STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world byJSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. A GREEK ANALOGUE OF THE ROMANCE ADVERB By Paul Shorey The original suggestion of this paper was merely the observa- tion of an interesting coincidence. The origin of the romance adverb in the instrumental ablative of mens with an adjective is familiar to every school boy. Laeta mente, Catull. 64. 237, is lietamente; constanti mente, ibid. 210, 239, is constamment. Sa- gaci mente, Lucretius 1. 1022, is, or rather might be, Italian sagacemente, French sagacement; obstinata mente, Catull. 8. 11, is ohstin6ment, ostinatamente; liquida mente, 63. 46, is, or rather might be, liquulement, Italian liqutdamente; mente maligna, 68. 87, is malignement; perversa mente, Commodian 1. 26. 24, is perversement; fera mente, ibid. 1. 28. 4, is fidrement; profana mente, Apuleius Met. 11. 5. 27, is, or might be, profanement, Italian profanamente; devota mente, Claudian 21. 232, is divote- ment; pura mente, ibid. 28. 608, is purement; integra mente, Lactantius 6. 1. 2, is entidrement. Tacitus Annals 4. 22. 8 has turbata mente. With memori mente, Lucret. 3. 859, Hor. Sat. 2. 6. 81, Manil. 2. 862, cf. memoriter. Apuleius 3Iet. 1. 6. 24 has dubia mente, 11. 6. 15 penita mente. Quintus Curtius 8. 3. 7 opposes with invidious antithesis fida mente and the adverb muliebriter. I should not have thought it necessary to cite even these random instances if, after my attention had been called to the matter by the comparative frequency of the idiom in Catullus,' I had not discovered, to my surprise, that neither romance scholars nor students of vulgar Latin have collected the examples in Latin literature of a construction so important for later developments. > 8. 11, obstinata mente; 62. 14, tota mente; 62. 40, tacita mente; 63. 46, liquida mente; 64. 71, perdita mente; 201, quali mente; 202, tali mente; 210 and 239, constanti mente; 249, mente immemori; 255, lymphata mente; 68. 37, mente maligna; cf. also, 64. 95, immiti corde; 100, languenti corde; 124, immenwripectore; 125, ardenti corde; 221, laetanti pectore; 295, sollerti corde; 116. 1, studioso animo, where studiose has been conjectured ; I am, of course, aware that in some of these instances the ablative is not strictly speaking modal. [Classical Philology V, January, 1910] 83 84 Paul Shorey The romance grammars, of course, all mention it. But I find nothing that goes much beyond the brief account in Nyrop Grammaire Historique de la Lcmgue Frangaise, Vol. Ill, pp. 292, 603: " . . . . a son point de d6part dans des expressions telles qne forti mente (Ovide) , bona mente factum .... (Quin- tilien), qui religionem devota mente suscepit (LeBlant Inscrip- tions religieuses de la Gaule, No. 436), concupiscit iniqua mente (Gr6goire de Tours)." "Sur le module de ces formules," he continues, we find brevi mente, rapida mente, and finally ricem- ment, nouvellement, dernidrement." There is need of a monograph on the subject which should include similar uses of corde, animo, ratione, pede, manu, modo, modis, etc. Meanwhile my colleague. Professor Beeson, gives me the examples cited below from later writers.^ This paper, however, is concerned not with Latin but with Greek. Scholars do not appear to have noticed, or to have 1 1 quote them without delaying to distinguish the cases in which mente retains all or much of its meaning, or even is construed as a locative. In a special study of the subject they would be retained but classed apart : Apuleius De Plat, et eiws dogm. ii. 24 : " sic enim fiet, ut omnes una mente sint iniquumque sibi factum nolint;" Florida xx: "ita ut contra in maleficiis etiam cogitata scelera, uon perfecta adhuc, vindioantur, cruenta mente, pura manu;" Met. i. 6. 24: "hunc talem quanquam necessarium et summe cognitum, tamen dubia mente propius accessl;" v. 23. 19: "sed dum bono tanto percita saucia mente fluctuat, lucerna ilia .... evomuit;" xi.5.27: " id sacrum nee sollieita nee prof ana mente debebis opperiri;" xi. 6. 15: "plane memineris et penita mente conditum semper tenebis mihi reliqua vitae tuae curricula ad usque terminog ultimi spiritus vadata." Arnobius Adv. nat. ii. 60 : "ad dominum rerum tota mente atque animo proficisci." Ausonius Ep. 14. 10 : " Set quae facunda de peetore Clementini Inspirant vacuos aliena mente poetas." Cassiodorus Varia iii. 31. 5: "adhibete nunc studia, praestate solacia, ut inquisi- tionern, quam debueratis petere, grata videamini mente eomplere;" iv. 20. 2: "Sed qiiia uos uti nullum volumus fraudibus suis, praesertim cum in dispendio pauperum detestabili mente versetur .... decernimus ut," etc.; v. 1. 3: "ut haec inter nos grata mente facientes ; " v. 13. 1 : " Studiura vestrum rei publicae grata mente debetis impendere quia," etc. ; v. 44. 3: "sineerem purgationem pura mente suscepimus;" vii. 35. 2: "Hinc est quod veniendi tibi ad comitatum iiduciam grata mente largimur ne," etc.; x. 14. 1: "Licet vobis sit insitum dominos vestros pura mente diligere;" X. 15. 1: "qui vobiscum sincera mente coUoquitur ; " x. 18. 1: "Kemedium, quod pro vobis, patres conscripti, pia mente traetavimus ; " xi. 3. 1 : "qui auctorem rerum illuminata mente conspicitis ; " xi. 28: "luvat bene meritorum votis beneficiis respondere vicariis, ut devotiore mente i)ossit obsequi, qui meruit anteferri ; " xii. 2. 5: " ut tributa indictiouis tertiae decimae devota mente persolvat ; " xii. 16. 2 : " debitum A Gkeek Analogue of the Romance Adverb 85 thought it worth while to mention, that in the periphrastic dic- tion of Greek tragedy a similar use of the instrumental or modal dative p€vl became almost a formula, though it could not give rise to a new form. In Aesch. Choeph. 772: yrjdova-rj pevi, cor- responding to aSeifiavTox; in 771, is virtually laeta mente or lietamente. In Choeph. 303: Tpoiap€vi is einrpoa-rjyopo)^ (Dionys. Hal. ) = affabili mente, which I presume will be found with search = affablement. In Aesch. Persae 374 : veiOapx^ <^pevl, X aKoap.af siquldem quod non potest evitari, prona debet mente semper offerrl;" i. 1. 6: "Qua- propter salutatlonis honorificentiam praeferentes prona mente deposcimus ne," etc. Carmina Epigraphica 301. 1 ff. : " Constantina deum venerans Ohristoque dicata Omnibus impensis devota mente paratis Numine divino multum Christoque iuvante Sacravi templum," etc.; 653. 1 ff. : " Cara pia coniux Yguia deditaque marito, Funeris tui causa tota nos mente dolemus ; " 750. 7 : " Tu tamen hinc spera caelum pla mente, fidelis." Codex Theodosianus xvi. 8. 18: "Judaeos .... sanctae cruois adsimulatam speciem in contemptum Christianae fidei sacrilega mente exurere provinciarum rectores prohibeant." Claudianus Car. xxi. 232 : "Ultro quin etiam devota mente tuentur Victorique favent ; " xxii. 142: "Quis cernere ouris Te vacuum potuit? quis tota mente remlssum Aut indulgentem dapibus, ni causa iuberet Laetitiae ; " xxviii. 607 : " meritis offertur inemptus Pura mente favor." Commodianus Itistr. i. 21. 2 : " Obscurati male aliena mente iurantes ; " i. 26. 24 : "Gens et ego fui perversa mente moratus ; " i. 28. 4 : " Tollit se in parte et fera mente moratur." Corippus Joh. iii. 379: " Justitiam nullus oompuncta mente secutus ; " iv. 266: " At pater exsurgens compnncta mente Johannes, Corde pio, genibus nlxis et poplite flexo, Supplioiter gemmas tendens cum lumine palmas. Ore canens haec verba refert ; " viii. 156 : " Has inter voces auctores criminis insons Turba trahit, vinctosque duoi pro crimine tanto Tradidit, et culpam devota mente piavit;" viii. 162: " devota mente fatentur Nunc hilares socii diotis parere iubentis;" 86 Paul Shorey is oboedienter, Italian ubbidientemente, and I dare say •jrei6dp')(a><; exists though not in the lexicons. In Aesch. Sappl. 775: €vyXax7a-(p ^pevi is Italian facondamente; in Choeph. 565 : (f>aiBpa pevi is giocondamente or allegramente; iXeq) (ppepi, Soph. Track. 763, is graziosamente. In Soph. frag. 579, -rrai^ova-r) pev( is giocosamente. 'A0o/Sa) pevi, Ar. Aves 1376, and araplSel <^pevi, Pind. P?/tt. V. 51, are intrepidamente (cf. trepida mente, Lucan 4. 701) ; ayvfiPcuTT^ pevi has no avyyopoK to match it, but is exactly paral- leled by the use of iyyevm in Soph. 0. T. 1225. Compare further Aesch. Ag. 895: a-rrev9riT(p pevr, Sept. 484 fiaivofieua <})pevi (lyric); Eumen. 986: fua (f>pevi [una mente, supra, p. 84, n. 1) ; 275: Be\Toypdq)pevr, 489: ixSiKOK p€aiv; Soph. Track. 103: •n-odov/Meva (ftpevl (lyric) ; 293: iravBUcp <^pevi; 264: aTrjpa In land. Justini iil. 406: "Quaeque suis aptata locis, rerumque favorem, In quantum potis est, devota mente notabo." Dracontius Romulea viii. 104-6 : "Verba fides pletas quatiunt mox corda parentum Admlssumque nefas generoea mente fatetur Pusus in ora rubor ;" De laud, dei iil. 577 : " Sacrilega quasi mente putem non omnia nosee Aut aliquid nescire deum," etc. ; vii. 113: "Unus [i. e. pontifex] erat Latialis mysticue aulae Alter apud Danaos sacrata mente dicatus." vii. 123 : " Quod licet exiguum tamen inter iura poetam Temnitie immemores facunda mente periti ; " Orestes 232: "Dum commune nefas aequali mente fruuntur (Egistue et Olytae- meetre) ;" Carm. xiv. 27 : "haec taediosa mente dum recogito." Portunatue Car. ii. 16. 41 : "Nee tetigit muetum, sed iniqua mente rotatur: Antea quam biberet, ebria turba iacet." Lex Salica (Hessels & Kern) xvi : " Si quis uxorem alienam vivo marito tulerit, malb. abtica: et in alia mente, arba theue en lanthamo, VIII M den. qui fac. sol. CC culp. iud.;" xxiii: "Si quie caballum alienum extra consilium domini sui ascenserit aut cabalicaverit, mal leodardo in alia inte borio sito sol XXXV culp. iud." in alia mente is explained as = Fr. autremente. Quintus Curtius viii. 3. 7: "At ilia purgare se, quod, quae utilia esse censebat, muliebriter forsitan, sed Ada tamen mente suasisset;" viii. 5. 5: "lamque omnibus praeparatis, quod olim prava mente conceperat, tunc esse maturum (ratus). A Geeek Analogue of the Romance Adveeb 87 pevi (lyric). To these may be added cases in which a virtual synonym is substituted, e. g., TXdfiovi V^v^a, Pind. Pyth. 1. 48; /xaXa aSoPTi v6]0ei, Simon, fr. 22; ■7rp6povi KapBia, Aesch. Suppl. 348; eKova-r) KapBia, At. ^2.1269; arpeaT^ KapSia, Aesch. Ag. 1402; iiaveur-g irpairCh, Eurip. Bacch. 999; '^Sofieva •^vxa, Eurip. frag. 754; and countless examples with v6^, ffv/Jio), KapBia, and "^VXV from Homer down. Though especially conspicuous in tragedy, this idiom like most things goes back to Homer. It is suggested by ^/seo-t /tat- vofievT/cn, II. 24. 114; oXot^ai, (f>pecrl, II. 1. 342; IBviricn irpairlSecrcnv, i. 6. i-TTuyTafieva)^, II. 18. 380, Odyss. 7. 92; and still more by the use of 6vfiq> in TerirjoTi Ovfi^, II. 17. 664 ; irpw^povi 0vp,^, 24. 140 ; aeicovTi dvfiq), 4. 43; tstXtjoti, Ovfim, Odyss. 9. 434; 11. 181; vrfXel Ovp,^, i. e. vrfXeSy;, Odyss. 9. 272, 287, 368; a€aipovi 6vfiw, 21. 302. Hes. Theog. 661 has arevel re vow kuI eirC<^povi, ^ovXri; Theognis 199, (f>LXoKepSei 6vp,^; 1125, vrfKel dvfiw; 1325, ev (cf. €vp6vaK, Pind. Fyth. 10. 40) . In the Attic Scolia aSoXa p€vl occurs in (5), but aS6Xa><; in (6), largely metri causa. Pindar has opOa pevl, 0. 8. 24; iXevdepa (f)pevi, Pyth. 2. 57; (lawofievac; (ppairlv, Pyth. 2. 26; arap^ei ^pevC, Pyth. 5. 51; X&ovLa pevi, 5. 101; ^porea pevi, frag. 61(33). And with sub- stitution of synonyms aXaOel vocp, 0. 2. 101; evfievel v6a>, Pyth. 8. 18; €k6vti v6a>, Pyth. 5. 43 and 8. 67; areXel v6a>, Nem. 3. 42; v6(p inevei, Nem. 7. 88; €vcre/3ei yvoiiJia, O. 3. 41; Kadapa yvmfjui, O. 4. 16; KapBia ryeXavel, O. 5. 2; dvfia yeXavel, Pyth. 4. 181; afioxOo) KapBia, Nem. 10. 30; rXafiovi yjrvxa, Pyth. 1. 48; fidXa aBovTi voa, Pyth. 6. 51. These examples are more than sufficient to illustrate the mere coincidence. But the whole question of the relation of these and similar modal datives to the adverb requires further discussion. Brugmann says that adverbs are mostly recognizable as "er- starrte" nominal or pronominal cases, or "erstarrte" syntactic word complexes in which the main component is a noun or a pro- noun. This is perhaps a formally true account of the older Greek and Indo-European adverb. It ignores the romance 88 Paul Shoeey adverb and the idiom here considered in which the noun is not the predominant component, but is merely used to carry the adjective or participle which it virtually converts into an adverb. Primitive man possesses and requires but few adverbs of manner. "The want of adverbs in the Iliad,''^ says Coleridge {Table Talk), "is very characteristic. With more adverbs there would have been some subjectivity, or subjectivity would have made them." On this hint, my student, Mr. C. E. Vance, has undertaken a study of the Greek adverb, in part statistical, in part lexicographical and stylistic. His collections already verify the presumption that Isocrates, most self-conscious of writers, would make nearly, if not quite, the largest use of adverbs and adverbial phrases. The function of the adverb may obviously be assumed by prepositional phrases, which multiply with the de- velopment of the language,' by the "internal" accusative, and by the modal datives considered in this paper. The wide adverbial range of the "inner" accusative is touched upon by Ktlhner-Gerth, § 410, and Monro Homeric Grammar, p. 129. They mention accusative formations such as aKjjv, aBrjv, \ir]v, etc., the adverbial use of the neuter adjective, singular and plural, and many adverbially used cognates. What we have to note here is that in dirptjKTOv iroXeixov iroKeixi^eiv the noun merely serves to carry the adverbial force of the adjective, precisely as in the datives to be considered. So vo3, f^'Opco, dpdirei., ffOevei, Xojco, vofim, ^im, o-To'Xp, etc., will serve equally well. The special frequency of the usage in the Attic drama is due to several causes: (1) the tendency of developing language to multiply adverbial phrases; (2) the lack of an adverb;'' or (3) the avoidance of an unpoetical adverb;' (4) the periphrastic 07/cos of tragic diction as created by Aeschylus and parodied by Aristophanes; and last, but not least, metrical convenience. An iambic or pyrrhic word preceded by a good mouth-filling adjective rounds out an iambic trimeter very neatly. In the case of T/aoVo) the usage is glanced at by Fraccaroli de Euripidis scribendi artificio. Professor von Wilamowitz, in a note inserted in the second edition of his Herakles, says on 1. 283 : "t^ avajKai^ rpoirq) sagt nicht viel mehr als Ta> avayKai^, ry avdyKr). Med. 751, fiedijcreiv eKOvcrim rpov^, Hel. 1547, eK^aXovrei SaKpva troirjTO) rpoTrcp, fast gleich einem hcova-ioK koI TrpocnronjTW .... doch nicht ganz; die Weise in diesem 'verstellter Weise' wird noch als Substantiv empfunden." Without aiming at ex- haustiveness, I have collected enough examples to show that the usage deserves further consideration. For rpoirm compare further Find. Nem. 8. 14: evl aiiv rpoTrcj) = fM>va>a %ejt)i; Nem. 4. 55: iroKefiia %€pi; Isth. 7. 65: a(f>vKT(p ^e/Jt; O. 7. 51: apiaTOTr6voi,<; xe/ao't'; Aesch. Suppl. 958: SeBco/idraiiiat S' ovB' €7&) afiMpa ^epi, i. e., on a grand scale, fi^'ioXeimt; Soph. Ajax 230: ■trapairXd.KTtp %e/Ji; Antig. 54: avrovpyq) x^P^ (^f. avT0(f>6vaK, Aesch. Suppl. 65; Sepf. 734, avT0KT6va><; ; and avOemoK, Eust; Eurip. Here. 839, av6evT7} (^oV^); Orest. 947, avroxeipi .... cr(j>ayfj^; El. 1378, XiTrapel %€/3i glossed by XnrapoK Kal avvexSxi; Trach. 923, avvTovtp x^/si, i.e., o-wi/roi'ftis; Oed. T^r. 140, ToiavTT) x^/"') well rendered by Jebb "with a hand as fierce," but virtually = wo-avTtB? ; frag. 598. 3, aypia xe/at; Eurip. Medea 612 [Rhesus 772), a66va> .... x^/'A virtually = a(/)^o'i'(B? (cf. a. /tern, Aesch. J-gr. 305, and a. XoVo), Aesch. Suppl. 321) ; Heracleid. IQ'i, ^lalcp x^i'P^^''''po'i ^Cav=^ ^laCayi; Medea 12SQ, Sva-fievea-Tepa Xepi; Apoll. Rhod. 4. 1349 has Kovrj x^pt. Here also belong oXt; x^/"'' Paroem., Vol. II, p. 192, and' Arjixvia x^P^i ibid. I. 110.' Very frequent is irohl. Already in Homer we have Biep^ -rrohX <}>evyefiev, Odyss. 9. 43, and in II. 18. 599 eirKnaiievoiai iroheaai = iTTiaTafiepa'}. Cf. Pind. Nem. 3. 42, cnpeKel irohi; Theog. 815, Kpareptp irohi; Aesch. Persae 95, Kpanrvw ttoBi; Prom. 279, e\a iroSi. In Soph. Oed. Tyr. 479, fieXea fieXecp ttoS/ is practically KaKO'; xaKW, and in frag. 227 airoirXijKTQ) ttoBi differs little from irapairXi^KTQ) x^PS Ajax 230, or from an adverb. In El. 162 ev(f>povi ^Tjfiari is a mere variation. In Euripides it is very common. Cf. Here. Fur. 1040 = Hippoly. 1243 = Phoeniss. 149; Hec. 1050; Orest. 136, 456, 1017, 1290, 1622; Helen. 868-69, 1449, 1516; El. 549; Hel. 555; Hec. 1039, Xaifr^pS, TToSt; Bacchae 1230; Phoeniss. 303, 834; Rhesus 697, aSeifiavr^ iroSi, i. e., aSet/xai/ra)?, Aesch. Choeph. 771; frag. 979. 3, ^paBei TToSt; Alcest. 586, a KOv<^; fr. 123, ra'xet 7re8t\^; Fr. Adesp. Nauck, 227, XaiOdpym ttoBi; Aristoph. Thesm. 985, evpiiOfiip iroBi; Theoc. 2. 104, ttoSI Kovcfxp; Apoll. Rhod. 4. 1164, oXq) iroBl (cf. Suidas and Paroemiograph., Vol. II, p. 557.^ For other ' Latin parallels, as e. g., sacrilega manu or dextra, could doubtless be found in large numbers. 2 There are doubtless many similar uses of pede in Latin, e. g., CatuU. 63. 21, citato pede; 30, properante pede. A Geeek Analogue op the Romance Adverb 91 parts of the body, cf. Find. Pyth. 9. 41, ayava 6pvi; Nem. 6. 59, €k6vti .... WTO); Eurip. Orest. 1317, '^a-v^w .... ofifiart; Alcest. 598, vorepo) ^ecfxlpcp. In Soph. Oed. Tyr. 1371, 6nnaa-i,v 7roioi,<; ^eirmv is only slightly more "plastic" than the ttw? .... /SXeTTwi' of Philodet. 110, and the indifference of the noun is illustrated by Longus' (II. 22) substitution of feet: iroioii iroalv aTreifii irapa top Trarepa .... dvev XXo'?;?. In Find. Pyth. 11. 35, vea Ke^aXa is therefore possible for vea Kea\d. Another group is composed of words for life, death, doom, fate, fortune, etc.; Soph. Track. 168, aXiw^r?) /3t^; El. 650, d^Xa/Sel ^ia>; Crit. fr. 1. 17, ac^^tVo) . . . . ^ia>; Hom. Odyss. 5. 70, Odvov olKTurra Oavdrq); Aesch. Ag. 1493, dcre^ei Oavdrai; Semon. 1. 18, hvaTrjvo) fiopa; Aesch. Persae 484, SvaKXeeaTdro) fiopo); Cho. 296, TTUficjiOdpTO} fiopa; Ag. 1495, BoXim fiopm; Soph. Aj. 1059 and Neophr. fr. 3, ala-x^o'TO} p-opw; Track. 1042, wkv- Trera fiopq); Aesch. Ag. 1230, KaKy tvxv; 1341, da-ivel Bai/iovi; 1647, irpevfievel tvxv, 1165, SvaaXyel Tw%a; Eumen. 93, evnr6nira> TVXV, Soph. O. C. 1585, aTrww .... tvx27 (cf. 1561-63, eTrt- TToV^ .... fidpoy) ; Eurip. ^eL 412, dveXiriaTa tvxj) ; 1143, dveXiriffTai'i Tv;j;at?; Tro. 631, evTuxeaTepm iroTfio); Orest. 79, deofievel iroTixcp; Sept. 899, Sixoia-iJ,aTi,; Aesch. Pers. 394, ev\jrvx

06v(j> nevei; Cko. 445, aKdnirT({) fievei; Prom. 49, BevTepa Xoy^; 46, airXw \o7^ (cf. 610, 975); Suppl. 580, d^jrvdel Xoyq); 321, aj>66va> X6ya>; Prom. 505, ^paxel Be fivOcp (cf. 641); 67, iyapdrei ade'vei; Adespot. 295, dXKifKp a-devei; Soph. Jjaa; 350, o/a^w vo'/i?); Aesch. yl*;. 594, yvvatKel

ist, Fall und Wendung. 92 Paul Shoeey Again, in many cases when an adverb is in fact used, we can see that a dative phrase would have served as well. We have already noted the interchange of aSo'Xo)? and aB6\q) tf)pevi in the Attic Scolia; and of eiricrTafievco<; and eirurTafievoun {iroZeaffiv) in Homer. In II. 15. 510, we have avToaxehirj; in 746, avT0(T')(^eh6v, and in 17. 294, avToa-xeBirjv. In Odyss. 7. 241, and Aesch. Ag. SirjveKem ayopevaai [uKoveiv) suggests Sir)veK€t \6y^ (Plat. Hipp. Maj. 301E), which does not happen to occur in tragedy. In Semon. 7. 107, irpoippovw^ differs little from 7rp6<})povi OvpM> in Hes. Theog. 586, or 7rp6povi KapZCa, Aesch. Suppl. 348. <^pevo. fi6pa)<}, Soph. Aj. 626 = ^pevrnXi]';, Aesch. Sept. 757 = (^pevonavri<:, Ag. 1140 = Xvo'o-aSt fioipa, Eurip. Here. 1024==^ p-aivo/Meva ^pevC, Sept. 484 = the Homeric <^pea\ fiaivofievrjcriv = fiaivofieva icpahCa, Medea 432, etc. In Prom. 730, OpaavairXdyxvo^ recalls Pindar's Bpaffeia .... Kaphia, Pyth. 10. 44. In Persae 374, oiiK aicoaixax: aXKci TreiOdpxo) (^pevi, adverb and phrase are, as often, co-ordinated. In Soph. 0. C. 490, aaTpopevo<;, 1302. And, not to multiply examples, other adverbs that suggest similar equivalents are ZvaeKXirayj, Prom. 60 ; a>p,opafffi6vQK, Persae 417; ■n-ayKpoTCK, Aesch. Suppl. 723; fiapva-T6v(o<:, Eumen. 794; aicrjhiffTCK, II. 24. 417; hvffXo^ax;, Eurip. Tro. 303; €vX6o>pahir,ai,; 15. 186, pir,. In Tyrt. 8. 13, ^t;jitw = bravely; in Aesch. Prom. 74, ^ia = ^iaia>aXeia, Oed. Tyr. 51 — aa, Persae 360, than Kpv^aiuK; eKovn vwrm, Find. Mem. 6. 59, than eicova-iax;; doXepm 'x^eifi&vi, Ajax 206, than 6o\e- pm; adeqi tto^I, Eumen. 541, than a^eo)?; Bvcr(f>dTq> xXdyya, Ag. 1152, than Bva-<;; Oavaa-i/j-m 'xeLpmp.an, Oed. Tyr. 560, than BavacrifMO^; (fvi') acraBda-Tq) ical Ta;^et irriBrifiaTi, Ajax 833, than acr^ahdcTTOD^ xal ra^^eiw? ; ajava .... 6vja [Sept. 953). The chief practical outcome, apart from two or three possible emendations,' is that we should beware of overinterpreting the noun in such formulas. In Soph. Track. 293, 294: TTWS 8' ovK iyto ^aipoifi' av avSpoi tvrv)^ KX.vov(ra Ttpaiiv riJvSe TravSiKia pari 'In Eurip. fr. 285. 7, 8 N., for iX/Sou Siolyav 0i\afu>v ^Surrov %«/>£ we might plau- sibly read iiSlffry xfpL In Hero. Fur. 1003, the verba desperafa iri \6poveiv (At. Nic. Eth. 10. 8), where the adverbial accusative Si\vaosi = dv6pmTriv(o'i (despite its slightly different con- notation). The absolute use of ioiKora is, of course, no difficulty. In Pind. Pyth. 9. 33, where the nymph Cyrene arap^el velKo<: A Greek Analogue of the Komanoe Advebb 95 ayei Ke^aKa, Professor Gildersleeve bids us "note the serenity of the heads of the combatants" in Greek art, and the plastic value of KCipaXa might be further illustrated by the lines in Keats' Lamia. Charioting foremost in the envious race Like a young Jove vnth calm uneager face. But aSeifidpTcp "ttoSi and arap^el (ppevi, Pyth. 9. 51, tend to reduce even this to formula. The much-debated %^ovta pept meant ' with such imperfect consciousness as the dead pos- sess,' then Pindar will be speaking like the Homeric Achilles. .... It is more in accord with Pindar's manner to regard xOovia as conveying a shadowy suggestion that the intelligence which belongs to the unseen world is of a different order from the intelligence of the living." This is perhaps considering the matter too curiously. Such suggestions are inevitable for us and may possibly have occurred to Pindar. But I think Professor Gildersleeve is nearer the truth when he simply glosses x^<""'? by inro x^ov6<;. Pindar is not concerned with either psychology or eschatology here. He is merely rounding out a line with a convenient formula. The "psychology" is that of Browning's Bishop: And then how I shall lie through centuries, And hear the blessed mutter of the mass, And see God made and eaten all day long. And feel the steady candle flame, and taste Good, strong, thick, stupefying incense smoke. As Lucretius says (3. 882): .... ilium se fingit sensuque suo contaminat astans. It would be easy to carry the idea of this study too far. Some of the cases cited are perhaps not strictly speaking adver- bial or modal. But if allowance is made for the wider range 96 Paul Shoeey of the Greek adverb in the expression of attendant circumstance and in proleptic or pregnant uses,' this objection will often dis- appear. In Aesch. Prom. 861, WKTi(f>povpijTa) Opda-ec is hardly more than nuitamment.'^ Tipovri .... ttoS/ (^Orest. 456) and yr]paim "TroSi [Phoeniss. 303) are practically yepovriKm. So Kadapaicp TroSt, Soph. Antig. 1142, and X"P* Traicovta, Aesch. Suppl. 1067, might be illustrated by proleptic adverbs. And in this way the formula might be stretched even to take in such cases as Xeva-t/xcp %e/3t, Orest. 863; Tv\^ iroSi, Phoeniss. 834; •jraWevKm ttoSi, Medea 1165; iKea-ia xept, Eurip. Suppl. 108; ^uKxeifp TToSi, Bacch. 1230; Kap^dva %€pt, Ag. 1061. In Aesch. Suppl. 606, yrjpaia (ppevi is not adverbial to our feeling. But neither is TratSt/cw? in Plato Lysis 211 A, which Liddell and Scott, misled by Cratyl. 406 C, accordingly misinterpret, and which Jowett is obliged to paraphrase, "in a childish (and affectionate) manner." A Greek poet might have turned it by iraihiKy (f>p€pi. But such considerations exceed the scope of this paper, which is not the maintenance of a thesis but the illustra- tion of a somewhat neglected phraseological habit of the Greek language. The University op Chicago > E. g. iKi)S(