This paper began with a seminar given in 1973 by David Gordon Mitten. His help and that of many others at Harvard, the University of Michigan and at the American School of Classical Studies in Athens has been of great value. In particular, I would like to acknowledge the helpful criticisms of Martin Robertson, Evelyn Harrison and Vincent Bruno; my obstinacies are of course my own.

The Niobid krater has been the object of perceptive discussion since its first publication. The works of earlier scholars referred to most often in the text are listed below and cited by the author's name alone:

Barron, J., ‘New light on old walls: The murals of the Theseion’, JHS xcii (1972) 20-45.

Christos, Ch., ‘Ho Polygnotos kai mia angeiographia me epeisodion ek tis Homerikis Nekyias’, AE (1957) 168-226.

Jacobsthal, P., ‘The Nekyia krater in New York’, Metropolitan Museum Studies v (1935) 117-45.

Jeppesen, K., ‘Eteokleous Symbasis’, Acta Jutlandica xl, no. 3 (1968).

Simon, E., ‘Polygnotan painting and the Niobid Painter’, AJA lxvii (1963) 43-62, with bibliography.

Six, J., ‘Mikon's fourth painting in the Theseion’, JHS xxxix (1919) 130-43.

¹ Louvre G 341: ARV ² 601.22, found at Orvieto: Helbig, W., Bulletino x (1881) 276–80Google Scholar. The vase and its painter are named after the scene of the killing of the Niobids on the reverse because of the ambiguity of the subject of the obverse, for which it often used to be called the ‘Argonaut krater’. The Herakles side of the krater overlaps both handle zones and shows greater care in the execution and so is clearly the obverse.

² Paus. x 25-31. See the discussions in Weickert, C., Studien zur Kunstgeschichte des 5. Jahrhunderts v. Chr. I. Polygnot (AAWB 1947 no. 8 [Berlin, 1950]) 9-14, 20Google Scholar; Simon 48, 51; Robertson, M., A history of Greek art (Cambridge 1976) 266–70Google Scholar.

³ Jacobsthal 124 despaired that an explanation of the scene would ever be found; other scholars have left the scene unidentified (see the bibliography in Simon 62). The lack of explicit clues, such as labels, may indicate that the scene imitates a model which was sufficiently well known to potential patrons that the painter did not feel the need to explain his scene (so Christos 279, and Jeppesen 27).

⁴ Henceforth, when different figures on the vase are referred to, the numbers assigned to them by Jeppesen, , Acta Archaeologica (Copenhagen) xli (1970) 158Google Scholar, fig. 3, here PLATE IIb will appear in parentheses. I would like to thank Prof. Jeppesen for providing me with this figure.

⁵ In Polygnotos' Nekyia at Delphi, Pausanias describes Antilochos as having one foot raised (x 30.3) and Hector sitting with his left knee clasped in both hands (x 31.5).

⁶ Zen. iv 28; see Wycherley, R., The Athenian Agora iii. Literary and epigraphical testimonia (Princeton 1957) 44 no. 96Google Scholar.

⁷ Among these are the four-section abdomen, the lined faces (6,10) and the three-quarter-view heads. (The reverse of the krater has one three-quarter-view head.) See the discussion in Barron 23-5.

⁸ The bibliography for the problem is listed chronologically by Simon 61-2. To be added to her list are: Löwy, E., Polygnot (Vienna 1929)Google Scholar; C. Weickert (n. 2); Christos (1957); Jeppesen (1968) and (n. 4) and Acta archaeologica xlii (1972) 110–12Google Scholar; Harrison, E., A Bull liv (1972) 390–402CrossRefGoogle Scholar; Barron (1972); Robertson (n. 2); Simon, E., Die griechische Vasen (Munich 1976) 133–5Google Scholar; Clark, R.J., Catabasis (Amsterdam 1979) 126Google Scholar; Kron, U., LIMC i (1981) 441Google Scholar, #229; Blatter, R., LIMC ii (1984) 597–8, #32Google Scholar.

⁹ Christos 179-95. Christos' interpretation appeared too late for inclusion in ARV ² or Simon's bibliography, and its length and publication in Greek have won it less consideration than it deserves. (It is cited only by Jeppesen [n. 4].) The argument presented here was substantially developed before Christos' article came to the author's attention. More recently, Christos' theory seems to have been altered along the lines presented here by Th. Karagiorga, AE (1972) 46.

¹⁰ Robert, C., Annali dell'Istituto di Corrispondenza Archeologica liv (1882) 280Google Scholar; Christos 182-5; Simon 49-50.

¹¹ Christos 182-93.

¹² Christos 179-81. Simon rejects a parallel argument, 46.

¹³ Originally identified by Gardner, E., JHS x (1889) 124Google Scholar; see the arguments in Simon 45.

¹⁴ As Christos 188 acknowledges.

¹⁵ Simon 44.

¹⁶ Harrison (n. 8) 394 implies this by pairing the goddess and the old man. Christos 192–3 says that the old man balances the warrior half hidden on the left (2), but this figure actually aligns with the horse's head.

¹⁷ Kenner, H., Weinen und Lachen (Vienna 1960) 38Google Scholar and Girard, P., REG vii (1894) 360–361Google Scholar discuss facial expressions.

¹⁸ Simon 47–48.

¹⁹ F. Brommer, Vasenlisten ³, 481-4; Hornbostel, W., Aus Gräbern und Heiligtümern (Mainz 1980) 126–7Google Scholar; Cressedi, G., EAA iii (1960) 217–19Google Scholar; Robert, C., Oidipus (Berlin 1915)Google Scholar passim. Oedipus appears occasionally with a beard before the sphinx (e.g on the well known kylix by the Oedipus Painter in the Vatican: ARV ² 451.1 and 1654; Robert, op. cit. 51, fig. 16) but he is always youthful. Particularly revealing here is the contrast between the youthful Oedipus with the sphinx and the shaggy Odysseus with Nausikaa on London E 156: ARV ² 1281 (recalls the Marlay Painter): CVA British Museum iv, pl. 34.1 (227). According to Höfer, O., Roscher iii, 735Google Scholar, the tradition of a wandering, elderly Oedipus dates from Euripides' Phoenissae at the end of the fifth century.

²⁰ See especially Touchefeu-Meynier, O., Thèmes Odyseéns dans Part antique (Paris 1968) 288–9Google Scholar; Schmidt, J. in Roscher iii 654–81Google Scholar; Paribeni, E., EAA vii (1966) 1046–51Google Scholar; Brommer, F., Odysseus (Darmstadt 1983) 110–11Google Scholar.

The problem of heroes without clear iconography is discussed by Robert, C., Archäologische Hermeneutik (Berlin 1919; reprinted, New York 1975) 39Google Scholar and more recently by Dusenberry, E., Hespetia xlvii (1978) 226Google Scholar. The Disney Painter oinochoe in New York, Metropolitan Museum 28.97.24 (ARV ² 1265.15 and 1688; EAA iii [1960] 141Google Scholar, fig. 172) presents a similar iconographical problem. A middle-aged archer with long hair and beard is shown alone. As Richter points out (Richter, G. and Hall, L., Red-figured Athenian vases in the Metropolitan Museum [New Haven 1936] 187, pl. 150, 152Google Scholar), his ‘untidy hair and ummartial looks suggest that this is no regular archer, but Odysseus’. Beazley seems to have originally suggested this interpretation (AV 447,9) and later added (ARV² 1688) ‘if Odysseus, an extract from a “slaying of the suitors”.’

²¹ According to Touchefeu-Meynier, (n. 20) 288, n. 7, Odysseus' pilos first appears in the last third of the fifth century BC. See Brommer (n. 20) 110-11.

²² Athens Agora P 18538, inscribed : ARV ² 611.40 and 1661 (manner of the Niobid Painter, ‘maybe an early, delicate work by the painter himself’); Para. 396; not in Touchefeu-Meynier. Corbett, P. (Hesperia xvii [1948] 189–90)Google Scholar suggests that this shows the hero during his mission to Skyros before the Trojan War; F. Brommer (AA [1965] 115-19) proposes Odysseus returning to Ithaka disguised as a beggar.

²³ Boston, Museum of Fine Arts, 34.79: ARV ² 1045.2 and 1679. This comparison is also made by Harrison (n. 8) 394 n. 29.

²⁴ Odysseus also stands with one leg raised in two later underworld scenes, the San Severo sarcophagus in Orvieto (Touchefeu-Meynier [n. 20], 140, no. 236, pl. 22.2) and a relief in Paris (Louvre 574: ibid. 137-8, no. 231, pl. 21.3). In both cases the hero faces to the right and so the left leg is against the background and therefore raised. A figure in a similar pose on a bronze relief is identified as Odysseus by Thompson, D., Hesperia xxxviii (1969) 242–51CrossRefGoogle Scholar.

²⁵ Compare the description by Pausanias (x 30.3) of the figure of Antilochos in the Delphic Nekyia. Many examples can be cited from later vase painting: Jacobsthal, P., Die melischen Reliefs (Berlin 1931) 190–2Google Scholar; Neumann, G., Gesten und Gebärden (Berlin 1965) 118–19CrossRefGoogle Scholar; and McNiven, T., Gestures in Attic vase painting (Diss. University of Michigan 1982) 58Google Scholar.

²⁶ Figure 5 also has his foot propped slightly. As will be seen below, this is in order to link him with figure 9, the youth with the helmet in his outstretched hand.

²⁷ So Six 132; Simon 50; Jeppesen 11. Harrison (n. 8) 392 says that the figure has not turned from the center, but is about to turn toward it. It is difficult to tell from the position of the feet just what he is doing, but I would interpret their position as showing that he is stepping up to the left. In this light, it is tempting to compare the description Pliny (Nat. xxxv 58) gives of a painting by Polygnotos that was in the Portico of Pompey, ‘in which there was some doubt whether he has depicted the figure with the shield as moving upward or downward’. (Trans., Pollitt, J. J., The art of Greece, [Englewood Cliffs, N.J. 1965] 96)Google Scholar.

²⁸ Ajax seems to be portrayed with a beard throughout Attic vase painting, whether playing draughts with Achilles, fighting, quarreling with Odysseus or falling on his sword. See in general, Friis Johansen (n. 31) 66, 172; Touchefeu, O., LIMC i (1981) 312–6Google Scholar; for the draughts-players, Schefold, K., JdI lii (1937) 68–70Google Scholar; on the suicide, Schefold, K., AK xix (1976) 71–7Google Scholar. Ajax has a beard on the Nekyia krater (PLATE IIId) and in the Delphic Nekyia (Paus. x 31.3). On a black-figure amphora by the Swing Painter in Munich (1494: ABV 308; CVA Munich vii, pl. 360) a figure in the underworld has been interpreted as Ajax turning away with balled fists.

Pausanias (x 26.3) mentions a snake on Menelaus' shield in the Iliupersis at Delphi and gives a dubious interpretation of it. No positive interpretation of this feature is possible: Chase, G., HSPh xiii (1902) 82–3Google Scholar; Robertson, M., ABSA 62 (1967) 10Google Scholar. A shield, though usually a Boeotian one, is an important part of Ajax's iconography in the Archaic period: Friis Johansen (n. 31) 66.

²⁹ Christos 185-6. See Touchefeu, O., LIMC i (1981) 256–74Google Scholar.

³⁰ Jeppesen (n. 4) 157.

³¹ So Johansen, K. Friis, The Iliad in early Greek art (Copenhagen 1967) 133, 178Google Scholar; Bocci, P., ‘Achille’, EAA i (1958) 25–33Google Scholar; Kemp-Lindemann, D., Darstellungen des Achilleus in griechischer und römischer Kunst (Frankfurt and Bern 1975) esp. 139–41Google Scholar; Kossatz-Deissmann, A., LIMC i (1981) 114–22Google Scholar and plates passim. Achilles is shown unbearded in contrast to the bearded Ajax on the latest (ca. 430 BC) of the series of draughtsplayers, a column-krater by the Hephaistos Painter (Berlin 3199: ARV ² 1114.9; Para. 452; JDAI lii [1937] 70Google Scholar, fig. 1; LIMC i (1981) pl. 100Google Scholar Achilleus 420).

That Patroklos undergoes a parallel change becomes evident from the discussion in Friis Johansen, op. cit., e.g. 230. The tradition is confused, however. Both heroes were shown as young and unbearded in Polygnotos' Nekyia at Delphi (Paus. x 30.3) and on a stamnos by the Kleophrades Painter (Villa Giulia 26040: ARV ² 188.63; see the interpretation in Friis Johansen, op. cit., 184-6, fig. 75). On the Sosias Painter's famous cup (Berlin F2278: ARV ² 21.1) Achilles is portrayed as unbearded while Patroklos has a sparse beard. Achilles has a sparse beard on the name-piece of the Penthesilea Painter (Munich 2688: ARV ² 879.1) but none on that of the Achilles Painter (Vatican 16571: ARV ² 987.1).

In the Symposium (line 180), Plato has Phaedrus argue that Achilles was the young, prettier ἐραοτής and not the έρώμενος of Patroklos, contradicting Aeschylus, so there was disagreement even in ancient times. See the discussion in Lenschau, T., ‘Patroklos (2)’, RE xviii 4 (1949) 2280–1Google Scholar.

³² Jeppesen 24, n. 19 and (n. 4) 160 demonstrates that he cannot salute Athena, but the actual object of his attention is not clear.

³³ The discussion of this episode by Clark (n. 8) 37-8 is especially insightful.

³⁴ Paus. x 29.1. For Theseus under the sea, see Bologna 303 by the Kadmos Painter: ARV ² 1184.6; Six 139-41; Barron 40-1; Robertson (n. 2) 256.

³⁵ That some artists followed a literary source closely is often assumed, but also well demonstrated, as in Friis Johansen (n. 31) 127, 188-91, 202-6, 226. These examples show a close dependence on the Iliad for the names and situations portrayed. See also Boardman, J., AK xix (1976) 11Google Scholar. Friis Johansen (n. 31) e.g. 190, 206, also gives instances where figures from a series of events are united in one scene to allow the artist to make a broader point.

³⁶ Webster, T. B. L., Der Niobidenmaler (Leipzig 1935) 15–16Google Scholar.

³⁷ Webster is followed by: Simon 43-4; Harrison (n. 8) 390; C. Picard, RA 1960, 110. For scholars who disagree with this opinion, see note 41.

³⁸ Silius Italicus, Pun. xii 784-5; Stat., Theb. viii 512-13.

³⁹ Ovid, Met. iv 432-80. Juno's purification after her return from the underworld recalls Apollo's fear of miasma from an encounter with death in the prologue of Euripides' Alcestis (22-3). Here we have a fifth century source, one closer in time to the Niobid krater, but one whose relevance is ambiguous. Apollo is worried about being inside with the corpse, suggesting that the rules of pollution affect him just as much as they do the average Athenian (Kurtz, D. and Boardman, J., Greek burial customs [London 1971] 146, 149–50Google Scholar). This is of doubtful relevance to the underworld, where even unburied corpses are left behind. In any case, while miasma is to be avoided if possible, it does have a standard remedy available in purification rituals. Compare Artemis' leave-taking with Hippolytos, Hip. 1437-8 and the Apulian vase in Taranto (RVAp i, 435.12a) where Athena watches with Achilles as Thanatos touches Memnon, who is accompanied by Eos.

⁴⁰ Compare Herakles' statement in Od. xi 623-6 and Helios' threat, Od. xii 382.

⁴¹ See Brommer, F., Herakles (Münster/Koln 1953) 43–5Google Scholar and in Studien zur griechischen Vasenmalerei (AK Beiheft vii) (Bern 1970) 50 n. 4; Felten, W., Attische Unterweltsdarstellungen (Munich 1975) 12–13Google Scholar; G. Roux, RA (1949) 904; and Beckel, G., Gotterbeistand in der Überleiferung griechischen Heldensagen (Bayern 1961) 17-18, 43–4Google Scholar. None of these sees a problem in Athena's presence in Hades.

An unidentified female figure in a Middle Corinthian version of the Cerberus scene has been identified as Athena: Payne, H., Necrocorinthia (Oxford 1931) 130Google Scholar, fig. 45c; Demargne, P., LIMC ii (1984) 958 #11Google Scholar.

⁴² Amburger, E. A., ‘Athena und Herakles in der Kunst der Antike’ (unpublished dissertation, Berlin 1949)Google Scholar as quoted by Rohde, E., CVA Gotha i, p. 44Google Scholar.

⁴³ As do Simon 46-7; Robertson, M., JWI xv (1957) 99Google Scholar.

⁴⁴ Altenburg 233: ARV ² 137.1; Hartwig, P., JdI viii (1893) 163Google Scholar with drawing, and Bielefeld, E., CVA Altenburg ii, pl. 67.2Google Scholar. On the traditions concerning the den of Cerberus, see Eitrem, S., ‘Kerberos’, RE ii 1 (1921) 278–9Google Scholar.

⁴⁵ A pelike in Kertch, not in ARV ² or Para.: Boltounova, A., Mélanges Michalowski (Warsaw 1966) 287–92Google Scholar, there dated ca. 375.

⁴⁶ Six 133.

⁴⁷ Felten (n. 41) 51–2; Christos 181; Simon 45; Webster (n. 36) 15.

⁴⁸ Odysseus only meets Herakles in Hades, because he was long departed before the Trojan War, as we know from the story of Philoctetes, who had inherited Herakles' bow and arrows and used them at Troy (Soph. Ph. 262; Apollod. iii 155.)

⁴⁹ See the discussion in Paus. i 17.4-6. Modern treatments are by Herter, H., RE² Suppl. xiii (1973) 1177Google Scholar, paragraph 102; Clark (n. 8) 128 and n. 7.

⁵⁰ Simon 46. Mingazzini, P., Atti AL ser. 6, i (1925) 413–90Google Scholar, illustrates many scenes where Herakles is wreathed. See also S. Karouzou, CVA Athens ii. III Hg. p. 10. Apotheosis is not the only meaning of such a wreath (cf. Wrede, W., MDAI(A) xlii [1916] 262–4Google Scholar), but it is the most likely here. Jeppesen 18 notes that the only other figure on the krater with such a crown is Apollo, on the reverse.

⁵¹ This suggests the melancholy mood of the scene which has been recognized by some scholars and denied by others. The problem is that sadness is difficult to isolate in a work so in line with Early Classical (‘Severe’) tradition.

⁵² Note that Hermes may easily have been added by the vase painter: Touchefeu-Meynier (n. 20) 286.

⁵³ New York, Metropolitan, 08.258.21: ARV ² 1086, 1; Jacobsthal figs. 6–10.

⁵⁴ So Jacobsthal 123; Richter, G., Attic red-figured vases. A survey (New Haven 1958) 130–1Google Scholar.

⁵⁵ See the list given by Oakley, J. in Ancient Greek and related pottery (Allard Pierson Series 5), ed. Brijder, H. A. G. (Amsterdam 1984) 125–7Google Scholar. Of these sixty examples only Bologna 298 (ARV ² 1018.62; Jacobsthal #9) and the krater from Pitchvnari in Georgia, Soviet (BCH xcviii [1974] 915–17Google Scholar, figs. 9-9a) have a divided scene in the top register. Even these are tightly connected by narrative. On the Nekyia krater, the placement of Hades at one end of the frieze and Persephone at the other argues for a comprehensive view.

⁵⁶ Friedländer, AA 1935, 31-2. Jacobsthal 130-1 n. 70, notes that the two appear in the Hell described by Socrates in Apology 41a.

⁵⁷ Jacobsthal 132; Friedländer (n. 56) 23–4, 29–32.

⁵⁸ Vienna, Kunsthistorisches Museum 1026: ARV ² 1087.2; Jacobsthal 132, figs. 14–15. Double register calyx kraters begin with the Niobid Painter (Jacobsthal 136–40), for example his krater in Boston, Museum of Fine Arts 72.850. The connection between double register kraters and monumental painting can best be seen in two kraters from Spina, both showing the Gigantomachy with very similar groups of figures, but in different formats. The fragmentary krater by a member of the school of the Peleus Painter (Ferrara in v. 2892, T300 VT) has the figures organized in registers, that perhaps by the Painter of the Woolly Satyrs (in v. 44893) spreads them across a ‘Polygnotan’ hillside: ARV ² 1041.7 and 1680, Para 446; Alfieri, N. and Arias, P. E., Spina (Munich 1958) p. 11. 66-7, 69–73Google Scholar.

It is possible that the Nekyia krater in New York may be copied from the same prototype as the Niobid krater, but this is difficult to prove. This was also suggested by Friedländer (n. 56) 23, n. 1.