Endnotes, Part 1

EN1.1 (Endnote to n1.181)

{94|95} If the Dioskouroi were thought of as buried beneath the earth at Therapne, their cult must, one assumes, have included the principal feature of hero cults, namely a grave. Unfortunately nothing is known of the twins’ cult at Therapne, which after Pindar is never mentioned again. Bölte RE ‘Therapne’ 2365 suggests that after Pindar’s time the twins’ cult was moved to other locations nearer Sparta, such as the Dromos and the Phoibaion, where shrines of the Dioskouroi are attested in Pausanias 3.14.6 and 3.20.2. The main sanctuary at Therapne was the “Menelaion” (so called first by Polybius in the second century BC), where Helen and Menelaus were worshipped as gods (not as heroes) according to Isocrates 10.63. Their worship at Therapne is attested for the sixth century BC by a story in Herodotus 6.61 about Helen’s cult, and by dedications to both Menelaus and Helen found in excavations of the Menelaion carried out by the British School in Athens in the 1970s (Catling 1977:36–37: there is one dedication to Menelaus, dated early fifth century; there are two dedications to Helen, not specifically dated but apparently older). Bölte RE ‘Therapne’ 2360–2362 argues that the Dioskouroi were once worshipped at the Menelaion, and he associates their burial place with the peak of the hill on which the shrine is located (RE ‘Therapne’ 2356, 2360). The British archaeologists do not discuss the Dioskouroi in their study of the shrine, which they date to the early fifth century BC (they also identify two predecessors of the shrine, which they date to c. 600 BC and c. 700 BC (Catling 1977:35–36; cf. Deoudi 1999:124–125). If the worship of the Dioskouroi did not take place at the Menelaion, there is no way of knowing where else in Therapne it might have taken place. The British excavators discovered a second hilltop shrine not far from the Menelaion, but the objects in the deposit of this shrine do not seem to concern the Dioskouroi (cf. Catling 35). The only other evidence relating to the twins’ sacred space in Therapne is the word kolōnā̂n in a badly preserved entry of Hesychius; it is a fragment of lyric {95|96} poetry attributed to Alcman by Diehl (fr. 8) and left unattributed by Bergk (adespota fr. 74) and Page (PMG 983 = adespota fr. 65):

τυίδε (τύδαι cod.)· ἐνταῦθα. Αἰολεῖς. τυδᾶν κολωνᾶν·
Τυνδαριδᾶν κολωνᾶν.

Whatever the correct wording of the quoted phrase (Page suggests τυίδ᾿ ἀν Τυνδαριδᾶν κολωνάν, “hither up the hill of the Tyndaridai”), the occurrence of kolṓnē, “hill, mound” in connection with the Tundarídai is perhaps significant in view of the word’s occasional meaning “sepulchral mound, barrow” (LSJ cite Sophocles Electra 894; cf. also Iliad 2.811). There is of course no guarantee of that meaning here, especially since Therapne itself is a hill (cf. Pindar’s reference to Castor as “dwelling in the high-placed seat of Therapne,” Τυνδαρίδας…ὑψίπεδον Θεράπνας οἰκέων ἕδος, Isthmian 1.31). If we assume that a tomb played a part in the twins’ cult, this may be why a tomb also plays a part in the myth of Castor’s death and return to life as told by Pindar in Nemean 10. After Castor is killed, Polydeuces finds Idas and Lynkeus “near their father’s tomb” (τοὶ δ' ἔναντα στάθεν τύμβῳ σχεδὸν πατρωΐῳ, line 66). They seize an “image of Hades, smooth stone” (ἄγαλμ' Ἀΐδα, ξεστὸν πέτρον, line 67) and hurl it at Polydeuces, but it does not hurt him. Polydeuces then kills Lynkeus with his spear and Zeus smites Idas with a thunderbolt. The significance of the tomb in this myth is two-fold: it symbolizes Polydeuces’ immortality (he is not crushed by the ágalma of Hades), and it marks the site where Idas and Lynkeus meet their end. The latter twins’ father Aphareus, a brother of Tyndareus (Stesichorus, cited by “Apollodorus” 3.10.3), was Messenian in origin, but his tomb must have been appropriated by the Spartans as a result of the Messenian Wars: Pausanias 3.11.11 reports its presence in Sparta not far from the bones of Orestes (3.11.10), which the Spartans had likewise expropriated from abroad (Herodotus 1.67–68; see Hiller von Gaertringen RE ‘Aphareus’ 2711); the Messenians, playing a similar game, claimed that Messenia was the birthplace of the Dioskouroi (Pausanias 3.26.3, 4.31.9; cf. Luraghi 2002:65). According to Pausanias 3.13.1 there was a mnē̂ma (memorial) of Castor in Sparta, and over it was a sanctuary of the Dioskouroi, which was said to have been built when the pair first received divine worship forty years after their battle with Idas and Lynkeus. This mnē̂ma may represent Castor’s grave (cf. Kearns 1992:66), and graves of Idas and Lynkeus were also to be seen nearby; Pausanias thinks that Idas and Lynkeus must have been buried in Messenia, and that only the Messenians’ long absence from their own country (after the Messenian Wars) can account for the Spartans’ claim to this pair’s burial place: ἔστι δὲ καὶ Κάστορος μνῆμα, {96|97} ἐπὶ δὲ αὐτῷ καὶ ἱερὸν πεποίηται· τεσσαρακοστῷ γὰρ ὕστερον ἔτει τῆς μάχης τῆς πρὸς Ἴδαν καὶ Λυγκέα θεοὺς τοὺς Τυνδάρεω παῖδας καὶ οὐ πρότερον νομισθῆναί φασι. δείκνυται δὲ πρὸς τῇ Σκιάδι καὶ Ἴδα καὶ Λυγκέως τάφος. κατὰ μὲν δὴ τοῦ λόγου τὸ εἰκὸς ἐτάφησαν ἐν τῇ Μεσσηνίᾳ καὶ οὐ ταύτῃ· Μεσσηνίων δὲ αἱ συμφοραὶ καὶ ὁ χρόνος, ὅσον ἔφυγον ἐκ Πελοποννήσου, πολλὰ τῶν ἀρχαίων καὶ κατελθοῦσιν ἐποίησεν ἄγνωστα, ἅτε δὲ ἐκείνων οὐκ εἰδότων ἔστιν ἤδη τοῖς ἐθέλουσιν ἀμφισβητεῖν, “There is also a memorial of Castor, and over it a temple has been constructed. For they say that the sons of Tyndareus first came to be considered gods in the fortieth year after their battle with Idas and Lynkeus and not before. The tomb of Idas and Lynkeus is pointed out next to the Canopy [assembly place]. It is likely from the story that they were buried in Messenia and not here; but the Messenians’ misfortunes and the time that has passed since they fled from the Peloponnesus have made many of the old traditions unknown to them even now that they have returned, and since they do not know it is open to those who wish to argue about them” (Pausanias 3.13.1–2).

EN1.2 (Endnote to n1.190)

Supporting the thesis of Sukthankar 1936/1937 that the Bhārgava family of priests thoroughly reshaped the Mahābhārata at some point in the tradition, Goldman 1977 studies the poem’s distinctive Bhārgava themes, including hostility toward the gods (Goldman 1977:113–128) and a preoccupation with violent death and a subsequent return to life (Goldman 1977:75–92). The ability to bring back to life is particularly associated with Śukra, also called Uśanas Kāvya. He is not the only Bhārgava with this ability, but he alone commands the mṛtasaṃjīvinī vidyā, the secret spell for bringing the dead back to life (Goldman 1977:88–92). Like Cyavana, Uśanas Kāvya occurs in the Rig-Veda, where he is a sage. In epic he, like Cyavana, embodies the Bhārgava theme of hostility toward the gods, for he is the priest of the Asuras, the demons who war against the gods. His counterpart Bṛhaspati, the priest of the gods, specifically does not command the mṛtasaṃjīvinī vidyā (Goldman 1977:125–126). What the theme of bringing back to life may have to do with the twin gods one can only speculate; at any rate the family’s hostility to the gods does not extend to the twins, with whom there is instead, in the case of Cyavana, a kind of well-meaning mutual rivalry. The Bhārgavas’ role in shaping the Mahābhārata may explain a basic feature of the twins’ role in the poem, namely the distinctions between them that are ignored elsewhere in Indic tradition. To the point here is another of the Bhārgava themes in the Mahābhārata, namely the family’s highly ambivalent relations with kṣatriyas, {97|98} especially in the case of Cyavana and his line (Goldman 1977:93–112). The twins straddle the division between priests and warriors, as is still evident in the contrasting affinities of the epic twins Nakula and Sahadeva: whereas Sahadeva is often paired with the priestly Yudhiṣṭhira, Nakula is often paired with the warrior Bhīma (see Wikander 1957:75; cf. also Dumézil 1968:80). One wonders whether this distinction between the twins reflects the Bhārgava preoccupation with relations between brahmans and kṣatriyas in their own family. Was it the Bhārgava reworking of the Mahābhārata that brought the distinctions between the twins Nakula and Sahadeva to the fore? The distinctive Bhārgava themes of the Mahābhārata had an earlier history to judge by one clear case: Cyavana’s power over Indra in the myth of the twins’ inclusion in the Soma sacrifice has an antecedent in the Jaiminīya-Brāhmaṇa 3.159–161. Here Indra objects when Cyavana makes an offering of a sacrificial ladle to the twins, whereupon Vidanvat, another Bhārgava sage, objects to Indra’s own ladle. A violent dispute erupts between the gods and the sages and the sages create the demon Māda, “intoxication,” whereupon the gods are forced to yield to the sages’ superiority; see Goldman 1977:166n11.

EN1.3 (Endnote to n1.204)

In a variant of the central myth of the Dioskouroi the conflict between the Dioskouroi and their cousins that leads to Castor’s death arises when the Dioskouroi carry off not cattle, but the daughters of Leukippos, who are betrothed to the Apharetidai (Theocritus 22 is the earliest literary source; other sources include the scholia to Iliad 3.243, the scholia to Pindar Nemean 10.60, Tzetzes, Hyginus, Ovid; cf. Frazer 1921, vol. 2, 30n4, on “Apollodorus” 3.11.2). The daughters of Leukippos are themselves cousins of both pairs of twins, for Leukippos is the brother of Tyndareus and Aphareus. From an Indo-European standpoint the marriages of the Dioskouroi to the Leukippides are secondary to their relationship with a solitary female figure who was at once their sister and common wife (see §1.55). With this in mind we may view the theft of wives by the Dioskouroi as patterned after their theft of cattle. In literature and art the theft of wives became more popular than the theft of cattle (see Pausanias 1.18.1, 3.17.3, 3.18.11, 4.31.9 for citations of artistic representations). “Apollodorus” 3.11.2 somewhat illogically combines both variants of the myth in one version: the Dioskouroi are said to have carried off the daughters of Leukippos from Messene, and to have had sons by them, before they carry out the joint cattle raid with their cousins; the Apharetidai are simply left out of account in this version until they participate in the joint cattle raid. The Leukippides had cult associations in Sparta; they had a sanctuary {98|99} of their own and priestesses, also called Leukippídes (Pausanias 3.16.1). Frazer 1921 on “Apollodorus” 3.11.2 makes the case for possible theriomorphism in this cult: an obscure gloss in Hesychius s.v. pōlía may mean that two maiden priestesses of the Leukippides were called “colts of the Leukippides.” It is possible that the Dioskouroi and their wives were themselves worshipped in the form of horses, but the only evidence is the name Leukippídes of the wives and the epithet leukópōloi of the twins. The epithet, which occurs in Pindar Pythian 1.66, is better taken as “owners of white horses” (cf. n1.158), but a doubt is raised by the Theban twins Amphion and Zethos, who are actually called “white colts of Zeus” in a fragment of Euripides’ Antiope (leukṑ…pṓlō tṑ Diós, fr. 223.127 Kannicht 2004); cf. Ward 1968:12, who cites parallels in the Latvian and Indic traditions (for Saraṇyū, who was said to have conceived the Vedic twins in the form of a horse, see n1.122).

EN1.4 (Endnote to n1.212)

The derivation of the name Lampetíē from a syntagma *lampeti-yā, “she who shines,” must exclude alternative explanations based on different forms from the same root. The verb lampetáō, “shine” (Iliad 1.104, Odyssey 4.662), is not relevant; Chantraine 1958:358 derives lampetáō from lámpō, like eukhetáomai from eúkhomai, “pray” (there is no related noun in either case; contrast naietáō, “dwell,” from [peri]naiétēs, “dweller,” Iliad 24.488); a form Lampetíē from the verb lampetáō would be unparalleled and difficult to explain. The adjective kallilampétēs, “beautifully shining,” in the phrase hḗlie kallilampétē, Anacreon fr. 106 Page [PMG 451] (for the vocative kallilampétē instead of kallilampéta cf. Hipponax fr. 139 Degani and Degani ad loc.) is formed like Homeric hupsibremétēs, “high thundering” (agent suffix -tēs); this formation has no relation with the form Lampetíē. The adjective khrusolámpetos, “shining with gold” (Hipponax fr. 79.7 Degani) contains a verbal adjective in -etos (cf. Hawkins 2004:203 and Chantraine 1933:299–300); an identically formed masculine name Lámpetos seems to be implied by the Homeric patronymic Lampetídēs (see below). Lampetíē could be derived from the name Lámpetos through an extended form Lampétios (cf. phílos and phílios); a Christian-era name Lampétios is in fact attested (Photius Bibliotheca 52, Bekker p. 13); cf. also the fragmentary name Lampeti- in IG V.2.175 (Tegea, undated), perhaps representing Lampetíōn (Boeckh CIG 1512, Pape-Benseler 1911 s.v.), this too to be derived from Lámpetos. But if we consider earlier evidence the name Lámpetos itself is suspect: the Homeric patronymic Lampetídēs does not mean “son of Lámpetos,” but “son of Lámpos,” on the explicit evidence of Iliad 15.526: Λαμπετίδης, ὃν Λάμπος ἐγείνατο φέρτατον υἱὸν, “Lampetides, whom Lampos {99|100} bore as his strongest son” (cf. n1.51 end). This, I think, is clear evidence that the name Lámpetos did not exist in the Homeric period; the name, which occurs in a fragment of Hellenistic poetry, must be a back formation from Lampetídēs (it occurs in the first line of a passage from the Foundation of Lesbos, perhaps of Apollonius of Rhodes, which is quoted by Parthenius Erotica Pathemata 21: ἔνθα δὲ Πηλεΐδης κατὰ μὲν κτάνε Λάμπετον ἥρω, “then Peleus’s son killed the hero Lampetos”); see Cuypers 2002/2003, whose interpretation of the fragment leaves little doubt that Lámpetos is derived secondarily from Homeric Lampetídēs (see esp. p. 129). What needs to be explained is the anomalous Homeric patronymic Lampetídēs; the regular patronymic from Lámpos would be *Lampídēs, which does not fit the meter; the normal metrical expedient in this situation would have been *Lampiádēs (cf. n1.51), but instead we have Lampetídēs. I suggest that to explain the form we need to take into account the relationship between the names Lámpos and Lampetíē as masculine and feminine equivalents (see n1.212); this would account for the anomalous patronymic by an analogy: Lámpos : Lampetíē :: Lámpos : Lampetídēs. If this is correct, it is the name Lampetíē that explains the name Lámpetos/Lampétios, and not the reverse. There is no other obvious explanation of the name Lampetíē. The abstract noun suffix -tíē survives in the Homeric noun akomistíē, “want of care,” Odyssey 21.284, but in Ionic this suffix regularly becomes -síē, as in Homeric huposkhesíē, “promise,” Iliad 13.369. Old abstract nouns with this suffix are mostly compounds, like the two forms just cited (see Chantraine 1933:83–84), although simple forms also occur, like Homeric eiresíē, “rowing” (see Chantraine 1933:85–86). To explain Lampetíē as an abstract noun meaning “shining,” like eiresíē, “rowing,” one would have to account for the lack of assibilation, which in akomistíē is phonetic (the few late examples cited by Chantraine 1933:83 in which -t- is retained in compounds are clearly not relevant).

EN1.5 (Endnote to n1.219)

Two-thirds of the examples of resolution of the name Nā́satyā (44 of 67) occur in syllables 5–8 of trimeters (eleven-syllable and twelve-syllable verses). Trimeters are divided into two segments by a caesura after the fourth or fifth syllable, and the two syllables after the caesura are regularly short. In the examples in question the name Nā́satyā follows the “early” caesura after syllable 4, and the normal rhythm in syllables 5–8 would be Náasatyā ( — — ); syllables 5–7 of the trimeter are called the “break,” and the normal break after early caesura is —, occurring in 40 percent of trimeters in the archaic period and more frequently thereafter (Arnold 1905:183). But while the rhythm Náasatyā ( — — ) would be “normal,” the rhythm Nā́satiyā ( — — ) is {100|101} by no means abnormal: a break after early caesura of the form — is among those next in frequency after the “normal” break, being one of three forms that Arnold calls “subnormal” (there are also “occasional” and “irregular” forms, which are still less frequent). Of the three “subnormal” forms the pattern — — occurs in about one-seventh of trimeters in the archaic period; Arnold does not give figures for the other “subnormal” patterns, — and , but the frequency of the pattern — at least seems to be on the same order (this is suggested by Arnold’s figures for the relative frequency of trimeters with long versus short seventh syllable, and by the greater frequency of — vs. after early caesura, at least in certain time periods; see Arnold 1905:183–184). My interpretation of these numbers is that it was precisely in this position (after the early caesura in trimeters) that the new rhythm Náasatyā ( — — ) replaced the old rhythm Nā́satiyā ( — — ) and became increasingly well-established. Of the remaining 23 examples of resolution of the name in other positions in trimeters or in dimeters (eight-syllable verses), there are seven examples in which the name occupies syllables 3–6 of dimeters, and this position strongly favors the rhythm Nā́satiyā ( — — ). The reason for this is that the rhythm of all Vedic verses is more fixed at the end of the verse than at the beginning, and in dimeters in particular the cadence (syllables 5–8) has an iambic rhythm ( — ) in over 94 percent of examples (figure calculated from Arnold’s table on p. 153, which includes “rather less” than half the dimeter verse of the Rig-Veda [Arnold p. 150]; my calculation excludes Arnold’s figures for trochaic Gāyatrī and epic Anuṣṭubh, which operate on different principles and have different cadences from other dimeters). Just as the rhythm Nā́satiyā (—  —) is strongly favored in syllables 3–6 of dimeter by the heavy preponderance of an iambic cadence in this verse-form, the rhythm Náasatyā ( — — ) is strongly disfavored in this position by the occurrence of a long fifth syllable in less than 0.7 percent of dimeters (figure again calculated from Arnold’s table on p. 153). There is thus no reason to think that the rhythm was anything but Nā́satiyā (— — ) in this position. But figures do not reveal everything. There are nine examples of resolution of the name in syllables 5–8 of dimeter, and in this position (the cadence) both rhythms are irregular and rare: the rhythm  — — (“syncopated”) occurs in less than 0.3 percent of examples and the rhythm — — is included in “other irregular forms” by Arnold (table on p. 153), all of which together account for about 1 percent of examples. What is striking is the large number of examples (nine) of the four-syllable form of the name in this position where either of the possible rhythms is irregular and rare. I hasten to point out that one of the nine examples is RV 2.41.7, which I have already argued is old because {101|102} of its clear contrast between the names Nā́satyā and Aśvínā in terms of cattle and horses, and because of its context in a list of gods of the three “functions” closely resembling the similar Mitanni list. Since the rhythmical tendencies of Vedic verse become more fixed as time goes on, with fixation occurring first at the end of the verse and gradually moving toward the beginning of the verse, the cadence of RV 2.41.7a will not have been irregular when it was first composed if it was composed early enough (cf. Nagy 1974:30–31, 36, who compares Greek dimeters); a double short in this position looks especially old (cf. Nagy 1974:31 with n13). RV 2.41.7 contains another possible sign of early composition, namely a short fourth syllable, which is only a third as common as a long syllable in this position (Arnold p. 151, Nagy 1974:159; Nagy 1974:157–159 considers the increasing fixation of a long fourth syllable as a factor in the replacement of an inherited phrase śráva[s] ákṣitam, cognate with Greek kléos áphthiton, by a newer equivalent ákṣiti śrávas in verse-final position in dimeter). I therefore posit that RV 2.41.7a contained the rhythm Nā́satiyā ( — — ) in verse-final position, and that this traditional verse served as the model for the other eight examples of resolution of the name in this position (RV 5.74.2b, 8.5.32c, 8.5.35c, 8.9.9a, 8.25.10b, 8.26.2b, 8.85.1a, 8.85.9a; note that four of these eight verses contrast the duals Nā́satyā and Aśvínā in different segments of a Gāyatrī just as 2.41.7 does: these four verses are 8.5.32, 8.9.9, 8.85.1, and 8.85.9). If this argument is correct, RV 2.41.7, which has already been crucial to this investigation of the twins’ two names Nā́satyā and Aśvínā, also supports by its rhythm the reconstruction *nasati-ya, “he who brings back to life and light,” of the first of these names. Here, with resolutions and scansion indicated, is the text of RV 2.41.7 (note the “regular” scansion of the second and third segments, each with iambic cadence and long fourth syllable, in contrast to the “irregular” scansion of the first segment, with non-iambic cadence and short fourth syllable):

gómad ū ṣú nāsatiyā / áśvāvad yātam aśvinā / vartī́ rudrā nṛpā́yiyam
—/ — — — — — / — — — —

The remaining seven instances of resolution of the name Nā́satyā include four in syllables 1–4 of trimeter, where rhythm is freest and both scansions are equally likely, and three in unique positions, all open to either scansion (syllables 8–11 and 2–5 in trimeter and 2–5 in dimeter). It remains to point out that in Van Nooten and Holland 1994 the metrically restored text of RV 2.41.7 is as above. {102|105}