p. A I ]3 K I L s T, b r F2 0 2 P 2, A2 L2j3 U, N : s.p.q. a , E r F 3 R 3 /4, y primum L I , U 3 fluctibus L sfro, F 2 , L 2 O r alnos P 4 4 vetat E I 5 trep. prim. ]unii primus tradidit F 4 J 2 0 I : protulit M 1 6 secreto F 1 8 atque levi R 3 tendere L 1 : tradere J 2 noto D ?, ]4 : notho ceteri

!2!

5

D E RAPTU P R O SERPINAE

1o

ast ubi paulatim praeceps audacia crevit cordaque languentem dedidicere metum, iam vagus inrumpit pelage caelumque secutus Aegaeas hiemes Ioniumque domat. II altius pro iam vagus ]4 irrupit cxl3y : exultat Isengr. marg. pelagus J 3 0 4, C 1 (a.c.) G 1 L 4 R 3 , R2/25 12 ioniasque L 10 W, F2,]2 L2 0 1 (p.c.) premit M 1 auctor est Burm. codicem quendam Sangallensem post v. 1 2 adiecisse hoc distichon : sic ego qui rudibus scripsi praeludia verbis f ingredior Stygii nobile Ditis opus. sed, ne longior sim, vox praeludium ab antiquis scriptoribus nusquam quod sciam usurpata spurios esse hos versus satis monstrat. praifationi .finem deesse credebat Gesner, quem secutus est Gramlewicz ( Qgaestiones Claudianeae') et, quamvis dubitanter, Koch (praef. ad ed. Teubner.). non assentior. vide adnotationes. '

·

!22

L I B E R PRIM U S

Inferni raptoris equos adflataque curru sidera Taenario caligantesque profundae Iunonis thalamos audaci prodere cantu mens concussa iubet. gressus removete, profani. iam furor humanos nostro de pectore sensus expulit et totum spirant praecordia Phoebum ; iam mihi cernuntur trepidis delubra moveri sedibus et claram dispergere limina lucem adventum testata dei ; iam magnus ab imis auditur strepitus terris templumque remugit Cecropium sanctasque faces attollit Eleusin. angues Triptolemi stridunt et squamea curvis colla levant adtrita iugis lapsuque sereno erecti roseas tendunt ad carmina cristas. ecce procul temis Hecate variata figuris exoritur levisque simul procedit Iacchus crinali florens hedera, quem Parthica velat incipit liber G I : Claudiani de r.p. liber primus incipit L 3 : Titulus Claudii Claudiani de r.p. primus liber incipit P I/3 : Liber I P 4 (man. recentiss. in marg.) : De raptu Proserpine Liber primus R7 (man. recentiss.) 2 tartareo L 5 , ]2 : trinacrio ]4 3 prodere J J R r (a.c.), C r F3 G r, R 2 : promere cx�y 4 Musa (cum augusta J 3 ) Gustqfsson concussa Isengr. : congesta cx�y : coniecta ) 4 : collecta P 2 (v.l.) : concreta ) 3 (v.l.) : augusta ) J : augusta C r, R 2 : commota laudat Dempster : incensa Baehrens 5 d.n.p.s. M2 R s/7, U: s.d. corpore n. L 2 : n.d. corpore s. P 3 8 diffundere L 2 limina R 3 , L 3 O r : lumina cx�y : culmina Isengr. : culmine Heins. : fulmina Meursius apud Burm. 9 adventus L6, F 3 dee Z (v.l.) 10 strepitus terris L6 M 1 T W, b r E 1, A2 J 2 L2 P 3 R25, N : terr. strep. L 5 : fremitus terr. cx�y : terr. frem. ljl : sonitus terr. d II cecropium M4 (a.c.) R 1 3 r t : cecropidum D K r sacrasque M r, O r R 2 5 attollit A I K r L6/IO M2, F2, 0 I P 4: extollit cx�y eleusin F 3 : -im N : -is ceteri 12 triptolemi D, C I G r ?, P 4 R 2 stridunt ] 3 , C r R 3 , E 2 P 4 R 2 : -ent cx�y 13 astricta a, E r F2 (p.c.) F 3 L r R4, y, N 14 tollunt R r4j26 15 trinis L 6 temas . . . figuras L r 1 6 lenisque nonnulli codd. : letusque a[3, F 4 ]2 L 2 O r U iacchus G4 K4 17 frondens e 2 r 17-18 tigris J velat D K r

!23

5

10

IS

D E RAPTU P R O SERPINAE

20

25

3o

35

tigris et auratos in nodum colligit ungues ; ebria Maeonius firmat vestigia thyrsus. Di, quibus innumerum vacui famulatur Averni vulgus iners, opibus quorum donatur avaris quidquid in orbe perit, quos Styx liventibus ambit interfusa vadis et quos fumantia torquens aequora gurgitibus Phlegethon perlustrat anhelis : vos mihi sacrarum penetralia pandite rerum et vestri secreta poli : qua lampade Ditem flexit Amor; quo ducta ferox Proserpina raptu possedit dotale Chaos quantasque per oras sollicito genetrix erraverit anxia cursu; unde datae populis fruges et glande relicta cesserit inventis Dodonia quercus aristis. Dux Erebi quondam tumidas exarsit in iras proelia moturus superis quod solus egeret conubiis sterilesque diu consumeret annos inpatiens nescire torum nullasque mariti inlecebras nee dulce patris cognoscere nomen. iam quaecumque latent ferali monstra barathro in turmas aciemque ruunt contraque Tonantem coniurant Furiae, crinitaque sontibus hydris L6 M2 T, E r , ]2/4 L2 O r U 18 nodos L r 19 meoniis . thyrsis a !3y : -io . . . -o C 3 , U ftrmat F 2 : ftgit a!3y : fulcit T (v.l.), L 3 20 famu­ 21 foribus pro opibus ]eep : oris latur a !3 , A2 C 2 E 2 P 4 : -antur a !3 y .

.

Baehrens damnatur Heins. olim 22 fuit pro perit F 14 humentibus F 3 : viventibus F 7 23 circumfusa V 2 spumantia J 4 : flagrantia u 24 vorticibus Isengr. phlegethon G 2 Q 25 hie incipit H sanctarum T 4 t 26 et nigri r 26-27 quod (pro quo?) vulnere Ditem / laesit Cud. 27 ulsit in notis Claver., unde ussit elicit fero L2 curru V 2/3 et Isengr. 28 hie incipit 0 3 letale F 14 29 genetrix R 7 (a.e.), F 4 : -itr- eeteri passu R s/7 30 fruges populis F r M r W, E 2 ] 4 : populo fr. A r T, E r F2 R4, ]2 R25, N : fr . populo K I leges pro fruges Claver., quod probabat Heins. cum pro et K I , ] 4 : cur p 2, quodHeins. volebat : ut Heins. 3 3 quia L s , L 3 34 con(n)ubii J 3, C 1 0 3 : con(n)ubio I.Jl: coniugiis R s/7 sterilisque c 35 dulcesque mariti M 2 : nudasque maritae in notis Claver. (maritae iam C r , marg. m. ree.) 38 aciesque ay 39 consurgunt 0 3 turbae c

124

LIBER PRIMUS

Tisiphone quatiens infausto lumine pinum armatos ad castra vocat pallentia Manes. paene reluctatis iterum pugnantia rebus rupissent elementa fidem penitusque revulso carcere laxatis pubes Titania vinclis vidisset caeleste iubar rursusque cruentus Aegaeon positis aucto de corpore nodis obvia centeno vexasset fulmina motu. Sed Parcae vetuere minas orbique timentes ante pedes soliumque ducis fudere severam canitiem genibusque suas cum supplice fletu admovere manus : quarum sub iure tenentur omnia, quae seriem fatorum pollice ducunt longaque ferratis evolvunt saecula fusis. prima fero Lachesis clamabat talia regi incultas dispersa comas : ' 0 maxime noctis arbiter umbrarumque potens, cui nostra laborant stamina, qui fmem cunctis et semina praebes nascendique vices altema morte rependis, qui vitam letumque regis (nam quidquid ubique gignit materies, hoc te donante creatur debeturque tibi, certisque ambagibus aevi 40 Tisiphone \jJ : Tys- D : Tes ceteri vett. infesto A 2 ] 2 numine C I 41 movet D ? K I pollentia L s 43 recluso L I 44 lassatis L6 46 ar(c)­ to D ? K I, A 2 ]4 : acto } 3 (a.c.), C I F 2, E2 R 2 : alto a, b i L I , Z, N : vincto 0 3 (v.l. marg. m.a.) : victo 0 4 : duro F4 (p.c.) : atro e2 m : vasto M 3 et 47 obruta M 2 vibrasset F 4 48 metuere C 3 , Goetz pectore M2/3 F 2 (s.c.) L I , P I, N 49 soliumque patris C 2 stravere F s s o vultu A I L S/IO R s/7, h i , A 2 Z : questu T : voto pl 51 sub lege V 2 5 2 o.£s.q.p.d. L 2 series R I 3 (a.c. ) R 2 2 fatali Gyrald. 5 3 cunctaque R I 3 adducunt F I fusis F I J 3, C I G I L4 0 3, N : pensis aj3y 54 conclamat J 3 M2 0 4, C I G I 0 3 , C2, N : proclamat C 3 M I , L 3 U 55 hie incipit R6 58 alternaque vices nascendi C 3, U modos pro vices J 3, b I 59 letum vitamque L I seris ] 3 (v.l.) in orbe C I 0 3 , N repensas r 61 certis (om. -que) a, C I F 3 G I 0 3 R6, y : M2, F4 (p.c. ex ubique) cunctisque D K I , J 4 : cunctis W : longisque e I : certique malebat Heins. -

125

40

45

so

ss

6o

D E RAPTU PRO SERPINAE

rursus corporeos animae mittuntur in artus) : ne pete firmatas pacis dissolvere leges quas dedimus nevitque colus, neu foedera fratrum civili converte tuba. cur inpia tollis signa ? quid incestis aperis Titanibus auras ? posce Iovern; dabitur coniunx.' vix ilia ; pepercit erubuitque preces animusque relanguit atrox quamvis indocilis flecti : ceu turbine rauco cum gravis armatur Boreas, glacieque nivali hispidus et Getica concretus grandine, pinnas disrumpit, pelagus silvas camposque sonoro flamine rapturus ; si forte adversus aenos Aeolos obiecit pastes, vanescit inanis impetus et fractae redeunt in claustra procellae. Tum Maia genitum, qui fervida dicta reportet, imperat acciri. Cyllenius adstitit ales somniferam quatiens virgam tectusque galero. ipse rudi fultus solio nigraque verendus maiestate sedet : squalent inmania foedo sceptra situ ; sublime caput maestissima nubes asperat et dirae riget inclementia formae ;

65

70

75

so

mutantur D M 2 ? R s/7, F2, ]4 L2 ortus F 2 (a.c.) 63 pactis Baehretls ne L s M 2j3 R s W, F2, L 3 P I (a.c.) R25 proelia D ? K I , ]4 rerum T com (m)it(t)e L6, F 3 : convelleJ. J. Scaliger 66 infestis T, R 2 : illicitas L s W, !3, Z orbem F 2 67 coniunx dabitur C 3 vix ilia ; pepercit Barth (ilia et R I4/26, uterque in marg. ), pro b. Heins. : vix ille pep. codd. ceteri : vix ilia peregit, lacunapost hunc vers. statuta, Reinhardt 68 elanguit Barth e suo libro 69 indoctus d rapto Z 70 minaci Claver. 71 gelida D F I K I, J 4 7Z disrumpit J 3, C I 0 3 : disrupit 0 I : irrupit F I4 : flare cupit cx!3y : flare iubet L7 R I7 : bella cupit Claver. : ire cupit Baehrens : num erumpit vel prorumpit? pelago M I : saltus 0 6 silvas pelagus L2 campos silvasque W 73 si fors L 5 aversus ] 3 , F 3 74 Eolos d : -us ceteri obiceat F I , ut1de obiciat ]eep : occlusit Gud. in auras P s (a.c. ) R 5/7, F 4 L 3 U : in antris V 2/3 75 ad claustra A I L ro, L I , L 2 : in castra K r M2, ]4, N: in saxa 04 76 tum A I J 3 L 6 W, !3y : tunc cx!3y maia natum U 77 affuit A I L 5 , F 2 heros M 1 , F 4 (p.c. ) 7 9 ille T , L 2 P I solio fultus K I , 0 2 P 2 R4, C 2 (solio fluctus !) ]4 L 2 Sz dirae ex, C2 H ]2 L2 R25 (p.c.) : durae cx!3y

6z 64 65 ]3

!26

LIBER PRIMUS

horrorem dolor augebat. tum talia celso ore tonat (tremefacta silent dicente tyranno atria ; latratum triplicem conpescuit ingens ianitor et presso lacrimarum fonte resedit Cocytos tacitisque Acheron obmutuit undis et Phlegethonteae requierunt murmura ripae) : ' Atlantis Tegeaee nepos, commune profundis et superis numen, qui fas per limen utrumque solus habes geminoque facis commercia mundo, i celer et proscinde Notos et iussa superbo redde lovi : " tantumne tibi, saevissime frater, in me iuris erit ? sic nobis noxia vires cum caelo fortuna tu1it ? num robur et arma perdidimus, si rapta dies ? an forte iacentes ignavosque putas quod non Cyclopia tela stringimus aut vacuas tonitru deludimus aures ? nonne satis visum grati quod luminis expers tertia supremae patior dispendia sortis informesque plagas, cum te laetissimus ornet

ss

90

95

100

83 horrorem ] 3 , C I G r L4 0 3 R 6 : terrorem aj3y : maerorem Cud. tum D K I, C 2 ]4 R 2 : tunc aj3y : nunc F7 84 magistro F 3 85 tartara W latratus triplices L 5, Z 86 cerberus R 3 87 cocytus a, C 1 E I F 2 0 3 , A 2 R 2 5 U Z tacitusque K r , H : tacitis (om. -que) ay 8 8 phlegethonteae F 4 89 tegcee M 1, J 2 : tegec aj3y 91 vehis pro facis P 4 caelo W, R25 92 i celcres proscinde Isengr. : i celer i proscinde e 2 et Heins. : i celer hinc et fmde ljJ : i ccler et precede T notos D ?, L2 ac cod. Bartholin. apud Burm. 93 hie incipit w fratrum a, F 3 , y, N : semper R7 (a c.) 94 iuris habes D K1 M1, ]4 U si K1 L6 M I P 5 (p.c. ?) R I (p.c.), C2 U sic vires noxia nobis R25 : vires sic noxia nobis L 5/6 (si), L2 Z 95 num 97 innoa, b I E I L4 R3/ 4, y : non aj3y 96 adeoque iacentes Heins. cuosque d quia H 98 spargimus L 5, ]4 (v.I.) L2 (m.a., p.c.) P I Z et] 3 , i3 vacuas F I M 3 0 4, Z , N : vanas aj3y : vano excerpta Laeti, laudat Pa"has. tonitru vanas M2 hominum pro tonitru Isengr. aures R 5 et Isengr. : auras aj3y : aras 0 2 (sscr.) P 2 (v.l.) 99 visum est ay quod grati a, L2 R25 U Z: superis quod A2, unde superi quod Heins. 100 tetrica Gustafsson : tristia Goetz postremae F 4 P I discrimina R25 : stipendia Jeep : sic praeinia Baehrens noctis T W, U : mortis F 9 101 infemasque F 5/I I r clarissimus A I L 1 0 : lcnissimus R 4, P I .

127

DE RAPTU PRO SERPINAE

105

no

115

120

Signifer et vario cingant splendore Triones ; sed thalamis etiam prohibes ? Nereia glauco Neptunum gremio conplectitur Amphitrite ; te consanguineo recipit post fulmina fessum Iuno sinu. quid enim narrem Latonia furta, quid Cererem magnamque Themin ? tibi tanta creandi copia; te felix natorum turba coronat. ast ego deserta maerens inglorius aula inplacidas nullo solabor pignore curas ? non adeo toleranda quies. primordia testor noctis et horrendae stagna intemerata paludis : si dictis parere negas, patefacta ciebo Tartara, Saturni veteres laxabo catenas, obducam tenebris solem, conpage soluta lucidus umbroso miscebitur axis Averno." ' Vix ea fatus erat, iam nuntius astra tenebat. audierat mandata pater secumque volutat diversos ducens animos, quae tale sequatur coniugium Stygiosque velit pro sole recessus. certa requirenti tandem sententia sedit. Aetnaeae Cereri proles optata virebat 102 lucifer W, b 1 stringant L7 103 thalamos ay, N quin etiam thalamum Lz nereida ] 3 (s.c.) M2, C I G I L4 (a.c.) 0 3 R6, O I glaucus C 1 0 3 : -urn L4 (a.c.) P2 (v.l.), A2 104 neptunus C 1 0 3 R 6 amphi­ trite C 3 F I , j3 , Az ]4 L 3 P 4 R2 : -em (-en L 5, N) aj3y 105 lassum T 106 referam pro narrem P 5 Latoia Heins. : latitantia D K I, J 4 latonia furta renarrem 0 4 : latonia furta referrem W, F 2 107 thetin a, L I 0 2 P 2 R4, y 108 natorum te felix j3 , J 2 P 4 : natorum felix te V 2 109 deforrni M 3 no poenas pro curas F 5 h R I 8 I I I nonne adeo Heins. 1 11-12 noctis f testor L6 M2j3 , C 2 L2 112 stagna a, E I F2, y : signa aj3y intemeranda L2, quod malebat Heins. : invidiosa L 10 1 13 dictis a, 0 3 (p.c.) R 3j4, y, N : dicto ] 3 K I M 3 W, j3y neges ] 3 R 5 ? R 7, C I G I L 4 0 3 R 6 115 solem tenebris T : tenebris lucem 0 5/6 n6 hie indpit B fulgidus Claver. 117 vix ita fatus a k 119 divisos Heins. animos ducens M2 R I h i (s.c.) 120 con(n)ubium G I L4 stygiosve L 5 T, F 3 L I O z P z R4, H P4 Rz pro lucc rec. C 3 , U : persolvere census B F 3 L I O z (a.c.) P 2, P 4 121 tandem haec Heins. c.r. mansit sententia tandem F4 P I 122 (h)en(n)ee C I 0 2,

!28

LIBER PRIMUS

unica nee tribuit subolem Lucina secundam fessaque post primos haeserunt viscera partus ; infecunda quidem, sed cunctis altior extat matribus et numeri damnum Proserpina pensat. hanc fovet, hanc sequitur ; vitulam non blandius ambit torva parens, pedibus quae nondum proterit arva nee nova lunatae curvavit germina frontis. iam vicina toro plenis adoleverat annis virginitas, tenerum iam pronuba flamma pudorem sollicitat mixtaque tremit formidine votum. personat aula procis : pariter pro virgine certant Mars clipeo melior, Phoebus praestantior arcu; Mars donat Rhodopen, Phoebus largitur Amyclas et Delon Clariosque lares ; hinc aemula Iuno, hinc poscit Latona nurum. despexit utrumque flava Ceres raptusque timens (heu caeca futuri !) [commendat Siculis furtim sua pignora terris infidis Laribus natam commisit alendam] 0 I : eunee R4 : etheree P s (v.l.) cereris L4 nitebat P s : vigebat U 123 subolem non ante ]eepium, ut vid. : sobolem codd., et edd. vett. sobolem tribuit C 3 L s , L2 R2 : tribuit prolem L6 R 5j7, L 3 : prolem tribuit C2 125-6 matribus extat J 124 requierunt R 7: cessarunt J 4 (gloss.) et Heins. complet N : reddit ]2 altior ]2 P 3 126 damnum numeri R 3 , R 2 127 vitulum a, B LI, L2 U Z nee K r , ]4 128 seva par. R3 qui MI necdum proterit P 2: non protriverit E r F 2 arvum P s : herbas R2o 129 lun. curvatur F ro : lun. curvantur P s (a c. ) : curvavit lun. Gr L 4 R 6 cornua a , G r L 4 R 6 , y 130 iam matura W thoris J 2 131 fama 133 certat ]4 L2 132 timet W : fremit K r , P 4 : premit P 2, F4 R25 C3 134 maior M2 constantior J 2 135 rhodopen "'' rodopen (-e) donat M 3 T 136 clarosque M r/2/3 (p.c.), C I L I (a.c.) 0 3 (a.c.) R 3 , R25 137 lat. nur. pose. O r utramque] 3 W, prob. Heins. 138 raptumque 0 4 tremens M 3 ignara futuri R I 3 139 om. A r (habet post 145) F r (add. marg. m.a. voce terris omissa) R I T w (add. marg. ; commisit et pignora leguntur) , B b r C I E r F 3 G r L r 02 (add. marg.) 0 3 P 2 R 3/4/6, A2 C2 H ]2 (add. marg.) L2 (habet post 241) P r (add. marg.) P3 Z; post 140 positus est in C 3 L 5/6 M r j2 0 4 R s/7, F 2 0 2 (marg.), F4 R25 U commisit L s w (marg.) : committit p 2 pig­ nora ex, F2 L4 0 2 (marg.), y, N : gaudia A r J 3 L w, L 3 (in ras.) : sydera P 5 (a.c.) 140 om. ] 3 M 3 , B b i (add. marg.) C r F 3 G r L I 02 (add. marg. ante 139) 0 3 (add. marg. ) P 2 (add. marg.) R 3 j6, E 2 (add. marg.) L 3 P 4 (add. marg. m.a.) R 2 ; .

9

129

HCD

125

130

135

140

DE RAPTU PRO SERPINAE

145

150

155

aethera deseruit Siculasque relegat in oras ingenio confisa loci. Trinacria quondam Italiae pars una fuit, sed pontus et aetas mutavere situm. rupit confmia Nereus victor et abscissas interluit aequore montes parvaque cognatas prohibent discrimina terras. nunc illam socia raptam tellure trisulco opponit natura mari : caput inde Pachyni respuit Ionias praetentis rupibus iras ; hinc latrat Gaetula Thetis Lilybaeaque pulsat bracchia consurgens ; hinc indignata teneri concutit obiectum rabies Tyrrhena Pelorum. in media scopulis se porrigit Aetna perustis, Aetna Giganteos numquam tacitura triumphos, Enceladi bustum, qui saucia terga revinctus spirat inexhaustum :flagranti vulnere sulphur et quotiens detractat onus cervice rebelli post 141 in W ethneis P I (s.c.) nat. lar. D K I , 14, ]4 committit K I, C 2 E 2 F4 P I : dimisit 1 5, h i : commendat A2 P 3/4 (marg.) alendam ay : ituram E I F 2 (a.c.) 0 2 (a.c.) R4, C 2 F4 P I : itura F I , F2 (p.c.) 14 0 2 (p.c.) 0 3 P 2, E2 P 3 141 om. F9, post 148 in e2 relegit in C 3 : relegerat M I 141-214 om. classis � omnes, F 4 H 1 2 (habet inter 24I et 242 ita dispositos : I 3 9/ 41/3/4/2/5 et cet. hie ordo postea correctus est) P I, N edd. plerique versui 1 3 9 s11ffragantur, 140 et 1 4 1 ut spurios eiciunt ; Baehrens, del. 140, rescribere voluit : aethera deseruit ; furtim sua pignora terris J commendat fidis Siculasque relegat in oras. sed de lata hac quaestione implicata vide comment. 143 pars magna R I w, A 2 : pars ima Gesner, quod et codd. nonnulli forsitan velint : pars iuncta Birt aetas F I , E 2 ]4 1 3 P 4 R 2 : aestus ay 144 divisere F 9 V 3 situs 1 3 rapuit K I M I, E 2 R 2 145 interluit D J 3 R 7 (a.c.) T , y : -fluit a, ]2/4 1 2 146 cognatis . . . terris Heins. 146-213 om. C 2 147 socia D ? F I M I/2 (p.c.), y : sociam ay raptam C 3 ] 3 (a.c.) 1 s P s (a.c.), y : rapta a, A 2 ]4 P 3 Z: ruptam FI M 3 , ]2 12 U (p.c.) : rupta a, OI trisulco 1 3 et Heins. : -am F I 1 6 M2, E 2 R2(25 : -a ay 148 opposuit F 1 P s R s/7, L2 O I U 149 repulit R s (v.l.) R7 (v./.) : despicit (? e respicit '!') in notis Claver. protentis ljJ : praeruptis R29 151 bracchia G 1 (m.rec.) k n R29 : brachia a�y : litora T dedignata K I , E 2 154 henna C 3 155 saxea U : saxo D K 1 , ]4 membra D K 1 , ]4 156 pectore D K 1 , ]4 : corpore R26 : gutture a G I (m rec ) k (marg.) Y2 virus M2, ]2 : fumum F 16 R 1 7 157 detrectat .

.

130

LIBER PRIMUS

i n dextrum laevumque latus, tunc insula fundo vellitur et dubiae nutant cum moenibus urbes. Aetnaeos apices solo cognoscere visu, non aditu temptare licet. pars cetera frondet arboribus, nullo teritur cultore cacumen. nunc vomit indigenas nimbos piceaque gravatum foedat nube diem, nunc molibus astra lacessit terrificis damnisque suis incendia nutrit. sed quamvis nimio fervens exuberet aestu, scit nivibus servare fidem pariterque favillis : durescit glacies tanti secura vaporis, arcano defensa gelu, fumoque fideli lambit contiguas innoxia flamma pruinas. quae scopulos tormenta rotant? quae tanta cavemas vis glomerat ? quo fonte ruit Vulcanius amnis ? sive quod obicibus discurrens ventus opertis offenso rimosa furit per saxa meatu, dum scrutatur iter, libertatemque reposcens putria multivagis populatur flatibus antra; seu mare sulphurei ductum per viscera montis oppressis ignescit aquis et pondera librat. K I, E2 L 3 : deiectat J 3 et Isengr. marg. : reiectat Heins. reflexa C 3 ? (ante ras.) M2 (v.l.) w et Claver. 158 d.l. -ve L io R s/7, R25 : l.d. -que a, L2/3 : l.d. -ve ]3 R I, A2 P 3/4 Z 159 vertitur F 14 : pellitur P 3 : tollitur Baehrens mutant C 3 P 5 montibus P 3 vers. ita refecit L 2 : vuolvitur et dubios (unde dubio) nutant fundamine turres 160 henneos F I : aerios vel aetherios 161 pars altera \jJ 162 nullo ter. A I excerpta Schottiana contingere U M 2 R I, Lz : t.n. ay : petitur n. a F 6 k Y2 163 vomit Isengr. marg. : movet codd. nubes L 2 164 molibus P 4 U : motibus ay : montibus L 6 T 165 fumificis T, Z flammisqueJeep mittit F I : miscet] 3 166 nim. quam. L2 167 faville Lz 168 caloris L6 M 3 T, Z 170 continuas Mz, P 4 171-8 ut supervacaneos reprehendit ]. C. Scaliger (Poetices libri septem, Lugduni 1561 , p. 322), ut spurios delevit jeep 171 rotent F I, E2 ]4 P 3 R2 Z cavernae Hertel, virgula post rotant posita. 172 glomeret D ? F I K I, E 2 ]4 R2 ruat D K I , ]4 amnis E z P 4 R 2 : ignis ay 174 otfensus rimata F I ruit a, ]4 Lz/3 R25 U : ruat D : furat L B : fluit F 7 ordo per saxa . . . rimosa A I D K I R s/7, ]4 R25 U 176 putria F I } 3 T, y : putrida ay montivagis F9 178 vibrat n .

IJI

16o

165

1 70

1 75

D E RAPTU PRO SERPINAE

ubi servandum mater fidissima pignus abdidit, ad Phrygios tendit secura penates turrigeramque petit Cybelen sinuosa draconum membra regens, volucri qui pervia nubila tractu signant et placidis umectant frena venenis : frontem crista tegit ; pingunt maculosa virentes terga notae; rutilum squamis intermicat aurum. nw1c spiris Zephyros tranant, nunc arva volatu inferiore secant. cano rota pulvere labens sulcatam fecundat humum : fl.avescit aristis orbita ; surgentes condunt vestigia culmi ; vestit iter comitata seges. lam linquitur Aetna totaque decrescit refugo Trinacria visu. heu quotiens praesaga mali violavit oborto rore genas, quotiens oculos ad tecta retorsit talia voce movens : 'salve, gratissima tellus, quam nos praetulimus caelo : tibi gaudia nostri sanguinis et caros uteri commendo labores. praemia digna manent : nullos patiere ligones et nullo rigidi versabere vomeris ictu; sponte tuus florebit ager; cessante iuvenco ditior oblatas mirabitur incola messes.' sic ait et fulvis tetigit serpentibus Iden. Hi e

18o

185

190

195

2oo

179 natam pro pignus R2o 181 turrigeram querit M r spumosa E 2 }4 R2 : squamosa Livineius 182 colla L 6 R r/s/7 T w 183 scindunt R2 : sulcant Goetz 184 frontes M r nitentes R26 185 interviret (? e v. 1 84) A2, prob. Heins., call. D. R . P. II. 1 o o et Cons. Stil. II. 429. 186 leves pro spiris Isengr. marg., unde lenes restituas transnant R 5/7 T, P 3 189 scindunt F I, E 2 } 4 : velant M2, U culmi Isengr. marg. : fruges cxy : messes J 3 L ro M 3 (v.l.) T 192 aborto ex, A2 }2 L2 R25 U 193 imbre genas a F 6 G I (m. rec.) k retorquet R2o : reflexit r 194 gemens pro movens 0 I (m.a. in lac.) 195 cui gaudia P 5, quod recipit Birt 196 claros G 2 et ventris caros V 2 197 post inarata manens L 2 198 vexacommitto R25 dolores L S bere } 3 (v.l.) : violabere M r 199 sponte sua R 2o iuvenca e 2 201 hec ait M 3 serpentibus attigit D K r , }4 : pergit serpentibus L 2 1 32

LIBER PRIMUS

Hie sedes augusta deae templique colendi religiosa silex, densis quam pinus opacat frondibus et nulla lucos agitante procella stridula coniferis modulatur carmina ramis. terribiles intus thiasi vaesanaque mixto concentu delubra gemunt; ululatibus Ide bacchatur ; tumidas inclinant Gargara silvas. postquam visa Ceres, mugitum tympana frenant; conticuere chori ; Corybas non inpulit ensem; non buxus, non aera sonant blandasque leones summisere iubas. adytis gavisa Cybele exilit et pronas intendit ad oscula turres. Viderat haec dudum summa speculatus ab arce luppiter ac Veneri mentis penetralia nudat : ' curarum, Cytherea, tibi secreta fatebor. candida Tartareo nuptum Proserpina regi iam pridem decreta dari : sic Atropos urget, sic cecinit longaeva Themis. nunc matre remota rem peragi tempus. fines invade Sicanos iden E 2 P4 R2 Isengr. : idam ay 202. hec L s , U aedes G4 Aldina templique verendi M 3 203 quam densis a F 6 G 1 (m. rec.) k R2o opacat a, ]2 L2 R25 U: ob umbrat C3 D F 1 L 10, A2 ] 4 0 1 Z (m.a.) 203-14 om. A 2 (add. marg.) E 2 L3 P 3/4 R2 Z (add. m.a. in marg. inf) 205 coniferis a, L2 R25 U : -feri 0 1 Z (m.a.) modulatnr a, A 2 ]2 L2 0 1 : -antur a, J4 R25 U Z (m.a.) sibila L6, R25 rami Z (m.a.) 206 rniro L6 : multo 0 4 207 tremunt e 2 ide D R r , J4 U : yde a, A 2 (marg.) J2 0 1 R25 Z (m.a.) : yda L 6 M 1 (p.c.) M2 P s (p.c.), Lz 208 turnidas ex, A 2 ]2 L 2 R25 : tirnidas a, J4 Or U Z (m.a.) garg. messes PS t 209 mugitns F r6 m n P9 r 210 enses A r L s 211 tonant Wakefield ad Lucr. II. 61 8 blandique a, L 2 0 r 212 dernisere F 9 cybebe edd. 213 intendit K r 0 4 P s, A 2 U Z (m.a.) : extendit a, ] 2 R25 : protendit O r : inclinat L 2 214 hoc F r L 6 M r , U summa dudum A r L IO R r T W , L 2 post hunc vers. v. 1 4 0 insertus est itl C 3 215 at sunt qui velint : hec G r : et L 6 M r/3 R r mentis veneri P s mentis . . . pandit F r P s (ment. ante ven.), L 2 : mentis . . . monstrat D K r , ] 4 : pandit . . . mentis C 3 L 10, R4 216 c.c.t.s.£ a, G r L4 0 3 R 3/6, y, N : c.s.t.c.£ a!3 y 217 nuptn R6, R 2 regi . . . nuptum R2o regi est J4 218 pridem est C 3 Fr R s/7 : dudum D K r , J4 219 iam matte L 2 220 peragi R4, A 2 : -it cx!3y 133

2.05

2.10

2.15

2.2.0

D E RAP T U PRO SERPINAE

225

23 0

235

240

et Cereris prolem patulis inludere campis, crastina puniceos cum lux detexerit ortus, coge tuis armata dolis, quibus urere cuncta, me quoque saepe soles. cur ultima regna quiescunt ? nulla sit immunis regia nullumque sub umbris pectus inaccensum Veneri. iam tristis Erinys sentiat ardores ; Acheron Ditisque severi ferrea lascivis mollescant corda sagittis.' Adcelerat praecepta Venus iussuque parentis Pallas et inflexo quae terret Maenala cornu addunt se comites. divino semita gressu claruit, augurium qualis laturus iniquum praepes sanguineo delabitur igne cometes prodigiale rubens : non ilium navita tuto, non inpune vident populi, sed crine minaci nuntiat aut ratibus ventos aut urbibus hastes. devenere locum, Cereris quo tecta nitebant Cyclopum formata manu : stant ardua ferro moenia, ferrati pastes, inmensaque nectit claustra chalybs. nullum tanto sudore Pyragmon nee Steropes construxit opus ; non talibus umquam spiravere Notis animae nee flumine tanto 221 prolem eer. L4 vaeuis R I 3 222 eandida A 2 0 1 purp ureos F I (p.c.) 223 pelle pro eoge Burm. L 5 M2, F4 L2 R25 direxerit a R 1 3 224 saepe vales L 5 R I, f3, F4 H P I/4 quieseant 0 2 (v.l.), L2 et Heins. 226 inaeeessum J 3 K I L 5, C I F 2, y 227 superbi C I : profundi M 3 228 moll. lase. U 229 mandata L7 R I J/I6/22/26 230 terret quae 0 4 R I, b I, y 231 passu P 5 232 eanduit Isengr. marg., prob. Heins. in orbem D K I , ]4 233 praeeeps a, F4 (p.c.) ]4 L2 O I U dilabitur A I M I W, A 2 L 2 P I : pre- R23 234 tute M2j3 P 5 , C 2 L 2 0 I U 235 nam erine R I3 236 maribus E 2 R2 eivibus 0 1 237 qua af3y nitebant af3, A2 ]4 P 3 : virebant af3y : manebant F I L5 M2, C2 Z : mieabant 0 3 (v.l.) : latebant R26, laudat Livineius 238 formata ] 3 (a.c.) M2, f3, F 4 : firmata af3y : fabrieata P 5 T, R25 Z : fundata '+' 240 numquamtanto '+' Pyraemon (-aehm-F I6) F 1 3 etedd. vett. jere omnes 241 nonster. L 6 eoniunxit F 2 (a.c.) : eommittit R 5/7 nee talibus a, b i 02 P 2, A2 L2 O I P 3 R2 242 folies pro animae M I, v.l. in ay : flammae Z (v.l. in marg.) non F 4 L2 flumine

1 34

LIBER PRI M U S

incoctum maduit lassa fomace metallum. atria cingit ebur ; trabibus solidatur aenis culmen et in celsas surgunt electra columnas.

245

Ipsa domum tenero mulcens Proserpina cantu inrita texebat rediturae munera matri. hie elementorum seriem sedesque patemas insignibat acu, veterem qua lege tumultum discrevit Natura parens et semina iustis

250

discessere locis : quidquid leve, fertur in altum ; in medium graviora cadunt ; incanduit aer ; egit :flamma polum ; fluxit mare ; terra pependit. nee color unus erat : stellas accendit in auro, ostro fundit aquas. attollit litora gemmis

255

fi.laque mentitos iamiam caelantia :fluctus arte tument. credas inlidi cautibus algam et raucum bibulis inserpere murmur harenis. addit quinque plagas : mediam subtegmine rubro obsessam fervore notat ; squalebat inustus limes et adsiduo sitiebant stamina sole ; vitales utrimque duas, quas mitis oberrat D ? }3 M r 0 4 R r (a.c. ), E r F 2 L r R 3 , L 3 (a.c.) P I/3/4 Z : flamine a, L4 (p.c.) 0 2 P 2 R4, y, N : fulmine W, !3: stamine C 3 : numine R2 1 243 laxa C 3 L 6 W, E 2 F 4 }4 R 2 : lapsa J J , C r G r R6 1assa maduit a F6 k Y2 fornace F4 (v.l.) et Isengr. marg. : cervice codd. 244 cinxit C r : vestit W, A2 (v.l.), recipit Birt 246 mulcens tenero C 3 D ? K r L IO M2, R4, }4 O r 248 series D K r , G r R6, A 2 H }4 O r (a.c.) U sedemque patemam L S 250 discernit C r , L2/3 R 2 : discrerit Heins. u t Pa"has. e t Heins. iussis M 3 252 ruunt T , R 2 iam canduit L s aether A I D K r M 2 W , B L I , }2/4 253 legit M2 (p.c.) P s (a.c.), B L r : cingit W : texit e r 254 non D K r L6, }4 inest D K r , E2 } 4 255 pingit aquas L s , L 2 256 subter pro iamiam D K r , }4, unde subito Burm. celantia olim intellegebam : curvantia Isengr. marg. tanta simulantia Burm. : iamiamque levantiaHeins. 257 tre­ mont laudat Pa"has. allidi F 7 algas a, G r L 4 0 2 (p.c.) 0 3 R6, y 258 insurgere F 7 R r 3 marmor Schrader 259 subtemine D ? M 3 , E r (a.c.), A2 P 3 nigro 'V 260 adustus D K I, b r , E 2 J 4 R 2 261 limns 262 vitalis r Baehrens fervebant R26 flamina } 2 : gramina e2 et Jeep obumbrat a F6 k R 2o

135

26o

DE RAPTU PROSERPINAE

265

270

275

28o

temperies habitanda viris ; tum fine supremo torpentes traxit geminas brumaque perenni foedat et aeterno contristat frigore telas. nee non et patrui pingit sacraria Ditis fatalesque sibi Manes ; nee defuit omen, praescia nam subitis maduerunt fl.etibus ora. Coeperat et vitreis summa iam margine texti Oceanum sinuare vadis ; sed cardine verso cernit adesse deas inperfectumque laborem deserit et niveos infecit purpura vultus per liquidas succensa genas castaeque pudoris inluxere faces : non sic decus ardet eburnum Lydia Sidonia quod femina tinxerit ostro. Merserat unda diem; sparso nox umida somno languida caeruleis invexerat otia bigis, iamque viam Pluto superas molitur ad auras germani monitu. torvos invisa iugales Allecto temone ligat, qui pascua mandunt Cocyti pratisque Erebi nigrantibus errant stagnaque tranquillae potantes marcida Lethes aegra soporatis spumant oblivia linguis : 263 habitata G2 tunc N : con- D ? K r, F2 G r, R2 : in a, C 2 E 2 L2 O r : sub 0 3 (p.c.) R4 264 texit 0 4, L I, J 4 265 celat A 2 constringit A r C 3 D K I P s, E2 H ]4 R2 : circumdat F 7 R 1 3 Y I telam C 3 L ro M r P s W, P 2 (p.c.) R4, H L2/3 266 fmgit K 4 R2r sacratria m (a.c.) : penetralia ' in ora libri vet.' Claver. dit. sacr. ping. W 267 non R r3 deficit R r7j26 268 nescia Burm. sed pro nam L6 R I 269 summo de D K I, ]4 : summa iam 0 2/3 (a.c.) : supremo R r (in subiuncto), 0 3 (p.c.), C2: suprema F I : i n summo T 4 tele F I, Lr 0 2 (sscr.) P 2 (sscr.) 270 simulare C I : curvare C 3 271 sensit D K I, J 4 : sentit F 7 h R I 3 r 272 desinit F 7 R r 3 273 incensa M 3 castique K r L s M2, E 2 ] 4 275 temperat W post hunc vers. lacunam statuit Jeep 278 vias M I R I w, Lr sup. pluto L 3 R 2 : pluto siculas D, ]4 in ]2 (m.a. in lac.) L2 Or auras ay, N: arces apy : oras Q (? m.a.) 279 torvosque F 2 : nigros A2 281 spatiisque K I, E2 J 4 (v.l. ) R2 282 potantes tranquillae F I M I , R25 283 pigra F. R. D. Goodyear soporiferis 04 spirant L ro, Cr, Z

LIBER PRIMUS

Orphnaeus crudele micans Cthoniusque sagitta ocior et Stygii sublimis gloria Nycteus armenti Ditisque nota signatus Alastor. stabant ante fares iuncti saevumque fremebant crastina venturae spectantes gaudia praedae. 284 Orphnaeus Parrhas. et G4 : orpheus N: orfeus R7 (a.c.) : omeus codd. Chthoniusque jeep (sed cf Schulze, Ortho­ plerique minans F I4 R2o graphica) : Aethonque Parrhas. (aetonque iam n) : ethonusque vel simm. codd. plerique sagittis M I 285 nycteus ljJ : nict(h)eus J J R I/7. h i L4 0 2, A 2 P 3/4 Z : nocteus M 3 : morpheus F I : mot(h)eus L6 M I T W, !3y, N 286 notis R I 3 alastor A 2 : alaster v el simm. ceteri 287 vincti sunt qui codd. velint seveque M 3 R s/7 (p.c.) : secumque F 7 288 spect. vent. R2o sperantes hi, ]4 (v. l.), quam lectionem Housman ad luv. 7.22 videtur probare : speculantes 0 4 : spirantes Goetz praemia A I C 3 R I , A2 P 3 Z palmae Z : lucis U

137

285

PRAEFATIO LIBRI S E C UNDI Otia sopitis ageret cum cantibus Orpheus neclectumque diu deposuisset ebur, lugebant erepta sibi solacia Nymphae, quaerebant dukes £lumina maesta modos. s

saeva feris natura redit metuensque leonem inplorat citharae vacca tacentis opem. illius et duri flevere silentia montes silvaque Bistoniam saepe secuta chelyn. Sed postquam Inachiis Alcides missus ab Argis

10

Thracia pacifero contigit arva pede diraque sanguinei vertit praesepia regis et Diomedeos gramine pavit equos, tum patriae festo laetatus tempore vates desuetae repetit ftla canora lyrae

15

et resides levi modulatus pectine nervos pollice festivo nobile duxit opus. Praefatio A2 (Pulmannus) : Liber secw1dus de r.p. F 2 : Liber secnndus F 4 (m.a., marg.) : PROLOGUS SCDI LIBRI G I : Prologus libri II L 3 (marg.) : Primus liber de r.p. explicit incipit prologus IJ P 3 : Liber ii de r. f P. P 4 (m. recentiss., marg.) : Incipit IJ ' liber W Heinsius praefationem alieno loco insertam ideoque fortasse tollendam opinatus est, etJeep audacter sane a libris de raptu seiunctam edidit. uterque iniuria, ut in prolegom. p. 94 demonstrare temptavi. 2 seposuisset Claver. : deseruisset L8 ebur F2 (a.c.), E2 (v.l.) : opus a�y 3 quaerebant R II amissa 0 I solamina e suo ms. Delrius 4 lugebant G2 O s/6 dulcis R I9/3o : tenues U 5 leonum Isengr. 6 iacentis L I opes d 7 istius F 4 P I eduri Claver. : Edoni Schrader 8 -iam . . . liram D ?, }4 Z : -ium . . . melos r 9 arvis ay : agris C I (p.c.) R6, O r 10 pacifica P s R7 (v.l.) II duraque (m.a.), N : oris a, C 2 }4 R25 Z 12 -eas . . . equas L6 sanguine Kr 0 4 M r P s, A2, N: nigraque R25 (s.c.), 0 2, A 2 } 4 P 3 pascit L 6 M r 1 3 tum M 3 0 4, �y : tnnc a �y 14 reparat Isengr. marg. : reparit 0 1 (a.c.) : coepit MS Pulmanni sonora A I M2j3 T, H L2 P 4 : movere MS Pulmanni 15 residens 0 4 ? R r , �. A 2 nervos . . . leni L 8 } 4 (a.c.) R 2 leni codd. nonnulli leni resides R s/7 moderatus R s/7 e t Gronovius (It1 Stat. Silv. Diatribe, ed. Hand, I pp. 1 02-3) 16 festino a, b r , y, N mobile w (a.c.), b r F2 L r 0 2 P 2 (a.c.), E 2 }4 O r 13 8

PRAEFATIO LIBRI SE CUNDI

vix auditus erat ; venti frenantur et undae, pigrior adstrictis torpuit Hebrus aquis, porrexit Rhodope sitientes carmina rupes, excussit gelidas pronior Ossa nives ;

20

ardua nudato descendit populus Haemo et comitem quercum pinus arnica trahit, Cirrhaeasque dei quamvis despexerit artes, Orpheis laurus vocibus acta venit. securum blandi leporem fovere Molossi vicinumque lupo praebuit agna latus. concordes varia ludunt cum tigride dammae, Massylam cervi non timuere iubam. Ille novercales stimulos actusque canebat Herculis et forti monstra subacta manu ;

30

qui timidae matri pressos ostenderit angues intrepidusque fero riserit ore puer : ' te neque Dictaeas quatiens mugitibus urbes taurus nee Stygii terruit ira canis, non leo sidereos caeli rediturus ad axes, non Erymanthei gloria mantis aper. P4 R2, N opus F r L6 W, 02 (v.l.) P2 (v.l.) : ebur al3y 17 sternuntur D ? K I R 7 (v.l.), E 2 J 4 R2 18 adfectis e suo ms. Delrius 19 sltlentis R I9/30 montes (-is y) D K I , ] 4 : cautes P S 21 desc. nud. L 4 2 2 mirtum pinus F I I 2 3 cirr(h)eique L6 M r R I (p.c.) R s/7, A2]2 P 3 Z despexerat C 3 L io W, 0 2/3 P 2 R3/4. H L 3 , N : dispexerat P s, prob. Buecheler despexit amores Schrader {despexit iam C I } 24 orpheis a, G r L r , y : -ei al3y cantibus acta e r : voce coacta R 6 25 lep. blandi F I : blandi cervam (cum securam) T 26 ieiunoque C2 latus . . . lupo L io, 02 P2 R4 : 28 massylam ( il ) ] 3 (p.c.) K I, I3, A2 E2 cerve latus . . . lupisjunii pr. R 4 nee a 29 stimulosque K r astusque (sc. Iuno11is) G2 et Hei11s. 31 qui ] 3 (p.c.) L6fro, L r , F4 P I : quod al3y : quo a, H ]4, N (p.c.) : cum n : u t 0 3 (p.c.) trepidae L s strictos ] 3 , C I G I L4 R 6 -eris O s, quod malebat HeitlS. 32 fcrox F4 P r : feros Barth puer . . . fero L ro P s , R4 -eris malebat Hei11s. 33 nee L ro T: non F r4 R ro mugit. auras 0 2 P 2 : mugit. edes a 3 4 nee a, 0 3 P 2 R6, y : non al3y 3 5 nee C 3 R r T, L2 U -eas C 3 W, b r , H arces C 3 L s M3 W, b r, H -

139

-

35

DE RAPTU PRO SERPINAE

solvis Amazonios cinctus, Stymphalidas arcu adpetis, occiduo ducis ab orbe greges tergeminique ducis numerosos deicis artus 40

et totiens uno victor ab hoste redis. non cadere Antaeo, non crescere profuit Hydrae, nee cervam volucres eripuere pedes. Caci flamma perit, rubuit Busiride Nilus, prostratis maduit nubigenis Pholoe.

45

te Libyci stupuere sinus, te maxima Tethys horruit, inposito cum premerere polo : firmior Herculea mundus cervice pependit ; lustrarunt umeros Phoebus et astra tuos. ' Thracius haec vates. sed t u Tirynthius alter,

so

Florentine, mihi : tu mea plectra moves antraque Musarum Iongo torpentia somno excutis et placidos ducis in orbe choros. 37 solvis C3 F I (p.c.) ] 3 (p.c.) L S/Io, C I P2, L3 Z (p.c.) : solus exl3y amazonio K I L6 P 5 (p.c.) R I (p.c.), G I L I/4 R6, F4 (p.c.) ]4 0 I 39 vm pro ducis Koch, coli. Ov. Trist. 4.7.16 nervosos Rzo : numero tot ].]. Scaliger. sed cf. Claud. c.m. 22.58 proicis W : dividis K I : decutis e2 : disicis Heins. olim 40 victo victor R I 3/I6jzo 41 nec M 3 P 5 , N nee F I L 5 M2/3 w, b i, L 2 42 nee C 3 F I K r 0 4 P 5, ]4 U : non exl3y cervum L io, L I 0 3 P z, F 4 P r R25 (p.c.) fort. recte ({{. Auson. xxxiii.4 Schenk/, et Anth. Pal. 16.92.4) : nessum M I/3 T, A2 Z (sed, ut obs. Heins., Nessus occisus inter Herculis labores, de quibus hie agitur, non numerandus est) celeres } 3 , C I G r L4 R 3j6, A2 U Z celeres cervam J 3 43 rubuit ex, b I, y : -itque o:l3y busiridis ara J 4 44 rubuit K I L6, E 2 (a.c.) ]4 R2 Z 45 libie L 6 timuere D K I , }4 : tremuere L 8 maxima tethis L 6 Mz, C I O z, Jz : maxima thetis exl3y : maximus atlas D K r , E 2 }4 R2 47 fortior D K I , ]4 celum L 2 48 -aruntque R s/7 : -avitque a 5 0 nam mea K I , }4 : qui mea M 3 : nunc mea L s 51 largo Rzo 52 placidos ex, E I F 2, y, N: -ido exj3y : -ito 0 3 (a.c.) Claver. : -itos Y I ab D F I K I L s , A2 C 2 ] 4 ore sonos D K r , }4

LIBER S E C UND U S Inpulit Ionios praemisso lumine fluctus nondum pura dies ; tremulis vibratur in undis ardor et errantes ludunt per caerula flammae. iamque audax animi fidaeque oblita parentis fraude Dionaea riguos Proserpina saltus

s

(sic Parcae volvere) petit. ter cardine verso praesagum cecinere fores ; ter conscia fati flebile terrificis gemuit mugitibus Aetna. nullis ilia tamen monstris nulloque tenetur prodigio. comites gressum iunxere sorores.

10

Prima dolo gaudens et tanto concita voto it Venus et raptus metitur corde futuros, iam d urum flexura Chaos, iam Dite subacto ingenti famulos Manes ductura triumpho. illi multifidos crinis sinuatur in orbes De r.P. lib. II A2 (Pulmannus) : Liber Tertius [sic] F4 (m.a. marg.) : secundus liber claudii claudiani de r.p. G r : incipit liber n' claudiani in raptu pros. L 3 : secundus liber incipit M 3 (marg.) : explicit primus [sic] liber incipit secundus P 3 : Claudiani poetae Alexandrini de r. p. liber secundus M4 R r9 I flamine Baehrens 2 clara dies C 3 R r , J 4 (sscr.) tremulus R6 vibrabat a : vibravit ed. Raphelengiana, laudat Heins. 4 animo 0 3 (a.c.) : anime H, N 6 ut parce R2o volvere (voluere) A r L ro M2 T, 0 2/3 P2 R4 : iussere aj3y : suasere e suo ms. Delrius tunc K r R s w, E 2 ]4 (v.l.) : sed O r 7 praesagum W, E r L r (p.c.) R6, ]4 L 3 P 4 : -e K r 0 4 (p.c.) R r (a.c.) R s/7, F2 R4, A2 E2 P 3 U Z : -ium aj3y tunc K 1 , }4 O r : tum w facti D K r R 1 , }4 (v.l.) 8 erma 0 2, H]2 (s.c.) 9 movetur R r9 10 gressum com. R2o : com. gressus A r L ro, F 3 0 2 P 2 R4 movere T II dolis 0 3 (p.c.) tanti A r F r K r M r T, R6 (p.c.), F4 (p.c.) ]4 L2 U concita 0 3 (a.c.) : con (s)cia aj3y : callida Claver., prob. Heins. voti A r F I T, R 6 (p.c.), U: facti K r , L2: fato Ls, } 2 : furti M r , F 4 (p.c.) ]4: furta F 4 (a.c.) cauti callida voti Heins. 12 meditatur w, L r , U : molitur]eep 13 dirum a, b r 0 2 P2 R 4, y : dudum a, C r L r 0 3 (a.c.) R 3 , P r (p.c.) R25 (a.c.) U (p.c.) : demum R 14/26 domitura Isengr. marg. 14 manes famulos F 3 , L 3 I S multiplices a, E r F2, y, N : -modos A r W, L 2 O r variatur K r, E 2 ]4 R2: curvatur N: nodatur Rr6 sinuatur crinis R25 orbis Rr (a.c.) I4I

IS

D E RAPTU P R O S ERPINAE

20

25

30

35

Idalia divisus acu ; sudata marito :fibula purpureos gemma suspendit amictus. Candida Parrhasii post hanc regina Lycaei et Pandionias quae cuspide protegit arces, utraque virgo, ruunt : haec tristibus aspera bellis, haec metuenda feris. Tritonia casside fulva caelatum Typhona gerit, qui summa peremptus ima vi get, parte emoriens et parte superstes ; hastaque terribili surgens per nubila ferro instar habet silvae; tantum stridentia colla Gorgonos obtentu pallae fulgentis inumbrat. at Triviae Ienis species et multus in ore frater erat, Phoebique genas et lumina Phoebi esse putes, solusque dabat discrimina sexus. bracchia nuda nitent; levibus proiecerat auris indociles errare comas, arcuque remisso otia nervus agit ; pendent post terga sagittae. crispatur gemino vestis Gortynia cinctu poplite fusa tenus, motoque in stamine Delos errat et aurato trahitur circumflua ponto. QEas inter Cereris proles, nunc gloria matris, mox dolor, aequali tendit per gramina passu 17 succingit T 21 cuspide LS flava L r 22 tiphona F3 L r , H P I/3/4 : tritona A2 F4 (a.c.) : phitona (fi-, phy-) a!3y : pythona O s/6 qui parte F I 23 ima viget parte moriens C I , unde veram lectionem expiscatus est Koch. sed antecessit Gusttifssotz, ut vid. suo Marte, et ima viget iam Heins. viderat : ima parte viget moriens codd. ceteri in parte sup. M 3 T W, b I, L 2 : pariterque sup. Jeep ima viget moriens partim partimque sup. Heins. moriens et parte sup. interpolatorem redolere existimat Baeltrens 24 ad nub. W : in nub. R2o giro (gyro) M 4 ? (a.c. ) n P9 u 25 habet K I M3 T, A2 E 2 J 4 P 3 R2 : erat a!3y tandem str. C I : raucum str. olim conieci. rauco cum tanto confunditur infra ad III. 242, et rauco stridore legitur apud Ovid. Met. 8.287, 1 4.1 oo 26 gorgonos J 3, C I G I R6 : -is a!3y obiectu R I, L I : obtectu K I L6, b i , H obumbrat ay 27 facies et ] 3 27-8 frater . . . f multus 29 facit pro dabat L 5 30 bracchia n Z 28 fratcr inest M 3 , F 3 p I : brachia vett. patent L I o R I euris r 33 vestis gemino L 5, L 3 36 inter quas M2 tunc M3, F2, C2: tum M2 37 tendens E 2 ]4 R2

LIBER S E C U ND U S

nee membris nee honore minor potuitque videri Pallas, si clipeum ferret, si spicula, Phoebe. collectae tereti nodantur iaspide vestes.

40

pectinis ingenio numquam felicior artis contigit eventus ; nulli sic consona telae fila nee in tantum veri duxere figuras. hie Hyperionio Solem de semine nasci fecerat et pariter, forma sed dispare, Lunam,

45

aurorae noctisque duces ; cunabula Tethys praebet et infantes gremio solatur anhelos caeruleusque sinus roseis radiatur alumnis. invalidum dextro portat Titana lacerto nondum luce gravem nee pubescentibus alte

so

cristatum radiis : primo clementior aevo fingitur et tenerum vagitu despuit ignem. laeva parte soror vitrei libamina potat uberis et parvo signatur tempora cornu. Tali luxuriat cultu. comitantur euntem Naides et socia stipant utrimque caterva, quae fontes, Crinise, tuos et saxa rotantem Pantagian nomenque Gelan qui praebuit urbi proserpina U gressu L6 M3 R 5/7, E I, }2 : cursu K I 38 non membris W, b I 39 si ferret spicula 0 2 P 2 (s.c.) R4 41 artis C 3 F I J 3 M2 P 5 (a.c. ), C I F 2 (a.c.) 0 3 (a.c.), C2 ]2 (a.c. ) ] 4 : arti aj3y : arte L I 42 nulle 43 veras P 5 (a.c.) T, R4 figuram D ] 3 (a.c.), j3y, N conscia F I L6 M I K I 0 4, U 44 de sanguine Junii primus 45 sed forma L io M 3 , A 2 lunam . . . forma a, J 2/4 U Z 4 6 noctisque vices L 4 (in ras.) ljJ tethys D : tethis L6, F 2, J 2 : thetis aj3y 47 sol. hanili J 4 48 caeruleosque R23 50 necdum F 2, C 2 R2 non pubesc. 02 52 pingitur C3 KI L 5/6 R I (p.c.), P 2, Z : tingitur R 9 tenero R 4 post hunc vers. dtjicit N 53 mve1 0 3 (a.c.) 54 signatur L io R I T, C I E I F 2 G I 0 3 , ]2 L 3 P 3 R25 : 55 euntes D, E x 56 naides F x , b x -antur aj3y cornua gyro W R4, R25 : naiades aj3y socie . . . caterve W : socia . . . corona D K I , }4 R2 stipantur L6, 0 2, H L2: -atur M 2 57 crinise (-ss-) a, C 1 E 1 F 2 G 1 L4, y : crisine aj3y : erisine 0 I : erasine D : Amasene Brodaeus apud Delrium quae saxa L7 R I4j26 : et stagna B moventem O x 58 pantagian } 3 , F 2, 0 1 : -am vel a aj3y gelan A 1 J 3 L 10 M xj2, i3 : -am vel -a aj3y -

1 43

ss

D E R A P T U P R O S ERPINAE

6o

65

70

75

so

concelebrant, quas pigra vado Camerina palustri, quas Arethusaei latices, quas advena nutrit Alpheus (Cyane totum supereminet agmen) : qualis Amazonidum peltis exultat aduncis pulchra cohors, quotiens Arcton populata virago Hippolyte niveas ducit post proelia turmas, seu flavos stravere Getas seu forte rigentem Thermodontiaca Tanain fregere securi ; aut quales referunt Baccho sollemnia Nymphae Maeoniae, quas Hermus alit, ripasque patemas percurrunt auro madidae : laetatur in antro amnis et undantem declinat prodigus umam. Viderat herboso sacrum de vertice vulgus Aetna parens florum curvaque in valle sedentem conpellat Zephyrum : ' pater o gratissime veris, qui mea lascivo regnas per prata meatu semper et adsiduis inroras flatibus annum, respice Nympharum coetus et celsa Tonantis germina per nostros dignantia ludere campos. nunc adsis faveasque, precor; nunc omnia fetu pubescant virgulta velis, ut fertilis Hybla invideat vincique suos non abnuat hortos. quidquid turiferis spirat Panchaia silvis, 59 nigra L4, F 4 P I camarina J 4 60 areth. fontes T 62 aduncis A I T w, h i E I F 2 (a.c.) L 4 R 4, A 2 ]4 L 3 P 3/4 : ademptis aj3y 63 arthon quotiens L 6 64 ducit niveas A 2 : viduas ducit Heins. : vacuas ducit Birt turmas . . . ducit M2 65 seu fonte O I 66 tanain E I R4, C2 R 2 : -im vel -i aj3y 67 bacho referunt P 5 68 hemus L 6 : hebrus L 10 virentes e I 69 circueunt A 2 madidas F 1 4 : gravide M 3, A2 P 3 Z : gravidas ed. Scaligeri Raphelengiana 71 umbroso ]unii primus de cespite \jl 72 (h)enna 0 2, O I 74 meatu D F I K I , ] 4 : volatu o:j3y 75 arvum L6, F 4 (p.c.) e t Jeep olim e coni. 76 turbas e t M 3 77 agmina M 3 , R 3 78 hue adsis P 5, P I U foveasque M 3 79 pinguescant C I v el his (sc. puellis ?) interpretatus est Heins. : novo R I 3 U nunc pro ut R I 3 : et G4 P7/8 p2 u floribus M 2 T So abnegct C 3 L 5 M2j3 P 5, F 2 0 2, 0 1 : deneget W 81 odoriferis M 3 , C I G I L4 R6, C 2 H : roriferis et floriferis 144

LIBER SE CUND U S

quidquid odoratus Ionge blanditur Hydaspes, quidquid ab extremis ales longaeva Sabaeis colligit optato repetens exordia busto, in venas disperge meas et flamine largo rura fove. merear divino pollice carpi et nostris cupiant ornari numina sertis.' Dixerat; ille novo madidantes nectare pinnas concutit et glaebas fecundo rore maritat, quaque volat vermts sequitur rubor; omnis in herbas turget humus medioque patent convexa sereno. sanguineo splendore rosas, vaccinia nigro imbuit et dulci violas ferrugine pingit. Parthica quae tantis variantur cingula gemmis regales vinctura sinus ? quae vellera tantum ditibus Assyrii spumis fucantur aeni ? non tales volucer pandit Iunonius alas, nee sic innumeros arcu mutante colores vv.ll. in R7 spargit cod. Bartholin. 82 odoriferus R25 large bland. R I 3 : blande largitur T 83 externis K I, J 4 longevus J 3, C I Sabaeis Isengr. marg. : colonis af3y : harenis ] 3 , C r (p.c.) : ab horis C r (a.c.) 84 optati D K r P s (a . c.), ]4 L 2 referens C 3 D K r W, b r , ]4 p rimordia C 3 F I L s busto Isengr. marg. : seclo af3y : secli D K r P 5 (a.c.), G r ? , ] 4 L z : celo F4 (p.c.) : fato R s/7, U : partu ] 3 M r W , C r F z (a.c. ) G r L4 R6, Rzs : leto Heins. optata . . . morte Baehrens 85 ex(s)perge ] 3 , C r L 4 : asperge r : diffunde (inf- T : eff- R 1 3 ) C z flumine R r (a.c.), [3, A 2 (a.c.) L 3 P I/3/4 sacra R s/7 86 fove ut merear a, B (ut sscr.), y tangi L 5, F2, L2 87 foveant J 3 , unde voveant vel faveant Heins. 88 madefactas (-us R22) 89 excutit k : concitat U L ro, 02 P2 R4, O r : manantes J. C. Scaliger (p.c .) 90 sequitur vemus 0 4 : veris seq. L s M 3 , L 2 : vernans seq. L6, Z : seq. vernans R 19 color D K 1 , ] 4 : calor n et ex suo Delrius 9 1 madidoque P 4, prob. Heins. : nimboque Baehrens : radioque Birt : nitidoque ego olim 93 induit D F I K I R I cf Sil. 4.12-13 violas dulci 0 4 tingit L6 94 non pro quae L6, O r : nee L s gemrnis . . . tantis Mz : tantis . . . donis Parrhas. 95 cinctura e r m 96 succis pro spurnis U (a.c.) fucantur E2 ]4 (v.l.) L 3 : fuscantur af3y : variantur Rz : ditantur M3 97 nee R r , 0 I talis K r volucris pandit A r (s.c.) D ? K r P s, F2, ] 4 O r (a.c.) : pand. volucer L ro M2, ]2 L 3 iunonius ales A I (s.c.) C 3 (s.c.) L s Mz, B L4 0 3 (p.c.) P 2 (p.c.), O r (s.c.) : iunonia pennas D K r , ]4 vers. ita refecerunt R s/7 = non tales alas pandit innonius ales 98 nee C 3 D K 1 M 1 R r , [3, L z ? 0 I : non af3y 10

1 45

HCD

ss

90

95

DE

100

105

no

ns

RAPTU P R O S ERPINAE

incipiens redimitur hiemps, cum tramite flexo semita discretis interviret umida nimbis. Forma loci superat flares : curvata tumore parvo planities et mollibus edita clivis creverat in collem; vivo de pumice fontes roscida mobilibus lambebant gramina rivis, silvaque torrentes ramorum frigore soles temperat et media brumam sibi vindicat aestu : apta fretis abies, bellis accommoda comus, quercus arnica Iovi, tumulos tectura cupressus, ilex plena favis, venturi praescia laurus ; fluctuat hie denso crispata cacumine buxus, hie hederae serpunt, hie pampinus induit ulmos. haud procul inde lacus (Pergum dixere Sicani) panditur et nemorum frondoso margine cinctus vicinis pallescit aquis : admittit in altum cernentes oculos et late pervius umor ducit inoffensos liquido sub flumine visus imaque perspicui prodit secreta profundi. [hue elapsa cohors gaudet per florida rura] diversos L2 variante 0 4, E 2 : monstrante R2 99 facto 0 4 R 5/7, E I , J 2 : ficto R 2 5 Z : picto Isengr. marg. 100 disiunctis L 5 intemitet A I ? C 3, F 4? (p.c.) J 4 et Jeep e coni. : -micat e I R2o 101 flor. for. loc. sup. L 5 flor. sup. H 103 colles R 10/20 vivi n et Jeep : niveo P I : missi laudat Parrhas. 104 humida (mob.) P 2 (sscr.) germina K I ripis A I C 3 105 ram. torr. L s : torpentes ram. L io M I P s R 5/7, C2 U (a.c.) 106 modico K I , C I, E2 (s.c.) }4 (s.c.) R25 vindicatj n : vendicat ceteri post hunc vers. lacunam statuit Jeep 107 accommoda pinus A 2 E 2 (s.c.) P 3 108 amata 0 3 iovis M I textura M 2 P 5 : pressura 0 4 , B cypressus L 5 , L2 109 nuntia laurus F I no estuat 0 2 P 2 hie aj3, E 2 }2/4 O I (a.c.) Z : hinc L 6 : et M I : in aj3y curvata R I 3 : stipata R I6 111 hinc . . . hinc F I , }2 ulmus A 2 1 12 coloni F 3 (cf. app. ad v. 83 supra) : priores A2 C2 113 margine tectus A2 P 3 114 clarescit R I 3/I6 : albescit Wakefield ad Lucr. II. 390 116 sub gurgite D K I L 5, J 4 117 promit A I : pandit T 118 hie vers., quem primus Bonnet damnavit, exstat in his vett. : C 3 D K I L s M2/3 R I/5/7, R25 ; in marg. legitur in A I (appictus a Pulmanno) F I M I (m.a.) w, G I (m. rec.) L4, 0 I ; post 1 1 9 positus est in U delapsa R 5/7 gaudens C 3 ? M 2 : gaudent F I4 k florea A I D K I M I R 5/7, L 4, R25

146

LIBER S E C U N D U S

Hortatur Cytherea legant : ' nunc ite, sorores, dum matutinis praesudat solibus aer, dum meus umectat flaventes Lucifer agros roranti praevectus equo.' sic fata doloris carpit signa sui. varios tum cetera saltus invasere cohors : credas examina fundi Hyblaeum raptura thymum, cum cerea reges castra movent fagique cava dimissus ab alvo mellifer electis exercitus obstrepit herbis. pratorum spoliatur honos : haec Iilia fuscis intexit violis ; hanc mollis amaracus ornat; haec graditur stellata rosis, haec alba ligustris. te quoque, flebilibus maerens Hyacinthe figuris, Narcissumque metunt, nunc incluta germina veris, praestantes olim pueros : tu natus Amydis, hunc Helicon genuit ; te disci perculit error, hunc fontis decepit amor ; te fronte retusa Delius, hunc fracta Cephisos harundine luget. Aestuat ante alias avido fervore legendi frugiferae spes una deae : nunc vimine texto ridentes calathos spoliis agrestibus inplet; 118 et 119 ita con.flavit 0 4 : hortatur citharea legant gaudent per florida rura 120 ether R2o 121 florentes L s M 3 T, A2 L 2 U Z 122 provectus

K r L6, R6, F 4 } 4 L 3 doloris a, E r F2 R4/6 (v.l.), y : cruoris J J , j3, L3 P 4 123 legit signa L2 tum J J , C r G r 0 2 P 2 : tunc aj3y 126 cavo M r , L 2 : cavae Jeep demissus a , C r E r F2 R4, y 127 astrepit w, 0 2 P 2, F4 H P I arvis M 3 (p.c.) 128 flavis F I : fulvis R 3 , F4 (p.c.) O r 130 velata rosis '+' 131 maerentem Paul 132 legunt C 3 F I K I L 5 M r , }4 : petunt M 3 (p.c.) tunc M 3 , Z germina L ro ? M I/2 P s T, E r L4, 134 te y: gramina aj3y 133 olim praest. U pueros olim C 2 L2 disci a, b I E I F 2 L4 0 3 (p.c. ), y : disci te aj3y percutit 0 I : perdidit '+' horror K r , C r G r , E2 H 135 hunc forme U (p.c.) fronde R I , F 3 0 3 , P 4 retusa a , C r E r P 2 (a.c.), y : recussa aj3, F 4 H J4 : revulsa A I : recisa L s R r , P 2 (p.c.) R4/6, C 2 O r R25 U (a.c.) : remissa P r 136 hunc strata e I cephisos p 2 et cod. Bartholin. : -us ceteri 137 avide J 3 138 texto L ro M r , j3y : textos a, b i F2 G I (p.c.) 0 3 R6, y 139 ridenti (cum textos) Heins. olim et Schrader foliis K I M 3 , O r 1 47

I0·2

uo

125

1 30

135

D E RAPTU P R O S ERPINAE 1 40

1 4s

1so

I SS

16o

nunc sociat fl.ores seseque ignara coronat, augurium fatale tori. quin ipsa tubarum armorumque potens dextram, qua fortia turbat agmina, qua stabiles portas et moenia vellit, iam levibus laxat studiis hastamque reponit insuetisque docet galeam mitescere sertis : ferratus lascivit apex horrorque recessit Martins et cristae pacato fulgure vernant. nee quae Parthenium canibus scrutatur odorem aspernata choros libertatemque comarum iniecta voluit tantum frenare corona. Talia virgineo passim dum more geruntur, ecce repens mugire fragor, confligere turres, pronaque vibratis radicibus oppida verti. causa latet ; dubios agnovit sola tumultus diva Paphi mixtoque metu perterrita gaudet. iamque per anfractus animarum rector opacos sub terris quaerebat iter gravibusque gementem Enceladum calcabat equis : inmania fi.ndunt membra rotae pressaque Gigas cervice laborat Sicaniam cum Dite ferens temptatque moveri I40 colorat w I4I quippe ipsa D, E2 ]4 I42 quae R23 I43 quae R23 portas stabiles E2 tollit R s/7 I44 lassat L6 R I (a.c. ) R s /7 w, C I , A 2 P I/3 U Z I4S insolitisque D K I , J 4 I46 recedit L 10 P 5 (a.c.) I48 nee quae R6, L 3 P 3 : I47 placato D K 1 , ] 4 O r fulmine M 3 , ]2 hec quae a[3y parthenios C 3 , R4, L 2 circumdat F 4 P I odores C 3 L w, 0 2 (p.c.) P 2 R 4 (p.c.) : odoris D , R 3 , E 2 ]4 R25 : odorum 0 4 et]eep e co11i. : odorans L2 149 t(h)oros nonnulli jeep, qui haec quae et toros (CIIm F I) itl vv. 1 48-9 legeret, lacunam inter toros et libertatemque statuit. tiOn haerebis si nee quae reccperis ISO innexa F 4 P I : inserta F7 R I 3 tantum voluit P s : voluit tandem A I K I M r/2 (a.c. ) , ] 4 O r P 4 : tandem voluit W IS2 recens D K r , ] 4 : frequens ]4 (v.l.) IS3 ipsaque 0 3 (v.l. ), tmde lapsaque Burm. vibr. de sedibus A r : verticibus vibr. F 2 velli G I R6 IS4 dubiosque F4 Pr agnoscit G I L 4 R6, R 2 ISS meritoque R25 : gratoque Koch IS6 namque C I animorum P 4 opacum J 4 : -is r trementem ] 3 K I L 5 IS7 sub tenebris tendebat r graviterque a L S ISS calcavit R I I6o tenens C 2 : gerens M 4 P 9 R 19

LIBER SECUNDUS

debilis et fessis serpentibus inpedit axem; fumida sulphureo prolabitur orbita dorso. ac velut occultus securum pergit in hostem miles et effossi subter fundamina campi transilit elusos arcana limite muros turbaque deceptas victrix erumpit in arces terrigenas imitata viros : sic tertius heres Saturni latebrosa vagis rimatur habenis d.evia fraternum cupiens exire sub orbem. ianua nulla patet : prohibebant undique rupes oppositae solidaque deum conpage tenebant. non tulit ille moras indignatusque trabali saxa ferit sceptro. Siculae sonuere cavernae ; turbatur Lipare; stupuit fornace relicta Mulciber et trepidus deiecit fulmina Cyclops. audiit et si quem glacies Alpina coercet et qui te, Latiis nondum praecincte tropaeis Thybri, natat missamque Pado qui remigat alnum. Sic, cum Thessaliam scopulis inclusa teneret Peneo stagnante palus et mersa negaret 161 et flexis r : et sulcis Cud. implicat G 4 P 8 t Parrhas. axes F 3 : actum Cud. 162 prolabitur D ? K I M 3 , Ez ]4 R 2 : prae- aj3, C z F4 P I R25 U : per- M r ,

C r , A 2 O r P 3 : de- a, b i E r F 2 (a.c.) L4 0 3 (p.c.), ]2 P 4 Z : di- A r , L z/3 I63 et velut P 5 occultum ] 3 , F 3 securus ] 3 M r , F 3 R 3j6, Z (a.c.) : se curru Rzo : sevum quom R r 3 prodit D K r , ]4 : tendit L ro : spargit Rzo e Rzo occulto se cursu spargit Hei11s. 165 elusos Isengr. marg. : inclusos codd. : inlusos Huxley (illusos Heins.) tramite in exempl. ed. Heins. in Mus. Brit. servato adnot. Wakiifield 166 ad arces L6 : in edes r 168 tenebrosa T, A2 L 3 P 3 R2j25 : nebulosa M 3 scrutatur D ? K r , E 2 ]4 169 fraternas R r6 per orbem a, E r Fz 0 3 R4, E2 F4]2 Lz P I : sub axem Mz T : per auras R I 6 170 montes T (cum oppositi v . seq.) 171 solitaque Iscngr. marg., unde solidaque Heins. : duraque F I K I , C I G r, E 2 ] 4 : siculaque aj3y : strictaque e 2 172 indign. teneri R s/7, L 2 173 tonuere J 3 (p.c. ) R s/7, 0 3 R 3 , P 4 U ? : tremuere M z R I (p.c.) T , b r F2j3, A 2 P 3 Rz Z 174 lipare a , h i E r F z, A z E2 P 3/4 Z : -is aj3y 175 reiecit malebat Heins. lumina D, ]4 176 audiit hec O r : audit et b r , H P4 R25 177 latiis Fr L s M 3 , F4 ] 2 (p.c.) L 3 P I (p.c.) Z : latis al3y triumphis W , 0 I 178 navigat R z Z (v.l.) 180 cum pro et F r negarent O z P 2 (a.c.) , y : vetaret ] 3 149

165

170

175

18o

D E RAPTU PRO S ERPINAE

185

190

195

200

arva coli, trifi.da Neptwms cuspide montes inpulit adversos : tum forti saucius ictu dissiluit gelido vertex Ossaeus Olympo ; carceribus laxantur aquae fractoque meatu redduntur fluviusque mari tellusque colonis. Postquam victa manu duros Trinacria nexus solvit et inmenso late discessit hiatu, apparet subitus caelo timor : astra viarum mutavere fidem; vetito se proluit Arctos aequore ; praecipitat pigrum formido Booten. horruit Orion ; audito palluit Atlas hinnitu; rutilos obscurat anhelitus axes discolor et longa solitos caligine pasci terruit orbis equos : pressis haesere lupatis attoniti meliore polo rursusque verendum in Chaos obliquo certant temone reverti. mox ubi pulsato senserunt verbera tergo et solem didicere pati, torrentius amne hibemo tortaque ruunt pemicius hasta : quantum non iaculum Parthi, non impetus Austri, non leve sollicitae mentis discurrit acumen. 181 montis 0 3 (a.c.) 182 tum D K I L6 R s/7, E I, A2 C2 E 2 }4 Z : tunc a!3y : cum M I , b i , H : nunc W 183 dissiluit a, E I F2 R4, y : se solvit a!3y : solvit se R22 184 fractoque M 3 W, F 4 J 2 (a. c.) P I R2j25 : factoque a!3y : subitoque L3 185 fluviique A I D F I (p.c.) K I L6, b i F2 (?p.c.), 186 tellus pro duros F 3 187 disce­ C2 (a.c.) J 4 L2 R2 : fluctusque 0 I dit M 3 , L4 (p.c.) 0 3 188 caelo subitus L w M 3 0 4 W, R 4, C2 H R25 : subitus caeli A 2 : subito caelo E 2 tremor M 3 189 polluit R I4/26 et Heins. e palluit Ls 190 booten Ugoletus : bootem (-e) a!3, A2 E2 P 3/4 R 2 : boetem a, b i L I 0 2/3 P 2, y 191 corruit R2 192 nitidos c axis R23 : arces 0 4 193 decolor laudatum in ed. Delph. solitos longa A2 R2 ferrugine M 3 , R2 194 pressisque F2 196 temptant a, C I G I L 4 R6, E2 ]4 L 2 R25 : pugnant L s M 2 197 mox ut F I4 turbato J 2 sens. pulsato L2 dorso L IO, 0 2 R4 199 velocius L 6, E I 200 tantum L7 R I4/26 afr ( ? = afri) iaculum pro iac. Parthi H 201 nee 20o-1 inleve R s, E2 R25 discurrit mentis R s : mentis discurrat Heins. terpolatos suspicatur Reinhardt ·

1 50

LIBER SECUNDUS

sanguine frena calent; corrumpit spiritus auras letifer; infectae spumis vitiantur harenae. Diffugiunt Nymphae; rapitur Proserpina curru inploratque deas. iam Gorgonos ora revelat Pallas et intento festinat Delia telo nee patruo cedunt : stimulat communis in arma virginitas crimenque feri raptoris acerbat. ille velut stabuli decus armentique iuvencam cum leo possedit nudataque viscera fodit unguibus et rabiem totos exegit in armos, stat crassa turpis sanie nodosque iubarum excutit et viles pastorum despicit iras. ' Ignavi domitor vulgi, deterrime fratrum,' Pallas ait, ' quae te stimulis facibusque profanis Eumenides movere tuae ? cur sede relicta audes Tartareis caelum incestare quadrigis ? sunt tibi deformes Dirae, sunt altera Lethes numina, sunt tristes Furiae te coniuge dignae. fratris linque domos, alienam desere sortem, nocte tua contentus abi. quid viva sepultis 202 frena madent L io 0 4, F4 (p.c.) corrupit R I9 203 violantnr U habene L I : aristae w, R4 et Schrader post hunc vers. non pauca excidisse existimat Bonnet 204 defugiunt M I : dissiliunt J 2 capitur M 2, R 3 205 deos e 2 T 4 ? gorgonos ] 3 , f', P 4 : -is af'y recludit R 1 3 206 in­ tenso R I4j26 : immenso L 2 : intecto c L 8 T 3 : infesto e 2 P 7 cornu D L6 R 5 w , 0 3 (p.c.) R4, ] 4 : nervo M I 207 in iras P 5 : utramque b I 208 victoris P 5 209 lacunam inter ille et velut statuit Jeep. perperam. iuvencum R2 Z 210 ceu leo C I pectora D K I M 2/3 04, ]4 fedat F2 211 excivit Burm., e gloss. excitavit in J 4 : exercet in adnot. P 2: effudit P 5 (a.c.) 212. turpis crassa F I sanie turpis L I, R2 213 excitat b I G I R6, ]2 : discutit C 2 vigiles r pastoris H despuit R I , L 2 P 4 214 vulgi domitor R I : domitor mundi D K I M 3, J 4 R2 teterrime D K I 0 4, b I 215 quo te e 2 m n r et Heins. : qui te 0 I : quis te J 4 (v.l.) facibus stimu­ lisque R 5 216 rapuere R2 tua ] 3 , C I G I L4 R6, F 4 P I recipit Birt 217 mundum D ? K I , ]4 incestare P 5 R I (a.c.) R s T, py : infestare a, h i R 3 , y : fedare R2o 218 sunt ibi R 5 dive M I P 5, L 3 : anime L 1 0 Lethae dativum maluit Heins. 218-19 verba inter deformes et furiae om. A I (add. marg. Pulmannus). 219 furietristes M 3 220 domus L 3 221 dampnatus T ISI

205

210

215

220

D E RAPTU PRO SERPINAE

225

230

235

245

adrnisces ? nostrum quid proteris advena mundum ? ' Talia vociferans avidos transire rninaci cornipedes umbone ferit clipeique retardat obice Gorgoneisque premens adsibilat hydris praetentaque operit crista; libratur in ictum fraxinus et nigros illurninat obvia currus missaque paene foret, ni Iuppiter aethere summo pacificas rubri torsisset fulrninis alas confessus socerum : nimbis hymenaeus hiulcis intonat et testes firmant conubia flammae. Invitae cessere deae. conpescuit arcum cum gemitu talesque dedit Latonia voces : ' sis memor o longumque vale. reverentia patris obstitit auxilio, nee nos defendere contra possumus ; imperio vinci maiore fatemur. in te coniurat genitor populoque silenti traderis, heu ! cupidas non aspectura sorores aequalemque chorum. quae te fortuna supcrnis abstulit et tanto damnavit sidcra luctu ? iam neque Partheniis innectere retia lustris nee pharetras gestare libet; securus ubique spumet aper saevique fremant inpune leones. te iuga Taygeti, posito te Maenala flebunt venatu maestoque diu lugebere Cyntho. 222 conteris R I6 : polluis L 3 P 3 226 praetentasque D, F2 aperit F 2 cristas D , F 2 : palla F I 4 vibratur a , C I, H L2 0 I i n altum A I 227 cur­ sus F I, b I (a.c.), E2 228 summo a, C I L4 0 3 (p.c.), y : vulso A 2 P 3 : celso J 3 : lapsus P I : missas L2 229 paciferas F I misisset F 3 fulminis a, h i L4 0 3 (p.c.), y: luminis af3y 230 con£ generum K I hymenaeon n et Heins. 23 1 signant V 2 232 attonite L s cessare Lz O I arcus w : iram F I : iras P 4 233 talemque . . . vocem L2/3 ita rifecit R I 6 : delia cum gemitu tales tonat ore querelas 234 sisque memor L6 235 contendere Q 239 aequalesque choros D K I , ]4 240 damnarunt I.J', prob. Hertel, p. 13 : turbavit H 241 iam nee Vz : iam non a expandere L2 silvis P 5 R s, H U 242 non F I L6 pharetram K r , ]4 portare Z licet P s ?, ]4 L2 0 I 243 sevumque A 2 premant J 4 245 maerebere L 2 1 52

LIBER SECUNDUS

Delphica quin etiam fratris delubra tacebunt.' Interea volucri fertur Proserpina curru caesariem diffusa Noto planctuque lacertos verberat et questus ad nubila rumpit inanes : ' cur non torsisti manibus fabricata Cyclopum

250

in nos tela, pater ? sic me crudelibus umbris tradere, sic toto placuit depellere mundo ? nullane te flectit pietas nihilumque paternae mentis inest ? tantas quo crimine movimus iras ? non ego, cum rabido saeviret Phlegra tumultu,

255

signa deis adversa tuli ; nee robore nostro Ossa pruinosum vexit glacialis Olympum. quod conata nefas aut cuius conscia culpae exul ad inmanes Erebi detrudor hiatus ? o fortunatas alii quascumque tulere raptores ! saltern communi sole fruuntur. sed mihi virginitas pariter caelumque negatur, eripitur cum luce pudor, terrisque relictis servitum Stygio ducor captiva tyranno. 247 ' Tertia distinctio ' notat F 14 rapitur w ? cursu P 5, L I 0 3 248 defusa R25 noto L4, ] 2 249 per nubila K I , E 2 ]4 (s.c. ) : in nubila m n R 1 1 rumpit A I C 3 L 5/10 P 5, !3y : rupit M2, H : tollit M 1 W , R 2 : fundit a:, F 2 G 1 L4 R6, L 2 R 2 5 U Z : tendit F 1 L6 0 4 251 i n m e R I W , L2 sic nos a: , E I (a.c.) F 2 O z P z R4, ]2 L z 0 1 U : sic vis L I 252 credere P 4 plac. toto L6 : clara plac. C 2 : tuto plac. Baehrens : toto licuit K 1 defendere L 1 253 pietas flectit L 5, L 3 : flectat pietas R 5 (p.c. ?), 0 1 nihilumque L I, A2 F 4 : nichilumne M 2 : nullumque a:!3y : nullumne (-ve) F 1 K 1 R s/7 W, E 1 (p.c.) O z P z R 3/4, y 254 novimus F 1 ] 3 M 3 R 5j7, b 1 F z/3 P z R 3 , C z 0 1 R25 255 rabido F4 et Heins. : rapido a:!3y phlegra A I D P 5 T, B L I 0 3 R 3 , E2 Z vers. sic refec. D K I , ] 4 : non ego te contra quateret cum phlegra (flegra) tumultu (-urn L 8 , T 3 ) 256 aversa n R 1 1 non F I K I , H J 4 257 glacialis vexit L 10, 0 2 P 2 R4, L 3 258 scelus pro nefas W vel cuius H noxia R2 noxae L 5 M 3 P 5, B G I (b) 0 3 R 3 , P 3 259 deducor W , Rz 260 alias 0 3 , ] 4 tulerunt L io M 2 T , A 2 C 2 261 saltim h luce a: , ]4 U (s.c.) 262 rapitur caelumque m R 1 I : pariter cum luce M 3 R s/7 : pariter cum sole w negantur malebat Heitzs. 263 cum sole A I L 10 M 2 0 4 T, B b 1 E I F 2 L 1 , C 2 ]2 Rz U (a.c.) Z 264 servitium a:!3y : infelix K I , ]4 ducor stygio 246 silebunt R 14/26

1 53

26o

D E RAPTU PRO SERPINAE

o male dilecti flares despectaque matris consilia ! o Veneris deprensae serius artes ! mater, io ! seu te Phrygiis in vallibus Idae Mygdonio buxus circumsonat horrida cantu, seu tu sanguineis ululantia Dindyma Gallis incolis et strictos Curetum respicis enses, exitio succurre meo, conpesce furentem, conprime ferales torvi praedonis habenas ! ' Talibu! ille ferox dictis fletuque decoro vincitur et primi suspiria sentit amoris. tum ferrugineo lacrimas detergit amictu et placida maestum solatur voce dolorem : ' desine funestis animum, Proserpina, curis et vano vexare metu. maiora dabuntur sceptra nee indigni taedas patiere mariti. ille ego Satumi proles cui machina rerum servit et inmensum tendit per inane potestas. amissum ne crede diem : sunt altera nobis sidera, sunt orbes alii, lumenque videbis purius Elysiumque magis mirabere salem cultoresque pios ; illic pretiosior aetas, aurea progenies habitat, semperque tenemus

:z6s

2.70

275

:zSo

:zSs

E I F2

triumpho c P7 266 imperia A I tardius L2 R2 267 te seu Phrygiae convallibus Heins. 268 sidonio F 5/I 5 R29 circumtonat W, prob. Burm. undique cantu F 3 270 despicis L s/6 271 exilio F I4 furentes R I4j26 272 opprime k : corripe p 2 fatales R I 3 rap­ toris ] 3 L s W 273 planctuque R s/7 274 sentit K I M I , B ? L I , E 2 H ] 4 Z : sensit a[3y : novit F 2 : movit e 2 : traxit W 275 tum ] 3 , [3 , E 2 F 4 ] 2 P 4 Z : tunc a, h i F 2 0 2 P 2 R4, y detersit a, G I (a) L4 R4j6, ]2/4 L 2 R2 U : deterget G I (b) et Birt 276 timorem A 2 : pudorem ] 4 278 varia a, G I R6, L 2 O I R25 : vacuo F I4 280 ipse M 2 : ilia D : nota L s : nempe v.l. in 0 2j3 machina mundi 02 P2 281 surgit per L 6 282 nec A I R 5 , F 4 (s.c.) J2 P I : non M I 283 ignes alii 0 2 (v.l.) 0 3 (v.l.) elysiosque D K I mirabile ] 3 P2 (v.l.) 284 darius Ls w, F 4 P I R I , h i F2 (a.c.) G I 0 3 (a.c.) R6, C 2 campum R s/7 W, U (p.c.) : campos 286 habiD K I , ]4 : lumen b i 285 illuc h i praestantior L I , R25

B LI

I 54

LIBER SECUNDUS

quod superi meruere semel. nee mollia denmt prata tibi ; Zephyris illic melioribus halant perpetui flares, quos nee tua protulit Aetna. est etiam lucis arbor praedives opacis fulgentes viridi ramos curvata metallo : haec tibi sacra datur fortunatumque tenebis autumnum et fulvis semper ditabere pomis. parva loquor : quidquid liquidus conplectitur aer, quidquid alit tellus, quidquid maris aequora verrunt, quod fiuvii volvunt, quod nutrivere paludes, cuncta tuis pariter cedent animalia regnis lunari subiecta globo, qui septimus auras ambit et aeternis mortalia separat astris. sub tua purpurei venient vestigia reges deposito luxu turba cum paupere mixtiomnia mars aequat-; tu damnatura nocentes, tu requiem latura piis, te iudice sontes inproba cogentur vitae commissa fateri. accipe Lethaeo famulas cum gurgite Parcas ; sit fatum quodcumque voles.' Haec fatus ovantes 287 que sup. M 3 tenuere a, J 4 tant R I4/26 prob. Heins. tenetur F I4 L2 R2 U Z non L 6 de(e)runt M r T, H et Heins. : desunt af3y 288 affiant A2 289 non L5 M 2, Z henna b i , O r 290 arbos c T2 perdives fr G2 P7 Q u Averni Isengr. marg., tmde Avemis Heins cf. Verg. Am. VI. 1 1 8 et 564 291 rarnos viridi T, R4, C 2 F4 P I : gravido ramos R 14j26 292 vi debis a, R 3 , L2 293 semper fulvis W semper pomis dit. fulvis H dotabere jtmii primus 294 gelidus pro liquidus C I 295 salis pro maris Claver. vertunt af3y : lati L7 R I4/I7/26 296 volvunt fluvii R25 : fluvii vertunt J 3 297 pariter cedent A I J 3 K I R s/7, 13, ]4 L 3 P 3/4 Z : pariter cedunt a , L I R3j4, y : etiam cedunt F I : pariter cedant G4 animantia L 5 regnis . . . cedent ]unii primus 300 ad tua F I4 R I6 : sed tua R I I veniunt A I L 5, L2 0 I : veniant R I : metuent R 14/26 fastigia F 1 et Barth. eadem variatio in codd. Manil. I. 61 6 (vide Housmammm ad loc.) 301 depos. cultu J 2 302 mors dampnat R2 303 bonis er 305 de pro cum F4 (p.c.) : in R I6 (p.c. ) : sub M I , E 2 ]4 (p.c.) 300 sit factum a, B ? L I , y : sitque ratum P 5 (v.l.), prob. Koch : sit sanctum vel pactum Heins. .

155

290

295

3oo

305

DE RAPTU PRO SERPINAE

310

exhortatur equos et Tartara mitior intrat. conveniunt animae, quantas violentior Auster decutit arboribus frondes aut nubibus imbres colligit aut frangit fluctus aut torquet harenas, cunctaque praecipiti stipantur saecula cursu mstgnem vtsura nurum. mox 1pse serenus ingreditur facili passus mollescere risu dissimilisque sui. dominis intrantibus ingens adsurgit Phlegethon : flagrantibus hispida rivis barba madet totoque fluunt incendia vultu. Occurrunt propere lecti de plebe ministri : pars altos revocant currus frenisque solutis vertunt emeritos ad pascua nota iugales ; pars aulaea tenent; alii praetexere ramis limina et in thalamum cultas extollere vestes. reginam casto cinxerunt agmine matres Elysiae teneroque levant sermone timores et sparsos religant crines et vultibus addunt flammea sollicitum praevelatura pudorem. Pallida laetatur regio gentesque sepultae •

3 15

320

325

0

0

•

velis ]3 K 1 L6, C r, E 2 ] 4 : voves Heins. hoc K r : sic o:, C 1, ] 2 L3 O r U (s.c.) Z 307 tenara P 5 (s c.), !3, H R2 mitius L 5 308 truculentior K r , ]4 3 09 discutit L 5 R r/5/7, A 2 0 r : excutit k nubilus]4 0 r 3 1 0 volvit 311 tartara cursu o:, E r P 2 (p.c.), y 312 ingentem harenas R s/7, Z Mr ille L 5 313 passus facili R 3 : faciem passus Burm. dubitanter : 315 phlegethon T, G r facili fassus F. R. D. Goodyear mitescere R r6 3 17 concurrunt L 5 : de316 tetroque e 2 : tostoque Heins. ruunt L ro F 7 R 1 3 : sue- F 14 properi L 5 , !3y : proprii L6 : subito ] 2 lecti K r L ro M2 (a.c.) P 5 (a.c.), C 2 ]2 (p.c.) ]4 0 1 : lecta o:!3y : leta L 5/6 RI W, C r : media F 1 : mixta Z de sede F 14 318 pars alacres L6 revocant altos L 5 currus . . . 319 ducunt emer. H ad altos W relocant Baehrens : removent Burm. pabula L6 320 trahunt ]2 321 thalarnis D Kr L 6 T, ]4 L 2 R25 : -os R 5/7, L4 (p.c.), ]2 U : -o R 19 certas R 2 : cunctis P 4 : cunctas b r attollere F r K I : extendere cod. Gentii apud Pulmannum 322 castam C 2 : caste L 2 duxerunt D K r , ] 4 : circumdant L 5 : cingebant e 2 ordine L r o 323 tenerosque ] 3 T W, b r L4 (a.c.), L2 vetant R ro timorem A 2 : dolores D K r , ] 4 : pudores T 2 326 regia letatur pallens G 4 t (regia pro .

LIBER SE CUND U S

luxuriant epulisque vacant genialibus umbrae : grata coronati peragunt convivia Manes. rumpunt insoliti tenebrosa silentia cantus ; sedantur gemitus ; Erebi se sponte relaxat squalor et aeternam patitur rarescere noctem. urna nee incertas versat Minoia sortes ; verbera nulla sonant nulloque frementia luctu inpia dilatis respirant Tartara poenis : non rota suspensum praeceps Ixiona torquet, non aqua Tantaleis subducitur invida labris ; [solvitur Ixion, invenit Tantalus undas] et Tityos tandem spatiosos erigit artus squalentisque novem detexit iugera campi (tantus erat !) laterisque piger sulcator opaci invitus trahitur lasso de pectore vultur abreptasque dolet iam non sibi crescere fibras. Oblitae scelerum formidatique furoris Eumenides cratera parant et vina feroci crine bibunt flexisque minis iam lene canentes extendunt socios ad pocula plena cerastas et festas alio succendunt lumine taedas. regio iam P I ) 327 genitalibus Z (a. c. ) 328 coronati celebrant L4 330 planctus 0 3 (v.l. pro gemitus) 331 pallor a infemam R s/7 clare­ scere L 10 M I, L I : discedere H : vanescere jeep olim 332 insertas 0 I 333 trementia L 5 M I : frequentia A 2 P 3 334 respondent C I, P 4 335 nee rota D K I R s/7, ]4 Z 336 om. L IO, ]4 (add. marg.) humida labris E2 337 ut spurium expunxit Heins. (sed non deest in L s , ut vir doctus asseverat) ; sed fieri potest ut ab ipso auctore sit conditus solvitur ixion et tantalus invenit undas L 10 P 5 R I 338 tityos Q : tytios M4 R 1 9 : ticios F 14 : titius a j3 y 340 verbum opaci impugnant Birt et Koch. perperam 341 invictus L6 lapso R s/7. C I : laxo 04, L I , R25 : lato R23 corpore L s M 1 R s/7 W 342 abreptasque ] 3 L6/IO M2 W, b 1 C 1 E I 0 3 , O J R25 Z : arrep tas que a j3y : arreptusque R 5/7 : abreptusque Reinhardt : abruptasque Heins. : directas­ que v.l. in R s/7, unde direptasque Burm. non iam R s/7 343 sceleris C 3 L 5, A2 R4 formidandique F I K I , E2 J 4 doloris J 2 : timoris K I L6 W, F2, ]4: tumoris Heins. 344 paterasque parant L 3 344-5 crine . . . f vina R 22 345 ore bibunt Reinhardt iam dulce W : iam leta L I 346 cerastes a j3 y 347 at K I L6 R s , E I , ] 4 : ac G 2 : nunc F2 faustas

! 57

330

335

340

345

DE RAPTU P R O S ERPI NAE

3 50

355

3 6o

365

tunc et pestiferi pacatum fi.umen Averni innocuae transistis, aves, f1atumque repressit Amsanctus : fixo tacuit torrente vorago. tunc Acheronteos mutato gurgite fontes lacte novo tumuisse ferunt, hederisque virentem Cocyton dulci perhibent undasse Lyaeo. stamina nee rumpit Lachesis, nee turbida sacris obstrepitant lamenta choris. mors nulla vagatur in terris, nullique rogum planxere parentes. navita non moritur f1uctu, non cuspide miles ; oppida funerei pollent immunia leti, inpexamque senex velavit harundine frontem portitor et vacuos egit cum carmine remos. lam suus inferno processerat Hesperus orbi ; ducitur in thalamum virgo. stat pronuba iuxta stellantes Nox picta sinus tangensque cubile omina perpetuo genialia foedere sancit ; exultant cum voce pii Ditisque sub aula talia pervigili sumunt exordia plausu : ' nostra potens Iuno tuque o germane Tonantis r et Heins. olim alio ] 3 , C r, E2 L3 R25 Z : alie af'y 348 nunc P r placatum L ro R r , L r , F 4 H O r (a.c.) P r R25 limen D ?, E 2 (p.c.) ]4 349 transitis K r P 5 R 5/7, 0 3 (a.c.), 0 1 (a.c.) fietumque R2o 350 am­ sanctus (an-) f'y : amxantus (an-) M 2 P 5 R r W, F2 L4, A 2 E 2 (p.c.) tacuit fixo a, br E r F2 G r R6, y : tacuit iusso C r L4 351 fiumine M r 352 maduisse R r 353 stagnasse ljl : sudassc T 2 : madidasse 0 I (p.c., m.a. ) : nudasse R 5 ?, C 2 : nutasse Paul 354 non K r L 5 M3 w, ]4 L 3 O r P r Z 355 obstreperent rupit a, b 1 C I, 0 I non L6, L 3 Z : ne D, A 2 E 2 P 3 D, A2 E2 ]4 R2 356 nullumque F 1 R 5 : nullaeque Isengr. sorores v.l. in 0 2 P 2 357 nec L6, 0 r ponto L 6 nec a, b r 0 3, E2 F4 L 2 0 I P I 358 gaudent R 1 3 359 implexamque 0 2, L 2 : irnpexosque (-umque R r9 359-60 post 347 ponere p.c.) D K r , ]4 crines (-em R r9 p.c.) D KI, ]4 voluit Baehrens 36o vexit R2o : legit Postgate, coli. Verg. Aen. V. 209 (in Class. Q!art. iv, 1 91 0, p. 262) 361 precesserat L 5 , U 362 thalamis C r, L 2 : -o H 363 stellatos E I , L 3 364 genialia C 3 L io W, b r L4 (p.c.) 0 2 P 2, F4 (p.c.) O r U (a.c.) : genitalia af'y federa L 5, R25 omine . . . foedera excerpta Schottiana sancit ] 3 T, !3, E 2 L 3 P 4 R 2 : sanxit a, F2 (p.c.) 366 exordia cantu B 367 nostra parens 0 2 R4, y 365 tum 0 3 (v.l.) I58

LIBER SECUNDUS

et gener, unanimi consortia discite somni mutuaque alternis innectite vota lacertis. iam felix oritur proles ; iam laeta futures expectat Natura deos. nova numina rebus addite et optatos Cereri proferte nepotes.' D W , P 2, J 4 : nostripotens vel nocte potens Heins. : nostra regensJeep : nostra, precor Birt 368 unanimis a, F 3 (a.c.) R6, C2 F4 O I P I R25 : unanimes M I, E 2 J 4 U (p.c.) : ignavi L6 discite !3, E 2 J 2 P 4 R25 : ducite a!3y lecti M I : regni Z 369 connectite L 3 vota lacertis P s R s/7 W, !3y : colla lacertis ay : brachia votis H 370 proles oritur L2 371 iam pro nova b I, C 2 munera F I2 p I 372 celo pro cereri L 2

I 59

370

LIB ER TERTIUS

s

ro

Iuppiter interea cinctam Thaumantida nimbis ire iubet totoque deos arcessere mundo. illa colorato Zephyris inlapsa volatu numina conclamat pelagi Nymphasque morantes increpat et Fluvios umentibus evocat antris. ancipites trepidique ruunt, quae causa quietos excierit, tanto quae res agitanda tumultu. ut patuit stellata domus, considere iussi, nee confusus honor : caelestibus ordine sedes prima datur ; tractum proceres tenuere secundum aequorei, placidus Nereus reverendaque Phorci canities ; Glaucum series extrema biformem accipit et certo mansurum Protea vultu. nee non et senibus Fluviis concessa sedendi Praefationem panegyrici de sexto cotiSulatu Honorii inter secundum et tertium librwn positam exhibent omnes quos a classi assignavi, et praeterea (vett. tantum hie recenseo) : b I (post III. 6o in marg. inf add.) E I F 2 (Prohemium tertii libri de r.p.) L4 0 2 P2, E 2 F 4 H J2/4 L2 O r P r/3 (liber secundus explicit incipit prologus IIJ libri) R25 U Z Inscriptiones recc. notabiles : Prologus in tertium et ultimum librum elegiaco metro scriptus R 19 : claudiani poete egyptiaci prefatio incipit in librum tertium r : Incipit quartus liber Q De raptu Proserpinae lib. m A2 (PulmantiUs) : Explicit prologus tertii libri E I : liber tertius de r.p. F 2 : liber III F4 (111. rec. marg.) : TERCIVS LIBER CLAVDIANI G r : incipit liber tertius K r (m.a. ) : Liber III L 3 (marg.) : explicit prologus incipit tertius liber P 3 : liber m de r. J P. P 4 (m. recentiss. marg.) 2 totosque Rs, C r F2 (a.c.) R4, Or R2 accersere Kr W, Er F 3 0 2 P2 R4, A 2 F 4 ]4 P r : accedere Rr, R 2 celo E 2 (a.c.) 3 zephyris b r G r (sscr. ) , C 2 F 4 P r : -os af3y perlapsa R 1 3j22 : prelapsa L r , C 2 ? F 4 H U : elapsa L s/6 M2, A 2 L 2 O r : pregressa T, R2j25 : transgressa K r , ]4: illata Junii primus 4 compellat M 3 6 vocatos R r , E2 7 eruerit R 14j26 agitata C r 9 non K r R r , 02, ]4 honos L6 M 3 R7, F 2, ] 2 L2 P 3 1 0 proc. tractum jtmii primus I I reverendaque a , E I F 2 L 4 0 2 (p.c.), H L 3 U Z : venerandaque F 7 : e t lucida af3y : e t livida L 2 : e t lurida J 3, A2 P 3 : et lubrica R 2 : et pallida L 6 : et candida L I : et caerula Isengr. marg. 12 extema]u nii primus 14 est pro et R2o fluv. sen. 0 2 P 2 1 60

LIBER TERTIUS

gloria ; plebeio stat cetera more iuventus, mille Amnes. liquidis incumbunt patribus udae Naides et taciti mirantur sidera Fauni. Tum gravis ex alto genitor sic orsus Olympo : ' abduxere meas iterum mortalia curas iam pridem neclecta mihi, Saturnia postquam otia et ignavi senium cognovimus aevi, sopitosque diu populos torpore paterna sollicitae placuit stimulis inpellere vitae, incultis ne sponte seges grandesceret arvis, undaret neu silva favis neu vina tumerent fontibus et totae fremerent in pocula ripae. haud equidem invideo-neque enim livescere fas est vel nocuisse deos-, sed, quod dissuasor honesti luxus et humanas oblimat copia mentes, provocet ut segnes animas rerumque remotas ingeniosa vias paulatim exploret egestas utque artes pariat sollertia, nutriat usus. Nunc mihi cum magnis instat Natura querellis humanum relevare genus, durumque tyrannum 17 naides J 3, C 2 : naiades a!3y placidi D K I, J 4 numina olympo est P 5 T, !3, L3 P4 vers. sic refecit R 2 : tunc gravis est solio genitor sic orsus ab alto 19 h inc incipit J I abduxere D J 3 P 5, !3, E 2 (a.c.) F 4 P I R 2 5 : adduxere a!3y 20 iam dudum w, F4 P I R2 iam n.m.p. V2 21 seriem cog. r: sensum cog. H cognoscimus C 3 w, R 4 : commovimus a F6 k et Schrader 22 lang (u)ore, J 3 M2, C I , Z 24 candesceret E 2 (p.c.) H R2 U (a.c.) : canesceret R 5/7 : concresceret a k R2o Y 2 agris T 25 neu C 3 J 3 L6 26 aut R 5/7 fluerent C 3 0 4, !3y : ne a!3y ne a, C r L I/4 0 3 R4, y L 5/10 M r W, O r : ruerent F 3 totae . . . vites W : tori . . . rivi Jeep 27 invideo a, L4 0 2 R6, y : invidia R 5/7, !3, P I (a.c.) P 4 U Z, J r : invidia est ]3 L 5 04 P s (a.c.), C r F 3 L r R4, C 2 ? (a.c.) F4 P r (p.c.) R2 nee a!3y 28 nee pro vel Z : aut H deo F r , ]4 (s.c.) : deis C 3 quid L I/4, ]4 29 excecat pro oblimat L 10 mentis M I 30 et segnes L6 : insegnes W : aud segnes 0 6, tmde haud segnes Heins. animas C r : annos R 2 5 3 2 atque C 3 L 5 M3, L3 artem Z 33 querellis c P6/9 y : querelis vett. 34 ' melius absit totus hie versus '-Barth dirumque K r, F4 P I/3 R25 Z, J r 15 stant Heins.

D K r , b r , ]4

II

18 tum ] 3 K r P 5 R r , !3y : tunc a!3y

I6I

HCD

rs

:zo

:zs

30

D E RAPTU P R O S ERPINAE

JS

40

45

so

inmitemque vocat regnataque saecula patri commemorat parcumque Iovem se divite clamat, qui campos horrere situ dumisque repleri rura velim nullisque exornem fructibus annum ; se iam, quae genetrix mortalibus ante fuisset, in dirae subito mores transisse novercae : " quid mentem traxisse polo, quid profuit altum erexisse caput, pecudum si more pererrant avia, si frangunt communia pabula glandes ? haecine vita iuvat silvestribus abdita lustris, indiscreta feris ? " tales cum saepe parentis pertulerim questus, tandem clementior orbi Chaonio statui gentes avertere victu ; atque ideo Cererem, quae nunc ignara malorum verberat Idaeos torva cum matre leones, per mare, per terras avido discurrere luctu decretum, natae donee laetata repertae indicia tribuat fruges, currusque feratur nubibus ignotas populis sparsurus aristas et iuga caerulei subeant Actaea dracones. 35 notat C 3 tempora R25 36 parvumque A ut Baehrens damnat F 14 37 qui Ar M I/3 T, br E r (a.c.) F2 L4 0 2 (p.c.), A2 L 3 P 3 , ] r : cur a[3y : quid C 3 P s (a.c.), R2, prob. Heins. : quod R s/7. R25 : quo Heins. 38 arva

C 3 , L3 nullisque F r ]3 L6 M r T, C r E r , A2 O r Z : nullis af3y : et nullis L ro, H : cur non R r6 exornans (om. -que) P4 frugibus M2 39 ipsam pro se iam C 3 genetrix Burm. : genitrix C 3 M 3 R s/7 (p.c.), ]2 L 3 O r : rectrix a[3y : nutrix ay : cultrix K 4 u 40 morem F2, L 3 : nomen R s /7 transire a[3, J 2 0 r P 4 R 2 41 traxisse polo mentem C 3 42 pererrent e i p2 43 devia L r o frangant a, 0 2 P 2 44 hecine L6, P 2 (a.c.) : heccine a[3y iuvet P 4 abdita F 2 (a.c. ?) , J 4 P 3 Z : adclita a[3y 45 talis G r R 6 : quales P r 47 p op ul os pro gentes L2 clivertere W chaonia . . . glande R2o 48 atque adeo M r T, [3y, J r 49 cum torva W, C 2 ]2 L 3 iugales P 3 50 rabido Baehrens percurrere M I : disquirere Heins. 51 decretum a, C I F 2 (a.c.) G r 0 3 R6, y : decretum est a[3y : decrevit F 3 , unde decrevi Heins. donec natae L io M r/3 0 4 R r , F 2 R4, H L 2 P 3 R 2 U, ] I 5 2 tribuat segetes 0 4 5 3 nubilus F2 : avius R r (a.c.), B F 3 54 iugales R 7 L I 0 3 R6, P 4 U ignaras R r6 populo L 5 : terris R2

!62

LIBER TERTIUS

quod s i quis Cereri raptorem prodere divum audeat, imperii molem pacemque profundam obtestor rerum, natus licet ille sororve vel coniunx fuerit natarumve agminis una, se licet ilia meo conceptam vertice iactet, sentiet iratam procul aegida, sentiet ictum fulminis et genitum divina sorte pigebit optabitque mori : tum vulnere languidus ipsi tradetur genero, passurus prodita regna, et sciet an propriae conspirent Tartara causae. hoc sanctum; mansura fluant hoc ordine fata.' dixit et horrendo concussit sidera nutu. At procul armisoni Cererem sub rupibus antri securam placidamque diu iam certa peracti terrebant simulacra mali, noctesque timorem ingeminant omnique perit Proserpina somno. namque modo adversis invadi viscera telis, nunc sibi mutatas horret nigrescere vestes, nunc steriles mediis frondere penatibus ornos. stabat praeterea luco dilectior omni laurus virgineos quondam quae fronde pudica (a.c.) 55 sed si L ro dirum 0 3 (p.c.) : durum F r , R25 56 moles pacemque T 3 : pacem molemque } 3 , C r 57 iste L 2 : ipse L s/6 0 4, L 4 P2, C2 58 aut L S natarumve C 3 L io M I/2 T, 0 3 (p.c.), A 2 : natarumque al3y 59 sed licet L6 M I (a. c.) R s, L I , C 2 : scilicet} 3 , R 2 ille R r , B ? R4, P I R25 : ipse 0 2 : ipsa F 3 P 2 (p.c.) : ista \jl conceptum P 2 (p.c.) R4, }2 (a.c.) Pr R25 semine R I : sanguine T, R25 iactet de vertice natam L s 6o iratam C I : iratum al3y aegida 13, E 2 L 2 P 4 : aegide al3y ictus J 3 L6, h i C I F 2 P2 (p.c.), E 2 }2 61 genitam L s/6 stirpe a, C I E I F2, }2 R25 62 tunc a, C I L4, y, }I diro pro vulnere 03 (p.c.) P 2 R4 languidus D J 3 KI M2 RI, b I C I, E 2 }4 P 4 R25 : saucius al3y ipse a: ictu 0 3 (p.c.) P 2 63 perdita R I , 0 I 64 respirent M 3 65 sanctum est A I L6/ro M2 0 4, B L I/4, R25 : pactum est R s/7 fluunt a cardine } 3 , 13, L3 P 4 R25 U 66 nutu }3 M2/3 (p.c.), b r , E 2 : motu al3y 67 armisonis R2o 68 iam cuncta F I M I , F 2 (a.c.) 69 noctisque (L I) timores (M 3 , B 0 2 P 2 R4, y) cum verbo ingeminant intransitive accepto coniungit Burm. 71 immensis K I , }4 : immissis \jl 72 iam sibi A 2 pallescere F I 7 5 quae quondam a, 7 3 med. ster. L 7 florere L6, R 2

163

I I -::1

ss

6o

6s

70

75

DE RAPTU PROSERPINAE

so

ss

90

umbrabat thalamos : hanc imo stipite caesam vidit et incomptos foedari pulvere ramos, quaerentique nefas Dryades dixere gementes Tartarea Furias debellavisse bipenni. Sed tunc ipsa sui iam non ambagibus ullis nuntia materno facies ingesta sopori : namque videbatur tenebroso obtecta recessu carceris et saevis Proserpina vincta catenis, non qualem Siculis olim mandaverat arvis nee qualem roseis nuper convallibus Aetnae suspexere deae : squalebat pulchrior auro caesaries et nox oculorum infecerat ignes exhaustusque gelu pallet rubor ille, superbi flammeus oris honos, et non cessura pruinis membra colorantur picei caligine regni. ergo hanc ut dubio vix tandem agnoscere visu evaluit, ' cui us tot poenae criminis ? ' inquit : ' unde haec informis macies ? cui tanta potestas 0 2 P 2, E 2 J 4 0 r : olim quae Isengr. fronte C r 76 umbrarat 0 4, L I R4, prob. Heins. ima F r6 : primo C 3 , E 2 (? a.c.) : vivo W cespite cesam C 3 W, 0 2 P 2, E2 ]4 L 3 : stirpe recisam Isengr. marg. cj. Verg. Aen. XII. 208 77 incisos 04 R s/7 : intactos R I crines M I w, R 4 : frondes R I6 78 quaerentique T et Isengr. marg. : quaesivitque ceteri 79 tartareas R I securi L io M I , F 3 0 2 P 2 R4, R25 78 et 79 delere voluit ].]. Scaliger So sed nunc T, E 2 : et tunc m R I I : sed tamen P 4, unde sed tandem Heii!S. ista h i 81 ingesta F I M I P 5 R I w, E I (a.c.) F2 L 4 R4, A 2 E 2 P 3 , J I : inserta K I L s/6 0 4 T, �. C2 F4 ]4 L2 P I : impressa A I L ro : iniecta M 2/3 W : oblata R s/7 sopori est L ro, ]4 82 tenebrosi Heins. obiecta F I L 6 M 2 T, A 2 P 3 R2, ] 1 : obducta A r C 3 P s R r, C I, ]2 : obsessa L r o : tecta L 5 , F2 83 ac P 5 duris F 2 84 nec P s R s/7 T, L r , R25 quondam 85 nee C 3 L6 P s T, B E I F2 G I R6, pro olim W antris T : horis R I y: non o:�y sub vallibus M I/3 : sub collibus M2 w enne 0 I : yde E I (a.c.), A2 E 2 (a.c.) 86 conspexere L 5, R4, O r : pro- 0 2 P 2 : in- F I : a­ M I 0 4 : accepere W : suscepere K4 (v.l.) 87 ignem M 2 : orbes A 2 8 8 exustusque R29 e t Heii!S. rubor o: , 0 I : pudor o:�y : color A I ipse R I 3/I6 superbus L6 M 3 , L I 90 calig. mundi e r h V2 91 tandem dubio vix B noscere R2o dubio tandem cognoscere (vix om.) F I vultu KI RI 93 facies o:, ]4 L 2 R2 cur tanta F I P 5 , L 2 : an tanta]4 facultas

LIBER TERTIUS

in me saevitiae ? rigidi cur vincula ferri vix aptanda feris molles meruere lacerti ? tu mea tu proles ? an vana fallimur umbra ? ' Ilia refert : 'heu dura parens nataeque peremptae inmemor ! heu fulvas animo transgressa leaenas ! tantane te nostri tenuere oblivia ? tantum unica despicior ? certe Proserpina nomen dulce tibi, tali quae nunc, ut cernis, hiatu suppliciis inclusa teror ; tu saeva choreis indulges Phrygiasque etiamnum interstrepis urbes ! quod si non omnem pepulisti pectore matrem, si tua nata, Ceres, et non me Caspia tigris edidit, his, oro, miseram defende cavernis inque superna refer. prohibent si fata reverti, vel tantum visura veni.' Sic fata trementes tendere conatur palmas. vis inproba ferri inpedit et motae somnum solvere catenae. L io 94 in te Reinhardt est post saevitiae add. M 3 , E I , A 2 95 trac­ tanda p 2 movere M I , E r 96 an mea tu a, F4 R25 : tune mea es r aut vana M I P 5 97 dura a, b I L4 0 2, y : dira a[3y : torva F I : dicta Delrii cod. 98 fulvos . . . leones a, E I , F 4 H ]2 L 3 U : sevas . . . leenas 0 2 P 2 et ]. C. Scaliger 99 cepere L 10 quantum J 4 100 non est pro certe F I M3 (p.c.), unde nonne estjeep 101 ccrnis a, B L4, y : ccrnor a[3y 102 feror a, b 1 E 1 (a.c.) F 2 L4 R6, 0 I (p.c.) R25 U : precor F I, unde premor Baehrens, quod iam in T 2 (marg.) : trahor codd. Parrhasii et Claver. in notis festa v.l. in R 5j7 : laeta Burm. 103 indulgens G4 V 3 (a.c.) et jeep phrygiasque C 3 ] 3 P s R I, [3y, J I : phrygias a[3y etiamnum E 2 : etiamnunc C 3 ] 3 R I T w, (3, A2 C 2 H L 3 R 2 : vel nunc a[3, F 4 O r P I : et nunc A I M I/3, E I (p.c.) R 6 (p.c.) : iam nunc D K r L io, ]4 : nunc nunc R 14j26 : seu nunc P 4 vel nunc ante phrygias in R I 3 instrepis J 3 R r T, b I C I E I (a c.) , L 3 R2, ] I : perstrepis H et Reinhardt 104 omni H pepulisti D J 3 K I R I , B R4, J 4 L 3 R2 : repulisti a[3y ex pectore (e l.jl) D : a pectore T, L2 105 tua nata F I L6, R 2 : tu nota a[3y : tu nostra A I , prob. Heins. : tu sancta D J 3 K I T, C I , C 2 E 2 H ]4 : t e nata Heins. t u nostra parens Reinhardt e t m e non M2 T 106 catenis D K I L io, L4, ]4 R 2 107 ferar ]2 108 vel tandem E 2 : vel saltern D K I M 2j3 T, J 4 : me saltern W : me tantum L 5 : vel tu me L6, F4 vel visura veni tantum L I 109 visque 0 4 110 excusscre D K I w, .

95

roo

ros

no

D E RAPTU PR O S ERPINA E

us

uo

125

1 30

obriguit visis ; gaudet non vera fuisse; conplexu caruisse dolet. penetralibus amens prosilit et tali conpellat voce Cybelen : ' iam non ulterius Phrygia tellure morabor, sancta parens ; revocat tandem custodia cari pignoris et cunctis obiecti fraudibus anni. nee mihi Cydopum quamvis extructa caminis culmina fida satis. timeo ne fama latebras prodiderit leviusque meum Trinacria celet depositum. terret nimium vulgata locorum nobilitas. aliis sedes obscurior oris exquirenda mihi. gemitu flammisque propinquis Enceladi nequeunt umbracula nostra taceri. somnia quin etiam variis infausta figuris saepe monent, nullusque dies non triste minatur augurium. quotiens fl.aventia serta comarum sponte cadunt ! quotiens exundat ab ubere sanguis ! larga vel invito prorumpunt £lumina vultu iniussaeque manus mirantia pectora tundunt. si buxos inflare velim, ferale gemiscunt ; h x , E2 ]2/4 O x 111 visu ]4 R2 nee 0 4 112. amplexu C 3 L6 P s (p.c.) W, F 2 altis pro amens J 2 : imis C 1 F 2, C 2 113 desilit R 16 : exL 3 Cybeben Bentleius et Cortius ad Luc. I. 6oo 115 sacra U tandem revocat L I O : revocat nam me F I M 3 : nam me revocat W : matrem revocat 116 tantis A I : cautis R2 : certis F I W, h x , C 2 P I R I 3/16j22 nostri IJI 117 non D K x L6j10, L 4 0 3 , ]4 R 2 constructa F 1 (p.c.) R r, L 2 : instructa 119 leviterque M2, L r , F 4 : fundata R 2 : fabricata L ro cavernis R2o F x R s/7 servet ]3 M 3 R s/7 W, E r (s.c.) , O r 121 aliis edes L 3 122. est quaerenda F 2 , 0 I : inquirenda w mihi est L ro W propinqui A 2 12.3 celamina nostra e r V 2 124 infesta a, F 3 0 3, L2j3 0 1 R25 : infecta F x , B L I 0 2 P2 R4, H P I 125 monent C 3 D K 1 M2, F2j3 G r R6, H ? 0 1 R2j25 : movent af3y omnisque M2j3 P 5 mihi pro non M2/3 P 5, J 2 127 exudat F I , L I 0 3 (a.c.) : exsudat L 3 R2 129 sunt qui inviseque habere videantur : immisseque D K x , ] 4 : inviteque A I L sfxo 04 P s (p.c. ), F4 merentia a, C x Ex (p.c.) L4 (p.c.), E 2 (a.c.) F 4 O x P I : lugentia U tempora R 14 laedunt K 4 : rumpunt R n 130 buxos D K x M I/3 (p.c. ) , 0 3, E 2 ] 4 : -us af3y : buccas W fa tale W gemescunt C 3 P s, 0 2 P 2 (a.c.), C 2 ! 66

LIBER TERTIUS

tympana s i quatiam, planctus mihi tympana reddunt. ah vereor ne quid portendant omina veri ! ah longae nocuere morae

!' ' Procul inrita venti

dicta ferant,' subicit Cybele : ' nee tanta Tonanti segnities ut non pro pignore fulmina mittat.

135

i tamen et nullo turbata revertere casu. ' Haec ubi, digreditur templis. sat nulla ruenti mobilitas : tardos queritur non ire iugales inmeritasque movens altemo verbere pinnas Sicaniam quaerit cum necdum absconderit Iden.

140

cuncta pavet speratque nihil. sic aestuat ales, quae teneros humili fetus commiserit omo allatura cibos, et plurima cogitat absens : ne gracilem ventus decusserit arbore nidum, ne furtum pateant homini, ne praeda colubris. Ut domus excubiis incustodita remotis et resupinati neclecto cardine postes O r R25 131 planctum D L w : gemitus R r 132 heu pro ah C 2 O r 133 (h)a(h) o:, hr : he (e) P s w, !3y, J r : (p.c.) : heus M I somnia M r W heu C 3 F r L 6 M r , F 2 0 2 (p.c.), F 4 P r : num W 134 suhiit F r P s R s , h r E r L4 0 3 R6, ]2 L 3 P r (p.c.) R 2 non o: , 0 3 (p.c.) R4, ]4 L 2 P 3 R2, J r : num Ls Rr tonantis Mr T 135 sumat pro mittat Isengr. marg. 136 i nunc et a k : i tantum et u tardata D K r 137 hoc R25 degre­ ditur 0 2 (a.c.) P 2 (a.c.), O r et Baehrens templo T 2 sat Heins. : sed (set) o:!3y : sihi U videtur F r R s/7, prob. Jeep 138 quer. tard. R s/7. L r nunc pro non T languere cod. Gentii apud Pulmannum dracones o:!3y 139 immissasque F7 R 1 3 (a . c. ) R r6j22 feriens pro -que movens W 140 dum L 2 necdum ] 3 L6 M 3 P s (a . c.) R r , !3y, ] r : nondum o:, B h r 0 3 , y ahsconderit ] 3 T , C r (p.c.) E r F 2 L4, y, J r : -erat D L s : ascenderit o:!3, C2 L 3 : -erat C3 K I L6, E 2 ? F4 ]4 P I : liquerit in notis Claver. iden (-em) M r , !3, E 2 (a.c.) P4 R25 : idam o:!3y 141 timet L 3 142 cum L s R r ten. fet. hum. M 4 R I3/I6 : fet. hum. ten. m n P 9 R 19 : fet. caros hum. r : ten. fragili fet. F4 (p.c.) commiserat L 5 R s/7. F2 02 0 3 (p.c.) P2, L 3 O r P r R 2 144 gracili] 3 , C r 0 2 (p.c.) R 3 , E 2 : fragilem A I C 3 L w 0 4, P 2 nidum . . . ventus 0 2 P 2 decusserit a F 6 R r6 R 19 Y 2 : discusserit vett. 145 neu L ro M2 pateant M 3 (a.c.) w, B : -eat F r M 3 (p.c.) 0 4 P s W, R4, HU, recipit jeep : iaceant o:!3y : -eat K r M2, F4 P r : fiant L6f ro R s/7. 0 2 P 2 pat. hom. furt. H: iac. hom. furt. R2o neu L IO M2 147 limine F 2 ! 67

1 45

DE RAPTU PRO S ERPINAE

Iso

I SS

I6o

I6S

flebilis et tacitae species apparuit aulae, non expectato respectu cladis amictus conscidit et fractas cum crine avellit aristas. haeserunt lacrimae, nee vox aut spiritus oris redditur, atque imis vibrat tremor ossa medullis. succidui titubant gressus ; foribusque reclusis, dum vacuas sedes et desolata pererrat atria, semirutas confuso stamine telas atque interceptas agnoscit pectinis artes. divinus perit ille labor, spatiumque relictum audax sacrilego supplebat aranea textu. Nee deflet plangitve malum ; tantum oscula telae figit et abrumpit mutas in fila querellas ; adtritosque manu radios proiectaque pensa cunctaquc virgineo sparsa oblectamina ludo ceu natam pressat gremio ; castumque cubile desertosque taros et sicubi sederat olim perlegit : attonitus stabulo ceu pastor inani, cui pecus aut rabies Poenorum inopina leonum 148 facies R I , U 149 exspectatae R2o et Claver. amictum a 150 evellit L6, b I R 4 : revellit F I L sfw M I R I , F2, ] 2 : repellit 0 3 (v.l.) 151 non R I , F 2 vox nee a, b i F 3 G I R6, L 3 R25 : vox non F 2 ori M 3 (p.c. ) 0 4 W , 0 3 (p.c.) P 2 , J 2 152 ex imis Isengr. marg. : a c totis L6 timor a , b I C I F 2 (a.c.), P 3 R 2j25, ] I 153 trepidant gloss. in R s/7 gressus titubant R2o 154 cum F 2 R4 hedes R I 156 interruptas F I L s W, F 4 : intermissas v.l. in R s/7 agnovit L s/6 T , b i F 2 , C 2 ]2 R25 : cognoscit F I R s/7, L 3 : cognovit W, R 2 159 hec deflet a, L I/4 0 3 ? R4, y plangitve P2 (p.c.) : -que ceteri tantum J 3 R7, C I G I 0 3 (a . c. ) R 3j6, A 2 ? L 3 : tamen a!3y : tandem e i : tunc F I , L I , J2 R2 telis R2 160 foot L s abrupit M I/2 W, C I L4, ] 2 L 2 : absumit Gesner mutas a!3, C2 (a.c.) J 2 L3 P I (a.c. ) P 4, ] I : multas a, C I L I/4, y : mixtas R I , F 2 : mcstas lj.l per fila M I : in verba F7 R I3/22 : in fine F I S querellas L6 : querelas a!3y 161 attritosque A I ]3 L io P s T, b i C I E I F2, E2 (a.c.) ]2 ]4 P4 R25 : astrictosque af3y : attractosque L s W 162 virginei . . . ludi Heins. 163 pressat gremio ] 3 L6 R sf7 T, !3y, ] I : pressit grem. a, L4 0 3 (p.c.) P2 (p.c.) R4, H]2 (p.c.) J 4 U: prensat grem. P I : grem. pressat R25 : grem. pressit L 5 R I , L2 carum­ que a!3y 164 toros sedesque ubi vel locos sellasque ubi Reinhardt ante L 6 : illa 0 4, A 2 P 3 , ] I 165 sic pastor 0 4, A 2 P 3 , ] I 166 cum R s/7

!68

LIBER TERTIUS

aut populatrices infestavere catervae ; serus at ille redit vastataque pascua lustrans non responsuros ciet inploratque iuvencos. Atque ibi secreta tectorum in parte iacentem conspicit Electram, natae quae sedula nutrix Oceani priscas inter notissima Nymphas. par Cereri pietas ; haec post cunabula dulci ferre sinu summoque lovi deducere parvam sueverat et genibus ludentem aptare paternis. haec comes, haec custos, haec proxima mater haberi. tum laceras effusa comas et pulvere canos sordida sidereae raptus lugebat alumnae. Hanc adgressa Ceres, postquam suspiria tandem laxavit frenosque dolor : ' quod cernimus,' inquit, excidium ? cui praeda feror ? regnatne maritus an caelum Titanes habent ? quae talia vivo ausa Tonante manus ? rupitne Typhoea cervix Inarimen ? fractane iugi conpage Vesevi Alcyoneus Tyrrhena pedes per stagna cucurrit ? T, E r , C 2 U : quem K r L6, }4 lupus K r L6, G r, C2 E 2 (ao co ) J4 luporum K r , 0 2 P2 167 carine L3 16S at ille a, E r F2 0 2 P2 R4, A2 C 2 (aoco) HJ 4 U, J r : a d illa a j3y serius ille M 2 pabula M r mirans G r (aoco ) 169 nee F 4 L 2 Pr vocat L6 170 utque K4 ubi a, Cr P 2, E 2 ] 2j4 P 4 : ita F r : in T, L 4 0 2, 0 r tecto secro F r in omo 0 2 P 2, 0 r latentem A r L ro 0 4 R r 171 aspicit K r 0 4 R s/7 W, b r, }4 U : respicit M 3 quae natae F I L6 0 4 174 ferre manu A 1 175 parentis J 2 0 I 176 matris Baehrens : semper R r habetur H 177 tum 0 2 P 2 : tunc aj3y : nunc M r , 0 3 , O r lacrimas R s , O r : lacrimis r diffusa A r L I O : con- F 4 : in- L 3 comas a , C r L4 R 3 , F4 ] 4 P r U : genas a j3 y effusa genas laceras P 4 laceras perfusa (vel affusa vel suffusa) genas, vel lacrimis offusa genas Heinso cano R I , B 02 P2 R3/4, } 2 : crines H O r 17S raptum M I , H deflcbat H 179 affata R r3/r6j22 ISO lassavit L6 fletusque F r w , H L 2 : fletumque e r : fremitusque P s, O r U quid R 2 ISI exitium C 3 W ferox A r M2 (aoco) P s, E 2 (aoco) ] 2 P 4, tmde quis pracdo ferox olim conieci vivitne A2 E2 0 1 1S3 rumpitne F2 Typhoia nonnu/li cervixne rupit F 10 h V 3 1S4 ruptanc c L 8 T 3 ISS alcyoneus M4 n 0 6 P 8 : alcioneus D ] 3 Kr M 2 R 1 , !3, A 2 P 3/4, } 1 per stagna o o tyrrhena K r L6fro, E 2 J 4 P 3 R2, J r recnrrit L6 0 0 o

0

169

I7o

I75

ISo

ISs

DE RAPTU P R O S ERPINAE

190

195

2oo

an vicina mihi quassatis faucibus Aetna protulit Enceladum ? nostros an forte penates adpetiit centum Briareia turba lacertis ? heu ubi nunc, ubi nata mihi ? quo mille ministrae ? quo Cyane? volucres quae vis Sirenas abegit ? haecine vestra fides ? sic fas aliena tueri pignora ? ' Contremuit nutrix, maerorque pudori cessit et aspectus miserae non ferre parentis emptum morte velit, longumque inmota moratur auctorem dubium certumque expromere funus. vix tamen haec : ' acies utinam vaesana Gigantum hanc dederit dadem ! levius communia tangunt. sed divae, multoque minus quod rere, sorores in nostras (nimium !) coniuravere ruinas. insidias superum, cognatae vulnera cernis invidiae. Phlegra nobis infensior aether ! Florebat tranquilla domus ; nee limina virgo linquere nee virides audebat visere saltus 186 aut L I R3 discussis e 1 187 propulit E 2 : extulit R 4

V 2j3 ignibus A I D L io M 3 0 4 enna 0 1 aut L IO, R25 189 heu a, C 1 G 1 , L 2 P 3 (p.c.) R25 : heus af3y ubi nunc ubi nata mihi F 2 : ubi nunc est nata mihi af3y : ubi nunc es nata mihi 0 2 P2, R 2 : ubi nata mihi nunc est R I , H : ubi nunc nunc nata mea est Y 2 190 quae vis volucres M 1 sirenas 0 4 R 1 , f3y : -es a, C 1 F2 L4 R 3j4, y, ] 1 191 hecine b 1 : heccine ceteri 192 timori b 1 , J 4 193 cessit a, J 2j4 0 1 P 3 (p.c.) U : cedit af3y nee R25 195 ' hie versus aberat a aspectum L6 M2 (p.c.)/3 (p.c.) vet. codice'-Claver. actorem A 1 ] 3 L 10, C 1 F 3 , E 2 P 3 R2 U (s.c.), ] 1 : raptorem C 3 F I K I M2 w, R4, F4 H 0 1 P I certumve P 2 (v.l.) ex­ ponere L 6 M2 (p.c. ) , ] 2 L2/3 P 3 U, ] 1 196 vix tandem R2 : vix tantum r u 197 hanc cladem dederit ] 3 L s/6 W, C 1 L4, C 2 : hanc dederit sortem D ? M 3 : hanc dederint cladem R 2 tangant Heins.: laedunt A I L 10 (p.c.) M 3 0 4 W, v.l. in 0 2/3 P2 198 quam rere non ita pauci : quas rere a 199 in vestras L6 conspiravere L 5 W querelas C 1 200 cognatae R I , C 1 E 1 F2 G r R6, y, J r : cognataque af3y sentis a e 1 h V2j 3 201 phlegra B L r infensior a, C r F2 L4 0 3 (p.c.) R4, A2 E 2 R 2 : infestior a, B, ]2 (p.c.) L 2 U : incensior A I F r R r (a.c.) R 7 T, f3, ]2 (a.c.) P 3j4, J r : offensior 0 4 aer 0 1 202 non e 1 203 viridis M 4 1 70

LIBER TERTIUS

praeceptis obstricta tuis. telae labor illi ; Sirenes requies ; sermonum gratia mecum, mecum somnus erat cautique per atria ludi : cum subito (dubium quonam monstrante latebras rescierit) Cytherea venit suspectaque nobis ne foret, hinc Phoeben comites hinc Pallada iunxit. protinus effuso laetam se :fingere risu nee semel amplecti nomenque iterare sororis et dura de matre queri, quae tale recessu maluerit damnare decus vetitamque dearum colloquia patriis procul amandaverit astris. nostra rudis gaudere malis et nectare largo instaurare dapes. nunc arma habitumque Dianae induitur digitisque attemptat mollibus arcum, nunc crinita iubis galeam laudante Minerva inplet et ingentem clipeum gestare laborat. Prima Venus campos Aetnaeaque rura maligno ingerit adfatu. vicinos callida flores ingeminat meritumque loci velut inscia quaerit nee credit quod bruma rosas innoxia servet, cernere M 3 silvas a k 204 astricta L I D R s/7 : obstructa A I F I M I P s (m.a.), L I 0 2 R 6 (a.c.) , y, ] I : instructa C 3 L 6 0 4 meis E I 206 nee non pro mecum M I cunctique F 2 : tantique R I (a.c.) R 5, 0 3 (a.c. ), J 2 0 I : tacitique A I L i o : castique F I K I L 6 M 3 , H : gratique L s W : solitique P s 207 dubium est quonam o:, b i F 2 (p.c.) R 3 , C 2 H latentem R I 6 : -es R I 3 208 prescient o:y 209 socias pro comites R I4j26 duxit C 3 M I , C I F 3 L 4 (p.c.) R4, prob. Heins. : sumpsit R2 210 s e laetam b I , L2 lusu r 213 maluerat R 2 vetitoque C 2 R2 e t Isengr. marg., prob. Heins. deorum F I L ID P s R s/7, R6, F4 H 214 alloquio P s (v.l.) patris F I }3 L6 M 2j3, F 3 LI, C2] 2 (a.c. ) ]4 L 2 U : procul a L io procul amandaverit (em-) b I G I ?, A2 P 3/4, } I : . . . -erat P 5 ? (p.c.), !3, C 2 L 3 R 2 : procul absentaverit K I M 3 0 4 R I : . . . -erat L6, R4, ]4 0 I : patriis mandaverit L ID horis K I 215 ilia rudis L 3 216 arma arcumque ] 2 (cj. Stat. Ach. I. 352) 217 digitis (om. -que) K I L s M3 R7 (a.c.) W, b i , J4 R25 attemptat o:, B E I F2 L I , y : attentat A I F I L6/Io, b i L4 R6, A2 ]4 P 3 R2 : aptavit L s R s/7 W, C 2 : aptatur Heins. 218 gaudente p I : ludente r 220 hen­ neaque 0 I 221 affatu o:, B F 2 0 3 (p.c.), A 2 E 2 (p.c.) ]4 R2, J I : affiatu 222 ceu nescia e I o:j3 y

171

205

210

215

220

D E R A P T U P R O S ERPINAE

ns

230

235

240

quod gelidi rubeant alieno gennine menses verna nee iratum timeant virgulta Booten. dum loca miratur, studio dum flagrat eundi, persuadet ; teneris heu lubrica moribus aetas ! quos ego nequiquam planctus, quas inrita fudi ore preces ! ruit illa tamen confi.sa sororum praesidia ; famulae longo post ordine Nymphae. Itur in aeterno vestitos gramine colles et prima sub luce legunt, cum rore serenus alget ager sparsosque bibunt violaria sucos. sed postquam media sol altior institit axi, ecce polum nox foeda rapit tremefactaque nutat insula cornipedum pulsu strepituque rotarum. nosse nee aurigam licuit : seu mortifer ille seu mars ipsa fuit. livor permanat in herbas ; defi.ciunt rivi; squalent rubigine prata et nihil adflatum vivit : pallere ligustra, expirare rosas, decrescere lilia vidi. ut rauco reduces tractu detorsit habenas, 225 booten F q : boot (h)em (-e) 224 gramine ex, L r , ]2 O r R2 U ] 3 P s , !3, A 2 E 2 O r (a.c.) P J/4, J r 227 hec L4 motibus Burm. 228 nequicquam gemitus W, 0 3 (v.l.) 229 abit illa R 2 230 online tendunt r 2 3 1 intrat i n R2 montes F 3 : campos ex, b I R 3 , R25 232 legit R r o dum R r sereno r 233 albet ' quidam codices habent'-Parrhas. 234 acrior R2 altior ex (s)titit (altius R 14/26) ex, 0 2 P 2, y : institit altior C 2 et Isengr. marg. : exstitit altior W : astitit altior b I axi J 3 R s/1 T, !3y, J r : axe ex, L4 P 2 R4, y : orbi R 2 : orbe IJI 235 atra pro foeda K r , F4 ]4 P r : alta D tegit K 4 : subit G 3 tremefactaque mutat R s , R 3 R6 (a.c.) , C 2 L 2 R25 : tremefacta remugit 0 I U (a. c.) 236 pulsu strepituque W, b I, prob. Heins. : strepitu pulsuque ex!3y : cursu strepituque A 2 : strepitu cursuque K r , ] 4 : fremitu pulsuque e 2 237 seu letifer C I ipse a F 6 k Y 2 : aestus Claver. 238 illa F I P 5 T , L I , C 2 R25 luror laudat Parrhas. permansit D ? K I M I : pervenit Claver. herbis K I M r , b r 240 et nihil ex, F 3 L4 0 2, ] 2/4 L 3 : et nil ex!3y affiatu A r K r M 3 (p.c.) R r , b r , ]2/4 L 2 U : affectum R 3 viruit R r3 .2.41 livescere P2 (v.l.) : marcescere L ro (p.c.) , U (p.c.) , et Koch : arescere vel flavescere Heins. olim 242 reduces rauco K I, E 2 J 4 tractu reduces P 5 , Or detorquet R s : deduxit L2 : detraxit G2 O s 0 6 P S 1 72

LIBER TERTIUS

nox sua prosequitur currum, lux redditur orbi. Persephone nusquam. voto rediere peracto nee mansere deae. mediis invenimus arvis exanimem Cyanen : cervix redimita iacebat et caligantes marcebant fronte coronae. adgredimur subito, casus scitamur eriles (nam propior cladi steterat) : qui vultus equorum ? quis regat ? ilia nihil, tacito sed laesa veneno solvitur in laticem : subrepens crinibus umor liquitur in roremque pedes et bracchia manant nostraque max lambit vestigia perspicuus fans. discedunt aliae ; rapidis Acheloides alis sublatae Siculi latus obsedere Pelori accensaeque malo iam non inpune canoras in pestem vertere lyras : vox blanda carinas alligat; audita frenantur carmine remi. sola domi luctu senium tractura relinquor.' Haeret adhuc suspensa Ceres et singula demens ceu nondum transacta timet; max lumina torquens ultra ad caelicolas furiato pectore fertur. 243 subsequitur R I : per- L6 P 5 , 0 2 P 2 currus R 5/7 244 nusquam est K I M I W : est nusquam L 5 : nulla est \jJ risere Burm. 245 undis pro arvis A 2, prob. Heins. 247 fronde P 5 (p.c.) , P 2 248 subito ]3 K I L6 T, b i C I L4, ]2/4 P4 R25 U (a.c.) : subitos Burm. : subitae v.l. in R7 : tacite o:j3y : taciti G I 0 3 R6 : tacitam M 3 , 0 I : propius A 2 P 3 casus J 3 L 5, b I F 2 R6 (a.c. ) , L 3 : casusque o:j3y : et casus U et Claver. casum . . . herilem L I scrutamur Isengr. marg. 249 cladi propior W steterat cladi M I : cladi fuerat R25 qui j3, F 4 R25 : quis o:j3y 250 qui j3, E2 R25 ista R 20 lapsa D, E2 (p.c.) ] 4 : lassa E2 (a.c.) P 4 dolore b I 251 latices 0 3 subrepens scripsi : subrepit (sur-) o:j3, F4 L3 P I R2j25, ] I : subrepsit (sur-) o:j3y 252 manus et R I , C 2 bracchia k m n : brachia vett. 253 perpetuus fons L 4 254 diffu­ giwlt H reliquae J 2 rapidisque V 2/3 255 caput e I h V 2 invasere F I4 : insedere Heins. 256-7 canoram / . . . liram L 5 258 obligat J 4 R2 259 domo F 2 (a. c.) R4 26o meret L 4 261 numquam J 2 fulmina (cum vultu) Buecheler 262 ultro b i , ] 2 : vultu o:j3y : multum U (v.l.) et Isengr. marg. : sic \jJ in pro ad Isengr. marg. : vocem om. R25 caelestes \jJ

!73

z4s

zso

zs s

z6o

D E RAPTU PR O S ERPINAE

265

270

275

arduus Hyrcana quatitur sic matre Niphates, cuius Achaemenio regi ludibria natos avexit tremebundus eques ; premit ilia marito mobilior Zephyro totamque virentibus iram dispergit maculis iamiamque hausura profunda ore virum vitreae tardatur imagine formae. Haud aliter toto genetrix bacchatur Olympo ' reddite' vociferans : ' non me vagus edidit amnis ; non Dryadum de plebe sumus ; turrita Cybele me quoque Saturno genuit. quo iura deorum, quo leges cecidere poli ? quid vivere recte proderit ? en audet noti Cytherea pudoris ostentare suos post Lemnia vincula vultus ! hos animos bonus ille sopor castumque cubile praebuit ? amplexus hoc promeruere pudici ? nee mirum si turpe nihil post talia ducit. quid vos expertes thalami ? tantumne relictus furiali IJl : furioso U : furibundo L6 ferri Mediceus Heinsii 263 ceu pro sic Schrader 264 achaemenio E r (p.c.) 0 2 P 2 265 avexit D, b r G I R6, E 2 ]2 R 2 5 : advexit a!3y : adduxit L 5 W, F 4 } 4 O r : avertit p 2 premit Heins. : fremit a!3y : furit M r , A 2 P 3 ista L7 R I4 (a.c.) R26 266 tantamque b I R 3 nitentibus F 4 P I 267 dispergit a, b I C I F 2 P 2 (p.c. ), y : dispersit a!3y : aspersit A2,} I iamiamque Claver. : nimiumque a!3y : timidumque P 5 et Heins. : vivumque A I L io R 5/7 : inhiatque Isengr. marg. hausura a, b I E I F 2 (p.c.) 0 3 (p.c.) P 2 (p.c.), A 2 L 2 0 I P 3 R 2, J I : haustura a!3y 269 genetrix c G4 K 4 : genitrix ceteri gen. toto L2 R2 toto bachatur mater olimpo J r 270 nec me F r, 0 3 (a.c.) 271 nec L 5/I o Cybebe Bentleius et Cortius 273 leges abiere K I , } 4 : leges cessere Heins. cecidere poli leges M r 274 profuit L6 276 dedit pro bonus R r ip se E I , 0 I : iste C 2 sopor ille bonus H 278 non L 10, C 2 mirum est Lw 279 om. R5 (add. marg. m.a ) R7 (add. marg. m.a.) : expungit 0 3 (vide infra) 28o-36o om. omnes classis !3 et praeterea A2 C 2 E2 F 4 } 2 (habet inter 1Il250 et 2 5 1 ) J 4 (add. ad .fin. libri) L2 (post v. 361 add. cum repetito v. 361 et seqq.) L3 O r (vv. 338-60 post 230 positi sunt; ceteri nusquam apparent) P 3/4 R25 Z, 28o-361 om. P r (add. marg. m.a.) R2 plurimi codd., non minus integri Jr quam decurtati, versum subditivum hie exhibent, ideo nimirum excogitatum ut sensum v. 279 expleret, lenemque ad v. 361 transitum efficeret. carent versu interpolato sequentes libri : F r J 3 M r /3 P 5 R I/5 T w (in C 3 vv. 221-80, in 0 4 vv. 224-95 desunt), F 3 G I R6, H P I ; ceteri aut omnis honor recti vobis. sic fata recedit vel .

174

LIBER TERTIUS

virginitatis honor, tantum mutata voluntas ? iam Veneri iunctae sociis raptoribus itis ? o templis Scythiae atque hominem sitientibus aris utraque digna coli ! tanti quae causa furoris ? quam mea vel tenui dicto Proserpina laesit ? scilicet aut caris pepulit te, Delia, silvis aut tibi commissas rapuit, Tritonia, pugnas ! an gravis eloquio ? vestros an forte petebat inportuna choros ? atqui Trinacria Ionge, esset ne vobis oneri, deserta colebat. quid latuisse iuvat ? rabiem livoris acerbi nulla potest placare quies.' His increpat omnes vocibus. ast illae-prohibet sententia patrisaut reticent aut nosse negant responsaque matri dant lacrimas. quid agat ? rursum se victa remittit inque humiles delapsa preces : ' ignoscite si quid intumuit pietas, si quid flagrantius actum quam miseros decuit. supplex deiectaque vestris simm. (singula in catalogo codicum reperies) aut hec ait et lectura faces altum nemus intrat vel simm. offerunt. versus hec ait etc. exstat ante 279 in K 3 L7, pro 279 in 0 3 (marg. e corr. vide supra), et post 279 in P 4 : versus omnis etc. occurrit post 279 in D K I L 5 M2, B h i C I E I F 2 L I P 2 R 3 R4, A2 C 2 E 2 F 4 ]2{4 L2{3 O I P 3 R2{25 Z, J I ; post 280 in L 6 W ; post 281 in A I L 10 ; et in marg. ascriptus in R 7, L4 0 2, U 28o hinc incipit V honos ex, J 2{4, V ? voluptas L 5 M I 281 veneris F I 5 h P7 (a.c.) sociis iunctae K I , J 4 : et sociis iunctae '+' sociae 282 in temp lis L 5 M 3 W hominum P 5 raptoribus Baehrens estis Jeep et Heins. : -es a e 2 k arvis e2 284 quem J 3 (a.c.) : quo H tenui dicto F I R 5{7, L 2 U (p.c.) , V : dicto tenui ex, H ] 2{4 : digito tenui U (a.c. ) : tenui digito Reinhardt ledit L 5{6 285 a caris T 2 : ex caris a k 286 permissas e 2 287 an gravis K I L 5{6 T W, J4, V : aut gravis ex, L2 U vestros aut D L IO M I/3 (a. c.) R I 288 atqui D ? P 5 Trinacrida m et Birt 291 dies K I L6jw, J 4 improbat r ambas a k R I3/I6 291-2 vocibus omnes f increpat W 292 illas A I K I M I/3 R I (p.c.) , ]2 U (sscr.), V : iste R 5 reverentia ex, J4 , V post hunc vers. sequitur i n U v . spur. omnis etc., tum v . 361 et seqq.; vv. 293-360 post v. 448 positi sunt 293 scire negant L 2 294 rur­ sum L 5/6 M 3 (p.c.) W, H ]2 (m.a.) L 2 : rursus a, ]4 U 295 delapsa ] 3 R 5/7 : dilapsa a, H ] 2 L2, V : devecta P 5 : demissa F I K I M I R I T, ] 4 296 actum J 3 K I M 2 R I , J4 U : actum est a, H J2 297 miseras decuit

175

28o

285

290

295

DE RAPTU PROSERPINAE

300

305

310

315

advolvor genibus. liceat cognoscere sortem, hoc tantum, liceat certos habuisse dolores. scire peto quae forma mali ; quamcumque dedistis fortunam, sit nota, feram fatumque putabo, non scelus. aspectum, precor, indulgete parenti ; non repetam. quaesita manu securus habeto, quisquis es ; ad:firmo praedam ; desiste vereri. quod si nos aliquo praevenit foedere raptor, tu certe, Latona, refer ; confessa Diana forte tibi. nosti quid sit Lucina, quis horror pro genitis et quantus amor, partusque tulisti tu geminos ; haec una mihi. sic crine fruaris semper Apollineo, sic me felicior aevum mater agas. largis nunc imbribus ora madescunt ; quid tantum dignum fleri dignumque taceri ? Hei mihi, discedunt omnes ! quid vana moraris ulterius ? non bella palam caelestia sentis ? quin potius natam pelago terrisque requiris ? D M 2, H : miseram decuit L s W, L2, V : decuit miseras K I L6 R sl7, ] 4 : decuit miseros L 10 R I ? (a.c.) : decuit miseram M I R I (p.c.) : deceat miseras T deiectaque Isengr. marg., prob. Heins. : miserandaque vett. quotquot praesto sunt excepto L2 (venerandaque) 298 advolvor pedibus W 299 novisse H doloris L 2 : furores a, J 4 : timores R2o 300 scire velim M I W quae­ cumque H : cuicumquc F wlu h dedisti R I 301 fortunae a, H ] 4 quamcumque daturac I Fortunam vel quodcumque dedistis I Fortunae Heins. si H et Claver. , prob. Jeep rapta K I, J 4 303 repeto J 3 305 vos a, H]4 L2 U, V munere a, J 4 306 tu saltern K I L s T W, L2 nam fassa U diana est m p i Q V3 307 tibi est L io M r 04 W quid horror ] 3 W , L 2 U : timor quis C 3 L6 0 4 : timorque \j.l 309 mihi est c L 8 T 3 310 tu felicius D ? M 3 311 nunc A I C 3 L slro 0 4, primus recepi : tunc a, H ]2l4 L 2 : num en? fletibus T verba largis tunc imbribus ora madescunt interpolata putat Baehrens, extra orationem Cereris locant Gesner, Birt et Koch 312 ' hunc versum non habuit vet. liber Claver. suspectum habebant Barth et Heins. , delevit ]eep quid tamen est C3 M 3 (p.c.) 0 4 : quid tandem Heins. : quid tantum est D ? M 3 (a.c. ) : quid tantum in- J 2 : quid tanto R2o dignum est K r fieri W, U : fletu R I (a.c.) R2o : flere est P7 tacere P 7 : fateri U (v.l.) : timeri P s 313 hei A r F I ]3 L IO T : heu a, H ]214 L 2 U, V (marg.) : ha P 5 314 num a, U 315 pelago natam L2 pel. terraque '-

LIBER TERTIUS

accingar lustrare diem, per devia rerum indefessa ferar. nulla cessabitur hora, non requies, non somnus erit, dum pignus ademptum inveniam, gremio quamvis mergatur Hiberae Tethyos et rubro iaceat vallata profundo. 320 non Rheni glacies, non me Riphaea tenebunt frigora, non dubio Syrtis cunctabitur aestu. stat ftnes penetrare Noti Boreaeque nivalem vestigare domum; primo calcabitur Atlas occasu facibusque meis lucebit Hydaspes. 325 inpius errantem videat per rura, per urbes Iuppiter ; extincta satietur paelice !uno. insultate mihi, caelo regnate superbi, ducite praeclarum Cereris de stirpe triumphum.' haec fatur notaeque iugis inlabitur Aetnae 330 noctivago taedas informatura labori. Lucus erat prope flavum Acin, quem candida praefert saepe mari pulchroque secat Galatea natatu, densus et innexis Aetnaea cacumina ramis qua licet usque tegens. illic posuisse cruentam 335 aegida captivamque pater post proelia praedam advexisse datur. Phlegraeis silva superbit T : terris pelagoque A I L r o : terra pelagoque R2o requiras K I L6 R I (a.c.) : requiram R I (p.c.) 318 nee e i nee M I w, L2 319 hibero P9 R2o 320 Tethyos fi : tethios L6 M2 : tetios P I (m.a.) : thetios o:, H J 2/4 L2 aut L2 lateat D M 3 velata C 3 M I/3 04 321 non . . . nec H U : nec . . . nee M 2 w, ]2, V glacies r (h)eni D M 3 Rhipaea G4 et edd. nonnulli 322 nee 323 fmes K I L6 R I , ] 4 : -em o: , H ]2 L 2 U, 0:, ]2 U syrtes dubio L 2 V 325 oceano fac. J 4 326 per castra K I, J 4 327 saturetur R2o : placetur ] 2 328 superbe L6 : supemo ]2 : superni M 3 329 de prole L6, ] 2 : de gente M I 330 sic fatur K I , ]4 notisque o:, L2 U allabitur H 331 montivago k informatura F I ] J L s M2 W, H L 2 : instauratura R I T : inflammatura o: , J 2/4 U, V 332 flavum cod. Saletani apud Claver. : flumen cett. quem o:, U, V, cum cod. Saletani : quod o:, H ]2 (m.a.) ]4 L2 profert FI M 2 P s, H ab hoc versu lib. quartum incipere putat Jeep 333 meatu T 2 334 cacum. silvis o:, H L 2, V 335 libet A 1 C 3 D M 3 , L2 336 patrem R I6 337 ad(d)uxisse D L i o M 3 R s/7, V : eduxisse 12

1 77

HCD

DE RAPTU PROSERPINAE

340

345

350

355

exuviis totumque nemus victoria vestit. hie patuli rictus, hie prodigiosa Gigantum tergora dependent, et adhuc crudele minantur affixae truncis facies, inmaniaque ossa serpentum passim cumulis exanguibus albent, et rigidae multo suspirant fulmine pelles ; nullaque non magni iactat se nominis arbor : haec centumgemini strictos Aegaeonis enses curvata vix fronde levat ; liventibus ilia exultat Coei spoliis ; haec arma Mimantis sustinet ; hos onerat ramos exutus Ophion. altior at cunctis abies umbrosaque late ipsius Enceladi fumantia gestat opima, summi terrigenum regis, caderetque gravata pondere, ni lassam fulciret proxima quercus. inde timor numenque loco, nemorisque senectae parcitur, aetheriisque nefas nocuisse tropaeis. pascere nullus oves nee robora laedere Cyclops audet et ipse fugit sacra Polyphemus ab umbra. Non tamen hoc tardata Ceres. accenditur ultro C 3 phlegraeis D ]3 M r 0 4, U 339 hinc F r T : sunt L ro hie D K r L 6 M r W, ]4 : et a, H ]2 L 2 O r U, V 340 tegmina R29 : corpora P 5 341 facies truncis D K r M 3 T, H J 4 342 ex. ardent R 5 343 merito pro multo T respirant P7 T 3 (p.c.) £amine 04 344 iactatmagna (sic) J 2 se iactat A r C 3 L sfro M 3 T, V arbos F r 345 fractos pro strictos c L 8 T2/3 , prob. Heins. 346 fronte] 3 K 1 R 5/7, L 2 gerit]2 347 Coei Parrhas. (' alii codices habent' ) : Zancli G 4 P 8 et Parrhas. : cac(h)i codd. cett. : Rhoeti Heins. rnimantis A r F r (p.c.) ] 3 M 2 T : minantis a, H ]2/4 U, V 348 omat ramos c V 3 : ramos onerat K 4 exustus F r R 5 ophion G 4, excerpta Laeti, Parrhas. : ofyon A r : osyon L ro R s/7 : ophias (of-) a, J4 349 his cunctis L6 M r , L 2 abies cunctis W 350 gessit 0 I 352 lassam F r ] 3 R r W w, H : lapsam a, L2 U, V : lapsum D K r , ]4 O r : lapsus F r6 m R I I/!7 maxima R s/7 353 unde L6 loci R r 3 : loco est L ro 354 -que om. C 3 0 4 P 5 (s.c.), O r non fas pro -que ncfas 0 4 P 5 (a.c.) : nee fas e 2 R2o 355 nullus pascit oves L6 R I ledcre robora a, H J 2 : robora cedere M2 W, U, V : robur scindere R s/7 357 nee a, L2 hinc W : est R s/7 turbata L6

tiBER TERTIUS

religione loci vibratque infesta securim (ipsum etiam petit ira Iovern) ; succidere pinus aut magis enodes dubitat prosternere cedros, exploratque habiles truncos rectique tenorem stipitis et certo pertemptat bracchia nisu. sic, qui vecturus longinqua per aequora merces molitur tellure ratem vitamque procellis obiectare parat, fagos metitur et alnos et varium rudibus silvis accommodat usum : quae longa est, tumidis praebebit cornua velis ; quae fortis, malo potior; quae lenta, favebit remigio ; stagni patiens aptanda carinae. Tollebant geminae capita inviolata cupressi caespite vicino, quales non rupibus Idae miratur Simois, quales non divite ripa lambit Apollinei nemoris nutritor Orontes.

36o

365

370

358 libratque D infensa K4 (v.l.) : incerta D : irata M2, U 359 seu pro etiam lsengr. marg. petit ira J. J. Scaliger: petitura K I L 6 R s/7 (p.c.) W, ] 2/4 : peritura R7 (a.c.) et Claver. in notis : feritura ex, H Ox U, V: lesura D M 3 T, L2: nocitura R I : fractura R29 et Jeep : petat ira Barth (cum incerta v. 3 5 8 ) : tactura Reinhardt : tritura Hertel : spretura Baehrens : feriatne Birt (cum incerta) pinos P 5 360 et G 4 et Parrhas. : an V et Heins. : ac ljl : at K I R s/7, ] 4 : vel F 7 minus a Y2 properat a , ]2/4 : cupida est jeep consternere r pinus W (a.c. in cedros) pinus ct cedros inter se mutare olim volebam post hunc vers. a F6 k habent omnis etc. et tum v. 389 et seqq. 361 (h)abiles L sfw 04 R7 (p.c.) W, b i R4, H L2 U, V: abiens A I ]3 K I M I T, E I F 2 L 4, y, J I : abies exj3, ]2 0 I R25 : obiens Gesner rectumque G x R6, P 4 post hunc vers. C 3 habet hec ait etc. 362 pertemptat C 3 F I ? P 5 , h i C I , C 2 ] 2 L 3 Z : pertentat ] 3 R I , L 2 R25 : praetemptat ex, ] 4 O I U : praetentat L 6 R s , L 4 : percenset M2 T, j3 , E 2 P 3/4 R2, ] 1 V ? : perlustrat A 2 bracchia n p I : brachia vett. visu codd. aliquot : nixu ljl post hunc vers. M I habet hec ait etc. 363 vecturam P 2 (p.c.), recipit Heins. 366 varios . . . usus R IJ/16/20 (et rud. silv. var. ace. us.) rudibus varium M I 367 cornua ventis A2 P 3 : carbasa velis R2 368 malo ex, b 1 E I F 2 L4 (itJ ras.), y, J I : 369 remigiis M2 malis L 6 : clavo F I M 2 0 4 R s/7. i3y malo est M I aptata H: est apta F 2: optanda C2 J 2: optata 0 I carine est M 3 : carinis C 3 0 4, R2 370 cupressi j3, P I (p.c.) P4 R25 : cupressus exj3y : cypressus A I (p.c.) L s M I P s, H hi11c deficit C I 371 non a, b i E I G I L 4 R6, y, J I V : nee H: in exj3y vallibus L2 372 in pro non L 2 : nunc A I R7, L I : sub K I 373 servator P 9 anapis T, E I , A 2 P 3 , J I

1 79

I2-2

DE RAPTU PROSERPINAE

375

380

385

390

germanas adeo credas, sic frontibus aequis extant et socio despectant vertice lucum. hae placuere faces ; pernix invadit utramque cincta sinus, exerta manus, armata bipenni, alternasque ferit totisque obnixa trementes viribus inpellit. pariter traxere ruinam et pariter posuere comas campoque recumbunt, Faunorum Dryadumque dolor. conplectitur ambas sicut erant alteque levat retroque solutis crinibus ascendit fastigia mantis anheli exuperatque aestus et nulli pervia saxa atque indignantes vestigia calcat harenas : qualis pestiferas animare ad crimina taxos torva Megaera ruit, Cadmi seu moenia poscat sive Thyesteis properet saevire Mycenis : dant tenebrae Manesque locum plantisque resultant Tartara ferratis, donee Phlegethontis ad undam constitit et plenos excepit lampade fluctus. Postquam perventum scopuli flagrantis in ora, 374 frontibus L6 04 P 5, 0 2j3 P 2, C2 H: frondibus af3y 374-5 astant / equis R27 375 ex(s)tant a, b I E I F2 L4, y, J r : a (d)stant A I K I L ro M I , f3y, V : constant C 2 socium R I , J I dispectant M I : pro- A I K I , J 4 vert. laurum P 4 376 velox pro pernix C 2 R25 invasit R r, L 2 378 alter­ nansque F 3 secat P 4 totis (om. -que) A I C 3 L 5 0 4 P 5, B 0 3 (add. -que sscr.), ]2 O r , } I innixa R I W trementis 0 3 (a.c. ) 379 ruinas ]2 380 pressere J I comas a, F 2, C 2 J 2 0 I : -am af3y posuere comam par. K I , ]4 campisque e2 383 anhelans Heins. 384 exceditque MS Pulmanni post 385 P 4 habet vers. omnis etc. 386 armare R I4 (non habet Vat. prim. Heinsii R26) : animarum 0 3 (p.c.) P 2 (p.c.) et Claver. in Coronide Miscella ad Claudianum Cap. 1 2 : fl.ammare Birt proelia D M 3 W, 0 3 taxus W 387 rapit M I : solet L 2 seu cadmi D L ro M 3 P s, G r R6, R25 limina K I , A 2 ]4 (p.c.) L 2 poscant }4 (m.a. ) : -it C 3 T, ]2? P4 388 properat af3y : tendat ]2 R25 390 Phlegethontis ed. Perusiana anni 1 481 (?) : phlegetontis, fl.egetontis etc. codd. ad R I, 0 3 (p.c.), A 2 H P 3 R 2 Z, J I : in af3y undam R I , A2 H P 3 Z, J r : -as M 3 , 0 3 (p.c.), L2 R 2 : -is C2 U (p.c.) : -a af3y 391 adstitit b I sus cepit F 2 : accepit K I L6, b I, J 4 392 perventum F I J 3 K I L s w, f3y, J r : perventum est a, b I F2 (p.c.) L4 P 2, C2 F 4 (p.c.) L3 U, V: est perventum }4 (m.a. ?) : pervenit L6, L2 ferventis =

1 80

LIBER TERTIUS

protinus arsuras adversa fronte cupressos faucibus iniecit mediis lateque cavernas texit et undantem flammarum obstruxit hiatum. conpresso mons igne tonat claususque laborat Mulciber : obducti nequeunt exire vapores. coniferi micuere apices crevitque favillis Aetna novis ; stridunt admisso sulphure rami. tum, ne deficerent tantis erroribus ignes, semper inocciduos insopitosque matiere iussit et arcano perfudit robora suco, quo Phaethon inrorat equos, quo Luna iuvencos. Iamque soporiferas nocturna silentia terris explicuere vices ; laniato pectore longas incohat illa vias et sic ingressa profatur : ' non tales gestare tibi, Proserpina, taedas sperabam, sed vota milii communia matrum, et thalami festaeque faces caeloque canendus ante oculos hymenaeus erat. sic numina fatis volvimur et nullo Lachesis discrimine saevit ! F 2 ad L 5 W, C 2 horam R r , L 2 : oras D K r 393 adversa ex, E 1 L4 02 R4, y, ] I V : aversa ex!3y fronde R r ? , 0 2, L2 cupressos O r R25 : 395 undantes . . . hiatus -us ex!3y 394 incendit R s/7, C 2 : inmictit a F I M r , R2 flammis F r M3 (p.c.) R r , E 2 ]4 L2 O r , J I obtexit P s (a.c.), b r , Z: texit MI 396 oppresso L 2, V ? : et presso M r mons ex, B F 3 L r 0 2/3 (p.c.), y, V : mox ex!3y sonat L6, F 3 : fremit W 397 exire ex, 0 3 (p.c. ) P 2 R 4, y, J I V : haerere ] 3 M r w ?, !3y 398 horrifici L 4 nituere L 2 : crepuere R 1 (p.c.) faces K I , V crescitque D : crepuitque P 5 (a . c. ) 399 henna 0 I stridunt C 3 ] 3 P s, !3, E 2 F4 P I R25 U, V : -ent ex, b I P 2, y, J r : -entque Birt fulgure D, F4 P r saxa A2 400 tum C 3 F r ] 3 , !3y : tunc ex, F 2 0 2 P 2, y, J r V deficiant L 3 U (a.c. ) ardoribus K r , ]4 402 perfundit P s, L 4 (a.c.) 0 2 R 3 , P 4 403 et luna G2 iuvencas H et Claver. : iugales J 3 L6/IO, R 3 , L2 404 et iam pro iamque J 4 somniferis ] 4 : sompniferas U soporiferis D K r L6fro, C2 L2 : soporatis K4 (v.l.) 405 corpore C2 406 progressa R s/7 : egressa Ps 407 tedas . . . tales a k portare R I3/I6 408 spectabam M I 409 set 0 3 (p.c.) : sed E 2, prob. Heins. 410 ante deos v.l. in 0 2 P2 nunc numine fati K r 411 ducimur K I , ] 4 : solvimur R 4 en ] 2 lachesis nullo K I W, ]4

181

395

400

405

410

D E R A P T U PR O S E R P I N A E

415

420

425

430

quam nuper sublimis eram quantisque procorum cingebar studiis ! quae non mihi pignus ob unum cedebat numerosa parens ! tu prima voluptas, tu postrema mihi ; per te fecunda ferebar. o decus, o requies, o grata superbia matris, qua gessi florente deam, qua sospite nusquam inferior Iunone fui ; nunc squalida, vilis ! hoc placitum patri. cur autem ascribimus ilium his lacrimis ? ego te, fateor, crudelis ademi, quae te deserui solamque instantibus ultro hostibus exposui. raucis secura fruebar nimirum thiasis et laeta sonantibus armis iungebam Phrygios cum tu raperere leones. accipe quas merui poenas. en ora fatiscunt vulneribus grandesque rubent in pectore sulci ! inmemor en uterus crebro contunditur ictu ! qua te parte poli, quo te sub cardine quaeram? quis monstrator erit ? quae me vestigia ducent ? qui currus ? ferus ille quis est ? terraene marisne incola ? quae volucrum deprendam signa rotarum ? 412 sublimis eram nuper L 5 414 tu sola D M 3 415 mihi a, b I F 2 0 2 P2, y, V : mei C 3 D M2, j3, F4 ]2 P I/4, J I ferebar F I , E 2 (11./.), R2 : videbar aj3y 416 species F I4 o tanta K I, J 4 : o sola M I 417 deum ] 4 nusquam C 3 , B E I G I L I R6, E 2 ]2 (a.c.) O I : numquam aj3y 418 parens pro fui Ls sordida e I h V 2 tristis M 3 , J 2 (m.a. in lac.) : vultus R 2 : visus Heins. 419 patti placitum R 2 placuit A I L6/IO M2, L2 0 I , V sed cur 0 3 (p.c.) R 4 , L 2 ilium a , E I 0 2/3 (p.c.) P 2 R6 (p.c.), y, J I V : ullum D ] 3 M 2 0 4, j3y 421 quod t e C 3 , R 3 , C 2 quae solam Heins. 422 opposui P 5 secreta R I fovebar L 3 423 nee mirum C 3 D arvis L I P 2 : antris L IO (p.c.) M I , ]2 O I 424 cum tu a, E I F 2 P 2 R4, y, J I V (p.c.) : cum te L6 M I/2 R I W, j3y, V (a.c.) raperere a, h i E I F 2 P 2 R4, y,] I V (p.c.) : rapuere aj3y 425 nunc oraJ 4 patescunt R s/7 426 ver­ beribus a k sub pectore E 2 R2 427 en pectus C 2 : en venter R I 3/I6 certo 0 4: pleno K 4 iam tunditur p I 428 sub sydere L6, F 2 429 quo me L 5 , L2/3 R2, V ducant J x 430 qui currus M 3 T, j3, A2 E2 O I P 3 , J x : quis currus a, F2j3 L x P2, y, V ipse aj3, C 2 E 2 ]2 L 3 R25, V ? : iste P s quis ferus ipse quis est raptor U 431 qua rotarum a, C 2 : viarum aj3y R4, quod malebat Heins. : quo c

! 82

LIBER TERTIUS

ibo, ibo, quocumque pedes, quocumque iubebit casus. sic Venerem quaerat deserta Dione. proficietne labor ? rursus te, nata, licebit amplecti ? manet ille decor, manet ille genarum fulgor? an infelix talem fortasse videbo, qualis nocte venis, qualem per somnia vidi ? ' Sic ait et prima gressus molitur ab Aetna exitiique reos flores ipsumque rapinae detestata locum sequitur dispersa viarum indicia et pleno rimatur lumine campos inclinatque faces. omnis madet orbita fletu; omnibus admugit, quocumque it in aequore, sulcis. innatat umbra fretis extremaque lucis imago Italiam Libyamque ferit : clarescit Etruscum litus et accenso resplendent aequore Syrtes ; antra procul Scyllaea petit, canibusque reductis pars stupefacta silet, pars nondum exterrita latrat. 432 ibo tamen L 5 M2, L2 vocabit r 433 error sic F 14 queret P 4 434 proficietne F r (?p.c.) P s , b r P 2 R4, F 4 H ]2 (a.c.) P r U, V : efficietne cx�y : deficietne M 3 , L2 435 ille color L7 R26 436 tandem D (v.l.), E 2 R2 438-48 om. omnes classis � (in B add. Petrus Danielis, in Gr m. rec.) et praeterea A2 (add. m.a . ) C2 E2 ]2 (add. m. tert.) L2j3 P I ( 438-40 add. m.a.) P 3/4 R 2 Z (add. m.a.), Jr 438 hec ait G 2/3 Q r y primos gressus G r (m. rec.) : primo gr. K r L s M 3 , A 2 (m.a.) : gr. primos r in ethna M2 439 excidiique L ro 440 depressa O s : diversa D M3, ] 2 (m.a.) R25 441 plenos Gr (m. rec.), ]4 O r Z (m . a.) : prono Livineius scmtatur R s/7 441-8 om. F4 (cf. P r ) 442 omnis perit orbita A r (a.c.) P s , prob. Heins. (cf. Stat. Ach. I. 236) : largo madet orbita R I J/I6 luctu G r (m . rec.), O r 443 quacumque excerpta Laeti et Claver. it (om. in) M I W, G I (m. rec.), ]2 (m. tert.) 0 I R25 : iit Aldina aequora Ar L 10 04: aera Claver. : aethere ed. Scaligeri Raphelengiana sulcis Z (m.a. ) : fulvis ex, H J 4 : silvis M I K I W, A 2 (m.a.) ]2 (m. tert., p.c.) 0 I R25 U , V : fl.uctus J 2 (m. tert., a.c.) : fl.ammis R 1 2 : 444 innatat L s, Z (m.a.) et Heins. : adnatat (an-) lucis excerpta Schottiana cxy : abnatat cx, J2 (m. tert.), V 445 libiemque 0 I (unde Libyenque conicias) petit G I (m. rec.) 446 respondent L6 M3 P s W, G r (m. rec.), Rzs : splendescunt M r 447 petit . . . procul P 6 et Aldina reclusis M 3 , ] 2 (m. tert., s.c.) : remotis T : relictis R25 : refl.exis M I (sscr.) : revinctis G I (m. rec.) 448 non pro nondum M2, H O r perterrita M 3 : territa L s , A 2 (m.a.), V surgit R29

435

440

445

D E RAPTU PR O SERPINAE

Subscriptiones quotquot in vett. libris adsunt : explicit liber claudiani A 1 C 3 R 7 W w, C 2 H 0 1 : fmit fmit liber claudian B : explicit liber claudiani de r.p. b 1 F 2 : explicit ovidius [sic] deo gratias amen E x : explicit F x 0 4 (m.a.) P s T, R 3 , ]2 (post v. 437· del. m. tert.) P I (m.a.), V : explicit liber claudianus F 3 :

explicit tractatus claudiani deo gratias amen F 4 : FINIT CLAVDIANVS DE RAPTV I G x (m. rec.) : deo gratias amen }4 (post v. 360) : explicit liber deo gratias amen K x : explicit liber claudii claudiani secundus [sic] de r.p. L 3 : explicit claudianus L 5, J 2 (m. tert. post v. 448) P 4 : explicit liber Durfey (?) q primo naves invenit et navigare I cepit L6 (m.a.) : finito libro sit laus et gloria christo L 10, R25 : hie liber est scriptus qui scripsit sit benedictus explicit claudianus M 2 : explicit libellus claudii claudiani 0 2 : explicit claudianus I de r.p. P 2 : explicit liber claudiani minoris R 4 : finit claudius claudianus de r.p. R6 : explicit liber claudii claudiani tertius U (marg.) Subscriptiones recentiorum notabiles : explicit claudiani liber florentini de r.p. F s : claudiani de r.p. tertius et ultimus finit F9 K 2 : Claudii Claudiani I Proserpina I Explicit I Deo Gratias I Amen n : fmis claudi claudiani de r.p. : eius quod invenitur. nam ultimum volumen apud nos non extat 0 5 (eadem subscriptio in ed. principe Celsani, tmde et P 8 mutuatus est (?) et Vossius transtulit in R7 (fo. 55 v) ) : claudii claudiani de r.p. liber tertius et ultimus explicit feliciter R 8 : finis claudiani poete florentini R I I : fmis est tragedie heroice claudii claudiani poete suavissimi Numquam fefellit veritas R 2o : clarissimi poete claudiani opusculum feliciter explicit R28 : Claudi Claudiani de r.p. imper­ fectum opus I explicit foeliciter r : Explicit liber Claudiani florentini de r.p. V 3 PROSERPINE

C OM M ENT A R Y

As a rule only matters affecting the constitution of the text are discussed in the Commentary. Points of purely literary interest (e.g. details of the myth} are omitted except where they have some bearing on the text, and parallel passages illustrative only of Claudian's familiarity with his predecessors are generally ignored. Those who seek further information on such topics may be referred to the works mentioned in the chapter entitled ' Sources and Style '. I do not normally comment (a) on readings which are obviously false, or (b) on pairs or groups of variants which are so nearly synonymous that attempts to discriminate between them must be mere hair-splitting. Where a number of variants appear to be equally good I regularly (but not invariably) print that one which has acquired squatter's rights in the text : an arbitrary procedure, but an unavoidable one. NoTE:

C O M MENTARY

Title. In his opening note on the epic Heinsius remarks : ' Raptum Proserpinae auctor, puto, inscripserat, non de Raptu. quod barbarorum commentum. Sic Tristia, Fasti, Amores, Ars amatoria, & similia, apud Ovidium, non de Tristibus, de Fastis, atque ita porro. ' How well attested, one may ask, is the traditional title De Raptu Proserpinae ? Although less than one-third of the vetustiores prefix to the first preface an inscription by the first hand or a second roughly contemporary with the first, and although there is considerable variety of wording in such inscrip­ tions as do occur, the phrase de raptu proserpinae is nevertheless incorporated in the titles of A r F2 G r L r L 3 O r and P 3 . It also appears internally in various places : before book I in L 3 P 1 and P 3 ; before the second preface in F 2 and P 3 ; before book II in G 1 ; before the interpolated third preface in F 2 ; and before book m in F 2 . Final subscriptions in the first or a contemporary hand are more numerous, being offered by rather more than half of the vetustiores, and in b 1 F 2 L 3 P 2 and R6 the phrase de raptu proserpinae is again present. In the recentiores both inscriptions and subscriptions regularly include the phrase de raptu proserpinae. None of the alternative forms of title, of which liber (claudii) claudiani is perhaps the most common, has the appearance of authenticity, and there is not a trace of Heinsius' Raptus Proserpinae. Such evidence as is forthcoming from medieval catalogues of manuscripts also supports a form of words including the preposition de. At St. Gallen 1 and Reichenau2 in the ninth century the title was claudiattus de proserpina (no weight should, I think, be laid on the variation proserpina for raptu proser­ pinae ) , and in the eighth-century catalogue of Charlemagne's court library is included the item incipit glaudiani [sic] deraptu proserpine lib. IIJ.3 The title de raptu proserpinae is thus defmitely attested as far back as the eighth century. That it was also current in, or immediately after Claudian's own time may, I believe, reasonably be inferred from the forms of title prefixed to other works of his. In our MSS the poems on the wars against Gildo and Alaric are entitled, not Bellum Gildonicum and Bellum Gothicum, but De bello Gildonico and De bello Gothico. It is of course conceivable that the author himself never gave definite titles either to these war poems or to his unfmished epic on Proserpine, and that the titles by which we now know these works were devised by his literary executors or earliest editor, but it seems the merest caprice to abandon a title to which all our evidence points 1 See G. Becker, Catalogi bibliothecarum antiqui, Bonn, r 8 8 5 , no. 1 5 . 3 3 1 . 2 See P . Lehmann, Mittelalterliche Bibliothekskataloge Deutschlands uttd der

Schweiz, Miinchen 1918, 1.266. 5 . J See Becker, op. cit. no. 20.

C O MMENTARY

at the dictates o fa subjective feeling about its inelegance. Whether Claudian wrote De Raptu Proserpinae or not we cannot know, and the best course seems therefore to acquiesce in the title offered by the MS tradition.

PREFACE TO B O OK I Implicit in this preface is an analogy between the progress of the ftrst mariner and that of the poet, who has gradually advanced from less ambi­ tious enterprises to a final culmination in the present work. On the implica­ tions of this analogy see the Introduction, p. 100. Gesner, followed by Gramlewicz (O!:!_aestiones Claudianeae, Breslau, 1 877) and Koch (Teubner ed. pp. liv-lv ), inclined to the view that the end of this preface was lost ; a rough idea of what was felt to be missing may be got from the undoubtedly spurious distich cited by Burman (see my app. crit.). But, while it is indeed true, as Koch contends, that Claudian normally underlines the point of such elaborate comparisons by appending a few lines relating them to the real situation, as for instance in the prefaces to D.R.P. II, In Rufinum 1 and the Panegyric on Honorius' Third Consulship, there is one notable exception in the preface to the Epithalamium for Honorius and Maria (adduced by Jeep, Acta, pp. 3 57-8), where the point of comparison between the mythological and historical situations is so obvious as not to need explaining. Similarly here, no one could fail to perceive that behind the sailor stands the poet himself, 1 and explicit elucidation of the implied analogy is quite unnecessary. I None of the three word orders is impossible, and there is no knowing which is the original. For the juxtaposition of inventa and primus c£ Claud. 18.161 inventas primus . . . II The Isengrin's reading exultat, for which c£ Ov. Met. I . 1 34f/uctibus ignotis exultavere (v.l. insultavere) carinae, is certainly elegant and attractive, but it can scarcely be said to be superior to the MS inrumpit (inrupit is disqualified by its tense). The operative parallel is Stat. Theb. 6.23-4 (cited by Heinsius) tum pontum inrumpereJretae (sc. biremes) f longius ereptasque oculis non quaerere terras (the fleet has previously been exercising on a peaceful lake, tranquillo . . . lacu ), a passage which may well have given Claudian the general idea for this preface. I C£ Nemes. Cyn. s B-62 ; Paul . Petr. Vita Martini n init. (adduced by Barth) ; Venant. Fort. Vita Martini 1 pref. (adduced by Birt). The use of nautical images to portray aspects of poetry is in general quite common : cf. Prop. 3.3 .22 and Butler and Barber's note ad loc., supplemented by Shackleton B ailey, Propertiana, p. 296 ; also Virg. Georg. 2.41, 4 n 6 ff. ; al. .

188

C OMMENTARY

B O OK I 1-19 Introduction : the poet's vision of the Eleusinian mysteries and cult figures. I The Claudian entry in the eighth-century library catalogue published by Becker (op. cit. no. 20) quotes this line with the spelling adjlata, not ajfiata, which is what most of our MSS have. In order to come to a reasoned decision as to which of these spellings to adopt, we must first consider what general principles of orthography should be applied in editing Claudian. Since the MS tradition of the D.R.P. begins comparatively late, one is not surprised to find that it offers no dear guide to the spellings Claudian is likely to have preferred. The orthography of the oldest Claudian MS now extant, however, the eighth-century Veronensis containing the carmina minora, and that of the late but invaluable Excerpta Florentina of Claudianus maior reveals such a measure of agreement with that of the best ancient witnesses, including, most notably, the Virgil MSS A F G M P R V (c£ Birt, p. ccvii), that one may legitimately assume that Claudian, a purist in diction and metre, was classically correct also in his orthography. On this assumption one may confidently adopt from his MSS those forms which the researches of Philipp Wagner, Lachmann and others have shown to be correct : see list (i) below. But what of those cases where the dissensions of the MSS reflect a real variation in ancient usage ? I think for example of the fluctuation between ex- and exs-, inl- and ill-, con- and com-, and so on. In such cases one cannot always be certain which forms Claudian favoured, nor indeed whether he was any more consistent in his practice than Virgil would seem to have been, and the most sensible policy is to collate the evidence of Claudian' s MSS and the ancient witnesses, especially the Virgil MSS, and to pick and choose as they seem to recommend, without expecting to achieve completely con­ sistent results. Thus, consideration of the available data suggests that Claudian preferred to drop s after the prepositional prefix ex- ; I therefore print without s. A prononnced tendency towards non-assimilation of in­ and a less prononnced though still distinct tendency towards non-assimila­ tion of ad- and con- is discernible ; I therefore accept non-assimilated forms where they are attested uruess there is reason to believe that they are less good than assimilated forms. 1 I may say here that I have nowhere introduced forms for which support is not forthcoming from Claudian's MSS.

' I think for example of inbuerejimbuere and adspicerefaspicere : c£ Ribbeck, Prolcgolllena, pp. 43 I and 400.

C OMMENTARY

A summary list o f the various forms I have adopted may now b e given , as follows : 1 (i) M iscellaneous :2 bracchia not brachia ; conubium not connubium ; hiemps not hiems; inclutus not inclytus or inclitus ; incohare not inchoare ; neclectus not neglectus ; nequiquam not nequicquam or nequidquam ; pinna not penna ; querella not querela ; quidquid not quicquid; religio (-osus) not relligio (-osus) ; summisere not submisere ; tum invariably before consonants ; vaesanus not vesanus. (ii) In- not assimilated : inlabi, inlapsus, inlecebra, inlidere, inlucescere, inludere, inmanis, inmemor, inmensus, inmeritus, inmitis, inmotus, inpatiens, inpedire, inpellere, inperfectus, inpexus, inpius, inplacidus, * inplere, inplorare, inportunus, inpositus, inprobus, inpune, inritus, inrorare, inrumpere. In- assimilated : illuminare,* imbuere, immunis. (iii) Ad- not assimilated : adcelerare, adjatus (subst.), a4firmare,* adjlatus, adgredi, adgressus, adpetere, adsibilare,* adsiduus, adsistere, adstrictus, adsum, adsurgere, adtritus. Ad- assimilated : accendere, accingere, acciri, accommodare, accommodus, alfixus, agnoscere, allaturus,* alligare, apparere, ascendere, ascribere,* aspectus (subst.), aspernari, aspicere, attemptare, * attollere, attonitus. (iv) Con- not assimilated : conpages, conpellare, conpescere, conplecti, conplexus (subst.), conpressus, conprimere. Con- assimilated: collectus, colligo, colloquium, commemorare, commendare, commercium, commissum (sub st. ), committere, communis, corrumpere. (v) S omitted after ex- : exanguis, exertus, exilire, expectare, expectatus, expirare, extare, extinctus, extructus, exultare, exuperare. Like Birt I always adopt the commoner -es endings of third declension words, though endings in -is are quite often found in the oldest MSS. I hope I need not apologise for writing consonantal u as v ; old habits die hard, and even Lachmann, Ribbeck and Munro, experts all in matters of orthography, were content to retain Ramus' v. 4 Congesta is read by most editors, but there is no general agreement as to how it should be understood, and not one of the various interpretations of it so far proposed can be said to carry conviction. Most commonly, and naturally, it is taken as nom. sing. with mens, but what does mens congesta mean ? The MS glosses coadunata (R 1 and T) and in se agregata ( 0 6) are far 1 All but seven of these words are found also in the old Virgil MSS ; the seven I mark with an asterisk. (atijirmo and allaturus, it may be noted, are confirmed by the Lucretius MSS.) Only one form, conpescere, conflicts with the evidence of the Virgil MSS, which have compescere; but conpescere is accepted by Housman in his Lucan. 2 This miscellany does not pretend to be exhaustive but does, I hope, include all those spellings which are sometimes erroneously regarded as discretionary.

B O OK I

from clear, and the fuller explanations put forward by Parrhasius (' intelli­ gatur retrorsus avocatus animus ad cogitandum '), Claverius (' Congesta mens vere est avvTJ6potcrj..u\v1l l.fi VXTJ ' ) 1 and J. G. Walch (' quasi cum se ipsa agens & aliarum rerum oblita ' ) , 2 not only seem scarcely justified by the Latin but also introduce an idea which, as Gesner observes, is out of place in this context : ' non profecto a corpore in se ipsam hie revocat animam noster : potius res a sensu remotas velut oculis admovet '. Gesner's own explanation, however, that the phrase

=

the Homeric

rrVKtvcxi q>p€ves,

is

hardly more plausible, for it is not by virtue of any native wisdom or cunning, such as the epithet 1TVKtv6s suggests, that the poet is enabled to sing of Inferni raptoris equos . . . but because he is inspired by Phoebus (vv. s-6 ) . Platnauer in the Loeb edition translates ' My full heart bids me . . . ', but this is very vague and I can find no evidence that isolated

congestus

mean ' full '. An entirely different line is followed by Birt, who takes

could ever

congesta

as ace.

neut. pl. and understands it as synonymous with breviter, comparing Capitol.

M. Ant. phil. 19.12 et quidem haec breviter et congeste and Anth. Lat. Riese 672.21 (not 18, as Birt states) congestos . . . labores. But whatever we may make of these supposed parallels (and congestos . . . /abores means anything but breves . . . labores) , the notion of brevity is surely absurd in view of the fact that, although incomplete, the epic already extends to some 1,170 lines. Nor, it may be added, is there much to be said for taking congesta as neut. pl., as Birt had advocated, but with the meaning ' compiled ', ' brought together ', with reference to the structure of the poem, (a) because the apposition of congesta to the synopsis of the story given in vv. 1-3 is weak and indeed uunecessary, and (b) because the verb congerere is elsewhere used of the mere stringing together of items or accumulation of examples without implying any artistic arrangement. In any case, the position of the participle in the sentence suggests strongly that it agrees with mens. If then, as I believe, congesta is corrupt, what word should replace it in the text ? The context, and especially vv. s-6, suggests that the reference here is to a mind touched by the supernatural afflatus of the god of poetry. If this is right, three variants seem

prima facie possible : augusta J 3 ; commota,

which

Dempster says he found in a Vatican MS but which I have looked for in vain ; and concussa, offered by the Isengrin edition. Commota and concussa are satisfactorily paralleled in the context of divinely induced ecstasy : for the former c£ Pacuv.

trag. (Ribbeck 3 rd ed. ) 423, and Sen. Ocd. 230-1 incipit Letoa vates spargere horrentes comas I et pati commota Phoeburn ; and for the latter cf. Sen. H.f 105-6 concutite pectus (sc. Herculis), acrior mentem excoquat I 1

In support he cites (inaccurately) Plato, Phaedo 6 5 c (T] l.fiVXTJ) ;\oyi3ETCXI . . . Kcx6' CXIiTf}v ylyVTJTCXI.

KaAAIO"TCX, (hcxv . . . cht �aAIO"TCX cx\JTf} z In his Leipzig edition of 171 5 .

I9I

C OMMENTARY

quam qui caminis ignis Aetnaeis Jurit. Augusta, however, for which c£ Ovid, Met. 15. 143 ff. (Pythagoras speaking) et quoniam deus ora movet, sequar ora moventem I rite deum Delphosque meos ipsumque recludam I aethera et augustae reserabo orawla mentis and Incerti carm. de lona (CSEL xxiii, p. 221 ) 1 3 (deus) praesagos agitans augusta mente prophetas, must refer to the mind of deity, but since there is no mention of any deity until we reach v. 6, it would surely prove obscure and troublesome here. Augusta being thus problematical, I opt for the Isengrin's concussa as having better credentials than Dempster's com mota. 6 For totum Jeep conjectured solum (for the confusion of these two words see Housman on Manil. 1.779). Emendation here is, however, unnecessary. Totum should be regarded either as transferred from praecordia, or, perhaps more likely, as meaning ' in full strength', ' in all his might', for which sense c£ Hor. Carm . 1.19.9 in me tota ruens Vm11s and Stat. Theb. 10.927-8 talia dicentem toto love fulmen adactum I corripuit. 8-9 The whole picture is shadowy and indistinct, gaining thereby in grandeur and solemnity, but at the same time making for some difficulty in deciding between lt�mina, /imina and wlmina in v. 8 and in determining who is meant by deus in v. 9. Lumina, if right, must taedae, as Parrhasius suggests (it cannot Phoebi radii, for the arrival of the procession from Athens and the mysteries themselves took place at night), and the scene thus vaguely depicted may be the arrival of Iacchus at the head of the colunm of initiates (c£ Foucart, Les Mysteres d'Eleusis, pp. 33 8--9), or one of the ceremonies in the telesterion. Reading lim ina, one may suppose that the picture is of the initiates standing in pitch darkness outside the telesterion and of the sudden flash of light as the doors are thrown open to reveal the torch-lit interior of the temple. If culmina on the other hand be preferred, the scene may be as imagined by Lenormant (Contemporary Review xxxviii (1 88o), 421 ), with the votaries outside the temple discerning the glimmer of the lighted interior through the aperture of the cupola roof (one might also think, rather less prosaically, of a sudden flash of divine light announcing the presence of deity), or perhaps it may be set within the telesterion, where the torches' light reflects off the ceiling as the ritual acts proceed. For the confusion in MSS of lumen-limen c£ Virg. Aen. 6.25 5 ; 828 ; Sil. 1.66 ; 15.563 ; al. It is possible that deus refers to Iacchus, for whom c£ Eurip. Ion 1074ff. and Aristoph. Ran. 324ff. Perhaps more likely, however, in view oflacchus' appearance at v. 16, it refers to Ceres (so Parrhasius and Heinsius). If Ceres in fact is meant, there is no need to read deae (found in Z and advocated by Heinsius ap. Burman, p. 937) : for the use of deus on the analogy of ij &e6s where there can be no question of allusion to a masculine divinity, see Langen on Val. Fl. 6.657. =

=

1 92

B O OK

I

For the very rare use of cernor with a prolative infinitive, c£ Avien. Arat. 16oo-1 si (sol) . . . radiorum luce carere / cernatur; also ibid. 165. II eleusin F 3/8/1 3 : eleusim N: eleusis the other MSS. On Greek nouns ending in -tv in the nom. sing. Priscian observes (Inst. gramm. 6.5.28 = G.L.K. 11.219.7) : sciendum quod omnia in '-i11 ' desinentia etiam in ' -is' pro­ ductamjiniuntur: ' Arin ' et ' Aris ', ' delphin ' et ' de/phis', ' Eleusin ' et ' Eleusis' Et notandum quod auctores Graecorum in ' -is ' magis terminatione sunt usi, nostri vero Jrequentius in '-in '. The forms Eleusin and Eleusis are both mentioned by Servius (on Virg. Georg. 1 .162), but in MSS of the Latin authors only the form in -in seems to have wide currency. Compare Ovid, Met. 7-439 (-is M N s') ; Her. 4.67 (G w) ; Fast. 4.507 (-is D B s' ) ; Stat. Theb. 2.3 82 ; 7.4I I ; 12. 1 3 2 (-is C) ; 627 ; Plin. N. H. 4.7(I I ).23 ; Sen. N. Q. 7.30.6 (-is J2M) ; Phaedra 8 3 8 (-is A) ; H.f 302 (-is A) ; H.O. 599 (-is A) ; Mela 2.3 .41 ; Firm. Mat. De error. 7·5· (The ace. E/eusin at Sidon. Carm. 9. 1 8 1 implies a nom. in -is; but the metre is hendecasyllabic, and no other ace. form would fit. ) In . • .

the present passage therefore I follow Heinsius, Gesner and Burman in reading Elettsin. 12 Priscian (Inst. gramm. 10.29 = G.L.K. 11.521) recognises both the second and the third conjugation forms of stridere, but the Virgil MSS (at

Aen. 4.689, 8.420, a/. )

mainly have third conjugation forms. I therefore

print stridunt here and at III - 399· Birt inconsistently adopts second conjuga­ tion forms in the D.R.P. and at 1 . 101, but third at 20.106.

16 For levis = imberbis cf. Hor. Carm. 4.6.28 levis Agyieu. Of the other variants available, Ienis is contextually weak, but laetus not inappropriate in view of ebria .

. . vestigia (v. 19).

The phrase firmat vestigia occurs at Virg. Aen. 3.659, Luc. 4.41 and Stat. Theb. 2. I I , and may well be right here ; so may Julcit vestigia, which

19

finds

support in Sil.

6. 79 lapsantis Jultum truncata cuspide gressus.

With both

verbs,Jirmat andfulcit, the subject may be either Iacchus, if Maeoniis (-io) . . . thyrsis (-o) be read, or Maeonius . . . thyrsus, which I adopt. Much more widely attested, though less satisfactory perhaps, is figit, which can be combined with Maeoniis (-io) . . . thyrsis (-o) to produce a passable line with the meaning ' he plants his steps with the aid of his Maeonian thyrsus ' : for Virg. Aen. 6.159 and Val. Fl. 4.394-5 ; also Lucr. 3 ·4·

figit vestigia c£

20-31 The poet invokes the nether deities and outlines the subject-matter of his composition. 21 The question here is : can opes reasonably be called avarae ? Jeep and Baehrens thought not, but the former's Joribus, though palaeographically easy (if that indeed be any commendation), produces the bizarre association

Joribus donatur, and the latter's oris seems little better.

The phrase

opes avarae

is undoubtedly bold, but it is no less defensible than opes avidae at Culex 8 r . O n both occasions one may suppose either that the epithet i s transferred 13

1 93

HCD

C OMMENTARY

from the person who seeks after wealth (Barth compares Sen.

H.f 782 avari

to the object of his search, or that to the poetical imagination the phenomenon of increasing riches suggested the notion of wealth greedy for

Ditis)

more wealth. Thus, Walch interprets avaris as meaning ' quum omnia ad se trahant ', and Gesner as ' quibus nihil decedit, sed accedit perpetuo '. 27 Claverius notes that his ' vetus liber ' read ulsit, for which he offers the easy correction ussit. Q!a lampade Ditem I ussit is perfectly good (c£ Stat. Silv. 1.2.79-80

ast illam summa leviter I lampade parcentes . . . strinximus 3 . 8.5-6 illius ex oculis, cum vult exurere divos, I accendit geminas lampadas acer Amor, both passages cited by Claverius ) , but there is no good reason for deserting the MS jlexit. For jlectere in the sense of ' to overcome ' compare D.R.P. II. I2-I 3 it Venus . . . I iam durumjlexura Chaos and Claud. c.m. 30.169-70 Hippomenes . . . I aurato volucrem jlexit Schoeneida porno. ' ferox : implacabilis quae vitae necisque ius habet & morientium precibus . . •

and [Tib.]

locum dat '-Parrhasius. 27fE For a similar confusion of moods in indirect question c£ Claud.

c.m. 29.2ff. where the following sequence of verbs is found : laborat . . . iubet .jluant . . . concutiat . . . ducat . . . tonent . . .jloreat. The phenomenon is dis­ cussed by Jeep, ed. maior dxviii ; also cf. K. S. II-494· 28 For quantas = quot c£ n.308 infra and Leumann-Hofmann-Szantyr, II. 207. 30 See Introduction, p. 84. At Virg. Aen. 4.5 8 legiferae (Cereri) is corrupted to Jrugiferae in some MSS. • .

32-47

Pluto, resenting his lack of a wife and children, prepares for war

against the upper world. 35-6 The combination of inlecebra, -ae with an animate objective geni­ tive is comparatively rare : ThLL, s.v. illecebra, vn.i.365, cites only Cic.

Catil. 2.4. 8 (iuventutis), Auson. xvi.26.5 Schenkl (viventum), and Salv. Cub. 1.43 (hominum). More commonly inlecebra, -ae is followed by a subj ective genitive, from which it is a reasonable inference that Claverius' maritae is a normalisation of the MS mariti. Claverius' other contribution at this point is nudasque for the MS nullasque, and this, whether it was intended to remove the pleonasm or merely to enliven the narrative, is certainly not what Claudian wrote. 40 The diphthong e1 in Greek proper names may in Latin be represented either by e or by i : c£ Servius' note on the name Thalia at Virg. Eel. 6.2 : latine ' Thalea ' debuit dicere, sicut Kv6epE1Cx ' Cytherea '; sed propter euphoniam contempsit ius regulae et ideo in graedtate permansit. Greek iota of course

remains unchanged in transliteration. In seeking to determine, therefore, whether Tesiphone, offered in the present passage by most of the MSS and printed by Birt and Koch, is a permissible form, we must first discover what is the regular spelling of the name in Greek.

1 94

B O OK I

Roscher, Lexicon der griechischen u. romischen Mythologie, s.v. Teisiphone, states categorically that the Greek spelling is Te1- not T1-, but a conjecturally restored inscription from Anazarba in Cilicia, which is all he can adduce in support of his assertion, can hardly be considered a reliable authority in matters of orthography. LSJ, s. v . T1a16VT) should probably be read ', but this view is shared neither by the Teubner editor, Wagner, nor by Frazer in his Loeb edition, both of whom print T1a1q>6VT) at 3 ·7·7• as also at 1.1.4. Stephanus recognises only the form T1a1q>6VT). An examination of all proper names listed in Pape-Benscler which are compounds of ' Tis-' reveals that the forms in Tel­ are confmed to inscriptions, and are thus to be considered vulgar, not literary. In the manuscripts only the form T1a1q>6VT) seems to have won recognition : c£ Norm. Dionys. (ed. Keydell) 10.40 ; 12.21 8 ; 44.208 ; Orph. Argonaut. (ed.. Bude), 968 ; Schol. Hom. Il. 9-454 (Dindorf 1.324 ; III.403 } ; and Harpocration, s. v. Ev1Jev1Ses (Dindorf, p. 140.16 } . On this evidence we might expect to find that Tisiphone is the only form recognised in literary Latin ; and our expectations are very largely fulfilled. Tisiphone is printed without comment, and without any hint of a variant Tesiphone, in the following places : Iuv. 6.29 ; Virg. Aen. 6.555 ; 571 ; 10.761 (and Servius ad loc. ) ; Ov. Her. 2.1 1 7 (and Planudes ad loc.) ; Culex 21 8 ; Val. Fl. 2. 194 ; 3 .214; 4.394 ; 410; 6. 179 ; 403 ; Sil. 2.53 0 ; 614 ; 1 3 .575 ; Stat. Theb. 8.346 ; Lactant. Plac. on Stat. Theb. 1 .477 ; 8.347 ; Hyg. (ed. Rose), Fab. p. 1 ; Carm. Verg. (PLM IV.199 Baehrens) 201.3 ; Lucil. (ed. Marx) 1. 13, v. 169 ; Coripp. Ioh. 3 . I I I. Side by side with these cases where the MSS are unanimous, or appear to be unanimous (orthographica are often ignored by editors), are others where both forms, Tisiphone and Tesiphone, are found in the MSS but editors are united in preferring the former, viz. Prop. 3 .5.40 ; Ov. Met. 4-474 and 481 ; Hor. Serm. 1 . 8.34; Sen. H.f 984; H.O. 1012 ; Luc. 6.730; Tib. 1.3.69 ; Stat. Theb. 1.59; 2.283 ; 4.213 ; 486 ; 6.514; 7.467 ; 8.66 ; 75 8 ; 9. 150; u.58 ; 208 ; 483 ; Petron. 121.120. At Dracont. Med. 481 and Coripp. Ioh. 4.326 the codices unici have Tes-, but edd. restore Tis-. From all this I conclude (a ) that e1 in the Greek name is a vulgar aberration from the literary -1-, due perhaps to the influence of pronuncia­ tion on spelling ; (b) that, as in Greek, so in Latin, the correct form, uni­ versally adopted by the authors, is -i-; and (c) that where the name appears as Tesiphone, we are probably justified in thinking in terms of ancient rather than medieval corruption : medieval scribes were notoriously helpless at handling proper names, and it is hard to believe that so many of them should independently have hit on the form Tesiphone, had it not been transmitted from antiquity. It is, I think, very likely that this corrupt form of the name circulated in Claudian's time, but not at all likely that an author -

195

-

C OMMENTARY

whose orthography, like his use o f language, reflects that o f the earlier writers of epic, should have admitted it into his writings. I therefore read Tisiphone with D recc. and the pre-Birtian editors. 42 Reluctatis is not passive from the very rare transitive relucto, as Lewis and Short maintain (c£ Birt, p. ccxxii), but comes from the deponent reluctor and does duty for the metrically intractable reluctantibus (so Barth). This line and the next one may be translated as follows : ' almost had the elements shattered their concord and resumed battle with struggling creation ' ; in other words, ' almost had chaos come again '. 46 For aucto cf. Val. Fl. 4·552 vigor novus auxerat artus. Only fractionally less good is vasto, which Platnauer seems to have had in mind in rendering aucto . . . corpore as ' his huge form '. The other variants are less satisfactory : vincto and arto (c£ Stat. Theb. 3 . 3 56 artam compagibus urbem) are otiose beside positis . . . de corpore nodis; alto is inappropriate to a recumbent giant ; duro is weak, and sancto obviously inept. Heinsius' advocacy of ancto is uncon­ vincing for the past participle of ango seems not to exist outside the pages of the grammarians. Pectore (M2 and M 3 ) is easier here than corpore (the other MSS), but the latter subsumes the former, so to speak, and can of course be understood to imply, with aucto, the act ofexpanding the chest. See Housman's note on the confusion of the two nouns at Manil. 4.923 . 48-67 Intervention of the Fates : Lachesis' address to Pluto. 50 Barth cites Ov. Met. 2.621-2 neque enim caelestia tingi J ora licet lacrimis to prove that the reading.fietu is impossible here. But what of D.R.P. 1.268, where we read of Proserpine : praescia . . . subitis maduerunt fletibus ora ? Weeping is of course characteristic of the suppliant : cf. Stat. Theb. 3 .687; 10.589 ; 1 1.749 ; Silv. 1.2.67. Fletu therefore is no less admissible than vultu, for which c£ Claud. 21.212. 53 Although the adjective ferratis naturally implies, in Gesner's phrase, ' firmitatem et velut rigorem fatorum ', an idea which squares as well with pensis as withJusis, its basic and usual meaning ' made of iron ' suggests that Jeep and Birt are right in preferring fusis to pensis, the reading of the older editions. 54 Reinhardt (quoted by Jeep, ed. maior, vol. n.clv-clvi) inferred from prima that the other Fates also interceded with Pluto, and supposed that a passage containing a speech by Clotho had fallen out after v. 67 (see my note on that verse). Arguing against this erroneous hypothesis, F. Baehrens (QE.aestiones Claudianeae, Diss. Miinster 1 885, pp. 23 ff.) aptly compares the use of prima at m.220 infra, and further observes that (a) the proportions of this book would be spoiled if this quite lengthy speech by Lachesis were followed by another one from Clotho, and (b) while Lachesis appears quite often in Claudian's pages, Clotho does not appear at all (and Atropos but twice). 196

BOOK I 62 Compare Hieron. Epist. 126.1.2 : an cotidie a deo fiant et mittantur in corpora (animae ) . No less attractive here than mittuntur is mutantur, for which c£ Ov. Trist. 5.2 b (3 ) .73 hinc dum muter; Ex Pont. 1 . 1.79 inque locum Scythico vacuum mutabor ab arcu ; ibid. 4. 14.7 terra qua muter ab ista (so Merkel with � Heins. ; Ehwald-Levy read quo mittar) ; Cic. Balb. 1 3 . 3 1 ne quis invitus civitate mutetur ; Plaut. Amphitr. 1.i. 122 ( v. 274) neque se luna quoquam mutat. In ortus (F2 ante corr.) may perhaps be supported by Paul. Nol. 1 8.78 post corporeos obitus, but it is probably due to a slip of the pen. For the idea in general c£ Pindar ap. Plato Meno 81 B. 64-5 Proeliafratrum I . . . committe is meaningful in itself, being equivalent to Jratres committe proeliaturos (for committere = ' to match' c£ Mart. 8.43 .3 victores committe, Venus, and for proeliafratrum c£ Virg. Georg. 1.3 1 8 omnia ventorum concurrere proelia vidi), but in context is hardly more satisfactory than foedera Jratrum I . . . committe, which Heinsius favoured, since both =

conformations rather imply the pairing of Pluto's brothers in a contest from which he stands aloo£ Pace Parrhasius, proeliaJratrum I . . . converte makes no sense. The true text is clearlyJoederaJratrum I . . . converte, which reiterates the idea expressed in v. 63. Really close parallels to the phrase Joedera . . . converte, ' overthrow the agreement ', appear to be lacking, but the following passages cited by ThLL, s.v. convertere, IV. 867, provide adequate support : Cic. Place. 3 8 . 94 non putant satis conversam rempublicam nisi in eandem impiorum poenam optime meritos cives detruserint; Tac. Ann. 3 .27 abolitis vel conversis prioribus (scitis) ; compare also the use of conversio at Cic. Place. 37·94 quanta in conversione rerum ac perturbatione versemur; Sest. 46.99 motus con� versionesque reipublicae ; Ph il. 1 1 . 1 1 .27 in tanta conversione et perturbatione omnium rerum. Scaliger's convelle is in no way preferable to converte ; indeed, the range of the two verbs, when used metaphorically, is virtually the same. 67-121 Pluto relents ; his peremptory message to Jupiter ; Jupiter's decision. 67 Birt and Koch read vix ille pepercit with almost all the MSS, but, as Barth observes, there is no point either in the phrase quamvis indocilis jlecti (v. 69) or in the following simile unless the general sense of 67££ is ' Ut subitus tyranni motus ad bellum, ita & repentina ad meliora consilia admoniti reversio '. Barth therefore conjectured vix ill a; pepercit, and ilia is in fact found in the margins ofR 14 and R26. Ilia is most likely neut. pl. ace., with dixerat or some such word understood : c£ Claud. c.m. 25.99 vix haec Hymenaeus; at ilia f . . . ; Val. Fl. 5.253 vix ea . . . cum (both passages adduced by Jeep) ; Claud. D.R.P. m. 196 vix tamen haec, mentioned by F. Baehrens (Zoe. cit.), is not quite parallel since haec is prospective. For the pause at the fifth trochee c£ Claud. c.m. 25.99, quoted above.

1 97

C O MMENTARY

It i s clear that pepercit must, like erubuit, refer t o Pluto. But what does it mean ? ' Franciscus Paraphrastes ' (in the Delphin eel.) ' clesiit furere explicat. ego verbis, quae ira antea effuderat, subintelligo, et idem esse puto, quod conticuit ' (Burman). Since Pluto has not yet uttered a single word, Burman's own explanation does not seem appropriate, but Franciscus' suggestion is very plausible, irae or a similar noun being quite easily supplied after pepercit. There is absolutely no need for Reinhardt's alteration to vix ilia peregit and assumption of a lacuna after 67 (the lost passage to begin on the lines of: flebilibus Clotho cum Jata est talia dictis) . 69-73 In 72 the majority of the MSS read flare cupit, while other MS variants are flare iubet, dismmpit, disrupit and irrupit. Claverius offers bella wpit, which is accepted by Scaliger, Barth, Heinsius, Gesner and Burman. Of the MS readings, flare iubet is plainly nonsensical, while flare cupit is crude and amateurish ; the probability is that flare is a metrical stop-gap suggested byflamine in v. 73 · From F 14's irrupit, disqualified by its tense, one may easily recover irrumpit, which could be taken with pelagus= ' tries to break into/on to the sea ', except that one's impression is that pelagus, like silvas camposque, is dependent on rapturus : c£ the group silvas armenta virosque at Virg. Aen. 12.688. Claverius' bella cupit is elegant and might conceivably be true, but one cannot help suspecting that it derives by conjecture from Stat. Theb. 2. 130 or 328 (c£ bella cient at Theb. 1 1 . 1 1 6) . Against disrumpit, adopted by Birt from C 1 0 3 , two main objections may be levelled: (i) that the structural organisation of the passage ceu turbine rauco etc. suggests strongly that the adj. phrase Getica concretus grandine pinnas is entirely self-contained, and that pinnas is not likely to be at once ace. of respect after concretus and direct object, as Birt, printing ceu . . . disrumpit without any kind of stop, obviously felt to be the case when he commented : ad Disrumpit suppleas ' pennas concretas ' ; and (ii) that on the evidence of other passages disrumpit pinnas ought rather to mean ' cleaves, tears in pieces, his wings ' than 'prises his wings apart ' (c£ Plaut. Bacch. 441 extemplo puer paedagogo tabula dirrumpit caput; Sidon. Carm. 1 3.9 collaque flammigenae dirrumpens fumida Juris ; Avien. Orb. terr. 1 1 1 7-18 femoris sub imagine partus f dirrupisse Iovem penetralia). But a slight extension of the meaning of disrumpere is perfectly possible, and the errro J
198

B O OK

I

ceu turbine rauco cum gravis armatur Boreas, gladeque nivali hispidus et Getica concretus grandine pinnas, prorumpit " ' pelagus szlvas camposque sonoro erumpzt, jlamine rapturus. 74 The form Aeolos is found also in the Veronensis at c.m. 22.35. For the spelling of Greek proper names see my note on rr. 136 infra. 86 Resedit is paraphrased as quievit in the Delphin edition, but means rather ' subsided ' : c£ Ov. Met. 15.272jlumina prosiliunt aut exsiccata residunt. 92 Praecede (T) possibly interpolated from Stat. Theb. 1 .293. 9 8 Some editors read stringimus and interpret it a s though i t were spargimus. Stringimus does not, however, mean ' iaculamur ' (so Parrhasius) . . •

}

.

or ' hurl' (Platnauer in the Loeb edition), but rather ' unsheathe ' ; for the opposite notion cf. Ov. Trist. 2.179 parce, precor, Julmenque tuum, fora tela, reconde. The variant spargimus is a clear case of simplificatory interpolation, suggested perhaps by some such passage as Ov. Met. 1.253 iamque erat in totas

sparsurus fulmina te"as. I prefer aures to at�ras for two main reasons : (a) according to ThLL, the verb deludere does not once have natural phenomena as its object (as would be the case if auras were read here), but on at least three occasions is applied to the human mind or senses (at Virg. Aen. 6.344 animum ; ibid. 10.642 sensus ; and Ennod. Carm. 1 .9.16 visus) and so may aptly be used of aures ; and (b) on a common-sense view, the statement ' we mock the idle ears of men with our thunder ' seems rather more meaningful than ' we mock the empty winds with our thunder '. C£ Ov. Met. 1.55 humanas motura tonitrua mentes. For the phrase vacuas aures c£ Hor. Epist. 1. 16.26 ; Ov. Am. 3 . 1 .62 (v.l. auras) ; Met. 4.41 (v.l. auras) ; 12.56 (v.l. auras) ; Lucr. 1.50; �intil. 10.1.32. There is nothing to be said for the Isengrin's hominum, for tonitru, the word it dis­ places, is quite indispensable, and aures alone, even without the addition of hominum, is perfectly intelligible. 100 Jeeps and Baehrens regard dispendia as corrupt, while Birt interprets it in the unparalleled sense of dispensatio. Like Parrhasius, I take it in its usual sense of ' loss ' : the underworld fell to Pluto because he was the loser in the

sors. Goetz and Gustafsson were puzzled by tertia, but their conjectures are unconvincing, and we can ill dispense with the transmitted epithet, which is best regarded as transferred from sortis for metrical reasons : c£ Sen. H.f 609 tertiae sortis loca ; ibid. 833 tertiae regem spoliare sortis ; Ov. Fast. 4.584

nupta Iovis fratri tertia regna tenet; Trist. 2.53 per mare, per terras, per tertia numina iuro ; Claud. D.R.P. rr. 167-8 tertius heres / Saturni. Parrhasius para­ phrases tertia supremae . . . dispendia sortis as ' ultima damna tertiae sortis '. 1 99

C O MMENTARY

108 The word order in this line i s perhaps confirmed by Dracont. Romul. 8.86 natarum turba coronat. IIJ Either dictis or dicto is possible. Birt' s reason for preferring the latter is quite ludicrous : 'nam unum iussum Iuppiter [sic] dedit Mercurio idque per singularem fortius exprimitur ' (p. cliv, n. w ) . n6 The adjective Julgidus, offered by Claverius, is extremely rare in Latin verse : ThLL cites only Q. Cic. Camt. Jrag. 15, Prud. Cath. 1.9, Cypr. Gall. Exod. 1062 and Hil. Gen. 67, apart from two appearances in Claudian, at 7. 1 1 and 14.26. In this instance it is almost certainly the product of conjecture. 122-78 Proserpine is concealed in Sicily ; description of the island and Mt Etna, with digression on the causes of volcanoes. 122 Aetnaeae or Hennaeae ? Heinsius makes out a case for preferring the latter reading, which is found in C 1 0 1 0 2 (R4), but although the association of Ceres with Henna is well attested elsewhere (c£ esp. Cic. Verr. 2.4.48, § 107f. ), a series of passages in the D.R.P. seem to suggest that in Claudian' s mind she was linked with Etna and therefore entitled to the epithet Aetnaea,1 just as at Stat. Silv. 5.3 .277 her daughter is called Aetnaea Iuno. After stating (1.141) that Ceres hid her daughter in Sicily, the poet turns aside to give a detailed description of the island as a whole, and its most prominent feature, Mt Etna, in particular. The section on Etna closes at 178 and is followed immediately by the words hie ubi servandum mater .fidissima pignus f abdidit. It is of course possible that hie refers vaguely to Sicily in general ; but the more natural interpretation is to take it as referring to the mountain. From this it seems to follow that Ceres' house is to be located on or near Etna. To the same conclusion point 1. 190 iam linquitur Aetna and, more strongly still, III. I 86, where Ceres, now returned home from Phrygia, talks of vicina . . . Aetna. Further confirmation is provided by r.237ff. and III . I I7, where Ceres' house is said to have been built by the Cyclopes who lived and worked beneath Etna. When finally it is observed that there is not a single certain mention of Henna in the whole of book I but that whole sections deal with Etna, it must be conceded that it is far more likely that at 1.122 Ceres should be described as Aetnaeae than as Hennaeae. The same choice between forms of Aetn- and Henn- has also to be made in four passages (rr.72, 289, m. 85, 220 ) depicting or alluding to the scene of the Rape. Once again a substantial majority of the MSS support Aetn-, and only 0 1 is consistent in presenting Hetm-. Among recent editors Koch accepts Hem1- on all four occasions, while Birt, after printing Aetn- twice 1 One may observe here that the confusion of Hennaeus and Aetnaeus is extremely common : c£ Luc. 6.293 ; 740 ; Sil. 1 .93 ; 214; I 3 .43o-- 1 ; 14.245 ; Colum. 10.270 ; Stat. Theb. 4· 124.

200

BOOK I and Henn- twice, subsequently decided in favour of Aetn- throughout (see his pre£ pp. cliv and clvi n. 12 ). Evidence for an Etna tradition of the Rape is provided by a number of authors : Hyg. Fab. 146 Rose, p. 106 ; Lact. Plac. on Stat. Theb. 5.347 ; Auson. p. 247, vv. 49 ff. Peiper ; Moschus 3 . 1 19 ff. ; Opp. Hal. 3 .488-9 ; Plut. Aet. phys. 917F; Schol. Pind. Nem. 1.20 ; Lydus, De mensibus, p . 162.19 Wuensch. C£ also Val. Fl. 5·344ff. and Stat. Aeh. 1.824-6. Was this the tradition followed by Claudian, rather than the better known one, represented by Ovid, according to which Proserpine was carried away from Henna ? Two passages of the D.R.P. suggest to me that it was. At III.43 8 ff. we read that Ceres, after lighting her torches from Etna's volcanic fires, prima gressus molitur ab Aetna I exitiique reos jlores ipsHmqlle rapinae I detestata locum sequitur dispersa viarum I indicia . . . , the clear implication here being that the Rape took place on or near Etna (so Parrhasius understands the passage : ' inde suspicari licet ad Aetnam montem rem gestam. nam Ceres adhuc erat in Aetna : quom rapinae locum detestabatur ' ) ; and at n.203-4 the laconic utterance rapitur Proserpina eurru is closely preceded by the words infeetae spumis vitiantur harenae, where harenae seems to mean ' lava­ dust ', as it does at n1.385 (c£ also Sen. Nat. quaest. 2.30.1 Aetna . . . ingentem vim harenae urentis effudit), and thereby to point to a location on or near the mountain. In the light of this evidence and of the poet's abiding preoccupation with Etna, I am inclined to regard the forms of Henn- which 0 1 constantly and other MSS occasionally offer as nothing more than manifestations of an attempt to bring Claudian' s version of the myth into line with that of Ovid in his Fasti and Metamorphoses. Apparent difficulties in the D.R.P. are easily explained : that Etna should be called parensjlorum at n.72 may seem strange, but c£ Plut. 917F, Anson. Zoe. cit. (jlorieoma . . . in Aetna), Hyg. Zoe. cit. (.flores legentem in monte Aetna) and Lact. Plac. Zoe. cit. (Proserpinam circa eawmen Aetnaejlores legentem) ; the herbosus vertex (n. 72) is obviously not the main summit of the mountain (c£ 1 .16o ff. ) but the crest of one of the smaller hills in the vicinity ; and the fact that Lake Pergus is said to be not far distant from the vertex (n. n2) is due no doubt to the influence of the Ovidian description of Henna (at Met. 5.3 85-6 ) , if it is not simply an instance of poetic indifference to geographical exactness. 127-8 Vitulum . . . qHi hardly less good than vitulam . . . quae : c£ Ov.

Fast. 4·459· 130 ' Matura toro mihi magis placeret non modo propter Virgilianam imitationem Aen. 7· 53 Jam matura viro, jam plenis nubilis atmis : sed quia vicina aliud significat, nempe puellam, cui jam parantur nuptiae '-Gesner. Vicina, however, might equally well imply ' whose marriage is close at hand ' (not necessarily ' in process of preparation' ) . The poet is stating simply that

201

C OMMENTARY

Proserpine i s of nubile age and i s already being courted. This idea i s perhaps more exactly expressed by vicina, which is in addition a much less obvious epithet than matura. It seems to me highly probable that matura is the secondary reading, suggested either by the Virgil passage quoted by Gesner or by Stat. Ach. 1 .292 virginitas matura toris ; c£ also Claud. 21.69 and c.m.

25.125-6. 138 Ignara (R 1 3 ) probably interpolated from Virg. Aen. 4.508 ; c£

also

Val. Fl. 7. 192.

139£[ Verses 139, 140 and 141 each convey roughly the same sense and fulfil precisely the same function in that they severally complete the sentence begun in 1 3 7 ; syntactically, however, they are totally unrelated to one another. Scaliger and Barth accept 140 and 141 together and reject 1 39, but the asyndeton between the verses is harsh and the repetition of ideas un­ pleasant. 1 It seems likely therefore that only one of the verses is genuine ;z and the problem is accordingly to decide which one. Considerations ofintrinsic and contextual aptness are in themselves hardly conclusive. Verse 139, to which Heinsius,3 Gesner, Birt and Koch give their support, is unexceptionable in sense and phrasing. Were it known to be spurious, however, one can see where the materials for its construction came from : pignora from 179, the variant gaudia from 195, and commendo from 196. Verse 140 is less well adapted to 137/ 3 8 /42. Whereas 139 clarifies the phrase ingenio confisa loci by its explicit mention of Siculis . . . terris, 140 leaves us without any clear notion of where Ceres has hidden her daughter ; and this we must have if the digression on Sicily is to appear relevant. Secondly, the ending commisit alendam could have been borrowed either from Claud. 28.583 infantem genitor moriens commisit alendum or from Ov. Met. 1 3 .43 1 (referring to Polydorus ; cf. Virg. Aen. 3-49-51), and alendam is perhaps not altogether appropriate to a girl who is iam vicina toro.4 Finally, infidis is not happily stationed ncar ingenio confisa loci. Verse 141 is at once the most interesting and the most problematical of the three verses : the most interesting in that it introduces the new idea aethera deseruit; the most problematical in that the verb relegat has no expressed object. This ellipse of the object is indeed peculiar, but if one believes that medieval scribes were in general more concerned to remove existing 1 No one has suggested accepting only 139 and 141. In this case the repetition would be still more unpleasant. 2 In view of the MS evidence (for which see below ) I do not think the p ossibility of author's variants arises. 3 In his second ed. 139 is the verse printed, but the note argues rather for 141-without however objecting to 139. 4 C£ the use of alendum at Virg. Aen. 3 . 5 0 ; Stat. Theb. 1 . 5 80-1 ; Ach. 1 .65 1 .

202

BOOK I

anomalies than to create new ones, the absence of some such word as natam, which may without too much difficulty be supplied from the context, is a better guarantee of the authenticity of the verse than of its spuriousness. The tangled evidence of the MSS, while hard to interpret in detail, nevertheless tends unequivocally to confum the view that 141 alone is genuine and the others false. All the MSS that omit 142ff. also omit 141, and conversely, all that have 142 f£ also have 141.1 Therefore, 141 was estab­ lished in the tradition before the emergence of Class !3. The presence or absence of 139 and 140, on the other hand, docs not invariably coincide with the presence or absence of 141 ff., and it must therefore be considered very doubtful whether those verses formed part of the tradition in the pre­ lacuna stage ofits development. Indeed, when one further notes that 139 and 140 are not fixed but variable in their position (139 follows 140 in more than ten vett. and 145 in A 1 : 140 follows 141 in W ; and both verses sometimes appear only as marginal additions), it must seem very likely that they are spurious intruders concocted after the loss of 14 1 ff. to complete the sentence begun in 137 and subsequently diffused by contamination throughout a large part of the tradition. 142-3 Trinacria quondam I Italiae pars una Juit. For una pars c£ Claud. 3 . 3 1 7 ; Ov. Met. 2·.426 ; 5.577-8 ; 9.20 ; 13.51 ; 14.482 ; Fast. 2.r56 ; Trist. 4.10.34; 5 ·7·4· There is absolutely no call for Birt's iuncta, which found favour with Koch (Ov. Met. 1 5.29o-1, which Koch cites in support, is no parallel) ; nor is Gesner's conjecture ima really necessary. Sil. 14. 1 1 writes Ausoniae pars magna iacet Trinacria tel/us : c£ R 1 A 2 w. For pontus et aetas c£ in general Val. Fl. 2.616-18. 146 Prohibeo has the meaning dirimo, separo also at Ov. Met. 3.448 ff. where Narcissus, addressing his reflection, remarks : nee nos mare separat ingens I nee via nee montes nee clausis moenia portis : I exigua prohibemur aqua. In the light of this passage Heinsius' cognatis . . . terris ( ? assuming the ellipse of Trinacriam, vel sim.) is seen to be UWlccessary. 147 There is little to choose between rap tam and ruptam : either reading could have arisen out of the other, with or without the help of the doublet rupit-rapuit in 144. For raptam, which I print, c£ Claud. 7.26ff. poscere partem I de spoliis, Scythicos arcus aut rapta Gelonis I cingula ; Ov. Met. 3 .730 (frondes) alta rapit arbore ventus; and for ruptam, with abl. of separation, c£ Val. Fl. 1.588--9 Libya cum rumperet advena Calpen I Oceanus. Trisulcam, for which c£ Claud. 17.203-4 trifidam . . . I Sicaniam, is perhaps more likely to be secondary than trisulco because of the possibility of 1 Except the fifteenth-century Riccardianus (F9), a MS of no importance whatsoever. In this case one must think in terms of the conjectural deletion of what was felt to be superfluous.

203

C OMMENTARY

assimilated terminations, but whichever reading i s preferred, the meaning remains the same : tribus ex partibus. ISI Bracchia is offered by the Virgil MSS almost invariably : see Ribbeck, Prolegom. p. 391 ; also Lachmann on Lucr. 6.434. The Isengrin's moveri perhaps comes from Stat. Theb. 8.123. ISS The participle revinctus is frequently qualified by phrases such as ad saxa (Ov. Met. 1 1 .212 ) , de poste (ibid. 10. 3 79) ; occasionally, however, it is used without such qualification, as for example at Ov. Met. 5.22 scilicet haud satis est, quod te spectante revincta est; Ars am. 3 ·429 quid minus Andromedae Juerat sperare revinctae. There is thus nothing objectionable in sauda terga revinctus. The text of D J 4 K 1, saxo membra revinctus, though perfectly possible, is probably interpolated. IS6 For pectore, the reading of D ]4 K1, Heinsius cites Val. Fl. 2.25. Vulnere, however, is better, because less trite. IS7-8 The reading detractat, to which Heinsius strangely objected, is aptly applied to Enceladus' attempts to throw off the burden of Sicily that weighs him down. The verb is often used of aniinals trying to throw off the yoke (c£ Ov. Am. 1.2. 14 detractant prensi dum iuga prima, boves; Prop. 2.3 ·47 3 b.3 ac veluti primo taurus detractat aratra ; Virg. Georg. 3-57 aut iuga detractans; Avian. Fab. 28.9 sed postquam irato detractans vincula collo), and is readily transferred to the confined giant. It is not easy to be sure what in dextrum laevumque latus is dependent on : whether detractat, if that word be taken to mean ' conatur excutere ' (so Parrhasius), or rebelli, if understood in the sense of ' twisting defiantly ', or both detractat and rebelli. For the adjective rebellis followed by a dependent phrase c£ Stat. Theb. 1 .3 5 £ and Sil. 17.347ff. The variant rejlexa, which is not appropriate here, was presumably interpolated either from Virg. Aen. 8.633 or Ov. Ars am. 3·779 or Stat. Ach. 1 . 3 82. x6off. C£ in general Sil. 14. 5 5 ff. 163 The Isengrin's vomit, whether it b e survival or conJecture, i s patently superior to the MS movet, which can scarcely mean ' emittit ', as Parrhasius supposed, and may perhaps have been prompted by motibus in the following line. The two verbs are commonly confused in MSS : c£, for example, Sil. 12.123 and 720. 171-2 Q!!ae tanta cavernas j vis glomerat ? This passage is discussed by Th. Hertel (De nonnullis Claudiani locis, Torgau progr. 1 848 ) , W. H. Semple (C.R. 1x (1946), 6x ff.), and E. L. B. Meurig Davies (C.R. lxiv ( 1950 ) , 95 ) . Hertel, supported by Semple, wanted to read cavernae, with scopulos taken chro KotvoO with rotant and glomerat, and cavernae a genitive dependent on vis. There is absolutely no need for emendation, however. Glomerat cavernas does not mean ' arches the caverns ' or ' collects together the caverns ' (two renderings set up as skittles by Semple, and easily knocked =

204

BOOK I

down on grounds of sense), but ' hurls out in a mass the rocks that make up the caverns '. Cf. Virg. Aen. 3 ·575-7 (cited by Semple) interdum scopulos avulsaque viscera montis / erigit eructans liquefactaque saxa sub auras / cum gemitu glomerat fundoque exaestuat imo, and Luc. 6.294-5 (cited by Meurig Davies) cum tota cavernas / egerit et torrens in campos dejluit Aetna. In the light of these two passages the tortuous expression quae tanta cavernae / vis glomerat (sc. scopulos) may confidently be rejected. On these two and the following six verses Claverius cites as follows from J. C. Scaliger' s Poetice : ' Scaliger . . . ait : Poetam importune quaerere ex persona sua causam incendiorum, quid enim, inquit, nunc agat physicum ? Si quem narrantem introduxisset, pateremur'. Scaliger's criticism met with enthusiastic support from Jeep (Acta, pp. 3 58-9 ) , who deleted the whole passage as an interpolation on two grounds : first, that the enquiry into the causes of volcanoes has nothing to do with the main narrative ; and secondly, that 171-8 sever the resumptive adverb hie from the lines up to and including 170, to which it is supposed to hark back. No one would deny that 171-8 constitute an irrelevant digression, but irrelevance cannot in itself be regarded as proof of interpolation. If these verses contained features of metre or language that went against the conventions elsewhere observed by Claudian, there would be some ground for suspicion ; but in fact they do not. Nor is hie particularly awkward after the digression, as it seems to me, for it could not possibly refer to any location except Etna. 174 Ojfenso . . . meatu. I know of no parallel to this unusual phrase, which may mean something like ' buffeted as it goes ', or ' striking the passage (along which it travels) ' ; or, metaphorically, ' with violent/angry onrush'. 179-213 Ceres, having entrusted her daughter to Sicily's safe keeping, sets out to visit her mother Cybele in Phrygia. 189 Fruges, 'fruits of the field', seems too general a word for this context, and messes derives almost certainly from 200 infra. The Isengrin's culmi, on the other hand, is well suited to the middle place between aristis and seges (culmi and seges are closely associated at Claud. 7-I99-20o ) , the progression now being ' ears ', ' stalks ', ' standing corn'. Platnauer carelessly prints fruges and translates culmi (other cases where his text is at variance with his transla­ tion may be found at 1.23 3 , 243, 11. 148, 111.137 ) . 194-5 For salve . . . f . . . tibi . . c£ Virg. Georg. 2.173-4. 208 Either tumidas, which I print, or timidas seems acceptable here : ' Gargara bends its proud woods in obeisance ' (or indeed ' its high woods ' since Gargara was the highest part of lda : for this use of tumidrts cf. Ov. Am. 2.16 ( 17 ) . 5 1 at vos, qua veniet, tumidi subsidite montes and Drac. De laud. Dei 1.163 colles tumidi iuga celsa supinant; in M3 tumidas is glossed as altas) ; or ' Gargara inclines its woods in fear (at the wild shrieks of Cybele' s devotees) '. Tumidas seems to me less obvious, and therefore slightly more likely to be .

205

C O MMEN T A R Y

right. The two adjectives are confused a t Ov. Am . 3.6 ( 5 ) .79 and S tat. Ach. 1.15 5 ; and at D.R.P. m.367 F 2 R 25 U Z have timidis for the correct reading tumidis. 212. There is still considerable difference of opinion among editors as to the correct way to spell Cybele's name when the metre demands a word of bacchiac shape. The question is, simply whether Cybebe must always be written, as Bentley and Cortius maintained, or whether the penultimate -e­ in Cybele may be lengthened metri gratia. The passages in Latin poetry where this question arises are :1 Anon. Aegr. Perd. 29 (-ebe the fifteenth-century codex unicus) ; Cat. 63.9 ; 20 ; 3 5 ; 84; 91 ; Claud. 15.120 ; 1 8.277 ; 20.280 (-ebe the Exc. Florent. : -ele, -elle the MSS) ; c.m. 30. 1 8 (-eben E: -elen, -ellen the other MS S ) ; D.R.P. 1.212 ; m.1 1 3 ; 271 ; Claudii Carm. de luna 732.9 Riese ; Colum. 10.220 ; Luc. 1 6oo (-eben Z M : -elen O) ; Maec. Carm.Jrag. 4 (-ebe MSS Diomed. G.L.K. 5 1 4 : -ele MSS Caes. Bass. G.L.K. 262 ) ; Opt. Por£ 27. 9 ; Phaedr. 3 - 17 ·4 (-ebe) ; 4. 1.4 (-ebe) ; Prop. 3 .17.35 (-ele, -elle 0 : -ebe !tali) ; 3.22.3 (-ele, -elle 0 : -ebe ltali) ; 4.7.61 ; 4. 1 1 .5 1 ; Sen. Tro. 72 (-ebe Th : -elae E : -ellae A) ; Sidon. Carm. 7.3 1 (-ebes P : -eles C T F : -ele M) ; Sil. 8.363 ; 9.293 ; 17.8 (-eben F : -elen 0 V : -eles L) ; S tat. Silv. p.223 (-ebe) ; Virg. Aen. 10.220 (-ebe M P : -ele recc. aliquot). In more than half of the thirty-one passages listed above the MSS show no knowledge of the form Cybebe, but offer only Cybele or Cybelle, the latter undoubtedly concocted to correct the supposedly false prosody ofthe former. Moreover, in only four passages, at most, is the form in -ele apparently un known. Clearly, therefore, there is a primafacie case to be made out for Cybele. The judgement of Servius (on Virg. Aen. 10.220 ) , that ' Cybebe ' . . . bacchius est; nam ' Cybele ' anapaestus est, has been taken in two ways. Rothstein (Prop. ed. I, vol. 2 ( 1 898 ) , 368 ) and Friedrich (in his comm. on Catullus) dismiss it summarily as a mere ' Schulregel '. Bentley, on the other hand, seemingly regards it as the whole truth, to judge from his comment on Luc. 1.600 : ' Puteaneus CYBEBEN : ut utique scribendum est, quoties media syllaba producitur. Ita Graeci Kvl3iJl31l ; numquam Kvl3ei\i\T'), vel Kvl3i]i\T'), sed Kvl3ei\T'). z One important error in this statement of B entley s tends to confirm the other view, that Servius is over-simplifying the issue. It is indeed true that the form Kvl3ei\i\T') does not occur, but at Nonnus , Dionys. 17.63 bothLudwich and Keydell print, from the MSS, the form Kvl3i]i\T') = Latin Cybele.J In the ••

.

­

'

1 This list is taken from ThLL Onomas ticon , s.v. Except where otherwise stated, all MSS have forms in -ele, -elle etc. z A similar pronouncement was made by Bentley's contemporary Cortius, also on this passage in Lucan. J Hertzberg (on Prop . 3 .22.3 ) alleged that the form Kvl3i]i\T') was found, but without citing evidence. Nonnus also has the fern. adj . Kvl311i\ls.

206

B O OK I

light of this admittedly isolated piece of evidence we are, I think, justified in accepting bacchiac Cybele as a legitimate alternative to Cybebe whenever it is offered by the MSS ; and although editors will doubtless continue to prefer Cybebe in cases where a choice has to be made between the two forms, all justification for restoring Cybebe conjecturally must be said to have gone. 214-45 Jupiter sends Venus to lure Proserpine out of doors. Accompanied by Pallas and Diana, Venus reaches Ceres' house of bronze. 232 The MS claruit (probably from claresco, for clareo is not found in poetry outside Ennius, Cicero and Iuvencus ), which carries the meaning ' was lit up ' (c£ m.445-6 infra clarescit Etruscum I litus, and Tac. Ann. 15.37; a t Claud. 10.185 and Val. Fl. 7.3 1 the sense o f clarescere i s rather ' to become visible ' ), is perhaps not quite so forceful as the Isengrin's canduit (' shone glittering white '), which Heinsius preferred, but it is by no means in­ appropriate here. Canduit might perhaps have arisen out of incanduit (v. 252 ) . 233 Astronomical reality is against the reading delabitur, for, as Heinsius observed, ' non . . . delabitur cometa, sed scintillans dilabi ac dispergi videtur. de stellis, quae delabi quandoque videntur, alia res est ' : cf. Sen. Nat. quaest. 7.19.fin. (cometae) dilabuntur nee diu durant . . . and 22 jtn . numquam cometes in imum usque demittitur neque appropinquat solo. But the simile has more relevance to the semita down which the goddesses descended (237 devenere locum) if de- rather than di- is read, and it will be conceded that although comets do not actually approach the earth, they may on occasion appear to do so. I do not think. there is any need to go as far as Gesner, who maintains that ' Cometen hie etiam aperte intelligit meteoron ', but it is certainly possible that Claudian had both the comet, harbinger of great disasters, and the meteor or shooting-star in mind at once, and that this picture is inten­ tionally composite. 238 Formata is perhaps the choicest reading here sinceJormare, while only rarely used ofbuilding (c£ Tib. 2.5.23-4 Romulus aeternae nondumformaverat urbis I moenia ; vv.ll. here, according to Lenz, are Jundaverat and .firmaverat), is commonly employed of engraving or casting in metal (cf. Claud. 1.98 ; 5·357 : 7.184; Ov. Fast. 1.239 ; Mart. 6.1 3 . 1 ; Stat. Silv. 3 . 3 .104) ; and Ceres' house was built of iron and steel. Also possible, but much more obvious, is Jabricata, for which c£ Sil. 1.444-5 Sidonia tecta IJemineaJabricata manu ; Virg . Aen . 9.144-5 moenia Troiae I Neptunifabricata manu ; Sen. Ag. 651-2 moenia divumJabricata manu, I diruta nostra (note that in all these passages, as here, the word manu is present). Less likely is .firmata, because.firmare normally refers to the defence or fortification of walls etc. already built. 240 Most of the older edd. of Claudian read Pyracmon both here and at 7.195, and Pyracmon is generally printed at Virg. Aen. 8.425 on the testi­ mony of Terentianus Maurus. The ancient Virgil MSS M P R, and Servius, 207

C OMMENTARY

however, all have Pyragmon, and this i s the form almost always found in Claudian's MSS. I follow the recent editors of Claudian in reading Pyragmon. 241 ff. Non talibus umquam I spiravere Notis animae. A feeble expression if animae means nothing more than venti, for in that case, as Burman observes, the sense would in effect be venti numquam talibus ventis spiravere. 1 Nor would animae = spiritus (of the Cyclopes) be appreciably less feeble. If the text is right, therefore, ani mae must, as is commonly assumed, imply the idea of ' bellows ' (note that the reading folies in fact appears as a gloss or variant in ay and thence as part of the text in M 1 ) . Virg. Aen. 8.403 quantum ignes animaeque valent is the only passage ever adduced in support of this inter­ pretation, and although animae there can readily = venti (so Nonius, p. 348 Lindsay), it is not hard to believe that Claudian might have been led by these words of Virgil to think that animae might represent folies. If change is necessary, and I do not think it is, animae should be ejected (it could have started life as a gloss on Notis) and folies placed in the text. In 243, recent editors print cervice, but with what meaning ? Sense and syntax both show that the word cannot refer to the Cyclopes, nor can it be taken with metallum in any meaningful sense, pace Barth. I can see no alter­ native but to acceptfornace from F 4 and the Isengrin edition. Construe now nee metallum lassaJornace incoctum tantoflumine maduit, and translate 'nor ever did metal, made molten in the weary furnace, flow in so vast a stream '. Lassa is here used proleptically to indicate the magnitude of the enterprise : c£ Stat. Silv. 1.1.4. The word incoquo seexns to be without parallel as a technical term of metallurgy (so ThLL, which provides but one parallel to the use of this verb in connexion with technological processes : Arnob. Adv. nat. 6.14 simulacra . . .Jornacibus itlcocta.figulinis) ; but since it is used of heating in order to bake, it is reasonable to allow that it might also be used of heating in order to melt, especially as the simple verb co quo is so used : c£ Plin. Nat. hist. 3 3 - 3 (I9).6o and Luc. 6.405. On maduit Gesner remarks : ' plane significat fluxit, ut ex adjecto nomine .fiumine apertum est, quod nomen pulchre tuetur Heinsius ex Aen. 8.445 .fiuit aes rivis, et aliunde '. Madeo is almost always used of liquids not actually in a state of flux, but c£ 11.3 1 5-16 below rivis I barba madet (Claud. c.m. 34.3-4, which Birt in his Index verborum, s.v. madeo, adduces as an instance of madere = 'fliissig sein', surely means nothing more than ' How did it come about that the miraculous crystal was both solid and liquid at the same time ? '). 1 One might add that, whereas such a phrase as animae Notorum could be justified on the analogy of ventorum animae in Accius, apud Non. p. 348 Lindsay (cf. also Arnob. Adv. nat. 1 .2 numquid suas animas expiraverunt venti ?), the converse, Noti animarum, which is implied inNotis animae, is as unparalleled, and indeed as unlikely, as animarum venti.

208

B O OK I

244 Vestit, the variant offered by W, may perhaps derive from Luc. 10. 1 19 ebur atria vestit. 246-75 Proserpine weaves a tapestry representing the whole of creation. 250 C£ E. R. Curtius, European Literature and the Latin Middle Ages, chapter 6 ' The Goddess Natura ', esp. p. 106, and note-p. 108. For iustis c£ Avien. Arat. 24 hie dispersa locis statuit primordia it�stis. 251 Discedere with dat. of direction towards seems to be unparalleled. The idiom is doubtless to be explained on the analogy of such verbs as ire and ruere which in poetry are quite often followed by such datives : cf. K.S. 1.320. 252 An examination of the cases where aer is attested by all MSS as against aether shows that Claudian uses the former word consistently of atmospheric air and the weather ; aether, on the other hand, while generally used of the upper air, does occasionally intrude into the territory of aer (c£ 26.3 79). In v. 253 we already have.flamma ( = the upper air), mare and terra ; what we still require is a word for the lower air. In a context such as this, the use of aether would surely give rise to misunderstanding, and we should therefore read aer. The two words are constantly confused in MSS : c£ Virg. Georg. 1.404 ; Sil. 13 .527 ; al. 253 'Egit jlamma polum est, flamma aetherea impulit choros astrorum, in motum egit (cf. Claud. 17. 102, the passage under discussion) . . . ' Uni­ cuique ex elementis attribuit proprietatem suam. Ut aeris proprium est To O!acpaves, in quo dies albet, aetherii ignis movere polos, maris fluere, terrae in ambiente aere pendere et vehi '-Gronovius, Observat. 2nd ed. Lugduni Bat. (1662), p. 6p. The discussion has moved on from consideration of the relative positions of the elements (vv. 251-2) to a summary outline of their properties. Egit is thus more apt here than legit. 254 Inest possibly suggested by Ov. Fast. 4.489 iam color t�nus inest rebus. For the use of in and the abl. instead of the simple abl. of material, cf. Claud. 1.98 electro Tiberis, pueriJormantur in auro ; Virg. Aen. 1 .64o-1 caelataqt�e in auro /fortiafacta patmm ; ibid. 6.32 bis conatus erat casus effmgere in auro ; Stat. Silv. 1.3 .5o-1 quidquid et argento primt�m vel in aere minori / lusit (manus). 256 ' Q!!omodo fila caelavcrint fluctus . . . ego prorsus non adsequor ', says Barth. But given that caelare may be used of embroidery (c£ Val. Fl. 5.6; Sil. 14.658), is there really any difficulty in the replacement of the usual personal subject by an impersonal one such as fila ? Alternatively, it might be possible to interpret the participle not as caelantia but as celantia ; the sense would then be ' the threads swell artfully, hiding at this very moment the fact that the waves are imitation' : in other words, such is the degree of realism that one might imagine the waves were real. This alternative inter­ pretation is at least as old as the thirteenth century, as appears from the following note in 0 2 P 2 : qt�amvisfila mentiantur }lt�ctus (non enim sunt veri I4

209

HCD

C O MMENTARY

jluctus), tamen celant illud mendacium quia praetendunt apparentiam verorum jluctuum . . . celantia mentitos, id est celantia mendacium. 259 On the spellings subtegmen and subtemen (the same variation occurs also at Claud. 8.601 ; Sil. 14.65 7 ; 15.43 3 ) , cf. Nettleship, Contributions to Latin Lexicography, pp. 590-1. 26o Inustus is supported by Luc. 9.852-3 zona rubens atque axis inustus f solis equis ; adustus by Sen. H.f 236 adusta medius regna quae torret dies. Neither reading seems superior to the other, but ad- may have come from adsiduo in the next verse. The same confusion is found also at Prop. 3 . 1 1.40. 269 Margo may be either masc. or fern. (c£ Charis. 1.65 K. ) , but Claudian treats it as masc. at D.R.P. rr. 1 1 3 and 18.216. 272 Desero is found frequently in Claudian with the normal dependent accusative (cf. 1 . 129 ; 213 ; D.R.P. rn.421 ) ; desino he seems never to have coupled with the ace., only with the infinitive. ThLL knows only three cases in Latin poetry where desino governs an ace. of the object : Virg. Eel. 5-19 ; ibid. 8.61 ; Sil. 12.725. Ov. Ars am. 2.725-6, which would make a fourth, is discussed and dismissed by E. J. Kenney in C. Q. n.s. ix ( 1959 ) , 256. At Sen. Ep. mor. 107.10 deseramus and desinamus are confused in MSS. 273 Liquidas is interpreted as puras by Parrhasius, rightly. C£ Servius on Virg. Aen. 8.402 where liquido (electro) is understood as puro, and Sen. Phaedr. 648 puras . . .genas. 275 Jeep supposed (Acta, p. 356 ) that a number of verses describing Proserpine's meeting with her divine visitors had dropped out after 275. Such a narrative as Jeep felt to be lacking here is in fact given in Electra's speech at rn.202 ff. infra, and it may thus be the case that Claudian never intended to go into detail at this point about the goddesses' encounter with Proserpine. I Nevertheless, there is without doubt some awkwardness in the abrupt transition from Proserpine to Pluto and his preparations for the ascent, and what is more, this first book is comparatively much shorter than the other books. It is certainly possible to explain these phenomena by assuming a lacuna after this verse, but if we conclude that there is a real problem here and not one that arises simply out of our misunderstanding the poet's intentions, we may have recourse to the less drastic hypothesis that book I was never fully finished but at Claudian' s death was left in an un­ polished draft. 276-88 Pluto makes ready to ascend to the upper world. 278 Superas . . . ad auras at Stat. Theb. 1.295, superas . . . ad arces at Val. Fl. 4·73· I As Foerster observes, Der Raub u. die Riickkehr der Persephone, p. 92 n. 2, the metamorphosis of Cyane, which is described in book rn , is similarly omitted in the narrative of book rr.

210

B O OK I 283 Pro£ Goodyear's conjecture pigra for aegra is both easy and attractive (c£ Stat. Theb. I0.89 limen opaca quies et pigra oblivio servant, and Silv. 1.4.56-7 subrepsit in artus I insidiosa quics ct pigra oblivio vitae), but it cannot be said to be needed. Aegra oblivia is the sad and sickly condition of oblivion to which all are reduced who are tainted by the deadly exhalations of Pluto's team : c£ II .2o2 3 infra. 284 The phrase crudele micans, printed by editors on the evidence of practically all the MSS, involves assuming the ellipse of oculis, or luminibus, or some other word meaning ' eyes '. There is no such ellipse, however, at Claud. c.m. 25.41 dulce micant oculi ; ib. JO. I2I utraque luminibus timidum micat; Ov. Met. 1 .498-9 igne micantes I . . . oculos ; Paul. Nol. Carm. 33·53 torvum oculis micans ; Maxim. Eleg. 5· r6 grata micans oculis ; and Anth. Lat. 1.2 C. 94 1 .47 Riese blanda micans oculis, the only other poetical cases known to me where mica describes the gleam in the eye. It is very possible therefore that it was because they found the present ellipse harsh that the scribes of F 14 and R2o came to write crudele minans, for which cf. D.R.P. II1.340-1 adhuc crudele minantur I affixae truncis facies ; Ov. Met. 8.467 crudele minanti ; Stat. Theb. 7· 29 crudele minor ; ibid. 6.429-30 the horse Arion aurigae Juriale minatur I efferus. But minans could be a mere slip of the pen. At Luc. 1.320 micantes is corrupted to minantes. For the proper name Aethon we are indebted to the edition of Parrhasius (and the fifteenth-century MS n) ; n apart, the MSS have ethonus (or further corruptions of this false reading). In my view Aethon is a wrong conjecture, suggested perhaps by Claud. 8.561 where it is the name of one of Aurora's horses (cf. Virg. Aen. I I . 89 and Ov. Met. 2.1 5 3 , where it is given to horses belonging respectively to the hero Pallas and to Phoebus). What we require here is a name which, like Orphnaeus, Nycteus and A/astor, has clear associa­ tions with the nether world; and this requirement is excellently fulfilled by Jeep's emendation Cthonius. 288 Spectantes here = expectantes, which is in fact found, contra metrum, in J 3 and R 7 : cf. Claud. 24.86; Liv. 9.10.5. The variant clausula praemia palmae comes from Virg. Aen. 5.70 ; c£ also Stat. Ach. 1 .69 gaudia palmae P : praemia palmae E B Q KR. -

P R E F A C E T O B O O K II The poet compares his return t o song a t the instance o f Florentinus to Orpheus' recounting the exploits of Hercules after the hero had delivered Thrace from the tyranny ofDiomedes. On the relevance of this preface see the full discussion in the Introduction, p. 94· In R26 there is a marginal note to the effect that praeter rem hie ponitur haec epistula, and 0 2 P 2 have the following marginal comment (added either by the first or by a roughly contemporary hand) : istud totum ab hoc versu usque ' Otia sopitis' et cetera usque 211

C OMMENTARY

ilium locum ' Impulit Ionios ' et cetera vitiosum insertum est, quia non est de hoc opere sed praefatio cuiusdam alterius sed eiusdem auctoris. 2 A majority of the MSS here offer opus, while an inconsiderable minority, comprising F 2 (a.c. ) and R2o, have ebur ; at v. 16 infra the positions are reversed, with ebur as the majority reading against opus. In both cases Birt and Koch side with the majority, while Heinsius, followed by Gesner and Burman, follows the majority in 16 but the minority in 2, thus printing ebur twice and omitting opus altogether. In the present verse deposuisset opus, though rather vague, might be defended by such phrases as deponere negotium (Cic. Ad Att. 7·5·5) and deponere studium (Cic. Sext. Rose. 17.49 ; Mur. 21.45 ), but ebur in 16 is clumsy (see ad foe.). I am therefore inclined to favour the minority readings in both verses, on the assumption that ebur and opus, being metrically equivalent and tolerably appropriate in either verse, were intro­ duced each into the other's home by the process of glossing, with the result that the original arrangement came close to being eradicated. 5 Leonem is more likely to be right than leonum, (a) because the generic singular provides a better balance to vacca in the next line, and (b) because the association of metuens with dependent ace., for which c£ Hor. Carm. 1 . 1 . 16, Ov. Am. 1 . 1 5.37 and Iuv. 14.96, is considerably less common than that with dependent genitive. 9 For Inachiis . . . Argis c£ Virg. Aen . 7.286, Val. Fl. 1 .107, 3.666, Stat. Theb. 1.66o. Also possible is oris, for which c£ Stat. Theb. 8.3 19. At Ov. Am. 1 . 10.5 Burman restores Argis for agris, arvis of the MSS. 15 For modulatus . . . nervos c£ [Tib.] 3 .4.39 hane (sc. lyram) . p le etro modulatus ; Paul. No!. Carm. 27. 100 lyrieas Jacili modulatus pectine ehordas (further exx. in ThLL, s.v. modular, VIII. 1247.24fE). Moderor, on the other hand, seems normally to be used not of playing a musical instrument but of producing sound : it is associated, for example, with sonus, cantus, earmina, numeri. Hor. Carm. 1.24.14 moderere . . .fidem, however, which would seem to be an isolated exception to the normal usage, may be used to support the reading moderatus, offered here by R 5/7 and e eonieetura by Gronovius. 16 Ebur can certainly mean ' lyre ', as Bitt rightly notes in his Index verborum (s.v. ebur), but in the phrase duxit ebur it must mean the ivory plectrum used for plucking the strings. If ebur be read in this verse, therefore, the sense of the whole couplet is : ' and stroking the strings of his lyre with smooth plectrum, he plied the ivory plectrum . . . ', and the pentameter adds precisely nothing to the hexameter. I therefore believe that ebur has no place here but is an intruder from v. 2, and that opus is the true reading. Taking duxit to mean ' he drew out ', and opus in the sense of ' theme ', we restore point to both pentameter and couplet. For opus used of a poetical theme, c£ Claud. e.m. 30.12jemineae virtutis opus, and Ov. Ars am. 3 . 3 3 8, where the story of Aeneas and the foundation of Rome is described in the words quo .

212

.

PREFACE T O B O O K II

nullum Latio darius extat opus; also the spurious distich appended to the preface to book 1 of the D.R.P. : sic ego qui rudibus scripsi praeludia verbis I ingredior Stygii nobile Ditis opus. For duxit c£ Hor. Serm. 1 . 10.43-4 epos . . . I . . . Varius ducit; also Ov. Trist. 1 . 1 1 . 1 8, 3.14.3 1-2, 5.12.63 . As regards the choice o f epithets in the pentameter, we shall not hesitate, I think, in preferring nobile to mobile (there is nothing in common between Orpheus' ' noble' theme and the mobile opus of the stone on the counting­ board at Petron. So) ; and the context speaks clearly for Jestivo as against Jestino-unless one supposes that Orpheus' song was rendered allegro vivace throughout. IS Heinsius compares Ov. Ex Pont. 3.3 .26 et coit adstrictis barbarus Rister aquis and Rutil. Namat. 486 (Rister) grandiaque adstricto.flumine plaustra vehit. 39 Koch objected to the substantive ducis on the grounds (a) that the visually identical verb ducis appears in the previous line, and (b) that ' mirum est Geryonem ducem praedicari, quasi ad militiam aevi Constantinorum pertinuerit '. His first point would not, I think, have troubled an ancient reader, and that Geryon should be called dux may be ' mirum ', but is perfectly possible : cf. ThLL, s.v. dux, V. I . 2322.15, where the noun is used of heroic figures, and V. I . 2324.75, where it applies to herdsmen (at Aug. Civ. Dei 10. 1 7 it denotes a cowherd, as in the present passage). 42 At Sil. 3·39 the MSS vary between (aeripedis) cervi and cervae. 52 In orbe of circular motion is paralleled at Prop. 3· 14. 10, to which Shackleton Bailey, Propertiana, p. 304, adds Manil. 1.299 and Drac. De laud. Dei 1 . 1 71. Placidos . . . choros is aptly used of such a theme as the myth of Ceres and Proserpine, perhaps with an implied contrast between this work and previous ones of a martial character : c£ Prop. 4.6.70. Ducis ab ore sonos, read by many early editors, is probably an instance of conjectural rewriting.

B O O K II I-54 Proserpine i s enticed abroad. Description o f the four goddesses, Venus, Pallas, Diana and Proserpine. 2 Vibratur is supported by Stat. Theb. 6.579 vibraturque Jretis caeli stellantis imago (cf. also Sil. 14.566 tremula vibratur imagine pontus), and there is absolutely no need for the Raphelengian edition's conjectural vibravit (for which c£ Sil. 2.664 ; Luc. 5.446 ; Val. Fl. 2.342 ; 582-3 ). 11 Claverius' callida is of suspect credentials and in addition produces a phrase of dubious Latinity, for whereas Venus' tantutn votum is a unique event, callidus with the genitive, to judge from the examples given in ThLL, conveys the notion of being experienced in something. Of the MS readings, conscia is very weak in comparison with concita : ' conscious of her mighty wish' is much less satisfying than ' spurred on by her mighty wish '. 213

C OMMENTARY

23 In most M S S this verse appears in the form ima parte viget moriens et parte superstes, which is just tolerable from the point of view of sense, but stylistically frightful. Something was clearly wrong, and in 1 878 Gustafsson neatly put wrong to right with his conjecture ima viget, parte emoriens et parte superstes. Precisely the same solution was later reached by Koch, working from the testimony of C 1 which gives the nearly correct version ima viget parte moriens et parte superstes. The only unusual feature in the line as restored by Gustafsson and Koch is the appearance of the colloquial verb emorior, found nowhere else in Claudian and shunned by all the epic poets except Ovid, who has it twice (at Met. 3 ·391 and Rem. am. 654). But emoriens makes such a pronounced improvement to the sense and style of this line that it must be right. For elision after the second foot see the examples given by Birt, p. ccxviii, among which are Claud. 8.3 3 8 and 22.455. 2 4 All the main editions prior to Jeep read notferro, but gyro, which can be traced back as far as a handful of fifteenth-century MSS. What gyro could possibly mean here, however, is a complete mystery to me. Claverius holds that it = rotunditas and Barth that it = orbis, but why should a spear be ' terrible ' just because its stock is round ? Gesner offers a more complicated explanation : 'gyrus ille terribilis a quo nomen invenisse creditur (n<xAi\ctv enim vibrare est) potest cogitari quomodo instar silvae habeat, multiplicatur enim ab illo celeri circumactu hasta, et non unam arborem, sed silvam exhibet : ut pruna in tenebris circumacta circulum ignitum refert '. But (a) I fmd no evidence to suggest that gyrus (or for that matter gyrare) could be used of the brandishing of a spear, (b) the idea of Pallas whirling her spear round so as to create the illusion of a forest is quite fantastic, and (c) the phrase surgens per nubila seems to suggest, not that the spear spiralled upwards (a grotesque picture), but that it rose up vertically through the clouds. If gyro is, as I incline to think, a false reading, its origin must remain totally obscure-unless it be due to a reminiscence of the clausula nubila gyro at Stat. Theb. 1 . 3 1 1 . 25 A s Barth notes, silvae here means 'ingentis arboris '. Parrhasius interprets tantum as ' solummodo ' and Platnauer translates ' only the Gorgon's hissing neck she hid . . . '. Pallas' other accoutrements, her helmet and spear, are open to view, but the Medusa's head she conceals, in Parrhasius' words, 'ne quam comitum uerteret in lapidem '. 26 The Greek ace. Gorgona is retained in Latin hexameter poetry for metrical reasons (cf. Virg. Aen. 8.43 8 ; Stat. Theb. 1.544; Sil. 9.462 ; Luc. 6.746) ; likewise the nom. plural Gorgones (c£ Virg. Aen. 6.289). In the genitive singular, however, the Latin form is also metrically possible, and the practice of the poets seems consequently to have varied : the Latin form Gorgonis appears at Prop. 2.25 . 1 3 , Ov. Met. 4.699, 5.180, Trist. 4.7.12, Manil. 1.359 and Sil. 3.3 14, the Greek form Gorgonos at Luc. 9.653 and 668.

214

B O OK II

Both forms are presented by Claudian's M S S here and at 205 infra ; I prefer the Greek form. Inumbro occurs again at Claud. 1.69, its synonym obumbro at 10.49 and D.R.P. 1.203 (v.l.). The choice is open, though one may suspect that the prefix ob- intruded from obtentu earlier in the verse. C£ Virg. Aen. 1 1 .66 obtentu frondis inumbrant. 41-2 0 2 P 2, reading arti, paraphrase as follows : eventus arti contigit per ingenium pectinis ; similarly the Delphin edition : ' success us numquam accidit arti felicior ipsa peritia texendi '. Gesner took pectinis not with ingenio but with arti, but the word order is against this. The simplest solution is to take ingenio as dative balancing telae (cf. m. 1 5 5-6 infra where telas is parallel to pectinis artes}, and read artis, which may either be coupled with eventus = 'artistic result' or be dependent on felicior (c£ Virg. Georg. 1 .277 (dies . . . ) felicis operum ; Sil. I .395 fe lix heu nemorum ; 4.729 o felix Jamae . . . nomen}. 47 For anili Heinsius c£ Ov. Met. 10.406 and Sen. H.O. 1761. This reading is, however, almost certainly conjectural. 54 The dausula cornua gyro (W} comes from Stat. Ach. 1.3 14. 55-70 Proserpine's retinue of river nymphs. 57 Crinise. The spelling of this proper name is fully discussed by Heinsius, ad loc. At Virg. Aen. 5 . 3 8 M P R all have Criniso. The form Crinisse, offered .by some of the MSS, might be supported from Nep. 20.2.4. 59 Camerina is the spelling of the oldest Virgil MSS at Aen. 3. 701 ; Camarina is known only from ' aliquot Pieriani ' (so Ribbeck). At Sil. 14. 198, however, all MSS apparently have the latter form. 62 Ademptis is inapposite here, for the pelta was regularly carried by the Amazons (cf. Virg. Aen. 1.490-1 and Ov. Ex Pont. 3 . 1 .96), and cannot therefore be described as ' captured, taken away ', the more so as nothing is said as to the persons from whom it might have been taken away. The origin of the variant ademptis is, I suggest, to be sought in a reminiscence of the clausula exultat ademptis at Ov. Am. 1 .10.29. Aduncis would seem not to be used elsewhere of the Amazons' half-moon shields, termed lunatae at Virg. loc. cit. and ibid. 1 1 . 663 and at Sil. 2.76 ; but just as a bent bow may command both the verb lunare (at Ov. Am. 1 . 1 .23} and the epithet aduncus (at Ov. l-Ier. 4-39), so the adjective lunatus may naturally be replaced by aduncus in descriptions of the pelta. 64 Niveus is so commonly used to denote the fair complexion of women that one may suspect that it is employed here simply to emphasise that Hippolyte's turmae are composed of women. The conjectures viduas (Heinsius) and vacuas (Birt) are unnecessary. 66 The form Tanain is found at Val. Fl. 1 . 5 3 8, Tanaim at Virg. Georg. 4.5 1 7. As usual I opt for the Greek spelling. 69 Auro madidae. C£ in general Claud. 17.287 (sonipes) madido iubas • . •

215

C OMMENTARY

adspergitur auro (sc. with the waters o f the Tagus, like the Hermus-c£ Virg. Georg. 2.137-, an auriferous river) and Ov. Met. I I .I45 arva rigent auro madidis pal/entia glaebis.In the present verse J. J. Scaliger conjectured gravidis (on the basis of gravide, the reading of A2 and others), and in Ov. Zoe. cit. Magnus suggested gravidis for madidis. But the two appearances of the epithet madidus support one another against emendation. 71-100 Ema appeals to Zephyrus to refresh her meadows and enhance their colours. 74 Meatus is used of the movement of the wind at D.R.P. I.174, Lucr. 6.301 and Luc. 9.453, and is much better here than the obvious volatu. 75 Annum here = anni proventum : c£ Virg. Georg. 1.224 anni spem credere terrae ; Luc. 3 . 70 magnum Libye tulit . . . annum ; Claud. 15.57 pensabam Pharium Gaetulis messibus annum ; 1 8.403 australem Arctois pensasset frugibus annum. 79 Formulae of the type velim facias are frequently found, but the combination of velis with an appositional subjunctive in the third person is very rare, Sil. 1 . 109 haec tua sit laus, nate, velis being the only other example known to me. So The MSS vary between abnuat (favoured by Heinsius and Koch) and abneget (printed by Birt) ; the same pair of variants is present also at Claud. 5.72 and c.m. 30.14 (also with ace. and in£, as here). From ThLL it emerges that both verbs are only rarely associated with ace. and in£ in poetry : abnegare at Sidon. Carm. 22.87-8 alone ( te . . I abnegat esse deum) ; and abnucre at Lucr. 3 .641 aetemam sibi naturam abnuit esse, Sil. 5.27-8 cum minus abnuerit noctem desisse viator, quam coepisse diem, id. 14.173-4 haud equidem dignum memet . . . I abnucrim, and Claud. c.m. 30.14 in . . . suos migrare virum non abnuit annos. Clearly there is little to choose, but it is possible that the comparatively rarer word abncget owes its existence here to the conflation of an original abnuat with a gloss neget. 83 In all his works Claudian reveals the doctus poeta's passion for proper names taken from geography, as well as from history and mythology, and where the opportunity offers itself, such names are regularly set in balanced clauses and phrases. C£, for example, r.r6off., 3 .1 07-8, 5 . r u ff., 7.69 ff. , 2I. 109ff., D.R.P. 1.134ff., m.pr ff. Consequently, I feel sure that the present symmetrical series quidquid . . . Panchaia, quidquid . . . Hydaspes should be completed, not by quidquid . . . colonis, but by quidquid . . . Sabaeis. The long accepted reading colonis I take to be a gloss on Sabaeis that has got into the text : c£ 112 infra where F 3 reads coloni for the proper name Sicani. See further the next note. 84 A good idea of what Claudian was trying to say in this verse may be gained from c.m. 27 on the Phoenix. In that poem a constantly recurring theme is that of life coming out of death, and variations on it crop up time and again : c£ 25 Jecunda morte ; 26 altemam totidem per Junera vitam ; 43-4 .

2!6

B O OK II

tumulum . . . I . . . bustumque sibi partumque futurum ; 5 1-2 renasci I exitio proprioque soles pubescere leto ; 57-8 iam sponte aematur I ut redeat gaudetque morifestin us in ortu m ; 67 victuri cineres ; and 102-3 praebetur origo I per cinerem, moritur te non pereunte senectus. Similarly at 22.414, the opportunity for a very brief description of the Phoenix sends Claudian's mind back to the selfsame theme, and there emerges the clause ubi fecunda reparavit morte iuventam. The frequency with which Claudian exploits this antithesis of life and death in contexts dealing with the Phoenix leads me to suppose that in the present line the phrase repetens exordia, ' seeking a new beginning ', 1 is to be completed by a contrast with the desirability of death. If i am right in this, partu and saeclo,2 which add nothing to repetetls exordia, must seem unlikely to be right. Apart from Heinsius' attractive but unnecessary conjecture leto (accepted by Koch and Platnauer), two readings give the sense we require here : jato (offered by R 5 R7 and U) and busto (the Isengrin edition), for which c£ Claud. 22.419--o-20 procul ignea lucet I ales, odorati redolent cui cinnama busti. As in the case of v. 83 (Sabaeis) , so here my inclination is to prefer the Isengrin's reading. The fact that both Sabaeis and busto recall c.m. 27.43-4 tumulum texens pretiosa fronde Sabaeum I componit, bustumque sibi partumque futurum may perhaps at first sight seem to raise the possibility of learned interpolation in the Isengrin, but there is an undeniable tendency on Claudian's part to use the same expression where the theme is the same : fecunda morte, for example, is found both at c.m. 27.25 and at 22.414, and extremo . . . ab Euro occurs both at c.m. 3 1 . 1 5 and at 22.417. Consequently, the recurrence of bustum and Sabae- in the present passage is in no way suspicious. 89 The use of maritare with objects such as glaebas in the sense of ' to fertilise' is a predominantly Late Latin idiom. Poetical parallels are : Mar. Viet. Aleth. 3 .64 in.findunt rastris et semine rura maritant; ibid. 395-6 rura maritat aquis . . . I . . . Iordanis ; Ale. Avit. Carm. 1 .278 lympha maritavit sitientis viscera terrae ; Dracont. Romul. 3 . 3-4 rore maritat I arva suo ; Ennod. Carm. 1 . 3 .1-2 tellus . . . I lacte maritatis tu rgida caespitibus ; and P.L.M. v.422 Baehr. carm. u 8.2 imbre maritatum vegetabat spiritus orbem. C£ also Pervig. Ven. I I maritis imbribus and Mar. Viet. Aleth. 2.167 imbre marito. 1 For exordia used without a qualifying genitive c£ Claud. 1.6-7 iam nova germanis vestigia torqueat annus I consulibus, laetique petant exordia menses and Dracont. De laud. Dei 2.29-30 qui das exordia rebus I et finem, sine fine parens exordia nescis. 2 For the syntax c£ Claud . 20. 150 aliis exordia sume rapinis. The gen. saecli

is patently a simplification. It may here be noted that Claudian has the plural forms of saec(u)lum no less than thirty-eight times, tlte singular not once. 217

C OMMENTARY 9 1 The correct interpretation o f this verse, and especially o f the phrase medio sereno, has puzzled many critics. Various attempts have been made to dispose of medio by emendation, without success. Baehrens' nimbo may be dismissed at once as it involves a patent contradiction in terms between sereno and itself; Birt's radio is disqualified by Latin idiom (where that noun signifies rays of light, it is always used in the plural) ; and my own con­ jecture nitido, for which c£ Mart. 6.42.8 nitidum vacat serenum and Sil. 5.58 nitido . . . sereno, I no longer feel to be necessary. Heinsius favoured madido, the reading of his Puteaneus ( = my P 4 ), but, although the corruption of an original madido to media is easy enough (at Mart. 7·37·6 the MSS divide between media and madida), the phrase madido sereno is very hard to explain. Heinsius himself compares Ennod. Carm. 2.149 en sine nube pluit sub tectis imbre sereno, I et caeli facies pura ministrat aquas, an inapposite parallel, 1 while Gesner adduces Claud c. m. 28.5 imbres serenos, also irrelevant, referring as it does to the fact that Egypt is watered not by rain but by the Nile. I consider it all but certain that madido derives from v. 88 madidantes. Whatever be the true explanation of this verse, there seems no call to jettison medio. Parrhasius comments : 'nam quom proximae quaeque regiones per anni tempus id est hyemem nubibus praegraventur solus Ager Ennensis sereno fruitur caelo quod in eodem tractu Syracusis accidit' (then follow references to, amongst others, Plin. N.H. 2.62. 153, who notes that Syracuse has some hours of sunshine every day, even in winter), and a little further on, ' Medio sereno. nam quom undique circum loca nubila sint . . . medium caeli spacium quod Ennensibus campis imminet serenum est '. This interpretation is not impossible, but does seem rather fanciful. Barth is perhaps more likely to be right in thinking that the meaning is ' totum quod inter coelum & terram est serenari'. Alternatively one may take media as signifying all that lies between the extremities of the convexa, between the horizons, the heavens being thought of here as a bowl upturned over the earth : this amounts in effect to saying that the whole sky is clear and bright. The position of the word sereno at the end of the verse and the sentence suggests strongly that it is being used substantivally here, as at Claud. 28. 542. 93 For imbuit c£ Avien. Arat. 128 (cum sol) inbuerit (Grotius' correction for the MS inbuit et) tremulo Tartesia terga rubore ; 727 multus rubor inbuit ora ; 1 576 ne sanguineus late rubor inbuat ora ; Lucr. 2.502-3 aurea pavonum ridenti imbuta Iepore I saecla ; Val. Max. 8.u ext. 7 spongeam (pictoris) omnibus inbutam coloribus. The variant induit would not, however, be impossible, as witness Sil. 4. 12-13 detersa rubigine saevus I induiturferro splendor. • . .

1 The couplet o ccurs in an ep ig ram entitled De fonte baptisterii S. Stefani et aqua quae per columnas venit, and caelum is the vaulted ceiling of the baptistry.

2!8

BOOK

II

96 The confusion in MSS between Jucare and fuscare is found also at Claud. 22.3 50 and Ov. Trist. 2.487. 97 Volucris Iunonia pennas (D ] 4 K r) suggested by Ov. Med. Jac. 3 3 · 99-100 Incipiens . . . hiemps i s translated a s ' young winter's sky ' by Platnauer : c£ Parrhasius, who interprets it as a periphrasis for ' Autumnus ', commenting that ' tradit Aristoteles post Autumnale aequinoctium qualibet hora diei arcum fieri, aestate non fieri nisi incipiente & inclinato iam die '. Alternatively, and to my mind more probably, incipiens . . . hiemps means ' the gathering storm '. Interviret. After mentioning the innumeros colores of the rainbow ( v. 98 ), the poet singles out the green element in the spectrum to do duty for all . Similarly at Aen. 4-70o-I, Virgil writes Iris croceis per caelum roscida pinnis and continues mille trahens varios adverso sole colores. C£ also Prop. 3 . 5.32 purpureus . . . arcus and Claud. c.m. 28.4 arcum variata luce rubentem, where only one colour is specified. The prepositional prefix inter- does not, I think, signify ' inter alios colores ', as Parrhasius opined, but goes closely with discretis . . . nimbis (dat.) : the rainbow's path ' glows green amid the parting clouds ' (so Platnauer translates), or ' amid the clouds it pierces/through which it shines '. Internitet, preferred by Jeep, is much the more straight­ forward, and therefore the less likely reading, and the verb is in any case unpoetical, being confined to the elder Pliny, Solinus, and Curtius, who has a particular liking for it. 101-50 Description of the scene near Etna. The goddesses and their attendants begin to pick flowers. 118 This verse appears, variously located, in only eighteen of the fifty­ five vett., and in a nineteenth case is unmetrically conflated with v. I I9. Bonnet (in Revue critique ( 1 875 ) , no. 27, p. 8 ) was the first to suspect that it might be a spurious stop-gap, and Birt and Koch then condemned it on the ground that it is not until v. 123 that saltus cohors invadit. Against the authen­ ticity of the verse, quite apart from its insecure position in the tradition, its bald phrasing, and the stylistically unpleasant repetition of the word cohors, is the important fact that its insertion in the text at one stroke completely alters the setting of the Rape. From v. 72 onwards the poet depicts with much colour and variety of expression the region ncar Aetna parens jlorum, and then (at 1 1 2 ) he goes off into a digression on the celebrated Lake Pergus, which he describes as 'not far from ' the scene of the flower-picking (haud procul inde). 1 1 8, however, has the effect of transporting Proserpine and her companions from Etna to Lake Pergus, quite contrary to the poet's intentions. An interesting fact about 1 1 8 is its close association with the ' liber Catonianus ' (on which see the Introduction, p. 69 ), no less than seven of the ten MSS that incorporate it before 119 being elementary readers. From this . • .

. . •

21 9

C OMMENTARY

i t i s a reasonable inference that I I 8 was concocted b y some medieval school­ master, with assistance from Virgil (the ending perjlorea rura occurs at Aen . 1 .430 ) and v. I24 infra. 126 Cavo (L2 M I ) is presumably a case of erroneous assimilation, but alvus is masc. in Calvus etc. : c£ Charis. 8o.fin.-8I K. 131 Paul's alteration of maerens to maerentem is quite unnecessary. Cf. I77-8 infra : te, Latiis nondum praecincte tropaeis f Thybri. 136 I admit the Greek forms of Greek proper names if and when they are offered by MSS. Unanimously transmitted are : 1. 136 De/on (cf. c.m. 5 3 . I I 5 ), II.34 Delos (c£ 8 . I 3 3 . I I .8, 24.256, c.m. 5 3 . 125 ) , 63 Arcton (c£ I 5.5 I I, 26.246, 329), I 89 Arctos (c£ c.m. 5 3 . I I , 3.325, 8.51, a/.) and 353 Cocyton ; transmitted by one or more, but not all, MSS are : 1.74 Aeolos (c£ c.m. 22.3 5), 87 Cocytos (cf. 5.467), 201 Iden ( = ill . I40), 207 Ide (c£ 10. 1 8 ) , n.26 Gorgonos ( = 205 ) , 5 8 Pantagian and Gelan, 66 Tanain (c£ 3 .324, 26.603 ) , I36 Cephisos (c£ 2.Io), 3 3 8 Tityos (c£ 5.5I5, 7.160 ) . In the two cases where the MSS show no trace of the Greek endings, I print the Latin ones : at n.61 Alpheus and III. 3 45 Aegaeonis. 148-50 Translate : ' Nor did she who tracks with her hounds the scent (of the game) on Mt Parthenius spurn their company, but her only desire was to bind her loose tresses with a garland.' Odoris was probably inspired by Virg. Aen. 4.I32 odora canum vis, and odorwn by the proximity of Parthenium and comarum (assimilation of endings). For -que introducing an adversative connexion after a negative clause c£ D.R.P. II. I72, Ov. Met. 2 . 8u, and Manil. 1 .877 (with Housman's note ad loc. ). 151-205 Pluto emerges from the depths of the earth and snatches up Proserpine. 161 Debilis etfessis serpentibus inpedit axem. The vv./1. sulcis and actum are found only in the Excerpta Gudiana and produce a text that is indeed, in Gesner's words, ' eruditior quam ut stupori librariorum tribui possit '. Gesner, whose defence of these variants involves taking serpentibus as a participle and sulcis in the extraordinary sense of ' tractus seu volumina serpentum ' (he compares Apul . Met. I I.J.IJ-14 (corona) dextra laevaque sulcis insurgentium viperarum cohibita), is doubtless right in refusing to believe that a medieval scribe could have thought them up ; but could they not easily have come from the head of Jacob Cujas, who is known to have penned some of the Excerpta himself? I find sulcis serpeutibus a grotesque substitute for the lucid Jessis serpentibus, and actum an unconvincing alternative to the dear and apt word axem, for which cf. Stat. Theb. 10.479 impedit axes. 162 It is dear from the preceding lines that Pluto's chariot and team pass directly over the body of Enceladus. The sense of this verse must therefore be something like ' vestigium rotae fumantis imprimitur tergo sulfureo ' (paraphrase in the Delphin ed. ) . But can such a sense be obtained by keeping 220

B O OK II

the compound praelabitur, which has hitherto held the field? The force of prae- in the compound praelabi, as in other verbs of motion having the same prefix, may be either ' ahead of ' (as at Virg. Aen. 7. 807 praevertere ventos) or ' beside ' (as at Hor. Carm. 4.14.26 qui regna Dmmi praefluit Apuli ; Virg. Georg. 3 . 1 So Alphea rotis praelabi fiumina Pisae), but never ' over ', the sense required here. Nor do verbs of motion with the prae- prefix seem ever to be followed by any case except the ace. : c£ K.S. 1.272. The younger Burman toyed with the idea of reading perlabitur (found in five vett.) and altering sulphureo . . . dorso to sulphureum . . . dorsum, adducing Virg. Aen. 1. 147 rotis summas levibus perlabitur undas (one might add Sil. 3 .4!0 quotiens perlabitur aequora curru ), but this expedient is perhaps rather drastic when prolabitur, which might well be accompanied by an abl. (cf. D.R.P. 1. 187 cano rota pulvere labens and in general K.S. 1.3 5o-1 ) , is to hand. For prolabitur confused with praelabitur c£ Cic. N.D. 2.44. n4; other instances ofpro- corrupted to prae- may be found at Cat. 64. 127, Sen. Tro. 1 82 and Ag. 222. 165 Platnauer translates the MS inclusos . . . muros as ' foe-invested walls ', but the ellipse of ab hostibus vel sim. is extremely harsh, and in any case the position of the participle suggests strongly that it be taken with the phrase arcano limite. Since inclusos cannot meaningfully be so taken, Heinsius con­ jectured illusos and, quite recently (in C.R. n.s. x ( 1960) , 8-9), H. H. Huxley independently proposed inlusos (the better spelling : c£ my note on 1. 1 supra ; at Virg. Georg. 2 . 464 inclusasque is v.l. for inlusasque ). Inlusos may well be right, but since Claudian has seven examples of eludere for one of inludere, I am inclined to prefer the Isengrin' s elusos, which gives the same meaning : ' walls cheated by a hidden passage '. C£ deceptas . . . arces in the next line. 171 Most MSS have the false reading siculaque, which clearly derives from sicule in 173 . A few offer duraque which might be right, but equally might have been suggested by duros . . . nexus in 1 86. The Isengrin has solitaque which is possible (' with their wonted barrier '), but more probably points, as Heinsius saw, to an original solidaque. (For the corruption of solitum to solidum, conversely, c£ Ov. Ars am. 1.94 and Culex wo). If solidaque is what Claudian wrote, the presumption is that siculaque came about through the scribe's eye slipping down two lines, and that duraque represents a conjectural attempt to get rid of siculaque. For so/ida conpage c£ Luc. 9.467-8 si so/ida Libye conpage et pondere duro / clauderet exesis Austrum scopulosa cavernis (adduced by Heinsius) and Sen. N. Q. 7-9.fin. terrae so/ida Jortisque compages. 173 Heinsius preferred tonuere (c£ Sil. 8.654), but the alliteration pro­ duced by sonuere is worth keeping. I79ff. c£ Herodotus 7-129. 180 For examples of negare vetare (J 3 in fact has vetaret) c£ Shackleton Bailey, Propertiana, pp. 82-3. Heinsius favoured negarent, = reatsarent. =

22 1

C OMMENTARY

C £ Claud. c.ttt . 40.16 claraque se vetito proluet Ursa mari. Discolor, ' of a different hue ' from the rutilos . . . axes of 192, comes here to mean much the same as decolor, niger. Conversely at Stat. Theb. IO. I I S-19, where preceding caeruleo makes discolor = Iucida vel sim. Discolor is confused with decolor at Val. Fl. 5.564 ; the reverse at Prop. 4.3.IO and Ov. Trist. 5.3 .24. 196 In support ofpugnant Heinsius cites Ov. Met. 5·349 nititur ille quidem pugnatque resurgere saepe ; also Stat. Theb. 6.394 and 3 .28. 200-1 Reinhardt regarded these two verses as spurious (see the preface to vol. 2 of Jeep's ed. maior), but I cannot imagine why. Their authenticity seems to me unquestionable. In this context quantum is roughly equivalent to quam cito, just as at Stat. Theb. 6.405-6 tantum = tam cito, and the prefix dis- in discurrit is of course strictly applicable only to the phrase mentis acumen. Cf. in general Gratt. Cyn. 204. It may perhaps be worth noting that the phrase mentis discurrit acumen is adopted verbatim by Claudian's imitator Corippus at Ioh. 8.4, and in a slightly modified form by another imitator, Dracontius, at Romul. 2. 76. 203 Harenae looks to be out of place if one imagines, as Schrader clearly did, that the Rape took place in the fields of Henna. If, however, the mise et1 scene is on or near Etna, harenae falls naturally into place in the sense of ' lava-dust' (c£ III.385 infra and note on 1.122 ) . After referring to Jeep's ' discovery ' of a lacuna between 1.275 and 276, Bonnet remarks (Revue critique ( 1 875 ) , no. 27, p. 8 ) : ' Mais il y en a une presque aussi manifeste entre II.203 et 204, OU le sujet meme de 1' epopee, 1'enlevement de Proserpine, est passe sous silence. 11 est dit, sans doute : rapitur Proserpina cums ; mais avant qu'elle fiit entrainee sur le char, il fallait que le dieu l'eut prise et l'y cut fait monter.' This puerile note met with a deserved rebuff from Birt, who in his app. crit. ad loc. pointed out that quam . . . desiderat ille de ipso raptu narrationem consulto omisit poeta quam elegantius proferret III v. 202 etc. ; his autem idem narrare non licebat. 205-46 Pallas and Diana attempt to defend Proserpine but are warned off by Jupiter. Diana's farewell to Proserpine. 209 Jeep postulated yet another lacuna between ille and velut because he felt the need for a main verb after the isolated pronoun ille. The text is, however, perfectly sound, and the ellipse involved in this simile can be paralleled exactly. Prop. 2.2.6--7, which Birt cites, is not, I think, apposite, but c£ Stat. Thcb. 4.363 f£ ille velut pecoris lupus expugnator opimi 189 193

decedit stabulis hue illuc turbida vcrsans lumina, si duri comperta clade scquantur pastores, magnique Jugit non inscius ausi.

222

B O OK

II

C £ also ac velut . . . a t Stat. Theb. 7-436ff., discussed b y A . Ker in C. Q. n.s. iii ( I953 ) , I 79, and see Housman's note on Luc. 7.I25. 2II Rabiem totos exegit in armos. P 2 and Gesner take exigere as signifying much the same as exercere, but it is stronger than that. The sense is rather ' has driven out, exhausted his fury on the heifer's flanks '. 212 C£ Stat. Theb. 6.402-3 nexusque et torta iubarum I expediunt (sc. ministri). 216 Pace Birt (p. cliv n. I I ) , sede does not need an epithet, nor does the addition of tuae make quae . . . Eumenides an ' insulsa interrogatio '. C£ Shackleton Bailey, Propertiana, p. 73, for discussion of a Propertian passage where ' the interrogative pronoun merely adds rhetorical force ' ; and to the exx. there given add Iuv. 6.29 (adduced here by Barth), Sil. 1 .443-4 and Claud. 24.58-9. My preference for tuae is determined by metrical considera­ tions. In his De re metrica ( I 894) , p. 208, L. Mueller remarks : ' etenim quo quisque perfectior arte Romanorum, eo magis cavent, ne tali numero " infandum regina iubes renovare dolorem " post tertium trochaeum potius quam post quartam arsin sensus fiat interstitium . . . Itaque cum severissimi Ionge in arte sint Ovidius Lucanusque et Claudianus . . . Lucanus ultimo volumine et Claudianus carminibus de consulatu Olybrii et Probini et de Manlii Theodori numquam habent incisam sententiam trochaeo tertio, at hercule saepissime omnes post quartam arsin.' What Mueller maintains for the two panegyrics mentioned holds good without exception for all Claudian' s hexa­ meter poetry (including the elegiac hexameters). Examples of weak pauses at the third trochee there certainly are (cf. D.R.P. II.78, m.30I ) , but there is not a single example of a strong pause at the third trochee such as that which editors, erroneously in my opinion, have introduced into this verse by reading tua. In view of the mention of stimulis and Jacibus profanis, I suggest that Eumenides here is more easily understood literally than figuratively, meaning ' madness', as Birt interprets it (Index nominum, s.v. Eumenides). 228 For paene with pluperfect subjunctive c£ K.S. I.I74 A 2 . There is no need to emend to Juit, as was suggested by W. B. Anderson in the margin of his copy of Gesner, now in my possession. The variation between celsus and summus occurs also in MSS of Val. Fl. 3 .462. 230 Confessus socerum, ' acknowledging that he was the father-in-law' (of Pluto, his brother) : c£ Virg. Aen. 2.59I alma parens, confessa deam ; Stat. Theb. 2.I22 confessus avum ; Dracont. Romul. I0.330 sic memit veniam gener11m confessus Achilles. Expressions parallel to confessus generum (K I recc.) are Dracont. De laud. Dei 1.3-4 moenia caeli I auctorem confessa su11m and Anon. Vales. 62.I2 quem (sc. .filium) ipsa ante confessa est (sc. mater). Parrhasius, reading Hymenaeus, commented : ' . . . nunc tonitrum quo matrimonium Iuppiter approbauit Hymenaeo poeta tribuit '. Heinsius, .

223

C OMMENTARY

feeling that ' mirum est Hymenaeum intonare, & Iovis Toii !'povTCJVTos officio fungi', emended Hymenaeus to Hymenaeon (which in fact is found in n), but by retaining the initial capital he did not improve matters. The correct solution is surely to jettison the capital. That done, it makes little difference whether we read hymenaeus . . . I intonat, ' his marriage song thunders out ', or hymenaeon . . . I intonat, 'he thunders out his marriage song ' (for hymenaeon as internal ace. after verbs like intonare cf. Ov. Met. 12.21 5 ; Sen. Tro. 202 ; Stat. Silv. 2.7.87-8), for the context shows clearly that Jove's marriage song is his thunder. 236 For vinci . . .Jatemur c£ Claud. 3 . 109-10 ipsa quidem Jateor vinci rapidoque magistram I praevenit ingenio and Stat. Ach. 1.483 cedit turba ducum vincique haud maesta Jatetur. 247-306 Proserpine reproaches her father and calls on her mother for help. Pluto endeavours to assuage Proserpine's grief by telling her of the power that will be hers as his wife. 249 For rumpit cf. Sil. 4-456 gemitumque ad sidera rupit; also Virg. Aen. 4-553 ; Val. Fl. 1.508 ; 4·42· For Jundit c£ Sil. 16.43 7 questus ad siderafusi. 252 A similar use of totus = ' utterly ' occurs at Ov. Met. 9.593-4 sub­ versaque toto I obruor Oceano. 255 Especially in view of saeviret, rapido is, I think, less appropriate than rabido, for which c£ Luc. 7. 145 non aliter Phlegra rabidos tollente gigantas, Stat. Theb. 2.81-2 qualia per Rhodopen rabido convivia coetu I Bistones aut mediae ponunt convallibus Ossae, Sil. 7.253 his dictis Jractus Juror et rabida arma quierunt, 16.409 rabidi certaminis. The rapidus-rabidus confusion is discussed by Shackleton Bailey, Propertiana, pp. 202-3 , and E. J. Kenney, C. Q. n.s. viii (1958), 66 and ix (1959), 248. 258 Heinsius rejects culpae as a gloss on noxae, for which he cites Sen. Phoen. 9 non video noxae conscium nostrae diem. But the alliterative clausula conscia wlpae, employed also by Val. Fl. 4.3 56 and Stat. Ach. 1.562, is, I feel, more attractive in this passage of impassioned rhetoric. 264 A number of parallels may be found for the association of active forms of ducere with a dependent supine (c£, e.g., Plaut. Cist. 90 pompam me spectatum duxit ; Poen. 19-20 dissignator . . . neu sessum ducat; Cat. 10. 1-2 Varus me . . . ad suos amores I visum duxcrat e foro ; Hor. Serm. 2.4.89 ducere me auditum perges quocumque memento), but for the passive forms of ducere combined with the supine the only example cited in ThLL is Porphyr. Hor. Epist. I. 17. 6o ad quem ( Osirim) il/o tempore iuratum homines ducebantur, a passage, inci­ dentally, that is deleted by Petschenig and Holder. The present servitum ducor would therefore seem to be almost, if not quite unique. Of the other vv.ll., servitium, which appealed to some early editors, cannot stand beside captiva ; s(a)evitum may be dismissed on grounds of sense ; and itifelix, though superficially attractive, is first attested in the interpolating group J 4 K I . 224

B O OK II 275 For detergit Birt cm�ectured deterget, now found in G 1. Elsewhere, however, Claudian's MSS ar� far from revealing an aversion to the third conjugation forms : at I 8.268 deterget is read by E e n V2 and -it by B C2 ; at I7. I9I n E have -es and /\ -is ; and at 20.375 -unt is offered by B as against the -ent of Birt's other MSS. Because of these dissensions among the MSS it is impossible to decide whether Claudian preferred the second or the third conjugation forms, and I therefore retain the generally accepted form

detergit. 287 Similarly at Luc. 1.349 the true reading derunt is corrupted in some MSS to desunt. 297-9 W. H. Semple, ' Notes on Some Astronomical Passages of Claudian', C. Q. XXXI ( I937 ) , I6I-9, compares Macrob. Somn. 1 .21 . 3 3

omnia haec, quae de summo ad lunam usque perveniunt, sacra incorrupta divina sunt, quia in ipsis est aether semper idem nee umquam recipiens inaequalem varietatis aestum. infra lunam et aer et natura permutationis pariter incipiunt, et sicut aetheris et aeris, ita divinorum et caducorum lima confinium est. 300 Barth conjectured Jastigia (which was subsequently found in F I ), but neither in the unusual sense ' sublimitas regia ' which he attributed to it nor in the more obvious sense ' roof ', ' house ', is Jastigia as forceful as vestigia, here = pedes, as also at Claud. 5.432, IO.I46, D.R.P. m.253 , Sen. Thy. I039. Stat. Silv. 5.2.28. Translate : ' kings clad in purple shall come and prostrate themselves at your feet '. 3o6-72 Pluto returns with Proserpine to the underworld. Preparations are made for their wedding. All torments are suspended, and the inhabitants of Elysium raise their voices to bless the union. 308 Truwlentior may be supported by Ov. Her. I 1.9 ut Jerus est multoque suis truculentior Euris, but violentus is an epitheton perpetuum of winds. Cf. Claud. 20.5 violentum . . . Caurum ; German. Arat. 404 11e pacem pelagi solvat violentior Auster; also Virg . Georg. 2.Io7, Aen. 6.3 5 5-6, Luc. 2.6I7, 7. I25. 313 For passus mollescere, ' deigning to become gentle ', c£ Claud. 8.3 /imina nee passi circum privata morari (fasces) and c.m. 32·5 passus corporea

ntlln en vestirefigura. 317 Parrhasius reads lecta de plebe, which he explains as an instance of hypallage, = lecti de plebe, but I doubt if this is possible. Barth understands lecta to mean ' huic officio destinata ', but plebs surely denotes the whole community, not a large contingent of ministri. I therefore read lecti, and with it propere. 318 Pars altos revocant currus. Gronovius in his Observationes cites Stat.

Theb. 3 .4I3-I4 pars meritos vertunt ad molle iugales J gramett et erecto currtl/11 temone supinant in support ofhis contention that revocant here means supinant ; and the gloss supinant in fact is found in Z (in R4 revocat is glossed as resupinat). No dose parallel to this usage is available, but it is far from

225

HCD

CO MMENTARY

unlikely. It may be, however, that revocant means nothing more than ' they back the chariots (into their stalls) '. Both interpretations are canvassed by Parrhasius. 321 ' Vestes cultas . . . tollunt in thalamum, h.e. ex illis sublatis ac suspensis thalamum, velut tentorium quoddam constituunt, sub quo lectus sternitur ' (Gesner). 331 Rarescere noctem. c£ Stat. Theb. 1 1.73-4 quantumque profundae I rarescunt tenebrae (of the underworld) and Sil. 13.550-1 laxata Iucida nocte I claustra. Clarescere is not so apt here as it is at Apul. Met. 4· 19 taedis, lucernis . . . clareswnt tenebrae. 337 This verse might, as Heinsius asserts, be ' ex superioribus duobus conflatus, ac proinde vel invitis libris expungendus ', but one cannot rule out entirely the possibility that it may be a genuine alternative to 3 3 5-6. I print it in square brackets not because I am convinced that it is spurious, but simply to indicate that it cannot stand in the text as long as 3 3 5-6 remain there. 340 Lateris . . . opaci. Latus may refer to the interior of the body as well as to its surface, as we see from Claud. 7. 102-3 altum I . . . latus, and opacum is aptly used to describe the entrails : cf. Ov. Met. 2.274 (fontes) qt�ise condiderant in opacae viscera matris (sc. Terrae), where opacae is transferred from viscera. 344-5 Vinaferoci I crine bib11nt. Reinhardt would alter crine to ore, but the Eumenides 'pro crine habent cerastas et hie crinis ipse bibere dicitur ' (so Koch, in Birt's app. crit. ad loc.) ; cf. Gesner's note : ' (vase ampliore) . . . in quo plures simul serpentes, quos pro crinibus habent, (hi sunt etiam socii cerastae) mergere capita bibendi gratia possint'. Elsewhere the Eumenides' crines or comae (i.e. their snakes) are said to keep silence (c£ Stat. Theb. 9. 153 iussi tenuere silentia crines) and to hear (c£ Stat. Silv. 5. 1.27-8 nil cantus, nil .fila deis pallentis Avemi I Eumenidrunque audita co11ris mulcere valerent) ; c£ also Val. Fl. 6. 176. 36o Vacuos egit . . . rernos. Postgate refused to allow that Charon con­ tinued to ply his oars while the rest of the underworld was making holiday, and conjectured legit, ' drew in', ' shipped'. The important word here, however, is vacuos, and provided that this is duly emphasised, there is no call for emendation. 364 Genialis seems better authenticated in nuptial contexts than genitalis, being associated with pacta at Stat. Ach. 2.68-9, iura at Stat. Si/v. 3 . 3 . 108 and Theb. 3 .689, joedera at Stat. Theb. 3.300,1 epulae at Claud. D.R.P. II.327, /cetus at Hor. Epist. 1 . 1 . 87, and thalami at Claud. 5.96 (again with v.l. genit-). Ecclesiastical prose apart, the only certain instance of the use of genitalis in the context of marriage is Iuvenc. 3 .485 genitali lege tororum. I therefore read genialia here. 365 Exultant cum voce may mean ' express their joy in song ' (for cum 1

So Mueller and Garrod,

with P; L. &

226

S. cite

the vulgate genitalia.

BO OK

II

voce = an instrumental abl., c£ �adrig. ap. Gell. 9.1 3 . 10 cum voce maxima conclamat), or alternatively, ' dance and sing ' (c£ 360 above egit cum carmine remos) . At all events, I cannot agree with Birt that ' cum non praepositionis, sed coniunctionis locum habet '. 367 In invocations patens seems normally to be qualified by mention of the sphere in which the deity's power is exercised (c£, e.g. , Stat. Theb. 4·747 and 6.63 3 diva patens nemorum). Exceptions however do occur : cf. Sen. Phaedr. 1 1 14-16 o nimium patens / Natura ; Oct. 377 patens Fortuna ; Stat. Silv. 4. 1.28 and Auson. Masella 3 78 Roma patens. Parens, found in D J 4 P 2 W, is not impossible, but most likely is a secondary reading suggested by parentes in v. 3 56 supra. For the corruption of patens to parens cf. Claud. 1 . 8 and 22.442 (where Birt refers to Markland's note on Stat. Silv. 4.1.28 ) . The pronominal adjective nostra is added to distinguish between the heavenly Juno and her infernal counterpart : c£ Claud. 10. 176 where Neptune is called noster Iuppiter by his sea nymphs. 368 Cf. in general Claud. c.m. 25.130 vivite concordes et nostrum discite munus. Koch, Adnot. crit. p. xviii, argues that Claudian preferred unanimus etc. to unanimis etc. This is very likely, as the MSS of the classical writers, notably those of Virgil, almost invariably offer only the second declension forms. 369 Vota here seems to be almost equivalent to ' loved one ' : c£ Dracont. Romul. 6.3 dulcia cantatis (sc. iuvenes) dum votis carmina vestris (at Ov. Fast. 5·459 and Cons. ad Liv. 39 votum signifies ' love ', ' affection'). The variant colla was doubtless suggested by reminiscence of the common clausula colla lacertis, found four times in Ovid and also in Sil. 9. 144.

B O O K III 1-66 Jupiter calls a council o f all gods and spirits a t which he justifies his former withholding of the bounties enjoyed by mankind under Saturn's reign and decrees now that Ceres shall give her corn to man when she has found her daughter. His subjects are, however, forbidden to reveal to Ceres the identity of her daughter's abductor. 2 Either arcessere or accersere is possible : c£ G.L.K. VI1.71.17 et in eo quod est ' arcesso ' et ' accerso ' putaverunt quidam differentiam esse . . . sed erravertmt. The word occurs only here in Claudian, so there is no means of deciding which form he preferred. 3 Most MSS offer Zephyros, and the ace. after inlabi may be supported by Sil. 8.454, Stat. Theb. 7.6, Avien. Orb. terr. 173, Arat. 74o-1. Claudian's normal practice, however, is to use inlabi with dependent dat. (so at 1 5.226, 24.367, D.R.P. III. 3 30; cf. Luc. 1 .475 ) , and the minority reading Zephyris is therefore more likely to be right here.

227

I5·2

C O M MENTARY

11-12 For reverendaque . . . canities c £ Prud. Perist. 1 1.137-8 reverendam I canitiem molli confovet in gremio ; Ambros. Epist. 16.5 ea est enim reverenda canities, quae est canities ani mae ; Rufin. Hist. monach. 9 (Migne 21 .426 A ) virum canitie reverendum ; Hist. 3 .6.7 nulla senibus pro canitie reverentia. Venerandaque, the reading of F 7, would be equally good here, as witness Sen. H.f 1248-9 perque venerandos piis J canos, Val. Max. 4·5 ext. 2 qui hominis aetate moti canos eius et annos . . . venerati sunt, Greg. Tur. Mart. r.6 apparuit eis veneranda canities senis. Heinsius, followed by the younger Burman, accepted the Isengrin's et caerula, but (i) although caerul(e)us is commonly used of sea­ gods and -goddesses (c£ the passage adduced by Heinsius : Cons. ad Liv. 435-6 illi caeruleum Panope matertera crinem I solvit), its association with canities would be bizarre, to say the least, and (ii) it does not balance placidus nearly so aptly as does reverendaque. Barth and Gesner read et Iucida, for which they might have compared Apul. Met. 5 . 1 6 candenti canitie lucidus, but neither tllis lection, nor the similar et candida, seems as good as reverendaque. 19 Heinsius aptly compares Luc. 7.3 1 1 di, quorum wras abduxit ab aethere tellus. At Virg. Aen. 3 .6or the true reading abducite is corrupted in M (ante corr.) to adducite. 24 In support of canesceret (R 5 R 7) Heinsius cites Ov. Met. r.no nee renovatus ager gravidis cane bat aristis ; at Met. 2.212, however, pabula canescunt = pabula rtruntur. 26 ff. In the editions ofBirt and Koch 27-9 are bracketed as a parenthesis and 30-2 are taken as dependent, like 24-6, on inpellere in 23 . But (a) sed . . . mentes must stand outside the parenthesis to facilitate the transition from negative to positive resolution, and (b) there is some awkwardness in the switch from imperfect subjunctives (grandesceret etc.) to present ones (provocet etc.), although less than if the present indicatives of the parenthesis had not intervened. I therefore follow Heinsius and Jeep in punctuating with a full-stop after 26, and keep haud equidem invideo outside the parenthesis (again with Heinsius and Jeep), taking it in the sense ' I do not indeed act out of spite ' rather than as ' I am jealous ' ; the words sed . . . provocet ut . . . egestas then follow naturally after the present indicative invideo with the meaning ' but to the end that want may stimulate . . . ' It may here be noted that Jeep's alteration of totae . . . ripae to toti . . . rivi in 26 is not necessary : c£ Stat. Silv. 1 . 3 . 107 debuit et jlavis Hermus transwrrere ripis. 27 Neque always (never nee) in the formulaic neque enim. Cf. Axelson, Unpoetische Wiirter, pp. 1 1 5 ff., for discussion of nec-neque. 29 Oblimare is extremely rare in a figurative sense, here = ' to dull'. In Solinus 16.3 ita . . . nomina Graeci permiscuerunt, ut . . . paene oblimaverint universa it means rather ' to obscure '. 33 The form querella is almost universal in the Virgil MSS and the Excerpta Florentina of Claudianus maior, and at Claud. c.m. 22.9 it is offered 228

B O O K III

by both the Mediceus and the Veronensis. I therefore accept i t here (from four recc.) and at 160 infra (from L 6 ) . See further Lachmann's discussion at Lucr. 3 .1015. 33-4 To Barth's comment on 34 ' melius absit totus hie versus ' Heinsius retorted ' nescio quid sibi velit vir eruditus, qui versum hunc expungendum censet'. So far as I can see, the only feature of 3 3-4 to which exception might be taken is the very unusual idiom mihi instat relevare instat mihi ut relevem (Tac. Ann. 1 1.34 affords a near parallel to this, unless Madvig' s emendations be accepted), but I cannot believe that this syntactical rarity is an adequate ground for the deletion of 34· 39-40 C£ in general Q.!!intil. 12.1.2 rerum ipsa natura . . . non parens sed noverca Juerit. 44 Addita, read by Birt and Koch ( ? because they knew abdita only from �). is very much the inferior reading. The same two participles are also confused, but in the opposite direction, at Sil. 1 5.68. 48 L & S, s.v. adeo, ad fin., observe that ' The assumption of a causal signi£ of adeo = ideo, propterea, rests upon false readings', and nothing in ThLL gainsays this observation. Since the sense required here is ' and therefore', I consequently read atque ideo, comparing Virg. Georg. 3 .212 and Val. Fl. 4. 128. The reverse corruption, adeo to ideo, occurs at Stat. Theb. 1.15, =

2 ! 9.

50 There is no need for Baehrens' rabido. Avidus luctus means 'insatiable grief ' : c£ Sen. Tro. 76off., where Andromache pleads brevem moram largire dum . . I . . . amplexu ultimo I avidos dolores satio. 59 ' Q.!!idam dicunt se licet ipsa meo conceptam et cetera quia pro Pallade dictum est. sed hoc non permittit linea sequens quia pro eodem sequitur genitum et saucius ictu. idcirco melius dicitur ipse et conceptum quamvis pro Pallade dicatur. masculinum enim genus protulit auctor quia praedixerat si quis audeat ' (note in 0 2 P2). This is much the line taken by Heinsius, who adds : ' nimis manifeste Pallas tangatur, si cum vulgatis legas, . . . ilia . . . conceptam '. But the reference to Pallas is quite unmistakable, and there is no justification here for a masculine termination as there is in v. 57 where the gender is determined by natus. In vv. 6r and 62 we return from mention of a particular deity to the generic idea, expressed initially by quis, and therefore to the masculine gender. 6o Heinsius, while printing the vulgate iratam procul aegida, opts in his note on this line for iratum, proCil[ aegide, which Gesner explains as ' iratum me, procul a sua aegide' : in other words, ' (shall feel) my anger when she is far from her shield'. Alternatively, iratum procul aegide might mean ' my anger expressed afar offby the thunder of my aegis '. In view ofthe repetition of sentiet, however, I am inclined to prefer the clearer and neater iratam . . . aegida, which aptly balances ictrtm Julminis in the parallel clause. .

22 9

C O MMENTARY

62 I print languidus a s being slightly less obvious than saucius, which may be due to scribal recollection of Virg. Aen. 12.5 saucius . . . vulnere. 64 P 2 comments : ' sciet an Tartara consentiant causae regis sui '. 6 6 In all the parallel passages that may be adduced for motu (c£ Cat. 64.204-6 adnuit . . . I quo motu tellus atque horrida contremuerunt I aequora ; Ov. Am. 3 .2.58 adnuit et motu signa secunda dedit; Met. 8.78o-1 adnuit his capitisque sui pulcherrima motu I concussit . . . agros ; Stat. Theb. 7.3-4 concussitque caput, motu quo celsa laborant I sidera ) , adnuit or a similar verbal idea is expressed just before it, but in this case we just have dixit. I think therefore that 11utu is the better reading here : c£ Ov. Fast. 2.489-90 nutu tremefactus uterque I estpolus (cited by Heinsius) and Virg. Aen. 9.106 adnuit, et tatum nutu tremefecit Olympum. 67-IIO Ceres is troubled by visions of ill omen. She dreams that she sees Proserpine herself, pallid and in chains, begging to be released from her captivity. 78 For quaerenti Heinsius compares Stat. Theb. 9-597-8 quaerenti nympha cruentas I Maenadas atque hostem dixit saevisse Lyaeum. 79 In an article in ALL IV.590 entitled ' Verbalform von Perfektstamme bei Claudian ', Birt notes that debellavisse provides the only exception to the ' rule' that ' Wahrend er [sc. Claudian] die sigmatischen Formen . . . stets zusammenzieht [e.g. 5. 122 umbrasse ; D.R.P. 1.47 vexasset] konserviert er dagegen in der " r-" Formen principiell den urspriinglichen Silbenbestand '. He therefore conjectured sic debellasse so that there should be no exceptions to his ' rule '. I am not convinced, especially when I observe that D.R.P. II p£ 48 lustrarunt is equally exceptional in Claudian. For debellare used figuratively c£ Claud. 3.378--9 debellatasque draconum I tonsa comas (Megacra) ; Stat. Theb. 7.86-7 debellatasque relinquit I Eurus aquas ; also Plin. N.H. 20.6(23 ).50 (alium) aconitum debellat; 22.23 (47).99 Jebel/at eos (fungos) et aceti natura contraria iis. 81 C£ Luc. 7.785 ingestaque Tartara somnis and Macr. Somn. 1 . 1 .2 rerum facies non dissimilia sign!ficans a Tullia11o Scipione per quietem sibi ingesta narratur. 88 Exhaustusque gelu pallet rubor. C£ Rhet. ad Herenn. 4. 10.14 ubi . . . iste vidisset scurram exhausto rubore ; also Stat. Theb. 8.215-16 corda levavit I exhaustus sermone dolor. Heinsius' conjecture exustus (now found in R29) lacks good parallel (the only relevant citations in ThLL, s.v. exurere, v.2. 2127.14ff., are from Plin. N.H., Vindic. med. and Ps.-Apul. ), but may be thought to derive some support from Ov. Trist. 3 .2.8 ustus ab adsiduo frigore Pontus and Luc. 4.52 urebant montana nives. 94 Reinhardt's alteration of me to te is attractive, but not, I think, absolutely necessary. An amply sufficient defence of the MS reading is provided by Parrhasius : ' In me quae Iavis soror & ex eo mater sum :

230

B O O K III

simulque matemus affectus exprimitur quae filiae cruciatu non secus ac suo movetur.' 103 So many are the variants available here (indulges and indulgens ; Phrygias and Phrygiasque ; etiamnum, etiamnunc, et nunc, vel nunc and iam nunc ; interstrepis and perstrepis) that no less than ten prima facie acceptable versions of this line may be constructed. While it is probably impossible to determine conclusively which of them is the one Claudian wrote, it is a reasonable assumption that the multiplicity of variants stems from a series of attempts to normalise or simplify a difficult or abnormal original. On this assumption the comparatively common perstrepo should be rejected in favour of the very rare interstrepo, 1 and of versions incorporating interstrepo that one should be preferred which exhibits the greatest metrical complexity. I therefore adopt, with the edd. vett., the version indulges Phrygiasque etiamnum interstrepis urbes which involves the combination, apparently unique in Claudian, of penthemimeral and hepthernimeral elisions with only trithernimeral caesura. This version was rejected by Jeep, ed. maior dxiv, precisely because of its metrical abnormality, but there are cases in Claudian where two elisions occur in one line (rr1.282 infra, 5.216, c.m.'·9.44), and instances of verses having only one weak caesura (m.252 infra, 15.3 16, 26. 1 8 3 ) or none at all (5. u8), and objections drawn from metre cannot therefore carry any weight. 106 Dejende here must have the sense of ' Iibera' (so Parrhasius), but I know of no parallel. 108 Vel tantum visura veni, ' at least come just to see me'. Vel here = saltern, which in fact is offered by some MSS in place of tantum : cf. Madvig's note on Cic. De fin. 4.16.43 . 110 For somnum solvere c£ Ov. Met. 3 .63o-- 1 veluti clamore solutus / sit sopor. The alternative phrase somnum excussere is paralleled at Ov. Her. 10. 1 3 , Ibis 154, Met. 1 1.677-8, and [Sen.] Oct. 123 . 111--69 Starting up from her sleep, Ceres pours out her fears to Cybele and insists that she must return to Sicily. On her return she fmds her house deserted and is stupefied by grie£ n6 For the variant cautis Heinsius adduced Sen. H.O. 6u intrant fraudes cautique doli. Cf. his conjecture at II. I I supra. 130 The plural forms of buxus hardly ever follow the fourth declension paradigm, Grom. 303.14 being the only certain case known to ThLL and Neue, Formenlehre 1.770. It is, I suppose, just possible that Claudian experi­ mented with the form buxus here, but since buxos is found in all his MSS at 20.286 (so Birt), and since elsewhere in classical poetry (e.g. Mart. 3 .20. 1 3 ) ' Here used in the unparalleled sense of ' cum strepitu peragrare ' (see ThLL, s.v.). The verb is used absolutely at Claud . 5 .303, but never again in classical poetry. On Virg. Eel. 9·3 6, where inter and strepere are traditionally divi­ ded, Servius notes that inter strepere sane et composite potest et separatim proferri.

2J I

COMMENTARY

the plural forms belong t o the second declension, I think there i s very good reason for preferring buxos. 137-8 Most MSS offer sed nulla ruenti I mobilitas. With this text we have to assume that it is not est but videtur that is to be understood. But this is a most unlikely ellipse. The best solution, I think, is not to replace ruenti by the very obvious interpolation videtur, as Jeep did nnder the influence of F I , but t o write Heinsius' conjecture sat for the MS sed. Perfect sense i s now restored : ' no speed is enough for her as she rushes along '. For the corruption of sat to sed c£ Luc. 8.3 14. 138 Tardos queritur non ire iugales. For the apparent contradiction, which T and R2o appear to have fonnd difficult (they read nunc for non), c£ Claud. 10.15 segnemque rotam non jlectere Phoebe (videtur) and c.m. 27.61 pigrosque polus non concitat axes. The variation in MSS between iugales and dracones occurs also at v. 54 supra. 143 Cogitat . . . ne . . . . An extremely rare construction which, if we may believe ThLL, is paralleled only at Cassiod. Hist. IO. I 3 . i 176A cogitabat enim, ne ulla propter eum seditio nasceretur. It may be, however, that cogitat is supported by aestuat (v. 141 ) : c£ Sen. Benef. 6.42. 1 verentis et aestuantis, ne in ullo officio sis tardior, Ammian. 29.5.15 imo . . . aestuans corde, nequid . . . omitteret, and Cassiod. Hist. I I . I 8 aestuabat, nequid mali Ardaburius sustineret. 144 For gracilis used of insubstantial and therefore vulnerable objects cf. Stat. Silv. 3 .2.78 ff.Jugit . . . ratis acta per undas I . . 1 . . . tot gracili ligna complexa timores, and for the confusion between gracilis and Jragilis c£ Calp. Eel. 2.87. 145 Jeep, ed. maior r.lxvi, adopts ne furtum pateat homini from F I, but I can find no evidence to suggest that Claudian would have allowed such a lengthening in arsi where there is no following pause. 151 Spiritus oris. For the phrase c£ Virg. Georg. 4.30D-I spiritus oris I obstruitur, Ov. Met. 15.303-4 spiritus oris I tendere vesicam so let and Repos. 19 intermixti captatur spiritus oris. The reading ori was doubtless produced by syntactical assimilation to redditur at the beginning of 152. 159 C£ Stat. Theb. 5·594 non lacrimas habet: ingeminat misera oscula tantum. 160 Abrumpit mutas in.fila querellas, ' she stifles her complaints in the cloth before they find utterance' : c£ Val. Fl. 7.388 ilia nihil contra vocesque abrumpit inanes. Gesner's absumit (' impendit, consumit in fila ') misses the point. 170-259 Catching sight of the nurse Electra, Ceres demands to know what has happened. Electra relates the events of the fateful day when Proserpine was carried off. 176 Proxima mater, as Buecheler rightly notes (in Birt' s app. crit. ad loc. ) , does duty for altera mater. For similar uses of proximus c£ Ov. Met. 8.595-6 o proxima mundi I regna . . . dixi, sortite tridentifer and Stat. Silv. 5.2.47 ille secundus apex bellomm et proxima cassis. .

• . .

232

BOOK

III

177 Although certain other conformations are probably possible, viz. lacrimis confusa genas (cf. Ov. Trist. 3.5.u ) and laceras diffusa comas (c£ D.R.P. n.248 supra),1 the vulgate laceras effusa comas seems as good as any­ thing : c£ Ov. Met. 1 3 .688 effusae . . . comas, Luc. 9.172 ejfusa comas and Val. Fl. 3 · 3 !4-15 laceras . . . I fusa comas. ' Hoc loco comas ferri nequit quia canos sequitur ', maintains Birt (p. clvi n. 7), but c£ Stat. Theb. 1 1.582ff. veteri stat sordida tabo I utraque canities, et durus sanguine crinis I obnubitfuriale caput. Cano, if it is not just a simple slip, may be due to a reminiscence of the phrase cano pulvere at 1.1 87 supra. 179-Bo Suspiria . . .frenosque by hendiadys for suspiriafrenata. C£ Claud. 28.265--6 ergo ubi praeclusae voci laxata remisit I frena dolor. 181 Cui praeda Jeror? ' Vehemens communicatio ', says Gesner : ' se praedam dicit ferri, cum abripitur ftiia.' Even allowing for Ceres' hysterical condition, however, I doubt whether she would be likely to say that she herself ' is carried off '. Perhaps the sense of this utterance is rather 'who boasts of having made me his prey ? ' Alternatively, Jeror may perhaps simply = sum : c£ Housman on Manil. 2.188 (with his Addenda in the final volume, p. 139 ) . C£ Sen. H.f u 86, where Hercules, surveying his slaughtered family, exclaims cui praeda iacui, and Claud. 26.85ff., where Alaric qui mente profundas I hauserat urbis opes, ultra victoribus ipse I praedafuit. 183 Typhoea and Typhoia are offered here by the MSS. Both forms are recognised by Servius (on Virg. Aen. 1.665 ) , but Typhoea is the one found in our Virgil MSS, which I follow. 189 For heu ubi, with hiatus, c£ Ov. Am. 3 . 8 (7 ) . 1 8, Her. 4. 1 50, 6.41, Fast. 3 .485, 5.465, Stat. Theb. 5.3 50, 613, 8.174, Silv. 3 · 5·44; note also heu iterum at Stat. Theb. 5·478 and heu ubinam at Theb. 9. 385. Jeep wrongly prints heus ubi, completely altering the tone and sense of the utterance : c£ ThLL, s.v. heus. 197 Utinam . . . dederit. The tense of the subjunctive seems to be un­ paralleled in the sense required, which is obviously ' 0 that the crazed army of the Giants had brought about this calamity', not ' . . . may prove to have . . . ' : c£ K. S. 1. 182-3 . 201 Infestior, for which c£ Ov. Trist. 1.4.26 and ThLL s.v. infestus, VII . 1408.22 ff. (esp. 39ff. ) , is perhaps more likely to be a secondary reading than infensior : at Virg. Aen. 10.521 infensam is corrupted in some MSS to infestam. L & S, s.v. incendo, accept the variant incensior as the truth, but this comparative form is found only in Porph. Hor. Serm. 1 .2. I2G-I and ( ?) Paul. 1 Lacrimis effusa, ' burs tin g into tears ' (c£ Virg. Aen. 2.65 1 nos contra effusi lacrimis), is not quite right for this context where what is needed is a phrase describing a state, not an action. Birt prints laceras infusa genas with the note ' sc. infunduntur genae lacrimis; atque placeret pro laceras si lacrimis rescriberetur ' , but infusa genas (with or without lacrimis) strikes me as rather dubious Latin.

23 3

C O MMENTARY

Nol. Carm. 28.107 (conjecturally restored), and in any case the sense i s not as appropriate here. 213-14 Either vetitoque dearum I colloquia or vetitamque dearum I colloquia is possible, though since vetitamque produces a very rare construction (paralleled at Stat. Theb. 12.558 quos vetat igne Creon), vetitoque may reason­ ably be regarded as a normalisation. In 214 the choice is between amandaverit and absentaverit. Amando com­ mands classical support (it is used several times by Cic. , and recurs in Tac. Hist. 4.56 and Gell. r2. r ad.fin. ) and appears elsewhere in poetry (at Paul. Nol. Carm. 19.428, 658) ; absento is unpoetical and late, with appearances at Cypr. Sing. cler. (CSEL iii) 30 and 32, Aug. Gen. ad /itt. (Migne xxxiv) 12.23, Ps.-Leo M. Serm. (Migne liv) 13.3, Prosp. In psalm. (Migne li) II9.6, and Cod. Theod. 12. 1 . 84. Neither word occurs elsewhere in Claudian, but perhaps the fact that an1ando has some classical and poetical currency should be thought to tell in its favour. 221 Adjatu should be read here, not ajjlatu. Cf . Claud. 5.367-8 Ru.finus sequitur, quo fallere cuncta solebat I callidus adfatu and Stat. Theb. 8.244-5 sociumque benignos I adfatus . . . passum ; at Val. Fl. 6.473 one of the properties of Venus' girdle is adfatus mali. 233 Albet makes its first appearance in Parrhasius' notes and, since Heinsius, has been generally adopted by editors, who compare Val. Fl. 2.73 albet ager, Hor. Carm. 1.4.4 nee prata canis albicant pruinis and Ov. Fast. 3 . 880 canuerint herbae rore recente. But the reading of all our MSS, alget, is quite unobjectionable : c£ Virg. Georg. 3-324-5 frigida rura I . . . dum gramina canent and Calp. Eel. 5-54frigida nocturno tanguntur pascua rore. 235 The variant atra may have been suggested by Virg. Aen. 5.721 or Stat. Theb. 1.346. 236 I feel sure that Heinsius was right in preferring pulsu strepituque rotarum to strepitu pulsuque rotarum, (a) because pulsus is regularly applied to feet or horses' hooves (c£ Virg. Aen. 6.591, 12.334-5, 533, Stat. Theb. 7. 120 ) but not to wheels, and (b) because strepitus rotarum (found in Claudian's imitator Dracontius at Romul. 7. 88 ) is better than pulsus rotarum. 237 Aestus does not in fact occur in C r, as Birt alleges ; the MS tradition is practically unanimous in offering ille (four dissident recc. having ipse), and the first appearance of aestus is in the edition of Claverius, who probably appropriated it from Lucr. 6. r r 3 8. The application of mortifer to ille, though unusual (the epithet is normally used oflethal states and forces, and especially of disease and plague, Prud. Perist. 3 .29--3 0 (iubet) iecur pecudis I mortiferis adolere deis being the only exception known to me), need cause no difficulty ; whereas it would, as Barth noted, be decidedly odd to say that the chariot was driven by a ' deadly exhalation '. 238 Luror is not, as Heinsius alleges, ' scriptorum praestantiorum lectio ', ·

23 4

B O O K III

but comes from Parrhasius' edition. Livor (all MSS), meaning a taint in vegetation (cf. Iuv. 2.8 1 ), is much the better word here. 241 For the rare use of decrescere c£ Aug. In psalm. 147. 19 crescant zizania, decrescat Jrumentum and Prob. Virg. Eel. 10.62 HAMADR YADES. nymphae pariter cum arboribus et crescere et decrescere traduntur. Marcescere, conjectured by Koch but in fact occurring in L 10 U, is wilikely in view of marcebant only six lines further on ; and Heinsius' arescere, for which c£ Cic. Oecon. apud Non. p. 450. 1 M. herbas arescere, Tac. Ann. 1 3 . 5 8 arescente trunco, and Firm. Mat. De err. xiii aeris ardore arescentibus Jrugibus, is no improvement on decrescere. 246 Cervix redimita iacebat: c£ Luc. 4-754 fossa iacet cervix and Stat. Theb. 10.306-8 huic languida cervix I in laevum cogente deo mediaque iacebant I colla replicta lyra. Redimita (sc. sertis) figures again without qualification at Stat. Theb. 10.101 hie haeret lateri redimita Voluptas, Ach. 1.612 vibravitque gravi redimitum missile dextra and Claud. 22.345 redimitae tempora Nymphae. 248 Subito = statim, as also at Claud. 20. 3 56 and c.m. 27.49. 251 In our MSS the theme solvitur in laticem is followed by two elabora­ tions, subrepit crinibus umor and in roretn . . . pedes et bracchia manant; between these stands the isolated verb liquitur, which is otiose in itself and both syntactically and stylistically unrelated to its context. I therefore read sub­ repens crinibus umor I liquitur, comparing for the phrasing Virg. Georg. 1 .43-4 gelidus . . . cum montibus umor I liquitur, and for the corruption of -ens to -it D.R.P. 1. 1 5 1 consurgens] consurgit P 7 and 175 reposcens] reposcit L6 0 1 ; similar is n1.26 1 torquens] torquet b 1 . 200-329 Ceres now rages against the goddesses, now pleads with them, but to no avail. Finally, she resolves to search the world for her daughter. 262 Perhaps the best form of this line is ultro ad caelicolas Juriato pectore Jertur, with ultro meaning sponte (gloss in J 2 ) , ' on a natural impulse '. Heinsius aptly compares 357 infra and Virg. Aen. 4.304 tandem his Aenean compellat vocibus ultro (sc. Dido). Of the alternatives to ultro, multum is in context weak, and vultu probably impossible in the absence of a qualifying adjective (c£ Val. Fl. 6.584, 7.292, Sil. 4.234 and 7·75 )· To accommodate vultu, Heinsius suggested adding et betweenJuriato and pectore and Buecheler emending lumina to Julmina, but neither is convincing. 265 Birt rejected avexit, which he knew only from Claverius, in favour of advexit, the reading of all his MSS. Avexit does however occur in seven vett., and as A. Ker pointed out in C. Q. n.s. vii (1957), 158, is much the better word here : all that the huntsman has done so far is to carry the cubs away from their mother ; he has not yet succeeded in conveying them to his king. Claverius very aptly compares Mela 3 · 5·43 (eques) ubi interceptos earum catulos citus coepit avehere, et . . . astu . . . unum . . . omisit . . At Virg. Aen. 1.512 avexerat is confused with advexerat. The adjectival phrase marito I mobilior Zephyro, with its allusion to the .

23 5

C OMMENTARY

theory o f impregnation by the wind (c£ Virg. Georg. 3 .271 ££, Colum. 6. 27.4££, Varro, R.R. 2.1.9 ) , seems to cry out for an accompanying verb of motion. Hence Heinsius' certain emendation of Jremit to premit: ,' swifter than the west wind that is her mate does the tigress pursue the huntsman '. For the corruption ofpremere tofremere c£ Virg. Georg. 3 .85, Ov. Met. 14.779, Luc. 9.3 32, Stat. Theb. 5.78, Macr. Sat. 5.17.6. Confirmation of the emenda­ tion is provided by Claud. 3 .227 Hyrcana premens raptorem belua partus (adduced by Dempster), Ambros. Hexaemeron 6.4.21 , Schenkl vol. 1.217-1 8 (cited almost verbatim in the medieval Physiologus and in the margins of 0 2 P 2 ) , and Alexander Neckam, De laudibus divinae sapientiae 9. 127-8 tigris sublato foetu velocior aura I instat atrox, sed nee segnius hostis abit. 266-7 Scriverius (Anecdota Philologica et Poetica, ed. Westerhovius ( 1752 ) , p. 1 3 ) saw in maculis a reference to theformido, a rope with coloured feathers attached to it which was found effective in snaring game : c£ Gratt. Cyn. 75 ££ This interpretation, however, is quite impossible : far from being hunted, the tigress is herself the huntress, and the only defence of the tremebundus eques lies in the mirror by which he hopes to delay her. The word maculae, here used by metonymy for pel/is maculosa (c£ Stat. Theb. 7-571 ) , may designate the ' markings ' of a wide range of animals, both wild and domestic : c£ Iuv. 15.160, Virg. Georg. 3.56, 4.91, Stat. Theb. loc. cit. and 2.128-9 qua/is ubi mtdito venantum mummre tigris I hormit in maculas. That the blackish stripes of the tigress should be called virentes is most curious. Conceivably tlus is a case of hypallage (c£ Sil. 5.569 viridissim11s irae), but since the leopard is similarly termed viridis at Claud. 1 7.305 and 24.345, a more likely explanation, perhaps, is that in all three passages the epithets viridis and virens have extended their range ofmeaning into that of caerul(e )us and come to signify ' dark', ' blackish', with reference to the spots or stripes of the cat family. Translate : ' diffuses all her anger from her dark stripes '. The nimiumq11e of most MSS adds nothing if taken with hattsllra ; if understood as intensifying profunda, it produces an intolerable hyperbole ; and distance forbids its being taken with tardatur. Other MSS offer vivumq11e and timid11mque, but the former of these is quite superfl.uous, and the latter otiose beside the tremeb11ttd11s of v. 265 ( ' neque est cur timor viri urgueatur Koch, pref. p. lix). Far and away the best reading is Claverius' iamiamque : the tigress is just on the point of seizing the huntsman when her attention is diverted by the sight of her refl.ection in the mirror he holds in front of her. For iamiam with future participle c£ Ov. Met. 12.588 and Claud. 8.6o ; for the form ha11s11ra (rather than haustllra) Jeep compares Virg. Aen. 4.3 8 3 . 273 For cecidere c£ Cic. Place. 2 . 3 Sll bsidia reipublicae, consi/ia, auxilia, iura ceciderunt, Lucr. 5.328 quo tot facta virum totiens cecidere, Sidon. Epist. 4. 17.2 apud limitem ipsum Latina iura cecidenmt, and Claud. 1 5.44-5 q11o Latiae vires urbisqrte potestas I decidit? Heinsius favoured abiere, comparing Claud. c.m. '­

B O O K III

22.4 quo sensus abiere pii? and Luc. 7 ·7 5 quo tibi fervor abit aut quo .fiducia

Jati? 280 The lacuna at m.28o-360 is discussed above, Introduction, p. 43 , in relation to Birt' s theory ofancient recensions of the poem and the judgement of Martinus de Vicomercato, the scribe ofJ 4, about the authenticity of the omitted verses. Most MSS that omit 280 etc. attempt to bridge the gap between 279 and 361 by the insertion, generally after 279, of one or other of two patently spurious verses, viz. omnis honor recti vobis? sic Jata recedit and haec ait et lecturafaces altum nemus intrat. Both of these verses presuppose a knowledge of the omitted passage, the latter because of its content, the former because of its incorporation of the word honor, found also in 280. From this one may reasonably infer that whoever concocted these bridge­ verses-and their distribution among the vett. suggests that they began life in exemplars of Class (3-shared Martinus' view that m.28o etc. was a spurious passage, and designed omnis honor e�c. and haec ait etc. as alter­ natives to it. In the course of time, however, both the bridge-verses and III . 280 etc. came to be transmitted side by side in most MSS of Class y and not a few of Class a. 281 lam Veneri iunctae sociis raptoribus itis? ' Do you now go at Venus' side, with ravishers for your associates ? ' The plural raptoribus may refer only to Venus, in which case the hyperbole is easy enough, or it may include a reference to the unknown auriga of 237 supra. The sociative ablative sociis raptoribus, iunctae raptoribus, may be paralleled from Claud. 10.22o-1 quidquid ab innwneris socio Stilichone tremendus I quaesivit genitor bel/is, Sil. 9.526 ajfatur virgo soda Iunone parentem, and Stat. Theb. 1 . 1 3-14 wr non expaverit ingens I Ionium socio casura Palaemone mater, and there is no need for Baehrens' sociae raptoribus, nor for the insertion of et, as in the text Veneri et sociis which Heinsius adopted from Ugoletns' edition of 1493 . 282 Sitie11s may be followed either by ace. or gen. : for the former c£ Claud. 5.232, 17.25 1, D.R.P. II p£ 1 9 ; for the latter c£ Claud. 1 7.297, Sil. 3 . 578. In this case hominem sitientib11s should be read : the plurals hominum and homines look to be normalisations of hominem humanum genus, for which Burman compares Claud. D.R.P. III . 145 and Ov. Fast. 1.3 37. 294 Rursum here, as at 20.30, to avoid the unpleasant sound produced by final and initial s juxtaposed : cf. L. P. Wilkinson, Golden Latin Artistry, pp. 1 3-14 and refs. At 19.2 rursum obviates consecutive endings in -us. 295 Inque hrmriles delapsa preces : c£ Ov. Ex Pont. 4. 1 5 . 3 3 scribere saepe aliud cupiens delabor eodem, Sen. Epist. 68.10 ad Epicureas voces delaberis. Also possible perhaps is devecta, for which c£ Prop. 4. 1 . 1 19 nutrc ad tua devehar astra and Claud. Mamert. Anim. 2. 12 p. 144.10 nos ad id usque loci disputationis ordo devexit. Demissa was probably suggested by remittit in the previous line. 297 Like Heinsius, I am inclined to prefer the Isengrin's deiecta to the =

=

16

23 7

HCD

C OMMENTARY

M S miseranda, which looks t o b e a gloss borrowed from Stat. Ach. 1.50 supplex miseranda, perhaps at the suggestion of miseros earlier in the line. For deiecta Burman compares Claud. 26.376 non deiecta malis, mixta sed nobilis ira and Stat. Theb. 3 · 3 1 5 .res super Argoficas haud sic deiecta videbis. 305 Gesner, while printing nos . . .Joedere, declares his preference for the vulgate vos . . . munere. The idea of the goddesses being bribed to keep silence is, however, too crude and undignified to countenance, and nos . . . Joedere should be retained. The sense of this and the next few lines is : but if the ravisher has, by entering into some compact of silence with Diana and Minerva, prevented me from learning the truth from them, you at any rate, Latona, might tell me what has happened, for you may have heard some­ thing from your daughter and you know what it is like to be a mother. 3II-I2 Verse 3 12 was condemned as spurious by Barth and Heinsius, partly, no doubt, because they found it difficult to comprehend, but primarily because Claverius had said that it was absent from his 'vetus liber '. But however Claverius' note be interpreted-and it may well be the case that he mistook the conjectural deletion of 3 12 ( ? by Cujas) for the recording of transmitted testimony-there can, I think, be no doubt that the verse is genuine. Gesner, Birt and Koch divide Ceres' address (295 ££ ) into two parts (.295-3 I I agas ; 3 12-29) and take the words largis . . . madescunt as a paren­ thetical comment by the poet e propria persona. But quite apart from its _almost intolerable abruptness, what could be the point of such a parenthesis ? We have already been told that the goddesses responsa . . . matri f dant lacrimas (293-4), so it must seem improbable that the poet should again interrupt Ceres' utterance to reiterate this point. And yet largis . . . madescunt must refer to the goddesses, or more exactly to Latona, for the only mean­ ingful way to take 3 12, in which those words are picked up, is as the expression of Ceres' bewilderment at the contrary behaviour of Latona in weeping and at the same time concealing the reason for her tears. If this interpretation of 3 12 is correct, it must be clear from the complementary relationship between that verse and the preceding one that the words largis . . . madescunt do not in fact form a parenthesis, but are part of Ceres' speech. With the replacement of tunc by nunc (for the confusion c£ rr. 132 supra), perfect sense is now restored to 3 I I -I2 : ' Copious tears now wet your cheeks, Latona ; but what is it that so merits tears and silence ? ' 320 It is tempting to read aut with L 2 instead of the et of the other MSS (for the frequent confusion of these two particles c£ D.R.P. 1.98, Virg. Aen. 6.609 and Manil. 1.460, 48 1 ) but not absolutely necessary since there are occasions in Latin literature where the copulative particles et and -que seem to do duty for disjunctives : cf. ThLL v.2. 894·3 o ff. ; Leumann-Hofmann­ Szantyr, rr. 484 ; Lofstedt, Komm. pp. 200-1 ; and the notes of Kroll and 23 8

BOOK III

Fordyce o n Cat. 45.6. Conversely, disjunctives occasionally occur with cop­ ulative force : c£ ThLL II. I 575-72ff. ; Leumann-Hofinann-Szantyr, II. 500 ; and Li:ifstedt, Komm. pp. 199-200. 321 Although the Greek word 'Pmcxicx (sc. op11 ) should properly have as its Latin transliteration Rhipaea, it seems that the Roman poets had a decided preference for the form Riphaea : c£ Vahlen's note on Enn. Sat. lib. incert. VIII : 'Pmcxicx opT] quae Graeci dicunt, Roman is Riphaei montes sunt.' The consensus of Claudian's MSS, both here and at 3.242, 7. 149, 20.151, 21.124 and 28.31, is for Riphae-, and the same is true of the MSS of Virgil at Georg. 1.240 and 3 . 3 82, Servius concurring. There cannot therefore be any justification for the spelling Rhipaea, which editors have introduced here. 323 Fines = regionem squares better with penetrare and nivalem . . . domum than does .finem terminum ( ThLL knows only two cases where the singular finis = regio : Hyg. Astr. 1.8. p. 27.22 and 28.16 Bunte, and it is not likely that there should be a third case here). 327 Extincta, sc. luctu : c£ Stat. Theb. 10. 803 tu miseram ante omnes •

=

properasti extinguere matrem. 330-403 Ceres cuts down twin cypresses in the sacred grove of Jupiter and lights them from the fires of Etna. 332 Against the phrase jlumen Acin, offered by all the extant MSS, the following points may be urged : (a) although short at Anth. Lat. 1 5 1 .2 Riese, the a in Acis is regularly long in the classical poets (c£ Ov. Met. 13.750, Fast. 4.468, Sil. 14.221). In view of the fact that Claudian's prosody con­ sistently reflects the conventions of the classical poets, it seems unlikely that he could have regarded Acin as an iambus ; and (b) the apposition ofjlumen and a proper name, though common enough in prose writers (c£ Caes. B. C. I.J 8.4jlumen Dubis), seems to have been avoided by the poets, who either make use of expressions like casta . . .jlumine Ladon (Claud. 24.260), or put the proper name in the genitive (cf. Virg. Aen. 6.659 Eridani . . . amnis ; 7· 714 jlumen Himellae ) , or adopt the adjectival counterpart of the proper name, where one exists (c£ Hor. Carm. 4.4.3 8 Metaurum jlumen and Ars poet. 1 8jlumen Rhenum ; Priscian, G.L.K. II. 169-70, mentions also Histrum, Tanagrum, Iberum and Vulturnum). I therefore rejectjlumen Acin in favour of jlavum Acin, which Claverius gleaned from an otherwise unknown ' codex Saletani '. For the frequent use ofjlavus in describing rivers, c£ Claud. c.m. 27. 83 and ThLL, s.v. jlavus, VI. 888.20-3 8, supplemented by Shackleton Bailey, Propertiana, p. 8o ; and for the position of the elision c£ Claud. 18.382 and 28.624. 347 Coei corrupted to Caci perhaps via the intermediate stage Caei ( Cei) : c£ Virg. Georg. 1.279 where Heyne notes that some MSS have Caeum for

Coeum. 352

The Isengrin's lapsum, found also in four vett., is very attractive

23 9

I6·�

C OMMENTARY

(c £Julcire ruinam and similar expressions at Claud. 8.460, Ov. Trist. 1.6.5, Ex Pont. 2.3 .60), but most probably represents an attempt to correct the contextually absurd lapsam (lassus is corrupted to lapsus at Virg. Georg. 4.449, Aen. 2.739). For lass(lm Heinsius compares Claud. c.m. - 27.3 1-2 ceu lassa procellis / ardua Caucasio nutat de culmine pinus. 359-6o

No edition provides a satisfactory text of 3 59· Claverius notes :

'PERITURA Iouem. Sic in manuscripto. Id est itura per Iouem . . . ', and per itura promptly found its way into the editions ofJ. J. Scaliger, Heinsius, Gesner and Burman. But what does it mean ? Burman adduced the phrase

ire per hastes and compared Ov. Her. 4.3 8 est mihi per saevas impetus ire Jeras, but these supposed parallels are not in point, and per itura remains, as it must remain, indefensible. The more recent editors, Jeep, Birt and Koch, return to the lection Jeritura, which figures in a number of MSS, b ut this is o bjec­ tionable because Claudian is otherwise impeccable in his prosody. 1 Equally objectionable on prosodic grounds is petitura. Of the other variants available, nocitura is never used by Claudian, or indeed any classical writer, with ace. of the object, and comes most probably from v. 3 54 ;fractura, conjectured by Jeep but in fact occurring in R29, gives poor sense ; and laesura, though pretty, is surely derived from laedere in v. 3 5 5 · The easiest and best emenda­ tion so far suggested is J. J. Scaliger' s petit ira, which I print . In general the cedar is not ' less knotted ' (magis enodis) than the pine, and it seems likely therefore that Parrhasius is right in commenting : ' Magis enodes. ad sui generis arbores comparatio fit : non ad pinum : que fusior inquit est inter Cedros & minus nodosa.' There is no need to transpose pinus and cedros (c£ Stat. Ach. 2.121-2) or change magis to minus (c£ Lucr. 2.53 3 , Ov. ExPont. 3 .9.23, Manil. 4. 1 10), as I was once inclined to do. Translate : ' she hesitates whether to cut down pines or fell the less knotted kinds of cedar '. 370 Cupressi , -os invariably in the Virgil MSS, which I follow here and at 393 infra. 386 The unusual phrase animare ( flammare) taxos is amply supported by the following passages adduced by Heinsius : Varr. Menipp. Jrg. 292 Buecheler (from the Marcopolis, not the Bimarcus, as H. states) noctilucam tollo, adJocu mJero, inflo, anima reviviscit ; Hor. Carm. 3.21 .23 vivae . . . lucernae ; =

1 The i inferitura is indeed shortened at Maxim. Eleg. 5 -97, as Birt notes (he might also have cited Drac. De laud. Dei 3 . 106), but such shortenings are not, I think, to be admitted in Claudian. All examples of unusual prosody adduced by Birt (p. ccxi) involve proper names, and most of them are excusable on metrical grounds : Academia at 17.94, as at Sidon. Carm. 1 5 . 120 ; Gerjion at 3 .294 ; at c.m. 5 1 .6 we should probably read Syracosius with Heinsius to avoid the unusual Syracusius; and at c.m. 22.17 the rare lengthening of i in Darium can easily be avoided by reading Dareum, for which c£ Ov. Ibis 3 1 3 . Such proper names apart, Claudian rigorously follows classical practice.

B O OK III

Plaut. Aul. 9 3 s i ignis vivet (additional exx. i n Wagner's ed. ad loc. ) ; and Ov. Met. 7·77 extinctaque.Jiamma revixit (so H.'s Arondelianus ; Merkel and Ehwald read reluxit with the other MSS). C£ also Sidon. Epist. 2.2. 1 1 ignis animatus (cited by Barth) ; Ov. Fast. 3 -427-8 ; 4-553 ; 6.291 ; Rem. am. 732 ; Met. 1 3.606. Barth preferred the variant animarum, with ruit now transitive, one presumes, in the sense of eruit (so the gloss in P 2), but the phrase ad crimina animarum is decidedly odd and could hardly convey the meaning ' ad puniendas aut lustrandas sontes & crimine pollutas animas ', as Claverius imagined. 391 Plenos . . .Jiuctus : c£ Prop. 1 .20.43-4 Jlumina . . . / . . plena. 393 Adversa Jronte refers presumably to the cypresses, not to Ceres (cf. apices, v. 398). Aversafronte, ' with her head turned away ', of Ceres, makes sense but is not so appropriate here. The two participles are constantly confused : adversus with aversus at Virg. Aen. 5-477 and Stat. Theb. 5-357; aversus with adversus a t Virg. Aen. 9.412 and 12.464, for example. 395 Flammis here has all the marks of a simplificatory gloss, suggested perhaps by Virg. A en. 6.218 aena undantia Jlammis. The genitive Jlammarum may either be regarded as dependent on ut1dantem, here used like abundantem (c£ K. S. 1.467-8, where the following passages are cited : Virg. Eel. 2.20 nivei quam lactis abundans ; Manil. 2.6oo abundant cuncta furoris), or be taken with hiatum, on the lines of Paul. Nol. Carm. 32.170 inform em contextae noctis hiatum. 399 The conjecture stridentque, which Birt proposes in order to exclude exceptions to a ' rule ' of clause connexion (see his pre£, p. ccxx, 1.9), is quite superfluous, since Claud. 20.163, which he would also emend away, provides an exactly parallel case of asyndeton. 404-48 Lamenting the harshness offate, and with bitter reproaches against herself, Ceres leaves Etna and begins her search by the light of her torches. 434 Pro.ficere. A predominantly prose word only rarely used by the poets (cf. Axelson, Unpoetische Worter, pp. 63-4). Claudian has it elsewhere at 28.533 and c.m. 9.36. 438 Prima here may have the force of an adverb (so Birt in his Index Vocabulorum, s.v. primus) or be regarded as transferred from gressus (so Parrhasius ; primos is in fact given by G 1 man. rec. and r) ; a third possibility, perhaps, is that it be taken with Aetna in the sense ' the foothills ofEtna '-an unusual idiom, but perfectly intelligible. That prima . . . ab Aetna could = ' a summo Aetnae vertice ' (Parrhasius' alternative suggestion) seems t o me rather doubtful : it is more natural to talk of a mountain ' beginning ' from its foothill s than from its summit. 443-4 Gesner, Birt and Koch read and punctuate as follows : .

Omnibus admtlgit. quocumque it in aequore, Julvis Adnatat umbra Jretis . . . To my mind, this is a clumsy and unsatisfactory text. (a) Since there is nothing in the context to justify Gesner's interpretation of omnibus admugit

C OMMENTARY

' Omnibus sibi obviis vel Dis vel Semideis . . . admugit, cum gemitu eos rogat de filia sua ', we should have to understand orbitis (from 442) with omnibus, as seems to be implied by Birt and Koch's punctuating with a semi­ colon after jletu (442). Stylistically, however, this cannot begin to compare with the Heinsian text omnibus admugit . . . sulcis, which obviates the rather inconsequential association of quocumque it in aequore with the clause immediately following, and in which the anaphora of omnis . . . / om11ibus introduces an elegantly turned theme and variation. The fact that sulci has already appeared at 426 need cause no more difficulty than the iteration of aequore (443 and 446), since both words are used in different senses on each occasion.1 For admugit c£ Stat. Theb. r r.6or-2, where Oedipus iacet immugitque cruentis / vulneribus. (b) Ad11atat umbra Jretis, ' her shadow floats alongside/towards the sea', is tantamount to saying that it floats over dry land, which is decidedly odd, for the goddess is constantly envisaged as being of superhuman size (c£, e.g., m.376ff) and it is natural that she should overshadow the sea just as the light of her gigantic torches reaches as far as Libya. I therefore prefer innatat to adnatat, which neither Gesner nor Birt is successful in defending.2 The text of 442-4 may now be translated as follows : ' Every rut is made wet by her tears ; wherever she goes over the plain, she groans over every furrow (left by the chariot's wheels). Her shadow floats on the waves . . . ' 447-8 The chief problem posed by these verses is the relation between antra procul Scyllaca petit (sc. lucis imago) and the words that follow. Gesner advances the interpretation 'Pars Scyllaeorum antrorum (neque enim hie unum monstrum fingit, sed, ut res est, regionem maris scopulosam) prae stupore fl.ammarum silet, reductis prae metu latratoribus illis canibus &c. ' I cannot agree, for if Claudian is not thinking in purely mythological terms, why are Scylla's canes mentioned at all ? It is certainly true that all the words used in 448 can individually be applied to inanimate objects (for exterrita c£ Virg. Acn. 3 .672ff clamorcm . . . quo . . . f contrcmuere undae pcnitusque extcrrita tel/us / ltaliae ; for latrat cf. Sil. 3 -471 latrantibus undis ; for stupcfacta cf. Sen. H.f 763 stupent ubi undac ; and for silet c£ Claud. D.R.P. r.84-5 tremcfacta silent. . . / atria), but their accumulation here suggests that they refer not to Scylla's cave but to her hounds. Canibus rcductis, as Birt rightly notes (pracf. p. ccxxiii), does duty here for as

1 Such repetitions are not uncommon in Latin poetry : cf. Shackleton Bailey, Propertiana, p. 9. At Ov. Met. 3 . 104 and I 07 sulcum, -is differ in number and case, but not in meaning. z The former's ' imago facium . . . propagatur ad litora ' and the latter ' s umbra Cereris in aequore quasi deae comes cum ea natat vel ei adnatat are both alike unjustified by the Latin as explanations of the text printed. Platnauer reads adnatat but seems to translate innatat.

B O O K III

the partitive genitive canum reductorum. Such substitution o f the ablative for the genitive is not uncommon in Claudian : c£ 2 1 . 3 52 non Jusum crebris hastilibus imbrem, where crebrorum hastilium could not be accommodated to the metre ; also 1 5 .416 praeciprws electa pube maniplos and 21.70 urgebat patrias suspenso principe curas, where the choice of expression dictates the syntax. See further K. S. 1.787.

24 3

BIBLIO GRAPHY

This list o f titles i s not exhaustive, but does include everything of value for the student of the D.R.P. Articles dealing with single passages are generally omitted ; standard works of reference are not mentioned at all.

De Raptu Pr�serpinae, editions ( I ) Separate J. Parrhasius. Milan, 1 500, second edition Milan, 1 505. L. Jeep. Turin, 1874. V. Paladini. Rome, 1952. (2) With Claudianus maior B. Celsanus. Vicenza, 1482. J. Camers. Vienna, 15 10. A. Francinus. Florence (Juntine), 1 5 19. F. Asulanus. Venice (Aldine), 1 523. M. Bentinus. Basle (Isengrin), 1 5 3 4. T. Pulmannus. Antwerp, 1 571. S. Claverius. Paris, r6o2. J. J. Scaliger. Leiden, 1603 . T. Dempster. La Fleche, 16o7. C. von Barth. Hannover, 1612, second edition Frankfurt, 1650. N. Heinsius. Leiden, 1650, second edition Amsterdam, 1 665. G. Pyrrho. Paris (in usum Delphini), 1677. J. M. Gesner. Leipzig, 1759. P. Burman II. Amsterdam, 1760. L. Jeep. Leipzig, 1 876--9 . T. Birt. Berlin, 1 892. J. Koch. Leipzig (Teubner), 1 893 . M. Platnauer. London and Cambridge, Mass. (Loeb series), 1922.

General Bibliography F/eckeis. Anna/. VIU (1 872), 634-{). idem, ]enaer Literaturzeitung 1 875, pp. 1 3 1-2. idem, Bursians ]ahresberichte for 1 877, p. 225. B A E H R E N S, F. , �aestiones Claudianeae, Munster, 1 8 85. B A E H R E N S, E . ,

B E R N E R T,

E.,

'Die �ellen Claudians in " De Raptu Proserpinae " ',

Philologus xcm ( 1 93 8-9 ) , 3 52-76. B O N N E T, M., Revue critique d'histoire et de littlrature, no. 27 (1 875), 5-9. C E R R A T O,

L. , ' De Claudii Claudiani fontibus in poemate De Raptu Proserpinae ', Riv. di.fil. IX (r88r), 273-395.

244

BIBLIO GRAPHY

C L A R K E, A . K.,

' Claudian's methods o f borrowing in D e Raptu Proser­ n.s. I (r95o-r ), 4-7. C R E M O NA, v., ' La composizione del " De Raptu Proserpinae " di Claudio Claudiano ', Aevum I948, pp. 23 1-56. D E LR I O, M. A., Notae in Claudianum, Antwerp, 1572. D I LKE, o. A. w. , ' Patterns of borrowing in Claudian's " De Raptu Proser­ pinae '", Revue beige XLIII (I965 ), no. I, 6o-r . E A T O N, A. M., The influence of Ovid in Claudian, Washington, I943 · F A B B R I, P., ' Claudiano in Sicilia e il ratto di Proserpina ', Raccolta di Scritti in Onore di Felice Ramorino, Milan, I927, pp. 9I-IOO. F A R G UE S, P., Claudien : Etudes sur sa Poesie et son Temps, Paris, I93 3 · F O E R S TER, R . , Der Raub und die Ruckkehr der Persephone, Stuttgart, I 874· G O E T Z, G. , Philofog. Anzeiger VII (1 875-6), I4I�. G R A M LE W I CZ, s . , O!!aestiones Claudianeae, Breslau, I877. G U N T H E R, c., De Claudii Claudiani comparationibus, Regensburg, I 894. G U S T A F S S O N, F W., Rh.M. XXXIII (I 878), 480-1 . H A U P T, M. , Hermes V (I871), 326. H A V E RFIE LD, F., ' Scholia on Claudian ', J.P. XVII, 27I-3 . H E R T E L, T., De nonnullis Claudiani locis, Torgau, I 848. J E E P, L. , Acta Societatis Philologae Lipsiensis I ii (I87I), 347ff. P. C.Ph. S.

pinae',

.

.

K O C H, J., D e codicibus Cuiacianis quibus i n edendo Claudiano Claverius usus est,

Marburg, I 8 89. ' Beitrage zur Geschichte romischer Dichter im Mittelalter 2. Claudianus ', Philologus XLIX (r 89o), 5 54££ M U E L L E R, L. , Jenaer Literaturzeitung nr. I 8 (1877), 284-5. M A N I T I U S, M.,

M U E L L N E R,

c.,

De imaginibus similitudinibusque, quae in Claudiani carminibus

inveniuntur,

Vienna, 1 893. ' De latinitate poetae Claudiani observationes ', Rh.M. xxxv ( 1 8 8o), 586-6o6. P A U L, T. G., O!!aestiones Claudianeae, Glogau, 1 8 57, and Berlin, 1 866. P O S T G A T E, J. P. , C.R. IX (1 895), 162--9. R O M A N O, D . , Claudiano, Palermo, 1958. S C A LI G E R, J. c., Poetices libri septem, Lugduni, 1 561, pp. 32I-3 . V I T E L LI, G., Riv. diji/. I (I 873 ), 330-4. V O L L M E R, F., the article ' Claudianus ' in R.E. V O L L R A T H, o. , De metonymiae in ClaudiiClaudiani carminibus usu, Weida, 1910. P A U C KER,

c.,

W A L C H, B. G., Uberioris commentationis de Claudiani carmine De Raptu Proserpinae inscripto specimen,

Gottingen, I 770. diss. Breslau, 1908.

WELZEL, A., De Claudiani et Coripp i sermone ep ico,

Z I M M E R M A N N, A. D e Proserpinae raptu et reditu Jabulas varias inter s e com­ paravit,

Lingen, 1 8 82. 24 5

IND I C E S INDEX N O M I N U M

Numeri versus denotant. Adhibentur haec compendia : a. = accusativus ; ab. = ablativus ; ad. = adiectivum ; c. = coniectura ; d. = dativus ; £ = femininum ; g. = genitivus ; m. = masculinum ; n. = netitrum ; nom. = nominativus ; p. = pluralis ; s. = singularis ; sub. = substantivum ; v. = vocativus ; v.l. = varia lectio. Achaemenius

Achaemenio m.264

(d.m.) Acheloides m.254 Acheron 1.87, 227 Acheronteus Acheronteos II.3 5 1 Acis Acin m.3 3 2 Actaeus Actaea m.54 (a.p.) Aegaeon Aegaeon 1.46 ; Aegaeonis Achelois

Acheron

m.345

Aegaeas 1 praif. 12 Aeolos (-us v.l.) 1. 74 Aethon Aethon 1.284 (v.l.) Aetna Aetna 1. 1 5 3 , 1 54, 190, II. 8, 72, 289, m.r 86, 399 (nom.) ; Aetuae m.85, 3 3 0 (g.) ; Aetna m.43 8 (ab.) Aetnaeus Aetnaeos r.r6o ; Aetnaeae 1.122 (d.s.) ; Aetnaea m. 22 0 , 3 3 4 (a.p.) Alastor Alastor 1.286 Alcides Alcides II praif. 9 Alcyoneus Alcyoneus m. 1 8 5 Allecto Allecto 1.280 Alpheus Alpheus II.61 Alpinus Alpina II. I76 (nom.s.) Amasenus Amasene II. 57 (c.) Amazonis Amazonidum II.62 Amazonius Amazonios II praif. 3 7 Amnis Amnes m. r6 (nom.) Amor Amor 1.27 Amphitrite Amphitrite 1. 104 Amsanctus Amsanctus II.3 50 Amyclae Amyclas 1. 1 3 5 ; Amyclis II. I 3 3 (ab.) Anapis Anapis m.3 73 (v.l.) Antaeus Antaeo II praif. 41 (d.) Aegaeus Aeolos

Apollinei m.373 (g.n. ) ; Apollineo m.3 1 0 (ab.m.) Arctos Arctos n. 1 89 ; Arcton n.63 II.6o Arethusaei Arethusaeus (nom.p.) Argi Argis II praif. 9 (ab.) Assyrius Assyrii II.96 (g.n.) Atlas Atlas II praif. 45 (v.l.}, II.191, m.324 ; Atlantis 1. 89 Atropos Atropos 1.2 1 8 Auster Auster II.308 ; Austri n.200 Avernus Averni 1 . 20 , II.290 (v.l.}, 348 ; Averno r. n6 (d. vel ab.) ; Avernis II.290 (c.ab.m.ad.) Apollineus

Bacchus Bistonius

Baccho n. 6 7 (d.) Bistoniam (-urn v.l.a.n.)

II praif. 8

Booten n. 190, m.225 Boreas 1.70 ; Boreae m.323

Bootes Boreas

(g.) Briareia m. r88 (nom.f.) Busiride II praif. 43

Briareius Busiris

Caci II praif. 43, m.347 (v.l. ) Cadmi m.387 Camerina Camerina (Camarina v.l.) II. 59 Caspius Caspia m. 105 (nom.f.) Cecropis Cecropidum I. I I (v.l.) Cecropius Cecropium 1. 1 1 (nom.n.) Cephisos Cephisos (-us v.l.) II. I J 6 Cerberus Cerberus 1.86 (v.l.) Ceres Ceres 1.138, 209, m. 179, 260, 357 (nom.) ; Ceres m.105 (v.) ; Cererem 1. 1 07, m.48, 67 ; Cereris Cacus

Cadmus

INDEX NOMINUM

(cont.) I.22I, 237, II.36, ID.329 ; Cereri I. I22, n.372, m.ss, I73 Chaonius Chaonio (-a v.l. ab.) m.47 (ab.m.) Chaos Chaos I.28, n. I 3 , 196 (a.) Cirrhaeus Cirrhaeas (-i v.l. g.m. ) n praej. 23 Clarius Clarios I. 1 3 6 Cocytos Cocytos (-us v.l.) 1.87 ; Cocyton n.3 5 3 ; Cocyti I.28 1 Coeus Coei m.347 ( ? c.) Corybas Corybas 1.210 Crinisus Crinise n.s7 Cthonius Cthonius I.284 (c.) Curetes Curetum n.270 Cyane Cyane n.6I , m. r9o ; Cyanen m.246 Cybebe Cybebe I.2 12 (c.), m.271 (c.) ; Cybeben m. n 3 (c.) Cybele Cybele 1.212, m. 1 34, 271 ; Cybelen 1. r 8 r , m.n3 Cyclopias Cyclopia 1.97 (a.p.) Cyclops Cyclops II. 175, m.3 5 5 ; Cyclopum I.238, n.250, m. n7 Cyllenius Cyllenius I.77 Cynthus Cyntho n.245 (d. ) Cytherea Cytherea 1.2 r6, n.u9, m.2o8, 274 Ceres

(sub.) Delia n.2o6 (nom.) ; Delia m.285 (v.) Delius (sub.) Delius n. r36 Delos Delos II.3 4 ; Delon 1. 136 Delphicus Delphica n.246 (nom.p.) Diana Diana m.306 (nom.) ; Dianae m.2r6 (g.) Dictaeus Dictaeas n praej. 3 3 Dindyma Dindyma II.269 (a.) Diomedeus Diomedeos n praej. 12 Dionaeus Dionaea n. s (ab.) Dione Dione m.433 Dirae Dirae n.2 r 8 Dis Ditem 1.26 ; Ditis 1.227, 266, 286, n.365 ; Dite II. I 3 , 160 Dodonius Dodonia 1.3 1 (nom.£)

Delia

Dryades m.78 ; Dryadum 1II.271 , 3 8 1

Dryades

Edoni n praej. 7 (c.nom.p.) Electram m.171 Eleusin Eleusin (-is v .1.) I. I I Elysius Elysium (-os v.l.) n.284 (a.m. ) ; Elysiae n.323 (nom.f.) Enceladus Enceladum n. r s 8, m. 1 8 7 ; Enceladi I. I 5 5 , m.123, 3 50 Erasinus Erasine n.57 (v.l.) Erebus Erebi 1.32, 281, II.259, 330 Erinys Erinys 1.226 Erymantheus Erymanthei n praej. 36 (g.m.) Etruscus Etruscum m.445 (nom.n.) Eumenides Eumenides n.216, 344 Eurus Euris II.30 (v.l. d.) Edonus

Electra

Fauni m.17 ; Faunorum m.3 8 1 (sub.) Florentine II praej. so Fluvius Fluvios m. s ; Fluviis m.14 (d.) Furiae Furiae 1.39, II.2 1 9 ; Furias m. 79

Fauni

Florentinus

Gaetula I. 1 50 (nom.£) Galatea m.3 3 3 Galli Galli s n.269 (ab.) Gargara Gargara I.2o8 (nom.) Gelas Gelan (-am v.l.) n.s8 Getae Getas II.6s Geticus Getica 1.71 (ab.) Giganteu s Giganteos r. I 54 Gigas Gigas II. 1 5 9 ; Gigantum m. 196, 339 Glaucus Glaucum m.r 2 Gorgo Gorgonos (-is v.l. ) n.26, 205 Gorgoneus Gorgoneis I!.225 (ab.f.) Gortynius Gortynia n.3 3 (nom.£) Gaetulus Galatea

Haemus n.68 (v.l.) ; Haemo II praej. 21 (ab.) Hebrus Hebrus II praej. r 8, II.68 (v.l.) Haemus

24 7

INDEX N O MINUM

Hecate 1.1 5 Helicon II. 134 Henna Henna (nom.) 1. 1 54 (v.l.), II.8 (v.l.), 72 (v.l.), 289 (v.l.), m. 186 (v.l.) , 3 99 (v.l.) ; Hennae m.85 (v.l. g.) Hennaeus Hennaeae 1. 122 (v.l. d.) ; Hennaeos 1. 160 (v.l.) ; Hennaea m. 220 (v.l. a.n.) Hercules Herculis II praef. 30 Herculeus Herculea II praef. 47 (ab.£) Hermus Hermus II.68 Hesperus Hesperus II.361 Hiberus Hiberae m.3 19 (g. ) Hippolyte Hippolyte II.64 Hyacinthus Hyacinthe II. 1 3 1 Hybla Hybla II.79 (nom.) Hyblaeus Hyblaeum II. I25 (a.n.) Hydaspes Hydaspes II.82, m.325 Hydra Hydrae II praef. 41 (d. ) Hyperionius Hyperionio II.44 (ab.n.) Hyrcanus Hyrcana m.263 (ab.) Hecate

Helicon

Lachesis 1.54, II.3 54, m.4I I Latiis II.177 (ab.n.) Latoius Latoia 1.106 (c.a.) Latona Latona 1.1 3 7 (nom.) ; Latona m.3o6 (v.) Latonius Latonia II.23 3 (f.sub.) ; Latonia 1. 106 (a.ad.) Lemnius Lemnia m.275 (a.) Lethaeus Lethaeo II.305 (ab.m.) Lethe Lethes 1.282, II.2 I 8 ; Lethae II. 2 1 8 (c.d.) Libya Libyam m.445 ; Libyae II praej. 45 (v.l. g.) Libycus Libyci II praef 45 (nom.) Lilybaeus Lilybaea 1. 1 5 0 (a.) Lipare Lipare II.174 Lucifer Lucifer II.121 Lucina Lucina 1. 123, m.307 (nom.) Luna Luna m.403 (nom.) ; Lunam II.45 Lyaeus Lyaeo II.3 5 3 (ab.) Lycaeus (sub.) Lycaei n. 1 8 (g.) Lydius Lydia 1.275 (nom.f.) Lachesis

Latins

Maenala n.244 (nom.) ; Maenala 1.230 (a.) Maeonius Maeonius 1. 1 9 ; Maeoniae n.68 (nom.) ; Maeoniis 1.19 (v.l. ab.m.) Maia Maia 1.76 (ab.) Manes Manes II.328, m.3 89 (nom.) ; Manes 1.41 , 267, n.14 (a.) Mars Mars 1.1 34, 1 3 5 Martins Martius II.147 Massylus Massylam II praef 28 Megaera Megaera m.3 87 (nom. ) Mimas Mimantis m.347 Minerva Minerva m.218 (ab.) Minoius Minoia II.3 32 (nom.£ ) Molossus Molossi II praef. 2 5 (nom.m.p.sub.) Mulciber Mulciber II.I75, m.397 Musa Musa 1.4 (c.nom.) ; Musarum II praef 5 1 Mycenae Mycenis m.3 8 8 (ab. vel d.)

Maenala

Iacchus 1.16 Ida Idam I.20 I (v.l.), m. r4o (v.l.) ; Idae II.267, ID.371 (g.) Idaeus Idaeos m.49 Idalius Idalia rr. I 6 (ab.) Ide Ide I.207 ; Iden I.201, m. 140 lnachius Inachiis II praef. 9 (ab.m.) lnarirne, Inarimen m. I84 Ionius Ionium I praef. 1 2 (a.n.sub.) ; Ionias I praef 12 (v.l.), I. I49 ; Ionios II. I Italia Italiam m.445 ; Italiae 1. 1 43 (g.) Iuno Juno I. Io6, 1 3 6, rr.367, m.327 ; Iunonis 1.3 ; Iunone m.418 Iunonius Iunonius II.97 ; Iunonia ibid. (v.l. £nom.) Iuppiter Iuppiter 1.2 1 5 , II.228, m. I , 327 ; Iovern I .67, m.36, 3 59 ; lovi I.93 , II.ro8, m.174 Ixion Ixion II.3 3 7 ; Ixiona II.3 3 5 Iacchus

INDEX N O MINUM Mygdonius

Mygdonio II.268 (ab.

m.)

Parthenius

Naides II. 56, m.1 7 Narcissum II. 1 3 2 Natura Natura 1.250, II.371, m.3 3 (nom.) Neptunus Neptunus II.I 8 I ; Neptunum r.w4 Nereius Nereia I.IOJ (nom.f.) Nereus Nereus 1.144, m. I I Nessus Nessum II praif. 42 (v.l. ) Nilus Nilus II praif. 43 Niphates Niphates m.263 Notus Noti m.323 (g.) ; Noto I praif. 8, II.248 (d.) ; Notos 1.92 ; Notis 1.242 (ab.) Nox Nox II.363 Nycteus Nycteus 1.28 5 Nymphae Nymphae II praif. 3 , II.67, 204, m.230 ; Nymphas m.4, 1 72 ; Nympharum II.76 Nais

Narcissus

Oceanus

Oceanum 1.270 ; Oceani

m.I72

Olympum II.257 ; Olympo u. r 8 3 , m. r 8 , 269 (ab.) Ophion Ophion m.348 Orion Orion II.I9I Orontes Orontes m.373 Orpheus Orpheus II praif. I (sub.) Orpheus Orpheis II praif. 24 (ab.f.) Orphnaeus Orphnaeus 1.284 Ossa Ossa II praif. 20, II.257 (nom.) Ossaeus Ossaeus II. I 8 3 Olympus

Pachyni 1.148 Pado II.178 (ab.) Pallas Pallas 1.230, II.39 , 206, 21 5 ; Pallada m.209 Panchaia Panchaia II.8r Pandionius Pandionias II. 19 Pantagias Pantagian (-am v.l. ) II. 5 8 Paphos Paphi II. 1 5 5 Parcae Parcae 1.48, II.6 ; Parcas 1!.305 Pachynos

Padus

Parrhasii rr. r 8 (g.m.) Parthenium rr.148 (a.m.) ; Parthenios ibid. (v.l. ) ; Partheniis rr.241 (d.n.) Parthicus Parthica 1.17 (nom.£) ; Parthica rr.94 (nom.n. ) Parthus Parthi rr.2oo (g.) Pelorus Pelorum 1. 1 5 2 ; Pelori m.2 5 5 Peneus Peneo II. r 8o (ab.sub.) Pergus Pergum II. I I2 Persephone Persephone rrr.244 Phaethon Phaethon m.403 Phlegethon Phlegethon 1.24, u.3 I S ; Phlegethontis m.390 Phlegethonteus Phlegethonteae I. 88 (g.) Phlegra Phlegra II.255 (nom.) ; Phlegra m.20I (ab.) Phlegraeus Phlegraeis m.3 3 7 (ab.f.) Phoebe Phoebe II.3 9 ; Phoeben rrr. 209 Phoebus Phoebus 1.134, I 3 5 , II praif. 48 ; Phoebum 1.6 ; Phoebi II. 28 (bis) Pholoe Pholoe II praif. 44 Phorcus Phorci rrr . I I Phrygius Phrygia m. I I4 (ab.) ; Phrygias rrr . 103 ; Phrygios 1.180, rrr .424 ; Phrygiis II.267 (ab.f.) Pluto Pluto 1.278 Poenus Poenorum rrr. r66 (m.ad.) Polyphemus Polyphemus m. 356 Proserpina Proserpina 1.27, 126, 217, 246, II. 5, 204, 247, m.7o, 83 , Ioo, 284 (nom.) ; Proserpina rr.277, m.407 (v. ) Proteus Protea m. 1 3 Pyragmon Pyragmon 1.240 Python Pythona n.22 (v.l.)

Parrhasius

Rheni m.32 1 Rhodope n praif. I9 ; Rhodopen 1. 1 3 5 Rhoetus Rhoeti m.347 (c. ) Riphaeus Riphaea m.321 (nom.n.)

Rhenus

Rhodope

249

INDEX N O MINUM

Sabaeis II.83 (ab.) Saturnia m.2o (a.n.) Saturnus Saturni I. I I4, II. I68, 28o ; Saturno m.272 (d.) Scyllaeus Scyllaea m.447 (a.n.) Scythia Scythiae m.282 (g.) Sicania Sicaniam II. I6o, III . 140 Sicanus Sicani II. I I2 (nom.sub.) ; Sicanos 1.220 (ad.) Siculus Siculi m.255 (g.m. vel n.) ; Siculas 1. 14I, 278 (v.l.) ; Siculae II. 173 (nom.p.) ; Siculis III . 84 (d.n.) ; Siculis 1.139 (d.£) Sidonius Sidonia 1.275 (ab . n.) Signifer Signifer 1.102 Simois Simois m.372 Sirenes Sirenes m.205 ; Sirenas III . 190 Sol Solem II.44 Steropes Steropes 1.241 Stygius Stygii 1.285 (g.n.) ; Stygii II praef. 34 (g.m.) ; Stygio II.264 (d.m.) ; Stygios 1. 120 Stymphalis Stymphalidas II praej. 37 (sub.) Styx Styx 1.22 Syrtis Syrtis m.p2 (nom.s.) ; Syrtes m.446 (nom.p.) Sabaei

Saturnius

Taenario 1.2 (ab.m.) Taenara II.J07 (v.l. a. ) Tanais Tanain (-im v.l.) n.66 Tantaleus Tantaleis n.3 3 6 (d.n.) Tantalus Tantalus II.337 Tartara Tartara n.3 I I (v.l.), 334, m.64, 390 (nom.) ; Tartara I. I I4, n.307 (a.) Tartareus Tartareo 1.217 (d.m.) ; Tartareo 1.2 (v.l. ab.m.) ; Tartarea m.79 (ab.) ; Tartareis II.2I 7 (ab.f.) Taygetus Taygeti II.244 Tegeaeus Tegeaee 1.89 Tethys Tethys II praej. 45, II.46 ; Tethyos m.320 Thaumantis Thaumantida III . I Thernis Themis 1.219 ; Themin 1.107 Taenarius Taenaron

Thermodontiaca II.66 (ab.) Thessalia Thessaliam n.179 Thetis Thetis I. I S O ; Thetin 1. 107 (v.l.) Thracius Thracius II praej. 49 ; Thracia II praej. IO (a.n.) Tbybris Thybri n.178 Thyesteus Thyesteis m.3 88 (ab. vel d.£) Tirynthius Tirynthius II praej. 49 (sub.) Tisiphone Tisiphone (Tes- v.l.) 1.40 Titan Titana n.49 ; Titanes m. I 82 ; Titanibus 1.66 (d.) Titanius Titania 1.44 (nom.£) Tityos Tityos II.3 3 8 Tonans Tonantem 1.3 8 ; Tonantis II. 76, 367, m. 134 (v. l.) ; Tonanti III. 1 3 4 ; Tonante m. 1 8 3 Trinacria Trinacria 1. 142, 1 9 1 , II. 1 86, III . 1 1 9 (nom.) Trinacris Trinacrida m.28 8 (v.l. sub.) Trinacrius Trinacrio 1.2 (v.l. ab.m.) ; Trinacria m.288 (a.n. ) Triones Triones 1.102 (nom.) Triptolemus Triptolemi 1 . 1 2 Tritonia Tritonia n.21 (nom. ) ; Tritonia m.286 (v.) Trivia Triviae II.27 (d. vel g.) Typhoeus Typhoea m. 1 8 3 (nom.f.) Typhon Typhona II.22 Tyrrhenus Tyrrhena 1. 1 52 (nom. £) ; Tyrrhena m. 1 8 5 (a.n.) Thermodontiacus

Venus 1.229, II.I2, m.220 ; Venerem m.43 3 ; Veneris II.266 ; Veneri 1.2 1 5 , 226, m.2 8 1 Vesevus Vesevi m. 1 84 Vulcanius Vulcanius 1.172

Venus

Zancli m.347 (v.l.) Zephyrum ii.73 ; Zephyro m.266 (ab.) ; Zephyros 1.1 86, m.3 (v.l.) ; Zephyris m.3 (d.) ; Zephyris II.288 (ab.)

Zanclus

Zepbyrus

250

G E N E RAL I N DEX THE NUMBERS REFER T O P A G E S

ablative used instead of genitive, 242 f. Acis, prosody of, 239 Alcuin, 67 Aldine edition, 77 n. 3 ancient recensions, theory of, 42 ff. animare, 240 f. archetype, arguments for existence of, 49 ff. asyndeton, 24I Baehrens, A., 9I Barth, C. von, 86 Bartholinianus, codex, 8 8 n. 2 Baudri de Bourgueil, 67 Bentinus, M., 77 £ Bernert, E., 107 £ Birt, Th. , 42 ff., 9I ff., 97 ff. book I, preface to, IOO £ book II, preface to, 94 ff. book III, Jeep's proposed division of, so Burman secundus, 90 buxus, declension of, 23 I

Camerina, spelling of, 2 I 5 Camers, edition by, 77 n. 3 and n. 6 carm. mitl. xxvii, 2I6 £ carm. min. xli, I O I £ Cerrato, L., I07 f. Charlemagne, court library of, 67 Chaucer, 74 Clarke, A. K., I09 Claudianus maior, definition of term, 4 n. I Claverius, S., 82 ff. contamination, 47, 57 ff., 6I ff. corruption of -ens to -it (-et), 2 3 5 Cremona, V., 96, I02 Crinisus, spelling of, 2 I 5

Cujas, ]., 82, 83 n. I , 8 5 Cybele, bacchiac scansion of, 2o6 £ Delrio, M. A., 82 detergere, conjugation of, 225 ' deteriores ', 47 f. Dilke, 0. A. W., 109 £ discedere, with dative, 209 disrumpere, I98 Dracontius, 66 D.R.P. : humanist lectures on, 75 ; language of, uo and n. 3 ; medieval commentaries on, 7I f. ; metre of, I I O and n. 4 ducor, with supine, 224

Eleusin, spelling of, I 93 elimination of MSS, 6I £ ellipse of object, 202 f. Etna and Henna, confusion between, 200 £ Excerpta Gudiana, 82 Excerpta Gyraldina , 82 n. I , 8 8 n. 2 Excerpta Laeti, 88 n. 2 Excerpta Schottiana, 88 n. 2 Fargues, P., 94, I07 Florentinus, 95 ff. Florilegium Gallicum, 72 Florus of Lyon, 67 Foerster, R., I06 f. Gaufridus of Vitry, I 7, 7I £ Gesner, ]. M., 90 Gramlewicz, S., I08 Greek proper names, declension of, I99, 2 I4, 2 ! 5 , 220 Guglielmo da Pastrengo, 74 Gyraldus, 82

GENERAL INDEX

Heinsius, N., 8 7 ff. , 94 £ Henna and Etna, confusion between, zoo £ Henri d' Andeli, 72 £ hiatus, 233

prosody, I IO and n. 4, 239, 240 and n. proximus alter, 232 Pulmannus, 82, 87 Pyragmon and Pyracmon, 207 £

incoquere, of smelting, 208 interpolations, alleged, 49 f. , 94 ff., 205, 226, 238 interstrepere, 23 I and n. Isengrin edition, 77 ff.

querella, spelling of, 228 £

Jeep, L., 33 ff., 89, 90 £, 94 ff. Junius, MSS of, 88 n. 2 Juntine edition, 77 n. 3 Koch, J., 82 n. 4, 83 n. I, 92, 99 lacunae, alleged, 49, I96, I98, 2IO, 222 liber Catonianus, 69 ff., 2I9 Livineius, 82

madere, 208 manuscripts : classifications of, 34 ff., 42 ff. , 54 ff. ; eli mination of, 6I £ ; interrelationships of, within Class a, 56 n. 2, within Class !3, 57 n. I , within Class y , 5 9 n . 3 ; the group D J 4 K I , 6o £ , 76 ; F I , 3 5 , 37 ff. ; L 1o, 71 Martinus de Vicomercato, 44, 237 metre, I IO and n. 4, I97, 2 I4, 223 , 23 1 , 239 moods, confusion of, 1 94

=

rabidus and rapidus, 224 recentiores : definition of the term, 6o and n. I ; relationship to vetustiores, 6I f. ; contribution to the text, 6I f. ; citation of, in app. crit., I I2 f. repetition of words, 242 and n. I revocare, 225 £ Ripa, Lucha, 75 Riphaea, spelling of, 239 Romano, D., 94, I 02 Scaliger, J. C., 205 Scaliger, J. J., 86 Schrader, J., 90 n. 4 Sidonius, 66 spectare expectare, 2 I I spurious verses, 202 £ , 2 I 9 £, 237 Stilicho, 95 ff. stridere, conjugation of, I93 subtegmen and subtemen, 210 =

Tisiphone, spelling of, I94 ff. title of poem, 1 87 f. Typhoeus and Typhoius, 233 transmission of D.R.P. with Claudianus maior, alleged, 53 £ unanimus and unanimis, 227

orthography, I I 3 f., I 89 £, I93, I94 ff., I 99 . 204, 207 £, 2IO , 2I4, 2 I 5 , 22� 22� 228, 23 3 . 239 Paladini, V., 92 £ Parrhasius, 77 perfect, syncopated forms of, 230 Petrarch, 74 Platnauer, M., 92 Postgate, J. P., 102 premit corrupted to .fremit, 236 prolabi confused with praelabi, 22 I

vetustiores : definition of the term 6o and n. I ; relationship to recen­ tiores, 6I £ ; citation of, in app . crit., I I2 £ Villani, Filippo, 75 viridis, meaning of, 236 vota, in a concrete sense, 227 ,

Walch, B. G., 90 n. 4 Zimmermann, A., 107