Department of the Classics, Harvard University Virgil's Ecphrastic Centerpieces Author(s): Richard F. Thomas Source: Harvard Studies in Classical Philology, Vol. 87 (1983), pp. 175-184 Published by: Department of the Classics, Harvard University Stable URL: http://www.jstor.org/stable/311257 Accessed: 10-11-2015 20:31 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact
[email protected].
Department of the Classics, Harvard University is collaborating with JSTOR to digitize, preserve and extend access to Harvard Studies in Classical Philology.
http://www.jstor.org
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
VIRGIL'S ECPHRASTIC
CENTERPIECES
RICHARD F. THOMAS
THE
ecphrasis, the exposition, that is, of a work of art within poetry, is marked by a feature as enduring as the tradition itself:' with varying degrees of precision, the poet was concerned to situate and relate to each other the details appearing in the work. This holds for real as well as "imagined"2 ecphrases. Later examples of the tradition display the most fastidious attitude toward this practice. So Ovid's Minerva, in contest with Arachne, wove a scene into each of the four corners of her tapestry: quattuor in partes
certamina work
with
oras (ioi).
she surrounded the quattuor addit (Met. 6.85). Appropriately, an olive wreath for a border: circuit extremas oleis pacalibus
Arachne responded with a border of flowers and ivy: ultima pars telae, tenui circumdata limbo, nexilibus flores hederis habet intertextos. (127-128)
Indeed, from Homer's shield on, it is the edge or border of the work of art that is most consistently defined. The shield begins (II. 18.483) and ends with the Ocean, which serves as boundary both to the account and to the world of the object: l Ev
T rTlOEL7Toraolo
t lya UoOvoS~'SKEavoLO vruvya 7Tdp 7TUadr77vCoKEOS 7TVjKa 7OL7Troo.
(607-608)
' Still the best treatment (and indeed the only comprehensive one) of ecphrasis
is P. Friedliinder, Johannes von Gaza und Paulus Silentiarius (Leipzig and Berlin 1912) I-103. Also, on the traditional nature of such descriptions, cf. W. Biihler, Die Europa des Moschos, Hermes Einzelschriften 13 (1960) 85-1 o8. 2 j. J. Kakridis ("Erdichtete Ekphrasen. Ein Beitrag zur homerischen Schild-
beschreibung," W.S. 76 [1963] 7-26) devised the term to distinguish the more complex instances of the tradition. In fact, of course, in that they are poetic constructs, all ecphrases are in a sense "imagined."
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
176
Richard F. Thomas
The Hesiodic shield ends on the same note (Asp. 314-315), and, presumably in part under the influence of these seminal instances of the tradition, subsequent ecphrases followed suit.3 Relative position is also prominent. Even where no overall structure is intended or can be inferred, the poet may define features or details in his description in relationship to other details. We thereby acquire an image of vignettes, if not of the entire structure. On Homer's shield armies surround cities (Trqv8' EtTipqv7TdALv UL6w aro Aacwv, II. 18.509), aTrpdrTTOL reapers are in distinct groups, and are followed by children (T7pEds' ap' atLaAAoSETrpEST
adrpo*Ta
aOE/7lraiSE
8paypLEUovrEs ....
554-555),
a the vineyard isOEaraaav" encompassed by a ditch KrTETOV, 564), 8i KuavE"qv 8' IEv group (TroLLV youth plays the lyre in the middle of a (&U_ 7rTaiS /aLoEOLTL /l'EpdEV KLOaPLIE,569-570), a pasture,river and reed-bed ALydE? d6pptkyyL ~ all lie adjacent (voytdvsE/ 7rrap posavov 8ovaK7La, r dp TrroTratkOV KEAo80VTra, 575-576) - and so it goes on. We have confined ourselves to selective examples from a single ecphrasis, but they are sufficient to demonstrate that the poet writing in this tradition has a concern for defining structure and relative position in the arrangement of the individual vignettes of his work of art.4 Within this general structural awareness, a specific position is clearly of potential significance - the center. The center of an object in effect defines, or permits visual reconstruction of, the work of art. The extended ecphrasis deals in two types of centrality: the middle of the entire work, and that of groups or scenes which form a part of the whole. For both types there is an introductory, formulaic tag: iv ~caac (E'vt•aaoEs) and in medio. The second type of medial reference (the central object of a group within the work) occurs frequently in the larger ecphrases;5 how3 Theocritus' cup has a lip of ivy (rd TorTL iV XEldAr KLatOr, 1.29), tap!jEhrat~4d0OL while Jason's mantle in the Argonautica has purple borders (aKpa 8 TrropvpEl? 7T"VT?7 7TAEV, 1.729). So the last element of Europa's basket: T ~ rdT waao oaEi ,dovarA rT TL cwKvaAogvvgul/XpvauEovraAcipoto rTEplaKETTE XE~Ata rapooLs, Mosch. Eur. 6o-6i (cf. Bdihler [above, n.i] io8 n.2). Finally, Virgil's cup: et nobis idem Alcimedon duo pocula fecit / et molli circum est ansas amplexus acantho, Ecl. 3.44-45. d & KadXrpOt/ OLOLO, 4 Examples, again selective, from other authors: Ldyay A s, Lg Hes. Asp. 172-173, 7rapa SJ AdiJEL re 'dPoo rE/EcJTIraaav, Asp. 195-196, 7rp 6' dv6pv, Asp. 270, irapA 6E'U tLLv 'AxA; ELcT7rKEL,Asp. 264, Trapd 8' Ev7rupyog 7or6'AhL Theoc. 7v, Asp. 296-297; 7rdp6E o cv6pSE/. . . VELKELOUoE7TEEorL, OPXOg/XPoVEO 6S' L 6AE ETdohAag, Apoll. 48; d ripywro 1.33-35, 6v' 8rl/qKEo70 tumidi late maris ibat imago/aurea, vwv interAw•TEKER, Virg. Aen. 8.671Arg. 1.736-737; haec 672, et circum argento clari delphines, 673. Cf. Hom. II. 18.507-508 KEi70 )' ap' LvpaaoLot 8co Xpvaoo rdaavra, 569-570 6' KtOdpt~E;Hes. Asp. 201-202 Ev gbOppLLyyLALYEY7/LM-EPdEV roimUv6' Ev lEaoootorLdTiE zOS Ka / KLOdpLtEA yE /ipav / L(' ov 7ToAAoL UVLg, [209-210] dpa tk Arro in mediis patrio vocat agmina sistro, 700 ECpl•p•Ev Virg. Aen. 8.696 regina aiVroi/ medio in certamine Mavors. saevit S6EA•ve;
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
Virgil's Ecphrastic Centerpieces
I77
ever, the history of the medial object of the entire work is more complex, and it is this type which will concern us. The Homeric shield has no centerpiece, that is, there is no item specified as being central, and the expression E'vpEuoooLUt, although it occurs twice with reference to the middle of groups within the shield,6 is not applied to any central object.7 This may be related to the fact that in general this shield defies complete definition. This is not the case with Hesiod; indeed, the first figure mentioned occupies the center of Heracles' shield: Evp/.aLc
a' aaivstavros~ Ei7V Pof
47rmaAwL oaCUoLCo7pt'
o0TLtbaTELOS~,
Aap7TopEvotaL 8EopKwS9.
(Asp. 143-144) So too Apolloniuss refers to the center of Jason's mantle, although only to the fact that it is of a solid color, not to its containing any pictorial detail :9
87 ya/p aKpa
Eaaa TE"RVKTO, otL /LEvE'pEvO •LErUUl 7T AEV. 7ropuvpE•7) 7T•rVTr (Arg. 1.727-728)
As in the case of the Hesiodic ecphrasis, this detail is specified at the outset. The Virgilian ecphrasis displays a change in attitude toward this feature, specifically a heightening of interest in the structural importance of the central object. This change is possibly a symptom of the Alexandrian 6 II. I8.507, 569 (above, n.5). 7 C. H. Whitman (Homer and the Heroic Tradition [Cambridge, Mass. 1958] 205) claims that the sun, moon, and stars are at the center of the shield (with the activities of man between them and the Ocean); while this is logically probable, it is not specified in the text. 8 Between the archaic and Hellenistic periods there is one instance of an ecphrasis with a medial object. The first shield of Aeschylus' Seven Against Thebes, which receives only a brief description, has the moon as its center piece: EXEL S' 6rreppov
f'rr' o717'oEto-TLoS -d1E,
•apoLa ou0pavov Se 7ravEA7rlvog EIl LeurErtVLEVOV" AaLzrpa aoKEL, arpWv, VVK-ro6dOBaAto, 7TrrpE-TEL' 7TpeaolLaTOV AAEydvOt' 6rT'
(Septem 387-390)
9 The most recent examination of Apollonius' ecphrasis is H. A. Shapiro, "Jason's Cloak," TAPhA 110 (1980) 263-286.
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
Richard F. Thomas
178
stress on the importance of but it occurred, I am certain, without rEXv7, the aid of any actual Hellenistic model."' The cups of the Third Eclogue, the modest beginnings of Virgil's interest in ecphrasis, are dominated by their centerpieces: M.
D.
pocula ponam fagina, caelatum divini opus Alcimedontis, lenta quibus torno facili superaddita vitis diffusos hedera vestit pallente corymbos. in medio duo signa, Conon et - quis fuit alter, descripsit radio totum qui gentibus orbem, tempora quae messor, quae curvus arator haberet? necdum illis labra admovi, sed condita servo. et nobis idem Alcimedon duo pocula fecit et molli circum est ansas amplexus acantho, Orpheaque in medio posuit silvasque sequentis; necdum illis labra admovi, sed condita servo. si ad vitulam spectas, nihil est quod pocula laudes. (Ecl. 3.36-48)
40
45
In both the rival cups centrality is stressed; the first has Conon and Aratus" at its center, the second Orpheus. Theocritus' cup, it should be noted, has no item in the middle.12 But there is a further development with Virgil. The medial reference itself appears in the center of the passage: Menalcas has the phrase in medio midway through his description of the cup (40), but he is capped by Damoetas, for whom in medio occurs at the medial caesura in the central line of his five-line response (46). So the work of art and the poem or passage in which it appears visually mirror each other; the poem in a sense is the object. Virgil was not to limit the technique to Eclogue 3. The great shield of Aeneas, answer to that of Achilles and the most perfect Roman ecphrasis, for all its individuality is in form completely traditional. Like
10 Indeed the surviving examples of the tradition in Hellenistic literature show no real divergence from their archaic models. 1 So I read quis fuit alter? Here see D. O. Ross, Backgrounds to Augustan Poetry : Gallus, Elegy and Rome (Cambridge 1975) 24. 12 As often Virgil has conflated his sources here. The notion of an exquisitely finished cup as a prize in a singing contest is, of course, Theocritean - as is the reference to a trim of acanthus: et molli circum est ansas amplexus acantho, 45 (cf. Theoc.
I.55
raviT4
6'/"
/E '7gr
yps
However,
Virgil has
rEpLer1aaoLt •rao an emphasis on tKGavOog). centrality, absent from the incorporated into his description Theocritean lines.
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
Virgil's Ecphrastic Centerpieces
179
the shield of Iliad i8, it contains centerpieces of separate vignettes,13 but unlike that shield, and in line with the new practice, Virgil's shield also holds an overall centerpiece, the battle of Actium: in medio classis aeratas, Actia bella, cernere erat, totumque instructo Marte videres fervere Leucaten auroque effulgere fluctus. (Aen. 8.675-677) Critics since Servius Auctus have asked only one question of in medio: utrum clipeo an mari? By analogy with the instances in the Third Eclogue, and, I think, following sense, we should answer, along with Heyne and Wagner and Forbiger, "of the shield," 14 although, of course, the reference could be to both. Perhaps surprisingly, what critics and commentators have not noticed is that these lines occur in the exact center of the description of the shield (49 lines preceding them, 51 following). As was the case with the Third Eclogue, but on a much grander scale, the central item, along with its formulaic referent (in medio), is placed in the center of the passage. Ut pictura poesis: once again the poem visually reflects the work it describes. In each of these instances, then, we see that Virgil created the practice15 of referring to the medial item in the middle of the actual description, and that this practice establishes an approximation between the work of art and the poem in which it appears. A further example of this, somewhat different in nature, occurs in the proem of the Third Georgic. Virgil here describes, in ecphrastic style, the templum he is to build - a metaphorical construct standing for a future poetic project of 13
Above, n. 5. None of the numerous recent studies of the shield (for bibliog-
raphy, see R. D. Williams, "The Shield of Aeneas," Vergilius 27 [1981] I i) deals with this ecphrasis in terms of the linguistic and dictional tradition to which it belongs - essentially our concern. 14 Conington appears to have misread Forbiger, ascribing to him the view that in medio refers to the sea. a possible instance of this feature which may have 15 It is worth mentioning influenced Virgil. In Catullus 64, at the center of Ariadne's soliloquy (36 lines from the beginning and 35 from the end), the deserted heroine turns from her own plight to consider the progress of the departing Theseus: ille autem prope iam mediis versatur in undis, Cat. 64.167. Imaginary or not, the tapestry and its contents (unlike the psychology of the protagonists) will have been quite simple: on one side Ariadne looking out to sea (Thesea cedentem celeri cum classe tuetur, 53) - with Dionysus subsequently visiting her; on the other Aegeus, also looking seaward (cum primum infecti conspexit lintea veli, 243). Between the two Theseus: mediis versatur in undis. In that this instance appears in a speech, rather than in the description of the work of art, it is perhaps qualitatively distinct from the Virgilian examples. Nevertheless, the coincidence does seem worth mentioning.
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
Richard F. Thomas
180
epic proportions. Whether or not the reference is specifically to the Aeneid (which I personally believe to be the case) need not concern us here. This temple, which itself appears in the center of the Georgics, will have as its own centerpiece the triumphant Octavian:16 in medio mihi Caesar erit templumque tenebit (Geo. 3.16) Once elsewhere in the corpus of Virgil there is a description of a temple. The palace of Picus (Aen. 7. 70-191) is twice referred to as a temple (hoc illis curia templum, 174; tall . . . templo, 192). It may be no accident that this structure, purely by line counting, is situated at the exact center of Virgil's epic, and that in it also there is a centrally placed figure, Latinus: ille intra tecta vocari imperat et solio medius consedit avito. (168-I69) It is, I think, with certain expectations that we turn to Virgil's other developed ecphrasis, the description of the murals in Dido's temple (Aen. 1.466-493). Replete with scenes from the siege and fall of Troy,'7 it is unique in the tradition of ecphrasis: only here is the onlooker (in this case Aeneas) actually a part of the work of art which he observes: se quoque principibus
permixtum
agnovit Achivis (1.488)
Like the shield, this work also has a three-line central element, at least one which in the description is centrally positioned (13 lines precede it, 12 follow):'8 interea ad templum non aequae Palladis ibant crinibus Iliades passis peplumque ferebant suppliciter, tristes et tunsae pectora palmis. (479-48 1) 16 It may not be merely playful to point out that Octavian (iuvenem) occupies the center of the central line (42) of the First Eclogue. 17 Again, relative position is noted (although this is less pervasive, since the layout of the murals is linear, and therefore needs little elaboration): Pergama circum, 466; nec procul hinc, 469; parte alia, 474. Also, as with other ecphrases
(above, n.5), there is a central item to one of the individual vignettes: Penthesilea furens mediisque
in milibus ardet, 491.
is Perhaps more important, four scenes precede this one (Greeks fleeing,
Achilles routing Trojans, attack on Rhesus, death of Troilus) and four follow it (Hector ransomed, Aeneas, Memnon, Amazons).
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
18
Virgil's Ecphrastic Centerpieces
At first sight disappointing, and in a way that should arouse our suspicions. On these murals are battle scenes involving Achilles, the night raid on the camp of Rhesus, the death of Troilus, Hector's body despoiled and ransomed, Aeneas himself, Memnon, and Penthesilea with her Amazons. Yet in the very center, in the place of honor, comes the adaptation of, or rather reference to, a brief scene from Iliad 6 Hector's injunction that the women of Troy attempt to appease Athena with the gift of a 7riTrhAo(II. 6.263-31 ). In Virgil's adaptation there is no prominent character (Hector's part is not referred to), and in this, as in general importance, it appears distinct from the vignettes that surround it. R. D. Williams, in an article on the murals,'9 perhaps understandably, has little to say on these lines. One word should alert us: peplum. Elsewhere in Latin it occurs twice in Plautus,20 in the Ciris (21), and in Statius (Theb. I0.56), with reference either to the robe of Athena or of Juno.2' Servius is strict about the meaning: peplum proprie est palla picta feminea Minervae consecrata . hodie tamen multi abutuntur hoc nomine. (ad Aen. 1.480)22 In referring to this peplum Virgil, I believe, invites us to recall the literary history of the word, and the associations it necessarily conjures up. At the center of his ecphrasis he has placed an object which itself would traditionally contain ecphrastic scenes. This is certainly true of the 7TE7rhosof Athena (cf. Servius' picta) and is even implicit in the language of Virgil's Homeric "source": iv6' aav otE7TAoLhoT7 rarolKLha c 1tovo23 E TVWv'
pya yvvaLK(oV
'EKagI. 60EpE&-9pov 'AO-v9,
EL~paL`wEvq 09~ KAALaros•V 7TOLKlAtcLutLV 78
/LEyLtaro~,
dalrip 8' cc g7TErAapTrEv.
(Il. 6.289-290, 19 "The
Pictures on Dido's
Temple
(Aeneid
1.450-493),"
151.
20 Merc. 67, fr. dub. et susp. 3 Lindsay 21 On
293-295) CQ io (1960) 145-
(ap. Serv. Aen. 1.480).
this see R. O. A. M. Lyne, Ciris, A Poem Attributed to Vergil
(Cambridge 1978) lo9-I 22 For its subsequent
o. use merely for "upper garment"
see Manil.
5.392 (s.v.
OLD). 23 The provenance Dido's temple.
of Homer's
makes it appropriate rrETAhog
in the setting
This content downloaded from 128.103.149.52 on Tue, 10 Nov 2015 20:31:17 UTC All use subject to JSTOR Terms and Conditions
of
182
Richard F. Thomas
The Homeric text does not elaborate, but this is the language of ecphrasis - which will not have escaped Virgil's notice. Indeed, in Iliad 3 Helen is at work weaving precisely the objects which Virgil was to place on Dido's mural:
94Ey #lLaVE, 7 LTyaOVYrv 8L'rTAaKa7rTopqvpE"qv, TroAiagS' vE7TraaacEv ckOAovg TpcOwv
8' L7r7aro'd4wvKaL'Axa(cJ
XaAKOXLT7V•V. (Il. 3.125-127)
To return to the 7TrTAog, one parallel will suffice, a contemporary description of the robe at Athens; the Chorus of Euripides' Hecuba is considering its captive fate: j HaAA'80os
v
'v 7TMAEL
749 KKaAAL'SLqpovs (T 'A~a?81'OU 'A~aTag
valags E KpOKEW7iEVT 6E6oprLatl&pa ir
