A List of Names from the Potter's Quarter Boegehold, Alan L Greek, Roman and Byzantine Studies; Spring 1974; 15, 1; ProQuest pg. 25 Korinthiaka Alan L. Boegehold I A List of Names from the Potters' Quarter THE texts of SEG XI 191 and 192 are fragmentary lists of names that were cut after firing into the surface of a Corinthian black-glazed vessel. These inscriptions have been of extensive interest since Agnes N. Stillwell published the editio princeps in 1933.1 Parts of two sides of the vessel-which may have been a kotyle-are preserved, one of which (SEG XI 191-henceforth A) was found in a closed, Early Protocorinthian context, the other (SEG XI 192-hence­ forth B) in disturbed fill. Although the sherds do not join, they surely come from the same pot and preserve parts of the same text (see PLATE 5, fig. 1). The phenomenon that has generated so much interest is the style and sophistication of the letters, for, if it were true that the vessel was made some time in the years 750-725 B.C., the Greek alphabet-at least the Corinthian form of it-had to have been in existence for at least a generation before. No one, the reasoning goes, could form such graceful letters while still trying to understand and use an absolutely new mode of thinking and expression. The style of decoration and the skill with which the letters were incised have also prompted comment, since both seem by some currently accepted canons of stylistic chronology to be anomalous at such an early date.2 1 A. N. Stillwell, "Eighth Century B.C. Inscriptions from COrinth," AJA 37 (1933) 605-10 [hereafter, STILLWELL]. Since the lines are exceptionally clean, Mrs Stillwell (p.605) sur­ mised that "the vase was refired after the letters were cut." Under 25x magnification, however, the incised lines show a roughness that is not consistent with lines on other Corinthian vases that are known certainly [0 have been fired after the letters were in­ scribed. I wish to dedicate these notes to my teacher and friend, Sterling Dow, there being no special occasion to commemorate. only the wish to record grateful recognition of his inspiring and generous passion for scholarship. I acknowledge here with thanks helpful criticism and suggestions I have received from Vi. M. Calder III, Ronald S. Stroud, Leslie Threatte, Charles K. Williams II and William F. Wyatt. 2 For earlier discussion and bibliography, see L. H. Jeffery, The Local Scripts of Archaic Greece (Oxford 1961) 120-21, 130 no.1 [hereafter, JEFFERY]; R. Arena, "Le Iscrizioni corinzie 25 26 KORINTHIAKA I offer here a speculative arrangement of the text with some observations and questions concerning its contents and date. [---]. [---] [---]~~aVTaC : XaL [---] [---]fI<€aC : 'AvyapLOc [---] [---]aVFLOC : EOIo : Katv[LOc :J XaLpta[---J. If the slab of volcanic stone from Thera (SEG XIV 522), whose inscribed names were cut sometime between ca. 650 and 550, carries a king's name, the Malcko of that stone was a distinguished person. See Jeffery 319,323 no.5. The name Maleko has been thought to be Phoenician, but see now O. Masson, Melanges de linguistique et de philologie grecques offerts Ii Pierre Chantraine (Etudes et commentaires 79, Paris 1972) 119-22. The last name might be XaLpux8ac or XaLptac. An integration of the texts of sherds A and B has not been pub­ lished before now, and yet there is reason to arrange them in this way even if the two sherds do not join. Line 5 of A has exactly the height of line 2 of B; the other corresponding lines, so far as can be determined, are well matched; and finally, as the examples I have given in the commentary to lines 4, 5 and 6 show, a continuous text can be eco­ nomically restored. Mrs Stillwell may have had something like this in mind, to judge from her disposition of the pieces in figure 1 of the editio princeps. An objection to this arrangement is that we have no way of knowing certainly how much space originally separated the two fragments. An effect of the hypothesis, if accepted, would be to reinforce the assumption that the phrase 'TOL MaAE90 is an item in a list of names. That is, when in the context as a whole names both precede and follow, the phrase is not likely to denote a Malekos who receives a gift or offering. And in fact authorities have been inclined in most discussions to construe MaAE90 as a genitive, although the identity of the "men (or sons) of Malekos" remains obscure. Certainty concerning special characteristics of Corinthian orthography and syntax will not be won as long as so little writing from archaic Corinth can be studied,' but the admittedly meagre evidence suggests that Corinthians usually spelled out such genitive endings, -OY. Of the four names preserved in full, three homonyms, viZ' Amyntas, Sokles and Malekos, are attested at Corinth or elsewhere in Greece during comparatively early times in contexts that imply pretensions to distinction. And Melantas, also the name of a king or eponymous 4 J. and L. Robert suggest that the scarcity of writing from pre-Roman Corinth results from frequent use of bronze for public documents. It is easily melted down and reused, and it disintegrates rapidly in Corinthian soil; see REG 84 (1971) 403 no.45. 28 KORINTHIAKA official at Argos, is a plausible restoration.5 There is not, however, enough known about nomenclature in the early years of Greek history to permit speculation concerning status with only names as a basis. To return to the question of chronology, at one time it seemed necessary for an understanding of the origin and transmission of the Greek alphabet to know when these sherds were inscribed. (I now include also SEG XI 193, a sherd found in the same context with fragment A. It bears an incised inscription [---] AEOL [ ___ ]6 and is therefore of equal importance.) Now, however, after R. S. Young's publication of inscriptions in an early, Greek-like, Phrygian alphabet at Gordion7 and other publications of other instances of early Greek writing, there is a larger literate context in which to set the Corinthian pieces. Correspondingly, even if all should agree on their date, their importance as evidence for the date and diffusion of the earliest Greek alphabet has been somewhat diminished.8 The Nestor Cup from Pithekoussai, dated 725 B.C., is often said to be about contem­ porary with them, and the letter forms of its inscription likewise show well established familiarity with writing.9 And other newly discovered writing from Pithekoussai shows a peculiar early feature that was once thought possibly to be even alien to Greek writing and in any event earliest of all and closest to the generally postulated Semitic exemplar.10 In any case, consideration of the date of the Corinthian sherds continues to be desirable, if only to make clear what is not known. Mrs Still well's evidence, the context in which two of the sherds were found, establishes a strong presumption that they are to be dated toward the end of the eighth century. There are, however, arguments for a lower date-and counterarguments. R. S. Young speaks of a triple band of added color on SEG XI 193, a feature that 5 Jeffery's evocation of an Italian origin for 'Avy&pws (Ancus, Ancharius, p.121 n.4) would if correct add another foreigner who bore a name that was distinguished in other circumstances. 6 M. Lejeune, REA 47 (1945) 106 n.2, objects that [lIpoK]A€OC [JftL] ought not to be restored here, because -A€WC is Ionic-Attic. In the same study, however, he cites the presumably Corinthian lIuTpoKMoc lftl from Krommyon in another context (p.l10 n.7) without noting its possible relevance to the protested restoration. 7 R. S. Young, "Old Phrygian Inscriptions from Gordion," Hesperia 38 (1969) 252-96. 8 See e.g. M. Lejeune, "La Diffusion de l'alphabet," eRAI (1966) 505-11. 9 Jeffery, P1.47; M. Guarducci, Epigrafia greca I (Rome 1967) 226--27, figs. 88a-b. 10 This is the alplta lying on its side; see Guarducci, op.cit. 225. ALAN L. BOEGEHOLD 29 could suggest a date well down toward the middle of the seventh century B.C., as does the expert incising technique.ll T. J. Dunbabin responded that if the added color were white, a date in the late eighth century was possible.1 2 J. L. Benson interprets the traces of "added color" as surface accretion that sometimes takes on a reddish or lavender tinge. It can be identified and distinguished from traces of added color because it follows no lines,l3 Another expert, however, explains the seeming added color as a phenomenon caused by uneven firing. And yet another characterizes the glaze of sherds A and B as Corinthian rather than proto-Corinthian and no earlier than 625 B.C. The conclusion to which this variety of expert opinion leads is that the evidence of fabric, glaze and technique is ambivalent when used for chronology in this particular instance. I shall examine separately R. Carpenter's argument for a lower date for these sherds in hope of excluding it from future discussions of these sherds in particular and of the early Greek alphabet in general. In a long and subsequently much cited review of Jeffery (AJP 84 [1963J 76-85), Carpenter attempts to show that Greeks formed letters in so regular and distinctive a way during the early centuries of literacy that an inscription of, say, the early sixth century is patently identifiable by shape and disposition of letters alone; so, too, writing of the late sixth century, the seventh century and the fifth. There is, he affirms, "a continuous sequence of changing alphabetic styles to which all inscriptions and graffiti should conform." But there were three obvious and formidable obstacles in the way of this optimistic prospect, viZ. the sherds under discussion here, the epitaph for the Corinthians who died at Salamis, and the late Geometric skyphos from ancient Pithekoussai that is claimed in a graffito by a Nestor. And so suitable dates had to be found for them. His arguments concerning the last two monuments have been countered and are now generally recognized as unacceptable, but there has not yet been a formal demonstration of the insufficiency of his treatment of the first.14 11 R. S. Young, Hesperia Suppl. II (1939) 227 n.2. Cf Jeffery 121. 12 T. J. Dunbabin, Perachora II (Oxford 1962) 29 n.l. 13 Benson will publish the pottery from the Potters' Quarter at Corinth. He made this observation while examining the sherd with me at Corinth in summer, 1969. 14 See now conveniently Meiggs-Lewis, SGHI no.I, pp.I-2 (Nestor's cup) and no.24, pp.52-53 (Corinthian epitaph). A. J. Graham in "Dating Archaic Greek Inscriptions," Acta of the Fifth International Congress of Greek and Latin Epigraphy, Cambridge 1967 (Blackwell, 30 KORINTHIAKA In order to show that the letters of sherds A and B belong to the early sixth century, Carpenter prints a facsimile of six letters (five letter forms), viZ. MOKI" 8M, with two other facsimiles directly below it, both indistinguishable from the first. The letters from the second two facsimiles Carpenter describes as having been taken from inscriptions "securely dateable to the early or mid sixth century." But method and data are wrong. To take five letter forms only, and these as stable as any in the alphabet, when seventeen are available, is to neglect too much evidence. Other letters, iota and digamma to be specific, are not the same in all these inscriptions. Among the five letter forms he does use, there are moreover differences to be seen when the inscriptions themselves are examined, notably in kappa and epsilon. And finally, the two inscriptions he has chosen for comparison are not securely datable to the early or mid sixth century, at least not by any means other than those whose efficacy he has been unable to demonstrate.1s To my own eye, nevertheless, there does seem to be a generic similarity-there are admittedly variations in detail-between the forms of letters on sherds A and B and those of two painted vase inscriptions. One is Polylaidas on a Late Corinthian kotyle; the other, Diomedes, Phoinix, Aias (bis), Nestor, Agamemnon and Teukros on a cup of the Cavalcade Painter dated ca. 580 B.C.16 The similarity, if there is such a thing at Corinth as distinctive forms of letters that belong exclusively to identifiable spans of time,17 points to a date Oxford 1971) 10, resists Carpenter's dating of sherds A and B but without detailed refuta­ tion. 15 The two inscriptions are a grave trapeza from Krommyon that Jeffery says is not "more closely datable than sixth century" (p.128) and the epitaph on the grave stele of Arniadas, which Jeffery dates "late 7th or early 6th c." (p.234 no.11). 16 For Polylaidas, see Arena p.94 no.44, p1.l3; for the cup of the Cavalcade Painter, Kunstwerke der Antike, Auktion 40 (13 December 1969) Miinzen u. Medaillen, Basel (a photograph also in Antike Kunst 14 [1971] pl.46.1). H. T. Wade-Gery ap. H. Payne et aI., Perachora I (Oxford 1940) 263, notes a curvature in lines of certain letters of A and Band observes that a like curvature can be seen in other early inscriptions, but the observation is too delicate to sustain a chronology. 17 M. Guarducci presents a vivid example of exigencies of belief in an identifiable stylistic development of the Corinthian alphabet. There is only one inscription written in Corin­ thian epichoric letters whose date can be established by reference to a securely dated historical event. It is the epigram written to honor the Corinthians who died at Salamis, which must therefore have been inscribed after 480, but probably not long after. There is no other such chronolOgical marker. But Miss Guarducci describes the writing of this unique monument as "un po' arcaizzante" (op.cit. [supra n.9] 178 n.3). A dogma, even if ALAN L. BOEGEHOLD 31 near the second quarter of the sixth century B. c. for sherds A and B. Since, however, that date is opposed by the context in which the sherds were found, and since expert judgements on fabric and glaze are at variance with each other, possibly the best way to signal the difficulty is to write the date 720 (?)-550 (?). BROWN UNIVERSITY August, 1973 long prevalent, by which all tall spindly letters are ipso facto older than the short square letters might well be abandoned in instances where the best evidence tells against it. A Neglected Gorgon Boegehold, Alan L Greek, Roman and Byzantine Studies; Spring 1974; 15, 1; ProQuest pg. 32 II A Neglected Gorgon RNALD S. STROUD in "The Sanctuary of Demeter and Kore on Akrocorinth," Hesperia 37 (1968) 321-22, publishes eight frag­ ments-all presumably from a single, large, open vessel­ that show parts of figures and objects in polychrome painting and a few letters in dark brown matt paint that seem to come from labels. The fragments are shown on PLATE 5, fig.2. One shows feathers of a great wing and the letters [---] B'IIO~ vae. On a second, there is a head in profile looking right. Eye, nose, curly hair and a wide head­ band with two small wings at the back are preserved. To the right of the face there are two letters, H B [---]. On a third fragment, there are the hind legs of two horses; on a fourth, parts of the belly of a chariot, a horse's tailer) and a five-spoked wheel. The fifth fragment of immediate interest has a fire and the letters [--- 1] 0 N [---1]. While no shape of a single letter from any of the three inscriptions permits sure identification of the alphabet as epichoric Corinthian, the letters B'IIO~ when taken together make sense only when they are read as Corinthian. For if they are read as [---]fivoc-that is, as letters from a standard Ionic alphabet-the result is a form that is not attested else­ where in Greek. On the other hand, [---]evoL, assuming they are Corinthian letters,! yields the last letters of a name that is possible and directly applicable to the picture of a wing. Before elaborating on what seems to me to be the correct restora­ tion, I note one other conceivable way to read the letters, not because I think it leads to a correct restoration, but because it introduces a question of the early Corinthian alphabet that needs to be clarified. Two of our leading authorities differ in their descriptions of the earliest Corinthian beta. L. H. Jeffery in The Local Scripts of Archaic Greece (Oxford 1961) 114 represents the earliest beta as t.P, followed by 1 The letters ON are not distinctive, and H B is ambiguous because it can be restored as though from a standard Ionic alphabet, e.g. "H,B[1)], or Corinthian 1i€[paKAic], hl[pa], hl[,B1)] etc. Stroud notes [--]£voc or [-]1)VOC, a possible restoration ofBV\o~ (op.at. 322 n.19), and H. Metzger's assertion (REG 83 [1970] 123) that Stroud did not recognize Corinthian letters could perhaps have been modified to acknowledge Stroud's note. For recent discussion of crooked iota at Corinth, see P. Amandry, "Collection Paul Canellopoulos (1)," BCH 95 (1971) 590. 32 BOEGEHOLD PLATE 5 Figure 1. hSCRIBEIJ SHERDS or: PROTOCORI:'o!THIAN KI\NTHAROS, SEC XI 204, reverse ALAN L. BOEGEHOLD 33 Ul and f". She does not consider the "doubtful betas" in IG IV 354 and 357, and her reasons for not doing so are sufficient (loc.cit. n.3). M. Guarducci, on the other hand, in Epigrafia greca I (Rome 1967) 170-71, describes \;\ as the earliest Corinthian beta (this is the "doubtful" one that Miss Jeffery refused to consider), followed by forms like those shoW'n above. She rejects IG IV 354 and 357, as Miss Jeffery had done, but turns to an inscription painted on a vase of the sixth century B.C. for evidence that beta existed in that form in the Corinthian alphabet. It is, she says, the sole example attested at Corinth. Then, to support her belief that this is the oldest form of Corinthian beta, she cites a graffito scratched into a proto-Corinthian pyxis, a sherd of which was found at Syracuse and dated to the beginning of the seventh century. There is, she argues, a beta (\;\) on the sherd. Since Corinthians colonized Syracuse in 734/3, and since \1\ is attested as beta later on, the figure on the sherd from Syracuse is the earliest form of Corinthian beta. (But cf Jeffery, pp.130, 263--65.) But both of Miss Guarduccfs examples are deficient, since in neither case is it satisfactorily clear that the letter she points to is an actual beta. To take the sixth-century example first, the so-called beta cannot in any way be distinguished from a nu. The name Daiphonos is written retrograde, and so the nu is properly oriented with the direction of the writing. And indeed H. Payne in Necrocorinthia (Oxford 1931) 158, while conjecturing that the figure was a "curious form" of beta, conceded that the name might in fact be Daiphonos and not Daiphobos. R. Arena in "Le Iscrizioni corinzie su vasi," MemLinc SER. VIII, 13 (1967) 115-16, is right to transcribe L1atovoc, for if the nu were a beta of the sort Miss Guarducci postulates, it ought to have been drawn in this particular case like a nu in writing that goes from left to right. The second so-called beta, that on the Syracusan sherd, is an arbitrary construction. The marks in question can be restored with equal credibility as Corinthian epsilon or eta (B) followed by some other letter like kappa, mu or nu (cf Jeffery 125). Given, therefore, only the evidence that Miss Guarducci cites, an uncommitted reader can justly reach a conclusion opposite to hers and reason that this form of beta did not exist at all in the Corinthian alphabet. And yet the form is indisputably attested at Selinus and at Melos,2 and so long 2 Selinus, IG XIV 268; cf W. M. Calder III, The Inscription from Temple G at Selinus (GRBM 4, Durham 1963) 54-55. MelDs, IG XII 3.1153; cf M. Guarducci, Epigrafia greca I (Rome 1967) 235. 34 KORINTHIAKA as it was used in related alphabets, there is a possibility that it was used also at Corinth. On the sherds from the Corinthian sanctuary of Demeter and Kore, the nu of 0 N is oriented canonically for left-to-right writing while that of Bv\O~ is backward.3 The figure \1\ therefore in shape and orien­ tation is quite satisfactory as a rare form of beta, and if we accept the letter as beta and read [---]et1oL or [---] et10c , words like "Epet1oc, €CPTJt10L or lPt>-'TJt1oc can be restored. But they are only mechanically possible and have no immediately discernible connection with the fragmentary pictures. Moreover, since a restoration can be proposed that is in accord with pictures, provenience and characteristic Corin­ thian orthography, it will be best for the present to put aside notions of an earliest Corinthian beta in the form proposed by Miss Guarducci. To return to what I propose as the correct restoration, let us suppose that BV\O~ should be read as [---]evoL. The label can then be restored [L8]ENOI (= EOevQJL) , an apposite name for a winged creature at Corinth, H ••• where the accepted formulae were created in which the gorgons and the gorgon story were clothed; and it was from Corinth that they passed to other parts of the Greek world" (Payne, Necro­ corinthia p.86). Hesiod, Theogony 276, names the three sisters: EO'vvw T' Evpva)'TJ Te M'8ovca Te >-.vypa 7TaOovca, and in our texts the spelling EO'vvw was recovered by learned conjecture (see M. West, app.crit. ad loc.) from a number of attested spellings that included inter alia EOevw and EOeLVW. EOevwL is a Corinthian form whose nominative singular ending in oWL is common on Corinthian vases.4 It is not in­ cluded among the spellings noted by H. Os tern, in Roscher, Lex.Myth. (1909) 1505-06 s.v. STHEINO, but perhaps it should be, especially in view of the gorgon on a late Corinthian oenochoe in Florence (Payne, Necrocorinthia p.165 no.33, p.325 no.1389; cf Arena pp.94-95 no.47) who is also labelled L8ENOI. She has a horseman on either side, one named Euphamos, the other, Polys. I do not know any story of gorgons in which these knights play a part,5 nor can I resolve the seeming incoherence of the fragments 3 Elizabeth Pemberton advises me per litt. that the fragment with the wing is upside down in the photograph. This mischance does not, however, affect our comments on the shapes and stances of letters within any of the three inscriptions. • The index in Arena, pp.139-41lists sixteen feminine names that end in -w,. On the form, see E. Schwyzer, Griechische Grammatik I (Munich 1939) 478 §6.2. 6 Sthenno and Euryale have long been judged creatures who have no real stories. They were created only to pursue Perseus. Demetrios I. Lazarides, reporting on work at Kavala ALAN L. BOEGEHOLD 35 under consideration, where fire, horses, a chariot and a gorgon come together. The head in profile, if part of a picture in which Sthenno, sister of Medusa and Euryale, plays a role, can be identified tenta­ tivelyas the head of Hermes, i.e., hi[pf'EC], who appears with Perseus and gorgons early on in the tradition. See e.g. Hans Besig, Gorgo und Gorgoneion (Berlin 1937) 49, and Fittschen, loc.cit. BROWN UNIVERSITY August, 1973 (Deltion 17 [1961/2] B' 238) notes fragments of a Corinthian krater (ca. 570 B.C.) that shows two gorgons in flight (p1.281 B); see also G. Daux, BCH 86 (1962) 835 fig.6 and 837. The picture as a whole was surely one of Perseus being pursued after his beheading of Medusa. It is risky to attempt readings from photographs, however ably reproduced, and so I note only the possibility that two of the three fragmentary, painted inscriptions can be restored yopy[ov£c] and [L'8£]VOL. For recent work and earlier literature on gorgons in early Greek art, see Hans von Steuben, Fruhe Sagendarstellungen in Korinth und Athen (Berlin 1968) 13-17, 94, and Klaus Fittschen, Untersuchungen tum Beginn der Sagendarstellungen bei den Griechen (Berlin 1969) 152-57. Aphrodite and Pothos on a Kantharos Boegehold, Alan L Greek, Roman and Byzantine Studies; Spring 1974; 15, 1; ProQuest pg. 36 III Aphrodite and Pothos on a Kantharos I N A REPORT on archaeological finds in Greece, Otto Walter noted at Corinth a black-glazed kantharos of the fourth century B.C. with a graffito. He read it as: 'Acppo8{:rac ;Aax£ Kat rAVKVC'TTOPOC, and surmised that Glykysporos was an otherwise unattested proper noun.l S. Weinberg later proposed three alternate readings, viZ. (1) 'Acppo8tTac ;Aax£ Kat YAVKVC II6poc, (2) 'Acppo8l'Ta c' ;Aax£ Kat YAVKtk II6poc, (3) ---1T6poc (sens. ObSc.).2 The four alternatives were reprinted in SEG XI 204 with a reference to G. Daux, who doubted that the postulated Glykysporos was a Greek name.3 I have seen no further discussion of the text, and yet correction of all four versions is demonstrable. The text should be printed: 'Acppo- 8l'Ta C' ;Aax£ Kat yAVKVC II600c, for the dot in the center of theta is visible, as is the clear difference between rho in Aphrodite and theta in Pothos. The letters can be seen in the two detailed photographs on PLATE 6. A stroke mistakenly interpreted as the vertical hasta of rho seems actually to be a slip by the engraver, who was scratching a round letter into a hard, fired surface. Attendant chipping of the flaky glaze also blurs the shapes of letters. The first three words, 'Acppo8l'Ta c' ;Aax£, represent familiar and well attested usage,4 and the last three add a second subject in a schema Pindaricon. The sense is: "Aphrodite has you as her portion, and sweet Pothos." The "you" who is addressed may have received the kantharos, but conceivably the line originally addressed another "you" in another context. It is hard to be sure whether a single line of iambic character is deliberately metrical or not, and usually context will be required for certainty. Yet the graffito does have a poetic flavor, and the line could have been transcribed without its first syllable. If, for instance, a monosyllable like &AA' is added to the beginning of the line, it can be scanned as an iambic trimeter.5 And in 1 AA 57 (1942) 143. 2 "A Cross Section of Corinthian Antiquities," Hesperia 17 (1948) 239 with pl.88 nO.6. 3 Fouilles de Delphes ill 3.2, p.187 n.l. 4 Cf Ar. Bcd. 999 p.a -r1Jv 'Ac/>polllrYjv 'ii p.' t!AaX£ KATJpovp.lVTJ and LSJ s.v. Aayxavw I 1. b lowe this observation to Alexander Turyn. 36 ALAN L. BOEGEHOLD 37 that case, we can perhaps imagine a lover taking the sentiment from a poet's work and making it one with his present.6 BROWN UNIVERSITY August, 1973 B Sir John Beazley in "Hymn to Hermes," AJA 52 (1948) 336-40, discussed Attic vase paintings in which he identified letters painted on papyrus rolls as fragments of poems. Cf H. Immerwahr, "Book Rolls on Attic Vases," Classical, Mediaeval and Renaissance Studies in Honor of B. G. Ullman I (Rome 1964) 17ff, and E. T. Vermeule, "A Love Scene by the Panaitios Painter," AJA 71 (1967) 313. In the present instance, the line seems to have been inscribed after the kantharos was painted and fired and so was presumably not an element of the potter's original intention. Cf Sappho's poem written on a potsherd, D. L. Page, Sappho and Alcaeus (Oxford 1955) 35. Several of the very earliest instances of Greek alphabetic writing are scratched into terra-cotta and are metrical. On the Geometric Dipylon oenochoe and Nestor's Cup from Pithekoussai, see e.g. Jeffery, LSAG 68ff, 235f. New discussions of both inscriptions continue to be published.