This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project to make the world’s books discoverable online.
It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books are our gateways to the past, representing a wealth of history, culture and knowledge that’s often difficult to discover.
Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book’s long journey from the publisher to a library and finally to you.
Usage guidelines
Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to prevent abuse by commercial parties, including placing technical restrictions on automated querying.
We also ask that you:
+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for personal, non-commercial purposes.
+ Refrain from automated querying Do not send automated queries of any sort to Google’s system: If you are conducting research on machine translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the use of public domain materials for these purposes and may be able to help.
+ Maintain attribution The Google “watermark” you see on each file is essential for informing people about this project and helping them find additional materials through Google Book Search. Please do not remove it.
+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other countries. Whether a book is still in copyright varies from country to country, and we can’t offer guidance on whether any specific use of any specific book is allowed. Please do not assume that a book’s appearance in Google Book Search means it can be used in any manner anywhere in the world. Copyright infringement liability can be quite severe.
About Google Book Search
Google’s mission is to organize the world’s information and to make it universally accessible and useful. Google Book Search helps readers discover the world’s books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web ai[http: //books . google. com/|
LS — OA A ee —
SS
HARVARD STUDIES
CLASSICAL PHILOLOGY
EDITED BY A COMMITTEE OF THE CLASSICAL INSTRUCTORS OF HARVARD UNIVERSITY
VoLuME VI
BOSTON, U.S.A. PUBLISHED BY GINN & COMPANY
LONDON: GINN & COMPANY LEIPSIC: OTTO HARRASSOWITZ 57 & s9 Lupcatr Hii Quer STRASSE 14
1895
PREFATORY NOTE.
THESE Studies are published by authority of Harvard University, and are contributed chiefly by its instructors and graduates, although contributions from other sources are not excluded. The publication is supported by a fund of $6000, generously subscribed by the class of 1856.
JAMES B. GREENOUGH, FREDERIC D. ALLEN, EDITORIAL JOHN WILLIAMS WHITE, |
COMMITTEE.
CONTENTS.
THE OPISTHODOMUS ON THE ACROPOLIS AT ATHENS... . I By John Williams White.
ARTEMIS ANAITIS AND MEN Tramvu, A VOTIVE TABLET IN THE BostoX MUSEUM OF FINE ARTS ... 1... e «© @©~)6 55 By John Henry Wright.
THE DATE OF LYCOPHRON. ... . . . © 1 ew te ew eS By William N. Bates.
Quo Movo [ACIENDI VERBI COMPOSITA IN PRAESENTIBUS TEMPORIBUS ENUNTIAVERINT ANTIQUI ET SCRIPSERINT . 83 Quaerit Mauricius W. Mather.
HOMERIC QUOTATIONS IN PLATO AND ARISTOTLE .. . . 153 By George Edwin Howes.
INDEXES. . 2. 2 6 ee ee ee ew ee we ew 239
THE OPISTHODOMUS ON THE ACROPOLIS AT ATHENS.!
By JOHN WILLIAMS WHITE.
N inscriptions of the fifth century and fourth century B.c. and in
Aristophanes, Demosthenes, and Lucian, references occur to
a structure on the Acropolis at Athens which is called simply
6 émicOodopuos, without further designation. The scholiasts, however,
on the passages in which the Opisthodomus is thus referred to and the ancient lexicographers define its situation.
August Boeckh believed that the Opisthodomus in question was the western chamber of the cella of the Parthenon, and maintained this view with vigour.?, So Leake,’? K. F. Hermann,‘ Boetticher,® Michaelis,® and many others. This has been and remains the gen- erally accepted view. It makes the “Parthenon” in the restricted sense,——the well-known treasure-chamber named in inscriptions,’
1 The following discussion was first made public in a lecture delivered at the American School of Classical Studies at Athens on March 1, 1894. It was eub- sequently privately printed in London, in a limited number of copies, under date of May 5, 1894. Contemporaneously, Professor Arthur Milchhofer of Miinster pub- lished in the PAslologus (Heft 2, 1894) a searching discussion of the same theme, in which he independently established the main contention of my paper, namely, that the Opisthodomus on the Acropolis at Athens was a separate building. On the secondary question of the situation of the Opisthodomus we held and hold divergent views. My reasons for regarding Professor Milchhofer’s view on this question untenable constitute a part of the present revision of my original argument.
2 Staatshaushaltung der Athener, 1886, 1. p. 517 ff., especially p. 519, note c. See also C/G. I. p.177 f.
3 Topography of Athens?, 1841, I. p. §59.-
4 Die Hypathraltempel des Alterthums, 1844, p. 27 f.
5 Philologus, 1862, xvil., plan; Untersuchungen auf der Akropolis, 1863, p. 165 ff.
8 Der Parthenon, 1871, p. 26f. See also p. 109.
7 See the important series, C/A. 1. 161 ff.
2 John Williams White.
—a part of the vews éxarduredos, and places within it the great chryselephantine statue of Athena. Ussing believed that the western chamber of the cella was the “Parthenon,” that the western portico was the Opisthodomus proper, and that the two together constituted the Opisthodomus of the inscriptions.' This is also Petersen’s view.? Kohler maintains that the statue stood in the Hecatompedos, not in the “ Parthenon,” but refuses to identify the western chamber of the cella, which he believes to have been the Opisthodomus, with the ‘‘ Parthenon.” *® Lolling also believed this to be the Opisthodomus, although he held new and revolutionary views in regard to the appli- cation of the terms “ Parthenon” and Hecatompedos.‘ Dorpfeld, on grounds independent of those on which Ussing had based his argument, concluded that the western chamber of the cella was the ‘‘ Parthenon,” and that, in official language, Opisthodomus always meant the western portico of the temple.° This view was adopted by Frankel,®° and is held by Frazer.? On his discovery of the. Hecatompedon,® Dorpfeld relinquished this view, and now main-
1 De Parthenone cjusque partibus disputatio, 1849. Also Griechische Reisen und Studien, 1857, p. 145 ff-
2 Mitth. d. Inst. Athen, 1887, x11. p. 70f. Petersen was the first to endeavour to establish the proposition that the same room might have in contemporary documents two official names, namely, ‘“‘ Parthenon ” and Opisthodomus. In the first edition of his Aleisterwerke der griechischen Plastik (1893, p-177) Furtwangler supported the same proposition, and declared for a complete identification of “Parthenon” and Opisthodomus. He has now relinquished this view, and believes that the Opisthodomus was a separate building. See his Masterpieces of Greek Sculpture, the English edition of his Mersterwerke, edited by Eugénie Sellers, London, 1895, p. 425 f.
8 Afitth. d. Inst. Athen, 1880, V. p. 89 ff., especially p. 100.
4 'A@nwG, 1890, II. p. 627 ff.
& Afitth. d. Inst. Athen, 1881, Vi. p. 283 ff.
6 In Boeckh, Staatshaushaltung>, 1886, 1. p. 106*, note 729.
7 Journal of Hellenic Studies, 1892-93, XU. p. 153 ff.
® The temple of Athena whose foundations lie close to the Erechtheum on the south. For Dérpfeld’s description of it as a structure, see Afitth. d. /nst. Athen, 1886, XI. p. 337 ff. See also Mstth. 1885, X. p. 275 ff. and Antshe Denkmdler, 1886, plates I., II. Dérpfeld himself names it “alter Athena-Tempel,” but this name seems to be misleading to those who do not believe that it was the oldest temple of Athena on the Acropolis. Petersen calls it “ Peisistratischer Tempel” (note 2 above), Frazer “ Pre-Persian Temple” (note 7 above). Dérpfeld main-
The Opisthodomus. 3
tains that the term Opisthodomus in the inscriptions and authors designates the three rooms constituting the western half of this temple, which, as is well known, he believes to have been still in existence in the time of Pausanias.' Finally, Fowler has advanced the original hypothesis that the Opisthodomus was the western chamber of the cella of the “‘ Parthenon,” that this room was doubt- less divided into three parts by two partitions of some sort, probably of metal, running from the eastern and western walls to the nearest columns and connecting the columns, and that the middle division of the three between the partitions was the “ Parthenon ” proper.?
I am unable to accept any one of these views, and venture again to present for consideration our sources of information about the vexed structure called the Opisthodomus. I purpose to discuss as the main thesis of this paper the following proposition :
The Opisthodomus on the Acropolis at Athens, referred to in inscriptions and in authors simply as 6 dxvOodop0s, was not a part of any existing temple, but was a separate building, complete in itself.
The current view, if I may so name it, would seem to be expressly contradicted by the testimony of the scholiasts and lexicographers. An important part of this testimony, with the original passages in the authors of which it is an explanation, is the following :
olov (oruws d¢ py OopvByoa por pydets, mpiv dv dravta cirw) dvépgay Syrou mpwnv tives Tov dmicOddonov. [Dem.] XIII. 14.
Anpoobévns ev tp mepi cvvragews: dvéptay Syrov mpwyv tives Tov émirGodopov. 6 olxos 6 dmiabev Tov vew THs "AOnvas ovTw Kadrcirat, év @ dreriPevro ra xpyjpata. Harpocration s.v. dmobddopos.
tains that the name which he has assigned to it is correct (Afitth. 1892, XVII. p. 158, note 1). Furtwangler also calls it the “Old Temple” (note 2 above). The official name, which will be used in this paper, is rd ‘Exaréyredor. See C/A. IV. p. 137 ff.; Aedrloy ‘Apx. 1890, p. 92 ff.; "AOn va, 1890, 11. p. 627 ff.
1 Mitth. d. Inst. Athen, 1887, Xu. p. 25 ff., 190 ff., 1890, XV. p. 420 ff. Dorpfeld’s view, both that the, Opisthodomus was in the Hecatompedon and that the latter was still in existence in the time of Pausanias, is accepted by Miss Harrison, Mythology and Monuments of Ancient Athens, 1893, pp. §05 ff., and by Miller, Amer. Jour. Arch. 1893, VU. pp. 500 and 528.
2 Amer. Jour. Arch. 1893, Vl. p. 10 ff.
4 John Williams White.
olkos GriaOev rod vew ris "AOnvas, év @ dwrerievro ra xpypara. F dy Ty axporoka roros évOa ro Sypociov dpyvptov dréxatro Kal 6 ddpos. Schol. FY [Dem.] xu. 14.?
Pépos Te THS axporoAews Tov AOnvaiwy, évba Fv To Tapteiov, Omobev tov tis A@nvas vaov, év @ dreriBevro Tra xpypata. Anpoobevys év re
mept ovvrdfews* dvéwkay Syrov mpdyv tives rov dricOddopov. Suidas $.v. dmvrOodopos."
iSpvadueD” odv adrixa par’, dAAa mepipeve,
rov IIAovrov, ovrep mporepov fv idpipevos
rov dmicOodopov det hvAdrrwy ris Jeov. Arist. Plut. 1191-1193.
érady Ta xpypara év ro dricOodopw dréxaro. pépos S€é €ore ris dxporroAews, vba éori rapseiov, Oricbev rod THs "APnvas vaod. Schol. R Arist. Plut, 1193.
GkAws. dricw rod vew THs KaAoupevys todiddos "AOnvas SiurA0vs rotxos (olkos ?) éxwv Oupay, Grov jv OyoavpopvAdxioy. érei ra ypypata ev Te omicGodopw aréxaro. pépos 5& ris dxporroAews. Schol. V Arist. Plut. 1193.°
omiaGev Tov lepod THs ‘AOnvas ro tov "AOnvaiwy Fv OycavpodvAaxioy. Schol. LB Arist. P/ut. 1191.
. Tapseiov THs moAews ev dxpordAa: Smobev rod ris “AOnvas ved. Photius 5.2. dmcrOodopos.
kat of rapiac ef av & dricOodopos éverpyoOn, Kai of ray ris Beod Kai
ot rav dAXwv Gedy, év rq olxjpare rourp hoav ews 9 Kpiots avrois éyévero. Dem. xxIv. 136.
loréov Gtt xpypata Kat Trav dAAwy iepov rav Geav Kat ris “AOnvas cxecro éy tive oixypart dricw THS axpoToAEws, TH KaAovpevy dricbodopy,
1 The statement, é¢v 77... dépos, is also in B.
2 dricOddop0s pépos THs dxpowbA\ews... xphyara. Schol. V Lucian Fug. 7. The Scholiast did not observe that Lucian was talking about an opisthodomus at Olympia.
8 See Diibner’s note (Schol. Graeca in Arist. 1841, p. 613), who also gives the variant in Par. 2821.—olkxos is the conjecture of Michaelis (Paus. descrip. arc. Athen. 1880, p. V1. add. ad c. 24, 47 &.7 a fine), and must be right. He compares Paus. 1. 26. 5, derAody ydp dors 7d olxnyua.
The Opisthodomus. 5
cai fody tives reraypévot Tapia eri ty pvAaxg Tovrwy. Schol. ATCV Dem. xxiv. 136.
év ry dxporoAa, ov Td Snpdciov dpyvpiov dréxaro [pos TE driaOoddpu | xai 6 @dpos. Hesychius s.v. drveOodopos (sic).?
The meaning of the interpreters here seems to be clear. Their testimony is that the Opisthodomus was a house, or a place on the Acropolis, or a part of the Acropolis, that lay behind the temple of Athena, and that it was used as a treasury.
These old Greek interpreters have been variously dealt with by modern writers who have discussed the Opisthodomus.’ By the most of the scholars named above they have been silently ignored, for whatever reason; by others they have been taken seriously‘; by others still their testimony has been rejected as worthless.’ In some instances it is impossible to tell by what interpretation of the Greek scholiasts and lexicographers some of the moderns arrive at the conclusion embodied in the current view.°
If in the interpreters as quoted above the words veds and ltepov mean ‘emple, it is possible to obtain the definition of Opisthodomus adopted in the current view only by attaching to émo6e the meaning
1 The Scholiast’s meaning, probably, is treasure that belonged both to the sanc- tuaries of the other gods and to Athena's. dxlaw rijs dxporéd\ews is as it stands nonsense, since it removes the Opisthodomus from the Acropolis altogether. Compare the definition in Pollux (IX. 40), 7d xarémcy rijs dxporébdews (sc. dv efracs) dria ObS80p0r.
2 Cf the schol. [Dem.] x11. 14 quoted above.
§ See, as to the credibility of this ancient testimony, p. 41 ff.
* By Boeckh, for example (C/G. I. p. 177 f.), whose interpretation is discussed Pp. 43 f.
6 By Michaelis, whose treatment of the evidence is considered on p. 12.
6 Leake, for example, cites the scholiasts and lexicographers, but gives no explanation how from their statements he reaches the conclusion that the Opistho- domus was the western apartment of the cella of the Parthenon. Doérpfeld also, who believes that the Opisthodomus was the rear part of the Hecatompedon, twice in interpreting the scholiasts and lexicographers translates the phrase SricGer Tov rijs "AOnvas vew, ‘hinter dem Tempel,’ ‘hinter dem Athena-Tempel.’ See Mitth. d. Inst. Athen, 1887, X11. pp. 34, 39. This is, I think, the right inter- pretation ; but it is difficult to see how, if the Opisthodomus lay ‘behind the temple,’ it was at the same time a part of it
6 John Williams White.
in the back part of. olxos Grobev rou ris "AGyvas vew would then mean a room in the back part of the temple of Athena. But this meaning of omo@e with the genitive cannot be established. It is recognized by none of the lexicographers. In order to express the desired meaning omo$e must be combined adjectively with the article; the genitive that follows is then partitive. Pausanias, for example. in telling where certain paintings are in the temple of Messene, daughter of Triopas, says (IV. 31. 11), ypapai 5@ xara rov vaov To omoGey of Bacttevoayrés ear Meconyys, i.c. in posta templi parte. |
It may be well to establish the uses of éricMe in this author, who naturally had occasion to use the word often. In Pausanias omobey may be used, as above, adjectively.'. Sometimes it is used adver- bially.? In the great majority of the instances of its occurrence, it is followed, as an adverb of place, by the genitive. In the most of these it clearly means échind; in some cases the meaning is indeterminable, because the statement is brief, and we have no other means of arriving at the facts; in no instance can it be proved that the word means 7a the back part of.
In the following cases omo@ey signifies, in my judgment, dchind : Kai opas trepeBadovro ‘A@Pyvaios tov xoAocoov dvaberres OricGe Tow vaov. 1. 18.6. Pausanias has just said that the peribolus of the temple (the Olympieum at Athens) is full of statues of Hadrian; but the Athenian colossus overtopped all the rest. The statue had a commanding position, facing the Acropolis. Cf vu. 9. 6; 30. 7; 30.8. ore 88 oriobey rot Avxetou Nicov pynpa. 1.19.4. Cf. I. 31. 3; ur. 16.6; VIII. rg. 10. pera ratta és ro rou Atos répevos éceAPovcr vaos dori Geas afios: Td 8¢ dyaApa ovx dLapydoby row Aros . . . oriabe 52 Tov waov Keira: fvAa nulepya... év d¢ aire Te vag Tpinpous dydxaras xarAxovv euBodrov. 1. 40. 4,5. The phrase é& 8 aire re vee makes it certain that the €vAa nycepya were not in the temple. ryyv 5 wmyy7y, 9 tori OmoGey Tov vaov. II. 5. 1. rovrov 5 (i.¢. rov Gearpov) oxiobey @xodopnra: oradiov xXevpa pia, dv€xyovod re airy Té Oearpov xai dyti épeioparos dvaXoyov éxeive xpopery., 11. 29. 11. ore S& THs oroas
11. 24. 5; I. 20.7; V. 10.8; VI. 5.6; VIL 45. 7; X. 1G 4. 23. 3. 3; V. 20. 2; VI. § 6; x. 26. 5.
The Opisthodomus. 7
ratrys GricGev yoga. I. 15.1. Cf 1.1. 3. SmiaGev 82 ris KaAxtoixov vaos ori “Adpodirys. I. 17.5. Kai OmwBev yuvy abrov. V. 17. 9. Cf. v.19. 6. eornxe 8& (4.¢. KAcooBévovs dppa) oricbev rov Ards rod dvareBéyros iro “EAAnvwy. VI. 10.6. Cf. X. 9. 9. tropévovres ris rdgews Gxiobey of oixérar rooade odio eyivovro xpnowmor. X. 1g. 10. The following are indeterminable, but that in them omofev means behind can hardly be doubted in view of the preceding clear instances of this meaning: II. 11.13 11. 13.7; V.15.73 VII. 22.7. In Pausanias érigw! is almost always an adverb accompanying a verb of motion ; the following example shows its meaning when followed by the genitive: éore & ris arods dricw ‘Adpodirys vads. VI. 25.2. It should finally be noted, as important in establishing the meaning of omoGey and driow with the genitive of place, that the counter-idea is generally expressed by zpo with the genitive, where by no contrivance can the preposition signify 72 the front part of.
In view of these facts, it is impossible to interpret omoev rov ris "AOnvas vew to mean in the back part of the temple of Athena.
But, as is well known, veds may signify c//a, as well as ‘emple, although, when the entire number of cases of the word’s occurrence is taken into account, this is very rare both in the literature and in inscriptions. If this signification of the word could be established for the phrase olxos dmuOev tov ris AOnvas ved, we should arrive at the meaning. demanded by the current view as to the situation of the Opisthodomus.
The question is limited to the use of veds in its actual application to temples of Athena on the Acropolis at Athens.? Fortunately
1 This form occurs in the schol. V Arist. P/ut. 1193 and in the schol. Dem. XXIV. 136 quoted above.
2 The consideration of the etymology of the word wpévacs is pertinent, but does not establish the meaning ce//a for vabs. That which rpévacs names doubt- less marked the first stage of development of the original vaés from a single room to a more complicated structure; at this time wpévaos meant (porch) before the temple. vabs and ddvrow were then identical. The original vaés kept its name when, with the permanent addition of the porch in front and the porch behind, that name received a larger application and designated (as it had exclusively at first) the entire structure. Cf Sduos and wpdbdopos (éy wpodduyp Sduov, Od. IV. 302), and the theatre-terms oxnr} and rpocxj»swov. The original etymological force of wpbraos is seen in its adjectival use, which is not uncommon. C/. Bwyods rpovdous,
8 John Williams White.
the successive labours of scholars have collected the existing literary and epigraphical evidence not only for the word veds but also for the other terms designating these temples and their parts,' and it is now not difficult to reach trustworthy conclusions in regard to their use and application. The law of use for vews and iepdv* requires that, when they have the limited sense ce//a, this shall always be clearly indicated etther by an added epithet or by the context. Such instances are surprisingly rare. I proceed to an examination of the evidence.
We meet first cases in which the old temple of Athena® is referred to as dpxatios or raAads vews,* where the epithet excludes the meaning cella for vews. Xenophon records that the old temple was set on fire ; Strabo contrasts it with the Parthenon. It would be as forced to suppose that veds means ce//a in any of these instances as in the scholium on Arist. Pax 605, the source of which is Philochorus, «at TO dyaApa TO xpvoovv ras ’"AOyvas éoraby eis Tov vewv Tov péyay, in which it is important to note the epithet. Here 6 péyas vees is the Parthenon, and yet we know that the statue was in the ced/a.
The meaning ce/a for vews is excluded also in the inscriptions that relate to the building of the old temple of Athena and to its restoration after the burning described by Xenophon; also in those referring to the setting up of stelae zapa rov vew and to the approach of the panathenaic ship.*
Aesch. Suppl. 494; THs povalns ‘AOnvalns, Hdt. vitl. 37; ‘A@ned xal ‘Epps, dvopafbueror IIpévaccr, Paus. 1X. 10. 2. Cf. Harp. s.v. wpovala: dd 7d xpd rod vaoi l8pbcGa:.
1 See Michaelis, Parthenon, 1871, pp. 285-317; Jahn-Michaelis, Puss. descrép. arc. Athen. 1880; Milchhofer in Curtius, Stadtgeschichte von Athen, 1891, pp- XX.-XXII.
3 lepdy occurs above, in the schol. Arist. P/ut. 1191. It occurs also in E. M. 5.u. Ori bbdopos.
8 In this paper “old temple of Athena” means the Erechtheum or its prede- cessor on the same site.
‘C/A. Iv. 1 C, 27 (p. 3f.); 1. 93,6; 1.748, 14; 163,9; 464, 6; 672, 43; 733 A, col. 11.6; Xen. Ae//. 1.6.1; schol. Arist. Lys. 273; Strabo 1x. p. 396. CIA. 11. 751 B, d, 19, and 758 A, col. 11. 8, do not belong here. See Lehner, Ueber die athenischen Schatsverseichnisse, 1890, p. 79.
6 C/A. 1. 60, 33 322, 1, 4, 8; IV. 321, col. III. 27 (p. 74 ff.); 11. 332, 44; 829, 35 Mitth. d. Inst. Athen, 1883, VU. p. §9, 25; C/A. Ill. 776, 3
The Optsthodomus. 9
There are other instances in which veds certainly means femp/e, although it is not easy to categorize them. Some of these refer to the Parthenon,! others to the old temple of Athena.* In some of them, although the object referred to was probably in a specific part of the temple, it is still clear that the temple as a whole was in the speaker’s mind when he used the word vews.? These are of the same nature as the quotation from Philochorus above (schol. Arist. Pax 605).*
There are three passages in dispute.® If in these 6 ris Modrados vews means the cella of the Polias, it should be observed that this results solely from the demands of the context,® for in the majority of
1 Aristot. Asst. an. VI. 24, p. 577 B, 29 (cf Aelian de mat. an. VI. 49); Philo- chorus in schol. Arist. Pax 605; Plut. de soll. an. 13, p. 970 B; Paus. I. 24. § and 8; Arist. 1. p. 548, 14 Dind.; schol. Dem. xx1I. 13; Hesych. s.v. ‘Exarévwedos; Lex. Pat. in Bull. Corr. Hell. 1877, 1. p. 149, ‘Exaréuredov; E. M. 5.v. ‘Exarduredov (cf Bekk. Anec. 1. p. 283, 15); Suidas s.v. ‘Exaréuwedos veds.
2 Hom. //. 1. 549; C/G. 6280 A, 31 = Kaibel Zp. Gr. 1046, 90; Plut. guaest. con. 1X. 6, p.741 B; Paus. I. 27. 2 and 4; Himer. Ze/. v. 30; Clem. Alex. Protr. III. p. 13 Sylb. (cf Apollod. 111. 14. 7); schol. Arist. £g. 1169; Eust. //. xxi. 451 (cf Hesych. s.v. Al8ots Bwydbs); Eust. Od. 1. 357; schol. Arist. Zys.759. The last two relate to the sacred snake, which was in the old temple. Hesychius (s.v. olxovpdy Edi») alone names an exact spot, and places the snake where we should expect to find it, év r@ lep@ ro6 'Epex6éws. This must interpret for us the onxés of Plutarch (Z7hem.10). In Dion. Hal. Ant. xiv. 2 similarly we find é» rou yryerois 'Epexbéws ry onxy, where he is speaking of the olive. See, for the snake, the passages in Jahn-Michaelis, aus. descr. arc. Athen. 1880, p. 27 (c. 27, 7®).
® Plut. Cim. 5; Anth. Pal. vi. 2 (Simonides); Xen. He//, 11. 3. 20. — If Paus. I. 24. 3 is to be taken into account, it belongs in this general category.
‘To this category belongs also Hdt. viii. 55, if one believes that there "Epex 0éos ynés refers to the whole building, as in the prevailing modern use of the word “Erechtheum.” If one does not believe this, but that the ofknyua 'Epéx Ger Kadotyeror Of Pausanias (I. 26. 5) is the double western half of the old temple, then the expression in Herodotus is not to be taken into account, since the present inquiry is limited to the investigation of the meaning of ved: and lepdp in their application to temples of Athena.
© Philochorus frg. 146 (Dion. Hal. de Din. 3); Paus. 1. 27. 1 and 3
6 See Michaelis’s discussion of the passages, Mitth. d. Inst. Athen, 1877, 1. p- 3! ff., with notes 25 and 27. His negative argument by which in Paus. 1. 27. 3 6 vews ris Todcddos is made to refer to the ce//a of the Polias has not met with acceptance.
10 John Williams White.
the instances of the occurrence of the phrase it is generally agreed that the reference is to the fempie of Athena. The argument from the context, in fact, led to this interpretation of these three passages.
There are two cases in which vews means ce//a. In the first of these* the inscription names the whole building, the Hecatompedon, and then its parts, 6 vews, rd xpormoy, and ra oixnpara ra ey re éxa- tourddy. Here the signification of vews is made clear by its colloca- tion with rpovjov. The second case is the well-known use of the word in the treasury-documents,® where it always has an epithet, 6 vews 6 éxarduredos. |
The facts for the use of iepdy are altogether similar. In some instances it signifies either sacred precinct or temple, generally the latter; in two it refers to the Parthenon.’ In none of these has it a more limited meaning than fempie.°
In a single case it means ce//a.'_ But here, as in the two cases of vews mentioned above, its meaning is made clear by the context, since it is interpreted by the preceding advrov.
Here, then, we have abundant instances, both literary and epigraphical, of the uses of vews and iepdv, extending from the earliest to the latest times, and among them all only three cases in which they certainly signify ce//a. In each of these three cases, moreover, the context or an added epithet makes clear that this is the signification. In the passages from the scholiasts and lexicographers, on the contrary, that are quoted above, no limitation whatever of the meaning of vews and iepov is indicated. And yet the especial purpose of these interpreters was to give a definition ; nor were they
1 Strabo 1x. p. 396; C/A. U1. 332, 44; Afitth. d. Inst. Athen, 1883, VIII. p. 59 25; Himer. £c/. v. 30; Clem. Alex. Profr. 111. 45, p. 13 Sylb.; schol. Arist. Zg. 1169; Eust. //. xx. 451 and Od. 1. 356.
2 C/A. Iv. p. 137 ff.
8 C/A. 1. 146, 157, 158, 159; U. 652 A, 15.
4 Hdt. v. 90; VIII. 41, 51 (425), 53, 54, 55 (#5); Thuc. 1. 126; Phot. s.v. raplac (quoting Aristot., frg. 4oz Rose, Bekk. Anec. p. 306, 7, and note Aristot. Resp. Ath. c. 47,1); Dion. Hal. Ant. xtv. 2; Hesych. s.v. Al8ots Bwyds; schol. Aesch. Il. 147.
§ Ps. Dicaearchus frg. 1, 1; schol. Pat. Dem. xx. 13 (Budi. Corr. Hell. 1877, I. p. 13)-
6 See Hesych. s.v. olxoupd» dg», and p. 9, note 2, above.
7 Hdt. v. 72.
The Opisthodomus. II
ignorant of the fact, had vews and iepoy seemed to them to be liable to misinterpretation, that the unmistakable aévrov, an Homeric word, and oyxos were ready to their hand.!
If, nevertheless, we seek to attach to vews and iepoy in these passages the restricted sense of ce//a, we encounter an unexpected difficulty. Schol. V Arist. P/ut, 1193 says that the Opisthodomus lay behind the vews of Athena Polias. Those, therefore, who hold the current view in regard to the situation of the Opisthodomus must either establish the worship of Athena Polias in the Parthenon or Hecatompedon, or reject the evidence. If the evidence is trust- worthy and if the term Polias designates, as is commonly believed, Athena of the Erechtheum or of the temple that preceded it on the same site, then we are forced, on the supposition that vews here has the restricted sense, to the conclusion that the Opisthodomus lay in the Erechtheum. But this is impossible. Boeckh saw these diffi- culties, and felt himself forced to declare that the scholiast had blundered,? although he himself accepted and in part sought to explain the remaining testimony of the scholiasts and lexicographers.
That in the phrase olxos or oixnpa drioOev Tov rHs ’AOnvas ved the words olxos and otxyjpa may signify Aouse will at once be granted. This is the first and common meaning of ofxos and is perfectly estab- lished for oixnpa, and if demanded by other considerations, namely, the use of émiofev and veds just discussed, it must be allowed. This use of olxos to denote a separate structure that was a treasury receives striking confirmation from the names officially recorded of four of the treasuries and magazines at Delos, *Avdptwy olxos, Nagiwv olxos, AnXiwv olxos, and Iwpivos ofkos.? In charge of the anathemata and materials stored in these were the leporoot, whose functions corre- sponded closely to those of the rapéac rs Geod at Athens.
The same word is used by Hesychius in defining @ycavpds, namely: cis dyadparwv Kai xpypatov [4] lepav droGecty olkos.
If the preceding discussion of the terms dmoGev, vews, and olxos is sound, we must either agree that the Opisthodomus was neither in
1 Hesych., Suidas, E.M.,svv.; Bekk. Anec. 1. p. 345, 25.
2 Staatshaushaltung?, 1886, 1. p. 517 f.
8 Bull. Corr. Hell. 1882, Vi. pp. 48, 87, 88, 91, 100, 135; 1884, VIII. p. 322; 1890, XIV. p. 509, note 3; 1891, XV. p. 141.
42 John Williams White.
the Parthenon nor in the Hecatompedon, but was a separate build- ing, or elue reject the testimony of the scholiasts and lexicographers as ty its situation. Michaelis does reject their testimony, declaring their explanation of the name for the most part worthless.' He makes an exception in favour of Harpocration, but the reasons for this are not apparent. In Harpocration’s definition, 6 olxos 6 dmwbev vu vew rhe ‘AOnvas otrw xadeira, one might be tempted to construe L bwurGey oluog rot veo, and render the back chamber of the temple; but this construction is excluded by the phraseology of the Epitome, dxuddiry b4 ourug & SmirOev rod vew ris "AOnvas olxos, where the genitive gob ves cannot be partitive. Michaelis’s rejection of the evidence acess to justify the conclusion that he did not believe it possible to interpret SrcrGey and vews in the manner demanded by the current view,
The testimony of these later writers receives unexpected con- fismation from an early and important inscription :
7 jo 8 “cepa dpyupo rd pev éx
rjés [ yevop Jev[ov rapcev-
cra [dv weprB}oAo[e roe omo~
Ojev ro rés ‘AOevaia[s dpxaio »
cjd du mora: CIA, IV. 1 C, 25-29 (p. 3 ff.).
This inscription, to which we shall return, says at least so much, if we accept the restorations,? that in the first half of the fifth century ic. treasure of the Eleusinian goddesses was kept in an enclosure behind the old temple of Athena on the Acropolis. This statement is strikingly similar to that of the scholiasts and lexicog- raphers quoted above, who say that the treasury was behind the temple of Athena, one of them that it was behind the temple of Athena Polias.
In 454 .c. the chest of the Delian Confederation was transferred to Athens, and from this date the funded treasure of the state, which consisted of the surplus of its yearly income and was kept on the
1 Der Parthenon, 1871, p. 293- ® Littenberger, S/G. 384, and Dorpfeld, A/stth. d. Inst. Athen, 1887, X11. p. 39. See p. 45: note 2.
The Opisthodomus. 13
Acropolis, was large." This was public money, Sydow, in contrast with sacred treasure, iena. The two funds were kept separate, but they were nevertheless both housed in the same place, in the keeping of the rapia ris Geod.
Now the theory that the public and sacred treasure of Athens was stored in the opisthodomus of the Parthenon fails to provide a place for it before the completion of that temple in 438 B.c. This fact is so formidable that once the advocates of the theory even resorted to the supposition that the opisthodomus of the Parthenon must have been completed and put to use as early as 454 B.c. when the Delian fund was brought to Athens, but this view is now abandoned.
A place, therefore, must be provided for the storing of these funds in the time before the Parthenon was built. This is conceded even by those who believe that the Opisthodomus was in the Parthenon.’
We have, further, excellent testimony to the existence of a treasury at Athens, which is mentioned in connexion with the Stoa Poecile and temple of Castor and Pollux in such a manner as to make it highly probable that it was a separate structure. The connexion in which it is mentioned makes it certain that it was at Athens.
Avxoupyos év ro epi rys lepeias’ mepi TloAvyvwrov rov (wypadov, @aciov pév rd yévos, viov §¢ xai pabyrod "AyAaopavros, Tuxovros 88 Tis "AGyvatwy wodsteias yroe érel ryv TlorxiAny oroay éypawye mpotxa, 4, ws érepot, ras év T@ Onoaps Kal T@ “Avaneiy ypadas, icropyxact dAXot re cai “Aprénov évy ro wept Cwypadwv nal "IdBas év rois wept ypaduxijs. Harpocration s.v. TloAvyvwros.
In Photius and Suidas (s.v. ToAvyvwros) and in Eudocia (340, ed. Flach, 1880), this reads as follows :
ovtos {wypddos piv Hv tiv réxvnv, Odoros S¢ Td yevos, vids Se Kal pabyrns “AyAaopavros, ruyov 8 7s "APnvaiwy wodtrelas, érel riv TloxtAyy oroay dvéypaiye wpoixa, 7, ws evar, tas év TP OyoavpG xai Tas év “Avaxeiy ypadds.
1 There was a fund before this time, but it was relatively small. See Thumser, Hermann’s Griech. Staatsalt.©, 1892, 1. 2, pp. 629 and 662.
2 For example, by Frazer, who says (Jour. Hellen. Stud. 1892-93, X11. p. 162)
that in this time the Athenians must certainly have had some strong place in which to store the public and sacred treasure.
a4 ah A ere a. at 2G. dalbecGoeloa Teun Semne = Suimors wotcuc
fee" 3. WLS Wee Lies UNegitumd le —stmuoav of JqsEs Mh. bsitiies We GaNCNes a we Mescum cu Mceim* Tie xo TURTOG, EA20cS0CR LA. Ula US Smearpus 1 ACTeNS WII was wR WI tanchgs WW Fut enedis Vas 1 segura: Juiuimy, ‘5s STsAgileGec 2° lle bed-auw. incl Unf ite TeEureduuses ar oVoGp. ade Leela Were Diicel Pwereepee LO Were semumre c1c- Tadime | SERTe Ulead We avs Wueifudis Teferencs Tu 1 ace for te Tide, 2 AnledsS Wick Was Pom imo nuenendent STwwtelc. fas Weisus Mediicdes QV Aarmuerstiua cunnut Taive Weed LA maa oe atimpertaa Juans. Toute Welt. Wat Ute Jtascteecmes of ite Acrepuds oct was
accu is 2 Weds Y Wie 2 xweDuraiet Mding .$ lLatthuitctea Jv inv MAT SCLOW if ela. calender, Df Mux Je ivavurieuw 3 (C ius Deira Ls aed ies S$ Beh teu lour im@mes 2 rte
-”~ oo ~~
. sae - -_ 7 2 — io. sa oe - ~_ ~let . “me: « n - = wemie’ ba wes Se ~oete Lf Tle MSS LEE GUULOU Ibe ec. “he ~GUStry
a ast lO
SAM be wu Tek Tes cur oTisvucupes Nomeces, TEM on Nuenres OG Tus, oTTE ine Mul we Tact Lend, Some 53.
Retz oS Te IARMLLen A Lay ime Gi Tdese Tuur Dassages caac che “aeiinc ude Mehiedes Noe “ic temerm /famoer of c2amvers of “She i2..2 liner Gi tue Paruienin cr i cue Hewaiompeuoa. ™ tne Luaivar. WILY is (Mev LLLk Lav olpacatien at ha cts caser 9 UCSus Gl Tie Terma reelided OV Demesxeacs ind Te spud Mun Gagited IV Lecian Te.die 19 1 Munding (nal Seu part ind
“has il .bst (2 2 where lediadetsi Clit te 2 MC OE 2 Creat
ene. lt 1S LEADIV WAaeTediifis Tat 1a] Paiiaecien saQvaid Tare ccen sel Te IB “Be cary oars Lf Lie Dura leikiurs (ipa ne dis
wre Neild 1E NW Bebicic 24 2eclhk Ta. Cui Cuwn iv cs. ln one
Serta, iw Seats tee oe OSE Sgt Vibe taco Ore Loree Pe a ol EY oe ee Se, 2 i Nw tele. a a Se I of) A a en) SC Y Pause “lias ol ce 2. amet “tye leat tici ff Ww deus “fe ilesyca.
The Opisthodomus. 15
may well wonder how Lucian imagined Timon to have set to work to dig through its massive walls. That would have been impossible, and yet in his answer Timon, although he denies the charge, recog- nizes the possibility of doing the thing of which he is accused.
There are two other references to an opisthodomus on the Acropolis :
rov yap dmiaOodopnov Tov Tlapbevovos arédagay airg (the Athenians to Demetrius) xardAvowy, xaxet dtarrav elye. Plut. Demet. 23.
dxyovre S& ovTw poe ycyverat dys dveiparwy rode: eddxovv evar pev "AOnynow dptt xarnpxws, oixety be éfomcbe ris dxporodews ey oixia @coddrov rod iarpov, elvoas 5° airnv mpwrnv mpos yAvov dvicxovra: rov 8 vew THS A@yvas SpacBat rov dmricOddopov dx aitys, Kai elvat woAd KaTw- Tépw THs akpoToAews THV oixiav. Aristides, I. p. 548, 14, Dindorf.
The disgraceful housing of Demetrius and his mistresses in the Parthenon is a well-known event. Plutarch’s record of it contains an implication of importance to the present discussion. When, namely, the Opisthodomus is referred to by Demosthenes, Aristophanes, and Lucian, no specification of its situation is necessary. It is sufficient to say 6 dmcoOddop0s. But Plutarch in designating the place in which Demetrius was lodged felt it necessary to name it rov dmricOddopov rou IlapOevavos. Aristides, likewise, who as the context shows undoubtedly refers to the Parthenon,’ says rov vew ris "AOqvas rov émcOddopov, not simply rov drygGodopov. The inference is that these two opisthodomi were not the same, and therefore that the Opisthodomus was not the western chamber of the cella of the Parthenon.’
———
1 dtéwicGe ris dxporédews means south of the Acropolis. Cf. Hdt. vu. 53, Eurpocbe xpd ris dxporé\ews. —I am indebted: for the reference to Aristides to Professor Edward Capps of the University of Chicago. Search might reveal other references to an opisthodomus on the Acropolis in other late writers. There is no such reference, other than those cited in this paper, in Homer, the Dramatists, Herodotus, Thucydides, Xenophon, Plato, the Orators, Aristotle, Diodorus Siculus, Lucian, or Pausanias.
2 Any objection to this conclusion based on the fact that Plutarch and Aristides were not, like Demosthenes and Aristophanes, Athenians, and that they wrote
«in a later age and needed, therefore, to be explicit in designating the opisthodomus that they mention, would hold equally against Lucian. Lucian had greater need
16 John Williams Whate.
The theory that the Opisthodomus on the Acropolis which was used as a treasury was a separate building is not contradicted by any references to it in the literature. The references to it in inscriptions are the following :
ovro S& rayuevovrwy éurdda év ro dricO[0d0]|uw ra Trav Oe@y xpjpara, 60a Suvarovy Kai datov, kai cuvavoryovrwv Kai cvyKAyovrev ras Jipas Tov dricOoddpov Kat cvoonpavecIwy Trois ray THs “AOnvaias rapias. CIA. I, 32 A, 15-18.
[éraday 8 ék r&]y diaxociwy ra[lAavrw lv, a és drodoow é[yydiora 6 djpos rots] aAAos Geois, a[rod08]y Tra dheropeva, ral mevecOw ra pey ris 'AOn|vaias xpypara [ev rq] eri Sefta Tov dmic[Oodopov, ra Se tav dAdwv Oleav ev ro ér dp[iorep|d. C/A. 1. 32 B, 20-24.
tade mapédocay of ra[pias Pwxiddys €€ Oijov Kai vvapxovres, émi X[rpa]lroxAdous dpxovros x[ai] éri ras Bovdgs 9 WaAleaorias rparos éypap |uareve, orparyyois mep[t THe ]Aorovvncov AnpooBeva ‘AAKcoEvous "Aqid[vaty éxi ris... 480s] rpvraveias reraprys [mputa |vevovons, tpiTy nplélpa tas mpuravecas é[ceAnAvOvas, ex rod dmoOjodopov. 4 4 4 C/A. 1. 273, ab, 16-20.
karabetva: év [woAa... Oley rov dmicPodopov éx[t THs ‘Irrw6 jwvridos aputaveta[s]. C/A. 1. 109.?
rapio[s rots Tav dAAwy Oey rots] eri Evx[rypovos dpxovtos, ols | Auwpobeo[s .. . eypappareve], xara yy[diopa rov Sypov mapedopev | trobéu[ evo. Ta xpypata ta ev tg] dmicod[onw...]. C/A. Iv. 225 € {p- 168), A, col. 11. 31-36.
[dro ]Oodopov. C/A. 1. 191, 3.
to be explicit than Plutarch. If Lucian, in the passage just quoted from his Zimon, where he is undoubtedly referring to the Athenian treasury, had said rdp éricbbdouov Tod Tlapfevwvos, as does Plutarch, the determination of the question now at issue would probably not be regarded doubtful. The fact that he does not do this, but uses simply the expression rév érceGddopoy, is in itself a strong argument that he is referring to some other opisthodomus than that named by Plutarch. ;
1 «‘ Derartige genauere Lokalbestimmungen sind uns ja in grosser Zahl erhalten, aber ich kenne kein Beispiel, in dem lediglich der Theil eines Tempels zum Aus- gangspunkt einer solchen Orientirung gemacht worden ware. Auch hier empfangt man (wie C/A. I. 32 bei duwéde dv Tp dwicbodéuy) unmittelbar den Eindruck, dass unter ‘Opisthodomos’ ein selbstandiger Bau zu verstehen sei.” Milchhofer, Philol. 1894, LIU. p. 358.
The Opisthodomus. 17 |
é& rov dmabodo[pouv}]. Sits. Ber. d. Berl. Akad, 1887, p. 1201, no. 45, If.
rade ev To dmaboddcpw ex THs KiBwrod THs Bpavpwv[ oe ]v- immxds Kexpvadros, éxnvia, Bevoripos Kapxivov dveOnxe. CIA. 11. 652 B, 23, 24. Repeated in C/4. 11. 660, 61, 62.
[émecbodo]pov. C/A. 11. 685, 2.
[orieO Jodopov. C/A. 1.720 A, col. 11. 6: draBodopos. Jd. B, col. 1. 32.
[r]av Ovpay rod éx[tcGodopov}. C/A. 1. 721 B, col. I. 19.
[rdde wpoorapédocay] rapice of éx[i Avxioxov apxovro|s rap tas tlots éxit UWvOodor[ov yxarKxoOyxns év re dm jicG0(d)o(u)o[e]* ox[edy xpjeuaora. C/A. 11. 721 B, col. If. 21-23.
The striking fact here, as before, is that the great treasure-house of Athens is referred to simply as 6 dmo@odon0s. Whatever other information about it we may be able to gather from these important records, there certainly is no implication in any one of them that the Opisthodomus mentioned was the western chamber or chambers of the cella either of the Parthenon or of the Hecatompedon.
The case, therefore, now stands as follows: The assumption that the Opisthodomus was not a separate building involves the rejection of the testimony quoted above of the scholiasts and lexicographers, who were drawing on good sources and whose special purpose was a definition. The authors and inscriptions, on the contrary, say nothing about the situation of the Opisthodomus because they unconsciously assume that this is known. On the other hand, the assumption that the Opisthodomus was a separate building, a fact clearly declared by the scholiasts and lexicographers, finds no contradiction in pas- sages in the authors or in inscriptions that refer to the Opisthodomus, provides a place for the public and sacred treasure before the build- ing of the Parthenon, and is further supported by independent considerations of weight. In other words this theory reconciles the evidence.
The direct discussion of the main thesis of this paper, that the Opisthodomus was a separate building, is now finished. The question of its situation still remains. If it was not the rear chamber or chambers of an existing temple, either the Parthenon or the Heca-
18 John Williams White.
tompedon, but a separate building, where on the Acropolis was it situated? This is in itself an interesting question, but there is a ‘stronger reason for its consideration. The proposition that the Opisthodomus was a separate building will be corroborated. if it can be shown with reasonable probability where it stood, and if that cémclusion is seen to conflict in no way with the ancient testimony that hay come down to us in regard to it.
[In the consideration of this question, as before, and for the same reason, we expect no help from the authors and inscriptions: the aheliasts and lexicographers, however, do give us information. They say that the Opisthodomus lay dchind the temple of Athena, and specifically behind the temple of Athena Polias. If, further, the gen- erally accepted! restoration of dpxaiov in C/A. Iv. 1. 28 be allowed, we have evidence that money was kept, although the Opisthodomus is not here named, behind the old temple of Athena, or) rov ris "AOqvaials dpxuiov velw éu worn, at least as early as 460 B.c.?
What was the temple of Athena Polias? Until very recently there was but one answer to this question. The term HodAcds, when used of the protecting goddess of Athens, was the epithet of Athena in her oldest temple on the Acropolis, and this oldest temple, 6 dpxatos vens, Was the Erechtheum or the temple that preceded it on the same site. The eastern chamber of the cella of this temple had been from carly times the shrine of the ancient wooden image of the goddess. Here, and nowhere else on the Citadel, she was worshipped under the title of “A@yva THoAcds.
(in the discovery of the Hecatompedon, Dorpfeld took issue with the prevailing view. The oldest temple on the Acropolis, he said, was the H[ecatompedon, not the Erechtheum, which was only a shrine of Erechtheus; the Hecatompedon was the original temple of Athena Polias, but not the only one; the temple of Athena Polias par excellence was the Parthenon.’
1 It is accepted among others by Kirchhoff, Dittenberger, Dorpfeld, Curtius, razer, and Furtwiangler.
4(Juoted In parton p.12. See also below, p. 45. Whether we read brid ]Jey or vérod]er docs not, as we shall see, affect the result
8 AMfitth. d. Inst. Athen, 1887, X11. p. 190 ff.
The Opisthodomus. 19
Dorpfeld’s attempt to wrest the names, o/d temple, temple of Athena Polias, from the Erechtheum has not met with acceptance. It has been considered and successively rejected by Petersen,' Curtius,? Frazer,®> and Michaelis,‘ all of whom maintain the traditional view that the oldest temple of Athena on the Acropolis was the temple of Athena Polias, and that this was the Erechtheum.® The discovery of the Hecatompedon modifies the traditional view only to the extent of making the designation of the Erechtheum as “the old temple” necessary at an earlier period than the building of the Parthenon, ée. it was “the old temple” as compared with the Hecatompedon, not with the Parthenon.
The question at issue is large, and its discussion has been able and searching. Strong arguments have been advanced in support of the traditional view.® Since they were made, a new element has been introduced into the discussion by Furtwangler, who in the remarkable book already mentioned’ has published a new and _ startling hypothesis in regard to the Hecatompedon. Neither his view nor Dorpfeld’s seems to me tenable, and I now purpose to state as briefly as possible the reasons why I am not able to accept either of them. It will be possible to consider the question here only in its most important aspects.
It is an essential part of Dérpfeld’s theory that, in the time prior to the building of the present Erechtheum, the worship of Erechtheus was maintained in a separate temple. He believes that the Heca- tompedon was “the old temple” and belonged exclusively to Athena, and that beside it, but separate from it, was the shrine of Erechtheus.®
1 Mitth. d. Inst. Athen, 1887, X11. p. 62 ff.
2 Stadtgeschichte von Athen, 1891, pp. 124, 151.
8 Jour. Hellen. Stud. 1892-93, XI. p. 153 ff.
* Altattische Kunst, 1893, p.16. See also Mitth. d. Inst. Athen, 1889, XIv. P- 349 (note the title).
5 Lolling also denies that the Parthenon could have been called the temple of Athena Polias ("A@»4, 1890, II. p. 661, note 4), but shares Dérpfeld’s opinion that after the Parthenon began to be built the Hecatompedon could be called dpxatos or wakads vews (p. 643).
6 See in particular Frazer’s able discussion, Jour. Hellen. Stud., 1892-93, XIII. pp. 167-185.
7 See p. 2, note 2. 8 Mitth. d. Inst. Athen, 1887, X11. p. 199.
20 Joun WVisitrms WVtaie.
Tins ew cannot 3e veconciea wrth che sv:dence af Homer in che “u:GWIng DJassage -
m9 ia | Adwees ayov. trat:ucroy sTasuedper.
aypaw Everyones acpwuyropes. ie sor | \omm
Qpewe. Luc Pevarnp. vexe 3 leOemes iperas.
<ao > @& ACmrys use. te & sun oT
@rUu 0¢ way Tarpaw?t au inracs _\agrm
corpus AGyousey TEMTEAAOMerer cnarres:
caw if yvyenover ~tos Terese Mereowers. 7. “1. =.16—=52.
Tae passage sstiniismes che ‘cllowing Gces. Accoreiny *o che Deets conception. 3f 1 Ime intecsdent co tne Tron War. iihena possessead 2 C2 “‘empie it lifiemsy ina in chs ‘empie. er 3jwn ‘emote, se 2siaoisned Erecntneus. Here. in he Doets ome. the Athenian youth dfered Erecatmeus sacmacs. in ine passage ‘t is important 7o 2cte “hal rere cefers 0 1 Ime Orer co the aime fF the main “ern dyer: “Nat aoe jecnotes permanent <smdilsamenr: thar pres (Gisputapiy Means emf: “AC “AIS Tempie was \thena’s: and that ws refers co Erecateus.. Ail scnoiars are agreed ‘iat the cemzi¢ seferrec co Was “he Jidest tempte of Athena on the Acropoiis. The s.aest tempie on che Acropous was, :Rerefore, a cempie of che ‘ont Worsmip af Athena and Erecatneus. The cradinon of this TOint WOYs0:9 wis Maintained *O “Me -atest Imes. A cerebrated passage im Sabu IX 10. p. 396° is as I chink, dec
sive igainsi che view “nat toe Hecarompedon was ~ the vid tempie.~? Darpieid .o0w maintans* shat Strapo $s scitement in this massage is
——— -- - _—_
Thoregeld gave cas 2. o>. aoter: + Cater a Kann man auca Jena Erechtheus versttenen. woe) “emer: Tam =3 Jewebnmucn wegen ies iunmitteibar vorner- yeneniien 6 47 TiOn ses tur raeta” What eauitor of Homer refers up cy : < rsfers to Erevntheus an ancient cradigun. Ane for che “est -€ reasons. Euls ‘ere 2ot offered cu Ataema, Suc henfers we hoch on rth ie. Thctek Sain 382 ch. sid ah. iv. coy. For a recente expression ot SpIMION “M1 inis FUCHCON, see Rohue. Prwte. 18Q4, LoD rtt. nore 1,
2 in ‘ateresting imsemptien id thas conpesn is Olu. ofS. FO. 31 i= Raabe, Ep. Gr. te4b. va. 01: eas 780 “Adwres vrer | Epydeows Saslia rey epsaredijae Tevee~ ov 5 usa ity.
'Ounted on p. 29
* Far aim earlier views. see Uta. 5. 2 026. Altes, i997. XE pp 43, tag.
kttema’ Tue ‘rauiten t2ar .
ww
The Optsthodomus. 21
entirely correct, but gives it a startling interpretation. By 6 dpyatos veas Strabo means the Hecatompedon; in this, not in the Erech- theum, was the lamp made by Callimachus! as well as the old dyaApa of the goddess. It was the original intention of the builders of the present Erechtheum that the old statue should be housed in its eastern chamber, but it never was placed there but remained in the Hecatompedon.? Pausanias in his tour of the Acropolis, Dorp- feld now believes,® entered the Hecatompedon from the Erechtheum at the point indicated in his book by the words, iepa pév ris ‘AGyvas xrA.;* he makes no mention at all of the eastern chamber of the Erechtheum, and indeed he had no occasion, for it contained nothing worthy of description. Only by this unparalleled break with tradi- tion in regard to the place of the lamp and ancient statue can Dorpfeld maintain his view that the Hecatompedon was ‘the old temple.’’ Scholars who are inclined to accept it must go further, and suppose that Strabo, who is evidently giving a categorical account of the temples of Athena on the Acropolis, fails altogether to mention the Erechtheum, which was certainly in existence in his day, for it still remains.
The difficulties that Dorpfeld encounters in these two passages, not to discuss at this time the remaining literary evidence, are very great. They disappear, if we assume that an Erechtheum, built on the site of the present Erechtheum, was “the old temple.” It will, perhaps, make the following argument more easily apprehensible, if I here state what I conceive the facts to have been. Homer’s plain language demands a temple of joint worship of Athena and Erech- theus. Such was the Erechtheum. This was the oldest temple on the Acropolis. In time, when the worship of the goddess had grown
1 On Callimachus, see Furtwingler, Afast. Greek Sculp., 1895, p. 437.
2In C/A. 1. 322, 1, érurrdras roid ved rod év wédea, ey § 7rd dpxaioy Ayaduya, Dorpfeld understands gera:. This is against the usage of Attic speech, and would be hard to parallel. The present écri is to be supplied, according to a well-known idiom, and the words prove clearly enough that at the date of the inscription (archonship of Diocles, 409-8 B.c.) the old image was already in the new Erechtheum, which was completed in the summer of 408 B.c.
® For his earlier view, see Afitth. d. Inst. Athen, 1887, X11. p. 52 ff., 210 f.
* Paus. 1. 26. 6.
22 John Williams White.
and her treasure had increased, the Athenians built her a great, new temple.! We are informed in a document whose trustworthiness is not disputed that its official title was ro “Exarouredov.? In con- trast with this newer temple, the Erechtheum was now “the old temple,” and it kept this designation to the latest times.» The Heca- tompedon, after its destruction by the Persians, was not rebuilt as a temple.‘ Its intended successor was the earlier Parthenon; its actual successor the Parthenon of Pericles. The Athenians, imme-
—— = — ——
1 This then Lecame ¢4e temple of Athena, although it was not the sole seat of her worship on the Acropolis, and from this time to the Persian wars any mention simply of the temple of Athena refers to it, unless the context makes clear that the krechtheum is meant. This doubtless is the temple meant by Herodotus in Vv. 72, 90, in describing events which occurred at the end of the sixth century. When, on the other hand, at this time the Erechtheum is meant, it is called 5 dpyxaios vews. Cf. Schol. Arist. Zys. 273. Such also are the references to the temple in the time just before and during the destruction of Athens by Xerxes. Cf. Vlut. Cim. 5; Herod. vill. 51, 53, 54- In passages, on the other hand, such as those that refer to the sacred snake, which by express testimony dwelt é» rg lepg r00 'EpexGéws, simply “the temple” is a sufficient designation for the .rechtheum, as in Herod. vill. 41. See p. 9, note 2.
2 See p. 2, note 8 end.
* For the passages in which the phrase 6 dpyatos (wada:ds) vews occurs, see p.%, note 4. This exact designation is, of course, not necessary where the con- text makes clear that the Erechtheum is meant. See note 1 above, and ¢f the building inscriptions cited on p. 8, note 5.
‘It may here be noted that Dorpfeld repeatedly asserts that Herod. v. 77 proven that the Hecatompedon was so rebuilt after the Persian wars (AZ7/#A. d. Inst. Athen, 1887, X11. pp. 31, 36, 200). One is at a loss to understand what he meann when he says (sdéd. p. 27) ‘an den Parthenon zu denken, wie es bisher geschah, ist nicht moéglich.” Who has believed that Herodotus refers to the Parthenon? Since Dorpfeld holds that the western chambers of the Hecatom- pedon were a treasury, and not the seat of a cult, in asserting that Herodotus refers to them in v.77 in the words rod peydpou roi rpds domdpyny rerpaypévov, he Iynoren the author's use of the word yudyapov. By uéyapow Herodotus refers to the ce//a of a temple where is established the worship of a god. So of the temple at Delphi, 1. 47, 65, vil. 140; of the temple of Hephaestus in Egypt, U. 141; of Zeun Ammon, tt. 143; of Demeter in Paros, vi. 134; and finally of Athena on the Acropolis, vitt. 53. It is extremely improbable, therefore, that in v.77 he applies the word to rooms used for semi-secular purposes. The Erechtheum, on the contrary, which at this time had been again rebuilt (see the next note), oxuctly meets the demands, for here, in the western half, was established the worship of Erechtheus.
The Optsthodomus. 23
diately after their return on the final withdrawal of the Persians, rebuilt the ancient temple of the joint worship of Athena and Erech- theus,' and this subserved alone the purpose of the worship of the goddess until the completion of the present Parthenon, for the earlier Parthenon was never completed. Toward the end of the fifth century the Erechtheum, which had been hastily reconstructed after the Persian wars, was replaced by the present structure,’ but it kept the name by which it had been known of ‘the old temple.” The traditions connected with its site were among the holiest posses- sions of the Attic race.
The most recent contribution to the discussion of the temples of Athena on the Acropolis has been made, as has been said, by Furt- wangler. He believes with Dorpfeld that the Hecatompedon was the first great temple on the Acropolis, but maintains that this was an Erechtheum, the original Erechtheum, where first was established the joint worship of Athena and Erechtheus. When the present Erechtheum was built, the Hecatompedon was entirely removed. Its traditions and epithets were transferred bodily to the new temple on the new site, which now, therefore, became “the old temple” of Athena. Previously the Hecatompedon had been “the old temple.” This view avoids many of the difficulties which beset Dérpfeld’s theory.
1 Cf. Herod. VIII. 55, quoted on p. 25. Writing sometime before 420 B.c., s.¢. before the present Erechtheum was begun, the historian says, “ There #s on the Acropolis a temple of Erechtheus,” and proceeds to speak of the olive and salt-spring as then actually existing in the temple. The present for: proves that an Erechtheum was in existence at the time of writing. Frazer, who fails to take this passage into account, says (Jour. Hellen. Stud., 1892-93, XIII. p. 169) that “after its destruction in 480 B.c., the Erechtheum was not, so far as we know, rebuilt till towards the close of the fifth century B.c.,” but he meets a real diffi- culty in explaining two inscriptions (C/A. Iv. p. 3; 1.93) which make mention of “the old temple” in such a way as to lead one to believe that the temple was at the time in existence. Further, there are two other passages in Herodotus (v.77, vill. 41) which are equally strong proof that the Erechtheum was rebuilt after the Persian wars, if one believes, as Frazer believes, that the Hecatompedon was not then in existence. In both these passages the historian speaks of the temple as existent, in the first at the time when he visited Athens, in the second at the time of writing.
2 It was begun in 420 B.c. and finished in 408 B.c. See Michaelis, Die Zeit des Neubaus des Poliastempels in Athen, Mitth. d. Inst. Athen, 1889, Xv. p. 349 ff.
24 John Williams White.
Furtwangler follows Dorpfeld in appealing to existing remains in order to prove the great age of the Hecatompedon. Its peristyle, he says, is not earlier than the time of Hippias, but the naos itself is considerably more ancient; moreover, there lie below it the remains of a yet earlier building.’ The philologist recognizes that such matters as this, the determination of the age of an ancient temple from its ruins, lie within the province of the expert archaeologist ; but when expert archaeologists fail to agree among themselves, he turns for decision to other evidence. Now, three archaeologists of distinction, not at this time to name others, have declared that the Hecatompedon dates from the time of Pisistratus.? Furtwangler’s further claim, which had previously been made by Dorpfeld, that no trace (with unimportant exceptions) exists of an older building on the site of the present Erechtheum, is a purely negative argument. This fact does not prove, of course, that such an older Erechtheum never existed. Such traces may in part have entirely disappeared, as he indeed acknowledges; they may in part be concealed by the existing building. The lack of them signifies merely that the exist- ence of an older building cannot be proved by its remains. The question whether there was such a building remains open.
We must appeal to the literature for a decision of the question raised by Furtwangler, whether the Hecatompedon was the original Erechtheum on the Acropolis, and thus “the old temple.” Here again we may be disappointed. The total number of references in Greek authors to temples of Athena on the Acropolis to the end of
1 Mast. Greek Sculp., 1895, p. 416. Furtwaingler here adduces no proof that this building, attested by slight remains, was a temple.—It has already been pointed out (p. 20) that the much-cited passage from the Iliad certifies to a temple of great antiquity, and this must be granted even if we suppose that the passage itself is not older than the sixth century B.c. It is difficult to follow Furtwiangler when he says (/.c.) that the passage “has in view the stately Hecatompedon with its double cella.” The tense of lAdoyra: shows indeed that the poet speaks of sacrifices made to Erechtheus in his own day, but they are offered in the ancient temple, and this he clearly conceives to have been in existence long before the Trojan War, a venerable structure about which centered the earliest religious traditions of the race.
2 Petersen, Alitth. d. Inst. Athen, 1887, XII. p. 62 ff.; Curtius, Stadigeschichte von Athen, 1891, p. 71 ff.; Michaelis, A/tattische Kunst, 1893, p. 16.
The Optsthodomus. 25
the fifth century B.c. is small. Further, in the nature of the case, the majority of these chance references will give us no information on the question at issue. In view of these facts, if any single reference is found to contain positive evidence, especially if this evidence is confirmed by other probable considerations, it must be allowed especial weight. Furtwangler’s view, for example, will be discredited, if at the time when he claims, as an essential fact, that the Heca- tompedon was the sole temple of Erechtheus on the Acropolis, it can be shown that a trustworthy Greek author testifies that there was another temple there devoted to his worship.
We have such evidence, I think, in Herod. vit. 51-55. The historian here describes the capture of the Acropolis, the sack of the Hecatompedon, and the destruction of everything on the Citadel by fire. The Hecatompedon, which was at this time fhe temple of Athena on the Acropolis, he calls ro ipoy (chap. 51 425, 53, 54), and specifically mentions its cella, rd péyapov,' as a place of refuge (chap. 53). He then continues (chap. 55), éore év ty dxpordAt ravry "EpexOeos rod yryeveos Aeyouevov elvac vnds, év TH éAaly Te Kal Odr\acoa ém. With these words he plainly introduces a new temple to the attention of his hearers. The 'EpeyOéos wns, here first named, is not the same temple that he has just mentioned repeatedly.? Herodotus, therefore, writing sometime before 420 B.c.,*° #¢. before the present Erechtheum was begun, testifies to two temples on the Acropolis, a temple of Athena and an Erechtheum.
1 For the use of uéyapor in Herodotus, see page 22, note 4.
2 It may be urged that rd lpéy in chap. 51-54 does not mean fempie, but sanctuary, te. Téyevos. The ambiguity of the word is well known. But even if we grant that ipéy has that meaning in these chapters, the force of the phrasing at the beginning of chap. 55 remains the same, for the Hecatompedon has been brought clearly before the mind of the hearer by the direct mention of its uéyapop in chap. 53. Nor can més in chap. 55 be given the meaning ce//a or chamber, and the word be made to refer to a part of the Hecatompedon, for in that case Herodotus would not have written éy ry dxporéut ratry, but would have said tor: dv re log robre (ic. TE ‘Exarourédy) "EpexOéos . . . mnés, or something of the sort. One could not speak of ‘“‘a chamber on the Acropolis” without immediate mention of the structure of which it was a part.
® Kirchhoff, Entstehungszeit des herodotischen Geschichtswerkes*, 1878, con- cludes that Herodotus died in 428 B.c. The absence of later allusions in his history shows that his death occurred at least before 420.
26 John Williams Whate.
This interpretation of the reference in "EpeyOéos vnos is made cer- tain by a statement in the passage itself. This statement proves that the temple called "Epex@éos vos cannot be the Hecatompedon. Herodotus says that the sacred tokens, the olive and the salt-spring, were “77 the temple.” His words are év rq (1.¢. €v r@ vy@) €Aaty xrA.! The tokens we know were on the low ground to the north of the higher plateau on which stood the Hecatompedon. They were cer- tainly not in that temple.* The difficulty caused by the language of the historian is so great that Furtwangler proposes to emend the passage and read onxos for vyos, but he offers no critical reason for the change, and none exists.* It is made to meet an exigency in argument and cannot be allowed. If allowed, it would bring a new element of confusion into the discussion in establishing a onxos of Erechtheus distinct and apart from his wyos.
The conclusion, adverse to Furtwangler’s theory, to which we have been brought is confirmed by other considerations which discredit the assumption that the Hecatompedon was the original Erechtheum. On this assumption the traditions connected with the early worship of Athena and with the worship of Erechtheus, and the epithets of the temple of their joint worship, must be transferred bodily at the end of the fifth century B.c. toa new temple. Not only is it inherently more probable that they had always belonged to the site to which a later age undoubtedly attached them, but positive con- siderations also make it extremely unlikely that they had previously belonged to the Hecatompedon. For example, 6 dpxatos vews was an official title of the Erechtheum in the fourth century.* This same term is applied to a temple on the Acropolis in reference to an event
1 On the place of the olive, see p. 39, note 6.
2 In speaking of the present Erechtheum (J/ast. Greek Sculp., 1895, p- 433), Furtwangler makes an important admission: ‘‘The choice of site was determined by the consideration of keeping near the sacred ‘tokens’; in fact, the temple was to be even more closely attached to these than its predecessor had been; the cleft in the rock was included within the building.” This is precisely what the passage quoted from Herodotus proves for the Erechtheum that preceded the present Icrechtheum.
8 Mast. Greek Sculp., 1895, p. 416, note 9.
4 C/A. 1.74 a, 14; 163,9; 464,6; 672, 43; 733 A, col. 11.6. The temple was not, then, called so simply ‘in common parlance ” (Furtwangler, #dd. p. 433).
The Optsthodomus. 27
that occurred before 500 B.c.! The title in the latter instance is meaningless on the assumption that the Hecatompedon was the only temple here at this time. Again, two inscriptions which date from the first half of the fifth century B.c. mention “the old temple” officially.* This was the Hecatompedon, it is asserted, and it was called “old” in contrast to the earlier Parthenon. But this Parthenon never got beyond its foundations, and further, we have unimpeach- able evidence that the official name of the Hecatompedon was ro “Exaropredov. These three references to an dpyaios vews are, on the other hand, perfectly applicable to an older Erechtheum on the site of the present Erechtheum, designated as “the old temple”’ in contrast with the Hecatompedon.
The assumption further that the Hecatompedon was the original Erechtheum involves a complete change of site for that building at the end of the fifth century. The sole reason alleged for this is that the Athenians desired to bring the temple into closer connexion with the ‘tokens.’* But this alleged fact would lead us rather to conclude that the tokens mark the original site of the temple. Moreover, it is a commonplace that Greek religious feeling demanded that a temple when rebuilt should occupy its old site. The exceptions, especially when at the same time the old temple was removed (the fact claimed in this instance), are few indeed, and there are always adequate reasons. Such reasons do not exist in this case. The alleged change of site, on the contrary, necessitated a vital change in the principle of construction, for the present Erechtheum stands on different levels; gave the temple a cramped situation hard upon the north wall of the Acropolis ; and reduced its size. And yet this is the temple, it is claimed, that was built by the opponents of the policy of Pericles to replace the stately Hecatompedon. Furtwdangler further argues that the Erechtheum in its interior arrangement repeats the Hecatompedon ;* but if this statement is to be used as an argument
1 Schol. Arist. Lys. 273.
2 C/A. IV. 1. c, 27 (p. 3 £.); 1. 93, 6.
® «Tt was in fact simply in order to attain this object that the position of the temple had been changed at all.” Furtwingler, Mast. Greek Sculp., 1895, p. 436. See also p. 433.
* Mast. Greek Sculp., 1895, pp. 433 ff.
28 John Williams White.
to prove that the former was the successor of the latter, comparison must be instituted between the two temples entire. The Hecatom- pedon was a Doric temple,’ built on one level; it was a peristyle, with porticos and entrances at the east and west. The Erechtheum is an Ionic temple, built on different levels; it is not a peristyle; it has a portico on the east, but none on the west, and quite irregularly, has porticos with entrances also on the north and south. The sole point of resemblance is the division of the cella into three compart- ments by cross-walls, and even this resemblance is disturbed by the difference of level in the Erechtheum between the two western chambers and the eastern chamber.? The two temples are not simi- lar, but strikingly dissimilar. The closer the comparison we make of them, the more we are impressed with the structural peculiarities of the Erechtheum. It is here important to note that to Dorpfeld the groundplan of the Hecatompedon seems to be strikingly similar to that of the Parthenon.’ This accords excellently with the view that the Hecatompedon, an exclusive temple of Athena, was the fore- runner, not of the Erechtheum, but of the Parthenon.
In view of these considerations the assumption that the Hecatom- pedon was the original Erechtheum seems to me improbable, if not impossible. It is more natural to suppose that from the earliest times there had stood on the site of the present Erechtheum a temple that had been built over the sacred tokens and had accom- modated itself to the original level; and that finally the present Erechtheum was built in imitation, not of the Hecatompedon, but of its own predecessors. This view is not contradicted by existing remains, is supported by positive evidence, does not force us to shift traditions and epithets, involves no change of site for the Erechtheum, and adequately explains its structural peculiarities.
owe ee
1 This is the opinion of its discoverer. See Mstth. d. Inst. Athen, 1886, XI. pp. 347 ff.
2 Furtwangler assumes also that the central chamber of the Erechtheum was divided by an east and west wall into two compartments. This would be another point of similarity, but he himself acknowledges that no traces of such a cross- wall exist. Afast. Greck Sculp., 1895, p- 433, note 5.
3 « Jedem wird sofort die grosse Aehnlichkeit zwischen diesem Grundriss und demjenigen des Parthenon auffallen.” Afstth. d. ust. Athen, X1. 1886, p. 340.
The Optsthodomus. 29
Another preliminary inquiry, important for the determination of the situation of the Opisthodomus, remains to be briefly considered. When ancient Greek writers, referring to a building on the Acropolis, speak of “the temple of Athena Polias,” which of the temples of Athena do they mean? As has been pointed out,’ Dorpfeld main- tains that the chief temple of Athena Polias was the Parthenon. This title, he says, was given also to the Hecatompedon, both before and after the building of the Parthenon, but she temple of Athena Polias was the Parthenon. He claims further that the official name of the Parthenon in the fifth and fourth centuries B.c. was either & vews OF 6 vews THs AOnvas ris ToAcddos.?
If these claims are valid, we shall reasonably expect to find them confirmed by the references to the temple of Athena Polias in Greek authors and inscriptions. The following are all the passages, so far as I know, in which the expression “temple of Athena Palias”’ occurs, with the exception of the passage whose proper interpretation we are now trying to reach.®
I. xvwv els tov ris TloAcados vewv eloeAPovca, xai dtoa eis to Tar Spdcrov, éxi rov Buwpov dvaBdoa rod “Epxeiov Auds, tov tid Ty éAaig, Kare- xaro. Philochorus frg. 146 in Miiller, Frag. Hist. Grace. 1. p. 408 (Dion. Hal. de Din. 3).
2. 708 dorvaird rérpa éoriv év rediy meptocxouperyn Kixrdw* evi 8 TH wérpg To THS AOnvas iepov, 6 re dpxaios vews THs ToAddos, év © 6 ao Beoros Avxvos, kai 6 TlapOevav ov éxoinoev ‘Ixrivos, év @ TO ToD Padiov épyov éXedhayrivov 4 ‘A@yva. Strabo 1x. 16, p. 396.
3. dye 8), mdpecpey yap vba xpiv: évratdd rov ev tp mpovdy Tis TloXados Sixdowper. 4 tépera SidBes 7a Babpa, Hpets 58 év rocovrw mpoo~- xuvyowuey tH Oew. Luc. Pise. 21.
4. Keira: 58 dy rp vag ris HoArddos “Eppuys EvAov, Kéxpomos elvos Ae ye- pevov dvafnpa. Paus. I. 27. 1.
1 See p. 18.
2“Wenn wir also von dem Tempel schlechthin oder von dem Tempel der Athena Polias lesen, so miissen wir annehmen, dass der Parthenon gemeint ist. Dieser war mithin der Haupttempel der Athena Polias.” M:tth. d. [nst. Athen, 1887, XII. p. 193. See also sdid. p. 196.
8 Schol. V Arist. P/ut. 1193, quoted on p. 4.
30 John Wilhams White.
5. wapGeva Svo rov vaov trys ToAados oixovow ov roppw, xadovor St "AOyvaia oas dppnpopovs. Paus. I. 27. 3.
6. ré dai “EptyOonos: ovyi év re vee Tis TloAcados xexndevra:; Clem. Alex. Protr. ut. p. 13 Sylb.
7. olos 8° 6 rips Hoduddos vews xai ro xAnotoy rot Hocadaves réeuevos- ourmyapey yap &a tev dvaxropwy tous Geovs dAAHAS pera THY dyudAay. Himenius, £c/. v. 30.
8. dvo cviv éri THs dxporoAews “APyvas vaot. 6 ras TloAsados xai ¥ xpuoereharrivn. yy aro trav Mydxwy oxvAwy xarexxevucay Padov wria- gayros. Schol. Arist. £g. 1169.
9. Bourys dx “EpeyGews 1o yévos exe, xai dr’ atrou xaXeirai re ADH rpc. yavos “EreoBovrdéda, of ry oyri azo row Bovrov. ovra rpoiorayra rou iepou THs év dxporoAa ‘AOywas trys TloAuddos. Schol. Aesch. 11. 147.
10. &o xara rovs raraots “AGyvyow Aidois xai "AdeAaas Fv Bwpos wep. tov rig [loAiddos “AGyvas vewy. xaba xui Havoartas iorope?. as ot pay wadaywyous. of 5¢ rpopots ris eas yeatohu daciy. Eustath. //. XXII. 451. p. 1279. 40.
IL. of dao, ws éxeiBer. xai oixovpds Spaxwy, purct rs THodsados, qyouy éy re vew THs Tlodiddos Kacraevos. Eustath. Ov. 1. 357. p. 1423, 8.
12. «ai dyuypayals avryy tov ypjayzpara& roy xara xpvrayeaay éy ornAy yoAx[y xai oTjou éjy axporo\a wapa Tov vew THs "AOywas THs Hof duddos}. Clef. 11. 332, |. 42 ff.
13. [orqou d€] atrov xai [ei ]xor[a] yoraqy eff trwow row Sypov wapa Tow vew Tov apyxatow rs A@nvas r[ ys Tlodddos]. Clif. 11. 464, L 4 ff.
14. [avaypayas 5¢ row ypayparer tloy xara xpvraveiay ds (o)TyAny MOL ivqy ro Wydkopus Koi Ta Gvopara Tov rupJerwr xai avabe jira éy axpo- woXtu rapa Tov vaoy THs AGn| vas rs [oAuddos, iva rovruy cuvreAouperwy J] .wyros 4 [yey]or[ eae] Sx" [abra]y wepl ratra ofroub} nal pudoworia] Mitth. d. Inst. Athen, 1883, VUL p. §9. 23 ff.
“ The temple of Atbena Polias” named in the first of the passages here quoted cannot be the Parthenon, because of the immediate proximity of the Pandroseum. In the second the language of the writer excludes that supposition. The same fact is true also in the eighth, however much the form of statement may have suffered in its transmission to us. In the fourth and fifth passages Pausanias
The Opisthodomus. 31
has already described the Parthenon and is now on the north side of the Acropolis. The temple named in the sixth passage also cannot be the Parthenon ; Erichthonius was identical with Erechtheus,! and the seat of the worship of Erechtheus was the Erechtheum. Simi- larly in the seventh, Himerius is speaking of the place of the joint worship of Athena and Posidon-Erechtheus.? The priestly family of the Eteobutads mentioned in the ninth cannot be dissevered from the Erechtheum.® So in the third the pronaos of the Erechtheum is meant, since Aeschines tells us‘ that the priestess of Athena Polias was chosen from the family of the Eteobutads. It will be claimed by nobody that the sacred snake spoken of in the eleventh passage had its dwelling-place in the Parthenon; and until we get proof to the contrary we have the right to conclude that Eustathius does not mean in the tenth passage by “the temple of Polias” any other temple than the one so named in the eleventh passage.®
If these conclusions are sound, no Greek author has called the Parthenon “the temple of Athena Polias.”
It is here instructive to note what the great temple ¢s called during the period of the writers quoted above. It had a perfectly estab- lished name. This was 6 IlapOevey. It is thus called by Demos- thenes, Heraclides (Dicaearchus), Strabo, Plutarch, Pausanias, Aelian, Philostratus, Zosimus, Marinus, a Scholiast on Demosthenes, Harpocration, Hesychius, the Etymologicum Magnum, and Suidas.® It is incredible that the Parthenon should, as it is claimed, have been the temple of Athena Polias, and should be mentioned so often in Greek authors, and yet that its so-called distinctive title should nowhere occur.
Dorpfeld claims that 6 vews was an official title of the Parthenon in the fifth and fourth centuries B.c. He fails to quote a single
1 See Rohde, Psyche, 1894, I. p. 128.
2 On Posidon-Erechtheus, see the passages in Jahn-Michaelis, Pasus. descrip. arc. Athen. 1880, p. 23 (c. 26, 27).
3 C/. Paus. 1. 26. 5; [Plut.] Vit. X.Or. 843 b.
4 Aesch. II. 147.
6 Dorpfeld also (AGitth. d. Inst. Athen, 1887, X11. p. 206, note) thinks that the Parthenon is not meant in the tenth passage.
6 See the passages in Jahn-Michaelis, Puus. descrip. arc. Athen. 1880, p. 13 ff.
32 John Williams Whiate.
inscription in which the Parthenon is called 6 veds. Even in Greek authors it is thus designated only twice,’ and both these authors, so ambiguous is the expression as they have used it, have here been charged with looseness of style. Again, the only inscmptions in which the phrase 6 vews trys "AOyvas trys Hoiuddos occurs are the two from which the twelfth and fourteenth passages given above are quoted. In the second of these two inscriptions, it will be observed, TloAsa8os is due to restoration. And yet on the basis solely of these two inscriptions, by combination with the inscription from which the thirteenth passage is quoted, where also HoA:ddos is due to restora- tion, Dorpfeld establishes o vews trys "A@yvas r7s ToAuddos as the other official name of the Parthenon. He adds, without proof, that this was probably its name also in current speech during the first century of its existence.” He seeks to establish this official title as follows. The designation 6 apyatos vews in official documents proves the exist- ence of a new temple. This was the Parthenon, 6 vews. In another official document (thirteen above) mention is made of an apyaios vews THS "AOyvas r7s LloAcddos. This name proves in like manner the existence of a new temple of Athena Polias, and in fact, he says, mention of a vews trys ‘A@yvas ris Uodddos (the phrase is not, it will be observed, 6 xacvos vews tys "AOyvas trys HoArados) occurs twice in official documents (twelve and fourteen above). Since now the apyatos vews tHs “AOyvas trys Hodddos is identical with the apxaios vens, it follows that the veas trys AOnvas trys oArddos is identical with the temple concisely called 6 vews. The latter is the Parthenon. The official title, therefore, 6 vews trys “AOyvas trys Hoduddos, belongs to the Parthenon.
This conclusion is unsound because it rests on unestablished premises. In the first place, the assumption is made that 6 apyxatos veas Means the Hecatompedon. But other scholars believe (and prove to their own satisfaction !) that “the old temple” is the Erech- theum. In that case “the new temple” is the Hecatompedon. Again, it does not follow that the antithesis to “the old temple of Athena Polias’’ is “the new temple of Athena Polias.’”’ This assumption begs
1 Xen. Hellen. 11. 3. 20; Aristot. hist. an. VI. 24, p. 377 5 29. 2 Mitth. d. Inst. Athen, 1887, Xl. p. 197 f.
The Opisthodomus. 33
the question. The expression 6 vews 6 apxatos THs ‘APyvas rys Todia- Sos may have arisen in some other way than the one assumed. The latter puts a relative emphasis on the word apxatos for which the war- rant is not forthcoming. Let us assume that the Erechtheum was called sometimes 6 vews 6 apxatos, sometimes 6 vews THs "AOnvas rHs ToArddos. The combination of the two titles would give, as Petersen long ago pointed out,’ 6 vews & apyxatos ris "AOnvas trys ToAddos, “the old temple belonging to Athena Polias.” This interpretation is, at least, as good a working assumption as the other.’ Finally, it does not follow that the temple named 6 vews ris ‘AOyvas ris MoArddos is iden- tical with the temple called simply 6 vews (if there was, indeed, such atemple). This claim rests on an assumption which also begs the question, namely that the phrases 6 vews trys A@yvas trys TWoAcddos and 6 vews are equivalent to 0 vews o xatvos THs AOnvas trys THodArddos and o vews 0 Katvos, which occur nowhere. If they did occur, everybody would agree that there were at least two temples of Athena Polias on the Acropolis, and the reader would be spared the present dis- cussion. But even if the conclusion did follow, we could not infer on that account that o vews rps '"AOnvas trys ToAcddos was a name of the Parthenon until it was established that 6 vews was an official designation of the Parthenon. No proof*of that, as I have said above, has been offered.
The only safe conclusions that can be drawn about the expression “the temple of Athena Polias” in the three inscriptions quoted above are that in the first the application of the phrase is indeterminable ; that in the second it cannot mean the Parthenon, since the temple is there called dpyatos ; and that in the third whatever indications we have point to the Erechtheum, since the girls there honoured were those who prepared the wool for Athena’s robe, and the peplus, as we know, belonged to the old statue in the old temple.’ But the inference is, of course, not certain that the stele in their honour was on that account set up beside that temple, although the fact is probable.
1 Afitth. d. Inst. Athen, 1887, X11. p. 65.
2 It is with reasonable certainty the true interpretation. C/ Strabo Ix. 16 (quoted on p. 29), where the contrasted titles of the temples are 6 dpyaios veds ris THodsddos and 6 Mapberwy.
3 See on the last point Furtwingler, Afast. Greek Sculp., 1895, p. 427.
34 Jokn Willtzms White.
It seems reasonable to conclude that the phrase “the temple of the Polias ” does not mean, in any of the fourteen passages quoted above, the Parthenon. The question is now narrowed to the two remaining temples. It is here proper to cail attention to the fact that we are certain of the existence of only one of these. the Erech- theum, at the time covered by the passages: we have to assume the existence of the Hecatompedon during the fourth century B.c. and the following centuries in order to bring it within consideration at all. Dorpfeld believes that the Hecatompedon was in existence at this time, and that it was, as it always had been, an exclusive temple of Athena; the worship of Erechtheus was confined to the Erech- theum, the present structure, which had replaced the earlier shrine.
The three inscriptions may be dismissed at once. They are indecisive, except so far as the considerations already presented in the case of the third make for the Erechtheum.'
The passages from the authors demand brief consideration. It should be noted that they all speak of «the temple of the Polias”’ as if there were only one such temple. They all indicate, I think, that this temple was the Erechtheum.
In the first the bitch enters the temple and goes down in order to get into the Pandroseum. The description exactly fits the plan of the Erechtheum. Its eastern chamber is on a higher level than its western chambers, and from the latter there was direct communica- tion on the same level with the Pandroseum. If the Hecatompedon is meant, we must conceive, since only one temple is mentioned, that the creature took a flying leap of over ten feet from the support- ing wall of the stereobate of the Hecatompedon at the north-west down into the Pandroseum. This is, at the least, an improbable meaning for Svea ¢is ro [lavdpoomov. The second passage has already been discussed.? In the third the Hecatompedon cannot be meant, because (besides the consideration urged above) the action of the dialogue would be impossible if the pronaos of that temple were its
1 If one believes, from independent considerations, that é dpyates reds always means the Erechtheum, then the second inscription becomes evidence; but it will be well, perhaps, to avoid the possible charge of the ‘ vicious circle.’
2 See p. 20 f.
The Opisthodomus. 35
scene. The Erechtheum would stand directly in the way.’ To identify the temple named in the fourth passage with the Hecatom- pedon is, if we follow Dorpfeld’s present lead,? to assume that the gdavoy and the lamp of Callimachus were not in the Erechtheum at all; and if on the other hand we suppose that Pausanias entered the Hecatompedon not after 1. 26. 5 but at 1. 27. 1* and thus place the écavoy and lamp in the Erechtheum, we must assume that Pausanias makes double application of the word Polias, once to the goddess of the old image in the Erechtheum, for it is clear that he refers to her as Polias in 1. 26. 6, a second time to the Hecatompedon. If we are convinced by these serious difficulties that by “the temple of the Polias ” in the fourth passage Pausanias means the Erechtheum, we establish its application also for the fifth, unless we are ready to believe that he speaks of two distinct temples, both as “the temple of the Polias,” within the short space of a dozen lines. The mention of Erichthonius (Erechtheus) and Posidon (Erechtheus)* in the sixth and seventh passages shows that here the Erechtheum is meant, not the Hecatompedon, a temple of the exclusive worship of Athena. In the eighth the very confusion of statement shows that the Scholiast closely associates in his mind the temple of the Polias and her ancient image,°® and therefore, if we suppose that by “the temple of the Polias”” he means the Hecatompedon, we must remove the image from the Erechtheum. The iepdv mentioned in the ninth must be
1 Dérpfeld claims that this passage refers to the Parthenon. For his three reasons, see Mitth. d. Inst. Athen, 1887, Xl. p. 198 f. To these it may be briefly answered that Lucian in the second century A.D. is not likely to have had intimate knowledge of the official phraseology of Attic inscriptions of the fifth century B.c., and further that the inscriptions do not add rfs Todd dos to the official form év ro Hpovnlw ; that any portico is large enough to accommodate an imaginary assembly (see Frazer, Jour. Hellen. Stud., 1892-93, XUI. p. 182 f.); and that to argue from the reference in the dialogue to the Pelargicon that Lucian must mean the Parthenon is to assume that the situation and extent of the vexed ancient forti- fication called the Pelargicon are satisfactorily known.
2 See p. 21.
3 So Miss Harrison, Afyth. and Mon. of Anc. Athens, 1890, pp. 508 f.
# See Frazer, Jour. Hellen. Stud., 1892-93, XII. p. 182.
5 See p. 31, notes I and 2.
6 Cf. schol. Dem. xx11. 13; schol. Arist. Pan. 187, 20 (Dind. II. p. 319).
30 John wWrllzams Wihzte.
the Erechtheum, for reasons already given.' The tenth and eleventh passages hang together. as has already been pomred out, and in interpreting the eleventh we have the express testimony of Hesychius that the snake was housed & rw ispe row Epexdus.* iv. in the Erechtheum. not m the Hecarompedon.
I conclude, therefore. that when in Greek writers we read of ~the temple of Athena Polias ~ we must understand by it the Erechtheum, and that this conclusion ts contradicted by no evidence offered by inscriptions.
This inquiry has been carried further by Frazer. in the article already often cited. He has mvestigated the word HoAas in all its applications. and in my opimion has gtven a final answer to the doubts raised by Dorpfeld. He has conclusively demonstrated that the word had a local connotation to the mind of a Greek when used with reference to the Acropolis at Athens. and has shown what this connotation is. In order to settle the question of the proper applica- tion of the term “Athena Poliis™ or ~the Polias” he collected the passages of classical writers bearing on the Athena Polias of Athens and all places in the Corpus of Attic Inscriptions in which the title eccurs. and gave them careful examinaton.* In some of these passages. as was to be expected. the application of the term Hoas is indeterminable: these passages furnish no indication whether the Athena referred to was the goddess of the Erechtheum, of the Heca- tompedon. or of the Parthenon. In the remainder the term is used with reference to the temple. to Athena in close association with Erechtheus. to her ancient image. to the peplus. to the priestess of Athena Polias. to the Errephori. to the sacred serpent. and the like. All the passages whose application ts determinable support the view that Athena Polias was the goddess of the Erechtheum. with the
I See p. 31. note > and Frazer. /vwr. Heilew. Stud. 1Sg2-g3, Xt. pe rt.
2 See p. } nuce =
3 [ had alreadv begun. with the same purpose, an independent collection of the passages before Frazer's article appeared. and am able to cuontribate from Greek writers the following slight additions to his extensive lit: Aclian, bur. Aust. 1 9; Locian. Sym. 22; Jf. Dun’. Meret. STL. 1; Schol. af co.: Clem. Alex. Pros. rv. g2 pr 5 Syib.: Apoll Ft desck. (quoting Aesch. IL 14>); Schol V Arise. Eg. 1169: Schol Hom. Ow x1v. 333; EM. 12. “Breqdowrddex. In completing my col- lecticns I have had the helpful assistance of Arthur S. Cooley of this University.
The Opisthodomus. 37
. exception of two. One of these countenances the view that she was the goddess of the Parthenon,’ the other may be so interpreted; * but the author of the first is a Christian writer living in Egypt in the second century A.D., the author of the second a twelfth century com- mentator on Homer, and each lays himself open to the charge of contradicting himself.‘
The result of Frazer’s investigation is instructive. The fact that the word Polias, when applied to Athena at Athens, had in the times for which we have literary evidence a distinct local sense con- fined to one place, shows what our conclusions must be for the earlier times to which our evidence does not extend. “Polias” was probably the oldest title of Athena on the Acropolis, and dates from ‘a remote age. The goddess had then a single temple on the Citadel, the temple of the Polias. The image within the temple was the image of the Polias. Thus the word became closely associated with a place. We have no reason for believing that when the Hecatom- pedon was built the worship of the Polias was transferred to that temple. We do know that in the earliest times and in the later times it was attached to “the old temple.” This conception, that the worship of the Polias remained in “the old temple,” in no way con- flicts with the supposition that the Hecatompedon also was a temple of the worship of Athena, containing its own image. It was fhe temple of Athena, but not the temple of the Polias. The Parthenon in the next century is an exact parallel in all particulars. The growth of the worship of Athena in the Hecatompedon would tend still more to localize the worship of Athena Polias in the old temple. After the destruction of the Hecatompedon in the Persian wars, the old temple gained in importance. It was doubtless the purpose of Pericles to transfer the worship of Athena Polias to the Parthenon. (That may have been the purpose also of the builders of the Heca- tompedon in an earlier age.) But conservative religious feeling and
1 Frazer (Jour. Hellen. Stud., 1892-93, XI. p. 184), who believes that the Opisthodomus was the western portico of the Parthenon (see p. 2), adds Schol. V Arist. Plut. 1193 (quoted on p. 4).
2 Clem. Alex. Protrep. IV. 47, p. 13 Sylb.
8 Eustath. Od. x1. 634, p. 1704, 30.
* See the discussion, on p. 31, of the sixth and eleyenth passages quoted above.
38 John VWwilliams White.
party strife combined to thwart him. The Parthenon was built, but his intention was frustrated. If it had been carried out, the new temple would have become “the temple of the Polias” and the Erech- theum would have been devoted to the exclusive worship of Erech- theus. We know in fact, on unequivocal authority, that the old image was not removed from the old temple.'! This temple remained, therefore, an important seat of worship of Athena on the Citadel ; and we are not surprised to learn by the consentient testimony of writers in the following centuries that it kept the name which first attached to it and continued to be called the temple of the Polias.
After this long, but necessary, consideration of the application of the expressions “old temple” and “temple of Athena Polias,” I revert to the question which occasioned it.”
What does Schol. V on Arist. P/ué. 1193 mean when he says that the Opisthodomus lay Jehind the temple of Athena Polias ?
If the front of the old temple of Athena, #.ec. of the Erechtheum, was at the east of the temple, as was generally true of Greek temples, the Opisthodomus must have lain to the west of it, behind the Pan- droseum, and must be sought for there. On this supposition there must have once existed at this place a substantial and independent structure, of the foundations of which, however, no trace has been brought to light by the recent thorough excavation of the Acropolis.
This brings us face to face with a question of great apparent difficulty. Namely, how is it possible that the treasury of Athens, a separate building as has been proved, was called an opisthodomus ?
Pollux (1. 6) after defining the uses of the word oyxos continues: To 5¢ xpo atrov xpodopos, Kai To xdromt omoaOodopos. Varro (de lingua Lat. v. § 160, ed. Spengel) gives the same definition: domus gractum et ideo tin aedibus sacris ante cellam, ubi sedes det sunt, Graect dicunt xposopov, guod post, GxicOcdopov. This use of éricGcSopos, to designate the back portico of a temple, is confirmed by its actual employment in the literature. It is thus applied to the western porticos of the temples of Zeus and of Hera at Olympia.*
1 See p. 21, nm 2. 2 See p. 18. 8 Paus. V. 10.9; 131; 15-3; 161; Lucian Hut.t; Fug. 7; de morte Peres. 32.
The Optsthodomus. 39
This established application of the word seenis to contradict hope- lessly the view that the Opisthodomus on the Acropolis was a separate building; but in fact it itself indicates the solution of the difficulty.
Whether in early times the northern or eastern portico of the Erechtheum was regarded the front of the temple cannot be surely determined ; but it seems probable that, at least in the time of the sources from which the scholiasts and Harpocration and the other lexicographers drew their information, the front of the temple was thought to be at the north. Here lay the broad portico through which Pausanias entered the temple.’ If the front of the temple was at the north, the Opisthodomus, which was situated émo6ev rov vew, must have lain to the south of the Erechtheum.
Here in fact we find it, the Opisthodomus of the old Hecatom- pedon, rebuilt, after the destruction of that temple in the Persian wars, to serve as it had served before the coming of Xerxes’ as treasury of the gods and of the state. The peristyle of the temple disappeared ;* its eastern chamber was not restored; the Opisthodo- mus, consisting of the three western rooms and western portico, was alone rebuilt. This was the Opisthodomus to which reference is made, in the times following the Persian wars, simply as o émo6o- Soxos ; the @yncavpds, probably, that was adorned with paintings by Polygnotus ;° the rapuetov of the scholiasts and lexicographers that lay “behind the temple of Athena.”® Not only its official but also
1 Paus. I. 26. 6 ff.
2 The mention of the rayula: in the Hecatompedon inscription (C/A. Iv. p. 137 ff.) makes this practically certain. See Lolling, 'A 0», 1890, 11. p. 647 ff.; Dorpfeld, Mitth. d. Inst. Athen, 1890, XV. p. 420 ff.; Frazer, Jour. Hellen. Stud. 1892-93, XIII. p. 162, note 24. For the pre-Persian rayla: rs deod, see also Hdt. vitl. 51. Furtwangler (Afast. Greek Sculp., 1895, p. 418) denies that the treasury was ever in the Hecatompedon.
3 So also Dorpfeld, Afstth. d. Inst. Athen, 1887, XII. p. 200.
4 See the plan, Afitth. d. Inst. Athen, 1886, XI. p. 337-
5 For the date of Polygnotus, see Brunn, Geschichte d. griech. Kiinstler, 1. 14 ff.
6 The language of Herod. viii. 55 shows that a Greek felt the Pandroseum to be a part of the Erechtheum, for within it was the sacred olive. It had no roof, but was doubtless enclosed by a wall on the north and west, just as it was shut
40 John Williams White.
its current name was 6 émo6ddou0s.1 Aristophanes, Demosthenes, and Lucian could so speak of it, without danger of confusion. There was another opisthodomus on the Acropolis, but when this was meant the speaker said 0 dmicOodopz0s trot Tlapfevavos or used an equivalent expression.’
This solution, which we owe to the insight of Ernst Curtius,®
in by walls on the east and south. There was direct communication between the north portico of the Erechtheum and the Pandroseum by a separate doorway, and the north portico was centered to the larger structure whose groundplan included the Pandroseum. The Opisthodomus was, quite exactly, ‘behind’ the temple, conceived in the sense seen in Herodotus.
1 Frazer (Jour. Hellen. Stud. 1892-93, XIII. p. 162 f.) thinks it remarkable, if the Opisthodomus was in use as a treasury after 480 B.c., that the first mention of it occurs in two decrees of 435 B.C., and draws the inference ex silentio that there was no Opisthodomus before the completion of the Parthenon. This apparently remarkable fact is accounted for by our lack of documents for the time between 480 and 435 B.c. The name Opisthodomus for the Athenian treasury does not indeed occur in any public document during that time, nor does any other. In fact we have only a single reference in all of these years to any f/ace on the Acropolis where money was kept (C/A. Iv. 1, quoted on p.12 and discussed on p. 45), and yet we know that the sacred and public money-treasure housed on the Acropolis was greater between 454 and 435 B.C. than ever afterwards. Frazer himself says (p. 162) that in this time the Athenians must certainly have had some strong place in which to store the public and sacred treasures, but comes simply to the conclusion that we do not know where this was. In like manner, the fact that the first literary mention of the Opisthodomus occurs in Aristophanes is not proof that it was not in existence during the whole of the fifth century. No author before Aristophanes whose works have come down to us had occasion to mention the exact place where the sacred and public money was kept. It would be as logical to conclude from the silence of these authors that there was no treasury at all as that it was not called 6 dwicOé5ou0s. Knowing that there was a treasury we are entirely justified, in the absence of all counter-proof, in concluding that the name by which it was designated in the last third of the fifth century and in the fourth century attached to it also in the previous time. I cannot agree with Frazer, further, in thinking that in the two decrees of 435 B.c. (C/A. I. 32, A, B, quoted in part on p. 16) the Opisthodomus is mentioned as if it were now for the first time to be used as atreasury. The references to it there (it is called simply 6 éric6dd0u0s) seem to me, on the contrary, to imply that it was a well- known place whose use was already established. See p. 46 ff.
2 Plut. Demet. 23, Aristides, I. p. 548, 14, Dind., quoted on p. 15.
8 Curtius, in the November session of the Archaeological Society of Berlin, 1890 (see Archacologischer Anveiger, 1890, p. 163): “Der alte Tempel nach der
The Opisthodomus. 4!
explains at once the same Opisthodomus in its application to the Athenian treasury and some apparently but not really contradictory testimony of the scholiasts. The tradition that this treasury was once in fact the rear chambers in the cella of an actual temple seems certainly to have been preserved during the centuries that followed the invasion of Xerxes. The name dmoOddon0s would serve to keep the tradition alive. Aristophanes and Demosthenes undoubtedly knew what the Opisthodomus was and were acquainted with its history ; and the interpreters who first explained their references to it must, with the great resources at their command, have had knowl- edge of the truth on so important a matter. Schol. V Arist. //ut. 1193' tells us that the Opisthodomus lay behind the temple of Athena Polias. The sources of the existing scholia on Aristophanes are acknowledged to be excellent. We cannot reasonably refuse to believe, however much we may regret the mutilated form in which the rest of the scholium has been transmitted to us, that the original author of the statement just quoted (Aristophanes of Byzantium or one of his pupils?) knew what he was writing about. He knew that in the time of the poet Aristophanes the Opisthodomus was a separate building, situated, as is here recorded, behind the temple of Athena, and he knew also why it was called émia@ddop0s, not because it lay behind her temple, but because it had once been the com- ponent part of atemple. But had the scholiasts and lexicographers whose comments and definitions have come down to us any knowl- edge of the true tradition? It is impossible to say. Two facts are here most important to note. First, it is certain that they knew what the normal opisthodomus was, namely the rear part of a build- ing. Temples with opisthodomi were extant for a long time. We have mention of the opisthodomus of a temple in so late an author
Zerstérung durch die Perser wurde nur in seinem Hinterhause wieder aufgebaut, um als Schatzkammer zu dienen. Es sind zahlreiche Zeugnisse vorhanden nach denen der Opisthodom ein selbstaindiges Gebaude der Akropolis gewesen sein muss.” See also Stadtgeschichte, 1891, pp. 132, 152. I came independently to the conclusion that the Opisthodomus must have been a separate building, convinced by the testimony of the scholiasts and lexicographers, but erred at first in sup- posing that it lay on the lower ground due west of the Erechtheum. Curtius’s suggestion has been received with favour. 1 Quoted on p. 4.
42 John Willams White.
as Achilles Tatius.' What we should expect to find, therefore, in the scholiasts and lexicographers is the definition of the normal opisthodomus. Secondly, what we do find in them in the main is something very different. I call attention again to the passages quoted on pp. 3, 4 of this article. Mulchhofer points out that it is remarkable that these scholiasts and lexicographers do not give us the definition that we should expect.2_ They define the orcOodop05 as a building that lay omo@e rot veo, they do not define it as ro GrioGey pépos rou vew. It 1s common to discredit these late writers, and to say that they ‘etymologized,’ but etymologizing here would have given the second definition, and that definition would have exactly described the sort of opisthodomus they knew about. Their persistent statement. therefore. that the Opisthodomus lay behind the temple of Athena, must be the record of a fact. This fact they had inherited from a trustworthy source. It 1s no paradox to say that it becomes even more credible as a fact, so contrary is it to what we should expect them to say. on the assumption that they did not fully understand it.
This statement that the Opisthodomus lay “behind the temple of Athena” appears persistently in slightly varying form in nearly all the ancient explanations of the ériodou0s that have come down to us. Sometimes there is added a brief definition of the normal opis- thodomus.* This should occasion no surprise. And sometimes there is evident confusion. which either results from the ignorance or carelessness of the writer or is due to too brief statement or to defective transmission of the text. This also should occasion no
1 Achil. Tat. Itt. 6.
2 Philol. 18a4, LIL. p. 359-
8 To the scholium on Arist. P/et. 1193 just cited is added after the main explanation the statement. ror érioGddouoy: 1rd SrurGer rod olxov, Fyovwy Tow race. Cf. the other brief scholia cn Arist. Plut. 1193: Td bwisGer rod Séuov. Dv. rode bwisGer rou vaow. Par. 2827. And on Arist. Plut. t1gt: ro» beusder olxor ris Geos. LB. So Photius s.r. érw@ddouos says first rd brurdey ravrds oichyaros, and then adds the alternative statement that the Opisthodomus was a state treasury on the Acropolis that lay SrwGer rou ris 'AOyyas ved. The brief definition ra Swisbey wavrds oix}paros is found also in E. M. 5.2. drurOédoues and Bekk. 4 mec. 1. p- 286, 26.
* Cf p. 4, note 2; p. §, note I. We find, as I think, an instance of confusion
The Optsthodomus. 43
surprise, although we must regret it, for it opens a field for doubt and discussion. Still, making due allowance for all difficulties of this sort, it seems certain that these late writers do contain a record of the true tradition about the Opisthodomus.
There remains, finally, a scholium which merits special considera- tion, because of the use that has been made of it and because, as it seems to me, it stands quite apart from the rest in its meaning. Schol. V Luc. Zim. 53! reads: ws tepdovdov SuaBddAXAwv trabrda pyow: o yap dmicOddopuos lepov: To dxiobey St Tov advrov ovtws éd€yero.” The point of view of the scholiast should here be noted. His mind is dwelling solely on the act of sacrilege. He adds the statement ro omobey S¢ rov advrov ovrws éA€yero in explanation of the previous predication tepov. He is thinking of the regular opisthodomus of a temple. Such an opisthodomus was a sacred place, because it stood behind the sanctuary of the god.
Boeckh® bases his belief that the Opisthodomus was the western chamber of the cella of the Parthenon on the scholium on Luc. Zim. 53- In two other passages to which he refers he thinks that vads is equivocal and must be interpreted by the single scholium on Lucian. This ignores, as we have seen, the established use of the word vads in its application to temples of Athena on the Acropolis.‘ The existence of the Hecatompedon, brought to our knowledge by Dorp- feld’s brilliant discovery, was not known to Boeckh. If it had been, he would not have maintained with such vigour against Osann® and
of facts in E. M. s.v. dwwOb8ou0s and Bekk. Amec. I. p. 286, 26. Here two defini- tions are confused, that of tke Opisthodomus (the Athenian treasury) and that of the normal opisthodomus (of any temple). The form of statement here should be compared with that in Photius. Schol. RY Dem. xxIv. 136, which, under the lemma raylat, reads, }» 3¢ 6 dricOb8opu0s 7d Exiorber Tis Oeod, leaves us in doubt what the scholiast means. 1d brisOer rfjs beod may signify the place behind the goddess in the sense of behind her temple quite as naturally as that of dehind her statue.
1 Quoted on p. 14.
2 The variant on this reads: Sri 4 dwirOb8opos lepby, 7d SwieOey To ddvrou ovrus Breyor, dy @ xal ra Snpbowa dwréxecro yphyara. (This should probably be: dr 6 dria bbdopos lepby- rd SrirBer (3e) xr.)
3 C/G. 1. p.177 f.
4 See p. 7 ff.
5 Sylloge Inscr. Ant. Graec. et Lat., 1834, p. 62, note 13.
44 John Williams White.
“Britanni nonnulli”? the thesis that there was no other opisthodomus on the Acropolis than that in the Parthenon.’
If the conclusion that the Opisthodomus was a separate building and that it consisted of the three western chambers and western portico that before the Persian wars constituted the western half of the cella of the Hecatompedon is correct, it should be confirmed by a consideration of the uses to which the Opisthodomus was put. If it fails, as thus constituted, to account adequately for any established facts, the conclusion is in so far invalidated.
The scholiasts and lexicographers, in the first place, call it a Tapuciov OF GncavpodvAdxiov. In it, by their testimony, were housed ra xpjpara, both the sacred treasure, ra iepa ypypara, ra xpypara Tov Gewy, and the public treasure, ra Snpdoia xpypara, To Sypociwov apyvpiov Kal o dpos.
Our earliest documentary proof of the existence of the Hecatom- pedon is the celebrated inscription already referred to.* This inscription names the rapia frequently,‘ and, although much muti- lated, evidently contained important prescriptions of their duties. Among these is specified ra oixyjpara [ra év ro éxar jopwredw avotyay [rots] rapias (11.17, 18). It is generally agreed that the chambers here referred to are those in the western half of the cella and that they were treasure-chambers.’ These are the rooms which, accord- ing to the conclusions to which we have come, constituted the treasury referred to as o émoGodopos in later times.
1 Dodwell, Classical Tour in Greece, 1819, 1. p. 345; Wilkins, Atheniensia, or Remarks on the Topog. and Buildings of Athens, 1816, p.g8, note. See also Stuart and Revett, Antiquities of Athens, 1825, i. p. 26, note d; p. 29, note h. Michaelis has a remark about the Englishmen: “ Die friiher vielfach beliebte Unterscheidung des Opisthodoms im Parthenon und eines zweiten Opisthodoms als selbstandigen Staatsschatzhauses hat jetzt nur noch historisches Interesse”’! (Der Parthenon, p. 27, note 85.) I must confess, with shame, that I knew nothing about the views of the Englishmen until my own views on the question were already formed and expressed in writing.
2 Sed nullus alius fuit usquam: opisthodomus est solius Parthenonis.” CIG. I. p. 177 f.
3 See p. 2, note 8, end.
#1.b (?), r(?), I. 3, 8, 13, 16, 18, 25.
§ See p. 39, note 2.
The Opisthodomus. 45
The inscription quoted in part on p. 12, which in date falls between 480 and 460 B.c.,' records the fact that treasure was kept in a precinct behind the old temple of Athena. The zepiéBodos here mentioned, if the lacuna has been properly supplied, was that of the Hecatompedon, in which at the time of the decree stood the restored treasury, and in this treasury the money in question must have been kept. That it was safely housed is certain, however general the phrase [éw wepsB]odo[«] may be; the treasure was not kept in the open.’
After the Persian wars to the time of the completion of the Parthenon the treasures iz 4ind of Athena must have been stored partly in “the old temple,”® partly in the treasury. In the latter was also the sacred money of Athena, and, as we have seen,‘ likewise that
1 See Dittenberger, 5/G. 384.
2It is by no means certain that [sep:8}édo[«] is right. The phrase [é» wepiB}édo[t] gives a suspiciously vague designation of the place where the money was kept. Curtius (Stadigeschichte, p. 132, note) conjectures [é» rec 6]édo[c], but this falls short bya letter. [6ri@]ev is Dorpfeld’s conjecture (Afstth. d. /nst. Athen, 1887, XII. p. 39), but he there fails to take account of the preceding odo. Kirch- hoff (C/A. Iv.1) and Dittenberger (S/G. 384) read [»éro@Jer, but this does not affect the conclusion which we have reached. See p. 39. The original of the remnant odo perhaps yet remains to be found. If the supposition expressed above, that the money referred to was housed in the restored treasury, is true, and if it is also true that the inscription read rep:Pbry, the reason why the place where the money was kept is mentioned in this general way, and not specifically by name, must be a matter of conjecture. It then becomes important to note that the inscription belongs to a time of confusion at Athens, just after the Persian wars, when it is quite possible that the name Opisthodomus, by which presumably the treasury was known before the destruction of the temple and which later became its fixed designation, had not yet been officially attached to this remnant of the temple. It must have seemed odd to the Athenians at first to designate as an opisthodomus a building which was now in fact a single struc- ture, although originally it had been, as part of an actual temple, a real opistho- domus. This seems a simpler and more natural conjecture than that in wrepPdrA we have reference, as Frazer conjectures (Jour. Hellen. Stud., 1892-93, XIII. p. 162), to a building temporarily erected to house the treasures till the permanent treasury was ready.
3 Some of the Persian spoils were still there in the time of Pausanias (I. 27. 1). For the proof that the old temple (Erechtheum) was rebuilt after the Persian wars, see p. 23, note I.
* C/A. Iv. 1 (p. 12) is a decree relating to the Eleusinian goddesses.
46 Join Williams White.
of some of the other gods. On the completion of the Parthenon the treasures in kind were transferred, 1s is well known. to the Pronaos, Neos Hecatompedos, and ~ Parthenun”™ of the new temple’ In $35-4 Were passed the two ceriebrated decrees C4. L 32 A, B) which brought all of the mnances of the state into order. In them we clearly perceive the influence of Percles. under whose careful financial policy Athens was preparing herself for the great struggle with Sparta that was to foilow. The provisions of these two decrees are met with singular fimess on the supposition that the public and sacred money was housed in the restored Opisthedomus of the Hecatompedon.
The monev now stored in the treasury had beccme 3 great sum (A 21%) This fact is confirmed Dy the testimony of Thucydides.* The amount of coined silver on the Acropoiis at the time when the decree was passed was gy0c tients.’ This included both the state-
+ Bat not the monev. Frazer's argument ( /uwr. Aeiicn. Stud. 1832-03, XIIL p- 153 £) for making she western purtico af the Parthenon the Opisthodomus (in brief. thar the treasury documents name four compartments where treasure was stored, and that if three of these were in the Parthenon the natural uference is thar che fourth, namelv che Opisthodomus. was also there) overlooks the important fact that the objects stored in these three compartments were essenually ditferent from the contents of the fourth. We have no evidence that anything except money was stored in the Opisthodomus from the completion of the Parthenon to the end of the Peioponnesian War. It was a treasury. and 3 treasury in which was kept a great sum of money. Against this view. moreover, is the testimony of Plutarch (quoted on p. 15). who tells as distinctly what the opisthodomus of the Parthenon was. Nobody supposes that Demetrius was lodged in the western portico.
* Thac. [4 [ 3.
+ Kirchhoff. Gescaseate ces Athensscten Staatssctutees, 1370, p. 22: Dittenberger. SiG. t4, note t.— The fact of the existence of this zreat treasure seems to me to be fatal to che view thar the Opisthodomus vas the western portico of the Par- thenon. (See Dorpfeid. who formeriy held this view. and Frazer. as cited on p. +) This portico was aitogether inadequate for the purpuse It is pertinent. further. here to add that it was open to view on three sides. where bronze zranngs between the columns were all thac barred a thieving pubiic. Can we believe, moreover, thar this great treasure was cramped into this smail space. and vet that the large room beyond it, the ~ Parthenon ™ in the limited sense. was left practically empty, containing, as the treasure documents show (C7/.4. 1. 101-173), only a relanvely - nal] number of articles emploved at festival-ume (* einige Dutzend Alinen und Scihle,” Petersen. Witts. 1887. XII. p. 4Q)?
The Opisthodomus. 47
reserve and the treasure of the goddess. The existence of so great a treasure in 435 B.c., which must have been the accumulation of years, necessarily implies the existence of a place for storing the money before the completion of the Parthenon.
It is noteworthy that the decrees assume certain facts. The fact of the existence of a treasury is taken for granted, just as that of the boards of the hellenotamiae, logistae, and rapia ray lepoy roy rns ‘A@yvaias. The treasury is named three times (A 15, 17, B 23), simply as 6 dmoOddopnos, in such a way as to imply that it was a well-known place in established use. The specification, further, of the duty of the hellenotamiae in B 18 ff., éx rav dopwv xararcOévar xara Tov énavrov Ta éxdorore yevopeva rapa Tois Tazlace Tay THs “AOnvaias, simply recognizes and emphasizes, as Kirchhoff has shown,! a previous practice. The one important new provision is the estab- lishment of the board of the rapia rav adddAwv Oewy (A 13 ff.). This necessitated certain rearrangements in the use of the Opisthodomus. From this time the tamiae of Athena are to store her money-treasure dy rm emi Sefca rov éricGodduov, the tamiae of the other gods év rq én’ dptorepd. This must mean, as Dérpfeld has already pointed out,? the room to the nght and the room to the left in the back part of the Opisthodomus. But since we know that there was a third treasure, and that it was large, which although in the care of the treasurers of Athena was still kept separate, we cannot but con- clude that it was stored in the larger chamber that lay in front of the two smaller chambers. It had probably been here from the first establishment of a state-fund. This use of the larger chamber explains the provision in A 15 ff. (quoted on p. 16). This provision, in which the words dca Svvarov xai dovov imply a limitation, means, as I think, that the rapiéa: roav dd\Awy Gedy are not to have access to the chamber in which the treasure of which they are in charge is stored except in company with the rapia: ruav ris “A@yvaias, not that they are always to be present when the other rooms are opened. To reach their own chamber they were obliged to pass through that set aside for the reserve fund of the state, which was in charge, under the authority of the state, of the more ancient and much more
1 Athen. Staatssch. p. 33- 2 Mitth. d. Inst. Athen, 1887, X11. p. 3.
48 John Williams White.
important board. It must not be forgotten that the state exercised absolute control over all of these treasures, although it employed the form of a fictitious loan when it drew upon the resources of Athena and of the other gods. The outward symbol of this author- ity was the key of the treasury held by the émordrnys rav rpvravewy, of whom Aristotle says,? rypet 5° otros ras Kreis ras Trav lepav? ey ols ra xypypara dori Kai (rd) ypaypara ty woka. The custody of the key did not imply responsibility for the actual management of the funds.
The view here advanced, that the Opisthodomus of the authors and of inscriptions was the Opisthodomus of the old Hecatompedon, rebuilt without peristyle and eastern chamber after the destruction of that temple in the Persian wars, is not accepted by Milchhofer. In his able and searching discussion of the Opisthodomus,® he first combats the identification of ‘ Parthenon” and Opisthodomus, a view that at the time of the publication of his article had just been reasserted by Furtwangler; he then advances positive argu- ments to sustain the proposition that the Opisthodomus was a separate building; and finally he states what he believes its prob- able situation to have been. He would place it at the eastern end of the Acropolis, where are remains which were once supposed to be those of the chalcothece.‘ His view as to the situation of the Opisthodomus has been accepted by Furtwangler.° To me it seems to be untenable, for the following reasons.
1 Aristot. Resp. Athen. 44.
2 The first of the two lepd here meant is the Opisthodomus. (The other was the Metroum, down in the town.) Cf Poll. vitt. 96: &xee 5é otros rip lepdy ras kets év ols rh xphuara xal rd ypdypara. Eust. Od. XVII. 455: rds re xdets (rap lepdv Rose) év ols rd xphuard elor pudrdrre xai 7d ypdupata ris wédews. Cf. also Suidas and E. M. s.v. éreordrys. Less exactly the argument to Dem. XXII. p. 59o. See Wachsmuth, Stadt Athen, 1890, I. 1, p. 338.
8 See p. 1, note I.
4 Where now stands the workshop of the Acropolis Museum. See the plan in the AeArlor ‘Apxatodoyexdy, 1889, nos. 16-19. ‘ Diese Oertlichkeit,” Milch- hoéfer says, ‘‘ wiirde allen Voraussetzungen entsprechen, die wir fiir das Schatzhaus ermittelt haben.” PAé/o/. 1894, LIII. p. 361.
& Mast. Greek Sculp., 1895, p. 425 f.
The Opisthodomus. 49
It is contradicted, in the first place, by such indications of the Situation of the Opisthodomus as we find in our authorities. They tell us that it lay behind a temple of Athena. No building at the eastern end of the Acropolis would be dchind either the Parthenon, or the Hecatompedon, or the Erechtheum. In placing the Opistho- domus at the eastern end of the Acropolis, Milchhofer thus rejects the testimony of the scholiasts and lexicographers. He thinks that these late writers are of doubtful authority in this question, but nevertheless, in arguing for the Opisthodomus as a separate building, calls attention to the remarkable fact that no one of them gives what we should think to be the most natural interpretation of the word dmiaOcdopos. They define it not as the rear part of a temple, but as ériow (Griobev) rov vew, Tov iepod, also omobey trys Oeod; we find no such expression as ro omoOev pépos rov vaod or omobey év re vag.} It should here be noted that in the very phrase to which Milchhofer calls especial attention they say unequivocally that the Opisthodomus was dchind thé temple. It was doubtless his feeling for the force of their testimony that led him at first to seek to place the Opistho- domus actually dchind a temple of Athena, namely to the west of the Parthenon.? He shows the same feeling for the force of dmoev in dealing with the early fifth century inscription already quoted.® He indicates his preference for the reading [émo6]ev, and adds that the peribolus with its treasury constituted “dann schon eine Art ‘Opisthodomos’ des alten Tempels.’’* This is full recognition of the fact that this earlier ‘sort of opisthodomus’ at least lay dehind the temple. But the other opisthodomus that followed, #4e Opistho- domus, he puts behind no temple.’
1 Philol. 1894, LIT. p. 359.
3 Jbid. p. 360.
3 See p.12. For its interpretation, see p. 45.
‘ Tbid. p. 357, note 8.
5 Milchhofer arbitrarily makes the inscription cited a terminus ante quem. Yet money had to be housed on the Acropolis both before and immediately after the Persian wars, and he himself believes that “the treasurers of Athena had been installed s# the Opisthodomus long before the Parthenon was built.” (/éid. p. 357.) How long before? To me it seems likely that the inscription refers to the real Opisthodomus, not to ‘a sort of Opisthodomus.’ See p. 45, with note 2.
50 John Williams White.
Milchhofer, as I have said, sought first to place the Opisthodomus behind the Parthenon, but here he encountered a substantial obstacle, the real chalcothece,' and relinquished his first suggestion. He adopts his final view with more confidence, although he says that certainty in the matter is not to be expected. Two considera- tions weigh with him. First, if we should put the Opisthodomus at the eastern end of the Acropolis, it would be an opisthodomus (rear- building) in the sense in which the Propylaea are the fore-court (Vorhof) of the Acropolis. But this is the gratuitous intrusion of a new point of view. In no ancient. reference to the situation of the Opisthodomus is there anywhere mention of the Propylaea; our authorities orient the Opisthodomus not from the Propylaea but from a temple of Athena. It is pure conjecture, therefore, to assume that the Opisthodomus got its name because it was at the ‘rear’ of the Acropolis as the Propylaea were at the front.? In this case, further, the name démiaOodon0s would stand without parallel. No- where in the literature, so far as I know, is the eastern end of the Acropolis called its ‘back part,’ nor is there intimation anywhere that a Greek ever thought of it as that. Secondly, it seems to Milchhofer in itself probable that such a structure as the Opistho-
1 See Afitth. d. Inst. Athen, 1889, XIV. p. 304 ff.
2 It should be noted that the word ITIportAaca does not mean “ front-gateway,” in the sense of gates in front of a space behind, but rather “ that which is in front of the gates,” in this case the wings and portico in front of the five doorways as one comes up from the we&t. The use of the adjective rportAaos establishes this fact. Neither the word IIporéAac nor the structure itself, which lies low and led to the rear of the temples, could have suggested to a Greek any such strong antithesis between the Propylaea and the high ground at the east fronting the temples, where stood the great altar, as lies at the basis of Milchhofer’s assump- tion. The natural means of orienting objects on the Acropolis is, as we have seen, not the Propylaea, but the temples. The probability, further, of such an antithesis as Milchhdfer assumes is weakened by the actual situation which he assigns to his ‘back-building.’ It does not lie on high ground on the eastern continuation of the axis of the Propylaea, but in the extreme south-eastern part of the Acropolis on ground that was lower than that to the north and to the west of it (see Jahn-Michaelis, Paus. descrip. arc. Athen. 1880, Tabula 11.), and with the peribolus of the Parthenon directly between it and the Propylaea. It is doubtful whether it could have been seen, even before the building of the present Parthenon, from the Propylaea.
The Opisthodomus. 51
domus was erected at the eastern end of the Acropolis on the site of a pre-Persian building used as a magazine and for purposes of administration... He would therefore identify the present remains as those of the Opisthodomus. This again is simply conjecture.’ There is a second strong objection to Milchhofer’s view. It contravenes the established application of the word dmicOddopos, which as defined by Pollux and Varro, and in fourteen instances of its use in authors, in each of which its meaning is certain, designates an integral rear part of a@ building’ It has no other application. Nothing, therefore, but convincing evidence should induce us to believe that the word could have been applied to a building on the Acropolis at Athens which always had been a separate structure.
1“ Es ist auch an sich wahrscheinlich, dass hier ein solcher Bau an Stelle vorpersischer Magazin- und Verwaltungsriume aufgerichtet worden ist.” /éid. p. 361. This statement leaves it doubtful where Milchhofer thinks the treasure was stored before the Persian wars.
2 This theory is not supported by schol. Dem. xxiv. 136 and Pollux 1x. 40, noticed on p. 5, note 1 above. Even if we accept Milchhofer’s suggestion that these may possibly rest on better ultimate authority than the rest of the notices (tbid. p. 359), they do not tend to establish his view. The phrase of the schol., olknpya orlaw ris dxpowrddews, “a house behind the Acropolis,” removes the Opistho- domus from the Acropolis altogether. Pollux doubtless means to say the same thing, “the (part) behind the Acropolis.” He here uses xarércy adverbially, as in the same phrase in his definition of onxés, quoted on p. 38.
3 For the definitions of Pollux and Varro, see p. 38. The word drw6d3op0s occurs in the following places in reference to the rear part of a temple: Diod. Sic. XIv. 41; Plut. Demet. 23 (see p. 15); Aristides, 1. p. 548, 14 Dind. (see p. 15); Paus. V. 10. 9, 13. 1, 15 3, 16.1; Lucian Aat.1, Fug. 7, de morte Pereg. 32; Anth. Pal. x11. 223, 4 (Strato); Achil. Tat. 111. 6; in reference to the back part of a house: Appian, de dell. civ. 1.20; in reference to the rear part of a buleuterium: Themistius, Xv. p. 234 Dind. All the occurrences of the word known to me have been cited in this paper. — In the course of his argument Milchhéfer unconsciously makes an important admission. He says (sbrd. p. 356), “ Von vorn herein wird Jedermann zugeben, dass lediglich der offiziell iiberlieferte Name ‘Opisthodomos’ darauf gefiihrt hat, das Schatzhaus fiir einen Theil des Tempels zu halten; auf Grund unserer sonstigen Ueberlieferung hatte man in dem ramefoy nur einen selbstandigen Bau vermuthen kénnen.” The fact remains that the treasury was called Opisthodomus. Of this apparently anomalous fact we find adequate explanation in the view advanced in this paper that the treasury was once an integral part of the Hecatompedon.
§2 Jokn Williams White.
Milchhofer’s sole support of his proposition is the declaration that the usage of Greek speech allows this employment of the compound word, analogously to the use of the word Aiaterhaus in German,' which may mean “a house in the rear” as well as “ the rear part of a house.”’ He adduces no proof. It is hazardous, of course, to say in etymologizing what Greek usage would not allow: but. in cases of doubt, one fact is worth many theories. Here the fact is certain that, as has been shown, the word according to its established application did mean fhe back part of a building, not a back building. Nor is it difficult to see precisely how the word oxioOddop0s came to have its established meaning. It has good Greek parallels, and in such a discussion as this Greek analogy carries much more weight than modern. Precise parallels are xpovaos, xpodopos, and xpooxy- nov, the etymology of which has already been discussed in its bearing on another question.? Etymologically, érccOcdopos signifies “the part behind the depos” (ro oxiabev rov dopov pépos), where Sopos signifies the main structure to which the émoOodopos is added. So xpddopos, “the part before the Sopos.’”’ and zpovaos, “ the part before the rags,” at the time when vads and aévroy were identical and signified all that there was in the way of a ‘temple.’ Then the idea conveyed by the second part of the compound was enlarged, so that Somos and wads came to include the zpodopos, xpovaos, and omofodop0s. Thus Homer can speak of the xpodopuos of a dopos, and Pausanias and Strabo of the zpovaos of a vaos.* So dmafddoxos came to mean “the back part of the dopos,”’ where dopxos signifies the whole structure. This is precisely the general definition of éxicOodoun0s given by the early lexicographers, namely ro GmiOev xavros oixyjparos.4 The émoGc- Sonos, mpodozos, and zpovaos were, then, integral parts of a whole, but what this ‘whole’ is, is defined with absolute certainty by the second part of the compound. A Greek could speak of the éroGe- Sopos, rpodSopzos, and zpdvaos of a house or of a temple ; but his sense, it may be an unconscious sense, of the etymological force of the words would have made it impossible for him to speak of the éw:OdSopos
1 /bid. p. 356.
2 See p.7, note 2.
3 Hom. Od. tv. 302; Paus. 1x. 4. 2; Strabo xvil. 28, p. 805.
4 Photius, E. M., s.v. dwurObdou0s ; Bekk. Anec. I. p. 286, 26. See p. 42, note 3
The Optsthodomus. 53
of an acropolis or of any ‘whole’ not indicated by the word itself. It would be as forced to suppose that the word éma6odop0s was so applied, meaning a rear-building of the Acropolis, as that arpddopos might mean @ front building of the Acropolis or mpovaos a front temple of the Acropolis. There is no shadow of authority for saying that the last two words were ever so used.
The case, then, may be summed up as follows against Milchhofer’s theory. First, it contravenes the ancient evidence, and assumes a Greek point of view for the orientation of buildings on the Acropolis which cannot be established. Secondly, on trustworthy testimony, the Opisthodomus (6 dxiaOddopos) in which the Athenians stored their treasure was a separate building, as Milchhofer himself believes ; but Greek usage of speech allowed the word démicOodopos to be applied only to a structure that was, or at some time had been, the integral rear part of a building. ‘ Ze Opisthodomus,’ therefore, must at some time have been an integral rear part of some building. The theory advanced in the second part of this paper is that this building was the Hecatompedon.
A VOTIVE TABLET TO ARTEMIS ANAITIS AND MEN TIAMU
IN THE BOSTON MUSEUM OF FINE ARTS.
A VOTIVE TABLET TO ARTEMIS ANAITIS AND MEN TIAMU IN THE BOSTON MUSEUM OF FINE ARTS.!
PiLaTE II. By JOHN HENRY WRIGHT.
I.
F the history of the votive tablet, figured on the accompanying plate and discussed below, nothing seems to be known, except that the slab came “from the Levant,” and has been in Boston for forty or fifty years.? It is now in the Museum of Fine Arts of that city, in the Room of Classical Antiquities ; its number on the register of acquisitions is S. 1695.° The dimensions are: greatest height, M. .72; width, M. .42; thickness, M..06. It is of coarse white marble, “ probably Asiatic.” ‘ On the upper part of the front surface there is a panel upon which are represented in low relief a man, child, woman, and second child (the last fragmentary), standing in a row from right to left, en face, each with the right arm raised from the elbow, palm turned out, and thumb separated from the fingers. The attitude is the familiar one of adoration.’ The man and the boys wear the short
1 The substance of this article was read before the American Philological Asso- ciation, July 12, 1894, at the Williamstown meeting.
2 My best thanks are due to Gen. Charles G. Loring, Director of the Museum, and to Mr. Edward Robinson, Curator of Classical Antiquities, for calling my attention to the tablet as soon as it was received, and for furnishing me with facilities for publishing it. A preliminary notice of the tablet appears in Mr. Robinson's Report to the Trustees for 1893, pp. 19, 20.
2 It was for many years exhibited among the curiosities of the old “ Boston Museum,” a hall attached to a theatre. In 1893, when the collections of the “ Museum ” were broken up, it came into the possession of Mr. Charles A. Cum- mings, who, on December 20, 1893, gave it to the Museum of Fine Arts.
4 Mr. Robinson.
5 Compare Sittl, Die Gebdrden der Griechen und Romer, pp. 147 ff. The atti- tude is as in Waddington-LeBas, Afonuments Figurds, p\.137, and on many
56 J. H. Wright.
sleeveless chiton and the long himation, the latter thrown under the right arm and over the left shoulder. The woman wears a short himation similarly thrown, and the long chiton. What appears to be an object held in the left hand of the boy in the middle is perhaps only a piece of the untrimmed marble,’ and the tassel-like appendage near the left hand of the woman is nothing more than the rudely-cut end of the himation. In artistic character this sculp- tured relief is crude, stiff, and lifelessly conventional. The design, as often on tablets of this sort, does not exactly fit the inscription, which speaks of only one child. . Hence we may infer that the slab with its relief was not made to order, but was selected by the devotees from a stock of ready-made stones as the one coming nearest to their needs.
Above the panel was probably once an ornamental coping, which has been sadly mutilated.? The left edge of the slab is likewise incomplete, being broken off from top to bottom.
Below the relief stands the inscription, in letters of the style current in Asia Minor, and especially in eastern Lydia, about 200 A.D. ‘ The inscription is perfectly preserved, except that each line lacks at the end from two to four letters.
"Apréwdse *Avaeire x[at Mn] vi Tiapouv Movaais B [kal] Kaddyevera 4 ovpBifos av-] tov wmrép Movoaiou told vod]
5 paptupourres tas S[uva-] pus tav Oeav arrédo[xar] Thy evynv: Erous aira- [un(vOs)] Aeiov 1.
other monuments, especially from Asia Minor.— Cf. Plutarch, Vist. Mar. 26, nutaro 8 xal Kdrdos dyuolws dvacxay ras xetpas xabepicay Thy rixny ris hutpas éexelyns, where xa@tepdcev is used in a specific sense, like d»éornoay, similar to that pregnant in droducey rhy ety jp.
1 On some reliefs of a similar character the adorants carry cakes in their hands.
2 On this coping may have been carved something to represent a part of a human body, the member cured by the divinities. Compare the four eyes carved on the coping above a similar row of figures, in Waddington-LeBas, Monuments Figurés, pl. 137 (no. 688, from Goerdis or Julia Gordus), and see below p. 72, nos. 6 and 7.
A Votive Tablet. 57
“To Artemis Anaitis! and Mén Tiamu: Muses, son of Muszs,? and Calligeneia his consort, on behalf of Muszs their son, in testimony to the powers of the gods, have paid their vow. In the year 281, the roth of the month Dius [a.p. 196#?].”
1 The Anaitis inscriptions known up to 1886, ten in number, have been col- lected and printed by S. Reinach, Chroniques d’ Orient, U1. 1885, pp. 105, 107 f.; 1886, pp. 155, 156 (from the Revue Archéologique, with additions). [According to Leemans’s facsimiles, on the stones from Kula in Reinach, Chronigues, 1886, p. 156, No. 2 (vA8), the date should be ov8, not od, and an ‘Avatr: should be inserted between Mnrpf and ’Afwrrnyp; and in no. 5, the date is ord, not opd.] Of these ten ’Avdirs inscriptions, one is from Hypaepa, near Odemish, undated ; two, undated, are from Philadelphia; the remaining seven are from Kula or vicinity, three undated, four dated respectively A.D. 160, 200(?), 235 (236), and 236 (237). To this number should now be added no. 11, our inscription, dated A.D. 196; nos. 12, 18, the two inscriptions from Kula (Macovla) published by Contoleon, Athenische Mittheilungen, XII. 1887, pp. 254, 255 (cf. E. L. Hicks, Classical Review, (Ul. 1889, p. 69), dated respectively A.D. 215 (216) and 244 (245); for no. 12 see p.72, no. §; no. 14, undated inscription from Philadelphia (not fram Tralles), Contoleon, Athenische Mittheilungen, X1V. 1889, p. 106. — Possibly also belong here (as no. 15) Michaelis, Journ. Hellen. S., V. p. 154, the proper name Ilar-dvaaé, ‘gift of Anaitis’(?); and (as no. 16) Waddington-LeBas, no. 6992, from Kula, undated: I would emend Myvoyévns beg AAIANH in it to 6e¢ ANAITI(H). The inscription has not been tested by a squeeze. To be sure there is a town Alia in Phrygia, but on coins we have AAIHNON.
Seven of these Anaitis inscriptions are now in Leyden, and have been published in facsimiles by C. Leemans, Verh. d. k. Amsterdam Akad. XVI. 1886 (1888), no. 7: Leemans also gives, from Reinach, the transliteration of all the remaining ones, except nos. 1 (from Odemish), 11, 14, and 16.
It will be noted that of these inscriptions, all of which the provenance is known, except one from Odemish, come either from Philadelphia (three), or from Kula or neighborhood (ten). — For Mén Tiamu inscriptions, see note on pp. 71-3.
2 On the spelling of this name in this inscription see the remarks on line 2. For Moveais (from Moveaios), Latinized Muses, cf. Barnaes = Barnaeus: Benseler, Curtius Studien, III. pp. 167, 168.
® Throughout this article it is assumed, in the reduction of dates, that the era of Sulla—which seems to have begun about July 1, B.c. 85, and not Sept. 24, B.c. 85, or 84 — was followed in Upper Lydia, as well as in Phrygia. Cf. Mar- quardt, Rom. Staatsverw., I. p. 337, and, for the date, Ramsay, //istorical Geog- raphy, p.452. (Where two dates are given in the reductions, the second must be adopted if the era is assumed to have begun Sept. 23, B.c. 85.) In the sequence of months Dius came first; Peritius, fourth; Dystrus, fifth; Xandicus, sixth; Artemisius, seventh ; and Daesius, eighth. — [But see below, p. 73.]
58 TH. Wright.
IL.
Judging from many other inscriptions similarly dedicated,! we may infer that one Musaeus, — who seems to have pronounced his name Muszes,— and his wife Calligeneia, on the serious illness of their son, or in some other grave stress, had made a vow to Artemis Anaitis and Mén Tiamu that if the boy were by these gods restored they would dedicate to them some aazathema. To the object thus dedicated they attached this tablet which commemorates the might of the gods named in it. |
The restoration and interpretation of the inscription offer no difficulties.
Line 1.— ANAEITL, in which EI has the value of I, and ree rersa, must have been felt as a dative for “Araires (perhaps ‘Avaira, cf. rdAcs woAa), though the regular form of the word would be ‘Avaircdks. So far as I have observed, "Avce‘r: does not occur except in a dative construction, and thus may be regarded as a dative form. Hence I would accent "Avaars not "Araar:, nor ‘Avadri. The following case- forms are found on the stones (a usually replacing the first .): nom., "Avaares: gen., Avairidos, “Araciris (sic: for "Avaeirios?): dat., "Avatredk. "Avuere, “Avae(d)re: acc., Ardariy.— The lost MH of [My]vi was prob- ably written as a monogram,” MI, or perhaps «[ai] was abbreviated ; there is not room for both words written out in full.
1 Anaitis (Artemis, Ged, seyd\y, whrnp, prefixed) is, in many of the dedicatory inscriptions, gratefully remembered for services of healing. as is also Mén Tiamu. Cf. Reinach, Chrunigues, thzg¢. Mén had a character in some respects like that of Asclepius. Thus Mén Caru had a medical school attached to his temple at Attuda (Ramsay, dmertcan Journad of Archacoiegy, IV. p. 277). and near his shrine at Ménos Come, between Carura and Attuda, were hot springs: Athen. IT. 434. The coins of Parlais in Lycaonia have as types not only the god Mén, holding the pine-cone and with cock at his feet, but also Asclepius and Hygieia : Head, Histuria .\Vaemorum, p. 590.
The Charite dedication, — Reinach, Chroniyues, 1886, p. 156: Aprémad Avac(fri Xapiry | ‘Aro\\wviov, wepleraya | cyotea cai Hasteica ord ris lepeias, edy dy. ‘To Artemis Anaitis, Charite. daughter of Apollonius, having had an accident, and having been by the priestess restored through exorcisms and incantations, has paid her vow,’ — when compared with Strabo, XV. 3. 15. p. 732, where mention is made of the priests of Anaitis (¢&¢doverr wpas), suggests something of the methods employed by the servants of the goddess in fulfilling her petitioners’ desires. Cf. Paus. V. 27. 5: drip adyos ... drdde: SdpSuga .. . driNeydmevos dx PiSXov.
2 Cf. MHNI in Waddington-LeBas, no. 668 ; also below, p. 72, no. 5.
A Votive Tablet. 59
Line 2.— We must not write Mymriduov as one word, with Boeckh, G. E. Benseler, Leemans, etc. The parallel forms, nom. Mav Kdpov, gen. Myvos Kdpov, and Myvos Tiauxov, make this impossible. Cf. Waddington-LeBas, /nscriptions de l’ Aste Mineure, no. 668. — Trdpov is an indeclinable word of uncertain origin and meaning ; see pp. 68 ff. — MOYSXAI® is, of course, not a graver’s blunder! for MOYSAIOX.? Examples of -s for and from -cos are frequent in Greek (and Latin) proper names on stones of the Roman era, and of -és for -éus in Roman inscriptions from the earliest date.* In by far the larger number of examples, which have been collected hy F. G. Benseler, the iota is not part of a diphthong as here, but instances are by no means wanting where -atos (-aeus) and -eos (e245) become respectively -ais (-a2es) and -as (-es). If these cases were lacking, we should either look upon Movgais as a graver’s blunder for Movoatos or be tempted to read Movoais (Musaés), with uncertain accent, as a trisyllabic word ; but with the examples in mind, and for other reasons sug- gested in the notes, it seems better to regard the final syllable as diphthongal, and to treat the word as a dissyllable.* The accent is
1 Some incisions on the slab between I and 2%, which might be taken for an attempt to insert an O, are nothing more than clumsily-cut parts of =.
2 The name Moveaios is very common on the stones. Note from Kula, date 244 (245) A.D., a dedication from one Adp(4dws) Moveatos to Myrpl *Avaelri: Athenische Mittheilungen, XII. p. 255 (Classical Review, III. 1889, p. 69). The encroachment of the Roman praenomen is significant of the later date.
*Cf. F. G. Benseler, Curtius Studien tur griechischen und lateinischen Gram- matsk, 111. pp. 149-183; also Lobeck, Pathol. serm. graec. Prolegomena, pp. 500 ff.; Blass-Kiihner, Ausfihkrliche Grammatik der griech. Sprache, 1. p. 400. This cor- reption is frequent in proper names in Greek, after about 50 A.D., but in Latin is much more ancient. Of the hundreds of examples collected by Benseler, the following occur where -aws has become -as (sometimes written -aes): Adp. das (but this was probably I'dis : cf. F. D. Allen, Harvard Studies, U1. p. 76), "AOhracs, "AOhvais, Elphvais, ‘Eorlass, “Aploras, “Avrats, "Iras, Kfwats, Dluas, Kérracs. The cases of -ewos becoming -ts (-e:s) are ZrpodGew (accus.), "Hdes, and perhaps Kpdreis, Eédpes. Professor Ramsay has seen EPMAIZ, gen. -I0T. Add also ‘Hpais, Waddington-LeBas, no. 678 (below, p. 71, no. 1).—I am not aware that Movcatos has ever before appeared in the shortened form.
4 In favor of the dissyllabic pronunciation, or for -a:s as against -ais, might be adduced the fact (e.g.) that "A@4#»aos becomes not only 'AOjras, but also *AGnvas (cf. ‘“Eppas from “Epyaws, etc.). The form ’Aé@yrds could not have arisen from a * APhvais.
60 J. H. Wright.
less certain. There is a twofold question here: the accentuation of the longer and that of the shorter form. The vulgate and tradi- tional accentuation of the longer form, traceable apparently to Aristarchus,' is Movoaios. But there are scholars who would give such proper names, to distinguish them from adjectival forms of identical spelling, the recessive accent, and would write accordingly Movoaos. Evidently Movoaios would become Movoais. and Movoaos would become Movoaus. The analogy of "A@jwuos (AGjvus ?) seems to favor Movoaos (Movoas); but the modern Greek Bap@odopais (from BapOoAopaios), and the accent of ‘A@yas, “Eppas,? etc., — which could have arisen only from a "A@yvaios, ‘Eppatos (not from "AOyvaos, “Eppaos), in both of which classes of cases the accent remains on the syllable that originally carried it, — speak emphati- cally, and in my judgment decisively, for Movoais.* — -B- after Muses indicates that Muszs is second of the name, Le., is in the second generation‘; thus in Waddington-LeBas, no. 656, we read &s
1 A scholium on //iad E 69, edited by Cramer (Anecd. Paris. III. p. 283) appears to contain Aristarchus’s doctrine on the accentuation of Moveaios and similar words. The matter is fully discussed by Lehrs, De Aristarchi studiis homericis,$ pp. 292 ff. Cf. also Herodian, fassime (Lenz’s index, s. Moveaios); Ellendt, Zex. Sophocleum,* p. 460.— On the accent of the shorter forms of words in ~s (from -os), see Benseler, sbi. pp. 182 f.
2 In’ AOnvas, ‘Eppzas, and similar words, we have a reversion to the accentuation of the adjectival form, the necessity of accentual differentiation — as in ‘A@y}racos, ‘ Athenaeus,’ and "A@yraios, ‘an Athenian’— not being felt where the spelling was different. — Indeed, though "A@}j»aos must stand, is it certain that "A@yjeas is right? Ought we not to make it ‘A@ypais, on the analogy of "A@yvas? And simi- larly should we not change into perispomena several of the proper names cited in the preceding notes ?
$ It would hardly be sound to argue that as correption must take place in unaccented syllables, neither of the two syllables in -ass would have been under the accent, and that we must accordingly write Moveaws and Motwass. In the first place, at the period of our inscription the pitch accent in Greek had not become a stress accent, and it is mainly the stress accent that works havoc with unaccented syllables ; in the second place, there are many examples of correpted syllables the former element of which was once under the accent. _
4 There are numerous instances in inscriptions of Roman date: e.g. A:opéouws B, Waddington-LeBas, no. 617; id., no. 647 ([H&Jo[Sdp]ou 6 rod "H\o[ Supov]), where three persons named Heliodorus are mentioned; #d., 734, 795, Hicks, Greek Inscriptions in the British Museum, U1. 1, 2, p. 213 (no. DLXXXVIII), etc.
A Votive Tablet. 6!
Hla:ruavov, but the sign for the ordinal or adverb is more frequent. Movoais 8 probably means ‘Muses, son of Muses.’ His own son, here referred to in rov tod, might have been designated as Movoais y.}
Line 3.— The H after Ka\Acyévea has been corrected from N. — ovpBios, which is the regular form on the stones, is preferable to ovpBia: there is a distance trace of O on the marble. Compare also Waddington-LeBas, no. 734 (C./.G. 3872).
Line 4.— For Movoaiov rov tov, see Waddington-LeBas, no. 682, from Goerdis (—Julia Gordus). There is hardly enough room for viov. Movoaiov ¥ is out of the question; and for Movoaiov rov + I can find no analogy. In Waddington-LeBas, no. 703 (from Kula), father and mother unite on behalf of their rd wo[Oevjov réxvov.
Line 6. — duvaycs for Suvduets, as in “Avaeire above. In a different sense the word occurs in Waddington-Lebas, no. 668 (from Kula, cited below, p. 72, no. 3).— Tov Oewy are the pair Artemis Anaitis and Mén Tiamu.
Line 7. — érovs oma, or 291, is probably of the era of Sulla? (85 B.c.), which seems to be the one followed in the part of Asia Minor where Artemis Anaitis and Mén Tiamu were together wor- shipped ; possibly, however, that of Actium is used (32 B.c., not 31 B.C.). — The word pyvos, ‘month,’ was probably indicated * either by its
1 An interesting inscription illustrating this usage is Waddington-LeBas, no. 743 (C.7.GC. 3952), Nexlay raida rap ed yeyorérwr vidv Taveavlov 8 rob Awruclov... éxipednr er Ilavoarlov rob [laveazlov y. This gives the stemma:
Dionysius | Pausanias
Pausanias B
Pausanias 7 Nicias Professor Ramsay has noted inscriptions with 8, and even ¢, with a name.
2 Franz established this era as beginning about B.c. 84 (C./.G. III. pp. 1103 f.), the year when Sulla reorganized the province of Asia (Appian, Mithrid. 61). Cassiodorus (Chron. p. 682) confirms the testimony of Appian: His consulibus — Cinna IV et Carbone II (=8.C. 84)—Asiam in XLIV regiones Sulla distribust. Cf. Waddington-LeBas, on no. 980 (Vol. III. p. 254). On the exact date of the beginning of the era see above, p. 57, note 3.
$ Mypts, as M in Waddington-LeBas, no. 683; as IW, id. no. 679; as M, id. no. 702 (from Kula).
62 J. H. Wright.
initial letter or by a monogram.—The month Atos, of the Mace- donian calendar, early adopted by the successors of Alexander for these regions, is the first month of a year which originally began soon after the autumnal equinox. Under the Romans the year appears to have begun ca. August 1. Hence our inscription may be dated in August, a.pD. 196 (or 249?). The relief, however, must be somewhat, if only very slightly, older; as we remarked above, it was already in stock when the inscription was cut.
ITT.
The provenance of the tablet may be placed with probability in the Karaxexavpery district in eastern Lydia, or, in modern terms, the vicinity of Kula, a large Turkish village in Asia Minor, about eighty miles east of Smyrna.’ This inference is based upon the fact that all of the other inscriptions known bearing the names of both Anaitis and Mén Tiamu have been traced directly to Kula or vicinity, not to speak of the large number of stones inscribed with one or the other of these names, which with few exceptions are from this region.? If it were possible to identify the marble of this slab with that quarried between Ghieulde and Sandal® (ancient Satala?), near Kula, this provenance would be established beyond a peradventure.
One is tempted to connect Kula with KoAdy, of which mention is made in classical writers. But apart from the consideration that the name KodAocy is given to several places in antiquity, Professor W. M. Ramsay has shown‘ not only that Kula does not occupy the site of
1 Interesting views of Kula are given by Texier, Description de l’Aste Mineure, I. pl. 50, p. 133, and by Hamilton, 4ssa Minor, Vol. II. p. 136. The volcanic cones, long extinct, that gave its character to the Catacecaumene, are prominent features of the landscape. Cf. Strabo, XIII. 4, t1, p. 628.
2 See p. 57, note 1, and pp. 72, 73. 8 Hamilton, Asia Afinor, II. p. 137.
4 About forty years ago Wagener saw and copied at Kula an inscription bear- ing the words 4 Kodonva@» xaroula, and dated A.D. ror. He published it in the Mémotres del’ Académie de Belgique, Mémoires couronnées XXX.; it has also been published in part by Keil, PAz/ologus, Suppl.-Bd. IT. p. 607 ff., and, more correctly, by Tsakyroglus in the Moveeioy (1876, p. 41; read Aawlov )’, not a’). The identifi- cation of Kula and Kodéy is based mainly upon this inscription, and is accepted by Wagener, De Witte, Waddington, Foucart, Von Diest, and others. Tsakyroglus
A Votive Tablet. 63
an ancient city,’ but also that the name ‘Kula’ is a Turkish word meaning ‘fortress,’ with merely an accidental resemblance to KoAcn, and that it was only in comparatively late times applied to this town. There are, however, as the remains show, several ancient sites a few miles to the north, northwest, and northeast of Kula (notably Sandal), and from one of these sites our slab may well have found its way— probably through Kula—to Smyrna, and thence to the United States forty or fifty years ago, when American trade with the Levant was much brisker than it is now. For many years past marbles of various kinds have been brought in to Kula from the adjoining regions, and are thus registered by epigraphists as from Kula. IV.
Our tablet raises a number of interesting problems, into the detailed discussion of which I will not enter, problems chiefly con- nected with the personality and cult of Artemis Anaitis and of Mén Tiamu.* These very names, on a slab dedicated by Greeks and dated by a Roman date, bear witness to the composite character of the religion of eastern Lydia and western Phrygia in the second century of our era.
thinks that the KoAé» here intended must have been on the site of Sandal, a town near Kula, and that the older name of Kula was IIdpyos, or Ilupyla, of which ‘Kula’ is presumably the Turkish translation. Professor Ramsay, however, has shown that the stone in question was brought to Kula (at least forty years ago), and not from Sandal, but from Injikler, a village in the mountainous Kara Tash district, eight hours north of Kula; accordingly he would place Kodéy in these hills (is- torical Geography of Asia Minor, pp. 123, 211, 432,456; private letter to the writer); so also Kiepert in 1894 (Formae, efc., no. 10). — Perhaps this mountain town may have been a colony from the region of the Myusy Kodén, near Sardis (Strabo, XIII. 4, 5, p- 625), taking thence its name. Though in inscriptions xcaroula often means ‘town,’ with no connotation of ‘colony,’ the latter sense is not always wanting ; cf. also Strabo, V. 4, p. 249. — Mr. Hicks (Classical Review, III. 1889, p. 69) seems to identify Kod\éy and Kula, but strangely enough puts Kula north of Lake Coloe, more than thirty miles from its true position.
1 Arundell (Asia Minor, I. p. 42), visiting Kula more than sixty years ago, before the interest in inscribed stones had caused them to be carried much from place to place, remarks that there is nothing in Kula to suggest an ancient site.
2 On the Anaitis inscriptions, see above, p. 57, noter. Fora collection of the Mén Tiamu inscriptions, see pp. 71-3.
64 J. H. Wright.
There appear to have been early worshipped ' in these regions a pair of mighty divinities of productive nature, orginally perhaps of an agrarian character, a female and a male god, the female regularly taking precedence. The former appears in literature and on inscrip- tions under many names," of which Ma was probably the native and most ancient name, which is regularly replaced by Myryp (with or without Geer). The latter was probably once a solar divinity. though subsequently he was transformed into a lunar god: he appears to have been known most anciently by the native populations as Ma-n, or Maen, though he, like the Great Mother, had also several less precise appellations.* It is possible that these two names in the
1 Professor W. M. Ramsay, than whom no one is better qualified, promises a chapter on Mén in his forthcoming Local History of Asia Minor, of which Vol. I. is now in press. To what he has already published on the subject of Anatolian religion and antiquities, in various archaeological journals and in his Aistertcal Geography of Asia Minor, | owe the germ of some of the suggestions here offered, especially concerning the pre€éminence of Mén in Asia Minor. But for the argu- ment on which this preéminence is based he must not be held responsible.
2 Some of the ancient names of the Great Mother are given by Strabo: Ol 32 Bepécurres, Dpvyav Ts Pidov xal awds of Sptryes cal rSv Tpewy ol repl ry “lony caroxourres ‘Péay yey xal avrol Timisor xal dpyidfover ravry. uyrépa xadouvres Gewr xal “Ayduorw cal dpvylay Oedy peyddny, dwd 8 Ty rérwr “Idalay cal AirSuphrny cal Zerudhrny wal Weoowwovrrlda xal KuSé\yv. Strabo, X. 3. p. 469.
® Stephanus Byz., s.v. Mdoravpa, asserts that among the Lydians Rhea was known as Ma. Cf. also an inscription from Byzantium : “Acca "Awo\\wrlov Myrpl Gesr Mg Movfyry xara eixhy ebyaporipa, cited by Foucart, Associations rels- gieuses chez les Grecs, p. 88.
4 «La religion de la Mere des Dieux avait son centre dans la Phrygie a Pessi- nunte. Elle parait a l’origine avoir consisté dans les rapports des deux person- nages divins, l’un feminin appelé Ma, d’ou peut dériver le nom de M#rnp; l'autre, masculin, appelé trés-anciennement Papas.” Foucart, sid. p. 88. — The evidence adduced for Papas as the ancient or native name of the male god is too slender to be convincing; Foucart cites an inscription of the Roman era (C./.G. 3817), from northem Phrygia, which reads Mawig Ad cwrijp edyhy cal ‘Hpaxdy douctry. Cp. also Ramsay, Journ. Hellen. S., III. p. 124, V. pp. 257 ff. There is, besides, Arrian, Fray. 30 (af. Eustath. //. E 408, p. 565: dxdévres els ra dxpa rar dpay Bidvvol dxédouv Ildwary rar Ala cal “Arti roy atréy), and Diodorus (III. 58. 4) cited below.
The antiquity of the name Ma-n, Maen, Mén, for this god is, however, well attested, as also its local use in Phrygia and Lydia. This attestation consists mainly in (1) the ancient local and legendary names Manes (Herodotus, I. 94), Manis (Plutarch, Jfor. 360), perhaps also Mjwr (from Ma-lwy?), etc., which point
A Votive Tablet. 65
beginning were but variations (feminine and masculine) on the same stem. In due time, however, the native or popular conceptions of the divinities fell under the influence of the religious traditions and beliefs of adjacent nations, especially of such peoples as came to
to a divinity, as Homeric Atoujdns points to Zevs; (2) the vast and early preva- lence of proper names in Asia Minor based on the same stem; (3) the early use of Mnpayoprys, on which see p. 67, n. 1; (4) a large number of inscriptions, beginning with the fourth century B.c. and extending far into Christian times, where Mén is honored, in many of which the Great Mother and Mén are coupled as are no other two divinities; (5) numerous coin types of a similar character; (6) and, finally, references in literature, to be sure for the most part of late date, to Mén as the distinctive god of the Phrygians.
In Lucian, /up. Trag. 42, we should probably read Spéyes 8 Myvi [for Mss. Mvp]; sc. Obowres. Cf. sbid. 8.
In classical literature the divinity most commonly named, in the ancient Phrygian religion, in association with the Great Mother, was Attis, but Sabazius was also known as her son. The line, however, cannot be sharply drawn between the personalities of Mén, Attis, and Sabazius. I am disposed to look upon each of them as originally only different and special aspects of the same divinity, though in later times they appear now and then to have assumed in the popular imagination independent existences; thus in Wagener’s relief Mén conducts a chariot in which Sabazius drives. (To Professor Ramsay’s kindness I owe a drawing of this relief, not yet published so far as I know.) Proclus (ad 7im. IV. 251) distinctly tells us that Mén was addressed as Sabazius in the most solemn ceremonials of the latter god; and on inscriptions (very late Roman) we find dedications to Attis Menotyrannus, where the identification is complete (Orelli- Henzen, /nscriptionum ... Collectio, nos. 1900, 1901, 2264, 2353). On certain coins Attis is represented with all the attributes of Mén (and wice versa); e.g., coins of Pessinus, of Roman date; Head, Astoria Numorum, p. 630; Guignault-Creuzer, Religions de l’Antiquité, II. 3, p. 951. On the imperial coins of the Carian Antioch we have in succession the head of Mén; Attis standing; the god ZNION, a male figure holding a branch (the equivalent of Sabazius). Of course this does not prove identification, but it points to it. At the same time, one must be care- ful not to urge what may be examples of very late conscious syncretism as evidence for an original identity.
In the solar characteristics of Sabazius I see a survival of the more ancient conception of the god, which through contact with Greeks and by the singular etymological perversion suggested above, was radically changed. — The epithet pecdvBpws, as applied to Mén (Sterrett, Papers of the American School at Athens, IT. no. 64, p. 94), should not be pressed in support of this view.
Full of suggestion is Diodorus III. 58. 4: of yap éyxdpio pudodoyoic: rd wadards yerécOa: Bacitda Ppvylas xal Avdlas Myfova: yihyarra 8¢ Acvdduny yervijca per
66 J. H. Wright.
reside among the primitive inhabitants : thus the female divinity now received, among other names, from the Persian colonists in Central and Upper Lydia, as from Persian rulers of a later date, that of the great Persian divinity Anahita.1 The Greeks, however, of the region thought of her sometimes as Artemis. Hence in later times, when devout persons of Greek descent residing in the Catacecaumene would worship the Great Goddess, they would often address her as Artemis Anaitis.’
The name and character of the male god also underwent serious modification. The name Maen, early bécoming Hellenized into the form Myv, was by popular etymology connected with pyy, py, ‘month,’ ‘moon,’ and then fancy transformed a god who may have had solar attributes into a lunar divinity,* and gave him, as appears in late art,* appropriate symbols (the crescent behind the shoulders over against the radiating solar disk that characterized his double, Sabazius). In this new relation he gained wide popularity, and his cult spread into regions that previously had not known the Phrygian god.° His vast significance is attested in many ways,
wadlov O\u .. . KuBé\ny [who was in due time loved by a youth] rd» wrpocayopevé- pavov pay “Arriv, vorepoy 3° éwixdnOévra Ildwa». Confused as all this is — the baldest kind of popular euhemerism, possibly the very doctrine of Euhemerus himsclf, who was one of the sources of Diodorus — is there not underlying Myfova a reference to Mén, if the word itself be not a blunder for Mra?
1 The towns slightly northward of Kula, which lay on the direct line of the ancient royal road to Persia, were a great seat of the Artemis-Anaitis-Métér worship. Cf. Ramsay, //istorical Geography, pp. 30,131; Journ. Hellen. S., IV. p. 385, TIT. Pp. §§.-——Into the questions either of a possible ultimate Babylonian origin for Anaitis, or of Semitic admixtures in her cult at various places, I do not enter. For the literature, cf. Cumont, Wissowa-Pauly, I. p. 2020. ‘
3 From the many passages I select only Pausanias III. 17, where we are told that the Lydians have the oldest statue of Artemis, worshipped by them, however, as Anaitis. At Philadelphia, and in Maeonia, she was known as Mirnp ‘Avdecris, Bulletin de Corr. Hellénique, VIII. p. 376; below, p. 72, no.6; above, p. 57, no. 13.
§ This etymological conception led the Romans of the Empire sometimes to translate Mén by Lunus (Spartianus, Vit. Carac. 6,7; cp. Chwolson, Ssadier, I. pp. 399 ff.). But on some Latin inscriptions, cited above, p. 65, the stem Mén is still used.
# See Waddington-LeBas, Monuments Figurés, pl. 132, no. 2, and below, p. 72.
6 A glance into the index of Head’s Historia Numorum (s. Mén), gives one a good idea of the geographical range of the cult of the god, from Istrus and
A Votive Tablet. 67
some of which have been hinted at in a former note (p. 64, n. 4), but there are one or two aspects of his worship that are of special interest to the classical scholar. If our examination of the evidence is trustworthy, we must believe that, as far back as the earlier years of the fourth century B.c., the priest of Mén was a familiar figure in the streets of Athens, and had vividly impressed himself upon the popular imagination.’
Panticapaeum on the western and northern shores of the Euxine to Laodiceia ad Libanum in Syria.— The prevalence of this cult in Thrace and in Thracian colonies is noteworthy as one of the many signs of the close connection between Thrace and Phrygia. Cf. Strabo, X. 3, 16, p. 471. On many Thracian coins and reliefs the god is represented on horseback, but he is similarly figured elsewhere. 1 The oldest inscription known to me relating to Mén is that published by Foucart, Bulletin de Correspondance Hellénique, 1V. p. 129 (Atovécws xal Bafb\ra 7T& Mnyl 1d lepdy dvéGecar). It is from the Peiraeus, and is dated, epigraphically, the second half of the fourth century B.c.— Apart from the names of persons, the earliest form in which the word occurs in literature is in Mynvaytprys. This word is a title of a comedy by Menander, fl. 320 B.c. (Athen. XI. 4728), and perhaps also of one by Antiphanes (so Bekk. Anecd. 88, 18; but Athen. XII. 553 c reads Myrpaydprns; cf. Kock, C.4./. II. p.74). From the presence of the related word Myrpaydprns (applied, for example, by Iphicrates to Callias: Aristot. Ret. III. 2, ro) likewise occurring as the title of plays, the emendation of Myvaytprns to Myrpaydprys, wherever in the manuscripts the former word is found, has often been proposed. But the word is guaranteed against such treatment by Hesychius’s 6 dxd rol neds curdywr [which I would emend to 6 dxrép rod Mypis curdyur, though 6 éri r@ My» is also possible; cf. Eustathius, Odyss. p. 1824, unrpa- yupreiy: TO... werd Tuprdvwr Kal ripwy Toobrwy wepuéva: cal éxl rT] Myrpl dyelpey tpopds,... 6 dori éwl rp ‘Pég}]. But for Eustathius one would be tempted to supply @doouvs with cusdywr, comparing Dem. /.Z. 281. The explanation of Suidas whereby dxd rod ynws means ‘monthly,’ is clearly a darkening of counsel, and Meineke’s suggestion that the first element in znvaytprns is Mhyn, ‘the moon- goddess,’ can hardly gain acceptance. Cf. also Clement of Alexandria, Protrept. II. p. 20, Potter; Dionysius Halic. Ant. II. 19, and Eusebius, Praep. Ev. I. 8.— The deportment of these dyépra: in public is described not only by Antiphanes in the fragment already cited, but also by many other writers, e.g., Plato, Resp. II. 364 B; Apuleius, A/cfam.8. Cf. Foucart, Associations religicuses ches les Grees, pp. 160 ff. The word dydprns, Latin Ariolus, in turn gave its name to a comedy by Philemon in Greek, and to one in Latin by Naevius. — The existence of these two words side by side, Myrpaytéprns and Myvaydprys, in the popular speech of the Athenians in the fourth and third century B.c., is quite intelligible, if we regard them as calling up the two characteristic divinities of Phrygia in their itinerant and beggar priests, — the Great Mother and Mén. The strange Phrygian religion
68 J. H. Wright.
The exact signification of the epithet Tidyov it is perhaps impos- sible to establish. Waddington, in his well-known note on Mén (Waddington-LeBas, no. 668), approves the suggestion that would connect the word with some Lydian root, comparing other words of similar termination (e.g., [Iptapos, Tevrapos, lépyapos). But is it hazardous to see in this word a local form of the great Babylonian (Semitic) divinity Tidmat’ (Tiam-tu, Tham-te; cf. Hebrew “4dm),
early impressed the imagination of the European Greeks: the erection of the Metroum in Athens in the fifth century, in expiation of the murder of a metra- gyrtes (Suid., s.v.), was a national recognition of its power; but yet deeper was the hold it took on the lower orders of society and on restless spirits, on the one hand giving rise to the formation of orgiastic thiasi, with their fanatical excesses, and, on the other, suggesting taking topics to the comic poets and to orators who often delighted in expressing themselves in ‘broad comedy. Cf. Strabo, X. 19, P- 471, FoAAa yap Tay Lender lepdv wapedétarro [the Athenians], wore kal éxwpip- 840ecar: xal 3h nal ra Opdxia cal rd Ppvywa. The drastic passage in Demosthenes (Cor. 260) describes in caricature a thiasus in honor both of the Mirnp ded» and her companion male divinity; the Scholiast distinctly informs us that the eof odBo refers to [Mén] Sabazius; though the dfs Arras immediately following properly refers to Attis, it here probably more specially connotes the Mother, and thus Strabo is right in saying that this scene contains ZaBdfia cal Myrpwa (X. 3, 18, p. 471). As the mother of Aeschines, a priestess and ruuraplorpa (Dem. Cor. 284), is here figured as a female Myrpaytprys, we may safely infer that Demos- thenes would represent Aeschines himself as a Myvaydprns, though of course in this tumultuous picture no attempt is made to keep the colors distinct.
1 On the Babylonian Creation Tablets the name of this being appears as Tiamat. In Hebrew the form becomes ¢4ém, usually rendered ‘waters, deep, depths.’ In Berosus, as corrected by Robertson Smith, the form Odyre occurs, with @¢dd\acca as the Greek translation; Damascius, however, gives Tadée — the transliteration of the same form in its variant Tiavat. Tiamat, fem., is the “construct” of Tiamtu (Tiamatu). Professor M. Jastrow, Jr., to whose kindness I owe some of the references to the literature of Tiamat, given below, writes me: “ As for the mythological con- ceptions associated with the Hebrew 44m, it is noticeable that it is never found in combination with the article, which is itself an indication that the consciousness of its being, or having been, a xomen proprium never died out. Again, such pas- sages as Ezekiel xxxi. 15, Psalms xxxiii. 7, civ. 6, Proverbs iii. 20, and others, prove that the Hebrews connected with the word views of a different order from the ordinary sense of ‘waters.’ In the passages quoted ¢4ém is the subterranean ocean that nourishes fountains and springs. It seems to me therefore that your identification receives added force from the mythological conceptions once con- nected with the Hebrew form, which warrant one in assuming a larger circle for the Tiamat conception than Babylonia. We may therefore expect some day to
A Votive Tablet. 69
demon of the subterranean waters (xarayfonos)? Now it is well known that epithets of Semitic origin are actually applied here and there in Western Asia Minor to divinities locally worshipped under Greek names; and that as a rule each of these is the equivalent of some more familiar Greek word.' Accordingly I would suggest that, in this inscription, Mén Tiamu is only another form of Miv xara- x9enos. Chthonic divinities are for the most part associated with fountains, i.e., with subterranean waters, and in especial with healing springs: the conception underlying the word Tiamat and its cog- nates is that of the vast world and underworld of waters, or of unorganized nature conceived as a watery abyss. As in Artemis Anaitis, Greek and Persian names —the latter colored by Semitic .myth— had been used to give a proper shading or depth to the native conception of the Great Mother, what is more probable than that a name distinctly Semitic might have been used in designation of the male divinity, especially when viewed in an aspect perma- nently characteristic of the Semitic god? If we are right in our explanation of the occasion of the dedication of the relief, — the commemoration of the might of Artemis and Mén in bringing back a sick child from the confines of the lower world, or in suffering it
find Téhaim (= Greek Tiay-) in a North Syrian inscription as the connecting link between the Babylonian and the Greek form.”
The fact that in the literary documents Tiamat is made a female divinity is not a fatal objection to our identification. As the personification and presiding genius of as yet unorganized nature, ruling over beings with bisexual characters, this divinity might originally have been conceived as without sex. In fact, on the monuments Tiamat is often represented with the attributes of a male being.
On the forms of the word Tiamat, see E. Schrader, Cunciform Inscriptions and the Old Testament (Eng. Transl.), I. pp. 6ff.: cf. also Jensen, Kosmologie der Babylonier, pp. 268, 269, and Sayce, Hibbert Lectures for 1887, p. 237. In the American Journal of Archaeology, VI. 1891, pp. 291 ff., Dr. W. H. Ward dis- cusses Tiamat in Babylonian art.—Cf. also Gunkel, Schop/fung und Chaos, pp. 16 ff., gor ff.
1 Examples are [Mh»y] Kayapeirys, on imperial coins from Nysa in Lydia, where the epithet is connected with Semitic (Arabic) Qamar, ‘the moon’ (Head, Listoria Numorum, p. 552), and [Zeds] ACEIC, on coins of Laodiceia ad Lycum in Phrygia, where Waddington and Longpérier, as Professor Ramsay reminds me, see in the epithet (= wjuwros) a Semitic stem, viz. that in the Arabic ‘aziz, ‘powerful’: Head, #6. p. 566.
70 J. H. Wright.
to return, — it must have been Mén of the lower world that was here addressed. And, as already intimated, in the polyglot terminology * of the period of the Mén Tiamu inscriptions, and in the Catacecau-
1 Most of the large number of epithets applied to Mén, — some of which are still unintelligible, — are geographical in character (Waddington-LeBas, no. 668). Among such as are not geographical may be cited .rvpa»vos, wdrpus Gebs, peod» Bpos, pwodédpos, Sows (?), caraxOdnos.
2 Professor G. F. Moore has called my attention to a curious bilingual inscrip- tion from Palmyra, which contains a name suggestive of Tiamu. He writes: “In Vogiie, /wscriptions sémitiques, no. 3 (cf. Mordtmann, Zestschrift der deutschen Morgen!. G. XXXI. 100), Mordtmann restored “3° "[4], Gad Thaimi. The Greek runs xar’ ros dva@éuara [Mada]xSHry xal Téxy Oaetos xal ['Arep]ydree. Gad Thaimi also occurs on a seal from Palmyra (Mordtmann, /.c.). An Old Testament scholar is reminded of Isaiah lxv. 11, where we read of lectisternia to Gad and Afen:. (The latter name is guaranteed against Lagarde's conjecture “35 == Nanaea, by the play on the name in “D1"s%) at the beginning of the next verse.) Evidently Meni, though not elsewhere found, was a Semitic (Aramaean) god of fortune, and by the side of Gad Thaimi, a Mn» Tidpouv might not seem inconceivable. We should, to be sure, not be much better off, for Thaimi is almost as obscure as Tiduov. Noldeke interpreted, ‘the Téx» of Thaimi’; Mordt- mann at first, ‘the Téx7 of the Thaimites’; afterwards he felt constrained by a series of proper names to take Thaimi as the name of a divinity (#6. XX XIX. 45 .). But the Greek GOacuetos is quite regular, and seems to preclude a combination with Tidyov. Mordtmann (#6. 44) combines, curiously at least, the Meni of Isaiah Ixv. 11 with... Menisgue magister on an altar from Vaison in Provence (Orelli- Henzen, no. 5862 [unless Afenis be a mistake for mentis, possibly Menis magister == Menotyrannus = My Tépavvos? read gaudebunt? J. H. W.]:
Belus Fortunae rector Menisque magister ara gaudebit quam dedit et voluit.
The corresponding Greek inscription is:
Evéuyripe tuxns Bhdtw Leboros Oéro Bwpdy ray év 'Arauela pynoduevos Noylwr.”
Though no one would venture to make Semitic Meni and Phrygian Mén identical in origin and nature, the resemblance between these two names would at least make easy, in regions where the two gods were known, the ascription to the latter of Semitic characters and epithets.
Perhaps an evidence of the fusion of Aramaean Meni and Phrygian Mén, at least in Ameria in Northern Asia Minor, may be seen in the language of Strabo XIT. 3. 31, p- §57, where the implication is that, in the oath of the kings of Pontus, Mh Dapvdxov and rbyn Bacchéws are identical in meaning. ‘aprdxov, though pos- sibly originally referring to a divinity, must here be used as a generic name for
em wm — -
A Votive Tablet. 71
mene, Mv xarayfovos might well have been sometimes known and worshipped as Myy Tidpov.’
‘king’ (cf. Kaiser from Caesar): hence Mi» and Téxy were felt to be the same. But this could be only because Meni, the Aramaean god, who was the god of fortune, or Téxn, had become identified in the popular mind with Mén, whose encroachments subsequently almost completely obliterate Meni as an indepen- dent divinity. — Other instances of the association of Téxn and Mén appear on the coins of Antioch Pisidiae (Mionnet, Supf/. VII. p. 102, no. 81; Stephani, Comptes Rendus, 1861, p.83), and of Laodiceia ad Libanum (Eckhel, III. pp. 366 ff.; Head, Historia Numorum, p. 663).
! The identification of Tiamu with Tiamat (Tham-te) appears to be asserted by a writer of the first century B.c., who was very familiar with Phrygian religious beliefs and customs, — Alexander Polyhistor. In a passage preserved to us in Syncellus, p. 52 (cf. Schone, Zuseb. Chron. I. 16), Polyhistor, quoting from Berosus, and interpolating explanatory remarks, informs his readers that according to Babylonian legends the demon of watery abyss, who was slain by Bel[-Marduk; Oubpwxa == 6 Mopdéxa}, was called in Chaldean Oaure, which [he tells us} in Greek translation is @d\acea, and is equivalent to the moon [perhaps, ‘according to the Phrygians is the moon’]. This last sentence becomes at once intelligible, if we assume that he was thinking of Mi» = Tidwovu as he wrote : — ZerAhvn: Mhv (Mfrm) = Tiduou: Oadure. Cf. Proceedings of the American Philological Association, 1894-5, Philadelphia meeting, no. 22; also Zestschrift fur Assyriologie, X.1 (1895); my note on “ Homoroka a Corruption of Marduk.”
NoTE. — Mén Tiamu Inscriptions.
Below are collected all the inscriptions known to me that contain the name of Mén Tiamu. There seem to be no coin types with the legend MHN TIAMOT. — The Movecioy xal BeBr\0Ohnn ris Evayyedxiis Zyo\js at Smyrna, in which nos. 6 and 7 were first published, is designated by Moveeiov. Facsimiles of nos. 1, 2, 3, and 8 are given by Waddington-Le Bas; of nos. 5 and 6, by Leemans; of no. 4, in our PLATE II, opp. p. §§.— For the literature of the Anaitis inscriptions, see above, p. 57, note I.
1. “Erous?] oun (?) dv€Onxay... cal Mnwds Tiduou xal Mnypds Ilerpaclrov rd [ayaA]ua rod Aovboou- ‘Iovkcards, Aduas, Edoxtpwr, ‘Ardloy ’Ardlw[vos, Térduws, SéxourSos, Lpwrébxryros, ‘Epyoxpdr(ns], Marpddwpos, Aduas, “Ardlas, ‘Epyoyérns, Adua[s, My] rddidos, "AwodAdmnos, "Hpais, 'Eppyoxpdrys, ‘AwodAwmos, ‘H[-yeu]wr, “Awod\Admos, “Aupior, Tptdwr, Medrrlyn, ‘Epyo[yé]yns, Necxduaxos, Mela, Mdpxos, Téxn, "Awé\das, ['Adé]Eavdpos, MeArivn, Lworparos, Nexlas, ‘IovNa.— From Goerdis (Julia Gordus). Dated a.p. 163 (164)? Waddington-LeBas, no. 678. Supply, after d»€Onxay, card Thr Td Gedy éxcrayhy, or words to the same effect (?).
72 J. H. Wright.
2. ‘lepda cupBlwors xal vewrépa car’ éwiraydy rob xoiplov rupdyvoy Aids Macdada- rnvot, cal Myvi Tidyov edy hy: ‘lovksands Mevexpdrov, Mevexpdrys Avodwpov, Awvrolows Tlawlov, ‘Eppoyérns ‘Epulewov, Aovxiws ‘Ovnoluov, Aroyérns T'dbxwros, Acoyérns Magluov, Tpddiuos ‘Epularsov, 'ArodAdmos ("ArodAwrlov), Oed8wpos (Geodiuspov), Map- mawds (Mapxiavov), Mévavdpos ‘Eppoyérov, ‘Epuoyévns Tarcavod, Myrpbdwpos Ededrlorov, "Aoxdniddns Maxpiavod, ‘Aokxdyriddys Atovolov, ‘Epyoyérns Tpodluov, [Z]éBndos ‘Eppoyévov. "Emiuednoapuévwr Tovar’ xal ‘Epuoyévov, frous ov<, u(nrds) Ado pov. — From Menneh, near Kula. Dated a.p. 171 (172). Stele representing Zeus aeto- phorus and Mén standing erect, with tunic, chlamys, and Phrygian cap, crescent behind his shoulders, a pine-cone (?) in his left hand, spear in his right hand, his left foot on the head of a bull that lies on the ground. Waddington-LeBas, no. 667: Monuments Figurds, pl. 136, no. 2. CJ.G. 3438 (Boeckh).
8. Kara rhy rév Oedy éxrirayhy lepds Sodu0s evyhy Ad Macdadarnrg xal Mavi Tidpov cal Myvi Tupdvvp éxédevcer rnpeicOar dwd huepdy 6. ef ris 3¢ rodrwr dreOhor dvayruicera: ras Surdus Tod Acbs. dred noapuévov Avcovuclou Awdwpovu xal ‘Eppoyérns Baneplov. €rovs orf, u(nws) Adorpov.— From Kula. Dated a.D. 172 (173). Stele with relief representing two busts, one with radiated head, the other with a crescent behind the shoulders. Waddington-LeBas, no. 668: Afonuments Figurés, pi. 136, no. 1; C.2.G. 3439 (Boeckh).
4. Our inscription; see PLATE II, and pp. 56 ff. Dated a.p. 196.
5. Ged "Avaelrs xal Myvt Tidpou Medrivn xat TAbcwy dwédwxay 7d lepowolnua edxapiorotvres. frovs T, u(nws) Zavdcxod.— From Kula. Dated a.pD. 215 (216). Marble stele: woman draped, with right hand raised in adoration. Leemans, Verh. d. k. Amsterdam Akad. XVII. 1886 (1888), no. III; Contoleon, Athenische Mittheilungen, XII. 1887, p. 255; E. L. Hicks, Classical Review, ITI. 1889, p. 69. This inscription has ETOTS:-TM-ZANAIKOT. Perhaps we should read grovus Th (unws) Zavdlxov = A.D. 255 (256), on the supposition that the sign for uns has been omitted. — Leemans’s Mcmridyou is a misreading of MENITIAMOT. Cf. above, p. 58; also p. 59.
6. OeG "Avacirs xal Myvi Tedpou Téxn xal Dwxpdrys cal’ Apwards cal Tpddipos ol "Auplov xal Dirsrn cal Twxpdria al ‘Aupddos wohoavres 7d lepowdnua el\acdpery Myrépay ‘Avdariy bwep réxvwr cal Opeunpdrwr Evypagov trrnoay. frovs TKa, (ds) Zavdixo0. — From near Kula. Dated a.p. 236 (237). Relief representing two breasts, two legs, and two eyes. Tsakyroglus in Movceioy, V. 1884-5, p. §4 (Reinach, Chroniques d’Orient, 1886, p. 155); Leemans, #6. no. IV.
7. “Apréuds ‘Avaeirs xal [M]nvt Tiduou MeArivn [dw ]ép ris ddoxAnplas [7rdv] roday edxhv [dxto]rnoev. — From Kula. Not dated; perhaps about 2204.D. With relief representing two feet. Moveefov, 1880, p. 127 (P. Foucart in Bulletin de Cor- respondance Hellénique, IV. p. 128).
Perhaps we should add, at least for comparison :
8. Aoves Mdpxos xar’ éwirayhy rot Geol drép ris owrnplas Asel Tiywaly dvéorncer, frous oka, u(nws) "Apreuolo[v].— From Menneh (near Kula). Dated 177 A.D. Waddington-LeBas, no. 669. If we regard the Arel Tiwaly as an attempt to render Mjv Tidyou into Greek words more intelligible to a Roman, the inscription
A Votive Tablet. 73
might be grouped with those relating to Mj» Tidyov. Of course, Tialy is not a translation of Tidyov, though it may here be a popular etymology for it. The phrase xar’ éxsray4y is very common in the Mén inscriptions. Cf. Zeds ZaBdsios in Wagener’s inscription, cited above, p. 62, note 4: Foucart, Associations veligieuses ches les Grecs, p. 71.
In Waddington-LeBas, no. 675 (from Kula, without date), I suggest M[»] éoly for Waddington’s dojy. (Cf. #6. no. 1670, [OJey doly xal decay.) The names of the dedicators of this slab, Manes and Banes, ard noteworthy.
POSTSCRIPT.
THE preceding pages were put into type in August, 1894, and went to the printer to be electrotyped in May, 1895. Hence it has been impossible, in the revision of the sheets, for me to make use of Professor Ramsay’s Citss and Bishoprics of Phrygia (Vol. I, Lycos Valley), the work referred to as in press on page 64 above, and issued only a few months ago. When Professor Ramsay was in the United States, in the autumn of 1894, he was good enough to glance over the first proof-sheets, and made many useful suggestions; hence his remarks, on p. 340, on my identification of Tiamu, and his criti- cism of the same. I trust that in its present more matured form my argument may commend itself to him.
In the Lycos Valley many topics, here only briefly adverted to, are fully and luminously discussed, with much fresh evidence: e.g., the ancient religion and divinities of Asia Minor, pp. 7 ff., 87, 105, 132 ff., 145, 169 ff., 262 ff., 271, 273, 292 ff.; Sabazius, Sozon, etc., pp. 140, 262 ff., 293 ff.; the Sullan era, pp. 201 ff. (which Professor Ramsay would now begin August 1, 85 B.c., and not July 1), etc., etc. Profes- sor Ramsay emphasizes the difference between the earliest religious stratum in Asia Minor, based upon a matriarchal social system, and that introduced by the conquering Phrygians, probably immigrant from Europe, where the male element is dominant. It should be borne in mind that in this article the word ‘ Phrygian’ has not been used in an ethnographical but only in a geographical sense, to desig- nate not alone the oldest inhabitants but also the same as they
74 J. H. Wright.
appear in history amalgamated with the Phrygians proper, and dominating in no small measure the religion and mythology of their conquerors.
While a few minor statements would have been modified and many important references to authorities, literary and epigraphic, would have been added had the Lycos Valley reached me in season, on the whole the main contentions of this article remain unaffected by it. Professor Ramsay, on p. 132, says: “These facts prove that the name Attes belongs to an older status of religious history than Men, ... and history confirms the inference that Kybele and Attes were the ancient Phrygian names for the Mother and the Son.” Certainly Attes is a more ancient word than Myyv in the Grecised form, though not necessarily more ancient than Maen (see above, p. 64; also Lycos Valley, p. 169). In spite of the testimony of Arnobius V, 6, to the effect that Attis is connected with Attagus, the Phrygian word meaning ‘goat,’ cited by Professor Ramsay p. 350, there seems much in favor of the view of Eduard Meyer that the name and myth of Attis is a Semitic importation, detachable from the circle of myths relating to the Mother, and hence later than the original native religion; see Geschichte des Alterthums, I, p. 308.
December, 1895.
THE DATE OF LYCOPHRON. By WILLIAM N. BATES.
HE date of the poet Lycophron has never been satisfactorily
settled. Writers on the history of Greek literature have been much at variance as to the period in which he lived. Some have thought that he flourished in the reign of Ptolemy Philadelphus (285-247 B.c.), others in the reign of Ptolemy Euergetes (247-221 B.C.), Others again have been satisfied with saying that he flourished about the middle of the third century. The reason for this diver- gency of opinion is apparent as soon as one looks at the evidence upon which our knowledge of Lycophron rests. The writers who make him flourish in the time of Philadelphus, base their statement upon the scraps of information which the scholiasts and lexicog- raphers have handed down about him ; while those who make him flourish in the time of Euergetes throw aside the testimony of the scholiasts and rely upon a passage in the Alexandra to show that he flourished at this later date. The passage in question (lines 1226-1280) is a prediction of the coming greatness of Rome, and these writers argue in a seemingly plausible manner that Lycophron could not well have written such a passage before the first Punic war.' The other writers, who say that he flourished about the middle of the third century, are simply trying to reconcile these conflicting statements.
This was the state of the case when in 1883 Wilamowitz under- took to settle the date of the composition of the Alexandra. He showed that the latest event mentioned in the poem is the murder of Heracles, the seventeen-year-old son-of Alexander the Great, in
1 Lines 1446-1450 were also appealed to as evidence, in spite of their enigmat- ical character. For a possible explanation of the references in these lines, see Wilamowitz, De Lycoph. Alex., p. 8 f.
76 William N. Bates.
the year 309. The part which Cassander had in this murder and in the previous murders of Olympias, Alexander’s mother, and of Roxana and her child is nowhere alluded to. Wilamowitz argued from this that the Alexandra was written while Cassander or his sons were in power, that is between 309 and 287. He showed furthermore that Lycophron had part of Timaeus’ history before him when he wrote the Alexandra, and Timaeus did not begin his work until 310. Hence he argued that it is more likely that the Alexandra was written between 300 and 2go than between 309 and 300. Wilamowitz also showed that the Alexandra was imitated by Dosiades in his Bwydés and hence was written before that poem, which he supposed to have been composed between 285 and 270. Susemihl in his Geschichte der Griechischen Litteratur in der Alexandrinerseit? shows that the Altar of Dosiades was probably written about 292~—290 and hence the Alexandra, which preceded it, was probably written about the year 295. Susemihl argues furthermore that the Alexandra is the work of a young man, and assuming that Lycophron was thirty or thirty-five years old when he wrote it, he concludes that he must have been born between 330 and 325.
Such is the present state of our knowledge as to the date of Lycophron. The most important point which has been gained is the fact that the Alexandra was written about 295. With this point settled, the passage in the Alexandra relating to Rome can have no bearing on the date of the poet and may consequently be disregarded. .
Additional evidence, however, can be presented which will estab- lish with much more accuracy the date of the poet.
The general period in which Lycophron lived is stated clearly by Tzetzes in his life of Lycophron,’? where he mentions him as the con. temporary of Ptolemy Philadelphus. This statement is repeated in an ancient scholium on line 1226 of the Alexandra,‘ and is inferred also from the statement of an anonymous writer on comedy published
1 Line 801.
3 Vol. I. p. 274, n. 28.
8 Tzetz. in Lycophr., ed. Miiller, Vol. I. p. 263 f. # Lycoph., ed. Kinkel, p. 179.
The Date of Lycophron. 77
by Cramer,! and from a fragment of Tzetzes published by Ritschl.? But we have more precise information than this. Tzetzes in the fragment just mentioned informs us that Alexander the Aetolian, Lycophron of Chalcis, and Zenodotus of Ephesus were hired by Ptolemy Philadelphus at the royal expense, the first to arrange the tragedies, Lycophron the comedies, and Zenodotus the Homeric poems and the works of other poets which had been collected at Alexandria.*> This statement is repeated in substantially the same form in the anonymous writer on comedy and in the scholium Plau- tinum.* These passages make it clear that the books which Lyco- phron, Alexander and Zenodotus were employed to put in order were those which Philadelphus and his father had collected and which formed the beginning of the Alexandrian library. This collecting of books had been going on for some years, and authorities agree that the books were brought together as a library at the very beginning of the reign of Philadelphus, that is between the years 285 and 283, when Philadelphus was king and Ptolemy Soter and Demetrius of Phalerum were still alive. The work of arranging the books must have been done before the library could be of use to any one. That is to say, this work must have been begun by 283 at latest. Moreover, as Tzetzes informs us that Philadelphus hired the men at royal expense, he must have done so after 285, when he became king. Consequently Zenodotus, Lycophron and Alexander must have begun their work of arranging the books during the years 285-283, or perhaps during the year 285-284.
This conclusion accepted, we have a definite point to start from in determining the date of Lycophron, for the dates of his two asso- ciates in the work can be estimated with some degree of exactness, and we should not expect Lycophron’s age to differ greatly from the ages of the other two. Zenodotus held the most important position
1 Anecdot. Paris., Vol. I. p. 6. Perhaps this is to be ascribed to Tzetzes.
2 Opusc., Vol. I. p. 206. See also p. 199.
3 "Anétavdpos 6 Alrwids xal Auxddpwv 6 Xadaideds, ddAd xal Zyrdsoros 6 "Edéocos re SradArdy Irodkeualy curwrnOévres Baciixds, 6 wey ras THs Tpaywslas, Auxédpwr dé ras rijs xcwpydlas BlBdous SidpOwoay, Znvdsoros 82 ras ‘Ounpelovs xal TGy Nowrdy TounTrwv «rr.
* Ritschl, Opuse., Vol. I. p. 5.
78 William N. Bates.
of the three; he had charge of all except the dramatic poetry, and was afterwards librarian. We should naturally expect him therefore to be older than Lycophron and Alexander. Now Zenodotus accord- ing to Couat! was born between 324 and 320, and Alexander? about 320; Susemihl*® gives the dates as about 325 and 315 respectively. Consequently if Lycophron was younger than Zenodotus, he must have been born after 325, but on the other hand, as he wrote the Alexandra about 295, he can hardly have been born as late as 315. His birth-year was probably not far from 320.
This date is in a measure confirmed by an additional fact which has been handed down about Lycophron, namely that he was a member of the Alexandrian Pleiad. This Pleiad, as we are informed in a scholium to Hephaestion,* consisted of seven tragic poets who . all flourished at the same time in the reign of Ptolemy Philadelphus. Their names are given by several authorities, but with some varia- tion. Lycophron is named as a member by all the authorities, and there is sufficient evidence for determining the other six members with considerable certainty.6 About the dates of two of the mem-
1 Histoire de la Poésie Alexandrine sous les trois premiers Ptolémées, p. 57. 2 /bid., p. 105.
8 Geschichte der Gr. Lit. in der Alexandrinerseit, Vol. I. pp. 188, 330-
# Vol. I. p. 199, ed. Gaisford.
6 The lists are as follows :
Schol. Heph. Schol. Heph. — Schol. Heph. Suidas. Tzetzes. I. p. 57, 10. I. p. 199. I. p. 199, codex S. Homerus Homerus Homerus Homerus Theocritus
Sositheus Sositheus Sositheus Sositheus Aratus Lycophron Lycophron Lycophron Lycophron Nicander Alexander Alexander Alexander Alexander Aeantides Philiscus Aeantides Dionysiades Philiscus Philiscus Dionysiades Sosiphanes Euphronius Sosiphanes Homerus Aeantides _Philiscus Philiscus Dionysiades Lycophron
It is noticed that all five authorities agree on Homer, Lycophron, and Philiscus ; four agree about the names of Alexander and Sositheus, and Alexander is still further confirmed by Eudocia, Viol., p. 62. Dionysiades is named in three of the lists, and likewise has the testimony of Strabo, p. 675 (Meineke, p. 941, § 15). This makes six of the members of the Pleiad about whom there can be no doubt. About the seventh member, however, authorities disagree. Aeantides has three authorities in his favor; Sosiphanes, two; while Euphronius, Theocritus, Aratus,
The Date of Lycophron. 79
bers, namely of Aeantides and Dionysiades, nothing definite is known; but the other four (#¢. omitting Lycophron) can all be shown to have flourished in the 124th olympiad, that is 285-281 B.c.! I have already shown that Lycophron was engaged in work in the Alexandrian library in the year 285-284. How long he was occu-
and Nicander have each one. The last three authors, however, were not tragic poets, and their names may consequently be disregarded. Likewise Euphronius ; for the only ground for believing that he wrote tragedy is the finding of his name in the scholium mentioned above. In favor of Sosiphanes we have the state- ments of Suidas and one of the scholia to Hephaestion. But Suidas tells us (s.v. Zwoupdyns) that this poet lived in the time of Philip of Macedon or of Alexander the Great, and furthermore that he died either in the 111th or 114th olympiad. In other words, Sosiphanes died about forty years before the other members of the Pleiad are said to have flourished. Consequently Sosiphanes cannot have been a member of the Pleiad, and the seventh member must there- fore have been Aeantides, who has the evidence of three passages in his favor. We must conclude, therefore, that the tragic Pleiad consisted of the following seven writers: Homer, Lycophron, Philiscus, Alexander, Sositheus, Dionysiades, and Aeantides, which is the list of members given by the scholium to Hephaestion, Vol I. p. 57, 10.
1 For Homer, see Suidas, s.v. “Ounpos: “Opnpos ypayparixds xal rpayydtOr rown- ths, 8d curnprOuhOn rots érrd of ra Sevrepeia ray rpgyixdy Sxover xal éxrAROncay ris Tihecddos. qxuatey ddupmidds pxd. Of Sositheus, Suidas says, s.v. Zwol6eos: ZwolBeos ... rdv rijs Wdeaddos els, dvrayworhs ‘Outhpov rol rpayixod ... dxudoas xara Thy pxd é\upmidda. Alexander Aetolus, as has been said, seems to have been born about 320-315, and in the year 285-284 to have been engaged in arranging the tragedies for the Alexandrian library. In 276 he was with Aratus and Antagoras of Rhodes at the court of Antigonus Gonatas, where he stood in high favor (see Vita I. of Aratus in Westermann, Siog., p. 54, and ¢f Droysen, Ffell., Vol. II. p. 197). Thus we can safely say that Alexander flourished as one of the Pleiad at the end of the 124th olympiad. Philiscus, the fourth member whose date we can fix, is described by Suidas (s.v. @:Aloxos) as rpayixds xal lepeds rov Asovbcou éxt rol Didaddrgou IroNepalou yeyords .. . fore Se ris Sevrépas rdfews ray rpayixdy ofrevds elor f° cal éxdhOnoay Idacds, erd.; and Callixenus, quoted by Athenaeus (V. p. 198 c), says, in describing as an eye-witness the great procession which took place when Philadelphus was crowned king in 285: ... ue" ous éwo- petero Dirloxos 5 roinrhs, lepeds wy Awvicov nal wdvres ol repli roy Ardvucoy rex virat. That is, Philiscus enjoyed a reputation as a poet in the year 285, or, in other words, may be said to have flourished in the 124th olympiad. Thus Homer, Sositheus, Alexander, and Philiscus are all shown to have flourished at the same time, as in fact we infer from the scholium to Hephaestion (I. p. 57, 10), and that time was the 124th olympiad, or 285-281 B.c.
80 William N. Bates.
pied in this work cannot of course be known, but it must have taken considerable time. Moreover we are informed that he wrote a work on comedy in at least nine books,’ which it seems likely was the result of his labors in the library. If, then, Lycophron flourished as a tragic poet with the other members of the Pleiad he must have done so towards the end of the 124th olympiad or about the year 281. Therefore if Lycophron was born at the date I have already shown, he must have been about forty years old when he began to flourish as a tragic poet; and this is certainly an age at which a literary man might very naturally be said to flourish.
As to the extent of Lycophron’s literary career we have no certain evidence. We are told by Tzetzes that he wrote sixty-four or forty- six tragedies, and Suidas gives us the names of twenty of them. These titles when added to his other works seem to imply a literary career of some length. How long he lived is not known. His death is mentioned in but oné place, namely in Ovid’s Ibis, where the poet says,*
Utque cothumatum periisse Lycophrona narrant Haereat in fibris fixa sagitta tuis.
This leads us to infer that Lycophron was killed by an arrow while engaged in some scenic representation. There are three scholia on the passage, but none of them appears to be ancient, and all simply confirm the words of the text without adding anything essential.
At first sight this seems to be all that can be gathered about the death of Lycophron; but after examining the Ibis carefully I think -we have good grounds for believing that Ovid took this allusion to Lycophron’s death from the Ibis of Callimachus. For in the first place Ovid himself says that he is imitating that poem.? How
1 Ath. XI. p. 485 d. 2 Lines 531, 532. $ Lines 55-60: Nunc quo Battiades inimicum devovet Ibin, Hoc ego devoveo teque tuosque modo. Utque ille, historiis involvam carmina caecis: Non soleam quamvis hoc genus ipse sequi. Illius ambages imitatus in Ibide dicar Oblitus moris iudiciique mei.
The Date of Lycophron. 81
closely he did this cannot of course be known, for no vestige of this latter work is extant; but the scholiast on lines 315-316 gives what purports to be a metrical translation of two lines from the Ibis of Callimachus which are very similar to the two lines of Ovid.’ If this scholium could be relied upon, we should have good grounds for believing that the imitation was a close one.* A second point is that in mentioning the death of Lycophron Ovid uses the words ‘utque narrant’ implying that he is quoting. Again, Lycophron was hardly known to the Romans —in fact he is mentioned but once again in all Latin literature — and it is hard to see why Ovid should have introduced his name here if it had not been suggested to him by finding it in the work he was imitating. For surely Lycophron’s death was much less terrible than many of the other misfortunes which he invokes upon the head of his enemy. Again, the enemy against whom Callimachus wrote his Ibis was Apollonius Rhodius, a poet like Lycophron ; and hence Callimachus might very fitly hold up the death of Lycophron as an example of what his fate might be. On the other hand Ellis has shown® that the enemy whom Ovid attacks was not a poet or literary man, but rather an informer. Thus Ovid can have had no particular reason for mentioning Lycophron’s fate unless he was repeating something which he found in Calli- machus. In view of these facts, although of course in a case like this certainty is impossible, I think we are justified in assuming
1 Ovid's lines are (315, 316): Utque necatorum Darei fraude secundi Sic tua subsidens devoret ora cinis. While the so-called lines of Callimachus run: Sic tu depereas sicut periere secundus Quos Dareus multo proruerat cinere. (secundi . . . proruerat, G.; secundus... obruerat, C. et Ask.) Ellis, in his commentary to this passage in his edition of the Ibis (p. 58), says: ‘‘Notabilis haec mentio Ibidis Callimacheae, nec video cur non genuina habenda sit.”
2 For a discussion of the Ibis of Callimachus, see Riese in Jahr. fur Phil. Vol. CIX. (1874), pp. 377-381, where Schneider’s theory that the Ibis was an epigram is completely refuted. Riese’s opinion is approved by Susemihl, Gesch. Gr. Lit., Vol. I. p. 351, n. 19, and by other scholars.
® Proleg. to the Ibis, p. xxii. fol.
$2 Wrlkam \. Bates.
that Ovid took the allusion to Lrcophron’s death from the Ibis of Calimachus.
If this ts granted, an approximate date can be established for the death of Lrcophron : for Lycophron must have died before the Ibis of Callimachus was written. Now it is agreed that this poem was Written about two vears before the hymn to Apolo! which Richter’ and Couar* think was composed m 243. Susemihi* says it was written either in this vear or m 263. and argues m defence of the karrer date. If. then. the hymn to Apoilo was written in 248, the Ibts must have been written about 250. before which time Lycophron must have died: or if we accept the date preferred by Susemihl, which certainly seems reasonable. Lrcophron must have been dead by 265. Bat he cannot have died very long before this date: for if he wrote the greater part of his tragedies after he left the Alexandrian kbrary. as 1s probable. the ume necessary for their composition obliges us to suppose that he hved untl verv nearly the vear m question.
To resume briefly. the life of Lycophron was about as foilows. He was born between 325 and 320. wrote his Alexandra about 295. was appointed to arrange the comedies in the Alexandnan library in 285-284: about 250 he was fiourshing as a tragic poet, and con- tinued as such down to the time of his death. which must have occurred before the year 250, and probably shortly before the year 265.
I See Sesemihl, Gesch. Gr. Lit. Vol IL. p 385. 2 51; also Apollon Vita I, Westerm.. Steg. p. 3.
2 Kalimachos, /freeace, p. 6 and fol.
3 Le Pees Alexand. pp. 229-233
§ Gesch. Gr. Lit. Voi I. pp 361. 362.
QUO MODO J/JAC/JEND/ VERBI COMPOSITA IN PRAE- SENTIBUS TEMPORIBUS ENUNTIAVERINT ANTIQUI ET SCRIPSERINT
QUAERIT
MAURICIUS W. MATHER.
I, PRAEFATIO.
UM has tactend: verbi praesentis formas quae praepositionibus subiunguntur pertractaverint viri doctissimi haud ita pauci, quorum in primis Wagnerum, Lachmannum, Corssenum, Guil. Schmitzium, L. Millerum honoris causa volo nominatos, adhuc tamen non fuit, qui omnibus inscriptionibus perscrutandis atque quem ad modum poetae Latini ab illius Livi Andronici temporibus usque ad extremum alterum saeculum p. Chr. n. haec verba composita ad versus accommodaverint quaerendo huius rei scrupulos dubitationesque multas tollere conatus sit. Mihi igitur in mente est, quae ex his fontibus duobus magna cum cura laboreque exempla derivavi, ea ita ordinata in extremo opusculo! enumerare, ut omnibus appareat quae testimonia sint antiqua ad hanc rem pertinentia, atque ut sit facultas iudicandi verumne ipse ego in meo commentariolo investigaverim. Soluta praeterea oratione scripti codices non nulli adhibiti, qui temporibus octavo saeculo p. Chr. n. superioribus exarati sunt, aliquid ad existimandum, quo modo haec verba scripsissent antiqui, obtule- runt. Post decimum autem saeculum qui scripti sunt libri minus, ut fit, proficiunt hac in quaestione; in his quoque saepenumero, prae- cipue renascentium litterarum temporibus, perversam per duas 3s litteras rationem videmus scribendi, quae non solum in quibusdam poetarum locis numeros laedit,*? verum etiam perpaucis in antiquio-
1 Cf. infra, p. 127 8qq. 2 Cf. ind. Plaut. b, pro cdnmicetis, cOniciam habet F conticstis, conticiam ; Moret. a, pro ddscét habent H e adiicit ; Mart., pro daicit habent b ¢ adiicit.
84 Mauricius. W. Mather.
ribus libris, in eisque admodum raro, occurrit, numquam autem in titulis. Atqui huius ipsius rationis auctores sunt grammatici Latini. Unde patet eos, cum non nihil prosint nobis quaerentibus, tamen praecepta scribendi non tam ex consuetudine hausisse quam ex similitudine; atque ideo iis quae docent necesse est ne nimis credamus cavere.
In editionibus auctorum Latinorum usque a tempore, quo formis describi coepti sunt libri Latini, ad medium huius saeculi sactendi verbi praesentis formas quae cum praepositionibus coniunguntur duplicata # paene constanter scriptas legimus, velut adsicit, deiicit et similia. Solae excipiuntur figurae in quibus e numerorum ratione primam apparet syllabam aut corripi aut per diphthongum, quem dicunt, cum insequente # coalescere, ut in Iuv. 15, 17 :—
Hunc déicét, saeva dignum veraque Charybdi, et in Verg. E. 3. 96:— Tityre, pascentes a flumine reice capellas.
Wagner primus, quod sciam, in editione Vergili carminum ad pristinam orthographiam revocatorum a. 1841 in lucem edita illam geminationem reiecit. Qua de re conferas illius operis p. 445. Cum autem unam / litteram ab antiquis scriptam putaret, duas tamen esse auditas adnotavit ad A. 12.308, ubi dixit diséct verbum esse scriptum sed disticit auditum. Etiam scriptas esse duas 7 litteras arbitratus Ritschl, vir doctissimus, verbum odziciunt et aetate et integritate par esse illis plurimus, universi, sequuntur vocabulis iudicavit in Proleg. Trin., 1848, p. 92. Brambach quoque in libro de Latine scribendi ratione, quem Die Neugestaltung der Latei- nischen Orthographie inscriptum a. 1868 edidit, quamquam unam / imperi aetate plerumque scriptam fatetur (p. 201, vs. 5), tamen cum vocali syllaba praefixa exiret, propter illum pleniorem sonum # litterae vocalibus interpositae duplicem # tantum non semper scriptam existi- mavit (vs. 19). Atque post consonante exeuntes praepositiones esse auditam # consonantem demonstrare adfirmat exempla a Gellio (4. 17) prolata, quorum prima syllaba positione producatur (vs. 11), atque hanc # consonantem scribi quoque iubere ipsum Gellium et Priscianum (vs. 24). Itaque qui lingua Latina hodie utuntur, eos
lactend: Verbi Composita. 85
iubet post vocalis duplicem #, post consonantes vel duplicem ponere vel simplicem (p. 202, vs. 7: ‘Es lasst sich demnach fiir unsere Orthographie die Regel aufstellen, dass nach vocalisch auslautenden Prapositionen das doppelte I zu schreiben sei; nach consonantisch auslautenden Prapositionen ist sowohl doppeltes als einfaches I gestattet, jenes nach der Theorie, dieses nach der vorherrschenden Praxis”). In libro autem qui inscribitur Hiilfsbichlein fiir Latei- nische Rechtschreibung a. 1872 in lucem emisso, § 20, II, unam # scribendam dicit et post vocalem et post consonantem.
At Lachmann ¢ duplicem non scripsisse antiquos cum perciperet (cf. Comm. in Lucr. p. 128: “ Zz contra veterum consuetudinem ’’), auditam tamen consonantem # putavit ubicumque producerentur syl- labae praefixae. Quem sonum ut significaret scribendo, unam autem atque unicam # litteram scriberet, 7 formam adhibuit. y7cé quidem, dejti, projci, conjte, objcitur, injce, adjceret, projctam scripturas in operis supra commemorati pagina 188 protulit. Quo autem modo haec enuntiari voluerit p. 136, si quid video, significavit his verbis: “Notissimum est cojcere sive coiucere.”’ Non alium igitur effecit sonum ac Wagner et Brambach. Fuisse autem ubi ex ze litteris constaret apud antiquos syllaba quae praepositionem subsequeretur, velut conteciant, obieciemus, alia, ex scripturis quibusdam conlegit quas satis paucas invenerat in libris manu scriptis; atque in sua Lucreti editione hanc scripturam aliquot locis restituit. Quibus de rebus conferas, si vis, illud commentarium, p. 128. Sed / litteram lingua Latina alienam, quamquam neque in sua scripta neque in editiones auctorum veterum adsumpsit Lachmann ipse, Vahlen tamen in ‘editione Ennianae poesis reliquiarum a. 1854 edita, et Ribbeck in prima editione scaenicae Romanae poesis fragmentorum, quam inter annos 1852 et 1855 edidit, omnibus locis adhibuerunt ubi produ- cendam primam esse syllabam ex numerorum apparet ratione. Huius rei exempla videas in altero opere in Annalium vs. 75, conjcit, in altero p. 72 prioris voluminis ubi est frojcit, posterioris autem p. 66, s#jci. Mox Lucianus Miller in libro quem scripsit de re metrica a. 1861, p. 249, dicit apud poetas dactylicos usque ad Pro- pertium # litteram horum actendi compositorum consonae habuisse instar; sonum autem cum non indicet, veri simile est non alium ac Lachmann eum voluisse. Ejius autem quod est zecso formas demum
86 Mauricinus WW. Matker.
imaperi aetate scripsisse non nullos putavit, quo facilius productionem praepositionis apud veteres poetas explanarent. Iilum Lachman- nianum sonum Guil. Schmitz quoque admisit ubi consonante finitur praepositio. Sin autem vocalis est extrema praepositionis littera, duabus vocalibus ita interpositae s consonantis sonum esse aliquando saltem mixtum, spreta ea volgari quadam synaloephe cucere, escere, vewere, al., docuit ille m quaestione de I geminata et de I longa, quae commentatio programmati gymnasi Marcodurani 2. 1860 inserta nunc commodissime reperitur in libro eius qui inscribitur Beitrage zur lateinischen Sprach- und Literaturkunde, apud Teubnerum a. 1877 impresso, p- 70 sqq- Hunc sonum sic repraesentat scri- bendo : “ proijecit (praet. temp.). coijicit, eijicit.” Postremo H. A. I. Munro, ut Angium denique nominem qui de hac re sententiam dixit, Lachmanno prorsus adsentiebatur. Namque cum adico vel abiecwo alia solas esse antiquas scripturas credidit, ut potest videri ex commentano in Lucretium 1. 34 et 2.951, tum priorem esse dictam et dicendam proinde quasi Anglice zéy:czv scriberetur, in libello qui inscribitur A Few Remarks on the pronunciation of Latin a. 1871 edito, p. g breviter exposuit.
Lexicis porro Latinis non semper possumus credere; quorum quod praecitpuum in nostra patria tenet locum, illud dico quod a Lewis et Short recognitum Harper's Latin Dictionary inscribitur, et illud minus a Lewis solo a. 18go editum, quod inscribitur Elementary Latin Dictionary, inducunt formas vel simplici vel duplici ¢ littera scriptas, sed quam inconstanti consilio rationeque perspici potest animadversis his exemplis : —
L. et S. s.v. abicio, dd%io or aéyic- (in the best MSS. abicio).
Lewis s.v. abicio, aduzo (a usu. long by position) or adsicio.
L. et S. sv. elcio, azo (or icio).
Lewis Sv. eicio, ¢cio (pronounced, but not written, &iicio).
L. et S. s.v. subicio, s#diciu (less correctly sudjtctv: post-Aug. sometimes sisd-).
Lewis s.v. subicio, sudscw (the first syl. usu. long by position ; often pronounced and sometimes written
sublicio).
lactends Verbi Composita. 87
His variis opinionibus doctorum nostrae aetatis virorum quadam in praefatione praepositis, iam demum veterum testimonia de hac re eO pergamus consilio inspicere, ut primum ea verba tractemus quorum praefixae syllabae consonante cadant, deinde postea quorum vocali.
II. JACIENDI VERBI COMPOSITA QUORUM PRIOR PARS CONSONANTE CADIT.
Modum volgarem haec iaciendi composita in praesentibus tempo- ribus dicendi quorum praepositiones consonante finiuntur temporum cursu se immutasse testimonio sunt poetae. Omnium enim formarum apud poetas repertarum qui ante Augusti obitum scripserunt, praeter . solas quattuor apud Plautum? et Naevium,’ produci primam syllabam vel licet vel necesse est. Nam cum maior pars eorum quae sunt exempla apud scaenicos poetas praepositionis mensuram non prae se fert,? non desunt tamen quae productionem sine ullo dubio osten- dant‘; atque quae in herois quidem reperiuntur, illa omnia produ- cenda sunt.’
Sed post Augustum mortuum, vel adeo fortasse paulo ante, alia consuetudo orta est, qua illa mensura, quam Plautum et Naevium quater certe adhibuisse diximus, volgaris est facta. Cf. Moret. 96 :—
Spargit humi atque ddzczt# . . . German. 196 :—
Qualis ferratos s#dicz# . . .
Conferas porro infra, p. 145 sqq., si libet, ut usum perspicias reliquo- rum. Hanc tamen correptionem, quamvis plerumque acceptam, omnino vitaverunt Valerius Flaccus * et Serenus Sammonicus,’ semel adsumpsit Statius*; neque productionem plane excluserunt alii. In sermone autem constanter correptas esse has praefixas syllabas con-
1 Cf. ind. Plaut. b; etiam p. 99.
2 Cf. ind. Rell. b.
3 Cf. ind. Plaut.c; Ter. b; Rell. c.
4 Cf. ind. Plaut.a; Ter.a; Rell. a
§ Cf. ind. Enn. a.
® Cf. ind. Val. a.
7 Cf. ind. p. 151; cf. etiam Ausonium, ib.
8 Cf. ind. Stat. b; cf. etiam Claudianum b, p. 151.
88 Mauricius W. Mather.
sonante exeuntes secundo saltem saeculo’ dilucide docet Auli Gelli narratio in 4. 17, cum dicit plerosque, ut apud veteres poetas ratio- nem numeri servarent, vocalem praepositionis produxisse. Quarto quoque saeculo correptionem testatur Marius Victorinus, quem in optimorum numero grammaticorum esse ducendum existimavit Ritschl in Museo Rhenano XIV (a. 1859), p. 302. Conferas illum locum (K. VI. 67. 17) quo dicit versum Vergilianum (A. 11. 354) qui incipit a verbis Adicias nec te dxépadov esse, veluti capite imminuto.
Quo modo igitur haec commutata ratio intellegi possit, qua, cum primo produxissent praefixam consonante finitam syllabam, deinde corripiebant, nunc tandem consideremus. Si unam é litteram scribe- bant, nullum obicit scrupulum haec contractio, sed quem tum fuisse modum et scribendi et dicendi arbitremur cum poetae productis uterentur praefixis syllabis?
Principio igitur grammaticorum, ut par est, videamus testimonia. Apud Quintilianum sunt haec verba (1. 4.11): “littera # sibi insidit ; consicet enim est ab illo sacét.”’ Gellius haec dicit (4. 17. 6 sqq.): “Sed neque of neque sub praepositio producendi habet naturam, neque item con, nisi cum eam litterae secuntur, quae in verbis constituit et confecitt secundum eam primae sunt, vel cum eliditur ex ea # littera, sicut Sallustius: /aenoribus, inquit, copertus. In his autem, quae supra posui, et metrum esse integrum potest et praepositiones istae possunt non barbare protendi; secunda enim littera in his verbis per duo 4, non per unum scribenda est. Nam verbum ipsum, cui supra dictae particulae praepositae sunt, non est icio sed iacio, et praeteritum non ‘cif facit, sed secit. Id ubi com- positum est, @ littera in # mutatur, sicuti fit in verbis szsiZio et incipio, atque ita vim consonantis capit, et idcirco ea syllaba productius latiusque paulo pronuntiata priorem syllabam brevem esse non patitur, sed reddit eam positu longam, proptereaque et numerus in versu et ratio in pronuntiatu manet. Haec, quae diximus, eo etiam conducunt, ut, quod apud Vergilium in sexto positum invenimus (vs. 366) :—
Eripe me his, invicte, malis: aut tu mihi terram Inice,
1 Cf. etiam ind. Inscr. a, ddics? in vss. hexametris a. 136 p. Chr. n. scriptis.
laciend: Verbt Composita. 89
sic esse imsice, ut supra dixi, et scribendum et legendum sciamus, nisi quis tam indocilis est, ut in hoc quoque verbo ## praepositionem metri gratia protendat.” Prisciani autem testimonium hoc est (K. II. 126. 18): ‘solet plerumque in compositione a in # converti, ut ‘cado incido,’ ‘facio inficio,’ ‘iacio inlicio’” ..... At contra haec Servius in commentario Vergiliano (4. 549): ‘‘odbsto, reicio, adicio ‘i’ habent vocalem sequentem quae per declinationem potest in consonantis formam transire, ut obsect, reteci.”
Ex his locis apparet grammaticos plerumque duas # litteras censu- isse scribendas, alteram consonantem, alteram vocalem. Quod si fit, positione longa redditur prima syllaba, atque sonus qui auditur est ille cuius omnes, qui de hac re recentioribus scripsere tempori- bus, fuerunt auctores.!
Nusquam tamen haec ratio grammaticorum firmatur ex inscriptio- nibus. Unum exemplum mihi repertum (sméice, cf. ind. Inscr. c) in Falsis continetur, in quibus haud scio an alia inveniri possint; non enim qua cura veras, ea falsas examinavi.
Neque in libris quidem manu scriptis est frequens geminata ¢ littera usque ad duodecimum vel potius proximum saeculum. Namque in poetarum libris haec fere sunt exempla, quantum quidem conligi potest ex variis scripturis quas optimae editiones suppeditant :—
Plautinorum? codicum in Lipsiensi uno est duplex ¢ saepe repertum, neque in hoc ubique. Hunc codicem, qui F volgo, L in Truculenti editione Schoelliana notatus est, saeculo XV scriptum ab Italo librario, verba poetae partim vere correcta, partim etiam magis cor- rupta atque licenter interpolata exhibere dixit Ritschl in Trinummi praefatione.* Decurtatus (C) saeculi XI, et ante rasuram Ursinianus (D) saec. XII habent scriptum odsicere in Mil. 619, sed in D alterum # est erasum. Iterum habet C odsicere in vs. 623.
Tibullianus * codex Guelferbytanus (G) saec. XV praebet dupli- catam # litteram in utroque exemplo éaciendi verbi praepositioni subiecti, conticit 1.8. 54, subiiciet 1. 5. 64.
1 Cf. sup., pp. 84-86.
2 Cf. ind. Plaut. a, b, c.
8 Schoell. edit., 1884, p. xiv. 4 Cf. ind. Tibul. a.
go Maurictus W. Mather.
Vergili' Palatinus-Vaticanus (P) saec. IV-V 3 bis exhibet, disiicst in A. 12. 308, ubi manu secunda deletum est alterum 4, et superiicit ib. 11.625. Tum schedae rescriptae Veronenses (V) saec. IV-V (?) habent in A. 7. 339 @ésiice ut videtur, supra scripta fortasse s littera. Cf. Ribbeck. ad h.l.
Ovidiani? codices sunt nulli ante nonum saeculum descripti. Haec # scripturae exempla in editionibus enumerata invenio, ad quae nimirum alia accedere possint; Bodleianus B. N. Rawl. 1oz (Q) saec. XV habet #7 sive #7 constanter in Tristibus. Holkhamicus (H, v. Owen. edit. p. xx) saec. XIII exhibet conncaendus in T. 3. 11. 46, Vaticanus 1606 (V) saec. XIII odsicrunt ib. 5. 10. 40, Ns codices incertae aetatis-(cf. Merkel. edit. 1841, pp. cclxxix et cclxxxil) adtice in F. 1. 189.
Moreti® carminis Vaticanus 3252 (B) saec. IX et Helmstadiensis 332 (H) saec. XV et Basileensis (e) saec. XV tradunt aduatur in vs.99. H « adticit in vs. 96.
Senecae‘ recensio interpolata (A), quae ad tempora medio XIV saeculo. haud superiora pertinet, adijce praebet in Med. 527.
Valeri Flacci® Monacensis (M) saec. XV, qui est optimorum huius poetae codicum, dés#ci¢ in 3. 162 manu secunda habet ex dzsicet.
Sili® Florentinus (F) saec. XV, qui secundus est ad Laurentianum omnium optimum, habet in 9. 538 dissiyce manu secunda scriptum ex aisice vel disire.
Iliadis Latinae’ codices Erfurtanus (E) et Leidensis (L), qui ambo saeculo XII scripti huius carminis sunt excellentissimi, tradunt alter aisticeret, alter dis™ceret in vs. 325.
Martialis® denique codices duo interpolati (b ¢) saec. XV profe- runt in ro. 82. 1 illud quod numeris obstat, adzicst.
Qui autem verba soluta scripserunt, in eorum codicibus ne unum quidem exemplum 7 duplicis credo inveniri posse, nisi in iis qui post octavum certe saeculum exarati sunt. Plures enim qui ante septimum saeculum sunt descripti, post autem partim rescripti sunt, quo modo
1 Cf. ind. Verg. a. 6 Cf. ind. Val. a. 2 Cf. ind. Ov. a. 6 Cf. ind. Sil. a. * Cf. ind. p. 145, Moret. 7 Cf. ind. Il. Lat. a
4 Cf. ind. Sen. b. 8 Cf. ind. Mart.
laciendi Verbi Composit. 91
eius verbi quod est sacere formas praesentis. qcae praepositionibus adfiguntur, nobis tradant. ex apographis quibusdam sive ex adno- tationibus criticis editorum cognovi, neque etiam seme! iam gem? nationem inveni. In ila enim ss'safer scripmra.’ quam habet in- Plini N. H. 13. 129 codex rescriptus Veronensis sive San-Paulinus saeculi IV-VI,? illud # supra positum recentior manus addidit Nam in ceteris horum compositorum exemplis. quae scnt hoc in libro, # simplex constanter scriptum legitur. Item m Gai Institutionibus 3- 119 codicem Veronensem saec. V-VI? non habere iliius @ vestigium recte iudicavit Lachmann, ubi dixit in sua editione, “ Duplex ; in codice esse non credo.” Quod enim a Studemundo traditum est in huius codicis apographo p. 159 in versus 11 fine, £.D/x(C/ ab initio proximi versus est perspicuum). id non est a codice descriptum, sed AD litterae solis schedis Goescheni debentur, et /x ex schedis Bluhmianis recipiuntur. In illo spatio x notato cum Biuhme olim exstitisse opinaretur, parum perspexisse oculos mihi certum videtur. Nam cum in extremo versu aliter potuit maculam pagina facile accipere, tum constanter in codice 4, non #, scribitur.* Horum praeterea codicum scripturas perspexi : —
Ciceronis libri de Re Publica Vaticani palimpsesti in lucem a. 1822 editi ab A. Maio (Auct. Cl. I), qui secundo vel tertio saeculo eum exstitisse suspicatus est (cf. o. c. pp. Ixv-Ixvi); Teuffel autem, § 184. 5, quarto saeculo attribuit.
Ciceronis in C. Verrem orationum partes complectentis Vaticani palimpsesti saec. IV (?) (cf. Teuffel. R. L.5 § 179. 6-11. 3) quem Maius a. 1828 edidit in Auct. Cl. II, p. 390 sqq.
T. Livi codicis rescripti Veronensis quem partes librorum III-— VI continentem Theodorus Mommsen a. 1868 descripsit et edidit. Saeculo IV eum esse scriptum censuit ille p. 158.
T. Livi codicis Puteani sive Parisini 5730, qui saec. V vel VI descriptus maximi est ad libros emendandos a vicesimo primo ad
1 Sillig. edit. 1855, VI. p. 10. 15.
2 Ib. p. xx et Teuffel. R. L-5 § 313. 6.
2 Cf. Studemund. Gai Instit. cod. Veron. Apographum, p. xx.
* Cf. Studemund. p. 320; deict, inicio, obicere, subicere formae satis frequenter occurrunt in codice.
Q2 Maurnicius W. Mather.
tricesimum (cf. apparatum criticum et prolegomenum Luchsianae editionis librorum XXI-XXX duobus voluminibus annis 1888-89 perfectae).
Codicis Vaticani n. 5766, in quo insunt Iuris Anteiustiniani Frag- menta quae dicuntur Vaticana, quem librum saec. V exaratum (cf. Mommsen. edit. p. 389) addita transcriptione notisque criticis edidit Th. Mommsen a. 1860. |
Frontonis codicis palimpsesti Bobiensis, qui ad saeculi VI initium pertinet (cf. Naber. edit. 1867, pp. xi-xii; etiam p. 280 ubi dicitur hic codex # pro # constanter habere in compositis sactendi verbi).
Codicis Fuldensis Novi Testamenti, qui, cum pertinet ad quintum decennium saec. VI (cf. Ern. Ranke. edit., 1868, p. viii), opus Hieronymi nobis tradit. In praeterito quoque ¢ litteram pro ze inter- dum habet codex, velut cierunt p. 109. 16 R.=Eu. Matth. 21. 39, et p. 118. 13 R.= Eu. Iohan. g. 34, alia (cf. Ranke. p. xxvii).
His igitur inspectis ne unum quidem exemplum # scripti inveni. Haec ergo geminatio quam raro sit scripta post consonante exeuntem praepositionem apparet. Nam quater tantum ante XII saeculum se mihi obtulit quaerenti, in Vergili videlicet P et V libris qui ambo ad IV vel V saeculum pertinent, in Moreti B qui IX saeculi est, in Plautique C XI saeculi, neque est frequens nisi XIV saeculo et XV. Cf. porro pp. 116-18 infra, ubi de hac re agitur cum vocalis est extrema littera praefixae syllabae. -
Itaque grammaticos fuisse solos huius scripturae auctores videmus, nec quod in sermonibus duplicem ¢ audirent, eo praecepta sua dabant, nam unum #/ dictum esse usque ab Augusto mortuo plane demon- strant cum correpta apud poetas prima horum syllaba verborum, tum ea quae narrant? Gellius (4. 17) et Marius Victorinus (K. VI. 67. 17) ; sed quia non solum similitudo poscere videbatur, ut acto verbum, cum praefixa esset praepositio, @ litteram in ¢ converteret,* verum etiam versus antiquorum poetarum productas ostendebant primas syllabas, id quod vix intellegi posse putabant nisi consonans # adesset, ut positione fieret productio.* Nec quisquam dubitare potest, quin,
1 Cf. pp. 88-89. 2 Cf. Quint., Gell., Prisc. locos supra, pp. 88-89, citatos. 8 Cf. Gell. ib.
lactendt Verbi Compostta. 93
quae duplicis ¢ in codicibus exempla exstant, omnia grammaticorum praeceptis debeantur; etenim cum, litteris iam renascentibus, fre- quentissima fierent, tum maxime, lingua Latina iam non viva vigente, nitebantur docti in grammaticis antiquis.
Duplicem 7 ergo in verbi saciendi compositis, quorum consonante finiretur prior pars, non scripsisse antiquos praeclare patet, neque quisquam, nisi Brambach? in priore libro nostris temporibus id scri- bendum docuit. Auditos tamen et consonantis et vocalis # sonos quoad productis uterentur primis syllabis, id docuerunt multi.
Huic autem doctrinae id obstat, quod temporibus de quibus agitur, id est antequam Augustus e vita excessit, geminationem eam littera- rum, qua altera consonans fieret, altera vocalis, tolerare noluerunt Romani. Itaque quae vocabula recto casu aus vel cius litteris termi- nantur, ea omnia genetivo singularis et nominativo pluralis az, ¢ litteris cadebant, dativo et ablativo pluralis as, es, numquam autem Git, cit, atts, eits.2 Neque posteriore quidem tempore nisi apud gram- maticos hoc alterum # accessit. Formae quae sunt fatris, auxiliis, similia, huc non pertinent, quoniam in eis utraque # littera vocalis habet instar ; atqui etiam ea antiquiores z¢/, zeés litteris terminaverunt.
Primum igitur ad id testimoni quod est in inscriptionibus animum quaeso intendas. In Corporis Inscriptionum primo volumine, quod titulos continet qui ante Caesaris obitum incisi sunt, cum verborum asus, etus \itteris finitorum genetivi singularis, nominativi dativi abla- tivi pluralis nullum sit exemplum, satis tamen faciunt haec exempli gratia citata: CI. II. 1129 PLEBEI*®; 1587 POMPEI; 4970. 51- 61 ATEI; 3695, a. 6 p. Chr. n., MAIS; IV. 1180, quae probabiliter
1 Cf. pp. 84-85.
2 Item volgus, eguos vocabula et quae sunt similia primo demum saeculo p. Chr. n. duplicatam # litteram patiebantur, ut vudgus, eyuus, alia. Cf. Quint. 1. 4. 11.
% 7 longam ne quis arbitretur indicare mixtum # consonantis et # vocalis sequen- tis sonum, nam non aliter sonari videtur quam # volgaris; sed ratione errorum, velut DICTATORI, CI. I. 584, non habita, productam vel vocalibus interposi- tam : litteram significat, velut in hoc eodem titulo Sulla dictatore inciso, exstat FELICI, et in 1079 et 1418. 19, quae incertae sunt aetatis, EIVS. Raro quoque / longa, ita ut # volgaris, scriptum legitur pro duabus # vocalibus, ut FILIS in CI. IT. 3877; MVniCIPI 3708; CONTROVERSIS in I. 199. 45, a. 117 a. Chr. n., et in Ter. Eun. 256-7 CVPPEDINARI, CETARI, LANI in A scribuntur, ubi numerorum ratio poscit ¢ duplicem.
94 Mauricius W. Mather.
Tiberi aetate incisa est, MAI, POMPEIS, [all]EI (cf. ALLEI in 1179); 1181, 1186, 1189 POMPEIS.
Libri quoque antiquissimi cum identidem duplicatam ¢ litteram habent, tum saepissime simplicem, velut in Livi codice Veronensi (cf. p. 91) sunt pledeis in 5. 2.13; Veisin 5.4.10; 5. 46. 4; §. 52. 10 (Beis); 6. 4.5; alia, Vesss autem in 5. 5. 10. cf. etiam Neue. Formenl. I.? gg.
Iam ea quae grammatici praebent ad hanc rem pertinentia con- sideremus. Probus igitur Berytius, qui Domitiano imperatore floruit, docet (K. IV. 104. 22) Auius Gat, hi Gat, his Gais, quaaumquam eum fuisse ipsum opinaremur, qui geminam # litteram scribendam iuberet. Apud Servium enim (Aen. 1.1) per unum # Zroiam, Graios, Atax scribi vetuisse dicitur' (cf. Ribbeck. Proleg. Verg., p. 138 et 174). Cassiodorius autem, qui sexto saeculo vixit, haec habet (K. VII. 206. 10): “Quidam huius Pompei Tarpei, hi Pompei Tarpei, his Pompeis Tarpeis per unum # scribunt; non nulli pusillo diligentiores alterum this addunt. Ego quoque tertium ? addendum praecipio, si enim, ut docui (vs. 6 infra, p. 108, citato), nominativus duo ¢ habet pro duabus consonantibus, haec perire nulla declinatione possunt, quae tamen omnimodis modernus usus excludit,” ubi verba illa, “quae tamen omnimodis modernus usus excludit,” ab illo adiecta videntur, reliqua vero ex Caeselli scriptis, qui saeculo altero p. Chr. n. floruit, sunt adsumpta (cf. Guil. Schmitz. Beitr., p. 73, adn.). Ac Terentianus Maurus qui hoc saeculo extremo floruisse putatur, in versus suos induxit Grdi* vocabulum, quam licentiam minus facile potuisset adhibere si ¢ consonans adfuisset.
Aliquid porro testimoni addunt illa Graiugena, Troiugena vocabula, quae, nisi Gratigena, Trotigena putidum et insolitum sonum dedissent, nullam sane causam sibi habuissent.*
1 Cf. tamen Prob. (K. IV. 221. 8 et 257. 17) infra, p. 108, prolatum.
3 Cf. Prisc. (K. II. 14. 10): Unde Pompeisi quoque genetivum per tria # scribe- bant, quorum duo superiora loco consonantium accipiebant, ut si dicas Pom fells ; nam tribus # iunctis, qualis possit syllaba pronuntiari? Quod Caesari doctissimo artis grammaticae placitum est. (Cf. Lachmann. Comm. in Lucr., p. 371.)
8 De Syll. vs. 453, K. VI., p. 339-
4Illae Z7rotiugenas, Graiiugenarum (Graliug. A) figurae, quas Lucretium 1. 465 et 477 scripsisse, codice Leidensi A saec. IX teste, censuerunt Lachmann et
Tactendt Verbt Composita. 95
Itaque solus, qui as, ¢# sonis favere videatur, est Caesellius. Priscianus enim (p. 94, adn.) id tantum agit ut rationem eius scrip- turae reddat quam male! arbitratur antiquorum esse. Namque Probus qui docuit Grasos per ¢ geminum oportere scribi, tamen Auius Gaz, hi Gai, his Gais iussit; atque Cicero cum aso, Matiam, Attacem scribere vellet,? nusquam dicitur Gait, Pompeiti, neque etiam Gait, Pompeii similiaque scripsisse. Caesar autem si re vera per triplicem t Pompeii et alia maluit scribere, eadem, qua Caesellius, cura com- motus duas # litteras quas in nominativo scribebat, ut illum pleniorem g consonantis sonum inter vocalis positae exprimeret (cf. p. 107 sq. infra), declinatione perire noluit. At si ita scripsit, nihil eo demonstratur consonantem / enuntiavisse. Enuntiaverit autem; mos modo grammaticus fuerit, qui se in usu volgari consuetudineque numquam firmavit.
Brambach igitur, cum in libro Die Neugestaltung der Lateinischen Orthographie, pp. 197-8 praecipit, ut ab iis, qui hodie Latine scri- bant, ## et és litteris exprimantur plurales nominativus, dativus, ablativus omnium vocabulorum quorum casus rectus aius vel eius litteris exeat, atque haec commendare studet non solum ex illo Prisciani loco et illo VERTVLEIEIS (CI. I. 1175) — quamquam hoc certo non demonstrat consonantem # litteram, excipiente # vocali, ut Vertuleiis, auditam esse, — verum etiam ex silentio, ut praedicat, grammaticorum de # et ¢s litteris scribendis. Sed in hac re mihi non videtur obtemperandum doctissimo viro.2 Namque quod testi- moniorum habemus ex inscriptionibus, libris, grammaticisque, id universum adversatur ei haec praecipienti. Concedit ipse quidem in titulis per unum / scripta interdum inveniri huius modi verba, atque laudat MAIS Or. 5614 a. 219 p. Chr. n., 6112 quod est I. R. N. 3571, a 387 p. Chr. n. POMPEIS Or. 5814, 6167 (5814=CI. IV. 1180; 6167 =IV. 1189, cf. pp. 93-94, sup.).
Munro, non consonantis ¢ sonum, vocali s sequente, nec vocalis quidem, conso- nante sequente, significant, sed consonantis tantum. Cf. p. 107 8q. infra.
1 Cf. Lachmann. in Lucr. p. 371; Ribb. Proleg. Verg. p. 138: sed Brambach (Lat. Orthog. p. 195) adsentitur Prisciano.
2 Cf. Quint. 1. 4. 11; Vel. Long. K. VII. 54. 16 infra pp. 107-08, prolatos.
* Cf. autem eius Hiilfsbiichlein fiir Lateinische Rechtschreibung § 14, ubi quasi in palinodia simplicem # dicit melius scribi.
96 Saurectus W. Mather.
Ergo mihi quidem certum esse videtur Romanos aéxat, adtict reliqua non magis dixisse quam scripsisse: Quintilianum autem et Priscianum (pp. 88-89 sup.), cum comsest et iniscio esse ab tactendo dicerent, noluisse # consonantem, praepositione ante se _posita, omnino amitti in figuris praesentibus, atque Gellium' non solum hoc eodem scrupulo commotum esse, sed ratione quoque, qua poetae antiqui haec verba ad versus accommodarent, quae ratio scilicet aliam mensuram praefixae syllabae poposcit atque ipsius tempore volgo audiebatur.
Haec igitur ratio pronuntiandi si spernenda est, nec vocalem praepositionis, cum consonante exit, produci licet nisi si s vel / littera x vel con praepositionem insequitur,” quo tandem modo arbitremur haec verba, libera re publica, dixisse Romanos?
Non desunt in libris, nec etiam in inscriptionibus, indicia quae persuadeant, id verbum quod est zzczv, cum praepositionibus con- sonante finitis subiungi inciperet, 2 litteram cum ¢ littera commutasse. Cuius rei inscriptiones unum exemplum praebent, conmsectan?* in CI. I. 198. 50, a. 123 sive 122 a. Chr. n (v. CI. p. 56).
In Plautinis* libris, etsi haec per ¢ facta scriptura non apparet, vestigia tamen eius quater deprehendimus. Poen. 1174 (octon. ana- paest.) : —
Fuit hédie operae pretitim quoivis qui amdbilitati animum daiceret
ubi adecerit habet F (quo de codice v. p. 89 sup.), adiceret A. Mil. 112: — Conitcit in navem miles clam matrém suam ubi contegit quod est in CD codicibus videtur ex coniecit corruptum. Truc. 298 :— . Ut pereat, ut eum ¢iciatis {n malam fraudem ét probrum
inleciatis (corr. illeciatis) habet A quod Schoell retinet, inzceatis BCD.
1 Ribbeck in Proleg. Verg., pp. 138-9 putat Gelli praeceptum totum esse Probi, vel saltem a Sulpicio Apollinari (cf. Gell. 4. 17. 11) ex huius commentariis excerp- tum. Cf. Kretzschmer. de Gell. Fontibus, p. go.
8 Cf, Gell. 2.17; 4.17.6.
§ Cf. ind. Inscr. a.
¢ Cf. ind. Plaut. a, c.
lactends Verbi Compostta. 97
Mil. 623 :— Edm pudet me tfbi in senecta odfcere sollicitidinem
obveccere in D fortasse ex odiecere est corruptum ; odicere¢ ante rasuram habet B codex, obstcere C. Terenti! codices unum exemplum praebent, Ad. 710: —
Itaque 4deo magnam mi é#sect¢ sua c6mmoditate cdram
ubi snzecit, quia A cum reliquis (¢ in ras. F) et Donatus in lemmate habent, videtur pro praeterito habitum esse ; quam prave autem, indicat cum contextus sermonis, tum proximus versus, ‘“ Ne impru- dens faciam” ...
In ea sententia, —
Fértior qui cipiditates ést quam qui hostes szdicit,?
quae a Vincentio Bellovacensi citata nunc commode inspici potest in Ribbecki Comicorum Fragmentis, p. 368. 49, sudecit traditum est a et y codicibus. Huic scripturae quamvis causae potuisset esse obscura -distinctio # brevis et ¢ brevis sonorum,’ quoniam tamen prima producta est syllaba, fortasse olim sudsecs¢ scriptum est, unde ¢ consonans incuria excidit.
Ciceronis‘* operis de Divinatione A et V codices, decimo saeculo descripti, habent adsecit in 1. 48. 106 (vs. hexam.) : —
A biecit efflantem et laceratum adfligit in unda. In Vergili® Aeneide 6. 421 : — Obicit. ille fame rabida tria guttura pandens
Priscianus (K. II. 349. 15) odsecét legit, testibus libris Lugdunensi et Sangallensi (odiecit G), qui uterque ad nonum saeculum pertinet.
T. Livium hanc scribendi rationem interdum certe adhibuisse bene testantur ex optimis codicibus non nulli. In ro. 8. 3:—
et nunc tribunus . .. quinque augurum loca, quattuor pontificum adiczt, in quae plebei nominentur eqs.
1 Cf. ind. Ter. a.
2 Cf. ind. Rell. a.
* Cf. Luc. 7. 574 ubi pro sudscs¢ habent non nulli libri ssdiget vel subegst. * Cf. ind. p. 138, Cic. a.
5 Cf. ind. Verg. a.
98 Mauricius W. Mather.
et in 10. 37. 14:—
Fabius ambo consules . . . res gessisse scribit . . . sed ab utro consule, non adicit eqs.
adiecit habent tres praecipui ad scripturam librorum a primo ad decimum confirmandam codices, Mediceus (M) saec. XI, Parisinus 5725 (P) saec. X, Upsaliensis (U) saec. X-XI. Cf. Madvig. Emend. Liv. edit. sec., 1877, p. 226 et Miiller-Weissenb. edit. Liv., 1890, app. crit. ad hos locos. Librorum autem a vicesimo primo ad tricesimum praecipuus codex, Puteanus sive Parisinus 5730 (P) saec. V-VI, habet in 22. 19. 2: — |
Hasdrubal ad eum navium numerum.. . decem adicit, quadraginta navium classem Himilconi tradit eqs.
adiecit, quod praesentis esse temporis indicat illud ¢rad:t verbum quod sequitur. Reiecta tamen hac scriptura, adtectis participium supponunt Luchs, Zingerle, Miiller in edit. oct. 1891. Item in 26. 19. 2 habet P codex udsecere pro subiere, cf. Luchs. app. .crit.
Plinium autem hunc modum scribendi usurpasse non credo, etsi in N. H. 7. sect. 1. 2 abiecit habet V codex (K a Silligio notatus) saec. XI, adicit reliqui. Eius enim tempore non iam sect, sed st moris erat in sactends verbi compositis.
Nec illorum vetustissimorum codicum, quos pp.g1-92 supra laudavi, ullus praebet huius rei exempla.
Ex Lachmanni commentario Lucretiano, p. 128, hoc excerpsi cuius prior pars consonante terminatur!: odsectemus in Digestis Florentinis, quae dicuntur, XLVII. 20. 3. § 1, hoc est, in libro Florentino, qui saeculo VI-VII descriptus, digesta Iustiniana continet. Conferas, si vis, Mommseni edit. 1870, vol. I. p. xxxx.
Schuchardt denique in libro, quem, Vokalismus des Vulgirlateins inscriptum, a. 1867 prodidit, vol. II. p. 4 haec profert ex codice Gothano Euangeliorum, qui saeculo septimo (cf. ITI. p. 4) scriptus est: adtecientur 44. b. 12; 319. b. 15; snsecient 362. a. 2. Haec exempla tametsi mihi non fuit facultas ut comprobarem, facile tamen
1 Illud consecsant, quod Lachmann ex Lege Servilia profert, non aliud est atque meum exemplum quod datur in Indice I, a. 2.
laciends Verbi Composita. 99
illi viro docto credere possumus, namque alias accuratissimum et fide dignissimum eius esse librum inveni.!
Quoniam igitur cum haec satis multa exsistunt exempla, quae abiécit, adiécit aliaque ostendunt, tum z consonans necesse est adfuisse, quo produceretur syllaba prima, mihi quidem videntur hae formae satis probari. Communis enim volgarisque usus, quo in compositis a brevis in # correpta est, ‘aciendi formas non nisi # consonante eiecta adficiebat ; nam aliter esset profecto auditus ille ¢ consonantis ante z vocalem positi sonus, quem linguae Latinae non fuisse satis plane nobis videmur supra pp. 93-95 indicavisse. Conferas etiam illud societatem vocabulum, quod, etsi in eo non agitur de # consonante, tamen per o litteram in ¢, non in ¢ correptam sonum difficilem durum- que vitavit.
Abiecio ergo, abiectt, adiecio, adiecit et cetera primo audita esse censeo cum taciend: verbo praepositiones inciperent praefigi, neque desita audiri certe inter poetas atque eos qui litteris humanitateque politiores erant usque ad extremam Augusti aetatem. Mutatio tamen illa, qua # littera fieret ab ¢, iam satis antiquo tempore incipiebat haec verba adficere. Quod cum fieret, # consonante omissa, iam praefixa syllaba non producebatur positione, et ddzcio, Gbicit similiaque audiebantur. Has formas autem dixisse hoc tem- pore imperitos solos rudesque, e quibus volgo oriuntur ea quae ex ingenio sunt linguae, est veri simillimum, quia ante Augusti obitum non nisi apud Naevium et Plautum exstant, quaterque tantum apud eos. Cf. Naev.? vs. 94, p. 23 R:—
Immé quos scicidi in iis conscindam atque a’dictam. Plaut.? Asin. 814: —
Praerfpias scortum amdnti atque argentum 0’dzctas. Merc. 932: —
Sdnus non es. quin pedes vos {n curriculum cd’nicttis. Rud. 769: —
Iam hércle ego te cont{nuo barba arrfpiam, in ignem cO’nsiciam.
1 Cf. etiam Seelmann. Aussprache des Lateins, p. viii. 2 Cf. ind. Rell, b. 8 Cf. ind. Plaut. b.
100 Mauricius W. Mather.
Pro hoc coniciam habent CD codices coiciam, itaque ilud conicitis in Merc. 932, ut etiam omnes formae conictenad: verbi quas infra in indice Plauti sub littera c enumerabimus, co:crenai potuerunt esse ; sed hae breviores formae minus placent! apud Plautum, quod nullo altero loco sunt codicibus sustentatae.? Illarum quoque omnium formarum quae sunt in Plauti indice c, Terenti b, Reliquorum c enumeratae, priores partes liceat, si quis velit, corripere. Sic dd:cto, dbiciam et reliqua, non adiecio, abieciam audiamus. Haud tamen probabile hoc esse mihi videtur, nam illa supra dicta quattuor exempla sola sunt, quae primae correptionem postulant; longae autem certo sunt hae syllabae apud Plautum decies, apud Terentium reliquosque scaenicos poetas octies.®
Cum igitur Naevius et Plautus et quivis alius correpta prima syllaba dixerunt haec verba, haud multum licet dubitare, puto, quin etiam scripserint adiao, abicit et alia similia per é litteras.* At non adic, sed adiecio plerumque hoc tempore et dici et scribi solebat, neque est illius ullum formae vestigium, exceptis illis quattuor exemplis, dum usque ad extremum saeculum alterum a. Chr. n. perveniatur. Ex anno 105 a. Chr. n. est nobis servata lex parietis faciendi, quae protcito® figuram continet. Sed septendecim annos prius in lege de pecuniis repetundis lata insculptum est conieciant.® Inter annos igitur 122 et 105 a. Chr. n. haec verba scribendi ita se consuetudo muta- verat, ut etiam in leges, quae veteres maxime amant formas, illa recentior per sc litteras facta scriptura admitteretur. Licet ergo conicere, ut mea fert opinio, illum sonum, qui his litteris significa- retur, satis frequenter paulo ante a. 105 auditum esse, eumque sane in eorum sermone, qui politioris humanitatis non expertes fuissent. Anno demum 44 a. Chr. n. sunt incisa in lege Ursonensi illa inicere et rciciantur,’ neque exstat eo interim spatio ullum in inscriptionibus
1 Cf. Lachmann. in Lucr. p. 188.
2 Cf. ind. Plaut.; etiam p. 122 inf.
8 Cf. ind. Plaut. a, Ter. a, Rell. a.
4 Sic etiam scribebant cum per diphthongum dicta sunt verbi zaciendi composita. Cf. p. 114 inf.
§ Cf. ind. Inscr. a, 3. Huius formae cum vocali exeat praepositio, nihil tamen hoc interest ad hanc rem.
6 Cf. ind. Inscr. a, 2. 7 Cf. ind. Inscr. a. 4 et 5.
lacitendt Verb: Compostta. IOI
exemplum. Itaque, quantum suspicamur, haec forma usque ab anno 105, vel paulo ante, volgari usu scribebatur, atque in inscriptionibus inde ab hoc tempore sola est quam reperimus, in codicibus nostris longe frequentissima.
Verum tamen quamquam iio, scat et reliqua volgo in compositis et scribebantur et dicebantur, tamen secio, :ecit ceteraque apud poetas certe eruditosque maxime valuerunt usque ad Augustum mortuum ; quas formas pronuntiasse eos satis indicant numeri, scripsisse autem, illae scripturae, quas supra pp. 96-98 laudavi. Plerumque tamen qui eorum opera descripserunt, volgari mori obsequentes, illas zeczo formas in éco.converterunt ; mox vetus scribendi modus fere in oblivionem adductus est.
Primum ? illius brevioris formae in herois exemplum dat Moretum carmen in versu 96: —
Spargit humi atque ddici#?. . .
Sed is, qui hoc carmen composuerit, cum productam syllabam numeri gratia in versu gg cuperet, non dubitavit illam veterem mensuram usurpare : —
Caseus ddicttur ...
id quod fortasse indicat veterem sonum nondum periisse, nisi forte mavis putare per imitationem scilicet veterum hoc fieri.
Eorum, qui aetate inferiores fuerunt, poetarum semper correptam primam syllabam adhibuerunt * Germanicus, Manilius, Seneca (sed excipiendae‘ fortasse dissicere verbi formae; cf. ind. a), Lucanus, Martialis, Iuvenalis. Duas habet Phaedrus in senariis formas, quae sublatione incipientes mensuram non significant. Et correptas et productas Silius induxit in versus praefixas syllabas; sin autem excipimus @ssice formam et suferiacit — quo in verbo in temporibus praesentibus simplicis saciendi formae subiunctae praepositionibus
1 Cf. ind. Ov. b.
2 Cf. ind. p. 145, Moret. a.
* Vid. exempla in indice, p. 145 sqq-
* Duo exempla habet ille (cf. ind. c) quorum prima syllaba producta an correpta sit incertum est. Haec duco in correptis, quoniam productionem nisi in a@ssrcere verbo non videtur Seneca usurpasse. Sed etiam hoc in verbo codicibus non credendum est. Cf. p. 126 inf.
102 Mauricius W. Mather.
mutari non videntur!— plures correptas habet. Valerius Flaccus autem et Statius Vergilium imitari ita studebant, ut illa productione paene constanter uterentur, non magis tamen Silio intellegentes quo modo veteres eam effecissent, nisi forte exemplaria habuerunt, quae veterem scribendi rationem conservarint ; itaque aut vocalem prae- positionis tractim pronuntiabant, aut una # littera scripta, sonum consonantis # et vocalis # sequentis in una syllaba efferebant. Semel per neglegentiam, ut opinor, illam sua aetate volgarem? correptionem adhibuit Statius in Thebaide 7. 4:—
Sidera proclamatque ddic# . . .
Eodem modo veterum consuetudinem imitati sunt Serenus, Auso- nius, Claudianus, alii; atque hic pariter cum Statio non caret volgaribus formis.
Haec poetarum testimonia si cum enarrationibus Gelli (4. 17) et Mari Victorini (K. VI. 67. 17), quas supra p. 88 protuli, com- paramus, plane apparet, ut mihi videtur, inde ab extrema Augusti aetate cum correpta prima syllaba constanter audita esse haec com- posita.
Ille autem vetus sonus, quo productio praefixae syllabae efficie- batur, sequentibus illis seco, seat aliisque, quamvis plane periisset inter eruditos litteratosque, ut praeclare demonstrat Gelli locus, ubi numeros apud veteres poetas observasse dicuntur alii vocali praepo- sitionis producenda, alii duabus ¢ litteris, quarum prior consonans esset, efferendis, est tamen putandus, mea quidem sententia, haud ubique omnino periisse, sed interim se propagasse aut inter eosdem imperitos rudesque homines a quibus éco et similia principium duxere * aut in provincialium linguis. Illae enim formae, quae p. 98 supra proferuntur ex Digestis Florentinis et codice Gothano Euan- geliorum, indicio sunt‘ illum zeco sonum in aliquibus locis multo post valuisse, quam Romae inter eruditos certe periisset. Licet
1 Cf. Verg. A. 11. 625 (ind. a) superiacit; Val. Max. 3. 2. ext. 7 superiaceretur habent omnes codd.; Plin. N. H. 7. 2. 2 § 21 superiacé codd. plerique, superars unus.
2 Cf. L. Miiller. de Re Metr. p. 291* = 250'.
8 Cf. p. 99 sup.
4 Cf. praeterea p. 113.
lactends Verbi Compostta. 103
enim Digestorum librarius describere vel imitari potuisset, quae in antiquis legibus viderat, non tamen facile arbitrari possumus, quae in codice Gothano sint exempla, vetere more scribendi adfecta esse, quoniam non ante medium alteri saeculi p. Chr. n. Euangelia Latine reddita sunt.
Unius rei restat ut mentionem faciam, quae ad has consonante exeuntis praefixas syllabas pertinet. Servius enim in opere de Finalibus (K. IV. 450. 12) haec dicit: “Item ex ipsis praeposi- tionibus ad et od et in et sub diversae in verbis ponuntur. Nam corripiuntur cum crescendo disyllabum reddunt, ut adit, obit, init, subit; indifferenter sunt cum trisyllabum faciunt, ut adicit, obicit, intcit, subtcit; producuntur tantum cum tetrasyllabum ex se reddunt, ut adicto, obicto, inicio, subicio”’ Haec igitur cum narret, patet in promptuque est ex usu eorum poetarum, qui dactylicos versus scribe- rent postquam cio sonus volgaris factus esset, argumenta eum deduxisse. Nam, enumeratione mea examinata, cum nullum ei adversatur exemplum, id facile. intellegi potest, quia in dactylicis numeris nulla ratio est, nisi praepositione producta, qua adscio forma et aliae similes in versus induci potuerint. Sin autem illa Naevi et Plauti exempla, ddiciam, dbicias, cinicitis, ciniciam (cf. p. 99 sup.) reminiscemur, Servi verba in loquendi consuetudine non niti praeclare apparebit.
Hactenus igitur, ut iam breviter recognoscam argumentum, haec conatus sum demonstrare: Cum consonante terminaretur praepositio, t2c10 scripturam et sonum apud antiquos in grammaticorum tantum praeceptis exstitisse, nec saepe esse adhibita, quoad litterae rena- scerentur saeculo XIV ; cio autem ferme scriptum legi et in inscrip- tionibus et in veterrimis libris manu scriptis, quamquam semel in illis, in his non numquam apud scriptores, qui ante Tiberium impera- torem vixerint, zecio forma aut exstet, aut vestigium sui reliquerit ; quoniam igitur per illud scio non significetur' sonus, quo intellegere possimus numerorum rationem apud poetas, qui praefixam syllabam produxerint, sin autem seco formam sumamus, facile illa ratio expli- cetur, veri simillimum esse secto non solum scriptum, verum dictum
1 Non adsentior Lachmanno aliisque qui duplicem sonum per unum é scriptum repraesentari docent. Cf. pp. 84-86 sup.
104 Mauritius W. Mather.
quoque esse a poetis politioribusque usque ad extremam Augusti aetatem ; sed scio tamen sonum se prodere iam in versibus Plauti et Naevi, atque ad extremum alterum saeculum a. Chr. n. ita frequenter audiri, ut scribendi ratio respondere sono inciperet, ab eoque inde tempore scio magis volgo scriptum quam ecto: Tiberio autem impera- tore zcio formam iam neque auditam inter eos qui litteris imbuti essent neque scriptam, attamen sive in multitudine imperita rudique, sive inter agrestis et rusticos semper remansisse; verum ico cum inter doctos moris esset continenter, poetas aliquot Vergilium alios- que veteres ita studiose imitatos, ut quam productionem praefixae syllabae apud illos perspexissent, eam in suos versus, quamvis ignari antiquae dicendi rationis, inducerent.
III. JACIENDI VERBI COMPOSITA QUORUM PRIOR PARS VOCALI TERMINATUR.
Iam de tactendi verbo agamus vocali finitis praepositionibus sub- iuncto. Praefixae syllabae, quae ad hanc rem pertinent, sunt septem numero, dé, ¢, prae, pro, co, re, tra. Sed prae praepositio nihil ad hanc quaestionem prodest, quoniam nusquam apud poetas neque in inscriptionibus reperitur. racicientes est apud Festum p. 249. 34 M. = 324. 34 Thewr., sed Columella 8. 17. 10 praciaciuntur videtur scripsisse, atque Praciacttur est Scauri codice Bernensi (B) saec. X traditum (cf. K. VII. 17. 6). Unde forsitan concludas hoc verbum, item ut id quod est suferiacio,’ simplicis verbi formam retinuisse.
Reliquas autem praefixas syllabas, quae omnes apud poetas repe- riuntur, numquam necesse est corripere. In scaenicorum versibus cum possint corripi hae syllabae haud raro, velut in Plaut. Stich. 360 :—
Pérnam et glanditim deicite. hic hércle homo nimium sapit.
Cas. 23 :— Etcite ex animo ciram atque alienum aés; papae. Mil. 205 :— Déxterum ita veménter eicz¢; quéd agat, aegre suppetit.
1 Cf. p. 102 et adn. 1.
lactendi Verbt Composita. 105
Pers. 320: —
Ego réiciam,; habe animém bonum. credétur; commodabo.
Ter. And. 382 :— Aliquam causam quam 6b rem eciat éppido. eicidt? cito.
Ph. 18 :—
Ille 4d famem hunc a sttdio studuit réicere et in aliis,) omnes tamen aut per se longae esse, aut cum sequenti syllaba per diphthongum coniungi possunt. Itaque quoniam hi poetae ipsi exempla praebent, quorum certo producendae sunt syllabae primae, velut Plaut. Asin. 127 : —
Sfcine hoc fit? foras aédibus me ¢ici ?
Laber. vs. 83, II. p. 292 R.: — Hoc voluit clipeum céntra pelvem prdici
et alia,” atque ceteri poetae constanter produxerunt aut per diphthon- gum enuntiarunt, credamus licet omnia illa dubia produci oportere. Qua ratione sum ipse usus in enumeratione mea.
Sed harum praepositionum vocales, cum diversas per se mensuras habeant, sunt tamen omnes adfectae consonantis s excipientis natura. Nam ut consonante,® ita vocali cum praepositiones caderent, saciends verbum non est dubium quin primae sonum litterae diu servaverit.*
Hae igitur sunt productionis rationes: de, ¢, pro praepositiones vocalis litteras habent natura longas, co et fra autem compensant productis vocalibus extremas litteras amissas,° sed re producitur
1 Cf. ind. Plaut. d, e, f; Ter. c,d; Rell. d, e, f ; etiam Lucil. b; Sen. d.
2 Cf. ind. Plaut.d; Rell. d; etiam Phaedr. b.
2 Cf. p. 96 sqq.
* Cf. inf. p. 110 sqq.
§ Cf. Gell. 2. 17. 8 sq.: “coligatus et conexus producte dicitur. Sed tamen videri potest in his, quae posui, ob eam causam particula haec produci, quoniam eliditur ex ea # littera (cf. 4. 17.6) ; nam detrimentum litterae productione syllabae compensatur.” Etiam cf. Verg. G. 4. 257, cénexae; A. 9. 410, conixus; ib. 1. 73, cénubio ; quibus in omnibus verbis co, non con, adhibitum esse apparet ex inscrip- tionibus librisque. Cf. Neue. Formenl. II.> pp. 865-67.
Alia est sententia non deductum esse illud co ex com (con), sed com (con) ex co, nescio quo addito. Confirmationem petit haec opinio ex linguis Germana et
106 Manrutas HW. Mather.
mea tintum ff iEtrera ccampemsanda, quantum / consonantis pro- pretate quadam. qua. cum ¢s est Imter vocalis posita, # vocalis litterae quast umbra e¢ mige ante eam audita. ut ov, Watia, sic adfecit antecedent®s Vocals miccrim, ct, etiam st Lia brevis fuit, produceretur.' Nam cum rzf pratpess} quibasdam verbis praefixa saepe adsimu- lata sit eorcm peumse Ecterze. velat nope.” rettuli, rellatum (Lucr. 2. 1901). reasmdese® (Cie. Rep. 2. 3. 14), radsacv (Lucr. 1. 228, etiam cf. Neue. Formenl II? 9. 923°. aliorum «cf. CI. I. p. 593), tamen numquam / litteram amissam compensat vocali producenda;? atqui ante / Consonantem rr semper producitur, ut in racet Verg. A. 5. 421; react? th. 11. 6352: rem cturs Ov. NL g. $13: retecte Sen. Thyest. 808; vewstat Sik fp. t22: rvs? wh. 15. 726: et aia mult
Gothica; nam alterics sr sviaha praetxa, alterics sc. cognatae sunt, ut videtur, cum cw. (Cf. F. Kiace Esym Worterd. der dectschen Sprache, 1889, s. v. ge-: “~ Verwandtschaft des Pratxes mit lat 1, case st wakrscheinlich.” Sed idem in edit. tert. 1553. scripseca: ist micht sicher.“ Cf. Germ. cesscin et Goth. ga- maces cum Lat. .veemacy (ose in sc de Bacchanalibus CI. I. 196, a 186 a Chr. n. qao tamen tempore omnino non geminabantur consonantes ; cf. secesus ese Bscsmd. VS. 4. wus? VS alia).
Potest tamen cows (.-<) primaria esse forma; nam in lingua Hibemica com, con, co pariter exstant. omnibusjue muneribas praepositionis funguntur, ita autem adhibentur ut crew ante medias ("98 ante 4) et vocals. cv ante tenuis ponatur (cf. Em. Windisch. Irische Texte mit Worterb, 1SSo, p. 450 s. v. 3 co, con-, et Irische Grammatik. 1579, §§ 97. 99 & 245 246 247 >). In Hngua Oscorum com (dom) solum exstat. Cf. «ow fretutus in tabulae Bante vs. 15 (Zvetaieff, Inscriptio- nes Italiae Inferioris ‘Dialecticae, 1556. p. co et tab III), com atrud (?) vss. 23-4, Aombened, kombennicis, (koe serstizncs, comferascuster (ib. pp. 120-2). In tab. Bant. vs. 16 com (preivatud) errore videtur pro ccm scriptum esse. Apud Sabellicos quoque 4um praepositio erat: cf. Avm adses (ib. p. 3). Inscriptiones Faliscae cuncaftum et comrras recdunt (ib p. 26). Corchris autem est ex Vol- scorum dialecto (ib. p. 20); atque Umbri, cum illud com (éws) praepositionem haberent (frequens post nomen suum ponitur cum a littera saepius omittitur, ut rerisco, esuku), in Compositis tamen co (4) tantum usurpabant; cf. comoltu, kumalts =‘ commolito,’ comehetz = ‘commota,’ conezgas, kunikas =‘conixus,' covertu, kuvertu = ‘convertito,’ £wuretiu = ‘convehito.” Vide Biicheler. Umbrica, 1883, pp. 210-11.
1 Recorderis quaeso quo modo sz littera ante / vel s, aut g ante ss vel # posita productam reddiderit antecedentem vocalem.
2 Cf. Corssen. Ausspr. IT.* pp. 466-8, qui aliorum alias rationes exposuit.
®§ Alia est Brixi sententia ad Capt. 918 elata, sed ibi recc/usit potuit scribi vel certe dici. Aliud remedium petiit Fleckeisen.
lactendt Verbi Composita. 107
/ consonanti hanc fuisse proprietatem cum non prorsus certum sit,’ veri tamen simillimum videtur. Cf. Seelmann. Ausspr. d. Lat. p. 231: “‘Gewisse momente lassen es nicht zweifelhaft erscheinen, dass bis zum IV oder V jahrh. n. Chr. die bessere volkssprache mitlautendes / und V als halbvocale, also=# und w# fortfiihrte,” item p. 230: ‘Die laute . . . wo also trotz des j-artigen beigerdusches eins... deutlich durchtont, nennen wir mit fug und recht ‘halb- vocale.’” At sane non semper. Nam neque cum ab initio verbi posita est, eo magis ultimam vocalem antecedentis verbi produxit (cf. Catul. 62. 3: Surger€é iam tempus... ), neque tum cum in medio vocabuli # vocalem excepit, subsequente quoque vocali in eadem syllaba, illam # antecedentem longam reddidit. zzugus enim, tritugus, quadriiugus et reliqua similia constanter brevibus primis syllabis utuntur. Sed si quam aliam vocalem atque # subsecuta est, tum tenuem vocalis #sonum ante se dedit, sed arte secum coniunctam, nec cum antecedente vocali in alia syllaba elatam ;? nihilque referre utrum in simplicibus verbis an in compositis stet,® re syllaba con- stanter longa satis demonstrat.‘ Atque sic factus est ille plenior,® pressior,’ latior,® pinguis® sonus # litterae, quo antecedens vocalis, si brevis fuit, evaderet longa, si vero producta, productior.
Hunc sonum ex vocali et consonante mixtum ut scribendo signifi- carent, duabus # litteris opus esse putaverunt non nulli. Inde sunt illa ato, Maiia, Aiiax, quae Ciceroni placuisse commemorant et Quintilianus, ubi dicit (1. 4. 11), “‘Sciat etiam Ciceroni placuisse aito Mattamque geminata / scribere; quod si est, etiam iungetur ut consonans’”’; et Velius Longus his verbis (K. VII. 54. 16): ‘Et in plerisque Cicero videtur auditu emensus scriptionem, qui et Aiiacem et Maiiam per duo #: scribenda existimavit; quidam unum esse animadvertunt, si quidem potest et per unum # enuntiari, ut scriptum
1 Cf. L. Miiller. de Re Metr. p. 292% (251 *).
2 Cf. Corssen. Ausspr. I.? p. 302.
§ Cf. Schmitz. Beitr. p. 77; dissentiunt Prisc. K. II. 14. 14. inf. p. 109 citatus et Corssen I. c.
4 Cf. etiam Mar. Victorin. K. VI. 35. 25: “quamquam et geminata ¢ scribere iubeamur, ut prosiecit” (in temp. praeterito).
6 Cf. grammaticorum locos infra prolatos.
* Cf. Pompeium K. V. 103. 33 sqq.
108 Mauricius W. Mather.
est. Unde illud quod pressius et plenius sonet per duo # scribi oportere existimat.” Eiusdem modi sunt etiam illa Lucretiana Trottugenas et Graitugenarum supra p. 94 adn. 4 laudata, atque Trowtanis in 1. 476, quae potius Ciceronis esse quam Lucreti equidem arbitrer. Illam enim geminationem usurpatam esse Lucreti aetate non probabile videtur, quoniam in CI. I. nullum est exemplum.’ In CI. I] tamen haec inter alia exstant aetatis incertae exempla: 1076 POMPEIIVS; 1923 EIIVS; 4587 CVIIVS; etiam haec, quae accuratius sonum prae se ferunt; EIIVS 1964. I. 41 et saepe: CVIIVS ib II. 10; et alia. Qua ex ratione scribendi prave factum est illud POMPEIIVS CI. IX. 3748 et alia quae sunt similia.? Plauti quoque Ambrosianus rara exempla habet, quorum videas exempli gratia MAIIORES Trin. 642.
Grammatici autem saepius poscunt ut haec # littera consonans inter vocalis posita per geminationem scribatur. Cf. Caesellium apud Cassiodorium (K. VII. 206. 6): “Pompetius, Tarpetius, et eiius per duo ¢ scribenda sunt et propter sonum (plenius enim sonant) et propter metra; numquam enim longa fiet syllaba, nisi per ¢ geminum scribatur.”’ Cf. etiam Diomedem (K. I. 428. 10): (positione longa fit syllaba) ‘“‘cum correpta vocalis desinat et interposita ¢ excipiatur a vocali . . . quoniam inter duas vocales duarum syllabarum posita # geminatur. Sic enim scribi per geminatam litteram metri ratione desiderat, si quidem potestatem tuetur duplicis consonantis.’’ Con- feras porro, si velis, Marium Victorinum in K. VI. 24. 21; 27.9; 35. 22: Maximum Victorinum K. VI. 197. 16.
Sed alii, duplicis soni haud ignari, unam scribi litteram voluerunt; veluti Probus (K. IV. 221. 8): “J littera duplicem sonum designat, una quamvis figura sit, si undique fuerit cincta vocalibus ”; itemque (257.17): “JZ littera cum fuerit in medio vocalium, ita ut consonans sit, duplicem sonum reddit”*; et Velius Longus (K. VII. 55. 2): “At qui Zroiam et Maiam per unum # scribunt, negant onerandam pluribus litteris scriptionem, cum sonus ipse sufficiat; hanc enim naturam esse quarundam litterarum, ut morentur et enuntiatione
1 EIVS tamen est satis antiquum ; cf. p. 93 adn. 3. 3 Cf. Schmitz. Beitr. p. 84 sqq. et Seelmann. Ausspr. d. Lat. p. 236. 3 At cf. Serv. Aen. I. r supra p. 94 laudatum.
lactends Verbi Compostta. 109
sonum detineant. . . . atque ipsa natura ¢ litterae est ut interiecta vocalibus latius enuntietur”’; etiam Donatus in K. IV. 368. 27 et Beda in K. VII. 229. 30.
Quamvis male’ opinati sint hanc # esse duplicem consonantem, atque ideo positione produci multas vocalis, quae re vera natura aut compensatione producuntur,’? tamen praeclare testificantur ¢ litterae vocalibus interpositae fuisse sonum duplicem.
Summam quandam omnium harum doctrinarum dat Priscianus, ubi duplicem et simplicem ¢ litteram distinguit (K. II. 13. 27): “Eté quidem modo pro simplici modo pro duplici accipitur consonante: pro simplici, quando ab eo incipit syllaba in principio dictionis posita, subsequente vocali in eadem syllaba, ut Juno, Juppiter; pro duplici autem, quando in medio dictionis ab eo incipit syllaba post vocalem ante se positam, subsequente quoque vocali in eadem syllaba, ut Afaius, peius, etus, in quo loco antiqui solebant geminare eandem # litteram et matius, pettus, eiius scribere, quod non aliter pronuntiari posset, quam si cum superiore syllaba prior 4, cum sequente altera proferretur, ut fei-ius, ¢t-ius, mat-ius”; addatur quod est in K. II. 14. 14: “Pro simplici quoque in media dictione invenitur, sed in compositis, ut inturia, adiungo, ectectus, retice® CVergilius in bucolico (3. 96) proceleusmaticum posuit pro dactylo: Tityre, pascentes a flumine rezice* capellas.”
At male non modo hoc tale discrimen fecit, verum etiam separavit® illa duo # in duas syllabas. Etenim in compositis non semper sim- plex est # littera, sed tum modo cum consonantem aut ¢ vocalem excipit, ut in adiungo, bitugus et similibus verbis ; neque separandus in duas syllabas est ille mixtus sive duplex sonus, namque aliter unum # omnino non suffecisset; atque Priscianus cum illa dicit, conatur tantum scripturam interpretari, quae non iam in usu fuit, cuiusque sonum ipse non intellexit. Cf. K. III. 467. 15: (in Zrofa
1 Cf. Schmitz. Beitr. pp. 73-4.
2 Cf. Aufrecht. Zeitschr. f. vergl. Sprachf. I. 225 sqq. et Corssen. Ausspr. I* 393 $99-
8 Sic optimi libri, resce H, ex reicere, Hertz.
$ Cf. Brambach. Lat. Orthog. p. 199 adn.
§ Ad eum tamen accedunt viri docti Schmitz (Beitr. p. 75) et Seelmann (Ausspr. d. Lat. pp. 235-6).
110 Mauricius W. Mather.
vocabulo) “solent . . . Latini . . . pro consonante duplici accipere s et eam a priore subtrahere syllaba et adiungere sequenti; quam- vis antiqui solebant duas 4% scribere et alteram priori subiungere, alteram praeponere sequenti, ut Zroia, Matta, Aitax.” Ibi quoque vehementer errat’ ubi dicit (K. II. 303.5): ‘ Idque in vetustissimis invenies scripturis quotiens inter duas vocales ponitur, ut cisus, Pom- peitus, Vultetius, Gaiius; quod etiam omnes, qui de littera curiosius scripserunt, affirmant.” Nusquam enim exstat id duplex ¢ in CI. I, hoc est in titulis aetate superioribus a. 44 a. Chr. n., neque est in libris frequens traditum. Qui igitur maxime id sustentarunt, erant grammatici, sed ne eorum quidem omnes, ut patet ex testimoniis quae supra protull.
Hac digressione iam finita, qua apparet # consonantem vocalibus interpositam effecisse suapte natura, ut, quae vocalis esset ante se posita, # vocali sola excepta, produceretur, nunc revertendum ad propositum nostrum.
Non modo ¢ consonantem saciendi verbi diu retentam esse post vocali finitas praepositiones satis significat constans harum omnium productio, verum etiam #ecio primo dictum et scriptum, cum verbum et praepositio in unum convenissent? plura indicia sunt quam in illis compositis quorum prior pars consonante cadit.
CI. IX. 782: proiecitad in lege antiqua?®
In Plautinis libris etsi nullum est certum exemplun, videtur tamen cieciam olim esse scriptum in Truc. 659 *: —
Hoc {ctu exponam atque é6mnis escidm foras
ubi pro ¢ic#ad D codicis habent B et C sectam (secia@ B). Lucilius ezecere® scripsit, si Noni (300. 25 M.) codicibus credendum est, in versu dactylico (XXVIIII. 106. M.) : —
Ezecere istum abs te quam primum et perdere amorem
1 Cf. Schmitz. Beitr. p. 83; Ribbeck. Proleg. Verg. p. 138.
2 Quo modo prius scripserint, quam facta sit verbi et praepositionis inter se coniunctio, exemplo est illud e#do sacito, quod erat in XII tabulis, teste Festo p. 313 M.= 452.7 Thewr. Cf. R. Schoell. Legis duodecim tabularum reliquiae, 1866, p. 115.
® Cf. ind. Inscr. a. 4 Cf. ind. Plaut. f. § Cf. ind. Lucil. b.
Taciendi Verbi Composita. III
hoc enim scriptum legitur in Lugdunensi (L) et ex correctione in Harleiano (H), qui uterque est LX saeculi, atque huc spectat fortasse ecicere scriptura quam ante correctionem habuit Bambergensis (B) saec. IX-X, unde profectum est id cccere codicis Guelferbytani (G) saec. X-XI. icere habent H:, Gen. (saec. X), B,, G:.
Varronem? quoque hanc scribendi rationem usurpasse eiusdem Noni codices testificantur. In versu enim, qui in 452. 9 M. citatus
est, Frigore torret vénatum esecft ieiunio véllicum
cum ect habeant LH.G, ezecz? est in BH:.
Lucreti? codices Leidenses (A saec. IX, B saec. x) habent eecs?, ceteri ccf, in 2. 951: —
Dispersamque foras per caulas esecz¢ omnis sed in 3. §13 :— Addere enim partis aut ordine frazecere aecumst
nulla varia esse scriptura videtur. Catullus® autem unum incertum exemplum praebet ubi fproiceet verbum in 64. 370: —
Proiciet truncum submisso poplite corpus
legitur; nam Proiecit in Ambrosiano (A) saec. XIV scriptum erat ante correctionem, et in Berolinensi (L) saec. XV.
Ex Vergili* codicibus habent sex hanc vetustiorem scripturam: Romanus (R) saec. IV-V, derecet in G. 1. 333 :—
Deiecit, ingeminant austri et densissimus imber A. 8. 428: —
Deiecit in terras, pars imperfecta manebat ib. 10. 753:—
Deitecit: at Thronium Salius, Saliumque Nealces.
Mediceus (M) saec. IV-V dezeceé in G. 1. 333.
1 Cf. ind. p. 136 Varro. 8 Cf. ind. Catul. b. 3 Cf. ind. Lucr. b. 4 Cf. ind. Verg. b.
112 Mauricius W. Mather.
Bernensis 184 (c) saec. IX deiecst in A. 10. 753, 11. 642 :— Deiectt Herminium nudo cui vertice fulva prowcit ib. 5. 776: — Proiecit in fluctus ac vina liquentia fundit traiecit ib. 9. 634: — Traisecit. I, verbis virtutem inlude superbis.
Codex Gudianus (y) saec. IX secunda manu, et Minoraugiensis (m) saec. XII, et ex correctione Bernensis 165 (b) saec. IX ¢rasecit ib. 9. 634. Videtur etiam Priscianus, vel potius fortasse posterior eius operis librarius, detect descripsisse in versu A. 10. 753 citando ; cf. K. III. 293. 16.
Ovidi codex Laurentianus 36. 12 (A) saec. XI-XII habet in margine deiecit in M. 11. 386, quamquam in versu est scriptum déssidit, atque in plerisque libris est dsssiczt aut disicit vel plane scriptum vel sat
indicatum : — Dissicit) (disicit) hos ipsos, colloque infusa mariti.
Stati? unus Bambergensis (B) saec. XI exhibet hoc genus formam, eiecitt quidem in Th. 6.770, ubi non nulli habent est aut eiicit, sed Puteanus (P) saec. X, Gudianus 52 (G) saec. XIV, Helmsta- diensis (H) saec. XV tradunt id quod sententia postulat, resce¢*: —
Et patria vigil arte Lacon hos vescét ictus.
Itaque illud ezecz¢t codicis B ex retecit decurtatum esse videtur. Liviani codices haec suppeditant: esecé# in 1. 41. 1: —
Tanaquil inter tumultum claudi regiam iubet, arbitros eicst et deecit in 1. 48. 3: — Tarquinius . . . medium arripit Servium elatumque ... per gradus deicit quae quidem aperte praesentis sunt temporis,* et deiecit in 1. 40. 7: — alter elatam securim in caput desecit, relictoque in vulnere telo ambo se
foras eiciunt
1 Cf. ind. Ov. a. 2 Cf. p. 114 inf. ® Cf. ind. Stat. c. 4 Cf. Madvig. Emend. Liv. edit. sec., 1877, p. 226; Miiller-Weissenb. edit. oct., 1885, app. crit.
lactendt Verbt Composita. 113
quod, cum praesentis esse putaverit Hertz,' potest esse praeteriti. Etiam in 22. 37. 9:—
ut praetor . . . classem in Africam fratceret
Puteanus (P) saec. V-VI habet frazeceret.
Haec praeterea dat Lachmann in commentario Lucretiano p. 128: e Caelio Antipatro Nonius 89. 6 desecit (‘“congenuclat percussus deiccet dominum”’), In Caesaris de bello Gallico comm. 4. 28 exemplaria duo (Parisinum 5763 (B) saec. X et Vossianum primum sive Lugdunensem 53 (C) saec. VI laudat Nipperdey, Holder autem B solum) desecerentur. In apocalypsi Iohannis 11. 2 unus codex “unum e meis”’) eece (cf. eius editionem Novi Testamenti. ice autem habet codex Fuldensis.’ v. edit. Ranke. p. 446.12).
Ex Schuchardti libro*® II, 4 haec adsumpsi: Digesta Florentina * IX. 2. 31 dewceret, IX. 3.1. § 4 et XLIII. 16. 3. § 12 desecerctur; IX. 2. 53 descerentur; VII. 1. 13. § 4 et XLIII. 16. 3. § 9 deiecere (infin.); VII. 1. 13. § 5 deiecerit (pro infin.). Codex Sessorianus Augustini Speculi saec. VIII-IX (cf. Schuch. III. 3) prosectetur.® Codex Bobiensis-Vaticanus Augustini Sermonum saec. VI-VII (cf. Schuch. l.c.) proiecere® Gothanus* Euangeliorum 38. b, 15 froiece.
Post vocali igitur, item ut post consonante, cadentes praepo- Sitiones iecio formam esse antiquiorem licet nobis arbitrari. Sed scto in his quoque compositis se mox ostendit in sermone, ut ex scaenicis poetis® apparet, non plerumque tamen ita elatum ut eitcto verbum, quod exempli gratia profero, in quattuor syllabas ééct-o distingueretur,® sed in tres ¢-ci-o quarum prima esset diphthongus.”
1 Cf. eius edit. ad h. 1. et ad 1. 41. 1.
2 Cf. p. 92 sup. 8 Cf. p. 98 sup.
* Cf. p. 98 sup.
5 Cf. A. Mai Novae Patrum Bibliothecae tom. I, 1852, partem sec. p. 28, vs. 27.
® Cf. ib. partem primam, p. 265. vs. 11.
7 Cf. p. 98 sup. 8 Cf. etiam p. 99 sqq.
® Contra éicio formam cf. sup. pp. 104-05 ; atque écso0 quamquam initio quidem veri est simillimum per quattuor syllabas auditum esse, non diu tamen haec ratio dicendi perstitisse mihi videtur. Operae est pretium fortasse illud deinde adver- bium recordari, in quo praefixa syllaba cum # littera sequente diphthongum effecit. Similiter saepius dehinc, deinceps, proinde, alia.
1) Cf. ind. Plaut.e; Ter. d; Rell. e.
114 Mauricius W. Mather.
Cf. Mil, 205 :— Déxterum ita veménter 2ic##, qudd agat aegre stippetit
Quin tu abs te socérdiam omnem 7éi’cis, segnitiem 4moves
Phorm. 717: — Nam si Altera illaec mdgis instabit, férsitan nos ré?’ciat ib. 18 : — Ille 4d famem hunc a sttidio studuit ré2’cere.
Sunt quoque alia! exempla quae per diphthongum sive synizesim, quam vocant, potuerunt dici. Haud tamen placet talis ratio; quae enim exempla certo sunt sic enuntianda apud hos scaenicos poetas non plus quinque inveniuntur, quae divisim necesse est enuntiari, quindecim.
Haec autem per diphthongum ratio dicendi, quoniam brevior commodiorque erat illa altera per secio, magis magisque valuit, dum a. 105 a. Chr. n. pervagati fuit usus. Namque cum sono apud Romanos obsecuta est scriptura, tum eo anno lex est incisa, quae habet froicito? scriptum. Apud poetas tamen et alios litteris instructos illud secio tam diu est identidem et dictum et scriptum, quoad remansit post consonante terminatas praepositiones, hoc est usque ad extremam Augusti aetatem.® Inde ab hoc tempore omnino non usurpatum est neque apud poetas neque apud alios. Sic enim non solum inter se constat ratio omnium sacendi verbi compositorum, et quorum consonantibus et quorum vocalibus cadunt praepositiones, verum huc quoque se vertunt vestigia earum quae zecio habent formarum.‘ Nam exstant haec aut in operibus quae sunt ante Augusti obitum composita, aut quae longe posteriora sunt; neque est ullum, quod sciam, exemplum apud scriptores priorum duorum p. Chr. n. saeculorum, uno tantum excepto apud Statium (cf. p. 112 sup.). Id ergo Statium ipsum scripsisse mihi quidem haud probabile
1 Cf. ind. Plaut. f; Ter. e; Rell. f.
2 Cf. ind. Inscr. a. 3. Nempe potuit hoc verbum per diaeresim enuntiari, sed cf. p. 113 et adn. 9. |
® Cf. sup. pp. 99, TOT.
4 Cf. sup. pp. 110-13.
laciend: Verbi Composita. 115
videtur ; atqui concedendum est profecto potuisse eum, quod in aliquo veterum scriptorum exemplari conspexisset, id imitari. Quae autem exempla in Novi Testamenti codicibus et apud Augustinum reperiuntur, ea, ut reor, indicant illius zecéo formas inter rudes rusti- cosque diu conservatas esse.!
Tametsi igitur, cum seco iam a Tiberio imperatore obsoluisset, in sermone nihil nisi breviorem formam per diphthongum expressam in usu fuisse maxime probabile videtur, in carmina tamen hanc quidem formam, at per diaeresim elatam, admiserunt poetae. Cf, German.’ 512: —
Rectus per medios decurrens fraicit axis
Phaedr.*® Append. 8. 20: —
Et véce molli; Ifcet? enim vero 2%cé et alia.‘
Sed quamquam diaeresim numerorum gratia, fortasse etiam vete- rum consuetudine adducti, plerumque adsumpserunt, volgarem tamen dicendi modum neque hi neque illi spernebant. Quae apud Plautum et Terentium sunt exempla supra laudavi. Quibus accedant haec: Lucil. 652 Lachm.: —
. ego animam Proi'ciam ut me amore expediam .. .
ubi incertissima est scriptura.’ Quod dedi est ex Lachmanni coniectura. Muinime tamen, etiam si ea accipietur, necesse est per diphthongum, frdiciam, enuntiare, alteram enim mensuram, /roiciam, facile accipit versus. Quod quidem maluit Lachmann ipse.® Lucretius’ duo exempla habet, 3. 877 : —
Nec radicitus e vita se tollit et ctcst Eicit enim sulcum recta regione viaque
1 Cf. p. 102 sup.
2 Cf. ind. p. 145 German. b.
8 Cf. ind. Phaedr. b.
4 Cf. ind. Manil. b; Val. b; II. Lat. b; Stat. c; Claud. c. § Cf. ind. Lucil. b.
6 Cf. Comm. in Lucr. p. 188.
7 Cf. ind. Lucr. c.
cr4 Huarcctus 4. Wither.
Caber* cm. Ta mo Rh —
Num oe” ne <recs” Jug? Junt xi scaenam 4cfero?
Jenzre je sxxa ves rut Tadere Cuimo. Ov.* F. g rsa — Pactm I0UL, Nuniment nanent, tam “222re* captam. Cirs.* ccd: — &-xere* of Tutmts virtue remmiere Wartem. Sen? Phoen. p29: —
Ro ter Boes or7t cee SIptim dias quamuam fir guogue oricix: per se wo svilzba poturt esse. Wal FL >. srg —
Nott 42200" Wuihis payer quale et Tora Uicts. Sac’ Th. 4 sy4:—
Teik main 92eiogue cames im Tunus Ios
Jaw autem Romances non idmisnse Smiik sme irgumenta eorum, quae suprz pp. Sy-g2 protux, aamcue deest if ot im ttulis ita in libris antiguissimis et solute cratumis ef vinctie. excepus modo PRe Vergilianis: neque stepe in poetirum certe codicibas ante XV saéciicm mnducebatur. Talem praeterea sonum. goo ¢ consonans s Vocali exceptum esset in eidem sylliba. abborrere 1 Romanorum consuetudine, sams, arbttror. ipparet ex testimunis supra p. 93 sqq. adlatis.
1Cf med Rell e * Cf ine. Cis. 2 CE£ ind. Verg. c * CE md. Sen d § Cf. ind. Hor. c *CE md Valco
$ Cf ined Ov. d * Cf. ind Sar d
laciendi Verbi Composita. 117
In poetarum libris manu scriptis quae huius formae tradita sunt exempla, quod comperire ex editionibus criticis potuerim,’ haec fere sunt: — .
Plautinus? F saec. XV (cf. p. 89 sup.) praebet resco in Merc. 908, deticiam in Stich. 355, resicsam in Pers. 320, reticere ib. 319.
Lucreti*® codices, etsi non exhibent ipsum /iczo, tamen in 3. 58 et 3- 497 eliciuntur, in 4. 945 eliciatur habent, unde Lambinus ecendi formas restituit, id quod Lachmann et Munro acceperunt. Nimirum recte. Horati enim codex Monacensis 14685 (C) saec. XI habet in S. 1. 6. 32 inliciat, et codex Gothanus (g) saec. XV dlcat* In Vergili quoque Gudiano (y) saec. IX super/icit® prima manu scriptum est in A. 11.625. Atque Livi Puteanus (P) saec. V-VI znx/icé dat in 30. 10. 16. Quae corrupta non tam, ut illud Graliugenarum® quod est in Lucreti codice A, oculo decepto debentur, quam auri male percipienti; non enim alia huius zscéo habent hi codices vestigia.'
Catullianus* Guelferbytanus (G) saec. XV geminum / constanter habet.
Vergilianorum® codicum habent trés huius formae exempla: Palatinus (P) saec. IV-V cotciunt in A. 10. 801, et de: t+ cit ib. 8. 428 in quo forsitan illius descit lateat vestigium. Romanus (R) saec. IV-V Prozice ib. 6. 835. Bernensis 184 (c) saec. IX ¢raz: cit ib. 10. 400 ubi # altera est erasa.
Ovidiana™ Defloratio Vincenti Bellovacensis saec. XIII dat eitcitur in T. 5. 6. 13. Codex m imperite interpolatus (cf. Merkel. edit., 1841, p. cclxxxi) ¢raticias in rasura habet in F. 4. 782. Codex Q (cf. p. go sup.) saec. XV habet ezyctur in T. 5. 6. 13, trasycerer ib. 5. 2. 33 ubi ¢raiicerer habent et Excerpta Politiani (A) et codex Oxoni- ensis (£), uterque saec. XV, et Berolinensis (0) saec. XIV.
Phaedri* Neapolitanus (N) saec. XV (?) et Vaticanus 368 (V) saec. XV-XVI dant escé in Append. 8. 20.
1 Si qui alii codices hanc formam habent, scilicet novissimorum sunt.
2 Cf. ind. Plaut. d, f. 8 Cf. ind. Lucr. b. 4 Cf. ind. Hor. a.
5 Cf. ind. Verg. a. 6 Cf. sup. p. 94, adn. 4.
7 Cum his placeat comparare Ter. Hec. 842, ubi pro conicias, quod est in codicibus, Palmer et alii com#/sctus voluerunt scribi. Cf. ind. Ter. b.
8 Cf. ind. Catul. b. 9 Cf. ind. Verg. a, b.
10 Cf. Vel. Longi locos inf. p. 118 citatos.
11 Cf. ind. Ov. c. 12 Cf. ind. Phaedr. b.
118 Mauricius VW. Mather.
Stati codex Dommerichianus (D), qui in libris pessimis interpolatis habendus est, praebet detsciat' in Achil. 1.311. Codices Behottiani, incerta aetate, caf? in Th. 6. 770.
Varronis librorum de lingua Latina codex Hauniensis saec. XV in versu Atti (430. I. p. 191 R.) 7. 65 prolato vrezces* tradit, rezices Gothanus saec. XV, retuis reliqui (excepto Florentino saec. XI, qui vewis habet) saeculo XIV infenores.
Gelli codicum interpolatorum pars in versu Pacuviano (94. I. p. 88 R.) 4. 17. 15 sumpto habent prviscit.*
Horum codicum nullus praeter Vergilianos PRc superioribus saeculo XIII temporibus descriptus est. Post consonante autem praepositiones finitas nullum ex codicibus quorum compen scripturas, praeter Vergili PV. qui uterque ad IV vel V saeculum pertinerent, Moreti B saeculi IX, Plautique C saeculi XI, saz scriptum habuisse dum ad XII saeculum perventum esset, supra p. 92 indicatum est. Quoniam ergo in Vergilianis solis vetustissimorum codicum hoc a0 invenitur, satis patet, opinor, id non a volgari usu sive loquendi sive scribendi esse ortum, sed ab alicuius more grammatici vel libran, atque a quo magis ortum putemus quam a Probo® illo Aristarcho Vergiliano?
Namque grammatici post vocalis, ita ut post consonantes (cf. pp. 88-—Sg sup.), quod sav verbi ¢ consonantem nullo modo amittendam esse crediderunt, per duo s haec composita scribi iubebant; atque Velius Longus quidem eo more adductus, quo consonans # vocalibus interposita per duo ¢ scripta est, tria ¢ memorat (K. VII. 54. 20): “Inde (he. a scribendo fzuvm, Matizm, Trotiam) crescit ista geminatio et incipit per tria # scribi qvzime.” Etiam (ib. 72. 4): ‘“Troia per # unum an per duo scribere debeam; et qe2¢ utrum per unum s an per duo an per tria . . . et sic fiat estict et witiait.”" Duo : laudat Priscianus in loco (K. II. 14. 14) supra p. 109 citato, iterumque in K. II. 126.13: “Solet plerumque in compositione a in 2 converti, ut ‘cado incido,” ‘facio inficio,” *iacio iniicio remae.’”
1 Cf. ind. Start. a, s. v. dessectat. 7 Cf. ind. Stat. c. s. v. recat. ® Cf. ind. Rell. f. * Cf. ind. Rell. d 5 Cf. p. o4 sup * Cf. pp. ro7-o$ sup.
* Cf. Schmitz. Beitr. pp. 73-79; Lachmann. in Lucr. p. 37t.
lacitendt Verbi Compostta. 119
Gellius in 4. 17. 15 testibus optimis codicibus proia?t scripsit, quam- quam post consonantes #ico scribendum modo dixerat,' nihilque erat causae cur post vocalis quoque idem non iuberet. Servius autem unum z plane demonstrat se, ut enuntiavisse, ita etiam scripsisse ; cf. ad Aen. 4. 549: ‘ Obicio, reiao, adicio ‘i’ habent vocalem sequentem, quae per declinationem potest in consonantis formam transire, ut obiec, reiect.”” |Accedat quod dixit ad Aen. 10. 473: “Fe naturaliter brevis est, et eam pro longa posuit. Sic alibi (G. 3. 389) :—
Aetce ne maculis infuscet vellera pullis.
Quod licet possit excusari, quia cum facit resect, inter duas vocales 7 posita producit superiorem, ut dicamus longam eam esse spe qua per declinationem longa futura est; tamen quia in hac re argumentum magis est quam ratio, dicamus ectasin factam, quae poetis plerumque conceditur.”
f/aao verbum igitur, cum praepositionibus quae vocalibus cadunt inciperet subiungi, in zecio se convertisse videtur. Plauti autem aetate in sermonibus certe haec ¢ littera propria mutatione porro convertebatur in # vocalem, id quod simul fecit ut # consonans reiceretur, nam # consonantis sonum # vocali subsequente non patiebantur Romani. Itaque praepositione cum illa # vocali in una syllaba plerumque elata, diphthongus est effecta, quae magis atque magis valens ad extremum alterum saeculum ante Christum natum ita volgo audiebatur, ut, scriptura apud Romanos sono obsequente, etiam in legibus, quae veteres formas maxime amant, repraesentaretur scribendo. Inde in volgari communique usu cum haec forma versa- retur, apud poetas tamen et alios politos litterarumque peritos illud seco usque ad extremam Augusti aetatem restabat. Tuiberio autem imperatore, non iam est hoc usurpatum, nisi, ut videtur, ab imperitis provincialibusve ; verum ‘cio cum in sermone omnibus iam inde fuit in usu per diphthongum enuntiatum, tum numeri gratia per diaeresim a poetis non numquam adhibitum est. .
1 Cf. 4. 17. 6 sqq. sup. pp. 88-89 prolatum.
120 Sauricius WV. Mather.
IV. De A LITTERA AD HIATCUM CORRIGENDUM ADHIBITA.
Ile tamquam hiatus, qui concursu ultimae vocalis praepositionum cum illo zw formatus est, raro in codicibus per 4 litteram insertam vitatur. Hoc & quamvis Lachmann? et L. Miller? putasse videantur aequale fuisse eius sw, perpaucorum mori librariorum, ut opinor, debetur. Nam neque in ttulis neque in illis solutae orationis codici- bus, quos supra pp. 91-92 laudavi, omnino occurmit: neque scilicet multo saepius in libris poetarum. Quae notavi exempla in quibus- dam editionum apparatibus criticis memorata haec sunt: —
In Vergili Mediceo (M) saec. IV-V fradsat* in A. 10. 400.
In Noni codice Lugdunensi priore ( L) saec. IX cwoAscere* p. 267 M., ubi est Afrani versus (216. II. p. 1g1 R-) prolatus. Cf p. 268 M. ubi in citando versu qui est (1gt. II. p. 188 R.):—
Cotcerat nescié Guid de ratiuncula
(sic L. Miller, e conere af codices plerique) wAwere dedit Li.
In Plauti codice Vetere (Bi sacc. XI uw Aw sam? in Stich. 349, unde &hine cum in CDF.
Catulli Oxoniensis (O) saec. XIV-XV reddit prvhices® in 55. 19, prohicere * in 64. 82.
Ovidiani libri aliquot, saeculo XIII vel XIV descripti, habent trahicerer® in T. 5. 2. 33. Quorum satis sit memorare Berolinensem (B) saec. XIII-AIV et bos NIIE saeculi, Guelferbytanum (sec. manu, G;), Holkhamicum (H). Parisinum $254 (R).
1 Cf. Comm. in Luacz. p. 128.
3 Cf. de Re Metr. p. 291° = 250°: “ Alii pro eo quod est :cze scripsere tecio vel hicso.”
3 Cf. ind. Verg. b.
4 Cf. ind. p. 135, Rell. f. 4 Cf. ind. Plaut. £
© Cf. ind. Catull. b. * Cf. ind. Ov.c.
Tactendi Verbi Compostta. 121
V. DE ILLIS VERBIS QUAE SUNT CoNIcio ET DISICIO.
a) Conicio.
Restat denique ut de verbis quae sunt conicio et disicio breviter agam. Nam ut de conicw prius verba faciam, non satis adhuc cognitum est, qua ratione con et co ab antiquis adhibita sint.’ Examinatis autem iis formis quae in inscriptionibus et apud poetas sunt, atque etiam orationis solutae non modo lis quas ego in vetus- tissimis libris manu scriptis inveni, verum etiam quas conlegerunt in libro suo Neue et Wagener, haec mihi videntur elucere: con prae- positionem plerumque, si quidem non constanter, adhibuisse poetas ; co autem formam, quae, ut veri simile est, pariter cum cox prae- ponebatur ei quod est seco verbo, tum, cum #eciv in icio converteretur, mox solam in sermone usurpari, atque, diphthongo cum / vocali plerumque effecta? magis magisque valuisse, ab eisque qui solutis
1 Cf. Lachmann. in Lucr. p. 136; Neue-Wagener. Formenl. II, pp. 864-65.
2 Et con (com) et co formas ante / et w litteras semivocalis praebent inscriptiones antiquissimae ; cf. illud senatus consultum de Bacchanalibus (CI. I. 196 (p. 43), a. 186 a. Chr. n.), in quo exstant haec: comtuovise v8. 13, coventiontd vs. 23 contoura(se] vs. 13; etiam CI. I. rort comiunxs, 1053 coniugt, 1064 et 1413 cotuge quae inscriptiones, etsi incerta aetate, sunt tamen inter vetustissimas. Vide alia exempla ap. Neue. II}. pp. 865, 867. Principio autem haud scio an illo co solo ante semivocalis usi sint Romani; comtio enim et cunctus (conctos in Carm. Arvali, CI. I. 28. 4) vocabula non ab conventio et contunctus possunt duci, verum ab coventio et coiunctus. Hoc si verum est, cotecio est primum adhibitum, deinde cum illo pariter comiecio (cf. ind. Inscr. a. 2, contectant, a. 123-2 a. Chr. n.); atque con praefixam syllabam plerumque scripserunt poetae, si quidem in codicibus possumus niti, cui non est mihi dubium quin secéends formas usque ad Tiberium imperatorem subiunxerint. Illud autem écio simul atque in usum venit, cum eo non saepe co, sed co, coniunctum est, quoniam sic poposcit consuetudo Romana. Namque ante vocalis co fere adhibitum est. Cf. cogo (ab coago), coactus, coo, coetus (ab coitus), coopto, alia. Excipienda sunt sola haec vocabula: comedo, comes, comitéum, cum iis quae ab his originem ducunt. Corripitur autem co syllaba (cf. p. 105 et adn. 5) cum eam excipiunt vocales; cf. Hor. S. 1. 2.9:—
Omnia conductis cdemens obsonia nummis. Verg. A. 3. 424: — At Scyllam caecis cdAsbet spelunca latebris.
122 Mauricius W. Mather.
verbis scriberent magna ex parte adhibitam esse.’ Cicero autem et Caesar et Livius, quantum quidem ex codicibus iudicare possumus, modo co modo con maluerunt. Potius tamen mihi quidem videntur illud coniecio semper adhibuisse.? Cf. adieit illud quod Cicero scripsit in libro de Divinatione 1. 48. 106 (cf. p. 97 Sup.); desece rentur quod est in Caesaris de bello Gallico commentariis 4. 28 (cf. p. 113 sup.) ; adsect? apud Livium ro. 8. 3, 10. 37. 14, 22. 19. 2; [s]sdiecere in 26. 19. 2 (cf. pp. 97-98 sup.). Cf. etiam p. 112 supra. Coniao autem aut qwiae ubi in codicibus eorum scmptum legitur, librariis debetur; sed illud praefixam syllabam, qua ipsi scriptores usi sunt, servavit, hoc volgarem formam admisit. Cf. Livi codicem rescriptum Veronensem,’ in quo cum coi in lacunam quadrat in 3- 13. 6, tum cots plane scriptum in 6. 2. 10. Codex Puteanus ‘ habet in 28. 3. 11 a prima manu cotercatur, quod secunda manus in conicerentur convertit, atque cotecus est pro coniectus in 25. 16. 22. Cf. porro Neue. l.c.
Sed apud poetas illud @ valde raro traditum est. In Plauti Rudente 769 coiciam*® habent CD codices, sed reliquis in locis huius poetae com solum legitur.£ Apud Terentium autem con est constans, atque in Hecyra 132 non est dubium quin id conieci, quod habent plurimi libri, praeferendum sit illi @éecs quod est in Bembino (A). Illud co quod est in Laberi et Afrani versibus' forsitan poetis ipsis non debeatur, sed Gellio® et Nonio,® qui hos versus sumunt. Namque con potest in omnibus restitui neque numeri laeduntur.
(Similiter ae syllaba. Cf. Ter. Haut. 825 :—
Ne ego hémo sum fortundtus ; dtemes té Syre. Verg. A. 1. 106: — Hi summo in fluctu pendent, his unda ddAsavns. Lucr. 2. 202: — Quin vacuum per inane worsen cuncta ferantur.)
Cum tamen eius éczo verbi formae huic co subiunguntur, non corripitur sed per diphthongum cum # sequente coniungitur. Cf. pp. 113 (cum adn. 9)-116.
? Cf. ind. Inscr. a. 6, 10; b. 3; Neue ID. pp. 864-65. Accedat etiam ex Plini codice rescripto Veronensi (cf. p. 91 sup.) 15, § 29 (= edit. Sillig. vol. VI. p. 240. 2) coictt forma. Serv. ad Aen. 9. 409 (411): “‘contctt antiquum est.”
2 Cf. p. 101 sup. ® Cf. p. ot sup. * Cf. pp. 91-92 sup.
§ Cf. ind. Plaut. b. * Cf. p. roo. 7 Cf. ind. Rell d, £
§ Cf. Gell. 16. 7. 5. * Cf. Non. p. 267. 34 et 36 M.
lactendt Verbi Compostta. 123
Item apud Lucilium haud scio an sit con! ponendum in 2. 25 et 29. 48 M. (=62 et 715 Lachm.), qui versus a Nonio 268. 5 et 506. 27 M. citati sunt. Con? dat Gellius 4. 17. 2 in Lucili versu qui est 11. 10 M. vel 342 Lachm.
Vergili libri con* formam exhibent constanter, nisi quod in A. ro. 646 et 5. 662 Romanus (R), in 10. 801 Palatinus (P) co habent. Ovidiani quoque libri con* semper reddunt, tametsi in M. 7. 245 ante correctionem co habuit codex Marcianus Florentinus 225 (M) saec. XI. Apud Ennium,® Lucretium,® Ciceronem,’ Tibullum ® nihil nisi con est traditum, neque licet nobis dubitare quin ipsi poetae contectendt formas scripserint. Sed post Augusti obitum cum illud tecio breviori ico formae cessisset, in con utendo, eoque longo, nihilo minus perstiterunt poetae.® Cuius rei sola duo sunt exempla, alterum apud Valerium Flaccum,” alterum apud Silium."
6) Disicio.
In verbo autem disiciendi de quo alias alii sententias dixerunt, eandem, qua sum in ceteris usus, rationem volo adhibere. Atque dissicio ne quisquam, oro, arbitretur aliud esse verbum ex illo secere'? compositum. Nam cum O. Ribbeck, vir doctissimus, hanc opinionem a. 1873 diserte protulerit in corollario comicorum fragmentorum p. xiii sqq., ac Prisciano sumendo 1oo2 P (K. III. 56. 18): “Sciendum, quod tunc dis praeponitur, quando sequitur ¢ vel / vel » vel s vel ¢ vel # loco consonantis, ut discumbo... differo... displicco... dissicio, dissero, distraho ... disiectus, disiungo,” ubi id dissicio enumerat ille cum dissero verbo, non cum disiectus, studuerit sententiam comprobare; minime, ut ipse quidem fatetur, sustentant codices hanc distinctionem, tum atsicere tum dissicere inconsulte ac temere exhibentes, neque id quod seco est in hoc solo composito sicio factum esse mihi facile persuadetur, eoque minus quod dsseco exstitit a dissecando. Itaque dissiciendi verbum, quod habent saepis-
a ce
1 Cf. ind. Lucil. b. 2 Cf. ind. Lucil. a.
3 Cf. ind. Verg. a. 4 Cf. ind. Ov. a. 6 Cf. ind. Enn. a. 6 Cf. ind. Lucr. a. 7 Cf. ind. Cic. b. ® Cf. ind. Tibul. a. * Cf. p. 102 sup. 1 Cf. ind. Val. a. 11 Cf. ind. Sil. a.
12 Secere quidem non minus se commendat quam /avere, sonere, tonere.
pa wezrrecis VW lather
SOE OK SEIT. Tom wind Sk om Tum ms det 2snacndi TESTE, IME OES SSEEE SMe moet = Goeryeas: et misscomes. F°rxt FITC Weevesse mtesnyea aime ili zeceres itttera- TOFES: 2M STOEL Som * assnerr. OsseEsmre © ~ vel * Sxsaz. separa, divide. “* sum: see. Gssetee. stage. dsc “*
“nue error ilo . ufesnm otum 2st suscemur= irc enim ararsus “mit “als: Tesmraira UM SIMs aredreie sviiabae Taaiis 5 won. mee. 2 <2 TNs af “rr iwer* Saca pauci TEISTeMnt, Wes sf Nuus du. Mud sum 22 iam post August: intum on ost ct fn quuemtu = om ao fems:. sed apud wees vereres amuse: slama Immitucttome wc emir Soran de mms xnlico yun cae aes. i wo Wei = rencen, pauzenup <= Ses smiious Tse Cwliu.* sims Rue ricmmtad fece-
TInt. vemum 5 ittere semimicurem winduerim. 12 yood facilius feccum est wi gsceco. Te Sorsse asi. noni Sssmili sono
essmir. Siu cn qwemans Iesccpemis cai ws rac scrberdi MRCS IPE SUNS Vers MUMPUSG “NT RAL iGyue (ok wour9 In usum Vemir uf vei Posmay, te Suraisse quicium pergemai urbitraretur id esse COMpUNCuM nls RMMics ver Jod 1 + Iftra meiperet.
Quam requens ec zemumite ¢ itterte Sauer. wdere pessumus Dr exempuis bows veri CumRUEFEMNS yume ex poets excerpsi.’ Samet edim im Xummm: umievwnG <xempuk cum. uf comimode inter se comparentur. 2ic Cumpunere 10et
=
ICE A Nat. Awe CL VWIIE pp rm
2Cf Hildebrand Glossarium [Lar Bidiiccrece: Fats. intiquisiimum saec. TX, p rz: Goece et Gancermann. Corpas Glows. Lat IW. p 352. 54.
8 Cf Goetz et Gundermann. [V.p 3319 Porte conferas Du Cange. Gloss. Med. et Inf Lat, sv.
* Cf. Wagner. ad Verg. A. t2. 308; Lachmann m Locr. p.1r25: L. Miller. de Re Metr. p. 2917 (250°).
§ C£. Gell. 4 17.
© Cf. Sen. Tro. 395, desszcat habet cod. Etruscus cf. p. 126 inf. In Apul. Met. VIIL. p. $31 dissicant et dissecant habent codd. Cf. Hildebrand. Gloss. Lat. p. 112 adn.
7 Cf. etiam Neue II.? 920-3.
laciendi Verbt Compostta.
1) Plaut. Curc. 424
2) Att. 348. I. p. 181 R. 3) Naev. 57. II. p. 16 R. 4) Caecil. 239. II. p. 74 R. 5) Lucr. 3. 639
6) Verg. A. 12. 308
7) Verg. A. 1. 70
8) Verg. A. 7. 339
9) Ov. M. 11. 386
10) Sen. Tro. 395 11) Sen. Agam. 896 12) Sen. Phoen. 343 13) Val. Fl. 3. 162 14) Sil. It. 9. 538.
15) Sil. It. 13. 444 16) Il. Lat. 325
17) Stat. Th. 1. 590
18) Stat. Achil. 1. 311
19) Stat. Th. 10. 69
aissicit. aissicit, aissicis. disstce.
aissicictur.
adissictt.
aissice.
aissice.
adissicit,
dissictt. dissicere. aissictte. aissicit. asice.
aissice. aisicere.
dissicit.
adissiciat.
Atssice.
125
dessicit BE, ailigit FEs. aiscicit Hi, dissicit Ha.
sic Putean., disce volgo.
beMa, distctt yM:, dissicit Pr, disicit P2, discidit R.
Rybcm, schol. Serv. 5. 683 ; dissicep M, disice Donati exempl.
RybyM2, dissicae c, disice Mz, distice, supra scripta for- tasse s, V. .
sic vel disicit codd. plerique ; dissidif X culus in marg. detecit.
dissicat E, dissipat A.
E, deicite interpolator E.
C, dtssicel P, disicet M1, disié- at M2.
LV, distre vel disice corr. in dissijce F, dissite OQ, discute volgo. |
Ch, discite O.
aisiiceret E, discideret FV, aivideret MN, ais™ceret L, aiscuteret B.
BG: M, disicit H Anglic., dis- sicet P, auscicit S, discdst Pal. 1.
PG, dissiceat Pc, atssotiat G, aeitctat D, discutiat ptH.
126 Mauricitus W. Mather.
Ex his igitur sunt novem (4, 6, 7, 8, 9, 13, 14, 16, 17) quae variis scripturis et dis et diss vel perspicue exhibent vel satis indicant: decem (1, 2, 3, 5, 10, II, 12, 15, 18, 19) quae, cum ass manifestum vel haud obscure indicatum habent, tum dais scripturae nullum dant vestigium ; unum denique (16) quod as solum habet, sed etiam in hoc forsitan ass in iis scripturis lateat, quae sunt in FVLB codicibus.
Quamquam hi ipsi poetae, ut arbitror, illam geminationem s litterae non adhibuerunt.’ Nam ante Augustum mortuum disiectends formas veri simile est esse scriptas, atque postea disiciendi, in quo sane dis syllaba corriperetur ; cuius rel exempla duo praebere Seneca? videtur in Agam. 896 : —
Distcere et hostem quaérit implicitis suum et in Phoen. 343 :— Distcite passim moénia in planim date quorum in utroque avs syllaba potest corripi. Sed in Tro. 395 :—
nubes... Arctoi Boreae d@issici¢t impetus
ubi versus Asclepiadei ratio postulat ut as syllaba producatur, non disicit scripsit poeta, verum dissecat vel dissicat,® si quidem codici Etrusco omnium praestantissimo credere possumus. Apud Valerium Flaccum autem et Silium et Statium, si eorum in saciend# compositis rationis meminerimus,* haud mira videbitur productio dis syllabae. Hanc vero productionem non duabus s litteris scribendis significavere, nam simplicis s exempla ipsa, quae sunt omnium optime codicibus sustentata, praebent Silius 9. 538 et Ilias Latina 325.
1 Cf. autem Kempf. Val. Max. edit. pr., 1854, p. 282, vs. 6 adn.; Fleckeisen. in Annall. Philoll. 1863, p. 199 adn.; Neue. II.? p. 21.
2 Cf. ind. Sen. a.
8 Cf. sup. p. 124, adn. 6.
§ Cf. sup. pp. 101-02.
Tactendi Verbi Compostta. 127
INDEX.
ENUMERATIO omnium praesentium praepositionibus subiunctarum sactenas verbi formarum, quae in Inscriptionibus et apud Poetas, quotquot inter annos 240 a. Chr. n. et 200 p. Chr. n. scripserunt, sunt repertae. Adduntur quae formae apud Ausonium et Claudia, num occurrunt.
I.
Ex Inscriptionibus sumpta exempla (CIL. — exceptis XI. 2, XIII- XV. 2 nondum editis ; — Ephem. Epigr. I.-VIII. 2).
@) Quorum tempora definiri possunt, secuntur : — (cf. p. 110) proiecitad (= proicito?) in lege Lucerina de luco tempore non certo sed anti- quo promulgata. CI. IX. 782; Eph. Epigr. II. p. 205. (cf. pp.96,100) conieciant in lege repet. a. 123 sive 122 a, Chr. n. CI. I. 198. 50. (cf. pp.100,114) proicito in lege parietis faciendi a. 105 a. Chr.n. CI. 1.577, 1. 12, 16 (p. 163) et X. 1.1781 (p. 218). inicere in lege Ursonensi a. 44 a. Chr. n. CI. II. Supplem. 5439. Ixi. 1; Eph. Epigr. III. p. gt. t. reiciantur CI. ib. xcv. 27; Eph. Epigr. IT. pp. 110, 224. (cf. p.121sq.) | coicerentur in titulo artificis de ossibus Hilarae sepeliendis a. 13 a. Chr.n. CI. VI. II, 9290. deicientes in Menologio Rustico Colotiano et Men. Rust. Vallensi, quae haud multo recentiora facta sunt quam Fasti nobis servati, qui omnes ex annis u.c. 723-804 (= 31 a. Chr. n.
128
(cf. pp. 121-22)
(cf. p. 88, adn.)
Mauricius W. Mather.
—§1 p. Chr. n.) orti sunt? CI. I. P. 359. XXII. a et b sub mense Decembr. vs. 15; VI. 1. 2305 et 2306 (pp. 637, 639).
adicere in titulo a Vespasiano ad milites in provinciam Baeticam a. 78 p. Chr. n. misso. CI. II. 1423. 11.
subici in decreto Domitiani ad Falerienses ex Piceno attinente a. 82 p. Chr. n. CI. IX. 5420. 8.
coicito in legibus civitatum duarum Hispana- rum temporibus Domitiani in aes incisis, CI. II. 1964, col. 2, 45 et 51.
adiciatur in titulo de exercitu aevo Hadriani insculpto. CI. VIII. 1. 2532, Frg. B, a), vs. 6.
Adicit in vss. heroicis a. 136 p. Chr. n. compo- sitis. CI. XIV. 2852. 15.
subiciantur in epistula a. 314 p. Chr. n., ut veri simile est, a Constantino ad Romae praefectum scripta. CI. V. 2781. 27; Eph. Epigr. VII. p. 416, tab. B, vs. 45.
adici in titulo sacro, qui in Ianiculo repertus, certe post Diocletiani tempora, pro- babiliter quinto saeculo iam ver- gente, incisus est. CI. VI. 1.1711. VS. 3.
adicit ib. vs. 17.
[reiciendos e Claudi orationis a. 48 p. Chr. n. habitae fragmento, quod in tabula aenea servatur. cf. ed. Monfalcon., 1851, t. IV. 8 (non vidi); Tacit. Nipp. edit. quart., 1880, II. p. 304, col. 11, 8.]
1 Cf. Ph. E. Huschke, Das Alte Rom. Jahr und seine Tage, p. 142.
Tactendi Verbi Compostta. 129
6) Quorum tempora non definiuntur, haec sunt : — adicias in praeceptis T. Flavi cuiusdam de ossibus et cineribus datis. CI. VI. 11. 8431. adiciatur CI. X. 1. 649.
(cf. pp. 121~22) coicito in fragmento Arimini effosso, quod Garrucci eius tabulae esse conicit, quae altera parte exhibet legem repetundarum. cf. contectant sup. CI. XI. 1. 364 a.
deiciat CI. X. 1. 1971.
inicere CI. XIV. 586.
traiciendas in titulo sepulcrali. CI. VI. 1. 10237, VS. 13. (In summa 21.)
¢) Haec exempla cum ad quaestionem nostram non pertineant tamen digna sunt quae laudentur: — (cf. p. 89) iniice in tabella marmorea, quae in compluri- bus libris titulos continentibus edita a Mommseno Falsis adsignata est. CI. X. 1. 204* (p. 10*).
OIECIOR in fragmento parvo ita mutilo, ut cum duo tantum vocabula sint certa, tum verbi nostri deperierit initium. Vix autem saciend: verbi potest esse compositum. Leblant supplet sic, [PRJO[TJEC[TJOR. CI. XII. 5385.
adiecit in fragmento pusillo VTVM:ADIECIT. Est sine dubio temporis praeteriti. CI. XII. 5309.
Il.
Exempla e Poetarum scriptis sumpta iam cum varilis scripturis disponamus, primumque quidem
130 Mauricius W. Mather.
PLAUTINA,
quorum expedit in primis ea proferre
a) in quibus syllaba praefixa consonante exit elusque Pproductio
metro confirmatur. cf. p. 100.
adiceret Poen. 1174.' (cf. p.122) cOnicit? Mil. 112.
cOnice Epid. 194. (cf.p.1258q.) dissicit Curc. 424. .
Tnicio Cas. 225. inicit Aul. 197. Inice Truc. 479. nice Pers. 88.
Gbicitur Merc. 339. Obicitur Pseud. 592.
A, adiecenit F cf. p. 96.
contigit B, contegit CD cf. p. 96, compegit F.
B, coniice F cf. p. 89, coée I.
dessicit BE, diligit FEs.
BI, initio E, initio FY.
inlice F (L cf. p. 89).
nice BCD, iniice F cf. p. 89.
obiicitur F oo“
aA, obiicitar F oo“ (In summa ro.)
6) Pauca sunt exempla quae correptam habent syllabam primam
consonante cadentem.
(cf. pp.99,122) cOnicitis Merc. 932. “s “ cé'niciam Rud. 769.
(cf. p. 99) Obicias Asin. 814.
¢) Plerumque anceps est primae mensura. Mbicio «Merc. 851. A’biciam Men. 555. A‘diciat Asin. 769. A’‘dicito Merc. 491. (cf. p.122) cdnicio Cure. 253. oa“ c6nicitur Poen. 69 eo cSniciam Cas. 342.
1 Ritschelianae editionis numeros sequor. 2Cf. omnibus Trin. 54; enécas Rud. 944 Pp. 274-
coniicitis F cf. p. 83. adn. 2.
B,coniiciamF “ “« coiciam CD cf. p. 122. obitias FE. (3-)
syllabae consonante exeuntis
abiicio F cf. p. 89. abliciam F“ “ aditiat BD.
A, abdicito F. conitio BEI. conlicitur F cf. p. 89. conitiam B, 9nitia E.
al. Vide Klotz. Altrém. Metrik,
lactendt Verbi Composita. 131
(ct. p.122) cOniciam Capt. 779.
conicite Cas. 386. cénicito Cas. 94. cO/nicere Trin. 238. Yniciam Yniciam Truc. 762.
Yniciam Cas. 589.
{niciat Pers. 71. Yniciatis Truc. 298. Inicite Capt. 659. inicere —_ Epid. 690. infcere Capt. 267. infcere Pseud. 407. {nicere Pseud. 643. Sbicio Cure. 567. Sbiciunt Cure. 531. Sbiciunt Trin. 1124. S’biciunt Pers. 470. Sbiciam Rud. 770 Sbiciet Epid. 664.
Sbiciemus Mil. 148.
Sbicias = Trin. 410. Sbiciatur Poen. 606. Sbicito
Sbicere Mil. 623. Sbicere Mil. 6109. Sbicere Most. 619.
Amph, 875.
Poen. 1235.
conitiam FE, coniiciam F cf. p. 89.
coniicito F cf. p. 89.
coniicere F “ «
Initium BE.
iniiclam F (L cf. p. 89), initiam rell.
A, inliciam F cf.p.8g, initiam EV.
initiat BCDbp, itiat Da, iniiciat F cf. p. 89.
BCD, inleciatis (corr. illeciatis) A cf. p. 96, iniiciatis F (L cf. p. 89).
inicite, icite in ras., B, iniicite F cf. p. 89.
A, incipere BCDF, iniicere ‘ pri- sca exemplaria’ Pii.
ACD, iniicere F cf. p. 89.
obitio E (corr. Es).
obitiunt E.
obiiciunt F cf. p. 89.
obiiciunt F “ «
A, obiiclam FF“ = *
obiiciet F eo 6
obitiemus B:CD, obiiciemus F cf. p. 89, obiciemus B:.
ABCD, obiicias F cf. p. 89.
obiiciatur F “co
obicito A
obiceret ante ras. B, obiicere C cf. p. 89, obveccere D cf. p. 97.
B et ex ras. D, obilcere C ct ante ras. D, item F, cf. p. 89.
obi BCD, obiici F cf. p. 89, obicere Miiller Pros. p. 538.
(34.)
132 Mauritius W. Mather.
@) Vocali autem exeuntes praepositiones in his producuntur,
dé‘ici Asin. 425. défcite' Stich. 360. ABCDF.
éfcite? Cas. 23. B, dicite V, diicite ras. ex discite E, ducite IF. (cf.p.105) @ici Asin. 127. pro ici Cist. 618. réfcio! Merc. go8. __reticio F cf. p. 117. (6.)
¢) Bis certe praepositio cum # littera sequente per synisesim, quam dicuat, coalescit. cf. pp. 104 sq., 114. ei cit Mil. 20s. ABCDF. reicis Asin. 254. _reice libri, reicis Lambin. (2.) J) In reliquis dudium est producta an per diphthongum cum # sequente praepositionis vocalis sit enuntianda. cf. pp. 104
Sq., 114.
dé‘iciam vel Stich. 349. A, de hic iam B, dehinc iam dei‘ciam CDF cf. p. 120.
déiciam vel Stich. 355. ABCD, deiiciam F cf. p. 117. dei’ciam
€icis vel Asin. 161. _ ras. unius litt. ante eicis D. éeicfs
€iciam vel Truc. 659. eicia D, iecia B, ieciam C cf. ei ciam p. 110.
€iciar vel Mil. 845. F, eicia BC, eici acellaria D. ei ciar
réiciam vel Pers. 320. A, reitiam C, reiiciam F cf. p. relciam 117.
r€icere vel Pers. 319. BCD, reicere A, reiicere F cf. rei cere . p. 117.
(7-)
1 His in exemplis atque in omnibus sequentibus Plautinis produci praeposi- tionem non demonstrant numeri. Verum tamen in talibus vocalem numquam corripi licet sine dubio adfirmare. Cf. p. 104 sq.
lactendt Verbi Composita. 133
TERENTIUS.
a) Terentius, quorum consonante finita prior pars producttur, haec suppeditat. cf. p. 100.
(cf. p.97) niecit Ad. 710. A cum rell. (ein ras. F), Donat.
in lemm. Obici Ad. 610b.
si’bice Ph. 387. (3-)
5) Correptas primas syllabas quae consonante cadunt cum non habeat poeta noster, ancipites tamen multas praebet. cf. p. 100.
Mbiciunda Ad. 744. abicienda A et, e in ras., D. (cf. p.122) cdnfcio Haut. 63. ic in ras C, conitio P.
cSnicias Haut. 292.
cénicias Hec. 842. ABCDF, conitias EP cf. p.117,
adn. 7. cS‘nicerem Ph. 190. cSnicito Ph. 166. cénfcere Eun. 547. infcere And. 140. Sbicerem Haut. 186. (9.)
¢) Vocali autem exeuntes praepositiones in his per se Jongam faciunt syllabam. cf. p. 132, adn.
éfcitur Ph. 673. éiciam Ph. 437.
éfciat And. 382. (ante caesuram versus.) éfciunda = Eun. 222. eiciinda A, ficiunda D. (4-) da) Per diphthongum haec necesse est dicere. cf. p. 132, adn., 114. rei ciat Ph. 717. rei cere Ph. 18. (2.)
¢) Unum exemplum licet duditare sitne per diphthongum legen- dum an membratim. cf. p. 132, adn., 114.
@iciat vel And. 382. (verbum a postremo proximum.) ei'ciat (1.)
134 Maurwius W. Mather.
RELIQUI SCAENICI POETAE.
Reliquorum, qui fabulas docuerunt, poetarum nunc indicabitur usus. @) Syllabas praefixas quae consonante finiuntur in his exemplis producendas declarant numeri. cf. p. 100.
(cf. p.125 sq.) dfssicis Naev. 57. II. p.16 R.’ ‘ “ dissicit Att. 348. I. p.181 R.* Ha, discicit Hx. “6 “ dissice Caecil. 239. II. p.74. Putean., disce volgo. inici Caecil. 262. II. p. 77. si’bicit Append. Sent. 49. sic volgo, subigit B,
IT. p. 368. subecit ay cf. p.97.
(5+)
5) Naevius solus correptam syllabam primam consonante cadentem exhibet.
(cf. p.99) &biciam Naev. 94. II. p. 23. (1.)
¢) In his éncerta est consonante exeuntium praepositionum men- sura. cf. p. 100.
Ab{cere Publil. Syr. 9. II. p.310. sic Ribb., aspicere libri.
A'dicis ex inc. fab. 86, II. p.127.
infcere Ennius 126. I. p. 31.
iniciendum Laber. 134. II. p. 298. (sine numeris citatum.)
Sbicitur ex inc. fab. go. II. p. 119.
Sbfcitur _—ex inc. fab. 57. II. p. 121. (6.)
ad) Vocali quae exeunt syllabae primae videntur in his producs. cf. p. 132, adn.
(cf. p.122) cdfcior Laber. 147. II. p. 300. “ « cdfcere Afran. 311. II. p. 204. éfcere Pacuv. 385. I. p.128. prdicit Pacuv. 94. I. p. 88. proicit Vat. Rott. Par.,
proiicit cett. cf. p. 118. (cf. p.105) pro‘ici Laber. 83. II. p. 292. (s.)
1 ©. Ribbeck, Comicorum Fragm., edit. sec., 1873. 3 O. Ribbeck, Tragicorum Fragm., edit. sec., 1871.
lactendt Verbi Compostta. 135
e) Synizesis, quae dicitur, semel apparet. cf. p. 132, adn., 114.
(cf. p.116) dei‘cis Laber. 119. II. p. 296. (1.)
J) Per diphthongum haec dicenda sint necne non liguet. cf. p. 132, adn., 114.
(cf. p.122) cd‘icere vel Afran. 216. II. cohicere L cf. p. 120.
coi’cere p. 191.
@iciebantur vel Enn.o. I. ei ciebantur p. 16.
pro‘iceret vel Afran. 347. II. proiceret Pp. 210.
réicis vel Att. 430. I. Flor., reiicies Haun., réi cis p. 191. reiices Goth., reiicis
rell. cf. p. 118. (4-)
ENNIUS (ANNALES).
Iam vero, ut reliquos poetas, qui omnes fere hexametros versus scripserunt, deinceps temporum ordine disponamus, Enni in Anna- libus usus indicetur.
a) Consonante terminatae praepositiones, ut constanter in herois usque ad extremam Augusti aetatem, producendae sunt.
cf. p. 87. (cf. p.123) cOnicit Ann. 61 M.=75 Vahl. inicit Ann, 172 M. =171 Vahl. (2.)
5) Solum exemplum, quod vocaéi finitam priorem partem habet, ea producta dicitur. cf. p. 105.
proiciunt Ann. 250 M. = 238 Vahl. (1.)
LUCILIUS.
Ex Lucili autem reliquiis excerpta sunt haec exempla, quorum in sex
136 Mauricius W. Mather.
a) Consonante cadunt praepositiones.
(cf.p.123) cOnicere XI.10 M. Obiciebat XI. 14 M. sibicit IV. 29 M. sic Salmas., sucit codd. sibicit XV. 37 M. sibicit XXVI.62M. (ad fin. septenari troch.). sibicit ?lib.incert. | subsit Fest. (summisit Paulus)
107 M. cf. 308 et 310 M., subrigit Lachm. in edit. vs. 1111. | _—— 6.) 6) In his autem vocaé: exit prior pars : —
(cf. p.123) cd'icis II. 25 M. cofcito ?XXVIIII.48 M. sic Lachm. in comm. Lucret. p. 136, coici aut conici codd. (sen. iamb.). défciam XXVIII.35 M. sic aut deiciunt codd. (sen.
iamb.). (cf.p.110) @iécére XXVITII. LH:, ecicere B:, eiccere Gr, | 106 M. eicere Gen. H: B: Gz. r€iceret XXII.6 M. sine metro citatum. (cf. 818 Lachm.). (cf.p.115) prdiciam 652 Lachm. perficiam XXVII. 22 M. ex vel Madvig. Advers. 1. 18, perei- proiciam ? ciam Non. codd. p. 296 M. (6.) LUTATIUS CATULUS. Giceret vs.4. Baehr. PLM. VI. 276. (1.)
M. TERENTIUS VARRO.
(cf.p.r11) @iécit Eumenid. IV. (cf. Vahlen. Varr. Sat. Menip. p. 125 Riese. Reliqq. Coniect. p. 183, frg. XXII). sic BH:, eicit LH2G.
(1.)
a) Consonante finitae praepositiones.
(cf. p. 123)
(cf. p. 125 Sq.)
lactendt Verbi Compostta. 137 LUCRETIUS. cO'niciunt 6.731. c6’nicere 1.751. co’nicere 2.121. cO'nicere 2.1073. coniciens 1.284. cd niciens 6.345. dfssicietur 3.639. O’biciens 5-755. (8.)
5) Vocaki finitae praepositiones.
dé‘icit
(cf.p.r11) @iécit
¢) Bis apparet diphthongus.
éicit
@iciuntur
@iciuntur @iciatur
éicere
pro iciunt
réicit ré iciat ~~ réiciat ré’icere
tra‘iciuntur (cf.p.111) tra’iécére
er cit el cit
A’bice
iniciens
5.1125.
2.951. AB, eicit Nicc., Flor. 31, Camb. 6.689.
3.58. sic Lambin., Lachm., Munro, eli-
ciuntur codd. cf. p. 117.
3-497. idem ac 3.58. oc 4«
4.945. eliciaturcodd. “ ‘*
4.1046.
.§-896. sic Turneb., Lambin., proficiunt codd. 1.34. B, Gottorp., reficit A, Nicc., Camb., al. 5-641.
6.81.
2.714. 3:757> 3.513. (14.)
cf. p. 115. 3.877. 4.1272. (2.) CATULLUS. 24.9. (vs. Phalaec.). 35-10. (vs. Phalaec.). (2.)
138 Mauricius W. Mather.
5) €iciunt 105.2. eliciunt G cf. p. 117. proicies 55.19. prohicies O cf. p. 120, proiicies G cf. p. 117. proiciet 64.370. O, proiiciet G cf. p. 117, proiecit A: L cf. p. 111.
pro‘icere 64.82. DC al., proiicere G cf. p. 117, pro- hicere O cf. p. 120. (4-)
CICERO!
Duorum exemplorum apud Ciceronem exhibet alterum_ certo Jongam priorem partem, alterum ancpitem.
a) (cf. p.97) abiécit de Div. 1.48.106. AV.
6) “ 123 codnfciet de Div. 2.5.12. (in arsi senari). (2.) TIBULLUS. (cf. p. 123) cOnicit 1.8.54. coniicit G cf. p. 89. sibiciet 1.5.64. subiicietG “ “ (2.) PROPERTIUS. a) @diciam 4.2.41. adice 4-11.77. iniciat 4-6.86. I nicere 4-7-34. Obicitur 3.19.1. Obiciat 2.30.25. si’biciet 1.7.20. (7-) 5) proicis 3.8.4. (1.)
1 Haec verba ex Tusc. 2.36 Thielmann in Walffl. Archiv IV. 600 in senarium voluit restituere, ‘ Impélluntur, feriuntur, dé/ciuxtir, cadunt.” Verum, ne asper- rimum dicam hunc esse versum, pro aetate Ciceronis omnino haud placet emenda- tio, quae primam syllabam compositi correptam reddit.
lactend: Verbi Composita.
139
VERGILIUS.
@) Consonante terminatur prior pars.
Adiciunt A. 8.304.
adiciunt 10.182. adiciam 12.837. a’dicias 11.354. cO'nicit 5.619. c6 nicit 7.347. co nicit 9.411. co nicit 10.646. co'nicit 10.657. CO nicit 10.891. cOniciunt 5.662. cOniciunt 6,222. cOniciunt 10.330. cOniciunt 10.8o0r. cOniciunt 11.194. co’nice G. 4.26. cOnicite A. 9.494. d{ssicit 12.308. dissice 1.70. dissice 7.339: fnicit 9-553- Pnicit 11.728. Yniciunt FE. 6.19. inice A. 6.366. 0 bicis 4.549. 0 bicit 6.421. 0 bicit 7.480.
co‘lcit c.
coicit R cf. p. 123.
coiciunt R cf. p. 123.
coliciunt P cf. pp. 117, 123.
be, disticit M cf. 7.339 inf., disi- cit Pay, disiicit P: cf. p. go, discidit R.
Rybcm, schol. Servi 5.683, dissi- cep M, disice Donati exempl.
Ryby, dissicae c, dis‘ice M cf. 12.308 sup., disiice ut vid., supra scripta fortasse s, V cf. Pp. go.
Re, incitat MPyb, incutit Heins. inciunt V.
obiecit Prisciani L, obiiecit G cf. p. 97.
140 Mauricius W. Mather. o’biciunt 2.444. o’biciunt 9-45. O’biciunt 9-379. 6 bicitur 2.200. Oo bicitur 5-522. Obicienda E. 3.7. sibicio A.3.314. sibicit E. 10.74. sibicit G. 2.19. sibiciunt 4.304. sibiciunt A. 2.236. si’biciunt 5.103. si’biciunt 7.110. sibiciunt 12.288.
(cf. p. 102.1) supériacit 11.625. superiicit P cf. p. go, superlicit
superinice G. 4.46.
6) Vocadi finitae priores partes. dé‘icis A. 11.665.
(cf. p. 111) dé@‘iécit G. 1.333. oo“ déiécit A. 8.428. oo dé@iécit 10.753.
(cf. p. 112) d@iécit 11.642.
66
dé‘ice G. 3.422. pro‘icis A. 11.361. proiécit 5.776. prd’ice 6.835. ré‘icit 10.47 3. r€iciunt 11.619. r€ice G. 3.389. traiécit A. 9.634.
6 traicit 10.400.
triicit 11.685.
y1,cf.p.117, suberigit R cf. Sil. 15.155 (p. 148 inf.). (43-)
RM, deicit ybe. R, deicit Mybc, de-i:cit P cf. p. 117.
Rez, Prisciani w, deicit MPV yb.
c1, deicit MPRyb.
Cc, proicit R. prolice R cf. p. 117.
y2cm, traifcit b, traicit M, translit ut vid. y1, transigit P, transadigit R.
trahicit M cf. p. 120, trai-cit (eras.
1) ccf. p. 117. (15)
lactendt Verbi Compostta. 14!
¢) Per diphthongum enuntiatum.
(cf. p. 116)
a)
6)
¢) (cf. p.116) dei’cere S. 1.6.39.
réice EQ 3.96.
HORATIUS.
a’bicito Ep. 1.13.7. 4 diciant C. 4.7.17.
iniciat 1.17.26. iniciat S. 1.6.32. G’biciebat 1.4.123. O biciet 1.6.69. O’biciet 1.6.107. o’bice Ep. 1.16.62. Oo bicere C. 3.10.3. prdo‘icit A.P. 97.
proicere S. 2.3.100.
OVIDIUS.
A'dicis A.A. 3.8. A‘dicis M. 11.285. a’dicit A.A. 3.152.
a'dicit M. 2.384. i dicit 7.266. a'dicit 11.637. a‘dicit 11.671. a'dicit 12.58.
a’diciunt 1.245. a’diciunt 7.121. a’diciunt 10.656.
Fdiciam Am. 2.13.25. 4’diciam M. 9.628. Adiciam Pont. 1.8.56.
(vs. Alcaicus.) inliciat C, illiciat g cf. p. 117, initiat ER: ut vid., z.
(vs. Asclep.) (9.)
(2.)
(1)
addicit AX.
addiciam FP.
142
(cf. p. 123)
éé 66
4 6
(cf. p. 126)
Slauricius W. Mather.
Hdicerem T. 5.5.25. addicerem HKV afnav,
a’dice Med. Fac. 63. a'dice 82. a’dice Rem. Am. 558. a dice 790. a’dice M. 6.182. ‘dice 14.319. a’dice T. 3.1.49. i'dice F. 1.189. dice 4-75. co nicit M. 7.245. co niciunt 5-42. co’niciunt 11.28.
co niciendus T. 3.11.46.
di{ssicit
{nicit inicit Tniciunt fniciunt fniciam fniciam i niciet iniciet inicias fnicias
M. 11.386.
Am. 3.9.20. M. 9.78.
A.A. 1.116. F. 6.515.
Am. 1.4.40. 2.5.30.
1.4.6.
T. 3.7-35-
Ep. 19.190. Pont. 3.4.101.
adijcerem Q cf. p. go, adducerem C _ (corr.), adderem D (corr. Dz).
addice BEOB{nyps, Fragm. Burn. 277, adiice Q cf. Pp. 90.
codd. plerique, addice & adiice Ns cf. p. go.
co‘icit M.
coniiciendus H cf. p. go, conviciendus O, conni- ciendus «, conijciendus Q cf. p. go, iniciendus F2V: (recte FV).
sic aut disicit codd. pleri- que, dissidit A cuius in marg. deiecit cf. p. 112.
innicit V, inniciet V3.
lactendt Verbi Composita. 143
fnicerem Ep. 12.158. Tniceret M. 3.389.
nice Ep. 8.16. inicere M. 1.184. Obicit " 3.516. 0 bicit 13.308.
Sbiciunt T.5.10.40. obiiciunt QV cf. p. go. Obicitur Am. 2.7.18. d’dicies 2.2.37. si bicio T. 4.1.74. subijcio Q co 4
(49.)
6) Correpta prima syllaba quae consonante cadit non occurrit apud
Ovidium. Namque illud dics, quod Merkel et Giithling in Pont. 2.3.37 acceperunt atque L. Miiller in libro de Re Metrica p. 291? (p. 250") defendit, nimirum adigé verbo locum dat atque cedit; scribitur enim in uno codice 8, Monacensi lat. 384 saec. XII, qui liber quamquam ad constituendam Ponticorum scripturam ponitur inter praecipuos, hic tamen non sequendus est contra ceteros, qui adig# tradunt. Neque Ovidius, qui aliter his consonante exeuntibus primis syllabis semper longis utitur, semel brevem passus esse potest putari. cf. p. ror.
déiicit M. 1.719.
€icitur T. 5.6.13. eiicitur deflor. Vincent. Bellov., eijicitur Q_ cf. p. 117.
prdicit Ep. 21.165.
pro‘icit M. 9.575.
proicit Pont. 3.4.97.
pro‘iciere I. 164. proitiere T.
pro‘iciare 294.
réicio Ep. 21.200.
réicit F. 1.436.
ré’iciat T. 1.1.66. reiciet Bp (corr. ps), proiciat
ERS, eiciat (in ras, d dispicitur) yw.
144 Mauricius W. Mather.
réicerer ré’ice réice ré‘ice ré‘icere réicere trdicit tré‘icias trd‘icerer
@) (cf. p. 116) déi’cere
M. 9.606. .
Am. 1.4.34. A.A. 1.695. M. 14.677. 2%s82. 9.51. 9.128.
F. 4.782. tralicias in ras. m. cf. p. 117.
T. 5.2.33. trahicerer BGaHR al. cf.
p. 120, traiicerer Ago,
traijcerer Q cf. p. 117,
transigerer V, traicerem
GO. (19.)
? F. 4.709. dicere habent codd. pleri- que, unde in Merkeli edit. min. Teub., 1884, @eicere restitutum et p. xxxix defensum est. de cvuce posuit Merkel ipse in edit. maiore. risrerz, quod est scriptura cod. Ursiniani (V) saec. XI, editortbus fere placuit.
(1.)
VONSOLATIO AD LIVIANM
petporam adscripta Nasoni unum exemplum praebet. quod comsue-
teakuaewa ithus sarvat. a hoe
adp (1.)
YX MABINGS
Quaue way dar cAcmpua
aAdRY
Bp Baa (1
lactends Verbi Compostta. 145
Sub finem Augusti imperi incipiebant poetae correpffas adhibere praepositiones quae consonante finitae subiunguntur taciends verbo. Hanc tamen novitatem nec omnes admisere, nec singuli constanter. cf. pp. 87, ror.
MORETUM.
Moretum carmen et veterem et novum ostendit usum. cf. p. ror. a) Abicit 96. sic aut adicit plerique codd., adiicit He cf. pp. 83, adn., go,
abdicat V. (1.) 5b) #dicitur 99. adiicitur BHe cf. p.go. = (1.)
CIRIS.
Ciris exemplum diphthongi habet.
(cf. p. 116) reicere ?118. sic Heins., dicere aut ducere codd. (1.)
AETNA.
Aetnam, cum post aliquanto componeretur, hic liceat ponere.
nice 405. C, isse H. (1.) GERMANICUS.
a) (cf. p. 101) sitbicft 196. obicit var. script. (1.) 6) (cf. p.115) tra‘icit 512. traxit et pertraxit var. scriptt. (1.)
MANILIUS. a) (cf. p. 101) Adic(e) 4-44. *(1.) 5) (cf. p.115) prodicit 4.259. (1.)
PHAEDRUS.
a) Consonante exeuntes praepositiones in arsi senari iambici positae ancipitis sunt mensurae; cf. p. ror.
146 Maazrictas bi Mather.
Ebiciet $-5-42- 6) Veeshk cadit praciixa syfinba. (cf. p. 115) €3a «Append. S.20. enci NV cf. p. 117. (1.) SENECA.
a) Comsomande quae terminantur pracpositiones producuntur. si quidem codicibas credere possumus, in formis tribus assi-
Gendt verdi. cf. autem pp. ror, adn. 4, 126. (cf. p. 126) dissicit Tro. 393 L. — dissipat A, dissicat E, corr. &,
= 404 P. cf. p. 124, adn. 6. (vs. Asclep.) “ “ dissicere Agam. 896 L. (sen. iamb.) = 954 P.
“-« dissicite Phoen. 343 L. sic E. deicite interpolator E co- a Oecd. fr. dicis. (sen. iamb.)
345 P. (3.) 5) Consonante cadentes praefixae syllabae his in exemplis corrs piuntur, chop. ror,
Xdicit > de Clem. 2.5.5. Thielmann in Wolff. Arches IV. p. 600.? idice Med. 277 L. = 277 P. (sen. iamb.) Xdice (bis) Med. 527 L. = 530 P. (sen. iamb.). adijce A cf. p. go. X'dice Med. 783 L. = 786 P. (sen. iamb.) i'dice Oed. 811 L. = 832 P. (sen. iamb.) i’dice Herc. Oet. 364 L. == 367 P. (sen. iamb.) Adicit Thyest. 727 L.==727 P. (sen. iamb.) 4dic(e) Med. 471 L. = 474 P. (sen. iamb.) dbicit Med. 496 L. = 449 P. (sen. iamb.) ddici Med. 237 L. = 237 P. (sen. iamb.) dbicf Herc. F. 434 L. = 438 P. (sen. iamb.) (11.)
i |,, am edit. Leo. 1878-79. P. = edit. Peiper. et Richter. 1867.
§ Ie senarium putat restituendum esse, “Maerér contundit méntes dércit cén- trakit.” Forsitan hoc recte coniecerit, sed alicui antiquo poetae quod tribuendum Sentit, periculum est ne erret, quoniam rarius apud antiques (cf. p. 99) haec fit ‘
lo.
lactendit Verbs Compostta. 147
¢) Consonante finitae priores partes dreves an longae sint numeris decerni non potest. cf. p. 101, adn. 4.
Hdicere Phoen. 201 L. = Oed. fr.201 P. (sen. iamb.) Sbicere Med. 497 L. =500 P. (sen. iamb.) (2.) ad) Uno in exemplo vocaii terminata praepositio productione sit legenda an per diphthongum est dudium. Haec tamen ratio se magis commendat.!
(cf. p. 116) prdiciet vel proi‘ciet Phoen. 426 L.= 64 P. (sen. iamb.) (1,)
OCTAVIA
fabula praetexta Senecae falso olim attributa, haud tamen ita multo recentiore tempore scripta, habet duo exempla quae cum eius con- suetudine consentiunt.
Adic(e) 125 L. = 130 P. (sen. lamb.)
sibicit 827 L. = 843 P. (sen. iamb.) (2.)
LUCANUS.
a) Consonante exeuntes in omnibus exemplis praepositiones sunt corripiendae. cf. p. 101.
dbicfs 8.796.
dbicit 9.188.
siibicft 7.574. subigit aut subegit var. scriptt. siibicf 8.740. subiit G:.
(4.) VALERIUS FLACCUS.
@) Consonante finitae praepositiones sunt constanter /ongae. cf.
pp. 87, 102. a‘dicias 7.508.
a’dice 8.41. (cf. p. 123) cOniciunt 6.271. (cf. p. 126) dissicit 3-162. C, dissicet P, disicet M, disiicit M2. cf. p. 99.
1 Cf. L. Miiller. de Re Metr., pp. 163-6? (150-3°).
148 Mauricius W. Mather.
fnicit 3-343-
Tniciunt 2.236.
o bicis 5-627.
6 bicit 6.679.-
obicit 7-460.
obicit 7-524
6 biciat 8.388. (11.)
+) Veesk cadentes praefixae syllabae im his producuntur. cf.
p. 115.
dé icit 1.191.
dé icit 2.330.
df icit 6.194
dF icit 6.218.
GE iit 6.552.
pro icit 3.537. (6.)
¢) Unum exemplum per svecsestr: dicendum est.
(cf. pr 116) deicit 7.514. nunc deicit vultus codd., Schenkl, Baehrens; deicit hinc vultus edit. Bonon. pr. a. 1.474. Thilo.
(1.) SILITS ITALICUTS. a) Comswnant quae exeunt praepositiones in his exemplis prods-
cuaiur. cf. pp. 101-02.
(ch p 123) CO nicit 23.306.
(cf. p 126) d¥ sice 9.53% distce LV. disire vel disice
corr. in dissijce F. cf. p. goa
dissite O, discute volgo. a 4&4 dissice 13394 Ch, discite O. cf. p 102, adm. 1) supe riacit 225.155. sic Bauer ed. r$92: superiicit vel potius supericit scriptum
esse tudicavit Wagner ad Werg. 11.623. suberigit codd. LOV. cf Verg. 1.625 (p. 130 sup.). subegerit F. super-
laciends Verbi Composita. 149
6) Ceteris in exemplis syllabae primae sunt confractione legendae. cf. pp. 101-02.
inicft 10.570. dbictt 4-149. siibicift 1.113. sibicf 13.298. (4.)
ILIAS LATINA.
Ilias Latina, quae fortasse a Silio scripta est, habet haec exempla duo : —
a) (cf. p.125sq.) disiceret 325 (Baehrens. PLM. III. p. 23).
disiiceret E cf. p. 90, discideret
FV, divideret MN, dis“ceret L,
discuteret B. (1.)
5) (cf. p. 115) trijicit 835 (PLM. III. p. 48). (t.)
STATIUS.
a) Consonante quae cadunt priores partes in omnibus, praeter unum, exemplis producuntur. cf. p. 102.
a’bicit Achil. 1.172. i bicit Th. 2.479. (cf. p. 126) dissicit 1.590. BG:M, disicit H Anglic., dissicet P, discicit S, discidit Pal. 1. oo dissiciat Achill. 1.311. PGs, dissiceat Pc, disso- tiet G:, deiiciat D cf. p. 118, discutiat ptH.
ou“ dissice Th. 10.69. Inicio 5.315. Tnicit? 3-434. Tnicit 6.194.
1 /nicit initio vs. Th. 9.807 omnes codd. habent, praeter Puteanum (P), codicum Statianorum excellentissimum, in quo ieci¢ ef, quod postulat contextus verborum. scriptum legitur.
150 Mauricius W. Mather.
Yniciam 1.242.
rnice 7.518.
inice 11.595.
sii bicit 2.189.
sibicit 5-672.
sibiciunt 3-716.
si biceres 1.74. (15-) 6) (cf. p.102) &dicf 704. (1.)
¢) Vocaii exeuntes praepositiones tantum non omnes faciunt per se syllabam Jongam. cf. p. 115.
d@icit Th. 6.650. déicit 12.368.
dé‘icit 12.743. prd icis 2.460. pro icit 1.388. pro ice 2.658. prd‘cite 3-643. ré icit 6.770. PGH, eicit Gabecr, eiecit B
cf. pp.112,114; elicit. codd.
Behottiani cf. p. 118. (8.)
@) Per diphthongum autem dicendum est unum exemplum. (cf. p. 116) reicit Th. 4.574. (1.)
MARTIALIS.
a) Omnium exemplorum primae syllabae comsonanée finiuntur corripundacgue sunt. cf. p. ror.
Adicit 4.54.9. Xdicit 10.82.1. adiicit b¢ cf. pp. 83, adn., go. si’bice 9.75.10. (scazon.) (3-) IUVENALIS.
a) (cf. p. 101) aAbicit 15.17. (1.)
lactendi Verbt Compostta. 151
SERENUS SAMMONICUS.
a) Hutus ad a. 200 p. Chr. n. florentis poetae utrumque exemplum habet productam syllabam praefixam. cf. pp. 87, 102.
a bice 113 (PLM. III. p. 112). adicies 463 (PLM. III. p. 128). (2.)
Posteriorum solos poetarum Ausonium Burdigalensem et Claudium Claudianum dignos puto qui hac in quaestione, quod versus Latinos numeris perfectis scripserunt, laudentur. Horum
AUSONIUS
@) ea tantum exempla praebet quorum priores partes exeunt consonante atque producuniur. cf. p. 102.
adiciam p.195. III. 2 [150.2] ed. Peiper. 1886; P- 49. 22; p. 101.12; p. 105. 23.
&'dicies p. 195. III. 2.
0 bicit p. 88. 17.
Obicitur _p. 89. 39.
si’dice p- 338. LXX. 2.
si/biciet pp. 91.9. (9.)
CLAUDIANUS.
Claudianus autem plerumque /ongas, bis tamen correptas habet praepositiones. cf. p. 102.
a) a‘dicias 1.141.
Obdicis 21.301.
O’bicit 26.613.
O’biciat 33-74-
si’bicit 18.358. (5.) 5) dbicfs 8.365.
stibicit 36.134. (2.)
¢) (cf. p. 115)
dé'icit
28.230.
(t.)
HOMERIC QUOTATIONS IN PLATO AND ARISTOTLE.
By GEORGE Epwin Howes.
INTRODUCTORY.
S it has not seemed wise to enter in this paper into the question
of the authenticity of the various works ascribed to Plato and Aristotle, I have here included all the quotations from Homer that are contained in any of the works edited under the name of Plato or Aristotle. For the text and variants of Plato I have relied, wherever possible, upon the collations given by Schanz (Platonis Opera quae Feruntur Omnia, 1875-). Unfortunately for classical scholars the edition of Schanz is still incomplete; the readings, therefore, of the following works only are taken from his text: — Alcibiades I., Alcibiades II., Amatores, Apologia Socratis, Char- mides, Convivium, Cratylus, Crito, Gorgias, Hippias Minor, Ion, Laches, Leges I.-VI., Lysis, Meno, Phaedo, Phaedrus, Protagoras, Sophistes, Theaetetus. From the edition of Stallbaum (Platonis Opera Omnia) have been taken the text and variants of the follow- ing: — Leges VII.-XII. (1859), Minos (1841), Philebus (1842), Respublica (1858). From Hermann’s edition (Platonis Dialogi, 1853) are quoted the passages in Axiochus and the Epistulae. In giving passages from Aristotle, greater uniformity has been possible by adopting for the complete works the text and collations of Bekker (Aristotelis Opera, 1831), and for the fragments the text of Rose (Aristotelis qui Ferebantur Librorum Fragmenta, 1870), both edited by the Berlin Academy. The principal other authors of whom critical use has been made have been quoted from the following editions :—- Homer, from La Roche (Homeri Ilias, 1873-1876, and Homeri Odyssea, 1867-1868); Hesiod, from Rzach (Hesiodi quae
154 George Edwin Howes.
Feruntur Omnia, 1884); Aeschylus, from Wecklein (Aeschyli Fabulae, 1885); Sophocles, from Jebb (Antigone, 1891; Electra, 1894; and Oedipus Tyrannus, 1893) and from Campbell (Sophocles, The Plays and Fragments, 1881; from this the Ajax is cited); and Euripides, from Kirchhoff (Euripidis Fabulae, 1867-1868). The editions of the Homeric Scholia used are those of Dindorf (Scholia Graeca in Homeri Odysseam, 1855; and Scholia Graeca in Homeri Iliadem, 1875-1877, containing the Scholia of Venetus A and of Venetus B) and of Maas (Scholia Graeca in Homeri Iliadem Townleyana, 1887-1888). Any departure from the text of the editions mentioned above has been indicated by a note.
I have thought that a study of the quotations from Homer found in our manuscripts of Plato and Aristotle might have a two- fold value, — it might show whether these authors quoted accurately or not, and it might possibly shed some light upon the Homeric text of theirday. It is evident at once that many difficulties beset our path. The mistakes of the scribes of the manuscripts of both Homer and of the authors quoting him have, of course, been numer- ous; and yet, if we assume that all the differences of reading between the passages quoted and the quotations are due to the mis- takes of these scribes, we beg the question at the outset, and admit that these authors quoted from the same Homeric text that we have to-day, and quoted accurately. Moreover, if we claim that all the variations, apart from those caused by the carelessness of scribes, are due to the practice of the ancients of quoting from memory, we again beg the question by assuming that none of the variants in the quotations has a real variant Homeric reading to depend upon. Besides, even if it should be granted that these authors may have quoted from memory, — an induction that does not necessarily follow because of a great difference between a passage quoted and the quotation, — this explanation would need to be used judiciously and not applied to every apparent case, for many passages that would seem at first sight to offer this as the most plausible explanation will on ‘careful study be explained in a much more satisfactory manner. But let me not anticipate too much. I wish, merely in a general way, to indicate some of the difficulties that confront us. It would be beyond the limit of this paper and of my ability to
Homeric Quotations tn Plato and Aristotle. 155
attempt to offer all the possible solutions of all the difficult questions that present themselves. It is my intention, while least of all wish- ing to dogmatize, to give what seems to me to be the most probable explanation of the various passages under discussion.
I. QUOTATIONS FROM THE DRAMATIC POETS AND HESIOD.
I have deemed it advisable to consider, somewhat briefly, the quotations of Plato and Aristotle from the dramatic poets and Hesiod, so that an impression, more or less distinct, may be formed of their general trustworthiness when quoting from other authors. In dis- cussing the quotations from the dramatists I shall consider those passages only that are found in the extant plays; for the fragments, except in special instances, necessarily fail to offer a satisfactory basis of comparison. |
Quotations from the Dramatists.
A. So far as I know, Plato gives but two quotations from the dramatists, apart from several references in which there is no attempt to quote the exact language.
1. Rep. 2, 362 A= Aesch. Sept. 580-581 : Plat. ov Soxeiy ddtxov GAXr’ elvan d6eAKv, Bobeiay droxa dia ppevos Kap rovpevoy, é€ Hs 7a xedva BAaordve. BovAcvpara, Aesch. fadeiav droxa da ppevds kapwovpevos, €§ 7s ra xedva BrAaordva BovAcipara. Plato has changed the nominative xaprovpevos to the accusa- tive, that it may fit the structure of his sentence.
2. Alcbiad. 1. 151 B= Eur. Phoen. 858-859: Plat.’ oiwvav €Ogunv, pyoi, xadAinxa orédn: dy yap xAvdon xeiuel’, dorep olafa av: Eur. olwvdv €Oéuny xadrAinxa oa orédy: dy yap xAvden xeined’, Gorep oloba ov,
1 Schanz, following Buttman, restores a4. Cod. B, xelyeOa; T, dtaxelpeda.
156 George Edwin Howes.
As od of the verse of Euripides is necessary for the trimeter, its loss from the manuscripts of Plato is probably due to a copyist, as we cannot suppose that Plato would have allowed such an unmetrical verse to stand in his text.
8. Aristotle has given us twenty-four quotations from the drama- tists. In seven! of these the manuscripts of Aristotle coincide with those of the poets, with the exception of a few unimportant variants evidently due to the carelessness of scribes.
The other seventeen will require some discussion.
1. Rhel. 3, 14 (p. 1415 b 20) = Soph. fatig. 223:
Aristot. drag, dpe péy oty orws orovdas vro. Soph. drag. dpe péy oty Gras Taxovs vro
The reading oxrovdys receives additional support from the scholiast who writes: ot rotro A€ya, ore pera orovdas doOpaivwr Epos ot weropevpar. The coincidence of the use of oxovdys by both Aristotle and the scholiast may, of course, be accidental; but it is at least striking and entitles the reading to a fair consideration. Even if Aristotle is quoting from memory here, as many suppose, the read- ing is not thereby invalidated. A man may quote from memory and still quote correctly. We might add that some of the editors — ey. Dindorf and Schneidewin — have adopted orovdys in their text.
2 (and 3). 2d. View 9. 9 (p. 1169b 7) and Mag. Afor. 2, 15 (p. 1212 b 27) = Eur. Or. 667: Anstot. Eta. Vic. 9 oray & Satnee e® ddg, ri SeT Gov; Anistot. May. .Mor. oray 3°6 Saiuew ® beds, ri SeT SXer; Eur.” oray 6'S Saiuew ev dds, Ti yoy Ser: The second quotation of Aristotle assures us — what we should otherwise readily have assumed — that the omission of 8° in the first quotation is merely a copyist’s blunder. Besides. it confirms the
1 De Mumio 6 (p ycob 25) == Soph. O. 7. 4-5; Rhet. 3 15 (p 1416a 30) = Eur. Aap. 612, Ret. 2. 22 (p. 13948 29) ae Eur. Mal. 294-297: Eth. Ema. 5, 1 (12352 16) ae Eur. Or. 235: Reet 3, 2 (p. 1405b 23) am Eur. Or. 1588S: Kéet. 317 (p. 2415 D 22) ae Eur. Zroud. 971; Rtet. 2, 23 (pr ry00b 23) = Eur. Zrossd. 990.
2 Cod. B has de¢ which Kirchho edits.
Homeric Quotations in Plato and Aristotle. 157
reading eZ, at least for Aristotle. Without going deeply into the question we may say that it is very doubtful whether the Attic poets ever used xp7 with a genitive. Besides, in Eur. Here. Fur. 1338 we have a similar verse — whether spurious or not:
Geot 8’ crav ripwoww, ovdiy Set hiror:
If genuine, this verse shows, a similar phase; if spurious, it was probably modelled after Eur. Or. 667. Again, the reading ri de? ¢iAwy is confirmed by Plutarch.’ Further, as already noted, the word de7 itself still appears in one manuscript of Euripides. So we should agree with Kirchhoff, Nauck, Paley and others in admitting de7 into the text of Euripides.
4. Rhet. 3, 6 (p. 1407 b 34) = Eur. /ph. Taur. 727: Aristot.? AdArov pév aide rorAvOvpor Scaxrvyal. Eur.® Sarrou pév aide TOAVOpyvos Sarrvyai,
The word roAv@pynvor, ‘much wailing,’ was long ago seen to be wrong. We are indebted to Aristotle for the true reading. The folds of the tablet were ‘many-gated,’ z.c. there were many leaves that might be considered to form the entrance to the tablet. Although Euripides has used an uncommon expression, it receives some justi- fication in the 5é@vpoyv of Pollux, 1V. 18: “Hpddoros pév A€ya SeA- tiov Siwrvxov, of 5¢ "Arrixot ypappareiov SiOvpoyv, xai Oupas ras wruyxas dxpe Svo, dra wrriyas xai rpixrvyov xai woAvrrvyxoyv; and later in X. 57: Ser rovs SeAria, ws eiwely ypappareidov S{Oupov 7 tpiwrvyov 7 Kal wred- vey xrvyev. The reading of Aristotle is, therefore, welcome, and is accepted by Kirchhoff, Nauck, Klotz, Paley and modern scholars generally.
5. Fol. 1, 2 (p. 1252b 8) = Eur. pk. Aul. 1400: Aristot. BapBépov 8° “EXAywas dpyxery elxds, Eur. BapBdpwv 8° “EdAnvas elxos dpxecv, dX’ ob BapBapovs,
1 Moral. 68 E.
2 Cod. Yb, 8éArov pla 82.
8 Edited by Kirchhoff, roAd@vpor.
# Edited by Kirchhoff, &pyew elxés.
158 George Edwin Howes.
The manuscripts of Euripides offer a metrical difficulty, namely a spondee in the odd foot of a trochaic metre. Ways suggested for avoiding the difficulty have been the cutting of the verse into two parts or the substitution of the Doric form dpxeyv. The discovery of the quotation in Aristotle, however, practically settled the matter in favor of the reading dpxecy eixos.
So far we have considered twelve passages, which show that Aris- totle’s quotations are entitled to great respect; seven of them are practically identical with the passages quoted, while the other five give readings superior to those found in our manuscripts of the poets themselves. The remaining twelve passages offer greater difficulties.
1. Rhket. 1, 13 (p. 1373 b 12) =Soph. Aatig. 456-457: Aristot.' od ydp te viw ye xdxOés, GAN’ dei rore {i tTovro, xoideis older & Grov davn. Soph. ov ydp rc vow ye xdyBés, GAN’ dei wore {yj ravra, xovdeis oldey ef crov ‘pavy.
Verse 456 is quoted by Aristotle again in RAeé.* 1, 15 (p.1375b1):
ov ydp Tt vuv ye KdyBés, GAN del wore.
In the two quotations of verse 456 the variants in the manuscripts of Aristotle contradict one another and thus corroborate the readings of the manuscripts of Sophocles. Possibly the word rotro was purposely written by Aristotle, that it might harmonize with his pre- ceding words: olov xai 4 SopoxAcovs "Avtiyovn qaivera A€yovoa, srt Sixarov drapnyévov Odiyyar rov [loAvveixyn, ws dvca dv rotro Sixacor: Otherwise its use must be due to the carelessness either of Aristotle or of the scribes.
2. Rhet. 1, 1§ (p. 1375 b 1) =Soph. Ankg. 456 and 458: Aristot.® ov ydp re viv ye xdxOés, GAA’ det wore. Tair’ ov éyw odx Eedrov dvdpos ovdeves. Soph.‘ od ydp rt viv ye KdyBés, GAN’ det wore ToUTwy éyw ovK enedAov, dvdpds ovdevds
1 Cod. QY?, re (for ye); Z, oe (for ye); Q, xal xOes. 2 Cod. A‘, rd» (for r:). 8 Cod. Ac, rdv (for rs); A‘, Aueddov. 4 Cod. L, éy' ode.
Homertc Quotations tn Plato and Aristotle. 159
Evidently this passage was so well known to his hearers or readers that Aristotle thought it unnecessary to quote it in full. It is quite possible that he may have used the word rovrwy, which is undoubt- edly right in the verse of Sophocles. A copyist might readily have changed this to ratr’ oty either carelessly, or because he thought that éxeAAov would most naturally be followed by an infini- tive, of which ratra would be the object.
3. Rhet. 3, 14 (p. 1415a 20) = Soph. O. 7. 774: Aristot. €uot rarnp Hv ToAvBos. Soph. poi rarip piv ToAvBos Av Kopivbios,
Little stress can be laid upon this passage, for Aristotle is rather referring to the verse than quoting it.
4. Rhet. 3,11 (p. 1411b 29) = Eur. /ph. Aul. 80:
Aristot.? robAcvbepov 8° “EXAnves dfavres rociv Eur. rovvrevbev otv “EAXAnves déavres Sopi,
This passage of Aristotle is clearly corrupt in the manuscripts. Possibly the word rooiv is involved in the corruption; it is surely more prosaic than dopi, and is probably wrong.
5. £th. Eud. 7, 1 (p. 1235 a 22) = Eur. Phoen. 539-540:
Aristot.? r@ mAdov 5° aiet roAguov xabiorarat TovAacaoy, €xOpas 0 Auépa KaTapxerat.
Eur.’ rw wou. 8° det woA€uov xabiorarat rovAaccov €xOpas O' ypépas xarapxera.
Except for the accent of é€y@pas one manuscript of Aristotle gives the same reading as the manuscripts of Euripides, and may preserve the correct tradition. The meaning of the last verse of Euripides, “and begins the hostile day,’’ is somewhat obscure, however. The reading of the manuscripts of Aristotle, ‘and the day begins hostility,” is about as intelligible and may possibly be
right.
1 Cod. Ac, robdetOepoy EAXnves; YbZb, roddebhepow 38° ErAnves; Q, 7rd AevOdpois 3° EAAnves. Bekker has edited rodvredbev oty “EXAnves. 2 Cod. Pb, dyépas. 8 Some cod., alel; C, rNelom.
160 George Edwin Howes.
6. Rhet. 3,17 (p. 1418b 21) = Eur. Zrvad. 969: Aristot.' rois Ocots xpera ovppayos yerpoopat. Eur.? rats Oeatoe xpera ovppayos yerpoopa The article rots is a purely grammatical blunder and cannot be attributed to Aristotle. That he had a feminine gender in mind is evident from his next words (also a quotation), éyw yap "Hpav. He may have used rats Oeots, which a scribe might think was a mis- take for rots Oeots. If he wrote rats Oeator, which the manu- scripts of Euripides show, it might have-been changed by a scribe, first to the common Attic rots Oeats, and later to rots Oeots. 7. Rhet. 3, 16 (p. 1417. a 32) =Soph. Aatig. 911-912: Aristot.* pyrpos 5° év adov xai ratpos BeByxorwy otx €or’ adeAdos Gatis dv BAdoro: ore. Soph. = pntpos 5’ év “Acdou xal xarpos KexevOorocy ox €or addeAdos Goris Gy BAdoror sore. The reading BeByxorwy may be due to the carelessness of Aristotle or it may have crept in as a gloss of some learned man,
who beside the xexevOorocv of his text wrote the corresponding expression BeByxorowv.
8 (and 9). het. 2,21 (p. 1394b 16) and £th. Eud.7,2 (p.1235b 21)== Eur. Zroad. 1051:
Aristot. (/et.)* ovdeis eparrys coris ovK det drAXci, Aristot. (2th. Eud.) ovdcis yap épacrips coris otk dei prc, Eur. ovK €or epaorys daTis ovK dei pircl.
In the passage of the Eudemian Ethics the word ydp has been introduced to join the statement more closely with the preceding words.
10. fol. 1, 13 (p. 1260a 30)=Soph. Ajax 293:
Aristot. yvvacxi xdopov 7 ary pepe, Soph. =ywvat, yuvacki xoopov F oryn pépet. 1 The verse is given thus in all the manuscripts. Bekker, however, edits rais
Geatcr. 2 Cod. BCG, Oeator; B, rats Geaicr. 8 Cod. Q, a» Braor9; ZbAC, dvaBrdorot. # Cod. Ac, dpacbels Soris obi Kal od.
Homeric Quotations in Plato and Anstotle. 161
11. Metaphys. 4, 5 (p. 1015 a 31) =Soph. Liz. 256: Aristot. ddd’ 4 Bia pe ratr’ dvayxdLe rocety. Soph. 4A’ 4 Bia yap radr’ dvayxdla pe Spay,
12. Rhet. 2, 21 (p. 1394b 4 and 6) = Eur. Hee. 864-865 : Aristot. ov« éorw.dvipdy doris dor’ éXevbepos 7 Xpnuarwv yap SovAds dori H ruyys. Eur. oux gore Ovntray doris dor édXevGepos: 7 Xenpdreov yap SovAds dorw 7 Tuyys,
Most of these last quotations of Aristotle, though differing from the manuscripts of the dramatists, contain — if we except palpable blunders evidently due to copyists—readings that are intelligible and quite possible. Some of them probably represent correct old read- ings ; for it would be singular if, where variants are found between the manuscripts of Aristotle and of these authors, he is right only
when grammatical or metrical difficulties prove the traditional read- ings of these authors corrupt.
Quotations from Hesiod.
After this rather brief reference to the passages of the dramatic poets quoted by Plato and Aristotle, we may perhaps with profit glance at the passages quoted from Hesiod.
A. At first sight Plato’s quotations from Hesiod seem to show great carelessness, as almost all of them give readings different from those contained in the manuscripts of Hesiod. Each of these pas- sages, however, will need to be considered separately.
1. Conviv. 178 B= Theog. 116-120:
Plat. avrap érera yat’ etpvorepvos, ravrwy dos dopares aie, 75° “Epos.
Hes. “Hrou pév rpwricra Xdos yéver’, airap éreara Tat’ cipvorepvos, ravtwv dos dodadés ale [dOavarwy, ot éxovc. xdpn vipdevros ‘OAUprov, | Tdprapda + nepdevra puxa xOovds evpvodeins, 75° “Epos,
162 George Edwin Howes.
In the words actually quoted the manuscripts of Plato agree with those of Hesiod. The omission of verses 118 and 119 will be dis- cussed later.
2. Lheaetet. 207 A = Op. e¢ D. 456:
Plat! éxarév 5é re Sovpud” dpdéns. Hes.? éxardv 5€ re Sovpar’ dudéys,
The difference here is mainly one of breathing, and therefore of little account, as manuscript traditions on such matters have small weight. We might say, however, that the best manuscript of Plato has dovpar’, which, if correct, would imply dudés — with smooth breathing — and thus cause a correspondence between the best manuscripts of the two authors.
3. Rep. 5, 466 C = Op. et D. 40: Plat. yvwoeror rov “Hoiodoy ore r@ ovte Hv copes A€ywv mréov elvai tus Butov wayTos. Hes.2 = Nipriot, ovde ivaciy, Gow wrA€ov yutou wayTos
Here it is evidently the purpose of Plato not to quote, but merely to refer to the passage of Hesiod. The two words yyecv ravrds are common to both passages, and there is nothing in the rest of the reference in Plato inconsistent with the manuscript readings of Hesiod. In similar language Plato again refers to the same passage in Leg. 3, 690 E.
4. Lysis 215 C= Op. ef D. 25-26:
Plat. «ai xepapeds xepape? xorée xai dordds dord@ Kal TTWXOS TTWXY,
Hes. Kai xepapets xepapel xorée xal réxrove TéEKT OY, Kol TTWXOS TTWYO POoven Kai dordos dordg.
These verses were variously quoted in antiquity. In one place — Pol. 5, 10 (p. 1312 b 5) — Aristotle gives the order xepape? xepapevs, though there it is rather a reference than a quotation. In three other instances,‘ however, he shows the traditional manuscript order,
1 Cod. B, 8ovpar’. 2 Cod. Mm,, dudtn ord; Mmg, audém; most cod., dovpad’ dudtys. 8 Cod. MZBAV, ovd" te. 4 Rhet 2, 4 (p.1381 b 16); Rhee 2, 10 (p. 1388 a 16); Lth. Lud. 7,1 (p. 1235 18).
Homeric Quotations in Plato and Aristotle. 163
which is confirmed by many other writers also. In Priscian' we find verse 26 quoted thus:
A Q a 4 s a, 4 Kal TTWXOSs TIWXD POovéa Kal TéxToMe TEXT WY.
This, in an indirect way, tends to corroborate the reading of Plato. For the last part of the verse as quoted by Priscian refers to the class of men (réxroves) which Plato would naturally have mentioned if he had finished his verse, since it is the only class referred to by Hesiod but omitted by Plato. Apparently, even in the remote past there were differences of reading, which may easily have arisen before the time of Plato, and even have crept into manuscript copies of Hesiod.
5 (and 6). Crat. 397 E and Rep. 5, 469 A= Op. et D. 121-123: Plat. (Crat.) abrap érady rovro yévos cata potp éexdrAuwper, of pev Satuoves dyvoi broxOdvioe xaXr€ovrat, écOr0i, dAcEixaxos, pudraxes Ovynrav avOpwrer. Plat. (Aep.) of pév Saipoves dyvoi émcxPdvioe rer€Bovary, écOAot, drckixaxot, pirdaxes peporwv dvOparwy Hes.’ abrap érady rovro yévos xara yata xaduype, roi péy Sainoves cio Atos peydAou da BovaAds €aOrot, érrxOoveor, pidraxes Ovynray dvOpwruy,
This is a difficult passage to settle satisfactorily. Plato differs not only from Hesiod but also from himself. Let us consider first those readings in which he consistently differs from Hesiod.
(a) ot pév. That this reading was found in the manuscripts of Plato in early times is seen from Eusebius,’ Hermogenes,‘ and others, who quote it thus from Plato. Lactantius,® though with a variant roi, quotes it thus from Hesiod. It may very well have stood in the manuscript of Hesiod to which Plato had access.
1XVIIL. 145, p. 169 K.
2 Cod. A has éwel xe, with 84 written above; cod. MC, éwel yer, with xe written above; almost all the other cod. have érel xev. ;
8 Pracp. Ev. 13, 11, p. 663 A.
4 Ed. Walz., Rhet. Graec. } p. 320.
§ Inst. Div. 2,14, 7.
164 George Edwin Howes.
(4) dyvoi. This word, too, is confirmed for Plato by Eusebius, Hermogenes, and Theodoretus.' It is suggested for Hesiod also by Plutarch,” who in wrongly quoting the verse as
dyvot éryxPonror pvdraxes Ovryrav avOpurrov, shows that the word dyvoi was somewhere in the sentence.
(c) dAcEixaxow. The testimony of Eusebius, Hermogenes, Theo- doretus, and Aristides*® shows that this is correct for Plato. Though Theodoretus is referring to the Cratylus of Plato, he thinks he is giving the words of Hesiod, for he prefaces his quotation with these words: 6 5€ ye ‘Hatodos repi rov xpvaod yevous radra épy.
Now let us look at those words that show Plato as differing from Hesiod and inconsistent with himself.
(2) émcxPoveore (Kep.); broxPdvior (Craz.). As Aristides alone, who seems to be quoting from the Cratylus, reads troxOovcor, while the other authors, including Theodoretus, who quotes from the Cratylus, give émrcxfoveo., probably Plato wrote in both passages éatxOovcor, which was early corrupted in the Cratylus to tro yOdviox.
(¢e) reXeBovorv (Rep.); xnadéovrac (Crat.). The authors quoting Plato are about evenly divided on these words. Probably these readings represent very old variants which may have extended back to old manuscripts of Hesiod.
In the case of one word, Plato agrees with Hesiod in one passage but disagrees in the other:
(/) peporwy (Rep.); Ovnrav (Crat., and also Hesiod). For the former word Eusebius offers his testimony, while the latter is con- firmed by many ancient authors. As the phrase peporwy dvOporwyv was a common ending for verses of both Homer and Hesiod,‘ it would have been easy for either Plato or a scribe to write pepowwy dvOpwrwy instead of Ovnrav dvOparwy.
In the Cratylus, Plato quotes one more verse than in the Republic. In this verse he differs from Hesiod in one phrase:
18, De Mar. p.gt5§ D. 3 Moral. 431 E. ® Vol. II. p. 171 (230 Dind.). ‘Cf. Hom. //. 1, 250; 3, 402; Hes. Op. e¢ D. 143, 180. Cf. also Plat. Leg. 3, 681 E, where he quotes from Hom. //. 20, 217, wéds pepdrwr dvOpwrwr.
Homeric Quotations in Plato and Anstotle. 165
(g) potp éxdArAupey (Crat.); yata xddAuwpe (Hes.). Theodoretus confirms potp for Plato, and evidently thought it correct for Hesiod. The phrase yata xdAvwe recurs in verses 140 and 156 of the Works and Days.. So it seems to me probable that Plato, if in verse 121 his text of Hesiod had read yata xdAvwe, would have quoted it so, even if he were giving the words from memory, as the repetition of the phrase would have impressed it upon his mind. Why is it not fully as natural, therefore, to suppose that the original reading of verse 121 of Hesiod was really potp éxdAvwe, which was early changed by a scribe to yata xaAvwe because of the repetition of that phrase in verses 140 and 156? ‘Taking all these points into consideration, I am inclined to believe that the text of Hesiod from which Plato quoted may have been:
avrap éready TovTo yévos Kara potp éxarduwpe KaX€ovTat
rot pév Saipoves dyvol ércyPdvioe reAcOovary
éxOrol, drckixaxos Piraxes Ovytay dvOpuruv.
7. Rep. 2, 364 C = OP. et D. 287-289:
Plat. os rHv pev xaxoryra xai Aadov dori éXéoOan pydius: Acin peév ddds, para 8° éyyVh vaie: ras 8° dperns Wpwra Oeot rpordpobev €Onxav
Hes. tyv pév roe xaxéryra xat lradov or éAéorbat pydius: drAlyn pev 680s, para 8° éyyvhe vaie. ras 8’ dperns pura Oeoi rpordpoSev €Oynxav
The word ws given by the manuscripts of Plato is nowhere con- firmed for this passage of Hesiod and undoubtedly merely connects the quotation with the preceding words of Plato: rovros 8 maor rots Adyors pdprupas woinTas émdyovrat, of pév Kaxias mépe evrrereias d.ddvres, ws, etc. Some scribe, however, supposing it the first word of the quotation, and finding that there were too many syllables in the verse, may naturally enough have retained ws and omitted rox.
In the variants Aefy and éA’yy we certainly have testimony strong enough to prove absolutely that there were old readings of Hesiod that are not found at all in our manuscripts of that author. For, although all the manuscripts of Hesiod here read éAdyy, the
’
166 George Edwin Howes.
reading Aeiy is confirmed (1) by Plato,’ who, in referring to this passage of Hesiod, again uses the word Aecy; (2) by Xenophon,’ who in turn is confirmed by Stobaeus*; and (3) by Plutarch.‘
8. Ley. 4. 718 E= Of. ef D. 289-292:
Plat.’ idpera Geot xpowdpoGey eOnxav dGarura, paxpos 5¢ xai cpfros olpos és avryy, wai Tpyyis TO xperor: éxyy 5° dis dxpov ixyat, pedis Sy “recta réAes, xarery rep éovva. Hes.* res 8° dpergs idpera Geol rporapoey. éOyxav dOavarac: paxpos St nai ophros oluos és airny Ku TPRY'S To Epetor> éxyy 5° cis axpow ixytrat, pyrddy bg €xecra weAXeu yarery rep eotea.
(a) The reading fxygac is confirmed by one manuscript of Xeno- phon," from whom Stobaeus® also quotes the same reading, and by one manuscript of Hesiod. while ixgrac is substantiated by Stobaeus® in another passage. The fact that Plato in the Protagoras” uses fx»rac is of no importance, for there he is merely referring to this passage of Hesiod and not quoting it: and, besides, he there adds ris to show that the statement is a general one, thus giving the same force to the sentence as if he had said icyac. I am inclined to think, therefore, especially as the form fxyrae offers dithculty in the way of interpretation, that «ya is the correct reading for Hesiod.
(4) The variation of $9 ére:ra and $y "recra needs no discus- sion, as it is a point to be determined by the judgment of the editor rather than by a particular manuscript tradition.
ic) The word ¢epew of the manuscripts of Plato is a mere blunder of somebody. For in referring to the passage again, Plato” Says: oruy O¢ res avtys cis dxpow Leyru, ppdcav Sgrara rede... .
d Zoey. 4. DIS E. 3 Mew. 2. 1, 20. " 8 Floral. 1, tol. 4 Vora... p. 77D. 3 The cod. have @épev (for réX\a). ° * Some cod. éw atrdz; Vat. 221, Gopar: Mm*. 32 Greve; mm’, djrare
V Mew. 2.1.20 Cod. A, ewes: the rest, Uwres
® Flord. 1.101. © Flore. 1, 1.
¥ yo Dz. D Prony. 310 D.
Homeric Quotations in Plato and Anistotle. 167
9g. Crat. 428 A= OP. et D. 361: Plat.) e& xai tis opixpov emi opixp~ xaradecn, Hes. Ei ydp xev xai opixpov émt opixpe xarabeto, The difference of these two readings is not easy to explain, unless we assume that Plato preferred to use another form for a general
statement. Still, both readings are metrical and possible, though a syllable is lacking in the verse in Plato.
10. Sep. 2, 363 B= OP.” et D. 233-234: Plat. 6 pév ras Spis rots dixaiocs robs Oeovs rocéiy dxpas pey te hépervy Badrdvous, péooas S¢ periooas: eiporoxor 5’ dues, Hyot, padrAois xaraBeBpiPact, Hes. axpyn pe re hépec Baddvous, péeoon St peAXiooas: eipordxot 8° Gres parrots xaraBeBpibacr-
Plato here adapts the verses of Hesiod to the structure of his sentence. There is nothing, however, inconsistent with the words of Hesiod.
11. Charm. 163 B= Op. et D. 311:
Plat.? guadov yap wap’ ‘Haiddov, os édn, épyov ovdey elvar dvecdos. Hes. "Epyov 8° oddev dvedos, depyin 5€ r° dvedos.
Here, too, the words are woven into the structure of Plato’s sen- tence.
There is another passage of Plato that, although it does not quote from Hesiod, makes such a reference to him as to entitle it to our consideration :
12. Crat. 402 B:
dpa ola dro Tov abroudrov abrov dudorépos pevpdtwv dvopara bér8a, weorep abl “Ounpos® 'Oxeavov re Oewv yéveotv drow Kai pytépa Tfiv- ofpar 88 cai “Hovodos. Aya S€ wov nai ‘Opdeds ore xrd.
The point of the passage for us lies in"the words ofpa: 8% xai “Hoiodos. Jowett,* who evidently thinks that Plato means that
1 Cod. B, opuxpotd (for opxpe). 2 Codices, 3” odder. 3 //. 14, 201. 4 In a note to his translation of this passage of Plato.
168 George Edwin Howes.
Hesiod had a very similar verse telling of ‘Ocean, the origin of gods, and mother Tethys,’ says: ‘The verse is not found in the extant works of Hesiod.” It seems to me, however, that Plato may have meant that Hesiod, too, describes Oceanus and Tethys as parents of (some) gods. I should agree with Hermann, therefore, in considering that Plato had in mind Zheog. 337:
Ty bis 8° ’OQxeavp Torapots réxe Suvjevras,
The words olfac 5é xai “Hoiodos surely offer considerable evi- dence in favor of the view that Plato quoted from memory. The natural interpretation would be: ‘I think Hesiod has such a verse, but I can’t recall it.’”’ Still, the words might imply merely: “I think Hesiod has such a verse somewhere, but I don’t know just where to look for it.” And we must bear in mind that looking for passages whose place was not tolerably well known, was a much more arduous process with the old rolls than with modern books.
Conviv. 178 B= Theog. 116-120. This passage was quoted a few pages above.' Plato is referring here to the antiquity of the god "Epws. It would not be surprising, therefore, if he omitted every- thing in the passage quoted that was extraneous to his purpose. But, since we find that Aristotle in quoting the same passage twice omits these same verses (118 and 119), we are led to believe that these verses may not have existed in their texts of Hesiod, especially as they are, for other reasons, suspected by many scholars.
A consideration of all of these passages leads me to think that Plato had a text of Hesiod different in many respects from ours; and that his variants must not thoughtlessly be dismissed as due to ‘lapse of memory.’
B. After this somewhat cursory treatment of Plato’s quotations of Hesiod, let us turn to Aristotle’s quotations of the same author. Three?’ of these, apart from very slight differences evidently due to scribes, give the traditional readings of Hesiod. The others I shall treat separately.
1 Cf. page 161. 3 Pol, 1, 2 (p.1252b11) and Oec. 1, 2 (p.1343a 21) = Of. et D. 405; Probl. 4, 25 (p. 879a 28) = Op. et D. 586.
Homeric Quotations in Plato and Aristotle. 169
4 (and 5). £¢h. Mic. 9,1 (p. 11644 27) and £¢th. Eud. 7, 10 (p. 1242 b 34) = OP. et D. 370: Aristot. (Z¢h. Vic.) év rots rovovrots 5° éviots dpéoxet TO picbos 5° avdpi. Aristot. (2th. Eud.) ucdos dvdpi dag. Hes.? Mic Gos 8’ avdpi pirw eipnyévos dpxios oro.
In the Eudemian Ethics the loss of 8’ makes the fragment unmetrical. The particle must have been omitted by a scribe. A knowledge of the real reading is shown in the passage of the Nicomachean Ethics.
6. Eth. Nic. 1, 2 (p, 1095 b 10) = Of. et D. 293-297:
Aristot.? ovros pév wavapioros 6s avTos mavTa vonon, éa6Xos 8’ ab xdxetvos Os ed eirovte wiOyrat. os S€ xe pyr avros voy pyT dAdov dxovwy év upg BadAnra, 6 S’abr’ dxpyos avyp.
Hes. = Ovros pév ravdpicros, os ait@ wdvTa voncy, ppacadpevos, td x erecta Kai és réAos you dpeivw° éoOr0s 5° ab xdxetvos, Os eb eirovre wiOyTat:
Os Sé xe py air@ voy pyr dAAov dxovwy év Ounw BddAnTaL, 6 Sadr’ dxpyos avyp.
The main difference in these passages is that the manuscripts of Aristotle give generally adros (verses 293 and 296) while those of Hesiod have aire. It seems clear that Tzetzes had in his manu- script of Hesiod atros in both places, for he says (on verse 293): wavdpioros éori, ds tis dd éavrovd xai atrodiddxrws To mpérov voi, and on verse 295: dyaos St xara Sevrepov Adyov, cai 6 py ad éavrod pay yvous, weoOeis 5¢ rots KaA@s ovpBovdAevovorw. Moschopulus seems to have read the same, for he says: yyovv ovros pey éoriv dpioros, os
1 Cod. B, y’; in many cod. this verse is omitted.
2 Cod. Lb Mb, aéré (verse 293); Lb, voéer; Mb, vofoe: (Mb adds ppaccdpevos 74 xtwara xal és ré\os foww duelrw); HaKbMbNbOb (ie. all the cod. except one), wel@yra:; Lb Mb, B4rnrac; Mb, ad.
3 Cod. Mm,, Z (with the gloss d¢’ éavrod), L, abr@ (verse 293); O, adrds; all cod., vohoe; B, low; ZA, eloly; M, dyelywy; M (verse 295) omits 3°; B, xA7' abre (verse 296); some, uhr' a’ry; Mmz,, ptr’ adrds (from avrds); many cod., vodes.
170 George Edwin Howes.
6: davrot rayvra voyoa ... os 5 dy pyre 5: éarrot voy pyr’ d\Aov axover «ri.
Since there are many writers some of whom quote atros and some etre, we may fairly infer. I think, that both readings go back to a very old period. Verse 294 of Hesiod is omitted by Aristotle and also by Aristides’ and by Clement of Alexandria.* It is quoted, however. by Stobaeus*? and by Andronicus Rhodius.* Whether Anistotle had it in his text of Hesiod or not is uncertain, for he might readily have omitted it as unessential to his quotation, even if he had had it. Sull the verse is open to suspicion and has been rejected by some scholars. ¢g. Brunck and Steitz.
7. Khet. 3.9 (p. 1409 b 28) = Of. et D. 265-266: Aristot’ of r’ aire anxa terya drap d\Agw xaxa reryer. y dt paxpa dvaSoAw Te roiyoarri naxioTy: Hes§ OU 7’ atre xaxa rerya dvap GUA xaxa revyer, y be nanyq BovAy tre Bovrevoarri xaxiorn.
These two verses do not properly belong in a collection of Aristotle's quotations. for they are a parody by Democritus, as Anistotle himself says: ecre yivera: 0 éoxee Angoaperos 6 Xios cis Me\ormrdyy roujourra dyri rev dyrucrpoduy dvafolds. of tT avre xrA.
8. Occ. 1, 4 (p. 13442 17) = OP. et D. 699:
Anistot.”’ wupGerucy b¢ yuneiy Cra ben xebva &dafys - Hes. wapbenany dé ype. ws x’ yen xedva S&dafys. Aristides.’ in referring to this passage of Hesiod. uses va. iva
has far the greater probability as the original reading, because it restores hiatus: Gu pyéea.
1 Vol. IL. p. 26 tp 31 Dind).
3 Paed. 3, 8 (p. 279).
3 Fiori’, 4, p. 252 (ed. Meineke).
4 Eta. Nic. Paraparasis i. O
§ Cod. QYDZD, avasoAyy; QO, caxiory yap.
© The word 7° edited by Kzach has no manuscript authority. Almost all the cod. have 7 or 0”.
T Cod. Md, ddaLfy.
® Vol. IIL. p. 33 (p. 41 Dind.).
Homeric Quotations in Plato and Antstotle. 171
9. Eth. Nic. g, 10 (p. 1170 b 21) = OP. et D. 715: Aristot.? éuperds epnrbat Soxet
pyre worvgervos pyr déetvos, Hes. Mynde rorveecvov pd’ dfecvov xaréerbau,
With the variants pyre and pydé we need not concern ourselves as they are often confused in manuscripts. If the manuscripts of Aristotle represent him correctly in this passage, it must be con- sidered either as a mere reference or as a misquotation, for the words woAvgecvoy and agecvov must be right in Hesiod, since they are in the same construction aS vexeorjpa (verse 716), which the metre demands.
10. Eth, Nic. 7, 14 (p. 11§3 b 27) = OD. et D. 763-764:
Aristot.? g@ypn 8 ov ri ye wdurav drddAvra, yvTiwa Aaoi wodXoi...
Hes. Pypn 8’ ov ris wduray dwroAAvtat, yvtiwa roAXol Aaoi PypiLover:
Of the variants ré ye and res we need say only that some manu- scripts of Hesiod® show ri, to which ye could easily have been added by a copyist, and one manuscript of Aristotle has ris without ye.
Demosthenes,‘ Aeschines*® and Dio Chrysostomus® give the order of words Aaoi roAdoi, while other writers confirm the traditional reading of Hesiod swoAAot Aaot. Of the five manuscripts of Aristotle collated by Bekker for this passage four read roAAoi alone, the other has of woddot. This oi, of course, may be the remnant of a previous Aaoi, but the weight of evidence would seem to suggest that Aristotle wrote merely woAAoi, thus completing the verse in accordance with our reading of Hesiod.
1 Cod. Lb, dfevos; LbOb add xandeolunp».
2 Cod. Kb, ovris rdpway; Kb has ol (for Aaol); the other cod. show no trace of the word.
* Cf. critical apparatus in text of Koechly-Kinkel for this passage.
$ De Fals. Leg. p. 417.
© Con. Timarch. p. 141.
8 Or. 37, p. 128 R.
172 George Edwin Howes.
11 (12, 13 and 14). ol. 5, 10 (p. 1312b 5), Kher. 2, 4 (p.1381b 16), Rhet. 2, 10 (p. 1388 a 16), Ath. Eud. 7, 1 (p. 1235 a 18) = OP. ef D. 25.
Aristot. (ol.) WS KEpapel KEpapLers
Aristot. (het. 2,4) xepapeds xepape’.
Aristot. (A/et. 2,10) nai xepapedis xepapel.
Aristot. (Zh. Eud.) xai yap xepapets xepape? xorda,
Hes. Kai xepapeds xepapel xorée kai réxrov TEKT WY,
From these last four passages of Aristotle it is evident that, when he was quoting only a part of a verse, or perhaps referring to it, he did not feel it necessary to give the exact language. A comparison of these four passages shows that Aristotle had here the same read- ing in his Hesiod as we find in ours.
15 (16 and 17). Phys. Auscul. 4,1 (p. 208b 30), Afetaphys. 1, 4 (p. 984 b 27), and De Xenoph. 1 (p. 975 a 11) = Theog. 116-120: Aristot.’ (Phys. Auscul.)
TAVTWY pev RpwTioTa xaos yever’, avTap éxara
yat’ eipvorepvos, Aristot.? (Metaphys.)
rdvTwy piv TpwTLoTa xaos yever, avTap érara
yat’ etpvorepvos,
HO depos, Os wavrecoe petampéewer AGavdrorocy, Aristot.® (De Xenoph.)
TpaTov piv ravrwyv pot xaos eyever, abrap érara
yat’ etpvorepvos, mavrwy eos adodarés aici,
nS épos, Os mdvrecoe perampérer A0avdaroicty, Hes. “Hroe pév rporiora Xdos yever, atrap érara
Tai’ eipvorepvos, ravrwy eos dodadés ale
[dbavdrwy, ot éxovat xdpy vipdevros ‘Odvprov, |
Tdprupa Tr nepdevra pvxy@ xOovds edpvodetys,
95° "Epos, os xdAAtoros év dbavarorot Oeotas
1 Cod. F, yévocr’. 3 Cod. Ha, adrd (for adrap); E, yéa yaia; HaAb, sows. 3 Cod. Ba Ra Va, alel3epos.
Homeric Quotations in Plato and Anistotle. 173
(2) In the first two passages, in which Aristotle is apparently intending to quote exactly, we read wavrwy, and in the third, where the inversion of words and the metrical difficulties in the first line indicate that —if our text is correct—the quotation really begins with atrap, we still find ravrwy. It seems to me the natural infer- ence is that Aristotle had wxayrwy in his text of Hesiod.
(6) The omission in the Metaphysics of verses 118 and 119 is of no great importance, for there Aristotle quotes only what is neces- sary to prove his point, namely that Hesiod made "Epus a ‘first cause.’ But still, as the general subject of this passage is the ‘first causes,’ the omission of Tdprapa from this list would seem to indicate that, even if Aristotle had verse 119 in his text of Hesiod, he did not interpret Tdprapa as one of the ‘first causes.’ In De Xenophane, where he quotes at greater length, the addition of verses 118 and 119 would materially change the force of é50s and yat’. The omission of these verses by Aristotle, therefore, indicates to my mind that they were not contained in his Hesiod.
(c) In two passages Aristotle gives wavreoot petamrpéerer aba- varotory, where the manuscripts of Hesiod read xdAXrcoros év &Oavarocot Oeotor. For the reading of Aristotle we find no support among ancient authors. It has been thought that Aristotle confused the verse of Hesiod with this verse of the Homeric Hymns!:
~ td @ ~ a’ tA mais €u0s, os xe Geotoe petampéror dGavarorory.
That view is perhaps possible; and yet it is not impossible that Aristotle gives us a variant reading that has elsewhere disappeared.
Taking all of Aristotle’s quotations together, I feel that they are tolerably accurate. Some differences between the quotations and the passages quoted are probably due to his carelessness; many are undoubtedly due to the blunders of scribes; but there is left a considerable number of differences that are best explained, I think, as coming from an earlier text tradition than is preserved in the extant manuscripts of the authors quoted, especially as I have proved, in a few cases at least, that a difference of text really existed. Feeling, therefore, that readings offered by Plato and
1 2, 149.
74 Gerzge Edain Howes.
Arstoce must mot be repected merely because they find no support im the munsesxripcs af the acthors quoted, but must be carefully cossatered a5 poskiuy giving independent testimony on many matters w texan. fet me iperasch my real subject.—the quotations from Homer.
Il. Qcorarmys Frow Homer. LET ERY.
A very casa. reading of 2 few of the scholia of Homer convinces ws that there were. even in very ey times many manuscripts of that auther, Some of te vartiat, taoceh they have disappeared eniirey from te Noy of Se text af the extant manuscripts, are SfL.. Preserved i= tae scheca = For exampie. in a scholion of Ven. A on Ua ta... we read: em Znvedures youoa €@ifaroy. .. . Tivés S& yanduevyy dyideavery, It cur extant mainuscripes we read éri- qurvey and c@.carer. Sut towtere €yidaaver. In the scholia in mansacrps HM. Disdcert: ce Odyssev 1. a3. we read: ruvés reuo S és Koyryy re white the minearipss of the Odyssey all show Srapryy re.
Amwwher proof of the aziguity of some cf the variants is found in Pipvres tragmeeaQy some cf wl zo Quck te 1 very oid date. I shall Speak Of omy twe of these farmersk
r. A papyrus fragment cortizing [iad 24. verses 127-804 Of the mary pecwiarttes of reaciag of cis fragment let me select two BS CKAMMER
yw? Ta verse i153 the fragment reads rees. Ven A and some Quher Manusempss read rad: the rest have rees An interlinear gloss of Ven. A tes us: eeres yews ree = The scholiast. then, kuew doth of these readings
y? Un werse gto the tragarent has 6:2 Ven A and many other manuscnys dave irda white the rest have 6.7) In a marginal schotiean of Vea. A we tod: ya O° avrees. Both readings then. were kowen to che schedast Now this fragment 5 assigned by La Roche? to the trsg century either Bo. or um) Hence we see that
> Somer. te Traders. BR ape
Homeric Quotations in Plato and Aristotle. 175
some of the Homeric variants are of high antiquity, and were known to the scholiasts.
2. The other fragment is of considerably greater importance for us. It is the one discovered recently in Egypt by Flinders Petrie.’ It contains portions of Iliad 11, verses 502-537. Though there remain only the first letters of some verses and the last letters of others, the fragment shows remarkable differences when compared with the traditional readings of the Iliad. Let us note these differences.
(a) Verse 515. Papyrus, raccwv: /., raccecv. Some of the ancients suspected this verse of Homer, as we see by the scholion in Ven. A: dOeretras, ... xai “Aptoropavys tponOérer* Zyvodoros 5¢ ovdé Eypaderv.
(6) Verse 520. Papyrus, ws: /1., vyas.
(c) Verse 528. Papyrus, xevO um: 11? xeto’ tm.
(7) Verses 529 and 530. Papyrus,’ xovpot +: Jl, imaqes and dAAyAOVS.
Besides, the fragment shows the endings of four verses not found in our manuscripts of Homer, — between 504 and 505 vongcey, between 509 and 510 yxts eAorvro,® between 513 and 514 vovo, and between 514 and 515 aAAovs.* Both Mahaffy® and van Leeuwen’ assign this fragment to the third century z.c. It is undoubtedly older than any other scrap of Homer that has come down to us, and offers invaluable suggestions with reference to early traditions of Homer. It preserves not only variants nowhere given in our Homeric manuscripts or scholia, but also traces of verses that have been lost to us. In view of these facts it will be unwise in our present investi-
1 Published by Mahaffy (with other fragments) in Flinders Petrie Papyri, Dublin, 1891.
2 Cod. Lips., «ei".
3So Mahaffy; but van Leeuwen says (.Waemosynxe 20, 1892, pp. 127-130), the fragment reads AT IVI or AT LU.
¢ Van Leeuwen thinks it is x7 ehocev.
§ This word, as van Leeuwen says, has been accidentally omitted by Mahaffy in his transcription.
© Athenacum, Dec. 6, 1890, no. 3293, p- 777-
7 Mnemosyne, 20, 1892, p. 127.
176 George Edwin Howes.
gation to reject without consideration any variants offered by ancient authors, even if they are unsubstantiated by any of our Homeric
manuscripts or scholia.
At last we are ready, I think, to examine thoughtfully the passages quoted from Homer by Plato and Aristotle. Some of these are only phrases or parts of verses, but I have included them in the list of
quotations.
Flato’s Quotations from Homer. A. No VARIANTS.
Plato gives many Homeric quotations that in the manuscripts used by the best editors show no readings different from those in the best manuscripts of Homer. ‘The minor differences of breathing, accent, adscript zofa and movable au I have not considered as variants, for any manuscript tradition on these points is comparatively modern and entitled to little consideration. There are fourteen of these quotations that show no variants.
I. Conviv. 219 A= 1. 6, 236:
xpvoea xa\xelwy
The entire verse is quoted by Aristotle.’
2. Phaedo 112 A= T//. 8, 14:
THAc par’ xe Babiorov td yOovds tor. BépeOpov:
3. Crat. 392 A=J//, 14, 291:
xaArxida xixAnoKover Oeoi, dvdpes 58 xuipuvdiy, This verse is quoted thus by Aristotle? also. 4. Rep. 3, 388 C= J//. 18, 54: wpot éyw Sern, apo Suvcapiorordxesa. 5. Afol. 283 D=V//. 18, 98: atrixa, pyoi, reOvalny 6. Gorg. 449 A= //. 20, 241 and often: evxopat elvat,
1 Eth. Nic. § tt (p.1136b 10). 3 Hist. An. 9, 12 (p. 615 b 10).
Homeric Quotations in Plato and Antstotle. 177
7. Protag. 340 A=TJI. 21, 308-309! pire xaciyvyre, cOévos dvépos duddrepol rep oXaper. 8. Rep. 3, 387 A=. 23, 100-101: Yux 58 xara xOoves, WUre Kamves, @XETO TeTpLyvia: 9. Ae. 3, 391 B= //. 23, 151: TlarpoxrAw pat, én, xduny dracapa péper Oa, 10. Alcibiad. II.’ 142 E= Od. 1, 34: twip popov dAye exovar. 11. Protag. 315 B= Od. 11, 601: Tov 5¢ per’ civevonoa, 12. Rep. 3, 390 B= Od. 12, 342: Ape 8° otxrurrov Gavéey xai worpov éxrmety 13. Epist. 7, 345 E= Od. 12, 428: Opp ére THv dAOnY dvaperpHoai xapvBdxy, 14. Theaetet, 170 E= Od. 16, 121: pada. pupior
B. SLIGHT VARIANTS.
To the list just given may properly be added twenty-one other passages, in which the variants, whether in the manuscripts of Plato or Homer, are slight and such as constantly arise from the careless- ness of scribes.
1. Rep. 3, 389 A=? 1, 599-600:
do Beoros 5° dp’ tvapro yéAws paxdpecar Oeotory, ws ov "Hdaorov da Swpata rovrvvovra.
2. Rep. 3, 389 E=J1.* 4, 412:
rérra, cLwry yoo, dup 5° exireifeo pvOy,
1So Schanz, though the codices of Plato have dwréppyopor. The codices of Homer vary between twép ubpor and bréppopor. That this difference was felt as a real variant is seen by schol. MQ (Od.): 0d cdpOeroy rd Uxep nbpor. Cod. B (Plato) has 4\y' xouveu.
3 Cod. G, 3° évapro. 3 Cod. A, rérya; N, ovy#.
178 George Edwin Howes.
3. Crat. 415 A= J//. 6, 265: Plat. py pe droyudcys péveos. Hom.? py p droyuwoys péveos, 4(and 5). Zheaetet. 152 E and Crat.* 402 B=//.414, 201 and 302: "Oxeavoy re Oeav yeveow Koi pytépa TrOiv 6. Rep. 3, 386 D= /2.* 16, 856-857 and 22, 362-363: Yuyy 5° éx peOewy rrapévy “Arddsde Be By xe, Ov TOTBOV yoowsa, ALTOvG’ dvdpoTyTa Kai HByv : 4. Phil’ 47 E=T/. 18, 108-109: | woAvppova rep xaXerHvat, Os TE TOAD yAvKiwy mEeALTOS Katara BopwEvorco — 8. Rep. 3, 386 D= J." 20, 64-65 : oixia 5 Ovyroics xai AOavaroor pavein opepdar€ , eipwevra, TA TE OTVyEoVTL Deol reEp° 9. Crat. 391 E= 178 20, 74: dv HavOov, pyoi, xadéovor Oeoi, dvdpes S& Yxdpavdpov 10. Rep. 3, 391 A= 1! 22, 20: yo av Tiwaipny, et por Svvauils ye wapein. 11. Leg. 7, 804 A = Od.” 3, 26-28: TyrA€uay’, GAA pev adres evi dpect oyor voncets, Gra. St xai Saipwv troPycera: od yap diw ov oe Dewy déxyte yevéerOar re Tpapéuev Te. 12. Leg. 3, 680 B= Od." g, 112-115: roto 5° ovr dyopal BovAnddpar ovre Oguores, dAX' of y' bYyAGv dpéwv valover Kdpyva dy orécot yAapupoict, Oewrorever 5¢ Exacros maidwv 75° dAdxwv, ovd" dAAnAWY dr€yovcrw.
1 Cod. T, droyudeys, emended to droynweys.
2 Cod. LO, péveos 5°; G, péveos 7’.
® Text, yéveciy pynow. Cod. T omits dyer». 4 Cod. H (verse 201), ‘ye.
® Cod G (verse 857), ddporfra ; Cant. Mor., ddporira.
6 Many cod., ws re wdpv. 7 Many cod., gavin, with and without sofas. 8 Cod. L Harlei., xduavdpor. ® Cod. D, ef ph. 10 Cod. K, ob8é Oedr a’. 11 Cod. I, of8’; DFIK, owéor; AQ, owéeor; K, yAagduppor; K, waldwy 7’.
Homeric Quotations in Plato and Aristotle. 179
The words Ocuoreva ... dAdywy are so quoted by Aristotle.! Once? also he refers to them, but with the reading raidwy 73° dA cyxov. 13. Rep.* 3, 389 D= Od. 17, 383-384: ot Sypuocepyoi éact, pavriv » inrypa Kaxav 7» TéxTova Sovpwy, 14. Rep. 2, 381 D= Od.* 17, 485-486: ° Beoi £eivorcww €orxores dAXAodaroict, mavtoio TeACovres ércotpwpact wroAnas ° 15. Soph. 216 C = Od 17, 486: émorpwdact roAnas, 16. Afinos,’ 319 B= Od. 19, 174: évernxovra, ToAnes, 17. Minos,® 319 B= Od.” 19, 178-179: ryote 5€. dyocy, ve Kywoods peyaAn rods, évOa re Mivws évvéwpos Bacideve Avds peydAov dcapiorys. 18. Jfinos, 319 D= Od." 19, 179: évvéwpos Bacireve Ards peydAou dapiorys, 19. Rep. 4, 441 B= Od.” 20, 17: ornbos St rAngas xpadinv Avirarre pvOw: 20 (and 21). Rep. 3, 390 D, and Phacdo,® 94 D = Od." 20, 17-18: otnOos Se wAnkas Kpadinv Hvirare pvOw- rérAal 89 Kpadtn, Kai Kivrepov aXXo tor’ érAys.
1 Pol. 1, 2 (p. 1252 b 22). 2 Eth. Nic. 10, 10 (p. 1180 a 28). ® Some cod. have 3npuoupyol. 4 Cod. QV, facor; D, udvrey y' (omitting 4); HIL, uderey y'9; M, uhd\wy and vp. Sobpwr. § Cod. Q, redéOworw;s V, rerdOwor; C, rédas. 6 Cod. C, réNas. 7 Some cod. have évvevfxorra. ® Most cod. have éverfxoryra. La Roche edits ¢yv}xowra, which is demanded by the metre. 9 Many cod., Kewods. 10 Many cod., roio:; very many, x»woods; La Roche edits K»wods; A (recent hand) CL, év»éopos ; many, dapirts; N, daporeds; S, 6 dpurrts. 11 Cf. note ro. 12 Cod. G, orhéea; H, Avxlrare, yp. jrlxare. 18 A few of the best cod. have 4#relrare. 14 Cod. G, ortGea; H, hvlxare, yp. h»lrawe; many cod. omit 8); N, &\Ad.
180 George Edwin Howes.
C. PLATO AGREES WITH THE BEST MANUSCRIPTS OF HOMER, THOUGH THERE WERE OTHER HOMERIC READINGS.
There are nineteen passages in which the readings of Plato agree with those of the best manuscripts of Homer, although other Homeric manuscripts or the scholiasts or Eustathius show that there were other' readings known to the ancients.
1. Rep. 3, 393 A= /0? 1, 15-16: xai eXigoero rayras ‘Axatovs, "Arpe(Sa & pddcora Siw xoopyrope Aaay Schol. Ven. A, — ore ruvés "Arpeidas. |
Though these verses are repeated in Homer,’ the context of Plato makes it clear that he is referring to the earlier passage, for he says, — Olc8" otv, ore péxpt piv tovrov tov éxav (then comes the quotation) A€yae Te atros 5 woinrns, a statement that would not be true if it referred to //. 1, 374-375.
2. Rep. 3, 389 E= V7. 1, 225: olvoBapés, Kvvds Gupat’ Exwy, xpadiny 5° éAddoro Schol. Ven. A, — ort Znvodoros rovrov tov rérov AOEéry KEV, éws TOU vai pa TOde OKHTTpoy (VeTSe 234). 3. Crat. 428 D=J//. 1, 343 and 3,‘ 1rog: apa rpocow xai éricgw. Schol. Ven. A (3, 108), —dwo rovrov éws rod Aevooa (Verse 110)
&Derotrrat orixo tpeis. It is uncertain to which passage of Homer Plato refers.
4. Hipp. Min. 370 A= 115 9, 312-313: €xOpos yap poe xeivos duds "Aidao ruAgow, Os x’ Erepov pty xevOy evi ppeotv, dro & cir, Eustathius seems to have had the reading Bda{y, for twice in explaining verse 313 he uses the verb Bala.
1 Under the head of other readings I have included verses which, though they are preserved in our manuscripts, were rejected by any of the old critics.
2 Cod. A, Mocero; and it is so edited by La Roche.
3 Jl. 1, 374-375. 4 Cod. E, rpbow xal érlew,
5 Cod. G, éxetvos; many cod., cedar; H, xedpn.
Homeric Quotations in Plato and Aristotle. 181
5. Leg. 10, 906 E= J//. g, 500:
Plat. AocBy Te olvov xvion Hom.* Ao:By re xvion
Schol. Ven. A, — é& d\Aw Aor Bas re Kvions Te.
Though Plato has amplified the expression Aor By re into Any Te otvoy, it is clear that he had our manuscript reading of Homer, for in the Republic® he quotes the verse in full just as we have it in our Homeric manuscripts.
6. Crat. 428 C= J/14 9, 644-645:
Alay Acoyevés TeXapwne, xoipave Andy, rayra Ti po Kata Oupov dcicw pvOncacbas.
Schol. Ven. A, — & rice rev tropynpatov Seypypevus deioao.
7 (and 8). Conviv. 174 D and Protag. 348 D= JI, 10, 224:
Plat. (Conviv.) *
aw re bu’, py, Epxopevw mpd 6 Tov Plat. (Protag.) and Homer‘ ow te bu" epxopeva, Kat Te mpd 6 Tov évoncer. Schol. Ven. A,— cvvepxdpevoe dio dvri rod cuvepxopevwr.
énoe 58 xy) vonoavres TO ¥ wpooTiOéact, Kaxas.
The first four words of this verse are twice quoted by Aristotle.’
The verse is parodied once by Plato.® g. Conviv!’ 214 B= J. 11, 514: intpos yap dyvnp rodAov dvragios dAXAwv: Schol. Townl., — Zyvd8oros & od ypdder: ‘Apiorodaryns 8 dber ei. 10. ep. 3, 390 C= J//. 14, 296: dirovs AnPovre roxjas Schol. Ven. A (/7. 1, 609), —AyOovro.
1 Many cod. have AoBp ye and xvlocy. * Many cod. have «plecy.
8 2, 364 E.
* Cod. G Mor, xolpay’ dya:dy; H, 8% (for rf); EL, éeleao.
® Best codices, 8800 (for 4 rob). ® Cod. G, épxouéra; L, xal roe.
7 Pol. 3, 16 (p. 1287 b 14) and £¢th. Nic. 8, 1 (p. 1155 a 15). 8 Alesbiad. I], 140 A. ® Cod. B, larpds.
182 George Edwin Howes.
11. Rep. 3, 391 A= VV. 22, 15: éBrAads p éexacpye, Geav dAowTarTE TayTer: Schol. Ven. B, — revés 88 SoAowrare. Schol. Townl., — of 58 ypa- gover SoAowrare. Eustathius (1254, 31), — revs 5 ypadovar SoAo-
wrare.
12. Rep. 3, 386 D= //' 23, 103-104: ® Toro, 7 pa Te €or. xai civ Aidao dopo Yuy xai ddwrov, drap ppeves ovK EW wapray: Schol. Ven. A (on verse 104),—évadoerorac éx rH ‘Odvoceias 6 arixos. Schol. Townl., — ypadera: 5¢ nai wacat.
13. «lxioch. 367 D= 1? 24, 525-526:
Ws yap érexAwoavro Oeoi SecAoior Bporotory, {wav dyvupevors,
The reading, dxvvpévors, of Plato and most of the manuscripts of Homer is confirmed by Stobaeus* and Plutarch.‘ The reading, dxvupévous, given by two manuscripts of Homer, seems to have been a real variant, for it is quoted for Homer in one passage of Stobaeus.°
14. Sep.® 3, 386 C= Od." 11, 489-491:
Bovroipny x éexrdpovpos tov Oyrevépev ddAw dvdpi wap dxAnpy, [~ wy Bloros woAds ein, | 7 wacw vexvero. xaTraPOipevocow dvaccey Schol. H, — rwés 5¢ wdpovpos, 6 dxddovbos, oix eo. Eustathius (1695, 36), — 70 d& BovdAnpyy x’ Ewapoupos derryv Exe: ypadyv. 7} yap érdpovpos ... ) TpiccvAdBws rapovpos. The reading, éxdpovpos, is verified by Plato® again in the Repub- lic, where he refers to these verses of Homer.
———
1A few cod., re; C, adrdp. 2 Cod. ES, dyvupépous.
8 Floril. 98, 50 and 98, 75. * Moral. 105 C.
& Floril. 124, 14.
6 In the best codices ¢ ... ef7 is lacking. These words were probably added by somebody who knew the passage in Homer.
7 Cod. I omits verses 489-535- 8 Rep. 7, §16 D.
Homeric Quotations in Plato and Aristotle. 183
15. Gorg. 526 D= Od) 11. 369: XPvoCor OKIET pow Cxorra. Gegsorevorra réxvooir. Schol. H (on verse 568),— ro@everat pexpt Tot ws cizwy ... aow (verse 627). A part of this verse is quoted again by Plato in the Minos.* 16. Char. 161 A= Od 17, 347: aides 5° ctx dyaby cexpnoy adpi sapeivat Eustath. (1823, 29), — rpotary. 17. Leg. 10, 904 E= Od. 19. 43: auryn ro dtxy dori Gewr. a “OArpror €xovor, Eustath. (1854. 45), — reves de ypagovow, atry ro dixy dort Geo. 18. Rep. 2, 363 B= Od* 19, 109-113: 9 Bacrn0s aurpores, core Geovdis eidcxias dvéxnot, pépyor de yuia peAciva rixty 5° guxeda pA, Gadacoa bt rapeyy ifs. Schol. H,— savra, ot pHAa The omission of verse 110 I shall consider later.
19. ep. 1, 334 B= Ow. 19. 396: xAexroorg O opKy Te
Schol. L (La Roche), — yp. [re] vow re (for 6° opxy re).
D. QUOTATIONS WoVEN INTO THE TEXT.
There are twenty-two passages in which Plato. while weaving a phrase or a verse of Homer into the structure of his sentence, shows that he had Homeric readings identical with ours; or at least the quotation has nothing inconsistent with our readings.
1 Many cod. have veatesoiy; S, rexteocv.
2 319 D.
8 Cod. A, rpolary (yp. wapeivas by a recent hand); G, alSes 3° odx dyabhy dno’ Cupera: dvSpl wpolery; M, yp. ono Euperac dvdpi rpolcry.
* Very many cod. have ricre; M, rlera: Al, wdévra (for uhda); H (in margin), wdarra, ob uRjda; A (recent hand), yp. unda; many, rapexe.
184 Geerge Edwin Howes.
1. Coazs-. 133 E refers w //. 2. G1: Piat. dpa yap Ty Tor cuparos avGa Aszyovri, ovrep Apa, of xe 7 a1 arorrapevos, Hom. eyer drorrapervos. emt d¢ yArats Uxvos dvjxer. 2. Accatiad. FM. 141 D refers to /7. 2, 303: Plat. otpa: S€ ce ota arpxoow dy ema ye xOc{a re xai Epeila yeyerqpern, " Hom! y@ila re cai rpecl, or és AtAida vies “Ayaiay 3. Phaedr. 260 A refers to /7.* 2, 361: Plat. Ovroe dzoBAnroy €xos ava Sd,
Hom. ov ro: deoBAnroy é€xos éoverat, orri xy Grew:
4. icthiad. [ 132 A refers to //. 2, 547: Plat. etxpoowros yap 6 rot peyadnropos Snpyos 'Epexdéws: Hom. éynpor “EpexOnos peyadrnropos, ow zor ‘AOnvy In the passage in Plato the epic form has been changed to the Attic "Epex@éws. Such a change — common in the manuscripts of Plato — is generally due, I think, to scribes. Here, however, as Plato rather hints at the passage than quotes it, he may well have used the form "EpexOéws himself.
5. Theactet. 194 E refers either to //. 2, 851 or to /7. 16, 554: Plat.® “Oray roiny AXdaoy rov T6 Kp y, 6 by éxyverey 6 racaodos TOTS, ; Hom. (//. 2, 851) WadAaydvey 5’ iyetro TvAapeveos Adorov xijp Hom. (/7. 16,554) wdpoe Mevoiriddew [larpoxAnjos Adaiov Knp°
6. Theaetet. 183 E refers to //. 3, 172:
Plat. Tappevidns 8€ por paivera, rd Tov “Opnpov, aidotds ré poe elvas dpa Secvos re.
Hom. aidotes ré pot doar pire Exupt Secvds re- 1 Cod. G, wpanta; C, wpolt’.
2 Cod. DGH, otre. ® Cod. xéap.
ad oe
Homeric Quotations in Plato and Aristotle. I
7. Alcbiad. I. 150 D refers to //. 5, 127-128:
Plat.’ dAAa Boxed por, dowep tH Acopydae Pyoiv tHv *'AOy “Opnpos dad trav 6fOarpov apereiy tnv adxAvy, 6« eb ycyvaooKxo. yey Gedy 782 Kai avdpa,
Hom.? dyAdy 8 ad ro dx’ déPOarApav ov, 7 mpiv exjey,
opp eb yryvaokys nutv Oedv dé Kai dvdpa. Schol. Ven. A, — ore Zyvodoros ypada nptv Geov 75° avOpwrov. 8. Crat. 407 D refers to //. 5, 221-222:
Plat. epi 5¢ dAAwv oy tivwy BovrAa rpoBadrrg€ por, dppa 6d: olor EvOvdpovos trou.
Hom. ddd’ dy’ epav dxéwv émByoeo, dfpa tna,
ofoe Tpaxoe Cron, erordapevor redioro 9. Rep. 5, 468 D refers to //. 7, 321:
Plat. xai yap “Opnpos tov evdoxipnoavra év rw ToACuw vor atv Aiavra épy Sinvexéeror yepaiperOar, —
Hom. veroroey & Aiavra Stnvexéeror yepapey
10. £pist. 7, 344D refers to //. 7, 360 or //. 12, 234: Plat. é€ dpa 8y of érera, Oeoi piv ov, Bporoi S€ dpevas wAc abroi. Hom. é€ dpa 8) roe érara Oeoi ppevas wAccay atrot. The author of this Epistle has changed the second person roe the third person oi, to adapt the quotation to his purpose. I might note the interjection of the expression péy ov, Bpo- 5é, which interrupts the metre.
11. €rito 44 A refers to //. 9, 363: Plat.? "ESdxe: ris por yur)... KaAddoar pe Kai eireiv: & Swxpar npari xev tToiTaTy BOinv epiBwroyv ixano.
Hom.‘ yparl xe rpiratre Bbiny épiBwrov ixotpnv.
1 Cod. B, yurwexor.
2 Cod. L, yeryrdéoxos; A and some others, yiwwoxos; H, yinbonet ; seve yiwweexys.
® Cod. D, S6olyy; B, S0eny.
4 Cod. D, Oly» 8’.
~
186
12.
13.
George Edwin Howes.
Gorg. 485 D refers to //. 9, 441:
Plat. ras dyopas, év als py o ronrns Tos dvdpas dperperets ytyvec Oa,
Hom.! ovd" dyopéwy, va r’ avipes dpixpemées redeovar.
Conviv. 179 A refers either to //. 10, 482 or to //. 15, 262:
Plat. wore opotoy edva ro dpiotrw pice, Kai drexvas, 6 édy "“Opnpos, pévos epmvetdoat vias TOY Fpwwy Tov Oeov.
Hom. (//. 10, 482) os ddro, re 8’ €urvevoce pévos yAavaams '"AOnvn,
Hom. (//. 15,262)? ws eizav éurvevoe pévos peéya momen Aawy.
14.
15.
16.
17.
Rep. 8, 566 C refers either to //. 16, 776 or to Or. 24, 40:
Plat. ‘O d¢ 59 xpoordrys éxetvos airas SyAov 8) ore péyas peyarAworti ov Ketrat,
Hom. (//.) xetro péyas peyadwori, AeAaopévos imxrocvvdwr.
Hom. (Od.)* xetoo péyas peyarwori Acdkacpevos irrocvwdwv.
Conziv. 174 B refers to //. 17, 587-588:
Plat. soujoas yap tov Ayapéuvova dtadepdvrus dyabov avdpa ra rodeutxd, Tov 6¢ MeveAcwy parOaxov alxunrny,
Hom.‘ olov 5) MevédAaov térpecas, os TO rapos rep
padBaxds aixpntys: viv 5’ otxerat olos deipas
Rep. 3, 388 A refers to //. 18, 23-24:
Plat. TdAw 579 ‘Ounpov re SenodpeOa.. . wy roy "AydAX€a, OcGs maida. ... pydt dudorépyot yxepoiv édorvra kovev alOarXoeooay xevapevov Kax kedhadrys:
Hom. dudoréepyar 8 xepoiv éAw Koviv aifarXoeocay
xevatro Kak Kehadns, xapiey 8 yoxvve rpdcwror:
Rep. 3, 388 B refers to //. 22, 414-415:
Plat.2 pde Ipiapov éyyis Oey yeyovora Arravevovrd re Kat xuAcvOopevov Kata Kémpov. éfovopaxArAndnv dvo- pafovr dvdpa éxacrorv.
1 Cod. GL, dyopdwy; D, rerébwer.
2 Cod. L omits this verse; Vrat b A, érveuge. 3 Cod. DL, peyadorr?. 4 Cod. H, trérpecoas; A, rdpos +e.
5 Some cod. have xudtwSodperory; some cod., xbarpov.
Homeric Quotations tin Plato and Aristotle. 187
Hom.! wdvras 8 Aitdveve kvALvSdpevos Kara KOrpoy, éfovopaxAndnyv dvopdlwy dvipa éxacroy:
18. Protag. 309 A refers either to //.? 24, 348 or to Od. 10, 279: Plat. ov ov pévros ‘Opypou éraverns ei, os éby xapteorarny yByv eva Tov trnvyrov, Hom. xpGrov txnvyry, Tov rep xapteoraryn 4 By
19. Phaedr. 266 B refers to Od. 5, 193:
Plat. rovroy dkwxw xaromobe pet ixvioy wore Geoto. Hom, xapwadipws: 68° érara per iyvia Batve Oeoto.
At first sight Plato’s words seem inconsistent with those of Homer. The differences, however, are easily explicable. The expression rovrov d:wxw, while not attempting to reproduce Baive, takes the place of it. The use of ¢yvcoyv in the singular avoids a hiatus before wore and preserves the rhythm. The word wore is used to sug- gest the comparison. The epic genitive Oeoto is sufficient to show that Plato had this verse in mind.
20. ep. 7, 516 D refers to Od. 11, 489-491:
Plat. 9 76 rod “Opyjpov av wewovOéva Kai adodpa BovrdcoOau éxrdpovpov éovta Onrevépev dArAw dvdpi wrap dxrAnpe
Hom.* BovAdoiuny x’ exrdpovpos ewy Onrevépev drAAg,
dvdpi wap’ dxAnpe, © uy Bioros xoAds dn,
21. Protag. 315 D refers to Od. 11, 582: Plat.® Kai peév 69 xoi Tavrardy ye eioetdov.
Hom.
kal pyv Tavradoy eicetdov yarter drye dxovra,
1 Cod. Vrat A, é\rrdvevee.
2 Cod. S, rpwrg.
* Cod. N, rpérw; QV, xapiéoraros; A (recent hand), yp. 8 (for ro6 rep).
* For variant readings and scholia cf. above p. 182, where this passage has already been given.
5 Cod. BT, es:3or, but B with a note in the margin to indicate that there is a mistake.
® Cod. H (first hand) K, cal yu»; C, rdvraddy 7 dower; many cod., xparép’ (for xaér’).
188 George Edwin Howes.
22. Lathes 201 B refers te Od. 17, 347: Plat. roy “Opnpov Soxet por xpyvar rpoBdrAXAcobau, os Epy obx dyabny eva aide xexpnpévy dvipi rapetvar Hom.) aidws 8 otx adyady Kexpnpévy avipi rapetvac. This verse is quoted exactly in the Charmides.*
E. ATTIC FOR Epic Worps.
In a few passages the readings of the manuscripts of Plato agree with those of the manuscripts of Homer, except that a few Attic have been substituted for epic forms. As Plato in quoting these or similar passages elsewhere has sometimes given the real epic form, we must attribute these Atticisms not to him, but to the scribes.
1. Laches 191 A= Jl? 5, 223 and //. 8, 107:
Plat. «at "Opnpos wov émacvav rovs rod Aivetou Grrovs kpatrva par’ évOa xai évOa Epy abrois érioracOa SiwKecv Hoe péeBer Gar -
Hom. xpaurva par’ évOa xai dvOa Siwxépev nde péBerOat-
The confusion of such forms as d:mKxecy and dcoxépev in the manuscripts of Homer is too common to call for comment here. The form 8ca«xecv in Plato may well be attributed to a scribe.
2. Soph. 268 D= //* 6, 211 and //. 20, 241: Plat. ravrys THS yeveds TE Kat aiuaros Hom. ravrys Troe yevens re xai aiparos evyoua eva.
The form xpaéinv is similar in its declension to yevejs. The former is quoted by Plato in the Republic® from the Iliad,® and also in another passage of the Republic’ from the Odyssey.* So Plato surely was familiar with this epic or Ionic declension. Besides, in a passage of the Republic’ where the same phrase is found, many
1 Cf. above p. 183, where the variants of this verse are given in full.
2161 A. 8 Cod. G, deter Gat.
4 Cod. G, 38% (for ro); D, yereds (here, too, probably through a fault of the copyist); H Cant, cal (for re xat); L, ro: xat; Lips, Mosc. 3, xat ebyopuar aluares el vai.
§ 3, 389 E. 6 3, 225. 7 4, 441 B. 8 20, 17. 98, 547 A.
Homeric Quotations in Plato and Aristotle. 189
manuscripts show ravrys ro. yevens. The natural supposition, there- fore, is that a scribe of Plato’s manuscript has changed the epic yevens to the Attic yeveas. The same scribe or another may then have changed roc to ris, that the full Attic usage might be shown. That Plato himself was not responsible for this change from ro: to 77s, is evident from the passage in the Republic? just cited, where we find the Homeric expression ravrys roe yeveds (OF yeveys).
3. Rep. 2, 364 D= V/V. 9, 497-501:
Plat.® orperrot 5€ re Kai Deot avrot, Kai Tous pey Ovolarae Kai ebywralts dyavatar Ao By Te xvicn Te Tapatpwrac’ dvOpwra Aowopevol, Gre Key Tits UrepPyn Kai dudpry.
Hom.?® wpr«és Frop dxav: orperrot 8€ re xai Geoi abrot, TOV TEep Kai peLwv dpery Tin TE Bin Te. Kai pév rovs Ovéerot xai edxwArAys dyavyace Ao Py TE Kvioy TE waparpwrao’ dvOpwrot Acoopevor, Gre Kev THs ErepBr Kai dudpry.
To verse 500 Plato refers in the Laws.* Of the omission of verse 498 I shall speak later. Perhaps a possible explanation of Ovofaror is that a copyist has carelessly written the more common Attic word in place of the epic and poetic @vos. Then the metre may have helped to produce the poetic or old Attic ending acox. The pev rovs of Homer has suffered inversion in Plato through somebody’s carelessness. The form edxwAats may be explained as an Atticism.
4. Rep. 3, 388 C= 7. 16, 433-434 :
Plat. af af éyey, dre por Saprydova dirrarov dvipav Moip’ $96 TlarpdxAoto Mevariddao Sapjva.
Hom.' & por éyav, 5 ré por Saprydova, Pirrarov dvdpav, poip rd TlarpdxAoo Mevorriadao Sapvat.
1 8, 547 A.
2 Many cod. (verse 497), ye (for re); many cod., serol 8 (before orperrot) ; some cod., dyargot; Flor. x, edxwA7s.
3 Cod. L, rpewrol; G, ebywraisw dyapaio:; A, ebywdhio’; CEGHL, xolocy; G, raparpwrdcw; H, drepBin; L, drepBely cal dudpro. Schol. Ven. A, — éy dd\r\y holBns re xvlons re. # 10, 906 E. 8 Cod. H, muy.
190 George Edwin Howes.
We see from this same section of Plato — Republic 3, 388 C — where we read spor éyw dary, cpa Svoupwrordcaa, that the exclamation spo: was known to Plato as Homeric: so I suspect that the un-Homeric ai ai is to be attributed to some scribe, who has substituted this common Attic form.
5- dfinos 319 D = Ou. 11. 569: Plat. ypvooty oxirrpor éxorra Hom.' ypvoceor oxpetpor €xorra. Oeyucrevovra vexvocw, Here we may consider either that Plato meant merely to refer to the words of Homer. without quoting them exactly, or — as seems to me more probable — that the Attic form yproaoty is due to a scribe.’
For in the Gorgias* Plato quotes the entire verse of Homer just as our Homeric manuscripts give it. 6. Rep. 3, 387 A= Ou. 24. 6-9: Plat. as 5° ore weerepides pryw arrpov Geowerioo tpoveou roréovra, exe Ké Tis droregyow Gppabor éx xérpys. dyad rT GAAyAHOW éxovTal, ws al rerptyviaa Gu’ yecay. Hom.‘ as 5° ore rvxrepides prvye dyrpov Georerino tpiLovou roréorras, éwei K€ Tis dworeoyow GppaGor éx werpys. ava r dAAnAYoww éxovrat, ws al rerpiyvia du yecay- Hpxe 5 apa oduy Here, too, I think we may say with great probability that a scribe has changed the epic #.0av— quoted by Plato from Homer — to the doubtful Attic yecay.
F. PLato’s VARIANTS SUBSTANTIATED.
In a few passages, where Plato has given us readings different from those of the traditional Homeric text, we find the most impor- tant vanants of Plato substantiated either by some of the manv-
1 Many cod. have vextesou; S, vexteou. 7 We might compare qgaoup for gxeor (Leg. 3 681 E = /7. 20, 218). * 526 D.
dvd * Cod. A, rpdfouem; KS, dua; M, dua; ADL, dAAdAnew; M, Sresra: with fxovra: above); A, rerpvyvias.
Homeric Quotations tn Plato and Aristotle. IQ!
scripts of Homer or by scholia of these manuscripts or by ancient authors.
1. Hipp. Min. 365 A=T/. 9, 308-314:
Plat. Acoyeris Aaepriddn, roAvpyyxav’ ‘OSvoced,
xen pev by Tov pvOoy dwyAcyéws droarely,
worep 5) Kpavéw re xai ws TeA€ecOar dia,
éxOpos ydp po xeivos suas "Atdao riAyovy,
ds x’ repow pav xevOy évi dpeciv, dAdo 5 ary.
avrap tywy épéw, ws Kai reTeXcopevoy dara. Hom.? d&oyers Aacpriddy, roAvpyyav ‘OSveced,
xen pay Oy Tov pvGoy darnrAcyéws droeireiy,
3 wep 89 hpovew te xai ws TEeTEACopEVOY EoTaL,
ws py por TpY{yTe wapypevos GAAoVey aAXos.
dxOpes ydp po. Keivos Spas “Aidao riAyow,
Os xX Erepov pav KevOn evi ppeciv, Ao 8 eixry.
atrap éywv épéw ws pot Soxel elvac dpiora:
There are several points in this passage that demand discussion.
(a) eowep. We note that all the manuscripts of Plato, with one exception, read wowep, while all of the manuscripts of Homer, together with one of Plato, read # wep. It may be that in this one manuscript of Plato has been preserved his original reading 3 wep, identical with that of Homer, but there are certain facts that discredit this view. For, as we shall see presently, this passage of Plato undoubtedly shows some old variants. Besides, the one manu- script of Plato that reads 7 wep is the only one that agrees with the Homeric manuscripts in reading rereAcopévoyv éorat; so I suspect that it has been revised to agree with Homer, especially as it shows many other remarkable agreements with Homeric manuscripts. So it is quite possible that eowep is an old variant.
1 Cod. S, jrep (for wowep); S, 4 yp. (for ws, verse 310); W, rered\derOa: (for reréecGa); S, rerehecutvoy fora: (for redéecOac). A variant reading, as often, is weber (for xevp). .
2 Cod. D omits 8 (verse 309); A and many others, xparéw (for dpordw) ; C (second hand), yp. ¢pordw; D, dori (verse 310); E, rpéfore; D, rpolfyre; A Townl., raphueros; A, yp. raphuern; many cod., cebOa; H, xedpn; many cod., épéw ws xal rerehecuvor fora (verse 314).
192 George Edwin Howes.
(5) «pavée. All the manuscripts of Plato give xcpavéw, while those of Homer give xpavéw or ¢poréw. Both readings are recog- nized by Eustathius (751. 5).—«pavéw 7 ppovéu, dyes yap ypa- g@erau. <Aristarchus favored ¢povréew, for in Schol. Ven. A we read, — *Apiworapyos 9 xep Sy Spovew, and in Townl.,—ai ‘Apiorapxou dpovéw.
(c) reXéecOac diw. The best manuscripts of Plato read rede- exGac die, while the Homeric manuscripts have rereXcopevor égrac. This Homeric reading is supported, as I have said, by one manuscript of Plato. If that correctly represents Plato’s original reading, then the two authors agree and no discussion is necessary. But the weight of evidence points to a real Homeric variant, rere exOacdiw. For in verse 314 Plato's manuscripts read rereXeopevov éorat, which is supported as a variant by many manuscripts of Homer, and by a scholion in Ven. A, — év dA@ ws nai rereXeop €vov éora:. It seems unlikely that two verses so near together should have the same ending, and so I think that reAX€eaGacr dim is the right reading for Plato in verse 310 and represents an old variant of Homer, which Plato has here preserved.
(2) Of the omission of verse 311 I shall speak later.
(¢) wai rereXeopevoy Egrat (verse 314). While most of the manuscripts of Homer show poz Soxei efvac dprora, the reading of the manuscripts of Plato is xai rereXeopevoy écrai, which is supported by many manuscripts of Homer, and by the scholion of Ven. A, — év dA ws cal TereAcopevoy Eo rac.
Taking the passage as a whole, I think that the variants of Plato gain sufficient confirmation from the manuscripts and scholia of Homer, to entitle the whole quotation to our thoughtful considera- tion, as probably representing an early version of Homer.
2. Hipp. Min. 371 B=WV. 9, 650-655 :
Plat.’ od yap zpiv roAcuoio pedyoopa: aiparcerros, rpiv y_ viov Lpidporo Sacdpovos, “Exropa Stov, Mupyudovwy éxi re xAoias Kai vnas ixéoba xretvovr "Apyeous, cata re PA€Eat wupi vyas: apdi S€ piv TH “py KAtoiy Kai val pedraivy "Exropa xai pepawra payns oxyoerOa iw.
1 Cod. S, euéfa:; Vindob. suppl 7, pdéfar
Homeric Quotations in Plato and Anstotle. 193
Hom.’ ov yap xpiv rorduoo pedyoopa alparcevros, xpiv y viov Ipidpoto Saidpovos, “Exropa. Stov, Muppudovwy ext re xAtcias Kat vaas ixéobas xreivovr ’Apyeiovs, xara re opveat wupi vaas. dui 5€ roe ty ey KALoin Kat vy peAatvy ° "Exropa xai pepawra payns oxnoer Oar diw.
That the reading @Aéface, given by the manuscripts of Plato, is a real Homeric variant, is shown from the scholion in Ven. A, — ovrws opvgat “Apiorapxos. olde 5é xai ryv PACEat ypadyy. ore yp. Kai xara re dA€~at. The word pev is due perhaps to carelessness on the part of the scribe. The difference between ‘ny and épy is too slight and too common to call for comment.
3. Axioch. 367 D=T/l. 17, 446-447:
Plat. ov pév ydp ri wor éoriv oifupwrepoy dvipds wavTwy oooa Te yaiav émmrveie Te Kat épre.
Hom.” ov pév ydp ri wov dori oifupwrepov dvdpos WavTwv, 600d TE yalav émt mveias TE Kat pre.
That the reading wor’ of Plato is probably correct for that author is shown by Stobaeus,’® who quotes these very words from Plato. In another passage,‘ where he gives the same words, Stobaeus seems to be quoting directly from Homer. If that is so, he helps to show that zor may be an old variant for Homer.
4. Leg. 3, 681 E= V7. 20, 217-218: Plat. dv rediw rerddcro, rods peporwy avOpwruv, GAN’ 20° ixrwpeius GKovy rodumidaxor “Tdns. Hom.* éy redie werdAtoro, TOS pEporwy advOpurwv, GAN’ 0° txwpeias GKeov worvwidaxos “ldys. The change from @xeoyv to @xovy is undoubtedly the work of a scribe.£ That Plato’s manuscripts had @xeoy in early times is clear from the fact that Strabo’ thus quotes from Plato:
e
1 Cod. G, opltac; L, dpyelous re xaracptta:; some cod., wh or np.
2 Cod. H Vrat d, olfvpérepoy; D, Sea.
3 Floril. 98, 75. * Floril. 98, 51.
5 Cod. G, dAX 0°; E, gxovy; many cod., roduriddxov.
® We have had a similar example of contraction in xpugoty in Jfinos, 319 D= Od. 11, 569; cf. above p. 1go. 7 13, 1, 25 (C 593).
194 George Edwin Howes.
& reke rerduoro. rolis peporey érOpurev, dU’ OH erupaes oxeoy reAdrridaxor “ldys.
This quotation of Strabo shows also that Plato wrote roduacda- xov. That this was a variant of Homer we know both from the manuscripts and from Schol Ven. A, — yp. roAuriédxor.
5. Lyses, 214 A= OW. 17, 218:
Plat «ied res rée épotoy dya Geos as Tov Gpowr Hom.' ows aici rev épotoy cya Geos ws Tow Gposor.
This same verse is quoted three times by <Anistotle. In two instances’? his manuscripts agree with those of Homer, but in the third* case he gives the same reading as Plato. It looks, therefore, as if there might have been, even in early ames, a variation between
e » @iek ros and ows aier
6. Ley. 6, 777 A= Ow. 17, 322-323:
Plat.‘ pcr yep Te voor. Gyo. drapecipera: ctpvora Zax dydper, ovs dv 8% mara Soxluow Rmap Agar.
Hom. qmow yep tr dperas droaivvra: etpvora Zeis dycpos. c¥r dy pir mara dowkoy Ruap Ayow.
At first sight it would seem as if the only explanation of the great difference between the quotation and the apparent original, would be to suppose a serious lapse of memory on the part of Plato. Fortu- nately for our investigation, and fortunately for a better idea of a possible explanation of such differences generally. we have the testi mony of both Athenaeus* and Eustathius, to show that Plato is here giving an old variant, of which there is no trace im the manuscripts or schoha of Homer. Athenaeus, to be sure, is quoting from Plato, and really substantiates the correctness of the text of Plato alone; but he seems to have accepted Plato's text as a correct Homeric quotation. Besides, we read in Eustathius (1766. 55). — quor yap vT dperqs drapetpera: cipvora Zevs dviper, ovs ay by ani eqs, orep
1 Cod. CDKL, és (for the second ws): M. &
2 Eth. Lua. 7.1 (p. 12384 7) and Lact. 1, 1r (p. 13716 26). 3 Way. Morai. 2 tt (p. r205 b 10).
* Cod. A (first hand), dwreyeiJeras
#6, 204
Homeric Quotations in Plato and Aristotle. 195
onpaive: Gri rapa Ticw dAAos evpyra: car’ érepoiay ypadyy, nusov ydp Te voov dxapeiperat. With this testimony before him even La Roche, who in his text of Homer feels obliged to follow the Homeric manu- scripts, is forced to admit, — ‘“‘id vero negari non potest, Platonem et Athenaeum in Homero suo scriptum reperisse ydp re vdov d7a- petperac... dvdpayv ovs av 5y.”?
G. NEw READINGS IN PLATO.
Now we come to those quotations from Homer which, although in individual variants they receive some confirmation from other sources, in general may be said to offer readings not elsewhere sub- stantiated. These passages, however, should be viewed in the light of the points already discussed. Note, for example, the passage just dismissed. From a chance quotation of Athenaeus and from a remark of Eustathius, we are able to maintain a position that might otherwise have seemed untenable. A very easy explanation of the following passages is to assume that Plato by quoting from memory has wrongly given readings that never existed as real Homeric variants. Now, although that supposition may be true in some instances, it must not be assumed beforehand. In fact, I think that, in view of the many passages already discussed in which Plato has evidently quoted with accuracy, it is fairer to assume — unless we find strong evidence to the contrary — that Plato, whether he quoted from memory or not, has given what was in his text of Homer. In some of the following passages there is not much to be said, except to note the points of difference between the quotation and the accepted text of Homer.
1. Hipp. Min. 370 C= f7. 1, 169-171: Plat. viv 8° dys S6invd’, dra wort Adiov eorw otxad’ iney ov vyvot Kopwviot, ovdé o° diw évOd5° dripos dav ddevos xal wAotrov ddvéev. Hom.’ viv 8° duu POinv 8’, eral } wodAdD héprepov eon oixad’ ipey atv vyval Kopwricw, odé 0° div évOa5" dripos twy adevos xai wAovrov ddvgey.
1In his Adnotatio Critica on this passage. 2 A few cod. have Adevor.
196 George Edwin Howes.
2. Rep. 3. 389 E= /. 3, 8 and 4, 431: Plat. ivay péven wveiovres ‘A xaz0i, ovyn Sadores onpdvropas, Hom. (//7.3, 8) oo 8 dp icay ocyy pevea rveiovres 'Ayaroi Hom.! (/7. 4,431) ovyy Sadeores onpdvropas: dudi 5&8 race It seems to me it would be unfair to Plato’s education and scholar- ship, to assume that he has ignorantly brought together two verses that belong to different books of the Iliad. More natural is it to think that these two parts of verses were given to illustrate the point under discussion, namely the proper relation of men to their leaders. The absence of a:yy may be due either to Plato, who may have preferred not to use the same word in two successive lines of his dialogue, or, as seems to me more likely, to a scribe, who on seeing two examples of o:y# may have thought one of them a mistake of his predecessor.
3. Rep. 2,379 E= V7. 4, 84: Plat. 00d" ws rapias ppiv Zevs dya0av re xax@y re réruxrat. Hom.? Zevs. és 7’ dyOpixwy rapins woN€poro rérvxrat. This seems to represent an entirely different version from the traditional Homeric text.
4. Rep. 3, 408 A=V/7, 4, 218:
Plat. 7 ob péuvyoa, Gri xai re Mevéiew éx rov rpavparos ov 6 Ildvéapos éBadev aly’ expvfqoarr’ éxi 5’ yma appar’ éxaccor,
Hom. aly’ éxurlyoas ex’ dp’ yma ddppaxa eidas
Taove, Apart from the verbal differences of these two passages, Plato gives us an entirely different account. According to the traditional version Machaon, after sucking the blood from the wound of Mene-
laus, applies an ointment. Plato. evidently through carelessness, makes Machaon and his brother attend to the wound.
1 Cod. H, Sed:dres. 2 Cod. M, d»Opdéras; G, ranulas.
Homeric Quotations in Plato and Aristotle. 197
5. Rep. 5, 468 D= /2.' 8, 162 and /7.? 12, 311:
Plat. xai yap pets é&v re Ovoias Kai Trois rovovros mact Tos dyaBovs, xa" doov Gy dyaBoi daivwyrat, kai dvpvows Kal ols viv 59 edeyopev risnoopev, mpos S€ rovras edpacs re kal xpésowy idé wAeios dexdeoory,
Hom. py re xpéaciy re idé rAcios Serdecor:
Since these words of Plato are not metrical, we must consider that he is giving the substance of the passage and not quoting, for we find here in Plato changes, additions and subtractions, as compared with the original Homer.
6. Hipp. Min. 370 B= JV. 9, 357-363:
Plat.® avpiov ipa Ari péfas, pyoi, cal race Oeoiory, vyjoas ev vias, éxnv dAade rpoepvocu, Sen, at « €OeAygoGa Kai ai Kev ror Ta peuyAy, Rpt par “EAAyorovroy én’ iyOvdevra rAeovoas vynas éuds, év 3° dvdpas épecoduevar pepawras ef 5é xev evrAoinv Sun KAvTos ‘Evvootyasos, npari kev rpiraty BGinv épiBwrov ixoiuny. Hom.* avpiov ipa Aut pégas nai aot Oeoior, vijoas eb yas, ery dAa St rpoepioow, Gyeat, Hv €0éA\ycGa Kai al Kéy ToL Ta peunAN, pt par “EAAnorovrov éx ixOvoevra mAcovoas vnas éuds, év 5° dvipas épeoceuevar pepaoras: ef S€ Kev ebrAoinv Sin KAvros évvoaiyatos, npari xe rpirary Pbinv épiBwrov ixoluny.
It would seem as if either Plato had in his Homer ai x’, or some scribe changed yv to af x’ to make it correspond to the ai xéy of the last part of the verse.
9. Alcbiad. Il. 140 A=J/7, 10, 224:
Plat. ow re dvo oxerropéeva Hom.® ovwv re 80" épxopédva, xal re mpd o Tov évonce, 1 Cod. G, 7’ 482. 2 Cod. G, re 482. 8 Cod. S, 4» (for af x’, verse 359). 4 Cod. G, Ipds; L, af x’ €0€\p00a; L Cant, peutre; Vrat b, neutro; S, peuhds;
G, é\fororrov; E, ef rep (for el 5¢ ev); D, BOlny 8’. 5 Cod. G, épxouérd; L, xal ros.
198 George Eduin Howes.
This verse of Homer is quoted entire in the Protagoras,' and the first part of it again in the Convivium.* In those two passages Plato gives the reading of our Homeric manuscripts. Here, however, it is clear that he is parodying what was evidently a well-known verse.
8. fon 538 C=. 11, 639-640 and 11, 630:
Plat. olny xpapveig, Gyciy, éxi 8° alyaov xvq Tupoy
KYHOTL yaAKeiy: rapa 8% xpdpvoy rore Gor: Hom.‘ (/7. 11, 639 and 640)
olny Lpapreig, di 8° alycioy avi rupiv
xwyyors yadxeiy, dri 8° dAgdira Aevxa wdAvre Hom.’ (/7/. 11, 630)
xXdAxaoy xaveov, dri 52 xpopvoy, roT®@ Spor,
Here it would seem as if Plato had confused the endings of two verses —630 and 640. But our impression is modified when we read in the Republic® these words: rexpaipopa: S€, Sri abrov of vieis dy Tpoig EtpvrvA rerpepery éx olvoy Ipduvaoy dAdira woAAa ém- wacQévra xai rupoy éxévobevra. Plato is here evidently referring to this same passage of Homer, and the words dAdgira rodda ém- waco Oévra Show that he knew that éwi 8° dAgcra Acvnad wrdAvve or some similar expression belongs to the Homeric passage. So I think the confusion was made through design and not through ignorance.
9g. Jon 539 B= JZ. 12, 200-207:
Plat.’ dpus ydp ogi éwnrAOe wepnoeuevar pepawory, aierés tyurérns, ér dptorepa Aadv eépyuy, hownevra Spdxovra pépwy ovixeror réAwpoy, {wor, dr’ doraipovra: xat ovxw AUOero xdppys. Kope yap airov é€xovra xara ornOos rapa Sapyy vweis driaw, 6 5° dd Dev xe xapale dAyjoas dduvynot, péow 8° eyxaBBar spiry- airés St xAayfas Erero wvays dvtpovo.
1 348 D. 2174D. _ § Cod. Vindob. suppl. 7, xrfore from xrjory. “Cod. G, xrforn; H, xvhore; S, xrtore. 5 Cod. D, xdAneor; G, xudveor. 6 3, 405 D.
7 The best cod. have éxlsow; Vindob. suppl. 7, évxduSan'; others, éyxafBad'|r; éx Vindob. suppl. 7, wéraro.
Homeric Quotations in Plato and Aristotle. 199
Hom.! dpus yap oquv érnAGe repnodpevar pepaadory,
aierds tyeuréerys ex dpiorepa Aadv eépywy, gownevra Spaxovra pépwy dvixeror réAwpov {wov, dr’ dowaipovra: Kai ov rw AnGero xdppns. xopye yap avrov dyovra xara ornOos rapa Sapiv ywhcis dricw: 6 8° dro Bev He xapale dAynoas ddvvyo1, péow 8° evi nad BBarXr spira, abtros 5¢ xAdyfas wérero rvois dveuoo.
With regard to the form in Plato éyx«aBBad we note that it is found in two manuscripts of Homer also. It is doubtful whether ¢éxero is a true ancient variant or a scribe’s blunder.
10. Leg. 4, 706 E= //. 14, 96-102:
Plat.?
Hom.®
Os Ké\eat woAguo.o ouverractos Kai duTys
vnas evo dApous aAad’ eAKeryv, Opp ert paAdov Tpwoi pév cixra yevyra eeASopevoca’ wep éurys, qptv 8° airs GAcOpos erippéry: od yap ‘Axarol TXyTovew woAgpov vywy Grad’ éXxopevawy,
GAN’ dxorarravéovewy, épwycover St xdppns:
évOa xe om BovdAn SyrAncera, of’ dyopevecs.
Os KéAeat ToAguoLo Gvvertadros Kai duTys
vnas évocdApors dra 8° éAxépev, Op ért paddoy Tpwoi pey eixra yévyra erixparéoval rep éurys, quty 8° aids oAcOpos emppéry. ob yap 'Axaoi oxyTovow moAepov yyw dra 5° éAxopevdwy,
GAN’ drrorartavéovow, épwycover 8 xdpuns.
évOd xe ot) BovAn SyAnocera Gpxape Aady.
There are several words in this passage of Plato that need to be
noted. (a) éAqKeey.
due to scribes.
This form is one of those Atticisms that are probably
1 Cod. G, dwfiGe; G omits fxorvra and adds ré after xara; S, ¢v; H Townl., ¢yxdéB8BaN ; many cod., cduBar ; G, rérero rorl rays.
2 Cod. A, ebo@rpous.
3 Cod. C, yévocro; S, érel cparéove:; C Barocc., éwippéwoe; Lips, éwippéwe; E, 31° Axoperdwy; H, drarraréovery; many cod., drorraydovew; S, bpxape dvdpwy.
200 George Edwin Howes.
(0) €eXSopeévorct. This reading is not so pleasant to me as the Homeric ércxparéovas, for it merely repeats the idea expressed in evxra. Still, it is intelligible, and Plato may have had it in his text of Homer.
(¢) roX€pov. This form is undoubtedly a mistake of the copyist, as the sentence demands an accusative as the direct object of ox7- govacy, and not a genitive of separation, for that is supplied by vyav.
(2) of dyopevers. This is one of the common verse-endings of Homer. In the Homeric text we have another common ending SOpxape Aaa@y. We shall have to ascribe the difference in Plato’s text either to Plato or his scribes, unless we may imagine that differ- ent rhapsodists may sometimes have used different verse-endings, in case the general sense of the passage was not affected thereby. Then Plato’s phrase might represent a real tradition.
11. Rep. 8, 545 D= Jf. 16, 112-113: Plat.’ 4 BovAa, dorep “Opnpos, eixwpefa rats Movoas elrediy jpiy, owws 57 xp@rov ordats eurece, Hom.? éowere viv poe potoa ‘OdAvuma dapat’ dxovcat, ormws 9 mpwrov wip Curece vyvolv Axaay.
Here in Plato, as often in the manuscripts of Homer, we read S7ws, where the metre demands a long penult.
The introduction of the word ordocs serves to parody this well- known verse of Homer.
12. Apol. 28 C= J/7. 18, 96:
Plat. avrixa ydp ror, dno’, peO” “Exropa worpos érotpos: Hom. atrixa yap ro drecra peO” “Exropa wdrpos érotuos.
In this passage dyai, which is generally extra metrum, has appar- ently crowded out the regular word éxetra, and has taken its place in the hexameter.
13. Apol. 28 D=J//. 18, 104:
Plat.® wapa vyvot xopwviaty axOos dpovpys. Hom. ddd’ Ha rapa vnvoiv érdccov dyOos dpovpys,
1 Bas 2, évéwece. 2 Cod. Lips, drws. ® Cod. B, xopwrnlecw (the « added above »m by a second hand); D, xopwryiew.
Homeric Quotations in Plato and Aristotle. 201
Whether here Plato has used, carelessly, an epithet copwviccy, in place of the traditional érwocov, or whether he gives us an old Homeric variant, is uncertain.
14. Conviv. 195 D= //. 19, 92-93: Plat.? THS pévO’ dwradoi modes: ov yap ér ovdeos widvara, dAX’ dpa 7 ye car’ dvépav xpdara Baivet. Hom.? otAopén: ry pev O° daradoi rddes: ob yap éx ovdet widvarat, GAN dpa y ye Kar dvipadv xpdara Baives That Plato really wrote ras seems clear from Stobaeus,® who, in quoting him, gives the verse as we find it in Plato. That Plato is here preserving an old tradition is seen from a scholion in Ven. A,—ovrws (i.e. ra) “Aptorapxos, ddA 5 trys pev 8° dradoi. This reading — r9s—has otherwise entirely vanished from the manuscripts of Homer with the exception of one manuscript, in which it is preserved, more perhaps through carelessness than actual tradition. Stobaeus shows us, too, that Plato wrote ovdeos. As both genitive and dative seem admissible here, Plato may well be preserving an old variant in the form ov8eos.
15. Rep. 3, 388 C= 7, 22, 168-169:
Plat. wore & roro Gava f irov avdpa Suwxdpevov wepi dary oPOarpoiory Spapa. pov 8° drAopvperar Frop:
Hom. @ root, 7 pirov dvdpa Siwxopevoy wepi retxos 6POurpotory Spwpac: epov 5° cAopUpera Hrop
dorv may be an old variant.
16. Crat. 392 E=//. 22, 507:
Plat. olos yap ogiy Epvro woAcy Kai Teiyea paxpd. Hom.‘ olos yap ogy épvoo wuAas xat reixen poxpd.
To adapt the verse to his sentence Plato changed the verb from second person to third, without any violence to the metre. To whom the change of xvAas to wroAcy is due is uncertain. The words are so similar that one might have been substituted for the other at almost any time.
1 Cod. T, ri\vara: (from wl8vara:); B, rndvaraz. 2 Cod. S, rs; G, Balvos. 8 Floril. 63, 36. * Cod. D, ydp uy.
202 George Edwin Howes.
17- fon 537 A=. 23, 335-340: Plat. KAcvOqvac 5é, pyot, cai airos évééore evi didpw RK ér dpurrepa rouv: drap rov Sefiov Umrrov xévoat GuoxAnoas, elfai ré of Hvia xepoiv. év vicoy Sé rot trmos dpiorepos éyxpinPOynrea, ws dy To. TAnpYy ye Sodocerat dxpov ixérOat KUKAOU otntoto: Aiov 8° dAdacOar eravpety.
Hom.’ atros bé xAcvOqvac €vrAéxryw évi didpw > 7 » * » S ~ 98 \ . @ NK ér apwrrepa touv: arap rov Sefidov Urrov
Ld ¢ , td ee 2 ’
Kévoat GpoxAjoas, elfai ré of Avia xeporv. év vioon 5€ rou irmos apiorepos eyxpinPOyre, ws dy Tou rAnpuvyn ye Soacoerat axpov ixécOax KUKXov moyroto: AiGov 8° dAdacGat eravpeiy,
It is uncertain whether the difference in the order of the first few words in Plato is due to him or not. For évéorw, however, we have a twofold testimony. In the first place it is given by one of the manuscripts of Homer. Besides, Xenophon*—though giving a different case of the word and adapting the sentence to the construc- tion of his own — gives this quotation:
avrov d¢ cAwOyva evérrov émi dippov
mK é€r dpiorepa rouv, adrap tov defcov Urirov
Kévoat SpoxAnoavr’ eifat ré of Avia xepai. But while he gives some support to évféorqw, he does not offer any for the order of words as given by Plato.
18. Rep. 3, 388 A= //, 24, 10-12: Plat. dA 54 ‘Opnpov re Senodueba xai ray dAAwv roinToy jy moveiv AxtAAéa, Oeds rratda, GAAor’ éri wAevpas xaraxeipevov, aAXore 8° abre Urriov, aAAore 5€ rpyvy, Tore 8° dpOov dvacrdyra wrwiLovr drvovr éri Ov adds drpvyéroio,
1 Cod. S, édvrdéxry; Vindob. suppl. 7, ae
2 Cod. D, éviéorw é; D, wom; many cod., éyxppOtrw; L Lips, rdsu; E, ddbooerat.
3 Conuiv. 4, 6.
Homeric Quotations 1n Plato and Anistotle. 203
Hom.' ddAdor’ éwi xrevpas xaraxeipevos, ddXore 5° abre Urrios, dAXore 5¢ mpnvys: roré 5° épOds dvacras Sevevern ddvwy rapa Oiv adds. ot5€ piv Qos
Since Plato is adapting these verses to the structure of his own sentences, he uses the accusatives xaraxeipevov, Urriov, mpnva, opOoyv, dvagravra and dAvovr in place of the corresponding nomi- natives. With the exception of dvaordavra, however, they do not violate the metre.
If rAwié{ovr in the manuscripts of Plato is what he really wrote, either he had rAwiLeox’ in his Homeric text, or else he introduced it for a parody. If the reading is corrupt, the suggestion of Heyne’ is a good one. He thinks that Plato does not offer a new reading, but is merely interpreting the verse of Homer, and that he wrote wpwilovr, which included the idea suggested in the last words of the verse, o05é perv Has. A scribe to whom the verb rpwi{w was unknown might easily have changed it to rAwi{w, a verb that does occur a few times.
The word drpvyéroso may have been added by Plato, to com- plete the verse metrically.
19. Jon 538 D=WJ//. 24, 80-82: Plat? 4 58 porvPdairy ixérAn és Bucoov Txavery, 9 TE KaT dypavAoio Boos Képas éppepavia épxera: wpnoryos per ixOior rpa pépovoa: Hom.‘ 9 dé podvBdaivy ixéAy és Bvocdv dpovcey, y Te xar dypavAow Boos xépas éu Be Bavia épxerar wpnorpoww ex iyGior xipa pepovea. (a) ixavev. Though this is a weaker word than dpovcey, it may, for all that, have stood in Plato’s Homer. (6) éppepavia. This is a form that might easily have been changed by ascribe from éuBeBavia; but we find that it was really
1 Cod. D (verse 11) omits 8; Syr (verse 11), 8); Vrat d, 3'ad; S, devebacn’.
2 Variae Lectiones et Observationes in Iliadem, vol. viii., p. 585.
8 Cod. Vindob. suppl. 7, rv0uér’ (for Buoody); t, rvdue¥ (in margin); S, xfjpa (for wijua).
* Many cod., Buddy; S, Bnbdy; D, éupenavia; G, dufeBaviaas.
204 George Edwin Howes.
recognized as a distinct reading. For it is so given in one manu- script of Homer and is referred to in a scholion of Ven. A, — éy Aw éppepavta.
(¢) xypa. This word, too, though it is considerably different in form from «pa and is not found in the manuscripts of Homer, is a real ancient variant. Proof of this fact is accidentally preserved for us by a scholion in Ven. A, — éva ray xara rodkas éw iyo. whpa $épovoa: droxov yap éx ixOvwv xypa A€yav. If it were not for this one scholion, we should be obliged to say of this reading, as of others, that it may represent a real variant, or it may be due to Plato’s carelessness.
20. Hep. 2, 379 C=/7. 24, §27-532:
Plat. Otx« dpa, Hv 5° éya, drodexréoy ovre ‘Opyjpov ovr dAXAov WounTrou Tavryv Thy apaptiay wepi Tous Oeois dvonrus dpapravovros Kai A€yovros, ws Socoi wiPor Karaxeiarar éy Acds ovde
kynpav éerreror 6 piv €cOAGyv, atrap 6 Serdar: Kai @ piv Gy pitas 6 Zets 0 dudorépey, GAXNorE péy TE KaK@ GO ye KUpera, GArAore 8° écMAa: @ 5° &y py, dX’ dxpara ra érepa, rov 5¢ xaxy BovBpworis éxi yOova Siav €Aavvet: Hom.? 527 Soi ydp re xi8or xaraxeiara: év Atos ovde 528 Swpwy ola Sidwar, xaxdy, Erepos St Edwy. 529 @ pe « dupikas Swy Zeis reprixépavvos, 530 Adore pey re xax@ GO ye KUpera, dAXore 8 éo0AG 531 @ O€ xe Tay Avypayv Sun, AwBnrov EOyxe- 532 Kat € xaxy BovBpworis eri xPova Sav éAawvei,
We feel pretty sure that Plato’s text has been transmitted to us correctly, for Eusebius,* in quoting the passage ws Sout... diay éAavve. from Plato, gives the same manuscript readings, except dudo- répwv So (for do dudorépwv) and ye (for re, verse 530). This might
1 Many cod., xaypltas; a few cod., dupltas or dvayiltas.
2 A papyrus fragment, xaraxeabe; cod. L, 3° (verse 528); cod. Townl. omits verse 528; L, x’ duttas; papyrus, dupetas; D (verse 531), 8oln; G Flor, BodSpwars.
8 Pracp. Ev. 13, 3 (p- 643).
Homeric Quotations in Plato and Anistotle. 205
appear to be one of the cases where Plato is quoting from memory. For as part of the passage is metrical and part is in Plato's own words, it looks as if he knew the exact words of only a part of the Homeric passage. If we admit this view, we must maintain that in verse 528, which appears so different in Plato, he thought he was quoting exactly. Then either this verse represents real old variants, or else Plato was greatly mistaken in what he thought he knew.
21. Rep. 4, 424 B= Od. 1, 352: Plat. dAX' ws oldv re partiora dvdAdrrev, poBovpévous, Stay ris A€yy, ws THY dodnv paAXAov éewippovdovow dvOpwmroi, gris decdovrecae vewrary dugiréAnrat,
Hom.' 7 ris dxovovrecat vewrarn duguréAnra.
In substituting decddvreac. for dxovdvreoa:, Plato is evidently parodying the verse somewhat.
22. Conviv. 220C= Od. 4, 242: Plat. ofoy 8° ad 10d" épege xai érAy xaprepos dvyp Hom. dAX’ olow 108° dpege xal érAn xaprepos dvip
23. Rep. 3, 390 A= Od. 9, 8-10:
Plat. wapamActat wou Tpawefat girov xai kpawy, webu 5° éx xpyripos apivoccwy oivoxoos popeyar xai eyxein Serderct,
Hom.® jpevor éfeiys, rapa 8¢ rAVOwor rpdwefa girov xal kpea@v, peOv 5° éx xpyripos ddvoccwv oivoxdos opéyat Kai éyxein Serdeoar:
The word waparActaz if not a real drag as I suspect it is, is surely very uncommon and would not come from a copyist, except by a most egregious blunder; nor would it result from ‘lapse of memory,’ it seems to me. It is more likely that the word was in Plato’s Homer.
1 Cod. E, 4 xev. 2 All cod. have aédrd; B, Eppege (with the pp where an erasure has been made). ® Cod. PS, rA#Gover; AMV, 82 (for 38” éx).
20H George Edarz Heaes.
zy Mew r2c A= Od 12. 453: Pliz- ea éoy “Opspos é reir refvemcw riv Taperiay civa, Acpew Tex aires. eri cuss rerrera tee éy “A:dor, Toi d& cum @overc. Hom.* om rerrectin: ret 6 cum aor.
In codex L of Homer, :ost as 2 the manuscripts of Plato, a scribe, thinking that rec shoud agree with eacai, bis changed it to the feminine form In the manusezipts of Piato the scribe has gone one step farther and given the Arce form «%
zs. Xp. 3. 336 D= VE 12. 495:
Pilar? ow rervcote. ra & cau &vaed- Hom.* ow resvrchun: rot & om doovecou.
Here. again. im Placo we hive 2 change sumilir to that m the pre- ceding passage. but the scribe has chanved merely the gender, and has not given the Atte form
26. fxrwer. 303 A= Gul rs, 242-245:
Plat. ror d° "Ap@unguoy mi Omri: TOY wen <a Gila Zevs Tr’ acywyos au Araliav rarroty @iloryr - ocd ixero yypacs order. Hom? oy rex op dcle Zevs t afywyos ani “Aralier wrarroiqy @ileryr - ocd icero yypaos ocdor,
In Homer the reiative ev his its antecedent in the preceding verse. Plato may well have preferred to make his sentence more complete by writing rev.
Plato may have had ravrocy in his Homeric text, but a simple
explanation of the form would be thar a scribe, supposing that @iAoryr was in the dative case, chanzed wasroiyr to rarracy.®
1 Schanz, in spite of ai im che best codices, has followed Cobet im editing rel.
2Cok L. rai: D. rw.
8 A few cod. rei.
* Cod. L, rei; D, ra.
§ Cod N, rév: D, devs tfor Zeist: L. cai cdyey: yp. cai ded ee.
* The best codices cf Stobaeus. who (Zurn. o3. 75) quotes this passage from Pilato. give the verses just as we have them in Homer.
Homeric Quotations in Plato and Aristotle. 207
27. fon 539 A= Od. 20, 351-357: Plat! Sacudvior, ri xaxdv rdde waoxere; vuxti piv dpewv | eiAvara: xepadai re rpdowrd Te véepGe Te yvia, olpwyn St Sébye, Seddxpuvras 5¢ zaperai: eidcAwv re rAdov rpdOupov, mAcin 5¢ Kai aiAy icnevov epeBoode Sd Copov: Hédros 8¢ otpavo éfarddwAc, xaxy 5° erdédpopev dxAvs ° Hom.’ d SetAod, ri xaxdv rode waoyere; vuxri piv dpéwy eiAvara: xepadal re mpoowmd re véepbe re yodva. oipwyy 52 Sede, Seddxpuvras 5¢ rapeai, alpar: 8° épfddarat rotxo. kadai re peoddpat eidaAwv 5% wrAéov xpoOupoy, mAcin 5é Kal avAy, ieuevwr "EpeBos 58 tra Cogov: WéeAros Se otpavod éfarcAwre, xaxy 5° éridédpopev dyAvs.
The expression d S8eAo¢ is so common in Homer that it must have been well known to Plato. Sacuoveos, too, though generally found in the singular, occurs in Odyssey 4, 774 in the plural. It is possible that it was a variant here in Odyssey 20, 351, though it may be a mistake of Plato’s.
yvita is a good Homeric word and may be a variant for this verse.
On the omission by Plato of verse 354 I shall speak in the next section.
The variants re and 5€ need no comment, for these words are often confused in manuscripts.
H. OMISSIONS IN PLATO.
There are four passages in which Plato in quoting from Homer has omitted a verse. These are: Hippias Minor 365 A*= Iliad 9, 308- 314; Republic 2, 364 D‘= Iliad 9, 497-501; Republic 2, 363 B’= Odyssey 19, 109-113; Ion 539 A*= Odyssey 20, 351-357.
These verses are, in a way, similar. For no one of them is neces- sary to the general sense of the passage in which it stands. The
1 Cod. W, duGs; best codices, 8¢43a:.
2 Cod. D, w Serol; KS, épé8arac; CDKQS, wrAéwy; A, éridéSpaper.
3 Verse 311 is omitted. 6 Verse r10 is omitted. 4 Verse 498 is omitted. © Verse 354 is omitted.
=c8 iuerz Scum Brwes.
exsiricim :f ht omssunm & at Kk ki cerum Plato may not fave het Spe vers pons ac. i:¢ we remember that in the Finders ese Surmenr roe tere. Ent passciy two, known in our ERENERIOCCS 2f SSiomer. ie mmm 0 Fors mar have omitted She verses ether cursos re Bocas 7. OF ae omission of them mrs te Sm Tr oes. cr wt pice =2e = te manuscnpts of Kiomer_ i, Sint verses ice echsmiy cmexaed by sombes.
= lM SoH Tas Fake
Fthere 36 m2 mipSkEre or Foic3 che: pam xs cargeeness demands. axvemore. Tt cs at Abctepdes CL“ it pecpeets to be from Homer, Ee OIC TESS Se TET polcaore Se scistznoe of five verses, only cmt of BER S fremct = cor Ecmerx mimascrcges. The passage in Plas. mx 2€ WR SS mesh. SS AS 25S: G_gew yap rors Tpeas eran recucrnrs Ges Dizerur. colrerrss Gertwa tas -
vq & cure & 2m relun cS qurs Owes etpzes acu Was cys Bor. fees escrors dureertn. etd cHelaz- pals pap rou Erqyfero Lues pg
The passage fas been thos restored 2 the Fiad-
i epdow 6° dttzrareem relperees darrentas. } arerye 6 ex reher artes Ocpor otparer acw
[pocas- mas 5° ov mr Get pacapes dzreovra, cai Lapos ari Lads erapelio [papcno}.
As the Alcibiades IT. belongs to the list of works that are probably spurious, we should not attribute much weizht to this passage as one of Plato's. But. as the work is probably of high antiquity, even if Plato did not write it, the passage will serve to illustrate once more the fact that there are old variants of Homer. and even whole verses, preserved for us by ancient authors alone.
1 149 D. 2 Cod. CD (second hand) EGHL, «slewys.
Homeric Quotations in Plato and Aristotle. 209
CONCLUSION WITH REGARD TO PLATO’S QUOTATIONS.
As this completes the list of passages in which Plato quotes from Homer, it may be well to emphasize the lesson of these quotations. Some scholars have thought that the differences between the read- ings of our Homeric manuscripts and those of Plato are best and most easily explained by assuming that Plato quoted from memory, and that his memory was very faulty. Whether, in general, Plato quoted from memory or not, is still a doubtful point. I am willing to admit that that supposition offers the best explanation of some few passages. If sometimes he quoted from memory and sometimes looked up the passage, we have no means of ascertaining which quotations are the result of one method and which are the result of the other. For if I have not shown that apparent mistakes cannot be taken as the test of that question, my paper has been in vain. Besides, as I have already stated, there is nothing at all inconsistent in quotation from memory and correctness. Let me say, as a kind of summary, that in general these are the reasons that influence me to believe that, whether he quotes from memory or not, Plato’s quotations are to be weighed very carefully, and not rejected merely because at variance with traditional readings. (1) Very many verses as quoted by him agree with our traditional text. (2) Many verses evidently owe their variants to careless copyists, who, in many instances, have changed the epic to the Attic form. That Plato is not responsible for these Atticisms is often shown by other passages, in which the same verses are given just as our Homeric manuscripts have them. (3) Some variants are supported either by Homeric manuscripts, or by scholia, or by ancient authors. Of scholia and authors we know that only a small percentage have come down to us. If more were extant, we should undoubtedly receive confirma- tion for still more of these variants of Plato. (4) Papyrus fragments in general, and the Flinders Petrie fragment in particular, show that ancient manuscripts had many readings far different from those that have come down to us from other sources. We should not be sur- prised, therefore, to find that Plato, or any other ancient author, pre- sents us with many variant readings. In fact, we should be properly surprised if they did scot show these variants. We might then
210 George Edwin Howes.
reasonably suspect that the readings they gave had been tampered with, to adapt chem to our later tradinon. Therefore I feel con- vineed, that to the exstence of Plato’s manuscripts we are indebted for a great many ancient Homenic readings that otherwise would have been lost to us.
Arsstute s Qavtatwns from Homer.
Let me pass on at once to the Homeric passages quoted by Anstode.
A. No Varrasts.
First I shall list those passages in which the manuscripts of Aristotle — so far as coilated by Bekker — show an entire agreement with those of Homer. with no variants for either author. There are twenty-eight of these passages.
ryand 2). Ket. 3, 14 (Pp. 1445 a 15) and Feet. 1g (p. 1456b 16) =f:
waquey aude Gen, 3. Foes, 23 yp. rg6r a red= 2 2 So: Clpqas méy perros: 4 De Mdunio 6 \p. 397 b 26)= Li. 5. 499 and /7. 5,754: . . . 12 fol. i, 12 (po 1239 b 13} = 77. 1. 5.94 and often: 6. Poet. 23 \p. 19613 16) = /Z 2, 1-2: Gos pay pa Gece re ani drapes edoy warren: Here, as often, Aristotle onnits part of a verse. 7. Sta. Nee. S 13 (po 116244 14) /7 2, 243 and often: “Aypepeurowe rapee Asaw S. Poet. 21 (p 21457 Bb wr) = Lh 2, 272: ¥ oy pepe “Odwrres érOAa depyor: go Rhet 1. 6 (:p. 1363.4 6) = 1. 2, 298:
‘gs
10.
II.
12.
13.
14.
Homeric Quotations in Plato and Aristotle.
Pol, 3, 16 (p. 1287 b 14) =. 2, 372: Tovovro: Séxa por ovpppadpoves. Frag. 143 (p. 1502 b 8) = //. 3, 298-300: Zed xvdiore peyuore Koi dOdvaror Oeoi ddAor, GrTorEepo, WPOTEpOL Urép GpKia wHpHveEtay, woe od’ dyxéparos xapddis peor ws dd¢ olvos. Probl. 9, 9 (p. 890b 9) = /7. 5, 75: yuxpov 5° Be xarxov ddovery, Poet. 21 (p. 1458 a 7)=T/. 5, 393: Sefirepov xara pafov Eth, Nie. §, 11 (p. 1136b 10) = /7/. 6, 236: Xpvoea xaAKelwv, éxaropBor évveaBoiwy,
The first two words are given thus by Plato’ also.
15.
16.
17.
18.
19.
20.
21.
Frag. 151 (p. 1503 b 26) = //. 7, 111-112: pnd’ er’ e epidos ced dueivove puri payerGar "Exropt
2II
Hist. An. 6, 21 (p. §75b 5) =. 7, 31§ and Od.? 19, 420:
dpgeva wevTacrypov Frag. 108 (p. 1495 b 10) = //. 9, 175 and often:
xovpo. O& Kpyrnpas éxecrepavro woroto Pol. 2, 7 (p. 126742 1)=J/7. 9, 319:
dv 5: in rysy que xaxds 75% Kai dx Odds. Rhet. 3, 11 (p. 1411 b 35) =f. 11, §74:
dy yain torayro AAaopeva xpods doa, Eth, Nic. 7,7 (p. 1149 b 16) = /7. 14, 214:
KeoTov ipayra
Rhet. 2, 21 (p. 1395 a 15) = 7. 18, 309:
Evvis "EvvdAuos,
1 Conviv. 219 A. 2 In the passage in the Odyssey we find a variant: cod. A, wlova yp. Apoeva.
212 George Edwin Howes.
22. Foet. 25 (p. 1461 a 28) =J/7. 21, $92:
xynpls veorevxrou KacoiTépoo,
23. Foe. 25 (p. 1461 a 23) =/7. 23, 328:
Td péy ov xararvGera: opBpg.
24. Soph. Elench.* 4 (p. 166b 4) = //. 23, 328:
TO pay ov xararvOera: oufpw.
A point of discussion among the ancients was whether in this passage of the Iliad ov was ov, ‘where,’ or ov, ‘not.’ Into this discus- sion I cannot go. :
25. Kher. 3, 14 (p. 1415 a 16) = Od. 1, 1:
Gyépa po. dvvere pore, 26. Frag. 165 (p. 1505 b 25)= Ou. 6, 6: ot odeas otvtaKorTo.
27. Rhet. 3, 14 (p. 1415 b 26) = Od. 6, 327:
Sos ps’ es Dainxas dAov AGeiy 9d" erceLvor,
28. De Anima 3, 3 (p. 427 2 26) = Ou. 18, 136:
Tolos yap yoos éorir. B. SLIGHT VARIANTS.
There are thirty-nine passages in which the variants of the manu- scripts of both Aristotle and Homer are few and slight, and undoubt- edly due to the carelessness of scribes.
1. Kher. 2, 2 (p. 1379 a 5) = //. 1, 82:
Ar® dda ye xui peromober €xa xéror: Hom.‘ dAAd re xui perorobey éxe xorov, 6ppa reAévoy,
2. Rhet? 1, 6 (p. 1362 b 35) = 44." 1, 255:
H xey ynOyou TMpcapos.
1 Cod. Ac Be, od.
2So edited by Bekker, although all the codices of Aristotle and of Homer have od. § Cod. YDZbAs, re; Q, Sppa redéooy (after xbroz).
4 Several cod., ye; S, dAN aye. .
® Cod. Yb, y964e9. © Cod. C, yatteat; H, r9bhedl.
Homeric Quotations in Plato and Anstotle. 213
3. Rhet.' 2, 2 (p. 1378 b 32) =/7. 1, 356: aripnoev: édXwv yap éxe yépas abros drovpas 4. Metaphys* 11, 10 (p. 1076.a 4)= JV. 2, 204: ovx dyaQov rodvKoipavin: els Koipavos tore. 5. Frag.* 143 (p. 1502 b 16) = //4 4, 65-67: EXOciv és Tpdwy xai "Axaav pvdromw aivyy, wepav 5° ws xev Tpwes trepxidaytas “Axa.ovs dpfwo. mporepor trip apxua SyAnoac Gat. 6. Rhet.® 3, 11 (p. 1411 b 35) = V/V. 4, 126: erurrécOar peveaivuv, 9. Frag. 13 (p. 14764 17) =J/1.° 4, 297-298 : imrjas pey mpara ov Urmowww Kal oxerpuy, aeLous 8° e€orev, 8. Eth. Mic. 3, 11 (p. 1116.a 25) = J. 8, 148-149: Ar.’ “Exrwp ydp more dyca évi Tpweoo’ dyopevuy, Tvdeidys tn’ dueto. Hom.’ “Exrup ydp wore dyca évi Tpweco’ dyopevwr: Tvdeidys tr’ éueio PoBevpevos Txero vijas. g. Pol’ 1, 2 (p. 1253a 5)=J//. 9, 63: ddpytwp, &éuioros, dvérrios. 10. Rhet.” 3, 9 (p. 1410a 29) = //." 9, 526: Swpnrot r’ éréAovro mapdppyroi tr’ éréeoon: 1 Cod. Ac omits dwrodpas. 2 Cod. E (also T in the margin) adds éerw. 3 Cod. E, rp@as; B, rpdocs. 4 Cod. O, eis; L omits 3’.
§ Cod. QYDZD), ériwraoGas.
® Cod. Vrat. b,c, Mosc. 1. 3, wpwriucra; M, rp@rov; CGH, éfbrio ber; E, dtériobe; N, &uwwrieGev.
7 Cod. Mb adds doBotpeves.
5 Cod. E, 8% (for yép); D, rpdecow; D, rvdeldns 3°; E, époto.
® Cod. Q, dpiprwp; SbT»,. dpitrup.
10 Cod. Ac, éré\orra; Ac, 3° (for the second 1’).
11 Cod. DH, rapapnrol.
214 George Edwin Howes.
rr, Fiwt.) 25 (p. 14614 12) = //. 10, 316: os Sy roc ados péy éyy naxds,
12. Hist. 10." 9, 44 (p. 629 b 22) =//. 11, §54 and //. 17, 663: omopera Te Serai, Tas Te Tpel Ccorperos TeEp,
13. Ra&et. 2, 21 (p. 1395 a 13) = /23 12, 243:
cis oleres dpicros durveoGa wepi ratpys, 14. A&et, 3. rr cp. rgirb 34) = 7/7. 13, 587: Ar. éxtar cords,
Hom.’ Gespyxos ywaror, dd 3° Ewraro wixpos dusts.
In the passage of the Rhetoric from which these words are taken Aristotle is giving examples of vividness in narration. Naturally, he quotes that part of the phrase that is especially pertinent.
1s. AKéet® 3.11 Up rgrza 7) =—/2" 13, 799:
avpre, @algnoerra: rpo mer fr GAA, atrap ér dA\a: 16, Hist. ta. g, 12 (p. 615 b 1r0)= /7. 14, 291:
xXeAniOa mucAQoxover Geor, ardpes SE xvpurdcy. 17. Rhet*® 311 (Pp. rgrz2a ry—= 40 15, 542:
aixay St oreproo decovro papewce. 18 Rael. 1. 11 (p. 1370b 11) = 7. 18, 109:
Sore WOAY yArcioy mOuros aaralafonaoo: 19. Radet™ 2, 2 (p. 1378b 5) = /7. 18, 109-110:
Gore WOAT yArcioy ROUuTOs caralafonarne
drépar ¢y oryGermy deferaz. 20. J¥et.™ 25 (p. 1461a 30) = //. 20, 234:
Adi oivoyoevery,
1 Cod. AS os Gre; BO at Ad rer: NA os (with 3§ re: omitted); AS ed ge.
2 Cod. Da, canadvas (with a gap left for the rest of the verse); Aa, Sacres; Ca, 3a8a:; PE, rpeis; Aa Saivas reer.
8 Cod. ES, dadvasda:. * Cod. PAa Ds Ea, atfiurdes.
* Cod. Q, dperrés. ® Cod. AS, penbuve.
§ Cod. L, &4 (for dé). % Cod. Yd, ovre.
* Cod. QYOZD, wads. 11 Cod. Zd, sere; AS, ergberuw.
7 Cod. L, gargedwrra. 3 Cod As, elsexecta.
Homeric Quotations in Plato and Aristotle. 215
21. Rhet. 1, 11 (p. 1370b 28) = /7? 23, 108 and Ou. 4, 183: ws daro, roto. dé raow id’ ipepov apoe yoo. 22. Rhet* 2, 3 (p. 1380b 29) = //4 24, 54: xwgnv yap 59 yaiav decile: peveaivwv. 23. Eth. Nic 7,1 (p. 1145 a 21) = 11° 24, 258-259: ovde dyxec dvdpos ye Ovyrod wdss Eppevar dAAG Deoio. 24. De Mundo’ 6 (p. 401a 4) = Od 5, 64: KAnOpy T° alyepos Te Kai ebwdys KuTdpioaos, 25. Frag. 162 (p. 1505 a 26) = Ou.’ 5, 93: xépacoe 5é véxrap épvOpdv 26. Frag. 165 (p. 1505 b 20) = Od.” 6, 4 and 8: ot xpiv ev wor évatov éd\gdycrawy. 27. De Mundo" 6 (p. 401 a 7) = Od.” 7, 115 and 11,” 589: Oxvat Kai Porat Kai pyAéa dyAaoKxaprot, 28. De Mundo 6 (p. 401 a 1) = Od. 7, 116: Ar.* = ovxai re yAvKepai xai édaiat,
Hom. ovxéa re yAvxepai xai éXatar rnAcOdwoa:.
1 Cod. QZb, 颒. 2 Cod. E, 颒.
® Cod. Q, deuign; the reading in AC is uncertain; cod. Q, uerdlw».
4 Cod. CD omit 84; H, Baiar ; H, pevtalrwy.
5 Cod. Mb, od yap; LbOb, ovdé yap.
® Papyr. (first hand), ovre, expueve ; cod. D, Orroto.
7 Cod. P (recent hand), x\epal; O omits re.
® Cod. LO, «relpy.
® Cod. ACER, xépace. e
10 Cod. DLMP, dAdurrdwy; B, dudiordwe.
11 Cod. O, &yx»m (but it has been corrected).
12 Cod. S omits this verse.
18 Though most cod. give 6x»a in these two passages of the Odyssey, La Roche edits, in both places, &yx»a:.
14 Cod. Q, ovaéas.
16 Cod. S omits the verse; many cod. have ovxat or cuxal; I, rndebdwoa; Schol. B.H.Q. (Od. 9, 425), rnd\eOdoveas.
216 George Edwin Howes.
29. Fol. 8, 3 (p. 1338 a 29) = Od.' g, 7-8: Sarrupoves 8° dva Swpar’ dxovaLwvra: dodo Hpevor éLeins. 30. Fol. 1, 2 (p. 1252 b 22) = Od.* 9, 114-115: Oemorevac St Exarrros wadwv 75 ddoxwv.
Plato* in quoting more fully from the same passage of the Odyssey ‘ gives the same text. The passage is referred to in the Nicomachean Ethics.’
31. Rhet.® 2, 3 (p. 1380b 23) = Od. g, 504:
gacba "OSvecna rrodkixophrov, 32. Hist. An.” 6, 21 (p. 575 b 6) = Od.§ 10, 19: Boos évvenpovo. 33. De Mir. Aus.® 105 (p. 839 b 33) = Od.” 12, 67-68: dAAd 6” Gpod wivoxds Te veWv Kal Copara dwrav xupad” dros hopéover wupds 7” dAoot OveAXat. 34. Eth. Nic." 2, 9 (p. 1109 a 32) = Od. 12, 219-220: Tovrou pév Kamrvov Kai Kiparos éxros depye By mistake Aristotle assigns these verses to Calypso, though they contain the advice of Circe and were uttered to his companion by
Odysseus. 35. Rhket. 1, 11 (p. 1371 b 16) = Od. 17, 218: ws alet Tov Gpuorov, 36. Eth. Lud. 7,1 (p. 1235 a 7) = Od. 17, 218:
€ , Qa @ » Q e a, @ ws alei Tov Guorov dyet Geds ws Tov GpoLov’
6 Cod. QYDZb, wrroNsropGos. 7 Cod. P, évvedrepos.
8 Cod. L, évpedpow.
#9, 112-115. ® Cod. Ba, dAnds, OveAat.
8 10, 10 (p. 11804 28). 10 Cod. QV, OvedXa.
11 Cod. MbNb, 7d ro6; Lb Ob, ws roi; Kb omits pes.
12 Cod. YbZb, rd; Q, 7d dpolw plrop rd. ds
18 Cod. CDKL, és (for the second ws); M, és.
1 Many cod., dxoudfovrac. 32 Cod K, ral8wy 7. 3 Ler. 3, 680 B.
Homeric Quotations in Plato and Aristotle. 217
Here the manuscripts of Aristotle agree with those of Homer, and disagree with those of Plato who, in the Lysis,’ quotes the same Homeric verse. In one passage,* however, the manuscripts of Aristotle show the same readings as those given in the manuscripts of Plato. In the Nicomachean Ethics * we find the verse of Homer referred to, but the first part of the verse is not quoted.
37. Eth. Nic! 4, 4 (p. 1122 a 27) = Od. 17, 420 and 19, 76°: modAdx Sooxov dAnTy * 38. Probl. 10, 36 (p. 894b 34) = Od. 20, 71: pyxos 8° érop’ “Apreus dyvy, 39. Khel.” 1,7 (p. 1365 a 30) = Od. 22, 347: avrodivdaxros 8° cipi.
C. AGREEMENT WITH THE BEST MANUSCRIPTS OF HOMER.
Now let me give those passages in which the manuscripts of Aristotle agree with the best manuscripts of Homer, although the existence of variants is indicated, either in the manuscripts or scholia of Homer, or by Eustathius. I have listed twenty of these pas- sages.®
1. Rhet. 1, 6 (p. 1363a 5) = /0. 2, 160:
xadde kev ebxwAnv Tpapw
Schol. Ven. A,— dmo rovrov (verse 160) éws tov év Tpoty droAorvro
(verse 162) &0erovvrar orixor tpeis.
2. Frag. 172 (p. 1506 b 31) = JZ7.® 2, 226-228: wAiai TOL yaAKod KdiLoia, TorAAGL & yvvaixes cigiy évi xALoins é£aiperos, ds Troe ‘Axacoi
apwriory dopey. 1214 A. 2 Mag. Mor. 2, 11 (p. 1208 b 10). 38, 2 (p. 1155 34). * Cod. Mb, 8éce. 6 Cod. S, 8¢ x6r’. 5 Cod. IK, roAAdais. 7 Cod. QYDZD> omit 3’.
® Where Aristotle has quoted a verse that we learn from the scholia was rejected by some one of the ancient critics, I have considered that he is in agree- ment with the original Homeric tradition, and that the rejection by the ancients constitutes a variant.
® Cod. Barocc. Mor., «ryolas; Cant., cdoly; L, cdclowes.
218 George Edwin Howes.
Schol. Ven. A,— 6re Zyvodoros ypader wretar & yvvacxay xai rors éfjs So 7 OérnKer. 3. Frag. 13 (p. 1476a 21) = //) 2, 554: Koopyou Urmous Te Kai dvépas domidiwras : Schol. Ven. A,—6re Znyvodoros dé rovrov (verse 553) tpeis orixous 7 Oér nKev. 4. Rhet.? 3, 12 (p. 14144 2) = 1/5 2, 671-673: Nipeds ad SupnOev, Nipeis "AyAains, Nipevs os xadAdcros. Schol. Ven. A (verses 673-675),— éx rav rpiav rots dio 7OérnKe Znvodoros, Tov dé pecov ovdEe Eypader. To emphasize the asyndeton in these verses of Homer, Aristotle quotes the first words only of each verse.
5. Frag. 144 (p. 1502 b 31) = 4. 3, 277:
neALds O° Os wavr’ épopds Kai dvr’ éxaxoves Schol. BQ (Od. 12, 374), — 9éAtos 8° as. 6. Frag. 146 (p. 15032 9) =/7/. 3, 454:
loov ydp ogdtv macw drnyGero xypr pedaivy: Schol. Ven. A (//. 9, 378), —odiar. 9. Poel. 25 (p. 1461 a 14)=/7.' 9g, 203:
Cwpdrepov 8% Képace
Schol. Ven. A,— ovrws xépace ywpis rov p. Schol. Townl., — d<ya tov p ‘Apworopavys xépace. Eustathius (746, 48),—ioréov 8 or ray Tia dvriypadwv To Képace Kepatpe daciv, and (1397, 64), — {wporepov dé xépace 7) Képacpe.
8 (and g). het. 2, 2 (p. 1378b 33) and Fol. 3, 5 (p. 1278 a 37) = //. 9, 648 and J//. 16, 59:
woe tw dtipyroy peravdort ny,
Schol. Ven. B (//, 16, 59), —6 wey ‘Apiorapxos ypdpa peravacr yy. Schol. Townl. (//. 16,59), — év ry Malo)oadtwwrixy Kat ry “Piavod pera- vaorev (cod. peravacrety).
1 Cod. L, xoopfoa 0’. 2 Cod. QYDZ), alctunber.
8 Cod. S, 3° aladunbev; L, ad éovpundev. 4 Cod. Bc, xepae; Na, xepees. 5 Cod. CD (by correction) HS, xépa:e.
Homeric Quotations in Plato and Aristotle. 219
10. Poet. 25 (p. 1461a 2) =J//, 10, 1§2-153: éyxea 5¢ ogiv 6p0" éxi cavpwrypos: Schol. Ven. A, —év rq "Apiorodavovs cavpwripas wAnOuvrexas. 11 (and 12). Fol.) 3, 16 (p. 1287 b 14) and Eth. Mie. 8,1 (p.1155 a 15) = J? 10, 224: ow re dU épxopevaw Schol. Ven. A (//. 3, 211), ow S€. Schol. Ven. A (77. 10, 224),— cuvepyopevor Svo dvi rod ovvepxopevwy. enor 88 py vonoavres TO ¥ apoorWeact, Kaxins. Plato, as we have observed, both quotes® and parodies‘ this verse.
13. Hist. An. 3, 3 (p. 513 b 27) = /7. 13, 546-547: dro d¢ pA€Ba wacay éxepoer, 4 v dva vara Oécovoa Staprepés adyév ixaver. Schol. Ven. A, — Znvodoros 51a 5¢ pAEBa. Schol. Townl., — Zyvodoros dca Se. 14. De Mundo 6 (p. 400a 19) = //." 15, 192: Zevs 5° day’ otpavov eiipuv év aidep xai veheAporv. Schol. Ven. A, —6ri Zyvodoros aimiv. Schol. Ven. A (//. 16, 365), — vedéeraorr. 15. Poet.’ 25 (p. 1461 a 20) =//. 18, 489 and Ou. 5, 275: oty 8° dppopos Schol. Ven. A (/7. 18, 483),— 6re Zyvodoros 7O€r nev dé Tovrov Tov orixou Ta Aowrd.
16 (17 and 18). Mag. Mor. 1, 20 (p. 11914 8), Eth. Eud?® 3, 1 (p. 1230a 20) and £¢h, Mic. 3, 11° (p. 1116.a 23) = 10." 22, 100:
TlovAvddpas po. xparos éAeyxetny dvabno et,
1 Cod. Tb, vr. 2 Cod. G, ouvepx our. 8 Protag. 348 D and Conviv. 174 D. 4 Aletbiad. IT. 140 A. § Cod. PDa, 4 da (for 7 7 dvd); P, Scauwapes; PDa, lxaver.
® Cod. L, 3° fiaxer; G omits év.
7 Cod. Na, a» pdvos. 8 Cod. Ph, rodvddyuas. ® Cod. Lb, rodvdduas; Mb, rpdrov; Kd, dvadhom; M>, érifiioe.
Cod. E, rodv8duas; C omits wou; E, rpdros 9 rparop.
Saar — iss ws ete Pre.. crcl t= eareyerec: (1260,
Soy. SN DL — sam bee ritics me rm Arveowes: ar ina & ~—s oymmer 62? t 8 Ib
ca Hasty rr wm ratte Set ia tte: es a SO. TOS Me EY Yes. TeerTwre-
= rteThws VW LTESs XT: TSE TEE.
Tmt oy § dew rassapss = wont = wis erdenaty Anstotle’s TESST T> BICC he wiles AC ZS lamer tr ite sructzcre of his own Sertem SEL Fo thise Gecimres = wiica sock passages either Sow The Sime Seecmrs Shit ruc Z lemess> am =usermpis offer. or, at B= TESS. EY DiC cpocmswert Sl tiese Teatirs
Az o@ & Teen omrs Larderez @r. cev ceoziy carareroy. Hoe eed? ebervem,. ira cox ceoaly caravrerce. 2. Fv gg RR rsMS Ars refers to 2. 234: Ar. "Quopens & raus Age ei c 2) 2657 dra roAircoipariny, Hom eva dysGexr relrcessavig-: es coparcs fore, This verse of Homer is coxted exicty by Aristotle elsewhere.‘ 3. Eth. Nec. 5. 13 ope rriSa 22> refers to /7. 3. 24: Ar. Gpocwws 3° od" idee | ci pwr EAagdor ¥ adypiory alya, Hom. ctpwer 9 éAagor cepaory § dypior alya, 4. Rhet. 3, 4 (p. 1406b 20) refers to /7. 10, 485-486: piv yap axy Tor AyrA\ea ws 5¢ A€wov excporcer,
cixeoy dorcy,
3 Cod. D, vais. 3 Cod. Q, xarelxero. § Cod. DL, cada (for xepoi). 4 Metaphys. 11, t0 (p. 1076 4); cf. above, p. 213.
Homeric Quotations in Plato and Aristotle. 221
Hom.! as 82 Adwy pyArnow donpayrocw érerOur, alyeow 7 dieoot, axa ppovéwy evo povon,
If this Homeric passage is the one referred to by Aristotle, there are several difficulties in his statement. In the first place, Homer is speaking of Diomedes, and not of Achilles as Aristotle says. This misstatement of Aristotle may be merely a slip on his part. Besides, the verb in the passage of Homer is évopovw, whereas Aristotle has éxopovw. The difference between the form of the two verbs, how- ever, is slight, and a scribe might easily have changed one to the other.’ Still, there may have been such a verse about Achilles some- where in Aristotle’s Homer, and he may be quoting it correctly.
5. Fol. 1, 4 (p. 1253 b 35) refers to /7.° 18, 376: Ar. dowep ta Aadddov daciv 9 rovs tov ‘Hdaicrov rpirodas, ovs Hyow O TroyTs avroparous Oetov SvecGat dyova, Hom. ddpa oi airéparos Oetov Svcaiar dyoava,
Eustath. (1148, 8), —S8voaiar’ dyava, } S¥covrat dyava. Schol. Ven. A,— dv rw Svcovrar dyava. év 8 rais eixasorépas kata Sopa veoiaro. Schol. Townl.,—év 8& rais eixacorépars Oeiov xara Sopa véow.wro. Though we cannot feel certain as to which reading, S8vcaiar, Svcovrat or S¥awvras, Aristotle had in his Homer, still we know he did not have the variant xara ddpa véorvro, which is cited by the scholiasts.
6. Eth. Nic. 10, 10 (p. 11804 27) refers to Od. 9, 114: Ar.* xai (9 éxacros ws Bovdrera, xuxrAwmxds Jeproredwv waidwy 75° dAcxov. Hom.° Oeprorever dé Exagros waidwy 75 d&rXdoxwy, Either Aristotle purposely wrote dAcxov for dAdxwy, or else some copyist made the change. For in another passage® Aristotle in quoting a part of these verses has left us 4A6xwyv in his text.
1 Cod. G, Vrat. A, Mosc. 3, évopotoa; CL, évopodcet.
21 might note an example of the confusion of these two verbs in //. 11, 747.
3 Cod. S, adréyara:; Townl., dueéar’; many cod. have dévorra; L Lips., 3écwrra:. * Cod. Mb omits fxacros. § Cod. K, wal3uy7. ¢* Pol. 1, 2 (p. 1252 b 23).
222 George Edwin Howes.
7. Eth. Nic. 8, 2 (p. 1155 a 32) refers to Od. 17, 218: Ar. of péy yap Gpooryrd riva riOeacw abryy Kai rovs Gpotous dtAous, ober TOV Opolov haow ws Tov Gpocoy, Hom.! us alei rov dpotor dya beds ws Tov Spotoy. This verse, either in whole or in part, is elsewhere quoted three times by Aristotle.’
8. L£th. Nic. 3, 11 (p. 1116b 26) refers to Od. 24, 318-319: Ar. irqruewraroy yap & Oupos xpos rots xivduvous, dbev «ai “Opnpos Spipt 8 dva pivas pévos Hom.* rov 5’ wpivero Oupds, ava pivas S€ of 4dn Spcpy pévos wxpovrupe pirov warép cicopdwvre.
These words belonging to different verses Aristotle has brought together —in a reference rather, than a quotation —to illustrate bravery or spirit. By the introduction of 8° the fragment has been made metrical, so far as it goes.
E. ARISTOTLE’S VARIANTS SUBSTANTIATED.
Of those passages in which Aristotle gives a reading different from the accepted one of Homer, there are many cases where Aristotle’s reading is substantiated either by manuscripts of Homer, by scholia, by Eustathius or by ancient authors.
1. Rhet. 2, 2 (p. 1379 2 4) =—//* 2, 196: Oupos 5¢ péyas écrit Storpehewy BacrAnwy
Schol. Ven. A,— ore Zyvddoros ypada dcorpedéwv BaactrAywy. ovTws évix@s ai “Aptordpxov. elyov dt xai ai yapiéorara: ovTws, dvev
1 Cod. CDKL, és (for the second ws); M, és.
2 Rhet. 1, 11 (p.1371b 16); Zth. Lud. 7,1 (p.1235a7); Mag. Mor. 2, 11 (p. 1208 b 10).
3 Cod. E, dvd piva; K, dvdpptva; S, dvdppiva; P, dvapplpas.
4 Cod. GL, Oupds yap; Cant., Ovuds 8h; ACES, dcotpedéos Bacitfos; DGHL, Siorpepéwr (or Scorpodpéwy) Bagthjwy. La Roche adopts the reading of Zenodotus instead of that of the best codices.
Flomeric Quotations in Plato and Aristotle. 223
tys Zyvoddrov. Schol. Townl.,— d:orpeptos Baorrjjos] ovrws al racas wAnv THs Zyvodorov. 2. Probl. 23, 23 (p. 9344 15) =. 7, 64: Ar) perdve S€ re xovros tx airov. Hom.* dpvupevoo veéov, peAdve S€ re wovros tr aitiis, Although only one manuscript of Homer gives airoi, the reading shown by Aristotle, we find from this scholion of Ven. A that it was a well-attested variant: "Apiorapyos wovrov da rod ¥, xai tr abrp, T7 pint. dAdo 58 wovros te avrov. 3. Hist. An. 6, 28 (p. §78b 1) =J//. 9, 539-540: Ar.® Opdwev éxi yAowny otv dypiov: 0882 égnet Onpt ye crroddyw drAAQ pi bAnevre. Hom.‘ dpoey ém xAovyy otv dypwv dpycddorvra, Os xaxa WOAXN Epdeaxery EOwy Olvgos dAgyy: Some scholars think that Aristotle has here confused this passage with Odyssey 9, 190-191 where we read: cai yap Gaipa réruxro® reAdpiov, ovde egixe dvipi ye avroddyy, dAAa pip tAnervre I cannot suppose that Aristotle has accidentally confused these two Homeric passages. For, as Homer in the passage of the Odyssey is speaking about the famous and peculiar Polyphemus, it seems to me absurd to claim that Aristotle, who we know wrote much about Homer and his works, was ignorant of this fact, or even forgot that these verses applied to the Cyclops. To me, Aristotle’s words have the ring of truth when he says: rév 8° dppévwy xai dypiwy of ropias peiLovs yivovra: xai yoXerwrepo, aowrep Kai “Opnpos troinoev Opeyev «rr. Besides, the reading of Aristotle is confirmed somewhat by Eustathius, who says (772, 46): ro & Gpoe xai dips eipyra: rapa TH yewypady, Opépev xi yAovrynv civ: wap @ xai orixos cipyra otros éxyypévos: ob52 eqer Onpl ye cirodayy GAAA pie bAHevre.
1 Cod. Ca Xa Ya, peralves.
2 Cod. S, weddvet; G Mor, wedacver; many cod., xévrov; Lips., Harl., Townl., Ven. B, aérg; G, atroid.
* Cod. P, Optyiv; Da, dd’ dyplw (but corrected).
4 Cod. C, xAodmp. § All cod. (except Ven. A), Gade érérucro.
224 George Edwin Howes.
Kai onpetwoar Kai rovro eis TO epi oTixwv Aawovrwy éx Tov Opnpov. Eustathius, a few lines below (772, 54), adds: xat ‘ApeororéAns 88 Kata Tov yewypadov yAovvyv civ Tov Topiav vod, Sexouevos TO Opépey exit xAovvnv civ dyptov Kai roy fs ypadévta orixoy zpos riorwow Tov ore Tov dbpévwv Kai dyptwy of ropuia: peiLous yivovrat Kat x Aerwrepor.
Eustathius, when he refers to yewypados, as he often does, means Strabo.! So, if we may believe Eustathius, Strabo too, though the reference cannot be found anywhere in his extant works, I think, quoted these verses from Aristotle and believed them to be Homeric. Therefore I think that Aristotle had these verses in his text of Homer.
4. Poet. 25 (p. 1461 a 26) = J, 10, 252: Ar. mapexnKkeyv St xréwv WE: Hom.® dorpa 8 59 rpoBéBynxe, rapgxuwxev St rrEwv vis Schol. Ven. A, —’Apiorapyos rapwixwxev. Schol. Townl.,— ovro, da Tov w Kata Tporny Tov 7 eis w. 5. De Fart. An. 3, 10 (p. 673 a 16) = /75 10, 457 and Od.§22, 329: Ar. pbcyyopévn 8 dpa rovye xdpy xovinow epiyOn, Hom. ¢$0eyyopévov 5 dpa rou ye xdpy xovinow épixOy. It is evident that there were two readings ¢0eyyouevy and
$Geyyopuévou, for there is a trace of both in the manuscripts both of Aristotle and of Homer; and, besides, Eustathius tells us (818, 4):
ypaderat pty, Oey yopery.
1 Perhaps one example will suffice to show this. Compare Zustatht: Commen- tarii 419, 21 (Geographi Graeci Minores, Vol. II, ed. Mueller) with Strabo 8, 8, 9 (c. 372). In Eustathius we read: 6 5¢ Tewypddos gol» Sri of vewrepor cal maddcora Maxeddéves xal Oerradol Apyos rd wredloy dacl. Strabo’s words are as follows: Apyos d¢ cal rd wedloy Néyerar wapd Tots vewrépots, wap ‘Outpy 8° 0b8° dwat> wddcora 8 olovrac MaxeSovixdy cal Gerrarcxdr elvpac.
2 Cod. Ac, wAédw; Be, wAéop.
* Many cod. have either rap¢ymxe or rapyxnxery; many cod., rréw.
4 Cod. E, pbeyyouévov; PZ, pbeyybuevov; EPSUY, rovde.
8 Cod. H, p6eyyouéva.
6 Cod. D omits this verse; L, p6eyyouévou, xdpn xorlyow eulydn; GR, rod de; MQ. rovde.
Homeric Quotations in Plato and Anistotle. 225
6. Eth. Mic. 7,7 (p.1149b 17) = JL. 14, 217: Ar.’ —_ xdpdacis, 7 7’ exXae voov wuKa rep PpoveovTos. Hom.” xdpdacis, 9 1 exAepe voov rixa rep PpoveovT wy.
7. Poet. 25 (p. 1461a 33)=J/7. 20, 272):
Ar? TH p €oxero xdAxeoy éyxos, Hom.‘ TH p doxero petAtvoy eyxos.
Schol. Ven. A,— dOerotvrat orixo: 8’. Schol. Townl.,—(verses 269-272) ovro: xai rponberovvro zap évias Tay codicray, dv éviots St epépovro.
8. Rhet. 3, 17 (p. 1418 a 8) = Od. 4, 204:
Ar. @® ptr’, érel roaa eles Go dv rervupevos dvpp, Hom. & @0)’, érei roca elras, oo dv rervupévos dvnp
g. Mag. Mor. 2, 11 (p. 1208 b 10) = Od. 17, 218:
Ar.® = aieé roe Tov dpocov dye Oeds ws Tov Gpotov. Hom.’ ws alet rév Spotov dye Oeds ws Tov dpoitov.
In two other passages® the manuscripts of Aristotle agree with the accepted Homeric reading ws aieé. Here, however, Aristotle disagrees with that reading, but agrees with the text of Plato.’ So it looks as if there were two old readings, one of which is given by Plato, and by Aristotle in this passage, while the other reading is preserved in the existing Homeric manuscripts and in the other two passages of Aristotle.
10. foet. 22 (p. 1458 b 29) = Od. 20, 259: Ar. = Sibpov daxéAoy xarabeis dAcynv re rparefav. Hom." 8idpov dexéArov rapadeis dAcynv re rparefav:
1 Cod. Mb, dpovedyrwy. 2 Cod. L, dpovdbnr dt” ® Cod. Be, 8° (for A’). 4 Some cod., xdAxeop. * Most cod., eles; BL (after erasure), rd’ fares; D, récca teres.
6 Cod. Mb omits roc. gs
7 Cod. CDKL, és (for the second ws); M, és.
8 Rhet. 1, tt (p. 1371b 16) and £th. Eud. 7, 1 (p. 123547). 9 Lysis 214 A.
19 All cod., 3ippo» re, the re evidently a mistake of scribes. A few cod., xaraGels.
226 George Edwin Howes.
F. HOMERIC VERSES OMITTED IN ARISTOTLE.
Naturally, perhaps, I should consider next the passages in which Aristotle’s readings receive no support elsewhere; but I prefer to pass over these for a time, and to take up those passages in which verses that are contained in our Homer are omitted in Aristotle.
1. ol. 3, 14 (p.1285 a 13) =//. 2, 391-393: Ar.. ov 5€ xk’ éyiay daravevOe paxns, ov of dpxiov éooeirar pvydayv Kivas 79 olwvors * wap yap é€uot Odvaros. Hom.* ov 5€ x’ éywy drdvevOe payxns e0€Aovra vonow ptpvalev rapa vyvol Kopwviow, ov of érara dpxtov éocetrar puyéey xvvas 75° olwvous.
In this passage Aristotle omits the last part of one verse and almost the whole of the succeeding verse. His object in quoting the passage is to show the authority of a leader to inflict death upon his men, if need be, in time of battle. Consequently he gives those words only that emphasize that part, omitting even the main verb of the sentence. In the Nicomachean Ethics® we find verse 391 given in full, though, to be sure, it differs somewhat from our Homeric text. Of the expression rap yap épuoi Odvaros I shall speak later.
2. Rhet. 3, 11 (p. 1413 a 28) = 17. g, 385-390:
Ar.‘ eioi 8&8 trepBodrai pepaxiwdas: oodperyta yap SyAovery.
d:0 dpyLopevor A€yover padiora:
ovd' & por té0a Soin ooa Wapabos TE Kons TE. xoupnv 8° ot yauéw "Ayapduvovos ‘Arpeidao, ovd' ei xpvoeiy "Adpodity xddXos épiLor, épya 8° ’AGyvarp.
Hom.* 008° e pot réca Soin doa Wapabds Te Kons Te, ovde kev Ws Ere Ovpov eudv meiva ‘Ayapeuvuy, apiv y dro racay epoi Sopevar Ovporyén AWByy. Kovpyny 5’ ov yapéw *"Ayapeuvovos "Arpeidao, ovd” ef xpvoein "Adpodity xddAXos épi{or, épya 8° *AOyvaty yAavxwmids ivodpapi Cor:
1 Cod. Ib, évetra:. 2 A few cod., 3° a» (for 3é x). 33, 11 (p. 1116. 34).
4 Cod. QZbAs, 8olys; Zb, Scoa; Q, Sas; Q, xdony; QYbZbAS, xpuch; Q, d6nvalys. 8 Cod. S, Cant., weioy; E, xpuoq; Vrat. b omits verse 390.
Homeric Quotations in Plato and Anistotle. 227
It might possibly be claimed that in this passage verses quoted by Aristotle had been carelessly omitted by a scribe, but I think that such a suggestion is not necessary or even probable. Aristotle is here quoting examples of exaggeration of statement. As the force of the first example lies wholly in verse 385, especially in the words Softy daa Wapabds re xoves re, Aristotle very properly omits the conclusion, which, no matter what its character might be, could not lessen the exaggeration of the condition. In the second exam- ple the three verses are so intimately connected that it is not easy to separate them. Even here, however, we notice that the last part of verse 390, yAavxwmrid: loodapi{o.r, has been omitted as unnecessary.
3. Poet. 25 (p.1461a 16) = /7. 10, 11-13: Ar. 10 & xara peradopay epyrat, olov .. . apa d¢ pow
nro. Or és wediov TO Tpurxov dbpnoeaer, atAG@v ovpiyywv 0" Spadoy.
Hom.! 4 rot or’ és wediov 7d Tparxdv dbpnoee, Oavpaley rupa rodAd, Ta xaiero "IXcdO: mpd, — avAwy ouptyywv Tr évorny cpadoy Tr dvOparur.
Here again those verses only are given by Anstotle which are necessary to illustrate his point. Of the variant 0’ é6padoyv I shall speak later.
G. VERSES NOT FOUND IN OUR HOMER.
There are many places in Aristotle where he shows a familiarity with verses of Homer that cannot now be found in Homeric manu- scripts. Such are the following:
1. Pol. 3, 14 (p. 1285 a 13) =J//. 2, 391-393:
Ar? ov 8d x’ dywy dxdvevOe payys, ov ol dpxiov écocira: pvyéey xivas 75° olwvors: wap yap épot Odvaros.
Hom.* oy 8€ x" éywy drdvevOe pays ¢Bédovra voyow © pupvdley mapa vyvot Kopwvicw, ov of erecta, dpxiov doodira: pvyéey xivas 75° olwvors.
In the Iliad, this last verse ends a speech of Agamemnon.
1 Cod. Cant., 74 xalovr’. 2% Cod. Ib, dceirax. * A few cod., 3° dy (for 3é «’).
228 George Edwin Howes.
I have already considered this passage of Aristotle. Now I wish to call attention to the words rap yap épot Odvaros, which Aristotle seems to have had in his Homer, but which are not found in our Homeric manuscripts.
2. Hist. An. 6, 28 (p. 578b 1) = J. 9, 539-540: Ar. Opépev ei yxAowvnv ovv adypiov: ovdé éwna Onpl ye acrodadyy, ddAdAa pip tAnevre. Hom.’ dpoev éxt xAownv ody dypiov dpy.ddovra, Os Kaxa TOAA’ Epderxey EOwv Oivnos dAwnr: Though I have already discussed this passage at some length,’ this second verse in Aristotle ought to be listed here, as it is not found in our Homer.
3. Rhet. 2, 9 (p. 1387 a 32) =/0. 11, 542-543: Ar.‘ xai tov arrow ro Kxpeirrom duguoByreiv, padvora pév ovv TOUS év r@ alta: dOev xai rovT’ eipyrat, Alavros 8° dAdave payny TeAapwnadao- Zevs yap of vepéoacy’, oT dpetvovs Suri paxorro. - Hom. Alavros 8° dddéave paynvy TeAapwvddao. [ Zeds ydp ot venecad’, or’ dpeivon uti paxorro. | Verse 543 is omitted in all the Homeric manuscripts, but it is substantiated by Plutarch,® who quotes it thus: Zevs yap Tor veuerg, Or’ ducivove Pwri payxoro.
The verse is given also in the Pseudo-Plutarch.*®
4. Eth. Eud. 3, 1 (p. 12304 19) =f. 22, 98-100: Ar. “Exropa 8’ aidws efAe: TlovAvédyas prot mparos éAcyxeinv dvabnoe. Hom.® 98 6yxOyoas 8’ dpa elre mpos ov peyaAntopa Oupov- 99 w por éywr, ef per xe wiAas Kat Teixea Suu, 100 TTovAvddwas por mparos édcyxelnv dvabyoa,
1 Cod. P, Oplyi»; Da, dAX dyplw (but corrected).
2 Cod. C, xdovmy. 6 Moral. 36 A. 8 Cf. p. 223. 8 De Vit. et Poes. Hom. 2, 1332. 4 Cod. AS, veuédcarne; QYOZ), veudrno’. 7 Cod. Pb, rorvdduas.
® Cod. H, réxea; E, wodvdduas; C omits no; E, rpdros 9 rpdrop.
Homeric Quotations in Plato and Aristotle. 229
The words “Exropa 8’ aidws efA«, if they existed in Aristotle’s Homer, must have come before verse 99, as they could not have formed a part of Hector’s soliloquy.
5 (and 6). £th. Me. 3, 11 (p. 1116b 26): — iryrixwtatoy yap & Oupds mpos Tots xevduvous, GOev Kat “Ounpos oOévos Eu Bare Ovp@ nai pévos cai Oupoy éyecpe
Of the words oOévos éuBare Ovy@ it might be said, that Aristotle had quoted them carelessly, since we find somewhat similar expressions in our Homer, as: pévos 5€ of éu Bare Ovyug,' and cOévos épBadr éxdory.* But, when we consider the expression pévos xal Oupow éyecpe, we find nothing in our Homer resembling it nearer than these words: pévos péya, OéArAye 5é Oupdv.® So I suspect that both of these quotations of Aristotle were made from verses in his Homer that are not found in ours.
Now, as is evident, I am treading on extremely doubtful ground. For there are some passages quoted by Aristotle as from Homer, that are entirely different from anything in our Homeric manuscripts. It might be suspected that Aristotle was quoting from other works, not now extant, that were sometimes attributed to Homer. But, so far as I know, with the exception of the Margites,‘* Aristotle mentions as Homeric only the Iliad and the Odyssey. Therefore, since we have abundant proof, as I have shown, that there were many verses of the Iliad and Odyssey that have not been preserved in our Homeric manuscripts, why may we not justly suppose that these quotations of Aristotle refer to verses of Homer that were found in some of the older manuscripts, but were either not known to the Alexandrine critics or else rejected by them, and hence were lost to our comparatively modern manuscripts?
7. Pol. 8, 3 (p. 1338 a 24): diorep “Opnpos ovrus eroinoey GAN olfov pév ore xaretv eri Batra Garelny. The insertion of a single short syllable in the second foot — for example, re — would make the hexameter complete. 1 71. 16, §29. 27], 11,11, and //. 14, 151. 8 71. 15, 594. § Aristotle refers to this, as a work of Homer, in Poet. 4 (p. 144%b 30).
230 George Edutn Howes.
S. De deta’ 1, 2 (p 4043 29): &s mades recporm rer Ongpov wos “Karep ceitr édArlodpeortor.
Anistotie thus refers to this verse in the Metaphysics’: ¢aoi & «ai vie “Opgper rarrys iyerre daceste: rv defer. ori éreigoe ror “Exropa, ws éLerry tre res rlryys. ceicGa: dd\Aodporeeorre,
So we can have no doubt, I think, that Anstotle’s quotation repre- sents part of an actus] Homenc verse. The same expression, too, is found in Theocritus,’ who sars :
was 8° éxi yaley ceit dAlod ver tor. g Are. New rpm rr16b 26): ityncerates yao & Gress Epes Trews auvéevers. chevy cui “Omapes ... mar €Leger eine:
The verb Sé¢w ts used br Homer, so far a5 our text shows, only in the expressions Attar wuav and 2 iveiter ivtti. Theocritus} how- ever, has the same expression that is quoted here m the Nicomachean Exh. namely jveser Aaal- dam 8 adap CLeoer eiga As Theo- entas knew his Homer wet and comed him freely. perhaps. this expression in his Idvi was horrowed directly from Homer. Why may he pot have had an liad ar Qdivsser m which this expression was used? This ad edinen of Homer in the possession of Theocritus might have contained also the expression «ceir” dlAdodperces, which Theocrnrss uses and which is: quoted in the De Anima’ It Would seem as of in these two passares Theocritms supported Anstode and his quotanons
wa Frag. isa op rsagh gg) in Schol Townl on /7 24, 420:
diiwares rexpey Tpayeara grez, o> Gyou “Aporerass ap_araws "Oncgper atreer 6¢ repiSperececa wrerdy
Thar this reading of Homer was jost in early times is shown by
the words of the scholiast thar follow: retre & re gare oF de
Seperar
1Cad SW, ws 6 kxcvap 2S 3 q@ 1900d 38 3 Jé 23, 129 €Jé. N13 * Cf the mnrevocens passage discusned.
Homeric Quotations in Plato and Aristotle. 231
11 (and 12). het. 3, 4 (p. 1406 b 20):
"Eor: St xai 4 ix perapopd: dtadeper yap pixpov' arav pév yap ciry Tov "AxiAAea ws 8% Adwy exdpovecer, cixwy ecru, drav & Adwv exopovce, peradopa:
The expression ws 5% A€wy éxdpovoey has already been dis- cussed. Whether that refers to a passage no longer found in our Homer or not, the words A€wy ércpovees, if they represent a quota- tion at all correct, must have come from a text of Homer different from ours; for in our Homer we cannot find any passage to which they would properly refer.
H. NEw READINGS IN ARISTOTLE.
There remain for consideration the comparatively few passages in which Aristotle, while quoting verses contained in our Homer, gives readings unsupported by other testimony. It will not be necessary to treat each one of these passages fully, though a few suggestions may very properly be made.
1 (and 2). Soph. Elench. 4 (p. 166b 6) and Poet. 25 (p. 1461 a 22) = //, 2,15 and 2, 32:
Ar. (Soph. Elench.) xai 1d rept rd évernoy tov ’Ayapéuvovos, ort ovx avrés 6 Zevs elrev SiSopev 5é of ebxos dpév Oat, BANG TO evurrvip ever AXero ddovat. Ar. (foet.) SfSopev Sé€ oi Hom. "Hpyn Axcoopery, Tpwecor St xyde’ EpHrrac.
Though the words 8iéSopev 5€ ror edxos dpéoGar are found in Iliad 21, 297, it is evident from the general sense of the passage in Aristotle, that he is referring to one of the earlier passages, and not to the later one.
3. Eth. Nic. 3, 11 (p. 1116.4 34) = //. 2, 391-393:
Ar sv 8 x’ yw dravevOe paxns TTaTTOVTA vonTw, ov ol dpxioy ¢ooedira pvyéey xvas.
1 Cf. p. 220. 2 Cod. Kb, &pxeory; Kd Mb, écetras.
232 George Edwin Howes.
Hom.! ov &€ x’ éyuw dxavevOe pdyys €OéXOvTA vonorw pipvaleay wapa vyvoi Kopeviow, ov of drara dpxiov écodira pvycay Kivas 95 olwvors.
In this passage it is stated carelessly that the verses were uttered by Hector, whereas they were really spoken by Agamemnon. That Aristotle was aware of that fact, is clear from a passage in the Politics,* where he correctly refers the words to Agamemnon.
The reading xrwocovra is metrically and grammatically correct.
4. Probl. 30, t (p. 953. 4 23) =/Z. 6, 200-202: Ar atrap éxei xai xeivos dxyyero racr Geoiow, TO 6 Karwredov TO AAnov olos dAaTo, ov Ovpoy xarédwy, rarov dvOpurey dAccivwr. Hom.* daAd’ ore 3% xai xeivos dxnyGero racr Oeotow, 9] TO O Kar rediov TO “AAmov olos GAGTo, ov Oupov xarédwy, watov dvOpwrwy dAccivey, 5. De Mot. sin. 4 (p. 699 b 37) = 77. 8, 20-22: Ar.® GAX” otk dy épvour’ €f otpavobey wediovie Znv vxaroy xavtTwy, ovd ef para woAAa xapowre® wayvres 8° éfarrecbe Get wacai re Géarva. Hom.® xavres 5° éarrecGe Geot wacat re Géorvas- GAN’ ovx Gy épioar® éf otpavobery rediov Se Zyv, Urarovy pyaoteap, ovd ef pdAa rodAd Kaporre. La Roche’ tells us that there has been no change in the order of verses of the Iliad or Odyssey from the time of Pisistratus down, and that the manuscripts of Homer show this fact. Here, however, in
Anstotle we find the order of verses changed. This change may be due either to Anstotle or to a scribe.
1 A few cod., 3° a» (for 3é x’).
23 14 (p.1285a 11).
8 Cod. t, obros (for xeivos); t, olov; t, caréder.
* Cod. C (after an erasure) and many others have «d«xeivos.
3 Cod. E, épécer’; P, uwaror uhorropa rdvrev; EP, é&dwrecOat.
6 Cod. C, wdvres +; S Mosc. 1, dv uw; G, épdowr’; HL Vrat. b, edggre; Schol. Ven. A, yp. xal rd@are.
1 Homerische Textkritik, p. 7.
Homerie Quotations in Plato and Aristotle. 233
6. De An. Gen. 5, 5 (p. 785 a 15) = #7. 8, 83-84: Ar.! iva Te wpwrat Tpixes Ura xpavip ¢urepvact, padrora dt xaipiov darcy. Hom.? dxpyyv xax xopudyy, 60e re zpwrat Tpixes Urry Kpaviw éurepvact, pddora 5€ xaiptoy tore.
7. Rhet. 1,7 (p. 1365 a 13) =J/7. 9, 592-594: Ar® = dcca xax dvOpwroact rede Tov dory dAwy: Aaoi pév POcvVAover, wor 5€ re wip duabwe, réxva. 8€ tr’ dAXAOL ayovor. Hom.‘ «75, 60° dvOpdirow: ra Toy dor dAwy’ dvSpas pev xreivovar, wordy S€ re wip duabuvea, réxva 5€ t' dAXAot dyovor Pabvfwvous Te yuvaixas. Schol. Ven. A,— dri Zyvodoros ypade réxva 8¢ 84104 dyovor. Schol. Townl., — not ypddover réxva St Stor dyover.
The expression Aaoi péev POcvvPovor is Homeric, and is found in Iliad 6, 327. It is not impossible that it may once have been used in Iliad 9, 593, just as Aristotle quotes it.
In the reading r dAAox Aristotle agrees with the Homeric manu- scripts, though the scholiasts, as we see, note a variant d7c01.
8. Poet. 25 (p. 14614 18)=/7. 10, 11-13:
Ar. rot or és rediov 76 Tpwixdy d0pyoaer, avAov ovpiyywv 0° dpaddy.
Hom.® 4 rot or és wediov 76 Tpwtxdv dbpnoee, Oavpaley wupa rodAd, Ta xaieto "IAcdA po, aiAay oupiyywv Tr évorny spaddv 7” dvOparuy.
Of the omission of verse 12 I have already spoken. I have now to treat of the word opadov. It seems to me probable that Aristotle had’ the expression in his Homeric text just as his manuscripts give it. In this passage Aristotle is speaking of metaphors. Now opadov with atAev and avpiyyeyv would give a much better example of a
1 Cod. SY omit rpdra:; Z, duweddxaci.
3 Cod. H, xdxxopugphy; C (first hand) omits re. 8 Cod. QYbDZ}b, Sea; QYbZb omit wéra.
4 Cod. G, «h8ea Seo".
5 Cod. Cant., 74 xalorr’.
234 George Eduin Howes.
metaphor than would évoxywy; for oaados generally refers to the din or uproar of men, whereas here by a transfer of meaning it would refer to the din or blare of trumpets.
9. Frag. 143 (p. 1502 b 4) = //. 10, 332: gnoi 5° *ApwroréAns ore od" 6 xounrAs A€ya ws eEmeapxyoay, xafaxrep éx” awry es aro xai p éxiopxoy spocery, Hom.' os daro xai p éxiopxoy éxepoce, tov &° dpobuver. Schol. Ven. A. — ovres xai da ror € ro érepocer ai Apurrdpyxov. We must assume that Aristotle intended to end the real quotation with éxcop«oy, or else that éx (or dr) has been lost through some
mischance, for the use of the simple verb spzooey renders the line unmetrical.
10. foct.? 22 (p. 1458 b 31) = //. 17, 265: juoves Boowory This expression of Aristotle is thus edited by Bekker. All the manuscripts, however, read twves Bo@orv. The error of the manu- scripts is probably due to a scribe, who would naturally suppose the subject of this verb to be persons rather than things.
11. Eth. Lud. 7, 1 (p. 1235 3 26) = //. 18, 107: Ar. ws épis éx Te Oewy xai dvOpwrwv droAcro: Hom. os épis éx re Oewy ix 7 dvOpwrwy dxdXorro, 12. Prob/. 26, 31 (p. 943 b 22) = Ou. 4, 567: Ar® — AX’ aiei Lepvpoo Scarveiovocy dnras Hom.* dX aici Cepvporo Ary xvelovros dyras For the reading of the Homeric manuscripts we have the testimony of many writers. The verse as quoted by Aristotle, however, would readily unite with the preceding verse of Homer, ot uqerds, ovr’ dp Xetpisy OATS ovTE wor GuBpos, if the succeeding verse were omitted, "Oxeavos dvinow dvapuxyev avOpwrovs.
1 Cod. H, éwel Spxov; a few cod., drwyoce; D, deradpoocey; H, dercsuoge ; C (recent hand), yp. éx. 2 All cod., fwres Bodocy.
8 Cod. Yas, del; s, dtarvéovow; Ca, durpal.
4 Most cod. have wrvelovras; M, wrelovrds ; H, wvelorras; Schol. HP, 7d xvelovros 8:4 rol 6 wpds 7d Sepdporo.
Homeric Quotations in Plato and Aristotle. 235
13. De Mundo, 6 (p. 400a 11) = Od. 6, 42-45: Ar.) QUA uprdvd", 60: pact Oeay os dogadres aici dupevar. ovr dvguown Tivdoceras ovTe ror OuBpy Severat, ovre xiwy éxiriAvatrat, dAAG par’ aldpy werratas, dvvépedXos, Aevxy 8° dvadédpopeyv alyAn. Hom.? OvdAvpzoyv 8°, 60: haci Oeav Sos dodarés aie éupevac’ ovr dveuowt Tiwaoceras ovTé wor opBpy Severas ovre xiwy émimiAvara, dAAG pad’ aibpy wenraras dvéepedos, Aevayn 5° excdédpoper aiyArAy: Of Aristotle’s form dvyédeAos nothing need be said, for that is found in many Homeric manuscripts also. dvadéSpopev is not supported elsewhere.
14. Poet. 22 (p. 1458 b 25) = Od. 9, 515: Arist. viv 8€ pn’ dio dAiyos re xai otridavds Kai dxixus, Hom.‘ viv 5€ p div dAiyos re xai obridavds xai dxixus Eustathius (1643, 7) says: ypagerac 5° &y riot xai decays. Since the three manuscripts of Aristotle have decdys or dydys, it seems to me probable that Aristotle had in his Homer and wrote decxys, which is read in a scholion of Homer and also in Eustathius. This could easily have been corrupted, in the text of Aristotle, into decSns and then dds.
15. Rhet. 3, 11 (p. 1411 b 33) = Od. 11, 598: Ar.’ —s atris ért Sdwedovde xvrivdero AGas dvadys, Hom.® airs dwecta wédov 52 xvdivdero AGas dvadys.
The difference in reading is really very slight, and the unmetrical ¢xi in Aristotle is probably due to scribes.
1Cod. P, obAuuwéy 0°; O, obdAunwor; Q, Sri; P, Soar; Q, vdaros (for eos); O, 0682 (verse 43); O, ofr’ ad xibves; P, dvégeda; Q, dvépedos; O, ddnh (for Nevch).
2 Cod. AK, ot3éror’; B, of8¢ 7 (verse 43); LPS, od8¢ (verse 44); most cod., dyvépedos ; Eustath. 1551, 5, dvépedos and dvptpedos; D, éxcdédpayer; Schol. H.P., "Prards aldo.
* This verse is so edited by Bekker, although one cod. (Na) has 4784s, and the other two (Ac Bc) have decd#s.
$ Cod. S, dvadxis; M, yp. decats.
® Cod. QYDZb, 3° éwl; Q, Sdwedby re. @ Cod. EQ, adhis.
236 George Edwin Howes.
16. Ret. 1, 11 (p. 1370b 5) = Od. 15, 400-401: Ar.! METa yap TE Kai aAyeot Téprerat avip BYYMEVOS, COTLS TOAAA TAO Kai woAAG EOpyy. Hom.” pywopevw: pera yap Te xai aAyeos Téprera: dnp, os tes 5% pada wodAa way Kai WOAN erarA4O7. The word éopyy given by Aristotle seems to me to harmonize
better with ra@y and with the general sense of the passage than the traditional érad0 7.
17. Fol. 8, 3 (p. 1338 a 26) = OW. 17, 385:
Ar. ot xaX€ovary dodoy pyoww. 6 Kev TéepayOW aravras. Hom.® 9 xai Oéoxcy dodorv, 6 kev réepryow deidov;
18. Probl. 30, 1 (p. 953 b 12) = Od. 19, 122:
Ar. xad pe dyoc Saxpv xAdav BeBapynpévoy olive. Hom.’ $7 5¢ daxpuxAway BeBapnora pe Ppévas oivy.
The first part of the verse as given by Aristotle is unmetrical. Were it not for the fact that the last part is metrical, we should think that Aristotle was merely referring to the passage and not attempt- ing to quote it. As it is, it looks as if he might have had a different text.
CONCLUSION WITH REGARD TO ARISTOTLE’S QUOTATIONS.
In general, now, what can be said of the trustworthiness of Aristotle’s quotations? Did he, by relying on memory, commit so many offenses that his variants are entitled to no consideration ? About him, as about Plato, I think we may say that there are occasional passages where the presumption seems very strong that he has quoted from memory and quoted wrongly. We cannot, how- ever, dismiss all, or even many, of his variants in that abrupt way. Whether he quoted from memory or not, for the following reasons I feel that his readings are entitled to a careful consideration, and
1 Cod. Ac, ponodyevos bre; Q, rdOor; QZ, fopye; Yb, dbpye. 2 Cod. D, powpéow; M, yp. pepvopéves.
8 Cod. R (and Eustath. 1824, 59), és a».
4 Cod. Xa, rAdvecy BeBapupé vor.
8 A (first hand) omits this verse.
Homeric Quotations in Plato and Anstotle. 237
that, where they differ from the traditional text of Homer, in most instances they probably give us variants of high antiquity.
1. Let me repeat again that we feel more strengly since the dis- covery of the Flinders Petrie fragment, that the modern manuscripts of Homer differ greatly from the earlier traditions.
2. Though I cannot go into the question as to whether Aristotle himself prepared a Homeric text for Alexander the Great, — for there seems to be conflicting testimony on this point, — it is clear from the titles of two of the works of Aristotle,! and from references to him in the scholia of Homer, where his readings or explanations are occasionally mentioned, that he paid a good deal of attention to the study of Homer, and hence must have been well acquainted with his poems.
3- Many of the quotations in Aristotle agree with our traditional Homeric readings. Many that disagree receive support either from Homeric manuscripts or from scholia or from ancient authors.
4. Many of the differences are undoubtedly due either to mistakes of scribes, or to the fact that Aristotle occasionally refers to verses without intending to give the exact words.
There then remains a comparatively small number of unsub- stantiated variants, to be attributed to a difference of text. Should we not expect that Aristotle, who lived so long before the Alexan- drine critics, would exhibit as many real variants as his manuscripts show?
1THpoBrvAfpara ‘Ounpexd (cf. Biographi Minores, ed. Westerman, p. 404, 77), and ‘Aropfpara ‘Ounpend (cf. Diogenes Laertius, 5, 1, 26).
GENERAL INDEX.
Abbreviation, 58. adbictam, 99. &bicit, 87, 101. abiécit, 97. Accent of 'Avaelr:, 58. of Moveaios, Moveas, 60. Actium, era of, 61. Odici, 102. adieceret, 96. adtecientur, 98. adsecit, 98. Adoration, attitude of, 55. Aeantides, tragic poet, 78 #. -aes from -aeus, 59.
Aeschylus, Plato’s quotations from, 155.
dytprns, 67 . Atitax, 95, 107, 110. @ito, 95, 107. -as from -acos, 59. “11S, -cius, words in, 93. Alexander Aetolus, 78 2., 79 #. Polyhistor, 71 #. Alexandra of Lycophron, date, 76. “Avaclri, 56. Anahita, 66. Anaitis, inscriptions relating to, 57 #. epithets of 57 f., 66, 72 #. cult, 58, 66. "Avdircs, inflection, 58. accent of dative, 58. Aratus, not tragic poe, 78 #. dpxaios (wadaids) reds, 8, 18, 21, 26 f. Ariolus, dyéprys, 67 7.
Aristotle, Homeric quotations in, 153 ff.
Artemis Anaitis, 56, 57 ff. worship, 66.
and Mén Tiamu, votive tablet, 55 ff.
Asclepius, Mén like, 58 .
"Aces [Zeus], 69 #.
Athena Polias, temple of, 11, 18, 29 f. temple of not the Parthenon, 30-33. nor the Hecatompedon, 34-36.
’"AOneas from ‘AOhrascs or ‘A@nvatos? 59 f.
Attagus, 74,
Atteo, Attis, 65 f., 68 #., 74.
"Arris offs, 68 2.
Attis, Menotyrannus, 65. Myth, 74.
Attitude of adoration, 55.
Asia Minor, religion, 63 ff., 73 £.
Ausonius, comp. of sacto, 151.
-B-, 60 f.
Babylonian Creation Tablets, 68 #.
Barnaes for Barnaeus, 57 #.
BATES, W.N., The Date of Lycophron, 75 ff.
Boeckh, on Opisthodomus, 1, 5 #., 11, 43 f.
Brambach, on -sszo etc., 84.
On -#25, 95.
Caesellius, on -ast-, -e##-, 94 f., 108.
Calendar, Macedonian, 62. sequence of months in, 57 #.
Calligeneia, 57.
Catacecaumene, 66 f.
Catullus, comp. of sacto, 137 f.
Cicero, comp. of facto, 138.
Claudian, comp. of sacso, 151.
co-, prepos., nature of, 105.
coicio, conicto, 121 ff.
coustctt, 118.
Te 3 iC Ee 2c Meme Actes Fula 1 Wage f Sectomoetce at Wn Cmerhvicm.c. oc mic © ieee 2t. Ft. 1 t
Sees, yea. O1s1c19, Hesrebs. 122% Oasiecs. Issrect. Tia Deorphei2. on Opisracdomss. 2£. 52 AR tempe. 14.95: s5%ecices te vem Of. 208 on Heratompedon, 13, 24 27. 3 om temole A Athena Polias. 13, 29; objections to view of, 29-7. Setipos, legis, 72 2. Surdpus (4urdpes), 56, 65, 72 2.
afors. f. Hcl, 114, 195. heere, U0. Cher sant, 10. OM tt, Ath, 082.
menuut. 272 Pormciceees. >-puiar. ay. 66.
E2aemesis. 2G 2.
est oscn. G «
Pagaramis. wt Cage poet. -9 =
Earcpices. Aristcce’s quotations from, | £ oe me
Piato'’s ;10tasj0ns from, 155 £
gda. SS 2.
Foocart. 64 2. 6- 2.728. Fowler, on Opisthodomas, 3 Frankel, on Opisthodomus, 2. Frazer, on Opisthodomus, 2, 40 #., 46 #. on ‘old temple’ and temple of Athena Polias, 19. on the word MHok:ds, 36 f. Furtwangler. on Opisthodomus, 2 #., 48. on Hecatompedon, 19, 23; objec- tions to view of, 24-28.
General Index.
Gai, Gaii, etc., Grai, Graii, etc., 93 ff., 110. Germanicus, comp. of sacio, 145. Gratiugenarum, 94, 108. Grammarians, testimony about comp. of tacto, 88 f., 92. Great Goddess, 58, 64, 66. Great Mother, 69.
Hf to remove hiatus, 120. Hecatompedon, names applied to, 2 #., 22, 27. inscription referring to, 3 #., 10, 44. not rebuilt after Persian Wars, 22 m. age of, 24. *Hpais from ‘Hpaios, 59 7. "Epuas, from” Eppacos or ‘Eppatos? 59%. f. Hesiod, Aristotle’s quotations from, 168 ff.
Plato’s quotations from, 161 ff. iepdy, meanings of, 5, 8, Io f., 25 #. lepowolnua, 72 #.
Homer, Aristotle’s quotations from, 210 ff.
Plato’s quotations from, 176 ff.
tragic poet, 78 #.
Homeric quotations in Plato and Aris- totle, 153 ff.
*Oubpwxa, a corruption of Marduk, 71 #.
Horace, comp. of sacto, 141.
HowEs, G. E. Homeric Quotations in Plato and Aristotle, 153 ff.
bis drrfs, 68 2.
bod, 56, 61.
Hygieia, 58 2.
casa, 58.
-icio, testimony for, 99 ff., 103, 113 ff. contracted w. preposit., 113 ff.
inieciatis, 96.
intectent, 98.
tnidcit, 97.
Injikler, 63 #.
241
Inscriptions: on votive tablet to Arte- mis Anaitis and Mén Tiamu, 56 ff. relating to Anaitis, 57 #.; to Mén Tiamu, 71 a. ff. dated, 56, 57 #., 71 #. from Byzantium, 64 #. Julia Gordus (Goerdis), 56 #., 61 #., 71 2. Hypaepa (near Odemish), 57 #. Kula (Menneh, Matopsla), 57 x., 59 #., 61, 62 #., 72 2. £. Palmyra, 70 #. Philadelphia, 57 #. 3 for -tos, 59. for -1%s, 59. Isaiah Ixv. (11), 70 #.
# consonant, effect on preceding vowel, 106 ff. doubled between vowels, 107 ff. Iaciendi Verba Composita, 83 ff. tacto, comp. W. prep. ending in conson., 87 ff. ending in vowel, 104 ff. comp. in inscriptions, 127 ff. -tecto, testimony for, 96 ff., 103, 110 ff. in vulgar and provincial Latin, 102, IIS., -ticto, testimony for, 88 ff., 103, 117 ff. Juvenal, comp. of sacéo, 150.
Kakdryévera, 56. Kayapelrns, [Mir] 69 2. Karax6émos, 69 f. Karaxexaupévn, 62. xaroxla, 63 7. xotplov, kuplov, 72 7. KoNéy, not at Kula, 62 f. Alurn Kodén, 63 #. Kula, 62 f. inscriptions from, 57 #., 72 #. ff. not KodAén, 62 ff. Ildpyos, Tlupyla, 63 #.
242
Lachmann, on -#cio etc., 85.
Lucan, comp. of sacso, 147.
Lucilius, comp. of saczo, 135 f. Lucretius, comp. of sacio, 137.
Lunus (M4»), 66 #.
Lutatius Catulus, comp. of sacso, 136. Lycophron, date, 75 ff.
Lydia, religion of eastern, 63 ff.
Ma, 64. Masiam, 95, 107, 10. Ma-n, Maen, Mén, antiquity of this god, 64 ff., 74. See also Mén. Manes, 64 #. Manilius, comp. of sacto, 145. Marduk, 71 #. Martial, comp. of tac#o, 150. Macdadaryry. Ai. 72 7. MATHER, M. W., Iaciendi Verbi Com- posita, 83 ff. matriarchal social system, 7 3. péyapor, meaning of in Hdt., 22 #., 25. Mén (see also Ma-n), epithets, 70 x. like Asclepius, 59 #. relation to Attis and Sabazius, 65x.; and Great Mother, 65 #. originally a solar divinity, 64, 66. range of cult, 66 m. f. priests of, in Athens, 67 #. Mén Caru, 58 #. f.; Pharnaca, 70%.; Tiamu, 57 ff. Artemis Anaitis and, votive tablet to, 55 ff. inscriptions relating to Mén Tiamu, 71 #. ff. Mnvl Tidpov, 56, 71 ». ff. Mnwis Tidyou, 59, 71 2. f. Mnmridyov, incorrect, 59. Mén Tiamu = Mi cxaxaxOdbmos ? 69 f. Mnvraytprns, 65 #., 67 2. £. comedy by Menander, 67 .; by Antiphanes ? 67 #. phen, Mhvn, 66, 67 2., 71 2.
General Index.
Meni, Aramaean Téyx, merged into Mén, 70 2. f. Menis magister, Ménotyrranus? 70 #. Myw» (from Ma-lwy?), 64 #. perdrSpios, 65 2. Myryp, ‘Avdizis, 57, 58, 66 2. (Ged»), 64, 68 ». Mnrpya, 68. Mnrpaytprys, 67 n. f. Metroum at Athens, 68 ». Michaelis, on Opisthodomus, I, 5 #., 12. on ‘old temple’ and temple of Athena Polias, Ig. on age of Hecatompedon, 24 . Milchhofer, on Opisthodomus, I #., 16 #., 48; objections to view of, 49-53. Monogram, 58, 62. Months in Macedonian calendar, 57 x. Mother, Great, names of, 64 #. See also M4rnp. Moveaios, frequent on stones, 59 #. Moveais from, 56, 58. accent of forms of, 60. Miller, Luc., on -ésczo etc., 85. Murro, on -##czo etc., 86. Musaes, 57 ff. from Musaeus, 58.
vews, meanings of, 5, 7-II.
Oath of kings of Pontus, 70 #. f. Sicias, 99. obiécere, 97. obteciemus, 98. ob:8cit, 97. olxos (ofcnua), meaning of, 11. Old temple, #.c. Erechtheum, 8 #., 18-23. passages referring to, 8#.,9#., 26f. bwricber (dricw), meanings of, 5-7. Opisthodomus, views on, I-3. a separate building, 3-17. passages relating to, 3-5, 12, 13, 14, 16f., 45, 46-48. situation of, 17 f., 38-40, 48-53.
General Index.
Opisthodomus — continued. the restored Opisthodomus of the Hecatompedon, 39 f., 44-48. meaning and application of the word, 38, 41f., 51-53. trustworthiness of tradition about, 41-43. other opisthodomi, 15, 40, $1 #. Ovid, comp. of sacto, 141 ff.
Papas, old Phrygian god, 64 #, 66 #. Parthenon, passages referring to, 9 #., 10 #. not the temple of Athena Polias, 30-33: earlier Parthenon, 22 f., 27.
Tlar-dva:ad, 57 7.
Petersen, on Opisthodomus, 2. on ‘old temple’ and temple of
Athena Polias, 19. on age of Hecatompedon, 24 a.
Ilerpaeirov, Mnyds, 71 7.
Phaedrus, comp. of sacso, 145 f.
daprdxov, Mie, 70 n. f.
Philiscus, trag., 78 n.
Phrygian, ancient religion, 64 f., 72 f. priests in Athens, 68 #. ‘Phrygian’ used in geographical
sense, 74.
Plato, Homeric quotations in, 153 ff.
Plautus, comp. of sacso, 130 ff.
Pleiad, Alexandrine, 78 f.
Ilokds, application of, 36-38. See also
Athena Polias.
Pompei, Pompetis, etc. 93 ff.
Pom peti, 94.
Pom petius, etc., 108, 110.
Popular etymology, 66, 73.
Praenomen, encroachment of Roman,
59 %. Prepositions w. sacto, 87 ff., 104 ff., 120, 121 ff. short before -scio, 99 fff. contracted w. -sc#o, 113 ff.
243
proiciet, 116.
prosectet, VII.
protest, 112.
protecttad, 110.
Propertius, comp. of taczo, 138.
Quantity of prepos. before sacio, 87 f., 99 ff., 104 ff.
Ramsay, W. M., 57-59, 62, 64 #., 65 ##.£., 73 ¢.
resce, 116.
veicere, 114, 116.
rétctat, 114.
réicit, 116.
rétcss, 114.
veiéctl, 112.
Religion of eastern Lydia and western Phrygia, 63 ff., 73 f.
Rhea, 64 #.
Ribbeck, on -s#czo etc., 85.
on disicto, 123 £.
- Ritschl, on -sscéo etc., 84.
Sabazius, 65 1., 73. relation to Attis and Mén, 65 . Zeds LaBd fos, 62 2., 73. TaBdfia, 68 x. Scenic poets, Roman, comp. of sacto, 1 34 f. Schmitz, W., on -ssczo etc., 86. Seneca, comp. of sacio, 146 f. Serenus Sammon., comp. of éacio, 151. Silius Italicus, comp. of sacio, 148 f. Sophocles, Aristotle’s quotations from, 156 f. Sosiphanes, not tragic poet, 78 #. Sositheus, 73 #. Sozon, 65 #., 73. Statius, quantity before -sczo, 102. comp. of sacio, 149 f. sibictt, 87. subidcere, 98. Sulla, era of, 57, 61 f., 73.
244
obpBus, 56, 61. cup flees, lepd, 72 #. Syncretism, 65 #.
Taiée (Tiavat = Tiamat), 68 x. Tehém (= Tham-te, Tiamat), 68 #. THpeia Gar, evy hy, 72 #. Terentius, comp. of sacto, 133 Thamte (Tiamat), 68, 70 f. Theocritus, not tragic poet, 78 #. Tiamat, 68 f.
without sex, 69 #. Tidpou, 56, 68 ff.
significance of (cara 6émos), 68. See
also under Mén Tiamu.
Tibullus, comp. of sacto, 138. Tisaly, Arel, 72 f. frastcere, U1. bratceret, 113 frarécit, 112. Trotia, 110. Trotiugenas, 94, 108. réxy Baciréws, Mhy Daprdxov, 70 n. £.
General Index.
' Vahlen, on -sicso etc., 85.
Valerius Flaccus, comp. of zaczo, 147 f. quantity before -sczo, 102. Variants, Homeric, antiquity of, 174 ff. Varro, comp. of azczo, 1 36. Vergilius, comp. of saczo, 1 39 ff., 145- Votive tablet to Artemis Anaitis and Mén Tiamu, 55 ff.
Wagner, on -stczo etc., 84.
White, J. W., The Opisthodomns on the Acropolis at Athens, 1 ff.
WRIGHT, J. H., A Votive Tablet to Artemis Anaitis and Mén Tiamu in the Boston Museum of Fine Arts, §5 ff.
Zenodotus, date, 78.
Zeus, 64 #. ‘Acas, 69 n. Macdadarnrés, 72 . bts. Lapdtios, 73 #. ourhp, 64 2”. Tesaios, 72 #. f.
INDEX OF CITATIONS.
Achilles Tatius (iii. 6), 42 ”., 51 #. Aelian, Vat. An. (vi. 49), 9”.; Var. Hist. (ii. 9), 36 #. Aeschines (ii. 147), 31 #., 36..; (schol. li. 147), 10 #., 30, 35- Aeschylus, Supp/. (494), 8 ». Anecdota Bekkeri (i. p. 283, 15), 9 #.; (p. 286, 26), 42 #., 43 My 52 #.; (p. 306, 7), 10 #5 (p. 345, 25), II #. Anecdota Paris. Crameri (i. p. 6), 77; (iii. p. 283), 60. Anthol. Pal. (vi. 2), 9 #.; (xii. 223), 51 #. Apollodorus (iii. 14), 9 #. Apollonius, Vst. Aesch. 36 n. Appian, Bell. Crv. (i. 20), §1 #. Aristides (i. p. 548, 14 Dind.), 9 #., 15, 40 #., 51 .; (schol. iii. p. 319 Dind.), 35 #. Aristophanes, Zg. (schol. 1169), 9 #., 10 #., 30, 35, 36%. Lys. (schol. 273), 8 »., 22 #., 27 #.; (schol. 759), 9 #. Pax (schol. 605), 8, 9 #. Plut. (1191-93), 4, 14 #.; (schol. 1191), 4,8%., 42”.; (schol. 1193), 4, 7 8.4 11, 29 2., 37 9., 38, 41, 420. Aristotle, De Anima (i. 2), 230. De Anim. Gen. (v. 5), 233: Hist. An. (vi. 24), 9 #., 32. 2.; (Vi. 28), 223, 228. De Mot. An. (4), 232. De Part. An. (iii. 10), 224. Eth. Exud. (iii. 1), 228; (vii. 1), 159, 234; (vii. 2), 160; (vii. 10), 169. Eth. Nic. (i. 2), 169; (ii. 9), 216; (iii. 11), 222, 229, 230, 231; (vii.
Aristotle — continued. 7), 225; (vii. 14), 171; (ix. 1), 169; (ix. 9), 156; (ix. 10), 171; (x. 10), 221.
Fragm. (143), 234; (159), 230; (402 Rose), 10 #.
Mag. Mor. (ii. 11), 225; (ii. 1§), 156.
Metaph. (i. 4), 172; (iv. 5), 161.
De Mundo (6), 235-
Occ. (i. 4), 170.
Phys. Ause. (iv. 1), 172.
Poet. (22), 225, 234, 235; (25), 224, 225, 227, 231, 233.
Pol. (i. 2), 157; (i. 4), 221; (i. 13), 160; (iii. 14), 226, 227; (viii. 3), 229>2 36.
Probl, (xxiii. 23), 223; (xxvi. 31), 234; (xxx. 1), 232, 236.
Resp. Athen. (44), 48 n.; (47), 10”.
Rhet. (i. 7), 2333 (i. 11), 236; (i. 13),
158; (i. 15), 158; (ii. 2), 222; (ii. 9), 228; (ii. 21), 160, 161;
(iii. 4), 220, 231; (iii, 6), 157; (iii. 9), 170; (iii. 11), 226, 235; (iii. 14), 156; (iii. 16), 160; (iii. 17), 160, 225.
Soph. El. (4), 231.
De Xenoph. (1), 172.
Berosus, af. Euseb. Chron. (i. 16), 71.
Caelius Antip. af. Non. (p. 89, 6, M), 113.
Caesar, 2. G. (iv. 28), 113.
Cassiodorius (p. 206), 94, 108.
Catullus (Ixiv. 370), 111.
Cicero, Div. (i. 48, 106), 97.
246
Clemens Alex, Proty. (iii 45, p. 13 Syib.), 9 *, 10 %.. 30, 31, 35) 57%: (iv. 47, p. 13 Sylb.), 37 =; (52. p- 15 Syib.), 36 «
Demosthenes (xiii. 14). 3 14%; (schol xii. 14), 4, 5 #; (xvii. 260), 68; Evil. 23.4), 68; (xix. 231), 67; (xxii p- 590), 48 «; (schol xxi 13), 9%, 10m. 35%; (xxiv. 136), + r4".; (schol xxiv. 136) 5 7 ©. $3 51%
Ps. Dicaearchas, fray. (1), 10 =.
Diodorus Siculus (iii. 5S, 4), 65 £; (xiv. $t), 5x =
Diomedes (p. 428. 10), 108.
eet
Dionysius Halic.. fat. (xiv. 2).9 a. IGM |
Draarch. (3)> 9 Bs 2
Etymol. Mag. s.r. davrov, tt =; "Exe-
réuwedor, 9 8; Crusrarys. 48 %.; | "EreoBourdda:, 36 =.; érurdédouos, | § w., $2 6, $5 By §2 8.5 Opnbs, IB. |
Eudocia (340 Flach). 13 Euripides, Hec. (864 f.), 161. Hf. F. (t 338), 157- Iph. A. (1400). §57- Iph. T. (727)» 157- Or. (667), 156. Phoen. (539 f-). 159 Tro. (969), 160; (1051), 160. Eustathius, //. (xxii. 451), 9 #-. 10 #., 30, 31, 36; Od. (i 356), 10 #.; (L 357): 9 8+ 30 31, 36, 37%; (xi. 634), 37 *-; (xvii 455). 45 *.
Gaius, /nst. (iii. 119), 91. Gellius (iv. 17), 84, 88, 92, 119.
Harpocration, s.v. édrwOd8opos, 3, 12; Tlok@yrwros, 13, 14; wpovala, 8 n. Herodotus (i. 47), 22 #.; (65), 22 #.; (ii 141), 22 #.; (143), 22.%.; (Vv. 72), 10 %., 22 #.; (77), 22 4-5 23%;
Index of Citations.
Herodotus — conttnned.
(go), 10 «., 22 #.; (Vi 134), 22%; (vii. 140), 22; (Viti 37), 8 =; (41), 10 @, 22 m, 23m; (51), ION, 22 @.. 25 39 @; (51-55) 25 and m.. (53)> 108, 15 *. 22 a. 25 and #-: (54). 10 @, 22 @., 25; (55), 9%, 10 %., 238, 25 and =, 26, 392. 45%
Hesiod. O. D. (25 f.), 162; (40), 162; (r2r f), 1635; (265£), 170; (287 ff.), 165; (289 ff.), 166; (293 ff), 169; (361), 167; (370), 169; (456), 162; (699). 170; (715), 171; (763 f.), r7 4.
Theog. (116 f€.), 168, 172; (337) 168.
Hesychius, sz. Gdvrev, 11 2.; Aldous Suudés, 9 2, 10 =; 'Exnrévredes, 9 #.; Oprarpés, 11, 14 2.; Myve- yoprys, 67; eixeupdy bgur, 9 #., 10 &.; Grie@bdouos, 5; oqads, II x.
Himerius, Eci. (v. 30), 9 ®, 10 8, 30, 3h 35
Homer, /7. (- 169-171), 195; (@ 15), 230; (32). 231; (196), 222; (391- 393). 226, 227, 231; (546-52), 20 and =. 21, 24 .; (547), 184; (549). 9 %.; (iit 8), 196; (iv. 84), 196 ; (218), 196; (431), 196; (¥. 223), 188: (vi. 93), 20 %.; (200- 202), 232; (211), 188; (274), 20 m.; (308), 20 =.; (vil. 64), 223; (360). 185; (vii. 21 f.), 232; (83 f.), 233; (107), 188; (162), 197; (548-552), 208; (ix. 308-314), 191; (357-363). 197; (385-390), 226; (497-501), 189; (539 f-), 223; (592-594), 233: (650-655), 193; (x. 11-13), 227, 233; (224), 197; (252), 2243 (332)> 2345 (457) 224; (485 £.), 220; (xi. 502-537), 175; (542 f.), 228; (630), 198; (639 £.), 198; (729), 20%; (xil.
Index of Citations.
Homer — continued.
200-207), 199; (234), 185; (311), 197; (xiv. 96-102), 199; (217), 225; (xvi. 112 f.), 200; (433 f.), 189; (xvii. 265), 234; (446 f.), 193; (xviii. 96), 200; (104), 200; (107), 234; (376), 221; (xix. 9zf.), 201; (xx. 217 f.), 193; (241), 188; (272), 225; (xxii. 98-100), 228; (168 f.), 201; (507), 201; (xxiii. 335-340), 202; (xxiv. 10-12), 203; (80-82), 203; (525 f.), 182; (527- 532), 204.
Od. (i. 352), 205; (iii. 382), 20 #.; (418), 20#.; (iv. 204), 225; (302), 7 %., 52%.; (567), 2345 (764), 202.; (v. 193), 187; (vi. 42-45), 2353 (ix. 8-10), 205; (114), 221; (190 £.), 223; (515), 2355 (x- 495): 206; (xi. 489-491), 182; (569), 190; (598), 235; (xii. 219 f.), 216; (schol. xiv. 533), 36.; (xv. 245f.), 206; (400f.), 236; (xvii. 218), 194, 225; (322 f.), 194; (385), 236; (xix. 122), 236; (xx. 259), 225; (351-357), 207; (xxii. 329), 224; (xxiv. 6-9), 190; (318 f.), 222.
Horace, Saé. (i. 6, 39), 116.
Inscriptions, AQn»a ii. (p. 627), 3 ”.;
44%. Bull. Corr. Hell. vi., viii., xiv., xv., II 2.
CIA. i. (32), 16 and #., 40 #., 46, 47; (60), 8 #.; (93), 82, 23 2. 27 #.; (109), 16; (146), 10 #.; (157), 10 .; (158), 10 #.; (159), IO #.; (161-175), I #., 46 ”.; (191), 16; (273), 16; (322), 8 #., 21 #., 3 n.; ii (74), 8 2, 26 ”.; (163), 8 ., 26 #.; (332), 8 #., 10 #., 30, 32, 33; (464), 8 #., 26”., 30 32, 335 (652), 10 #., 17; (660), 17; (672), 8 #., 26 %.; (685), 17;
247
Inscriptions — continued.
(720), 17; (721), 17; (733), 8 #., 26 #.; (751), 8 2.; (758), 82.; (829), 8 #.; iii. (776), 8 #.; iv. (1, p- 3f.), 8 »., 12, 18, 23.”., 27 #., 40”., 45 and #., 49 and #.; (p.74 ff.), 8 #.; (p. 137 ff.), 3.., 10%, 22 %., 39", 44and#.; (p. 168 ff.), 16.
C/G. (6280), 9 #., 20 #.
AeArlor Apx. 1890 (p. 92 ff.), 37.
Dittenberger, SIG. (384), 12 #., 45%.
Kasbel, Ep. Gr. (1046), 9 #., 20 #.
Mitth. d. Inst. Athen, viii. (p. 59), 8 #., 10 %., 30, 32, 333 xii (p. 39), 12 %., 45 %.
Sits.-Ber. d. Berl. Akad. 1887 (p. 1201, 45), 17.
CIL. i. (198, 50), 96, 100; (577), 100, 114; ix. (782), I10.
Laberius (119, R), 116. Lexicon Patmiun, s.v."Exaréuredor, 9x. Livy, i. (40, 7), 1125 (41,1), 1125 (48, 3), 112; x. (8, 3), 97; (37, 14), 98; xxii. (19, 3)» 98; (37, 9), 1133 xxvi. (19, 2), 98. Lucian, Dsal. Meret. (vii. 1, and schol.), 36 x. Fug. (7), 38 #.; (schol. 7), 4%., 42 #., 51 7 Hdt. (1), 38 2-5 $1”. Mort. Pereg. (32), 38 #., 51 #. Pisce. (21); 29, 31, 34, 35 %. Sym. (32), 36 #. Timon (53), 14, 16 #%.; (schol. 53), 43 and n. Lucilius (xxix. 106 Miill.), 110. Lucretius, i. (465), 943 (477): 943 iL. (951), 111; iii. (523), 111; (877), I1§; iv. (1272), 115.
Marius Victorin. (p. 67, 17), 88, 92.
Naevius (94 R), 99.
248
Ovid, F. (iv. 709), 116.
fb. (§5 f£.), 80; (schol. 315 f.), 81; (531 £.), 80. M. (xi. 386), 112.
Pausanias (i. t. 3), 7; (3- 3), 6 #3 (17-
2-4), 14 #.; (18. 6), 6; (19. 4), 6; (24. 3), 9 #-; (24. 5). 6 #.; (24. 5 8), 9 #3 (26. 5), 4 Mp 9 My 31 7, 35; (26. 6), 21 #., 35 39 %.; (27-1), 29, 35) 45 %.5 (27. 1, 3), 9-5 (27. 2, 4), 9 4-3 (27- 3)s 9 Ms 30, 35; (40. 4, 5), 6; (ii. §. 1), 6; (11.1), 73 (13-7))75 (20.7), 6 #.; (29. 11), 6; (31. 3), 6; (iii. 15. 1), 73 (16. 6), 6; (17. 5), 73 (iv. 31. 11), 63 (v. 10. 8), 6 #.; (10. 9), 38 #., 51 2.; (13. 1), 38%, §1 #.; (15. 3), 38%. 51 #.; (15.7), 7; (16.1), 38 %., 51 #.; (17. 9) 73 (19. 6), 7; (20. 2), 6#.;
(vi. 5. 6), 6 #.; (10. 6), 7; (19), -
14 #.; (25. 2), 7; (viii. 9. 6), 6; (14. 10.) 6; (22. 7), 73 (30- 7)» 6; (30. 8), 6; (45. 7), 6.5 (ix. 4. 2), §2 #.; (10. 2), 8 #.; (x. 9.9), 73 (11. I, 2, 5). 4%; (19. 4), 6 .; (19. 10), 7; (26. 5), 6 ».
Philochorus, frag. (146), 9 #., 29, 30, 34.
(See also 8, 9 and #.)
Photius (5.v. drieObdopu0s, 4, 42 #., 52 2.;
TloAvyvwros, 13; Talat, 10 #.
Plato, Adc. i. (132 A), 184.
Alc. ii. (140 A), 197; (149 D), 208.
Apol. (28 D), 200; (28 C), 200.
Axtoch. (367 D), 182, 193; (368 A), 206.
Conviv. (178 B), 168; (195 D), 201.
Crat. (392 E), 201; (397 E), 163; (402 B), 167; (428 A), 167.
Epist. vii. (344 D), 185.
Hipp. Min. (365 A), 191; (370 B), 197; (370 C), 195; (371 B), 192.
Jon (537 A), 202; (§38C), 198; (538 D), 203; (539A); 207; (539 B), 198.
Index of Citattons.
Plato — continued. Lach. (191 A), 188. Leges (681 E), 193; (706 E), 199; (718 E), 166; (777 A), 194.
Lysis (214 A), 194; (21§C), 162.
Meno (100 A), 206.
Minos (319 D), 190.
Phaedr. (266 B), 187.
Resp. (364 B), 67; (364 C), 165;
(364 D), 189; (379 C), 204 ; (379 E), 196; (386 (C), 182; (386 D), 206; (387 A), 190; (388 A), 202; 388 C), 189, 201; (389 E), 196; (390 A), 205; (405 D), 198; (408 A), 196; (424 B), 205; (466 C), 162; (468 D), 197; (469 A), 163; (545 D), 200.
Soph. (268 D), 188.
Theaet. (207 A), 162. Plautus, 4s. (814), 99-
Mere. (932), 99-
Ail. (112), 96; (623), 97-
Poen. (tt74), 96.
Rud. (769), 99.
Truc. (298), 96; (659), r10. Pliny, . 7. (vii. 2), 98; (xiii. 129), gr. Plutarch, Crm. (5), 9 ”., 22 #.
Demet. (23), 15 and #., 40 ., 46 n.,
5t a.
Quaest. Con. (ix. 6, p. 741), 9 #-
Soll. An. (13, p. 970), 9 #.
Them. (10), 9 #.
Vit. X. Or. (843), 3U #.
Vit. Mar. (26), 56.
Pollux (i. 6), 38, 51 #.; (viii. 96), 48 #.; (ix. 40), 5 #., 53 7.
Priscian, vol. II, K. (p. 14, 10), 94 #.3 (Pp. 13, 27), 109; (Pp. 14, 14). 109, 118; (p. 126, 18), 89; (p. 303) 5), 1103 (p. 34915), 97; vol. ITI, K. (p. 56, 18), 123; (p. 467, 15), 109.
Probus, (p. 104, 22), 94; (p. 221, 8), 108; (p. 257, 17), 108.
Index of Citations.
-Quintilian (i. 4, 11), 88, 107.
Scholia, see under Aeschines, Aristides, Aristoph., Demosth., Homer, Lu- cian, Ovid.
Seneca, Phoen. (426), 116.
Servius, ad Aen. (iv. 549), 89, 119; (x. 473), 119; de Final. (p. 450, 12), 103.
Sophocles, 4%. (293), 160.
Ant. (223), 156; (456 f.), 158; (911 f.), 160.
Elec. (256), 161. : Statius, 7hed. (iv. 574), 116; (vi. 770), 112.
Strabo (ix. 16, p. 396), 8 #., 10 #., 20, 21, 29, 33 #-» 345 (x. 3, 18, p. 471), 68; (x. 3, 19, p. 471), 68; (xiii. 4, 5, p. 625), 63; (xv. 3,15, p- 738), 58; (xvii. 28, p. 805), 52 #.
Suidas, s.v. Advroy, 11 #.; ‘Exarduredos yews, On.; émurrdrns, 48 #.; Mn- vaytprns, 67; dri Ob8ou0s, 4; Ilodb- yrwros, 13; onxds, 11 9.
249
Terence, Ad. (710), 97.
Themistius (xv. p. 234 Dind.), 51 #.
Theocritus, Jd. (xx. 15), 230; (xxii. 129), 230.
Thucydides (i. 126), 10 #.; (ii. 13), 46 ”.
Tzetzes, in Lycoph. 1. (p. 263), 76 #.
Valerius Flaccus, (vii. 514), 116. Varro, Ling. Lat. (v. § 160 Spengel), 38, sin. ap. Non. (p. 452, 9 M.), III. Velius Longus, (p. 54, 16), 107; (p. 54, 20), 118; (p. §5, 2), 108; (p. 72, 4), 118. Vergil, Aen. (v. 776), 112; (vi. 421), 97; (viii. 428), t1r f.; (ix. 634), 112; (x. 753), 111 f.; (xi. 642), 112. Cirss, (118), 116. Ecl. (iii. 96), 116. Geor. (i. 333), TI.
Xenophon, Hellen. (i. 6.1), 8 8.3 (ii. 3 20), 9 #., 32 #.
mi 3 6305
007 2b 454
aura
| a
s6eT
ASOTOTEUd Twoysseto ut setprys prvarey
“ga Tesoot Agteueatun pawasey = 9%,