Skip to main content

Full text of "The Journal of philology"

See other formats 


ἡ Ὡς 

at ΠΝ ἷ 

ΠΡ share 

"Ὁ ὶ 
4 


FLY R baeh 
a ΠῚ 


ae 


STE 
ἧς asta 
ῃ 


rit 
ἐδ 
τυ 


ΡΝ 
Pana It APA ᾽ 
Sa ae 


yaa 
Au ntf 
Ses 


yy 


Bly 
ee 


Wyn he 
et Soy, 
eet er 

nf 


pte 
on Feet 
Heit 

te, 


es 


es Mi 
ig 


asthe 
Rena 


a eS 


= 
oe 


aa eye 
Bess 


oe 
ΤῊΝ 
ar, 


gh ds 
ay 
© 


p 


ὶ 


ee te ἐδ ΤΟ Ὺ, 
Le Pr cng ne Nem ig 


- bes el 
=~ ES A Selle ot 
Le 


4 
Ἰ 


Spree 
i ΕΝ 


t 
ΙΑ ἤν τὰ 


are 
ΠΕ 


Fgh 


Che Journal 


OF 


PHILOLOGY. 


EDITED BY 


W. ALDIS WRIGHT, M.A. 
INGRAM BYWATER,~MA, 
AND 


HENRY JACKSON; Luirt. D, 
VOL. XVI. 


London and Cambridee: 
MACMILLAN AND CO. 
DEIGHTON, BELL AND CO. CAMBRIDGE. 


1888 


CONTENTS. 


No. XXXI. 


Emendationes Propertianae. A. 2. Housman 
Miscellanea Critica. A. Palmer ; 

Life and Poems of Juvenal. H. Nettleship 

Notes in Latin Lexicography. H. Nettleship. 

The Title of the Second Book of Nonius. H. Nettleship 


On the Hebrew Root Y¥P and the Word YiSprd. W. Robertson 


Smith . ; ‘ F “ : ‘ ξ 
On the Fragments of the Fadia See Poem contained in the 
Herculanean Papyri. R. Ellis . 
Kin and Custom. F. B. Jevons 
Pausan. VIII 16 ὃ 5. John Εἰ. B. Mayor 
Seneca ep. 19 ὃ 3. John E. B. Mayor 
Ovid Met. ΤΥ 139—141. John E. B. Mayor : 
A Lost Edition of Sophocles’ Philoctetes.. John Masson 
Lucretiana. J. P. Postgate 
The Pugio Fidei. 8. M. Schiller-Szinessy 
A Roman MS. of the Culex. Robinson Ellis . 
Aristarchos’ Reading and Interpretation of Iliad N 358—9,  _W. “Leaf 
Servius on Aeneid IX 289. H. Nettleship 


112 


114 
124 
131 
153 
157 
160 


iv CONTENTS. 


No. XXXII 


Noniana Quaedam. J. H. Onions 

Notes on Propertius. F. A. Paley 

Adversaria. H. Nettleship 

Lexicographical Notes. F. Haverfield / 
The Numasios Inscription. H. D. Darbishire 
Aeschyleae Hugh Macnaghten easy: 
On the Date of Calpurnius Siculus. R. Garnett 
Notes on Juvenal. John E. B. Mayor 

Notes on Martial, Book III. John E. B. Mayor 
The Agamemnon of Aeschylus. A. E. Housman 
Note on Emendations of Propertius. A. E. Housman 
On the Aetna of Lucilius. Robinson Ellis 


Conjectures on the Aetna, Culex and Ciris. R. Unger . 


Cic. Acad. Prior. XXV 79, 80. Robinson Ellis 


THE JOURNAL 


OF 


PHILOLOGY. 


EMENDATIONES PROPERTIANAE, 


I SEE no hope of completing a presentable commentary on 
Propertius within the next ten years; but in the mean time I 
trust that the following list of corrections may be found of 
service to scholars. For my own sake too I have some desire 
to put my conjectures on record, as I am for ever seeing them 
forestalled by other students: Mr Konrad Rossberg in vol. 127 
of Fleckeisen’s annual has bereft me of no less than nine. 
True, it is agreeable enough to have one’s results confirmed by a 
scholar who stands next to Mr Baehrens and Mr Palmer at the 
head of living Propertian critics; but I should like to retain 
something of my own. As many readers are apt to fancy that 
the textual critic proposes alterations out of pure gaiety of 
heart and not because the vulgate wants altering, I have 
added an examination in detail of the first elegy; ‘ne mea 
dona tibi studio disposta fideli, Intellecta prius quam sint, 
contempta relinquas’. I employ Mr Baehrens’ mss and nota- 
tion, 


1 i. Between 11 and 12 are lost two such verses as these: 
‘multaque desertis fleuerat arboribus, | et modo submissa casses 
ceruice ferebat’. 


Journal of Philology. vou. xvi. | 1 | 


2 THE JOURNAL OF PHILOLOGY. 


1123 tune ego crediderim wobis et sidera et amnes | posse 
Cytinaeis ducere carminibus] et manes et sidera uobis. 

1 i 33 in me nostra Venus noctes exercet amaras] me non. 

I ii 9 aspice quos summittat humus formosa colores] morosa. 

1 ii 13 litora natiuis persuadent picta lapillis] superant de- 
prcta. 

I ii 23 non illis studium wulgo conquirere amantes DVN, 
aquirere AF] fulgore anquirere. 

111. Between 6 and 7 should be inserted 1 ii 9—12 as 
follows: ‘qualis et Ischomachi Lapithae genus heroine, | Cen- 
tauris medio grata rapina mero, | marcori Ossaeis fertur Boebei- 
dos undis | uirgineum primo composuisse latus, | talis’ eqs. 
Mercurio satis FN, Mercurioque satis DV, Ossaers Burmann. 

I iii 37 namque ubi longa meae consumpsti tempora noctis] 
nempe. The interrogation at the end of 38 should be re- 
moved. 

I iv 19 nec tibi me post haec committet Cynthia] se. 

I iv 24 et quicumque sacer qualis ubique lapis] quaeret. 

I iv 26 quam sibi cum rapto cessat amore deus] decus. 

I v 9 quod si forte tuis non est contraria nostris| uerbis. 

I vi 26 hane animam eatremae reddere nequitiae] hue... 
extremam. 

I vii 16 quod nolim nostros ewiolasse deos| eualwsse. 

I vii 23 and 24 should be placed between 10 and 11: Mr 
Baehrens has seen that they are now out of place. 

I viii 13 atque ego non uideam tales subsidere uentos] laetos: 
13 and 14 should be placed after 16 with Scaliger. 

I vill 22. Read ‘de te | quin ego, uita, tuo limine, nostra, 
querar’. werba Mss. ) 

I ix 32 nedum tu possis spiritus iste leuis] tutus erit. 

I x1 6 ecquis in extremo restat amore locus] amor tecore. 

1 xi 15 and 16 should be placed between 8 and 9. 

I xi 22 aut sine te uitae cura sit ulla meae AFN, an DV] 
Perhaps haut or hau. 

I xiii 12 nec noua quaerendo semper amicus eris] iniquus 
Guietus rightly, except that Propertius wrote inicus. 

I xv 25 and 26 should seemingly be placed before 33, where 
tam tibi should be read with Mr Palmer after Madvig. © 


EMENDATIONES PROPERTIANAE. 3 


1 xv 29 multa prius uasto labentur flumina ponto] Perhaps 
aucta. 

I xvi 9 nec possum infamis dominae defendere noctes] woces - 
thus no transposition is required. 

I xvi 23 me mediae noctes, me sidera prona iacentem | frigi- 
daque Eoo me dolet aura gelu] noctis. 

I xvii 3 nec mihi Casiopae solito wsura carinam DV and 
nearly so AFN] Castoreast stella inwisura. 

I xvii 28 mansuetis socio parcite litoribus ADVN, thoribus F] 
pectoribus. 

I xviii 15 tua flendo | lumina deiectis turpia sint lacrimis] 
flentis. 

I xviii 23 and 24 should be placed between 6 and 7 with an 
tua quod altered to a tua quot as in the interpolated Mss. 

I xviii 27 diuini fontes et frigida rupes] dumetz sentes. 

I xix 18 illic formosae ueniant chorus heroinae] formosus. 

1 xix 16 e¢ Tellus hoc ita iusta sinat] wt. 

I xx 3 and 4 should be written thus: ‘saepe inprudenti 
fortuna occurrit amanti | crudelis: Minuis true erat Ascanius’. 
dixerat O, diaerit N. 

I xx 24 raram sepositi quaerere fontis aquam] Perhaps 
sacram. 3 

I xx 30 et uolucres ramo submouet insidias] armo. 

I xx 52 formosum Nymphis credere wisus Hylam ON, rursus 
V m. 2] ni wis perdere rursus Mr Palmer rightly, except that 
rusus should be read. 3 

I xxii. Between 8 and 9 should be inserted II xxx 21 and 
22; after 10 should be placed Iv i 65 and 66: ‘si Perusina tibi 
patriae sunt nota sepulcra, | Italiae duris funera temporibus, | 
cum Romana suos egit discordia ciues | (sic, mihi praecipue 
puluis Etrusca dolor, | tu proiecta mei perpessa es membra 
propinqui, | tu. nullo miseri contegis ossa solo) | spargereque 
alterna communes caede penates | et ferre ad patrios proelia 
dira lares, | proxima subposito contingens Vmbria campo | me 
genuit terris fertilis uberibus. | scandentes quisquis cernet de 
uallibus arces, | ingenio muros aestimet ille meo’. praemia MSs. 

111 5 siue illam Cois fulgentem incedere coccis] zwwat: thus 
no transposition is required, 


1—2 


4 THE JOURNAL OF PHILOLOGY. 


11 11 and 12 should be placed after 14, and cwm in 11 
should be changed to tum. 

11. After 38 should be inserted 11 ix 33 and 34: ‘Theseus 
infernis, superis testatur Achilles, | hic Ixioniden, ille Menoe- 
tiaden ; | Caesaris et famae uestigia iuncta tenebis: | Maecenatis 
erunt uera tropaea fides ’*. 

11. After 56 should perhaps be placed xv 31—36 which 
have no. business in their present situation. 

Π ii 9—12 belong, as I have said, to I 111; their present 
place, between 8 and 13, must have been originally occupied by 
two such verses as these: ‘aut patrio qualis ponit uestigia 
ponto | mille Venus teneris cincta Cupidinibus’. 

II iii should be joined to ii; but iii 1—8, between 4 and 5 
of which should be inserted with Scaliger 1: 1 and 2, are a frag- 
ment which has no business here. 

II iii 11 and 12 should be placed between 16 and 17, and 
the whole passage written thus: ‘nec me tam facies, quamuis 
sit candida, cepit | (lilia non domina sunt magis alba mea), | nec 
de more comae per leuia colla fluentes, | non oculi, geminae, 
sidera nostra, faces, | nec siqua Arabio lucet bombyce puella | 
(non sum de nihilo blandus amator ego), | ut Maeotica nix 
minio si certat Hibero | utque rosae puro lacte natant folia, | 
quantum quom posito formose saltat Iaccho’ eqs. 

II iii 45 and 46 (Hertzberg Haupt Palmer=iv 1 and 2 
Mueller Baehrens) have no business where they now are; no 
more have iv 5 and 6 (H. H. P.=15 and 16 M. B.). 

II vi is a patchwork of these fragments: 1—8; 9—14; 15— 
26 after which we should seemingly with Mr Heydenreich 
place 35 and 36; 27—34 (so Lachmann); 37—40 (the same) ; 
finally 41 and 42 should be placed with Mr Baehrens after 
vii 12. 

II vii 16 non mihi sat magnus Castoris iret equus] nawus. 

II vill 8 and 4 are out of place; so are 11 and 12. 

II vili 21—24 have no business here and should perhaps be 
placed after xxviii 40. 

II viii 30 cessare in ¢ectis pertulit arma sua] Teucris. 

Il ix 7 wiswra et quamuis numquam speraret Vlixem] Per- 
haps wsurum. 


EMENDATIONES PROPERTIANAE. 5 


II ix 12 et dominum lauit maerens captiua cruentum | ap- 
positum flauis in Simoenta uadis| Simoente. 

Π ix 15 cwm tibi nec Peleus aderat nec caerula sister FN, 
quom tibt DV] cur tum or quot tum. 

II ix 18 tunc etiam felix inter et arma pudor] otva tune. 

II ix 29 and 30 should be placed between 20 and 21. 

II ix 44 nunc quoque 6718, quamuis sis inimica mihi] era’s. 

II x 2 et campum Haemonio iam dare tempus equo] campum 
et Maeonio. 

II xii 6 fecit et humano corde uolare deum] haut wano. 

II xiii 1 non tot Achaemeniis armantur Etrusca sagittis O, 
armatur ΝῚ armatus Eruthra or Erythra. 

II xiii 38 quam fuerant Phthii busta cruenta uiri] funere 
quam. 

Ir xiii 39 and 40 should be written ἀμοϑι ‘tu quoque si 
quando uenies adfata (memento) | hoc iter, ad lapides, cara, ueni 
memores’. ad fata MSS. 

Il xiii 45 nam quo tam dubiae seruetur spiritus horae] 
Perhaps aurae. 

II xiii 48 cui si tam longae minuisset fata senectae | Galli- 
cus Iliacis miles in aggeribus] caelicus. 

II xiii 55 illic formosum tacuisse paludibus, illuc | diceris 
effusa tu, Venus, isse coma DVN, plaudibus F] ciuisse a planc- 
tibus: ciwisse Mr Baehrens. 

II xiv 5 salwum cum aspexit Orestem ΕἾΝ, suwm saluum DV] 
cum saluum. 

Il xiv 7 and 8 should be written thus: ‘nec sic, cwm inco- 
lumem Minois Thesea uidit, | Daedalium lino cuz duce rexit — 
iter’. The MSS omit cum in 7 and have cum for cuz in 8. 

II xiv 29 and 30 should be written thus: ‘nunc a te, mea 
lux, pendet, mea litore nauis | soluat an in mediis sidat honusta 
uadis’. ad te and weniet MSS; seruata an FN, seruata in DV. 

1 xv. The verses of this elegy should be arranged as 
follows: 1—8, 37—40, 9—24, 49 and 50, 29 and 30, 27 and 28, 
25 and 26, 51—54. 25 and 26 have already been placed after 
28 by Mr Palmer. 41—48 should be placed after xxx 18; 
31—36 perhaps after i 56. 

11 xv 1 should be written thus: “10 me felicem, 7o nox mihi 


6 THE JOURNAL OF PHILOLOGY. 


candida, io tu | lectule’. o0...0...0 F, 0...0...e¢ ὁ N, ah...0o... 
et o DV. 

II xv 37 quod mihi si tecwm tales concedere noctes | illa 
uelit] interdum. 

II xvi 13 and 14 should be placed after 28; 17 and 18 after 
12; 29 and 30 after 46; 41 and 42 after 111 x1 38. 

II xvii 13 and 14 should be placed after 2; after 4 should 
be placed xxii 43—50. The verses 5—12 and 15—18 are a 
fragment of another poem. 

II xviii 5 quid si iam canis aetas mea caneret annis] Per- 
haps marceret ab. | 

II xviii 9 illum saepe swis decedens fouit in ulnis | quam 
prius adiunctos sedula lauit equos] prius...stadis functos: no 
transposition of verses should be made. 

II xviii 23 nwne etiam infectos demens imitare Britannos | 
ludis et externo tincta nitore caput] tune...uadis. 

Π xvii 29 and 30 should be placed after 24, and in 29 
deme: mihi should be written with Perreius. 

II xviii 33 and 34 have no business here. 

II xviii 37 and 38 should be placed at the end of xix, which 
see. 

II xix 5 nulla neque ante tuas orietur rixa fenestras] ulla. 

It xix 17—24 are no part of this poem. 

II xix 18 me sacra Dianae | suscipere et Venert ponere uota 
iuuat] Perhaps Veneris. 

II xix 27 and 28 should be placed after 32; after 28 should 
be placed xviii 37 and 38. 

II xix 31 quin ego in assidua mutem tua nomina lingua] 
The sense required is ‘quin ego tua crimina metuam uelut in 
assidua turba’. In 29 sic should be changed to set with Munro. 

II xx 8 nec tantum Niobae bis sex ad busta superbae | solli- 
cito lacrimans defluit a Sipylo] os. 

Ir xx 35 hoc mihi perpetuo zus est] haec...laus. 

II xxi 12 eiecta est tenws namque Creusa domo] eiecit 
Aesonia. 

II xxii 43—50 should be placed after xvii 4, as I have said. 

Ir xxii 1 cui fut indocti fugienda et semita uulgi FN, et 
omitted by DV] cui fugienda fuit indocti semita uulgi. 


EMENDATIONES PROPERTIANAEL. 7 


II xxiii 4 ut promissa suae uerba ferat dominae] praemissa. 

It xxii 23 and 24 should be placed after xxiv 4. 

Il xxiv 4 aut pudor ingenuus aut reticendus amor] a pudor, 
ingenuus reiciendus amor. 

II xxiv 8 urerer et quamuis, nomine uerba darem] urerer et 
quamuis non bene, uerba darem. 

Ir xxiv 51 fz tibi nos erimus] hie. 

I xxv 35 at si saecla forent antiqus grata puellis] gratis 
antiqua. 

ΤΠ xxv 41 uidistis pleno teneram candore puellam, | uidistis 
fusco] Perhaps niweo. 

I xxv 43 uidistis quandam Argiua prodente figura, | uidistis 
nostras Ὁ, guadam N] patriam Argiuas. 

mI xxv 45 alaque plebeio uel sit sandicis amictu] aeque. 

11 xxv. Before 47 at least two verses have been lost: the 
passage may have run thus: ‘quin tu wulgares, demens, compescis 
amores | in poenamque uagus desinis esse tuam, | cum satis una 
tuis insomnia portet ocellis | una sit et cuiuis femina multa 
mala’. 

Π xxvi 23 non si Cambysae (cambise) redeant et flumina 
Croesi] tam inuist. | 

II xxvi 31 and 32 should be placed before 29: after 28 two 
verses have been lost: the passage ran thus: ‘svue iter in terris 
dominae sit carpere cura | terrestrem carpet me comitante wiam ; | 
unum litus erit positis torus unaque tecto | arbor, et ex una 
saepe bibemus aqua. | seu mare per longum’ eqs. sopitis MSS, 
omitting torus. | 

I xxvi 54 nec umquam | alternante worans uasta Charybdis 
aqua] uacans Ayrmann rightly, except that Propertius wrote 
wocans, a form preserved by F in Iv 11 19 ‘ mendax fama woces’. 

II xxvii 7 rursus et obiectum flemus caput esse tumultu O, 
fletus ΝῚ 5468 tu. 

tm xxviii. After 2 should be placed 33—38: 33 and 34 
were so placed by Passerat. 

II xxviii 9—32 are no part of this poem. 

mI xxviii 40. After this verse should perhaps be placed viii 
21—24, as I have said. 

It xxviii 51 uobiscum est cope] If Mr Rossberg’s Cretw for 


8 THE JOURNAL OF PHILOLOGY. 


troia in 58 and Jacob’s Beli for phebi in 54 are correct, I propose 
Hesione. 

II xxviii 57 and 58 should be placed before 111 xviii 25. 

II xxviii 61 and 62 should be punctuated thus: ‘redde etiam 
excubias diuae nunc ante iuuencae | uotiuas, noctes et mihi 
solue decem ’. 

II xxix 27 ibat et hinc castae narratum somnia Vestae] 
Perhaps a. 

II xxix 36 signa uoluptatis non iacuisse duos. uoluntatis FN, 
uolutantis F man. 2, nec Ν] uolutantis concalwisse. 

ΤΠ xxx. Here are three elegies or fragments of elegies : the 
first 19 and 20, 1 and 2, 7—10, 3—6, 11 and 12; the second 
18—18 (here insert xv 41—48), 37—40; the third 23—30, 
33—386, 31 and 32. 21 and 22 should be placed after I xxii 8, 
as I have said. 13 eqs. have already been separated by Mr 
Heimreich and 23 eqs. by Lachmann from the verses which 
precede them in the Mss: 31 and 32 have been placed after 36 
by Mr Rossberg. 

II xxx 35 si tamen Oeagri quaedam compressa figura | Bis- 
tonis olim rupibus accubuit] figurae. 

II xxxil. Before 1 should be placed 7 and 8. 

II xxxii 5 cwr uatem Herculeum deportant esseda Tibur O, 
curua te N| curnam te. 

Il xxxli 15 and 16 should be written thus: ‘et leuiter 
lymphis lato crepitantibus orbe | quam subito Triton ore recon- 
dit aquam’ or ‘aqua’. tota...wrbe cum Mss, toto...orbe Heinsius. 

I xxxii 25 and 26 should be placed after 30. 

_ II xxxii 32 et sine decreto uiua reducta domum est] de. 

Il xxxli 37 hoc et Hamadryadum spectauit turba sororum 
DVN, non ΒῚ wos. 

II xxxil 41 and 42 have no business here and should perhaps 
be placed after M1 xiii 12. 

II xxxli 43 and 44 should be placed after 46. 

II xxxii 61 should be written ‘quod si tu Graias winces 
imitata Latina’. tuque es FN, siue es DV: I fancy O had 
ἔν 68. 

II xxxill 6 quaecumque illa fuit, semper amara fuit] Per- 
haps quodewmque. 


EMENDATIONES PROPERTIANAE. 9 


II xxxiv 12 posses in tanto uiuere flagitio N, posset et in F, 
posses et in DV] posses tun. 

Π xxxiv 31—54 should be thus arranged: 51—54, 41 and 
42 (so Munro), 39 and 40 (Munro), 31 and 32 (Munro), 43 and 
44, 33—38, 45—50. There should be a comma, not a full stop, 
at the end of 38. 

Π xxxiv 40 Amphiaraeae prosint tibi fata quadrigae | aut 
Capanei magno grata ruina Ioui? N, magno omitted by O] «rato. 

II xxxiv ὅθ me iuuet hesternis positum languere corollis] mi 
lubet...posito: retain Vergilio in 61. With 59 begins a new 
elegy. 

II xxxiv 83 nec minor his animis, aut sim minor ore, cano- 
rus | anseris indocto carmine cessit olor] hic...ué sit. 

III ii 24 annorum aut τού pondere uicta ruent N, ictu pon- 
dera F, ictus pondere DV] ictus pondera. 

III iii 41 nil tibi sit rauco praeconia classica cornu | flare N, 
praeconica O| Perhaps Phoenicia. 

tI iv 4 should be written ‘Thybris, et Euphrates sub tua 
iura fluet’. Tygris N, Tigris O: fluent NO. 

Im iv 18 et subter captos arma sedere duces] cautos: 17 and 
18 should be placed before 15 with Mr Keil. 

111 v 9 corpora disponens mentem non uidit in arte] arto. 

Im v 11 nunc maris in tantum uento iactamur] ponto. 

111 v 15 uictor cum uictis pariter miscebitur umbris] misce- 
tur in. 

m1 v 40 and 42 should exchange places. 

ΠῚ vi 3 and 4 should be placed after 8. 

III vi 28 et lecta exectis anguibus ossa trahunt DVN, exactis 
ΕἼ exuctis. 

ΠῚ vi 40 me quoque consimili inpositum torquerier igni | 
iurabo bis sex integer esse dies] ἴρ86. 

lI vil. The verses of this elegy should be arranged thus: 
1—10, 43—66, 17 and 18, 11—16, 67—70, 25—32, 37 and 38, 
35 and 36, 19 and 20, 33 and 34, 21—24, 39—42, 71 and 72. 
17 and 18 were placed after 66, 67—70 after 16, 25—28 after 
70, 35 and 36 after 38, 39—-42 after 24 by Scaliger; 43—66 
after 10, 11 and 12 after 18 by Mr Baehrens. Further, 51 and 
58 should exchange places as Mr W. Fischer bids them. 


10 THE JOURNAL OF PHILOLOGY. 


III vii 60 attulimus longas in freta uestra manus] nocuas: 
the sentence is interrogative. 

ΠῚ viii 12 e¢. Veneris magnae uoluitur ante pedes] haec: a 
full stop should be placed at the end of the verse: 13—18 have 
no business here. 

III viii 35 and 36 are out of place. 

iI ix 9 gloria Lysippo est animosa effingere signa N, fingere 
ΟἹ] ecfingere. 

II ix 16 Praxitelem propria windicat urbe lapis] wendit at: 
uenditat Hertzberg. 

Il ix 25 Medorum pugnaces tre per hostes] Perhaps pugna 
rescindere postes. 

ΠῚ ix 33 and 34 should be placed after 111 38, as I have said. 

ΠῚ ix 49 and 51 should exchange places. 

lI x 23 tibia nocturnis succumbat rauca choreis] continuis. 

ΠΙ xi 13—16 should be placed after 20. | 

ΠῚ xi 17 Omphale in tantum formae processit honorem] 
Perhaps Maeonis. 

ΠῚ xi 36 and 40 should exchange places as Lachmann bids 
-them ; then after 38 should be inserted 11 xvi 41 and 42: the 
passage should run thus: ‘haec tibi, Pompei, detraxit harena 
triumphos: | nulla Philippeost agmine adusta nota. | issent 
Phlegraeo melius tibi funera campo; | nec twa 810 socero colla 
daturus eras: | Caesaris haec uirtus et gloria Caesaris haec est, | 
illa, qua uicit, condidit arma manu’. wna Philippeo sanguine 
and wel tua st MSS. 

ΠῚ x1 47—68 should be arranged thus: 51—58, 65—68, 59 
and 60, 47—50. 67 and 68 were placed before 59 by Passerat. 

ΠΙ xi 55 and 56 should be written thus: ‘non hoc, Roma, 
fui tanto tibi ciue uerenda’ | diverat assiduo lingua sepulta 
mero. dixit et MSS. | 

Il xi 70 tantum operis belli sustulit una dies] tanti...bellum. 

1Π xii 25 castra decem annorum et Ciconum mons Ismara 
calpe| mersa...clade or caede: domita...clade Eldick. 

ΠῚ ΧΗ 9 haec etiam clausas expugnant arma pudicas] 
Euhadnas. 

1 xiii 12, After this verse should perhaps be placed 
Il xxxii 41 and 42, as I have said. 


EMENDATIONES PROPERTIANAE. If 


ΠῚ xiii 19 and 20 should be written thus: ‘et certamen 
habent, letum quae uiua sequatur | coniugi’. leti...coniugiwm 
Mss : editors put the comma after Jetz. 

lil xiii 35 hinulei pellis totos operibat amantes] Perhaps 
lentos: in 37 laetas should be read with F. 

III xiii 39 corniger atque dei uacuam pastoris in aulam | dux 
aries saturas ipse reduxit oues] die. 

ΠΙ| xiii 43—46 I fear have no business here. 

ΠΙ xiv 15 and 16 should be placed before 11; Scaliger 
placed them before 13. 

ΠῚ xvi 20 sanguine tam paruo quis enim spargatur amantis | 
improbus? eaclusis fit comes ipsa Venus] ewsuctis. 

ΠῚ xvi 21 quod si certa meos sequerentur funera casus] 
cursus. 

III xvii 12 spesque timorque animo uersat utroque modo] 
anvmae. 

ΠῚ xvii 24 Pentheos in triplices funera grata greges] Per- 
haps carpta. 

ΠῚ xviii 10 errat et in westro spiritus ille lacu] inferno. 

ΠῚ xviii 19 and 20 Attalicas supera uestes atque omnia 
magnis | gemmea sint ludis] ostra zmaragdis...Indis, 

ΠΠΙ xviii 21 sed tamen hoc omnes, huc primus et ultimus 
ordo| manet...inus. 

ΠῚ xviii 25. Before this verse should be placed τι xxviii 57 
and 58, as I have said. 

ΠῚ xvili 29 and 30 should be placed after Iv vi 34. 

III xviii 31—34 should be written thus: ‘at tibi nauta, pias 
hominum qui traicit umbras, | hac animae portet corpus inane 
uia, | qua Siculae uictor telluris Claudius et qua | Caesar ab 
humana cessit in astra uice’. huc...twae MSS. 

ΠῚ xix 25 and 26 should be placed after 28: thus tamen in 
27 can be retained. 

1Π xx 19—24 should be placed before 15: Lachmann placed 
19 and 20 there. 

III xx 25 qui pactas in foedera ruperit aras] Perhaps tacta 
Sic...ard. 

Il xxii 3 Dindymis et sacra fabricata tuuenta Cybelle O, 
inuenta N] in caute. 


12 THE JOURNAL OF PHILOLOGY. 


1Π xxii 15 and 16 should be placed before 7 and written 
thus: ‘siqua et olorigeri uisenda est ora Caystri | et quae 
septenas temperat unda uias’. θέ siqua NF man. 2, at siqua O. 
gua NO. 

1Π xxii 25 Albanus lacus et socii Nemorensis ab unda N, 
sotii F, sotiis DV] foliis Nemorensis abundans. 

ΠῚ xxii 41 hic tibi ad eloguiwm ciues] Perhaps adloquiwm. 

ΠῚ xxiii 14 an tu | non bona de nobis crimina ficta iacis] 
carmina. 

lI xxiii 17 and 18 should be punctuated thus: ‘et quae- 
cumque uolens reperit non stulta puella | garrula, cum blandis 
dicitur hora dolis’. 

ΤΠ xxiv 9 and 10 should be placed after 12 thus: ‘haec ego 
non ferro, non igne coactus, et ipsa | naufragus Aegaea uerba 
fatebor aqua. | quod mihi non patrii poterant auertere amici | 
eluere aut uasto Thessala saga mari, | correptus saeuo Veneris 
torrebar aeno’ eqs. 

ΠῚ xxiv 19 Mens Bona, siqua deo es, tua me in sacraria 
dono] adeo. 

1vi7 Tarpeitus. Tarpetius N] Tarpeiius. So iv 1 I should 
write Tarpeiiae for the tarpelle of F, and iv 15 Tarpeza for the 
carpella of the same MS. 

Iv i 19 annuaque accenso celebrare Palilia faeno FNV, cele- 
brate D] celebrante. 

1vi28. After this verse should be inserted x 21 and 22; 
after 29 should be placed in reverse order x 19 and 20: the 
passage will run thus: ‘nec rudis infestis miles radiabat in 
armis: | miscebant usta proelia nuda 5866, | picta neque in- 
ducto fulgebat parma pyropo: | praebebant caesi baltea lenta 
boues. | prima galeritus posuit praetoria Lycmon, | nec galea 
hirsuta compta lupina iuba. | idem eguos et frenis, idem fuit 
aptus aratris, | magnaque pars Tatio rerum erat inter oues’. δέ 
and eques MSS. 

Iv i31—56. Out of these verses, 33—36 should be placed 
after x 26; the rest should be arranged thus: 37 and 38, 55 
and 56 (so L. Lange), 31 and 32, 45 (write hinc with Heinsius) 
and 46, 39 and 40 (write huc with Messrs Baehrens and Palmer), 
47—52, 41 (write alos with Schrader) —44, 53 and 54. 


EMENDATIONES PROPERTIANAE. 13 


Iv i 31 hinc Tities Ramnesque uiri Luceresque coloni O, 
solont N] seuerc. 

Iv 1 50 diait Auentino rura pianda Remo] Perhaps dizerat 
α uentis non rapienda. 

Iv i 57—70, when 65 and 66 have been removed and placed 
after I xxii 10, compose a prooemium to i (1—56), ii, iv, vi, ix, 
x, and should be arranged thus: 61—64, 57—60, 67—70. 

Iv 1 81 eqs. should be written thus: ‘nunc pretium fecere 
deos et fallimus auro | (Iuppiter !) obliquae signa iterata rotae | 
felicesque Iouis stellas’ eqs. fallitur Mss. 

Iv 1 85 and 86 should be placed after 108: 83—86 were 
placed there by Scaliger. 

Iv i 87 and 88 should with Scaliger be placed before 71: 
88 should be written ‘et maris et terrae regna superba canam’. 
longa sepulcra MSS. 

Iv i 120 incipe tu lacrimis aequs adesse nouis] miraclis. 

IV i 124 et lacus aestiuis intepet Vmber aquis] non tepet. 

Iv i 143 and 144 should be placed before 141. 

IV ii 2 accipe Vertumni signa paterna dei] regna. 

Iv ii 4. After this verse should be placed 49—56 in the 
following order: 51—54, 49—50, 55 and 56. There should be 
a comma at the end of 4, a full stop at the end of 52. 

Iv ii 12 Vertumni rursus credidit esse sacrum] credis id. 

IV ii 35 est etiam aurigae species Vertumnus et eius eqs. | 
Perhaps mentiar. 

IV ii 39 pastorem ad baculum possum curare] da baculum, 
pastor me possum ornare. pastor me Ayrmann. 

Iv iii 7—10 te modo wderunt iteratos Bactra per ortus, | te 
modo munito Sericus hostis equo, | hibernique Getae pictoque 
Britannia curru, | ustus et Eoa decolor Indus aqua] Itwraeos 
uiderunt...arcus...Hyrcania...tusus. 

Iv ili 11 should be written thus: ‘haecne marita fides et 
[primae] praemia noctis’. hae sunt pactae mihi DV, et pacate 
mihi F, [et pactae mihi O I fancy], et parce awa N, [et prae mia 
the archetype I fancy]. 

Iv iti 29—62 should be arranged thus: 43—50 (so Mr 
Luetjohann), 29—32, 55 and 56, 33 and 34, 51 and 52, 35—42, 
53 and 54, 59—62, 57 and 58, 


14 THE JOURNAL OF PHILOLOGY. 


Iv iii 48 cum pater altas | Africus in glaciem frigore nectit 
aquas] caelicus. 

IV iii 62 succinctique calent ad noua lucra popae] lustra. 

IV iii 63 ne precor ascensis tanti sit gloria Bactris] accensis. 

Iv iv 17 and 18 should be placed after 86: Broukhusius 
placed them after 92. 

IV iv 47 cras, ut rumor ait, tota pugnabitur urbe] pigrabitur. 

Iv iv 71 and 72 should be placed after viii 52. 

IV iv 82 pacta ligat, pactis ipsa futura comes] coeptis. 

IV iv 87 prodiderat portaeque fidem patriamque iacentem|] 
Perhaps patrem. 

Iv v 19 and 20 should be written thus: ‘ewercebat opus, 
uerbis hew blanda, perinde | saxosam atque forat sedula gutta 
uiam’. exorabat...ceu...perure Mss, feratque V, que ferat DFN, 
forat Messrs Rossberg and Palmer. 

Iv v 21 si te Koa derorantum iuuat aurea ripa O, dorozan- 
tum N] topazorum. 

Iv v 29—62 should seemingly be arranged thus: 59—62 
(so Mr Luetjohann), 41—44, 47—58, 45 and 46, 31 and 32, 
29 and 30, 33—36, 39 and 40, 37 and 38. 

IV vi 26 armorum et radiis picta tremebat aqua ON, que for 
et V man. 2] radwsque icta. ita Heinsius. | 

Iv vi 34. After this verse should be placed 1m xviii 19 and 
20, as follows: ‘non ille attulerat crines in colla solutos | aut 
testudineae carmen inerme lyrae, | sed quali aspexit Pelopeum 
Agamemnona uultu | egessitque auidis Dorica castra rogis | (hic 
olim ignaros luctus populauit Achiuos | Atridae magno cum 
stetit alter amor), | aut quali flexos soluit Pythona per orbes’ 
eqs. 

Iv vi 45 and 46 should be placed after 52. 

Iv vi 49 guodque vehunt prorae Centaurica saxa minantis] 
Perhaps quot. 

Iv vi 81 siue aliquis pharetris Augustus parcet Eois] Per- 
haps aequus. 

IV vii 4 murmur ad extremae nuper humata wiae] He 
should have written Tibure ad extremam...uianv. 

Iv vii 23 at mihi non oculos quisquam inclamauit euntis] 
eunt, 


EMENDATIONES PROPERTIANAE. 15 


IV vii 55—58 should be written thus: ‘nam gemina est 
sedes turpem sortitague ueram, | cwmbaque diuersa remigat 
omnis aqua: | una Clytaemestrae stuprum uel adultera Cres- 
sam | portat mentitam lignea monstra bouis’. per amnem 
turba MSS. 

_ Iv vii 64 narrant historias, pectora nota, suas] sancta. 
Iv viii 1 disce quid Esquilias hac nocte fugarit aquosas] nocte 
hae furiarit. 

Iv viii 4 hic tibi tam rarae non perit hora morae] Perhaps 
gratae. ) 

IV viii 9 and 10 should be placed after 12. 

IV viii 39 should be written thus: ‘wnguentum, tibicen erat, 
crotalistria, phimus’. Nile twus...phillis Mss, 

Iv viii 52. After this verse should be placed iv 71 and 72 
as follows: ‘nec mora, cum totas resupinat Cynthia ualuas. | non 
operosa comis sed furibunda decens | illa ruit, qualis celerem 
prope Thermodonta | Strymonis abscisso pectus aperta sinu’. 

Iv ix 21 dixerat; et sicco torquet sitis ora palato] at. 

Iv ix 29 populus et longis ornabat frondibus aedem] glaucis. 

Iv ix 31 δυο rwit in siccam congesta puluere barbam]  Per- 
haps in sicca ruit...labra. 

Iv ix 60. Write ‘haec lympha puellis, | auia secreti limitis 
unda, fluit’. wna MSS. | 

Iv ix 70 Hercule exterminium nescit inulta sitis] I had 
conjectured ‘Herculea (extremum) nec sit inulta sitis’; but 
perhaps ‘ Herclei exterminium nec sit’ is right. 

Iv x 19 and 20 should be placed after 1 29, and 21 and 22 
after i 28, as I have said. 

Iv x 23 and 24 should be placed as Passerat bids before 27 ; 
after 26 should be placed 1 33—36 with Mr Lucian Mueller’s 
transposition of 34 and 36: the passage should run thus: 
‘necdum ultra Tiberim belli sonus: ultima praeda | Nomentum 
et captae iugera terna Corae. | quippe suburbanae parua minus 
urbe Bouillae | ac tibi Fidenas longa erat ire uia;| et stetit 
Alba potens, albae suis omine nata, | et, qui nunc nulli, maxima 
turba Gabi. | Cossus at insequitur’ eqs. 

Iv x 87 di-Latias iuuere manus. Romuleas F, and D in 
marg.| di Remulas, . 


16 THE JOURNAL OF PHILOLOGY. 


Iv xi. The verses of this elegy should be arranged thus: 
1—18, 47—54 (in 49 write wmbra with Eldick), 19 (retain aut) 
—32, 43 and 44, 33—36, 45 and 46, 37—42, 55—62, 97 and 98, 
65 and 66, 99 and 100, 69 and 70, 73 and 74, 63 (retain te...te) 
and 64, 75—96, 67 and 68, 71 and 72 (write torum with 
Schrader), 101 and 102. Mr Baehrens has placed 71 and 72 
after 68. 

Iv xi 15 damnatae noctes et uos uada lenta paludes | et 
quaecumque meos implicat unda pedes] testes. 

Iv xi 40 quique tuas proauo fregit Achille domos O, proawus 
V man. 2] ‘quique tuas proauus fregit, Auerne, domos’ Munro: 
write proauos. 

Iv xi 50 turpior assensu non erit ulla meo] accensu. 

Iv xi 87 coniugium, pueri, laudate et ferte paternum] 
durate. 

I hope I have managed to keep my neighbour’s goods out 
of this catalogue, but I dare hardly expect it: at the very last 
moment I have cancelled an amendment of II xxxii 23 which 
I find was made forty years ago by Schneidewin and has been 
neglected by everyone since. ΤῸ anyone who will enable me to 
restore misappropriated discoveries to their rightful owner I 
shall be honestly indebted. I now go on as I promised to 
comment on the text of the first elegy. 


Cynthia prima suis miserum me cepit ocellis 
contactum nullis ante cupidinibus, 
tum mihi constantis deiecit lumina fastus 
et caput inpositis pressit Amor pedibus, 
donec me docuit castas odisse puellas 5 
improbus et nullo uiuere consilio. 
et mihi iam toto furor hic non deficit anno, 
cum tamen aduersos cogor habere deos. 
Milanion nullos fugiendo, Tulle, labores 
saeuitiam durae contudit Iasidos. 10 
nam modo Partheniis amens errabat in antris 
ibat et hirsutas ille uidere feras; 
ille etiam psilli percussus uulnere rami 
saucius Arcadiis rupibus ingemuit. 


EMENDATIONES PROPERTIANAE. 17 


ergo uelocem potuit domuisse puellam : 15 
tantum in amore preces et bene facta ualent. 
in me tardus Amor non ullas cogitat artes 
nec meminit notas, ut prius, ire wias. 
at uos, deductae quibus est fallacia lunae 
et labor in magicis sacra piare focis, 20 
en age dum dominae mentem conuertite nostrae 
et facite illa meo palleat ore magis. 
tunc ego crediderim uobis et sidera et amnes 
posse cytalinis ducere carminibus. 
et uos, qui sero lapsum reuocatis, amici, 25 
quaerite non sani pectoris auxilia. 
fortiter et ferrum saeuos patiemur et ignes, 
sit modo libertas, quae uelit ira, loqui. 
ferte per extremas gentes et ferte per undas, 
qua non ulla meum femina norit iter. 30 
uos remanete, quibus facili deus annuit aure, 
sitis et in tuto semper amore pares: 
in me nostra Venus noctes exercet amaras 
et nullo uacuus tempore defit Amor. 
hoc, moneo, uitate malum: sua quemque moretur 35 
cura neque assueto mutet amore locum. 
quod si quis monitis tardas aduerterit aures 
heu referet quanto uerba dolore mea. 


ἄς ἃ Among all the four thousand verses of the poet there is 
not a sounder or simpler than this. Not only are such locutions 
as ‘deiecit Jumina’ for ‘effecit ut lumina deicerem’ frequent in 
both tongues—see for example Hor. epist. I 5 22 ‘ne sordida 
mappa conruget naris’ and Eur. Hel. 1122 πολλοὶ δ᾽ ᾿Αχαιών... 
“Avdav μέλεον ἔχουσιν, τάλαιναν ὧν ἀλόχων κείραντες ἔθειραν---- 
but the very words of Propertius are closely imitated in Ouid. 
her. XI 35 ‘erubui-gremioque pudor deiecit ocellos’. Again, 
inasmuch as ‘lumina fastus’ is not Latin, the genitive here is 
of course the genetiuus qualitatis cum epitheto: ‘constantis 
lumina fastus’ = ‘constanter fastosa’ just as Hor. carm. 111 7 4 
‘constantis iuuenem fide’ =‘constanter fidelem’. Why then 
Fonteine should write in his margin ‘tum me constantis deiecit 
Journal of Philology. vou. xvt. 2 


18 THE JOURNAL OF PHILOLOGY. 


culmine fastus’, why Burmann should propose ‘tum mihi con- 
stanti deiecit lumina fastu’, why Mr Paley should say ‘the 
expression is a remarkable one’, why both he and Mr Palmer 
should repeat Burmann’s conjecture a hundred years late, why 
Hertzberg should be driven to Tartara leti and Mr Postgate to 
Roby 1304, I am unable to discern. 

5. Fonteine does not himself say why he desires ‘ cunctas’ 
for ‘castas’, but Mr Baehrens prolegg. p. XLVII gives his own 
reasons for adopting the conjecture: these I will examine. 
In 11 3 1 an imaginary censor is made to address Propertius 
thus: ‘qui nullam tibi dicebas iam posse nocere, | haesisti: 
cecidit spiritus ille tuus’: these words, says Mr Baehrens, 
evidently refer to some passage in book I. But where in book 
I can this boast be found? nowhere: it must therefore be 
imported. This Mr Baehrens thinks he can do by writing 
here ‘ me docuit cunctas odisse puellas’. He states his reason, 
of which more anon, for deeming ‘castas’ corrupt, and proceeds 
‘immo omnes omnino feminas propter unius duritiem Pro- 
pertium tum odisse innuere uidetur u. 30 (ferte per extremas 
gentes et ferte per undas, gua non ulla meum femina norit 
wer)’. Now the phrase ‘cunctas odisse puellas’ can mean 
either of two things: it can mean either to dislike women 
because you are indifferent to them, or to dread them because 
you are too susceptible: either of these two things, I say, it 
can mean, but it cannot mean both at once. In the latter of 
the two senses it will be appropriate enough to this elegy into 
which Mr Baehrens wants to bring it; but of course it will 
then be of no service whatever to Mr Baehrens as an equiva- 
lent for the ‘nulla mihi iam potest nocere’ of 11 8 1. In the 
former of the two senses it will tally precisely with ‘nulla mihi 
lam potest nocere’; but then it can by no possibility have a 
place in this elegy. This elegy is written by a man desperately 
in love: first he invokes magicians to turn his mistress’s heart 
and colour her face paler than his own; failing that, he invokes 
his friends to cure him of his slavish attachment by surgery 
and cautery: ‘nullo uacuus tempore defit Amor’ he says; and 
we are asked to believe that he said in the same poem ‘nulla 
mihi iam potest nocere’! That very verse 30 to which Mr 


EMENDATIONES PROPERTIANAE. 19 


Baehrens appeals is his confutation: why must the poet be 
fleeing to the ends of the earth ‘qua non ulla meum femina 
norit iter’? quia omnes feminae nocent. I may add that the 
Pompeian inscription C. 1. L. Iv 1520 ‘candida me docuit 
nigras odisse puellas’ affords an indication, slight indeed, but 
still an indication, that the adjective here was at any rate not 
‘cunctas’ but a descriptive epithet such as both ‘nigras’ and 
‘ castas ’ are. | 

It remains to consider whether Mr Baehrens’ objection to 
‘castas’ is better supported than his advocacy of ‘cunctas’, 
He writes ‘uulgares meretrices qui sectabatur, is sine iusta 
causa querebatur de tristi Venere noctes in se exercente amaras 
(u. 83)’. Lachmann will answer him better than I: praef. p. 
XXIV ‘tu ne dubita quin poeta se, Cynthia et castis puellis 
relictis (hoc erat illud: peccaram semel et totum sum pulsus 
in annum), iam per totum annum uiles quaerere et sine con- 
silio queratur uiuere, aduersa tamen Venere et Cynthiae de- 
sertae memoria animum assidue subeunte. hunc uerum sensum 
esse certius fit ex his eiusdem carminis uersibus: hoc, moneo, 
uitate malum. sua quemque moretur cura, neque assueto mutet 
amore locum’. A very little consideration would have been 
enough to convince a scholar of Mr Baehrens’ acumen that 
‘castas’ was unimpeachably right and ‘cunctas’ the idlest of 
guesses. 

11, But if critics have shewn morbid alertness above, they’ 
are cast into a deep sleep when they come to this verse. If a 
poet in the year 26 B.c. or thereabouts writes ‘Milanion was 
lately roaming in the dells of Arcadia’, he writes nonsense; 
yet no other meaning does the Latin tongue permit these 
words to bear. For if modo is to mean ἐνίοτε μέν it must be 
answered by an ἐνίοτε δέ in the shape of a second modo or of 
some other competent adverb such as nunc, rursus, interdum, 
saepe, aliquando, non numquam; and of course efiam in 13 is 
not a competent adverb. If rules like this, built up by wide 
and orderly induction, are to be overthrown at the bidding of 
fourteenth century MSS, goodnight to grammatical science. 
The Mss of Propertius exhibit the solecism four times in all, 
dutifully followed in every instance by all modern editors but 


2—2 


20 THE JOURNAL OF PHILOLOGY. 


Mr Baehrens, and by Mr Baehrens in two instances. The 
verses ΠῚ 24 9 sqq. run thus: ‘quare ne tibi sit mirum me 
quaerere uiles: | parcius infamant: num tibi causa leuis? | et 
modo pauonis caudae flabella superbae | et manibus dura frigus 
habere pila | et cupit iratum talos me poscere eburnos | quae- 
que nitent Sacra uilia dona Via. | a peream si me ista mouent 
dispendia’ eqs. Here, setting grammar aside, it is manifest 
and was remarked by Scaliger that 11 sqq. have not the 
remotest connexion in theme with the preceding verses: 11 
—16 are a fragment truncated of its head and inserted in a 
wrong place. Mr Baehrens therefore rightly marks a lacuna: 
his fellows print the lines as if they were coherent and gram- 
matical. Mr Baehrens again is the only modern editor whose 
text of 1 11 1—5 is Latin or sense: this is the vulgate: ‘ec- 
quid te mediis cessantem, Cynthia, Bais, | qua iacet Herculeis 
semita litoribus, | et modo Thesproti mirantem subdita regno | 
proxima Misenis aequora nobilibus, | nostri cura subit memores 
a ducere noctes?’ This was corrected long ago by the Italians 
of the renascence and again by Scaliger: Propertius wrote in 
4 ‘et modo Misenis aequora nobilibus’: ‘ proxima’ is the in- 
terpolation of a scribe who not perceiving that‘ subdita’ was 
to be repeated from the hexameter imagined ‘ Misenis’ to be 
without construction. Mr Baehrens most justly points out 
that the corruption has robbed ‘mediis Bais’ (= mediis inter 
aequora Thesproti regno subdita et aequora Misenis subdita) 
of its meaning, and, he adds, ‘effecit ut plane singulariter ei 
quod legitur u. 3 et modo desit cui respondeat’; but alas, we 
have already seen that the phenomenon is not unique even in 
Mr Baehrens’ text. In 11 14 9 sqq. the Mss order the verses 
thus: ‘nunc ligat ad caestum gaudentia bracchia loris, | missile 
nunc disci pondus in orbe rotat; | gyrum pulsat equis, niueum 
latus ense reuincit | uirgineumque cauo protegit aere caput, | 
qualis Amazonidum nudatis bellica mammis | Thermodontiacis 
turba lavantur aquis, | et modo Taygeti, crines aspersa pruina, | 
sectatur patrios per iuga longa canes; | qualis et Eurotae 
Pollux et Castor harenis’, That 15 and 16 ‘et modo...eanes’ 
are out of their place is evident from the manifest continuity 
of 17 and 18 with 18 and 14; hence Scaliger followed by 


EMENDATIONES PROPERTIANAE. 21 


Mr Baehrens places 15 and 16 after 12. But there too they 
separate lines which unmistakeably cohere (latus ense reuincit 
protegitque aere caput qualis Amazonidum turba, quae in 
Thermodonte lauatur)'; and modo remains solecistic. The 
right place for 15 and 16 is after 10: ‘missile nune disci 
pondus in orbe rotat, | et modo Taygeti’ eqs. 

To return to our starting point: not only does syntax un- 
veil a fraud, but I find too an external token that the Mss are 
cheating us. The verses 9—16 are closely imitated by Ovid, 
ars am. If 185—192. 


quid fuit asperius Nonacrina Atalanta ? 185 
subcubuit meritis trux tamen illa uiri. 
saepe suos casus nec mitia facta puellae 
flesse sub arboribus Milaniona ferunt. 
saepe tulit iusso fallacia retia. collo, 
saepe fera toruos cuspide fixit apros. 190 
sensit et Hylaei contentum saucius arcum ; 
sed tamen hoc arcu notior alter erat. 


Now here 187—190 cover the same ground as 11 and 12 in 
Propertius: 11 in Propertius has the same theme as 187 and 
188 in Ovid, the disconsolate wandering of the ill-used lover 
(see too ars am. I 731 ‘ pallidus in Dirces siluis errabat Orion’) ; 
12 in Propertius has the same theme as 189 and 190 in Ovid, 
the hard work of the chase at Atalanta’s side. But in Ovid 
the two things are duly discriminated as happening one at 
one time, the other at another: to read Propertius you would 
fancy both happened at once. To be brief, with Ovid and 
Latin grammar for guides I infer that two verses have been 
lost between 11 and 12; lost through the recurrence of ‘modo’ 
in the same part of each hexameter. I have manufactured 
these stopgaps: 


nam modo Partheniis amens errabat in antris 
multaque desertis fleuerat arboribus, 


1 So, for example, Hor. epist. 1 16 purior ambiat Hebrus’=‘ut nec fri- 
12 ‘fons etiam riuo dare nomen ido-_ gidior nec purior sit Hebrus, qui 
neus, ut nec frigidior Thracam nec Thracam ambit’, 


22 THE JOURNAL OF PHILOLOGY. 


et modo submissa casses ceruice ferebat 
ibat et hirsutas eqs. 


12. The task of essaying to shew that ‘uidere feras’ has 
any meaning suitable to this place is undertaken by Markland 
and Lachmann: truly ‘si Pergama dextris defendi possent, 
etiam his defensa fuissent’. They cite many passages and 
might have cited more to prove that ‘uidere’ can be used in 
the sense of ‘adire’ or ‘experiri’: yes, so it can, but with 
this marked limitation, that the substantive which is its 
object must signify either a place or a condition. Thus on 
the one hand you have wdere turbatum nemus, Tartara, wasto 
sub antro Scyllam, ignota flumina, alium Phasin, insanum 
forum ; on the other widere mortem, casus marinos, tanta mala, 
nihil infesti, alios menses, aliwm annum altricemque niuem 
festinaque taedia witae: thus Propertius might have said lustra 
uidere ferarum had he so chosen, but say wdere feras for 
encounter wild beasts he could not. To this conviction Hein- 
sius, Burmann and Mr Baehrens have borne witness by their 
conjectures ; and in the fulness of time the verse has been most 
acutely corrected by Mr Palmer : 


ibat et hirsutas comminus ille feras. 


This phrase is copied word for word by Ovid fast. v 176 ‘in 
apros | audet et hirsutas comminus ire feras’, and its sound 
he again echoes ex Pont. I 5 74 ‘aspicit hirsutos comminus 
Vrsa Getas’; Propertius repeats the construction If 20 22 ‘aut 
celer agrestes comminus ire sues’ though lI 1 26 he writes 
‘fluminaque Haemonio comminus isse uiro’. But how got 
‘comminus ille’ changed to ‘ille uidere’? In this way. But 
for the position of the single letter s there is virtually no 
difference between comminus and conuisum: now this trans- 
portation of a letter to some distance is a common freak in 
MSs much older and better than ours: Verg. Aen. Iv 564 
uarios MP suario Ε', georg. Iv 71 aeris M aries P; and in ours 
too: I 6 34 accepti pars ON acceptis par Prop., 11 3 18 Adriana 
ON Ariadna Prop., 29 36 uoluntatis FN uolutantis Prop., m1 13 
31 uetustas F uestitas DVN, Iv 11 53 cuius rasos O cui Sta 
suos (sacra) Prop. The scribe then who found himself confronted 


EMENDATIONES PROPERTIANAE. 23 


with the unmetrical verse ‘ibat et hirsutas conuisum ille feras’ 
preserved the sense, such as it was, of ‘ibat conuisum feras’ 
and mended the metre by writing ‘ille uidere’. 


13. Volscus amended ‘psilli’ to ‘Hylaei’ by the light of 
Ovid’s imitation quoted above ‘sensit et Hylaei contentum 
saucius arcum’; Aelian too and Apollodorus agree that Atalanta 
was assaulted by Hylaeus. Some have been dissatisfied with 
this as straying too far from the Mss, and Hertzberg has proposed 
‘ille et Phyllei’; but Phyllec rami might be the club of a 
shepherd, a satyr, a river god or Pan himself as well asa centaur, 
with nothing in the context to point the allusion. I explain 
the corruption as follows: Hylaez, written dei, was changed to 
ili: now the confusion of tlle with ipse is perpetual, as 11 4 17 
(27) alle NV ipse DF, πὶ 21 6 ile DV ipse FN, 11 28 26 illa 
ON where ipsa must in my opinion be read (ipsa sepultura facta 
beata tua): I imagine then that lz stood here in some ancestor 
of our MSS, that a reader emended it from another MS thus 


8 7 
ill and that the next copyist misunderstanding the correction 
inserted the letters ps in a wrong place and gave us psilla. 

VN uulnere, AF arbore. Lachmann has shewn that ‘percussus 
uulnere rami’ is irreproachable Latin; but that is not enough: 
the ‘arbore’ of half the Mss has to be accounted for: till that is 
done, nothing is done. This end is admirably achieved by Mr 
Baehrens’ correction ‘uerbere’, which would be corrupted with 
about equal ease into ‘arbore’ and into ‘ uolmere’: in Ouid. met. 
Iv 726 the Mss vary between ‘uulnerat’ and ‘uerberat’. The 
phrase ‘percussus uerbere’ will be illustrated by Ouid. met. XIV 
300 percutimurque caput conuersae werbere uirgae’ and Iuuen. . 
XV 21 ‘tenui percussum uerbere Circes’. 


16. Such is the efficacy in love of prayers and service ren- 
dered. Prayers! where has he said a word about prayers? 
They are not in the received text, there is no room for them in 
the lacuna which I have detected, there is no trace of them in 
Ovid’s paraphrase. Those who to defend the credit of a scribe 
will impute any imbecility to a poet are, I suppose, capable of 
maintaining that Propertius here forgot what he had just said 
and imagined he had said something else. But even this loop- 


24 THE JOURNAL OF PHILOLOGY. 


hole is blocked by the careful and orderly planning of the entire 
passage: Propertius says (9, 10) that Milanion won Atalanta 
nullos fugiendo labores ; then in 11 with the explanatory nam he 
proceeds to say what these labores were, namely (11) patient 
endurance of her cruelty, assistance (12) in the hunting field, 
hard knocks (13, 14) encountered in her defence; therefore (15) 
he won her; such (16) is the efficacy of...and deeds of merit. 
If the poet put preces in that gap, well might he cry to his 
friends ‘quaerite non sani pectoris auxilia’: the ergo of 15, the 
tantum of 16 pointedly invite attention to what has preceded, 
and there has preceded not a word, not a hint of preces. And 
yet this flagrant discrepancy has run the gauntlet of Scaliger, 
Heinsius, Hemsterhuys, Markland, Schrader and Lachmann, 
half a dozen of the greatest names in criticism, and has only 
been detected by the vigilance of Fonteine. Fonteine’s con- 
jectures are now first given to the world in Mr Baehrens’ edition: 
many of them of course are the mere guesses which we all jot 
down in our margins simply to help us take up the thread of 
thought to-morrow where we drop it to-day, and although Mr 
Baehrens does well to print them entire, still most of them are 
necessarily worthless; but the residue betoken one of the most 
acute intellects that have ever been bent on the study of 
Propertius. Fonteine then proposed instead of ‘ preces’ to read 
‘fides’, a word most appropriate in itself and strongly confirmed 
by 11 26 27 ‘multum in amore fides, multum constantia prodest:| 
qui durare potest multa, et amare potest’, where the pentameter 
too recalls the ‘nullos fugiendo labores’ of our elegy. But, it 
may be said, the change is violent. No, that is not so: the 
same confusion recurs in Tibull. [Lygd.] m1 4 64 ‘tu modo cum 
multa bracchia tende prece’ G ‘fide’ AV, 6 46 ‘aut fallat blanda 
subdola lingua prece’ G excerpt. Paris. ‘fide’ AV. Then turn 
to Hor. ars poet. 395 ‘dictus et Amphion, Thebanae conditor 
urbis, | saxa mouere sono testudinis et prece blanda | ducere 
quo uellet’: even if you do not feel that ‘ prece’ is a trifle ludi- 
crous, still you will confess it is unique: neither Amphion nor 
Orpheus is elsewhere depicted as beseeching stocks and stones 
to follow him: the dead things are brought by the mere charm 
of song. Having regard then to carm. I 12 11 ‘blandum et 


EMENDATIONES PROPERTIANAE. 25 


auritas fidibus canoris | ducere quercus’ and 24 13 ‘quid si 
Threicio blandius Orpheo | auditam moderere arboribus fidem’ 
I follow Peerlkamp when he reads ‘ fide blanda’. 

19. I conceive that so far as Latinity is concerned the 
words ‘ deductae fallacia lunae’ may bear any one of three mean- 
ings. First, they may mean false pretence of bringing down the 

‘moon: a sense peremptorily forbidden by the context. Mr 
Lucian Mueller points out that Propertius cannot look for help 
to those whom he holds and asserts to be impostors, and that 
this argument is clinched by the ‘tunc ego crediderim’ of 23: 
Propertius now doubts whether the power of magic be real or 
no, but turn Cynthia’s heart and he will believe. Secondly 
then, ‘ deductae fallacia lunae’ may legitimately mean deceiving 
men by bringing down the moon on the analogy of Ouid. met. 
ΧΙΠ 164 ‘deceperat omnes, | in quibus Aiacem, swmptae fallacia 
uestis’. But plainly this sense is no better than nonsense: if 
magicians bring down the moon as men believe them to do, then 
men are not deceived. Equally absurd is the third possible 
sense of the words, deceiving the moon and bringing her down. 
I know that ‘Pan deus Arcadiae captam te, Luna, fefellit in 
nemora alta uocans’, but in what sense do magicians fallere 
lunam ? what conceivable deceit can they employ? manufacture 
a ‘cerea effigies’ of Endymion I suppose and lay it out on moun- 
‘tain-tops. The truth is that those who read and fancy they 
understand this passage translate ‘deductae quibus est fallacia 
lunae’ as Mr Postgate does, ‘ who bewitch the moon into coming 
down’. But the words cannot bear that meaning. Bewitchment 
comprises several departments, and of these departments fallacia 
is one: Prop. Iv 5 14 ‘sua nocturno fallere terga lupo’, Ouid. » 
met. ΠῚ 1 ‘iamque deus posita fallacis imagine tauri | se con- 
fessus erat’, Verg. Aen. I 683 ‘tu faciem illius noctem non amplius 
unam | falle dolo et notos pueri puer indue uoltus’, georg. IV 441 
‘omnia transformat sese in miracula rerum, | ignemque horribi- 
lemque feram fluuiumque liquentem. | uerum ubi nulla fugam 
reperit fallacia, uictus | in sese redit’: there you have bewitch- 
ment which is fallacia. But it does not follow that you can 
use fallacia in season and out of season as an equivalent for 
bewitchment. To lure the moon from heaven, ghosts from the 


26 THE JOURNAL OF PHILOLOGY. 


grave, the standing corn from a neighbour’s field, is not fallacia 
but, as L. Fruterius and J. M. Palmerius 300 years ago perceived, 
pellacia. Seruius on Verg. buc. ΨἼΠ 99 quotes from the twelve 
tables ‘neue alienam segetem pellexeris’, Pliny hist. nat. XVIII 
6 8 § 41 has ‘ceu fruges alienas pelliceret ueneficiis’. Now to 
shew the facility of the corruption: Verg. Aen. 11 90 pellacis M, 
Velius Longus, Donatus, Seruius, fallacis P, Charisius ; georg. IV 
443 fallacia PRV, phallacia M, pellacia ὃ (cod. Bern. saec. IX), 
‘fallacia, legitur et pellacia’ Philargyrus. Munro on Lucr. II 
559 ‘placidi pellacia ponti’ says ‘Virgil has the adj. pellax: 
these two appear to be the only good writers who use the words’: 
yes, but it is appearance only: if the Mss of Horace were as 
trusty in such matters, or the Mss of Propertius in any matters, 
as the mss of Virgil and Lucretius, it would be another story. 
Horace in carm. ΠῚ 7 professes to tell Asterie news of her absent 
Gyges: he lies awake all night weeping for her; and yet his 
hostess Chloe is in love with him and her minister tempts him 
by recounting her sighs, tells him what peril Bellerophon and 
Peleus incurred through continence, ‘et (19) peccare docentis | 
fallax historias monet’. Now fallere can indeed signify seduction 
followed by desertion, but it is of course always the woman who 
in this sense fallitur, not the man: the reverse is absolutely 
meaningless in Greece, Rome or England. In these lines of 
Horace fallax can have but one meaning: it must mean that 
the ‘ nuntius’ intends ‘ mentiri noctem, promissis ducere aman- 
tem’, thus flatly disobeying the ‘sollicita hospita’ who sent him 
on his errand, and giving Gyges no chance to put his ‘ constans 
fides’ to the proof: the whole poem is stultified. Write pellaa 
with Bentley and all is straightforward. Finally in Prop. Iv 1 
135 we read ‘at tu finge elegos, fallax opus, haec tua castra, | 
scribat ut exemplo cetera turba tuo’: well, it is true that ‘docere 
qua nuptae possint fallere ab arte uiros’ is part of the office of 
elegy, but who could catch the allusion here with nothing to 
point it? the phrase would more naturally mean ‘a slippery 
task’; a sense which is most inappropriate. The pentameter 
speaks loud for Heinsius’ pellax, ‘a fascinating, alluring task’; 
and it seems to me that Ovid with his ‘imbelles elegi, genialis 
musa’ imitates ‘elegos, pellax opus’. To return then to the 


EMENDATIONES PROPERTIANAE. 27 


first elegy, I can feel not the slightest doubt that ‘deductae 
pellacia lunae’ is what Propertius wrote. The construction of 
the sentence, I should add, is rightly explained by Mr F. Leo in 
vol. 35 of the Rheinisches Museum as ‘uos quibus labor est de- 
ductae fallacia [pellacia] lunae et alter labor sacra piare’. 

But what is ‘sacra piare’? Hemsterhuys tells us ‘sacra 
prare usu uetusto nihil aliud quam sacra pie sollemnique ritu 
facere’. But Mr Lucian Mueller and others have rightly observed 
that ‘sacra piare’ in this sense is no peculiar office of magicians 
but common to all sacerdotes and indeed to the head of every 
Roman household. The mention is demanded of some magic 
portent answering the ‘pellacia lunae’ of the preceding verse. 
What this portent should be we shall be better able to judge 
when we have discussed verses 23 and 24. 

Sidera et amnes ducere carminibus. Scores of times, when 
the ancients tell us of the wonders wrought by magic or by 
music, do they employ the verb ducere, its compounds and its 
synonyms; scores of times do they employ the substantive 
amnes and the other substantives which mean streams and 
rivers: never, save in this single place, do they employ the 
phrase amnes ducere. Here are the dealings of magic with 
rivers: Verg. Aen. Iv 489 ‘haec se carminibus promittit... 
sistere aquam fluuiis’, Tibull. 1.2 46 ‘fluminis haec rapidi 
carmine uertit iter’, Ouid. am. 1 8 5 ‘illa magas artes Aeaeaque 
carmina nouit | inque caput liquidas arte recuruat aquas’, 
m 1 25 ‘carmine dissiliunt abruptis faucibus angues | inque 
suos fontes uersa recurrit aqua’, her. VI 87 ‘illa refrenat aquas 
obliquaque flumina sistit’, met. VII 153 ‘uerbaque ter dixit... 
quae concita flumina sistunt’, 198 ‘adeste | quorum ope, cum 
uolui, ripis mirantibus amnes | in fontes rediere suos’, remed. 
amor. 257 (he disclaims magic) ‘ ut solet, aequoreas ibit Tiberi- 
nus in undas’, Petron. 134 ‘his ego callens | artibus Idaeos 
frutices in gurgite sistam | et rursus fluuios in summo uertice 
ponam’, Sen. Med. 763 ‘cantu meo...uiolenta Phasis uertit in 
fontem uada | et Hister in tot ora diuisus truces | compressit 
undas omnibus ripis piger’, Luc. Phars. vi 472 ‘de rupe pepen- 
dit | abscissa fixus torrens, amnisque cucurrit | non qua pronus 
erat’, Sil. Punic. vit1 502 ‘ Aeetae prolem...stridoribus amnes | 


28 THE JOURNAL OF PHILOLOGY. 


frenantem’, Val. Fl. Arg. vi 443 ‘ mutat agros fluuiumque uias’, 
Claud. in Rufin. 1 159 ‘uersaque non prono curuaui flumina 
lapsu | in fontes reditura suos’, Appul. met. I 3 ‘magico susur- 
ramine amnes agiles reuerti’, 8 ‘saga, inquit, et diuina, potens... 
fontes durare, montes diluere’, Apoll. Rhod. Arg. πὶ 532 καὶ 
ποταμοὺς ἵστησιν ἄφαρ κελαδεινὰ ῥέοντας. Here are the deal- 
ings of music with rivers: Verg. buc. VIII 4 ‘quorum stupe- 
factae carmine lynces | et mutata suos requierunt flumina 
cursus’, Hor. carm. I 12 9 ‘arte materna rapidos morantem | 
fluminum lapsus’, Prop. 11 2 3 ‘Orphea delenisse feras et 
concita dicunt | flumina Threicia sustinuisse lyra’, Ouid. fast. 
1 84 ‘quae nescit Ariona tellus? | carmine currentes ille tene- 
bat aquas’, met. XIV 338 ‘et mulcere feras et flumina longa 
morari | ore suo uolucresque uagas retinere solebat’, Calpurn. 
11 15 ‘affuerunt sicco Dryades pede, Naides udo, | et tenuere 
suos properantia flumina cursus’. Thus amnes sistere, amnes 
uertere, come over and over again: amnes ducere never. Now 
this cannot be accident, for lunam ducere, sidera, segetes, wmbras, 
saxa, quercus ducere, are for ever recurring: what then is the 
reason? The reason is the simplest in the world. Music and 
magic work miracles, invert the order of nature: thus Lucan 
Phars. vI 437 ‘Haemonidum...quarum, quidquid non creditur, 
ars est’: this he proceeds to illustrate, ‘calido producunt nubila 
Phoebo | et tonat ignaro caelum Ioue’, ‘uentis cessantibus 
aequor | intumuit: rursus uetitum sentire procellas | conticuit 
turbante noto, puppimque ferentes | in uentum tumuere sinus’, 
‘Nilum non extulit aestas, | Maeander derexit aquas, Rhoda- 
numgque morantem | praecipitauit Arar. submisso uertice mon- 
tes | explicuere iugum: nubes suspexit Olympus. | solibus et 
nullis Scythicae, cum bruma rigeret, | dimaduere niues’; so 
Appuleius |. 1. ascribes to his witch the power ‘caelum deponere, 
terram suspendere, fontes durare, montes diluere, manes subli- 
mare, deos infimare, sidera extinguere, Tartarum ipsum illumi- 
nare’. Such miracles, such inversions of nature, amnes sistere 
and amnes uertere are; but amnes ducere is nothing of the sort: 
it is one of the commonest operations of Italian agriculture : 
Virgil’s graceful picture of the process is familiar to everyone. 
A man would no more dream of invoking incantations to amnes 


EMENDATIONES PROPERTIANAE. 29 


ducere than to shave his chin or cook his dinner; and when 
this every-day work of the farmer is coupled with the ‘sidera 
ducere’ of the magician, the absurdity is doubled. There are 
those who, if we had ‘amnes et sidera ducere’, would take sanc- 
tuary at the shrine of Zeugma and pretend that ‘sistere’ or 
‘uertere’ might be mentally supplied to ‘amnes’; but as ill 
luck will have it the order of the words is ‘sidera et amnes 
ducere’ and retreat in that direction is cut off. Propertius then 
did not write what the Mss give: what did he write? No feat 
of magic is more renowned than the evocation of departed 
spirits: Lucan in Phars. v1, Statius in Theb. Iv, Silius in Punic. 
XIII raise the dead to life till they tire the reader to death, and 
Valerius Flaccus has a brief episode of the sort at the end of 
Arg. 1: pages might be filled with allusions scattered through- 
out the poets, but I here content myself with passages where 
the power of magic over the dead is coupled with its power over 
the heavenly bodies. Such are Verg. Aen. Iv 489 ‘haec se 
carminibus promittit...uertere sidera retro, | nocturnosque 
mouet manis’, Hor. epod. 17 78 ‘polo | deripere lunam uoci- 
bus possim meis, | possim crematos excitare mortuos’, Tibull. 
I 2 45 ‘hanc ego de caelo ducentem sidera uidi, | fluminis haec 
rapidi carmine uertit iter, | haec cantu finditque solum manesque 
sepulcris | elicit et tepido deuocat ossa rogo. | iam tenet infernas 
magico stridore caterwas,|iam iubet aspersas lacte referre 
pedem’, Ouid. am. 1 811 ‘sanguine, siqua fides, stillantia sidera 
uidi, | purpureus Junae sanguine uultus erat...17 euocat anti- 
quis proauos atauosque sepulchris | et solidam longo carmine 
findit humum’, met. vil 205 ‘iubeoque tremescere montes | et 
mugire solum manesque exire sepulcris;| te quoque, Luna, 
traho’, remed. amor. 253 ‘me duce non tumulo prodire iube- 
bitur umbra, | non anus infami carmine rumpet humum, | non 
seges ex aliis alios transibit in agros | nec subito Phoebi pallidus 
orbis erit. | ut solet, aequoreas ibit Tiberinus in undas, | ut 
solet, in niueis Juna uehetur equis’, Sen. Herc. Oet. 460 ‘mea 
iussi prece | manes loquuntur...468 carmine in terras mago | 
descendat astris luna desertis licet’, Val. Fl. Arg. vi 447 
‘quamuis Atracio Junam spumare ueneno | sciret et Haemoniis 
agitari cantibus wmbras’, Claud. in Rufin. I 146 ‘noui, quo 


30 THE JOURNAL OF PHILOLOGY. 


Thessala cantu | eripiat Jwnare iubar...154 saepius horrendos 
manes sacrisque citaui | nocturnis Hecaten, et condita funera 
traxi | carminibus uictura meis’. To these passages I should 


add 


tune ego crediderim et manes et sidera uobis 
posse Cytinaeis ducere carminibus, 


Verg. Aen. IV 34 manis GMPR amnis F, 490 manis MP amnas F, 
Ir 296 manibus FMP amnibus V: in our Mss too this inversion 
of two consecutive letters is frequent: 1 3 27 duait ON for duct, 
14 24 alciont D for Alcinor, 11 6 6 Phyrne DVN for Phryne,8 39 
marte ON for matre, 13 55 paludibus DVN plaudibus F, 28 29 
herodias DV for heroidas, 111 5 35 palustra F for plaustra, 7 61 
alcinoum F alcionum DV, 13 24 ipa F for pia, 55 et ON for te, 
15 41 parta for prata, Iv 5 74 caltra for clatra. When the 
unmetrical ‘ crediderim et amnes et sidera uobis’ was thus pro- 
duced, there was nothing for it but to arrange the words as 
they stand in the Mss to-day: similar transpositions for metre’s 
sake will be found at 1 9 18, 10 2,13 38, 23 1, 1v 2 39,3 7,81. 

In 24 the good mss have Cytalinis, Citalinis, Cythalinis, 
Cithalinis, which all come to the same thing and have all alike 
no meaning ; the bad Mss have the impossible forms Cytaeinis 
or Cytainis: scholars have conjectured Cytaei tuis, Cytaeaeis, 
Cytaines, Cytacacis. But the correction which is at once nearest 
to the Mss and most appropriate in sense is Hertzberg’s Cyti- 
naeis. Hertzberg, whose confidence in his own conjectures 
usually bears an inverse proportion to their value, did not place 
it in his text; and it seems thus to have escaped subsequent 
editors, until Mr Postgate, who again at 116 13 has recalled 
an excellent but neglected emendation of Scaliger’s, has most 
properly accepted it. I say ‘CytIN-AEis’ is nearer to ‘CytAL- 
INis’ than are any of the other conjectures, because this per- 
mutation of syllables is one of the commonest phenomena: the 
first instances which occur to me are Verg. Aen. x1 711 rapu M 
for pura and Hor. carm. I 36°17 trespu 6 for putres: so in Prop. 
12131 write su-per-a-nt-de-picta for per-su-a-de-nt-picta: but 
I will now cite only examples where this change is accompanied 
by the change of one letter, as here of E into L: τι 82 17 


EMENDATIONES PROPERTIANAE. 31 


falleris ON for fallis et, 34 53 restabit eritipnas F restawerit undas 
DV, 11 5 24 sparsit et F sparserzt DV, 23 21 retulzt et F rettule- 
rit DVN, tv 1 106 umbraque ne O for umbrawe quae. As to the 
word, Hertzberg cites Lycophr. 1389 Λακμώνιοί τε καὶ Κυτι- 
ναῖοι, Kédpor and Steph. Byz. Κύτινα᾽ πόλις Θεσσαλίας, ὡς 
Θέων ἐν ὑπομνήμασι Λυκόφρονος" ὁ πολίτης Kutwaios: now 
in the palinode to this elegy 111 234 9 and 10 you have ‘quod 
mihi non patrii poterant auertere amici | eluere aut uasto 
Thessala saga mari’. And this emendation will at once confirm 
my correction of the hexameter and derive confirmation thence. 
Necromancy, above all other forms of magic, was Thessalian : 
see Stat. Theb. m1 141 ‘Thessalis...cui gentile nefas hominem 
reuocare canendo’ and 559 where ‘Thessalicwm nefas’ stands 
kat ἐξοχήν for necromancy. 

So we have settled the reading of 23 and 24: now we are 
better equipped for discussing ‘sacra piare’ in 20. Just 88. 
‘deductae pellacia lunae’ tallies with ‘sidera ducere’ so we 
shall expect ‘sacra piare’ to tally with ‘manes ducere’. This 
expectation will be strengthened if we observe how frequently 
pare is used with manes or the like: Iv 7 34 ‘fracto busta 
piare cado’, Verg. Aen. v1 379 ‘ossa piabunt’, Ouid. fast. v 426 
‘compositique nepos busta piabat aui’, met. vi 569 ‘piacula 
manibus infert’, x11I 514 ‘hostilia busta piasti’, Cic. in Pison. 
16 ‘a me quidem etiam poenas expetistis, quibus coniuratorum 
manes mortuorum expiaretis’: in Petron. 137 Burmann per- 
haps rightly reads ‘expiare manes pretio licet’ for ‘manus’. 
That manes piare would be a natural accompaniment of manes 
ducere is shewn by Cic. in Vatin. 14 ‘cum wmferorum animas 
elicere, cum puerorum extis deos manes mactare soleas’. Now 
turn to ΠῚ 1 1 ‘Callimachi manes et Coi sacra Philetae, | in 
uestrum, quaeso, me sinite ire nemus’. You cannot ask the 
sacred rites of Philetas for permission to do this or that, least 
of all when in the same breath you address the same request to 
the spirit of Callimachus. I hold it to be as certain as aught 
in these matters can be that in 1 1 20 and 1 1 1 either ‘sacra’ 
means ‘manes’ or else it is the corruption of another word 
which means ‘manes’. That ‘sacra’ stood for ‘manes’ was 
maintained by Dousa in the former place and by Broukhusius 


32 THE JOURNAL OF PHILOLOGY. 


in the latter; but this contention they entirely failed to estab- 
lish by examples, and indeed it seems inconceivable that ‘sacra’ 
could come to have any such signification. So I infer that 
‘sacra’ is in both places a corruption of the same word; and 
that word I think has been restored by Fonteine in the one 
place and by Mr Baehrens in the other: ‘fata’, No two words 
I suppose are more commonly confused than fata and facta: 
see 11 28 26, Iv 1 71, 11 70: and how easily ‘facta’ would 
become ‘sacra’ may be seen from the following blunders all 
culled within the compass of seven lines: 29 ferre A for ferte, 
30 semina A for femina, 31 remanere A for remanete, 34 desit 
AN for defit, 35 uitare F for uitate. Forcellini cites Mela for 
fatum = umbra, but I think we can find better authority than 
Mela. In Hor. carm. I 24 15 sqq. we read ‘num uanae redeat 
sanguis imagini, | quam uirga semel horrida | non lenis precibus 
fata recludere | nigro compulerit Mercurius gregi?’ and it is 
usual to explain ‘fata recludere’ with Lambinus as ‘fati nexus 
et necessitatem resoluere ac rescindere’. But there is no sem- 
blance of authority for sucha use of ‘recludere’: if ‘fata’ here 
means destiny, then the words ‘fata recludere’ can only mean 
what ‘pandere fata’ means in Luc. Phars. vi 590, namely 
‘aperire futura’; and this meaning is totally foreign to the 
context. I believe then that ‘fata recludere’ =‘ Orcum reclu- 
dere’: the lexicons will shew that the use of ‘recludere’ and 
its synonyms in regard of the infernal regions is very frequent. 
Again in Luc. Phars. v1 652 ‘nam quamuis Thessala uates | wim 
faciat fatis, dubium est, quod traxerit illue, | aspiciat Stygias, 
an quod descenderit, umbras’ the context seems to indicate 
that fatis = manibus. I think then that Propertius enjoys his 
own again when Fonteine writes ‘in magicis fata piare focis’. 
25. Hemsterhuys’ correction ‘aut’ for ‘et’ is adopted by 
Lachmann, Hertzberg, Haupt, Mr L. Mueller and Mr Baebrens: 
three English editors, Mr Paley, Mr Palmer and Mr Postgate, 
all retain ‘et’ and do not seem to have the faintest suspicion of 
its entire absurdity. ‘I am surprised’ Mr Paley gravely says 
‘that Lachmann, Hertzberg, Miiller, and Kuinoel showld have 
admitted, and Jacob approved, aut wos, the conjecture of Hem- 
sterhuis’; Mr Postgate acquiesces: ‘et, as Paley rightly with 


EMENDATIONES PROPERTIANAE. 33 


the mss for aut edd.’; Mr Palmer reads ‘et’ in silence. These 
three scholars award the poetry of Propertius commendation 
which I think too high; yet they impute to him without scruple 
the stupidity of praying that Cynthia may begin to love him 
and that he may cease to love Cynthia. If the impossibility of 
the MS reading is not made plain by this naked statement of its 
sense, pages of argument will be vain; nor can I hope that 
those who are deaf to Hemsterhuys will listen to me. 

33. I agree with Mr Baehrens that 34 means Cupid is 
never idle, never absent: if you take ‘uacuus amor’ to be un- 
gratified passion you get the wonderful circumlocution absence 
of fruition is never absent. If then Amor in the pentameter is 
the god, Venus in the hexameter is the goddess. But ‘nostra’ 
has thus no meaning, and is altered accordingly by Francius to 
‘dura’ and by Mr Baehrens to ‘ maesta’: the latter appositely 
quotes I 14 15 ‘nam quis diuitiis adwerso gaudet Amore? | nulla 
mihi tristt praemia sint Venere’. There is however another 
difficulty unremoved: it must I think be conceded to Mr Post- 
gate that ‘in me noctes exercet’ is harsh and quite unexampled. 
I propose to abolish both difficulties at once by this very slight 
alteration : 


- 


me non nostra Venus noctes exercet amaras 


‘noctes amaras’ being then ace. of duration like Horace’s ‘longas 
pereunte noctes’, The use of meus tuus suus noster uester = 
secundus is well known: in 111 13 56 Mr Baehrens rightly reads 
with O ‘te scelus accepto Thracis Polymestoris auro | nutrit in 
hospitio non, Polydore, two’ against the ‘pio’ of N; and this 
makes it all the stranger that he should miss the same use in 
τ 25 31. There we find ‘tu tamen interea, quamuis te diligat 
illa, | in tacito cohibe gaudia clausa sinu; | namque in amore 
suo semper sua maxima cuique | nescio quo pacto uerba nocere 
solent. | quamuis te persaepe uocet, sepelire memento’; and 
Mr Baebrens to my surprise says ‘malim nouo’: suo = secundo, 
see 27 ‘mendaces ludunt flatus in amore secundi’. Thus then 
‘non nostra Venus’ is ‘aduersa Venus’. The change of non, 
abbreviated n, to in is easy: in IV 5 9 F has tinducere for. non 
ducere, and in Iv 1 124 I think all our ss have the same error: 


Journal of Philology. vou. xvt. 3 


34 THE JOURNAL OF PHILOLOGY. 


‘qua nebulosa cauo rorat Meuania campo | et lacus aestiuis 
intepet Vmber aquis’. Since ‘intepet’ is not a Latin word 
Mr Lucian Mueller writes ‘si tepet’, Mr Baehrens ‘ut tepet’: 
better than either, if I am not mistaken, will be ‘non tepet’. 
Their chill in the heat of summer is the natural praise of 
streams and lakes: ‘te flagrantis atrox hora Caniculae nescit 
tangere, tu frigus amabile fessis uomere tauris praebes’; and 
Pliny epist. vill 8 describing the source of the Clitumnus, 
perhaps this very lacus Vmber, says ‘rigor aquae certauerit 
niuibus, nec color cedit’. The alteration of ‘me in nostra 
Venus’ to ‘in me’ would be demanded by grammar and metre 
alike, and would perhaps be helped by the occurrence of ‘in 
me’ at the beginning of verse 17. 

Of Mr Postgate’s conjecture ‘in me nostra Venus uoces 
exercet amaras’ ‘against me my darling plies her bitter speech ’ 
I am at a loss to know what to say. There is some justice in 
Mr Baehrens’ contention that the estranged and obdurate 
Cynthia can hardly be called by the endearing name ‘nostra 
Venus’; but that is nothing: the alteration makes nonsense of 
the whole elegy from beginning to end. Mr Postgate tells his 
readers on p. XXII of his Introduction that Lachmann’s explana- 
tion of the circumstances of this poem seems to him unquestion- 
ably correct: he holds, that is, and in my opinion rightly holds, 
that this poem was written when Propertius had been banished 
from Cynthia’s presence for a year; and yet he makes her ‘ply 
her bitter speech against’ Propertius, from whom she was as 
many miles asunder as Hypanis is from Eridanus of the Veneti! 
If this is the attention to context with which conjectural emen- 
dation is practised, no wonder that many students of the classics 
regard it as a game played merely for the amusement of the 
conjectural emendator. The corruption is not even new: ‘uoces’ 
is the reading of the codex Hamburgensis, where everyone 
hitherto has left it lying justly contemned among a hundred 
other blunders almost equally worthless. But I imagine that 
these considerations will have occurred ere now to Mr Postgate 
himself, or will have been pointed out to him by his friends. 

36. So far as I am aware no one has even attempted to 
prove by examples that mutare locum can have the metaphorical 


EMENDATIONES PROPERTIANAE, 35 


sense to be inconstant which is here required; nor do I believe 
there are examples to prove it. And even were such a sense 
established, still Propertius durst not employ it here, A reader 
fresh from ‘ferte per extremas gentes et ferte per undas’ and 
“uos remanete’ must necessarily at first sight be tempted to 
take ‘mutet locum’ literally ; indeed in some commentaries it 
is actually so explained, despite the hopeless shipwreck of sense 
which such an interpretation causes. Markland with his usual 
acumen first detected the fault; but his alteration of ‘locum’ 
to ‘nouum’ is a violent remedy. Mr A. Otto in the Philolo- 
gische Wochenschrift for 1884 has proposed ‘torum’ comparing 
Iv 8 28 ‘mutato uolui castra mouere toro’, and five or six years 
ago I noted down the same conjecture and the same parallel: 
this I mention not as wishing to wrangle with Mr Otto for 
ownership but merely because some weight is justly given to 
such coincidences, The confusion of ὁ and ὦ I need not ex- 
emplify ; for r and c see 11 6 21 capere F for rapere, 25 45 
sandyris V for sandicis, 34 33 rursus ON for cursus, UI 6 36 
cursu DV for rursus, Iv 1 83 capacis F for rapacis, 89 Arria FN 
Accia DV, 4 12 foco ON for foro, 72 fertur for pectus, 8 53 
recidere D for cecidere. In Sen. Here. Fur. 21 ‘escendat licet | 
meumque uictrix teneat Alemene locum’ I think ‘ escendat’ 
shews Bentley’s ‘torwm’ to be right. Then in Ouid. met. XI 
471 sqq. ‘ut nec uela uidet, uacuum petit anxia lectwm | seque 
toro ponit. renouat lectusque locusque | Alcyonae lacrimas et, 
quae pars, admonet, absit’ will any student of Ovid's style deny 
that the parallelism requires ‘lectusque torusque’? if he does, 
let him mark the next verse and ask himself whether pars loci 
or pars tori is the better sense. 


A. E, HOUSMAN, 


October, 1886. 


MISCELLANEA CRITICA, 


TERENCE Eun, 4. 4. 31. 


Py. Hic est vetus, vietus, veternosus senex, 
Colore mustelino. PH. Hem, quae haec est fabula ? 


Bentley has a characteristic note on this passage proposing 
‘stellionino for mustelino, because weasels or cats (mustelae) are 
not all of the same colour, nor the same colour on all parts of 
the body. It is difficult to see how an old man could be com- 
pared to a lizard’. But Bentley thought he was backed up by 
the remark of Donatus that Menander here had yaXewrns* a 
lizard, stellio, and he supposed that Terence either mistook it 
for γαλῆ, mustela, or intentionally substituted the latter word. 
Others have surmised that Menander wrote γαλεώδης (γαλεός--- 
εἶδος), but it is more unlikely still that an old man would be 
likened to a fish. I suggest that Menander did write γαλεώδης 
γέρων but that γαλεώδης was here formed, comically, from γαλῆ---- 
of#—and that Terence either mistook it to come from yaAn7— 
εἶδος or what is more likely wrote odore (or cum odore) mustelino, 
Readers of Aristophanes do not need to be reminded of the 
smell of the γαλῆ. 


Heauton Timorumenos 4. 1. 32. 


Mi Chremes, peccavi, fateor: vincor. Nunc hoc te obsecro 
Quanto tuus est animus natu gravior, ignoscentior, 
Ut meae stultitiae in iustitia tua sit aliquid praesidi. 


1 Liddell and Scotts. γαλεώτης trans- cat,’ but I do not see that this render- 
late y. γέρων an old man ‘as grey as a__ing is justified, 


MISCELLANEA CRITICA., 37 


In the second line I read: 
Quanto tuust animus natura gravior, fi ignoscentior. 


Bentley saw that an imperative was wanted before zgnos- 
centior, but he quite rewrote the verse to bring in tanto es. The 
imperative fi is used quite thus: Plaut. Cure. 1. 1. 87: fi mi 
obsequens: Pers. 1. 1. 39: fi benignus. 


PLAUTUS, Curculio 5. 15. 


TH. Nec mihi quidem libertus ullus est. Ly. Facis sapientius 
Quam pars lenonum libertos qui habent et eos deserunt. 


I do not here address myself to the question whether Jatro- 
num should be read for lenonum or not. <A difficulty occurs to 
me, which seems not to have struck anyone else, as to sapientius. 
It is evident that wisdom is not at stake when a man deserts 
his freedman: that may or may not be wise according to cir- 
cumstances. But it is always a breach of duty. I thercfore 
propose : 

facis pientius 
Quam pars lenonum(?) libertos qui habent et eos deserunt. 


1 am aware that pientiws occurs nowhere else. But if a 
comparative of pie existed in the time of Plautus, that compara- . 
tive certainly was pientius. Cicero, as is well known, censured 
Mark Antony for using the superlative piissimus, and the super- 
lative pientissimus is frequently found in inscriptions. This 
would point to a comparative pientior, pientius. As to the 
propriety of pie with desero there can be no question. Cf. Ter. 
Ad. 3, 4. 13: Si deseris tu periimus—Cave dixeris: Neque 
faciam, neque me satis pie posse arbitror. : 


Persa 434 [3. 3. 30]. 


Ubi quid credideris, citius extemplo a foro 
Fugiunt, quam ex porta ludis quom emissust lepus. 


One of Plautus’s many jibes at the frequent suspensions of 
payment on the part of Roman bankers. What gate is meant 
in the second line? The gate of the circus, says Ussing. Whither, 
then, did the hare run so fast? Into the arena? This would 


38 THE JOURNAL OF PHILOLOGY. 


be to meet his foes, the dogs: for surely the hare was enlarged 
for a coursing match. Or away from the circus, outwards? This 
would be prevented, and the spectators would not see it, even if 
it did occur. I suggest that the hare was carried in a cage or 
hamper into the middle of the arena, and enlarged there for the 
course: and I read ex sporta. The sporta was a box or hamper 
of wicker-work used for various purposes, and doubtless of 
various sizes: fishing-baskets for instance were called sportae: 
fowls were fattened shut up in sportae, so tightly shut in, as only 
to be able to exsert their beaks to peck their food: a fragment 
of Sallust tells us that on some occasion unknown to us dogs 
were let down from walls in sportae: though in this passage it 
has been proposed to read panes for canes without the slightest 
warrant. Acidalius was doubtless right in reading in Mere. 5 
4. 28 Jam obsecro hercle habete vobis cum sportis cum fiscina 
where the mss. give the utterly unintelligible porcis. ‘Take 
and keep her baskets and hamper!’ the metaphor being perhaps 
from the sale of a donkey or mule. 


Persa 815 [5, 2. 39]. 


Do. Ego pol vos eradicabo. PA. At te 1116, qui supra nos habitat 
Qui tibi male volt maleque faciet. Non hi dicunt, 
verum ego. | 


The first verse must be scanned, if the above reading is 
sound, as an octonarius. But the next verse and those which 
follow are septenaril. We should probably read habet for habitat, 
in the same sense, a sense which habet frequently bears in 
Plautus. It may be necessary to give supera for supra, but 
supra may have been the only scansion known to Plautus. 


Rudens 847 [3. 6. 20]. 


Post huc redito atque agitato hic custodiam. 
Ego hunc scelestum in ius rapiam texulem +. 
Age, ambula in ius. 


Plesidippus drags Labrax before the praetor, to institute 
proceedings for the recovery of Palaestra whom he had bought, 
and given a deposit for, but whom the leno was attempting to 


MISCELLANEA CRITICA. 39 


earry off by force. Such an action would be called, in Greek 
I think, ἐξούλης Sixn*, and in the corrupt erulem we should, I 
think, recognise ἐξούλης. There are several Greek words in the 
play translitterated, as anancaeo 361. Then dica may have 
fallen out at the end of the line under diam at the end of the 
previous line: nothing is commoner than this. 


Ego hunc scelestum in ius rapiam, exules dica 
or dicam in apposition with the sentence. dicam would be quite 


right here: see Aul. 4. 10. 30: Iam quidem hercle te ad prae- 
torem rapiam, et tibi scribam dicam. 


Rudens 811 [8. 5. 42]. 
Iam hoc Herculi est Veneris fanum quod fuit. 


Read fit for est instead of inserting modo after Veneris with 
Ussing. fio is proper in metamorphoses. 


Trinummus 885 [4. 2. 43]. 


Si ante lucem ire occipias a meo primo nomine 
Concubium sit noctis prius quam ad postremum perveneris, 


This sycophant must have had even more names than the 
pious farmer, who chose twenty-six names from Scripture, be- 
ginning with every letter of the alphabet, wished to give his 
son. We shall greatly improve tle metre and add to the hyper- 
bole by inserting ztere after ire: ‘to go on a journey’ from his 
first name to his last is surely more forcible than simply ‘to go.’ 
I say after ire, because there is a lacuna in B, of six letters 
according to Ritschl. The archaic ablative itere may easily have 
been omitted after ire. Ussing inserts Ritschl’s maid of all 
work, hercle, after 176. 


Truculentus 503 [2. 6. 27]. 


Iam magnust? iamne eit ad legionem? ecquae spolia rettulit ? 


ASTAPHIUM. 
Ere nudius quintus natus quidem illic est. St. Quid postea? 
1 Harpocration quotes ἐξούλης from τάττεται τοὔνομα καὶ οὐχ ὡς οἴεται Και- 


the comic poet Phrynichus : ὅτι δὲ ἐπὶ κίλιος μόνος τῶν ἐκ καταδίκης ὀφειλόντων, 
παντὸς τοῦ ἐκ τῶν ἰδίων ἐκβαλλομένου καὶ Φρύνιχος ἐν Προαστρίαις δῆλον ποῖει. 


40 THE JOURNAL OF PHILOLOGY. 


Spengel’s reading of the first line as given above seems 
correct. It appears strange that Gerrae! ‘nonsense!’ does not 
seem to have struck anyone for the corrupt ere in the second 
verse’. 


EuRIPIDES, Medea 886. 


ἡ χρῆν μετεῖναι τῶνδε τῶν βουλευμάτων 
καὶ ξυμπεραίνειν καὶ παρεστάναι λέχει 
νύμφην τε κηδεύουσαν ἥδεσθαι σέθεν. 


νύμφην κηδεύουσαν is, as Mr Verrall remarks, a not easily 
intelligible expression. Probably we should read κηπεύουσαν. 
There would be intentional bitter coarseness in both ξυμπε- 
paivew and κηπεύουσαν, though both words in another context 
might be quite free from anything of the kind. κηπεύειν is a 
Warigidean word; here it seems that the beginning of ἥδεσθαι 
immediately fallen caused the change of κηπ- to κηδ-. 


A. PALMER. 


PS. Dr Jackson has kindly sent me the following in- 
teresting letter : ? 
Crort Corracr, Barron Roap, CaMBRIDGE, 
1 Dec., 1886. 
MY DEAR SIR, 

Here is a parallel to one of the emendations which you 
sent me the other day. “For these critics have often presumed 
that that which they understand not is false set down: as the 
Priest that where he found it written of St Paul, Demissus est 
per sportam mended his book, and made it Demissus est per 
portam, because sporta was an hard word, and out of his reading.” 
Bacon, Advancement of Learning Il xix=p. 414, Ellis and 
Spedding. 

Yours very truly, 


HENRY JACKSON. 


1 Erre (éppe) is another possibility. Cf. apage. 


LIFE AND POEMS OF JUVENAL. 


Ir is sometimes necessary to distinguish between the 
position which an author holds in the world of letters at large, 
and that which a nearer consideration of the circumstances 
of his life and times would dispose the student of history 
to assign him. The literary reputation of Juvenal is a case 
in point. The scourge of a- corrupt age, the master of moral 
indignation, the great representative of the most original 
production of the Latin genius; such is the idea of Juvenal 
which may be said to have prevailed, and still to prevail, 
in the modern literary world’. Iam far from saying that such 
an estimate is false, but I think it partial and inadequate. 
Take Juvenal at his own estimate, assume that the pictures 
which he draws of contemporary life are in the main correct, 
study him alone and leave the younger Pliny and Quintilian 
and Suetonius and the inscriptions unread, and the ordinary 
view of Juvenal becomes the natural-one. _ But literary criticism 
must in the present day be based upon history; and studied 
historically the position of the famous satirist will, if I am not 
mistaken, appear to be a peculiar and personal one, and his 
satires, though containing a large element of truth, to represent 
the partial and exaggerated views natural in such circum- 
stances. } | 


1 This view seems in the main to be 
that of Professor Mayor, if I may judge 
by the preface to his new edition. I 
‘wish it clearly to be understood that, 
while I venture to differ from Mr 
-‘Mayor’s general estimate of Juvenal’s 
moral position, I cannot adequately 
express my admiration for his edition 


and indeed for his many unique con- 
tributions to Latin scholarship and the 
history of Latin literature. I suppose 
that in wealth of learning and fresh- 
ness of interest combined, Mr Mayor 


‘holds a position occupied by no scholar 


since Casaubon. 


42 THE JOURNAL OF PHILOLOGY. 


It is strange that so little should be known about the life of 
so celebrated a writer. The biographies prefixed to his satires 
in the manuscripts are as numerous as they are unsatisfactory. 
Of these lives there are nine, seven of which are printed 
by Otto Jahn in his edition of 1851. An eighth was published 
from a Harleian ms by Riihl in the Neue Jahrbiicher of 1854; 
a ninth, which I am sorry to say adds nothing to the inform- 
ation conveyed by the others, I have myself found in a 
Bodleian manuscript of the thirteenth century. In point of 
Latin style, and presumably therefore of antiquity, the best of 
these memoirs is that printed by Jahn as No. 1. The author 
imitates the style of Suetonius, but not his clearness or 
accuracy. Of this memoir Borghesi rightly observes that 
Suetonius could never have written in so unsatisfactory a way 
of so distinguished a contemporary. 

The biographies all agree that Juvenal was the son or ward 
of a freedman, that he was born at Aquinum, that he practised 
declamation till middle life (ad mediam aetatem’) and that 
he was banished in consequence of an attack made upon an 
actor. The date of his birth is variously given in the three 
memoirs which mention the fact. Two (2 and 9, Canon.) put it 
in the reign of Claudius Nero (Claudius), the other in that 
of Nero Claudius (Nero). The accounts of his exile present 
equally serious discrepancies. For while one tradition (Lives 
1, 2, 4, 7 and Schol. Τὰν. 4 38) represents him as banished 
to Egypt, another (5, 6) relegates him to Scotland, or the 
Scottish border. Again, in the accounts of the time and cir- 
cumstances of his exile there are irreconcilable differences. 
According to (1) he was banished in his eightieth year, and 
died soon afterwards; according to (4) he was banished by 
Domitian’, and remained in exile, altering and enlarging his 
satires, till he died in the reign of Antoninus Pius: according to 


1 For media aetas Mr Mayor quotes Marcus mediis Antonius annis Primus. 
Plautus Aulularia 157, and Phaedrus 2 So Schol, Iuv. 4 38: Schol. 7 92 
223: we may add Celsus 13 inediam makes him banished by Nero: Schol. 
facillime sustinent mediae aetates, mi- 15 27 says he was in Egypt, not that 
nus iuvenes, minime pueri et senectute he was banished thither. 
confecti; Martial 10 32 3 talis erat 


LIFE AND POEMS OF JUVENAL. 43 


(5) he was banished by Trajan to Scotland and died there soon 
afterwards. 

The tradition however is uniform that the pretext for 
his exile was furnished by some verses which he had written 
against the pantomimus Paris, a favourite of Domitian. The 
verses were, it is stated, inserted into the seventh satire 
(v. 90 foll.) Quod non dant proceres, dabit histrio: tu Camerinos 
et Baream, tu nobilium magna atria curas? Praefectos Pelopea 
facit, Philomela tribunos. This story is confirmed to a certain 
extent by some lines of Sidonius Apollinaris (Carm. 9 270—275) 
Non qui tempore Caesaris secundi Aeterno incoluit Tomos reatu, 
Nee qui consimili deinde casu Ad vulgi tenuem strepentis auram 
Lrati fuit histrionis exul. 

If any reliance can be placed on these words of Sidonius, if 
indeed we can be sure that they refer to Juvenal at all, and not 
to some other poet, then Juvenal must have been banished for 
having said or written something not only offensive to an actor 
but unpopular with the pit and gallery. If the actor was Paris 
the favourite of Domitian, the date of the poet's exile must be 
placed in or before A.D. 83, for Paris was put to death in that year’. 

It is in truth impossible to make anything out on this point 
from the biographies and the scholia. The compilers of these 
notes may have got hold of the fact that Juvenal was banished, 
but have confused their tradition with the import of the lines 
(7 87 foll.) Llle et militiae multis largitus honorem Semenstri 
vatum digitos circumligat auro: Quod non dant proceres ete. 
This passage, however, really contains no reflection whatever on 
any actor: the indictment, if any, lies against the aristocracy. 
The mention of Paris and of an actor here may have led 
to these verses being connected with the story of the histrio 
who caused Juvenal’s banishment. The words satira non 
absurde composita in Paridem pantomimum poetamque semens- 
tribus militidis emitatem (1) to which (2) adds poetamque 
Statium are clearly a mere plagiarism from the text. 


1 Friedlander has pointed out that old names, There was a Paris in the 
actors often took the names of cele- . reign of Nero, as well as in that of 
brated predecessors, asshopsinmodern Domitian, and three more afterwards. 
times sometimes continue to bear the 


44 THE JOURNAL OF PHILOLOGY. 


But another circumstance may have contributed to form 
this tradition. An inscription found at Aquinum contains 
a dedication to Ceres made by a D. Iunius Juvenalis, flamen of 
Vespasian, and holding some appointment (whether that of 
tribunus or praefectus cannot be ascertained, for the word 
is lost) in a cohors Delmatarum. Now, as in A.D. 103 the 
cohors quarta. Delmatarum was in Britain, while an unnum- 
bered cohors Delmatarum was there in 105, and the prima 
cohors Delmatarum in 124, scholars have been inclined to 
suppose that Juvenal was actually, as some of the memoirs 
say, at one time in Britain in a military capacity. It should 
however be added that the fifth cohors Delmatarum was in 
Germany in the year 116, and that as the number of the 
cohort to which the inscription attaches its D. Iunius Iuvenalis 
cannot be recovered, there is really no evidence on which 
we are justified in connecting Juvenal with Britain. 

The inscription of Aquinum, then, throws no real light 
on the question of Juvenal’s banishment. And it should be 
added that if Juvenal was sent to Britain in 103 or 124 contra 
Scotos, as the memoirs say, sub honore militiae, this must have 
taken place either under Trajan or under Hadrian. The first 
alternative is not impossible, though I do not know that there 
is any other evidence of the northern tribes of Britain having 
menaced the frontier during the reign of Trajan. But it is 
inconceivable that Hadrian should have committed to an aged 
literary man (for aged Juvenal must have been in 124) the 
command of a cohort on a dangerous frontier. 

If useless with regard to the question of the banishment, 
the inscription may, perhaps, be turned to account in another 
way. The Juvenals whose name it bears was a flamen of 
Vespasian. So far as this fact goes, it affords a presumption 
that the inscription was put up in the reign of one of Ves- 
pasian’s immediate successors, i.e. either of Titus or Domitian. 
If the Zuvenalis of the inscription is the poet, he must then, in 
the reigns of Titus or Domitian (79—96), have attained the 
age qualifying him for the post either of tribunus or praefectus ~ 
cohortis. : 

Let us consider whether any light can be obtained by 


LIFE AND POEMS OF. JUVENAL. 45. 


interrogating the memoirs with the help of such internal 
evidence as is afforded by the satires themselves. Were we 
dealing in this way with Vergil, Horace, or Ovid, we should 
meet, in all probability, with no difficulty. But Juvenal’s 
manner is at times so unreal that it is impossible for the 
reader to be sure whether the poet is referring to contemporary 
events or only professing to do so. In the first satire, for in- 
stance, he speaks of Tigellinus as a formidable person (pone 
Tigellinum: taeda lucebis in alla, ὅσο.) and suggests therefore 
that he is writing in the reign of Nero. Yet it is clear that 
the piece cannot have assumed its present form until after 
100 A.D. in which Marius was condemned for his misgovern- 
ment in Africa. (Haul ab octava Marius bibit et fruitur dis 
Lratis.) 

There are however some undoubted marks of time in the 
satires which I will at once mention, taking the latest first and 
working backwards, 

The latest is 15 27 (assuming the satire to be really 
Juvenal’s) nuper consule Iunco. Iuncus was consul in the year 
127. 

There is some doubt about 13 16, stupet haec, qui vam post 
terga reliquit Sexaginta annos, Fonteio consule natus’. A Fonteius 
Capito (the praenomen is lost) was consul with C. Julius Rufus 
A.D. 67, and this would bring the thirteenth satire down to 127: 
but C. Fonteius Capito was consul in 59 with C. Vipstanus 
Apronianus. As Fonteius was the first consul in 67 and would 
therefore give his name to the year, recent commentators refer 
the verse of Juvenal to 67: but this is not a necessary in- 
terpretation. The reference may be to C. Fonteius Capito, 
consul 59: for though the Fast: Consulares make him second 
consul after Apronianus, Pliny (H. N. 7 84) and C. I. L. 6 2002 
quote his name first. The alternative dates for this satire are 
then 127 and 119. 

_ The 398th line of the sixth satire (instantem regi Armenio 
Parthisque cometen) is rightly referred by all commentators to 
the comet of 113 A.D. The earthquakes mentioned in the same 


“1 Friedlander refers stupet to Ju- birth in674.p. But surely stupet re- 
venal, and therefore puts the poet’s fers to Juvenal’s friend. 


46 THE JOURNAL OF PHILOLOGY. 


passage may be those which took place in Galatia in 113, 
including perhaps that of Antioch (Α.Ρ. 116: Dio 68 25). 

The eighth satire seems to have been written not very long 
after 100 a.D., for it speaks of the trial of Marius as recent 
(cum tenues nuper Marius discinzerit Afros, v. 120). And the 
same remark applies, as I have said, to the first satire. 

The fourth satire purports at least to have been written not 
very long after Domitian’s death, and the same may be said 
of the second. The lines (29—30) qualis erat nuper tragico 
pollutus adulter Concubitu, qui tum leges revocabat amaras, &e. 
can hardly have been written in Domitian’s life-time. While 
the expression (v. 160) modo captas Orcadas et minima con- 
tentos nocte Britannos shews that the memory of Agricola’s 
British campaign was still fresh in the writer's memory. 

We have thus obtained definite marks of time from about 
96 to 127 A.D. It should be added that the first, third, fourth, 
eighth and tenth satires contain vivid reminiscences of Nero’s 
reign’, while Otho figures in the second. These reminiscences 
suggest that Juvenal was, during Nero’s reign (54—68 A.D.), 
of an age to be keenly alive to what was going on in Rome. 

Let us now proceed to consider another source of evidence. 

There is no doubt that Juvenal and Martial were on terms 
of intimate friendship’, and Martial died, at about the age of 
sixty, in 101 or 102 A.D. In the twenty-fourth poem of his 
seventh book Martial says Cum Iuvenale meo quae me com- 
mittere temptas, Quid non audebis, perfida lingua, loqui? Te 
fingente nefas Pyladen odisset Orestes, Thesea Pirithoi destitu- 
asset amor: Tu Siculos fratres et maius nomen Atridas, Et 
Ledae poteras dissociare genus. In the same book we have an 
epigram (91) addressed to Juvenal himself, De nostro facunde 
tibt Iuvenalis agello Saturnalicias mittimus, ecce, nuces. The 
date of Martial’s seventh book is 92 4.D.2 At that time he 


1 e.g. 1 fin. Tigellinus: 3116 Barea Friedlander as follows: Books 1 and 
(66 a.p.): 3 251 Corbulo: 8 211—212 τι, 85—86 a.p.: m, 87—88: Iv, De- 
Seneca and Nero, cember 88: v, autumn 89: vi, 90 

2 One of the biographies (3) notices (summer or autumn): vu, ὙΠ, 92, 93: 
the fact: Romam cum veniret et Mar- τχ, x (1st edition), 94-96 (December) : 
tialem suum non videret. x (2nd edition) 98 : x11, 102, . 

3 Martial’s epigrams are dated by 


LIFE AND POEMS OF JUVENALE. 47 


knows Juvenal intimately and calls him facundus. This word 
has been taken as implying that Martial only knew of him as 
a teacher of rhetoric: but such a limitation is not necessary. 
Facundus is by writers of this period applied to eloquent writers 
as well as eloquent speakers or declaimers: Horace A. P. 41 
Cui lecta potenter erit res, Nec facundia deseret hunc nec lucidus 
ordo. Statius Silv. 1 4 28—30 seu plana solutis Cum struis 
orsa modis, seu cum tibi dulcis in artum Cogitur, et nostras 
curat facundia leges (whether you write prose or poetry): 
Martial 5 30 3 facundi scaena Catulli (of Catullus as a writer 
of mimes): 14 185 facundi Maronis: a strong instance, as Vergil 
was notoriously a bad speaker. Quint. 8 1 3 τη Tito Livio, 
mirae facundiae υἱῦο, putat imesse Pollio Asinius quandam 
Patavimtatem. There is nothing, then, to stand in the way 
of supposing that Martial knew of Juvenal as a writer in 
92 A.D. 

The only other allusion to Juvenal is in Martial’s twelfth 
book (18), written in 101 or 102 A.D. Dum tu forsitan in- 
quietus erras Clamosa, Iuvenalis, in Subura, &e. 

The intimacy between Juvenal and Martial need not, of 
course, of itself exclude the supposition that Juvenal was much 
the younger man. But taking the evidence as a whole, I doubt 
whether it is necessary to suppose that there was a difference 
of more than ten years between “the ages of the two poets. 
One very remarkable circumstance, which so far as I know has 
not been fully considered by the writers on this subject, seems 
to me to shew that Martial and Juvenal must have been intim- 
ate not only as men, but as writers: that they sympathized 
in their views of literature and saw a good deal of each other’s 
literary work. The circumstance to which I allude is the 
remarkable correspondence between Martial’s epigrams and the 
satires of Juvenal, a correspondence apparent not only in their 
view of literature, but in the subjects they treat, the persons 
they mention, their language and expression, and their general 
tone. This consideration is always of great importance when 
we have to deal with the history of Latin literature. The cor- 
respondence I allude to points to one of two conclusions: either 
that Juvenal, writing some twenty years after Martial’s death, 


48 THE JOURNAL OF PHILOLOGY. 


took a pleasure in imitating his friend’s poetry: or that like 
Calvus and Catullus, Vergil and Horace, Martial and Juvenal 
were much in each other's confidence, working and it may 
almost be said thinking together. 

Before pronouncing in favour of one or the other conclusion, 
it may be well to quote the following passages: 

(1) Their view of literature. 

Martial 4 49 Nescit, crede mihi, aids sint epigrammata, 
Flacce, Qui tantuwm lusus ista rocosque vocat. Ille magis ludit, 
qui scribit prandia saevt Tereos, aut cenam, crude Thyesta, 
tuam, Aut puero liquidas aptantem Daedalon alas, Pascentem 
Siculas aut Polyphemon oves.. A nostris procul est omnis 
vensica libellis, Musa nec insano syrmate nostra tumet. ‘ Illa 
tamen laudant omnes, mirantur, adorant’: Confiteor: laudant 
alla, sed ista legunt. 

Martial 8 3 17 Scribant ἰδία graves nimium nimiumque 
severt, Quos media miseros nocte lucerna videt. At tu Romano 
lepidus sale tinge libellos: Agnoscat mores vita legatque suos. 
Angusta cantare licet videaris avena, Dum tua multorum vincat 
avena tubas. 

Martial 10 4 Qui legis Oedipodem caligantemque Thyesten, 
Colchidas et Scyllas, quid nisi monstra legis? . Quid tibi raptus 
Hylas, quid Parthenopaeus et Attis, Quid tibi dormitor proderit 
Endymion? Eautusve puer pinnis labentibus, aut qui Odit 
amatrices Hermaphroditus aquas? Quid te vana iuvant miserae 
ludibria cartae? Hoe lege, quod possit dicere vita, Meum est. 
Non hic Centauros, non Gorgonas Harpyiasque Inventes: homi- 
nem pagina nostra sapit.. 

Juvenal 1 52 Haec ego non agitem? sed quid magis? 
Heracleas Aut Diomedeas aut mugitum labyrinth Et mare 
percussum puero fabrumque volantem? 85 Quicqud agunt 
homines, votum timor ira voluptas Gaudia discursus nostri est 
farrago libelli. 

(2) Subjects treated. 

Philosophical debauchees. 

Martial 1 24 Aspicis incomptis illum, Deciane, capillis, 
Cuius et tpse times triste supercilium, Qui loguitur Curios, 
adsertoresque Camillos: Nolito fronti credere, nupsit heri. — 


LIFE AND POEMS OF JUVENAZ. 49 


12 42 Barbatus rigido nupsit Callistratus Afro &c. 

7587 Quaere aliquem Curios semper Fabiosque loquentem, 
Hirsutum et dura rusticitate trucem: Invenies: sed habet tristis 
quoque turba cinaedos: Difficile est vero nubere, Galla, viro. 

9 27 6 Curios, Camillos, Quinctios, Numas, Ancos, Et quid- 
quid umquam legimus pilosorum Loqueris sonasque grandibus 
minax verbis, Et cum theatris saeculoque riwaris. Occurrit 
aliquis inter ista si draucus &c. 

9 47 Democritos, Zenonas, inexplicitosque Platonas Quidquid 
et hirsutis squalet imaginibus, Sic quasi Pythagorae loqueris 
successor et heres, Praependet sane nec tibi barba minor, &e. 

Juvenal 2 1 [0]. Ultra Sauromatas fugere hinc libet et 
glacialem Oceanum, quotiens aliquid de moribus audent, Qui 
Curios simulant et Bacchanalia vivunt. Indocti primum ; quam- 
quam plena omnia gypso Chrysippt invenias, nam perfectissimus 
horum est, St quis Aristotelen similem vel Pittacon emit, Et 
tubet archetypos pluteum servare Cleanthas. Frontis nulla fides: 
quis enim non vicus abundat Tristibus obscenis? castigas turpia, 
cum sis Inter Socraticos notissima fossa cinaedos. Hispida 
membra quidem &e. 

2 129 traditur ecce viro clarus genere atque opibus vir &e. 

Neglect of the liberal professions by their proper patrons. 

Martial 1 107 Saepe mihi dicis, Luci carissime Luh, 
‘Scribe aliquid magnum: desidiosus homo es:’ Otia da nobis, 
sed qualia fecerat olim Maecenas Flacco Vergilioque suo: 
Condere victuras temptem per saecula curas Et nomen flammis 
eripuisse meum. In steriles nolunt campos iuga ferre tuvencr: 
Pingue solum lassat, sed vuvat upse labor. ; 

4 46 Saturnalia divitem Sabellum Fecerunt: merito tumet 
Sabellus, Nec quenquam putat esse praedicatque Inter causidicos 
beatiorem. Hos fastus animosque dat Sabello Farris semodius 
faubaeque fresae, Ht turis piperisque tres selibrae &c. 

3 38 Quae te causa trahit vel quae fiducta Romam, Sete ? 
aut quid speras aut petis inde, refer. ‘Causas’ imquis ‘agam 
Cicerone disertior ipso, Atque erit in triplict par mihi nemo 
foro. LEgit Atestinus causas et Civis ; utrumque Noras, sed neutrr 
pensio tota fuit. ‘Si nihil hinc veniet, pangentur carmina nobis ; 
Audieris, dices esse Maronis opus. Insanis: omnes gelidis 

Journal of Philology. vou. Xvt. 4 


50 THE JOURNAL OF PHILOLOGY. 


quicunque lacernis Sunt ibi, Nasones Vergiliosque vides. ‘Atria 
magna colam. Vix tres aut quattuor ista Res alut, pallet 
cetera turba fame. ‘Quid faciam, suade: nam certwm est vivere 
Romae. Sti bonus es, casu vivere, Seate, potes. 

5 16 11 Sed non et veteres contenti laude fuerunt, Cum 
minimum vatt munus Alexis erat. 

8 56 5 Sint Maecenates, non derunt, Γ᾽ lace, Marones το. 

8 82 5 Fer vates, Auguste, twos: nos gloria dulcis, Nos tua 
cura prior deliciaeque swmus. 

The whole of the seventh satire of Juvenal might be taken 
as an illustration of these lines; see especially the lines 1—12: 
53—70: 105—123. 

The vulgar and niggardly patron: perhaps the individual 
referred to by Pliny Ep. 2 6°. | 

Martial 3 49 Vetentana mihi misces, ubi Massica potas: 
Olfacere haec malo pocula quam bibere’. 

3 60 Cum vocer ad cenam, non iam venalis ut ante, Cur mihi 
non eadem, quae trbi, cena datur? Ostrea tu sumis stagno saturata 
Lucrino, Sugitur inciso mitulus ore mihi. Sunt tube boleti, fungos 
ego sumo suillos: Res tibi cum rhombo est, at mihi cum sparulo &e. 

12 36 Lnbras quattuor, aut duas amico Algentemque togam 
brevemque laenam &c. Pisones Senecasque Memmiosque Et 
Crispos nuht redde, sed priores &c. 

Juvenal 5 30—110 may again be taken as a companion 
picture to all these sketches. 

The unsociable gourmand. 

Martial 7 59 Non cenat sine apro noster, Tite, Caecilianus. 
Bellum convivam Caecilianus habet. 

Juvenal 1 140 quanta est gula, quae sibi totos Ponat apros, 
animal propter convivia natum! 


1 Longum est altius repetere, nec re- 
fert quem ad modum acciderit, ut homo 


eligendi, sed ne ius esset recusandi, 
aliud sibi et nobis, aliud minoribus ~ 


minime familiaris cenarem apud quen- 
dam, ut sibi videbatur, lautum et dili- 
gentem, ut mihi, sordidum simul et 
sumptuosum. Nam sibi et paucis opima 
quaedam, ceteris vilia et minuta pone- 
hat, Vina etiam parvis lagunculis in 
tria genera discripserat, non ut potestas 


amicis (nam gradatim amicos habet) 
aliud suis nostrisque libertis. Pliny’s 
second book of letters is dated between 
97 and 100 a.p. 

* See also Martial 1 20, 2 43, 4 85, - 
611. 


LIFE AND POEMS OF JUVENAL. 51 


The mam who burns his own house for the sake of the 
contributions made for him after the disaster. 

Martial 3 52 Empta domus fuerat tibi, Tongiliane, ducentis: 
Abstulit hance nimium casus in urbe frequens. Conlatum est 
deciens. Rogo, non potes tpse videri Incendisse tuam, Tongiliane, 
domum ? 

Juvenal 3 220 Meliora et plura reponit Persicus, orborum 
lautissimus, et merito 1am Suspectus, tamquam ipse suas incen- 
derit aedes. 

Life at Rome. 

Martial 4 5 Vir bonus et pauper linguaque et pectore verus 
Quid tibi vis, urbem qui, Fabiane, petis? Qui nec leno potes nec 
comissator haberi, Nec pavidos triste voce citare reos: Nec 
potes uxorem cari corrumpere amici, Nec potes algentes arrigere 
ad vetulas &c. . 

Juvenal 1 38 optima summi Nune via processus, vetulae ἢ 
vensica beatae: 55 cum leno acciprat moechi bona. 

3 41—50 Quid Romae faciam? mentiri nescio, librum Si 
malus est, nequeo laudare et poscere, motus Astrorum ignoro, 
funus promittere patris Nec volo nec possum, ranarum viscera 
nunquam Inspexi, ferre ad nuptam quae mittit adulter Quae 
mandat, norunt ali &e. 

Rome and the country. 

Martial 12 57 Cur saepe sicct parva rura Nomenti Laremque 
villae sordidum petam, quaeris &c. 

Compare Juvenal 3 239 foll. 

Women and their habits. 

Martial 6 6 Comoedi sunt tres, sed amat tua Paula, Luperce, 
Quattuor: et κωφὸν Paula πρόσωπον amat. 

Juvenal 6 73 Solvitur his magno comoedi fibula. 

Martial 6 7 Iulia lex populis ex quo, Faustine, renata est, Atque 
imtrare domos vussa Pudicitia est, Aut minus aut certe non plus 
tricesima lua est, Et nubit decimo iam Telesilla viro. Quae 
nubit totiens, non nubit: adultera lege est: Offendor moecha 
sumpliciore minus. 

Juvenal 6 224 Imperat ergo viro, set mox haec regna relin- 
quit, Permutatque domos et flammea conterit, inde <Avolat et 
spreti repetit vestigia lecti. Ornatas paulo ante fores, pendentia 


4—2 


52 THE JOURNAL OF PHILOLOGY. 


linguit Vela domus et adhuc virides in limine ramos. Sic crescut 
numerus, sic fiunt octo mariti Quinque per autumnos, titulo 
res digna sepulchre. 

Martial 7 67 4 (Philaenis) Harpasto quoque subligata ludit 
Et flavescit haphe, gravesque draucis Halteras facili rotat 
lacerto το. 

Juvenal 6 246 Endromidas Tyrias et femineum ceroma Quis 
nescit, vel quis non vidit vulnera pali? το, ib. 420 magno gaudet 
sudare tumultu Cum lassata gravi ceciderunt bracchia massa &e. 

Martial 10 68 Cum tibi non Ephesos, nec sit Rhodos aut 
Mytilene, Sed domus in vico, Laelia, patricio, Deque coloratis 
nunquam lita mater Etruscis, Durus Aricina de regione pater ; 
Κύριέ μου, μέλε prov, ψυχή pov congeris usque, Pro pudor! 
Hersiliae civis et Egeriae. Lectulus has voces, nec lectulus audiat 
omnis &e. 

Juvenal 6 185 Nam quid rancidius, quam quae se non putat 
ulla Formosam, nisi quae de Tusca Graecula facta est, De 
Sulmonensi mera Cecropis &e....... Quotiens laseivum interventt 
illud Ζωὴ καὶ ψυχή &e. 

Martial 2 66 Unus de toto peccaverat orbe comarum Anulus, 
an certa non bene fixus acu. Hoc facinus Lalage, speculo quo 
viderat, ulta est, Et cecidit saevis icta Plecusa comis. Desine 
iam, Lalage, tristes ornare capillos &c. 

Juvenal 6 490 Disponit crinem laceratis ipsa capillis Nuda 
umero Psecas infeliv nudisque mamillis. Altior hic quare 
cincinnus? taurea punit Continuo flexi erimen fucinusque ca- 
pilla &e. 

_ (3) Persons’. | 

Thymele and Latinus: Martial 1 5 5 qu Thymelen spectas 
derisoremque Latinum: 5 61 11 quam dignus eras alapis, 
Mariane, Latin. 9 29 (his epitaph). Juv. 1 36, 6 44. 

Fronto: Mart. 1 56, 5 34: Juvenal 1 12 Frontonis platani. 
Mommsen thinks this is the consul of A.D. 96. 

Chione: Mart. 1 35 7 al. Juv. 3 136. 


1 1 should perhaps have said names, dence is no less striking, and tells, 
as many of the names in Martialand though in a different way, in favour of 
Juvenal are doubtless fictitious. But my argument. 
even where this is the case, the coinci- 


IIFE AND POEMS OF JUVENAL. 53 


Pontia: Mart. 2 34, 4 43: Juv. 6 638 (where see the 
scholia). 

Tongilius: Mart. 2 40, Juv. 7 130. 

Cordus the poet: Mart. 2 57, 3 15, 5 23, 5 26: Juv. 1 2, 

3 208. 
| Pollio the singer: Mart. 3 20 18, 4 61, 12 12, Juv. 6 387, 
7 176. 
Paris the pantomimus, Mart. 11 13, Juv. 6 87. 

Catullus the mime-writer: Mart. 5 30, Juv. 8 186, 13 111. 

Hamillus the schoolmaster: Mart. 7 62, Juv. 10 224, 

Glaphyrus the flute-player: Martial 4 5 8, Juvenal 6 77. 

(4) Words and expressions. 

Mart. 1 20 4 boletum qualem Claudius edit edas. Juv. 5 147 
boletus domino, sed quales Claudius edit &e. 

Mart. 1 76 14 steriles cathedras. Juv. 7 203 vanae steri- 
lisque cathedrae. 

Mart. 1 92 9 pasceris et nigrae solo ndore culinae. Juv. 5 
162 captum te ndore suae putat ille culinae. 

Mart. 2 1 4 hoc primum est, brevior quod mihi carta perit: 
1047 quid te vana tuvant meserae ludibria cartae? Juv. 118 
periturae parcere cartae. 

Mart. 2 43 9 tu Lnbycos Indis suspendis dentibus orbes. 
Juv. 11 122 latos nisi sustinet orbes Grande ebur. 

Mart. 4541 cut Tarpeias liceat contingere quercus. Juv. 6 
387 an Capitolinam deberet Pollio quercum Sperare. 

Mart. 5 44 11 antiquae venies ad ossa cenae. Juv. 8 90 ossa 
vides rerum vacuis exsucta medullis. 

Mart. 6 505 Vis fiert dives, Bithynice? conscius esto: Nil 
tibi vel minimum basia pura dabunt. Juv. 3 49 quis nunc 
diligitur nisi conscius Xe. 

Mart. 6 60 10 victurus geniwm debet habere liber. Juv. 6 
562 nemo mathematicus genium indemnatus habebit. 

Mart. 6 71 3 tendere quae’ tremulum Pelian Hecubaeque 
maritum Posset ad Hectoreos sollicitata rogos. Juv. 6 325 
quibus incendi tam frigidus aevo Laomedontiades et Nestoris 
hirnea possit. 

Mart. 8 21 3 placidi numquid te pigra Bootae Plaustra ve- 
hunt? Juv. 5 23 pigri serraca Bootae. 


54 THE JOURNAL OF PHILOLOGY. 


Mart. 9 35 Scis quid in Arsacia Pacorus deliberet aula: 
Rhenanam numeras Sarmaticamque manum: Verba ducis Dact 
cartis mandata resignas, Victricem laurum quam venit ante 
vides: Scis quotiens Phario madeat Tove fusca Syene, Scis quota 
de Libyco litore puppis eat. Juv. 6 402 Haec eadem novit quid 
toto fiat in orbe, Quid Seres, quid Thraces ugant...... Instantem 
regi Armenio Parthoque cometen Prima videt &c. 

Mart. 9 73 9 Srange leves calamos, et scinde, Thalia, libellos. 
Juv. 7 27 frange miser calamos, vigilataque proelia dele. 

Mart. 10 25 5 nam cum dicatur tunica praesente molestu 
&e. Juv. 8 235 quod liceat tunica punire molesta. 

Mart. 13 64 1 succumbit steril frustra gallina marito. Juv. 
3 91 quo mordetur gallina marito. 

Mart. 10 87 10 Cadmi municipes ferat lucernas: 14 114 
Hane tibi Cumanae rubicundam pulvere testae Municipem misit 
casta Sibylla suam. Juv. 14 271 municipes Lovis advexisse 
lagonas. 

Two things should be observed with regard to these coin- 
cidences: first, that they are of a kind which points rather to 
independent handling of the same themes by two intimate 
friends than to imitation by the one of the other’s work: 
secondly, that they for the most part occur in the first nine 
satires of Juvenal; the great majority, indeed, in the first 
seven. The most natural conclusion is that during the greater 
part of Domitian’s reign Martial and Juvenal virtually worked 
together. This inference would agree with the tradition of 
the biographies that Juvenal was a professor of declamation 
usque ad mediam aetatem. For supposing his youth to have 
fallen in the reign of Nero and his death to have taken place 
(say) 127 or 128 A.D., his media aetas would begin about 85, 
not long before the publication of Martial’s first two books. 

It does not follow, of course, because Juvenal had writ- 
ten satire in Domitian’s reign, and shewn it to Martial and 
perhaps to other friends’, that he had published anything so 
early. In their present form, at any rate, it is probable if not 
certain that most of his satires are later than Domitian’s death’. 


? As to Quintilian? who says (101 qui olim nominabuntur. 
94) sunt (satirici) clari hodieque, et 2 Teuffel’s solution is as follows 


LIFE AND POEMS OF JUVENAL. 55 


I have said that many of the earlier satires are, in my 
opinion, to be assigned to the later years of Domitian. It will 
no doubt be asked whether the seventh satire, Et spes et ratio 
studiorum in Caesare tantum, does not belong to the age of 
Trajan or Hadrian? Undoubtedly this is the opinion of most 
modern commentators, including Mr Mayor. But it is evident 
that this hypothesis lands us in considerable difficulties. The 
setting of the piece is, in any case, taken from the ‘time of 
Domitian, for Statius and Quintilian are spoken of as if. alive, 
and the good fortune of Quintilian, indeed, as quite recent 
(exempla novorum Fatorum transi). But the commentators, 
for some reason which I cannot comprehend, seem to have 
an invincible repugnance to applying the line Ht spes et ratio 
ἄς. to Domitian. In-no case is the saying truer than in that of 
Domitian that the evil which men do lives afterthem. For the 
crimes of his later years I am not attempting to apologize. 
But it is only just to say that they were committed by a man 
whom suspicion and terror had driven to the verge of frenzy. 
Domitian was probably not a man of strong head, and it should 
never be forgotten that the historians of his reign belonged to 
the senatorial party. It is abundantly clear, however, even 
from their evidence, that his administration of the empire was 
that of a careful and conscientious ruler’. The provinces were 


(Studien und Charakteristiken pp. 413 
—415). ‘* Dass Iuvenal seine Satiren 
unter Domitian nicht verfasst hat, 
sondern erst unter Traian,...geht aus 
seiner ersteren Satire...positiv hervor.” 

**Thr (1.6. der Satiren) Stoff die 
Zeit des Domitian ist.” 

ἐς Perspektivische Zeichnen scheint 
seine” (Iuvenal’s) “Sache nicht zu 
sein; die grdéssere kiinstlerische Ruhe, 
das Masshalten, die versdhnte Stim- 
mung, den weiteren Gesichtskreis und 
die epische Glatte, welche sich daraus 
hatte ergeben sollen, dass es etwas 
Vergangenes, hinter ihm Liegendes, 
ist, was er schildert, hat er nicht eintre- 
ten lassen, sondern den gleichen Eifer 
aufgewendet wie wenn er noch mitten 


stiinde in dieser grauenvollen Zeit, 
und jeden Augenblick dadurch zu lei- 
den hitte. Ueberhaupt hat ihn jene 
Differenz zwischen der Zeit in welcher 
er schreibt, und der, welche er dar- 
stellt, nicht viel Kopfzerbrechen yge- 
kostet; er ignoriert sie einfach.” 

1 Suetonius, Domitian, 2 simulavit 
et ipse mire modestiam, imprimisque 
poeticae studium, tam insuetum antea 
sibi quam postea spretum et abiectum, 
recitavitque etiam publice....4. Instituit 
et quinquennale certamen Capitolino 
Tovi triplex, musicum equestre gymni- 
cum, et aliquanto plurium quam nune 
est coronarum. Certabant enim et prosa 
oratione Graece Latineque, &c....Cele- 
brabat et in Albano quot annis Quin- 


56 THE JOURNAL OF PHILOLOGY. 


well governed in his reign, and justice well administered. 
Profligate and cruel in private life, he yet shewed in some 
of his legislation a real concern for humanity and public morals. 
Though himself an indifferent general, there is no evidence 
that he was not concerned to make good military appointments. 
That he had an honest intention to encourage literature, so far 
as to do so seemed compatible with the security of the Empire 
and the preservation of private morality, there can be no 
doubt. He took steps for the restoration of libraries and 
the copying of texts. The calumny of Suetonius, that he 
read nothing but the commentarw and acta of Tiberius, is 
refuted by the undoubted fact that he read Martial, and was 
indeed concerned to keep that brilliant writer within the 
bounds of decency. And if he read Martial he probably read 
Turnus and Statius. The agon Capitolinus', or five-yearly 
contest of artists and men of letters on the Capitol, and the 
similar trials of skill at the emperor’s Alban villa, no doubt 
must have done something to encourage poetry and rhetoric, 
even if we believe Pliny (Paneg. 54) that they resulted largely 
in flattery of the emperor. 

To state the matter quite fairly, we should probably say 
that to encourage literature was an honourable tradition of the 
early empire. Here, as in politics, the princeps took upon 
himself the functions of the old aristocracy. There was no 


quatria Minervae, cui collegium insti- 
tuerat, ex quo sorte ducti magisterio 
fungerentur ederentque eximias vena- 
tiones et scaenicos ludos, superque ora- 
torum ac poetarum certamina. 

8. us diligenter et industrie dizit, 
plerumque et in foro pro tribunali extra 
ordinem:; ambitiosas centum virorum 
sententias rescidit ; recuperatores, ne 86 
perfusoriis adsertionibus accommoda- 
rent, identidem admonuit : nummarios 
iudices cum suo quemque consilio no- 
tavit. Auctoret TR. PL. fuit aedilem 
sordidum repetundarum accusandi iudi- 
cesque in eum a senatu petendi. Magis- 
tratibus quoque urbicis provinciarumque 
praesidibus coercendis tantum curae ad- 


hibuit, ut neque modestiores unquam 
neque iustiores extiterint: e quibus 
plerosque post illum reos omnium cri- 
minum vidimus. Suscepta correctione 
morum licentiam theatralem promiscue 
in equite spectandi inhibuit: scripta 
famosa vulgoque edita, quibus primores 
viri ac feminae notabantur, abolevit, &e. 

1 Statius Silvae 3 5 28: tu me nitidis 
Albana prementem Dona comis, sancto- 
que indutum Caesaris auro ἄορ. 4 2 66 
Cum modo Germanas acies, modo Daca 
sonantem Proelia, Palladio tua me ma- 
nus induit auro. 4522 hic mea carmina 
Regina bellorum virago Caesareo deco- 
ravit auro: 5 3 228 si per me serta 
tulisses Caesarea donata manu. 


LIFE AND POEMS OF JUVENAL. 57 


deliberate intention on the part of the emperors to crush the 
freedom of speech as such: a poet or orator was safe so long as 
he remained on neutral ground. None the less, of course, is it 
true that the springs of all nobler writing were gradually 
choked up, as the aristocracy declined from its ancient power, 
position, and independence. For the production of great works 
expansion of soul is necessary, nor could minds of high powers 
and sincere emotion be content with the hackneyed themes of 
mythology or the trivialities of social intercourse. But, after 
all, the main burden of Juvenal’s seventh satire is not so much 
the encouragement of literature by the court as its neglect by 
the nobility, its natural patrons; and this is a point upon which 
Martial, writing mostly under Domitian, insists with almost 
wearisome iteration. Sint Maecenates, non derunt, Flacce, 
Marones, and so on. If the satire under discussion is to be 
allowed to have any life and meaning it must surely be assigned 
to the reign of Domitian. 

It remains to be asked whether there is any evidence 
that Juvenal was banished from Rome, and if so, when he was 
. banished ? 

The fact is asserted by all the memoirs, though they differ 
as to the place of exile. I will now mention the only other 
evidence which seems to me to bear upon the point, and 
of this I must confess that little can be made. Juvenal was 
probably in Rome in the year 92 and 93, when Martial com- 
pleted his seventh book, in which, as we have seen, he addresses - 
Juvenal twice. It is, however, noteworthy that Martial does 
not again address Juvenal till the year 101, five years after 
Domitian’s death. Can the reason of this be the absence 
of the exile from Rome? If so, it may well be that Juvenal 
was one of the large number of persons whom the last years of 
Domitian drove from the city and from Italy. 

Let us now, leaving the question of chronology, endeavour 
to form an idea of the social surroundings into which Juvenal 
was born, and to examine whether his satires are a faithful 
reflection of them. 

It is not too much to say that modern city life on a large 
seale, the highest development of European civilization in its 


58 THE JOURNAL OF PHILOLOGY. 


best and its worst forms, has its first example in the Rome 
of the first century A.D. In the history of moral progress, 
eighteen hundred years would sometimes appear to be a mere 
cipher. 

I am not, be it well understood, wishing to understate 
the differences between ancient and modern life as a whole; 
but even taking all these into account, it remains true that 
Rome was the first great capital city in Europe, exhibiting in 
its society all the features of the struggle for wealth, that is, for 
power and position, which is the main characteristic of modern 
life when left at repose from war or revolution. The central 
fact which should be grasped in looking at the Roman society 
of the early Empire, as contrasted with that of the last two 
centuries of the Republic, is the comparative instability of 
its distinctions. The disorder, the want of public security 
of the last period of the Commonwealth, had endangered com- 
merce, and thus helped to maintain the landed aristocracy 
in an assured position. With the Empire came peace, and 
their chances to all and sundry. Nune patimur longae pacis 
mala, says Juvenal regretfully in his sixth satire (286). While 
the aristocracy was wasting its strength in futile struggles with 
the court, and many noble families were becoming impoverished, 
the honest merchant and the unscrupulous adventurer, Roman, 
Greek or Oriental, were pushing to the front and using their 
new social and political opportunities. The situation was 
much aggravated by the existence of slavery. A peculiar 
character was given at this time, and at Rome, to this curse of 
the ancient world. Quantities of slaves of all known nations — 
and all characters were brought, from one reason or another, to 
Rome. Their disproportionate number tended, in one respect, 
to alleviate their condition and prospects. Emancipation was 
easy and common. It let loose upon society a number of 
persons who had lived and meant to live by their wits, often 
not inconsiderable, men who had done and suffered everything, 
with the vices of slavery and without the virtues of freedom, 
supple, serviceable, wicked. “A serving-man, proud in heart 
“and in mind, that curled my hair, wore gloves in my cap, 
“served the lust of my mistress’s heart and did the act of 


LIFE AND POEMS OF JUVENAL. 59 


“darkness with her: swore as many oaths as I spake words, 
“and broke them in the sweet-face of heaven: one that slept in 
“the contriving of lust and waked to do it. Wine loved I 
“deeply, dice dearly, and in woman out-paramoured the Turk. 
“False of heart, light of ear, bloody of hand; hog in sloth, fox 
“in stealth, wolf in greediness, dog in madness, lion in prey.” 
In Shakespeare’s portrait we seem to recognize the coarser 
forms of the Calvisius Sabinus (Sen. Ep. 27 5 foll.) the Hostius 
Quadra (Sen. N. Q. τ 16) the Zoilus and the Trimalchio of 
Seneca, Martial, and Petronius; men for whom the court, 
in case of need, had its favours, ladies their commissions, men of 
letters their filthiest verses. The traditions of Italian manli- 
ness and dignity were violated at every turn by the influx 
of foreign vice and the shamelessness of foreign adventure. 
The mere presence of the Orientals irritated and alarmed 
Roman feeling. The hunt for wealth, the rush from step 
to step of the social ladder, was fierce and undisguised*. There 
was no end to the accumulation of large fortunes and the 
formation of immense landed estates. The desire of pleasure 
gratified itself by every refinement of luxury; the multitude of 
slaves gave facilities for the gratification of every form of lust. 
Impudicitia in ingenuo probrum est, in servo necessitas, in li- 
berto officitum, is an opinion quoted by the elder Seneca. A 
loosening of the older social conventionalities began even in 
the circles of the Roman nobility, who sometimes for their own 
gratification, sometimes to please the court, would forget the 
proprieties of a former day and turn actors, gladiators, charioteers. 
Women enjoyed their share of the general freedom, and while 
the more serious among them plunged into literature or law, or 
became devotees of some foreign religion, others patronized 
actors and gladiators, or pursued other and more questionable 
forms of an emancipated life. Meanwhile the life of the 
_ capital exercised its irresistible attraction upon the provinces. 
Men streamed to Rome, with hopes, more or less slender, of 
making a livelihood by honest means. They might succeed, 


1 Schol. Iuv. 5 3 Sarmentus...incer- ut pro equite Romano ageret, et decu- 
tum libertus an servus, plurimis forma  riam quoque quaestoriam compararet, 
et urbanitate promeritis co fiduciaevenit See especially Pliny 33 88 32—34. 


60 THE JOURNAL OF PHILOLOGY. 


and make a name in literature or politics; they might fail, and 
become the restless and degraded dependents of one or more of 
the great houses. 

This is the dark side of the picture; what is there to set 
against it? This century, if characterized by the beginning of 
remarkable social changes, saw also the beginning of a religious 
and moral evolution no less remarkable. In the upper and 
better educated class philosophy and the higher culture were 
producing considerable moral results. Philosophy and religion 
are in this unfortunate, that while their practical manifestations 
in ordinary life are often unrecognized even by honest observers, 
any clever cynic can detect their counterfeit. Philosophy also, 
as Bernays has well pointed out, lay in the ancient world under 
a peculiar disadvantage. It was for the most part revolutionary 
and opposed to the existing forms of social life. Postremo nemo 
aegrotus quicquam somniat Tam infandum, quod non aliquis 
dicat philosophus, is the verdict of healthy Roman common 
sense as expressed in Varro’s Saturae (Humenides fr. 6). No 
doubt, as the social evolution implied in the change from Greek 
to Roman life worked itself gradually on, the antagonism became 
less pronounced. The organization of the Roman empire was, 
to a certain extent, a realization of the Stoical ideal; at any rate, 
it had broken down the conception of isolated city life, and sub- 
stituted for it the conception of a larger society. An active 
performance of the duties of a citizen was not inconsistent—far 
from it—with the profession of a Stoic or Academician. None 
the less had the philosophic profession, as a whole, a strong 
tendency, at the period which we are considering, to isolate its 
followers if not from the duties, at least from the interests of 
ordinary life, and devote them to the contemplation of an ideal 
morality. Stoicism, the most influential theory in the first 
century, had a pronounced influence in this direction. That 
philosophers of any independence of character were looked upon 
with suspicion both by the government and by society lay in 
the nature of things’. E’rrare mihi videntur, says Seneca (Epist. 
73 1), que existimant philosophiae fideliter deditos contumaces 


1 Seneca Epist. 5 2 Satis ipsum no-  tatur, invidiosum est... Intus omnia 
men philosophiae, etiamsi modeste trac- _dissimilia sint: frons populo conveniat. 


LIFE AND POEMS OF JUVENAL. 61 


esse et refractarios, contemptores magistratuum et regum eorumve 
per quos publica administrantur. The prejudice extended to 
men who professed to represent a sound and common-sense 
view of educated life and conduct, men like Quintilian, Martial, 
and Juvenal. These could only see that there were not a few 
hypocrites among the professors of philosophy (Quint. 12 3 2, 
Sen. Ep. 29 2, Juv. 2 and Mart. ll. cc.). 

If philosophy was doing much to hold a lofty ideal of life 
before the eyes of those among the cultivated classes whose 
intellect and moral sense were capable of accepting its teaching, 
Judaism found its way from the Jewish quarters into the great 
houses, and was popular, nay, even fashionable, among rich and 
high-born ladies. But of the great revolution which was silently 
preparing itself among the lower orders, binding together the 
poor and oppressed into a new society, with principles of con- 
duct, a mode and object of worship, and hopes for the future 
unknown or imperfectly known before, the upper classes, in Rome 
at any rate, knew nothing. Christianity was to them no more 
than a form of Judaism. 

In the presence of social phenomena so absorbingly inter- 
esting, what is Juvenal’s attitude? Are his pictures of contem- 
porary life to be trusted? Does he, in his character of moralist, 
represent the highest effort of contemporary thought ? 

In a sense in which Juvenal-did not intend the words, 
difficile erat saturam non scribere. The satura was not properly 
an attack on vice and folly, though Juvenal did his best to 
encourage the idea that it was, but a sketch of life and cha- 
racter. The Romans had a natural aptitude for this kind of 
writing, not because they were more spiteful than the Greeks, 
but because they had a larger sphere of experience, and a 
greater knowledge of the ars vivendi. At the time which we 
are now considering, the artist had abundance of materials, nor 
is it surprising that during these years two eminent poets, 
Martial and Juvenal, refused to have anything to say to the old 
mythologies, and turned to real life for their models. Turnus, a 
third excellent writer of the time and a satirist like Juvenal, 
has been so unfortunate as to leave to posterity nothing but his 
name, which is coupled with that of Juvenal by Rutilus Nama- 


62 THE JOURNAL OF PHILOLOGY. 


tianus (1 603). Martial (11 10) says of him Contulit ad saturas 
ingentia pectora Turnus: and again (7 97 7) Turni nobilibus 
libellis. | 
Juvenal was the native of a country town, Aquinum, and 
had been brought up in the house of a rich libertinus, whether 
as his son or fosterchild is unknown. In position he exactly 
resembled his contemporary Turnus, who, if we may believe a 
notice preserved in Valla’s scholia to Juv. 1 20, attained great 
influence in the courts of Titus and Vespasian. He seems to 
have been in Rome from his childhood upwards (3 84 et nostra 
infantia &c.). Thus, though an Italian by birth, he was a 
Roman by education, and as a consequence became a Roman in 
sympathies and antipathies. Several passages shew that for 
some time at least he was a cliens, in the later sense of the word, 
that is, a poor dependent on great houses: 1 99 iubet a praecone 
vocart Ipsos Troiugenas, nam vewant limen et ipsi Nobiscum: 
3 187 praestare tributa clientes Cogimur, et nitidis augere peculia 
servis: and so Mart. 12 18 Dum tu forsitan inquietus erras το. 
The statement of the memoirs, that Juvenal practised de- 
clamation till middle age, is abundantly confirmed by the tone 
of his compositions. The touch of the declamator is every- 
where. There is no need, with Ribbeck (Der echte und umechte 
Iuvenal) to separate the declamatory satires, such as the tenth, 
from the rest’. Some pieces evidently contain several rhetorical 
loct or passages of description well tricked out and loosely 
strung together. Such are, for instance, the picture of Otho 
2 99—109: of Eppia, 6 82 foll.: of Messalina, 6 114 foll.: of 
Lateranus, 8 146: the verses on Cicero, Marius and the Decii 
8 231—268: on Seianus, 10 56 foll., and others in the same 
satire. The composition again is sometimes that of a rhetor- 
ician, loose, inharmonious, inconsistent. The first satire is 
a series of incoherent complaints: wnde illae lacrimae? A 
married impotent, an athletic lady, a barber rich enough to 
challenge the fortunes of all the patricians: the Egyptian 
Crispinus with his ring, the lawyer Matho in his litter: the 
infamous will-hunter, the robber of his ward, the plunderer of 


1 Teuffel’s answer to Ribbeck (Studien und Charakteristiken Ὁ. 414 foll.) is 
well worth reading. 


LIFE AND POEMS OF JUVENAL. 63 


the provinces: the pander husband, the low-born spendthrift, 
the forger, the poisoner; all these are hurried together in no 
intelligible order, and with the same introductory cum hoc fiat, 
and the same conclusion in several variations non scribam 
saturam? Then at v. 81 the satire seems to open again and 
promise a description of various vices, but instead of this we 
have an elaborate complaint, extending over many lines, of the 
poverty of the nobility, with a description of the hardships of 
a client. The ill-proportioned piece concludes with a promise 
to write against the dead, and the dead are to be (if we are 
to suppose any coherence at all in the peroration) those who 
lived before the days of Nero. Yet the satire in another 
passage (exul ab octava &c.) purports to have been written 
after 100 A.D. 

Juvenal’s most elaborate effort is the sixth satire. <A very 
brief analysis of the first part of this celebrated piece will 
discover the badness of its composition 1—59: Do not think 
of marriage, few women being both chaste and fair: 60—113 
do not look for a wife in the theatre: all ladies prefer actors 
and gladiators: 114—135 Messalina’s habits are described: 
135—160 no men love their wives, but only their wives’ for- 
tune or beauty: 161—183 a perfect wife would be intolerable : 
184—199 it is very bad in a lady to talk Greek: 200—224 a 
wife is always a tyrant: 225—230 she will marry as often as 
she likes: 231—241 the daughter-in-law is corrupted by the 
mother-in-law : 242—245 there is a woman in every lawsuit: 
246—267 ladies are often very fond of gymnastics: and so on, 
and so on, 

In fact, with all its brilliancy of execution in detail, the 
piece, as far as composition is concerned, is a mere chamber of 
horrors. The main theme, that it is madness to marry because 
a good wife cannot be found, is not so much worked out as 
illustrated by a series of pictures quite unconnected, and argu- 
ments sometimes inconsistent. The gist of the argument seems 
to be that women are either very bad or very good, or too 
learned, or too athletic; but in truth there is no argument 
properly so called, but a string of sketches, which give the 
impression of having been drawn not from a wide observation 


64 THE JOURNAL OF PHILOLOGY. 


of life, but from particular and notorious cases. An instance of 
Juvenal’s desire to produce effect at the expense of consistency 
is to be found in his treatment of the passion of women for 
athletics and for law, in the second and in the sixth satires. In 
the second satire, where his object is to exalt women at the. 
expense of men, Favonia is made to say luctantur paucae, come- 
dunt colyphia paucae: Numquid nos agimus causas, civilia twra 
Novimus? &c. But in the sixth satire (242 foll., 246 foll., 
352 foll.) a directly opposite impression is conveyed. 

Rhetoric, as Matthew Arnold well says, is always incon- 
sistent, and this is the inconsistency of the rhetorician. A 
corresponding unreality tinges many of Juvenal’s utterances as 
a moralist. We have seen that in the first satire he expends 
much the same amount of indignation on the nouveau riche of 
a barber as on the most abandoned criminals, and that the 
grievances of the poor client, perhaps his own grievances, occupy 
a place out of all proportion to their moral importance. Still 
more strikingly conspicuous is this perversity of judgment in the 
second and eighth satires’. In the eighth, after some hundred 
and fifty verses of excellent quality in all respects, Juvenal strikes 
off into an indignant tirade against the nobleman who is too 
fond of horses, the nobleman who acts on the stage, and the 
crimes of Nero: which are, it would seem, his murders of his 
mother and his relations; and, as a climax, his love of music 
and the drama. Are we reading De Quincey’s Art of Murder? 
or is further evidence needed that Juvenal is only half a 
moralist, that irritation against social improprieties is almost 
as strong an element of his invective as genuine anger against 
vice? That with such a point of view he should have no 
theory of life but that of the most superficial common sense, 
that he should see little in philosophy but a solemn imposture, 
is only natural (14 120). Nemesis overtakes him, however: 
he has nothing to say against slavery nor against the games 
of the amphitheatre, though Seneca (see Epist. 7 and 47) com- 
pletely condemns them. 


1 In the second (v. 143), after men- fuscina Gracchi, Lustravitque fugam &e. 
tioning a case of unnatural vice, he 2 It is interesting to compare this 
goes on Vicit et hoc monstrum tunicati satire with Seneca’s forty-fourth epistle. 


LIFE AND POEMS OF JUVENAL. 65 


_ Nor can this capriciousness be defended on the ground that 
Juvenal is not a moralist but a humorist. If he falls short of 
the simple philosophical elevation of Persius, he is equally 
incapable of the light and plastic touch of Petronius. From 
Juvenal we hear what people on particular occasions have done; 
but we know nothing of their personality; he cannot draw a 
character, he cannot laugh. Think of Juvenal’s Virro and then 
of Petronius’s Trimalchio; the one is a figure cut out in paper, 
the other a living man. The inconsistencies of the sixth satire 
might be defended in a humorist ; he would be in his right in 
saying that a licentious wife or an over-virtuous wife are 
equally objectionable. But this ground is not open to the 
moralist, who is bound to defend virtue against all cavil. 

In fact, Juvenal is at his best not when he is lashing vice, 
but when he is in the vein of grave and simple moral expos- 
tulation. The tenth satire is perhaps too declamatory to be 
taken as a specimen of his best work: the thirteenth and four- 
teenth are better, defaced by none of the faults which I have 
mentioned, and carrying the reader along from point to point 
with sweetness and dignity. 

The style of Juvenal, the influence of which is so familiar 
in modern literature, is, so far as we know, new in satire. 
While Persius imitates Horace, and makes at least a clumsy 
attempt to preserve the form of a dialogue, Juvenal, in most of 
his pieces, throws this entirely aside, and casts his ideas into 
the mould of the Vergilian epic. Fingimus hoc altum satira 
sumente coturnum Scilicet, et finem egressi linguamque priorum 
Grande Sophocleo carmen bacchamur hiatu Montibus tgnotum 
Rutulis caeloque Latino (6 634). Taking these words out of their 
context, we might accept them as a description of Juvenal’s man- 
ner, which, like all we know of the man, is elevated, serious, and 
unbending. He is a perfect master of his metre, a perfect master 
of expression within the limits of his ideas. But his ideas, and 
the way in which he marshals them, are those of the poetical 
declaimer, not of the poet. Facit indignatio versum: verses, 
yes ; but not poetry. It would be difficult to quote from Juvenal 
one really poetical line. But he is a great metrist, a master 
of points, a rhetorician inspired by the love of his calling. His 


Journal of Philology. you. xvt. 9 


66 THE JOURNAL OF PHILOLOGY. 


arrangement is often bad: it is his glittering language which 
arrests attention. It is this, far more than the coherence 
or truthfulness of his workmanship, which has won and will 
maintain his position in literature. There is a genuine and 
passionate rhetoric which seems almost to reach the strain of 
poetry ; this is the gift of Juvenal, which we should do ill to 
underrate. But we should do equally ill to mistake it for 
anything higher than it really is, or to put too much confidence 
in a writer honest indeed, but soured by poverty and disap-- 
pointed ambition, who, with whatever brilliancy of detail, does 
not pass beyond the bounds of a somewhat narrow experience, 
mingles righteous anger with much personal irritation, and 
gives, after all, an exaggerated picture of a peculiar phase of 
ancient life. 


H. NETTLESHIP. 


NOTES IN LATIN LEXICOGRAPHY. 


[Words marked * are not to be found in the dictionaries of Georges 
(7th edition) or Lewis and Short. ] 


Ablaqueo. The right form of the word is probably ablacuo, 
preserved in the text of Varro R. R. 1 29 1. 

* Acherontinus Hercules, C. I. L. 9 947. 

Adfectio, absolutely, in the sense of affection; add OC. I. L. 9 
1592 (of the age of Commodus) pati rarae adfectionis: Ib. 
1612 (both at Beneventum). 

Amlbitiosus in aliquid: add to the instances in Georges 
Seneca Trang. 1 14 ambitiosus in verba. 

Assulatim. The reading assultatim given by B in Plautus 
Captivi 832 may possibly point to a form astulatim, from 
astula. 

-* Aularius, = aulicus, Gloss. ap. Mai Cl. Auct. Vol. 6. 

Avidus futuri, Horace A. P. 172. This difficult phrase 
‘should apparently mean anxious about the morrow, if we may 
argue from the parallel passage in Seneca Epist. 32 4: O quando 
illud videbis tempus quo scies tempus ad te non pertinere? quo 
tranquillus placidusque eris et crastint neglegens et in summa 
tut satietate. Vis scire quid sit quod faciat homines avidos 
futurt? Nemo sib contigit. 

Caballus as distinguished from equus. Add Lea Metalli 
Vipascensis (Ephemeris Epigraphica 3 to p. 167) asinos asinas 
caballos equos. 

Caperro. On p. 344 of my Lectures and Essays I have 
argued that this, and not capero, is the right form of the word. 


5—2 


68 THE JOURNAL OF PHILOLOGY. 


I have since found that the word is so written in Varro L. L. 
7 107 according to the authoritative MS., and also in the two 
best MSS. of Martianus Capella 5 509. 

Civitas in the sense of social feeling. This important usage, 
which, as far as I know, is unnoticed in the current lexicons, is 
attested by the following passages: Cicero pro Murena § 74 
horribilis oratio, sed eam...civitas ipsa respuit: Ib. § 77 haec 
omnia ad rationem civitatis st derigas, recta sunt. This may 
also be the meaning of the word in Pliny Epist. 1 14 9, cum 
publicos mores atque etiam leges civitatis intueor, quae vel in 
primis census hominum spectandos arbitrantur. 

Depositio in the sense of death. Add to the instance in 
Georges C. 1. L. 9 1370, 1872, 1376, 1383, 1386, 1397 
(Aeclanum). 

Dignatio= esteem, regard. Add C. I. L. 9 729, 1681: dig- 
nationem sensi; per dignationem suam. 

Immunis probably = disobliging, the opposite of munis: 
Plautus Trinummus 1 Amicum castigare ob meritam noaiam 
Inmoene est facinus: Cicero Laelius § 50 non est amicitia... 
immunis neque superba. 

*Innullare, to set at nought: Bobbian translation of St Mark 
9 12, ut multa patiatur et innulletur. (P. 5 of Wordsworth and 
Sanday’s Old Latin Biblical Texts, no. 11.) 

Inoffensibilis cursus, smooth, without stumbling, Cledonius 
p. 9 Keil. 

Instructus -%is, equipment: add Servius Aen. 5 402 quibus 
(caestibus), quorum instructu. 

lus commune in a popular sense=common morality: 
Sidonius Epist. 2 10 nimiwm qui supergressi ius fasque com- 
mune summam beatitudinem eaxistimant summam potestatem. 
Ius gentium=law universally recognized. Ib. 1 7 cum Bur- 
gundionibus ture gentium Gallias diwidi debere confirmans. 

*Lausia, apparently =a stone tablet: Lew Metalli Vipasc. 
54. Hiibner explains the word by comparing it with the 
Spanish losa, Portuguese lousa or louzia,= tabula lapidea. 

Lncet aliquid alicui de aliquo. This construction is found 
Aen. 6 502, cur tantwm de te licuit, which Conington says has 
not been illustrated. I have found an instance in Seneca De 


NOTES IN LATIN LEXICOGRAPHY. 69 


Const. Sapientis 10 3, sapiens a nullo contemmitur...nullique 
tantum de se licere renuntiat sibi. 

Lumino. Add C. I. L. 3 45 (Thebes in Egypt, Α.Ὁ. 134) 
prolata Oceano luminat alma dies. 

Notare = to convict, with genitive of the offence: add Pom- 
peius p. 201 Keil, notavit (grammaticos) imperitiae. 

*Olli=tum (locative of ollus). The existence of this word 
is implied by Servius twice, in his notes on Aen. 1 254 and 
5 10: compare perhaps Probus Appendix p. 199 Keil, olim, 
non olt. 

*Paracentia Minerva = Berecyntia, C. I. L. 9 1539, 1540, 
al. saep. 

Pectenarius (sic), C. I. L. 9 1711 (Beneventum). 

Publica opinio, the general opinion, Servius Aen. 5 527, 
6 136. 

*Recisamen, a chip, Lex Metalli Vipasc. 29, ex recisaminibus 
ramorum. 

Recuro, to repair: add Lex Metalli Vipasc. 43, vestimenta 
γάτα vel recurata. 

Remeabilis, returning: add Servius Aen. 5 251, flexuosa et 
in se remeabilis purpura. 

Rutramen, rubbish, earth ὅσο. (what is dug up by a rutrum): 
Lex Metalli Vipasc. 47, pulvis ex scaurits et rutramina. 

*Scaurarius, one who has to do with the slack or scoriae of 
a mine: Lex Metalli Vipasc., seripturae scaurariorum et testa- 
riorum. ‘The spelling scauria, which occurs in the same lex 
(47), is unknown to the lexicons. 

Splendor, splendidus, in the sense of spotlessness, spotless. 
Cicero Rep. 2 § 69, wt sese splendore animi et vitae suae sicut 
speculum praebeat civibus ; Cluent. § 46 (Aletrinatiwm) muni- 
cipum, in quibus quantus splendor sit, quam prope aequabilis, 
quam fere omnium constans et moderata ratio vitae...nemo... 
ignorat. Plane. § 30, hunc tu vitae splendorem maculis aspergis 
wtis? Horace 4 Od. 7 21, et de te splendida Minos Fecerit 
arbitria: where the Cruquian scholia say cum de te absque ullo 
fuco aut obscuro odi invidiaeve livore splendide et lucide 
iudicavertt. Livy 3 35 9, nequaquam splendore vitae pares de- 
cemviros creat, 


70 THE JOURNAL OF PHILOLOGY. 


Tignuarius: add to the instance of this form given by 
Georges C. I. L. 9 2213 (Telesia) 2339 (Allifae). De Vit gives: 
others. 

* Ubertumbus. Apparently. the epithet of a place outside the 
limits of a particular mine, whence the same ore can be pro- 
cured. Lex Metalli Vipasc., ew aliis locis ubertwmbis. 

Usurpo, to use a word in a wrong sense, Servius Aen. 5 145; 
‘carcere’ usurpavit: to make an incorrect statement, Ib. Aen. 
7 706, usurpat hoc. 


H. NETTLESHIP. 


THE TITLE OF THE SECOND BOOK OF NONIUS. 


THIS is given by Mss, and editions as De honestis et nove 
veterum dictis. Honestum dictum in the writings of a gram- 
marian or lexicographer means a good word, as opposed to 
a bad or vulgar expression: Servius Aen. ὃ 107 on inter opacum 
nemus: (‘Inter’) est honesta elocutio, ut si dicas ‘inter cenam 
locutus sum’, id est ‘per cenam’. But there is a difficulty in δέ 
nove. In the first place, an ancient author would not naturally 
be spoken of by a late grammarian as speaking nove: in the 
second place, it is difficult to see how De honestis et nove vete- 
rum dictis can be Latin. One would have expected novis. 
I would therefore raise the question whether the true title of 
the book is not 


ECLOGE 
De honestis veterum dictis. 


Ecloge might easily have been corrupted into e¢ nove, and 
then have crept into the text. 


H. NETTLESHIP. 


ON THE HEBREW ROOT YP AND THE WORD ΡΟ. 


THE word ΝΘ is taken by lexicographers and comment- 


ators to mean an inside corner or reentrant angle. But when one 
tries to realise the description of the eastern wall of Jerusalem 
in Neh. iil, a passage in which the word repeatedly occurs, 
doubts arise as to the adequacy of the received interpretation. 
In attempting a more precise determination of the sense of the 
word one is led to look at several points of some interest, partly 
etymological, partly exegetical and topographical. 

The Hebrew Bible exhibits only one certain verbal form of 
the root Y¥), viz. the Hiphil Y¥p in Lev. xiv. 41, which plainly 


means to “scrape” a wall’, The sense of the root thus indic- 
ated is confirmed by the nominal form niyype in Isa. xliv. 19, 


for though most of the representatives of exegetical tradition 
fail us in this passage’, the Targum has spin (DTN) 
=Syr. Jad 8} or flat (P. 5. 1134), LeX%) (Bar Ali ed. 


Hoffmann 3517), Ar. a-<51 , that is, σμίλη, scalprum, a scraping 
or graving tool. And this rendering must be correct, for the 
context demands the name of the instrument used by a wood- 
earver to give human form to a block after the pattern has 
been marked upon it by lines and punctures. In this sense 


ysp answers to Arabic ree in éclas “fine dust” (Qamis), 


niyyp , Exod, xxvi. 23, is only 2 LXX and Pesh. are puzzled and 
another pronunciation of Mip¥pd (Ew. shorten the verse to avoid the diffic- 
215 a), and nipypnn. Ezek. xlvi. 22, ulty. Aquila and Jerome have ἐν περι- 
ee se γωνίοις (or περιγωνίσκοις), in angulari- 


is deleted by puncta extraordinaria bus—a mere guess from Ji¥P = γωνία. 


and so omitted by the versions. 


72 THE JOURNAL OF PHILOLOGY. 


and thus it seems very possible that My 'sp, κασία, means 
simply “powdered” fragrant bark, and is not to be compared 


with the Arabic synonym daw “decorticated.” 
In Lev. /. c. the Hiphil of ΝΡ is followed in the same sense 


by the Hiphil of A¥P (¥P7 ver. 41, Mi¥PT ver. 43). Neither 


Hebrew usage nor the analogy of the cognate tongues affords 
any support for ascribing the sense “scrape” to A¥P, and a 
_ change of verb in the same context seems highly improbable. 
On the other hand there are familiar instances of a corruption 
of text arising from the dropping of an } in pronunciation, such 


as ΠΡΟ) for ΤΡ.) (Amos viii. 8 Kethib), ‘tah for jtayd (Ps 
xxviii. 8), DOES jv for DDN jWy’ (Hos. vii. 6, cf. Deut. 
xxix. 19); and the ἢ of NSP, which makes the chief difficulty 


in applying this solution to the case before us, may have come 
from the FN following. That the text is not sound but corrupted 
from FN ΡΠ might indeed be strongly suspected from 


the punctuation of the first syllable with 7 instead of d@ The 
possibility of such a change in infinitives Piel and Hiphil is 
affirmed by gramiarians, and less cautious writers like Bottcher 
offer long lists of cases in which the thing occurs. But by far 
the greater number of cases cited by Bottcher (Lehrbuch, p. 226) 
are manifestly perfects standing in dependence on a construct 
case (cf. Philippi, Status Constructus im Hebrdischen, p. 79 sqq.), 
or at least could be regarded as such by the tradition followed 
by the punctuators. A shorter and much better chosen list is 
given by Ewald § 238d, who judiciously remarks that in all 
cases this infinitive is found after a noun or a longish (i.e. 
nominal) preposition, and so in a position which would readily 
admit of a finite verb. But even Ewald’s list may be at once 
purged of seven out of the fourteen passages cited (Lev. xiv. 46 ; 
Num. xxi. 35, where "NW may be taken as intransitive and 
9 as its subject—note also that LXX read 7) for 35), on 


which reading NWN is a perfect even if it is taken trans- 
itively; Deut. xxvii. 55; 1 Kings xv. 29, where the suffix is 
objective; 2 Kings x. 11, 17, cf. Jer. xxii. 14 and. Philippi wt 


THE ROOT νὰ AND THE WORD γ᾽}. 783 
supra ; Jer. li. 33). Again pny in Jer. xxxi. 32 is not to the 


point, for seghol represents % as well as i, and the Assyrian 
punctuation in Cod. Petrop. actually has 4". This leaves only 
six cases, including our passage. Of these 1 Chron. viii. 8 is 
thoroughly unintelligible and lacking in the Syriac—moreover 


certain copies of LXX point to a reading ney; and in 
Jer, 1. 34 the second of the two forms PJM and PAW was 


certainly taken as a perfect by the oldest interpreters (LXX, 
Aquila). Again in Deut. vii. 24 (23) the true reading of the LXX 
is ἐξολοθρεύσῃ, answering to a reading DMix THY (see 


Field’s Hexapla in 1.), and similarly in J ak x1. 14 ἕως ἀπώλεσεν 
αὐτούς answers to ὩΣ TY or MIN WHY Ty, in either 
of which the perfect is correct. Here the Massoretic text simply 
arises from a fusion of these two texts without application of 
the correction necessary to make the conflate text grammatical, 
and in like manner in Deut. vii. 24 the vowel of the perfect in 

TW is nothing more than a reminiscence of a reading in 
which there was no suffix. The same explanation may be 
safely extended to the one passage not yet mentioned, Deut. 
xxvill. 48, though here we cannot tell whether the LXX had 
Tow or ὙΦ. And finally in the passage with which we 


started no one would take yon for anything but a perfect un- 
less it were followed by Disp. The result of this enquiry, 


then, is that there is no sure example of i for ἅ in the penult 
of infinitives of intensive and extensive stems in Hebrew, and 
those who are familiar with the state of the Hebrew text as a 
whole will find it far easier to believe that five or six isolated 
cases of i for & have arisen by conflation of readings or other 
mistakes than that they rest on an original and true tradition. 
There is no doubt that the Massoretic punctuation represents 
with scrupulous fidelity a fixed and elaborate tradition of pro- 
nunciation which is much older than the vowel signs them- 
selves, and which in some cases, as appears in the examples 
that have just been cited, even embodies elements older than 


1 In 2 Sam. iii. 13 the text is corrupt. LXX read NN3N ON 32. 


74 THE JOURNAL OF PHILOLOGY. 


the present consonantal text. But the simple fact that the 
Massorets undertook to give a pronunciation for every word in 
the text, and by implication professed to understand every 
passage of the Old Testament, including many that are totally 
corrupt, shews that the tradition on which they went cannot be 
primitive. Hitherto systematic criticism of this tradition, which 
is the foundation of Hebrew Grammar, has hardly kept pace 
with the criticism of the consonantal text, but both are alike 
indispensable to real progress in Old Testament philology. 
Returning from this digression let us observe that while the 
root idea of “scraping” suits the other O. T. words derived from 
Y¥P, it affords no conceivable explanation for ΡΟ. And 
as Hebrew ¥ corresponds to three original Semitic consonants, 
represented by Arabic .», ὁ and 4, Aramaic ¥, Y and { © 
respectively, it seems not unlikely that two distinct roots are 
united in y¥p. In point of fact the root of yi¥pd is generally 
identified with Aramaic YO), and so the word is taken as 


meaning literally “a place where something is cut off” or ends 


abruptly (Arabic νὰ ὁ. At first sight this seems objection- 
able because the { of the Aramaic root is found also in the 
Arabic bi, and therefore not ¥ should appear in Hebrew 
also. The regular Arabic equivalent of yp = Aramaic yop 
would be a non-existent #43. It is however to be observed 
that the combination &3 appears to be avoided in Arabic. 
There is but one certain example of it, viz. ba) = Ppa; for 
43, and &4i, which Freytag gives from the Qamiis, are not 


acknowledged by Jauhari and seem to be known only from 
obscure traditions in which there were various readings with 


Ὁ instead of & (see the 7) al-Aris). It is therefore not 
impossible that -43 stands for 243 under the influence of 
the ,3. The same phonetic change can perhaps be traced in 
another case. The Arabic biw “fall” can hardly be equated 
to Heb. pe, but may be compared with 1p’: for which the 


THE ROOT ΝΡ AND THE WORD Yi¥P. 75 


etymological sense of “refuse” (bin, d&bli.) is altogether suit- 
able. Here however the Aramaic fails us, so that we have no 
means of testing our equation; and altogether the material 


available for an induction as to the behaviour of Ὁ after 3 is 
too scanty to allow us to say more than this: that the deriv- 


ation of yiypo from a root YSp = yop -- εὉ is worth con- 


sidering if it is borne out by the usage of the word’. 

As regards the usage of the word the oldest evidence is that 
of the book of Ezekiel. In Ezek. xlvi. 21 sq. the ΓΊΝΩ of a 
court seem to be its four corners viewed from inside, and if 
this passage stood alone we should find no occasion to ques- 
tion the adequacy of the usual interpretation of the word. 
But in view of other passages, presently to be discussed, it 
ought to be observed that in this place the four corners are 
represented as cut off by walls so as to form four smaller courts, 
or cantons as they may be called in the language of heraldry. 
And in the Mishna also ΝΡ is used not of an angle as such, 
but of a canton or space cut off from an angle to contain a 
chamber (Tamid iii. 3; Middoth τ. 5). Now in Ezek. xli. 22 it 
is plain even as the text stands, that the Miyypp of the altar 
are a part of its structure. And this comes out more clearly 
when we note that the reading b YOYSP) is conflate, and 
that one of the earlier texts, represented by LXX, Tgm., read 
b DIYS, καὶ κέρατα εἶχε, while the other and better text, 
which is followed by Hieronymus, omits 1b, and with the 
necessary correction of Ἰ) δὲ for SN (LXX βάσις) gives the 
sense “and its corner posts its base and its walls were of wood”. 
The sense of our word which the context demands here is con- 
firmed by the Rabbinical use of yispd for a “pillar (corner 
post) of the law”; cf. στύλοι in Galatians ii. 9. The posts of 
the altar, to which the boards of the framework are nailed, must 
of course stand inside the framework, and so will cut cantons off 
the angles as seen from inside, in exactly the same way as the 
small courts in chap. xlvi. cut cantons off the great court. 


1 It may be added that the combination ob is also unknown in Arabic roots. 


76 THE JOURNAL OF PHILOLOGY 


The next evidence is found in the description of the taber- 
nacle (Exod. xxvi. 23, 24; xxxvi. 28, 29). There is no doubt 
that the tabernacle is to be conceived as having the same pro- 
portions as the temple, thirty cubits long and ten broad. And 
it is natural to take these as the outside measurements of the 
boarded walls. In that case the arrangement of the boards can 
be explained simply enough. On each side there are twenty 
boards making up the full length of thirty cubits. The six 
boards for the end make nine cubits and thus just fill up the 
line between the side walls if we assume the latter to be half 
a cubit thick. But some kind of strengthening is wanted at 
the angles, and this is provided by two boards serving “as 
corner posts for the tabernacle at its inner part” (xxvi. 23). In 
verse 24 we should probably read ΙΝ a second time in 
place of ὩΣ ; at all events the versions take these words in the 
same sense. The verse will then mean that the two corner 
boards are twin pieces to the extreme boards of the end wall, 
z.e. are applied face to face with them inside, so as in fact to 
double the thickness of the end wall for the space of a cubit 
and a half from each corner. Thus thickened from top to 
bottom the end wall comes up against the first ring of the bars 
that lock the boards of the side walls together. The side walls 
of the tabernacle, we must assume, are to be set up first and 
bolted together by their bars. Then the end wall is set up, 
and when the corner pieces are placed they come close against 
the first ring (z.e. the ring nearest the end) and keep the bars 
from sliding. If this explanation is correct the yispio of the 


tabernacle is precisely similar to that of Ezekiel’s altar; and no 
other view seems to allow of the two corner boards being similar 
to the others, or to give a reasonable sense to ver. 24. 

We have still to consider the yspd in the wall of Jeru- 
salem. If the word means no more than the point at which 
the wall turns and forms a reentrant angle it is plain that the 
same turning cannot be referred tu in Neh. i. 20, 21 and in 
verses 24, 25 of the same chapter. But from 2 Chron. xxvi. 9 
it appears that “the Y¥PD” was as familiar a feature in the 
fortifications of Jerusalem as “the corner gate” or “the valley 


THE ROOT yxp AND THE WORD yp. "7 


gate” [Tyropoeon gate]. And from the same passage it appears 
that it was crowned by a tower, which must be identical with 
that mentioned in Neh. iii. 25 sqq. In Neh. ui. 19 we find 
that the eastern wall of the city, running along the face of the 
Kidron valley in a direction approximately northerly, reaches 
a point “in front of the ascent to the armoury at the yispid”. 
As this ascent must have gone northwards up the hill, and the 
wall evidently did not continue to run alongside of it, the 
natural assumption is that the line of fortification now turned 
eastward, and at verse 24, at a distance which can hardly be 
less than two or three hundred feet, we come to another corner, 
but also are once more at the Ὁ. Or rather we are still 
there, for the line extending “from the Θ᾽ (ver. 20), “as 


far as the yispd” (ver. 24), 1.6. as far as the yispd reaches 


(cf. the use of YW as a conjunction in the sense of Latin dum), 
appears to run along the whole length of that feature. And as 
there is here a corner (735) the next section of the wall must 
again have a more northerly direction. This piece (ver. 25) 
faces the yispi and the tower by which, as we have learned 


from Chronicles, the yixpd was crowned or commanded. That 


tower, it is obvious, was not part of the outer circuit of the wall, 
but belonged to an inner line of defence consisting of fortific- 
ations belonging to the upper palace. In v. 26 we must write 
navn for DD’) I, and then it appears that the next 
succeeding piece of wall, repaired by Pedaiah and the Nethinim 
of Ophel, still lay opposite (7.e. under) the great tower, and 
ended at a point east of the water-gate. The water-gate there- 
fore was not in the outer wall but belonged to the inner fortific- 
ations connected with the palace. And finally in verse 27 
another group of builders still work under the great tower and 
carry on the wall to its junction with the wall of Ophel. 

We have here a complicated topographical problem, the 
solution of which can be effected only by repeated trials on 
a contoured plan of the site. Without carrying the reader 
through this process I will state the only solution which I have 
found to answer the conditions. The Haram area as it now 
exists has been levelled up by retaining walls both on the south 


78 THE JOURNAL OF PHILOLOGY. 


and on the east. The temple and the royal palace adjoining it 
(and this palace as distinguished from that of David lower 
down the hill must be meant by the upper palace of Neh. 
111, 25) cannot be supposed to have had such vast substructures 
as now exist; but even Solomon’s buildings, not to speak of the 
additions made by his successors, cannot be placed anywhere on 
Mt Zion (I use this word in the Biblical sense, not in the 
incorrect modern one) without the use of banking and a retain- 
ing wall (cf. 1 Kings vii. 10). But it appears to follow from 
the passage before us that before the exile and at the time of 
Nehemiah the plateau of the royal buildings and temple was 
not carried out to the 8.E. angle of the Haram area, Here on 
the contrary there was a reentrant angle against which the bare 
hill-side formed a kind of buttress. This is the way in which 
most level ground could be got with least building, if, as is 
evident from Nehemiah’s description, the prison court of the 
palace lay on the slope of the hill, overhanging the Kidron 
valley, where the ground falls away to the south and east. To 
make this plain I introduce a plan of a proposed reconstruction 
of the temple plateau prepared for the article TEMPLE in the 
Encyclopaedia Britannica. Most of the details do not concern 
us here; the parts with which we have to do are as follows. 
QP is the southern retaining wall of the Haram area from — 
Robinson’s arch (Q) to the Triple Gate (P). The eastern part 
of this wall existed in the time of Nehemiah, having been built 
to support the arsenal or House of the Forest of Lebanon (/). 
Then from P to # the retaining wall ran northward, and as the 
hill rises the depth of the wall constantly decreased, till near # 
the level of the rock corresponded with that of the plateau, and 
the retaining wall disappeared. But at # a new retaining wall, 
parallel to PQ, began (that of the prison court) and ran east- 
ward to G. Along its course the hill falls away and the retain- 
ing wall became deeper and deeper. At Ο' it again turned 
north, and at the corner presented the appearance of a huge 
projecting bastion. This solid bastion, with the upper works 
which would naturally be constructed on it, I identify with the 
great projecting tower of the palace which has so prominent 
a place in Nehemiah’s description. It will now be seen that 


THE ROOT yp AND THE WORD PSPS. 79 


atwec P and @ the hill lies like a buttress against the re- 
taining wall, and that at J, where it reaches the level of the 
inner plateau, a gate would naturally be placed. Here there- 
fore, in the wall PF and very near to 1, I place the water gate, 
which from Neh. iii. 26 appears to have faced the east, and 
which, from its name, must have opened on a road descending to 


200 450 600 Fret 
209 300 40υ Cusits 


the Kidron valley. Nehemiah’s wall, running northward from 
Siloam at some height above the valley, reached a point near P 
where an ascent led up to the site of the old arsenal (H), 
This ascent I identify with the “staircase at the wall above 
David’s house” (Neh. xii. 37) which led straight up from the 
lower part of the city to an open space (Neh. viii. 1) inside 
the water gate. That the water gate lay on the eastern 


80 THE JOURNAL OF PHILOLOGY. 


side of this space seems to be implied in xii. 37. The 
arsenal of course had been burned down by Nebuchadnezzar. 
From Neh. iii. 20 to iii. 27 inclusive the wall described is 
an outwork, enclosing the buttress of rock between P and @ 
(on which there were houses), and giving additional security 
to the important water gate. From a comparison of verses 
19, 24 and 25, it appears that the ΡΟ was enclosed in this 
outwork and was near to every part of it. The word therefore 
can hardly mean anything else than the buttress of rock, con- 
ceived as a great natural corner post sustaining the reentrant 
angle of the plateau’. The line taken by the outwork must 
have been determined by the contours of the hill and cannot be 
guessed at with precision without new excavations. The rock- 
contours laid down on the plan from the Palestine Exploration 
surveys are largely conjectural. But from the staircase to the 
“corner” (M357) it must have run nearly parallel to LG, and then, 
instead of turning due north, it must have still had some east- 
ing. For the part immediately beyond the corner is said to 
face both the rock buttress and the great tower at G (verse 25). 
And as the point due east of the water gate is not reached till 
verse 26 this implies that the line of the wall in verse 25 faced 
the south side of the great tower and therefore had something 
of an easterly direction. The part of the outwork spoken of in 
verse 27 may, on the other hand, have run more nearly north 
and south, for it evidently faced the east side of the great 
tower, ending in the wall of Ophel. Ophel, or rather “the 
Ophel”, in the Old Testament is a fortress (Isa. xxxii. 14; 2 
Chron. xxvii, 3, xxxiii. 14) and apparently the royal stronghold 
(Micah iv. 8). From our passage one is led to view it as a 
fortress forming the north-east part of the royal buildings, and 
so we must understand that on reaching the Ophel wall the 
outer wall of the city terminated in the fortifications of the 
palace plateau. In like manner, in the time of Josephus the 


1 On the analogy of Ezek. xlvi. it but this explanation, while topographi- 
might equally well mean the whole cally it comes to the same thing, does 
space between the outwork and the ποῦ explain the expression ‘‘ opposite 
reentrant angle of the retaining walls, the Y)Sp'”’ in ver. 25, 


THE ROOT Ρ AND THE WORD isp. SI 


first wall of Jerusalem terminated at Ophel in the eastern wall 
of the temple area (B. J. v. 4, § 2). 

In confirmation of this argument it is worth noting that the 
part of the wall which I treat as an outwork was not included 
in the procession of either of the companies described in Neh. 
xii., in the story of the dedication of the wall. The company 
that took the northern half of the circuit went no farther than 
the prison gate, which may be placed in the prison court of the 
royal buildings, and the other company, coming from Siloam, 
went straight up the stairs at P. On the view now developed 
this is quite intelligible; the outwork was of secondary import- 
ance, the main fortifications lying behind it and consisting 
essentially of the great retaining walls, which had not been 
destroyed by Nebuchadnezzar, and indeed were practically 
indestructible. 

The final result of this examination is that a ah ype) always 
implies something of the nature of a support or buttress against 
the inner side of an angle where two walls meet, whether that 
support be a mere post in a wooden structure, a solid buttress 
of natural rock, or what I may call a hollow buttress formed by 
two walls enclosing a court or chamber. 

In its form the word is a nomen loci; therefore it 
‘primarily means the place where such a feature occurs, and 
only in a secondary sense the support or buttress itself. 1 
there is anything in the equation YEP = Ἐ 83 the term may be 
explained etymologically by observing that, where a post or 
buttress occupies the inside corner of a court or the like, the 
sharp angle of the two walls is cut off. 


W. ROBERTSON SMITH. 


Journal of Philology. vou. xvt. 6 


ON THE FRAGMENTS OF THE LATIN HEXAMETER 
POEM CONTAINED IN THE HERCULANEAN PAPYRI. 


A BRIEF account of these fragments will be found in Bahrens’ 
Poetae Latini Minores τ. 212—214. Cf. Riese A. LZ. π΄. pp. 
3—5. My own attention was first called to them as far back as 
1863, when, owing to the efforts of Prof. Gomperz and Mr 
H. O. Coxe, the box in which Hayter’s transcripts (made early 
in the century at Naples) were preserved, was, after prolonged 
search, discovered, and found to contain a treasure of no ordinary 
kind. In June of 1863 I copied the 8 folios in which Hayter, 
one of the most expert of draughtsmen, has reproduced the forms 
of the Roman letters, and the smallest particulars of the writing 
of the first century A.D. with a minute exactness which proves 
how well he was fitted for his task. The publication in 1885 
of Prof. Walter Scott’s Fragmenta Herculanensia (Clarendon 
Press), in which he has availed himself of Hayter’s Greek tran- 
scripts, and to which Mr W. M. Lindsay has added by way of 
appendix a most careful facsimile of the Latin 8 folios above- 
mentioned, gives me an opportunity of printing some remarks 
which a re-perusal of the poem has suggested. 

Fragm. A. 7. 

The ordinary prolongation of the letters R and 4 as exhi- 
bited in the transcripts hardly leaves room for five letters. 
Hence it is more probable that ADsIDUO should be read than 
apsipuos. This is the only point of doubt which the v. 
admits of. 


CALLIDUS . ADSIDUO TRACTANDO . INMUNERE . MARTIS 


POEM IN THE HERCULANEAN PAPYRI. 83 


Fragm. Β. 4. 
In 1863 following Hayter’s transcript I supplied the missing 
letters as follows: 


MAGIS . QUAM . SJ NOS JSTA LATERENT 
and I still think it may be so. 


5. /UM/UPER IUS . PELUSIA OENIA . CAESAR 

The letter before 1US is not L but T. The antithesis Pelusia 
suggests LATIUS. Whether CUPERET or SUPERET (? superat) pre- 
ceded, I leave it for others to elicit. Merkel’s supplements will 
be found in his larger edition of the Tristia and Ibis p. 332. 
I will here propose a conjectural restoration of this passage, 
which though closely following Ciampitti’s differs in some im- 
portant particulars. 

CUM CUPERET LATIUS PELUSIA MOENIA CAESAR 

COEPERAT IMPERIIS ANIMOS COHIBERE SUORUM 

QUID CAPITIS IAM CAPTA? IACENT QUAE SUBRUTA UERBIS 

SUBRUITIS FERRO MEA MOENIA? QUONDAM ERAT HOSTIS 

HAEC MIHI CUM DOMINA PLEBES QUOQUE . NUNC SIBI 

VICTRIX 
VINDICAT HANC FAMULAM ROMANA POTENTIA GENTEM 


Of the restitutions I am responsible for LATIUS, SUBRUTA 
UERBIS, GENTEM. Most of the rest with the punctuation is 
Ciampitti’s, whose view of the passage I hold to be sub- 
stantially right. Caesar tells his soldiers not to look for any 
great resistance from the Alexandrian rabble and their queen. 
There was a time when Cleopatra and her people were real and 
formidable foes: now the Roman power was victorious and 
might assert its right to superiority over a fallen foe. 


Transcript C. 3. 

The first word of this v. is lost. The next two are unmis- 
takably ALEXANDRO THALAMOS, ‘Then follows /NER/RE . DE- 
/RUM. Obviously this is not intrare, but onerare. 


4. I have no doubt that Kreyssig’s conj. twmultus, which 
quite suits the truncated remains of the letters, is right. 


8. This v. is sufficiently made out in all but one word, 
Multa vetustatis nimio honoris. 
6—2 


84 THE JOURNAL OF PHILOLOGY. 


Of the lost word c ERAT remain, ie. according to 
Hayter’s transcript : Ciampitti read c EDAT, and thought » 
the word was CONCEDAT. The space however takes up more 
than three letters ; and it seems likely that CONQUAERAT should 
be read. 

Transcript Ὁ). 2. 


The 4th and 5th words of this v. were, I think, VETZRIS 
CULP4AE ; the space is too large to make CURAE probable, 


3. Here our transcript is invaluable. It supplies the word 
indubitably. 

QVA/ IGITUR . SEGNIS . /T//NNUNC. QUAERERE . CAUSAS . 
1.6. of course ETIANNUNC. 


4. The first word was perhaps not PARTHOS but PARTHICA. 
The passage would construe if we read the two vv. thus 
est mihi coniunx, 
Parthica, si posset, Phariis subiungere regnis 
Qui statuit, nostraeque mori pro nomine gentis, 


Cleopatra is speaking of Antonius, who has identified his 
cause with her own and that of Alexandria. But statwit (Ciam- 
pitti) is very doubtful. Lindsay gives very clearly QUI.. 5 then 
a lacuna of 4 or 5 letters, then an imperfect letter which may 
be U or N, then NIT: and so Hayter. 


Transcript Εἰ. 1. 

///CTUMQ/////]/M QUO NOXIA TURBA COIRET. 

At the beginning of the y. not fewer than 3, perhaps not 
more than 4 letters are lost. Possibly ERECTUMQUE LOCUM. 
The poet seems to describe the ἠρίον or raised tomb to which 
Cleopatra betook herself before her death, and into which she 
caused serpents to be conveyed (ποῖα turba). The subject 
changes in v. 2 | 


PRAEBERETQUE . SUAE . SPECTACULA . TR/STIA . MORTIS. 
to Cleopatra. 


Transcript Εἰ, 2. , 
AUT PENDENT IS CERUICIBUS . ASPIDE . MOLLEM . 


ove NS i gid ined er rne 


POEM IN THE HERCULANEAN PAPYRI. 85 


Possibly GAU are the missing letters. The space would thus 
be sufficiently filled, as each of these three letters sprawls as 
written in the papyrus elsewhere. 

Transcript G. 5. 

The immediately preceding vv. describe Atropos eying the 
unhappy Cleopatra as she wavered between diverse forms of 
death. 

_ 3. Haee regina gerit. procul hance occulta uidebat 
Atropos inridens inter diuersa uagantem 
Consilia interitus. 
Then the transcript gives, as I read it, though the 0 may 
have been a A, 
QUAM TAM . OUA FATA MANERENT 
This must be, I think, 
QUAM TAM NOUA FATA MANERENT 
ΑΜ and T which I have italicized are, notwithstanding much 


of them is lost, beyond doubt: but the loss of N before ὁ 
(through which Lindsay’s transcript marks an irregular line, 


hardly traceable in Hayter) has no other parallel in the 


fragments. 


6—S8 
Ter fuerat reuocata dies, cum parte senatus 
Et patriae comitante suae cum milite Caesar 
Gentis Alexandri ad moenia uenit 
Signaque constituit, sic omnes terror in artum 


Of v. 8 the words Gentis Alexandri are certain. The last 
word is given by Hayter VENIS, by Ciampitti VENIT which must 
of course be right. Of AD MOENIA, the last stroke of the A, the D, 
the E and Ν᾿ remain: yet no real doubt can exist about this either. 
But the word between is open to much dispute. Hayter gives 
CA/EN/, but of the letter after A enough is figured to prove that 
it was Ror N. Ciampitti gives c...EN. each of the dots roughly 
representing a letter. If Hayter was right in giving cA, the 
nearest word possible is CARENA, the adj. formed from Carae, 
Carrae, or Carrhae. Steph. B. Κάρραι πόλις Μεσοποταμίας, ἀπὸ 


86 THE JOURNAL OF PHILOLOGY. 


Kappa ποταμοῦ Συρίας, τὸ ἐθνικὸν Kappnvos ἢ καὶ Kappaios’. 
It is however difficult or impossible to see how Augustus can have 
been marching to Carrhae at this time, when Cleopatra was still 
meditating suicide, but had not yet killed herself; afterwards he 
returned home through Syria and might (though I know of no 
passage which states that he did) visit Carrhae. It had 
occurred to me that some part of the walls of Alexandria 
might have been called the wall of the Carrhenes ; but this is 
mere conjecture. If we are to trust Ciampitti, the obvious word 
is CURRENS; and this well agrees with the description of the 
march in the preceding vv. ‘Three morns had passed, and 
Caesar was already at the walls of Alexandria.’ Currere of a 
rapid journey as in Phil. 11. 12. 30, ad legiones Brundisium 
cucurrertt. X. 5.10 is tamquam extruderetur a senatu in Mace- 
doniam et non contra prohiberetur proficisct, ita cucurrit. 


1 It is noticeable that the whole vy. 19. 91 τῶν ἐν Kappas κατῳκισμενων 
is explicable on the hypothesis that Μακεδόνων, hence Gentis Alexandri 
Carrhae is alluded to. A colony of would be strictly appropriate. 
Macedonians had settled there. Diod. 


R. ELLIS. 


KIN AND CUSTOM, 


Kin and Custom go together and imply each other, as do 
Law and State. Law is the enactment of the State: Custom 
is the habit of the Kin. And as Custom precedes Law, so the 
State is preceded by kin or sib associations. The earliest form 
of the State is modelled on that of the sib associations, out of 
which it is developed: and the first laws promulgated by the 
State are but the old Customs committed to writing. We may 
therefore expect to find in the organisation of the State traces 
of the sib associations of which it is an artificial extension; and 
from the Laws of the State we may expect to recover the 
Customs which regulated the life of the Kin at a time when 
Law and State were not. 

As the political States into which the various members of 
the Aryan stock formed themselves were posterior to the dis- 
persion of the Aryan family, it is a necessary inference that our 
Aryan forefathers were held together by sib associations, and 
were governed by Custom. The customs, by which they lived 
in their original Aryan home, they carried with them into their 
new settlements; these Customs became their Laws, and in 
their laws we still find traces of those customs. 

The resemblance between the Gortyna Code and Attic law 
in the matter of inheritance ab intestato, ἀγχιστεία, is too close_ 
to be accidental and not close enough to admit of the theory 
that either set of provisions was imitated from the other, The 
only admissible explanation is that both had a common origin. 
We may therefore fairly use one to explain the other. The 
legitimacy of this, the comparative method, is not likely to be 
disputed in England at least; for, thanks to Sir Henry Maine, 


88 THE JOURNAL OF PHILOLOGY. 


the comparative study of ancient law has been placed on a 
wide and firm foundation. In Germany, too, it might have 
been expected that the comparative method would have been 
applied to the study of Greek law, for as long ago as 1813 
Bunsen made an attempt to explain Solon’s provisions in the 
matter of ἀγχιστεία by a reference to the regulations of Manu 
on the same subject. He said, De jure hereditario Athenien- 
sium, pp. 111 and 112, “in antiquissimo atque sanctissimo illo 
Indorum legum codice, qui a Menu, homine et rege primo, 
Bramae filio, compositus esse fertur, illum juris hereditarii 
Attici successionis ordinem clarius quam in ipsis Solonis legi- 
bus expositum, ejusque rationem et totius juris hereditarii 
sanctitatem declaratam esse monstrabimus.” Yet this fertile 
suggestion has never, to my knowledge, been worked out 
either in England or in Germany. In England the com- 
parative method has been applied exclusively to the elucid- 
ation of Roman law. In Germany the study of Greek law 
has been followed without reference to comparative law. Leist 
indeed has in his Graeco-Italische Rechtsgeschichte (Jena, 1884) 
at last made a valuable contribution to the subject, though 
its value is unfortunately diminished by the fact that he has 
limited himself to Graeco-Italian and Hindu law; and Jacob 
Simon in his edition of the Gortyna Code (Wien, 1886) has 
some references to Slavonic and Germanic law; but both 
like other Germans seem quite to ignore the work of writers 
like Sir Henry Maine. In France, E. Caillemer, the most lucid 
writer who has ever dealt with Greek law, denies many of 
Bunsen’s conclusions, ignores M. Fustel de Coulanges and re- 
jects his method. Where Caillemer disagrees with Bunsen 
Caillemer is usually right. But though Bunsen’s conclusions are 
sometimes wrong, his method was not wrong. It was faultily 
applied. If to this we add the fact that Bunsen overlooked 
many, most, of the points in Greek law and politics which at 
once exemplify and confirm his theory, we shall have the reason . 
why his keen conjecture has never been worked out in Germany 
or France. 

All discussions about ἀγχιστεία, or inheritance ab intestato, 
must start from the words of Sir Henry Maine, Ancient Law, 


KIN AND CUSTOM. 89 


p. 177, “in all indigenous societies, a condition of jurisprudence 
in which testamentary privileges are not allowed, or rather not 
contemplated, has preceded that later stage of legal develop- 
ment in which the mere will of the proprietor is permitted 
under more or less of restriction to over ride the claims of his 
kindred in blood.” Before Solon, as we have known for some 
time from Plutarch, Sol. 21, the power of bequest was unknown’. 
And now we have in the Gortyna Code an example of a body 
of law in which testamentary principles are not contemplated. 
No mention is made of a Will in the Code, but the order of 
succession to property is fully regulated. As however these re- 
gulations closely resemble those which in the law of Solon, 
Dem. c. Macart. 1067, regulate inheritance ab intestato, it is 
clear that the provisions made in Solon’s law were not Solon’s 
invention entirely, but were simply the embodiment and modific- 
ation of the customary order of succession, common to Athen- 
ians, Cretans, Hindus and all Aryans alike. Solon’s enactment 
and the Gortyna Code may have been the first laws in Greece 
which stated the ἀγχιστεία. But custom precedes law, and the 
custom regulating ayysoreia comes down from Aryan times. 
Now, to ascertain the nature of the custom, and the modific- 
ations introduced into it by Solon and at Gortyna respectively. 
The Gortyna Code not contemplating a Will declares that on 
the death of a man his property passes to his children if he 
has any, his grand-children or his great-grand-children*. The 


1 Meierand Schémann’s denial of this 
fact (der Attische Process p. 572 ed. Lip- 
sius) seems to me perfectly gratuitous. 


τέκνα, τούτος ἔκ[εν] τὰ κρέματα. This 
passage of the Gortyna Code—dis- 
covered of course after Leist had 


There is not the least trace of the ex- 
istence of any testamentary power at 
Athens, however limited, before Solon. 
Dem. 6. Sept. 102 is not inconsistent 
with the introduction of this power by 
Solon, ὁ μὲν Σόλων ἔθηκε νόμον ἐξεῖναι 
δοῦναι τὰ ἑαυτοῦ ᾧ ἄν τις βούληται; and 
Plut. Sol. 21 is inconsistent with the 
pre-Solonian existence of the power of 
bequest. 

3. νυ, ἢ ὃ κ᾽ ἀπ[ο]θάνει ἀνὲρ... αἱ μὲν κ᾽ 
εἶ τέκνα @ ἐς τέκνον τέκ[ν]α ἃ ἐς τούτον 


written his Graeco-Italische Rechts- 
geschichte — completely refutes his 
statement, p. 74, that the limitation 
to the third degree of direct descend- 
ants is ‘‘wholly unproved and impro- 
bable.” He says on the same page 
that it is ‘‘decidedly erroneous” to 
imagine that the sui heredes of Roman 
law were limited in any such way. 
But the analogy of the Greeks and 
other nations entitles us to ask for his 
evidence, 


9O THE JOURNAL OF PHILOLOGY. 


so-called Code of Manu, also not contemplating a Will, says, 
ΙΧ, 185, “Not brothers, nor parents, but sons are heirs to the 
deceased.” And in 186 it goes on to say that the fourth in 
descent (i.e. the great-grand-son) is heir in default of nearer 
descendants. At Athens also a man could not make a Will if 
he left legitimate children; and the limit here too was the 
fourth in descent (Isaeus, p. 216). In all three codes, the 
children might divide the property (Gortyna Code v. 30, Manu 
Ix. 104, [saeus, Philoc. ὃ 34), With the Hindus.the brothers 
might if they liked continue to live together on the undivided 
estate (Manu 1x. 105, Gaut. xxv. 1). And this was probably 
the original custom. 

In default of children, grand-children and great-grand- 
children, at Gortyna the deceased was succeeded by his brother, 
his brother’s son or his brother’s grandson®. At Athens the 
order of succession, ab intestato, was the same®, If the brother's 
stock failed, then at Gortyna the sister, her children or her 


grand-children succeeded *. 


1 At Athens, as we shall see, the 
children frequently continued to live 
on the undivided estate. 

2 v.13 αἱ δέ κα μέτις ef τούτον, ἀδελ- 
πιοὶ δὲ τὸ ἀποθανόντος κἐκς [ἀδ]ε[ ἈπΊ]δν 
τέκνα ἂ ἐς τούτον τέκνα, τούτος ἔκεν τὰ 
κρέματα. 

3 Dem. 6. Macart. 1067 ἐὰν μὲν ἀδελ- 
φοὶ ὦσιν ὁμοπάτορες, καὶ ἐὰν παῖδες ἐξ 
ἀδελφῶν γνήσιοι, τὴν τοῦ πατρὸς μοῖραν 
λαγχάνειν, ἐὰν δὲ μὴ ἀδελφοὶ ὦσιν ἢ 
ἀδελφῶν παῖδες, τοὺς ἐξ αὐτῶν κατὰ ταὐτὰ 
λαγχάνειν. From this we may perhaps 
infer that at Gortyna too it was neces- 
sary that the ἀδελπιοί should be ὁμοπά- 
ropes and the τέκνα be γνήσιοι. That 
the heirs, if e.g. there were two bro- 
thers or two brothers’ sons, divided 
equally is patent, and is further im- 
plied by κατὰ ταὐτά. 

4 εἰ δὲ κα μέτις εἶ τούτον, ἀδευπιαὶ δὲ 
τὸ ἀποθανόντος κἐς ταυτᾶν τέκνα é ἐς τὸν 
τέκνον τέκνα. The daughters in this 
case were heiresses (v111. 40) and were 


So too at Athens’. 


According to 


married to the next of kin (vm. 15). 

5 This is distinctly implied in Solon’s 
law, which after the words quoted 
above in note 3, goes on: κρατεῖν δὲ 
τοὺς ἄῤῥενας καὶ τοὺς ἐκ τῶν ἀῤῥένων ἐὰν 
ἐκ τῶν αὐτῶν ὦσι καὶ ἐὰν γένει ἀπωτέρω. 
The words ἐκ τῶν αὐτῶν cannot mean 
“from the same father and mother,” 
for the law has already said that for a 
man to inherit from his brother it is 
only necessary that the brothers should 
be ὁμοπάτορες. Caillemer (Droit de 
Succession p. 93) therefore gives up the 
reading as meaningless, adopts the 
reading ἐκ τούτων from the parallel 
passage in Isaeus, and violently inter- 
polates (ib. 104) the words ἀδελφὰς καὶ 
παῖδας between ἀδελφῶν παῖδες and ἐξ 
This is unscholarly and un- 
necessary. Τῶν αὐτῶν is masculine 
and ἐκ τῶν αὐτῶν means “from the 
same male ascendants.” The law 
says: if there are brothers and sisters 
descended from the same father or 


7A 
αὐτων. 


KIN AND CUSTOM. gi 


the Code of Manu in defauit of sons the deceased was succeeded 
by his brother or his brother’s son or grandson’. Here it is of 
the highest importance to notice the language of the Hindu 
Code, which says (1x. 187), “Τὸ the nearest sapinda after him 
in the third degree the inheritance next belongs,” i.e. after the 
deceased. The identity of the sapinda, with the ὀργεῶνες, 
I have shewn elsewhere*. Sapindas were those persons of male 
sex who united in offering a funeral oblation (a cake) to their 
deceased House-Father; and who continued in Aryan times, 
and in India do continue to this present day, to live together in 
what modern Anglo-Indian lawyers call a Joint Undivided 
Family. If one of these sapindas in his turn died, his share in 
the joint property and his acquests went “to the nearest sapinda 
after him,” i.e. in the first instance to his brother, and then to 
his nephew, provided always that the property did not pass 
further than the third degree from the deceased House-Father, 
1.6, did not pass further than his brother’s grandson. Manu 
expressly says, l.c., “for three ancestors is the funeral cake 
ordained.” At Athens the dpyedves were a man’s descendants 
to the third degree: and tpeyovia or the Triple Descent has 
left traces elsewhere in Attic law. Both at Athens and Gortyna 
the right of inheritance was limited to descendants in the third 
degree, as we have already seen. The same limitation is found 
in other Aryan peoples; and is undoubtedly the survival of an 
Aryan custom, Caillemer therefore is wrong (p. 13) in suppos- 
ing that property could descend further than the third degree : 
Leist (p. 24) is wrong in saying that it passed only to the 
deceased’s nephew, who was descended from the. House-Father 


grandfather, the brothers exclude the 
sisters. This implies (1) that in de- 
fault of brothers, sisters being heir- 
esses succeeded: (2) that κρατεῖν τοὺς 
dppevas only applied to the issue of 
males not to the issue of females (to 
whom as we shall see a different rule 
applied). 

According to Caillemer’s reading, ἐκ 
τούτων, the rule κρατεῖν τοὺς dppevas 
κιτ.λ. would only apply to those des- 
cended ἐκ τούτων, i.e. from the persons 


already mentioned, that is from bro- 
thers of the deceased. But the law 
certainly applied to those who were 
descended from the father of the de- 
ceased, i.e. the deceased’s brother ex- 
cluded his sister. 

1 Leist, p. 24, overlooks this and the 
law of Solon; and excludes the brother’s 
grandson, 

2 Development of the Athenian De- 
mocracy (Griffin and Co,: London), 


Q2 THE JOURNAL OF PHILOLOGY. 


(the deceased’s father) in the second degree. The principle 
which regulated the customary order of succession in Aryan 
times—there was no legal order then—is simplicity ‘itself when 
once understood: if the deceased were himself a House-Father, 
the head of a Joint Undivided Family, he was succeeded by one 
of his descendants, to the third degree, i.e. by his son, grandson 
or great-grandson. If the deceased were not himself a House- 
Father he was succeeded by one of the Joint Undivided Family 
to which he belonged’, i.e. by one of the descendants of his 
House-Father: and as the Joint Undivided Family was limited 
to three degrees, the successor to the deceased was his brother, 
nephew or grand-nephew. 

In Aryan times the question when a man died was not 
who should inherit his property? for in the Joint Undivided 
Family there was no separate property, but who should per- 
form his funeral rites? and when the deceased was a House- 
Father, who should succeed to the office? Finally, when the 
family divided, the question of property arose, and property 
followed the duty of offering the funeral oblation. 

In default of direct descendants to the third degree, and in 
default of dpyedves, or those who had joined the deceased in 
making a funeral offering to his House-Father, 1.6. in default 
of brothers, nephews and grand-nephews, the deceased was 
succeeded, according to the Custom as incorporated in Solon’s 
law, by relatives on the father’s side “as far as cousins’ sons,” 
μέχρι ἀνεψιῶν παίδων. Under the same circumstances, the 
deceased was according to the Custom as embodied in the 
Code of Manu, 1x. 187, succeeded by the Samanodocas. The 
Samanodocas are persons not members of the Joint. Family 
but related to it. They are not descendants of the head of 
the Joint Family, the House-Father, but-are related to him. 
They are descendants of his father. They are called Sama- 
nodocas, offerers of a joint libation, because when the House- 
Father dies they join in offering a libation at his funeral. 


1 So too at Rome: “On the failure the nearest person or class of the kin- 
of the sui (or directdescendants whohad dred who was or might have been un- 
never been emancipated) the Nearest der the same Patria Potestas with the 
Agnate came into their place, that is, deceased.” Ancient Law, 199. 


ee = a Oo 


KIN AND CUSTOM. 93 


In Athens they were called ὁμογάλακτες, because the joint 
libation was—not, as among the Hindus, water but—milk. 
Now when the deceased is not himself a House-Father—and 
the Custom, embodied in the Codes of Gortyna, Solon and 
Manu, after providing for cases in which the deceased is a 
House-Father, proceeds to contemplate this contingency—he 
is as we have seen succeeded in the first instance by his 
Sapindas or opyedves, the descendants of his House-Father 
to the third degree; next, failing them, by his Samanodocas 
or ὁμογάλακτες, that is by the descendants of his grandfather, 
to the third degree. And the descendants of his grandfather to 
the third degree are his uncle, his cousin and his cousin’s son. 

The Gortyna code provides that, in default of direct 
descendants and of ὁμόκαποι, as the opyedves or sapindas 
were called. in Crete (Arist. Pol. τ. i. 7), the deceased shall be 
succeeded by “the proper persons, whosoever,” ois κ᾽ ἐπιβάλ- 
λει, O70 K εἶ, VI. 23. And commentators on the Code com- 
plain that it does not specify who “the proper persons” are. 
Obviously it was perfectly well understood in Gortyna who 
were designated by this expression, or the Code would have 
been more explicit. It is impossible, with the Codes of Manu 
and Solon before us, to doubt that the Gortyna Code incor- 
porates the same Custom as do they. The reason therefore 
why it was unnecessary to use any more precise expression in 
the Gortyna Code was that the Custom was still so firmly 
fixed that explanation was unnecessary. I conclude therefore 
that the Gortyna Code, in default of descendants to the third 
degree and of ὁμόκαποι, called the ὁμογάλακτες to inherit, 1.6, 
the uncle, cousin and cousin’s son of the deceased. 

According to Hindu law, failing Sapindas and samanodocas, 
the sagotras or sakulyas were called to inherit (cf. Gautama 
XVIII. 6, xIv. 13, Apastamba Il. vi. 14, with Haradatta’s com- 
mentary). The sagotras are those bearing the same name 
with the deceased, 1.6. members of the same γένος. As opposed 
to the sapindas and samanodocas—the near kin or agnates— 
they are the remote kin or gentiles. That is to say, whereas 
the sapindas are the descendants of the father of the deceased, 
and the samanodocas of his grandfather, the sagotras are the 


94 THE JOURNAL OF PHILOLOGY. 


_ descendants of his great-grandfather, his great-great-grand- 
father and so on. But as it is difficult to prove such remote 
degrees of relationship it seems to have been the Aryan 
custom to accept possession of the same name, or residence in 
the same village-community, as the deceased, as evidence of 
kinship with him. And the reason is that the village-com- 
munity consisted of off-shoots from a single Joint Undivided 
Family (cf. Dev. of Ath. Dem. p. 29 ff.). Aristotle, Pol. 1. i, 
regards the κώμη or village-community as the natural exten- 
sion of the οἰκέα or Joint Undivided Family. ‘This was the 
case with Saxon settlements in England: “all the primitive 
villages in whose name the patronymic syllable -ing occurs 
were originally colonised by communities united either really 
by blood or by the belief in a common descent,’ Stubbs 
Const. Hist. 1. 81 and cf. Kemble, Sazons I. 580 and 
App. A. 

Thus the members of a village-community possessed the 
same patronymic; and so did members of the same γένος. 
Consequently it is immaterial whether we say that—in 
default of near kin—the deceased was, according: to the 
Custom, succeeded by members of his yévos or of the village- 
community to which he belonged. At Gortyna the latter 
expression was chosen, tds Fouxlas of τινές κ᾽ ἴοντι ὁ κλᾶρος, 
v. 26. 

The distinction between the Near and the Remote Kin 
is common to all branches of the Ayran family: and as its 
origin is not to be looked for in the enactment of any Aryan 
law-giver, or in any convention arrived at by the Aryans; we 
must seek to explain it by the conditions under which the 
Aryans lived, The Aryan village-community consisted of a 
collection of Joint Undivided Families, descended from a 
common ancestor. A man was near a-kin to the members of 
the Joint Undivided Family to which he belonged. His 
Remote Kin were members of other Joint Undivided Families. 
Property passed, according to the Custom, in the first instance 
to the Near Kin, and only in the second instance to the 
Remote Kin. Now as long as the Joint Undivided Family 
was the only mode of family life practised, there could be no 


KIN AND CUSTOM. 95 


doubt who were a man’s Near Kin. But when migration from 
the original home began to break up this mode of family life, 
it became necessary—for the first time—to define the Near 
Kin. And different branches of the Aryan family seem to 
have adopted different definitions. About the principle there 
was no doubt. A man’s Near Kin were those who would have 
belonged to the same Joint Undivided Family, would have 
been under the same patria potestas, had they lived according 
to the old mode. Again there was no doubt as to the limits 
of the Undivided Family and the patria potestas. Three 
generations gave the limit unanimously observed by all 
branches of the Aryan family. But the question, under whose 
patria potestas would the deceased have been? was answered 
in different ways. The Greeks assumed that he would have 
been in the first instance in the Joint Undivided Family, in 
the potestas, of his father: and consequently that in the first 
instance his property would descend to his father’s descendants 
to the third degree, to the opyedves as they were called at 
Athens, or ὁμόκαποι as they were called in Crete. Failing 
ὀργεῶνες, the Greeks assumed that the deceased would. have 
been in the Undivided Family of his grandfather; and that 
therefore his grandfather's descendants to the third degree, 
the ὁμογάλακτες as they were called at Athens, would inherit. 
But this was the limit of the Near Kin according to. Greek 
notions. Other members of the Aryan family however, as for 
instance the Italians, assumed that the deceased would have 
been in the Joint Undivided Family of his great-grandfather ; 
and therefore that his heirs were his great-grandfather’s 
descendants to the third degree, i.e. second cousins. 

But although the Greeks and Romans differed as to where 
the ἀγχιστεῖς as the Greeks called the Near Kin, or the Agnates 
as the Romans called them, ceased and the Remote Kin began, 
they agreed in limiting the Near Kin to the third degree, the 
fourth person, in descent, from the ascendant in whose potestas 
they assumed the deceased to have been: that is they limited 
the Near Kin to the members of the Joint Family to which 
the deceased was assumed to belong, 

But Leist (p. 82), says that to speak of the ἀγχιστεῖς “as 


96 THE JOURNAL OF PHILOLOGY. 


Greek agnates” is “totally unjustifiable’.” He denies as a 
matter of fact that they were agnates; and as a matter of 
theory that the patria potestas had in Greece reached the 
development necessary for the evolution of the idea of agnation. 
The ἀγχιστεῖς group according to Leist (p. 82) is based on 
the obsequium due to parents; the idea of agnation on the 
patria potestas: the difference between the ἀγχιστεῖς group 
and the group of agnates is due to the development of an “ en- 
tirely abstract theory” in Latium, and to a mode of counting 
relationship peculiar to Italy (p. 23): the existence of the 
patria potestas at Rome reduced to one level son and daughter, 
who divided the paternal inheritance: the non-existence of the 
patria potestas in Greece resulted in the total exclusion of the 
daughter by the son from the inheritance. 


Now in Greece no woman could be the head of a household. — 


Every woman was in the Hand of some κύριος. No woman 
had any standing before the law. Only children of the same 
father could divide an inheritance. Children of the same 
mother (and different fathers) could not inherit from each 
other. Only children of the same father could inherit from 
each other. It must therefore be admitted that at first sight 
it is somewhat surprising to find that according to Leist the 
Near Kin were agnates in Rome and were not agnates in 
Greece. “Qui per feminini sexus personas cognatione conjun- 
guntur, non sunt agnati, sed alias naturali jure cognati,” 
Gaius 1. 156. And, says Leist, even amongst the ἀγχιστεῖς 
“416 Verbindung durch Frauen vollgeniigend fiir das Erbrecht 
ist,” (p. 83). In proof of this assertion he refers to Dem. 
c. Leoch. In this case, as Leist truly says, both defendant and 
plaintiff were issues of females. But, I reply, the defendant 
claimed the property as being the adopted son of the deceased. 
The plaintiff did not profess to be an agnate or one of the 
ἀγχιστεῖς. He claimed under the provision of Solon’s law 
which in default of ἀγχιστεῖς calls the Remote Kin to inherit 


1 As however he has already said on the essential differences between it 
Ρ. 71 that ‘‘wir sind vollberechtigt von and the agnation of Roman law, it is 
einem griechischen Rechte der Agna- difficult to know exactly what his real 
tion zu reden” if we always remember _ opinion is. 


KIN AND CUSTOM. 97 


(ἐὰν δὲ μηδετέρωθεν ἢ ἐντὸς τούτων, TOY πρὸς πατρὸς ἐγγυτάτω 
κύριον εἶναι). Leist’s contention, therefore, that the ἀγχιστεῖς 
included persons not agnates, falls to the ground. 

But it may be said that both at Athens and Gortyna in 
default of brothers and brothers’ sons, sisters and sisters’ sons 
were called to inherit’; and that therefore the latter were at 
once ἀγχιστεῖς and cognates, The reply is that only a woman 
having neither father nor brother could inherit. Such a woman 
according to the definition of the Gortyna Code, vu. 40, is 
an heiress: and the next of kin was bound to marry her. If 
he was already married, he must divorce his wife. If the 
heiress was already married, she must leave her husband. |The 
children therefore resulting from the marriage of the heiress 
and the next of kin were agnates, and did not inherit “ per 
feminini sexus personas.” The children of the heiress by her 
former husband were not heirs to the estate. They were not 
ὁμοπάτορες or ἐκ τῶν αὐτῶν with her children by the next of 
kin, They were not ἀγχιστεῖς or agnates: and if they ever 
claimed the estate, as did the defendant in Dem. c. Leochar., 


1 As regards Athens this inference 
seems to me to clearly follow from the 
much vexed clause in Solon’s law xpa- 
τεῖν δὲ τοὺς ἄῤῥενας καὶ τοὺς ἐκ τῶν ἀῤῥέ- 
νων, ἐὰν ἐκ τῶν αὐτῶν ὦσι, κἂν γένει 
ἀπωτέρω. With the interpretation 
given by Meier and Schémann, der 
Attische Process, p. 586 ed. Lipsius, I 
cannot agree, although it receives the 
weighty support of Lipsius, against 
* whose profound knowledge of Greek 
“law it is presumptuous to match one- 
self. According to M. and §S., the 
-uncle and aunt of the deceased are ἐκ 
τῶν αὐτῶν (whatever that may mean), 
and consequently the aunt was exclud- 
ed by the uncle or his sons. Again the 
uncle’s son and the uncle’s daughter 
are ἐκ τῶν αὐτῶν, and consequently the 
uncle’s daughter was excluded by the 
uncle’s son or his sons’ sons. So far 
I agree with M. and 5. But when M. 

and S. go on to say that the same 


Journal of Philology. you, xv. 


principle did not apply to the brothers 
and sisters of the deceased, that there 
is ‘‘eine allerdings auffillige Abweich- 
ung” in the principle applied to 
uncle and aunt and that applied to 
brothers and sisters, I must dissent. 
Brothers and sisters were certainly ἐκ 
τῶν αὐτῶν, and sisters were certainly 
excluded by brothers or brother’s sons. 
There was no variation in the principle. 
It was applied in exactly the same way 
to brothers and sisters as to uncles 
and aunts. 

There remains the important ques- 
tion whether a sister was excluded by 
an uncle of the deceased. As I state 
in the text, when a sister was the only 
member of the Joint Family left on 
the Joint Estate—when she had neither 
father nor brother—she was an heiress, 


and would be necessarily married by 


the next of kin. 


98 THE JOURNAL OF PHILOLOGY. 


they did so not as ἀγχιστεῖς but as τοὺς πρὸς πατρὸς ἐγγυτάτω, 
and in default of all other claimants. 

As Leist has got the facts wrong, we shall not be surprised 
to find that his explanation is wrong also. In Greece as at 
Rome the patria potestas was developed to such an extent that 
son and daughter were reduced to one level and divided the 
inheritance: for in the Gortyna Code it is provided that (with 
the exception of the houses in town and the live-stock) all the 
property shall be divided, a son receiving twice as much as 
a daughter *. 

The inclusion of second cousins in the “cognatio,” so far 
from being peculiar to Latium or due to any specifically Italian 
mode of reckoning, is found in many Aryan peoples. In the 
“Ancient Laws of Wales” (11. 427) we find “the ancestors of 
a person arc his father and his grandfather and his great grand- 


1 I am inclined to think this may 
have been the case at Athens even in 
some cases with the Joint Estate. The 
clause in the Gortyna Code which pro- 
vides for the division of property, is 
quite separate from that which regu- 
lates the order of succession. Now we 
do possess in Solon’s Code the clause 
regulating the order of succession ; but 
owing to our defective authorities we 
do not possess the clause providing for 
the division of property. But I think 
we may infer the existence of such a 
clause from Isaeus de Philoctemonis. 
From that speech we learn that 
Euctemon’s property was divided be- 
tween his two grand-sons and one 
grand-daughter in such a way that the 
grand-daughter received one fifth of 
the property. This has caused great 
difficulty to writers on Greek law. 
Some imagine that daughters and 
mothers divided the property between 
them—but Caillemer, D.S.L. 53, has 
exploded that notion. ‘Others, includ- 
ing Caillemer, incline to reject the words 
πέμτου μέρους in favour of the conjec- 


ture ἐπὶ μέρους. But if we assume that 


at Athens as at Gortyna property was 
divided between male and female issue, 
as 2: 1 we seem to have a satisfactory 
explanation of the difficulty. 

As I have stated above, the provision 
in Solon’s law, κρατεῖν τοὺς appevas καὶ 
τοὺς ἐκ τῶν ἀῤῥένων, seems to apply only 
to τοὺς ἐκ τῶν αὐτῶν, i.e. ‘descendants 
of the same male ascendants”; and 
consequently the provision, κρατεῖν τοὺς 
dppevas x.7.d., Only applies to. the issue 
of males. To the issue of females, as 
on the analogy of the Gortyna Code I 
conjecture, a different provision ap- 
plied, viz. that the male issue of females 
received twice as much as the female 
issue of females. That is why the sons 
of Euctemon’s daughter each received 
twice as much as the daughter of his 
daughter, 

From Dem. 6. Leochar. we see that 
the issue of females could claim the 
estate not as ἀγχιστεῖς but only in de- 
fault of all other heirs and as πρὸς πα- 
τρὸς ἐγγυτάτω. From Isaeus de Phil, 
we see how the estate was divided be- 
tween τοὺς πρὸς πατρὸς ἐγγυτάτω. 


KIN AND CUSTOM. 99 


father: the co-inheritors are brothers and cousins and second 
cousins.’ The same is the case in the laws of the Langobards, 
of Rothar and of Bavaria, and with the Irish Fine. The Saxons 
on the other hand seem to have alone coincided with the Greeks: 
for the Saxon Maeg like the Greek ἀγχιστεῖς extended only to 
first. cousins. 

To say as Leist does that the group of ἀγχιστεῖς has a 
different basis from that of the agnates; because the former 
is based on the obsequiwm of children to parents, the latter on 
the father’s potestas over his children, seems to me fine-drawn. 
The obsequium of children to a deceased parent consisted in 
offering the sacra to his spirit. According to Aryan Custom 
only to males and only by their male issue could such sacra 
be offered. The duty of offering the sacra, and the inheritance 
of the estate went together : and as in Greece no woman could 
offer or receive the sacra, could be the head of a household 
during her life or its House Spirit after her decease, we must 
I think conclude that the ἀγχιστεῖς system was agnatic. 

To specialise is undoubtedly often good, but there is one 
method im which specialisation is dangerous—and that is the 
comparative method. It is because Leist has confined himself 
practically to Greek and Roman law that he has fallen into 
the error of imagining that there is something peculiar in 
the Roman view of the Near Kin,. It is this vain attempt to 
specialise in a comparative science that has led him to limit 
the principle of the “three descents” to the Hindus and the 
Graeco-italians (p. 24); to say that its existence in Hindu 
law “proves nothing for the Greeks and Romans’” (p. 74), and 
that it is obviously the refinement of a later age (ib.). Now it 
is true that there is much in “the sacred laws of the Hindus” 
which is of late date. These “sacred laws” are not revealed, 


1 As however he admits that the ἀγ- 
xtorets and the Agnati find their ex- 
planation in this principle; and as his 
denial of the applicability of the prin- 
ciple to direct descendants in Greek 
law is wholly disproved by the Gortyna 
Code, it is difficult to follow him. He 
admits that the limit of ascendants was 


three: refuses to admit the same limit 
for descendants. He admits that the 
acceptance of an inheritance involved 
the performance of the sacra; refuses 
to admit that the transmission of the 
inheritance was regulated by the same 
principles as the transmission of the 
sacra (p. 74). 


7—2 


100 THE JOURNAL OF PHILOLOGY. 


as the orthodox in India believe; nor are they laws which 
were ever enacted and enforced as were the laws of Solon. 
The so-called Code of Manu or Manava-dharma-sastra, as 
Prof. Max Miiller has shewn, has its source in the Sitras, 
and is at the least post-Vaidic. The Sitras themselves belong 
to the end of the Vaidic period, and are but metrical versions 
of the prose treatises of the Brahmanas, which are themselves 
studies of the Vaidic hymns. Thus as the Code of Manu says, 
11. 6, “the roots of the Law are the whole Veda.” 

But if this is the case the value of these ‘‘ sacred laws” as 
evidence for Aryan times is much weakened. But I venture to 
think that the Manava-dharma-sdstra and other Smritis draw 
on a much more valuable source of information than the Vedas; 
and that is the actual Customs themselves, which have persisted 
in many cases with singular tenacity in India to the present 
day. The Sitras, the immediate sources of the Smritis, were 
the possession of individual families and frequently take their 
names from the families to which they belonged. The Code 
of Manu or Manava-dharma-sastra is called by Prof. Max Miiller 
“the last redaction of the laws of the Manavas.” Now it is 
in a Brahmanic Kin that we may naturally look to find primi- 
tive Customs handed down with considerable fidelity. And 
we find in the Code of Manu, 11. 6, Custom placed by the side 
of the Veda as “a root of the Law.” The various collections 
of sacred laws were then attempts to codify Customs, inherited 
from Aryan times, and to harmonise them with the Brahmanic 
interpretation of the Vedas. This comes out very clearly in 


Apastamba, I. vi. 15, “by this discussion the law of custom. 


which is observed in particular countries or families has been 
disposed of.” This passage is interesting as showing that 
Customs survived in full force in the more conservative Kins; 
and that primitive Aryan customs came into collision, at times, 
with the teaching of the Vedas. 

The Code of Manu although in its present form probably later 
than Apastamba, Gautama, Vasishtha Baudhayana or Vishnu, is I 
am inclined to think a better source of information for primitive 
Aryan Customs than they, because the Manavas seem to have 
been a kin particularly conservative of the Customs they in- 


—— 


KIN AND CUSTOM. IOI 


herited from a time antecedent to the Vedas. Whether any 
given Sitra represents an original Custom or is the refinement 
of a later age is a matter for investigation and a question of 
evidence. The customary order of succession to an inheritance 
has obviously been often modified in the Sttras for the aggrand- 
isement of “spiritual teachers”; but the primitive Custom can 
often be recovered even in the Sitras by a comparison of the 
customary provisions in default of sons, and of the Sraddhas, 
where the “spiritual teachers” had no material inducement to 
tamper with the Custom. In the next place a comparison of 
the Custom followed by other branches of the Aryan family 
enables us to distinguish between Sfitras derived from Customs, 
and from other sources. Finally, as Kin and Custom go toge- 
ther and imply each other, a primitive Custom will be found to 
accord with and be explained by the mode of life prevalent 
when the State had not yet displaced sib associations. 

Now the principle of the “three descents” is commen to 
Teutons and Celts as well as to Greeks, Romans and Hindus. 
The traces of its influence are numerous in the political organ- 
isation of the Greeks, as I have shown elsewhere. It is found 
in Athens and Ionic cities as well as in Crete and Dorian cities. 
It is at the base of the Athenian conception of the yévos, was 
long the condition of citizenship and still longer the condition 
of office at Athens, and in other states (not mentioned by name) 
was even in Aristotle’s time still the condition of citizenship. 
It defines the limit of the obligation to pursue the blood-feud 
in the time of Homer as well as of Dracon, and in Iceland as 
well as in Greece. It explains the hereditary provisions of the 
law of Athens and Gortyna alike; and explains the Greek 
custom hitherto unexplained of naming a son after his grand- 
father. 

We may therefore accept the principle of the three descents 
when it occurs in the Manava-dharma-sastra as primitive Custom 
and not as the refinement of a late age: and we may confidently 
look to find its explanation in the mode of life of the original 
Aryans. That explanation is afforded by the Joint Undivided 
Family. To say that Leist denies the existence of the Joint 
Undivided Family would be inaccurate, for so far as appears he 


102 THE JOURNAL OF PHILOLOGY. 


does not seem to have heard of it. But he does deny that the 
form of Aryan society was Patriarchal, and, relying on Zimmer's 
Altindisches Leben, he asserts that “regelmassig ziehen die 
Neuvermahlten ins neuerrichtete Haus” (p. 65), and that a son 
when he married erected a house of his own and began to be 
independent. As English scholars know, the precise reverse 
of this is the state of the case amongst the Hindus at the 
present day, where the descendants of one father live together 
to the third generation in a Joint Undivided Family. The 
Slavonic House Community shows at this time the same forma- 
tion as the Joint Family of the Hindus. The newly-married 
son's “bride lives under the common roof and the common 
rule” (Dixon, Free Russia, τι. 40). In both cases the Family 
forms a corporation: and the primitive Aryan Household un- 
doubtedly also formed a corporation, for traces of the corporate 
character of the Household are to be found in all Aryan nations. 

Here I propose to offer some proof of the existence of the 
Joint Undivided Family and its corporate character in Greece. 
In Homer, in the Iliad the married sons of Priam together with 
their wives dwell under Priam’s roof; in the Odyssey, Nestor’s 
house includes his sons and daughters-in-law, and Menelaus © 
brings home a wife for his son. The Gortyna Code, Iv. 24, 
provides that a father is at liberty but is not compelled to 
divide his property between his sons. Here it is evidently 
implied that the usual course was for the Family to remain 
joint and undivided during the lifetime of the father; and we 
may reasonably compare the provision of Hindu law, Baud- 
hayana 11. 11. 3, that “a father may divide his property equally 
among all.” After the decease of the father, the sons, according 
to the Gortyna Code, might continue to live in a joint undi- 
vided family, or not, as they liked, v. 30. In Athens the law 
was the same, as we learn from Dem. 6. Leoch: In this case 
Euthymachus had three sons, who on the death of their father 
continued to live on the joint undivided property. Eventually 
one of them died; and the remaining two still continued to live 
on the undivided estate (ἀνέμητον οὐσίαν, ὃ 10). Then one 
of them married, brought home his wife, and still the two bro- 
thers, the wife, and the child by the marriage, continued to 


KIN AND CUSTOM. 103 


live in one joint undivided family, even till the daughter was 
old enough to be offered in marriage by her father to her- uncle, 
The offer was declined and the family continued to live toge- 
ther till the death of the old bachelor (c. Leochar. 10). Here 
I wish to note in passing that the daughter was considered to 
be and is called by Demosthenes (δ 17) the jomt daughter of 
both brothers; and that she was given in marriage even- 
tually and dowered by the brothers jointly (ἐκδιδόασι τὴν 
ἀδελφὴν τὴν ἑαυτῶν ὁ Μειδυλίδης καὶ ὁ ᾿Αρχιάδης). The sup- 
position of polyandry is impossible, for the daughter as we have 
seen was at one time offered by her father to his brother as 
wife. 

In Aesch. c. Timarch. § 102 we find another case of a family 
remaining joint and undivided after the father’s death. In this 
case three brothers lived together; one married and brought 
home a wife; and they continued to live together, until first 
one and then another died, leaving behind them one son, joint 
heir to the whole estate. Another Joint Undivided Family 
occurs in Isaeus de Philoctemonis. In this case Euctemon had 
a son Philoctemon, who married, continued to live on the joint 
estate, having no separate property of his own. Eventually 
Philoctemon in default of sons adopted an heir, Chaerestratus ; 
and subsequently died, Euctemon still being alive. During 
the life of Euctemon, Chaerestratus lived on the joint estate, 
and only at the death of Euctemon was entitled to claim it’, 

Joint Undivided Families so far from being uncommon at 
Athens were so frequent that creditors had to ascertain at the 


1 These instances seem to me to dis- 
prove the assertion made in Meier and 
Schémann, der Attische Process, Ὁ. 5384 
ed, Lipsius, and accepted by Leist, p. 
96, (made also by Boeckh Kl. Schr., tv. 
145, which I have not been able to con- 
sult) that an Athenian father ceased to 
have potestas over his son when the son 
attained the age of 18. The passage 
from the Magna Moralia τ. 33, 16, ὅταν 
ἤδη λάβῃ τὴν τοῦ ἀνδρὸς τάξιν καὶ xwpic- 
θῇ ax’ αὐτοῦ, τότ᾽ ἤδη ἐν ἰσότητι καὶ 
ὁμοιότητί ἐστι τῷ πατρί, might support 


Meier and Schémann’s view if it did 
not contain the very important modi- 
fying clause καὶ χωρισθῇ am’ αὐτοῦ. As 
it is, the passage clearly implies that a 
son did not pass out of his father’s 
Hand at the moment and by the mere 
fact that he attained τὴν τοῦ ἀνδρὸς τάξιν. 
It was further necessary that he should 
χωρισθῇ am’ αὐτοῦ. 

Ar., Ν.Ε. vu. vii. 2, is, as Lipsius 
says, inconclusive until it can be shown 
to be referring to ‘feinem Attischen 
Beispiele.” 


104 THE JOURNAL OF PHILOLOGY. 


decease of a debtor whether his sons had divided the estate or 
were living on it jointly (Dem. c. Huerg. et Mnes. ὃ 34, πότερα 
μεμερισμένος εἴη πρὸς τὸν ἀδελφὸν ἢ κοινὴ ἡ οὐσία εἴη αὐτοῖς); 
and joint estates in certain cases formed a special class of 
exemptions from liturgies (Harpoc. Κοινωνικοὺς av λέγοι τάχα 
μὲν τοὺς avéuntov οὐσίαν ἔχοντας ἀδελφούς, ὧν ὁ μὲν πατὴρ 
ἐδύνατο λειτουργεῖν, οἱ δὲ κληρονόμοι τῶν ἐκείνου καθ᾽ ἕνα 
τριηραρχεῖν οὐκ ἐξήρκουν) and, in cases where a single heir 
refused to accede to a division of the joint estate it was only 
by resorting to the law, εἰς δατητῶν αἵρεσιν (cf. Hesych. s. v. 
δατεῖσθαι and Gortyna Code v. 40), that his brothers or co-heirs 
could obtain a division of the property. 

In Sparta, as at Athens, both during the life and after the 
decease of their father, brothers lived on the joint estate. If 
one brother alone married, his son was accounted the son of 
them all, and was their joint heir, just as at Athens, as we saw 
from Dem. c. Leochar. 17, the daughter of one brother was con- 
sidered the daughter of the joint family, and was heiress to the 
joint estate. That this was a primitive Aryan Custom we may, 
I think, safely infer, for we find it converted into Hindu law. 
Vasishtha, Xvit. 10, says, “If amongst many brothers who are 
begotten by one father, one have a son, they all have offspring 
through that son” (cf. Vishnu, xv. 42). The Joint Undivided 
Family persisted in Sparta long after it had disappeared in 
other parts of Greece, and thus Polybius, misunderstanding the 
practice, was led to imagine, where brothers lived on the joint 
estate, and one alone had a wife, that the wife was common. to 


all the brothers*. On the authority of this passage in Polybius’ 


modern writers on primitive marriage have rashly concluded 
that polyandry was practised among the Spartans. Doubtless 
Polybius was confirmed in his error by the fact that amongst 
the Spartans, as amongst many other Aryan peoples, a husband 
in default of sons called in his brother or other near kinsman to 
raise up seed unto him. This practice however has its founda- 


1 Precisely the same mistake, due to ascribes polyandry to the ancient 
the same cause (a misinterpretation of Britons, B. G. v. 14, ‘maxime fratres 
the institution of the joint undivided cum fratribus.’ 
family), is made by Caesar when he 


. 
ee ee 


KIN AND CUSTOM. 105 


tion not in polyandry, which was unknown to the Aryans, but 
in the paramount necessity according to Aryan ideas of pro- 
viding sons to offer the usual sacra to the House Spirit. 

Before dismissing the subject of the Joint Undivided Family, 
I must point out that it is erroneous to assume as Leist does, 
p. 65, that the family must—to be a joint family—have dwelt 
in a single house*. In the Gortyna Code it is expressly. provided 
that on the death of the father “the houses in the town” go to 
the sons. Therefore the joint estate which was undivided be- 
fore the decease of the father and might, at the option of his 
sons, remain undivided, included several houses. In passing 
I may note that this expression “the houses in the town” at 
once confirms and is explained by the conclusions already 
established from the practice of the Norse and Germanic 
nations. According to this practice, arable and pasture land 
was the common property of the Maeg or Near Kin. In the 
town were situated the houses of the kinsmen. To each house 
was attached certain rights over the arable land and of common 
appurtenant and common pur vicinage. At Gortyna land was 
held in common, for succession to every kind of property except 
land is provided for by the Code; the houses were situated in 
the town and like the cattle and implements used in cultivation 
were the property of the respective Households. 

The corporate character of the Household comes out strongly 
in the Gortyna Code. A large number of its provisions are 
directed to the defence of the corporation against both its own 
members and outsiders, to the definition of the rights of its 
members, and to the distinction of what belongs to the corpora- 
tion from what was the acquest of any of its members. By 
Hindu law the acquests of a brother do not become part of the 
property of the corporate family, Gaut. xxvii. 30, Manu Ix. 206, 
Yagnavalkya 11.138. So too the Gortyna Code allows a son 
power over his own acquests, VI. 5, and protects them from the 
corporation, represented by the father, vi. 9 and 32. On the 


1 Doubtless in many cases the family the joint life of the family gave special 
had a single house. If Hesiod singles facility for this form of adultery, O. et 
out adultery with a brother’s wife for 17). 329. 
special denunciation, it was because 


106 THE JOURNAL OF PHILOLOGY. 


other hand the Code protects the corporation’s property : a son 
may not pledge or alienate any thing belonging to the corporate 
family, γι. 4; the amount of a dowry’, Iv. 50, and of κόμιστρα, 
Li. 37—40, is fixed by the Code; more may not be given, and 
what is given must be given in the presence of three witnesses, 
vi. 1. Further the Code protects the corporation against 
strangers: if a stranger marries an heiress, the inheritance re- 
verts to the next of kin, vil. 52; if an adopted son dies without 
children, the property again reverts to the kinsmen, XI. 6. 

By the Gortyna Code the wife’s acquests also are assured to 
her: she has the option of dividing or not dividing them be- 
tween her children, Iv. 24; if she dies, they go in certain cases 
to her daughters, Iv. 45, as is the case in Hindu law, Gautama 
XXVIII. 24; if she dies childless, her acquests revert to her 
own kin, 111. 31; in case of divorce, 11. 1—-15, or of the death 
of her husband τη. 19 the Code guarantees her acquests, but at 
the same time prohibits her, under penalties, from taking any 
of the corporate property, and inflicts a double penalty on a 
stranger who assists her to do so. 

In Attic law the corporate character of the family reveals 
itself in the fact that an inheritance was divided per stirpes, 
not per capita: nephews for instance took the share which 
would have fallen to their father, had he been alive; in other 
words the corporation, not the individual members thereof, in- 
herited, and whether the father was alive or not, whether he 
left several sons or one only, the amount which the corporation 
could claim was the same. Again, at Athens as at Gortyna the 


corporate property was protected from strangers who thought to 


become possessed of it by wrongfully marrying an heiress. In 
both states the properties in such cases reverted to the kin 
(Dem. pro Phorm. 954).. Further the law at Athens as at 
Gortyna protected the property of the corporate family against 
both its members and strangers: a son of Pericles borrowed 
money from a stranger, and the creditor, so far from being able 


- 1 At Athens too the father was pro- from Plut. Sol. 20 μὴ εἶναι φερνὰς, ἀλλ᾽ 


hibited, in the interests of the corpora- ἱμάτια τρία καὶ σκεύη μικροῦ νομίσματος. 


tion, from giving a dowry larger than ἄξια, ἕτερον δὲ μηδὲν ἐπιφέρεσθαι τὴν 
custom or law allowed, as we may infer γαμουμένην. 


an ttt Lio iii 


KIN AND CUSTOM. 107 


to recover his money from the property of the corporation, 
rendered himself liable to an action for having lent the money, 
_Plut. Per. 36. It is to be noted that the son of Pericles was 
not a minor, he was married ; and though he had attained his 
majority and had taken a wife, he was still a member of the 
Joint Undivided Family, had no property of his own, and was 
in the Hand of his father. If this tale of Plutarch’s were the 
only instance of the existence in Greece of the Joint Undivided 
Family, if we only had Plutarch as an authority to prove that 
ἣν married son might be in the Hand of his father, and that the 
law protected the corporate family, we might hesitate. But 
taken in connection with the evidence already quoted, the story 
Plutarch tells is a cumulative proof which we may reasonably 
accept. 

The existence of the patria potestas in Greece is denied 
by various German scholars. The apparent instances of the 
power they would explain as being not the exercise of any right 
on the part of the father but as the discharge of a duty by the 
son. This seems to me to be an inversion of the facts, and an 
inversion due to the neglect of Comparative Law in Germany. 
If the power of the Hand were peculiar to the Romans, we 
might with Leist regard the patria potestas as a development 
peculiar to Roman Jaw; but as the power of the Hand was 
acknowledged by all Aryan peoples; we must regard the differ- 
ence between the extent of the patria potestas in Rome and in 
Greece as due to a weakening of the original Hand in the latter 
country, and not to a strengthening of it in Italy. Further, the. 
obsequium theory of the Greek father’s power does not explain 
all the facts, for the Greek father exercised some rights which 
cannot, from the nature of the case, spring from any duty on 
the part of the son. Finally the Hand of the father, i.e. of the 
representative of the corporation, whether actually the pro- 
creator of the Joint Family or one of several brothers, shows in 
Greece the distinctive characteristic of the patria potestas: the 
son has no rights as against the Father, κατὰ πατρὸς μὴ εἶναι 
δίκην, ) THY τῆς παρανοίας, Sop. Div. 382. 

With the Greeks, as with all other Aryans, the father had 
the right to decide whether a new-born child should or should 


108 THE JOURNAL OF PHILOLOGY. 


not be allowed to live’. This is an example of a right on the 
part of the father which cannot have sprung from any duty on 
the part of the son; and the obsequium theory breaks down at 
the threshold. If the father decided to accept the child it 
came into his Hand’, and there are traces in Attic law that 
originally the father continued to exercise the cus vitae necisque 
as long as the child remained in his Hand (cf. Plut. Sol. 23, and 
especially 13 πολλοὶ δὲ καὶ παῖδας ἰδίους ἠναγκάζοντο πωλεῖν᾽ 
οὐδεὶς γὰρ νόμος ἐκώλυε). But from the earliest times it seems 
to have been a Custom that in grave matters the House-father 
should act in the presence of the Family Council. The new- 
born child was accepted or rejected by the father in the 
presence of this Council. This was the case at Gortyna, I. 50: 
and probably originally at Athens also, though eventually the 
Phratry took the place of the Family Council, as at Rome did 
the Clan Council’ (Dion. Hal. 11.15). But whereas at Rome 


1 According to Aelian V.H. τι. 7, at 
Thebes fathers might in case of ex- 
treme poverty only sell their new-born 
sons or daughters. 

* The expressions ‘‘in the Hand,” 
**to come into the Hand,” may be due 
to the father’s taking the child into his 
hand, and by that act at once acknow- 
ledging his paternity and asserting his 
power. Or possibly—unless the sug- 
gestion be thought fanciful—the ex- 
pression may be derived from the pri- 
mitive practice of counting degrees of 
relationship by the fingers. The fingers 
of both hands just represent all the re- 
lations who could according to Aryan 
notions be ‘‘in the Hand” of the House- 
father. Thus, if the middle finger 
represents a father, the first and third 
fingers are his sons; the little finger 
and thumb their sons. And taking the 
thumb to represent myself, the fingers 
in order represent my father, grand- 
father, uncle and cousin. Taking both 
hands together, if the middle fingers 
represent brothers (sons of a father not 
represented—because not being—in the 


Hand) and the other fingers their des- 
cendants as before, we can count on 
the two hands together all possible 


degrees of relationship which could 


exist between the Near Kin; and can 
tell at once whether any given relative 
**comes into the Hand” of the House- 
father. 

The Greeks limited the ἀγχιστεῖς to 
those who could be counted on a single 
hand. The Romans included in the 
Agnates not only father, grandfather, 
uncle, cousin: they went on to the 
other hand and counting from the 
middle finger, included great-uncle, 
first cousins once removed, and second 
cousins, 

3 At Gortyna also the Clan Council 
in several matters took the place of the 
Family Council, e.g. in witnessing the 
presents given by a husband to his 
wife, 11. 20, or the amount of a dowry, 
vi. 1. But whereas at Rome, as in 
India and elsewhere, the number of the 
Clan Council was five (see Hearn Aryan 
Household, 129) at Gortyna it was 
three. For other traces of the Court 


KIN AND CUSTOM. 109 


the Clan Council seems to have constituted but a slight check 
on the power of the father, at Athens the Phratry came to 
exercise such control over him that if he admitted that the 
child was his child he was compelled to adopt it. 

Over his children the Greek father had complete power. 
He gave his daughters in marriage to whom he would; and 
provided, as does Menelaus in the Odyssey, his sons with wives. 
He could disown his son (ἀποκηρύττειν) for good cause; and he 
could give his son into the Hand of some other person (ἐκποιεῖν) 
who wished to adopt a child. The power of the father over his 
son did not cease, as we have seen, either when the son married 
or came of age. Whereas the father could compel the son to 
support him, the son had no legal rights against the father; 
and generally no member of the Household could set the law in 
action against the head of the Household’. 

Writers on Attic law have been much exercised on the 
question what place in the order of hereditary succession the 
father of the deceased took ; for he is not mentioned among the 
heirs in the law of Solon. Nor is he mentioned in the Gortyna 
Code. And the reason is plain: he was not one of the heirs. 
_As long as the father lived, the son was in his Hand, lived on 
the Joint Estate and had no property in it, to bequeath or 
to leave. If further proof is necessary it may be found in 
the law itself. The law provides that if the deceased have no 
brothers but nephews, the nephews “take their father’s share.” 
This implies that the nephews could not inherit in their father’s 
lifetime. Generalising this case we arrive at the conclusion we - 
have already reached by another road, viz. that a son could only 
succeed to the Estate at his father’s death—which is why the 
father could not inherit the Estate from his son. 

In conclusion the close resemblance between the Garis 
Code and the law of Solon in the matter of inheritance shows 


“οὗ Five in Roman customs, see Dion. 
Hal. τι. 75 and Gaius1. 20. The Fa- 
mily Council at Athens seems to have 
consisted on certain occasions of the 
whole γένος, οἵ, Ath. v1. 245 who says, 
at a wedding τοὺς κεκλημένους μὴ εἶναι 


ἐπὶ τοῖς τριάκοντα. ‘The thirty” evi- 


dently refers to the γένος which con- 
sisted theoretically of thirty γεννῆται. 
1 Thus in Aesch. c. Timarch, 102 
Arignotus seems to have had no legal 
redress against Timarchus, who when 
he became head of the household de- 
clined to support his blind uncle. 


IIo | THE JOURNAL OF PHILOLOGY. 


us the meaning of the statement of Plutarch (Lyc. 4) that Solon 
borrowed some of his laws from Crete, or of Ephorus (ap. Strabo 
γι. 260) that Zaleucus borrowed some of his laws from Sparta, 
some from Crete and some from Athens. Some resemblance 
between these laws there must have been. But that Zaleucus 
borrowed his laws is only the conjecture of Ephorus. The real 
reason of the resemblance is that which accounts for the resem- 
blance between the Gortyna Code and the Code of Solon: the 
laws were but the common Customs of the Greecks—indeed of 
the Aryans—written out and invested to some extent with the 
sanction of the state. Hence the resemblance of the laws of 
Androdamas for the Thracian Chalcidians (Ar. Pol. τι. viii. 9), 
and of the laws of Charondas (Diod. x11. 18) to the code of 
Solon’. While in confirmation of the inference that all these 
codes consisted mainly of Customs, we have the fact that the 
laws of Epimenides and of Charondas related to the Kin 
(Ar. Pol. τ. ii. 6). 


1 And of the Twelve Tables to Greek codes. The theory of imitation is un- 
‘necessary in this case also. 


F. Β. JEVONS. 


PAUSAN. VIII 16 § 5. 


ς / VAgs / ‘ > / / > \ 
Εβραίοις δὲ “EXé€vns Ἔπάνῆχος ἐπιχωρίας τάφος ἐστὶν 
/ 4 “Ὁ > 5 ’ δ 
ἐν πόλει Σολύμοις, ἣν ἐς ἔδαφος κατέβαλεν ὁ Ῥω- 
7 , , Ν b] a / \ 
μαίων βασιλεύς. μεμηχάνηται δὲ ἐν TO τάφῳ τὴν 
, ¢ / \ 3 a / ’ ‘ 
θύραν, ὁμοίως παντὶ οὖσαν τῷ τάφῳ λιθίνην, μὴ 
’ , \ a e / a \ 
πρότερον διανοίγεσθαι πρὶν ἂν ἡμέραν τε ἀεὶ Kai 
Ἂν > / ᾽ , / ¢ U 
ὥραν TO ἔτος ἐπαγάγῃ τὴν αὐτήν" τότε δὲ ὑπὸ μόνου 
τοῦ μηχανήματος ἀνοιχθεῖσα καὶ οὐ πολὺ ἐπισχοῦσα 
/ ὃ 3 5 / lal \ ‘ “ \ 
συνεκλείσθη δι᾿ ὀλίγης. τοῦτον μὲν δὴ οὕτω. τὸν 
δὲ Vv / > a  ς τι / Ἁ > 
ἑ ἄλλον χρόνον ἀνοῖξαι πειρώμενος ἀνοίξαις μὲν οὐκ 
Μ δὲ ’ | U , 
av, κατάξεις δὲ αὐτὴν πρότερον βιαζόμενος. 


H. van Herwerden in an excellent paper of emendations 
on Pausanias (Mnemosyne, n. s. XV, 1887, 67) says truly ‘ Mire 
dictum δε ὀλίγης pro δι’ ὀλίγου et pleonastice post verba ov 
πολὺ ἐπισχοῦσα. He leaves to others the solution of the 
riddle : ‘ Requiri videtur quod significet πάλιν, sed quod lateat 
non reperio. Context and the ductus litterarwm suggest δι᾽ 
“EAYTns for δι’ ὌΛΙΓης". 

I am tempted to cite three-passages in which Herwerden 
convicts the latest editors of neglecting indisputable corrections 
of Porson’s. If Englishmen, as custodians of the honour of the 
English school, would expose every instance of such carelessness 
as it occurs, our current texts would be far purer than they are. 


v 14 § 6 (Herwerden p. 62): 
γράφομεν δὲ καὶ ἡμεῖς ἐν τοῖς λόγοις τοῖς Λετρι- 
ναίοις. Promisso stetit vI 22 § 5, hic vero scripsit 
γράψομεν, ut correxit iam Porsonus, sed nemo au- 
divit. Alia vero haud pauca in Pausania emendavit 
criticorum sagacitas et diligentia, neglecta et spreta 
ab editoribus. 


1 [Prof. vy. Herwerden in a post-card accepts this emendation. ‘Rem acu 
tetigisse mihi videris.’] 


112 


THE JOURNAL OF PHILOLOGY. 


vit 11 § 2 (p. 67): 


Pessime spernitur Porsoni correctio delentis verba 
Tov κριὸν Tov ἑψόμενον, vel idcirco non ferenda, quia 
τὸν ἑψόμενον dici nequit pro τὸν ἑψηθέντα.  Prae- 
terea nihil falsius est quam, quod perhibent, κριόν 
iam per se significare vetulum arietem, et. verissime 
Kuhnius correxit γέροντος. παλαιὸν δὲ κατασφάξασα. 


χ 82 § 9 (ρ. 72): 


Sine controversia 6 Porsoni correctione edendum 
\ er > , ’ \ x «ς , \ 
fuerat καὶ ῥάων εὐζώνῳ ἀνδρὶ «ἢ; ἡμιόνοις τε Kal 
“ 39 ἃ \ v > \ ww \ / ἃ 
ἵπποις ἐπὶ τὸ ἄντρον ἐστὶν ἄνοδος τὸ Κωρύκιον, ubi 
aeque manifestum est vitium quam vitii origo. Cur 
tandem miseros lectores vexant editores sequenda in 
talibus pessimorum codicum auctoritate scilicet, quae 
plane nulla est ? ᾿ 


JOHN E. B. MAYOR, 


SEN. ep. 19 § 3. 


Madvig advers. 11 468: 


ep. 19 § 3 (ubi Seneca Lucilio dicit iam non liberum 
esse, postquam in tantam hominum notitiam venerit, 
penitus se in otium abscondere): Ut in extrema 
mergaris ac penitus recondaris, tamen priora 
monstrabunt. Non hoc agitur, in quid se mergat, 
sed quando. Scrib. Ut in eatremo, h. 6. in extrema 
vitae et studii parte. 


No note of time is required, nor is in extremo used for ‘at 
last’, but rather ad extremum or extremum or extremo. But 
mergaris imperatively requires an addition, corresponding to 
the penitus with recondaris. For the constr. cf. Flor. m 30 
(=Iv 12) ὃ 38 aquilam mersam in latebras balte. ‘Though 
you should bury yourself in earth’s remotest corner, in the 
deepest concealment, your past will discover you’. 


JOHN E. B. MAYOR. 


OVID MET, IV 139—141. 


MADvIG adv. I 25 inserts in the first chapter of his first 
book, which is supposed to contain only certain emendations, 
the following conjecture : 


de Thisbe Pyramum lugente haec leguntur: 


et laniata comas amplexaque corpus amatum 
uulnera suppleuit lacrimis fletumque cruore 
miscuit. 


Quid sit uulnera supplere lacrimis, nemo dixit di- 
cetue, ut intellegi possit; nam quod Gierigius idem 
dicit esse atque fletum cruore miscere, ea non 
enarratio est’. Ovidius scripserat subleuit (lauit), 
hoe est, quod ΧΠῚ 490 dicitur, lacrimas in uulnera 
fudit. 


A glance at Forcellini s. ἃ. sublino will shew that this is a 
most unfortunate conjecture. The word cannot be used seriously 
in the sense ‘to bathe with tears’. Ido not think however that 
Haupt is right in rendering supplewit ‘filled’. It is rather 
‘made good’, ‘eked out’; when the streaming wounds began to 
fail, her weeping supplied the lack of blood; there flowed out 
a mixture of tears and gore, 


JOHN E. B. MAYOR. 


1 (Madvig’s note) Statius, tumidus repleta habet; sed longe aliud esset 
poéta et omnia ridicule exaggerans, suppleta, ut lacrimae pro uulneribus 
Theb. πὶ 180 uulneraaltalacrimis  essent. 


Journal of Philology. vou. xvi. 8 


A LOST EDITION OF SOPHOCLES’ PHILOCTETES. 


THE British Museum contains a copy of Turnebus’ edition of 
Sophocles (Paris, 1553) which belonged to Lambinus. At the 
foot of the title-page is written ‘Dionysii Lambini’ and below it 
Διονυσίου AapBivov. <A note on the fly-leaf mentions that the 
volume contains MS. notes by Lambinus. These notes, which 
are copious, are on a single play, the Philoctetes, of which 
Lambinus would appear to have meditated an edition. He gives 
(1) conjectures of his own, (2) a number of conjectures by 
Auratus, to whom we owe the brilliant emendation of 556 audi 
σοῦ νέα, (3) occasional conjectures by Turnebus not known to us 
from his edition and (4) Latin translations of many passages 
(especially difficult ones), often giving the explanations of 
Auratus and Turnebus side by side with his own. In many 
cases the notes are mutilated, the margin having been cut off 
by the bookbinder. 

Whence did Lambinus obtain these readings of Auratus and 
Turnebus? Certainly not, so far as we can ascertain, from any 
printed work of theirs. In his edition of Lucretius Lambinus 
tells us that, when he was in difficulty as to any passage, he 
used to consult other scholars and, in particular, his intimate 
friends, Auratus and Turnebus. “ Viros doctos, mihi notos ac 
familiareis....Galliae nostrae atque adeo totius Europae princi- 
pes, collegas meos, Adrianum Turnebum et Joannem Auratum 
retuli: nonnullas meas conjecturas cum 615 communicayvi: eorum 
sententias et quasi responsa, quamvis interdum non satis mihi 
probata, summa fide ipsis auctoribus reddidi*’ The three 


1 Epistola ad Lectorem, p. xxv. (Ed. 1583). 


ie a 


A LOST’ EDITION OF SOPHOCLES’ PHILOCTETES. 115 


great scholars appear to have habitually worked together’ 
and must have owed much to this intellectual fellowship. 
Genuine scholars do not dread but rather seek the full daylight 
of the most open criticism. Moreover, men like Lambinus and 
Auratus do not live in fear lest their emendation, if communi- 
cated, should be stolen by another. It is clear from the notes 
we have transcribed that Lambinus must have had before him 
MS. remarks by both his friends on this play of Sophocles. 
Auratus, many of whose readings are quoted in these notes, 
impressed his contemporaries, though he published very little 
work of the kind, as a man of brilliant genius in the way of 
scholarship’. After a somewhat adventurous life, three years 
of it spent in the ranks as a private soldier, he was elected 
Professor in the College of Paris, in which both his friends 
occupied chairs. The notes which we print below compel us to 
assign to Lambinus and Auratus, but chiefly to Auratus, ἃ 
number of notable emendations which have been accredited to 
later scholars. _ 

We quote the text of Turnebus with which, unless otherwise 
specified, Aldus agrees. 

Metrical argument 1. 1, ἐν χρυσῇ] Pachuics writes in mg. 

“ yvovons, Auratus’.” Edd.“ 

Phil. 1. 22. ἔχει] mg. “ ἐκεῖ fortasse” Lambinus. Assigned 
to Canter. So Hermann, Schneid. and (in note) Dindorf. 

29. τύπος] mg. κτύπος Lambinus, Edd. 

L reads κτύπος. 

38. καὶ ταῦτά γ᾽ ἄλλα θάλπεται | ῥάκη mg. ἀλέᾳ Aurat. 

42. προσβαίη] mg. “προβαίη. D. L.” 

106. After θρασύ Lambinus adds mark of interrogation. 
So Edd. 


1 See for example Lambinus’ note 
on Luer. vr. 1135, ‘‘ Cum haec scrip- 
sissem in prima editione, non longo 
post tempore venit in mentem Adriano 
Turnebo, et Ioan. Aurato, mihique cum 
una essemus, neque corruptum, neque 
coruptum legi posse...sed cruentum.” 

2 Lambinus, in dedicating to Auratus 
the 6th Book of his Lucretius, expresses 


the most enthusiastic admiration of 
his friend’s genius, 

3 Auratus’ name is here written in 
full: also Aurat., Aur. and A, 

4 Edd, appended indicates that the 
reading appears in the text of two © 
representative editions, that of Pro- 
fessor Campbell (Vol. τι. 1881) and 
that of Dindorf (Oxon. 1860). 


8—2 


116 THE JOURNAL OF PHILOLOGY. 


126. δοκῆτ᾽ ἔτι text and almost all MSS.] δοκῆτέ τι A. 
(i.e. Auratus), Edd. Assigned to Canter. 

129. ἀν] ἐν Lambinus. 

148. πρὸς ἐμὴν del χεῖρα προχωρῶν] mg. “Aur. προσορῶν. 
sunt enim versus anapaestici.” 

166. πτηνοῖς iots] mg. “Aur. rrnvoictp” (sic. Last letter 
is indistinct). 

189. ὑπόκειται]ὔ ὑπακούει Auratus. See Lambinus’s note ἡ. 
This conjecture is assigned to Musgrave. 

266. τῆσδ᾽] τῆς Auratus, Edd. 


δ᾽ ἀνδροφθόρου ἐχίδνης) Aur. δενδροφθόρου [δ᾽ and avdp 


underlined in text. Too lively an imagination must here have 
misled Auratus !] 

324. θυμῷ γένοιτο χεῖρα] θυμὸν γένοιτο χειρὶ Lambinus. 
So all Edd. since Brunck. 

423. é&npuxe] Aur. ἐκκήρυξε. Lambinus alliter] ἐξείρηκε. 
[L. has yp. κἀξεκήρυξε a man, antiqua. | 

509. τύχοι] τύχη Aur. 


σε 

554. ἀμφί σ᾽ οὕνεκα Turn. and MSS.] mg. “ ἀμφί σου νέα 
Aur.” (cod being struck out and ce written above) “ita postea 
legere censuit (1.6. Auratus). ἀμφὶ σοῦ νέα Lamb.” | 

(Only one MS. gives a trace of the correct reading.. I has 
yp. audio ὃν ἀντὶ Tod περὶ cov. Of course ἀμφὶς ὄν is a mis- 
reading for ἀμφὶ σοῦ". The writer of this gloss thought ἀμφὶ 
with Gen. unusual and explained it by περὶ σοῦ. 

Is it the rareness of ἀμφὶ with Gen. which led Auratus to 
spoil his own conjecture? Aldus reads ἀμφὶ σοῦ “vexa.) 

571. ἔσω] Aur. ἐγὼ, Edd. Assigned to Canter. 

635—6. ὡς... ὁρίζει] ὁρίζῃ Lamb., Edd. Assigned to Brunck. 
“ ὡς nam, sed, Auratus: ut, Lambinus.” 

639. ἐπειδὰν πνεῦμα TOUK πρῴρας ayn] ἀνῇ, remissior fuerit, 
Lambinus. Edd. Assigned to Pierson. 

648. éve] Aur. ἔπε, vel ἔνε ἐπὶ τῆς veds. (Lambinus adopts 
the latter view. ἐπὶ is reading of London ed., Wakefield and 
Musgrave. Dindorf and Prof. Campbell rightly retain ἔνι.) 

1 Page 119 of this article. Bas ἄνω for βασάνῳ (O. T. 493) and εἰ 


2 A mistake characteristic of the πόντος for εἰπόντος (Phil, 1141). 
scribe of I who is capable of writing 


A LOST EDITION OF SOPHOCLES PHILOCTETES. 117 


690. κλύζων] κλύων J. A. (ie. Joannes Auratus), Edd. 
Assigned to Canter’. 

694. βαρυβρώῶτ᾽] Tur. βαρυβοόντ᾽. 

698. εἴ tis ἐμπέσοι, | φορβάδος ἔκ ye yas ἑλεῖν] mg. Tur. 
εἴ τις et ἑλών. εἴ Tis S(ubaudi) aiuas. 

Aur, εἴ τιν΄... «φύλλα... ὥστε ἑλεῖν. 

(ἑλὼν Turn. So Reiske and Schneidewin. εἴ τιν᾽ (i.e. φύλλα) 
ἐμπέσοι Auratus. So Reiske, Musgrave and Brunck.) 

774. θάρσει. προνοίας Turn.| Aur. θάρσει προνοίας οὕνεκ᾽, 
οὐ δοθήσεται Edd. 

780. καὶ εὐσταλὴς] « Edd. 

782. μ᾽ ἀτελὴς εὐχή] μ᾽ ἀτελῶς εὔχ [one letter cut off] 
Lamb. μὴ μάτην Turn. 

(Both Lamb. and Turn.’s readings imply εὔχῃ. The reading 
μὴ μάτην εὔχῃ is assigned to Camerarius.) 

791. κεφαλὴν] mg. κεφαλλὴν, Edd. So Aldus. 

794. ᾿Αγάμεμνον] Aur. 2 ᾽γάμεμνον. 

830. avtéyois| Lambinus erases y and writes in mg. “Tur. 
Ar.” 

831. αἴγλαν] Aur. ἀχλὺν. So Reisk. Wakef. 

837. καιρός τοι πάντων γνώμαν ἴσχων] Lambinus inserts 
a comma after πάντων and adds “ coniunge καιρὸς πάντων." 

(Musgrave conjectured καιρός Tor πάντων" γνῶμα δ᾽ ἔχων.) 

934. προσφωνεῖν Turn. Aldus} Lambinus deletes ν. So 
the MSS. L. and A. Edd. 

952. σχῆμα] Aur. habitatio, non probo. Idem emend(at) 
ὦ σχῖσμα, non probo. 

992. τιθείς] τέθης Aur., Edd. Assigned to Porson. 

1032. πῶς θεοῖς εὔξεσθ᾽, ἐμοῦ | πλεύσαντος, aiPew ἱερά ;] 
Aur. ἐξέστ᾽. (sic.) Does this mean ἐξέσται 2 Pierson, whom 
Musgrave and Schneidewin follow, reads ἔξεστ᾽. 

1043. ὡς ζῶ] ὃς Aur. Assigned to Reiske. 

1067. οὐδὲ cod φωνῆς ἔτι | γενήσομαι προσφθεγκτὸς] 
Auratus apparently conjectured tov, as we may infer from the 
following in mg. “Lamb. non compellabor abs te: Aur. aliter. 
male. nem....(neminis) vox compell(abit) te.” 


1“ κλύων I, Vat. V4.” Campbell. 


118 THE JOURNAL OF PHILOLOGY. 


1099. εὖτέ ye παρὸν φρονῆσαι |...εἵἴλου. εὖτέ] εὖ τε 
Auratus. Lambinus adds “εὖτε pro ὅτε: Aur. εὖ φρονῆσαι," 
ie. he connects εὖ with φρονῆσαι. 

1128. φίλων Turnebus, Aldus] φίλον Lambinus, Edd. and 
the MSS. 

1140. δίκαιον] “ δικαίου fortasse” Lambinus, who adds 
“ Auratus putat d/casov superare.” 

1149. μ᾽ οὐκέτ᾽] Aur. μηκέτι. 

1829. ἐντυχεῖν] ἂν τυχών Lambinus. Assigned to Elmsley. 
ἂν τυχεῖν Porson, Dindorf. 
ο΄ 1380, 6s] ἕως Lambinus. So Scaliger, Valckenaer, Her- 
mann. 

1332—3. πρὶν av ta Τροίας medi’ ἑκὼν αὐτὸς μόλῃς 

καὶ τῶν παρ᾽ ἡμῖν ἐντυχὼν ᾿Ασκληπιδῶν] 

ἂν τυχὼν Lambinus’, 


1360—1. ols yap ἡ γνώμη κακῶν 
μήτηρ γένηται, τἄλλα παιδεύει κακά' 
mg. “Aur. κακή. Both κακῶν and κακά are underlined. 


1365. ot] L. x’ (standing for καὶ οἵ), Aur. of τε. 

1448, γνώμῃ ταύτῃ τίθεμαι] “al. γνώμην ταύτῃ. Lambi- 
nus inmg. Assigned to Toup and adopted by Dindorf. 

1461. γλύκιόν Te ποτόν] al. Δύκιόν. Lambinus. So all Edd. 

In the scanty notes at the end of Canter’s Sophocles (ed. 
1593) the following readings are attributed to Auratus, viz. 266. 
“ Assentior J. Aurato τῆς δενδροφθόρου reponenti.” 556. ἀμφὶ 
cov νέα. 652. εἴ μοὶ (after Aldus). Canter also mentions, 
without saying whence they are derived, the following: 126. 
“ Videtur scribendum δοκῆτέ te.” 571. “videtur scribendum 
ἐγὼ. 688. “malim κλύων." 782. “malim, si divinare licet, 
μὴ μάτην εὔχῃ, τέκνον. 1365. οὗτε τὸν. These form almost 
the whole of Canter’s notes on Philoctetes. It is curious that 
all are found in Lambinus’ MS. and all, except 782, are there 
assigned to Auratus. 

Of the explanatory notes the following appear to be the 
most of interest. 

2, ἄστειπτος) Lucr. nullius ante trita solo. 13—14. κἀκχέω 


1 In order to avoid construing ἐντυγχάνω with a Dative. 


A LOST EDITION OF SOPHOCLES’ PHILOCTETES. 119 


τὸ πᾶν | σόφισμα. Virgil. omnis Effusus labor. σόφισμα] Aur- 
atus, entreprise: Ego, artificium, artem, consilium callidum et 
ingeniosum, finesse. 39. ῥάκη, βαρείας του νοσηλείας πλέα. 
τοῦ Turnebus in text] mg. του. τινὸς vel ἀνδρὸς vel voc... 
159. ἀμφίθυρον, biforem Aur.: bipatentem L. 165. φύσιν] 
νόμον, morem, quia consuetudo longa vertitur in naturam ; 
itaque per naturam nomen significat longi temporis consuetu- 
dinem. Lamb. 178. οἷς μὴ μέτριος αἰὼν qui sunt prorsus 
inopes rerum ad victum necessariarum, quibus non sunt me- 
diocres facultates ad vitam agendam. 183. μοῦνος ἀπ᾽ ἄλλων] 
separatus ab aliis, vitam agens solivagam et solitariam. 


187—190. a δ᾽ ἀθυρόστομος 
ayo τηλεφανὴς πικρᾶς 
οἰμωγᾶς ὑπόκειται. 

ἀθυρόστομος] garrula Aur. Ego ludibunda et iocosa. 189. 
Aur. vel leg. πικρὰς οἰμωγ.. ὑπόκειται, excipit, vel ἠχὼ πικρᾶς 
οἰμωγᾶς vel leg. πικρὰς οἰμωγὰς ὑπα..., id est respondet. 
[Plainly this last conjecture was ὑπακούει. 

199. ἐξήκοι χρόνος] existat, adveniat vel praetereat. 408. 
ἐς τέλος] unquam. 470. ἱκέτης ἱκνοῦμαι} supplex venio vel sup- 
plex supplico Aur. 509. τύχοι] τύχη Aur. quae nullus meorum 
amicorum perferat. Tur. subaudit λέξας,.. λέξας τύχῃ. 552. 
τῶν ἴσων] cum nactus essem pares comites et tuitores. Al. 
τῶν δικαίων. 635. ὡς] nam, sed Aur.ut L. dpifec] opifm L. 

691. ἵν᾽ αὐτὸς ἦν πρόσουρος] vel expositus ventis vel 5101 
ipse vicinus. (Is there any authority for the first meaning ? 
Wakefield explains πρόσουρος ventis expositus, while Musgrave 
conjectures, πρόσαυρος auris expositus.) 

696. αἱμάδα. T. profluvium sanguinis: Au. tabem cruen- 
tam. 786. ὦ πούς, οἷά μ᾽ ἐργάσει κακα] Lambinus writes in 
mg. ἐργάζῃ but afterwards strikes it out, feeling probably that, 
as Professor Campbell says, “a great evil, perpetually recurring, 
is ‘most in apprehension’.” 789. ἔχετε] T. habetis, tenetis: 
Aur. ἐπέχετε. 799. συλλαβὼν] mg. βοηθῶν Turn.: corrip(iens) 
Aur. 


800. τῷ Λημνίῳ τῷδ᾽ ἀνακαλουμένῳ πυρὶ 
ἔμπρησον'" 


120 THE JOURNAL OF PHILOLOGY. 


ἀνακαλουμένῳ] Tur. ἀνακαλοῦντι. qui m’appelle. (Schol. 
παρακαλοῦντι. Turnebus’ explanation, “the fire calls me,” 
makes Philoctetes in his delirium of pain turn fanciful—He 
imagines the glare of the volcano to be lit for a solemn sign to 
himself.) 

861. παρακείμενος] affinis L. ...ositus (i.e. expositus) Aur. 
non probo. ὁρᾷ nihil videt. 868. οἰκούρημα] Aur. minis- 
terium. lL. eodem loco et in insula, in eodem quod non disces- 
serat Pyrrhus et ejus socii. 884. συμβόλαια] Aur. σύμβολα 
(the word has been struck out by Lambinus) improbum. Tur. 
συνθῆκαι. 904. οὐδὲν ἔξω τοῦ duTevcavTos] nihil alienum ab 
eo qui te procreavit. 928. ὦ πῦρ σὺ] non Pyrrhus. 950. ἔτ᾽ 
ἐν σαυτῷ γενοῦ] ad ingenium tuum redi; esto apud te. 1041. 
ἀλλὰ. ..«ποτὲ] saltem: aliquando tamen. 1049. εἴ μοι παρείκοι) 
si mihi commodum sit. Si mihi occasio detur L. aliter A, 
1082. ὡς] L. ut, quam: Aur. ergo. 1085. συνοίσῃ] comes 
eris. 


1093. εἴθ᾽ αἰθέρος ἄνω 
/ > / \ / 
πτωκάδες ὀξυτόνου διὰ πνεύματος 
ἕλωσί p. 


Ἐπτωκάδες. Αὐτ. ruinae. (We need not suppose that Auratus 
conjectured some word different from mrwxddSes. Ruinae is 
merely a new interpretation of the rare word πτωκάδες, which 
he renders as if from the root of πέπτωκα, and meaning the 
same as πτώματα.) 

1111. μοι.. ὑπέδυ] fefellere me. 1116. τάδε] κατὰ τάδε 
vel appo. πότμος καὶ τάδε. 1133. ὧδε] hic. 1140. ἀνδρός τοι 
τὸ μὲν εὖ δίκαιον εἰπεῖν] Auratus putat δίκαιον superare: 
fort(asse) δικαίου. 1144. εὐφημοσύνᾳ Turn.) consilio. 1149. 
μ᾽ οὐκέτ Aur. μηκέτι. ne amplius.....ad fugam: prope accedite. 
L. ne co....te (conicite) vos amplius in fugam. 1154. φοβητὸς] 
Aur. terrendus: ego, terribilis. 1167] ἀδαὴς] nescit se habere. 
1169] ᾧ ξυνοικεῖ] is cum quo habitat. 

[Is not this passage made plain by putting a comma after 
ἔχειν and taking ἄχθος in apposition to κῆρ, adans being used in 
a passive sense applying to the disease, “beyond our know- 
ledge”? 


OE ESS a ΓΤ ee | 


et ee ele the 
tye * 


A LOST EDITION OF SOPHOCLES PHILOCTETES. 121 


οἰκτρὰ yap (sc. Kip ἔστι) βόσκειν, ἀδαὴς δ᾽ 
ἔχειν, μυρίον ἄχθος ᾧ ξυνοικεῖ. 

“For the disease is a piteous one to support and there is no 
learning to assuage it, an endless trouble to the man on whom 
it has settled.” Does not ἔχειν mean to ‘assuage’ or ‘check’ 
rather than to ‘endure,’ which is too near the meaning of 
βόσκειν; Cf. ὀδύνας ἔχειν 1]. 11. 848.)] 

1194. ἀλύοντα χειμερίῳ | λύπᾳ] eum qui angatur et aflic- 
tetur dolore aestuante. 1202. εὖχος] Aur. volnus. male. 1235. 
κερτομῶν] Lamb. ut mihi cor uras (?) verbis asperis : Aur. quasi 
diceret παίζων. non placet. 1259. ἐσωφρόνησας] male Aur. 
consulto fecisti. 1327. πελασθεὶς} qui accesseris ad. 1364. ἐς 
τροίαν] ad Graecos qui Troiam obsident. οἵ γέ] alii hoc referunt 
ad ea quae sequuntur, hoc modo. et iure te eis socium bellum 
adiunges qui et, &c. 1441. his non paruit Neoptolemus. 
1443. ἡ yap εὐσέβεια συνθνήσκει βροτοῖς] comes est mortalibus 
morientibus et comitatur mortuos. 1446. χρόνιος] longo inter- 
vallo conspectus. 1463. δόξης... ἐπιβάντες) nunquam exopinati. 

Ronsard included Auratus in his ‘ Pleiade’ of French poets. 
Is it not, in a limited form, ‘imagination’ which enables Auratus 
to make conjectures like ἀμφὶ cod νέα 556, προσορῶν 148 and 
σχῖσμα 952, even though calm judgment may reject some of 
these? The commentaries of scholars like Schneidewin and 
Professor Robinson Ellis show a similar faculty. Thus Schnei- 
dewin’s interpretations of Pindar and Sophocles show, very 
notably, genuine poetic feeling, Schneidewin rarely giving a 
sense beneath his author, though often one which is not borne 
out. To make a trustworthy conjecture, one which does not jar 
with the context and show ‘like a raw touch on a picture ill- 
restored’, it is necessary to enter into the poet’s mind, to feel 
as he felt, to see as he saw. Thus alone can we decide in a 
corrupt passage what ought to be said, and what on the other 
hand a poet like Sophocles never could have written. But this 
gift, certainly akin to the poetic faculty, is seldom united with 
sufficient scholarship, so that good conjectures are rare enough. 
For example Auratus’s well-known conjecture haedulei (for 


1 Sophocles, ed. by Professor Campbell, Vol. 1. p. 107, 2nd Ed, 


122 THE JOURNAL OF PHILOLOGY. 


Haediliae Hor. Od. 1. 17), Lambinus’ unquestionable pectus (for 
tempus, Lucr. 1. 46), Ellis’s silicumque senescere petras’* 
(Luer. v. 312)—all these show a faculty entirely wanting in the 
average critic who, after first postulating the ‘unglaubliche 
Verdorbenheit’ of the Sophoclean MSS’, proceeds, in obedience 
to the precept—‘ Du sollst vor Handschriften nicht niederfallen,’ 
to manufacture new readings after the most approved methods. 
The emendation monitura (for moritura) of the dying Dido, is 
a fair sample of such work. The language of a great poet is 
a living thing. Each sentence of Sophocles forms an organic 
whole and, like a beautiful living body, must be reverently 
handled. Editors like Mr Blaydes generously offer us the 
choice of three, four or a dozen emendations, each of different 
or opposite meaning, on the same passage. Almost as well 
might we offer a sculptor the choice of a dozen differently 
fashioned limbs or features from which to complete the same 
statue. A hard saying indeed, but a noble rule to aim after, is 
the precept of Ritschl—‘ There is not more than one right.’ 


1 Journal of Philology, Vol. m1. p. in red, yp. λόγοις. The latter reading 


267. Might we venture, combining 
Ellis’s conjecture with another by 
Munro, Aeraque (adopted in his 2nd 
Edition), to suggest the following ?— 
Aeraque proporro silicemque senescere 
cernis. 

2 No doubt the MSS. of Sophocles 
are, in some respects, disappointing. 
They tantalise us by seeming to lead up 
to a common point, beyond which they 
refuse to carry us. Yet all the MSS. 
have by no means yet been thoroughly 
sifted. For example, while collating 
the Oxford MS. Laud. 54, I observed 
the following remarkable reading. At 
Ajax 330 the true reading λόγοις seems 
to have been entirely lost out of the 
MSS.,since all which have been hitherto 
collated read 


φίλων γὰρ οἱ τοιοίδε νικῶνται φίλοι. 


Laud. 54 also has φίλοι in the text, 
but above is added by the gloss-hand 


has been restored by the editors on the 
sole authority of Stobaeus. From 
what source did the scribe of Laud. 54 
derive this correction which, if derived 
from another MS., would seem to carry 
us beyond the point from which the 
universal corruptions date? Professor 
Ellis has very conclusively shown 
(American Journal of Philology, Vol. 
1. p. 401) that the value of a ΜΗ, is 
not necessarily in proportion to its 
age. For example, the well-known 
Datanus of Catullus, written after 
1470, ranks ‘‘in the first class of 
Catullian codices.” There exist be- 
yond question several 14th century 
MSS. of Sophocles which do not con- 
tain an ordinary 14th century text, 
but have been copied directly from 
older MSS. and represent a much 
earlier text. From later MSS. of such 
a kind valuable readings may yet be 
gained for the text of Sophocles. 


- γε. he ων. ω .᾿ ἊΨ vs Sed ee aT et 
ee wee leeches 
| 


A LOST EDITION OF SOPHOCLES’ PHILOCTETES. 123 


The notes we have printed are interesting if only as showing 
on what intimate terms the three great scholars worked together. 
Moreover, students of Greek literature owe just enough to 
Auratus to wish that they owed more. It is evident that the 
light which Auratus could flash on a corrupt or difficult place is 
very different from that which an ordinary scholar holds in his 
hand. It may be less steady, but it is far brighter and far 
more suggestive. 


JOHN MASSON. 


LUCRETIANA. 


LUCRETIUS I 356 


quod nisi inania sint, qua possent corpora quaeque 
transire ? haud ulla fieri ratione wideres. 


We should, I think, unquestionably accept Munro’s repunctua- 
tion. But his change of possent to possint is by no means so 
convincing, and indeed is discountenanced by the neighbouring 
wideres. The combination of present in protasis and imperfect 
in apodosis stands precisely on the same footing as in V 276 
‘qui nisi retribuat recreetque, omnia iam resoluta forent’ ; 
which, however, according to him, is the sole instance of such a 
sequence in Lucretius. In this latter statement he leaves out 
of sight another passage where the same construction has been 
emended away. 
In 11 1033 sqq. 


omnia quae nunc si primum mortalibus essent, 
ex improuiso si sint obiecta repente, 
1035 quid magis his rebus poterat mirabile dici 
aut minus ante quod auderent fore credere gentes ? 
nil ut opinor; ita haec species miranda fuisset. 


The MS. reading, it is clear, is corrupt; and either essent or 
sint must go. Munro elects for keeping essent and changes sint 
to nunc. Orelli and Lach. keep sint and make the*much easier 
alteration extent, and I believe they are right. It is well known 
that the earlier Latin writers, and especially the poets, were not 
so strict or so consistent as later ones in the use of secondary 
tenses to express unrealized conditions. In particular a special 
preference seems to have been extended to δὲ sint, ni sint from 


nthe ΣΟ EME ras oat) io te 
ah Rages: % 


selenite te ne, eee erm 28h ee 8 
eS WR iar σα Cyr te ον», ey 


εὐ, Ἂ" 


LUCRETIANA, 125 


their metrical convenience; cf. Kiihner Lat. Gr. 1. § 314. 2. 
See besides the present passage Catull. 6. 2 Flaui, delicias tuas 
Catullo nei sint inlepidae atque inelegantes, uelles dicere nec 
tacere posses, Tib. 1. 8. 22 carmina ni sint, ex umero Pelopis non 


nituisset ebur. 


1 469, 470 


namque aliut Teucris (Munro, codd. terris), aliut 
regionibus ipsis 
euentum dici poterit quodcumque erit actum. 


In the lines 449—482 Lucretius is occupied with showing 
that neither coniuncta ‘inseparable properties,’ nor euenta ‘ acci- 
dents,’ exist by themselves. Having explained the difference 
between the two (449—458) he shows that time (an ewentwm) 
cannot exist per se (459—463). Next he points out that the 
use of the word esse, as in T'yndaris rapta est, bello subactae 
sunt Trovugenae gentes, must not mislead us into thinking that 
they exist; because ‘ea saecla hominum, quorum haec euenta 
fuerunt, irreuocabilis abstulerit iam praeterita aetas’ (464—468). 
That is his first argument: the ewenta cannot exist because the 
subjects have perished. His second argument is that the euenta 
could not have existed at all unless both matter and space had 
existed, which he proves by an example, the ‘events’ of the Trojan 
war (471—477). This shows that all events (res gestae) have 
no separate existence in the sense in which body and space 
have, but are only accidents of them (ut merito possis euenta 
uocare corporis atque loci res in quo quaeque gerantur 482). 
The lines 469, 470, which only break the argument in their 
present position, being quite unconnected with what follows 
and what precedes, wili be useful and appropriate as an illus- 
tration of what is meant by an euentum corporis as distinct 
from an euentwm loci, a distinction which does not emerge till 
481, 482.  Teucris, as Munro points out, stands for corporis and 
regionibus for loci. It seems clear then that, unless 469, 470 
are a marginal addition, they should be placed after 482. 


1 884 consimili ratione herbis quoque saepe decebat 
cum lapidi in lapidem terimus, manare cruorem ; 


126 THE JOURNAL OF PHILOLOGY. 


et latices dulcis guttas similique sapore 
mittere, lanigerae quali sunt ubere lactis. 


Reading for latices in line 886 Bruno’s certain conjecture salices 
(the bitter [amarae] food of cattle and sheep, cf. Virg. Ecl. 3. 
83, 1. 70, Lucr. 2. 361), we come to 887, which is hardly satis- 
factory as it stands. The MS. evidence is nearly balanced 
between quali sunt B and qualis sunt A. Munro, following 
Lachmann, reads quali sunt and translates ‘that waters should 
yield sweet drops in flavour like to the udder of milk in sheep,’ 
an involved construction not justified by the parallel of Prop. 1. 
2.21. I believe that swnt is a corruption for DANT; qualis is of 
course the acc. plur. and lactis the gen. after it. 


11 20 ergo corpoream ad naturam pauca uidemus 
esse opus omnino quae demant cumque dolorem. 
22 delicias quoque uti multas substernere possint 
gratius interdum, neque natura ipsa requirit 
si non aurea sunt iuuenum simulacra per aedes, etc. 


Munro has shown so much acuteness in divining the meaning 
and restoring the proper punctuation that we cannot help 
feeling a little surprised that he has left 22 as it stands and 
takes (in his last edition) neque in 27 in the sense of non. The 
argument may be briefly stated thus. ‘The wants of nature 
are few and limited to the removal of pain (21, 22). Though 
the addition of luxuries may produce greater pleasure, yet (a) 
on the one hand (neque) nature does not miss their absence 
(2428), while simple natural delights can supply their place 
(29—33), nor (8) on the other hand (nec 33) are they of any 
value in curing pain (33—36).’ 

In the reading of 23 there are two difficulties. First there 
is no nominative to possint, and none can be supplied from the 


context. The anticipation of one from 24 and following seems. 


out of the question. Secondly, substernere is only used of 
persons, or things personally regarded ; and certainly cannot be 
taken to refer to the ‘golden statues etc.,’ which are the deliciae 
themselves. Both difficulties disappear with the simple correc- 
tion Possis. “ Although you may pillow a man on a multitude 
of luxuries so as to produce occasionally a greater satisfaction, 


——— ee 


1 60». eee »- νῶν, γ΄. .. 
' 


LUCRETIANA., 127 


yet”—Observe how every word is telling, multas... possis...inter- 
dum, ete. substernere is used in its simple and primary sense of 
laying or placing something under something else, as in Ter. Ad. 
4, 3. 12 ‘ex ara hinc sume uerbenas tibi atque eas substerne.’ 
Here it is pleasure’s bed of roses that is meant; and the grassy 
couch of v. 29 ‘prostrati in gramine molli’ suggests itself as a 
natural contrast. 


II 98 partim interuallis magnis confulta resultant. 


As Lachmann’s erroneous interpretation ‘conferta et con- 
glomerata’ still appears to hold the field, I trust I may be 
allowed to refer to my proofs that confulta, the MS. reading, 
is to be interpreted ‘rebound when pressed together’, in my 
note in the Journal of Philology 1x p. 64 on Prop. 1. 8. 7 and 
in the Appendix to my Select Elegies of Propertius. 


11 180 nequaquam nobis diuinitus esse creatam 
naturam mundi: tanta stat praedita culpa. 


The words tanta stat do not occur in the MSS. They have 
been introduced into the text by Lach. from v 199 where the 
verses occur again. There is an obvious objection to them 
here, that they do not explain the MS. corruption quam- 
quam. This objection is almost entirely removed if we read 
QVANTA (=quod tanta), an idiomatic use which requires no 
illustration. 


1π 647 et simul in pugnae studio quod dedita mens est. 


Lachmann, followed by Munro, changed simul to semel, 
without reason as it appears to me. Lucretius gives two reasons 
for the man’s unconsciousness of his loss of the limb: (1) there 
has not been time for the mind to feel it, vv. 645—646 ; (2) the 
mind is, at the same time, too much absorbed in fighting to 
notice it, v. 647. Bockemiiller retains stmul but disarranges 
the whole passage. 

UI 941 uitaque in offensust. 

The MSS. have in offensost, which of course cannot stand. 
Lachmann and Munro follow Lambinus in changing it to 
in offensust, an unexampled meaning of offensus. This is the 
stranger, as in offensasr the classical phrase lay ready to their 


128 THE JOURNAL OF PHILOLOGY. 


hands, being actually mentioned by Lambinus and illustrated 
by Munro. 


Iv 642 id quibus ut fiat causis cognoscere possis. 


The same three scholars are again agreed in changing the 
order to ut quibus id. It is difficult to resist this consensus. 
But the involved order requires no defence after Munro’s note 
on ΠῚ 843, and the emphatic position of the 2d is as proper as 
that of 7s in 1 177. I may note in passing that the four 
lines immediately preceding, which Munro connects with their 
context by reading extetqgue ut serpens for est itaque ut serpens 
in 638, appear to have been left by the poet himself in an 
unfinished condition. 

Iv 1152 

et quae (sc. uitia) corpori’ sunt eius si guam petis ac uts. 

Munro following Lach. thus reads for the MS. praepetis 
(A ete. precis B). But has it been proved that praepetis 
cannot stand? It gives the right sense, ‘quam prae als petis’ : 
and Festus (s. v. praepes) has it in one of the derivations which 
he quotes for praepetes aues ‘quod ea quae praepetimus, indi- 
cent’. Bockemiiller retains praepetis but absurdly changes ac 
wis to ac Ut. 


V 1117 sqq. 

quod si quis uera uitam ratione gubernet, 

diuitiae grandes homini sunt uluere parce 

aequo animo; neque enim est umquam penuria parui. 


There are some difficulties in these lines which have not 
been properly realized by the editors. The first is in the latter 
part of 1119 which means, as Munro rightly translates, ‘never is 
there any lack of a little’, although the first and obvious con- 
struction is to take est and parui together ‘is worth’ or ‘costs 
little’. There is another and more important objection. What 
Lucretius is insisting on here is the well-known commonplace 
that the poor and contented are as well off as the wealthiest. 
But there is nothing said about poverty in the lines. A man 
who lives parce does so voluntarily. parce wiuere may be 
said just as well of the rich as of the poor; and it seems to me 
that Munro’s translation ‘a frugal subsistence joined to a con- 


LUCRETIANA. 129 


tented mind is for him great riches’ is practically an admis- 
sion that the received text is unsatisfactory. For ‘a frugal 
subsistence’ is something different from ‘living frugally’. What 
again is the necessity of adding aequo animo? We need not 
go far for examples of the regular phrase Hor. Od. 2. 16, 18 
‘uiuitur paruo bene’ etc., ΤΊ. 1. 1. 25 ‘ contentus wiuere paruo’. 
Cic. Paradow. 6. 51 is an excellent commentary on the general 
sense: ‘non esse cupidum pecunia est; non esse emacem 
uectigal est. contentum uero suis rebus esse maaximae sunt 
certissumaeque diwitiae’. If we read PARVO for parce, we get a 
clear idea expressed appropriately, we remove the ambiguity or 
at least the awkwardness of the construction of parwi and 
we restore to neque enim its proper function of adding to 
as well as explaining what has gone before. The wealth in 
contentment is not only great (maximae Cic.) but lasting 
- (certissimae Cic.) ‘He is very rich who can live contentedly 
on a little. Yes, and he can never want that little’. 


γι 1022 huc accedit item (quare queat id magis esse 
haec quoque res adiumento motuque iuuatur) 
quod simul a fronte est anelli rarior aer 
factus inanitusque locus magis uacuatus, 
continuo fit uti qui post est cumque locatus 
aer a tergo quasi prouehat atque propellat. 


The sense of this passage is clear.- The movement of the iron 
is assisted by the impulses given to the air behind it when a void 
is created in front. It is in the precise explanation of the two 
lines 1022, 1023 that the difficulty lies. Munro’s interpretation 
can only be gathered from his translation : “Moreover” [= huc 
accedit item...quod] ‘(to render it more feasible, this matter also 
is helped on by external aid and motion) as soon as etc.’ ‘This 
matter’ is undoubtedly the attraction of the iron to the magnet. 
Then what is the point of saying ‘this matter also’? What 
other matter is there before us? Surely the also must refer to 
some contributory cause, so that we must alter a letter and read 
HOC (for hec). res iwwatur, it may be added, has its own ap- 


1 It will be noticed that here, as have been corrupted, as is so frequently 
also in 1 22, the last letters of a line the case in Lucretius. 


Journal of Philology. vou. xv. 9 


130 THE JOURNAL OF PHILOLOGY. 


propriateness in an indefinite statement, cf. 111 910 ‘ad somnum 
si res redit atque quietem’; and hoe with quoque has its proper 
reference to what is to come. , 


vi 1194 (Lucretius is giving the signs of approaching death) 
frigida pellis 
duraque inoretiacet rectum frons tenta tumebat. 

This is the reading of B and the Vienna fragment, A has 
inhoretiacet rectum, a further corruption. Nonius (p. 181. 27) 
has in ore iacens rictu. Under the MS. rectum and Nonius’ 
rictu lay rictum (neut.) the correction of Lambinus. rictus (or 
rictum) is well illustrated by Munro from Shakespeare, the 2nd 
part of Henry VI, “See how the pangs of death do make him 
grin.” It means the mouth wide open and the teeth disclosed 
as in hearty laughter (Hor. Sat. 1. 10. 7 risu diducere rictum). 
The rest of the sentence must be restored by combining the 
wmoretiacet of the MSS. with the inoreiacens of Nonius. The 
archetype must have had in ore tacet or TACENS.which latter is 
to be preferred on account of the neighbouring imperfects. The 
picture which Lucretius desires to call up is the ghastly one of 
the sight and suggestion of laughter without its sound. The 
dying man seems to laugh, but you hear nothing. Contrast this 
with the beautiful picture in Prop. ΠῚ 29. 13 hic equidem 
Phoebo uisus mihi pulchrior ipso marmoreus tacita carmen 
hiare lyra; the statue is only marble, but it seems to sing 
‘with parted lips and all but speaking lyre.’ We must add that 
the words of Hippocrates (Munro ad loc.) seem to require us to 
punctuate so that frigida pellis duraque in ore may be taken 
together. 


J. P. POSTGATE. 


THE PUGIO FIDET’. 


It is very intelligible that Christians, being convinced of 
the truth of their religion and feeling the happiness of that 
conviction, should try to bring the heathen within the pale of 
Christianity. These efforts become even more intelligible if 
one takes into consideration the distinct and strict injunction 
of Jesus of Nazareth to his disciples (Matt. xxviii. 19): ‘Go 
and teach all nations.’ But if the heathen, who had never seen 
the light of Christianity, were desirable objects of conversion, 
what must those be who profess the Mohammedan religion, 
thousands and thousands of whose forefathers had actually 
been Christians? (Let one think only of the once flourishing 
and glorious churches in Africa and elsewhere, which were 


1 The titles of the book in its two 
editions run respectively thus: 

1, Pugio Fidei Raymundi Martini 
Ordinis Praedicatorum adversus Mau- 
ros, et Judaeos nunc primum in lucem 
editus...cum Observationibus Domini 
Josephi de Voisin... Parisiis M.pc.11. 

2. Raymundi Martini Ordinis Prae- 
dicatorum Pugio Fidei adversus Mau- 
ros et Judaeos, cum observationibus 
Josephi de Voisin, et Introductione Jo. 
Benedicti Carpzovi...Lipsiae...cIo Ioc 
LXXXVII. 

Carpzoyv the editor of the 2nd edi- 
tion secured the services of an eminent 
artist to represent the author’s view 
given in the Prooemium m1. (p. 2): 
‘Opus tale componam, quod quasi 
Pugio quidam praedicatoribus Christi- 
anae fidei atque cultoribus esse possit 
in promptu, ad scindendum quando- 


que Judaeis in sermonibus panem 
verbi divini; quandoque vero ad eorum 
impietatem atque perfidiam jugulan- 
dam, eorumque contra Christum per- 
tinaciam, et impudentem insaniam 
perimendam.’ Only that the ‘Poniard 
of the Faith’ is apparently not repre- 
sented as cutting the bread of the 
Divine Word, but as being ready to be 
plunged by the hand of an unseen 
man into the heart of a luckless rabbi, 
engaged in prayer-and study, and hay- 
ing on prayer-cloak and phylacteries. A 
Mohammedan evidenily dead, or feign- 
ing death, lies on his face at the 
rabbi’s feet. We must confess the 
missionaries of our day are, if not 
more sincere, less bloodthirsty than 
Carpzov and those of his time, who 
were apparently anxious to convert 
the Jews ἃ tout prix. nN 


9—2 


132 


ruined, in part at least, by the conquering followers of Mo- 
hammed!) Now, although the fathers of the Jews had not been 
for more than two thousand years idolaters and, of course, were 
much less renegades from Christianity, the desire on the part of 
the Christians to convert the Jews must be the more intense 
since the New Testament exhibits Jesus of Nazareth as the 
Messiah of the Jews in the first instance ; a doctrine which the 
Jews, though silently, most emphatically, reject by their simple 
continued existence as Jews. We do not wonder, therefore, 
that certain Christians of high standing, notably the superior 
clergy, have, at all times, supported such a conversion-move- 
ment. For a long time however these efforts were accompanied 
by war and bloodshed, as regards the heathen and Mohamme- 
dans and by oppression and persecution as regards the Jews. 
In the end these efforts proved, on the whole, unsuccessful. 
If the Christians could not often conquer the heathen and very. 
rarely the Mohammedans, owing to their material strength, they 
could still more rarely conquer the Jews in spite of their ma- 
terial weakness, largely compensated as it was by their spiritual 
strength, their Scriptures and their Rabbinical writings. Those 
therefore who were bent upon the conversion of the Mohamme- 
dans and Jews endeavoured, as they ought to have done ori- 
ginally, to convince the former by their own Qoran and the latter 
by their own Scriptures, their own Targums, their own Talmuds 
and their own Midrashim. Leaving the Mohammedans out of 
our discussion this is exactly what was tried in the north of 
France in the first half, and in the north of Spain in the second 
half, of the thirteenth century, and later on also’. Unfortun- 
ately however these efforts, though they were the efflux of 
sincere religious feelings on the part of the promoters were 
not so on the part of their agents in this conversion-move- 


ment. The Pope of Rome, the King and the Bishops of 


THE JOURNAL OF PHILOLOGY. 


1 In 1413—14 a religious disputa- 
tion took place at Tortosa between 
Hieronymus de Sancta Fide (a con- 
verted Jew) and some of the mosi 
eminent rabbis of the day. It must be 
confessed that in spite of their learn- 
ing and their piety, they did not make 


such a defence as they might and 
ought to have done. The imposing 
presence of the Pope (Benedict XIII., 
although he with his fellow-Popes 
was eventually rejected), past persecu- 
tions and fear for the safety of the 
community rendered them spellbound, 


THE PUGIO FIDEI. 133 


France, the Confessor of the King of Aragon were, no doubt, 
sincere; but the two Jewish converts, who were employed 
by them in France and Aragon respectively and most of 
-those who followed suit were unquestionably forgers and rogues 
sof the e deepest dye. In 1240 a certain Nicolaus (the Jewish 
convert Donin) pretended to be prepared to prove to the Jews 
the divinity of Jesus and the truth of Christianity. By the 
influence of the Pope, and the superior clergy of France, he 
succeeded in bringing about a public disputation with the 
Rabbis before the king, the queen-mother and other high per- 
sonages of the realm. When defeated by R. Yechiel of Paris’, 
Donin-Nicolaus accused the Talmud of containing blasphemies 
against Jesus and the mother of God; an accusation which 
finally (1244) resulted in twenty-four waggon-loads of copies of 
the Talmud being publicly burnt”. A somewhat similar, though 
less fatal, occurrence took place a little less than twenty years 
afterwards in the north of Spain. In 1263 Jayme 1. of Aragon, 
prevailed upon by his confessor (of whom more anon) ordered 
a public disputation to take place between Jewish and Christian 
doctors on the merits of their respective religions. A Dominican 
friar, a Catalonian® by birth, Fray Pablo by name, a converted 
Jew, of attainments even more moderate than those of Donin- 
Nicolaus, was chosen as champion to oppose no less a personage 
than Rabbenu Mosheh Ὁ. Nachman (the famous Nachmanides’‘). 
Fray Pablo maintained that the Talmud and Midrash testified 
to the divinity of Jesus. When disgracefully defeated, as was 
to be expected, he accused the Jews of having in their religious 
books blasphemous matter against the Founder of Christianity. 
Defeated in Aragon, he obtained from the king, by the influence 
of his confessor, a commission to preach Christianity in the 
synagogues of the Jews in those parts of Provence which be- 
longed to Jayme I.; and the Jews were commanded to 
deliver to him such books as he might want for that pur- 
pose. There we leave Fray Pablo for a while in order to 


1 For more about this great rabbi 3 See Touron, Histoire des hommes 
see Schiller-Szinessy, Catalogue,1.pp.  illustres de VOrdre de Saint Dominique 
79, 246. ,..Paris, 1743, 4to. p. 484. 

2 See Sepher Shibbole Halleqet Has- 4 See Encycl. Brit. Vol. xx, under 


shalem (ed. Buber), p. 252. Ramban. 


134 THE JOURNAL OF PHILOLOGY. 


say a few words about his protector, the king’s confessor. 
This was no less a personage than the sometime General 
of the Order of Saint Dominic, the collector of the Papal De- 
cretals, the well-known saint of the Roman Church, Raymundus 
de Pefiaforte. Although the miracle of his sailing on his 
cloak with his stick for a mast from Majorca to Barcelona’, 
on which part of his title to saintship rests, may be ques- 
tioned by heretics, the other and chief part of this title, his 
cordial hatred of Judaism, if not of the Jews, cannot be dis- 
puted. Raymundus de Pefiaforte prevailed upon his royal 
penitent to nominate a commission of five (in reality of six, as 
Fray Pablo was their guide) to search all the Hebrew books in 
his dominions and to remove all matter objectionable to Chris- 
tianity from the Jewish writings’; a command which they 
executed with a vast deal of annoyance and loss to the luckless 
Jews. Four members of this commission, of whom one was our 
saint and another the Bishop of Barcelona, were only so nomin- 
ally, as they knew little or nothing of Hebrew; the fifth was 
Raymundus Martin, the reputed author of the Pugio Fidei in 
its present state® and of the Capistrwm Judaeorum (MS. Bo- 
logna). Raymundus Martin* was born at Subirats, a small 
town in Catalonia, between 1225 and 1230. He entered the 
Order of Saint Dominic between 1243 and 1248. In 1250 he 
was certainly nominated one of eight friars who should devote 
themselves to the study of Oriental languages, especially 
Hebrew and Arabic, for the conversion of the Jews and Moors’. 
Whilst nothing whatever is known of their teacher in Arabic 
(although there is evidence that Raymundus Martin knew 
something of that language) and nothing worth speaking 
of is left of their controversial literature against the Moors®, 


1 See Touron ut supra, p. 44. 

2 See Ibidem, p. 492. 

3 See note above, p. 131. 

4 The name of the reputed author of 
the Pugio Fidei was Ramon Martinez 
and in his convent he was called Ray- 
mundus Martin, the name ‘Martini’ 
arose no doubt from the wrongly-ap- 
plied Latin genitive. A similar mistake 


is continually made on the continent 
with respect to our Castle or Castell 
(sometime professor of Arabic) who is 
called by several writers Castelli ete. 

5 See Quétif (Echard=Eckhard) 
Scriptores Ordinis Praedicatorum...... 
Lut. Paris. 1719, folio p. 396, col. 2. 

6 See Pugio pp. 429 (534) which 
contains an Arabic proverb; 452 (565) 


THE PUGIO FIDEI. 135 


there can be little question that. their teacher in Hebrew and 
Rabbinic was the before-named Fray Pablo, and that besides 
the Capistrum Judaeorum the Pugio Fidei is the only piece of 
literature resulting from the efforts of that par nobile fratrum, 
the Dominicans’, Fray Pablo and Raymundus Martin, as we 
shall see lateron. Now this Pugio Fidei contains, by the side of 
genuine matter, numerous and most shameless forgeries ; a fact 
observed by others before us, and brought home to the English- 
reading public by two of the present writer’s hearers, Messrs 
Jennings and Lowe? in their Commentary on the Psalms (Ap- 
pendix to Psalm ecx.). For this they were taken to task by 
the late Dr Pusey (The fifty-third chapter of Isaiah, Oxford, 
London, and Leipzig, 1877, 8vo.). Now, Dr Pusey, if he ever 
read the Pugio Fidei, certainly did not closely examine in the 
original the Hebrew and Rabbinic passages to be found therein. 
Nor could the late Dr Zunz, who also defends Raymundus 
Martin (Gottesdienstliche Vortrdge, Berlin, 1832, 8vo. pp. 287— 
293), have read the Pugio even in Latin (and of course much 
less in Hebrew), as we shall presently show. The others, both 
in England and abroad, who declare the contents of the Pugio 


which contains a piece of the Moreh of than the Jesuits, in consequence of 


Maimonides; 587 (749, 750) which 
contains extracts from the Qoran and 
other Moslem books etc., all in the 
original Arabic, in Hebrew letters. 
(1)It will have been seen from the 
above account, that all the deadly 
enemies of the Jews in Aragon were 
Dominicans; but their hatred was not 
confined to Aragon. The Jews driven 
out from England in 1290 also owe 
their misfortunes to the Dominicans, 
who were incensed against them on 
account of Robert de Redingge (also a 
Dominican) becoming a Jew and mar- 
rying a Jewess. In fact what the 
Jesuits were, and are, to the Protes- 
tants, the Dominicans were, and are, 
to the Jews. Only the Dominicans, 
being chiefly preachers, however emin- 
ent, have always had less influence 


their incomparable scholarship, and 


_their being chiefly the instructors of 


the young. 

2 Messrs Jennings and Lowe are 
certainly to be blamed. Not because 
they suppressed the name of him from 
whom they had obtained this infor- 
mation, since one known to be the 
pupil of another need not mention 
his master’s name in giving informa- 
tion (see Τὶ, Y. Berakhoth 11. 1); but 
they are to be blamed for not giving, 
in accordance with the Mishnah 
(Eduyyoth 1. 3) the ipsissima verba of 
their teacher. Had they done so they 
would have saved themselves some 
trouble and annoyance, and the incor- 
rect statement, in their second edition, 
that Raymundus Martin was a ‘sound 
Hebrew scholar,’ 


136 THE JOURNAL OF PHILOLOGY. 


to be genuine deserve no separate consideration, since their 
assertions rest on a second-hand foundation. 

It will be convenient to divide the charges brought by us 
against the Pugio Fidei into several classes. We shall trace 
some of the forgeries, bringing proof positive that they are such. 
We will then show that Raymundus Martin, owing to his 
ignorance of Rabbinic and even of Biblical Hebrew, could not 
have been himself the inventor of these forgeries, and we shall 
finally show that their perpetrator was not merely a rogue but 
a buffoon. 


1, 
Six proofs of forgeries pure and simple. 


Hitherto the proof against the genuineness of the contents 
of the Pugio Fidei has only been of a negative character, viz. 
that the books from which they are stated to have been drawn 
could not be found. Now such a proof was justly rejected by 
Dr Pusey, as indeed it cannot be convincing to any philosophic- 
ally-trained mind. For, if not hitherto found, these books may 
yet be found, and if never found they may yet have existed at 
one time, though now irretrievably lost. But our proofs will be 
the more convincing since we have succeeded in actually finding 
the passages referred to up to the point of their falsification ; 
since we have succeeded also in showing that these forgeries, 
whether consisting in omission or commission, commencing at 
certain points, betray themselves by their clumsiness, and finally 
since we have succeeded in producing unquestionable testi- 
mony that these deviations from the texts from which they 
have been taken not merely were not, but cowld not have been, 
the readings of the originals. 


1. Pugio, p. 284 (354) v. 
This passage (said to have been taken from the Bereshith 
Rabbah of ὦ. Mosheh Haddarshan)’ consists of two pieces 


1 Through the kindness of Mr 5. as an indefatigable and disinterested 
Buber of Lemberg (who is well-known editor of several pieces of the ancient 


THE PUGIO ΡΙΡΕΙ. 137 


copied with slight variations from Pesigto Rabbathi (Pisqo 
XXXII. "ΔΝ ΔΝ in fine). Its purport is to show that, un- 
like man, who uses for wounding one means and for healing 
another, the Lord heals by the very means with which He 
wounds, even as He pays ‘measure for measure’ (7433 "7D 
i175) i.e. wounding with the same means with which sin had 
been committed. As an illustration “virgin” (ADIN) is given 
with three proofs from Scripture For the sin, Pesigto Rab- 
batht has Joel i. 8, whilst Pugio has Ezekiel xxiii. 3’; for the 
punishment, both have Lam. v. 11; and for the consolation, both 
have Jer. xxxi. 21. Pugio adds to this last verse another (22), 
applying the word man (733) in the name of two Rabbis to 
King Messiah, etc. But this latter passage could not have been 
in the Bereshith Rabbah of R. Mosheh Haddarshan. For, in 
the first place neither he, nor any other learned Jew , believed 
that_ th the Messiah was to be born of a virgin; and secondly if 
one carefully reads this passage fathered on Rab Huna in the 
name of Rab Iddi (or, as the Majorca Codex adds, and R. Je- 
hoshua b. Levi) one finds that if } naps (Jer. xxxi. 22) was the 
Virgin and 933 was King Messiah the evidence from Judges v. 8 
has not the least force, or even meaning. Nor does the poor 


literature of the Jews as Pesigotho etc.) 
we have before us a copy of the so- 


called Bereschit Rabbathi_of Rabbi 
Mosheh Haddarshan. We can posi- 


tively assure the reader that the late 
learned Rabbi S. L. Rapoport in this 
respect first deceived himself, and then 
deceived Zunz who in his turn deceived 
many others in declaring the contents 
of this MS. to be Rabbi Mosheh Had- 
darshan’s. It is most certainly not 
the work of: Rabbi Mosheh Haddar- 
shan, although it is no doubt an early 
Midrashic commentary on the book of 
Genesis. In a general way we must 
caution the reader against the con- 
jectures into which Rapoport’s genius 
led him, against the notices of Zunz 
founded on these conjectures, and 
against the buildings reared by the 
idle on their idols’ foundations, At 


all events this so-called Bereschit Rab- 
bathi does not throw the least light 


on the Pugio; the only piece it has 


in common-with it is on the death of 
Moses (MS. on xxvi1. 17). Jellinek who | 
copied this piece directly from Rapo- 
port’s MS. (Bet Hamidrash, v1. Vienna, 
1877, 8vo. pp. XXII., XXIII.) seems not 
to know that it is to be found with 
variations in the Pugio 308, 309 (385). 

1 In the interests of truth we must 
say that the text of the Pugio Fidei 
has a better (and, no doubt, the only 
correct) reading for the first link of 
the evidence (Ezekiel xxiii. 3) than 
Pesiqto has (Joel i. 8). It is a matter 
of surprise that neither Abarbanel 
(Yeshw’oth Meshicho, Part τι., Specula- 
tion 3, Chapter 3), nor Friedmann, 
the latest editor of Pesigto Rabbathi, 
notices this superiority. 


138 THE JOURNAL OF PHILOLOGY. 


translator know what to do with this passage as applied, but is 
in the greatest perplexity how to render psy ποῦ ἿΝ for he 
gives the whole verse thus hesitatingly, ‘ Eligit Deus nova, tune 
debellabit vel tunc debellans portas: vel tunc debellatio, vel 
expugnatio portarum, vel opinionum, vel turpitudinum, vel 
immundiciarum.’ , 


2. Pugio, p. 317 (397) v. 


This passage is found at least twice with slight verbal 
alterations in Hebrew literature, once in the Zargum (Sheni* on 
Esther i. 1), and once in the so-called Pirege de-Rabbi Eliezer 
(XI. in fine). In each case however King Messiah is the ninth 
and God is the first and last. In the Pugio Fidei the words 


yw ΠΣ spn). which are substituted for "YYYN bor, 


have the effect of identifying King Messiah with God Himself. 
But such cannot be true when fathered on old Jewish literature. 
For although both this particular Targum and the Ptrege 
de-Rabbi Eliezer are by hundreds of years posterior to the rise 
of Christianity, the Midrash about the ‘ Ten Kings’ embodied in 
them is, no doubt, anterior to it; and, unluckily for the forger, 
was certainly known in the middle of the first century of the 
Christian era. See 1 Corinthians xv. 28 which runs thus: And 
when all things shall be subdued unto him, then shall the Son 
also himself be subject unto Him that put all things under him, 
that God may be all in all. 


3. Pugio, p. 337 (421) xxi. in medio; comp. p. 645 [828] v1. 
in medio. 


This is a shameful concoction of a piece of the ordinary 
Bereshith Rabbah, cap. xc. of Psalm Ixviii. 4, 6, with sundry bits 
from various Prophets put in, the references to these latter 
being given. Each of these things is true and correct in its 
original place; but the connection of the whole, which is to 
prove that the name of God means King Messiah, the mis- 


1 Abarbanel (ut supra, cap. 6) ap- Sheni as one of the sources of this 
parently knows nothing of the Targum quotation. 


THE PUGIO FIDEI. 139 


application of the Rabbinic phrase Δ TYD9N (which is 
absurdly rendered docet dicendum, and framed as an interroga- 
tion, although the word does not precede it), and the argu- 
mentation based on it are such sheer nonsense that an enlight- 
ened mind like that of R. Mosheh Haddarshan could not have 
put them together. 


4. Pugio, p. 593 ὧν fine (759) v. 


Few forgeries are so audacious and at the first moment so 
convincing as this. This passage is to be found, up to the 
evidence quoted from Scripture, verbatim in one of the oldest 
Midrashim (Zkhah Rabbatht on Lam. v. 2). For some time 
one does not observe that the chief force of the forgery consists 
in the omission of one word (DN)) and in the giving to the 


word redeemer (Osan) a Christian sense whilst, as is well 
known, the Jews use it in the sense of one who delivers from 
a material foe or trouble. Moreover the word Redeemer here 
distinctly refers to Esther, who then had neither father Nor 
MOTHER. The evidence itself however is worthy of a genuine 
statement, a genuine passage and a genuine doctrine; for 
although it refers in the first instance to Esther, if the Jews had 


had a right to apply the term by to their Messiah, the 
Christians would have had an equal right to apply it to Jesus 
of Nazareth, their Messiah. 


5. Pugio, p. 674 (866) XXXVIIL. 


We need not say that this passage is not to be found in the 
Siphere, which is said to be, Pugio 669 (859), ‘valde authenticus 
apud eos’ (scil. Judaeos); and much less is it to be found at p. 121, 
as Dr Wiinsche (Leiden des Messias, p. 65) has it; but up toa 
certain point it certainly is to be found in Siphro, xi. ὃ 10 
(Ed. Weiss, Wien, 1862, folio, leaf 27*, col. 1),in Yaigut, τ. § 479, 
and in Rashi and Leqach Tob on Lev. v. 17, where however, what- 
ever the verbal deviations in the before-named works may be, 
none has ‘the merit of the Messiah,’ nor the illogical argument 
in connection with it. The genuine passage in the 2nd and 3rd 
centuries ran thus: 


140 THE JOURNAL OF PHILOLOGY. 


‘Rabbi Yose (the Galilean) says: if thou wishest to know 
the reward of the pious in the world to come, go and learn (it) 
from the first Adam who had only been commanded one single 
negative law which he transgressed ; see how many deaths were 
decreed against him and his generations and against the genera- 
tions of his generations to the end of his generations. Now 
which measure is greater? Is the measure of goodness greater 
or the measure of punishment? Surely thou must say the 
measure of goodness (is greater). Now, if notwithstanding the 
measure of punishment which is smaller, so many deaths have 
been decreed against Adam and against his generations and 
against the generations of his generations to the end of all 
generations, anybody who abstains from eating of a sacrifice 
offered with an improper thought, or of one left beyond its 
proper time, and who fasts on the Day of Atonement, how 
much more will he obtain merit for himself and his generations 
and the generations of his generations to the end of all genera- 
tions ?’? (Compare Bensly, The Missing Fragment of...The Fourth 
Book of Ezra, Cambridge, 1875, 4to. page 28, note 6.) For 
more than one reason we abstain from entering into a religious 
controversy with the late Dr Pusey. Only we must make two 
exceptions, one is here and the other will come presently. (1) 
Regarding this passage. Whilst Dr Pusey is perfectly correct 
from his Christian point of view in saying that the merit of the 
Messiah is greater than keeping a negative commandment, he is 
totally wrong from an ancient Jewish point of view, according to 
which it was held that Adam brought misery upon his descend- 
ants by not keeping one single negative commandment, whilst 
the pious, by keeping ever so many negative commandments 
bring merit on themselves and on their generations to come. The 
fact is that the skill of the forger deserted him in this particular 
passage, and his forgery is absolutely proved by his leaving intact 


the words pps by ΓΟ) which have no meaning if, as 


Dr Pusey rightly says, from his point of view, the merit of the 
Messiah is higher than the keeping of negative commandments. 
But the most absurd of all is that King Messiah is not 
merely introduced in the translation as suffering (AIYNIN) but 


THE PUGIO FIDEI. [41 


also as fasting (ΣΝ ΓΙΘΓΊ), which shows that there originally did 
stand something of fasting, 1.6. abstaining from food on the Day 
of Atonement. (2) The other exception we are obliged to 
make is, as regards Dr Pusey’s appeal to Ben Addereth, by 
whom he means Rashba (3 Encycl. Brit. xx.). Had the Pugio, 
so argues Dr Pusey, contained forgeries of Rabbinic passages, 
Rashba surely would have pointed them out as such, and that 
he (Rashba) had seen the Pugio Pusey knows apparently from 
Dr J. Perles’s most interesting monograph: R. Salomo ὃ. 


Abraham ὃ. Adereth, Breslau, 1863, 8vo. pp. τὸ note 2, 
ΔΓ note 1, }’® note 1, 8’ note 1, 9”’3 note 1. Now Dr 
Perles can be scarcely correct in asserting that, in answering 
a Christian’s attacks made on Rabbinic Judaism, Rashba had 
Raymundus Martin in his eye. Rashba no more mentions the 
Christian assailant’s name than that of the Mohammedan 
assailant whom he also refutes. But the points mentioned in 
the before-named five passages were then in everybody’s mouth. 
We are quite sure moreover that Rashba only knew of these 
attacks by hearsay, for however liberal-minded a rabbi Rashba 
was for his age, he nevertheless would not have allowed an 
‘heretical’ work, like the Pugio Fidei, to be in his house, 
Rashba must have known of the connection which had existed 
between Raymundus Martin and Fray Pablo, an DINAN 
Sw, a “WS and a “WD (a converted Jew); and this 
fact alone would have precluded the toleration of the Pugio 
Fidei by him. Had Rashba (who knew neither Arabic nor 
Latin, but was certainly a most eminent Rabbinic scholar) seen 
the Hebrew-Rabbinic quotations of the Pugio he would have 
instantly discovered the forgeries contained in them and have 
rent his garments (comp. 2 Kings xviii. 37; Is. xxxvi. 22) on 
reading blasphemous matter by a Jew, but would not have 
answered them (T. B. Synhedrin, leaf 38 b): we in our days are 
satisfied with merely exposing the forger’s nefarious transactions... 


6. Pugio, p. 683 (877) v. 


This passage is found verbatim in the ordinary Bereshith 
Rabbah, cap. Lv1. (on Genesis xxiii. 4), After the word 9B, 


142 THE JOURNAL OF PHILOLOGY. 


towards the end, is added quae est rex Messias sicut dictum est 
Psalm. \xxx. 8. Domine Deus exercituum converte nos: ostende 
faciem tuam et salvabimur. Hucusque Glossa. The Hucusque 
Glossa is doubly untrue, for in lieu of the genuine piece a sub- 
stitution is made. After this could Zunz, if he had read the 
Pugio Fidei either in Latin or in Hebrew, and Pusey, if he had 
read it in the original, have maintained the genuineness of its 
contents ? 


IT. 


Sia proofs of the ignorance of the translator, pure and simple. 


1. Pugio, p. 317 (396) π|. 


He translates NYT733 (the name ofa place in Babylonia) 
by “in Nahardea,” id est “in flumine conscientiae” ! 


2, Pugio, p. 332 (416) xv. 

He translates the passage (not quite exactly quoted) “BX 
min mada> ap syn ye Ny (Comp. T. B, Synhe- 
drin, leaf 43 a, Amsterdam ed. of 1645), by Diait Ula, Jesus 
Nazarenus propinquus fuit regno, id est familriae regiae. In 
reality this is impossible. M\7 mabn> Δ} ) means because 
he was held in favour by the government, that is, by Pontius 
Pilate who represented Caesar; a view fully borne out by the 
Gospel writers (Matt. xxvii. 19—24; Mark xv. 14; Luke xxiii. 
4, 14, 15, 22; John xviii. 38, xix. 4, 6, 12, 14, 22). Unfortu- 
nately, a certain Membre de l'Institut made a similar mistake 
in translating this passage’, and has been the cause of mis- 
leading others. 


3. Pugio, p. 549 (697) ΧΙ. mm medio. 


He makes that wonderful and strange discovery that *\w3 
(Hosea ix. 12) means “incarnatio mea,’ whilst every child ac- 
quainted with Hebrew knows that DS )w3, however 
written, means “ when I depart from them.” 


1 Derenbourg, Essai...Paris, 1867, p. 349, note 2. 


THE PUGIO FIDEI, 143 


4, Pugio, p. 664 (852) ΧΧΙ. 


Et tu Domine es scutum pro me, gloria mea et exaltans caput 
meum. Gloria mea eo quod habitare fecisti divinitatem tuam in 
medio nostri et exaltans caput meum pro eo quod facti sunus 
rei tibi exaltasti caput (1. nostrum) id est dedisti nobis suspen- 
stonem capitis sicut dictum est etc. The miserable translator 


did not know that MBA 7D (1. ΚΞ) ΚΞ ΠΘ wMY nnn 
wen yon > AND. BN signifies: whilst we had, by our 


sins against thee, incurred. the penalty of decapitation’, thou 
erantedst unto us an elevation of the head. It is however 
scarcely fair on our part to charge Raymundus Martin with 
ignorance when some Jewish scholars made the same mistake. 
We will therefore not count this but substitute another number 
for it, calling it 


4b. Pugio, p. 669 (859). 


In the text used by him or his teacher the two words 


sbynd rps in the last line but one are out of place. Our 
ignorant translator of course has no idea that such is the case, 
and thoughtlessly translates, ‘Simile est et Numer. 12, v. 12. 
Qui in egressu suo de utero matris nostrae, et edit dimidiam 
partem carnis nostrae. Oum in lege scriptum sit, dici potest, 
DE DEO loquitur, sed scriptura mutavit, et posuit, matris suae, 
et carnem suam. Hucusque Traditio’, thus referring the pas- 
sage to God, which is impossible ! 


5. Pugio, p. 671 (861) ΧΧΎΠΙ. in medio. 


Here our translator tries his hand a little at Aramaic, 
showing that he ought to have been not merely silent in two 
languages (Hebrew and Rabbinic) but also in a third (Aramaic) ; 


1 This is, no doubt, not the bona fide 
Biblical meaning, but a Midrashic ex- 
planation. Nevertheless Friedmann 
(Pesiqto Rabbathi Pisgo x. (NWN "3 
note 89) is not quite correct in the 
distinction he makes here between the 


Biblical and the Rabbinic meaning of. 
WNIT NDI (DI in the Hiph’il); comp. 
Ezek, xxi, 31(26), where FOYT Ὁ ΠῚ is 
absolutely parallel with NDIY9N VA 
and is accordingly rendered both by 
the Authorized and Revised Versions. 


144 THE JOURNAL OF PHILOLOGY. 


for he translates the well-known mond Nm, Isaiah [11]. (“ he 
shall be despised”) by propter hoc erit ad bonam Annuncta- 


tionem, as if it were mead (a messenger of good tidings). 


6. Pugio, p. 713 (918) ΧΧΙΙ. 


There is a Talmudic passage here, which of course must be 
taken allegorically, translated in the following singular manner: 
Dixit R. Aggaeus: Primi araverunt et seminaverunt et sar- 
culaverunt et messuerunt et arconisaverunt et trituraverunt et 
ventilaverunt et purgaverunt et moluerunt et apposuerunt et non 
est nobis os (5) ad edendum. Now here evidently either the 
5 stands for a 3, as in both Talmuds, and the word was AQ, 
“nothing whatever to eat,” or the ™ stands for a ΠΣ) and it (FS) 
means “we have no bread to eat;” but in no case can it 
mean they had not a mouth to eat, for these rabbis had just 
been talking with their mouths and thus, on the principle of 
the Cartesian Cogito ergo sum, they must have been conscious of 
having a mouth. Blasius Ugolini (Thesaurus, ΧΎΤΙΙ. Coll. Xcmmt. 
XCIV.) is nearer the truth, for although his Hebrew is also 75 
he nevertheless translates it by panis. A little later on there 
is another Talmudic passage given, in which the Rabbis in 
their great humility (comp. Ps. lxxiii. 22) say, δὲ primi fuerwnt 
homines ; nos sumus asini: et non sicut asinus R. Chaninae ; 
nec sicut asinus R. Menachem (1. Pinechas) fila Jair etce., on 
which the writer of the Pugio makes the following charitable 
remark: Asininitatem hance pastorum, 1. 6. magistrorum Judae- 
orum etc. We leave it to the reader to decide whether the 
Rabbis of the Talmud or their traducer deserve to be charged 
with this asininitas. After this could Zunz, if he had read the 
Pugio Fidei, have said of Raymundus Martin that he was “ew 
tiichtuger Gelehrter” (G. V. p. 288), and Pusey if he had read 
the original (Introd. wt swpra) have said: ‘there is no doubt of 
his ability’ ? 


THE PUGIO FIDEI. 145 


ITI. 
Siz proofs of forgeries and ignorance combined. 


1. Pugio, p. 222 (277) 1x. 


We are treated to two passages (Midrash Rabbah on Gen. 
xvill. 22 and Midrash Tehillim on Ps. xvill. 36) where the 
Tiqqune Sopherim are mentioned, and of course among these is 
the alleged falsification of Ps, xxii. 17, 985 sicut leo instead of 
ΝΘ or YS foderunt. This enumeration of the latter among 
the Tigqune Sopherim is a clear forgery. We know from Qimchi 
(in loco) who charges the Christians (perhaps referring to the 
Septuagint) with inventing this reading, that it could not have 
been in the Hebrew text. Nor could it have been in R. Mosheh 
Haddarshan, as is apparently implied in Pugio, p. 547 (696) Xt, 
where it is given as reported by one R. Rachmon, of whom more 
anon. For Qimchi lived in the very place where R. Mosheh Had- 
darshan had lived only one hundred years before ; a place where 
his works, if anywhere, must have been well known. Moreover 
S75 or SND no more could signify “piercing” than the Tar- 
gumic {1} could signify “ piercing” (353 means “digging, 
hollowing out,” and ΓΤ) means “ biting”). The reading (N35 is 
indeed to be found in several of the MSS. collated by Kennicott 
and probably in others besides. But anybody can see that the 
reading \"N5 originated naturally in the elongation of the " into 
a }, for in the root M3 there certainly is no X. There is, in- 
deed, to be found in the printed ‘editions, and perhaps also in 
some manuscripts a little earlier than these, a piece of Massoreth 
Parva, running thus ger "NA PSDP ‘3, which would give 
colour to ‘85 meaning something different from “like unto a 
lion.” It ought to be remarked, however, that not only are 
these last two words not to be found, as far as is known, in a 
manuscript of high age, but the age of this Masoretic note 
itself has yet to be ascertained. Moreover, there are actually 
manuscripts of great age lying before us in which the “N35 is 
not even spelt with a , (7370) but with a - (AMD). Let us add 


Journal of Philology. vow. xvi. 10 


146 THE JOURNAL OF PHILOLOGY. 


that this piece of Massoreth, as it stands in Ginsburg, Massorah. 
I, p. 106, Rubric 1079, together with the next Rubric, implies 
no diversity of meaning at all and ABSOLUTELY PRECLUDES A 
DIVERSITY OF SPELLING. ΑΒ regards the Targumic }‘M3) we 
wish to draw the reader’s attention to the fact that in this 
Aramaic word we have a second translation of the Hebrew 
ΒΤ ; a practice frequent in all Targums, e.g. Genesis iii. 21, 
where “VY is rendered “skin,” and again “glory” (see Adler, 
Nethinah Lagger, in loco), as if there had stood “δξ. The real 
meaning of the phrase Tigqgune Sopherim is, of course, entirely 
unknown to, and absolutely misrepresented by, the author of the 
Pugio Fidei, whoever he was; and not even really learned men 
Jews or Christians have quite grasped its real signification. 


2. Pugio, p. 452 (563) 1. 


Here is a passage which testifies no less to the incapacity 
than to the audacity of the forger. Can anybody, who is in 
the least acquainted with Rabbinic literature, believe that any 
Rabbi would teach so monstrous a piece of nonsense, ay of 
idolatry, as is here attributed to R. Mosheh Haddarshan, 
that the Lord should have commanded the angels to worship 
the first man*? Let one only read the genuine Bereshith 
Rabbah (cap. vit.), where R. Hosha’ya says: At the time 
when the Holy One (blessed be He!) had created the first 
man the ministering angels erred and wished to say before him 
(man): ‘Holy!’ This may be compared to the case of a King 
and his Stadtholder who were in the same carriage. The peo- 
ple wanted to address the King, ‘Domine!’ [}%3\7] (or recite 
a hymn [})J7] to him); but they did not know which of the 
two it was. What did the King do? He pushed his Stadt- 
’ holder out of the carriage and all thereby knew that this was 
only a stadtholder. Thus also when the Holy One had created 
the first man the ministering angels mistook him for God and 
wanted to recite before him, ‘Holy!’ What did the Holy One 
(blessed be He!) do? He threw upon him a deep sleep, and all 

1 This monstrous piece is not in once in it, as 727%, WAIN, 7%. 


Rabbinic, but in imitation Hebrew, i.e. Moreover 127 is used instead of ION, 
the ἡ conversivum occurs more than ete. 


THE PUGIO FIDEI. 147 


knew that this was a mere man. This is what is written in 
Tsaiah ii. 22: Cease ye from man whose breath is in his nostrils; 
for wherein is he to be accounted of ? 


3. Pugio, p. 520 (657), towards end of Ch. 11. 


After having treated in the whole of the chapter of the 


names of God and especially of that of pow (‘Peace ’) he quotes 
a passage which he pretends to have read in the prologue to 
Ekhah Rabbatht (where of course it is not to be found). In 
this passage, which occurs in one shape or another more than 
once in other Rabbinical writings (Vayyigra Rabbah, 1x. &c.), 


he purposely leaves out in the text the word δ Ὁ and in the 
translation the corresponding word, in order to show his par- 
ticular cleverness’ by explaining it in a bracket as if it was his 
invention (and in a certain sense an invention it certainly is). 
All this is done in order to be able to interpolate a saying 
attributed to so early a teacher as R. Yose Haggalili that the 
Messiah also is called ‘Peace’; an application which not every 
Rabbi would admit, as the words ‘and He called his name, 
&e. Prince of Peace, are generally referred by the Rabbis 
to king Hezekiah. Now it ought to be remarked that whilst 
R. Yose Haggalili is introduced in the Midrashim as say- 
ing that God’s name was ‘Peace, it is not reported that he 
said that Messiah’s name also was ‘ Peace, although: the occa- 
sion lay very near, for he speaks in more than one place of 
Messiah immediately after God. Such a statement that Mes- 
siah’s name also was Peace is, indeed, distinctly made in Pereg 
Hasshalom ; a piece of literature which is, however, apocryphal. 
For although the sayings to be found therein are not only on 
the whole true but mostly extracts from the Talmuds and the 
Midrashim, the saying that Messiah’s name also was ‘ Peace’ is 
anachronistically attributed to R. Yose Haggalili, a pre-Mishnic 
teacher of the time of Hadrian, who is supposed to make his 
remarks & propos of those of R. Chiyya bar Abba who lived in 

1 How capable he was of translating to mean ‘id est quando persolvit quod 


Rabbinic will be seen from the follow- natura requirit’! 
ing example: ἢ, Dip is explained 


10-—2 


148 THE JOURNAL OF PHILOLOGY. 


the times of Diocletian! Moreover, the Yalqut apparently 
knows nothing of this Pereg Hasshalom, nor was it found in the 
genuine Machzor Vitri, no doubt now lost, as all Machzors 
going under that name (including that otherwise precious MS. 
[ Add. 27, 200—1] preserved in the British Museum) are of much 
later date and contain much later literature. Nor, indeed, was the 
Pereq Hasshalom, seemingly, before the eyes of either the scribe 
or the owner of the famous Talmud MS. preserved at Munich. 
It seems to have been compiled late in the x1th century either 
in Provence, by a Jew, or in Aragon, by a converted Jew, 
where the author of the Pugio Fidei, whoever he was, saw it 
and copied out this sentence, leaving out purposely the words 
Bar Abba after R. Chiyya, a 2nd century teacher, so as to make 
the anachronism less glaring. (It is not impossible that he who 
did the forging part in the Pugio Fidei is the author of this 
whole sentence.) But one might ask, Why did the author of 
the Pugio Fidei quote a passage from a book where it is not to 
be found when he could have quoted it from a book where it 
is to be found? To this we give the following answer. In 
the first place because of the high antiquity and authority of 
the Ekhah Rabbathi, between the composition of which and - 
that of the Pereqg Hasshalom many centuries elapsed; and 
secondly because of the craft of the forger who anticipated this 
very question, and expected that people would acknowledge his 
quotation as genuine, although it is not to be found in the book 
which he pretends to quote, seeing that it is to be found in 
one from which he might have quoted it. 


4. Pugio, p. 565 (719) vit. in medio. 


This is not merely a forgery, but a piece of ignorance. The 
reference is given to T. B. Baba Bathra, and one R. Rachmon is 
mentioned, who must have lived later than R. Mosheh Had- 
darshan, seeing that he quotes the latter, comp. p. 548 (695) ΧΙ, 
whilst here he is apparently a Babylonian teacher mentioned 
in the Talmud. There are other anachronisms in this pre- 
tended Talmudic passage, which if it occurred anywhere in 
Baba Bathra must be at leaf 25a. The fundamental ideas how- 
ever are to be found more than once in Rabbinic writings. 


THE PUGIO FIDEI. 149 


5. Pugio, folio 620 (794) xxx. sub finem. 


Here R. Rachmon is introduced as a teacher who lived 
anterior to the composition of the Midrash Rabbah. He is 
professedly quoted from the smaller (or ordinary) Bereshith 
Rabbah (cap. XLVII. in fine on Gen. xvii. 27) where although 
the first portion of the quotation is to be found no such person 
is mentioned ; but see the next division (IV.). 


6. Pugio, folio 664 (852) ΧΧΙ. in medio. 


Here is a piece of shameless forgery with ignorance com- 
bined. In the second quotation (i.e. from the Tanchwma) the 
word YX" is left out (although it is inserted on the margin) and 
id est exaltatio capitis becomes thus suspensus or sn = cruci- 
fizus! Now could Zunz and Pusey, if they had read the Pugio 
Fidei in the original, have maintained the genuineness of its 
contents and the learning of Raymundus Martin ? 


IV. 


Proof of the irreverence of the forger. 
Pugio, p. 335 (419) xx. 


Here is a passage that professes to be written out of the 
Bereshith Rabbah of R. Mosheh Haddarshan and which, on the 
whole, is kindred to Pesigto Rabbatht Pisqos XXXVI. (8 PP) 
and XXXVIL (W°YWN WW); but neither here does it occur, nor 
there could it have occurred that the son of David, Messiah 
(Χριστός), was irreverently introduced as ‘loving the daughters 
of Israel,’ for the very reference given to Cant. iii. 9, 10 proves 
the direct contrary. The writer who was, or rather pretended to 
be, a Christian, ought to have had regard to the passage in 
the New Testament (Luke xxiii. 27), which testifies of the real 
love the daughters of Jerusalem (ie. of Israel) bore to him 
whom he professed to call his master, and runs thus: ‘And 
there followed him a great company of people, and of women 


150 THE JOURNAL OF PHILOLOGY. 


which also bewailed and lamented him,’ showing that it was 
rather the daughters of Jerusalem (or Israel) who bore the love 
(just as it is given in Canticles). 

It is clear from the foregoing that the trinelabad of the 
forged passages in the Pugio Fidei could not have been the 
author of them. For although somewhat clumsily, they are 
yet too cleverly, done to be the work of the translator. Whilst 
the author of the Pugio Fidei as such is perhaps the translator 
of these forgeries also the forger himself must be another person. 
Who then was he? We have not far to seek for him; there can 
be little doubt that it was Fray Pablo. (1) Smarting under 
the defeat inflicted on him in 1263 by Nachmanides, Fray Pablo 
furiously hated his former coreligionists. (2) Fray Pablo, if 
we may judge from the known to the unknown, was, after 
his conversion, scarcely what may be called a good Christian. 
(3) It was Fray Pablo who, travelling in Provence with the 
commission to preach to the Jews, etc. (see above, p. 135), no 
doubt obtained among other books the Midrashic commentary 
on the Five Books of Moses by R. Mosheh Haddarshan (probably 
identical with the JWT AWD "Δ by ἡ ; compare Rashi 
on Num. vii. 18, etc., and the sn mvp "Δ ‘37; comp. 
Rashi on Gen, xxxv. 18, etc.). Now the Yesod of R. Mosheh 
Haddarshan, being unlike ‘Rashi’ an absolutely Midrashic 
interpretation, Fray Pablo had but little to alter, omitting 
here and there a word or two, adding here and there a word or 
two especially at the end of a genuine passage, and occasionally 
only inserting two or three lines in the middle of a passage. 
Understanding Rabbinic well, although quite unable to write 
it, his weak points are chiefly seen not in his omissions or slight 
additions, but in these insertions. (4) Fray Pablo then, there 
can scarcely be a doubt, is the sole author of the forgeries pure 
and simple, and joint author of the forgeries combined with 
ignorance. He is again the sole author of the various inven- 
tions to be found in the book, of which the following is not the 
least diverting. 

It cannot have escaped the attentive reader of this article 
that a certain R. Rachmon is several times introduced in 


THE PUGIO FIDEI. 151 


the Pugio Fidei, now as apparently living in pre-Midrashic 
times, or in pre-Talmudic times, or in post-Talmudic times, and 
now as late as the 11th century or even later still. This 
R. Rachmon is found in the Pugio Fidei several other times 
also (as pp. 534 (676), 620 (794), 660 (847), 665 (854) three 
times, 671 (862), 679 (872), 682 (877), 729 (928), 741 (955), ete.). 
But who is this ubiquitous personage with a life as long as that 
of the Wandering Jew? This R. Rachmon is in one sense an 
invention of the forger, since he never lived in pre-Midrashic 
times, or in pre-Talmudic times, and in another sense a 
reality, since he certainly lived not merely in post-Mishnic 
and post-Talmudic times and after R. Mosheh Haddarshan, but 
even as late as the second half of the 13th century. We will 
put at once an end to the reader’s suspense. This enigmatic 
R. Rachmon is no Rabbi, but is none less than Rahmon, 1.e. 
Ramon or Raimond; in full, Raymundus Martin himself, the 
supposed author of the Pugio Fidei in its totality! What would 
Zunz and Pusey, if they were here alive now on earth, say to 
this extraordinary dénouement ? 

Now having convincingly proved, as we believe, that neither 
Zunz nor Pusey could have read the Pugio Fidei in the original 
Hebrew, and admitting that the latter may have possibly read 
it in Latin, we proceed to give unquestionable proofs that 
the former could not even have read much of the translation 
with any attention. Zunz published, as is known, in his Zert- 
schrift (Berlin, 1822, 8vo. 1. pp. 277—384), a life of Rashi, in 
which (p. 279) he declaims against such Jews as derive their 
Hebrew information from Buxtorf, charging the latter with 
inventing a wrong solution of the " in the word ”Y%, that 
letter signifying ΓΝ and not 9M. In his Die Goittes- 
dienstlichen Vortrdge etc., which, as is seen from above, came 
out in 1832 and in which Zunz corrects ever so many of the 
mistakes made by him in Rashi’s life, that mistake ascribing 
to Buxtorf the invention of "7" (instead of ΠΝ) is certainly 
left. In 1839, indeed, Zunz had found out that Buxtorf was 
not the inventor of the before-named wrong solution. He then 
charged (15. Ann. 1. p. 336) Sebastian Miinster with being the 
originator of this error. This mistaken view he held at all 


152 THE JOURNAL OF PHILOLOGY. 


events down to 1875 (see his Collected Works, Berlin 111. p. 104). 
But had he read any considerable portion of the text of the 
Pugio Fidei, in the translation, with the slightest attention, he 
could not have failed to see that the " in Rashi had been solved 
as 7 as early as, if not earlier than, the 13th century (see 
Pugio, pp. 252 (315), 264 (829), 283 (353), 303 (378), 309 (386), 
311 (389), 321 (401), 326 (407), 329 (411), 336 (420), 343 bis (429, 
430), 353 (441, 442), 358 (448), 372 (466), 373 (467), 376 (470), 
378 (471), 382 (476), 390 (487), 413 (515), 450 (562), 488 (613), 
569 (726), 611 (781), ete. Now the so-called foolish fabling 
and wicked rabbis certainly exhibit a higher sense of justice 
and morality when they say (Shemoth Rabbah cap. XLV1): 
‘Woe unto people who testify concerning things they have not 
seen” and (T. B, Niddah, leaf 7”): “One must not say to a man © 
who has not seen the New Moon that he should come and 
testify (to its appearance), but one says to a man who has seen 
it (Come and testify !).” 


S. M. SCHILLER-SZINESSY. 


Postscript. The above article is substantially identical with 
“ Raymundus Martini” originally written for Vol. xx. of the Hncy- 
clopaedia Britannica and subsequently withdrawn out of consideration 
for the learned doctor of Berlin who was then in his ninety-second 
year. But Dr Zunz, as Dr Pusey, is now with God, where nothing 
but pure and absolute truth can prevail, and we thought it our duty 
to give the whole truth. For if one owes some consideration to the 
living, one certainly owes to the dead nothing but truth (On doit des 
égards aux vivants, on ne doit aux morts que la vérité). 


A ROMAN MS OF THE CULEX. 


Amone the MSS in the Corsini Library in Rome is one 
~numbered in the old printed catalogue 64 (now 43 F. 5), 
which, as I hope to show presently, holds a unique position 
among the MSS of this corrupt poem. It is in parchment, 
consisting of 84 leaves. The order of its contents is as follows: 
1. Franc. Petrarchae Eclogae. 2. Claudianus de Raptu Proser- 
pinae. 3. Prudentii Psychomachia. 4. Epistola Saphos poetisse 
ad Phaonem amasium suum feliciter. 5. Calpurnii Eclogae. 
6. Culex. 7. Aetna 1—6. It was therefore written after the 
publication of Petrarch’s Eclogues, and may belong to the late 
14th or early 15th century. ,The capitals in which the titles 
or headings of each poem are written are in red; as also the 
names of the interlocutors in the Eclogues of Petrarch and 
Calpurnius. In form the MS is a very tall octavo; a large 
margin of more than an inch is left on the right of the text; a 
smaller (about 4 inch) on the left. Each page, when written 
uninterruptedly, has 35 lines. The writing is clear and the 
pages very clean. 

I shall mention at once what gives it a unique place among 
the MSS of the Culex. vv. 366—3868 are written in all the 
earliest MSS (those at Paris, the Cambridge, and Vatican 3252) 
as follows: 


365 Mucius et prudens ardorem corpore passus 
Legitime cessit cui fracta (facta Vat.) potentia regis 
Hic curius clare socius uirtutis et ille 
Flaminius deuota dedit qui corpora flamme. 


As far as I know, no one has yet been able to explain the 
meaning of 366. Now the Corsini MS gives it as follows: 


154 THE JOURNAL OF PHILOLOGY. 


Cui cessit lidithime facta potentia regis 
Le. Cui cessit Lydi timefacta potentia regis, 


a reading which at once makes all clear. The Lydian King is 
of course the Etruscan Porsena. I should suppose that there 
can be no doubt that this is right; and if so, the form of the 
corruption points to a writing in which g and d were easily 
confused. 

The following are also passages where our MS perhaps points 
to the true reading: 

65 lapidum nec fulgor in ulla 

Cognitus utilitate manet. 


Corsini ulna. Read ulnae, and explain of jewelled bracelets 
worn on the arm. 
87, 88 
Ille colit lucos, illi panchaia tura 
Floribus agrestes herbe uariantibus adsunt. 


Corsini herbas and addunt. Here it is difficult to decide 
between the two possibilities opened by Cors. 


(1) Floribus agrestes herbae uariantibus addunt, 
(2) Floribus agrestes herbas uariantibus adsunt ; 


but one or other appears to me indubitably right. 
92 Quolibet ut requie uictu contentus abundet. 


Cors. requiem. Read 
Quolibet ut requie ἐπ uictu contentus abundet, 


1.6. ut requie abundet contentus in uictu quolibet. In this way 
the awkwardness of the two ablatives requie uictu intersecting 
each other requie abundet, contentus quolibet uictu is avoided, 
and the construction becomes clear. 


98—100 
Talibus in studiis baculo dum nixus apricas 
Pastor agit curas et dum non arte canora 
Compacta solidum modulatur harundine carmen. 


> 


Cors. et nondum dum arte canora. Read therefore δέ dum 
nondum arte canora, ‘and whilst with no tuneful art as yet, 


A ROMAN MS. OF THE CULEX. 155 


i.e. with a simple music that had not yet reached the dignity 
of art. 


122 et dulci fessas refouebat in umbra. 
Cors. feras.. Read fetas, sc. capellas. 


177 naturae comparat arma 
Ardet mente, furit stridoribus, intonat ore. 


Cors. computat, which agrees particularly well with the cata- 
logue raisonné of the snake’s different preparations for attack ; 
his inward fury, his angry hissings, the loud sounds or cries 
by which he expresses his rage. 


185 Qua diducta genas pandebant lumina gemmis. 


Cors. gemmas. Possibly for gemmans, which Schrader con- 
jectured, constructing the word with pupula in 186. 


262 Aduersas perferre faces all the early MSS. Corsini 
alone has preferre, rightly. 


311 Ipsa uagis namque Ida potens (patens) feritatis et (ab) 
Ida 


Ida faces altrix trepidis prebebat alumnis. 


Cors. Ipsas uagit. Probably therefore Ipsa sudis. The 
rest of the v. I have already corrected in the American Journal 
of Philology. The whole is now clear. 


Ipsa sudis namque Ida parens feritatis et ipsa 
Ida faces altrix trepidis praebebat alumnis. 


It will be observed that here again d and g have been con- 
fused. 


332 The epithet of Carybdis in this v. which in some 
MSS appears as ranolea, in others as metuenda, in Cors. is 
uerida: but I do not know what to elicit from this. 


352 hic modo letum 
Copia nunc miseris circumdatur anxia fatis. 


So most MSS, Jaeta edd. generally. Cors. has letam, i.e. I 
think laetans. The pres. part. has its full and proper force 
= quae modo laetabatur. 


156 THE JOURNAL OF PHILOLOGY. 


357 Cors. gives this v. thus: 
Omuis in equoreo fluit atia naufrage luctu. 


379, 380 non inmemor audis 
Et tamen ut uadis dimittes omnia uentis, 


So most MSS. Cors. has dimitteres. I explain this as 
follows. The poet wrote: 


Ut tamen audieris, dimittes omnia uentis. 


The er was transferred from audieris to dimitt es. To 
attempt any more explicit statement would be rash: but not 
only is the meaning thus quite clear, but the mode by which 
the corruption crept in intelligible. 

In the Aetna, v. 1 is given in Cors. 

Aetna mihi ruptisque caui fornacibus ignes. 


5 is omitted, but at the bottom of the page is written, 
inclosed in a border, seu te Cinthds. 


ROBINSON ELLIS. 


ARISTARCHOS’ READING AND INTERPRETATION OF 
ILIAD N 358—9. 


THIS passage stands in our texts as follows : 


\ ’ A 4. i / 7 
τοὶ δ᾽ ἔριδος κρατερῆς καὶ ὁμοιίου πολέμοιο 

a > U > Fae ΦυΔι , f : 
πεῖραρ ἐπαλλάξαντες ἐπ᾽ ἀμφοτέροισι τάνυσσαν. 


With other unimportant variants La Roche notes for τοὶ δ᾽: 
οἱ δ᾽ D, τὼ δ᾽ L. 

The Didymean Scholia are, 

358. τοὶ] οὕτως ᾿Αριστοφάνης. ἄλλοι δὲ οἱ δ᾽ ἔριδος. 

359. ἐπ᾽ ἀμφοτέροισι] διχῶς ᾿Αρίσταρχος, καὶ ἐπ᾽ ἀλλή- 
λοίσιν. ὃν δὲ δι’ ἀμφοτέρων τὸ λεγόμενον, ὅτι 6 ἸΤοσειδῶν καὶ 
ὁ Ζεὺς τὸν πόλεμον τῇ ἔριδι συνέδησαν, τὸ πέρας τῆς ἔριδος καὶ 
πάλιν τὸ τοῦ πολέμου λαβόντες καὶ ἐπαλλάξαντες ἐπ᾽ ἀμφο- 
τέροις, ὥσπερ οἱ τὰ ἅμματα ποιοῦντες, τόδε ἐπὶ τόδε. οὕτως 
᾿Αρίσταρχος. 

Schol. V. ᾿Αρέσταρχος...δ᾽ ἔριδος, ΜΡ ΟΝ τοί, 


It will be observed that the scribe of Schol. V. has inge- 
niously omitted the critical word which distinguished the 
reading of Aristarchos from that of Aristophanes. Rémer, 
followed by Ludwich (Aristarch’s Homerische Textkritik, i. p. 
357), inserts of into the lacuna. This, I believe, is demonstrably 
wrong. 

Aristarchos, it will be observed, understood the passage to 
mean, “They (Zeus and Poseidon) knotted together a rope of 
strife and a rope of war,” But evidently, if Aristarchos meant 
the pronoun to refer to the two gods, he must have read τώ: 


158 THE JOURNAL OF PHILOLOGY. 


believing as he did the Attic origin of Homer, in support of 
which he emphasized the use of the dual as common to the 
Attic and Homeric dialects, he can hardly have failed to admit 
it in a passage where it is not only inoffensive but actually 
prevents the ambiguity of a reference to a quite different 
subject, the Τρῶες and ᾿Αχαιοί mentioned just above—an am- 
biguity which profoundly affects the whole interpretation of the 
passage. Indeed nothing more is needed than the first Scho- 
lion quoted to prove that at least he did not read oi δ, Had 
he done so it is quite impossible that Didymos should have 
classed him, almost contemptuously, among the anonymous 
ἄλλοι to whom that reading is ascribed; there is only one 
other parallel for such an extraordinary course, viz. on O 459, 
Ζηνόδοτος μάχης, ἄλλοι δὲ μάχην, and that we may safely 
conclude is corrupt. See Ludwich, i. p. 125. If then he read 
neither of δέ nor τοὶ δέ, he must have το τώ. This is found 
in La Roche’s L (Vindobonensis quintus, no. 105 in La Roche’s 
Homerische Textkritik, p. 476), a MS. which occasionally stands 
alone in preserving Aristarchean readings; a striking case is 
found a few lines farther on (399), where L alone has the 
Aristarchean αὐτὰρ ὃ ἀσθμαίνων (for 6 y’)’. 

It follows then that the original Scholion of Didymos on 
358 must have been οὕτως ᾿Αρίσταρχος τὼ δ᾽ ἔριδος. ᾿Αριστο- 
φάνης τοί, ἄλλοι δὲ οἱ δ᾽ ἔριδος. The error is simple enough, 
depending on the similarity of name—a frequent source οὗ con- 
fusion in the Scholia—aided perhaps by the fact that the 
original note was appended to a text which had τώ; when 
transferred into A, which has τοί, the οὕτως was no longer 
applicable. Little stress however can be laid upon the last 
suggestion, for it is well known that the scribe of A continually 
copies out the οὕτως of Didymos without noticing whether or 
no it suits the text before him (Ludwich, i. p. 143). However 
this may be, the conclusion is equally certain, that Aristarchos 
wrote τώ, not τοί or οἱ. 

It will now be seen that we have two distinct readings from 


1 The ‘‘Lipsiensis” is so closely to have coincided here also. That L 
connected with L that we may pro- only is quoted is doubtless due to im- 
bably assume the readings of the two _ perfect collation of Lips. 


ARISTARCHOS’ READING OF ILIAD N 358—9. 159 


which our text is conflate. One is τὼ δ᾽ ἔριδος... ἐπ᾽ ἀμφοτέ- 
ροισι τάνυσσαν, the other is τοὺ (or οἱ) δ᾽ ἔριδος... ἐπ᾽ ἀλλήλοισι 
τάνυσσαν. In the first τώ means Zeus and Poseidon, who 
stretch the rope of strife over both parties; in the second τοί 
means Trojans and Achaians who strain the rope of war for (or 
against) one another. The metaphor in the first case will be 
taken from the idea that the gods govern the movements of the 
battle by invisible cords fastened to the two armies, at which 
they pull alternately (ἐπαλλάξαντες, or was there not very 
likely a variant ἐπαλλάξαντε, of which no mention has come 
down to us? The hiatus, though legitimate, would probably 
be fatal to it), For this I may refer to my note on H 102. In 
the second case the metaphor is from the “tug of war”; the 
two armies are regarded as fastening themselves together by 
a rope, and pulling one another backwards and forwards. Of 
these the first is undoubtedly to be preferred, for in all the 
other passages where the mention of πείρατα in this and similar 
phrases occurs it is always in connexion with divine interference. 

A third explanation is that of Aristarchos, which may be 
briefly explained, though it is not likely to be accepted; as 
there remains a doubt as to how he came to take ἀλλήλοισι 
and ἀμφοτέροισι as virtually identical in meaning. From the 
Schol. of Didymos on 359 it would seem that he explained 
“Zeus and Poseidon knotted together (ἐπαλλάξαντες) a rope of 
strife and a rope of war, and drew them tight over one another” 
(ἀλλήλοισι), or “over both ends” (ἀμφοτέροισι) (ὥσπερ οἱ τὰ 
ἅμματα ποιοῦντες, τόδε ἐπὶ τόδε, as Didymos explains). There 
is however the Scholion of Aristonikos, on 359, ἡ διπλῆ, ὅτι πα- 
parrAnyopet’, δύο πέρατα ὑποτιθέμενος, ἕτερον μὲν ἔριδος, ἕτερον 
δὲ πολέμου, ἐξαπτόμενα κατ᾽ ἀμφοτέρων τῶν στρατευμάτων. 
This would indicate that while explaining ἀλλήλοισι as above, 
he took ἀμφοτέροισι to mean “over both armies,’ a far more 
natural explanation, if not the only possible one. In a question 
of interpretation we are probably right in preferring Aristo- 


nikos to Didymos, when, as here, there is a discrepancy 
between them. 


1 This word is not given in the last edition of L. and 5. It means of course 
** speaks metaphorically.” 


160 THE JOURNAL OF PHILOLOGY. 


It was only after the conclusion of the above enquiry that it 
occurred to me to refer to the Codex Townleianus in the 
British Museum; as the Scholia V. are copied from this, there 
was some chance of finding the missing word giving the reading 
of Aristarchos. It was with no small gratification that I found 
the required proof of the correctness of my conjecture. The 
MS. says quite plainly ᾿Αρίσταρχος τὼ δ᾽ ἔριδος ᾿Αριστοφάνης 
τοί. It is earnestly to be hoped that the Oxford edition of 
these important Scholia will not be much longer delayed. 


WALTER LEAF. 


SERVIUS ON AENEID ix. 289. 


“Audentior” ut “sapientior”, ab appellatione, non a par- 
ticipio futura cum: Terentius...“ ignoscentior”’. 


For the corrupt futwra cwm, which has puzzled the editors, 
I propose to read figuratum. 


H. NETTLESHIP. 


NONIANA QUAEDAM. 


Nonivs I—III. 


[Mss. containing the whole : 


F = Florence xiv, 1 (1—111) (9th or 10th cent.). 
H = Harleian 2719 (9th or 10th cent.). 

L = Leyden Voss F 73 (9th or 10th cent.). 

P = Paris 7667 (1 and u—p. 140) (10th cent.). 

V = Wolfenbiittel Gud. 96 (10th cent.). 


Extract Mss. : 
C = Paris 7666 (10th cent.). 
D = Paris 7665 (10th cent.). 
M = Montpellier 212 (10th cent.). 
O = Oxford Bodleian Canon. Lat. 279 (11th cent.). 
X = Leyden Voss 116 (11th cent.). | 


11 3 (Afranius 378 R.). 


Si possent homines delenimentis capi, 
Omnes haberent nunc amatores anus. 
Aetas et corpus tenerum et morigeratio, 
Haec sunt uenena formosarum mulrerum. 
Mala aetas nulla delenimenta inuentt. 


The Florence Ms. has in for δὲ in the first line, which tends 
to support Schoppe’s conjecture nz. Sz might easily be due to 
a correction of nz to nisz (nt with sz superscribed), but it is 
difficult to see how si can have been corrupted to im, though 
I notice that Schoell writes δὲ for the Mss. in Plaut. Rud. 
Prol, 22. 


Journal of Philology. vou. xvt. 1l 


162 THE JOURNAL OF PHILOLOGY. 


ib. 18 (Novius 37 R.). 
Operaeque actor, cantor, cursor, senium sonticum. 


opereque F X. opere quae HL PVC. 
auctor V*. actor FH LP V’CX. 


It is possible that V* preserves a trace of the original read- 
ing, which may have run 


Operaeque actor, awctor, cantor, cursor, seniwm sdénticum. 


This reading completes the metre, and it may be urged in 
its defence, (1) that actor and auctor are constantly found in 
juxta-position, (2) that Novius seems to have had a peculiar 
penchant for alliterating pairs of words; v. the fragments of 
Ribbeck passim. 


111 2 (Pacuvius 301 R.). 


Metus, egestas, maeror, senium exiliumque et senectus. 
et gestas C. 
For senectus Ribbeck reads desertitas, a ἅπαξ λεγόμενον. 
I would suggest grandaeuitas, of which senectus may well be a 
gloss, ousting the original word. Grandaeuwitas is used else- 
where by Pacuvius and occurs twice in the fragments of Attius, 
v. Nonius 116. 12—20. 
The line would then run | 
Métus, egestas, maéror, senium, extliumque et grandaé- 
urtas. 


v 16 (Plaut. Aul, 422 R.). 
Ita fustibus sum mollior magis quam ullus cinaedus. 


submollior F HL PV*CX. submolior V*. sum mollior Mss. 
Plaut. 

maser magis quam Plaut. BD. magis miser quam F J. 

mulius F* HLPCX. ullus ΕἾ V with the Mss. of Plaut. 

It is very tempting to suggest 


Ita fistibus sum méllior quam millus cinaedus. 


mullus cinaedus has a peculiarly Plautine ring, especially 
coming from the lips of a cook. An etymological connection 


NONIANA QUAEDAM. 163 


between mullus and mollis may possibly be intended, and the 
confusion in the Plautine Mss. (due perhaps to submollior) 
seems to point to some doubt about the reading. 


γι 16. Tilicere est proprie illaqueare, Naeuius, Lycurgo 
(Naev. 29 ἢ... 
alis 
Sublimen alios saltus illicite, ubr 
Bipedes uolucres lino linquant lumina. 


Inlicere i CX MO. illicere FH L?> PV Ὁ. 

inlaqueare L'CX DMO. tllaqueare FHL’ PV. 
subumem L’. 

aliis saltos ἘΝ. 

inlicite CX. inlicitae L’. <illicite FHV. illicitae L’ P. 
ubipedes V". 

linqguant H*PVCX. lgquant F ΗΠ. 


I would suggest 


Inlicire est proprie inlaqueare, Naeuius, Lycurgo, 


Sublimen altos sdltus inlictte, ubi 
Bipedés uolantes lino linquant limina. 


In support of this reading the following considerations may 
be urged, (1) That Nonius explains the word as meaning inla- 
queare, (2) That «nlicire is a strictly natural formation (ef. 
tnretire) meaning to surround with threads, nooses or springes, 
the linum of the next line, (3) That it restores both metre 
and sense in the first of the two lines. Professor Havet re- 
gards alis and alios as a dittography of altos, a view with which 
I entirely agree, and which is strongly supported by the variants 
of F. The metre of the second line is restored by the easy 
change of wolucres to wolantes. The corruption is a very 
natural one and occurs as early as the capital Mss. of Vergil, 
e.g. Aen. VI 728 R. gives wolucrum for wolantum. 


ib. 20. Pellices a graeco uocabulo significantiam sapientes 
inflecam putant, quasi πάλλαξ, hoc est ut παλλακίς. Quod si 
hoc non est, wana compositio nominis uideri potest. 
11—2 


164 THE JOURNAL OF PHILOLOGY. 


This is Quicherat’s reading. The important Mss. variations 
are 

Peliceos, all. 

quasi pellex (sic) all, except H mg., which has pallez. 

ut pallais H7LPVCDMO. ut pallaces X. ut pellacis 
et 6 

conpositio F H 1 Ὁ. 

hominis all. 


I would read 


Pelices a graeco uocabulo sapientes inflecam putant. Quod 
si hoc non est wana conpositio hominis uideri potest. 


Nonius is suggesting two alternative derivations of pelices 
either from the Greek, hoc est ut παλλακίς, or from the Latin, 
quasi pellex, 1.6. wana conpositio hominis. For this use of wana 
ef. Non. 417. 1, where wanwm is glossed by insidioswm, subdolum. 
Hoc est ut pallacis and quasi pellex are two marginal glosses 
which have found their way into the text, hoc est ut pallacis 
referring to the first, quasi pellex to the second derivation. 


ib. 32 (Attius 382 R.). 


Sed mémet caluor. uds istum tussi dcius 
Abstréhite. 


Ribbeck and Quicherat insert wt before iussi, but the change 
is not really necessary, cf. Aen. vit 156, festinant vussi. 
DMO have extraite. 


vil 8. Frigere est...sussilire cum sono uel erigi et exilire, 


exilire H?PV. exitari FH’ LC. excitari X. exagitari 
DMO. 

The various readings of the Mss. seem to point to an original 
excitart, corrupted to exitars (F H' LOC), corrected rightly to 
excitari (X), wrongly to exilire (ΗΠ PV). ψαστίαγὶ in DMO 
represents excitart corrupted to exatari, and corrected by the 
superscription of cz. 


ΧΙ 27 (Caecilius 115 R.). 


Suppilatum est aurum atque ornamenta omnia. 


NONIANA QUAEDAM. 165 
So Quicherat and Ribbeck. The Mss. offer the following 


variations : 
subpilatum FHL P*. suppilatum P? V. 
est eum all. 
ormenta F H* L. 


Does not subpilatum est eum represent subpilat uestem, cor- 
rupted to subpilatu estem, and then corrected to subpilati est 
eum? So 179. 17 for funestat ueste tonsu F has funestatu é 
(est) et tonsu. 

The line may well have run 


Subptlat uestem, aurum dtque ornamenta dmnia. 


This reading satisfies the metre, and awrwm is constantly 
‘used by the comic poets in combination with westem, when 
speaking of a lady’s belongings. 

ΧχΙ 26 (Turp. 104 R.). 

Turpilius, Leucadia. 

Ei pert! uaden ut osculatur cariem? non illam haec pudent? 


So Quicherat. Ribbeck reads num hilum illa haec pudet. 
The Mss. have leucadia ciperi all. 

non H’. num F WLP YV. 

illum illa ec pudet FH LP γ᾽. haec pudet V’. 

I would suggest - 


Périi! Viden ut dsculatur cdriem? non illum écpudet? 


The cz before perit seems to be nothing more than a ditto- 
graphy of a, the a of leucadia being corrupted to ci, and cor- 
rected by the superscription of a. So again zJlum illa is merely 
a dittography, illum corrected to illam or vice versa. 


Xxv 17. Silones superciliis prominentibus dicti, significa- 
tione manifesta, Varro γνῶθι σεαυτὸν (207 B). Nonne non unum 
seribunt esse grandibus superciliis silonem quadratum quod 
Silenus hirsutis supercilis fingeretur. 

The last five words though found in all the Mss. are gene- 
rally, and no doubt rightly, considered to be a gloss. non unwm 
again, which can hardly be genuine, has been altered by Pro- 
fessor Buecheler to hominem, by Professor Nettleship to Silenum. 


166 THE JOURNAL OF PHILOLOGY. 


Professor Nettleship would also read czlones and cilonem for 
silones and silonem, arguing that silones would not mean super- 
ciliis prominentibus. But would czlones mean this? If connected 
with ciliwm it would surely mean not superciliis prominentibus, 
but ciluis prominentibus, i.e. with projecting eyelids. It seems 
however to admit of doubt whether czlones is really connected 
with cilium. The following authorities may be cited to determine 
the meaning of the word: Paul. Diac. 43. 10 Cilo cui frons est 
eminentior, ac dextra sinistraque uelut recisa uidetur; Charis. 102. 
1 K. Cilones quorum capita oblonga et compressa sunt; Placidus 
25. 5 D. Cilones quorum capita oblonga ; Gloss. Labb. Cillo 
προκέφαλος, μακροκέφαλος, and Cilones gofoi; Caper de orthog. 
2242. 20 P. Cilo est angusto capite cur hoc contigit in partu ; 
Velius Longus orth. 2234. 19 P. Cilones homines uocantur 
angustt capitis et oblongi, and so Cassiodorus 2287, 33 P. 

These citations seem to shew that cilo means, not a man 
with projecting eyebrows, but rather a man with a pointed, 
sugar-loaf, cut-away head, what is colloquially known as a 
“grocer’s forehead.” I cannot help believing that the clue to 
the meaning of the word is to be found in the explanation 
recisa preserved by Paulus, cut frons dextra sinistraque uelut 
“recisa” widetur, and it is to be noticed that Varro and Ovid 
give a similar explanation of ancile, Varro L. L. vi 89 Ancilia 
dicta ab ambecisu, Ov. Fast. τι 377 Idque ancile uocat quod ab 
omni parte recisum est. I would then retain the Mss. silones and 
silonem; cf. Plin. XI 37, 158 cognomina simorum silonum, Gloss. 
Labb, Silo, simus. The word seems indeed to have a peculiar 
appropriateness here, as “snubnosedness” is a recognized pro- 
perty of Silenus, cf. Lucr. Iv 1169 Simula silena ac satura est. 
The explanation superciliis prominentibus seems, as often, to 
have found its way into the text from the margin, the word being 
significatione manifesta and requiring no further elucidation, 
cf. Non. 36, 14 Hmungi ex manifesta significatione manat. If 
the significatio is manifesta is it not futile to mention it? For 
non unum I would write nanwm. The open ὦ has been written 
u, as often, and then nunum has been corrected to fiwnum (non 
unum). The quotation will then scan as a line and a half of 
trochaic tetrameters, and the whole passage will run 


NONIANA QUAEDAM. 167 


Silones dicti significatione manifesta, Varro γνῶθι σεαυτόν. 


Noénne nanum scribunt esse grdndibus supércilics, 
Stlonem quadrdtum, 


a description reminding one closely of the picture given of 
Labrax, Plaut. Rud. 317 R. 


Recaluom ac silonem senem, statutum, wentriosum, 
Tortis supercilus, ce. 
XxviI 24 (Varro Sat. Men. 432 B). 
de lacte ac cera tarentina. 


Read terentina, the old spelling, of which a trace is preserved 
in V (cetera retina Ν᾽, cetera rentina V*), cf. Macrob. 111 18, 13 
Nuc terentina dicitur...De qua in libro Fauorint. Quod quidam 
Tarentinas oues uel nuces dicunt, quae sunt terentinae, a tereno, 
quod est Sabinorum lingua molle. 


xxvii 12. Varro Sesqueulixes. Quo cum ire uellemus, 
obuius flare. Uli corport aerinas compedes impositas urdeo, 


This is the reading of H’P V. H"* has quoque ire. ἘΠῚ, read 
sesque ire, omitting ulixes quocum. F* has quocumque ire, which 
is clearly right. conpedesFHL*. «inpositas L’. 


Read (with Buecheler, Varro 473), 


Varro, Sesqueulize; Quocumque tre uellemus, obwius flare. 
Ubi corpori aerinas conpedes inpositas uideo. 


ib. 18. Coagulum a coagendo, quod est colligendo. 
Varro. Hoc continet coagulum conuiura. 
Read colligando: so uolantes and uolentes are constantly 
confused. 
XXXIV 23 (Plaut. Mil. Glor. 4) 
Praestringat oculorum aciem in acie hostibus. 
aciem in aciem F H’ L. 


Perhaps the simplest way of restoring the metre would be 
to substitute fostibus for hostibus, cf. Paul. 84, 5 antiqua dicebant 


168 THE JOURNAL OF PHILOLOGY. 


Jostem pro hoste. So Professor Nettleship has suggested fordeo 
for hordeo, Plaut. Asin. 706 R. 


Demam hercle iam de hordeo tolutim ni badizas., 


xLviml 10. Varro, ταφῇ Μενίππου. Ut antiqui nostri in 
domibus latericiis paululum modo lapidibus suffundatis. 


Ut should probably be omitted. The Mss. give 


tafe menippu antiqu ἘΠ H*. tantiqui H?L PV. santiqu 
(sic) ἘΦ ἘΠ nostri in rightly, nostrun ἘΝ Η Γ΄, nostrum H’ L* ΡΥ, 
laynditus all. 


ib. 21. Parochus: a graeco tractum est nomen, quod uehicula 
praebeat; ὀχήματα enim graece latine uehicula appellantur. Et 
est officii genus quod administrantibus paret. Varro, Sesque- 
εἶχε; Hic enim omnia erat ; idem sacerdos, praetor, parochus ; 
denique idem senatus, idem populi caput. 


The Mss. have the following variations. 


Paracos all. 

praeeat or preeat all. 

hoc schemate, sechemate, or scemate, all. 

ertt H". 

paratos F H* L. paracos Η PV. 

populus all. 

Kaput (sic) F. kaput H’. 

Editors are agreed in changing praeeat to praebeat, but 
praeeat seems clearly right. The explanation given shews that 
Nonius derives the word not from παρέχω, but from παρὰ 
ὄχος (unless the passage from quod—appellantur is a marginal 
gloss); and it is surely more pointed and therefore more 
Varronian to say, that a man is his own outrider, than that he 
is his own postmaster. 

The erit of Η is a mere slip, corrected by the same hand. 
Caput is probably nothing but a marginal gloss on proboscis, 
the next lemma but one, which has found its way into the text. 
In F it is written with a capital and looks like a later addition. 


XLIx 6. Varro sexagest (490 B), Inuenisse se cum dormire 


NONIANA QUAEDAM. 169 


coepisset tam glaber quam socrates, esse factum ericium e pilis 
albis cum proboscide. 


claber ¥" L. 
socratis all. 
caluum esse factum ericium all. 


The passage occurs again 106,13. There the Mss. all give 
socrates galbam esse factum. 

I would suggest tam glaber quam socratis calua, “As smooth 
as Socrates’ bald-pate,” which accounts fairly well for the 
variations of the Mss. and adds greatly to the force of the 
passage. 


Lvil1l. Agilem, celerem, ab agendo, Sisenna Hist. 11. 
Agilem dari facilemque uictoriam neque fossas aut bellum 
remoratum. 


So the Mss. Editors alter to wallum remoraturum, but the 
text will construe as it stands, and surely gives a more vigorous 
sense. A general addressing his soldiers is supposed to say, 
* You have before you a victory in the open field (agilem and 
facilem are meant to suggest agendo and faciendo), not siege- 
works and lingering hostilities.” 


ΙΧ 5. Nefarw proprietatem Varro patefacit a farre, quod 
adoreum est, in quo scelerati uti non debeant ; non triticum sed 
far. Hoc quoque idem adsignificat, quod qui indigni sunt qua 
uiuant, nefarw wocantur. 

adhorreum Li. adhoreum L’?. adoreum FHP VD*MO. 
adreum C Ὁ X. 

qu FHL? D?MO. in quo HL'PVCD*X. 

[Ὁ is very tempting to retain the adhorreum of L, and read 
quod adoreum est, id est adhorreum, quo. The im may fairly 
represent the lost τά, and Varro may easily have intended to 


connect the word with ad horreo (quo scelerati utr non debeant); 
non triticum sed far seems a mere gloss. 


LX1 19. Quod est inter duos sulcos elata terra dicitur porca, 
quod ea seges frumentum porricit. 


The Mss. of Nonius give secus, those of Varro seges. For 


170 THE JOURNAL OF PHILOLOGY. 


porricit, the Mss. of Nonius give porrigit, those of Varro porrigit 
or porrigat. Festus and Paulus 238 and 239 say Porcas quod 
porrigant frumentum. 

Seges and porricit are both very difficult to explain. Should 
we not retain secus and read quod ea secus frumentum porrigitur, 
“because it stretches side by side with the rows of corn”? 


(Fest. and Paul. 218 and 219 quod porcant aquam frumentis 
nocere.) 


LxIl 12. «dda enim graeci ligna dicunt ut Homerus. 


The reference seems to be to Homer, Hymn Merc. 112, 
Πολλὰ δὲ κάγκανα κᾶλα κατουδάιῳ ἐνὶ βόθρῳ 
Οὖλα λαβὼν ἐπέθηκεν ἐπηετανά. 
Lx 5. Hnnius lib. xvi. gruma derigere dixit degrumare 
ferrum. Lucilius lib. 11. 
Viamque 
Degrumabis uti castris mensor facit olim. 
derigere H? LP. dirigere F H’ V. 
degrumart FHL P*V. degrumare P*. degrumauis FH LY. 
grumauis P. ut castris FH LP V. 


I would read 
gruma derigere dixit degrumari ferrum. 


Lue. lib. 111. Viamque 
Degrumatus uti, &e. 


retaining the deponent form in each case. All the Mss. but 


P' give degrumari, and degrumatus is as near as degrumabis to 
the Mss. degrumauis. 


LXV 7. Cicero Alcionibus, 
Hune genuit claris delapsus ab astris, 
Praeuius aurorae, solis noctisque satelles. 


Read Lucifer hunc genwt, &c. 

The reference is clearly to Lucifer (Hesper-Phosphor, sols 
noctisque satelles), who was father of Ceyx, the husband of 
Alcyone. 

Cf. Ov. Met. x1 271 Lucifero genitore satus. 

ib. 346 Lucifero genitus. 


ee 
—_—— ~~ 


NONIANA QUAEDAM., 171 


LXxIU 30. Atri dies dicuntur quos nunc nefastos uel posteros 
dicunt. 

For wposteros Bongars suggested «mprosperos, Quicherat 
praeposteros. Posteros is however right. The day following 
the Kalends, Nones and Ides was an unlucky day, so that 
posterus dies came to mean dies nefastus, cf. Macrob. I 16. 24 


- Pontifices statuisse postridie omnis Kalendas, Nonas, Idus 


atros dies habendos, ib. 23 post sacrificvcum die postero celebratum 
male cessisse conflictum. 


LXXv 21. Abscondidit pro abscondit. 
So Β΄ alone rightly, other Mss. have abscondit pro abscon- 
didit or abscondit pro abscondit. 


LXXvit 15. 

Betere, id est tre, Varro...betere cusstt. 

Pacwuius, vos...betite. 

Idem...prohibet betere. 

Buecheler (Varro 553), and Ribbeck in the two passages 


of Pacuvius, read baetere, but it may be doubted whether the 
Mss. here do not rather favour the spelling betere, which is read 


- by all the Mss., including in one instance the Ambrosian, in the 


two passages where the word occurs in Plautus. 

In the lemma here the readings are 

betere WLP V. baetere FH*CDMO. 

In the passage from Varro 

baetere FL? CD MO. betere H? PV. Obretere Ἐ“. baretere 
H'.  baeterat L’. 

In the first passage from Pacuvius 

bibite FH LP’ V. bibibite P’. 

In the second betere FH LP V. 

In the lemma the weight of authority is rather in favour of 
betere. Baetere is supported by F and the extract Mss., but has 
against it the powerful group H* P V supported by L. In the 
passage from Varro baetere is supported by F* L and the extract 


6 
Mss., betere again by Η PV. The britere οἵ ἘΠ᾿ points to bitere 


(bitere corrected to betere), the baretere of H* is a misunder- 
ae ; 
standing of bretere in F. 


172 THE JOURNAL OF PHILOLOGY. 


In the first passage from Varro bibite points clearly to 
bitite, the bibibite of P looking like a recorrection to bitite. In 
the other passage the whole Mss. authority is in favour of betere. 
The general result seems to point to bitere corrected to betere 
and baetere. It must be remembered that for books 1—im H* 
can no longer be regarded as an independent authority. 


LSxx Ot 
Bellosum, bellicosum, Coelius, Tum bellum suscitare conari 
contra bellosum genus. 


So Quicherat: the Mss. read 


Tantum bellum siscitare cénarit aduersdrios 
Contra bellosim genus, 


and while Coelius is supported by FL and the extract Mss., 
caecitlius is given by ΗἾ PV. Editors seem generally agreed 
that the language is not that of a comedian, and adopt the 
reading coelius. But it is surely singular that a passage from 
a historian should be in perfect trochaic metre, and it must not 
be forgotten that, when H’P V are opposed to FL, they are 
right in some seven cases out of nine. On the whole it seems ~ 
safer to restore the passage to Caecilius. The language is no 
doubt above the level of ordinary comedy, but it is perhaps not 
incompatible with the recognised gravitas of Caecilius. 


LXxx1 10. Lwuciliws Xxx. 
Clauda una est pedibus cariosis mensa liboni. 
So Quicherat. Lucian Mueller reads 
Plauta una est pedibus cariosis mers Lnbiteinar 
or menstrua libans, 


supposing the reference to be to some old hag of the day. 
Pedibus cariosis however seems to suggest a table, or some 
similar article of furniture, rather than a human being. 


The Mss. readings are as follows : 
plauda FLP’?V. plaun P’. 


mensu all, 


zabino F*. libano ἘΠ H? LP YV. 


NONIANA QUAEDAM. 173 
Clauda may be right, but the plawn of P* seems to point to 


n 
an original plauda, i.e. plauda corrected to planda or plancla, 
(cl and d being for all practical purposes interchangeable). 
Plancla would be a diminutive of planca a plank or board, a 
word which is not found in extant Latin literature earlier than 
Palladius. It was however in use in old Latin as we learn from 
Paul. 231. 3 Plancae tabulae planae. 

Starting from mensu iabino the reading of Ἐπ, which is 
regularly to be preferred when it stands alone, mensula suggests 
itself at once. Bino seems to represent the name of some 
proverbial or notorious pauper, the Jrus temporuwm. I would 
suggest mant, m and b being frequently interchanged; cf. Persius 
ΥΙ 56 praesto est miht Manius heres. So in explaining the 
origin of the proverb multi Mani Ariciae, Festus 8. v, says, 
Sinnius Capito ait turpes et deformes significare. 

The line would then run 


Plancla (clauda?) una est pedibus cariosis mensula Mani. 


LXXXIV 4. Conspicillum unde conspicere possis, Plautus 
Medico. 


In conspicillo obseruabam pallium adseruabam. 


So Quicherat. The Mss. give conspicilium without ex- 
ception, and conspicilio, except L’, which has conspilio. The 
majority of Mss. have in conspicilio adseruabam pallium obser- 
uabam, FL omit adseruabam pallium, O reads Plautus obser- 
uabam medico in consp. &c. These Mss. varieties seem to 
suggest that in the original Ms. obseruabam was omitted, and 
then restored from the margin, ousting adseruabam pallium in 
I" L, and being inserted in the wrong place in Ὁ. Perhaps 
the simplest way of restoring metre and sense may be 


Hine in conspicillo obseruabam ddseruabam pédllium. 
ib. 6. Colustra lac nouwum in mammis, Luc. lib. vu. 
Hiberam insulam fomento omnicolore colustra. 


Columnum F*°. Columnam L'. Colustra lumnam cet. 
lacconere guumere mammis F*. lacchonere iunmi mammis 
cet. 


174 THE JOURNAL OF PHILOLOGY. 


Bera ¥*. Beram L*. Hiberam cet. 

imsuam V". 

Colustra F°. colustra F’ H?L PV. 

The extract Mss. omit columnum altogether, and give simply 
Colustra Laberius. 

I have previously argued that we have here a combination 
of two glosses, and this view is strongly corroborated by the 
evidence of F* and the extract Mss. Colwmnum should probably 
be read with F*. It may be an old genitive plural either from 
columen or columnum, and Nonius could hardly quote columnam 
as honeste or noue dictum. Lacconere gium looks like lacunar 
regium, a gloss introduced from the margin and, like many of 
the explanations in this book, not to be attributed to Nonius. 
The rest of the line seems very uncertain, but conjectures should 
start from bera, the reading of ἘΠ, not from hibera or hiberam, 
which looks like an obvious correction. 


ΧΟΙΧ 24, Desabulare perfodere, &c. 

Desubulare FH LP V, all the extract Mss. have desuberare. 

Is desuberare a further corruption of desubulare, or is it not 
rather the relic of a lost lemma, containing, like the one follow- 
ing and the two immediately preceding, a quotation from the 
Satires of Varro? Desuberare would mean “to remove the 
bark,” to abrade, just as discobinare immediately preceding 
means “to take off the sawdust,” to graze. This suits the 
explanation perfodere, and is exactly parallel to our English 
slang-phrase “to bark your shins.” 


C4. Ht alio loco idem (Varro) (588 B). 
Quid mihi somnus si dormitio tollitur. 
Quo mihi somnos. Buecheler. 


Quo mihi somno FH LP V with the extract Mss., except 
C, which alone reads somnos. 
Will not somno stand, on the authority of Verg. Aen. Iv 99, 


quo nunc certamine tanto? 


and Hor. Ep. 1 5. 12, 
Quo mihi fortuna si non conceditur uti? 


which latter is precisely similar in form to the present passage. 


NONIANA QUAEDAM. 175 


It may be noticed that the Mss. of Horace vary between 
quid and quo, fortuna and fortunas. 


ΟΥ̓ 4. Varro Bimarco (47 B.). 
Ipsum propter uia liberti semratrate. exequiantur. 


So Buecheler with the Mss. Vix libertz however seems 
somewhat suspicious. Should we not read ww sex liberti 
(VIXVILIBERTI)? The change is a slight one, and adds 
much to the picture. 

ib. 28. Laberius in Centonarto. (Lab. 24 R.) 


Quare tam arduum 
Ascendas? an concupisti eugium scindere. 


The simplest way of restoring the metre seems to be 


Quare tam drduum 
Ascendas? an céncupiwstc eigium mi scindere, 


or etigium rescindere. 


ox 6. (Pomp. 176 R.). 
Verum ili ualent qui luctantur leonibus. 


Ribbeck we luctantur cum. 
I would prefer 


Vérum illi [ualidé] ualent qui Victantur lednibus. 
The phrase walide ualere occurs Plaut. Pers. 426, and the 
double alliteration seems to add to the vigour of the line. 
cxvi1 15. (Pomp. 145 R.). 7 
gaudet si cur quid bona. 
sicut Ἐ5 alone, rightly. sicut ΕἸ H LP V with extract Mss. 
quid extract Mss. rightly. qu FHLPYV. 


cx1x 3. Deum H* isa mere copyist’s blunder, F has deam 
with the other Mss. 


ΟΧΧΙ 12. Hostire est conprimere, recedere, dictum ab hostia. 
Hostire, offendere, cedere. 
conprimere, cedere ἘΠ᾿ alone. 


I feel no doubt that we should read here 


176 THE JOURNAL OF PHILOLOGY. 


Hostire est conprimere, caedere, &c. 
Hostire, offendere, 40. 


Caedere by dittography of re has become recedere, caedere 
has then been suggested in the margin, and the correction has 
found its way into the wrong place of the text. F* has it in the 
right place in the first line, but has not deleted it in the second. 


CXXVI 31 (Laberius 65 R.). 
Aequum animum indigna iniquat contumelia. 
So Ribbeck. The Mss. have 
indignat F7 LD MX. indignant ἘΠῸ. indignanti H’ P V. 
I would suggest 


Aéquum animum indigndtione iniquat contumélia. 


CXXVII 17 (Caecilius 136 R.). 


Inbera essem vamdiu 
Habwissem ingenio si sto amatores maha. 


tamdiu siston abutores mihi ἘΝῚ, (cet. om.). 
iamdiu siston habuissem ingenio ἘΝ. 
tamdiu habuissem ingenio si ston H’ P V. 
Read 
Si istdc habuissem ingénio amatorés mehe. 


The n of siston probably represents the h of habwissem, which 
the reading of ΕἾΤ, (abutores 1.6. abuissem amatores) shews to 
have disappeared from the original Ms. 


CXxIx 25 (Ennius 327 R.). 
inimicitiam atque amicitiam in frontem promptam gero. 
in fronte ἘΝ ὧν frontem cet. 


cxxxvil 6, Mercatis pro mercatibus, Sall. Hist. V. Ceteri 
negotia sequebantur familiaria legatorum aut tribunorum et pars 
sua, commeatibus mercatis. 

mercatibus H? V. mercantibus cet. 

Is mercatis used in a passive sense, the explanation being 
altogether wrong, or should we accept the explanation mercan- 
tibus and read commeantibus mercatis ? 


NONIANA QUAEDAM. 177 


ΟΙ, 17. £t eo plectuntur poetae quam suo uitio saepius 
Ductabilitate nimia uestra aut perperitudine. 


Quicherat alters ductabilitate to ducti wilitate, but ducta- 
bilitate is surely right. 


Dictabilitate dnimi nimia uéstra aut perperitidine 
restores the metre and gives excellent sense. 


cui 31. Fateor, sed cur propter te pigrem haec aut huius 
dubitem parcere capiti. 

pateor FL. proferre Ἐπ, proterre H*. propterre ἘΠῚ,. 

proferre seems clearly right. In all the Mss., more especially 
L, ὁ is a constant corruption of f, and in one instance at least 
L reads scaptis for scafis. The line may either be scanned as it 
stands as an anapaestic octonarius, or may be read as a trochaic 
tetrameter by transposing pigrem and haec with Ribbeck. 


Fateor: sed cur préferre haec pigrem αὐ huius dubitem parcere 
Capiti. 

cLiv 3. Prouidere, mouere, ecferre. 

Ὑ' has prouedere, which tells somewhat in favour of my 
suggestion progredere. 


ib. 5. Puellascere, ecfeminari uel eutrescere, Varro Bais. 
Quod non solum innubae fiunt communis, sed etiam ueteres puellas- 
cunt, et multt puert puellascunt. 

ueteres puellascunt is read by all the Mss. Buecheler (frag. 
44) reads repuellascunt. I would prefer puerascunt, We have 
a gradual declension. The inubae become communes, the 
ueteres, puert, the puert, puellae, 


ib, 14 (Pomponius 168 R.). 
Quidam apud forum praesente testibus mihi uendidit. 


So Ribbeck. The Mss. have quidem for quidam, Should 
we read Quine, n and d being often interchanged ? 


CLVI 25. Varro Synephebo (514 B,). Verere ne manu non 
mittat cum tot romae mendicari honestos audissemus, Dare possis 
mittet quod si non mittet fugies si me audies, Cum tempus 
reuocat ea praecox est fuga. 

All the Mss. give cum toctro medicarios. 

Journal of Philology. you. xvt. 12 


178 THE JOURNAL OF PHILOLOGY. 


Buecheler reads cum tot romae uicarios honestissimos dare 
possis, mittet, &c. which gives excellent sense but is rather wide of 
the Mss. I would suggest mendicarier. 

It may be noticed that cxxxvil. 22, H’? V alone have men- 
dicarier, mendicaries F LPC. mendicares D. mendicaes M. 
The change from mendicaries to mendicarios is a very easy one. 


CLVII 22 Varro (324 B.). <wtaque st plures dies inter medicr 
discessum et aduentum pollinctoris interfuerunt. . 

ἘΠ alone medict. medicis L*. medicos ἘΠ H’ L? V. 

medici seems to have been corrupted to medicos through the 
influence of the neighbouring inter, though it is possible as 
Prof. Havet suggests that medicos represents medicor: 1.6. medt- 
corum. 

CLxIII 3. Pingue positum pro imperito et inepto, Cicero 
Acad. 1111. Quod ipsi Antiocho pingue widetur et sibt ipsum 
contrarium. 7 

So Quicherat. For tmperito the Mss. have inpedito or in- 
pedimento. | 

Inpedito is surely preferable. 


CLXXI 24 (Varro 1 B.). 


Ita sublimis speribus 
Tactato wuolitantis altos nitens trudito. 


The Mss. read tactato nominatuo (nominatiuo V*) uolitantis, 
for which Buecheler suggests : 
iactato homines at uolitantis, 


it seems however perhaps more probable that Quicherat is right 
in regarding nominatiuo as a marginal note on sublimis.. Has 
not animos fallen out after the antis of uolitantis? The pas- 
sage will then run : : 
lta sublimis spéribus 
Idctato uoliténtis animos dltos nitens truvdito, 
altos being proleptic after trudito. The idea recalls Pindar, 
Pythian Ode vitr 89—91 
| μεγάλας 

ἐξ ἐλπίδος πέταται 
ὑποπτέροις ἀνορέαις. 


NONIANA QUAEDAM. 179 


CLXXxII 6 (Attius 176 R.). 

Quorum crudelitatem nunquam wulla explet satias san- 
guinis. 

The excellent authority F* has cruditatem (crudelitatem 
with the el dotted out), making a trochaic tetrameter, and this 
is surely right, as there seems a special appropriateness in the 
combination of cruditatem and satias. 

CLXXIv 34 (Afranius 177 R.). 

Vide ut puellam curent, conforment cube. 

Inde ut Ribbeck. The Mss. give 

uidead FL. wide at H’ V. 

Is not the simplest correction 

Vidednt puellam ctérent, conformént ube, 
i.e. tell them to see that they have the bride ready. 
CLXXvi1I 18 (Pomponius 23 R.). i 
Clandestino tacitus taxim perspectaurt per cauum. ᾿ς 


The excellent F* alone has cauwum (cautum with the ὁ dotted 
out). all the rest cautum. 


ib. 22 (Pomp. 179 R.). 

Iam istam caluam colafis comminussem testatim tibr. 

Sok. EF’ LV omit caluam, -Is not the unusual rhythm 
purposely introduced, like the alliteration, to add to the energy 
and distinctness of the words ? 

CLXXIXx 17 (Attius 86 R.). 

Sed quaenam haec est mulier funesta ueste tonsu lugubri. 

funesta V’ alone. funeste V*. funestat cet. 


CLXXXVII 8. demptionem uel decerptionem. 
decerptionem Ἐδ rightly. decreptionem cet. 


cxcll 3 (Att. 322 R.). 
Cum Scamandriam undam salso sanctam obtext sanguine. 


So Quicherat and Ribbeck (the Mss. have obtexus), but 
surely obteat is a singular word to use of dyeing a river with 
blood. Should we read obtinai from obtinguo? The word does 

12—2 


oa: THE JOURNAL OF PHILOLOGY. 


not seem to be found elsewhere, but obfuco and similar com- 
pounds are frequent. Obtinai is even nearer to the Mss. than 
obteat as the lost n may be represented by the wu of obtexus. 


ib. 33 (Pac. 397 R.). 
Postquam calamitas plures annos aruas caluitur. 


I would read conplures (€plures) for the sake of the allitera- 
tion, 


cc 28. Praeterea Tuscis resoluta crine capillus. 


For Tuscis the Mss. have tusis, which is no doubt a corrup- 
tion of fusis. We should read apparently 


Praeterea fusus resoluta crine capillus. 


ΟΠ 30 (Pac. 382 R.). 
Voce suppressa, mutato fronte, wultu turbido. 


For mutato the Mss. have innato or innata. Many emenda- 
tions have been offered, but no one seems to have suggested 
minato, which makes excellent sense, and requires the slightest 
possible alteration of the Mss. 


cov 5. Transgressus fluwiam quae secundum H erculaneum 
ad mare pertinebat. 

pertinebat F°H V. perfinebat F" L. 

Is perfinebat a corruption of pertinebat, or was the original 
reading perfluebat, corrupted to perfinebat, and corrected to per- 
tinebat? Perfluebat seems the exact word required, on the 
other hand pertinebat is rather nearer to the Mss. 


ccxv 13. Varro ὄνος λύρας (368 B.). Ht id dicunt suam 
Briseidem producere, quae evus nerura tractare solebat. 

So Buecheler. F* alone gives nerwia and tractare, other 
Mss. have neruzas and traciare. 


CCXVI 8 (Varro 549 B.). 
Nec multinummus piscis ex salo captus. 


Nec multunummus Buecheler. F* alone has multinwmmus 
(multumnummus with one stroke of the u, and the m, dotted 
out). Other Mss. multwm nummus. 


NONIJANA QUAEDAM. 181 


cexx 9. Infriasse sent papauerem. 
The Mss. have infrias sene here, Charis. 83.26 K simply 
infriasse. We should surely read infriassere papauerem. 


cox:xr 25 (Afranius 115 R.). 


Pulere hoc wncendi rogum. 
Ardet, tenetur: hoc sepulcro sepelvet. 


sepelict FL. sepelicet Ἠ" saepe licet V. 

The rogum in question seems to be that of a person que 
periturus est amore. I would suggest, adopting the reading of 
H? V, which is regularly to be preferred to that of F L, 


Ardé, tenetur: héc sepulcro sépelietur: tlicet. 


taking the whole line in a metaphorical sense. The copyist’s 
eye might easily pass from elz to «le. 


ΟΟΧΧΙΧ 12. Turdi masculini sunt generis ut plerumque 
lectum est. Feminint Varro,Quinquatribus. Tu medicum te audes 
dicere cum in eborato lecto ac purpureo peristromo cubare wdeas 
aegrotum et eius prius aluum quam τύλην subducere malis. 

So Buecheler, Varro 447. 

plent F°. pledi ΕἸ H’L V. 

conta FL. contra H’ V. 

caudes audes F H LV. 

cum in ἘΠῚ cum? ἘΠῊ ΤΙ, ΚΞ 

purpureo peristoro Ε΄ ΗΝ. purpureo peristo ἘΝΤ,. 

prius album all. 

quam typen all. 


It seems clear that a passage, probably from Varro, con- 
taining an example of turda, feminino, has dropped out, together 
with another pair of masculine and feminine words, the feminine 
of which is represented by typen in the example. The clue 
to this passage seems to be given by Martial 11. 16. The 
Zoilus of the day is shamming ill in order to shew off the 
magnificence of his bed-furniture. Varro is supposed to say 
to his medical attendant, “ What! you call yourself a doctor, and 
fail to see that you must remove his love of display, the cause 
of his disease, before you can deal with his (pretended) consti- 


182 THE JOURNAL OF PHILOLOGY. 


pation, which is only the effect.’ Remove the cause and the 
effect will disappear with it. 3 

Starting as usual from F*, I would suggest [Typhus mascu- 
lino. Typhe] feminino, Varro, Quinquatribus, Tu κλινικόν te 
audes dicere, cum in eborato lecto ac purpureo peristromo cubare 
utdeas aegrotum, et eius prius aluum quam typhen subducere 
malis. 

Isid. xvi 9, 101 has Typhus herba quae se ab aqua inflat, 
unde etiam ambitiosorum et sibi placentium hominum tumor 
typhus appellatur. τύφη in Greek is the name of a plant used 
for stuffing, but I do not know that it is ever used in a meta- 
phorical sense. Contra seems an obvious conjecture for conta, 
and caudes audes simply a dittography of audes. Te audes has 
been corrupted to taudes and caudes, and the duplicate reading 
is due to a correction superscribed : KAvve corresponds fairly well 
to Pleni, though it is possible that the vocative of a proper 
name (Plini?) has been combined into one word with κλινικόν.. 


CCXxx 25 (Attius 348 R.). 


Iphidamantem necr 
Dare, turbat uulgum, ewitat moeros dissicit. 


difidamantem FH. difidam antem L V. 
ne cidere turbat uulgum ambigua accius eurtat all. 


The words ambigua accius are clearly introduced from the 
previous example. Is it not also practically certain that ne 
cidere (nect dare?) is a gloss on euttat? Omitting these words 
the line will run 


Tfidamantem euttat, uulgum tirbat, moeros déssicit. 


coxxx1 19. Sall. Hist. lib. I: Ht mox Fufidius, adueniens 
cum legionibus, postquam tantas spiras, haud facilem pugnantibus 
uadum &e. 

Editors seem generally to alter tantas spiras, but the pas- 
sage is quite good Latin as it stands. We know from Festus 
330 s.v. that ϑρίγα was used in old Latin of a troop of men, 

Ennius quidem hominum multitudinem ita appellat cum ait, 
Spiras legionibus nexunt. ! 


J. H. ONIONS. 


NOTES ON PROPERTIUS. 


As the first Editor of Propertius with an English commen- 
tary, I may be allowed, perhaps, the privilege of making, after 
a long interval of years, a few-further remarks, partly bearing 
on Mr Housman’s “Emendationes Propertianae” in the last 
number of the Journal, but mainly as supplementary to his 
many ingenious and plausible corrections of a difficult and 
corrupt text. 

I have no intention of criticising Mr Housman at length; 
but I shall venture to express a doubt whether, among many 
shrewd and thoughtful suggestions, he has not proposed to 
disturb some of the received readings without sufficient cause. 
Not to go beyond the first elegy, i. 1, 19—24, on which six 
verses he has written eight pages of notes (p. 25—33), Iam by 
no means convinced of the soundness of his alterations. The 
editions give 

At vos, deductae quibus est fallacia lunae, 
Et labor in magicis sacra piare focis. 


I should not myself accept here either pellacia for fallacia, or 
Jata piare for sacra piare. Still less, in the following lines, can 
I approve the proposed reading, 


Tune ego crediderim et manes et sidera vobis 
Posse Cytinaeis ducere carminibus. 


Here the received reading is 


Tune ego crediderim vobis, et sidera et amnes 
Posse, &ce. 


“ Then would I put faith in you, (when you boast) that you can 
make both the stars and the rivers go which way you will,” 1.9. 


184 _ THE JOURNAL OF PHILOLOGY. 


the stars fall from the sky and the rivers go backwards from 
their sources. I see no objection to ducere in the sense of 
“making them follow your will.” But I have a serious objec- 
tion to Mr Housman’s emended verse, which, as it appears to 
me, would mean, or ought naturally to mean, “Then would 1 
entrust both ghosts and stars to you.” 

In any other sense, vobis is surely out of place. My objec- 
tion to pellacia is, that it means “allurement,” and that “ coaxing 
the moon to come down from the sky” is not the correct 
description of the irresistible power of magic spells. I think 
fallacia practically means “ trick,” “cunning art,’ “exercise of 
secret power.” I should not be too hard on a poet who is 
writing under the constraint of metre. Again, pellacia is a 
word barely in use. 

As for piare, which is a favourite word with Propertius, I 
cannot doubt that it represents ἁγέζειν, to consecrate on the 
altar offerings to Hecate. 

In 33 of the same elegy, where in me seems the ablative, 
“in my case,” it would be an improvement to read leges exercet 
amaras, for noctes, albeit Ovid (I think) has noctis amarior 
umbra est. Mr Housman reads me non nostra, ἕο. I suppose 
nostra is the Venus of both the lovers. 

In the next elegy, 1 should award the highest praise to his 
fulgore anquirere (I should prefer acquirere) amantes, for vulgo 
conquirere (aquirere). Mr Housman should have pointed out, 
in his own favour, that cultw and falso candore preceding mean 
the very same thing, viz. personal adornment and “sham 
beauty.” 

Here (i. 2, 13) I will introduce a suggestion of my own. 


For 


Litora nativis persuadent picta lapillis, 


I would read persqualent, “are all rough with coloured pebbles.” 
The compound does not occur; but squalent is evidently the 
word wanted. The emphasis is on nativis, pebbles produced on 
the spot, not imported like the Indian gems. 

The most serious difficulty I feel is the acceptance, on any 
grounds of probability, of Mr Housman’s wholesale system of 


NOTES ON PROPERTIUS. 185 


transposition. This is, indeed, periculosae plenum opus aleae, a 
method to be followed with the greatest caution and judgment. 
It is conceivable that a poet, in the furor scribendi, may add 
here and there a couplet that reverts to a former sentiment, 
and which might have been more aptly placed in some other 
position. But it is inconceivable that any transcriber, finding 
the true and logical sequence of verses or paragraphs in his 
copy, ABCD, should perversely and illogically write in quite a 
ditferent order BDCA, 

Only one just possible theory occurs to me to account for 
such violent disarrangements having come down to us. The 
writings of Propertius may have been collected and edited at 
first from “scrappy ” documents, in which added or interpolated 
lines, or some “ scripta in tergo,” were copied into wrong places. 
But how a couplet, say, in Bk. ili or iv, can be carried back to 
some elegy in Bk. i, with anything like a reasonable probability, 
this, I confess, I cannot see. 

Dismissing Mr Housman with the assurance that I have 
studied his valuable paper (transpositions and all) with the 
greatest care, I will add a few suggestions of my own. 

In i. 15, 29, 


multa prius vasto labentur flumina ponto, 


the context requires orta, “rivers shall have their source in and 
flow from the sea instead of towards it,’—dvw ποταμῶν πηγαί. 
If orta were wrongly read or wrongly written ulta, it would 
certainly pass into multa. Mr Palmer reads alta, which is an 
epitheton otiosum ; and muta does not seem probable. Compare 
111, 19, 6, Fluminaque ad fontis sint reditura caput. 

In 1. 16, 9, 


nec possum infamis dominae defendere noctes, 


where Mr Housman proposes voces, I suggest novas, “ the insults 
brought on me by a disreputable woman.” To this word has 
inter refers just below, i.e. “one of these insults is to be called 
crudelis,’ &c. Has inter (noctes) certainly has no meaning, 

_ 1 cannot help singing a paean of εὕρηκα on the perplexing 
distich of the same elegy, 23—4; 


186 7 THE JOURNAL OF PHILOLOGY. 


Me mediae noctes (noctis H.), me sidera prona iacentem, ἡ 
Frigidaque Eoo me dolat aura gelu. 


Read, verberat aura,—which, as it seems to me, the context 
absolutely requires,—and (perhaps) me madidae noctes. A tran- 
scriber, mistaking the quantity of 600, and reading frigida eoo 
me | verberat, as in Virg. Georg. i. 288, terras irrorat eous, may 
have led the next copyist to seek a synonym of verberat m 
Horace’s fuste dolat (Sat. i. 5, 23). A good Latinist would 
say, “me pater, me frater iniustaque premit noverca,” or “in- 
lusta me premit noverca,” but not “me pater, me frater, iniusta- 
que me premit noverca.” 
The passage in 1. 19, 16—20, is exceedingly difficult :-— 


et Tellus hoc ita iusta sinat. 
Quamvis te longae remorentur fata senectae, 
Cara tamen lacrimis ossa futura meis. 
Quae tu viva mea possis sentire favilla! 
Tum mihi non ullo mors sit amara loco. 


Here I propose, “at Tellus haec ita iusta sinat,” Le. sinat 
fiert, where tusta, as in 11]. 7, 9, are “funeral obsequies.” The 
Earth, as a power influencing both lives and deaths, is asked to 
allow Cynthia to survive even the aged poet, and to be. suffi- 
ciently strong and able to throw incense, to mark her affection, 
on his funeral pile. Compare 11. 13, 29, 


Osculaque in gelidis pones suprema labellis, 
Cum dabitur Syrio munere plenus onyx. 


Here therefore I would place a comma at sinat, and regard 
the true syntax to be sinat—futura esse. Then me remorentur, 
and lacrimis tuis, and suffire for sentire. In iv. 8, 84, the poet 
uses this verb meaning “to fumigate with incense,” θειοῦσθαι 
(Eur. Hel. 866), and Virgil (Georg. iv. 241) has suffire thymo. 
The meaning of the passage thus amended is as follows:—* May 
earth grant that my obsequies may be conducted thus” (the 
poet seems writing in a serious illness); “that however long I 
may live, my bones may yet be bedewed by your tears, and 
that you, yet in life (and aged as you may be), may be able to 
throw incense on my pyre.” 


NOTES ON PROPERTIUS. 187 


In ii. 9, 7, for “visura et quamvis nunquam speraret Ulixen,” 
where Mr Housman writes, “perhaps visurum,” I read venturum, 
which the context seems absolutely to require. 

In the beautiful verses on the loves of Aurora and Tithonus 
(ii. 18, 12), for lawit equos the epithet sedula suggests pavit, i.e. 
gave them a bait before starting on the journey. Possibly the 
doctus poeta had in view 1]. viii. 564, 


“ \ a ‘ > ’ \ 3 , 
ἵπποι δὲ Kpt λευκὸν ἐρεπτόμενοι καὶ ὀλύρας, 
ἑσταότες παρ᾽ ὄχεσφιν ἐύθρονον ἠῶ μίμνον. 


In ii. 11, 6, for humano corde volare dewm, which is unintel- 
ligible, I suggest humana sorte, “in a human shape,” “under 
the conditions of humanity.” For deuwm is contrasted; Cupid, 
though a god, carries bow and arrow like a human creature. 
So in 11. 20, 12, a company of little Cupids (turba minuta) 
exclaim, intereat, qui nos non putat esse deos. 

In ii. 28, 27, we should read quo sis (for sit) formosa periclo. 
Cynthia, conversing in the under world with the heroines whom 
Jupiter has loved, will tell Semele the risk she (Cynthia) has 
incurred by her beauty; and Semele, says the poet, from her 
own experience, will believe it. 

In ii. 31, 9, 


Hic equidem Phoebo visus mihi pulchrior ipso 
Marmoreus tacita carmen hiare lyra, 


read hic quidam. “A certain person” means a statue of Augus- 
tus himself, attired as Phoebus, and playing the lute. Perhaps 
this suggests a new point in the description of the Palatine 
Library, which this ode commemorates. Mr Palmer reads 


Hic Phoebus Phoebo visus mihi pulchrior ipso, 


but the point. of this is not clear, as the god himself is men- 
tioned below (16), “ Pythius in longa carmina veste sonat.” 
In iti. 4, 3, 
parat ultima terra triumphos, 
Tigris et Euphrates sub tua iura fluent, 


Mr Housman reads Thybris, et—fluet. I think we may improve 
on his suggestion by reading 


188 © THE JOURNAL OF PHILOLOGY. 


Thybris, ut Euphrates sub sua iura fluat. 


Here “Tiber” is in exegesis to ultima terra, i.e. Italy as opposed 
to Parthia. Thus the Tiber desires that the Euphrates should 
become subject to it. 

In iii. 6, 28, possibly we should read unguibus for anguibus ; 
the witch scratches wp bones with her long nails. But exectis 
cannot mean, I suppose, “uncut,” and haud sectis does not seem 
probable. Mr Housman reads exuctis anguibus, which I do not 
understand. 

In ui. 18, 10, 


errat et in vestro spiritus ille lacu, 


Mr Housman reads inferno. Rather, perhaps, infesto, “ haunted,” 
—a very “weird” verse. 

I conclude by offering a meed of praise to Mr Housman for 
his clever correction of iv. 7, 4, Zibure ad extremam nuper 
humata viam, for murmur ad extremae viae, which is nonsense. 
Her burial at Tibur, on the banks of the Anio, where the road 
ended, is indicated in ver. 85 (if we so read with Mr Palmer), 


Hic sita Tiburna iacet aurea Cynthia terra. 


F. A. PALEY, LL.D. 


δῆς thw ~~ ὧν 


ADVERSARIA, 


Cato. 


Origines 11 27 Jordan. Lavini boves immolatos, priusquam 
caederentur, profugisse in siciliam. So the Ms. of Daniel’s 
Servius, which preserves this fragment on Aen. 10 542. In 
silvam is Brisson’s emendation for in siciliam, and this the 
editors have accepted: I would suggest, in order to account for 
the first syllable of siciliam, that in vicinam silvam is the right 
reading. 


Horace. 


Carm. 2 2 5 Vivet extento Proculeius aevo, Notus in fratres 
animi paternt. Rutilius Lupus 1 5 (p. 5 Halm) has the fol- 
lowing story of a certain Proculeius to illustrate the figure 
ἀνάκλασις, or taking a word intentionally in a different sense 
from that in which it was meant. Huius modi est vulgare 
illud Proculeianum. Proculeius cum filium suum moneret et 
hortaretur, audacter ex bonis ipsius sumptum fuceret, quas in 
res vellet atque opus esset, nec tum denique speraret libertatem 
licentiamque utendi futuram cum pater decessisset, cut vivo 
patre promisce omnia licerent, filius respondit, non esse se opus 
saepe eadem oratione monert, nec se patris mortem expectare. 
Cut Proculeius pater subiecit, ‘Immo oro, inquit, ‘meam mortem 
expectes, nec properes moliri ut velocius moriar. The anecdote 
is told in a much abridged form by Quintilian (9 3 68). 

If the Proculeius of Horace may be identified with the 
young Proculeius of Rutilius, then animi paterni will gain 
greatly in meaning, ‘his father’s well-known generosity.’ Pro- 
culeius treated his brothers (we must suppose) as his father 
had treated him. 


190 THE JOURNAL OF PHILOLOGY. 


Livy. 

2 21 4 Tanti errores implicant temporum, aliter apud alios 
ordinatis magistratibus &c. After errores I propose to insert 
res, otherwise implicant has no object. 

3.5 14 Ut Romam reditum et tustitium remissum est, 
caelum visum est ardere plurimo igni, portentaque alia aut 
obversata oculis aut vanas exterritis ostentavere species. After 
aut some word seems required to balance obversata oculis: 
I propose to insert audita, which might easily have dropped out 
after aut. Thus we get an intelligible sentence, ‘ portents either 
seen or heard of.’ 


Servius. 


Aen. 10 664 ‘Lille autem Aenean’ legunt: quo si ita est quia 
Aeneas ‘multa virum demittit corpora morti. For quo si ita 
est Scholl has proposed quod st ita est, reading for guia Aeneas... 
morti qui Aeneas...morti? The right reading may perhaps be 
quod stultum est, quia &e. 

Aen. 10 705 Face praegnans Cisseis regina Parin creat. 
Daniel’s Servius has the following corrupt note: ‘Parin creat’ 
plus est quam si diceret ‘face praegnans incendit paret. Pos- 
sibly for plus est quam si diceret ‘face praegnans incendia parit’: 
the commentator meaning that the mention of Paris’s name 
after the epithet face praegnans is a more powerful touch than 
the use of such a phrase as incendia or ignes tugales (Aen. 7 
320) would have been. 


Digest. | 


2 4 20 Sed etiam ab inea et balneo et theatro nemo dubitat- 
in ius vocart licere. Ab inea may perhaps stand for ab Iano. 


Lucan. 
1 314. 


Scilicet extremi Pompeium emptique clientes 
Continuo per tot sociabunt tempora regno ? 


Mr Haskins, though apparently with some hesitation, refers 
extrem to ‘the distant kings in Africa and the East who were 


νον 


ADVERSARTA. ΙΟΙ 


under the influence of Pompeius’ I confess my inability to 
make sense of extremt, and would suggest that it is a corruption 
for hesterni, ‘clients of yesterday.’ This would, I think, fit in 
excellently with the context, and would be a parallel, as far 
as the expression goes, to Persius’s contemptuous hesterni 
Quirites. 


3 558. 


Tunc in signifera residenti puppe magistro 
Brutus ait: Paterisne acies errare profundz, 
Artibus et certas pelagi? 


The commentators offer no satisfactory explanation of et 
certas: nor do I see how the words are to be translated. Per- 
haps et certas is a corruption for expertas. 


7 139. 


Tune omnis lancea saxo 
Erigitur. 

Mr Haskins says erigitur =‘is straightened.’ Can erigo 
have this meaning? I had conjectured derigitur, when I found 
that it is confirmed by Usener’s Berne Commenta, saxi pondere 
curvamen dirigitur contortae lanceae et recurvae. 


Velius Longus. 


P. 49 Keil. ‘JZ’ vero littera interdum eailis est, interdum 
pinguis, ut in eo quod est ‘ prodit’ ‘vincit’ ‘condit’ exilius volo 
sonare. 

For volo sonare I suggest vult sonarz. 

P.52. Nam quod ea hoc quoque existimant quidam colligi 
posse consonantem esse (‘h’ litteram) et adsignificantem, quod 
aut accedens aut recedens immutat significationem, siquidem 
aliud est hira, aliud tra, canam exemplum. YVelius Longus goes 
on to observe that the meaning of words is sometimes changed 
by a change of quantity (as in pila and pila) and sometimes by 
a change of accent (as in cércewm and circtim). The mere fact 
of the change of meaning in such cases as tra and hira is, there- 
fore, not a strong argument for claiming h as a letter. 


102 THE JOURNAL OF PHILOLOGY. 


No satisfactory emendation has been offered for canam 
exemplum, nor am I at all confident in the one which I am 
going to propose, contrarium ponam exemplum, ‘I will give an 
instance which makes the other way’: though this would give 
the sense required. 

P. 63. Jn eo quod est ‘ expectatius’ duplicem scriptionem 
quidam esse voluerunt, ut quotiens cum verbo quod est ‘ expecto 
opperior’ praeposita haec pars orationis fuerit, ‘e’ et ‘a’ littera 
contenta sit; quotiens autem cum verbo quod est ‘ specto ludos, 
necessaria esse ‘s’ litterae insertior sit, ut in ipsa quoque scrip- 
tione ambiguitas deducatur eius quod est ‘expectare illum volo’ 
et ‘ spectare mihi placet,’ 

Read perhaps necessariam esse ‘s’ litterae insertionem, ut... 
ambiguitas deluatur. 


Sergius on Donatus. 


P. 520 Keil. Consonantes autem sunt (i et u) cum alits voca- 
lLibus in una syllaba praeponuntur, aut cum ipsae inter se in una 
syllaba coniunguntur. Nisi enim et prior sit et in una syllaba 
secum habeat coniunctam vocalem, non erit consonans ἃ vel τι. 
Nam ‘Julius’ et ‘Iarbas’ cum dicis, + consonans non est, licet 
praecedat, quia in una syllaba non habet coniunctam vocalem, sed 
in altera consequentem. For Julius read Lulus: for in Lulius 
the first τ is consonantal, and the writer can hardly be referring 
to the second 2. 


Vergil Aen. 12 158. 


On the words conceptumque excute foedus Conington re- 
marked ‘It is just possible that there may be a reference to 
the physical sense of conceptum, and that excute may mean 
“render abortive,” as, though no instance is quoted of the word 
in that sense, it would be sufficiently appropriate.’ I have just 
found a parallel in Scribonius Largus (p. 2 Helmreich), medica- 
mentum quo conceptum excutitur. 


H. NETTLESHIP, 


LEXICOGRAPHICAL NOTES. 


THE following words, from the fifth volume of Keil’s 
Grammatici Latini, are wanting in the last edition of Georges’ 
Handwérterbuch : : 

nector: Macr. Exc. Bob. 655. 8, ἀγχονιστής" nector. 
Stephens’ Greek-Latin Glossary (in Labbé) has ἀγχονιστής" 
necator (a known word). There is the same confusion between 
nectus and necatus, cf. Osbern in Mai 8. 371 and Hagen Suppl. 
Ixvi. 10. ᾿Αγχονιστής seems to occur in no Greek lexicon. 

passivoneutra: t.t. in Phocas 431. 8. ‘Neutropassiva’ is 
well known. 

paumentum: Tul. Exe. 324. 10, pro pavimento. Compare 
caueo cautus; pavimentum occurs late, and an inscr. gives 
paimentum, for which see Seelman’s Aussprache p. 218. 

piper: Cons. 392. 4, ‘i in African Latin’. 

polibo: Pomp. 225. 12, polio et poliam et polibo, nutrio et 
nutriam et nutribo. Nutribo is quoted elsewhere (eg. by 
Cledonius and Palaemon), but polibo occurs here only. See 
Neue It. 448 foll. 

praedormio: Macr. Exc. Bob. 637, 15 ne without ref. 
De Vit cites Augugtine and the Latin of Irenaeus. 

quia, quoniam, quod, after scio, praemoneo &c.: Pomp. 224. 
2, 225. 3, &., Macr. 633. 5. Οὗ Rénsch p. 402, Drager 11. p. 
231 foll., Avianus I 2 with Mr Ellis’ note. 

rosum: de Dub. Nom. 589; Labbé (Cyr.) ῥοδόν" rosum. 
Cf. Du Cange; de Vit refers to the Digest. 

salmentum: ‘‘verdirbt fiir salsamentum Charis. 265. 16; 
Tul. Exe. 324. 10 and 327. 15; Cons. 392.10” Georges. And 
so Donatus Keil 4. 392. 15, and some glosses, But 1]. 327.15 
quotes ‘Caper antiquissimus doctor’ (100 A.D.) as saying ‘salmenta 

Journal of Philology. vow. xvt. 13 


194 THE JOURNAL OF PHILOLOGY. 


ne timeas proferre, quia latinum est’ (Keil 7. 101. 8), and adds 
‘salmentum est purgamentum maris, salsamentum vero salsugo 
dicitur, in qua liquescit sal ut sales condiantur. Compare — 
Gl. Amplon. Oehl. 379. 53 and 13, salmentum: quod salibus 
condimus; Ball. Gl. &c., salmentum: confectio piscium; Pla- 
cidus (ed. Hagen), salsamenta sunt omnes res salsae, ut pisces et 
sardae. (Cp. Osbern in Mai 8. 558 and Hildebrand 5. υ. 
salamentum.) The grammarians’ notion that salmentum is 
contracted from salsamentum, as if through an intermediate 
salamentum, though upheld lately, is absurd. Seelman’s 
examples of ‘s’ dropping out (Aussprache p. 318) are quite 
different. 

scutris, scutrilla: Cons, 346. 30, without ref. Possibly the 
words are confused forms of the common ‘scutra, scutrillus’ 
Pomp. 164, 24 ἄο. De Vit, overlooking Consentius, quotes 
only ‘scutrilla’ from Osbern in Mai 8. 567, but there Mai prints 
scrutilla, which we should probably read scutella (cf. Hagen, 
Suppl. 239. 10). Weinhold in W6lfflin’s Archi iv. 180 adds 
nothing to the explanation of these forms. 

stetim: Cons. 392. 15 barbarismi, ut si quis dicat...stetim 
pro statim. See Seelman, p. 171. 

stiria (n. pl.): de Dub. Nom. 590. 15, stiria dicuntur ab 
stillis, quae Vergilius genere feminino, Varro neutro dixit. The 
first part of the note comes from Verrius (Paul. ex Fest. 345. 
3 M.), but it is common in grammarians, and does not, therefore, 
prove the correctness of the second part, the reference to Varro 
and the gender. The diminutive ‘stirillum’ would point to 
a neuter form, but it appears in many glossaries (e.g. Ball.) as 
sterillum, and Lowe (Prodr. p. 18) holds this to be the proper 
spelling. Weinhold (Genuswechsel der Deminutiva, Archiv iv. 
179—188) does not seem to mention stirillum, which is in Du 
Cange (= caprae barba). 

subineptt: used by Macr. Exc. Bob. 632. 3, qui putant... 
dixere dualis esse numeri, subinepti sunt. . 

subtegumen: Aug. reg. 501. 12, quoted without ref. 

torve: quoted by Pomp. 290. 19, torvum pro eo quod est 
torve, cf. nave naviter (of which latter new exx. are Gotz Gl. Ter. 
519, Gl. Ampl. 353. 8 Oehl., Bodl. Glossary) and Neue τι. 654. 


LEXICOGRAPHICAL NOTES. 195 


tumax: Kutych. 454. 1; ars anon. Bern. in Hagen Suppl. 
74. 23; quoted without ref. The word occurs also in Diefen- 
bach’s Lat. German Glossary, ‘qui cito emittitur’, By the side 
of this word 

tumex: σμώδιξ Stephens’ Lat. Gr. Gloss., and 

tumia: τύμπανον Labbé (Cyr.), deserve quotation here. 

turturella (for -illa): de Dub. Nom. 592. 4, ‘ Pollio et alii.’ 
The ‘Pollio’ need not be taken for certain, as the tract often 
seems to ascribe wrong authors to words. Turturilla is other- 
wise quoted only from Seneca (see Archiv iv. 184). 

verbialis : = verbalis, t.t. in Pomp. 149. 1, Cledon. 37. 7. 
De Vit’s ref. to Eutych. 450. 28 is wrong. For the form cp. 
artificalis, artificialis (see Journal vol. xiii. p. 81). De Vit gives 
verbialiter also. 

utrum...aut: Pomp. 149. 10, utrum sarsor dicimus aut... 
aut...2? So num...vel num in Cons, 368. 22, inspicere num 
possit...vel num...in alio intellectu...esse possit. There is, 
then, some authority for the ‘utrum...ve’, which Peerlkamp 
put down as Horatian, | 


F, HAVERFIELD. 


_ 13—2 


THE NUMASIOS INSCRIPTION. 


Towarps the end of last year a tomb was opened at 
Praeneste by Helbig and Diimmler, in which was discovered a 
gold fibula bearing a very interesting inscription—the interest 
due to the form being considerably increased by its being 
supposed to date from before 509 B.c., which makes it much 
older than any other Latin inscription. 

An abstract of the paper which Diimmler read before the 
‘archiiologische Institut in Rom’ appeared in the Wochenschr. 
fir klass. Phil. of Jan. 26th, 1887 (No. 4, col. 121). The 
inscription is briefly noticed in Wolfflin’s Archiv fiir Lat. Lew. 
1887, Pt. I p. 143. More important than these however is a 
paper by Biicheler in the Rheinisches Museum (Vol. χται p. 317), 
and it is the latter which forms the immediate occasion of the 
present article. 

The inscription as given by Biicheler runs: 


lol SAWVAVM:419 A'8 4:5 84:49: S0IMA WA 


manios - med - vhe - vhaked - numastiot 


[After the second vh are visible traces of an upright line, 
apparently a mistake corrected. ] 

In discussing such an inscription the three heads to be taken 
up are, palaeography, phonology, and morphology, inasmuch as 
the matter is unimportant, 

1, Palaeography. 

a, The direction of the writing is retrograde. Hitherto 
the canon had been (e.g. Hiibner, Wiiller’s Handbuch, Vol. 1 
p. 496) that this order was only used in devotiones when the 
meaning was to be intentionally obscured. 


THE NUMASIOS INSCRIPTION. 197 


b. The form of the letters. 

If any one will compare the characters used here with those 
on the Dvenos inscr., which stands next in antiquity, he 
cannot fail to be struck by the differences they present, par- 
ticularly in the signs for ὦ, h, o and 8, Again, a glance at 
the Greek inscriptions on western soil, either directly in Roehl 
(507—550) or by means of the tabulated results in Kirchhoff 
or Hinrichs (Handb. |. c. p. 416), will suffice to show that the 
similarity of their alphabets to this is very marked indeed and 
becomes more so in the Italian examples, so much so that we 
are fairly justified in describing this as a Greek alphabet, 
borrowed but not naturalised. The importance of this result 
will appear below. 

c. The words are divided by double points instead of a 
single one as common later. In ‘vhevhaked’ the first syllable is 
divided from the rest by :. This seems merely to bear out the 
evidence for the hysterogene nature of the form. Biicheler’s 
parallel im - perator, with tribarakat - tins’ (Cipp. Abell.), 
medicat-inom (Tab. Bant.) and καταλείπον : ταῦ (I G A 321), 
are only additional evidence that the constant element of the 
word was kept mentally distinct from the variable terminations 
and prefixes. 

2. Phonology. 

a. The combination ‘vh’. 

This is the transliteration which Biicheler suggests, meaning 
presumably Latin v, not German or English. In this I follow 
him, but justify the transliteration on very different grounds. 

His view is, in brief, that this collocation represents a stage 
in the process by which the voiced aspirates dh, bh, gh passed 
in Latin into f Now a principle of the utmost importance for 
directing the science of language is this: that no phonological 
change is to be considered as fully established until each step 
has been shown to be natural on phonetic grounds. It is the 
recognition of this which forms one of the merits of the modern 
school; it is this which decides the superiority of ‘sonant 


1 Bartholomae (Bezz. Beitr, x11 82) 2 This has also been explained as 
denies that this has any phonetic a mistake. 
value, 


198 THE JOURNAL OF PHILOLOGY. 


nasals’ over ‘nasal insertion’ as an explanation of the same 
set of phenomena. 

This canon Biicheler violates when he speaks of dh, bh and gh 
taking a parasitic v between the sonant and the aspirate, and 
when he goes on to say that from the dvh, &c.thus resulting, either 
d, &c. fell off, leaving vh (of which this is an instance) which 
passed into f, or dv, &c. fell off, leaving h only. Even allowing 
the possibility of a parasitic v after a velar guttural, which is 
phonetically explicable, the combination would be ghv and not 
gvh, while for the palatal, dental, and labial series such a 
hypothesis is entirely unjustifiable. 

It may be urged that as long as a phonetic explanation of 
the change of the sonant aspirates to f, &c. is wanting, the 
insertion of hypothetical stages is permissible. It will therefore 
be scarcely out of place, if I attempt here an explanation of 
the change on phonetic lines. 

A sonant aspirate contains in itself two antagonistic ele- 
ments. While the vocal chords are stretched the current from 
the lungs is more or less impeded and softened. On the other 
hand the expulsion of a strong breath tends to force the chords 
apart and leave the passage free. Hence a sonant aspirate is 
an unstable combination and practically unknown in Europe 
except in the Irish pronunciation of English. 

When this unstable sound is modified it must be done in 
one of two main directions, (1) in favour of the sonant, or (2) in 
favour of the aspiration. The latter again has two subdivisions, 
(a) the sonant may become its corresponding surd, (b) the 
aspiration may prevail entirely and produce a sound which 
may be called ἢ if we remember that ἢ is wholly indeterminate. 
The only other modification worth mentioning is that adopted 
by Teutonic in which the concession to the aspirate is made, 
not in the larynx, but in the mouth, resulting in 2, 3, αἰ, w* 
respectively. Of these methods (2 α) is best exemplified by 
ancient Greek. Latin adopted (1) in the middle of a word 
where a strong breath is in any case difficult. Hence the rule 


1 The value of these signs is that seem to have been employed in Latin 
given by Sievers, Phonetik, p. 127. unless the v in nivis is an attempt to 
This method of reduction does not represent 3. 


THE NUMASIOS INSCRIPTION. 199 


for Latin that gh, dh and bh medial pass into g, d and b 
respectively. On the other hand, at the beginning of a word 
the lungs are full and a strong expiration is easy; here then 
(2 δ) is the method adopted. But the indeterminate h resulting 
therefrom, although indeterminate in writing, is by no means 
so in speech, its value being fixed by the following considera- 
tions. If the sound which it replaces was gh the only tendency 
to contact will be at the back of the mouth. Hence the breath 
will have a slight guttural character and will be accurately 
represented by h. On the other hand if the sound replaced 
was dh or bh, the tendency to contact is at the front of the 
mouth, the lips are approached and a bilabial sound is formed 
which bears most resemblance to f. Hence we may formulate 
the rule that gh initial, in Latin, passes to h*, but bh and dh 
become αὶ Further, the f which comes from bA will be purely 
a bilabial sound, and a very slight severance of the lips would 
render it indistinguishable from h, whereas the f from dh would 
naturally be more labiodental in character and so be stable. 
The table given by Stolz, ὃ 58 (Handb. τὶ p. 177), should there- 
fore be modified as follows for initial sounds : 


Indo-Eur. gh dh bh 
Lat. h Ἢ Ἃ 
h 


If this account of the process is reasonable, the necessity for 
assuming a parasitic v falls to the ground, while a direct argu- 
ment against Biicheler’s view is supplied by the rule that dh 
medial passes into d. The sequence he seems to assume is 
I.. Eur. *dhédhé, Latin *dvhédvhé, then *vhévhd, but I. Eur. 
*dhédhé would give Latin *fédé (Osth. Perf. p. 207). 

I believe the explanation of this orthography to be given at 
once by the nature of the alphabet, which was shown above to 
be Greek. That being so we have here to deal with a translitera- 
tion. But in a Greek alphabet of this date, no sign for f existed 
and the engraver would find some difficulty in representing the 
sound. The nearest equivalent would obviously be a breathed 


1 Cases of ‘f’ are to be regarded as borrowed. Osthoff, M. U. 4. 99. Brgm. 
Grundriss, § 389. . 


200 THE JOURNAL OF PHILOLOGY. 


digamma and as such 1 regard this combination. To mark that 
a sound is to be breathed ἢ is used as a diacritic by the Greeks 
themselves, e.g. 6 = breathed p (Brgm. 1. c. § 266) and Fh as here 
(I. G. A. 131). This breathed F would differ very slightly from 
bilabial f and its regularly passing into ἢ in Greek (Brgm. 
1. c. ὃ 166) supports what was advanced above on the second 
change of f from bh’. That no distinction was made in writing, 
between f bilabial and f labiodental, is not to be wondered at. 
Of course when the alphabet had been naturalized and v was 
the equivalent for F, a double sign for f was cumbrous and 
unnecessary”, and Biicheler, with great probability, suggests 
that while Latin retained F, the Umbro-Samnite alphabet sim- 
plified in the other direction and 4 modified as © for h was 
retained as 8 for fi 

The vocalism of fefaked depends too closely on the mor- 
phology to be discussed here. 

b. The proper name Numasius. 

The difficulties presented by this word are not inconsider- 
able. In classical Latin we have the two forms Numisius and 
Numerius. Is this connected with either of them or with both? 

Of course it is possible to call in the aid of that deus ex 
machina, the anaptyctic vowel. It is also possible with Bii- 
cheler to regard -asius -isius -esius as unimportant variations of 
suffix. Perhaps however a more legitimate explanation may be 
found. . 

Roman proper names are formed from a comparatively small 
number of simple roots. These may or may not have a meaning 
that we can trace—more frequently the latter is the case, for 
the very fact of a word becoming a proper name implies that it 
ceases to be used with a connotation. To these simple roots are 
added a great variety of suffixes both primary and secondary. 


1 This confusion of bilabial f with 
h is also illustrated by CIL 1. 1501, for 
Jordan’s explanation (Krit. Beitr. p. 
50 sqq.) is hardly plausible. 


* An interesting confirmation of this. 


is yielded by a close inspection of 
the inscription. If Biicheler’s copy is 
accurate, the engraver had actually 


omitted the diacritic after the second 
F, and only perceived it after beginning 
the K. Thereupon he turned the A 
into ΗΕ and erased the upright line 
which was to have formed the kK. This 
shows that the hk was necessary but 
awkward, 


THE NUMASIOS INSCRIPTION. 201 


An example οὗ a root with traceable meaning is ‘manus’ ‘a 
good man’, from which come Manius Manilius Maneius, &c.; of 
an obscure root is Titus, forming Titius Titinius Titidius Titu- 
rius, &e. One such old root-name was Nuwma—probably the 
oldest masculine -a stem in the language—which may be con- 
nected with νόμος. With the suffix -so (see Pauli, Alt. It. Stud. 
1 53) this gives Nwmdso- secondary Numdsio-, with which the 
later Audasio- Equasio- and the like cannot be compared on 
account of the doubtful quantity of their -a-. 


The chief forms of the names are 
i” Latin. 


1. Numasius. 

2. Numsius Momms, U. 7. Diall. p. 252. Corss. Etr. 
Π 14. 

Numpsius Momms. ]. 6. p. 197. CIL1 1211. 

Numisius classical and frequent. 

Numaiirius, Numiirius CIL tv 2313.., 

Numerius classical. 


B. Oscan. 


1. Νιυμσδιηις Zv. 160. 
2. niumsis Zv. 57. 
3. niumerio Zy. 24 (late and Latinizing), 


SEs ah 


C. Etruscan. 
1. Numsi. Corss. ]. ¢. 


The antiquity of this inscription (as well as the regularity of 
the formation itself) would lead us to assume Nwmasius as 
the original for all these forms. In Oscan and Etruscan the 
short vowel was syncopated as usual, and in this form, if we 
may trust A. 2 and 3, the name was borrowed by Latin. 
Another of the names derived from Numa was Numitor, in 
which ὁ is regular, being in an unaccented syllable before a 
dental (Stolz, § 23, no. 4). It is then not surprising if Nu- 
masius and Numsius give way to Numisius. The name in this 
form is found in classical times, and, with other names in -sius, 
has been much discussed. Jordan (Krit. Beitr. p. 104 foll.) con- 
cludes that the tradition which makes Roman names in -rvwus 


202 THE JOURNAL OF PHILOLOGY. 


rest upon older forms in -stus is unfounded, and that all names 
in -stus are to be considered as borrowed. The prevailing view 
however is, that the tradition is correct, but that names in -sius, 
in classical times, are to be explained as Jordan does. I scarcely 
think enough allowance is made for the natural stability of 
spelling in the case of proper names (cf. Le Maistre in modern 
French) which moreover are specially subject to family caprice, 
tradition and even politics, as in the case of Clodius. It is not 
to be wondered at then that a certain family should retain 
Numusius although for the most part the rule of rhotacism pro- 
duced. Numirius. The subsequent change of ὁ to e before r is 
regular (Brem. Grds. ὃ 33. 1). 

Except for the explanation of the other forms, however, there 
is no need to insert any stages between Numasius and Nu- 
merius, for Numarius would of necessity receive e from numerus— 
Quintus Sextus, &c. and Quinctius Sextius, &c. giving the pro- 
portion both for praenomen and for nomen. See also Varro ap. 
Non. 352. 29 qui celeriter erant nati fere Numerios praenom- 
nabant. 

Numisius gives in all probability the suffix -%sius which 
forms names in -trius (e.g. Papirius from Papius'), to analogy 
with which may be due Numiirius in A. 5. In the case of 
Apisius and the like, the quantity of the ὁ depends on whether 
they are formed from Apus or Apius (Appius). 

One name seems to have followed step by step the develope- 
ment of Numerius. This is Valerius, which has Vala, also an 
a- stem, as base. That Numerius set the analogy, is certain 
from its superior antiquity, as shown by its twofold use. If the 
above considerations have any weight we must correct the 
statement of Festus (p. 23 Miill.) that the older form of Valerius 
was Valesius. It shculd be Valsius after Numisius, which 
would become Valerius as above. The existence of a suffix 
-€sius is questionable, for, in the examples quotable from the 
Corpus, both the quantity of the -e-, and its forming part of the 
suffix, are doubtful. 


1 Jordan 1. δ. seems: to make Nu- does to Papius, which is impossible 
misius stand to Nummius as Papesius from the difference of quantity. 


THE NUMASIOS INSCRIPTION. 203 


3. Morphology. 
a. The character of the language. 


This I have assumed to be Latin, on the authority of Bii- 
cheler, whose argument is that it lies between Latin and Oscan 
and is shown not to belong to the latter by the presence of 
‘med’ for which, to judge by analogy, Oscan would have 
‘miom.’ 

b. ‘med.’ 


This accusative with the ablative ending -d shows that even 
in the sixth century B.c. Latin had its case-formation consi- 
derably broken down and confused. 

c. ‘fefaked.’ 

That this cannot represent an original reduplicated form of 
the root dhé- was shown above, by the consideration that 
*dhédhé would in Latin become *fédé, while the interpunctua- 
tion points to the same conclusion, From the cognate forms 
the following may be suggested as a possible history: /dhé 
Perf. Ist sg. *dhédhé-a =*dhédhé (Sanskrit dadhd-u, Greek 
τέθη-κα); as in Greek, this was probably strengthened by the 
deictic particle -ke or -ka (the later -ce) either as an inde- 
pendent developement or inherited by each from the common 
stock. The case for Greek is stated in Osthoff, Perf. p. 326 
foll. As in Greek also the & spread into the other tense systems, 
as the aorist dhék = @nx* whence the classical perfect féc-7, and 
present dhdk- with the short vowel originally in plural only, 
but levelled into the singular. This may be considered as pre- 
dialectic, for it is found in Oscan fe-fac-ust fe-fac-id (with 
hysterogene reduplication as here) fakurent, &c., in Umbrian, in 
classical Latin as in facio, and in this form. Biicheler thinks 
that the erased line points to a form feficit analogous to tetigit 
which the engraver was going to write but changed his mind. 
Perhaps the erasure is better explained as above. (Note 2, 
p. 200.) 

The ending -ed, its character, and the quantity of the e, 
have been fully discussed by Osthoff (Perf. pp. 205—232) whose 


1 This equation is Bartholomae’s (K. Z. xxvii. 255). 


204 THE JOURNAL OF PHILOLOGY. 


conclusion is that this “3. sing. auf -éd” is to be regarded as 
“alte ur- und gemein-italische neubildung zu der 3. plur.’ 
This therefore brings nothing new to bear upon his result. 

d. Numasioi. 

The chief direct testimony to the use of this dative in -oz in 
old Latin, corresponding to the Greek -@, had hitherto been 
that of Marius Victorinus(G. L. 17. 20) and it had been consi- 
dered doubtful by some, e.g. Jordan, Krit. Beitr. p. 241. 

The collateral evidence however, the analogy of Greek, and 
of Oscan forms like Nuvlania, was very strong at least for its 
existence in Italy. This however proves it to have lasted into 
Latin. 

The general results of the discovery may be summed up as 
follows. It supplies a step in the history of alphabets, as it 
shows the process of borrowing from Greek in actual operation 
and as yet incomplete. It throws the separation of dialects in 
Italy to a very early date, as Latin is not only separated, but is 
considerably advanced towards its classical stage—of course this 
is assuming the accuracy of the ascribed date. It gives evidence 
of the existence of a single-name epoch and shows that that 
name might be a patronymic, thus explaining the existence of 
such among praenomina. Its bearing on other inscriptions is 
naturally unimportant from both its brevity and its antiquity, 
but it has a slight connection with the interpretation of the 
‘Dvenos.’ Pauli (Alt. It. Stud. 11 foll.) makes two assumptions 
—that the retrograde order was not primarily due to desire of 
secrecy, but was a relic of an older custom,—and that the dative 
masculine in Latin ended in -oz. These are no longer assump- 
tions but are actually in evidence, and to this extent the dis- 
covery supports his view. 


H. D. DARBISHIRE. 


AESCHYLEA. 


Ἑρμῆ χθόνιε πατρῷ ἐποπτεύων κράτη 
σωτὴρ γενοῦ μοι ξύμμαχός τ᾽ αἰτουμένῳ" 
ἥκω γὰρ ἐς γῆν τήνδε καὶ κατέρχομαι. 
τύμβου δ᾽ ἐπ᾽ ὄχθῳ τῷδε κηρύσσω πατρὶ 
κλύειν ἀκοῦσαι. Choephori, 1—5. 


The difficulty of the first line is well known. It is quoted 
by Euripides in the Frogs of Aristophanes as illustrating the 
charge of obscurity which he brings against Aeschylus. Euri- 
pides himself in that passage suggests one interpretation of the 
line. Dionysus proposes a different explanation. It has been 
assumed by every commentator with the exception of Aristar- 
chus and Hermann that the rendering suggested by Euripides 
is malicious and absurd: while on the other hand the inter- 
pretation of Dionysus has been generally accepted by modern 
editors. It seems to me that the discussion in the Frogs does 
undoubtedly throw some light upon this passage—only it does 
so not by suggesting a meaning which we may accept, but by 
enabling us to set aside certain renderings of the passage 
which need not be further considered. May we not take it 
for granted that any explanation offered by a comic poet in 
elucidating a tragedy would be intentionally and comically 
false? Now with regard to the rendering suggested by Eurip- 
ides almost every editor admits this. It is agreed that the 
misinterpretation is intentional. But what are we to say of 
the suggestion of Dionysus? Are his other answers to be 
taken seriously? Let us see. No one will maintain that his 
suggestive explanation. of κλύειν ἀκοῦσαι (Frogs 1175) is ad- 
vanced seriously. Surely the reason alleged τεθνηκόσιν γὰρ 
ἔλεγεν.. οἷς οὐδὲ τρὶς λέγοντες ἐξικνούμεθα is only intended 


206 THE JOURNAL OF PHILOLOGY. 


to be absurd. And it is worth noticing that Mr A. Sidgwick 
in his edition of the Choephori in view of the general 
absurdity of the answers given by Dionysus assigns the lines 

ov δῆτ᾽ ἐκεῖνος, ἀλλὰ τὸν ᾿Εριούνιον 

Ἑρμῆν χθόνιον προσεῖπε, κἀδήλου λέγων 

ὁτιὴ πατρῷον τοῦτο κέκτηται γέρας 


(Ar. Ran. 1144—46), which he takes to be a serious explana- 
tion, not to Dionysus but to Aeschylus. The suggestion in the 
mouth of Aeschylus would certainly be more deserving of 
réspect, but I cannot think such an arbitrary change justifiable. 
Inasmuch then as we owe these explanations to Euripides and 
Dionysus under Aristophanic treatment I cannot think them 
entitled to any respect. Could the Athenians have been ex- 
pected to laugh if Dionysus after his previous perversely absurd 
suggestions had gravely propounded a perfectly correct render- 
ing of this line? It remains to consider whether either of these 
two interpretations gives a good sense. The first we may at 
once dismiss: for we cannot extract from πατρῴα κράτη so far- 
fetched a meaning as that suggested by Euripides ὡς ὁ πατὴρ 
ἀπώλετο αὐτοῦ βιαίως. The sense given by the other render- 
ing is hardly more satisfactory. For by translating “Hermes 
god of the lower world superintending the powers of your 
father Zeus” we introduce confusion into the passage. πατρῷ᾽ 
in line 1 should, I think, in some way answer to πατρί in 
line 4. Orestes is here thinking of his own father only, and 
any reference to the source from which Hermes holds his powers 
seems to me quite beside the point. 

A third rendering has been suggested by Conington, who 
takes πατρῷα κράτη to mean “the palace of my father.” He 
also supposes that Orestes is addressing a statue of Hermes by 
the tomb. I do not think this at all probable. Agamemnon 
was buried δαΐαις ἐν ἐκφοραῖς possibly out of sight of the 
palace, and the assumption of a statue to Hermes near the 
tomb seems arbitrary. 

In all these renderings it is assumed that the vowel elided 
in πατρῷ isa. I believe it to be ε and punctuate thus: 


‘Ez a θό aA? b] , / 
PH) x Ovle, TATP®@ , ET OTT TEUDV ΚραΤΉ,; 


AESOHYLEA. 207 


translating “‘Hermes of the lower world, god of my fathers, 
overseer of victory.” 

Hermes is thus addressed as (1) χθόνιος, (2) πατρῷος, 
(3) ἐναγώνιος. Then in the following lines σωτὴρ cippayds 
Te answer to ἐποπτεύων κράτη, πατρί (line 4) to matp@e, and 
τύμβου δ᾽ ἐπ’ ὄχθῳ τῷδε to χθόνιε. Now there would be 
no ambiguity in the verse when duly pronounced by the actor, 
even if we suppose that the short vowel was elided in speaking. 
Probably however it was not elided but pronounced fully thus, 
Ἑρμῆ χθόνιε, πατρῷε, ἐποπτεύων κράτη. If the verse be read 
so, there is I think a decided advantage gained on metrical 
grounds—as it is usually read there is an abrupt pause after 
the second foot which spoils the rhythm. The obscurity of 
which Euripides complains certainly exists, and might have 
been removed by writing κράτος for κράτη. But fortunately 
we can illustrate from Aeschylus (Supplices 962) his preference 
for the plural κράτη. There we find εἴη δὲ νίκη καὶ κράτη τοῖς 
ἄρσεσιν, where as here the plural is preferred without metrical 
necessity. We find also κράτεσιν ἀρσένων in Supplices 388, 
-but that is not so strong a case. I conclude then that this 
preference for the plural form κράτη is the cause of such 
obscurity as exists, but here as in Choeph. 553, αἰνῶ δὲ κρύπτειν 
τάσδε συνθήκας ἐμάς “I bid these conceal my counsels,” there 
would be no ambiguity when the line was spoken. Now as to 
the meaning of watp@e. Hermes was a Pelasgian god wor- 
shipped at Argos (Supplices, “Ἑρμῆς ὅδ᾽ ἄλλος τοῖσιν ᾿ Ελλήνων 
νόμοις), and Agamemnon boasts (Hom. Il. Bk. m. 72) that he 
received his golden sceptre through Thyestes, Atreus and Pelops 
from Hermes himself. Orestes therefore is specially justified 
in addressing Hermes as πατρῷος. (For πατρῷος applied to 
Hermes elsewhere cf. Lysias p. 104. 16, ἀσεβεῖν περὶ τὸν 
“Ἑρμῆν tov αὑτοῦ πατρῷον.) For the sense I have given to 
ἐποπτεύων κράτη overseer of victory we may compare Choeph. 
475 ὦ γαῖ aves μοι πατέρ᾽ ἐποπτεῦσαι μάχην, and 1. 583 


\ fal 
τὰ δ᾽ ἄλλα τούτῳ δεῦρ᾽ ἐποπτεῦσαι λέγω 
U > lel 3 
ξιφηφόρους ἀγῶνας ὀρθώσαντί μοι, 


where Conington supposes Hermes himself to be intended. 


208 THE JOURNAL OF PHILOLOGY. 


κράτη may mean either victory, plural for singular (cf. Sup- 
plices 962, and for the meaning Isthmia vu. 5 ἀέθλων κράτος 
“victory in the games” Fennell), or feats of strength, mighty 
deeds. To conclude then, I propose this rendering on the 
following grounds: it gives a sense suitable to the context, 
it is not suggested as an explanation by Euripides or Dionysus, 
it may be thought to improve the rhythm, and lastly it bears 
out the charge of ἀσάφεια brought against Aeschylus by 
Euripides. 


129—132 
κἀγὼ χέουσα τάσδε χέρνιβας βροτοῖς 
λέγω καλοῦσα πατέρ᾽ ἐποικτείροντ᾽ ἐμὲ 
φίλον τ᾽ ᾿᾽Ορέστην πῶς ἀνάξομεν δόμοις. 


Here I would read κελεῦσαι for καλοῦσα, translating “ And 
I pouring these libations to mortals bid my father in pity for 
me and loved Orestes command us how we are to rule the 
house,” 


152—154 ἵετε δάκρυ καναχὲς ὁλόμενον 
ὀλομένῳ δεσπότᾳ, 
πρὸς ἔρυμα τόδε κακῶν κεδνῶν τ΄. 


If this chorus should be divided, as seems probable, into 
στροφὴ and “ἀντιστροφὴ, ἔρυμα will not scan. Consequently 
Hermann reads πρὸς ἕρμα yas, and Weil Hartung and Schoe- 
mann all make changes. Paley, too, thinks that ἔρυμα is 
corrupt. I do not believe that the expression “a barrier of 
good and evil” would convey much meaning to a Greek, and 
sense and metre will both be improved if we read πρὸς εὔγματα 
τάδε, a very slight change, translating “ Let fall the tear, with 
reference to these prayers for good and evil.” LElectra’s prayer 
is divided into two parts, (1) a prayer for good for herself and 
her brother, (2) a prayer for evil on her enemies. This is 
obvious even if we omit the lines 145—6 which distinctly 
assert it. The chorus is intended as a résumé of Electra’s 
speech and naturally emphasises the fact of its division into 
two main parts, 


AESCHYLEA, 209 


277—78 τὰ μὲν yap ἐκ γῆς δυσφρόνων μειλίγματα 
βροτοῖς πιφαύσκων εἶπε, τάσδε νῷν νόσους. 


This passage has been variously emended. Retaining νῷν 
to whom can we refer it? To Orestes and Electra? Impossible. 
Everywhere Orestes alone is considered responsible. I believe 
that Orestes and Pylades are here referred to. The oracle 
would be delivered to both, even though Orestes alone was 
concerned in the consequences. There is a further difficulty in 
δυσφρόνων μειλίγματα for which I should prefer to read 
δύσφρον᾽ ὧν μειλίγματα translating, “ For the hostile influences 
springing from the earth, propitiations of which in his oracles 
he told to men, these [he told] to us as diseases, namely 
leprosies.” Thus the oracle threatens diseases to which all men 
are liable, but which they can avert by propitiations which will 
not avail Orestes. 


283—285 ἄλλας 7 ἐφώνει προσβολὰς ᾿Ερινύων 
ἐκ τῶν πατρῴων αἱμάτων τελουμένας 
ὁρῶντα λαμπρὸν ἐν σκότῳ νωμῶντ᾽ ὀφρύν. 


These lines are usually transposed or emended. Would it 
be possible to take ὁρῶντα as the accusative after ἐφώνει and 
προσβολὰς ᾿Ερινύων as governed by ὁρώντα I should then 
translate—*“ And he spoke of me seeing clearly, though asleep, 
other onsets of the Furies brought to pass from a father’s blood.” 
ἐν σκότῳ νωμῶντ᾽ ὀφρύν I take to be merely a poetical equival- 
ent for sleeping. 


481—2 κἀγὼ πάτερ τοιάδε σοῦ χρείαν ἔχω 
φυγεῖν μέγαν προσθεῖσαν Αἰγίσθῳ υ-- 


(The last word of the line is lost in the Mss.) For this we 
have the following scholium ὥστε φυγεῖν tas ἐπιβουλὰς 
Αὐγίσθου τιμωρησαμένην αὐτόν. Paley laying stress on this 
proposes to read 
κἀγὼ πάτερ τοιάδε' σοῦ χρείαν ἔχω 
φυγεῖν με γῆν προσθεῖσαν Αἴγισθον δίκῃ. 
But Orestes had prayed “Father, give me the lordship over 


your house.” Is it not then a contradiction for Electra to say 
Journal of Philology. vou. xvi. 14 


210 THE JOURNAL OF PHILOLOGY. 


“T too make a like request, namely that I may get safe out of 
the land.” I should read φυγεῖν μ᾽ ἄγην translating “I ask of 
thee that having punished Aegisthus I may escape all malice,” 
1.6. not only the odium that the deed may excite among men, 
but also the νέμεσις of the gods. The scholiast wrongly 
supposed the dyn to refer to Aegisthus and so paraphrased it 
tas ἐπιβουλὰς Αἰγίσθου (Stephanus gives invidia φθόνος as 
equivalents to ayn). 


639—648 τὸ δ᾽ ἄγχι πλευμόνων ξίφος 
διανταίαν ὀξυπευκὲς οὐτᾷ 
διαὶ Δίκας" τὸ μὴ θέμις γὰρ οὐ 
λὰξ πέδοι πατούμενον 
τὸ πᾶν Διὸς 
σέβας παρεκβάντες οὐ θεμιστώῶς. 

Paley’s translation “for the irreligion of one who has law- 
lessly transgressed the majesty of Zeus is not trampled by it 
under ground (1.6, is not slighted)” is quite impossible. The 
participle πατούμενον cannot stand for the indicative. 

I propose to read 

TO μὴ θέμις yap 

ov λὰξ πέδοι πατούμενον 

mot ἂν Διὸς 

σέβας παρεκβὰν πέσοι θεμιστῶς 
and translate “for unrighteousness not being trampled under 
foot will one day after transgressing the majesty of Zeus fall 
by a righteous doom.” By the change of ποτ᾽ dy for τὸ πᾶν 
and πέσοι for τες ov sense and construction are made plain, 
and we get a forcible contrast between τὸ μὴ θέμις and θεμισ- 
Tos instead of the intolerable repetition τὸ μὴ θέμις, οὐ θεμισ- 
tos. It is possible of course to retain τὸ πᾶν: in that case 
πέσοι expresses a wish, “as for unrighteousness...when it has 
transgressed all the majesty of Zeus may it fall by a righteous 
doom.” 

In the antistrophe the Mss. give 


τέκνον δ᾽ ἐπεισφέρει Sipace 
αἱμάτων παλαιτέρων 


AESCHYLEA. ΦΙΊ 


τείνει μύσος 
χρόνῳ κλυτὴ βυσσόφρων "Ἐρινύς. 

Relying on the scholiast’s note (ἐπεισφέρει δὲ τοῖς οἴκοις 
τέκνον παλαιῶν αἱμάτων, 6 ἐστι, τίκτει ὁ φόνος ἄλλον φόνον) 
I propose to read 

τέκνον © ἐπεισφέρει δόμοισιν 
αἷμ᾽ αἱμάτων παλαιτέρων᾽ 
τίνει μύσος 
χρόνῳ κλυτὴ βυσσόφρων ᾿Ἐρινύς. 
1014 νῦν αὐτὸν αἰνῶ νῦν ἀποιμώξω παρών. 

If we keep this reading we must suppose αὐτὸν to refer to 
Agamemnon. Against this there are two strong objections, 
(1) Orestes has said no word in praise of Agamemnon, (2) there 
is nothing to justify our referring αὐτὸν to Agamemnon unless 
indeed it be πατροκτόνον in the line below. But further a 
reference to Agamemnon does not harmonise with the context. 
The three previous lines speak of the ¢apos—the line that 
follows also refers to it. It seems therefore strange that a line 
should be inserted between the two on an entirely different 
subject. 

I believe that αὐτὸν must therefore be wrong (cf. 1. 991 
where αὐτὸν is obviously a mistake for αὐτὸ) and should read 


νῦν αὐτὸ φαίνω, viv ἀποιμώξω παρών 
“Now I bring it to the light, now I bewail it standing on the 
spot.” At the same time it must be admitted that Ag. 1543—5 
ἢ σὺ τόδ᾽ ἔρξαι τλήσῃ, κτείνασ᾽ 
ἄνδρα τὸν αὑτῆς ἀποκωκῦσαι ; 
τίς δ᾽ ἐπιτύμβιος αἶνος ἐπ᾽ ἀνδρὶ θείῳ ; 


may be thought to justify the Mss. reading here. 


1044—5 ἐγώ δ᾽ ἀλήτης τῆσδε γῆς ἀπόξενος 
ζῶν καὶ τεθνηκὼς τάσδε κληδόνας λιπών. 

Here the remedy is simple and I think certain. I read ζῶ 
for ζῶν and translate “But I a wanderer and banished from 
this land live, though I be dead, in these stories which I leave 

14—2 


283 THE JOURNAL OF PHILOLOGY. 


behind me.” The change from ζῶν to ζῶ is the slightest 
possible and we cannot wonder, if ζῶ καὶ τεθνηκώς was the 
original, that it should have been changed to ζῶν καὶ τεθνηκώς. 
And now as to the meaning. 

If it be said that this opposition between life and death 
is more in the manner of Euripides than Aeschylus it will be 
sufficient to refer to a passage in this same play which strikingly 


illustrates the proposed correction. I give it here in full, 
1, 503—6 


μὴ ᾿ξαλείψῃς σπέρμα Πελοπιδῶν τόδε" 
οὕτω γὰρ οὐ τέθνηκας οὐδέ περ θανών" 
παῖδες γὰρ ἀνδρὶ κληδόνες σωτήριοι 
θανόντι. 


It is not merely that there is a general resemblance between 
the two passages: the very words correspond. In the one case 
it is children who are κληδόνες σωτήριοι to the father, for so 
he dies not though he be dead: in the other it is κλήδονες, 
but not children, which will keep alive the memory of Orestes 
in spite of death. 


Agamemnon 182—3 
δαιμόνων δέ που χάρις βιαίως 
σέλμα σεμνὸν ἡμένων. 

βιαίως is nonsense, and has been variously emended. I 
propose to read δ᾽ aids which gives an excellent sense by the 
change of one letter only. 

The meaning of the whole passage is shortly this “It is to 
Zeus that songs of victory must be sung: for it is due to the 
grace of the gods that the Trojans have learnt wisdom by 


suffering.” In support of the reading proposed I may quote 
Supplices 573, 605 


Ζεὺς αἰῶνος κρέων ἀπαύστου... 
v " ς 7 ͵7] / 
οὔτινος ἄνωθεν ἡμένου σέβει κάτω. 


1227—30 νεῶν τ᾽ ἔπαρχος ᾿Ιλίου τ᾽ ἀναστάτης 
οὐκ οἶδεν οἷα γλῶσσα μισήτης κυνὸς 
λέξασα κἀκτείνασα φαιδρόνους δίκην 

ἄτης λαθραίου τεύξεται κακῇ τυχῇ. 


AESCHYLEA. 213 


The difficulty of this passage, on which much critical ingenuity 
has been spent, is well known. Before accepting Madvig’s 
emendations φαιδρὸν οὖς and δήξεται, which make a tongue 
stretch out a cheerful ear, it may be worth while to consider if 
we cannot get a better sense from the words as given in the 
Mss. Keeping to the Mss. reading I propose a new rendering 
of the passage. It is I believe generally allowed that the words 
λέξασα κἀκτείνασα contain some reference to Agamemnon’s 
answer to Clytemnestra, 1. 914—916, 


Λήδας γένεθλον δωμάτων ἐμῶν φύλαξ 
ἀπουσίᾳ μὲν εἶπας εἰκότως ἐμῇ: 
μακρὰν γὰρ ἐξέτεινας. 


This being so, it is ὦ prior not improbable that the following 
words (φαιδρόνους δίκην ἄτης λαθραίου) may also have some 
connection with Clytemnestra’s speech or Agamemnon’s answer. 
Now, looking to the closing words of Clytemnestra’s speech we 
read 


910—3 εὐθὺς γενέσθω πορφυρόστρωτος πόρος 
ἐς δῶμ᾽ ἄελπτον ὡς ἂν ἡγῆται δίκη" 
\ > ΜΝ \ 3 ivf 7 
τὰ δ᾽ ἄλλα φροντὶς οὐχ ὕπνῳ νικωμένη 
θήσει δικαίως σὺν θεοῖς εἱμαρμένα. 


There is treachery in this proffered justice of which Aga- 
memnon thinks not, and well may Cassandra say that the king 
“knows not the true meaning of the words of the accursed 
hound.” It is to this δίκη and this δικαίως that Cassandra refers 
when she speaks of the tongue proffering justice. But Cassan- 
dra does more than this—she defines the nature of this justice— 
it is δίκη ἄτης NaPpaiov, a justice which consists in a dark 
crime, There remains a slight difficulty as to the meaning of 
ἐκτείνασα. I can find no instance of ἐκτείνω with the meaning 
of holding out or proffering, but the somewhat similar sense 
which it has in the words πρὸς κέντρα κῶλον ἐκτενεῖς seems at 
least to prove that it was not exclusively used in the sense of 
“to prolong”; and it is difficult to believe that ἐκτείνω could 
not be employed with the meaning “to proffer.” I therefore 
take δίκην to be the object of ἐκτείνασα and translate thus: 


214 THE JOURNAL OF PHILOLOGY. 


“The captain of the fleet and the destroyer of Ilium knows not 
what words the tongue of the lustful she-hound has spoken, 
with a gay heart proffering a justice which is a hidden crime, 
and will accomplish with an evil success.” 

The repetition of δίκη (1. 911) δικαίως (1. 913) may be 
thought to justify the translation “dwelling on a justice which 
is a crime,” but I prefer to give ἐκτείνω the simple meaning “to 
proffer.” 


Agamemnon 1625—7 


γύναι σὺ τοὺς ἥκοντας ἐκ μάχης νέον 
οἰκουρὸς εὐνὴν ἀνδρὸς αἰσχύνουσ᾽ ἅμα 
> \ Ὁ ’ > > / , 
ἀνδρὶ στρατηγῷ τόνδ᾽ ἐβούλευσας μόρον ; 


Many corrections of this passage have been proposed ; τοῦ γ᾽ 
ἥκοντος (Auratus), τοῦ δ᾽ ἥκοντος (Stanley), τῷδ᾽ ἥκοντος (Tyr- 
whitt), γνοῦσ᾽ ἥκοντος (Heusde), but none of these appear 
satisfactory. I propose to read 


/ a » SW 
γύναι σὺ τοῦ σ᾽ ἥκοντος. 


This correction explains the corruption, for if τοῦ σ᾽ were written 
tous, ἥκοντος would inevitably be changed to ἥκοντας, though 
the passage is thus reduced to nonsense. The accusative after 
ἥκω is common in Aeschylus, ef. Prom. Vine. 749 ἐνθ᾽ ᾿Αμαζόνων 
στρατὸν ἥξεις. We find also ws αὐτὸν ἥξοι μοῖρα in Sophocles, 
The repetition of σὲ after συ is not otiose but has a peculiar 
force, “ Agamemnon came home from battle, and came to you: 
and yet you killed him.” I should translate thus: “ Wife thou of 
him who is come but now from battle to thee, hast thou the 
keeper of his house while bringing shame upon thy husband’s 
bed at the same time plotted for a warrior husband this death?” 
—(or we might take ἅμα to mean “in concert with Aegis- 
thus”). Mr A. Sidgwick objects to these lines on the ground 
that Clytemnestra does not come on the stage till 1. 1654. I 
see no reason for this supposition, When Clytemnestra 
finished speaking, 1. 1576, Aegisthus undoubtedly came upon 
the stage, but it seems to me preposterous to suppose that 
Clytemnestra at once left it—surely such an action on her part 
would be the reverse of a compliment to her lover. On the 


AESCHYLEA. 215 


other hand the appeal to Clytemnestra seems to me not inap- 
propriate. The chorus first address a dignified remonstrance to 
Aegisthus, 1. 1612—16. He answers by furious invective and 
threats. In disgust the chorus turn to Clytemnestra and 
address a last appeal to her. But Aegisthus, provoked at this 
want of respect to himself, answers for her with renewed threats. 
All this seems to me perfectly natural, and involves no violent 
change. 

One passage from the Eumenides may serve to ‘conclude 
these comments. 


751—754 
πεμπάζετ᾽ ὀρθῶς ἐκβολὰς ψήφων, Eévor, 
τὸ μὴ ἀδικεῖν σέβοντες ἐν διαιρέσει. 
γνώμης δ᾽ ἀπούσης πῆμα γίγνεται μέγα, 
βαλοῦσά τ᾽ οἶκον ψῆφος ὥρθωσεν pia. 

This last line has always proved a stumbling-block. The 
scholiast explains ἡ ψῆφος δὲ βαλεῖν τε καὶ ὀρθῶσαι οἶκον 
δύναται----ὃι ὙΘΤῪ feeble sense even if it could be extracted from 
the words. Others propose to take μία ψῆφος βαλοῦσα in the 
sense of εἷς ψηφιζόμενος βαλὼν, which seems almost impossible. 

I believe that the line has been interpolated with the ob- 
vious purpose of obtaining a false symmetry in the passage. It 
has been thought desirable to provide a cheerful antithesis for 
πῆμα μέγα, and in consequence we have got this verse which, 
though intended to serve a plausible purpose, is really useless 
or worse. Apollo has said “If a vote is absent, much mischief 
may be caused.” The statement is perfectly general and in- 
cludes a vote of condemnation as well as a vote of acquittal. 
We want nothing more, the sense is complete: “Count the 
votes carefully, the absence of a vote may make all the differ- 
ence.” But some one who misunderstood the passage stumbled 
at πῆμα, and by way of a cheering contrast produced this 
singularly unfortunate line. 


HUGH MACNAGHTEN. 


ON THE DATE OF CALPURNIUS SICULUS. 


THE interesting question of the date of this Roman bucolic 
poet, Calpurnius Siculus, has been revived by the publication 
of Mr Keene’s excellent edition of his eclogues. It would be 
determined if scholars could agree respecting the identity of 
the young emperor celebrated by him, whose accession at an 
early age, coincident with the appearance of a comet, closed 
a period of civil war and general insecurity ; whose predecessor 
had been a scourge to his subjects in general and to the senate 
in particular; and who himself exhibited spectacles in the 
amphitheatre. Until the appearance of Sarpe’s dissertation 
in 1819, this emperor was universally identified with Carinus, 
a traditional opinion which will not bear a moment’s exami- 
nation. Sarpe’s view that he was Nero has been adopted by 
most subsequent editors and literary historians, notwith- 
standing the dissimilarity of the circumstances attendant upon 
Nero’s accession to those described by Calpurnius. In an 
article in the Encyclopaedia Britannica (1876) I ventured to 
point out the difficulties of the Neronian hypothesis, and to 
suggest that Calpurnius’s young Emperor was Gordian the 
Third, who came to the throne at the age of thirteen or sixteen, 
succeeding a ferocious tyrant especially inimical to the senate, 
and closing a period of civil strife; whose accession nearly 
coincided with the appearance of a comet; who exhibited 
games in the amphitheatre; who, like Calpurnius’s Emperor, 
was guided by a virtuous minister, and whose personal character 
and literary tastes agree with the description of the poet. For 
a fuller discussion of the subject I must refer the reader to 
my article, begging him to correct the atrocious misprint of 
Maximus for Maximin, and content myself here with adducing 


THE DATE OF CALPURNIUS SICULUS. 217 


some additional considerations, suggested by Mr Keene’s argu- 
ment on behalf of the Neronian theory, which had previously 
escaped my attention. 

Mr Keene identifies the comet mentioned by Calpurnius 
(L, 77, sq.) with “that which appeared at the end of the reign 
of Claudius, and beginning of that of Nero.” Now we know 
that this comet appeared while Claudius was still on the 
throne, for Suetonius enumerates it among the praesagia 
mortis ejus. But we do not know that it was still visible at 
Nero’s accession, and, as we shall see immediately, there is 
strong reason to believe that it was not. Pliny certainly does 
not say, as Mr Keene understands him, “that the comet was 
visible for a considerable time.” His words are Sidus...princi- 
patu ejus adsidwum prope ac saevum. This of course does not 
indicate that one and the same comet was visible during the 
greater part of Nero’s reign, but that appearances of comets 
were frequent at the period. Granting, however, that this 
comet may have been visible for a long time after Calpurnius 
wrote, we know from himself that when he composed, or more 
probably published, his poem, it had only been seen for twenty 
days: a space of time surely insufficient for the prognostic- 
ation of the Emperor’s death, the fulfilment of the prophecy, 
the installation of his successor, and the composition and public- 
ation of so finished and elegant.a poem. It further deserves 
to be noted that Calpurnius’s description of his comet is quite 
at variance with Pliny’s. Pliny calls it sedus terrificum. Cal- 
purnius expatiates on its mild lustre, its perfect orb, free from 
gaps or rents (sine vulnere plenus, contrast Webster's “rough- 
bearded comet” and Milton’s “horrid hair”); its general un- 
likeness to comets ominous of ill :— 


“Numquid utrumque polum, sicut solet, igne cruento 
Spargit et ardenti scintillat sanguine lampas? 
At quondam non talis erat.” 


In fact, however, there is very good reason to believe that 
the comet which was supposed to have announced the death 
of Claudius preceded that event by four months. There is 
no evidence, so far as I know, of a comet having appeared in 


218 THE JOURNAL OF PHILOLOGY. 


October, A.D. 54: but in Williams’s “ Observations of Comets 
extracted from the Chinese annals” (1871), one is recorded to 
have appeared in June of that year. (Mr Williams, by a slip 
of the pen, says A.D. 55; but if, as he states, it appeared in 
the thirtieth year of a Chinese cycle commencing with A.D. 
25, the year of its appearance must have been A.D, 54.) This 
may with the greatest probability be identified with the comet 
deemed to have prognosticated the death of Claudius. Comets 
visible in China appear to be generally visible in Europe also: 
see Williams, passim, and Fleming’s Travels in Manchu Tartary, 
where (p. 36) there is a representation of the appearance, as 
observed in China, of the great comet that startled Europe in 
July, 1861. There is, therefore, good reason for believing that 
the comet of 54 appeared in June, not in October, while we 
know from the Chinese observations (Williams, p. 21) that one 
was seen in September 238, about two months after Gordian 
had ascended the throne. We shall now see that the first 
eclogue of Calpurnius was almost certainly written in this 
very month of September, and not in October, as it must have 
been if Nero had been the emperor celebrated by him. 

He marks the period of composition with great precision. 
It is declinis aestas, “waning summer,” as Mr Keene renders 
it, “early autumn, before the summer heats are past.” The 
heat is so great that a cap is not sufficient protection :— 


“Torrida cur solo defendimus ora galero ?” 


This squares very ill with the month of October, even in 
Italy. But there is a closer indication of date. The vintage 
has reached that early stage when the must is expressed from 
the gathered clusters :— 


“Quamvis et madidis incumbant praela racemis, 
Et spument rauco ferventia musta susurro.” 


The earliest date for the commencement of the vintage in 
Italy mentioned by Columella is Sept. 2 for hot districts: the 
latest Oct. 14, but only frigidis regionibus. Most generally 
(pluribus regionibus) he says, it takes place between Sept. 17 
and Sept. 28 (De Re Rustica, lib. XL, cap. 2). The last day of © 
the holiday allowed for the purposes of the vintage was Oct. 15. 


THE DATE OF CALPURNIUS SICULUS. 219 


Now the young prince in whose honour the eclogue is thus 
composed at an early period of the vintage is already on the 
throne. (forsitan Augustas feret haec Meliboeus ad aures.) 
He cannot therefore be Nero, whose accession did not take 
place until Oct. 13, but may very well have been Gordian, who 
became Emperor A.D. 238; in September of which year, and 
therefore at the usual time of the vintage, a comet appeared 
which was visible for forty-one days (Williams, and article in 
Encyclopaedia Britannica). 

The very difficult passage, Hcl. 1., 84—88, is thus rendered 
by Mr Keene, “Aye, for a very god will take on his strong 
shoulders the weight of the Roman empire, so unshaken that 
neither will a thundering crash be heard as the world passes 
to its new ruler, nor will Rome decree that the dead should 
be deified in accordance with their deserts, before that the be- 
ginning of the new reign can look back upon the close of the 
last.” This is an admirable translation, but it suggests the 
inquiry, why should the deification of the Emperor’s prede- 
cessor be so pointedly referred to when, according to the entire 
previous tenor of the eclogue, he had been so unworthy? and 
why should a single person be spoken of as Penates? Both 
difficulties are explained on the hypothesis of the identity of 
the Emperor celebrated by Calpurnius with Gordian the 
Third. Gordian was not the_immediate successor of the 
tyrannical Maximin: the ephemeral reign of two good Em- 
perors, Maximus and Balbinus, had intervened. Gordian’s own 
relatives, the first and second of the name, acknowledged as 
emperors by the Senate, had also perished nearly at the same 
time as Maximin himself. Either or both of these Imperial 
pairs might with great propriety be described as Penates. 

I learn from Mr Keene’s preface that Calpurnius has already 
been referred to the age of Gordian by the late Mr Greswell. 
I was not aware of this when I wrote in the Encyclopaedia 
Britannica, nor have I been able to ascertain in which of 
Mr Greswell’s writings the suggestion is made. 


R. GARNETT. 


NOTES ON JUVENAL. 


DuRING the two years which have elapsed since I printed 
the additions to my first volume, I have gleaned the following 
illustrations. 

I 13 ADSIDVO LECTORE on the ablative see Wopkens on 
Justin XVIII 2 2. 

116 p. 95 1. 5 fr. foot, read dictaturam. cf. with Sulla’s 
retirement that of Lydiades, Paus. vit 27 12. 

I 27 p. 99 f. (on purple) Lact. Iv 7 6 regiae dignitatis as- 
sumptae indumentum. Sagittar. on Justin I 3 2. 

I 84 NOBILITATE cf. ind. and the concrete use of ciuitas, 
paucitas. 

1 38 NocTIBvS Heins on Ov. a. a. It 38. 

» 99 BEATAE Suet. Dom. 33. Gron. on Liv. xxiv 8 § 3. 

» 51 LYCERNA Kopp on Martian. Capella ὃ 2 fin. p. 13. 

» 55 LENO dig. xLvu 5 29 pr. mariti lenocinium coercutt. 
§ 3 qui quaestum ex adulterio uxoris suae fecerit, plec- 
titur; nec enim mediocriter delinquit, qui lenocinium in 
uxore exercuit. 

1 57 VIGILANTI STERTERE NASO Ov. r. a. 499 saepe ego, ne 
biberem, volut dormire uideri. 

1 59 60 CARET OMNI MAIORVM CENSV Nep. Pel. 1 ὃ 4 in 
quibus Pelopidas pulsus patria carebat. Suet. Vit. 2 Quintus 
caruit ordine, cwm auctore Tiberio secernt minus idoneos sena- 
tores remouerique placussset. 

I 61 p. 112 1. 6 fr. foot read tr. v 6 10. 

, 62 SE IACTARET AMICAE Wopkens on Justin xxIv 53 nec 
minus ferociter se legatis quam inter amicos iactauit. 

I 68 GEMMAE Oy. P. 1110 7 sis licet oblitus pariter gemmae- 
que manusque (‘seal and handwriting’). 


NOTES ON JUVENAL. 221 


170 see the story of the poisoning of Britannicus Tac. an. XIII 
16 illic epulante Britannico, quia cibos potusque eius delectus ex 
ministris gustu explorabat, ne omitteretur institutum aut utriusque 
morte proderetur scelus, talis dolus repertus est. wmnoxia adhuc 
ac praecalida et libata gustu potio traditur Britannico ; dein, 
postquam feruore aspernabatur, frigida in aqua adfunditur 
uenenum, guod ita cunctos evus artus peruasit, ut wow pariter 
et spiritus raperentur. Cf. Justin x11149,. Macr. Sat. vir 6 6 
si uero aconitum ipsum cum uino tritum potur datum sit, 
nulla curatio a morte defendit. 7 tunc enim uinum natura fri- 
gidum admixtione sui frigus auxit uenent. 

I 86 GAVDIA DIscvRsvs hence Coripp. lust. 11 74 saltatus 
risus discursus gaudia plausus. 

I 97 FACIEM PRIVS INSPICIT not merely as a protection from 
fraud was examination of visitors enforced Suet. Vesp. cited on IV 
64 (p. 402). id. Claud. 35 salutatoribus scrutatores semper 
apposuit, et quidem omnibus et acerbissimos. DCass. LX 3 3 all 
visitors both men and women searched ‘ μή τι ξιφίδιον ἔχωσιν᾽. 
Spartian. Seu. 6 § 2. 

I 99 A PRAECONE VOCARI Martian. Capella § 63 Kopp tunc 
LTanus in limine militesque [ours ante fores regias constiterunt ; 
ingressuros etiam cunctus nominatim uocabat Fama prae- 
conans. 

1 100 TROIVGENAS Tac. an» .Iv 55 ne Ilienses quidem, cum 
parentem urbis Romae Troiam referrent, nisi antiquitatis gloria 
pollebant. XII 58 Nero, aet. 16 to gain a reputation for elo- 
quence, causa Iliensium suscepta Romanum Troia demissum et 
Iuliae stirpis auctorem Aeneam aliaque haud procul fabulis 
uetera facunde exsecutus perpetrat ut Ilienses omni publico 
munere soluerentur. Galba (Suet. 2) traced his pedigree to 
Tuppiter and Pasiphae ! 

1110 p. 340 1. 6 fr. foot, lemma NE not NEC. 

» 115 virtvs Kopp on Martian. Capella § 7 f. p. 33 a. 

» 120 121 DENSISSIMA LECTICA Ov. tr. v 10 19 densis- 
simus hostis. 

I 129 130 TRIVMPHALES...TITVLVS Suet. Dom. 15 e basi 
statuae triumphalis titulus eacussus ui procellae in monu- 
mentum proaimum decidit. 


222 THE JOURNAL OF PHILOLOGY. 


I 130 ARABARCHES already by Marcile (on Suet. Vesp. 6) 
identified with Tiberius Alexander. 

1135 OPTIMA SILVARVM hence Avian. 29 13 14 namque awh 
agrestem cupiens ostendere uitam | siluarum referens optima 
quaeque dabat. 

ΠῚ 9 RECITANTES Macrob. Sat. vil 3 7 carries the practice 
up to Cicero Octauius, qui natu nobilis uidebatur, Ciceront reci- 
tanti ait: ‘non audio quae dicis’. 

11 9 p. 174 Ov. P.1 5 57 58 gloria wos acuat ; vos, ut reci- 
tata probentur | carmina, Pieris inuigilate choris. I 5 37 39 
40 dic tamen, 0 tuuenis studiorum plene meorum, | ...ecquid ubi 
aut recitas factum modo carmen amicis, | aut, quod saepe soles, 
exigis ut recitent. 

mr 9 p. 174 1. 11 fr. foot, for ‘ib. (ie. ex Pont.) read 
trist. 

» 71 ESQVILIAS XI 51 (not 71). 

» 77 SCHOENOBATES Sid. c. xx 301. 

,, 84 85 NOSTRA INFANTIA CAELVM HAVSIT AVENTINI Oy, tr. 
Iv 8 25 tempus erat nec me peregrinum ducere caelum. 

lI 85 BACA SABINA see Der Oelbaum.- Eine kulturhisto- 
rische Skizze von A. Hedinger. Prag 1886. pp. 14. 

ΤΠ 104 IACTARE MANVS Quintil. cited on 99 (not 9). 

» 108 CREPITVM DABIT Lucr. VI 109 carbasus ut quondam 
magnis intenta theatris| dat crepitum malos inter tactata 
trabesque. Aug. serm. 28 4 (Mai). 

III 124 PERIERVNT TEMPORA Luc. Ix 233 perierunt tem- 
pora witae. 

tr 189 Sen. prou. 5 § 2 detestabilis erit caecitas, si nemo 
oculos perdiderit nisi cui eruendi sunt. itaque careant luce 
Appius et Metellus. Loss of sight in attempting to save a 
mother (Liban. Iv 739). 

1 142 Q@VAM MVLTA MAGNAQVE Macrob. Sat. 11 15 8 
quam multi magnique auctores. 

1 151 NON VNA VI 218. vit 218. Tyrrell in Classical Rev. 
1 51 a. Luc. τι 466 haud unum. 

mi 153—5 EXEAT...cvivs Luc. vill 493-4 exeat aula | qui 
uolt (read with Grotius uolet) esse pius. 

1π 186 cf. Paus. vii 20 3. 


NOTES ON JUVENAL. 223 


tr 190 RvINAM 197 INCENDIA Suet. Vesp. 8 deformis urbs 
ueteribus incendiis ac ruinis erat. 

tr 198 PoscIT AQVAM the passage of Cic. is imitated by 
Justin v 1 5 omnia Graeciae regna uelut ad exstinguendum 
commune incendium concurrunt. So xiii 5 4. Cf. Ix 3 5, 
XIV 5 6. 

111 209 210 VLTIMVS AVTEM AERVMNAE CVMVLYVS Maguire 
(Hermathena 1887 168) compares Oy. m. XIV 472 cumulum- 
que Capharea cladis. 

III 226 HORTVLVS HIC PVTEVSQVE BREVIS Ov. P. 18 45 46 
(hortos) quos ego nescio cut colui, quibus ipse solebam | ad sata 
fontanas, nec pudet, addere aquas. 60 et dare, iam sitiens quas 
bibat hortus, aquas. 

ΠΙ 230 EST ALIQVID Ov. P. π 8 9. 10 39 (cf. 1 55 sunt 
quiddam oracula uatum). 

III 238 DRVSO VITVLISQVE MARINIS Sen. const. sap. 17 1 
Chrysippus ait quendam indignatum, quod illum aliquis uer- 
uecem marinum dizerat. | 

111 260 OBTRITVM VVLGI PERIT OMNE CADAVER Suet. Nero 5 
(of Nero’s father) in uiae Appiae wico repente puerum citatis 
wwmentis haud ignarus obtriuit. 

m1 278 seq. Suet. Nero 5 (Nero’s father) Romae medio 
foro cuidam equiti Romano liberius twrganti oculum erutt. 

ΠῚ 280 MOX DEINDE Wopkens and A. Gronov on Justin 
13 4. 

Π| 285—287 see J. Miller, Die Beleuchtung im Alterthum. 
I. Die Beleuchtung bei den Griechen. Affschaffenburg 1885. 
pp. 57. 

111 297—301 see the story of Nero’s drunken frolics Tac. an. 
ΧΠῚ 25 e.g. wbi...quidam permissa semel licentia sub nomine 
Nerons multi proprus cum globis eadem exercebant, in modum 
captiurtatis nox agebatur ; Iuliusque Montanus senatorit ordinis, 
...congressus forte per tenebras cum principe, quia wi attempt- 
antem acriter reppulerat, deinde agnitum orauerat, quasi ea- 
probrasset, mori adactus est. 

P. 218 1. 7 for ‘Bc.’ read ‘A.D.’ 

Iv 15 Macrob. Sat. 11 12, now numbered m1 16 9. 

» 19 PRAECIPVAM CERAM see Rein Privatr. 806. Brisson, 


224 THE JOURNAL OF PHILOLOGY. 


Dirksen and lexx. under praeceptio, praecipio, praecipuus. Suet. 
Galba 5 sestertiwm namque quingenties cum praecipuum inter 
legatarios habuisset. Sid. ep. v1 12 (p. 354 f. Baret) illud autem 
debert tibi quodam, ut wrisconsulti dicunt, praecipui titulo, nec 
tuus poterit wre pudor infitias. 

Iv 21 SPECVLARIBVS Marcile on Suet. Tit. 10 p. 135 Burm. 

» 99 VENDERE MVNICIPES SILVROS Macrob. Sat. vir 3 6 est 
autem loedoria (λοιδορία) huius modi “ oblitusne es quia salsa- 
menta uendebas?’ 

Iv 57 PRAEDAM Macrob. Sat. 11 15 1 sed...ad praedae 
marinae transire luaum Liciniorum me nomen admonutt. 

Iv 59 PROPERAT Tac. h. 11 62 ex urbe atque Italia irri- 
tamenta gulae gestabantur, strepentibus ab utroque mari wtin- 
eribus. 

Iv 69 IPSE CAPI VOLVIT Mart. vill 78 11 12 nunc implere 
sinus securos gaudet et absens | sortitur dominos, ne laceretur, 
auis. 

Iv 71 Antiochus II was surnamed θεός, and so (according to 
prol. Trog. Pomp. 42) was Tigranes. 

Iv 71 DIS AEQVA POTESTAS Ov. tr. IV 8 52 aequantem su- 
peros emeruisse wrum. Iambl. uita Pyth. § 259 ἔσους μακά- 
ρεσσι θεοῖσιν. | 

IV 79 INTERPRES LEGVM SANCTISSIMVS Dempster on Coripp. 
Justin Iv 4. 

IV 81 CRISPI IVCVNDA SENECTVS Martian. Capella § 3 poetae 
...secutt caecutientts Maeonii suauiloquam senectutem. 

Iv 89 90 CONTRA TORRENTEM Archiv fiir lat. Lexikogr. 
Iv 25 26. 

Iv 90 91 LIBERA VERBA Ernesti on Suet. Vesp. 9. 

, 96 OLIM Justin ΧΧΧΥΙΠ 3 1. 

» 98 FRATERCVLVS GIGANTIS Symm. ep. I 3 3 Baws re- 
motis arbitris otiabar. eo postquam rumor adlatus est, terrae 
filios conuentre, oppido cauimus, ne sobriam solitudinem nostram 
sodalitas plebeia fuscaret. Blitter f. bayer. Gymn. XVI 235. 

Iv 108 109 AMOMO, QVANTVM VIX REDOLENT DVO FVNERA 
Ov. P. 1 9 51—53 tlle tabi exsequias et magni funus honoris | fectt 
et in gelidos versit amoma sinus, | diluit et lacrimis maerens 
unguenta profusis, 


NOTES ON JUVENAL. 225 


Iv 125 TRIVMPHI Mart. vill 78 3. 

,, 1386 VICIT SENTENTIA Justin ΧΠῚ 2 13. 

» 138 FALERNO Varro res human. ΧΙ in Macrob. Sat. 111 16 
12 ad uictum optima fert ager Campanus frumentum, F a- 
lernus uinum. 

V 14 ΙΝΡΥ͂ΤΑΤ Lue. vill 657 (not 567). Benecke on Justin 
XXXVIII 6 5. see esp. Tac. ἢ. τι 85 unde metus et ex metu consi- 
lium posse imputari Vespasiano quae apud Vitellium excu- 
sanda erant. 

V 36 37 QVALE VINVM THRASEA HELVIDIVSQVE BIBEBANT 
BRYTORVM ET CASSI NATALIBVS Tac. ἢ. Iv 8 Marcellus said con- 
stantia fortitudine Catonibus et Brutis aequaretur Heluidius. 

Υ 50 pecocta Lamprid. Heliog. 23 8 montem niuium in 
wridario domus aestute fecit aduectis niuibus. Macrob. Sat. 
VIL 5 32 memineritis tamen lepido me conuiuio adesse, non 
anxio: nec sic admitto warietatem, ut luxum probem, ubi quae- 
runtur aestiuae niues et hibernae rosae. 

V 94 DEFECIT NOSTRVM MARE Macrob. Sat. 11 16 10 nec 
contenta illa ingluwes fuit maris sui coptis. See ind. noster. 

v 94—96 Macrob. Sat. vit 5 32 dum magis ostentuc quam 
usur serutur, siluarum secretum_omne lustratur et peregrina 
maria sollicitantur. 

Υ 99 MVRENA Aristoph. frogs 475, Poll. νι 65, and Varro in 
Gell. vi (v11)16 5 Tartesia. Vitellius (Suet. 13) in his vast platter 
(clipeus Mineruae) served, among other things, muraenarum 
lactes. Lamprid. Heliog. 23 8 muraenarum lactibus...in locis 
mediterraneis rusticos pauit. 

V 107 FACILEM SI PRAEBEAT AVREM Ov. P. 11 9 25 Iuppiter 
orante surdas si praebeat aures. Hor. ep. 1 1 40 st modo cul- 
turae patientem commodet aurem. 

V 114 ANSERIS IECVR in the clipeus Mineruae (Suet. Vit. 
13) scarorum iecinora. 

Vv 117 TVBERA XIV 7 (not xiv 1). 

,, 192 SIMILIS DIS beneficence divine Wiener Studien Ix 
(1877) 199. 

V 134 Ex Macrob. Sat. vil 3 21 ipse me mendicum fecit ex 
diwite. Justin xiv 4 3 wos me ex uictore uictum, wos me ex 
imperatore captiuum fecistis. 


Journal of Philology. vou. xvt. 15 


226 THE JOURNAL OF PHILOLOGY. 


vit 19 LAVRVM Kopp on Martian. Capella § 10. 

» 29 VENERIS MARITO Oy. P. Iv 10 55 duas terras, Asiam 
Cadmique sororem., 

Vil 29 HEDERIS Kopp on Martian. Capella § 10. 

» 45 QVANTI SVBSELLIA CONSTANT Suet. Vesp. 19 cnterrogatis 
palam procuratoribus, quanti funus et pompa constarent. 

Υ 49 LITVS VERSAMVS ARATRO Archiv fiir lat. Lexik. Iv 27. 

» 14 75 NON HABET...HABET Sen. prou. 3 § 1 potest enim 
miser dici, non potest esse. 

vil 78 CAPIVNT PLYS INTESTINA POETAE anth. Pal. x1 207 
1 καὶ τρώγεις ὅσα πέντε λύκοι, Taye. 

vil 91 ATRIA Sen. ep. 19 § 11 alioquin habebis conuiuas, quos 
ex turba salutantium nomenclator digesserit. errat autem qua 
amicum in atrio qguaerit, in conuiuio probat. Marcile on Suet. 
Vesp. 5. 

vil 92 Ov. tr m 507—510 quoque minus prodest, scaena 
est lucrosa poetae, | tantaque non paruo crimina praetor emit. | 
inspice ludorum sumptus, Auguste, tuorwm: | empta tibi magno 
talia multa leges. 

vit 121 VINVM, QVINQVE LAGONAE like apposition with 
milia? Benecke on Justin xxIv 7 9. 

Vir 126 QVADRIIVGES Suet. Dom. 13 Janos arcusque cum 
quadrigis et insignibus triwumphorwm per regiones urbis tantos 
et tot exstruxit, ut cuidam Graece inscriptum sit: arci (ἀρκεῖ). 

vil 155 CoLor Ov. tr. 1 8 (9) 63 64 ergo ut defendi nullo 
mea posse colore, | sic excusari crimina posse puto. Luc, 1x 
207 Haskins. Serv. Aen. Iv 613. 

Vil 162 QVIDQVID 1D EsT Ov. P. πὶ 3 73. 

» 163 DELIBERAT AN...AN in the sense ‘whether...or’ 
Justin xxiv 71 deliberauit an...an wero. 

vil 170 VETERES CAECOS Spartian. Hadr. 25 3 uenit et de 
Pannonia quidam uetus caecus ad febrientem Hadrianum 
eumque contingit. 4 quo facto et ipse oculos recepit et Hadri- 
anum febris reliquit. 

vil 173 AD PVGNAM QVI RHETORICA DESCENDIT AB VMBRA 
Ov. τ. a. 152 wade per urbanae splendida castra togae. tr. Il 
12 18 cedunt uerbost garrula bella fori. ἢ Iv 188 et fora 
Marte suo litigiosa wacent. P. Iv 6 29 Marte forensi, _ 


NOTES ON JUVENAL. 227 


vil 180 LvTo Suet. Vesp.5 mox, cum aedilem eum C. Caesar, 
succensens curam uerrendis uiis non adhibitam, luto dtussisset 
oppleri congesto per milites in praetextae sinum, non defuerunt 
qui interpretarentur, quandoque proculcatam desertamque rem 
publicam ciuili aliqua perturbatione in tutelam eius ac uelut in 
gremium deuenturam. 

Vu 183 RAPIAT CENATIO SOLEM Maguire (Hermathena 1887 
166) cites Stat. s. 11 4 98 sie att et speculum reclusit imagine 
rapta. Claud. nupt. Hon. et Mar. 106—108 (every wall is 
polished) speculi nec uultus egebat | tudicio. similis tecto mon- 
stratur in omni, | et rapitur quocumque uidet, dum singula 
cernit. 

vil 196 EDERE VAGITVS Ov. τὴ. xv 466-7 aut qui uagitus 
similes puertlibus haedum | edentem iugulare potest. Quintil. I 
1 21 futurus eloquentissimus edidit aliquando uagitum et ἰοφιὶ 
primum incerta uoce temptauit et haesit circa formas litterarum. 

vil 199 vENTIDIVS Bernegger on Justin XLII 4 10. 

» 202 CORVO RARIOR ALBO Ov. P. 1113 95 96 st dubitem, 
faueas quin lis, ὁ Maxime, dictis, | Memmonio cycnos esse 
colore putem. White blackbirds in Paus. vit 17 3 4. 

VII 206 GELIDAS CICVTAS Sen. prou. 3 12 male tractatum 
Socratem iudicas, quod illam potionem publice mixtam non aliter 
quam medicamentum immortalitatis obduxit et de morte dispu- 
tauit usque ad ipsam? male cum allo actum est quod gelatus 
est sanguis ac paulatim frigore inducto uenarum uigor con- 
stitit ? 

vit 210—212 METVENS VIRGAE ET CVI NON ELICERET A conj. 
Ov. tr. IV 10 65 66 molle Cumdineis nee inexpugnabile 
telis | cor mihi, quodque lewis causa moueret, erat. Sen. de 
ira 117 2 telum firmum perpetuwm obsequens nec anceps nec 
quod in dominum remitti posset. τι 29 2 est aliquis malignus 
et qui amicitias cohaerentes diducere uelit, cet. Suet. Vesp. 
5 secundum (ramum) praeualidum ac prolixum et qui 
magnam felicitatem portenderet.—B ind. Ov. met. 11 64 ardua 
prima wa est, et qua wix mane recentes | enituntur (Riese 
enitantur) equi. Sen. const. sap. 3 2 contigit illi res uulgaris 
et quae discitur ipsa iniuriarum assiduitate, patientia. ep. 
71 14 mens hebes et quae se corport addixit. Grammarians 


15—2 


733 THE JOURNAL OF PHILOLOGY. 


and commentators have not, to my knowledge, observed the 
corresponding use of adverbs with ut consecutive Pers. 2 8 haec 
clare et ut audiat hospes. Suet. Vit. 10 magnifice et ut 
ostenderet. | 

Vil 212 CITHAROEDI MAGISTRI Wiener Studien Ix (1887) 183. 

» 284 NVTRICEM ANCHISAE Macrob. Sat. vil 3 13 nee nega- 
uerim philosophos quoque incurrisse nonnumquam per indigna- 
tionem hoc genus scommatis. nam cum regis libertus ad nouas 
diuitias nuper erectus philosophos ad conuiuium congregasset et 
irridendo eorum minutulas quaestiones scrre se welle dixisset 
cur ex nigra et ex alba faba pulmentum unius coloris edatur, 
Aridices philosophus indigne ferens: ‘tu nobis’ inquit ‘ absolue, 
cur et de ngris et de albis loris similes maculae gignantur.’ 

vil 240 vicisvs Justin 11 412 duae his reginae...uicibus 
gerebant bella. Stat. s.1v 9 48—50 quid? si cum bene mane 
semicrudus | illatam tibi dixero salutem | et tu me uicibus 
domi salutes? Th. vim 422. xm 458. Manil. m 153 alter- 
nant genus et uicibus wariantur in orbem. π|ι 553 554 et 
modo dest aliud, modo adest uicibusque recedit | aut redit. 
671 cetera nunc urgent uicibus, nunc tempora cedunt. Known 
to Forcellini, but not to Riddell-White or Lewis-Short, this 
usage is confined to a few authors; sometimes we find alternis 
u., mutatis u.; most commonly in uicem (esp. frequent in Celsus), 
im uUrces. : 

ΥΙι 241 OCVLOS TREMENTES Sagittar. on Justin 13 2 (p. 128- 
9) oculorum lasciura. 

vil 242 Vitellius ruled by jockeys Suet. 12 pr. Tac. ἢ. τι 87. 

X 358 EXTREMVM INTER MVNERA (vol. I p. 466) Ov. tr. IV 
51 0 mihi dilectos inter pars prima sodales. v 12 25 26 tem- 
pore qui longo steterit, male curret et inter | carceribus missos 
ultimus zbet equos. P. 1 5 1 alle tuos quondam non ultimus 
inter amicos. IV 10 57 quos inter maximus omnes. VM. VI 
9 14 f. (of Marius) quem si inter miseros posueris, miserrimus, 
inter felices felicissimus, reperietur. 


JOHN E. B. MAYOR. 


NOTES ON MARTIAL, BOOK III. 


FRIEDLANDER’S edition has laid a solid foundation both for 
text and interpretation, but much remains to be done. A 
commentary, combining all that is valuable in Rigault, Rader, 
Hérault, Marcile, and other early scholars, together with much 
that is still in manuscript, or dispersed in adversarva and 
periodicals, would be a boon to students. Meanwhile I wish 
to shew my gratitude to Friedlander by printing such of my 
marginal notes as have escaped him. Some of them may have 
been anticipated, but for the present purpose, of furnishing a 
supplement to Friedlinder’s great work, it seems unnecessary 
to ransack many volumes in order to guard against repetition. 
I omit what I have already published (Classical Review 1 56— 
58) in a review of Friedlander, and begin with book 11, as I 
have at press an edition of the first two books. 

II 1 1 QVIDQVID ID EST ΥἹ 08 11. Iuv. γι 162 η. Ov. P. 
Il 3 73. 

lil 1 6 DEBET ENIM GALLYM VINCERE VERNA LIBER luv. Ix 
9—11 certe modico contentus agebas | vernam eguitem, conviva 
toco mordente facetus | et salibus vehemens intra pomeria natis. 

ΠῚ 2 Ellis on Catull. 11. Munro elucidations of Catullus 
p. 5. Becker-Rein Gallus 1° 381 --- 4, 

ΠΙ 2 4 CORDYLAS MADIDA TEGAS PAPYRO IV 10 2 pagina 
dum tangt non bene sicca timet. Catull. 95 7 8 at Volusi 
- annales Paduam morientur ad ipsam | et lawas scombris saepe 
dabunt tunicas. The shabby bundle of Greek writers on 
miracles, which Gellius (Ix 4 §§ 1—5) picked up for an old 
song at Brundisium, must have been on the way to the fish- 
monger’s. 

url 2 5 VEL TVRIS PIPERISVE SIS CVCVLLVS Ath. 374°” 
Anaxandrides the comic poet, being hot-tempered, when a play 


230 THE JOURNAL OF PHILOLOGY. 


of his lost the prize, did not recast it, as most did, but λαμβά- 
νων ἔδωκεν εἰς TOV λιβανωτὸν κατατεμεῖν. 

111 2 6 FAVSTINI FVGIS IN SINVM? ΒΑΡΙΒΤΙ. ΙΧ 5 1 nubere 
vis Prisco: non miror, Paula: sapisti. XI 106 4 transis hos 
quoque quattuor ? sapisti. 

ΠΙ 2 7 CEDRO PERVNCTVS Becker-Rein Gallus 11° 376. 

» » 9 10 PICTIS LVXVRIERIS VMBILICIS ET TE PVRPVRA 
DELICATA VELET Ellis on Catullus 22 7. Becker-Rein |. ο. 377 
—382. 

ul 4 7 8 POETA EXIERAT: VENIET, CVM CITHAROEDVS ERIT 
luv. vil 176 n. 

ΠῚ 5 2 AN SATIS VNVS ERIT? V 19 14 forsitan unus erit. 
Cf. Ovid there cited. 

ul 5 7 8 EST ILLI CONIVNX, QVAE TE MANIBVSQVE SINV- 
QVE EXcIPIET Anthol. Pal. x1t 208 5 (Strato to his book, 
destined for a boy) πολλάκι φοιτήσεις ὑποκόλπιον. 

111 6 4 LIBAT FLORENTES HAEC TIBI PRIMA GENAS Luv. III 
186 n. pp. 201, 378. 

ΙΠ 9 1 VERSICVLOS IN ME NARRATVR SCRIBERE CINNA 
Drager hist. Synt. 117° 454. Tac. Agr. 7 22. With this epigram 
cf. Lessing’s no. 86. 

m1 11 6 Munro on Lucr. Iv 1152 retains the comma after 
Thaida. 

ut 12 cf. x 49 and anthol. lat. 796 R (an epigram formerly 
printed as Martial’s iv 78; cf. Ellis in Journ. of Philol. x 191) 
where the point is just the same as here : 


ad cenam Varus me nuper forte uocaut : 
ornatus diues, paruula cena furt. 

auro, non dapibus decoratur mensa ; ministre 
apponunt oculis plurima, pauca gulae. 

tunc ego ‘non oculos, sed uentrem pascere uent: 
uel tu pone dapes, Vare, uel aufer opes’. 


111 12 3 RES SALSA EST BENE OLERE ET ESVRIRE I ὅθ 4 
tam male cum cenem, cur bene, Flacce, lawor? Lue. de mere. 
cond. 28 τὸ μὲν yap λιμῷ ξυνόντα καὶ νὴ Δία ye διψῶντα 
μύρῳ χρίεσθαι καὶ στεφανοῦσθαι τὴν κεφαλὴν ἠρέμα καὶ 
γελοῖον. 


NOTES ON MARTIAL, BOOK 11]. 231 


mi 12 4 5 QVI NON CENAT ET VNGVITVR, FABVLLE, HIC 
VERE MIHI MORTVVS VIDETVR luv. Iv 109 n. pp. 234, 410. 
Muret u. 1. πὶ 19. Ov. ἢ IV 853 arsurosque artus unxit. P.1 
9 52 53 dilwt et lacrimis maerens unguenta profusis | ossaque 
uicina condita texit humo. Auth. Pal. x1 8 1—3 μὴ μύρα, μὴ 
στεφάνους λιθίναις στήλαισι yapifou | μηδὲ TO πῦρ φλέξης" ἐς 
κενὸν ἡ δαπάνη.  ζῶντί μοι, εἴ τι θέλεις, χάρισαι. ib. 19 8 4 
μυρίσωμεν | αὑτοὺς, πρὶν τύμβοις ταῦτα φέρειν ἑτέρους. 

ΠῚ 13 1 Dvm Ellis on Catullus 44 9. 

» 15 2 CAECVS AMAT cf. 39 2 quam bene lusca widet! 
Iuv. Iv 114 n. pp. 235, 411. This epigram is translated by 
Sherburne. 

I 16 5 6 SED TE, MIHI CREDE, MEMENTO | NVNC IN PEL- 
LICVLA, CERDO, TENERE TVA from Hor. s. I 6 22 vel meritto, 
quomam in propria non pelle quiessem. The ass in the 
lion’s skin Luc. pise. 32. 

ΠῚ 17 5 DIGITOSQVE ADMITTERE VISA EST V 78 6 7 ponetur 
digitis tenendus unctis | nigra coliculus uirens patella. Ov. a. 
a. Il 755—6 carpe cibos digitis (est guidam gestus edendz) | 
ora nec immunda tota perungue manu. 

ut 17 6 MERDA 38. 1 83. 

» 19 1 PROXIMA CENTENIS OSTENDITVR VRSA COLVMNIS 
Firmic. math. vit 10 p. m. qui buxeas arbores tondens in 
beluas fingat. So Becker-Rein Gallus 1° 45 takes the bear 
here, not as Fr. of a bronze figure. 

lt 19 2 EXORNANT FICTAE QVA PLATANONA FERAE. XII 50 
1 daphnonas platanonas. Prop. IV (v) 8 75 tu neque Pom- 
pela spatiabere cultus in umbra. 

lit 20 1 CANIVS in the tract appended to Jerome (XI 2 333° 
ed. Vallarsi, Ven. 1771 4to) ‘ Valerius Rufino ne ducat uxorem ’, 
I find Canninus (sic) a Gadibus Herculis, poeta facun- 
diae lenis et iucundae, reprehensus est a Liuio Poeno (1), 
graut et uxorato historiographo, quod multarum gauderet amo- 
ribus, his uerbis, which are not worth repeating. 

111 20 4 AN QVAE NERONI FALSVS ASTRVIT SCRIPTOR? [05. 
ant. xx 8 ὃ 3 has an important passage on the falsification of 
Nero’s history by biassed historians. 

ΠῚ 20 7 AN IN COTHVRNIS HORRIDVS SOPHOCLEIS Fr. cites 


232 THE JOURNAL OF PHILOLOGY. 


Verg. and Ov. but not Iuv. vI 636 grande Sophocleo carmen 
bacchamur hiatu, nor Prop. 0 (111) 33 41 desine et. Ace 
conponere uerba cothurno. 

til 20 11 SPATIA ARGONAVTARVM see Obbar on Hor. ep. 1 
6 26. 

tr 20 15 TITINE THERMIS AN LAVATVR AGRIPPAE? Fr, 
cites Becker's topography 286 (read 686) in his ἢ. on 1 14 15. 
Becker omits Suet. Tit. 7 ἢ amphitheatro dedicato thermisque 
tuxta celeriter exstructis. ib. 8 (p. 239 48 Roth) ne quid popu- 
laritatis praetermitteret, nonnumquam in thermis suis admissa 
plebe lawit. Charisius (p. 93 28 K) finds an excuse for bringing 
the three baths into his grammar: deriuationis uero tanta est 
inaequalitas ut conprehendi non possit. nam cum sit Agrippa, 
mulierem Agrippinam dicimus. thermas vero Agrippinianas. 
item cum sit Nero ut leo, pelles leoninas, thermas Neronianas 
dicimus, item cum sint Titus et lupus similia, thermas Titinas 
ut pelles lupinas non dicimus, sed Titianas. 

ΠῚ 20 19 ΒΑΙΑΒ [πγ΄ ΠΙ 4 π. pp. 171, 346. χι 49}. xi 80n. 

ΠῚ 20 20 PIGER ΠΥΘΒΙΝΟ NAVCVLATVR IN STAGNO Plin. 
pan. 81 quantum dissimilis illu qui non Albani lacus otium 
Baianique torporem et silentium ferre, non pulsum saltem 
fragoremque remorum perpet. poterat, quin ad singulos ictus 
turpt formidine horresceret! itaque procul ab omni sono in- 
concussus tpse et immotus religato revinctoque navigio non secus 
ac piaculum aliquod trahebatur: foeda facies, cum popult 
Romani imperator alienum cursum alienumque rectorem velut 
capta nave sequeretur. Becker-Rein Gallus  142—152. Fried- 
lander 1° 108 6. 

ΠῚ 21 1 FRONTE NOTATA Hemst. on Luc. Timon 17 f. 
ὥσπερ στιγματίας δραπέτης πεπεδημένος. Luv. X 183 n. XIV 
24 n. (esp. Pont. vit. Cypr. 7 confessores frontium notatarum 
secunda inscriptione signatos). Wallon hist. de l’esclavage 1* 
288—291 has several examples, including this, of servile fidelity. 

11 23 1 2 OMNIA CVM RETRO PVERIS OBSONIA TRADAS, CVR 
NON MENSA TIBI PONITVR A PEDIBVS? anth. Pal. x1 11 3—6 
ἀλλ᾽ ἐκάλουν σε μόνον: σὺ δ᾽ ἔχων χορὸν οἴκοθεν ἥκεις | dp- 
χηστῶν, αὐτοῖς πάντα διδοὺς ὀπίσω. | εἰ δ᾽ οὕτω τοῦτ᾽ ἐστί, 
σὺ τοὺς δούλους κατάκλινον, ἡμεῖς δ᾽ αὖ τούτοις πρὸς 


NOTES ON MARTIAL, BOOK ΠΙ. 233 


πόδας ἐρχόμεθα. 205 οὐδὲν ἀφῆκεν ὅλως, Διονύσιε, λεί- 
ψανον Αὔλῳ | Εὐτυχίδης δειπνῶν, ἦρε δὲ πάντ᾽ ὀπίσω" | καὶ 
νῦν Εἰὐτυχίδης μὲν ἔχει μέγα δεῖπνον ἐν οἴκῳ, | wn κληθεὶς δ᾽ 
Αὖλος ξηροφαγεῖ καθίσας. 207 καὶ τρώγεις ὅσα πέντε λύκοι, 
Tape, καὶ τὰ περισσά, οὐ τὰ σά, τῶν δὲ πέριξ, πάντα δίδως 
ὀπίσω. πλὴν μετὰ τοῦ κοφίνου τοῦ πρὸς πόδας αὔριον ἔρχου, 
πρίσματα καὶ σπόγγον καὶ σάρον εὐθὺς ἔχων. 

lI 24 3 TVSCVS...HARVSPEX Bentley Works, ed. Dyce 11 
435. Τὰν. x1 62 n. 

ΠῚ 24 9 INGENS IRATIS APPARVIT HIRNEA SACRIS see a 
learned note of Casaubon on Suet. Aug. 82 pr. 

ΠῚ 24 14 quoted in anth. lat. 127 9 10 R solus vera probas 
aucundi uerba poetae: | DVM IVGVLAS HIRCVM, FACTVS ES IPSE 
CAPER. Gron. obs. 11 15 first rightly explained the line. 

it 25 1—3 51 TEMPERARI BALNEVM CVPIS FERVENS, 
FAVSTINE, QVOD VIX IVLIANVS INTRARET, ROGA, LAVETVR, RHE- 
TOREM SABINEVM from Macho in Ath. 580* (Diphilus log.) 


νὴ τὴν ᾿Αθηνᾶν καὶ θεοὺς ψυχρόν γ᾽, ἔφη, 
Γνάθαιν᾽, ἔχεις τὸν λάκκον ὁμολογουμένως. 
ἢ δ᾽ εἶπε, τῶν σῶν δραμάτων γὰρ ἐπιμελῶς 
> ’ \ ΨΟΝ \ , > / 

εἰς αὐτὸν ἀεὶ τοὺς προλόγους ἐμβάλλομεν. 


Cf. ib. 579°. More in Hermann (Becker’s Charikles 1° 192). 

11 26 cf. τι 43. Translated by Sherburne. 

» » 48 COR SOLYS HABES 27 4. 118 6 tune ego te credam 
cordis habere nihil. vu 78 4 habes nec cor, Papile, nec 
genium. ΧΙ 84 17 unus de cunctis animalibus hircus habet 
cor. Cic. de or. πὶ ὃ 61 hine discidiuwm illud exstitit quasi lin- 
guae atque cordis, absurdum sane et inutile et reprehendendum, 
ut alii nos sapere, αὐτὶ dicere docerent. Liv. perioch. 50 cum 
tres legati ad pacem inter Nicomeden et Prusiam faciendam a 
Romanis missi essent, cum unus ex tis caput multis cicatricibus 
sartum haberet, alter pedibus aeger esset, tertius ingenio socors 
haberetur, dixit Cato, eam legationem nec caput nec pedes nec 
cor habere. Suet. Nero 2 in hunc [Cn. Domitium] diait Ln- 
cinius Crassus orator, ‘non esse mirandum, quod aeneam barbam 
haberet, cui os ferreum, cor plumbeum esset’. id. Caes. 77 f. 
eoque arrogantiae progressus est, ut haruspice tristia et sine corde 


234 THE JOURNAL OF PHILOLOGY. 


exta quondam nuntiante, ‘futura’ diceret ‘laetiora, cum uellet ; 
nec pro ostento ducendum, si pecudi cor defuisset’. This 


passage of Caesar may have suggested Avian. 30 9—14 (where 
see Ellis) 


tunc domini captum [aprum] mensis dedit ille superbis, 
in uarias epulas plurima frusta secans. 

sed cum consumpti dominus cor quaereret apri, 
iumpatiens fertur quod rapuisse cocus, 

rusticus hoc vustam uerbo compescuit tram 
afirmans stultum non habuisse suem. 

‘nam cur membrorum demens in damna redisset, 
atque uno totiens posset ab hoste capi?’ 


Auson. epigr. 48, 49 (= 8, 60 Peiper) he who says reminisco, 
faceret cor, si cor haberet cet. 

ΠῚ 29 Jahn’s Persius p. 138. Τὰν. vit 140 anulus ingens. 
Petron. 67 eo deinde peruentum est, ut Fortunata armillas suas 
.. .detraheret...etuam periscelides resolut...Trimalchis iussit af- 
ferri...et ‘widetis’ inquit ‘ mulieris compedes’. 

mt 30 Becker-Rein Gallus n° 166—7. 

» » 2 QVID ROMAE, GARGILIANE, FACIS? [uv. ΠΙ 41 n. 
p. 851 quid Romae faciam? 

II 30 8 FVSCAE PENSIO CELLAE Becker-Rein I’ 15. 

» » 4 QVADRANS Becker-Rein 111° 105. 

» ol 4SVSTENTATQVE TVAS AVREA MASSA DAPES Schnei- 
dewin and Fr. read massa with mss. TQ, for mensa of the 
others, because ‘golden or gilt tables are nowhere mentioned’. 
Yet Musonius (Stob. fl. 85 20) says κλῖναι μὲν ἐλεφάντιναι καὶ 
ἀργυραῖ ἢ νὴ Ala χρυσαῖ, τράπεζαι δὲ παραπλησίας ὕλης. 
Mart. τχ 28 ὅ ut Mauri Libycis centum stent dentibus orbes | et 
crepet in nostris aurea lamna toris. XIV 89 (‘mensa citrea’) 
accipe felices, Atlantica munera, siluas: | aurea qui dederit 
dona, minora dabit, where the wooden tables are surely com- 
pared not with plate, but with gilt tables. sustentet (like 
sustinuisse in XIV 91 2, cf. Hor. Iuv.) is more naturally pre- 
dicated of a table, or the leg of a table, than of plate (though 
Pliny is cited for this latter use in lexx. sustineo). Nor is massa 
in itself very tempting. 


> Wellies δ ΠΥ. ΜΥ͂Σ] 


NOTES ON MARTIAL, BOOK III. 235 


Ill 82 4 NONDVM ERIT ILLA CANIS luv. X 271 n. 

» 99 2 LIBERTINA MIHI PROXIMA CONDICIO EST Plaut. trin. 
455. Ter. Andr. 79. hee. 241.  Cic. Phil. 2 ὃ 99 ἢ.  p. Cluent. 
§ 42. Plin. xxxvi§ 20. Iustin1 6 ὃ 6. x17 ὃ 8. Quintil. 
decl. 257 f. sic homint inter principes nostrae ciwitatis numerando 
coept bona esse condicio. 

ΠῚ 35 1 ARTIS PHIDIACAE TOREVMA CLARVM IV 39 4 solus 
Phidiaci toreuma caeli. xX 87 16 donet Phidiaci toreuma 
caelt. 

Ill 35 2 PISCES ASPICIS: ADDE AQVAM, NATABVNT cf. Ov. 
her. 13 153—4 


crede mihi, plus est quam quod videatur, vmago: 
adde sonum cerae, Protesilaus erit. 


1 36 cf. τ 70. 108. 

» » ὅ 6 LASSVS VT IN THERMAS DECIMA VEL SERIVS 
HORA TE SEQVAR X 70 13 14 balnea post decimam lasso 
centumque petuntur | quadrantes. 

It 38 cf. vitt 56 

» » 4 ATQVE ERIT IN TRIPLICI PAR MIHI NEMO FORO luv. 
ΧΙΠ 135 ἢ. 

Il 88 11 ATRIA MAGNA COLAM XII 68 1 2 matutine cliens, 
urbis mihi causa relictae, | atria, si sapias, ambitiosa colas. 
Becker-Rein Gallus 11° 161. 

mit 39 οὗ 15. vit 49. 

» 40 translated by Sherburne. 

» 41 Becker-Rein Gallus P 40. τι 322. 

,» 42 4 QVOD TEGITVR MAGNVYM CREDITVR ESSE MALVM 
magnum Schneidewin and Fr. from one ms. the maius of the 
others is much more forcible, as may be seen from Bernegger’s 
citations on Justin xiv 1 § 2. 

Itt 44 cf. 1 89. 

» » © NON DIPSAS see Obbar on Hor. ep. 1 17 30 
Ρ. 378. 

, 18 VIR IVSTVS, PROBVS, INNOCENS TIMERIS Plut. de 
garrul. 4 ἔ (p. 504) ὁ δ᾽ ἀδόλεσχος πανταχοῦ ληρεῖ, ἐν 
ἀγορᾷ, ἐν θεάτρῳ, ἐν περιπάτῳ, μεθ᾽ ἡμέραν, νύκτωρ" ἔστι δὲ 
θεραπεύων τῆς νόσου βαρύτερος: συμπλέων τῆς ναυτίας andé- 


236 THE JOURNAL OF PHILOLOGY. 


στερος" ἐπαινῶν τοῦ ψέγοντος ἐπαχθέστερος. ἥδιόν γέ τοι 
πονηροῖς ὁμιλοῦσιν ἐπιδεξιοῖς ἢ χρηστοῖς ἀδολέσχαις. 

ΠΙ| 45 6 NEC VOLO BOLETOS, OSTREA NOLO: TACE cf. Ix 35 
11 12 to a newsmonger tolle tuas artes ; hodie cenabis apud me | 
hac lege, ut narres, nil, Philomuse, noui. anth. Pal. x1 
10 τὸν τοῦ δευιπναρίου νόμον οἴδατε: σήμερον ὑμᾶς, | Αὖλε, καλῶ 
καινοῖς δόγμασι συμποσίου. οὐ μελοποιὸς ἐρεῖ κατακεί- 
μενος" οὔτε παρέξεις | οὔθ᾽ ἕξεις αὐτὸς πράγματα γραμ- 
ματικά. 

111 47 10 ILLIC CORONAM PINGVIBVS GRAVEM TVRDIS XIII 51 
(‘turdorum decuria’) texta rosis fortasse tibi uel diuite nardo, | 
at mihi de turdis facta corona placet. 

m1 50 cf. vi 48. 

»» 9» 9. DEPOSVI SOLEAS Becker-Rein Gallus 11° 165. Bot- 
tiger kl. Schr. 11 206. Ter. haut. 124 adsido: accurrunt serut, 
soccos detrahunt. 

mt 50 6 MENSA SECVNDA Gell. ΧΠῚ 11 ὃ 6 neque non de 
secundis quoque mensis, cuiusmodi esse eas oporteat, praecipit. 
his enim uerbis utitur: ‘ bellaria’ inquit ‘ea maxime sunt mel- 
lita quae mellita non sunt: πέμμασιν enim cum ππέψει societas 
infida’. quod Varro hoc in loco diait ‘ bellaria’, ne quis forte 
in ἰδία uoce haereat, significat id uocabulum omne mensae 
secundae genus. Ov. met. Ix 92 mensas, felicia poma, 
secundas. 

Ill 54 CVM DARE NON POSSIM, QVOD POSCIS, GALLA, ROGAN- 
TEM, | MVLTO SIMPLICIVS, GALLA, NEGARE POTES. Friedlander 
has not thought it worth while here to record Madvig’s con- 
jecture (aduers. IL 163) ‘fuit: quod dare non possim, cum 
poscis, Galla, rogantem, cet. (nihil in epigrammate obsceni 
est.) The last remark might hold good, but for the general 
character of the epigrams addressed to Galla. Fr. therefore 
here, as in the parallel epigram [1 25, is justified in giving 
certainly to rogo and nego, possibly to dv also, the common 
erotic sense. The text is sound: ‘since I cannot grant what 
you demand as the price of the favour which I ask, you might, 
Galla, say Wo more any ’. Her terms are only a round- 
about way of saying ‘ No’. 

ΠῚ ὅθ and 57 water sold at Ravenna of dig. XxxIV 11 si 


NOTES ON MARTIAL, BOOK III. 237 


alimenta fuerint legata, dict potest, etiam aquam legato inesse, 
st in ea regione fuerint legata οὐδὲ venumdari aqua solet. 
ib. 14 ὃ 3 in Africa and Egypt agua venalis est. Ancient cities 
were generally most copiously supplied with water by aque- 
ducts; for these Ravenna was afterwards indebted to Theoderic. 
Sidonius (ep. 1 5 p. 184 Baret) says of the town: huc cum 
peropportuna cuncta mercatur tum praecipue quod esurt com- 
peteret, deferebatur: nisi quod, cum sese hinc salsum portis 
pelagus impingeret, hinc cloacal pulte (a graphic touch!) fos- 
sarum discursu lintrium ventilata, et upse lentate languidus 
lapsus wumorts nauticis cusyidibus foraminato fund glutino sor- 
didaretur, in medio undarum sitiebamus: quia nusquam 
uel aquaeductuum liquor integer uel cisterna defae- 
cabilis uel fons irriguus uel puteus illimis. On the 
salt marshes see Sil. vir 601—2 quique graui remo limosis 
segniter undis | lenta paludosae proscindunt stagna Ravennae. 
Sid. ep. 1 8 (p. 195 B) in qua palude.indesinenter rerum omnium 
lege peruersa muri cadunt, aquae stant...sitiunt uiui, natant 
sepulti. A recent traveller in Spain states that in many parts 
of Andalusia wine is cheaper than water: in Rioja (district of 
Pamplona) mortar is mixed with wine (A. v. Seefeld in Ver- 
einsblatt fiir Freunde der natiirlichen Lebensweise, no. 90, 
Nordhausen 15 Nov. 1876, p. 1432). 

ΠΙ 58 cf. x 51. 79. 

» » 1 2 BAIANA VILLA...NON OTIOSIS ORDINATA MYRTETIS 
Bentley and Obbar on Hor. ep. 1 15 2—7 (Baias...myrteta). 
Cels. 11 17 pr. sudor etiam duobus modis elicitur: aut sicco 
calore aut balneo. siccus calor est et harenae calidae et laconict 
et clibant, et quarumdam naturalium sudationum, ubi e terra 
profusus calidus wapor aedificio includitur, sicut super Baias 
in myrtetis habemus (these last words super—habemus in Il 
21 p. 107 2 D have rightly been rejected as a gloss). 

ΠῚ 58 7 ET MVLTA FRAGRAT TESTA SENIBVS AVTVMNIS Ov. 
met. Ix 89—92 et nymphe, ritu succincta Dianae, | una min- 
strarum, fusis utrimque capillis, | incessit totumque tulit prae- 
diwite cornu | autumnum. XIV 660 suspiciens pandos au- 
tumni pondere ramos. οἷ. ὀπώρα. ; 

ir 58 14 NOMENQVE DEBET QVAE RVBENTIBVS PINNIS Luv. 


238 THE JOURNAL OF PHILOLOGY. 


ΧΙ 139 n. Sen. ep. 110 8 12. Polemii Siluii laterculus (p. 267 
Mommsen) fincopter. 

ΠῚ 58 16 PHASIANA Iuv. ΧΙ 139 n. Sen. ad Helu. 10 8 3. 
Suet. Vit. 3. 

ΠῚ 58 24 NON SEGNIS ALBO PALLET OTIO Copo Hor. s. 11 2 
20—22 tu pulmentaria quaere | sudando: pinguem witiis album- 
que neque ostrea | nec scarus aut poterit peregrina iuvare lagois. 
ef. Obbar on Hor, ep. 1 14 24. 

ΠῚ 58 31 CAPILLATI X 62 2. Petron. 27 pr. senem caluwm 
...anter pueros capillatos ludentem pila. — 

II 58 36 SOMNICVLOSOS GLIRES Non. p. 119 Laberius in 
Aquis caldis etiam hic me optimus somnus premit, ut oppri- 
nutur σ 115, Ammian. xxvill 4 ὃ 13 Val. poscuntur etiam in 
conuiuits aliquotiens trutinae, ut adpositr pisces et uolucres pon- 
derentur et glires, quorum magnitudo saepius replicata non sine 
taedio praesentium. 

ΠῚ 58 42 NEC AVARA SERVAT CRASTINAS DAPES MENSA Luv, 
XIV 129 n. 

ΠΙ| 58 47 FVREM PRIAPO NON TIMENTE Lact. Π 4 ὃ 2 (after 
citing Hor.s.1 8 1 seq. οὗ § 3, and ὃ 4 where he cites Verg. 
g.1v 110; see the commentators on both places) quis non sit 
tanto hoc custode securus? fures enim tam stulti sunt, ut Priapi 
tentiginem timeant; cum aues ipsae, quas terrore falcis aut 
inguinis abigt existimant, simulacris fabrefactis, id est hominum 
plane similibus, insideant nidificent inquinent. 

11 60 7 CLYNIBVS Plin. x ὃ 140. 

» » 8 IN CAVEA MORTVA PICA Petron. 28 f. super limen 
autem cauea pendebat aurea, in qua pica uaria intrantes salu- 
tabat. 

lI 61 translated by Sherburne. 

» 63 ef. II 7. 

» »» 8 BELLYS HOMO EST, FLEXOS QVI DIGERIT ORDINE 
CRINES X 65 6 tu flexa ntidus coma uagaris. 

ΠῚ 63 5 GADITANA Tuy. ΧΙ 162 n. 

»» 4 8 INTER FEMINEAS TOTA QVI LVYCE CATHEDRAS 
DESIDET Justin XxI 5 ὃ 4 non contentus...conspici in popinis 
lupanaribusque, sed totis diebus desidere. 

Π| 63 6 BRACHIA VOLSA X 65 8 leuis dropace tu cotidiano. 


NOTES ON MARTIAL, BOOK LIU. 239 


mt 63 13 QvID NARRAS? Caecil. ὅθ R. Ter. Andr. 461. 
eun. 672. Ph. 136. ad. 448. 559. cf. Ph. 685 quid ergo 
narras? quidegonarrem? 935 quaeso quid narras? ad. 
557 quid malum ‘bone wir’ mht narras? Andr. 477 ego quid 
narres nescio. 734 quid narres nescio. So quid narrat? 
qud istic narrat? quid ea narrat? The comic vocabulary has 
been too much neglected by expositors of Martial. 

lit 65 2 DE CORYCIO QVAE VENIT AVRA CROCO on awra see 
Munro Lucr. τὶ 851. 

ΠΙ 65 8 QVOD MADIDAS NARDO PASSA CORONA COMAS XI ὃ 
10 quod modo diuitibus lapsa corona comis. 

ΠῚ 66 3 LAVRIGEROS AGERES CVM LAETA TRIVMPHOS Claud. 
3 cons. Hon. 12 inter laurigeros aluerunt castra triumphos. 

II 67 4 CELEVMA Sid. ep. 11 10 (with Savaro p. 158). VIII 
12 (Sav. p. 544). Serv. Aen. mr 128. Eddius wit, Wilfr, 13, 
Hier. ep. 14 ad Heliod. § 10. See Faber’s thesaurus, 

ul 67 7 INTERIVNGIT Π 6 16. 

» 68 cf. 69. V 2.. priap. 1. 

» » 2 SCRIPTA INTERIORA Cic. ep. ad Qu. fr. m1 1 ὃ 18 
quod interiore epistula scribis. DL. v § 4 raiava, ὃς ἔνδον 
γέγραπται. 

III 68 5 DEPOSITO POST VINA ROSASQVE PVDORE Hor. ep. I 
9 11 12 quodsi | depositum laudas οὗ amici tussa pudorem. 
Mart. x 19 19—21 (in the comissatio) haec hora est tua, cum 
furit Lyaeus, | cwm regnat rosa, cum madent capilli: | tunc 
me uel rigidi legant Catones. 

ΠῚ 68 11 81 BENE TE NOVI I 112 2 nunc bene te noui, 
Hor. ep. 118 1 Obbar si bene te noui. Fr. cites Ov. 

ΠῚ 69 7 8 AT TVA, COSCONI, VENERANDAQVE SANCTAQVE 
VERBA A PVERIS DEBENT VIRGINIBVSQVE LEGI I 35 1—3 uersus 
scribere me parum seueros | nec quos praelegat in schola 
magister, | Corneli, quereris. 

π| 70 ef. 1 73. 

» 12 3 TIBI PANNOSAE DEPENDENT PECTORE MAMMAE 
Jahn’s Persius p. 174. moret. 34 Forb. cacens mammas. 

ΠῚ 72 4 SVLCOs VTERI Claud. in Eutr. 1 110 vamque aeuo 
laxata cutis sulcisque genarum | corruerat passa facies rugo- 
sior wud, 


240 THE JOURNAL OF PHILOLOGY. 


lil 74 2 3 NVMQVID TONSOREM, GARGILIANE, TIMES? QVID 
FACIENT VNGVES? this is explained by xiv 36 (‘ferramenta 
tonsoria’) tondendis haec arma tibi sunt apta capillis; | un- 
guibus hic longis utilis, illa genis. cf. Obbar on Hor. ep. 1 7 
50 51 adrasum quendam uacua tonsoris in umbra | cultello 
proprios purgantem leniter ungues. Theophr. char. 26 (29. 
Jebb) the oligarch goes about, ‘his hair daintily trimmed, his 
nails delicately pared’. 

ΠῚ 74 4 RESINA when Fr. says that resina was ‘offenbar’ 
(i.e. by inference from this passage) an ingredient in the de- 
pilatory mixture, he forgot the express testimony of Iuv. VII 
114 ἢ. resinata iwuentus. On the habit generally add Suet. 
Dom. 22. Lamprid. Heliog. 31 ὃ 7. Capitol. Pertin. 8 ὃ 5. 

ul 74 5 st PVDOR EsT Tuy. 1 154 n. pp. 198, 373. 

» 75 3 ERVCAE aphrodisiacs also in Verg. moret. 86. Plin. 
ΧΙΧ § 44. ἴυν. 1x 134. priap. 46 8. 47 6. 

ΠῚ 75 3 BYLBIQVE SALACES Apul. vil 12 aqua decoquant 
bulbos, qu Veneris usum quaerunt. 

ml 75 4 SATVREIA Ov. a. a. 11 415—6 sunt qui praecipiunt 
herbas, satureia, nocentes | sumere. iudiciis ista wenena mes. 
Classed with eruca by Cels. 1 31 as a diuretic. Magerstedt, 
Bilder aus der rém. Landwirthschaft (Sondershausen 1863) 248 
—251. 

ΠῚ 76 4 possis Hor. epod. 12 15. 

»» »» »» HECABEN proverbial example of old age 32 3. X 
90 3—6 where also she is contrasted with Andromache. ἴαν. 
x 272 τὰ priap. 12 1 quaedam wnior Hectoris parente. 
anth. Pal. x1 67 a crone of 800 years, τρυφερὴ Aat Κορων- 
exa Bn, | Σισύφου ὦ μάμμη καὶ Δευκαλίωνος ἀδελφή. 

ΠῚ 77 1 BAETICE 81. 

» » 1—3 TVRDVS...LEPVS...SECTAE QVADRA PLACENTAE 
v1 75 1 2 cum mittis turdumue mihi quadramue placentae, | 
siue femur leporis. ΧΠῚ 92 inter aves turdus, st guid me 
tudice certum est, | inter quadrupedes mattea prima lepus. 
luv. V 2 n. pp. 244, 417. 

ΠῚ 77 4 PHASIS XIII 72. 

» » ὅ CAPPARIN Ath. 497! 567°. Polyaen. str. Iv 8 ὃ 32 
p. 142 1 Woelfflin. Schneider on Colum. ΧΠῚ 3 ὃ 54 p. 597. 


NOTES ON MARTIAL, BOOK II. 241 


ur 79 2 PERFICERE cf. the goddess Perfica Arn. Iv 7 pr. the 
jest of Heliogabalus (Capitol. Maximini 4, with Saumaise). 

ΠῚ 80 DE NVLLO QVERERIS, NVLLI MALEDICIS, APICI: RVMOR 
AIT LINGVAE TE TAMEN ESSE MALAE interpreted by Minuc. 28 
§ 10 etiam ille qui de adoratis sacerdotis wirilibus aduersum nos 
fabulatur, temptat in nos conferre quae sua sunt. ista enim 
impudicitiae eorum forsitan sacra sunt, apud quos sexus omnis 
omnibus membris prostat, apud quos tota impudicitia uocatur 
urbanitas: qui scortorum licentiae inuident, qui medios uiros 
Jambunt (Mart. 111 81 2) libidinoso ore inguinibus adhaerescunt ; 
homines malae linguae etiam si tacerent. See Davies 
there. In anth. Pal. x1 155 4 (ἀλλ᾽ ἐάλω ποιῶν ἔργα κακο- 
στομάτων) the last word is a conjecture for κακοστομάχων, 
suggested to Boissonade by our verse. He should have shewn 
that the form is possible. Compounds of αἷμα, γράμμα, λῆμα 
cet. do make -atos, but κακοστόματος must leave the honours 
of the lexicon to κακόστομος. 

iI 81 2 Minue. just cited. Ellis on Catull. 80 6. 

ae eee 1. 10. 

» »- ὅ GALBINATVS Suet. Galba 3 Torr. 

» » 9 CVSPIDESQVE LENTISCI VI 74 3. Marquardt Privatl. 
1 320 11. Bottiger, Sabina 157. Becker-Rein, Gallus τῇ 191. 
Petron. 33 pr. (with comm.) ut deinde argentea pinna dentes 
perfodit. 

it 82 11 FLABELLO Plaut. trin. 253 Wagner flabelliferae. 

» » 12 FVGATQVE MVSCAS MYRTEA PVER VIRGA Sen. lud. 
10 ὃ 3 qui wobis non posse uidetur muscam excitare. de ira 
ir 25 § 3 Lipsius quid est enim, cur...musca parum curiose 
fugata in rabiem agat? Arist. wasps 597. knights 59 60 
ἀλλὰ βυρσίνην (parody on μυρσίνην) ἔχων | δειπνοῦντος ἑστὼς 
ἀποσοβεῖ τοὺς ῥήτορας. 

ΠῚ 82 15 DIGITI CREPANTIS SIGNA cf. Beda h. e. Iv 3 (p. 96 
7) sonitum manu faciens, with my n. p. 305. 

ΠῚ 82 23 cocTA FVMIS MVSTA MASSILITANIS Desjardins, 
Gaule 1 444—5. 

ΠΙ| 82 24 OPIMIANVM luv. v 30 n. 31 n. 34 n. pp. 249, 420. 

» » 32 MALCHIONIS οὗ the Trimalchio of Petronius (Rhei. 
Mus. II 69). ᾿ 


Journal of Philology. vow. xvt. L6 


242 THE JOURNAL OF PHILOLOGY. 


ΠῚ 83 1 VT FACIAM BREVIORA MONES EPIGRAMMATA II 77. 
VI 65. . 
I 85 οὗ 1 83. 
» 913 HVIC COMES HAEREBAT Ellis on Catull. 26 5. 
ΠΙ 92 VT PATIAR MOECHVM, ROGAT VXOR, GALLE, SED 
' VNVM. | HVIC EGO NON OCVLOS ERVO, GALLE, Dvos? Fr. 
remarks 
Das dauernde Verhiltniss mit einem Liebhaber, also gleich- | 
sam einem zweiten Manne (Sen. ben. 111 16 § 3 matrimonium 
uocari wnum adulterium) galt fiir schlimmer als Ehebruch 
mit mehreren. 


On the parallel epigram (v1 90 moechum Gellia non habet 
nist unum. | turpe est hoc magis: uxor est duorum) Fr. in like 
manner 

Da nimlich das Verhiltniss einer Frau mit nur einem 
Ehebrecher als eine Art zweiter Ehe galt, so ist diese 
Bigamie noch schimpflicher als Ehebruch mit mehreren. 


Let us read a little more of Seneca’s text: 


numquid iam ullus adulterii pudor est, postquam eo uentum 
est, ut nulla uirum habeat, nisi ut adulterum irritet? argu- 
mentum est deformitatis pudicitia. quam inuenies tam 
miseram, tam sordidam, ut illi satis sit unum adulterorum 
par, nisi singulis diuisit horas? et non sufficit dies omnibus, 
nisi apud alium gestata est, apud alium mansit. infrunita 
et antiqua est, quae nesciat matrimonium wuocart unum 
adulterium. 


In the last words we plainly hear the emancipated women: 
to be content with a single adulterer is as old-fashioned and 
‘slow’ as to be a Lucretia. No hint here that in public esti- 
mation a wife with a single paramour was guiltier than one 
with ten. Mart. vi 90 has no point if it is not a paradox. 
‘Gellia has but one gallant’ is her plea: ‘ Nay, the scandal is 
only the greater; she is wife to two,’ half her time, half her 
care, is given to another than her husband; the personal rivalry 
between Paris and Menelaus comes out more strongly. 

Now come back to 11 92. The gist of it is: ‘only one has 
sinned’ says she; ‘only one shall suffer; say I. ‘My wife, 


NOTES ON MARTIAL, BOOK 17]. 243 


Gallus, begs me to condone her infidelity: to all the world but 
one, she is the faithful wife. When Fr. takes sed as in 1 117 7 
(‘aye, and what is more, Luv. Iv 27, v 147 n. pp. 236, 434, ind. 
vol. 11s. v. sed; add Ov. tr. tr 11 45, v 5 24) he misses the 
force of wnwm ‘only one.” Then hinc ego cet. ‘She confines | 
herself to one; so do I.’ ‘This one, what shall I do to him?’ 
‘Shall I not, Gallus, gouge out his two eyes?’ a single pair of 
eyes for one rival, a score for a score: flagrantior aequo non 
debet dolor esse wrt, nec wulnere maior. I take duos as ‘only 
two,’ ‘a single. pair,’ and eruo like the ind. pres. in quid ago? 
The threat effodiam oculos is common in comedy. 

ΠῚ 93 5 ARANEORVM CASSIBVS Anmibros. hexaiem. v ὃ 24 si 
araneam, guae tam subtiliter ac docte laxos casses suspendit in 
JSoribus, sapientiae non reliquit immunem. 

ΠῚ 93 8 MELIVSQVE RANAE GARRIANT RAVENNATES Sid. ep. 
I 8 (p. 195 Baret) zta tamen quod te Rauennae felicius exsu- 
lantem, auribus Padano culice perfossis, municipalium rana- 
rum loquax turba cercumsilit. 

ΠΙ 93 15 see I 34 8 (not, as Fr., 34 38). 

» » 2D TALASSIONEM XII 42 4 Talasse. -Catull. 61 127 
Ellis. | 

ΠῚ 93 26 VSTORQVE TAEDAS PRAEFERAT NOVAE NVPTAE 
Catull. 59 5 Ellis wstore. [Sen.] Oct. 570 hic mihi iugales prae- 
ferat taedas deus. Petron. 26 pr. tam Psyche puellae caput 
enuoluerat flammeo, 1am embasicoetas praeferebat facem. 

ΠΙ| 95 1 NVMQVAM DICIS HAVE, SED REDDIS, NAEVOLE, SEMPER 
Vv 66 1 saepe salutatus numquam prior ipse salutas. Hor. 
ep. I 7 66 occupat et saluere iubet prior. 

ΠῚ 95 7 8 ORE LEGOR MVLTO NOTVMQVE PER OPPIDA NOMEN 
NON EXPECTATO DAT MIHI FAMA ROGO I 1 4—6, 

mt 95 13 Pers. 1 87. 

» 97 1.RVFE cf. 100. Iv 82. 


JOHN E. B. MAYOR, 


16—2 


THE AGAMEMNON OF AESCHYLUS. 


THANKS to Wecklein it is at length possible to study Aeschy- 
lus in comfort. Next to an accurate collation of the cardinal 
Mss, a complete register of the conjectures of critics is the 
student’s prime requisite. Nothing short of a complete register 
will serve: no man can be trusted to sift good from bad: some 
editors do not know a correction when they see one, others 
through childish jealousy of this scholar or that ignore his 
discoveries, the most candid and the soundest judgment is 
human and errs. The time lost, the tissues wasted, in doing 
anew the brainwork done before by others, and all for lack of a 
book like Wecklein’s Appendix, are in our brief irreparable life 
disheartening to think of. — 

In the ensuing pages I have ποῦ set down all or nearly all 
the corrections which I imagine myself to have made in the 
Agamemnon: I know how easily one is satisfied with one’s own 
conjectures. I have arraigned the Mss only where their delin- 
quencies can be made as clear as daylight, and I have proposed 
only corrections which I think may possibly convince others as 
well as myself. For instance, however confident I may feel 
that in v. 17 Aeschylus wrote not ὕπνου but πόνου, still I have 
to own that the former can by hook or by crook be defended, 
and that the indications which suggest the latter are not de- 
cisive ; so I leave the reader in peace. I need hardly say that 
I have not broached cunjectures on a tithe of the passages I 
think corrupt: diagnosis is one thing and healing another: let 
us keep the precept ἢ λέγε τι σιγῆς κρεῖσσον ἢ σιγὴν ἔχε. 

The numeration is Wecklein’s, which for this play tallies 
with Dindorf’s. 


THE AGAMEMNON OF AESCHYLUS. 245 


ind 
v ‘ ς ’ 
ἄστρων κάτοιδα νυκτέρων ὁμήγυριν 
καὶ τοὺς φέροντας χεῖμα καὶ θέρος βροτοῖς 
λαμπροὺς δυνάστας ἐμπρέποντας αἰθέρι, 
’ 7ὔ ¢/ / ? / nr 
ἀστέρας, ὅταν φθίνωσιν, ἀντολάς τε τῶν. 


I know the stars and the rulers of the seasons, the stars to 
wit. This of course is one of those sentences which a poet does 
not write; so mest editors with Pauw and Valckenaer bracket 
v. 7 for spurious. It is a good riddance, that I see, but I do 
not see on what principles of criticism it can be justified: the 
Aeschylean archaism τῶν never came from the workshop of an 
interpolator. Fault has of course been found, ὥστε σύγγονον 
βροτοῖσι τὸν πεσόντα λακτίσαι πλέον, With the initial dacty]l, 
and when the faultfinders have got rid of Cho. 215 καὶ τίνα 
σύνοισθά μοι καλουμένῃ βροτῶν; 984 of the same play ἥλιος 
ἄναγνα μητρὸς ἔργα τῆς ἐμῆς, Sept. 640 ὦ θεομανές τε καὶ 
θεῶν μέγα στύγος and fr. 290, 4, Dind. ἥλιος ἐν ἣ πυρωπὸς 
ἐκλάμψας χθονί, then they may be heard: not before. But 
Hermann and others, who suppose themselves to have rescued 
the verse by trifling with the punctuation, ascribe to commas a 
cabalistic virtue which did not reside in the seal of Solomon. 
Mr Margoliouth writes 5—7 as follows: καὶ τοὺς φέροντας 
χεῖμα καὶ θέρος βροτοῖς | λαμπροὺς δυνάστας, ἐμπρέποντας 
αἰθέρι | ἀστέρες ὅταν φθίνωσιν, ἀντολαῖς τε τῶν : the λαμπροὶ 
δυνάσται, he says, are the Pleiades. If a year of sleepless 
nights has taught the watchman so little astronomy that he 
singles out this nebulous cluster from the host of heaven to call 
it λαμπρός, he is a signal confutation of his creator’s favourite 
doctrine, παθήματα μαθήματα. Nay Mr Margoliouth’s own 
witnesses turn round and testify against him: ὀλίγαι καὶ 
ἀφεγγέες, ἐπισκέψασθαι apavpai, ‘ignis uix tenui longe face fit 
spectabilis’ say Aratus and Auienus; and to set against this 
damaging evidence Mr Margoliouth can find nothing better 
than the following citation: ‘Cic. Progn. 356 fugiet cum lucida 
visus Plevas.’ Now the employment of lucida, by Cicero or by 


1 Pers, 287 and Soph. Aiax 1331 seem to answer Mr Verrall’s objections to 
this verse. 


246 THE JOURNAL OF PHILOLOGY. 


any one else, as an epitheton ornans for Plecas in a context 
which does not pit these stars against their fellows, concerns 
these verses of Aeschylus not a jot; but what is ‘Cic. Progn. 
356’? If Mr Margoliouth has access to 356 verses of Cicero’s 
Prognéstica, he is more fortunate than the rest of the world 
who know only 27. Truth to tell, however, the words which he © 
ascribes to Cicero were written in the seventeenth century after 
Christ by Hugo Grotius. If Mr Margoliouth cares for Cicero’s” 
account of the Pleiades, here it is: Phaen. 27 ‘omnis parte lo- 
catas | parua Vergilias tenut cum luce uidebis’, 37 ‘hae tenues 
paruo labentes lwmine lucent’. And let the Pleiades be as 
brilliant as you will, masculine they will never be: even Cie. 
Progn. 356 does not present us with lucidus Pleias. 

The passage is I believe to be righted, not by the change of 
a single letter, but by a simpler remedy; the simplest which 
can be applied to the text of any poet Greek or Roman. We 
should have heard no evil of the initial dactyl if the Ms gave 
the verses thus: 


ἄστρων κάτοιδα νυκτέρων ὁμήγυριν, 

\ / b] / > / 
λαμπροὺς δυνάστας ἐμπρέποντας αἰθέρι, 6 
καὶ τοὺς φέροντας χεῖμα καὶ θέρος βροτοῖς 
ΕἸ ’ὔ ¢, / ’ / ΄ 
ἀστέρας, ὅταν φθίνωσιν, ἀντολάς τε TOV: 


οι 


The watcher is grown acquainted with the stars, which he 
likens to a congregation of princes, and chiefly with the down- 
setting and the uprising of those which bring men winter and 
summer, the stars of the zodiac. These, by which he reckons 
the passage of his year’s vigil, are singled out from the other 
stars by καὶ, as in Pers. 751 the god of the sea is singled out 
from the other gods whom Xerxes fought against when he 
bound the Hellespont: θεῶν δὲ πάντων @et’, οὐκ εὐβουλίᾳ," 
καὶ Ἰ]οσειδῶνος κρατήσειν. It should be said that the trans- 
position must have taken place before the time of Achilles 
Tatius, who quotes vv. 4—6 in the traditional order. 


49—59. 


U ᾽ A Ὁ...» ’ 
τρόπον ALYUTTL@V, OLT ἐκπαγλοις 
w , “ f - 
ἄλγεσι παίδων ὕπατοι λεχέων 50 


THE AGAMEMNON OF AESCHYLUS. 247 


στροφοδινοῦνται, 
πτερύγων ἐρετμοῖσιν ἐρεσσόμενοι, 
δεμνιοτήρη 
/ > / > ; 
πόνον ορταλίχων ὀλέσαντες" 


or 
Or 


ὕπατος δ᾽ ἀίων ἢ τις ᾿Απόλλων 
ἢ Πὰν ἢ Ζεὺς οἰωνόθροον 

γόον ὀξυβόαν 

τῶνδε μετοίκων ὑστερόποινον 
πέμπει παραβᾶσιν ᾿Ἐρινύν. 


The learner of Greek, in quest of probable or even plausible 
reasons for believing that ὕπατοι. λεχέων summi cubiliwm means 
ὑπὲρ λεχέων super cubilia, is dismissed by Mr Paley to these 
references: “ἐσχάτη χθονὸς Prom. 865, ὑστάτου νεὼς Suppl. 
697, ὕπατος χώρας Ζεὺς inf. 492’. The first two of these 
passages, πόλις ἐσχάτη χθονός and οἴακος ὑστάτου νεώς, prove 
to him what he could well believe without proof, that such a 
phrase as θρυγκὸς ὕπατος τείχους a coping which rs the highest 
part of a wall is Greek; but since vultures on the wing are not 
the highest part of their eyries the information does not help 
him. Had he been referred, say, to a passage where a fish 
following a ship is called ὕστατος νεώς, then he would have 
been helped; but Greek literature contains no such passage: 
such a fish is ὕστερος νεώς To the third reference he turns 
with keen interest, because it is manifest that Mr Paley’s 
translation of ὕπατος χώρας must differ widely from the usual 
rendering. But no: Mr Paley translates like everyone else 
‘supreme over the country’; and the learner of Greek returns 
with a touch of resentment from his fool’s errand. 

I propose παίδων ἀπάτῃ λεχαίων, because their brood 18 stolen 
away. The phrase παίδων λεχαίων finds an exact parallel in 
Sept. 278 δράκοντας ὥς τις τέκνων ὑπερδέδοικεν λεχαίων 
δυσευνάτορας πάντρομος πελειάς, the phrase παίδων ἀπάτῃ in 
Soph. Ant. 630 ἀπάτας λεχέων ὑπεραλγῶν, wroth that he rs 
cheated of his bride. My reading is rather an interpretation 
than an alteration of the Ms text: confusions of ἀπ- and ὑπ- 
are to be counted not by scores but by hundreds; and for 
century on century οὐ was identical with » in pronuncia- 


248 THE JOURNAL OF PHILOLOGY. 


tion, and e with a. In the passage which I have cited 
from the Septem, λεχαίων had to be restored by Lachmann: 
the Ms there as here gives λεχέων. As for the scansion 
of λεχαίων as an anapaest, the penultimate az of πετραῖος is 
shortened by Sophocles in lyrics, Ant. 827, of παλαιός and 
δίκαιος by Euripides in senarii, Hl. 497 and Cycl. 274, of 
yepavos by Sophocles in lyrics, Ὁ. C. 200, and by Euripides in 
anapaests more than once, of de/Aavos by Sophocles in lyrics, 
El. 849, and by Aristophanes in senarii over and over again. 
In Soph. Ant. 1240 the MS gives τὰ νυμφικὰ | τέλη λαχὼν 
δείλαιος ἐν Αἰδου δόμοις : the conjecture εἰν is not to be dreamt 
of: Sophocles seems to have written λαχὼν ἐν Ἅτδου δείλαιος 
δόμοις τέλη Or δόμοις ἐν “Ardov δείλαιος τέλη λαχών. Finally 
in the Agamemnon itself, v. 723, is found εὐφιλόπαιδα καὶ 
yepapois ἐπίχαρτον: which now is the more prudent, to confer 
on γεραροῖς an alien and unexampled’ meaning, unexampled, for 
Supp. 675 proves nothing at all, or to suppose that here, as in 
the same word in Eur, Supp. 43 (yeparév Markland, yepapov 
MSS), a scribe confused two. letters which in old uncials can 
hardly be distinguished, I and P? I take the second alterna- 
tive: uiris doctis aliter uisum. 

But another check awaits us in v. 58. The dissension 
about the meaning of τῶνδε μετοίκων is of long standing. The 
scholiast refers the words to the nestlings, and renders ὑπὲρ 
τῶν μετοικισθέντων νεοσσῶν, a Version which of course is 
peremptorily forbidden by τῶνδε. Another explanation is given 
in the scholion on Soph. 0. C. 934 and reappears, somewhat, 
curtailed, in Suidas: Αἰσχύλος.. ἐν ᾿Αγαμέμνονι... μετοίκους... 
εἶπε τῶν ὑψηλῶν τόπων τοὺς οἰωνοὺς...ἀντὶ τῶν ἐνοίκων. But 
obviously a poet who writes thus, a poet ὅς χ᾽ ἕτερον μὲν κεύθῃ 
ἐνὶ φρεσὶν ἄλλο δὲ εἴπῃ, cannot hope that his audience will 
understand him: he might as well call the birds πελειάδας 
ἀντὶ αἰγυπιῶν. Mr Paley says that the parent vultures are 
called μέτοικοι to contrast them with the μέτοικοι of Athens 
who could obtain redress at law only through προστάται; an 
allusion frigid in itself, and so carefully obscured that even after 
Mr Paley has-told us it is there one scans the Greek for it in 
vain. : 


THE AGAMEMNON OF AESCHYLUS. 249 


It will be conceded that a copyist who found in his ex- 
emplar the letters τωνδειμετοκων would be likely to make 
Greek of them by transposing the single letter ἡ to the place 
it holds in the text to-day. Such transpositions, intentional or 
unintentional, are common enough: in one play I notice these 
three: Supp. 22 ἱεροστέπτοισι for ἐριοστέπτοισι, 278 λέγοι 
πρόσως for λέγοις πρόσω, 961 ἔσθιε μὲν through cOupev for 
ἔοιγμεν. But if my hypothetical copyist had been so faithful 
or so dull as to write what he read, criticism would before now 
have restored, letter for letter, a phrase which seems to me the 
most appropriate in the world, τῶν αἰνοτόκων. The substitu- 
tions 6 for a, ev for ἐ, w for v and ε for o are so common, not in 
Aeschylus merely, -but all of them in most Greek Mss and some 
of them in all, that I will not fill with illustrations the pages 
which might be filled; but take two instances where the av of 
aivos by changing to de: has wrought further mischief: in Soph. 
O. C. 212 τόδ᾽ ; αἰνὰ is restored by Wunder for τόδε : δεινὰ, and 
in Eur. Med. 640 προσβάλοι μ᾽ αἰνὰ by Verrall for προσβάλοιμι 
δεινὰ. The terms αἰνοτόκος, αἰνοτόκεια, αἰνὰ τεκοῦσα are 
especially applied to parents rendered wretched by the calami- 
ties of their children: J]. A 414 ὦμοι τέκνον ἐμὸν, Ti νύ σ᾽ ἔτρε- 
gov αἰνὰ τεκοῦσα; says Thetis to Achilles; Oppian. Hal. v 
526 μητρὶ παρ᾽ αἰνοτόκῳ, the mother dolphin whose young one 
the fishermen harpoon ; Nonn. Dion. 11 160 αἰνοτόκοιο θεημάχον 
οὔνομα νύμφης, and XLVIIL 428 Τανταλὶς aivotoxeva, Niobe in 
both places. But let me ask especial attention to the employ- 
ment of αἰνοτόκεια in Mosch. Iv 27 where Megara relates the 
death of her children at the hands of Heracles: ὡς δ᾽ ὄρνις 
δύρηται ἐπὶ σφετέροισι νεοσσοῖς | ὀλλυμένοις, ovat’ αἰνὸς ὄφις 
ἔτι νηπιάχοντας | θάμνοις ἐν πυκινοῖσι κατεσθίῃ" ἡ δὲ κατ᾽ 
αὐτοὺς | πωτᾶται κλάζουσα μάλα yd πότνια μήτηρ | ...ὡς 
ἐγὼ αἰνοτόκεια φίλον γόνον αἰάξουσα | μαινομένοισι πόδεσσι 
δόμον κάτα πολλὸν ἐφοίτων. The poet who wrote this was 
imitating first and foremost Jliad B 308 sqq., but he would 
naturally remember also Aesch. Sept. 278 and this passage of 
the Agamemnon: that he did remember this last, I find another 
indication besides aivotoxea. In v. 2 of the poem you have 


1 See too Cho. 841 δειματοσταγές for αἱματοσταγές. 


250 THE JOURNAL OF PHILOLOGY. 


ἐκπάγλως ayéovoa, in v. 72 ἐκπάγλως ὀλοφύρομαι, in Vv. 98, 
δειμαίνω...ἐκπάγλως : in ν. 2 is a variant ἀχέεσσε: if ἐκπάγλοις 
ἀχέεσσι is to be read, it is neither more nor less than a repro- 
duction of the phrase which Blomfield has in Ag. 49 restored 
to Aeschylus, ἐκπάγλοις ἄλγεσι. The MS reading ἐκπατίοις 
is undoubtedly a word which might have existed, though in 
face of the resemblance between yA and τὸ there is nothing 
approaching proof that exist it did; but in this passage it gives 
a totally inadequate sense, and ἐκπάγλοις as well as αἰνοτόκων 
seems to gain support from Moschus. 


97—-103. 


τούτων λήξασ᾽ 6 τι καὶ δυνατὸν 

καὶ θέμις αἴνει 

παιών τε γενοῦ τῆσδε μερίμνης, 

ἣ νῦν τοτὲ μὲν κακόφρων τελέθει, 100 
\ te a > Ν a2 

τοτὲ δ᾽ ἐκ θυσιῶν ayavn φανθεῖσ 

ἐλπὶς ἀμύνει φροντίδ᾽ ἄπληστον 

τὴν θυμοφθέρον λύπης φρένα. 


Fortunately I need not demonstrate that ν. 103 is corrupt. 
For my own part, if I could believe ὕπατοι λεχέων to be Greek 
or τῶνδε μετοίκων to be sense, I could believe τὴν θυμοφθόρον 
λύπης φρένα to be a paroemiac; but some scholars seem to 
find it the harder feat. Wecklein records over a score of con- 
jectures, not one of which affords a plausible explanation of the 
phenomena presented by the Ms. Those phenomena, as well as 
the scholion, are explained if Aeschylus wrote this: 


ἐλπὶς ἀμύνει φροντίδ᾽ ἄπληστον 
θυμοῦ, λυπησίφρον᾽ ἄτην. 


θυμοῦ is of course to be construed with ἀμύνε. But the 
scholiast construed ἄπληστον θυμοῦ, and therefore paraphrased 
the words ἄπληστον θυμοῦ λυπησίφρονα by ἥτις ἐστὶ θυμοβόρος 
λύπη τῆς φρενός, rendering the adjective λυπησίφρονα by the 
phrase ἥτις ἐστὶ λύπη τῆς φρενός, the phrase ἄπληστον θυμοῦ 
by the adjective θυμοβόρος. Some reader of Aeschylus, under’ 
the same misapprehension, wrote θυμοφθόρον in the margin of 
his copy: that θυμοφθόρος and θυμοβόρος were to Byzantine 


THE AGAMEMNON OF AESCHYLUS. 251 


ears identical in meaning is shown by Hesychius θυμοβόροιο" 
ψυχοφθόρου, and Photius θυμοβόρος" ἡ τὴν ψυχὴν διαφθείρουσα. 
A subsequent copyist took this marginal θυμοφθόρον to be a 
correction of θυμοῦ, and corrected accordingly. Hardly more 
than a wrong division of the letters was needed to convert the 
unfamiliar λυπησιφρονατην into λύπης φρένα THY; and since 
τήν could not stand at the end of the line it was transplanted 
. to the beginning, where it flourishes to-day. Hesiod Op. 795 
πεφύλαξο δὲ θυμῷ | τετράδ᾽ ἀλεύασθαι φθίνοντός θ᾽ ἱσταμένου 
τε | ἄλγεα θυμοβορεῖν, and Theognis 1323 σκέδασον δὲ μερίμνας | 
θυμοβόρους, were perhaps the passages which suggested to Aes- 
chylus his ἀμύνειν θυμοῦ and his φροντίδ᾽ ἄπληστον, and to the 
scholiast his θυμοβόρος. 

I learn from Wecklein that Avrncidpova was detected three 
centuries ago by Scaliger: how he completed the verse I do not 
know. Similar compounds with similar force are θελξίφρων 
Eur. Bacch. 404 and ῥηξίφρων Hesych. 


131—135. 
χρόνῳ μὲν ἀγρεῖ 
Πριάμου πόλιν ἅδε κέλευθος 
πάντα δὲ πύργων 
κτήνη προσθετὰ δημιοπληθῆ 
μοῖρ᾽ ἀλαπάξει πρὸς τὸ βίαιον. 135 


Most editors adopt in v. 134 the conjecture πρόσθε ta which 
appears in the Florentine apograph; but how to translate it 
they cannot agree. Half take πρόσθε in a temporal sense, 
which makes Calchas a lying prophet: if the wealth of Troy 
was exhausted before its fall, how comes it that Cassandra was 
πολλῶν χρημάτων ἐξαίρετον avOos? Half, giving κτήνη its 
usual meaning of cattle, construe πρόσθε πύργων in front of the 
walls, which leaves the lines no point whatever: the Iliad 
shows that after nine years’ siege neither Trojans nor Greeks 
were lacking in flocks and herds, whether those flocks and 
herds were or were not πρόσθε πύργων. So if πρόσθε is local 
the lines are trivial, if it is temporal they are not true. But 
against both renderings lies the further and fatal objection 
that they refer the sentence to circumstances of the leaguer 


252 THK JOURNAL OF PHILOLOGY. 


when it must of necessity be referred to the taking of the 
town. It was the fall, not the siege, of Troy which depended 
on averting an aya θεόθεν. 

For these or for other reasons Pauw and Hermann prefer 
to write πρόσθετα, which indeed is virtually the reading of the 
MS, aS many scribes preserve in compounds of @erds the ac- 
centuation of the simple adjective. But Weil justly observes 
that πρόσθετα cannot, as Hermann would have it, mean con- 
gesta, which is by no means the same thing as additicia. If 
then there were no other meaning of πρόσθετα the MS must be 
abandoned. But there is another and a most appropriate 
meaning. Over and over again in this play the Trojan war 
is likened to a lawsuit in which the Greeks prosecute the 
Trojans: 41 Πριάμου μέγας ἀντίδικος Μενέλαος, 458 προδίκοις 
᾿Ατρείδαις, 539 ὀφλὼν γὰρ ἁρπαγῆς τε καὶ κλοπῆς δίκην | τοῦ 
ῥυσίου θ᾽ ἥμαρτε κτλ., 804 δίκας γὰρ οὐκ ἀπὸ γλώσσης θεοὶ 
κλύοντες ἀνδροκνῆτας ᾿Ιλίου φθορᾶς | ἐς αἱματηρὸν τεῦχος οὐ 
διχορρόπως ψήφους ἔθεντο" τῷ δ᾽ ἐναντίῳ κύτει | ἐλπὶς προσ- 
eles χεῖρας οὐ πληρουμένῳ: in 537 too συντελής seems to 
mean paying joint penalty. Now πρόσθετος, as the lexicons 
will show, has the technical meaning addictus, surrendered to 
a creditor. It is noticeable that the scholiast explains κτήνη 
by κτήματα, and that this very phrase κτήματα πρόσθετά τινι 
ποιεῖν occurs in Boeckh’s Corp. Inscr. 2691. I think then that 
πρόσθετα is not only sound but exceedingly apt. 

But πύργων κτήνη is a strange phrase. The wealth of the 
Trojans, the wealth of Troy, I could understand: the wealth of 
the fortifications, no. And this is not the only difficulty. The 
strophic verses answering 131—135 are 110—114: 

ὅπως ᾿Αχαιῶν 
δίθρονον κράτος, Ελλάδος ἔβας 
ξύμφρονα τὰν γᾶν, 
πέμπει ξὺν δορὶ καὶ χερὶ πράκτορι 
θούριος ὄρνις Τευκρίδ᾽ ἐπ᾽ αἷαν. 
How are we to amend τὰν γᾶν! Blomfield writes ταγόν, 


Hermann tayav: I am bound to suppose that these scholars 
attached some meaning to the phrase a unanimous captain, but 


THE AGAMEMNON OF AESCHYLUS. 253 


what that meaning may have been I cannot divine. Neither is 
it possible, in Greece or anywhere else, for two persons to com- 
pose one Tayds or one tayns: in v. 41 ἀντίδικος is of course 
Μενέλαος, not Μενέλαος ἠδ᾽ ᾿Αγαμέμνων. If with Dindorf we 
write ξύμφρονε Tayo we get a meaning, but we stray some 
distance to get it. The correction which is instantly suggested 
by the requirements of the sense is as old as the earliest 
apographs of the Medicean: tayav. Hesychius has tayais- 
ἀρχαῖς, ἡγεμονίαις, which is precisely the meaning wanted: the 
two Atridae compose one tay as they compose one κράτος. 
Dactyls are often varied with tribrachs by Pindar in Doric 
melodies, just as lyric anapaests are so varied in passages like 
Eur. J. JT. 130, where see Monk and Dindorf. In the notation 
of J. H. H. Schmidt, which is familiar to Englishmen, the verse 
ξύμφρονα tayav will be -:vv<|-, see his Leitfaden ὃ 12. 
But of course ταγάν in the strophe and πύργων in the anti- 
strophe cannot live together: which shall be the victim? The 
inappropriate, not the appropriate word. 


πάντα δὲ Φρυγῶν 
κτήνη πρόσθετα δημιοπληθῆ 
μοῖρ᾽ ἀλαπάξει πρὸς τὸ βίαιον. 


The confusion of an aspirate with its tenuis is among the com- 
monest of those errors of the ear to which copyists are subject: 
an apposite and undisputed instance of a for ¢ is Cho. 417 
πάντες for φάντες. Another of their favourite tricks is to 
reverse the order of two consecutive letters: disregarding such 
perpetual confusions as θράσος and θάρσος, κραδία and καρδία, 
I take the following examples from Aeschylus alone: P. V. 934 
προσδάρκοι for προσδράκοι, Pers. 689 ῥοθιάζοντες for ὀρθιά- 
Covtes, Supp. 372 ἐκπνοεῖν for ἐκπονεῖν, 703 θεαί τ᾽ for θείατ᾽, 
Ag. 117 ἀργίας for dpyds, 797 πόνος for πνόος, 1204 βαρύνεται 
for ἁβρύνεται, Cho. 270 κἀξοθριάζων for κἀξορθιάξων, Hum. 
260 χερῶν for χρεῶν. It is interesting to note that Mr 
Margoliouth has conjectured Τευκρῶν, as it counts for something 
that two minds should independently require the same meaning. 

The adjective δημιοπληθῆ is one of those many poetical 
compounds in which the second element is purely ornamental : 


254 THE JOURNAL OF PHILOLOGY. 


just as ἀρσενοπληθῆ in Supp. 29 means simply ἄρσενα, so 
δημιοπληθῆ here means simply δήμια; it would be wrong I 
think to say that it means even δήμια πολλά. But be that as 
it may, Mr Margoliouth by defending ἁβροτίμων προκαλυμ- 
μάτων is estopped from impugning κτήνη δημιοπληθῆ. 


415. 444. 


λιποῦσα δ᾽ ἀστοῖσιν ἀσπίστορας _ ptr, 
κλόνους TE καὶ λογχίμους ναυβάτας θ᾽ ὁπλισμοὺς 
ἄγουσά T ἀντίφερνον “Xia φθορὰν 41 
βέβακεν ῥίμφα διὰ πυλᾶν 
ἄτλητα τλᾶσα: πουλὺ δ᾽ ἄνστενον 
LANG > / , a 
τάδ᾽ ἐννέποντες δόμων προφῆται" 
‘id ἰὼ δῶμα δῶμα καὶ πρόμοι, 
ἰὼ λέχος καὶ στίβοι φιλάνορες. 420 
πάρεστι κοίτας ἀτίμας ἀλοίδορος, 
ἄπιστος ἐμφανῶν ἰδεῖν. 
/ μιν / 
πόθῳ δ᾽ ὑπερποντίας 
Ul / , > U 
φάσμα δόξει δόμων ἀνάσσειν. 
εὐμόρφων δὲ κολοσσῶν 425 
ἔχθεται χάρις ἀνδρί, 
ὀμμάτων δ᾽ ἐν ἀχηνίαις 
ἔρρει πᾶσ᾽ ᾿Αφροδίτα. 
.) U \ / 
ὀνειρόφαντοι δὲ πειθήμονες ant. 
πάρεισι δόξαι φέρουσαι χάριν ματαίαν. 480 
μάταν γὰρ, εὖτ᾽ ἂν ἐς θυγὰς δοκᾶν ὁρᾷ, 
παραλλάξασα διὰ χερῶν 
/ bd > Ud 
BéBaxev ὄψις ov μεθύστερον 
a > an [4 / ? 
πτεροῖς ὀπαδοῖς ὕπνου κελεύθοις. 
‘ \ > »Μ b] / Μ 
τὰ μὲν KAT οἴκους ἐφεστίους ἄχη 435 
ANG > \ \ lal > ς / 
τάδ᾽ ἐστὶ καὶ τῶνδ᾽ ὑπερβατώτερα" 
τοπᾶν δ᾽, ap “ἕλλανος αἴας ξυνορμένοις, 
ποθεινὰ τλησικαρδίοις 
δόμων ἑκάς που πρέπει. 
πολλὰ γοῦν Ouyyaver πρὸς ἧπαρ' 440 
οἵους μὲν γὰρ ἔπεμψαν 
*O\ > \ \ A 
οὐδὲν, ἀντὶ δὲ φωτῶν 


THE AGAMEMNON OF AESCHYLUS. 255 


, \ Ν > € / 
τεύχη Kal σποδὸς εἰς ἑκά- 
στου δόμους ἀφικνεῖται. 


414. κλ. τε καὶ λ. ναυβάτας θ᾽ H. L. Ahrens, κλ. Xr. τε καὶ 
ναυβάτας Flor. 417 πουλὺ Arnaldus, πολὺ Flor. ἄνστενον 
scripsi, ἀνέστενον Flor. 418 τάδ᾽ Auratus, τόδ᾽ Flor. 421 
κοίτας ἀτίμας scripsi, συγᾶς ἄτιμος Flor. 422 ἄπιστος ἐμφανῶν 
Margoliouth, ἅδιστος ἀφεμένων Flor. 429 πειθήμονες scripsi, 
mevOnuoves Flor. 431 és θιγὰς δοκῶν scripsi, ἐσθλά τις δοκῶν 
Flor. ὁρᾷ Scholefield, ὁρᾶν Flor. 435 ἐφεστίους Vossius, ἐφ᾽ 
ἑστίας Flor. 437 τοπᾶν scripsi, τὸ πᾶν Flor. “Ed Xavos 
Bamberger, “Ελλάδος Flor. 438 ποθεινὰ τλησικαρδίοις scripsi, 
πένθεια τλησικάρδιος Flor. 439 éxas mov H. L. Ahrens, 
ἑκάστου Flor. 441 οἵους G. C. W. Schneider, ods Flor. 
ἔπεμψαν scripsi, ἔπεμψεν Flor. 442 οὐδὲν scripsi, oidev Flor. 
“, To save space I have written down this passage at once in 
the form to which I propose to bring it: I will now render an 
account of the changes made. 

417. I think every edition reads here πολλὰ δ᾽ ἔστενον 
from Triclinius’ conjecture ; but how then arose the reading of 
the uninterpolated ms? I restore the metre simply by restoring 
the epic forms which the copyist translated into the common 
dialect. The form πουλύ should not be denied to Aeschylus: 
he has πολεῖ and πολέα, both Sophocles and Euripides have 
πολλός, and πουλύπους is the regular Attic form: for ἄνστενον 
see v. 1552 κάππεσε, κάτθανε. I will not quarrel with any- 
one who prefers πολλὰ, but ἄνστενον must I think be read. 

418. It seems that δόμων προφῆται is taken to mean the 
seers belonging to the household; but the verses 419—434 
needed no seer to utter them, for not one word of prophecy do 
they contain: they contain merely, as Mr Paley says, specula- 
tions on Menelaus’ state of mind. And what is more, I altogether 
deny that προφήτης in tragedy can mean μάντις. If I may 
trust the lexicons, προφήτης and προφῆτις occur in tragedy 
eleven times. In eight of these instances the meaning interpres 
is beyond all doubt, the gen. of the person or thing interpreted 
being expressed or implied: Aesch. Hum. 19 Διὸς προφήτης, 
Eur. Or. 364 Νηρέως προφήτης, Bacch. 211 προφήτης λόγων, 


256 THE JOURNAL OF PHILOLOGY. 


551 Διόνυσε, σοὺς προφήτας, Rhes. 972 Βάκχου προφήτης, Lon, 
321 and 1322 Φοίβου προφῆτις, 42 κυρεῖ.. .προφῆτις ἐσβαίνουσα ᾿ 
μαντεῖον θεοῦ, where, even if θεοῦ is not to be taken ἀπὸ κοινοῦ 
with προφῆτις and μαντεῖον, the word is correctly employed as 
the title of Apollo’s interpreter, the Delphic priestess. There 
remain, besides the present passage, two others where προφήτης 
is construed seer: these I will now examine. In Aesch. Ag. 
1083 sq. we read: ἦμεν κλέος σου μαντικὸν πεπυσμένοι, | NMED 
προφήτας δ᾽ οὔτινας ματεύομεν : of course ἠμὲν has been foisted 
in from the line above. The second verse is rendered by Mr 
Paley and almost all other commentators we are not on the look- 
out for prophets: a version which not only imputes to the 
coryphaeus highly uncivil and rather impious language, but is 
entirely uncalled-for. The words are explained with perfect 
correctness by the scholiast, though no one but Blomfield has 
listened to him: τοὺς λέξοντας ἡμῖν περὶ σοῦ" αὐτοὶ yap αὐτό- 
mTat γινόμεθα. Cassandra has just scented in the palace the 
banquet of Thyestes; and the coryphaeus exclaims: We had 
been told of your divining power before, but now we seek for 
none to fell us of it: we witness it at first hand. Aeschylus 
probably wrote ἤδη προφήτας δ᾽ οὔτινας ματεύομεν : but that is 
by the way. Finally I come to Sept. 596 sqq.: οὕτως δ᾽ ὁ 
μάντις, υἱὸν Οἰκλέους λέγω, | σώφρων δίκαιος ἀγαθὸς εὐσεβὴς 
ἀνὴρ, | μέγας προφήτης, ἀνοσίοισι συμμιγεὶς | θρασυστόμοισιν 
ἀνδράσιν... Διὸς θέλοντος ξυγκαθελκυσθήσεται. Here the ren- 
dering seer is necessary if the text is sound. But the fact that 
this is a solitary exception, for in Ag. 418 as I said at the outset 
this meaning even if permissible would be inappropriate, is of 
itself some presumption against the soundness of the text; and 
it is not the only presumption. The words εὐσεβὴς ἀνὴρ occur 
at the end of v. 589 only eight lines above; and the nearness 
of that verse not only makes the repetition in v. 597 unpleasant, 
but also, if Aeschylus wrote σώφρων, δίκαιος, ἀγαθὸς, εὐσεβὴς, 
θεοῦ (or θεῶν or Atos) μέγας προφήτης, explains the corruption 
by the wandering of the scribe’s eye from the latter εὐσεβὴς to 
the former. I therefore, to return to my starting point, give to 
προφῆται in Ag. 418 precisely the sense it has in ν, 1084: 
δόμων προφῆται are of λέγοντες ἡμῖν περὶ δόμων, purveyors of 


ee ee eee ΝΣ να: 


THE AGAMEMNON OF AESCHYLUS. 257 


gossip about the royal family: of course οἶκος αὐτὸς, εἰ φθογγὴν 
λάβοι, σαφέστατ᾽ ἂν λέξειεν, but in default of that the Argive 
people had to get their news through προφῆται, retailers at 
second hand. 

421. Here, first of all, the metrical dissension between 
strophe and antistrophe has to be removed. If the scansion of 
the two lines as handed down be compared 


str. ὌΠ Ψ af ΘΟ ΩΝ ΟΡ 421 


ant. υπυπυυππυπυ - 437 


it will be seen that the simplest and most rhythmical cure will 
be to alter the quantity of the 8th syllable in the strophic and 
of the 5th in the antistrophic verse. The latter change is 
merely the change of a letter, “EXXavos for “Ελλάδος, and this 
easy and graceful emendation of Bamberger’s has naturally 
found much favour. Lighted by this and by Mr Margoliouth’s 
beautiful restoration of v. 422 I have corrected the strophic 
passage thus: there he stands, reviling not his dishonoured bed, 
believing not what is plain to see. Menelaus does not upbraid 
his wife’s unfaithfulness, nay he refuses to believe her unfaith- 
ful. πάρεστι sc. ὁ ἀνὴρ τῷ λέχει, as is readily understood from 
λέχος and φιλάνορες in the preceding verse. For the form 
atiwas compare in the first place ὑπερποντίας only two lines 
below, and also Sept. 105 εὐφιλήταν, 761 ἁρπαξάνδραν, Pers. 
600 περικλύστα, Ag. 1104 Evvastia, Cho. 68 παναρκέτας, 99 
petaitiat, 617 ἀθανότας, Hum. 268 μητροφόνας, 792 δύυσοιστα; 
perhaps the ἀτίτᾳ of the MS means ἀτίτῃ not ἀτίται in Ag. 72, 
and θελκτηρία is to be read in Cho. 666 ; in Supp. 63 Hermann 
with high probability writes κιρκηλάτας ; in Ag. 796 I should 
be disposed to read viv δ᾽ οὐκ am’ ἄκρας φρενὸς οὐδ᾽ adirys | 
εὔφρων πνόος εὖ τελέσασιν, as ἀφίλως εὔφρων is a phrase which 
conveys no meaning to me. The transmutation of κοιτὰς into 
iccirac and the consequent loss of the letters ic in the sequence 
Tlapectiiccirac are errors of the easiest sort: the confusion 
of o with c and of τ with r I need not illustrate; but a word 
on the confusion in Aeschylus of x with ic. In Cho. 896 is 
found ὠκὺ for ὧι σὺ (Robortellus), in Sept. 927 κακὸς for ἴσος 
(Weil) through xos,in Hwm.178 ἐκείνου for εἶσιν οὗ (Kirchhoff), 


Journal of Philology. vow. xvt. 17 


258 THE JOURNAL OF PHILOLOGY. 


in 864 ἱδρύσηι κάρη for ἱδρύσηις ΓΆρη (Stephanus) ; an especially 
noticeable instance is Cho. 160 where, for =avOsxa, the MS has 


pee ct : that is, the scribe wrote τ in error and added καὶ above 
as a correction, and this « was corrupted by successive eopyists 
first to ἐσ, then to yo. In Ag. 106 I think Mr Margoliouth 
tries to extract the right sense from the corrupt ἐκτελέων, 
but his conjecture is unsatisfactory: Aeschylus seems to have 
written ἀνδρῶν eis τελέων, though the context is so obscure 
that we cannot speak with certainty’. There is less doubt 
however about a scholion on the same passage explaining 
θεόθεν πειθώ by τὴν εἰς θεοὺς πειθώ: not even a scholiast 
could suppose that θεόθεν meant εἰς θεούς : he wrote ἐκ θεοῦ. 
As to v. 422, Mr Margoliouth’s alterations are very slight: 
for the confusion of II] and A see Sept. 654 προσεῖπε for 
mpoceioe: such transpositions of letters as adeu for εμῴφα 
are common enough, whether accompanied, as here, by the 
addition of a letter, or by the subtraction of a letter as in 
μακιστῆρα for μαστικτῆρα Supp. 475, or by the change 
of a letter as in éxpetpovpevos for τεκμαρούμενος Soph, O. 7. 
795. 

424, Those commentators whose opinion I can ascertain 
take φάσμα to be nom.: I think it is acc., the subject of 
δόξει being the same as the subject of πάρεστι, Menelaus, The 
tense is what may be called the conjectural future, = οἶμαι 
δοκεῖ, methinks he sees in fancy a wraith queen of the palace, 
just as in v. 349 ἀφύλακτον εὑδήσουσι πᾶσαν εὐφρόνην = οἶμαι 
εὕδουσι, for the night is passing away as Clytaemestra 
speaks. We have the same idiom in English: he will be 
crossing the Channel by now ; and it is Latin too: luu. 1126 
quiescet. 

429. I suppose we are all in the habit of thinking πενθή- 
μονες a very poetical epithet and are ready to resent its ex- 


1 T would now read κύριός εἰμι Opoetv = goliouth) and Soph. O. T. 1196 (Heim- 
ὅδιον κύρος αἴσιον, ἀνδρῶν εἷς τελέων: soeth): that the mss of Aeschylus and 
κύρος hap is to κύρμα as mpayos to Aristophanes should agree in error is 
πρᾶγμα : the scholiast explains τὸ cuu- a strange accident due to the extreme 
βὰν αὐτοῖς σημεῖον ἐξιοῦσιν. For the rarity of the word. 
confusion with κράτος see Ag. 10 (Mar- 


ET ὰ  ππαΑΙεασσ τ ΊΎ ΣΕ ΎνΎν ΆῊὰ 


THE AGAMEMNON OF AESCHYLUS. 259 


pulsion as the act of a Vandal. Let us see: the word must 
mean one of two things. It may mean of sorrowful aspect : 
this is so thoroughly aimless that I suppose I may dismiss it 
at once: there is no assignable reason why the visions of 
Helen should always or ever wear a sorrowful look. It may 
mean, and it is commonly taken to mean, causing sorrow. This 
is. not pointless like the other rendering; but it is something 
worse. That the visions cause sorrow is true; but how do 
they cause it? Not by their arrival—that causes joy—but by 
their departure. To call them πενθήμονες at the outset is 
to anticipate and utterly to ruin the exquisite turn of ματαίαν 
following hard on χάριν and echoed by μάταν, the key-note 
of the mournful cadence ensuing. If I had my own taste only 
to trust to, I would hold my peace; but hear Euripides. He 
in Alc. 348—356 is imitating this passage: the κολοσσῶν of 
Aeschylus suggests to him σοφῇ δὲ χειρὶ τεκτόνων δέμας τὸ 
σὸν | εἰκασθὲν ἐν λέκτροισιν ἐκταθήσεται | ᾧ προσπεσοῦμαι 
«.7.r.; and then he goes on ἐν δ᾽ ὀνείρασιν  φοιτῶσά μ᾽ 
εὐφραίνοις av’ ἡδὺ γὰρ φίλους | Kav νυκτὶ λεύσσειν, ὅντιν᾽ 
ἂν παρῇ χρόνον. That is just what I said above: the visions 
themselves give joy, while they stay: what gives sorrow is their 
evanescence. I think then that whatever Euripides found in 
the text of Aeschylus he did not find πενθήμονες : the word 
that was in the text I will try to recover from another imitator, 
Propertius was familiar with this stasimon of the Agamemnon: 
one famous passage suggested to him ur 12 13 ‘ neue aliquid 
de te flendum referatur in urna: | sic redeunt, illis qui cecidere 
locis’; and in writing the poem whose mangled remains they 
call the Queen of Elegies he naturally resorted to this locus 
classicus for the griefs of the widower. There the dead Cornelia 
speaks thus, 81 sqq. ‘sat tibi sint noctes, quas de me, Paule, 
fatiges, | somniaque in faciem credita saepe meam; | atque, 
ubi secreto nostra ad simulacra loqueris, | ut responsurae singula 
uerba iace. In these lines simulacra comes from Aeschylus’ 
κολοσσῶν ; and somnia in faciem credita meam, Latin of an 
audacious sort which no Roman durst permit himself except 
he were translating or mistranslating Greek, reads to me like 
a rendering of ὀνειρόφαντοι πειθήμονες δόξαι, visions which 


17—2 


260 THE JOURNAL OF PHILOLOGY. 


persuade him that they are Helen’. Let me add that Euripides’ 
ἐν ὀνείρασι φοιτῶσα awakes in me more than a suspicion that 
he read in his Aeschylus not ὀνειρόφαντοι but ὀνειρόφοιτοι. For 
this confusion see P. V. 684 where the Medicean has νυκτίφαντα 
and other Mss νυκτίφοιτα, and Ag. 82 where it is impossible 
to say whether we ought to read ἡμερόφαντον with Triclinius or 
ἡμερόφοιτον with H. L. Ahrens for the corrupt ἡμερόφατον. 
But ὀνειρόφαντοι is faultless in itself, so I keep it: πενθήμονες 
is not faultless. 

431. As this line runs in the Ms it has no construction, 
aud the usual remedy is Scholefield’s ὁρᾷ. But if this mends 
the grammar it does little mdeed to make sense. To begin 
with: I thought we were talking about Menelaus ; but who is 
this τὸς to whom we are now introduced? To drift off into 
a generalisation is murder to the noble verses;.and I trust 
there is no man so void of discrimination as to adduce the 
usage in Soph. H/. 1406 βοᾷ τις ἔνδον, or Ar. Ran. 664 ἤλγησέν 
τίς. Secondly, ἐσθλά is good things, a term which may indeed 
include, wretchedly inadequate though it be, the apparition of 
an absent wife, but which includes a thousand things besides, to 
all of which the sequel must perforce apply and cannot apply 
without extreme absurdity. Extremely absurd I call it to say 
that whenever a man sees good things in sleep they slip through 
his hands: suppose he sees the council of the elders, or the 
temples of the gods, or the gods themselves, is it through 
his hands that they slip when they vanish away? Thirdly, the 
words διὰ χερῶν demand that some mention of an attempt 
to grasp the apparitions shall have preceded. Keck’s conjecture 
εἶτ᾽ ἂν.. «ὁρᾶν is designed to meet this last difficulty, but leaves 
the others untouched. The reading given above, I hope, re- 
moves them all: when he looks to touch the phantoms. For 
ὁρῶ és τι see Eur. frag. 161 Dind. ἀνδρὸς δ᾽ ὁρῶντος εἰς Κύπριν 
νεανίου | ἀφύλακτος ἡ τήρησις, I. A. 1624 στρατὸς πρὸς πλοῦν 
ὁρᾷ. For δοκή = δόκησις see Hermann’s note on the preceding 


1 πειθήμονες is supported also by dpa γ᾽ ἔχει σύγκοιτα τὰ δάκρυα κἀμὸν 
Meleager Anth. Gr. v 166 dpa μένει ὄνειρον  ψυχαπάτην στέρνοις ἀμφι- 
στοργῆς ἐμὰ .λείψανα καὶ τὸ φίλημα | βαλοῦσα φιλεῖ; 


μνημόσυνον ψυχρᾷ θάλπετ᾽ ἐν eixacia; | 


ee a ὐὰὸ ἐν 


THE AGAMEMNON OF AESCHYLUS. 261 


verse: he there proposes to substitute δοκαί for δόξαι, but 
Ahrens’ transposition in the strophe seems the more rhythmical 
amendment. I cannot doubt that Karsten rightly introduces 
this word in v. 970 δοκὰν δυσκρίτων ὀνειράτων for the unin- 
telligible δίκαν of the Ms: Euripides seems to imitate the phrase 
in the δόκημα ὀνείρων of H. F.111. For @yn do not see the 
lexicons: it is not there. Aeschylus however had no foreknow- 
ledge of this circumstance, nor would such foreknowledge have 
deterred him from the use or coinage of a word which was his 
by indefeasible birthright to use or com. Scholars who dispute 
either of these propositions are free to tell us that we must 
not introduce new words: scholars who do not dispute them 
are not free to tell us so. It is of course necessary that the 
would-be emender of Aeschylus should be grounded in the 
elementary laws of the Greek language; but so much being 
granted, the addition of new ἅπαξ εἰρημένα to the scores 
already registered is not merely safe but imperatively necessary. 
It is manifest that the more unfamiliar a word is to a copyist 


the more likely is he to corrupt it; and thus no word runs 


such risk of corruption as a word which occurs but once. Now 
for the history of the error here. The incessant confusion of 
Γ and T produced eo@itas; this by the inversion of three 
letters became ec@a tis; the scribe then hastened to make 
Greek of the no-word eo@a, if indeed the A be not a mere 
iteration of the A. This inversion of three letters is a corruption 
with examples of which mss, Greek and Latin, abound; but 
I confine myself to the Medicean ms. Here then I find 
Aesch. P. V. 55 λαβὼν for βαλών, Ag. 762 κότον for τόκου, 
1367 μυθοῦσθαι for θυμοῦσθαι, Hum. 500 προσμένει for προσ- 
νεμεῖ, 719 μένων for νέμων, 730 δαίμονας for διανομάς, Soph. 
Ai. 1307 λέγων for γελῶν, Ant. 718 θυμῷ for μύθῳ, 965 ἠρέθιζε 
for ἠθέριζε, El. 567 ἐξεκίνησεν for ἐξενίκησεν, Phil. 680 ἔλαβ᾽ 
ὁ for ἔβαλεν, 1429 ἐκβαλὼν for ἐκλαβών, O. C. 475 βαλών 
for λαβών. In Aesch. Pers. 164 sq. we read καί με καρδίαν 


; 3 ’ δ τς > > ¢ ™ > a a 3 “Ὁ > a 
ἀμύσσει φροντίς" ἐς δ᾽ ὑμᾶς ἐρῶ | μῦθον, οὐδαμῶς ἐμαυτῆς 


οὖσ᾽ ἀδείμαντος, φίλοι, | μὴ KTA., Where ἐμαυτῆς has no mean- 
ing; neither in Weil’s conjecture μῦθον οὐδαμῶς ἐμαυτῆς οὐδ᾽ 
ἀδείμαντον can I get any satisfactory sense from the words 


262 THE JOURNAL OF PHILOLOGY. 


οὐδαμῶς ἐμαυτῆς : I propose és δ᾽ ὑμᾶς ἐρῶ, | θυμὸν οὐδαμῶς 
᾿ ἐμαυτῆς οὖσ᾽ ἀδείμαντος. Often too, as in our case of ATI for ITA, 
this inversion is united with the change of one letter into another 
letter of like shape or sound: Aesch. P. V. 397 προθυμεῖσθαι 
and προμηθεῖσθαι, 448 βαθύς for βυθός, Ag. 1605 ἐπὶ δέκ᾽ ἀθλίῳ 
for ἔλιπε καἀθλίῳ, 1621 γῆρας for piyos, Cho. 470 ἑκάς for ἄκος, 
Soph. 0. 7. 48 προθυμίας and προμηθίας, O. C. 550 ἀπεστάλη 
for ἐφ᾽ ἁστάλη. The alteration of δοκᾶν to δοκῶν is no marvel: 
a Doric gen. of this declension may count itself lucky if the 
scribe neither translates it to the common form nor transmutes 
it to an ace. sing. With the adscript iota ὁρᾶν is in our MSS 
barely distinguishable from ὁρᾶν. And now turn to Milton's 
sonnet on his late espoused saint and see how, though the dust 
of centuries lay thick upon the page of Aeschylus, one great 
poet unwittingly repeated the very phrase of another: ‘ But lo, 
as to embrace me she inclined, I waked, she fled, and day brought 
back my night.’ 

434. πτεροῖς and κελεύθοις cannot both have been written 
by Aeschylus; but I hardly know which of the rival conjectures 
to accept: perhaps Karsten’s κελεύθων is the simplest. 

436. ὑπερβατώτερα is just defensible, only just; though 
the numerous editors who retain it do not seem to be aware 
that it needs defence. The word ὑπερβατός nowhere else 
means passing or anything like it; and if it did, you would 
expect τῶνδ᾽ ὑπερβατά passing these, not τῶνδ᾽ ὑπερβατώτερα 
more passing than these. Still ὑπερβατός undeniably might 
have an active sense, and might possibly, by the same inaccuracy 
which gives us more transcendent in English, have a compara- 
tive and superlative. But that Aeschylus wrote this when by 
writing ὑπτερκοπώτερα he might have written what was more 
forcible, more like himself, and in our Mss most easily corrupted 
to ὑπερβατώτερα, I doubt. 

437—439. If τὸ πᾶν Evvoppévors could mean πᾶσε τοῖς 
ξυνορμένοις, if ξυνορμένοις πρέπει could mean πρέπει, TO τῶν 
ξυνορμένων μέρος, if πένθεια could mean πένθος, if τλησι- 
κάρδιος could mean καρδιόδηκτος, if δόμων could mean ἐν 
δόμοις, if these five impossibilities were possible, then the sense, 
if sense it can be called, which commentators elicit from these 


a 
3. ἃ 


THE AGAMEMNON OF AESCHYLUS. 263 


three lines, would be elicited legitimately. But as things are 
it is elicited by casting down the foundations of Greek. In 
particular, the verse πένθεια τλησικάρδιος is perhaps the most 


appalling sight which the Mss of Aeschylus have to shew. 


When σέβεια, πάθεια, θυμία, πραξία and the like are words, 
then πένθεια will be a word: till then it is a mere collocation 
of letters. δυσπένθεια is a word: it is found in no Greek 
author, but it is a word: πένθεια would remain no word were 
it found in fifty Greek authors. Scholars who tamely accept 
from a scribe such monstrosities as this, or as εὔπραξις in v. 
267, are not entitled to laugh when Prof. Newman presents 
them with ὁσιογέννημα. But admit for an instant the inad- 
missible, admit that wév@eva can mean πένθος: to a substantive 
of that meaning the adjective τλησικάρδιος can by no pos- 


‘sibility be applied, neither can τλησίφρων, ταλακάρδιος, 


ταλάφρων, ταλαίφρων, ταλασίφρων, τλήθυμος or φρεσὶν τλή- 
pov: both their two significations forbid it. As to the inter- 
pretation of the passage as a whole, all commentators, I believe, 
supply ᾿Αχαιοῖς with ξυνορμένοις. The antithesis between κατ᾽ 
οἴκους ἐφεστίους and ἀφ᾽ “EdXavos αἴας pointed by μὲν in v. 
435 and δὲ in v. 437 should have taught them to supply not 
᾿Αχαιοῖς but ᾿Ατρείδαις. With this for a clue, vv. 437—9 can 
be mended by the slightest of changes: These are the sorrows 
of the hearth, ay and worse than these; but, I guess, since quitting 


Hellas, these sorrows look like joy in their eyes, amid their suffer- | 


ings far from home. Great as was the misery of the Atridae 
in their deserted palace, yet their misery at Troy, to see their 
followers dying around them and to hear the threatening mur- 
murs of the survivors, is so much greater that they wish the 
former sorrow back: it seems desirable by contrast. So Soph- 
ocles says O. Οἱ 1697 πόθος τοι καὶ κακῶν ap ἦν Tis: Euripides 
puts the same thought in still stronger language, 770. 431 sqq., 
δύστηνος οὐκ οἶδ᾽ οἷά νιν μένει παθεῖν" | ὡς χρυσὸς αὐτῷ τἀμὰ 
καὶ Φρυγῶν κακὰ | δόξει ποτ᾽ εἶναι. It is of course to Mr 
Verrall’s paper in vol. IX. of this Journal that I owe τοπᾶν, 
a lost verb which has the meaning of the cognate τοπάξειν. 
For the infin. of a like verb used in like manner (τοπᾶν = ὡς 
τοπᾶν) see Soph. O. 7. 82 ἀλλ᾽, εἰκάσαι μὲν, ἡδύς : 80 very often 


264 THE JOURNAL OF PHILOLOGY. 


δοκεῖν ἐμοί. If any one is afraid of the word I counsel him to 
take Karsten’s τόπων, which also makes good sense. Since ὁ 
and ¢ are hardly to be known from one another, the only dif- 
ference between ποθεινά and πένθεια is the position of v: of 
this corruption I spoke in my note on v. 58. As for the correc- 
tion éxas που, in which I find myself forestalled by Ahrens, a 
becomes τ if the ink of the two downstrokes runs together: see 
Wecklein’s app. crit. on v. 1416 “εὐτόκοις, forte ut videtur ex 
εὐπόκοις factum, g.’ 

440. Those who render θυγγάνει as if it were χωρεῖ should 
favour us with a parallel. If the words are sound, as I think 
they are, we must understand αὐτῶν with Blomfield. I would 
not alter θυγγάνει, nor would I propose πόλλ᾽ ἀγοῖν. 

441 sq. The sum total of the changes which I have made 
here is no greater departure from the Ms than Porson’s insertion 
of rus after ydp, and the sense I surely improve: in the vulgate 
you must strain οἶδεν to make it mean μέμνηται and then your — 
imagination must furnish ἄγνωτα with τεύχη καὶ σποδὸς, to 
extort your antithesis. The construction of my text will be 
τοιοῦτον μὲν yap οὐδὲν, οἵους ἔπεμψαν ἑκάστου δόμοι, εἰς 
αὐτοὺς ἀφικνεῖται, τεύχη δὲ καὶ σποδὸς ἀντὶ φωτῶν : for unto 
each one's home there returns nought in the semblance of those 
whom it sent forth, but arms only and ashes in lieu of men. The 
frequent confusion of « and v accounts for the corruption of οἵους 
and οὐδὲν both, and the inflexion of ἔπεμψαν was accom- 
modated to that of οἶδεν, since metre forbade the converse error. 


498—504. 

, 5» ᾽» > ᾽ n Ud > ε lal , 
KNPUK aT ἀκτῆς τόνδ᾽ ὁρῶ κατάσκιον 
κλάδοις ἐλαίας" μαρτυρεῖ δέ μοι κάσις 
πηλοῦ ξύνουρος διψία κόνις τάδε, ὅ00 
ς yA eS ς wv / , τ 
ὡς οὔτ᾽ ἄναυδος οὔτε σοι δαίων φλόγα 
ὕλης ὀρείας σημανεῖ καπνῷ πυρὸς 
ἀλλ᾽ ἢ τὸ χαίρειν μᾶλλον ἐκβάξει λέγων--- 
τὸν ἀντίον δὲ τοῖσδ᾽ ἀποστέργω λόγον. 

The coryphaeus catching sight of the herald sees also in the 


distance a cloud of dust which he supposes to be raised by the 
returning army; and the return of the army means something 


THE AGAMEMNON OF AESCHYLUS., 265 


decisive, either victory or defeat. The crew of Agamemnon’s 
ship, if Aeschylus followed Homer, would be 120 men; and 
these, together with an ἁμαξήρης θρόνος for Agamemnon and 
Cassandra, would raise in clear dry southern air a cloud of dust 
to be seen a great way off. No doubt to us the allusion seems 
obscurely worded; but I fancy the Attic audience recognised an 
old friend. Of the plays of Aeschylus only a tithe has come 
down to us, but in that tithe we find Supp. 186 ὁρῶ κόνιν, 
ἄναυδον ἄγγελον στρατοῦ, and Sept. 79 sqq. μεθεῖται στρατὸς 
στρατόπεδον λιπών. | ῥεῖ πολὺς ὅδε λεὼς πρόδρομος ἱππότας. 
αἰθερία κόνις με πείθει φανεῖσ᾽ [ ἄναυδος σαφὴς ἔτυμος ἄγγελος. 
How many repetitions of the phrase lie foundered in the wreck 
of antiquity we cannot tell; but it may be guessed that by the 
time the poet wrote this play—three years before his death—he 
had so familiarised his hearers with the conception of κόνις as 
an ἄγγελος στρατοῦ that he could dispense with an explicit re- 
minder. The addition κάσις πηλοῦ Evvoupos is mere ornament 
like the αἰόλην πυρὸς κάσιν of Sept. 481. What coherent sense 
those scholars who take κόνις to be the stains of travel on the 
herald’s dress suppose themselves to extract from the passage, I 
have vainly tried to ascertain. 

My business however is with the σοι of v. 501. Mr Margo- 
‘liouth observes ‘ coz tibi, Clytemnestrae. That cou means tibi 
is very true, and it is equally indisputable that only Clytaem- 
estra can here be signified. But that is the very reason why 
σοι cannot be right; for it is as certain as anything about 
Greek plays can be certain that Clytaemestra is not now on the 
stage. The conjectures tov, μοι, Tor and γ᾽ αὖ bear witness to a 
due appreciation of the difficulty; and if the reader is satisfied 
with any one of them he will not trouble himself about me 
when I propose ὡς οὐκ ἄναυδος οὗτος ἀνδαίων φλόγα ὕλης 
ὀρείας σημανεῖ κτλ. The phrase ἀνδαίων φλόγα is chosen to 
recall the phrase in the ἀγγαρήιον at v. 317 πέμπουσι δ᾽ ἀν- 
δαίοντες ἀφθόνῳ μένει | φλογὸς μέγαν πώγωνα. In the Mss 
of Aeschylus the wrongful omission or insertion οὗ ν after a vowel 
is exceedingly common: it would seem indeed that the practice 
of denoting ν merely by a superscript line, usually confined to 
final syllables, must in some ancestral codex have prevailed in 


266 THE JOURNAL OF PHILOLOGY. 


all parts of words alike. The confusion of a with the diphthong 
ot, which is simply a resolved into its constituent elements, is 
of course a very frequent error. The facility with which ay 
might so pass through a to ox is illustrated by the confusions of 
-pavtos and -φοιτος which I adduced on ν. 429. Thus οὗτος 
ἀνδαίων is scarcely distinguishable from οὔτε cor δαίων, and 
οὔτε of course demanded the change of οὐκ to οὔτ᾽ : in fact οὐκ, 


οὔτ᾽ and οὐδ᾽ are for ever interchanged even when such excuse 
is absent. 


550—552.. 
KH. ποθεῖν ποθοῦντα τήνδε γῆν στρατὸν λέγεις. 
XO. ὡς πόλλ᾽ ἀμαυρᾶς ἐκ φρενός μ᾽ ἀναστένειν. 
ΚΗ. πόθεν τὸ δύσφρον τοῦτ᾽ ἐπῆν στύγος στρατῷ; 


The last word is of course precisely the reverse of what we 
want : we want πόλει or the like. The only conjecture which 
merits consideration is Heimsoeth’s λεώ (he writes it λεῷ), on 
which word he supposes στρατῷ was a mistaken gloss. My 
objection to this is that if Aeschylus wrote Xe he was gravely 
in fault for using so ambiguous a word: λαός in the Lliad is the 
regular name for the army at Troy. I should therefore much 
prefer to read πάτρᾳ. The close likeness of one form of w to 
στ is notorious, the confusion of a and ὦ common enough, and 
such transposition as has here been suffered by p I have already 
illustrated: watpa, otatpa, στρατῷ, στρατῷ is the facilis de- 
scensus. 7 


560—563. 
μόχθους yap εἰ λέγοιμι Kai δυσαυλίας, 
omapvas παρήξεις καὶ κακοστρώτους--τί δ᾽ οὐ 
στένοντες οὐ λαχόντες ἤματος μέρος: 
τὰ δ᾽ αὖτε χέρσῳ καὶ προσῆν πλέον στύγος. 


‘In a word, what was there we had not to complain about, 
or that we did not get for our daily share?’ This is Mr Paley’s 
translation of τί δ᾽ οὐ στένοντες κτλ. How ἤματος μέρος part 
of a day comes to mean daily share he makes no attempt to 
explain, nor does he say a word about the difference in tense of 


THE AGAMEMNON OF AESCHYLUS. 267 


στένοντες and λαχόντες ; but these are trifles, 1 pass them by. 
What rivets my attention is the absurdity of the question thus 
put in the herald’s mouth. What was there that we did not get 
for our daily share? Why, the things that they did not get for 
their daily share were like the sands of the sea for multitude. 
Their grievance was precisely this, that they got for their daily 
share no good thing whatever. But, to quit the translation for 
the text, of course one fatal objection which disposes at once of 
the MS reading and of half the conjectural essays is that the 
pendent nominative participles contravene not merely grammar, 
that is nothing, but the elementary rules of writing. An irregul- 
arity in grammar like Cho. 518 ta πάντα yap τις ἐκχέας av” 
αἵματος | ἑνὸς, μάτην ὁ μόχθος is not only legitimate but may 
be used with fine effect; but the man who thinks that because 
a nominative absolute can replace a genitive absolute therefore 
a participle can replace the principal verb of a sentence, is not 
destined to succeed in criticism. The most plausible of the 
conjectures which address themselves to the cure of this evil is 
perhaps Mr Margoliouth’s orévovtas, ἀσχάλλοντας. This how- 
ever though not very far from the MS is yet not very near: the 
change of the inflexions, in particular, is hard to explain: that 
I do not like the nature of the aposiopesis may be due merely to 
the fact that I have my own nostrum, which 1 will now pre- 
scribe. When I glance below at vv. 568—572 χειμῶνα δ᾽ εἰ 
λέγοι τις...ἢ θάλπος... .---τί ταῦτα πενθεῖν δεῖ; παροίχεται πόνος 
I cannot stitle the suspicion that in ν. 561 the words τί δ᾽ οὐ 
are simply the corruption, the very easy corruption, of τί δεῖ, 
breaking off the conditional sentence in a manner exactly 
parallel; and on this hint a touch or two will correct the lines. 


-- τί δεῖ 
, s / iA , 
OTEVOVTOS εὖ λαχόντας NT ATOS Mepos ; 


what do men of right temper want with a mourner? The con- 
struction is the familiar one of αὐτὸν yap σε δεῖ προμηθέως : 
for τέ δεῖ στένοντος compare Hum. 94 καθευδουσῶν τί δεῖ; for 

ἂν , \ 3 / a 04 ᾽ 
εὖ λαχόντας = τοὺς εὖ λαχόντας see ν. 39 μαθοῦσιν αὐδῶ κοὺ 

n / = Γ 

μαθοῦσι λήθομαι and a dozen more passages in Aeschylus. The 
phrase εὖ λαχόντας ἥπατος μέρος finds a counterpart in v. 391 


268 THE JOURNAL OF PHILOLOGY. 


εὖ πραπίδων λαχόντα: the poet might have written εὖ mpa-. 
πίδων λαχόντα μέρος in that place or εὖ λαχόντας ἥπατος in 
this without a whit of difference to the sense. λαχεῖν τινος 
without μέρος is the more common, but you have Theogn. 353 
εἰ γνώμης ἔλαχες μέρος and Soph. Ant. 918 οὔτε του γάμου] 
μέρος λαχοῦσαν: μεταλαχεῖν τινος but also Eur. Supp. 1078 
μετέλαχες τύχας Οἰδιπόδα, γέρον, μέρος : μετέχειν τινός but 
also Ag. 512 μεθέξειν φιλτάτου τάφου μέρος : μέτεστί τινος 
but also Eur. J. 7. 1299 μέτεστι χὐμῖν τῶν πεπραγμένων μέρος, 
wrongfully suspected by Nauck and Wecklein. Throughout 
tragedy the ἧπαρ is the part of the mind or soul which feels 
regret and remorse : 6 ev ἥπατος λαχὼν then is the man whose 
ἧπαρ is proof against the excess of these emotions, the man 
who is not the victim of self-tormenting regrets for the irreme- 
diable past, of the ἄλγος παλίγκοτον disapproved in v. 576 if 
we there accept the correction of H. L. Ahreas as those must 
who are not prepared to invent a new meaning for παλίγκοτος. 
The herald therefore, as J understand him, checks himself in the 
midst of his recital with the reflexion that men of a right and 
happy temper of soul, like those to whom he speaks, do not 
want to hear unavailing lamentation over the past. But then 
at v. 563 other hardships recur to his mind and spring to his 
lips: then he checks himself again at v. 572 with a similar 
reflexion. 

It will be seen that the incessant confusion of ε and o is 
responsible for the change of δεῖ στένοντος εὖ into δ᾽ οὐ στέ- 
vovtes ov: the change of inflexion in λαχόντας was due of 
course to the στένοντες thus produced. The confusion of II 
with that form of M in which one slightly curved line replaces 
the two diagonal strokes is chargeable with the errors ὅπως for 
ὅμως in v. 980, βλέπει for βρέμει in v. 1015 and δυσπαθῆ 
for δυσμαθὴ in v. 1254: besides the ἤματος for ἥπατος of our 
passage I think I detect another instance in the play. In v. 
1432 sqq. καὶ τήνδ᾽ ἀκούεις ὁρκίων ἐμῶν θέμιν" | μὰ τὴν τέλειον 
τῆς ἐμῆς παιδὸς Δίκην κτλ. the use of ὁρκίων for ὅρκων is 
improper, and though ὅρκων θέμις might perhaps stand for 
ὅρκος θεμιστός lawful oath yet it is quite without point to 
call the oath which follows either lawful or unlawful: Cly- 


THE AGAMEMNON OF AESCHYLUS. 269 


taemestra simply swears that she is not afraid. When I con- 
template this verse there rings in my ear another, Cho. 498: 
καὶ τῆσδ᾽ ἄκουσον λοισθίου Bons, πάτερ. Should we not write 
καὶ τῶνδ᾽ ἀκούοις ὁρκίων ἐπῶν, Oéur? The restorations are all 
easy, ὦ for ἡ, ot for εἰ, 7 for w, + for t Mr Margoliouth reads 
καὶ τήνδ᾽ ἀκούειν ὁρκίαν φήμην θέμις : this meaning might be 
obtained at less expense by καὶ τῶνδ᾽ ἀκούειν σ᾽ ὁρκίων ἐπῶν 
θέμις, but I should still prefer the reading given above. 


886—8 94. 


na a / re ΨΞΙΙ. 3 / ‘ 
νῦν, ταῦτα πάντα τλᾶσ᾽, ἀπενθήτῳ φρενὶ 
/ > ΝΝ v , > 2 \ a / 
λέγοιμ᾽ av ἄνδρα τόνδ᾽ ἐγὼ σταθμῶν κύνα, 
σωτῆρα ναὸς πρότονον, ὑψηλῆς στέγης 
στῦλον ποδήρη, μονογενὲς τέκνον πατρὶ, 
καὶ γῆν φανεῖσαν ναυτίχοις παρ᾽ ἐλπίδα, 890 
t 3 > a > / 
κάλλιστον ἦμαρ εἰσιδεῖν ἐκ χείματος, 
/ an -Ὁ 4 
ὁδοιπόρῳ διψῶντι πηγαῖον ῥέος. 
\ \ > a 3 - “ 
(τερπνὸν δὲ τἀναγκαῖον ἐκφυγεῖν ἅπαν.) 
τοιοῖσδέ τοί νιν ἀξιῶ προσφθέγμασιν. 


In these lines as they stand three serious faults have been 
pointed out. Firstly, it is plain that when you call a man by 
seven προσφθέγματα you do not, if you have regard to style, 
connect the fourth and the fifth by «ai leaving the rest ἀσύν- 
Sera. To mend this fault Blomfield proposes γαῖαν for καὶ 
γὴν in v. 890, a conjecture which gets some support from Hum.. 
758 where Dindorf’s γαίας for καὶ γῆς seems a probable emenda- 
tion. Secondly, that Aeschylus did not put v. 893 where it 
now stands, severing v. 894 from the προσφθέγματα to which it 
refers, is evident to every one who understands, I do not say 
the art of poetry, but I say the art of writing respectable verse. 
To mend this fault Enger inserts v. 893 as a parenthesis 
between 886 and 887. Thirdly, I cannot help feeling, with 
Hermann and Meineke, that the superlative κάλλιστον in 
v. 891 as an epithet to ἦμαρ gravely impairs the force of the 
phrase. To mend this fault the conjectures γαληνὸν and yavu- 
στὸν have been proposed, but of course are only valuable as 
testifying a perception of the difficulty. 


270 THE JOURNAL OF PHILOLOGY. 


I have hit on a device, which seems at least as simple as 
- Blomfield’s and Enger’s, to mend all three faults at once. 
I propose to remove the four verses 890—893 from their pre- 
sent seat, so that μονογενὲς τέκνον πατρί shall be followed by 
τοιοῖσδέ τοί νιν ἀξιῶ προσφθέγμασιν, and to insert them nine 
lines lower down, with one slight change, in the following 
order : 


> \ / / / 
εὐθὺς γενέσθω πορφυρόστρωτος πόρος 


ἐς δῶμ᾽ ἄελπτον ὡς ἂν ἡγῆται δίκη. 902 

κάλλιστον ἦμαρ εἰσιδεῖν ἐκ χείματος, 891 

ὁδουπόρῳ διψῶντι πηγαῖον ῥέος 892 

καὶ γῆ φανεῖσα ναυτίλοις παρ᾽ ἐλπίδα" 890 
\ \ > a > ad ¢ 

τερπνὸν δὲ τἀναγκαῖον ἐκφυγεῖν ἅπαν. 893 


When the word ἄελπτον has passed Clytaemestra’s lips, its 
ring of menace is so clear to her guilty ear that she hastens to 
obscure its real significance by resort to the familiar saw that 
unexpected pleasure is sweetest. Sweetest, she says, is sunshine 
after storm, water to a traveller athirst, land ahead when the 
mariner least expects it; and indeed there is pleasure in all 
escape from stress of fortune: therefore, she implies, Agamem- 
non’s home-returning, in which both elements of pleasure, 
Tavaykaiov ἐκφυγεῖν and τὸ ἄελπτον, are combined, is sweetest 
of all. It is noticeable that her speech at the end of this 
episode is closed with a similar digression on a word, vv. 963— 
965: she has said ἀνδρὸς τελείου, and she bursts out Zed Zed 
τέλειε, TAS ἐμὰς εὐχὰς τέλει. In v. 891 κώλλιεστον, transformed 
from attribute to predicate, is now without offence: the corrup- 
tion in v. 890 arose from the adhesion to φανεῖσα of the initial 
ν of ναυτέλοις, whence the further change, by assimilation, of 
γῇ to γῆν. 

I seem to myself to find external confirmation of this arrange- 
ment in two passages imitated, I think, from this. One is the 
distich, ascribed to the name Asclepiades, which Blomfield cites : 
ἡδὺ θέρους διψῶντι χιὼν ποτὸν, ἡδὺ δὲ ναύταις | ἐκ χειμῶνος 
ἰδεῖν εἰαρινὸν στέφανον, almost a paraphrase of the verses as 
I write them. The second is Eur. fr. 552 Dind. ἐκ τῶν ἀέλπττων 
ἡ χάρις μείζων βροτοῖς | φανεῖσα μᾶλλον ἢ τὸ προσδοκώμενον. 


THE AGAMEMNON OF AESCHYLUS. 271 


‘Vs. 2 aut spurius aut corruptus ’ says Nauck : ‘corrupt, surely 
not spurious’ says Munro: neither spurious nor corrupt, I must 
take heart to say. Construe φανεῖσα with ἐκ τῶν ἀένπτων, and 
for the pleonasm μείζων μᾶλλον see Hec. 377 θανὼν δ᾽ ἂν εἴη 
μᾶλλον εὐτυχέστερος | ἢ Cav. 


990---1009. 


μάλα βροτοῖσι τᾶς πολλᾶς ὑγιέας str. 
\ 
ἀκόρεστον τέρμα. νόσος yap 
γείτων ὁμότοιχος ἐρείδει" 
καὶ πότμος εὐθυπορῶν 
ἀνδρὸς ἔπαισεν «ἄφνω 
/ v “ 
δυστυχίας» ἄφαντον ἕρμα. 

Ν \ \ \ t - 
καὶ τὸ μὲν πρὸ χρημάτων 995 
κτησίων ὄκνος βαλὼν 

/ > 3 ’ / 
σφενδόνας ἀπ᾽ εὐμέτρου 

> " / / 
οὐκ ἔδυ πρόπας δόμος 
πημονᾶς γέμων ἄγαν 
οὐδ᾽ ἐπόντισε σκάφος" ' 1000 

4 
πολλά τοι δόσις ἐκ Διὸς ἀμφιλαφής τε καὶ 
> ’ , 3 a 
ἐξ adoxwy ἐπετειᾶν 
νῆστιν ἤλασεν νόσον. 


\ 3 3 \ A ¢/ \ / 
τὸ δ᾽ ἐπὶ γᾶν ἅπαξ πεσὸν θανασίμου ant. 
πρόπαρ ἀνδρὸς τίς μέλαν αἷμ᾽ ἂν 1005 


πάλιν ὠἀγκαλέσαιτ᾽ ἐπαείδων ; 
τοῦδε τὸν ὀρθοδαὴ 
A / ,’ / 
τῶν φθιμένων ἀνάγειν 
\ ’ , 
Ζεὺς ἀπέπαυσ᾽ ἐπ᾽ εὐλαβείᾳ. 


Vv. 990---4 and 1004—9 I have restored provisionally to 
show the drift of the whole passage, but without firm faith that 
I am giving the precise words of Aeschylus. The metres of 
strophe and antistrophe now correspond, except twice in ana- 
crusis, where correspondence is not required; though perhaps 
γείτων in v. 992 is a gloss on πελάτας. How much of the 
change is my own and how much borrowed I leave the curious 
reader to seek from Wecklein. But on vv. 995—1000 I hope 
I have something definite to say. : 


272 THE JOURNAL OF PHILOLOGY. 


With the pendent nominative ὄκνος βαλὼν I do not quarrel : 
harsh it may be, but it has ample warrant. But anyone who 
will consult the lexicons will see that of all the Greek words 
which we render in English by fear, ὄκνος is in this place the 
least appropriate: ὄκνος connotes shrinking and sluggishness, 
not the alertness and presence of mind which saves the ship 
by casting away the cargo. Further I should like to know what 
sense editors attach to σφενδόνας ἀπ᾽ εὐμέτρου from a sling of 
just proportions. Mr Paley renders ‘by a well-calculated 
throw’; on which I have to remark, first, that no engines of 
torture will wring that meaning from the Greek, and secondly 
that any such meaning is entirely irrelevant to the context. 
If you want to hit a mark, then you require a well-calculated 
(εὔστοχος) throw: but it is news to me that you need precision 
of aim to pitch your goods overboard. And setting aside these 
details I wish to point out that the general sense yielded by 
vv. 995—997, or perhaps I should rather say extorted from 
them, is not the sense which the context demands. 

So far as I can discover the general opinion about the con- 
nexion of vv. 995—1009, that opinion is demonstrably wrong. 
All editors, I believe, put a full stop where I have put a colon 
after σκάφος in v. 1000, and all, I presume, summarise with Mr 
Sidgwick thus: ‘a labouring boat may be saved (995—1000), a 
famine averted (1001—3), but blood once -shed is irrevocable 
(1004---9). 10 will not be denied, when it is once asserted, 
that this would require a connecting particle in v. 1001. 
When Mr Paley translates ‘doubtless too a good supply...puts 
an end to famine, and when Dr Kennedy translates ‘And 
truly gifts abundant...have brought...famine to an end, those 
scholars forge this necessary link on the anvil of their imagina- 
tion: in the Greek they do not find it. Mr Sidgwick says ‘the 
connexion is abrupt: perhaps τοῦ is corrupt. But a right 
account of the connexion will show that τοῦ is quite sound. 
The chorus in vv. 990—4 lay down the doctrine, recurring in 
fifty passages of Greek verse and prose, that over-great pros- 
perity brings ruin, strikes im mid career on a reef. Yet (995— 
1000: καὶ = atque) if a man then sacrifice his substance he may 
avert utter destruction from his house: God (1001—38) can 


THE AGAMEMNON OF AESCHYLUS. 273 


restore him substance enough for his wants. But (1004—9) if 
that which is sacrificed be not man’s substance but man’s life, 
that none can restore. This seems clear and coherent sense, 
and the only sense which the form of the passage permits: vv. 
1001—3 then have reference to 995—7. Now let me ask 
attention to the words νῆστιν νόσον in v. 1003: what brings 
this mention of famine? Does famine come from throwing 
overboard a part (τὸ μὲν) of your cargo? No: from throwing 
overboard the whole. 


καὶ γόμον πρὸ χρημάτων 
/ \ \ 
κτησίων κενὸς βαλων 
΄ 3 > > / 
σφενδόνας am εὐμέτρου 
οὐκ ἔδυ πρόπας δόμος. 


Yet if the house cast overboard, till nought be left, its freight of 

possessions from a capacious sling (in English metaphor with 
unstinting hand), it sinks not utterly. The epithet εὐμέτρου 
now has its proper force of just proportions, that is, large 
enough to hold the entire freight. It is usually said that πρὸ 
and βαλὼν are in tmesi: I should prefer to call πρὸ the ad- 
verb. The change of romon to to men I need not explain; but 
perhaps I should say a word on the corruption of κενός. In my 
note on v. 133 I gave instances of two consecutive letters re- 
versed: here we have this error combined with alteration of one 
out of the two letters. This mistake is rendered the easier by 
the custom of writing one letter over another for brevity’s sake. 
The rule requires that the superscript letter should be read as 
the latter of the two, but it often happens that an ignorant 
or inattentive scribe will misinterpret the abbreviation when 
copying it out at length. This is of course the more likely to ᾿ 
occur if one of the two letters has been corrupted: when the 
correct order no longer gives a Greek word the copyist tries to 
obtain one by inversion. Similar errors to ὄκνος for κενός are 
Ag. 980 ὑμνῳδεῖ contra metrum for wov@de?, Cho. 661 λεχθεῖσιν 
for λέσχαισιν, 700 δυσσεβείας for δ᾽ εὐσεβείας, Supp. 230 
κρέκω for κίρκων, Sept. 452 εἰσημάτιστα for ἐσχημάτισται, 682 
αἰσχρά for ἐχθρά, a mistake which recurs in Soph. Phil. 1284, 
ἔχθιστος fur αἴσχιστος. 


Journal of Philology. vou. χυι. 18 


274 THE JOURNAL OF PHILOLOGY. 


But we are not out of the wood yet. The phrase πημονᾶς 
γέμων is, in its proper place, which is not here, good sense: the 
phrase πημονᾶς γέμων ἄγαν is, and I am glad to see Weil 
thinks so too, ridiculous: as if there were such a thing as πημο- 
vas γέμειν μετρίως! But further: whether my reading of vv. 
995—7 be accepted or no, it is equally certain that the ship is 
there represented as laden not with πημονή but with χρήματα 
κτήσια. Therefore we have to say that the χρήματα them- 
selves are here called πημονή as leading to disaster by their 
too great abundance. Now perhaps there are places where 


wealth can be called πημονή, but this is a place where it can- 


not: it cannot be called πημονή when it has just been called 
ὑγίεια. And the word ἄγαν survives to make nonsense of the 
present text and to tell us that just as Aeschylus began with 
the danger of ἡ πολλὴ ὑγίεια, so he here described the sinking 
ship as γέμων ἄγαν not πημονῶς but ὄλβου or the like, γέμων 
Tov πολλοῦ ὄλβου. If the reader will turn to the parallel pas- 
sage Sept. 753 sqq. he will find, I think, strong confirmation 
both of the correction I have made in vv. 995 and 996 and of 
the correction I am about to make in v. 999. The passage is 
this: τὰ δ᾽ ὀλοὰ πενομένους παρέρχεται, | TpoTpupva δ᾽ ἐκβο- 
λὰν φέρει | ἀνδρῶν ἀλφηστᾶν ὄλβος ἄγαν παχυνθείς. Here 
πρόπρυμνα (or πρόπρεμνα) ἐκβολὰν, rendered by the scholiast 
ὅλου τοῦ φόρτου ἐκβολήν, is precisely my γόμον πρὸ κενὸς 
βαλών; and ὄλβος ἄγαν παχυνθείς is δόμος παμονᾶς γέμων 
ἄγαν. : 

The word παμονή, which the lexicons do not contain, is to 
Tapa as πημονή to πῆμα, χαρμονή to χάρμα and πλησμονή to 
πλῆσμα. Against the entire family of words akin to πέπαμαι 
the copyists πνέουσιν ἄσπονδον “Apn. The verb itself, which is 
common enough, usually escapes with no worse injury than the 
mis-spellings πέπαμμαι and ἐπασσάμην, but even the verb 
sometimes perishes. Thus in Soph. 0. C. 528 Nauck has to 
restore δυσώνυμα λέκτρ᾽ ἐπάσω for ἐπλήσω, and Wecklein 
"Avda μόνον φεῦξιν ov πεπάσεται for οὐκ ἐπάξεται in Ant. 362. 
The word πάτωρ has escaped death— Asda φεῦξιν πέπατα:--- 
only in Photius who has waropes: κτήτορες : elsewhere πατήρ 
has swallowed it: Hesych. πάτορες [W. Dindorf, πατέρες MS}: 


THE AGAMEMNON OF AESCHYLUS. 275. 


πλούσιοι, Eur. fr. 654 Dind. χρημάτων πολλῶν κεκλῆσθαι 
βούλεται πάτωρ [W. Dindorf, πατὴρ Mss] δόμοις, Phoen. 473 
ἐγὼ δὲ πάτωρ [Munro, πατρὸς MSS] δωμάτων προὐσκεψάμην | 
τοὐμόν τε καὶ τοῦδ΄. The word πολυπάμων has escaped in 
Iliad δ 433 and in Hesych. πολυπάμονος" πολλὴν κτῆσιν 
ἔχοντος, and πολυπάμων" πλούσιος, πολλὰ κεκτημένος, πολυ- 
χρήμων, πάματα γὰρ τὰ χρήματα; but it has perished in Soph. 
El. 515 οὔτι πω] ἔλιπεν ἐκ τοῦδ᾽ οἴκους | πολύπονος aixia, 
where the scholion τοὺς πολυκτήμονας δόμους points as 
Schneidewin has seen to the reading οἴκους πολυπάμονας. 
The word βουπάμων had to be restored for ᾿βουπαλίων by 
Valckenaer in an epigram of Leonidas Tarentinus (51 tom. 1, 
p. 167 Anth. Gr. Jacobs). Hesychius offers other corruptions : 
ἐπιπαματίδα |Maussacus,.ériatida MS|* τὴν ἐπίκληρον ; ἐμπά- 
pove [Hemsterhuys, ἐμπαγμῷ MS)’ πατρούχῳ; αὐτοπάμονα 
{Hemsterhuys, αὐτόπομα Μ5] ἐπίκληρον. The word πᾶσις 
survives only in Hesych. πᾶσις" κτῆσις, but I propose to 
restore it once at least to Euripides. In Andr. 192 sqq. the 
heroine thus ridicules the jealousy of Hermione: εἴπ᾽, ὦ veavt, 
τῷ σ᾽ ἐχεγγύῳ λόγῳ | πεισθεῖσ᾽ ἀπωθῶ γνησίων νυμφευμά- 
των; ὡς τῆς Λακαίνης ἡ Φρυγῶν μείζων πόλις | τύχῃ θ᾽ 
ὑπερθεῖ, κἄμ᾽ ἐλευθέραν ὁρᾷς ; [ ἢ τῷ νέῳ τε καὶ σφριγῶντι σώ- 
ματι | πόλεώς τε μεγέθει καὶ φίλοις ἐπηρμένη | οἶκον κατα- 
σχεῖν τὸν σὸν ἀντὶ σοῦ θέλω; Τὸ 15 ;plain that in v. 197 the 
words πόλεως τε μεγέθει cannot be right: the disparity between 
the one πόλις and the other has already been dealt with in 
v. 194; so Brunck amends the sense by writing πλούτου. A far 
slighter change suffices: moAewc stands merely for teaewc, which 
is macewc with one letter misplaced. In Soph. £1. 837 sqq. is 
this sentence: οἶδα γὰρ ἄνακτ᾽ ᾿Αμφιάρεων χρυσοδέτοις ἕρκεσι 
κρυφθέντα γυναικῶν" καὶ νῦν ὑπὸ γαίας πάμψυχος ἀνάσσει. 
The two last words are diversely interpreted to mean πασῶν 
ψυχῶν ἀνάσσει or ἀθάνατος ἀνάσσει or πάντι σθένει ἀνάσσει. 
The advocates of each rendering are so triumphantly successful 
in the easy task of exploding the other two that I can pass the 
question by to say that I should change one letter and write 
παμοῦχος ἀνάσσει he is lord and king: see Hesych. παμῶχος" 
ὁ κύριος, the Doric form. ἐπίπαμα is preserved, wrongly spelt, 
18—2 


276 THE JOURNAL OF PHILOLOGY. 


in the Theocritean scholia, but not in Eur. J. 7. 414 The 
‘chorus there enquire concerning Orestes and Pylades are they 
merchants who sail the sea φιλόπλουτον ἅμιλλαν | αὔξοντες 
μελάθροισιν; | φίλα yap ἐλπὶς γένετ᾽ ἐπὶ πήμασιν βροτῶν | 
ἄπληστος ἀνθρώποις, | ὄλβου βάρος οἱ φέρονται | πλάνητες ἐπ᾽ 
οἶδμα πόλεις τε βαρβάρους περῶντες. The greater part of this 
antistrophe corresponds very accurately with the strophe; but 
the verse φίλα γὰρ ἐλπὶς κτλ. differs greatly from the strophic 
verse 399 tives ποτ᾽ dpa τὸν εὔυδρον δονακόχλοον, and moreover 
makes no sense. I would restore meaning and correspondence 
thus: φίλα yap ἐγένετ᾽ ἐλπὶς ἔν τ᾽ ἐπιπάμασιν ἄπληστος 
ἀνθρώποις, where ἐπιπάμασιν is my own conjecture, the other 
alterations borrowed: dear unto men is hope, and insatiable in — 
acquisition. Finally, the word vaya, outside lexicographers and 
scholiasts, seems to be found only twice in Greek literature : 
Theoer. Fistul. 12 and Dosiad. Anth. Pal. 15. 25. 5, pointed out 
by Valckenaer Animadv. ad Ammon. lib. 3, cap. 7- But either 
I am thoroughly mistaken, or two more instances are to be dis- 
interred from the text of Aeschylus. In Sept. 926 sqq. the 
chorus lament over Eteocles and Polynices who have ended ‘by 
mutual slaughter their contention for sovereignty: πικρὸς δὲ 
χρημάτων | ἴσος δατητὰς “Apns, ἀρὰν | πατρῴαν τιθεὶς ada). | 
ἔχουσι μοῖραν λαχόντες, ὦ μέλεοι, | διοσδότων ἀχέων" | ὑπὸ δὲ 
σώματι yas | πλοῦτος ἄβυσσος ἔσται. The verse διοσδότων 
ἀχέων should answer metrically to διατομαῖς οὐ φίλαις in the 
strophe. Meineke’s διαδότων parted between them appears to be 
the first step towards emendation: for a and oo confused see 
Porson on Eur. Hec. 788. But now are we to write ἀφίλοις with 
H. Voss in the strophe, or alter dyéwy in the antistrophe ? 
Assuredly the latter; for ἀχέων, quite apart from metre, does 
not give a right sense. The ἀρὰ πατρῴα whose fulfilment these 
lines describe was (773 544.) σιδαρονόμῳ διὰ χερί ποτε λαχεῖν 
κτήματα, not ἄχη. Aeschylus seems to have written διαδότων 
παμάτων: some perversely ingenious reader chose to regard 
this as Doric for πημάτων, and signified his opinion by writing 
ἀχέων above it: then the gloss, as usual, expelled the genuine 
word. With the μοῖραν λαχόντες παμάτων thus restored com- 
pare vv. 890sq. ἐμοιράσαντο δ᾽ ὀξυκάρδιοι κτήμα θ᾽ ὥστ᾽ 


THE AGAMEMNON OF AESCHYLUS. 277 


ἴσον λαχεῖν. Last of all comes an instance of πᾶμα from the 
Agamemnon itself. Look at vv. 1567 sqq. 


ἐγὼ δ᾽ οὖν 
ἐθέλω δαίμονι τῷ Πλεισθενιδῶν 
ὅρκους θεμένη τάδε μὲν στέργειν 
δύστλητά περ ὄνθ᾽" ὁ δὲ λοιπὸν ἰὼν 1570 
> nA / Vv A 
ἐκ τῶνδε δόμων ἄλλην γενεὰν 
/ U ’ / 
τρίβοι θανάτοις αὐθένταισιν. 
κτεάνων τε μέρος 
βαιὸν ἐχούσῃ πᾶν ἀπόχρη μοι 
μανίας μελάθρων 1575 
ἀλληλοφόνους ἀφελούσῃ. 


The τε of v. 1573 is plainly insufferable. Auratus’ δὲ is suffer- 
able but still an encumbrance: the connecting particle should 
be ydp or there should be no connecting particle. Nor is this 
the only objection I feel: wav ἀπόχρη μοι anything suffices me 
I could understand, and I could understand μέρος βαιὸν ἐχούσῃ 
ἀπόχρη μοι tt suffices me to have a small portion; but μέρος 
βαιὸν ἐχούσῃ πᾶν ἀπόχρη pot is a string of words which I am 
unable to construe, for wav is not the same thing as παράπαν. 
Mr Paley translates ‘I am content to keep even a small part 
out of all my possessions’: that is to say, he renders πᾶν as a 
gen. plur. agreeing with κτεάνων: He does not translate τε, but 
ignores it. And now for the remedy. It must be observed 
that we have clear evidence of dislocation in this passage: the 
generally accepted μανίας μελάθρων ἀλληλοφόνους is Erfurdt’s 
correction for δ᾽ ἀλληλοφόνους μανίας μελάθρων, in which the 
meaningless 6’ is an insertion to cure the hiatus caused by the 
- displacement. I propose a similar transposition and the change 
of one letter: 


βαιὸν ἐχούσῃ 
ge ᾽ ῇ 
πᾶμ᾽ ἀπόχρη μοι κτεάνων τε μέρος. 


The likeness of μ to ν, and the fact that a Byzantine copyist, if 
he knew the word at all, knew it only in the form πάμμα, make 
the alteration as easy as an alteration can well be. 


278 THE JOURNAL OF PHILOLOGY. 
190. 


? 7 =? \ / ae \ / , 
αλλ nv παλαιστῆς KapT EOL πνέων χάριν. 


“παλαιστής a suitor, lit. a wrestler; one of Aesch.’s pic- 
turesque and bold words’ writes Mr Sidgwick, representing, lL 
suppose, the general opinion. That it is bold to say wrestler 
when you mean suitor I cordially agree; but in what way it 
is picturesque, what picture it should present, I cannot guess. 
The term would be perfectly right and apt on the lips of 
Marpessa: to her Apollo really and truly ἦν παλαιστὴς κάρτα 
πνέων χάριν, When he contended with Idas for her hand. But 
story knows no rival of Apollo’s in the suit of Cassandra; and 
the only picture which παλαιστής could present to an audience 
not specially informed beforehand is the picture of one beating 
the air. I should like then to substitute for a word which can- 
not here mean suitor a word which can: πελαστής. This is of 
course a legitimate formation from πελάζω, and is preserved by 
Ammonius: the cognate πελάτης is employed in the required 
sense of temptator by Sophocles Phil. 678 τὸν πελάταν λέκτρων 
ποτὲ τῶν Διὸς Ἰξίονα. Having regard to the constant inter- 
change of e and au the reader will see that this is an instance of 
that inversion of three consecutive letters which I illustrated on 
v. 431. The very same confusion occurs in Eur. 1.17΄. 881, where 
one of the two Mss which contain the play gives πρὶν ἐπὶ ξίφος 
αἵματι og πελάσαι, the other παλαῖσαι. 


13821—1325. 
ἅπαξ ἔτ᾽ εἰπεῖν ῥῆσιν ἢ θρῆνον θέλω 
ἐμὸν τὸν αὐτῆς. ἡλίῳ δ᾽ ἐπεύχομαι 
πρὸς ὕστατον φῶς, τοῖς ἐμοῖς τιμαόροις 
ἐχθροῖς φονεῦσι τοῖς ἐμοῖς τίνειν ὁμοῦ 
δούλης θανούσης εὐμαροῦς χειρώματος. 1395. 


Once more I fain would speak. my own harangue or dirge is 
a deplorable specimen of style.. But bad as it is there is some- 
thing worse, Hermann’s ov for 7; an alteration which not only 
fails to remove the totally inappropriate ῥῆσιν but even intro- 
duces a fresh defect: it is, as Weil says, most certain that if 
Aeschylus had written ov θρῆνον he could not have added ἐμὸν 


a 
j 

Ν 
᾿ 
᾿ 
a 
Ἢ 
. 
; 


THE AGAMEMNON OF AESCHYLUS, 279 


τὸν αὐτῆς. In justice to Hermann it should be said that he 
himself put forward the conjecture with little confidence: he 
would marvel at its present vogue. Nor do I find among the 
other guesses enumerated by Wecklein a single proposal which 
repairs the passage with any critical probability. | 

I seem to find here the traces of a strange hallucination in 
the scribe, incredible to those who have not studied MSs and 
unnoticed by many of those who have, but not rare in Latin and 
not unknown in Greek: I mean the wholesale permutation of 
the letters which constitute a word. The letters ρησινηθ are the 
letters which in their proper order constitute the word ἠριθνής. 
The word, I say, for νεοθνής and ἡμιθνής on the one hand, and 
ἠρυγέρων and ypuyéveca (Aesch. fr. 346 Dind.) on the other, will 
vouch for ἠριθνής dying rathe, though the lexicons know it not. 
This correction I think will add force to ἅπαξ ἔτι: since she 
must perish, and perish before her time, the double bitterness of 
her fate cries for a second dirge; so she enters the palace to 
chant it and there at last κύκνου δίκην | τὸν ὕστατον μέλψασα 
θανάσιμον γόον | κεῖται. 

Were I to illustrate this error of permutation as fully as I 
might, and as I hope to do some other day, I should have to 
stray far from the Agamemnon, so I content myself with one 
more instance which the text of the play will furnish: vv. 
1537 sq. 


Δίκα δ᾽ ἐπ᾿ ἄλλο πρᾶγμα Onyaver βλάβας 
πρὸς ἄλλαις θηγάναις μοῖρα. 


I have written down these lines with Hermann’s θηγάνει for 
the unmetrical θήγεν which has come from λήγει in the line 
above, and with the necessary correction of βλάβης to the 
Doric form. In addition to these changes it is usual to read 
for metre’s sake θηγάναισι, and Δίκαν for the sake of a construc- 
tion. The sense thus elicited is given by Mr Paley as follows: 
‘Fate is whetting (the sword of) Justice upon another whet- 
stone, for a new business of harm.’ Justice I find in the Greek, 
but as for her sword, προχαλκεύει Mr Paley φασγανουργός : the 
words mean, as Dr Kennedy renders, ‘ Fate is sharpening Jus- 
tice’; and the picture of this august divinity whetted on a hone 


280 THE JOURNAL OF PHILOLOGY. 


like some article of cutlery must be to any mind a ridiculous 
image and to the mind of Aeschylus an impious one. Beyond 
comparison the best conjecture ventured is Musgrave’s Δέκα... 
Onyavais μάχαιραν, which gives precisely the sense required ; 
but we can come even nearer to the Mss: 


Δίκα δ᾽ ἐπ᾿ ἄλλο πρᾶγμα Onyaver βλάβας 

πρὸς ἄλλαις θηγάναισιν ἄορ. 
Allow for the confusion of ν with yw, and the letters of waop and 
of wotpa are the same. 

I return to vv. 1321—5. In 1322 Jacob’s ἡλίου is necessary 
and now generally accepted. Than the ensuing sentence there 
is, it will be confessed, no sentence in tragedy more indisputably 
corrupt. But in my opinion there is also no sentence in tragedy 
more simply and certainly corrigible. The corrections which 
I am about to propose occurred to me the very first time I read 
the passage with attention. I am. therefore not surprised to 
find from Wecklein that the main points of the emendation 
were anticipated years before: surprised I am that these ob- 
vious corrections are suffered to. lie ‘Orci tradita thesauro’, 
while scholar on scholar pours forth conjectures which rival one 
another in rashness and lack of meaning. In particular, the 
number of distinguished critics, from Musgrave onwards, who 
‘ have mistaken τίνειν for τίνεσθαι, is confounding. 

First we must have a subject for τίνειν : this, since tive 
after all is not τένεσθαι, will be ἐχθροὺς...τοὺς ἐμούς. Next we 
must have an object for τίνειν which shall also furnish the gen. 
δούλης with a construction: this we shall seek in the otiose 
and misplaced φονεῦσι; and there it is, φόνευσιν. Thirdly, 
ὁμοῦ would tell us, if common sense did not, that Cassandra’s 
prayer is not the absurd one that her own avengers may avenge 
her, but that certain destined avengers of some one else may in 
avenging him avenge her also: therefore ἐμοῖς in 1323 1 is corrupt 
and shall be replaced by véors.. 


ἡλίου δ᾽ ἐπεύχομαι. 
πρὸς ὕστατον φῶς, τοῖς νέοις τιμαόροις 
ἐχθροὺς φόνευσιν τοὺς ἐμοὺς τίνειν ὁμοῦ 
δούλης θανούσης εὐμαροῦς χειρώματος. 


THE AGAMEMNON OF AESCHYLUS. 281 


The rare dovevow—it is not in our lexicons though φόνευμα 
is—was mistaken for the familiar φονεῦσιν, and the adjacent 
accusatives were thus attracted into the dative, a change ren- 
_ dered the easier by the likeness of v tow. The corruption of 
νέοις into ἐμοῖς is an example of the error which gave us ὄκνος 
for κενός in v. 996 and the other blunders cited in my note 
there: precisely the same alteration is found in Eur. Med. 1388 
σὺ δ᾽, ὥσπερ εἰκὸς, κατθανεῖ κακὸς κακῶς πικρὰς τελευτὰς 
τῶν νέων γάμων ἰδών, where νέων is Weil’s necessary correction 
οἵ ἐμῶν. Of these changes, ἐχθροὺς... τοὺς ἐμούς belongs to 
Pearson, φόνευσιν to Bothe, νέοις only to me. 


1456—1458. 
ἰὼ παρανόμους Ἑλένα, 
μία Tas πολλᾶς, τὰς πάνυ πολλὲς 
\ λέ ’ « \ T aS 
ψυχὰς ολέσασ ὑπο Tpo:g. 


The metre of v. 1456 is usually mended by iterating fo 
with Blomfield and writing zapavovs with Hermann. This 
conjecture is so generally accepted that παράνους is received 
into the lexicons solely on the strength of it. Yet I neither 
understand how zrapavous produced the portentous παρανόμους, 
nor do I find much point or even much sense in the epithet. 
I am willing to take Hermann’s word that Helen was crazy; 
but this was not the place for saying so: her distraction of mind 
is one thing, her destruction of life another. Let us try to find 
something a trifle more appropriate. Everyone remembers the 
play on the name ‘EXév7y in v. 693, ἑλέναυς, EXavdpos, ἑλέπτολις. 
Now there is another ἔτυμον of the name which Aeschylus 
could hardly overlook and which exactly suits the context here. 
If we write 


ἰὼ παρὰ πῦρ ὄνομ᾽ οὖσ᾽ “Ἑλένα, 
we shall write what the loss of πυρο after παρα would transform 
to παρανόμους. The facility of this loss is shown by the error 
παραφόροιο for πυροφόροιο in an epigram in the Medicean Life 
of Aeschylus. The construction παρά c. acc. is the technical” 
phrase by which grammarians indicate the derivation of one 
word from another. The derivation of ‘EXévy from éXavn a 


282 THE JOURNAL OF PHILOLOGY. 


firebrand must, as I said, have been specially tempting and is 
here specially appropriate with reference to oXécaca: fire, as 
our newspaper writers are aware, is the devouring element. 
I think I find the same etymology in Euripides. In Tro. 891 
sqq. Hecuba is warning Menelaus against the charm of Helen: 
ὁρῶν δὲ τήνδε φεῦγε, μή σ᾽ ἕλῃ πόθῳ" αἱρεῖ yap ἀνδρῶν 
ὄὀμματ᾽, ἐξαιρεῖ πόλεις---80 far the ἔτυμον is ἑλεῖν ; but then 
she goes οῃ--πέμπρησι δ᾽ οἴκους : surely that is a glance at 
ἑλάνη. 


1476—1485., 


KA. viv δ᾽ ὥρθωσας στόματος γνώμην 
τὸν τριπάχυντον 
δαίμονα γέννης τῆσδε κικλήσκων. 
ἐκ τοῦ γὰρ ἔρως αἱματολοιχὸς 
νείρει τρέφεται, πρὶν καταλῆξαι 1480 
τὸ παλαιὸν ἄχος, νέος ἰχώρ. 
ΧΟ. ἡ μέγαν οἴκοις τοῖσδε 
δαίμονα καὶ βαρύμηνιν αἰνεῖς, 
φεῦ φεῦ, κακὸν αἶνον ἀτη- 
ρᾶς τύχας ἀκορέστου. 1485 


‘Before the old woe ceases, the new blood flows’ is Mr 
Sidgwick’s rendering of vv. 1480-1; and the verb flows is in- 
disputably necessary to the sense but indisputably absent from 
the Greek. Therefore, and because νεέρει, when altered into 
veipa, is quite superfluous, it seems to me that we should trans- 
pose that corrupt word; should read ἐκ τοῦ γὰρ ἔρως αἷμα- 
τολοιχὸς | τρέφεται, and endeavour to get the verb flows from 
velper, πρὶν καταλῆξαι | TO παλαιὸν ἄχος, νέος ἰχώρ. 

᾿ εὐ 

I think νείρει is simply ee ῥεῖ is the common gloss on the 
rarer vd. Hesych. vae ῥέει ; vaovor ῥέουσι; νῶντα᾽ ῥέοντα ; 
ναέτωρ᾽ ῥέων; ναρᾶς" ῥευστικῆς ; νᾶμα ῥεῦμα; vavas’ Tas ῥυ- 
Tas; νασμούς" ῥεύσεις ; νασμῶν" ῥευμάτων. As the copula is 
desirable and would readily perish in the sequence -tau καὶ νᾶι, 
I suppose Aeschylus to have written 


THE AGAMEMNON OF AESCHYLUS., 283 


ἐκ τοῦ yap ἔρως αἱματολουχὸς 
\ a \ A 
τρέφεται, Kal va, πρὶν καταλῆξαι 
’ 
τὸ παλαιὸν ἄχος, νέος ἰχώρ. 


The antistrophic verses answering vv. 1482—5 are these, 
vv. 1506—9, 
ὡς μὲν ἀναίτιος εἶ 
τοῦδε φόνου τίς 6 μαρτυρήσων ; 
πῶ πῶ; πατρόθεν δὲ συλλή- 
πτωρ γένοιτ᾽ ἂν ἀλάστωρ. 


To reconcile the metres of 1506 and 1482 is a problem which 
has caused much torment: the many conjectures recorded by 
Wecklein are all violent or ineffectual. But the first step 
towards emendation has I think been taken by Schuetz, who 
reads in the antistrophe ὡς μὲν ἀναίτιος εἶ σύ : the pronoun, if 
not absolutely necessary in poetry as-it would be in prose, is at 
any rate an improvement; and the metre now approximates to 
that of the strophe. The likeness of εἰ to cy makes the loss of 
the latter easy to understand: the same loss has happened in 
Supp. 950 where everyone now accepts Bothe’s εἴσει σύ τ᾽ 
αὐτὸς for εἰσθύγαυτος : that means, cy was absorbed by εἰ which 
afterwards became 61. The strophe I propose to amend thus: 


5 / > / a 
ἢ μέγαν εἰκόσι ταῖσδε 
δαίμονα καὶ βαρύμηνιν αἰνεῖς 


verily a great and vengeful demon is he of whom thou speakest 
in these parables, that is, in the metaphorical language of vv. 
1479—81 : εἰκών a metaphor occurs in Aristophanes and Plato. 
Virtually εἰκόσι and οἴκοις differ only in the order of their two 
final letters: when the inversion (see on v. 133) had taken 
place, the good scribe justly proud of knowing the gender of 
οἶκος completed the corruption by writing τοῖσδε. 


1531. 
τείσας ἅπερ ἦρξεν. 


μεταβολὴ πάντων γλυκύ : let us play the conservative for 
once. Wecklein has recalled attention to Spanheim’s very 
attractive conjecture ép&ev, certainly a more just and pointed 


284 THE JOURNAL OF PHILOLOGY. 


opposition to τείσας. But I think it well to sound the warning 
that ἦρξεν would seem to have been read here by Euripides. 
See Eur. fr. 825 Dind. τιμωρίαν ἔτεισεν ὧν ἦρξεν κακῶν and 
H, F.1169 τένων δ᾽ ἀμοιβὰς dv ὑπῆρξεν ᾿Ἡρακλῆς. 


1590—1597. 


ξένια δὲ τοῦδε δύσθεος πατὴρ 1ὅ90 
᾿Ατρεύς, προθύμως μᾶλλον ἢ φίλως, πατρὶ 
τὠμῷ, κρεουργὸν ἦμαρ εὐθύμως ἄγειν 
δοκῶν, παρέσχε δαῖτα παιδείων κρεῶν. 
τὰ μὲν ποδήρη καὶ χερῶν ἄκρους κτένας 
ἔθρυπτ᾽ ἄνωθεν ἀνδρακὰς καθήμενος 1595 
ἄσημ᾽ ὁ δ᾽ αὐτῶν αὐτίκ᾽ ἀγνοίᾳ λαβὼν 
ἔσθει βορὰν ἄσωτον ὡς ὁρᾷς γένει. 

In ν. 1591 προθύμως is condemned not merely by its own 
absurdity but by the presence of εὐθύμως in the next verse: 
I should write without hesitation προσηνῶς. This seems to 
give just the sense required, and of course the difference be- 
tween σὴν and θυμ is palaeographically nothing. 

Vv. 1594—7 I have written down just as they are in the 
MSS, with one exception: I have of course accepted Dindorf’s 
don’ ὁ δ᾽ for ἄσημα δ᾽. Unless this change be made, the 
subject of ἔσθει will of necessity be Atreus: when Mr Paley 
writes ‘and Thyestes’ and when Dr Kennedy writes ‘so my 
sire, they are translating the ὁ δέ which they exclude from 
their texts, not the MS reading which they print. Mr Paley, 
Dr Kennedy, Mr Sidgwick, Mr Margoliouth, retain the solecism 
ἀνδρακὰς καθήμενος uiritim sedens. Casaubon’s ἀνδρακὰς καθη- 
μένοις and Wecklein’s avdpaxas δατούμενος are Greek : they are 
most obscure, and so far as they do yield a meaning that 
meaning would seem to be that Atreus gave the murdered 
children for meat not to Thyestes only but to the rest of the 
company as well; but still they are Greek. But in no tongue 
save the tongue of Soli can one person καθῆσθαι ἀνδρακάς, any 
more than he can form himself in square to receive cavalry. 
Because Suidas, quite correctly, renders ἀνδρακάς by χωρίς, we 
should not therefore jump to the conclusion that whenever we 
mean χωρίς we can say ἀνδρακάς. 


THE AGAMEMNON OF AESCHYLUS. 285 


Whatever else in this passage may be sound, I think ἀν- 
δρακάς mnust be corrupt. That word necessarily imports an 
allusion to the other guests at the banquet. Now if that 
allusion were introduced at all, which was not needful, it should 
at least have been made more intelligible. The mention of the 
guests starts our curiosity to know how Atreus contrived to set 
the children’s flesh before Thyestes alone among their number; 
and our curiosity is not gratified. What I propose then is this: 


τὰ μὲν ποδήρη Kal χερῶν ἄκρους κτένας 

ἔθρυπτ᾽ ἄνωθεν ἄνθρακος καθημμένου 

aon . 
Once let @ become ὃ, as in v. 988 ψύδη for ψύθη, nothing but 
ἀνδρακάς could ensue; then καθημμένου bereft of its substantive 
must change its inflexion: the remaining error yw for py recalls 
the converse blunder in v. 1418 λημμάτων for ἀημάτων. θρύπ- 
τειν 1s a technical term in cookery, see lexx. sub voce. ἔνθρυπτα 
and θρύμματα : it means properly to mince a solid, usually 
bread, into a liquid, thus forming a pulp. It would appear 
from v. 1082 ὀπτάς τε σάρκας πρὸς πατρὸς βεβρωμένας that 
such parts of the bodies as were not plainly recognisable for 
human were roast: the tell-tale hands and feet were, I pre- 
sume, boiled in a λέβης. The lines thus emended seem to have 
been imitated by Euripides in his account of another ἀνθρω- 
πομάγειρος. In Cycl. 244 sqq., σφαγέντες αὐτίκα | πλήσουσι 
νηδὺν τὴν ἐμὴν ἀπ᾽ ἄνθρακος θερμὴν ἑλόντος Sait ἄτερ Kpea- 
νόμου, | τὰ δ᾽ ἐκ λέβητος ἑφθὰ καὶ τετηκότα, You have ἄνθρακος 
as here, αὐτίκα... ἑλόντος to recall Aeschylus’ αὐτίκα... λαβών, 
and τετηκότα to recall ἔθρυπτ᾽...ἄσημα. It is true that ἄνθραξ 
is there in opposition to λέβης, as avOpaxia is in opposition 
both to boiling and to roasting in v. 358 ἑφθὰ καὶ ὀπτὰ καὶ 
ἀνθρακιᾶς ἄπο (broiled) χναύειν... μέλη ξένων. But ἄνθραξ is 
used of boiling in vv. 373 sq. ἑφθά τε δαινύμενος μυσαροῖσιν 
ὀδοῦσιν | ἀνθρώπων θέρμ᾽ ἀπ᾿ ἀνθράκων κρέα. The compound 
καθάπτειν does not seem to occur elsewhere in the sense I 
give it here; but that is nothing: drew kindle is warrant 
for καθάπτειν kindle thoroughly. The tragedians prefix with 
great freedom the intensive ἐξ and κατά: thus xatavy@ for 


286 THE JOURNAL OF PHILOLOGY. 


avy@ is used once by Aeschylus, never by anyone else ; and here 
he prefers to say ἄνθρακος καθημμένου while Thucydides tv 100 
is content with dvOpaxas ἡμμένους. ἄνωθεν c. gen. is found 
fifteen lines above, in a passage to which I will devote a word 
for its own sake, vv. 1578 sq. 


/ a bu a a / 
φαίην av ἤδη νῦν βροτῶν τιμαόρους 
θεοὺς ἄνωθεν γῆς ἐποπτεύειν ἄχη. 


Mr Paley and many others construe γῆς ayn: Auratus’ instinct 
told him that this phrase was unsuitable, and he therefore con- 
jectured dyn, which many accept. But anyone who will turn 
to Eur. fr. 959 Dind. ἔστι, κεἴ τις ἐγγεχλᾷ λόγῳ, | Ζεὺς, καὶ θεοὶ 
βρότεια λεύσσοντες πάθη will see that Euripides found ἄχη in 
his Aeschylus and construed it with βροτῶν. γῆς therefore 
depends on ἄνωθεν. 

In Wecklein’s list of conjectures I find these: ἔκρυπτ᾽ ave- 
θεν ἄνθρακας καθειμένος Tyrwhitt, ἔθρυπτ᾽ ἄνω θεὶς ἄνθρακας 
καθημμένους Abresch, ἄνθρακος ‘anonymus’ with what context 
I know not. I cannot extract much sense from any of these 
readings; but it is right that I should mention conjectures 
which verbally resemble mine so nearly. 


1654—1665. 

KA. μηδαμῶς, ὦ φίλατ᾽ ἀνδρῶν, ἄλλα δράσωμεν Kaka. 

ἀλλὰ καὶ τάδ᾽ ἐξαμῆσαι πολλὰ, δύστηνον θέρος. 1655 

πημονῆς ἅλις δ᾽ ὑπάρχει μηδὲν αἱματώμεθα. 

στείχετε δ᾽ οἱ γέροντες πρὸς δόμους πεπμωμένους τούσδε, 
ἔρξαντα 
ἔρξαντες 
εἰ δέ τοι μόχθων γένοιτο τῶνδ᾽ ἅλις γ᾽ ἐχοίμεθ᾽ ἂν, 
δαίμονος χηλῇ βαρείᾳ δυστυχῶς πεπληγμένοι. 1660 


πρὶν παθεῖν] \ καιρόν: χρὴν τάδ᾽ ws ἐπράξαμεν. 
) 


ὧδ᾽ ἔχει λόγος γυναικὸς, εἴ τις ἀξιοῖ μαθεῖν. 

ΑἹ. ἀλλὰ τούσδ᾽ ἐμοὶ ματαίαν γλῶσσαν ὧδ᾽ ἀπανθίσαι 
KkaxBareiv ἔπη τοιαῦτα δαίμονος πειρωμένους, 
σώφρονος γνώμης δ᾽ ἁμαρτῆτον κρατοῦντα. 

XO. οὐκ ἂν ᾿Αργείων τόδ᾽ εἴη, φῶτα προσσαίνειν κακόν. 1665 


In considering the well-known difficulties of vv. 1657—9 
I will begin with the hypermetrical τούσδε of 1657. To discard 


THE AGAMEMNON OF AESCHYLUS. 287 


this with Auratus does not explain how it got into the text; to 
insert it in the next verse with Weil demands the extrusion of 
some other word. Now it is to be observed that just as we have 
two syllables too many at the end of v. 1657, so have we three 
too few at the end of v. 1664. Not only this, but I notice that 
while the verse to which τούσδε is tacked ends with πεπρω- 
μένους, v. 1663, the next-door neighbour to the defective verse, 
ends with a word of almost identical appearance, πειρωμένους. 
I guess then that τούσδε is the missing end of v. 1664, and that 
either the end or the beginning has been misplaced through the 
homoeoteleuton of vv. 1657 and 1663. And indeed Hermann 
and others have already seen that v. 1664, supplement it how 
you will, is misplaced. Take vv. 1662—4 in Mr Paley’s trans- 
lation: ‘ But to think that these men should thus gather the 
flowers of their vain tongue against me, and have uttered such 
words, challenging their fate, and so fail in sound judgment’ 
(ἁμαρτεῖν Casaubon) etc. Was ever such an impotent sequel as 
these words form to the two foregoing verses? Two enemies 
are in the heat of an envenomed altercation, insults and 
menaces flying to and fro: a friend exhorts them to be calm ; 
and one of them bursts out ‘But that this man should fail in 
sound judgment’! No: it is not thus that mankind talk. 
Take this v. 1664 away, and 1665 follows appropriately on 
1663: now let us see what can be done in the neighbourhood 
of τούσδε. Of the two Mss which are here our authorities the 
Florentine alone gives δ᾽ ἁμαρτῆτον: the Venetian omits it, 
leaving ἃ blank space. This indicates that in the common 
parent of both the mss these letters were barely decipherable ; 
so it will not be rash to alter one letter more than was altered 
by Casaubon. I should place v. 1664 between 1656 and 1657 ; 
and in the corrupt tradition 


σώφρονος γνώμης δ᾽ ἁμαρτῆτον κρατοῦντα τούσδε 
I suggest that ἁμαρτῆ stands for ὁμαρτεῖ and τούσδε for τοῖς λε: 
| σώφρονος γνώμης δ᾽ ὁμαρτεῖν τὸν κρατοῦντα τοῖς λεῴς. 
With σώφρ. γν. supply ἐστίν: to ὁμαρτεῖν I give the sense which 
προσχωρεῖν has in Eur. Med. 222 χρὴ δὲ ξένον μὲν κάρτα 
προσχωρεῖν πόλει. In the Antigone of Euripides, where the 


288 THE JOURNAL OF PHILOLOGY. 


position of the τύραννος was canvassed, occurred the line, 
Jr. 172 Dind., δεῖ τοῖσι πολλοῖς τὸν τύραννον avdavew, which 
looks to me like a paraphrase of the verse I give to Aeschylus. 
I am sorry to deal in this guesswork, but it was necessary to 
handle the matter in order to justify my rejection of τούσδε 
from v. 1657, to the correction of which I now proceed. Madvig 
and others have seen that δόμους πεπρωμένους destined home 
has no meaning which suits the context: “ πεπρωμένοι cur 
domus appellentur causa iure quaeritur nec reperitur’: it could 
signify nothing but “Avdou δόμους. True, you can invest it 
with some sort of sense by accepting Franz’s conjecture στεῖχε 
καὶ od xot γέροντες ; but how a scribe could mistake καισυχ for 
red, and how without gross superstition we can believe that 
scribes who made mistakes like this have preserved uncor- 
rupted a single word that Aeschylus wrote, I do not know. 
Science here furnishes a correction so obvious, and so appro- 
priate to the lips of Clytaemestra in her part of peace- 
maker, that far from being surprised to find it anticipated by 
Ahrens, I am surprised not to find it anticipated by Auratus. 
The scribe who corrupted στείχετ᾽, αἰδοῖοι γέροντες into the 
present reading of the Mss merely, for the hundredth time, sub- 
stituted ¢ for as, and wrote οὐ once when he should have 
written it twice. For the rest of the line the most plausible 
conjecture by far is Madvig’s: πρὸς δόμους, πεπρωμένοις, | πρὶν 
παθεῖν, εἴξαντες. This, though I do not like parting with the 
familiar juxtaposition of παθεῖν and ἔρξαι, is excellent sense so 
far as it goes; but now what are we to make of the sequel 
καιρὸν χρῆν τάδ᾽ ὡς ἐπράξαμεν! Of course καιρόν must be 
altered into an infinitive, but what infinitive ἢ Heath’s αἰνεῖν, 
which Madvig would like, is very wide of the Mss: Hermann’s 
ἀρκεῖν, which Madvig accepts, is near to the Mss but very wide 
of a satisfactory meaning. I have seen no suitable and probable 
word suggested, and can suggest none myself. It seems to me 
that each of the vv. 1657 and 1658 is a complete sentence. 
The former is this : 


στείχετ᾽, αἰδοῖοι γέροντες, πρὸς δρόμους πεπρωμένους. 


See Eur. Med. 1245 ἕρπε πρὸς βαλβῖδα λυπηρὰν βίου. Cly- 


THE AGAMEMNON OF AESCHYLUS. 289 


taemestra counsels the elders to betake themselves to the new 
course of life to which it has pleased God to call them, sub- 
mission to the rule of Aegisthus and herself. The wrongful 
omission or insertion of p after a mute is very common; com- 
monest after 8 and τ, but common after all mutes: for δρόμος 
and δόμος confused see fr. 374 Dind., where the Mss of 
Theocritus’ scholia vary between πρόδρομος and πρόδομος, and 
Eur. Andr. 1099, where both δρόμοις and δόμοις are found. 

In v. 1658 I suppose πρίν to be the adverb (= πρότερον), 
not the conjunction. When ép£avres is found in the Florentine 
MS and ép£avra in the Venetian, I think it is the most rational 
inference that ἔρξαντας was in the Medicean: καιρόν must in 
any case be altered for metre’s sake, as the singular ἔρξαντα 
is indefensible. But already we have restored, sense to the 
verse, and metre may be restored thus: 

- καιρόν 
πρὶν παθεῖν ἔρξαντας ὥραν χρῆν, τάδ᾽ ὡς ἐπράξαμεν. 
Hesych. ὥρα" καιρός ; ὧραι" καιροί; ὥρῃ" τῷ καιρῷ; καθ᾽ 
ὥραν" κατὰ καιρόν. The adverbial use of τὴν ὥρην = tusto tem- 
pore occurs in Herod. Π. 2: καιρόν itself is used in that sense in 
Soph. Az. 34, 13816 and Eur. Hel. 479, but καιρόν is perhaps 
thus employed only with ἥκω or verbs of that meaning. Of 
course I cannot promise that ὥραν was the very word on which 
καιρόν is a gloss; but that καιρόν is a gloss, luckily detected by 
metre, I have no doubt. I render you should have exchanged 
blows earlier, in season, when we did this deed. Strictly I sup- 
pose πρίν belongs to παθεῖν, ὥραν to ἔρξαντας. In the same 
meaning which I here give to παθεῖν ἔρξαντας (smite and be 
smitten) Euripides Phoen. 480 uses κακόν te δρᾶσαι καὶ παθεῖν. 

In v. 1659 almost all editors now accept Martin’s δεχοίμεθ᾽ 
ἄν. But manifestly this of itself is not enough to amend the 
line. To say εἰ μόχθων γένοιτο ἅλις the moment after you 
have said πημονῆς ἅλις ὑπάρχει is so obviously inconsistent 
that there is a general consent against the genuineness of ἅλες. 
Donaldson proposes and Paley approves ἄκος, which makes 
good sense. But the verse is to be corrected with much less 
change than ἄκος, δεχοίμεθ᾽ av. The reading which I propose 
is really almost identical with that of the Mss: 


Journal of Philology. τ... xvt. 19 


290 THE JOURNAL OF PHILOLOGY. 
εἰ δέ τοι μόχθων γένοιτο τῶνδ᾽ arn, στεγοίμεθ᾽ ἄν, 


n for ἐ, τ for γ, y for y. Should there be any way to ward off 
these ills (civil war), ward them off we should. We moderns 
know only ἄλη error; but the Greeks knew also ἄλη tritura 
akin to ἀλέω tero and ἄλη defensio akin to ἀλέω defendo. The 
existence of the verbs might support this surmise, even 
were there no other proof; but it happens that both these lost 
substantives occur in the Agamemnon, though obscured in one 
place by the corruption of the copyists and in the other by the 
mistranslation of the commentators. In v. 204 the winds that 
blew at Aulis are called βροτῶν ἄλαι, which is supposed to 
mean causes of wandering to men: a less happy name for winds 
which prevented the Greeks from sailing and kept their fleet on 
the shore it would need some ingenuity to devise. The true 
rendering is suggested by v. 207 ἄνθος κατέξαινον ᾿Αργείων 
τρίβῳ (so I should arrange the words, making no change in 
the antistrophe but ῥείθροις for ῥεέθροις): βροτῶν ἄλαι are 
grindings or tribulations of men, winds that wear men away 
ἀπλοίᾳ κεναγγεῖ. In ν. 1659 ἄλη is akin to ἀλέω defendo, a 
verb preserved, I think, only in Hesych. ἄλεε: φύλασσε: I 
imagine that ἀλέομαι wito is originally part of the same verb: 
compare too ἀλέη, ἀλεωρή, ἀλεύω, adé~w. The verse means 
then εἰ μόχθων γένοιτο τῶνδε φυλακὴ, φυλασσοίμεθ᾽ ἄν, but 
Aeschylus chooses poet-like to vary his words. 
Here therefore are the verses as I would write them: 


στείχετ᾽, αἰδοῖοι γέροντες, πρὸς δρόμους πεπρωμένους. 
\ a », δ “ ‘~ ¢ 3 , 
πρὶν παθεῖν, ἔρξαντας <dpav>, χρῆν, Tad ὡς ἐπράξαμεν. 
> / , / al Ν / » ἃ 
εἰ δέ τοι μόχθων γένοιτο τῶνδ᾽ ἄλη, στεγοίμεθ᾽ ἂν, 
/ nr 7 an , 
δαίμονος γηλῇ βαρείᾳ δυστυχῶς TETANY MEVOL. 


A. E. HOUSMAN. 


NOTE ON EMENDATIONS OF PROPERTIUS. 


[I said on p. 16 of this vol. of the Journal of Philology that 
I feared some of the corrections there proposed had been fore- 
stalled by others. I am to blame that this is true of a larger 
number than one could wish, mainly through trusting to my 
memory of Burmann’s notes instead of giving them a fresh 
perusal. I now make restitution: I xx 24 sacram Rutger- 
slus, II ix 7 wiswram (uisurwm is an error) Paley, xxviii 62 
punctuated so by Postgate, xxxiv 12 posses tun and 40 zrato 
Heinsius, ΠῚ viii 12 haec Liuineius, xvi 21 cursus Markland, 
xvil 24 carpta Heinsius, xviii 21 manet Palmer, xxii 15 siqua 
et Heinsius, Iv 11 12 credis id Postgate, vii 23 ewnti: Reland. 
The three living scholars will, I hope, accept my apologies. 

Further, the following proposals have more or less in 
common with my own, and ought to be mentioned: I iii 37 
nempe δὲ Burmann, II viii 30 Teucros Passeratius, ix 12 apposito 
...Svmoente Guietus, x 2 campum Maeonio and xxi 12 excepta 
Aesonia est Heinsius, I vi 28 exsuccis unguibus Burmann. 

Let me here subjoin a few conjectures accidentally omitted 
from the paper of which I speak: 1 i 53 an in me for siue, 
Ii xxxil 9 quid wubet for cum widet, ΠῚ i 32 terra for Troia, 
ll xix 17 more parentis for tempore matris, IV iv 83 ascensum 
monstrat dubio for mons erat ascensu dubius, ΤΥ viii 13 fuerunt 
for fuerint. 


A. E. HOUSMAN. 


19—2 


ON THE AETNA OF LUCILIUS. 


DuRING the last year, in which the Aetna, as one of the 
poems included in the Appendix Vergiliana, has occupied a 
great deal of my attention, I have made a renewed and most 
careful study of Munro’s edition. It is not too much to say 
that this is, in the eyes of scholars trained up to the present 
level of philological criticism, the only edition which is self- 
sufficing and in any sense complete. For the text of the Aetna 
depends so largely on the one reliable and throughout uninter- 
polated ms, Cambridge Kk v 34, that the publication by 
Munro of its readings for the first time in 1867 marks the real 
moment at which the poem became, in the true sense of the 
word, intelligible. Up to that time the Mss known were com- 
paratively late in date and corrupted proportionately; the 
Cambridge codex seems to be as early as cent. x. Only one 
other can claim anything like the same antiquity, the Stabu- 
lensian fragment (S) at Paris. -Munro did not know this; but 
Bahrens has given its readings in his edition (PLM. IL. 
p. 88 sqq.). These agree closely with the Cambridge Ms 
(Munro’s a, Bahrens’ C), so closely as rarely to throw much 
additional light on the disputed or obscure passages. Whether 
any third codex that can rank with these two is lurking in the 
libraries of Europe, I cannot say: in Rome where I examined 
five mss of the Aetna, to which I may now add one at Naples, 
all written in cent. xv, nothing of the kind greeted my re- 
searches ; indeed only one of all the six was sufficiently free 
from interpolation to deserve collating, Vat. 3272". 

1 Fortune has not befriended us here: off his work after copying the first 
for the excellent and absolutely original six lines. It is however noteworthy 
ms of the Culex in the Corsini’ that in v. 1 this ms gives ruptisque 


Library at Rome, was to have con- caui not ruptique cauis. 
tained the Aetna; but the scribe broke 


ON THE AETNA OF LUCILIUS. 293 


Of the Gyraldinus, a suppesed codex of very early date, 
which has long been lost, but of whose readings on vv. 138—287 
we still possess the tradition (see Munro p. 31 sqq., Bahrens 
pp. 6—8, cf. Wagler de Aetna poemate quaestiones criticae. 
Berlin 1884), I purposely omit saying much: for so many of its 
variants are irreconcileable with the text of C'S, and are so very 
like the ingenious conjectures of some modern scholar, as to 
throw suspicion on the integrity of the whole of them. At any 
rate it seems safer always to start from CS as the basis for our 
reconstruction of disputed passages; for there is nothing to 
prove any such early depravation of the text of these two Mss as 
must be admitted if many of the reported variants of the 
Gyraldinus are right. 


61 iam patri deatera Pallas 
Et Mars saeuus erat, iam cetera turba deorum 
Stant utrimque tdeust ualidos tum Luppiter rgnis 
Increpat et wcto proturbat fulmine montes. 


So C; the Stabulensian fragm. has de//, There can be no 
meaning in deus, but it is very doubtful what it represents. 
Haupt’s secus is clever, but seems to me a little prosaic. 
Bahrens’ tuens gives an idea of divine unconcern alien to the 
feeling of, the passage; possibly uerens, of which the first 
syllable may have been lost after -gue, may be the word. For 
uicto I confess I prefer the old Italian correction acto to wictor, 
which Munro accepts from two MSS γ and e. 


66 atque umpius hostis 
Praeceps cum castris agitur, materque vacentis 
Impellens uictos. tum pax est reddita mundo 
Tum liber cessat went per sidera caelum 
Defensique decus mundi nunc redditur astris. 


I differ from Munro in his view of these vv. (1) As to the 
meaning of Impellens, ‘rallying’ M. Surely the sense cannot 
be this; rather Earth urges on to flight her prostrate children, 
the Giants, 1.6. urges them to rise and take to headlong flight. 
(2) v. 69 Munro prints 


Tum liber cessat: uenit per sidera: caelum, 


294 THE JOURNAL OF PHILOLOGY. 


translating ‘then peace was restored to the sky, which then 
was free and at rest; this peace came by the help of the stars: 
heaven and the honour of the sky’s defence are now assigned 
to the stars’. If caelwm of CS is correct, it is almost impossible, 
I believe, to disconnect it from the preceding words, ‘ heaven 
is seen appearing through the stars’, 1.6. as the giants clear 
off from the face of heaven, the stars reappear and the inter- 
spaces of sky become discernible. Yet it is also possible that 
cessat represents cessata a participle which occurs in 384, 


Si cessata diu referunt spectacula uenti, 
and the passage may then run 


Tum Inber cessata uenit per sidera: caelum 
Defensique decus mundi nunc redditur astris. 


Bacchus bore a conspicuous part in the conflict with the Giants, 
as Horace tells us C. 11. 19. 21—24, and his progress amid the 
now resting stars would be a natural way of expressing that this 
strife was ended. 


74 Haec est mendosae uulgata licentia famae 
Vatibus ingenium est, hinc audit nobile carmen. 
Plurima pars scenae rerum est fallacia, wates 
Sub terris nigros wderunt carmine manes. 


For scenae Vat. gives scenica. May not the right reading be 
sed enim? By this we gain the retention of rerwm (uerum 
Munro), and an excellent sense; sed enim refers to mendosae. 
For uiderunt carmine, which is very odd Latin, I suspect the 
poet wrote finverunt, unless indeed Bahrens’ luserunt is more 
probable. 


80 Hi Tityon poena strauere in iugera foedum, 
Sollicitant alle te circum, Tantale, poena 
Sollicitantque siti. 


Many points are open to question here. (1) poena in 80 must 
be wrong, as zwgera could hardly stand by itself thus barely; 
yet Haupt’s strauere nouena in iugera is to my mind less likely 
than what has found its way into many Mss, including Vat. 
3272, septem strauere in tugera: for this septem can hardly be 


ON THE AETNA OF LUCILIUS. 295 


a 15th century correction, since most schoolboys were taught 
then, as all are now, that Tityos covered nine plethra. I think 
therefore that septem may be the right reading. (2) In v. 81 
I would not alter circum; and poena is just intelligible, if we 
suppose it to be explained by sitz; but Bahrens’ cena is very 
plausible, preferable I think to pomo, pomis or Munro’s poma. 
Is it possible that 2/7 is an error for mili, ‘an unreal repast’ 
which instantly eludes his grasp? Cf. Varro L. L. ix. 77 nihil 


argumentum ‘a nugatory argument’. 


83 Minos tuaque Aeace in umbris 
Iura canunt idemque rotant Iaionis orbem. 
Quicquid et internis falst sibi conscia terrent. 
Nec tu terra satis: speculantur numina diwum. 


So CS: but in 85 for sibi conscia the Helmstadt Ms gives 
consortia. Here again I lean to the reading of the inferior 


or 
authority : conscia would explain the corruption. But terrent 
must be wrong: I believe it to be a corruption of adhaerent, a 
favorite word with Seneca, e.g. Ep. 65. 18. 
I would write the v. then | 


Quicquid et infernist, falsi consortia adhaerent 


‘whatever belongs to the world below, some association of false- 
hood is inseparable from it’. 


96 Non totum et solido est. 
So CS: obviously for 
Non totum ex solido est. 


This is, I think, certain: et for ex is one of the most common of 
all corruptions. 


98 utque animante 
Per tota errantes percurrunt corpora uenae 
Ad witam, sanguis omnis qua commeat idem. 


So Munro, perhaps rightly; except that for idem I would write 
eidem, 1.6. animanti, ‘by which passage all the blood passes to 
and fro in the body of one and the same creature’. 


296 THE JOURNAL OF PHILOLOGY. 


102 Scilicet aut olim diwiso corpore mundi 
In maria et terras et sidera, sors data caelo 
Prima, secuta maris. 


Munro regards this awt as answered by sive in 110, and so 
Jacob. It is I think an error for ante; a similar case is to be 
found in Catalept. 1. 5 Venerit aut tibi for which I would read 
Venerit ante tibi, ‘suppose Delia has at some time in the past 
come to you: what good does the announcement of that past 
arrival do me now ?’ 


105 et qualis aceruus 
Exilit inparibus iactis ex tempore saxis 
Vt crebro wntrorsus spatio uacuata charibdis 
Pendeat in sese. 


Exilit is here the opposite of residit, just as in Sen. Epist. 
66. 11 (virtutes satorum animaliumque) exiliunt residuntque 
‘spring, shoot up’; the idea in either case is the suddenness 
or rapidity with which the stone heap is formed. Vat. 3272 
has a remarkable variant for caribdis (which may point to 
a less outré word), carambos: possibly coronis ‘the apex’, ‘the 
last stone that completes the pile’, τὸ τελευταῖον τῆς οἰκοδομῆς 
ἐπίθεμα (Hesych.). 

116 non est hic causa dolendi 


Dum stet opus causae. 
So CS: docendi and causas Munro. I doubt both changes, 


‘There is no cause for complaining (of our ignorance), provided 
only the effect of the (unknown) cause is permanent’. 


120 Nam ille ex tenui uocemque agat apta necesse est 
Cum fluwio errantes arcessant undique uenas 
Et trahat ex pleno quod fortem contrahat amnem. 


So C, and so S except that it has cum fluwia and fontem. 
The last two vv. Munro writes thus 


Confluuia errantes arcessant undique uenas, 
Vt trahat ex pleno quod fortem contrahat amnem. 


Rightly, I imagine, except that contrahat is slightly harsh after 


ON THE AETNA OF LUCILIUS. 297 


trahat in the same v. Béahrens gives comparat. Possibly con- 
rogat. The first I propose to emend as follows: 


Non ille ex tenui uolens ueget: arta necesse est. 


Tile se. torrens, ‘be sure, that torrent does not change from 
a puny stream into boisterous vigour: there must be pent-up 
confluents that collect from every side their wandering ducts, in 
order that the torrent may draw from a full source the supply 
of a vigorous stream’. Non for nam is already in Vat. 3272 as 
well as other mss. If this restitution is right, the lost v. after 
119 must have been something of this kind, 


Inualidus solet atque alio se erumpere fortem. 
Cf. Sen. N. Q. vi. 8. 


140 Cernis et in siluis spatioque cubilia retro 
Antraque demissa pedibus fodisse latebris. 


So CS: the Gyraldinus is said to have had spatiosa and 
demersas penitus f. latebras. I would call attention to the fact, 
here very palpable, that these readings are exactly such as 
a modern emender of the passage might propose: and would 
suggest that the really lost original was not this, but perhaps 


Cernis et in siluis spatioque cubilia tecto 
Antraque demissas penitus fodisse latebras. 


Vat. 3272 has regtro, which might be a corruption of tecto. 
The reported reading of Gyr. is no doubt neater; but is it true ? 
Certainly there are many reasons for distrusting these reported 
variants elsewhere. It is, for instance, nearly incredible that 


161 Fallere sed nondum tibi lumine certaque retro 
(so CS) should have been corrupted from 
Falleris et nondum certo tibi lwmine res est 
as reported from Gyr. Many possibilities occur, e.g. 
Falleris et nondum tibi lumine certa liquet res, 


none perhaps sufficiently convincing to supplant the now 
generally received reading of Gyr., yet enough to increase the 
suspicion with which we approach this authority. Indeed, to 


298 THE JOURNAL OF PHILOLOGY. 


take the most crucial instance of all, one which like all others 
I have long been accustomed to consider convincing as to the 
substantial authenticity of at least part of these readings of 
Gyr., 227 


Ingenium sacrare caputque attollere caelo 
for 
Sacra peringentem capitique attollere caelum 


of CS and most Mss, I do not think it impossible that some 
other and quite different version came from the poet, e.g. 


Sacra patris ridere*, caputque attollere caelo 


‘to laugh at the rites of Father Jove, and lift our head to the 
sky’, 1.6. instead of worship and adoration, to bear a bold front 
- and look into Father Jove’s sky. But, though this inquiry is 
a not unprofitable one, and though I believe the view, that the 
reported variants of Gyr. are mostly conjectures of the 16th and 
17th centuries, to be more than tenable, it would take too much 
time to carry out this examination in detail in an article not 
directly devoted to such a purpose. 


144 Tu modo subtiles animo duce percipe curas 
Occultamque fidem manifestis abstrahe rebus. 


Munro says on this “abstrahe etc. must mean ‘draw from 
things seen belief in the unseen’.” Surely this is not neces- 
sary: abstrahe is not ‘draw from’ but ‘ withdraw’ or ‘abstract’ 
from the visible workings of nature the hidden principle 
which we are to accept as the law of her working. So occultas 


causas in 179. 


146—149 


Nam quo liberior, quoque est animosior ignis 
Semper in inclusis, nec uentis segnior ira est 
Sub terra penitusque mouent hoc plura necesse est 
Vinela magis soluant magis hoc obstantia pellant. 


1 ingentem, urgentem, rigentem are  ridente which would=ridere, much as 
the variants; the latter would be a  turbant=turbare in 168. 
corruption of rigente, and this of 


ON THE AETNA OF LUCILIUS. 299 


Munro reads hic plura, making the apodosis begin at necesse 
est. I have always found myself pausing at this, mainly 1 
think from the inequality of the two clauses. May not mouent 
be the seat of obscurity, and an error for dolent? (dolare). 


nec uentis segnior tra est 
Sub terra penitusque, dolent hoc plura necesse est, 
Vincla magis soluant, magis hoc obstantia pellant, 


‘and in such proportion as the winds are equally quick (as the 
fire) to shew their fury under ground and deep below, in such 
proportion they must needs scoop out more ground, must so 
much the more break the fastenings loose, so much the more 
remove what stands in their way’. Cf. Seneca N. Q. vi. 24 
(Motus est) subter et ab imo. 


162—4 
Namque illuc quod cumque uacat hiat impetus omnis 
Et sese introitu soluunt adituque patents 
Conuersae languent wires animosque remittunt. 


Munro here follows Gyr., which necessitates the hypothesis 
of a lost v., to say nothing of the awkwardness of the rhythm in 
the supposed restoration. It is safer, in my opinion, to keep to 
the outline of CS as written above. With very little alteration, 
we might read 


Namque illuc, quodcumque uacans hiat, impetus omits, 


‘for the whole force of their onset is towards any point where 
there is an open vacuum’. «luc qguodcwmque = ad id quodeum- 
gue: uacds would readily pass into wacdt, this into wacat. 


165 Quippe ubi contineat uentosa qua quaeque morantis- 
In wacuo desint. 


So C, continuat S; this must be not gui teneat (Gyr.), but 
quod teneat (Haupt); and so I see Bahrens prints. In the rest 
of the passage Gyr. seems to be right in defit for desint, and 
nearly right in wentos aquasque, if, as seems likely, Munro’s 
acuatque is the corrupted word. 


300 THE JOURNAL OF PHILOLOGY. 


180 Plurima namque patent ili miracula monti 
Hine uasti terrent aditus merguntque profundo 
Corrigit hinc artus penitus quos exigit ultra. 


I fancy that arcus, not artus, is what the poet wrote. ‘On 
another side it calls in the arches which it carries out internally 
to a point beyond’, i.e. elsewhere Aetna presents to the eye the 
appearance of arches terminating externally, after stretching to 
some distance inwards. Porrigit of Gyr. is of course a slight 
change, but I doubt its being right, and all Mss extant seem to 
agree in corrigit. Cf. 347 ualidoque absoluerit arcu. 


203 sqq. 
Ipse procul magnos miratur Iuppiter ignes 
Neue sepulta nour surgant in bella gigantes 
Neu Ditem regni pudeat, neu Tartara caelo 
Vertat ™m occulto tantum premit omnia dextra 
Congeries operit saxorum et putris harena. 


So C except that in 206 it has vertant, in 207 operis. The 
only thing wrong is deatra, for which Gyr. had, as reported, 
omniaque extra. If this was right, OMNIAQ. EXTRA may have 
been the intermediate step. Or must we read omnia ad 
extra? ad for at is common enough. At any rate at suggests, 
what is required, the opposition of the external appearance 
of Mount Etna to the far greater, but concealed, workings 
within. 


212 sqq. 
Hac causa expectata ruunt incendia montis. 
Spiritus inflatis nomen, languentibus aer. 
Num prope nequiquam par est wolentia ; semper 
215 Ingenium uelox illi motusque perennis. 
Verum opus auxilium est ut pellat corpora; nullus 
Impetus est wpsi, qua spiritus imperat audit. 
Hic princeps magnoque sub hoc duce militat agnis. 
So I would write this difficult passage, retaining the readings 


of Οἱ with the exception of Hac for Haec 212, montis for mortis, 
ib.; uiolentia for uolentia 214; corpora for corpore 216; Hic for 


ON THE AETNA OF LUCILIUS. 301 


Hinc, magnoque for magnosque 218. All these changes are ad- 
mitted by Munro, with whom however I differ in the constitution 
of the first two vv., especially 213. Munro like all the other 
critics that I have seen considers the text of this v. corrupt, and 
changes it to spiritus inflabit momen languentibus acre. To me, 
from the first moment I read the line to the present time, a 
conviction of its soundness has asserted and reasserted itself : 
the poet parenthetically introduces a piece of scientific nomen- 
clature. ‘This is the reason why the rush of fire in Etna never 
comes as a surprise’, namely, the working of the winds inside 
the mountain. ‘These winds when inflated are called spirit 
(Sen. N. Q. τι. 1. 3 cum metus terrae spiritu fiant, spiritus autem 
aer sit agitatus), when in subsidence, air. (Without their agency, 
fire alone can do nothing.) For it is almost of no effect that 
the two elements are equal in violence: true, fire has a natural 
velocity and continual motion, but then it needs some auxiliar 
to set bodies in motion: by itself it has no velocity; only where 
air bids, it follows obediently’. Yet though the passage may 
be so construed, there is an undeniable harshness in the absence 
of any word to express with distinctness what are the two forces 
of which par est urolentia. Munro from Gyr. substitutes fammae 
for semper: a far easier remedy would be to read igni for alli in 
215, which has this besides in its favour, that the recurrence of 
the same word at an interval of three lines (igni 215, ignis 218) 
is a marked feature of the poem. Seneca has a very similar 
passage N. Q. vi. 21 Nobis quoque placet hunc spiritum esse qui 
possit tanta conart, quo nihil est in rerum natura potentius, nihil 
acrius, sine quo nec illa quidem quae uehementissima sunt, ualent; 
ignem spiritus concitat: aquae st uentum detrahes inertes sunt : 
tune demum-impetum sumunt, cum illas agit flatus. 

226. nosse fidem rebus of C may be right ‘to know the 
amount of trust we can give to things’; what to accept as 
demonstrated by them. The dative would be justified by the 
construction of fidere. 

254—6 

Nam quae mortalis spes quaeue amentia maior 

In Iows errantem regno perquirere uelle 

Tantum opus ante pedes transire uc perdere segnes ? 


302 THE JOURNAL OF PHILOLOGY. 


So C, with est written over spes. I would write 


Nam quae mortali superest amentia maior? 
290 Seu forte flexere caput tergoque feruntur. 


Read fortes sc. uenti, ‘or waxing bold have rounded the 
head (turned the point) of Etna, and sweep on behind it’. 


293 
Nam ueluti sonat ora duc tritone cancro C 
diu canoro eye: 
sonitura dius tritona canoro Vat. 3272 }° 


Without attempting an examination of the previous cor- 
rections of this v., I will offer my own conj. based on Vat. 
3272, 


Nam ueluti sonat urna ciens Tritona canorum. 


Vrna, I think, would be quite a proper term for the hydraulic 
box which the poet is here describing. This machine seems to 
have sounded by setting in motion an apparatus communicating 
with a trumpeting Triton. 


In 294 wictusque mouere Spiritus I still think that mouerz is 
more probable. 7 


316, 317 
Atque haec in uacuo st tanta potentia rerum est, 
Hoc plura efficiant infra clusique necesse est. 


rorum (Jacob) seems right, though the word is a strange 
one under the circumstances. But in 317 Vat. has a v. 1. which 
is worth noticing, introclusique. Possibly then intra clusique. 


337 Non illam uidet Aetna nec ullo intercipit aestu 
Obsequitur quacumque wubet leuis aura reditque. 


May not widet be right? Aetna has no eye for, ‘takes no 
notice of’ this cloud, which is unaffected by the agitation of the 
mountain and simply drifts with the breeze. 


339  Placantes etiam caelestia numina ture 
Summo cerne ἄπο, uel qua liberrimus Aetna 
Inprospectus hiat, tantarum semina rerum 
Si nihil irritet flammas stupeatque profundum. 


ON THE AETNA OF LUCILIUS. 303. 


Munro takes Aetna as masce. here, quoting Solin. v. 9. 1 
cannot think it possible that in this one passage our poet should 
permit himself a licence which neither he nor any other poet 
seems to have taken elsewhere. Schrader’s Inprospectus is a 
very slight change and perfectly intelligible, ‘even at the point 
where the view inside the crater opens most freely on Etna’, 
1.6. at the very point where the agitation of the voleano is mest 
perceptible and seen most uninterruptedly. flammas I take to 
be the participle flammans; as an appositional accusative it 
is inconceivably harsh. 


351 Sparsa liquore manus sacros ubi uentilat rgnis 
Verberat ora tamen, pulsataque corpora nostris 
Incursant, adeo in tenui um causa repellit 
Non cinerem stipulamue leuem non arida. sorbet 
Gramina, non tenus plantis humus excita predas. 


The only thing in these five vv. which is questionable is the 
last word predas, for which however there are many variants, 
though mixed with such confusion as to give but a faint light. 
I quote them from Munro exit humus apredas 6, exit humor aprv- 
das ε, exit humor apndas y. These point to an unusual word, 
probably apludas, ‘bits of chaff’, for there is not to my know- 
ledge any thing to prove that the first a of apluda was different 
from the first a of aplustre, long or short indifferently. But why 
adeo in tenmi uim causa repellit should be changed into adeo in 
tenuist, uim causa repellit (M.) I cannot see. The meaning is 
perfectly clear, and the construction though more condensed 
than usual, legitimate,i.e. adeo in tenut causa est quae uim repellit, 
‘in so small a matter lies the cause of this repulsion of force’: 
namely, in the sprinkling of water on the hand, and the rapidity 
with which the hand whirls round the lustrating fire : two things 
slight in themselves, but enough to allow the human body (nos- 
tris) to feel the impact and charge of these natural bodies or sub- 
stances unharmed. Non cinerem, &c. returns, I think, to the 
main point of the sentence, the illustration of the seeming calm 
at the top of Etua from the undisturbed condition of the human 
countenance when fire is rapidly whirled round close to it in 
the ceremony of lustration. The nominative to sorbet is perhaps 


304 THE JOURNAL OF PHILOLOGY 


humus, ‘no ashes or light stubble, no wisp of dry grass, not the 
lightest chaff that the feet have stirred from the ground, is 
(drawn into the vortex of the fire and then) reabsorbed by the 
ground’, 


374 sqq. Saepe premit fauces magnis extructa ruinis 
Congeries clauditque uias luctamI NE ab imo 
Et sPisso ueluti tectoS sub pondere praestat 
Aut simill teneT oCcursu, cum frigida monti 
Desidia est tutoque licet dEsceNdere VEntiS. 


The capitals here mark the deviations from C. I am 
responsible for tectoS, teneT’ oCcursu, dE'scenNdere: the other 
corrections have been made before. C€ gives in 376 tecto, in 
377 similis teneros cursu, in 378 discedere montes. In 376 tectos 
sc. uentos (372). By occursu I mean ‘stoppage’, ‘ obstruction ’. 
Two causes are assigned for the intermittent violence of Etna: 
(1) an accumulation of rock which blocks up the’passage at the 
bottom and keeps the winds imprisoned wnder it, (2) a similar 
obstruction which meets the winds,on their way downwards into 
the crater during periods when the volcano is inactive. 


385 Nunc superant quaecumque regant incendia siluae 
Quae flammas alimenta uocent quid nutriat aethnam 
Incendi poterunt. 


So C, the only variant of any consequence is quod of several 
Mss for quid. In 385 M.’s rigant seems to me better than any 
other conjecture yet offered; but I greatly doubt his wocant = 
uacant, especially as C gives fammas not fammis. Why should 
not wocent be taken literally? ‘Every form of aliment meant 
to call up the flames, that Aetna feeds, may now be kindled’: 
substituting therefore guot for quid or quod of Mss, nutriat 
Aetna for nutriat ethnam. Superant I would translate ‘ become 
overpowering ’. 

393, 4 Atque hance materiam penitus discurrere fontes 
Infectae eripiantur aquae radice sub ipsa. 


Munro rumpuntur for eripiantur. I have before suggested 
what I think nearer to the letters of the word, crispantur. De 
Rooy in his clever Spicilegia Critica (1771) shows that erispus, 


ON THE AETNA OF LUCILIUS. 305 


crispari are particularly used of water. ‘Crispari eleganter 
dicitur aqua, cum breuioribus undis quasi trepidat. Hinc 
crispt undarum motus, Auson. Mosell. v. 194, eodem fere modo 
et arenam crispari dicit, v. 63. Minuce. Fel. c. 3, Lt ut semper 
mare, etiam positis flatibus, inquetum est, etst non canis spumo- 
sisque fluctibus exibat ad terram; tamen crispis torosisque 
ibidem erroribus delectati perquam swmus’. 


397 Quin etiam warie quaedam sine numine saxa 
Toto monte liquant. 


sine alumine M.: more probable, I think, is sine lumine ; 
substances which submit to the action of fire but without taking 
light, in opposition to sulphur, bitumen, &c. This ware is 
perhaps the right word in 184, Inter opus nectunt warie, where 
C gives uaries. 


425 sqq. Cerne locis etiam similes arsisse cauernas. 


This passage to 447 has not, I think, been understood. If I 
am not mistaken, the poet means that places near volcanic 
regions sometimes show traces of similar volcanic action, e.g. 
the coast on the mainland opposite Aenaria (Ischia) and the 
island Strongyle and Hiera not far from Mt Aetna. On this 
view it will not be necessary to add his in 425 after etvam 
(Munro), or to correct Loerts (Wagler): locis is sufficiently 
explained by simzles ‘observe again that fires have broken out ἡ 
in caverns corresponding te particular regions’: dic (426) will 
then mean in those extinet volcanoes, where the fire has died 
out from absence of the lapis molaris or lava stone: lava di 
Vesuvio as it is sometimes called by the Italians. His first 
illustration is from Aenaria, | 


4.29 Dicitur insidis flagrans Aenaria quondam, 
Nune extincta super testesque Neapolin inter 
Et Cumas locus et multis iam frigidus annis 
Quamuis aeternum pinguiscat et ubere sulphur 
In mercem legitur, tanto est fecundius Aetna. 


If 430 is rightly given by C and other MSs, super may 
possibly mean ‘at the top’,*i.e. covered over with grass and 
trees which prove that it has become extinct: it seems to me 


Journal of Philology. vou. xvt. 20 


306 THE JOURNAL OF PHILOLOGY. 


impossible to construct it with the following clause (so Munro). 
Somewhat similar in the form of antithesis is a passage of 
Lucan Vi. 355 Atque olim Larisa potens, ubi nobile quondam 
Nunc super Argos arant. But I suspect a corruption, possibly 
silet. In 431 my emendation pingui scatet ubere is accepted 
both by Munro and Bahrens though Herr Wagler has not 
condescended to notice it in his treatise of 1884. There is no 
reason to doubt the genuiueness of insidizs. Sudden outbreaks 
of nature’s most terrific workings are to this day characteristic 
of Ischia. Witness the frightful earthquake by which Casamic- 
ciola was in July, 1883, reduced in a few seconds to a heap of 
ruins. Those ruins are around me as I write this article; and 
so great is the insecurity still felt (a minor shock had preceded 
in 1881) that the prosperity of this lovely island will be, it is to 
be feared, seriously damaged for some years to come. In 1736, 
when de Serionne published his translation of our poem, Ischia 
had been long tranquil: his text gives indiciis. In reference 
to the particular point dwelt upon in the Aetna, it is interesting 
to notice that in the first century of the Christian era no 
voleanic eruption had taken place for so long that the memory 
of it was a mere tradition. dicitur flagrans Aenaria quondam, 
Nunc extincta. Yet Julius Obsequens, the author of the little 
treatise de Prodigiis, says that at the time of the outbreak 
of the Social War Aenariae terrae hiatu flamma excita in 
caelum emicuit. We may perbaps infer that this was a very 
short outbreak, which occasioned a momentary wonder, but no 
permanent impression. (See Johnston Lavis’ excellent Mono- 
graph on the Earthquakes of Ischia, Naples, 1885.) The won- 
derful 14 miles of lava blocks near the town of Ischia date, I 
believe, from the great eruption of M. Epomeo in 1302. 


439—443 
Insula durat adhuc, Vulcani nomine sacra, 
Pars tamen incendi maior refrixit et alto 
Tactatas recipit classes portuque tuetur 
Quae restat minor et diues satis ubere terra est 
Sed non Aetneis wires quas conferat εἰ. 


Such I believe to be the right punctuation and constitution 


ON THE AETNA OF LUCILIUS. 307 


of this passage, in which Scaliger’s durat adhuc for durata of 
Mss is beyond all praise. v. 440 is surely not to be written as 
M. gives it, 


Pars tamen incendi: maior refrixit et alto 


for this involves (1) making Pars a definition of Jnsula, the 
island, that is a part of it, (2) treating ncendi as an infinitive 
depending on durat: both of which hypotheses are forced and, 
to my view, impossible. I simply follow Scaliger here. v. 443 
I give after de Serionne: οἰ is the smaller, still volcanic, part 
of the Vulcanian island, which though active cannot compare in 
strength with Aetna. Οὗ gives Aethnet. 


450—452 
Nam circa latera atque imis radicibus Aetnae 
Candentes efflant lapides disiectaque saxa 
Intereunt uens. , 


Munro says ‘Zvt. wen. must mean ‘immiscentur uenis 
Aetnae’ but I know no other instance of this use of the word’. 
Surely this is not the meaning: the sense is that at the bottom 
of Mt. Etna stones may be seen smouldering with their pores, 
i.e. with the heat still alive but gradually dying out. The 
construction is exactly ,parallel to Sophocles’ φθίνουσα μὲν 
Kadviw ἐγκάρποις χθονός, Φθίνουσα δ᾽ ἀγέλαις βουνόμοις 


O. T. 25, 26. 


452. 454 manifesto ut credere possis 
Pabula et ardendi causam lapidem esse molarem 
Cuius defectus reiunus colligrt ignis. 


So C, and no change I think is called for. Translate ‘you 
may feel sure that these smouldering red-hot stones are attri- 
butable to the presence of the lava-stone, whose leavings the 


starved fire gathers up and burns fer want of a larger supply of 
fuel’, 


457 Haud equidem mirwm facie que cernimus eatra 
Si lenitur opus restant: magis uritur illic 
Sollicitatque magis uicina incendia saxum 
Certaque uenturae praemittit pignera flammae. 


20—2 


308 THE JOURNAL OF PHILOLOGY. 


I would offer, but only as a tentative re-constitution of 
457, 8, the following : 


Haud equidem mira in faciem, quae cernimus extra, 
St lentur opus, restant: magis uritur illic, &e. 


‘Not indeed that the effects we see outside the mountain, 
if the voleanic working is toned down, offer anything of per- 
manent interest as curious: the stronger burning of the lava, 
its more potent solicitation of the fires near it, is in the other 
centre, within the crater’, 


461 sqq. | 
Nam simul atque mouet uires turbamque minatus 
Diffugit ex(t)emploque solum trahit ictaque ramis 
Et graue sub terra murmur demonstrat et ignes. 


M. marks a lacuna after 461: but there would seem to be 
another after 462. ramis perhaps represents raris, with which 
Soraminibus in the lost v. might have agreed, cf. 566. In 463 
denuntiat (Jacob) for demonstrat et is very plausible. 


469—472 
Illine incertae facies hominumque figurae 
Pars lapidum domita stanti pars robora pugnae 
Nec repit flammas hinc defensus anhelat 
Atque aperit se hostis decrescit spiritus illic. 


So C. The Helmstadt Ms gives recipit for repit, rightly: 
and defessus for defensus. Almost all editors change hostis to 
host. I would retain it, and write the vv. thus 


Pars lapidum domita, stanti(s) pars robora pugnae, 
Nec recipit flammas: hinc indefessus anhelat 
Atque aperit se hostis, decrescit spiritus illic. 


Bahrens already has hic indefensus. 

(1) Why should not Lucilius lengthen @ before st, as so 
many other poets have done? (2) robora has every mark of 
genuineness. ‘Part of the stones present the sturdy strength 
of a standing fight, resisting all approaches of the flames: on 
one side the enemy (the fire) pants unweariedly, and opens out 
its forces, on another its violence is abating’. 


ON THE AETNA OF LUCILIUS. 309 


489 sqq. 
Nune siluae rupesque notant haec tela solumque 
Ipsa adiutat opes facilesque sibi induit amnis 
Quod si forte cauis cunctatus uallibus haesit 
Vipote inaequalis uoluens perpascitur agros 
Ingeminant fluctus et stantibus increpat undis. 


Wernsdorf’s rotant for notant (489) is accepted by Munro and 
looks right. Ipsa is a mistake not for tpswm, but ipse 86. 
amms. Opes are the materials which swell the lava current; 
are readily taken in by it and form part of its onward course, 
instead of arresting that course as might be expected. In 492 
is not in aequalis to be written? ‘inasmuch as it rolls over level 
fields, it grazes freely there’, 1.6. when it comes to a level 
surface with nothing to arrest its course, its velocity and freedom 
increase. Jngeminat of the 15th cent. Mss is probably right: 
but wcrepat I think is ‘loudly calls to its standing waters’ to 
come on, rather than as explained by M. 


498, 9 Paulatimque tgnes coeunt ac flammea messis 
Eeutur facies. 

Such is certainly the right punctuation: as by degrees the fire 

combines into a molten mass, it loses the SEPORTANCE of a 

waving field of flames. 


-506—8 uerum impetus ignes 
Symaetht quondam ut ripas traiecerit amnis, 
Vie iunctis quisquam fivo dimouerit illas. 


Lucilius here contrasts the impetuous onset of the lava- 
flood, which was sufficiently strong to carry it over the bed of 
the river Symaethus, with the utter immobility of the same 
lava-stream when hardened and solidified afterwards. Hence wt 
is ‘though’: alas are the banks which no effort of human skill 
can, afterwards, part clear again from the immovable lava-mass 
which now crosses them. But iwnctis, though retained by M., 
is so extraordinarily harsh that I think it must be wrong, and I 
would read for it wncis, grappling irons or grips which might 
naturally be used for hauling up heavy weights, or getting 
stronger hold upon them. fiwo may be right, though M.’s faxo 
is very clever. 


310 THE JOURNAL OF PHILOLOGY. 


532 Quin wpsis quondam Siculi cognomina saais 
Inposuere +fridicas et tam ipso nomine signant 
Fusilis esse notas. 

The variants frichas, fricas, phricas point with some dis- 
tinctness to a digammated word, perhaps Fputds or Ἐρύδας (ef. 
ῥύδαν διαῤῥύδαν) from ῥεῖν.. 

5387—540 
Cogitet obscurt uerissima dicta libelli 
Heraclite, tur, nihil insuperabile gigni 
Omnia quae rerum natura semina iacta 
Sed nimium hoe mirum. 

Scaliger’s ab igni for gigni is accepted both by Munro and 
Bahrens and seems indubitable. But 539, 540 I would write 

Omnia quae rerum natura semina tacta, 
Seminium hoc mirum.. 


‘This (fire) is the marvellous seed-ground (nursery) of. all the 
seeds of things planted in the course of nature’. The construc- 
tion is, omnia quae rerum semina natura vacta (sunt), hoc mirum 
seminium (esse): omnia is.of course an attraction. into the case 
of quae. 
547 sumilique obnoaia sorte.. 

Lexicographers should take note of this abl. It is quite 
possible that it is a rare, but still correct, construction. Such 
an abl. after obnoxius occurs in the Digest (Forcellini). 


555, 7 quae tanta. putas incendia nostris 
Sustentart opibus, tantis fornacibus Aetna 
Vritur ac sacro numquam. nec fertilis igni 
Sed non qui nostro feruet moderatior usu, 
Sed caelo propior. 
quantis for tantis is an old and necessary correction :. for nec 
in 557 editors are content to write non. Possibly the origina 
reading was numquam haec non fertilis ignt.. : 


568 sqq. Magnificas laudes operosaque uisere templa 
Dimtus hominum aut sacras memorare uetustas: 
Traducti materia et terris per proaima fatis 
Currimus. . 


ON THE AETNA OF LUCILIUS. 311 


If laudes is an error for aedes it is a rare one, to say nothing 
of the tautology with templa following. Till something better 
is proposed, I prefer to explain it, somewhat on the analogy 
of res antiquae laudis et artis (Verg. G. I1.), as ‘glories’ 1.e. 
monuments which have become famous in the world for their 
magnificence. Sacras I believe is a corruption of arcas, 
‘coffins’ of ancient worthies, such as were shown particularly 
in Egypt. Maria is De Rooy’s convincing emendation of 
materia: for the old conj. traduce materia, which I have 
found in a MS. of the Naples Museum, cannot be right even as 
Latin, to say nothing of the unusual rhythm. Whether terris 
(? terras) is right, or is a mistake for certis, it is hard to decide. 
For the infinitives wisere memorare after currimus, see my note 
on Avianus ΧΧΙΙ. 1, 2. 


586 philomela canoris 
Euocat in siluis et tu soror hospita tectis 
Acciperis. 


For Huocat in I would write Plorat (It)yn. A similar 
depravation attaches to this unfortunate name in Cul. 252 
Quarum uox Ityn edit Ityn, which the oldest Vatican Ms 
(Bembo’s) presents in this strange shape, Quarum wox rt in 
edytyn. 


612,613 Viedum castra putant hostem mouisse, tremebant. 
Et vam finitimae portas-euaserat urbis. 


Jacob wrote tremendum for tremebant, a weak and improbable 
conj. The mss have rightly preserved tremebant: ‘scarcely had 
they begun to think the enemy was on the march, and already 
they were trembling at his approach’. 


619 Et quod cuique fuit cari fugit ipse sub illo. 
Caesar B. G. v. 33 of a rapid flight, quae quisque eorum 
carissema haberet ab impedimentis petere atque arripere 
properaret. 3 
621—623 
Cunctantis uorat ignis et undique torret auaros 
Consequitur fugisse ratis et praemia captis 
Conerepat. 


312 THE JOURNAL OF PHILOLOGY. 


Le Clere, Concremat, ingeniously. 


623, 4 haec nullis parsura incendia pascunt 
Vel solis parsurat dees. 


Munro pies. But deis may be right: the flames spare the 
gods alone, i.e, those whom the gods interfere to save for their 
piety. 

627, 8 
Aspiciunt pigrumque senem matremque + senemque 
Eheu defessos posuisse in limine membra. 


Bahrens senentem: rather sequentem: the que is out of 
its place. : 


629—632 
Parcite auara manus dites attollere praedas 
Illis diuttiae solae materque paterque 
Hane rapitis praedam: mediumque exire per ignem 
Ipso dante fidem properant. 


rapitis is my conj. for rapies of Mss. The poet bids the 
selfish majority of the Catinaeans, who had carried off their 
valuables, to spare this more precious burden, the father and 
mother whom their sons had saved instead of property: ‘this 
is the only booty you can seize’. The que of mediwmque marks 
an act which is an immediate attestation of the piety of the two 
brothers and the visible interference of the gods in their behalf. 


637  Dextra saeua tenent laeuaque incendia feruent. 


I have no doubt this is what Lucilius wrote :. in an immense 
proportion of cases I have found dextra written as a trisyllable 
dextera even when palpably contra metrum. Saeua agrees with 
incendia : tenent is ‘are in occupation’. | 


ROBINSON ELLIS. 


Casamiccrona, IscHta, 
June, 1887. 


As an Appendix to the above, I add some conjectures on 
the Aetna, as well as on the Culex and Ciris, which were sent 
to me by the well-known Professor Robert Unger of Halle in 
November of this year. 


57 sq. 


(1) 


AETNA. 


Quod fremat imperium (fremat, ut Stat. Theb. 11. 576 
et, ut videtur, Sedul. 1.. 196. imperium, ut Val. 
Max.1. 1.9, Arntz. Maximian. Pan. 13, 5, p. 343, 
Barth. Stat. vi. 315 p. 4632). 


Seu te Cynthus habet seu Delo est gratior Arna 
Sive tibi Zenedos potior. 


Iam nova. Pierio properent a fonte sorores 


Pocla. 


Quis non Argolico deflevit Pergama in igni (Nicandr. 
fr. 62: ἐν πυρὶ---πάτρην). 


Compositam et tristi natorum funere matrem. 


Aversumque diem sparsumque 6 semine dentis (= Prop. 
11. 21, 30,33. 6 semine, ut Stat.: Martisque e 
semine Theron. dentis, ut Val. Flacc., Lucan., 


Claudian.). 


Quidquid in Aetna actum, iam facta est fabula: carmen 
Fortius (ignotas molimur pectore curas), 

Qui tantz motus, opera δὲ quae tanta perennes 
Explicet introrsum. flammas. 


Pelion Ossa gravat. 


infestus cunctos ad proelia divos, 
Praenotat amotis, qua Tethyos aequora, signis. 


Iupiter et telo metuit dextramque corusca 
Abiunctus flamma. 


Hic magno tonat ore pater geminantque /fuwrentes 
Undique discordi sonitus molimine venti. 


314 
61 sq. 


65 sqq. 


74. 
76. 


80. 


81. 


84. 


86. 


107. 


128. 


THE JOURNAL OF PHILOLOGY. 


patri iam dextera Pallas 
Et Mars laeva ierat, iam et cetera turba deorum: 
Stant, wt cuique decus. Vastos tum Jupiter ignes 
Increpat et tunctos proturbat fulmine montes. 
(laeva ierat = in laevam, dextram partem, ut Ovid. 


Plin. al.) 


Ilicet invectae verterunt terga ruinae (Senec.: terga 
dare alicui). 

Infestae divis acies atque zgnibus hostis 

Praeceps ceu flagris agitur materque fatiscit 

Componens cunctos: tum. pax est reddita mundo, 

Tum liber cessata venit post foedera Phoebus 

Defensique decus mundi dus redditur astris. 


Haec est ventosae vulgata licentia famae. 


Plurima pars scenae rerum est fallacia: vates 
Sub tenebris nigros finxerunt carmine Manes. 


Mentiti fauces Stygias umbrasque canentes (cf. Simo- 
nid. Hor. Stat. al.). 


Hi Tityon poena stravere in iugera fetwm (Sinid. 
c. VL 3 p. 111 sq.). 


Sollicitant malo te siccum, Tantale, plena (Sin. Iv. 
1 p. 62sq.). 


Quid, quod et ulterius falsi contagia aberrant 
Necdum terra sat est? 


Nec metuunt oculos alieno advertere mundo: 
Norunt bella deum, norunt abscondita furtis 
Coniugia. 


Ut crebro introrsus spatio recava acta Charybdis (re- 
cavus Stat. Avien. Prudent. Alcim. Avit. Paulin). 


Quid, si intus versos emittat terra canales 
Hospitium in fluidwin 1 
(fluvius: fluidus codd. Lucret. 1. 596 cf. Cort. 
Lucan. VI. 89, p. 16; hospitium Plin. N. H. νι. 
18, 22. Pallad. R. R. 1. 17, 2). 


129. 


132. 


146 sq. 


150 sq. 


158 sq. 


162 sq. 


114. 
168. 


AEBTNA. 315 


—sunt semina nulla profecto 
Fontibus et rivis non stat via pigraque tellus 
Conferta in solidum: segni sub pondere cessat. 


Condita si redeunt, si quae clam condita serpunt 
(=Senec. N. Q. VI. 8 quo illum putas abire nisi in 
obscura terrarum. clam serpunt, ut in Here. 
Fur. 186 nimium pectore forti; Stat. os flatu 
paene inviolabile tunctus. Sil. xiv. 425). 


Nam quo liberior quoque est animosior zmpes 
(Asper enim in clauso nec ventus segnior irae est 
Sub terra penitus remanens), pia tura necesse est 
Victa magis solvat, magis hoc obstantia pellat. 
(in clauso, Virg. Senec. Columell. Impes, Priscian. 
vi. 10, 55: Gloss. Labb. p. 88). 


Nec tamen in privos exit collecta canales: 
Vis animae: flatu acre ruit, qua proxima cedunt 
(acre, ut Sallust. Manil. Sulpic. al.). 
Obliquansque. secat, qua jfissa tenerrima, claustra. 
Sed summis si forte putas. concrescere caulis 
Tantum opus ex subitis alimenti incursibus, ora 
Namque imis quacunque wgent in hiatibus, omnes 
En sursum. introitu assiliunt ostioque patenti 
Consertae languent vires animosque remittunt. 


(IT) 
CULEX. 


Floribus-fragrantibus (Cir. Comm. p. 290). 
haec cura est subdita cordi, 
Qualibet ut requie, victu quum venter abundat, 
Iucundoque Jevet languentia corpora somno (Cir. p. 
245). 
Posterius poenam vatum est memorare futuram. 


Tendebant tarde venientis ad humida nisus (Cir. p. 
293). 


316 
172. 


174. 
176. 
202. 
216. 
233. 
240. 


242. 
200. 


265. 
911, 


378 sq. 


THE JOURNAL OF PHILOLOGY. 


Edita purpureo luces iaculatur amictu 
Adspectwmque negat (Cir. p. 222 et Cinnae fr. p. 5). 


Vectabat sese circum loca, quum videt aegre (Cir. 
p. 299). 


obvia ad ornum 
Saevius arrepens infringere. 


Iam quatit ew biiugis oriens Erebo cita equos Nox. 


video en flagrantia taedis 
Limina quam livent infaustis conscia templis (Cir. 
p. 293). 


Quam tristes circa densentur in ultima poenae! (vel 
(ua). 
visu uvidus amni 
Restat. (Cir. p. 290). 


resolutus in ultima 


saeva marito 
In Chaleodoniis sat Mortis cura morata est. 


Ecce Ithaci conjux supra est, genus Icarionis 
Femineum omne indepta decus: m. (Cir. p. 197). 


flamma arva cremante 


(Stat. ἘΠ]. Ult. p. 179). 


(Complicatam sententiarum rationem his evolvimus: 
quum tu mihi causa mali sis nec conscius tibi facinoris 
nequaquam tolerabilis: etsi hoc, quod gravius acerbius- 
que adversus te dici oportuit, ita audis, ut ne nunc 
quidem iustitiae memor gratam voluntatem testificeris 
(v. 227, 230), contingat tamen, ut ipsum somnium 
alta mente conditum teneatur. Quas sententias codi- 
cum vestigia pressius sequendo licet his verbis com- 
plecti :) 
@uum mihi tu sis causa mali nec conscius ausis 
Haud tolerabilibus: si iwris hoc immemor audis, 
Sit tamen alte adytis demittere somnia mentis. 
(Cir. p. 289.) 


Digredior: nowam immeritus luo: tu cole fontes. 


303. 
305. 
315. 
324. 
326. 


359, 


CIRIS. , 317 


(III) 
CIRIS. 


Alcathoi, Phoebi usque decus: namque affuit illi, 
Unde etiam citharae voces imitantur acutas. 


Sistere et indomitas virtute retundere mentes. 


Candida caesaries (frondebant tempora lauro), 
Sed roseus—. 


Aurea sollemni comptum οἱ fibula ritu 
Morsu habilem tereti nectebat dente cicadae. 


Nec veri haec dotis custodia vana fuisset 
(Haec mora erat). 


ne perdita 


Aurea acus gracili solvisset corpore pallam! 
Omina, quae retinere gradum cursusque morari 
Possent—hoc tantum—vellem obvia semper haberes. 


caeli speculatur honorem. 
Sordibus et scaeva patiar tabescere labz. 
quove icta malo hoc exordiar ore ? 
capta arce avecta nequivl, 
Tam grave servitium, tam duros passa labores, 
Effugere? adsistam exitium crudele? malorwm 
Summam nec nobis aequum et senioribus ullum 
Vivendi capiam pretium et decus ? 
Unde alii affulsisse ferunt. 
Dictynnam dixere tuo de culmine lunam. 
Saepe tremo. 
Sin est, quod metuo, per munia alumna—. 
Per te saxa precor, per flumina mitis Eleuthus, 
Ne tantum en facinus tam cruda mente sequaris. 
Communemque timere deum vult regis amicos, 
Nunc se 1886 in vetitum; ast orbum flet maesta parentem, 
Cum Iove communes cut non datum habere nepotes. 


318 


374. 
384. 


409. 


THE JOURNAL OF PHILOLOGY. 
Inde nigro geminata Iovi feralia sacra. 


Illud imprimis mihi laetandum iure esse video, quod 
Tu Rhauct nomen restituendum censuisti. Haec enim 
in commentaris nostris M. 5 p. 249 sqq. scripta extant: 
Sed enim brevi diluxit id ipsum omnes veritatis numeros 
continere, quod nec alius quisquam in ullo posuit 
discrimine et qui in examinando codicum, quibus usus 
est, pondere frustra laboravit, Ribbeckius p. 46 “ prava 
coniectura” illatum esse dicere ausus est. Rhaucus 
enim Cretae urbs fuit famae nequaquam obscurae: testes 
sunt Meursius Cret..1. 15, p. 58 et quem non neglexit 
Gronovius Scylac. Peripl. p. 42, Holstenius Steph. Byz. 
p. 270 (cui ipsi emendati versus Lycophr. 1304 laus 
debetur illa quidem a Bachmanno p. 264 Hoeckio Cret. 
I. p. 433 vindicata). Tanta vero elus vocis novitas 
plerisque visa est, ut alii (Vatic.) vacui spatii aliquid 
relinquere, quam non intellectum vocabulum chartae 
mandare, mallent, alii scribendo depravatum rauci (ravst, 
raphci) ad hane quae iam perplacuit speciem (rephaht) 
reveht deducerent, quum non promptum magis, quam 
consentaneum -esset hoc redintegrari: Rhauct moenta 
(arx v. 290), unde novo appareret documento non 
vulgares in eo scriptore litteras fuisse, de quo tot docti 
minus bene sentire consueverunt. Itaque nos non 
temere eruisse videmur rem ut vetustate oblitteratam, 
ita maxime pertinentem ad pernoscendam fortunam 
Carmes filiaeque, quas Rhaucum urbem (nam alii Caeno 
memorant Wess. Diod. v. p. 392, 44) patriam sedem 
(v. 385, 290) habuisse iam pro comperto est. Iam vero 
hoc deliberato ac constituto sequitur, ut non minus certa 
arte enucleem, quod adhuc omnibus difficile fuit ad 
excutiendum. Quod enim librorum consentiens auctoritas 
praebet: moenia crescant (crescat), id ad hunc statum 
revocare, moenia restant, nullius negotii est, siquidem 
videntur permutatae litterae ὁ et s (rescant) genuisse 
illud crescant. 


Vos, o Hmathia... 


CIRIS. 319 


441. Nancta queror necem. Et illa quidem communis: at ulla 
 Ossibus injecta tellus tumulabit arena ? 


443. Mene inter Mnotas ancillarumque maniplos 
Venales inter famulari munere fungi. 


451. Aequoreae pestes, wmitamina corpora montis. 
469. Et notas aegui heu frustra respectat Athenas. 
477. Anguineamque sinit Tenwm undiferamque Seriphum: 


478.  Fertur et infestis iactatur ad ultima ventis, 
Cymba velut magnas sequitur quum parvula classes, 
Afer at hiberno bacchatur in aequore turbo, 
Donec tale decus formae vastarier Euris (Austris)—. 


R. UNGER. 


POSTSCRIPT. 


[It is due to other critics as well as to myself to state that 
some of the corrections proposed by Prof. Unger have been 
anticipated. Aet. 49 grauat by Jacobs and Munro, 69 cessata, 
77 finwerunt by my own article in the present number of the 
Journal, written five months before Prof. Unger’s conjectures 
were sent to me. 

I observe too that, no doubt inadvertently, Prof. Unger has 
repeated a conjecture of mine on Cir. 175 honorem, and suggested 
what I think no improvement of another Cir. 361 cuz non datum 
where I had conjectured qui non dat. Both honorem and qui 
non dat were printed in the first no. of the American Journal of 
Philology for 1887, of which I sent a copy to Prof. Unger, and 
which he has quoted on Cir. 384 Rhauct. On the other hand 
Prof. Unger has forestalled Mr Hildebrandt’s lucem caculatur 
Cul. 172. The reference on p. 318 ‘in commentariis nostris M. 
5 p. 249’ is to Prof. Unger’s unpublished remainder of his 
commentary on the Ciris, the first portion of which was pub- 
lished as a pamphlet in 1886.—Ropinson ELLIs.] 


320 THE JOURNAL OF PHILOLOGY. 


CIC. ACAD. PRIOR. xxv. 79, 80. 


Tu autem te negas infracto remo neque columbae collo 
commouert. Primum cur? nam et in remo sentio non esse id 
quod uideatur, et in columba pluris uderi colores nec esse plus 
uno. Deinde nihilne praeterea diaimus? Manent ella omnia, 
tacet (so Reid, laterat, lateat, vaceat MSS) wsta caussa: weracis 
suos esse sensus dicit. Igitur semper auctorem habes, et ewm 
quit magno suo periculo caussam agat. Ho enim rem demittit 
Epicurus, si unus sensus semel in uita mentitus sit, nulla wumquam 
esse credendum. Hoc est uerum esse, confidere suis testibus et 
+inportata insistere. 

The last sentence is ironical, as Reid shews in his Transla- 
tion, ‘This is candour, to rely on your own witnesses!’ The 
irony of the passage will be perfectly maintained if for the 
corrupt inportata we write in torquata, ‘to take a firm stand on 
an appeal to the ring-dove’s neck’, TZorquatus as an epithet of 
ring-doves is found in Prop. Iv. 5. 63, Sed cape torquatae, Venus 
o regina, columbae Ob meritum ante tuos guttura secta focos, 
Mart. x1ul. 67. 1, Inguina torquati tardant hebetantque palumbi. 
The omission of the substantive in our passage would be 
perfectly intelligible after columba twice mentioned in 79. 


ROBINSON ELLIS. 


CAMBRIDGE: PRINTED BY C. J. CLAY, M.A, ἃ SONS, AT THE UNIVERSITY PRESS. 


ἂν 
oO 
ro 
Oo 
ri 
rs 
ot 
me 
i 
ο 
ri 
c 
SS 
τὰ 
τὰ 
cS 
> 
ἐκ 
Ξ 
« 


| Uae fT | University of Toronto 


1] 


Library 


DO NOT 
REMOVE 
THE 
CARD 
FROM 
THIS 

PO CKET 


Acme Library Card Pocket 
Under Pat. ‘‘Ref. Index File’’ 
Made by LIBRARY BUREAU 


Roe 
Beet ae ¥ 


δέ “τ