Xian ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 2017 Ruobing Xian An Etymological Note on Homeric ὑπόδρα Ruobing Xian INDARIC LYRICS, as is well known, preserve some archaisms of the IE root *derk ̑- (‘see’, Gr. δέρκοµαι).1 Less established is the view that “epic has its own fossils as- sociated with the root, e.g. ὑπόδρα.”2 This article aims to illuminate the prehistory of ὑπόδρα with special reference to Vedic material. The adverb ὑπόδρα is a compound of the prepositional prefix ὑπό and the second half δρα (< *d̥rk ̑ ), the zero-grade of the IE root *derk ̑-.3 The attestation of the word is entirely restricted to the ὑπόδρα ἰδών formula. According to the analysis of Holoka,4 the formula is used twenty-six times in 1 On IE *derk ̑- see H. Rix, Lexikon der indogermanischen Verben: die Wurzeln und ihre Primärstammbildungen2 (Wiesbaden 2001) 122. In contrast to the tradi- tional view that the aorist participle δρακείς found in three Pindaric pas- sages (Nem. 7.3, Pyth. 2.20, fr.123.3) is to be understood as derivative from of an otherwise unattested aorist passive in -η, B. Forssman, “Δρακείς,” MSS 16 (1964) 17–19, has convincingly shown that it should be traced back to the originally athematic paradigm of the root aorist active, which is frag- mentarily attested in Indo-Iranian cognate forms. Recently, T. G. Barnes, “δρακείς, δέδορκε, and the Visualization of κλέος in Pindar,” HSCP 107 (2013) 73–98, has pointed to another archaism of IE *derk ̑- in Pindar: the perfect form δέδορκε(ν) (Ol. 1.93, Nem. 3.84, Nem. 9.41), which serves to visualize the κλέος of heroes, reflects IE *dedórk ̑e as a stative-intransitive perfect meaning initially ‘is seen, is visible’. Barnes further argues that the archaisms in Pindaric δρακείς and δέδορκε(ν) go back to the tradition of choral lyric. 2 Barnes, HSCP 107 (2013) 76 n.15. 3 For a survey of such compounds in Homer see E. Risch, Wortbildung der homerischen Sprache (Göttingen 1974) 364–365. 4 J. P. Holoka, “ ‘Looking Darkly’ (ὑπόδρα ἰδών): Reflections on Status P 262 AN ETYMOLOGICAL NOTE ON HOMERIC ΥΠΟΔΡΑ ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 the Homeric epics, always serving as speech introduction. It has long been noted that ὑπόδρα is etymologically connected to Vedic u p a-d ̥r ́s ́- (‘sight, appearance’),5 as argued by Chantraine: “Dans l’adverbe homérique ὑπόδρα (ὑπόδρα (ϝ)ιδών ‘qui re- garde en dessous’) il faut sans doute reconnaître un nom-racine répondant au sanskrit upad ̥rç-.”6 Frisk also suggests a root-noun *upo-d̥rk ̑- as pre-form for both ὑπόδρα and Vedic upa-d ̥r ́s ́-. He argues that ὑπόδρα is “[a]us *ὑπό-δρακ zu ὑποδέρκοµαι und mit aind. upa-d ̥r ́s ́- f. ‘Anblick’ formal identisch, wohl eig. Neutr. einer adj. Bahuvrihibildung in adverbieller Funktion.”7 This etymological correspondence has been widely acknowledged in etymological dictionaries8 and historical grammars alike.9 Yet in examining Homeric ὑπόδρα and Ṛgvedic u p a-d ̥r ́s ́- in more philological detail, we encounter difficulties. First, there is a great semantic disparity between the two words: while in the Homeric epics ὑπόδρα (ἰδών) throws into relief the gaze of a furious speaker, in the Ṛgveda upa-d ̥r ́s ́- is merely used two times (VIII 102,15c; IX 54,2a) in the phrase su ̄́rya ivopad̥r ́k: the first re- fers to the footprint of Agni (VIII 102,15 padáṃ devásya mīḷhúṣo / ánādh̥rṣṭābhir ūtíbhiḥ / bhadrā ́ su ̄́rya ivopad̥ŕk, “The footprint of the generous god, with his unassailable forms of help, is an auspicious sight, like the sun”), and the second calls Soma a ___ and Decorum in Homer,” TAPA 113 (1983) 1–16, at 3 n.6. 5 On the meaning of Vedic upa-d ̥r ́s ́- see J. Wackernagel and A. Debrun- ner, Altindische Grammatik II.2 (Göttingen 1954) 8; L. Renou, Etudes védiques et pāninéennes VIII (Paris 1961) 87; Th. Krisch, RIVELEX, Rigveda-Lexikon II (Graz 2012) 339 (“Determinativkompositum/verbales Rektionskomposi- tum (Nom. rei actae) aus dem Adverb/Präverb úpa ‘(her)zu, zu, (hin)zu’ (s.d.) + d ̥r ́s ́- f. ‘Anblick’ (s.d.), oder + -d ̥r ́s ́- ‘sehend, sichtbar, aussehend’ ”); S. W. Jamison and J. P. Brereton, The Rigveda: The Earliest Religious Poetry of India III (Oxford/New York 2014), who translate upa-d ̥r ́s ́- as ’sight’. 6 P. Chantraine, La formation des noms en grec ancien (Paris 1933) 4–5. 7 H. Frisk, Griechisches etymologisches Wörterbuch II (Heidelberg 1970) 972. 8 M. Mayrhofer, Kurzgefasstes etymologisches Wörterbuch des Altindischen I (Hei- delberg 1956) 105; R. S. P. Beekes, Etymological Dictionary of Greek II (Leiden 2010) 1536. 9 E. g. Wackernagel and Debrunner, Altindische Grammatik II.2 4. RUOBING XIAN 263 ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 sight like the sun (IX 54,2 ayáṃ sū ́rya ivopad̥ŕg / ayáṃ sárāṃsi dhāvati / saptá praváta ā ́ dívam, “This one is a sight like the sun; this one runs to the lakes, along the seven slopes, to heaven”).10 Second, the prepositional prefix ὑπό in ὑπόδρα seems to be functioning quite differently than in the case of úpa in upa-d ̥r ́s ́-. As Schindler and Scarlata have convincingly shown, úpa hardly modifies the meaning of the compound upa-d ̥r ́s ́- (=‘appearance, sight’).11 On the contrary, Homeric ὑπόδρα is unmistakably linked to a facial expression, namely “looking (out) from be- neath (scil. beetling or knit) brows,” underpinned by the prefix ὑπό.12 Moreover, the Homeric ὑπόδρα and Ṛgvedic upa-d ̥r ́s ́́- differ in that the former is an adverb only attested in combina- tion with the aorist participle ἰδών whereas the latter is a root- noun. In this short note I wish to draw attention to Vedic upadraṣṭár- (‘onlooker’) that has not been included in etymo- logical discussions of ὑπόδρα. It will be argued that comparison of Homeric ὑπόδρα ἰδών and Vedic upadraṣṭár- sheds new light on the prehistory of both items whereby the issues raised above simply disappear. Vedic upadraṣṭár- (‘onlooker’) is a Nomen agentis of the pre- 10 Transl. Jamison and Brereton 1214 and 1278. The syntactic problem raised in the verses quoted has been discussed by G.-J. Pinault, “Le substantif épithète dans la langue de la Ṛk-Saṃhitā,” in E. Pirart (ed.), Syntaxe des langues indo-iraniennes anciennes (Barcelona 1997) 125 and 139–141, who translates the phrase su ̄́rya ivopad ̥r ́k as ‘soleil-spectacle’, ‘soleil-image’, ‘aspect- soleil’. 11 J. Schindler, Das Wurzelnomen im Arischen und Griechischen (diss. Würzburg 1972) 26; S. Scarlata, Wurzelkomposita im Ṛg-Veda (Wiesbaden 1999) 229. H. Hettrich, A. Casaretto, and C. Schneider, “Syntax und Wortarten der Lokalpartikeln im Ṛgveda, IV: I. Allgemeines, II. úpa, III. áva,” MSS 64 (2004 [2010]) 17–130, at 80, note that in the Ṛgveda the local particle úpa in combination with the verbal root darś draws “keine sichtbare semantische Modifizierung des Verbs nach sich.” 12 See Holoka, TAPA 113 (1983) 4 n.8; cf. Th. Rakoczy, Böser Blick, Macht des Auges und Neid der Götter (Tübingen 1996) 45: “Das [ὑπόδρα ἰδών] hebt sich deutlich ab von dem sonst ‘üblichen’ geraden Blick ins Gesicht des Gegenüber (e.g. ἐσάντα ἰδών).” 264 AN ETYMOLOGICAL NOTE ON HOMERIC ΥΠΟΔΡΑ ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 positional prefix úpa and the verbal root darś (‘see’),13 the same roots inherited from IE as found in Greek ὑπόδρα. Moreover, as a Nomen agentis Vedic upadraṣṭár- provides a well-matched parallel with the Homeric phrase ὑπόδρα ἰδών, as they are “semantisch vergleichbare Bildungen,” as pointed out by Tichy.14 The Vedic word upadraṣṭár- first occurs in the Atharva- veda, the second-oldest text of ancient India, and is attested several times in the Brāhmana literature as well. Commenting on the occurrence of upadraṣṭár- in AV XI 3.53, Tichy observes that the word often designates a ritual expert in full command of the rite. While looking on the ritual performance an upadraṣṭár- is in a position to take notice of the deviations and failures of the ritual performer.15 This interpretation is further supported by a passage in MS I 9.7(2),16 where upadraṣṭár- explicitely refers to a Brahmán, a ritual expert stricto sensu.17 To be sure, upadraṣṭár- can also be used in other contexts. Kim has rightly shown that it “ist im YV häufig mit den Göttern (und Dämonen) in Verbindung gesetzt, z.B. mit Vāyu (TS III 3,8,5), mit Agni (KS XXXVI 13: 80,4-5), mit Nirṛti 13 On the verbal root see M. Mayrhofer, Etymologisches Wörterbuch des Altindoarischen I (Heidelberg 1986–2001) 704–706, ‘einen Blick auf etwa werfen’; on the etymology of Vedic úpa (‘herzu, zu, hin, hinauf ’) and Greek ὑπό see Mayrhofer 218–219. 14 E. Tichy, Die Nomina agentis auf -tar- im Vedischen (Heidelberg 1995) 19. However, Tichy did not give any further account for this correspondence. 15 Die Nomina agentis 154 n.20: “Das [upadraṣṭár-] heißt wohl: wer bei einer rituellen Handlung zusieht, die er auch selbst beherrscht, d. h. wer Ab- weichungen und Fehler bemerken könnte.” 16 See K. Amano, Maitrāyaṇī-saṃhitā: Übersetzung der Prosapartien mit Kom- mentar zur Lexik und Syntax der älteren vedischen Prosa (Bremen 2009) 345, who also argues that upadraṣṭár- is “wohl die Bezeichnung für einen, der bei einer rituellen Handlung zusieht, die er auch selbst beherrscht, so daß er Fehler finden könnte.” 17 On the meaning of Brahmán in the classical Vedic rite see J. P. Brere- ton, “Bráhman, Brahmán, and Sacrificer,” in A. Griffiths and J. E. M. Houben (eds.), The Vedas: Texts, Language and Ritual (Groningen 2004) 325–344. RUOBING XIAN 265 ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 (MS III 2,4: 20,3).”18 In addition, the occurrence of upadraṣṭár- in a newly edited hymn (AVP 8.15.3) describes gods as critical spectators examining the behavior of mortals by using the power of their gaze.19 Moreover, Kim has also observed that the occurrences of upadraṣṭár- occasionally describe a furious onlooker who brings about damage against the offender.20 In JB I 138, 13, for instance, upadraṣṭár- is linked to a ritual expert who might curse a ritual wrongdoer.21 Such features of Vedic upadraṣṭár- strongly invite us to compare it to the Homeric ex- pression ὑπόδρα ἰδών. The first two occurrences of the formula ὑπόδρα ἰδών in the Iliad are sufficient to demonstrate its semantics. At Il. 1.148 ὑπόδρα ἰδών is linked to Achilles when he feels that he is being wronged by Agamemnon and so responds to him with a ‘dark look’ (schol. D Il. 1.148 ὑπόδρα ἰδών· δεινὸν ὑποβλεψάµενος). Odysseus aggressively criticizes Thersites at Il. 2.246–264, because Thersites, despite his low social status, spoke to Aga- memnon ill-judged words (2.243–245): ὣς φάτο νεικείων Ἀγαµέµνονα ποιµένα λαῶν, Θερσίτης· τῷ δ’ ὦκα παρίστατο δῖος Ὀδυσσεύς, καί µιν ὑπόδρα ἰδὼν χαλεπῷ ἠνίπαπε µύθῳ· As argued by Holoka, ὑπόδρα ἰδών “conveys anger on the part 18 J.-S. Kim, “Die Nomina agentis auf -tar- im Vedischen: Funktion und Gebrauch,” in I. Balles and R. Lühr (eds.), Indogermanische Nomina agentis (Berlin 2007) 71–160, at 87–88. 19 According to A. Lubotsky, “PS 8.15. Offense against a Brahmin,” in A. Griffiths and A. Schmiedchen (eds.), The Atharvaveda and its Paippalādaśākhā: Historical and Philological Papers on a Vedic Tradition (Aachen 2007) 23–33, at 23, this hymn, “which represents an appeal to the Brahmins to stay united in their protest when one of them is abused,” is “unique in its subject mat- ter.” On this hymn see also J.-S. Kim, Die Paippalādasaṃhitā des Atharvaveda: Kāṇḍa 8 und 9 (Dettelbach 2014) 140–152. 20 Kim, in Indogermanische Nomina agentis 88. 21 See W. W. Caland, Das Jaiminīya-Brāhmana in Auswahl (Amsterdam 1919) 40–41; H. W. Boderwitz, The Jyotiṣṭoma Ritual: Jaiminīya Brāhmaṇa I, 66-364 (Leiden 1990) 78. 266 AN ETYMOLOGICAL NOTE ON HOMERIC ΥΠΟΔΡΑ ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 of a speaker who takes umbrage at what he judges to be rude or inconsiderate words spoken by the addressee … dark looks signal irritation and resentment and are meant to stop short an offender against social decorum.”22 In addition, Cairns points out that Homeric ὑπόδρα ἰδών “presupposes a claim to super- iority only in the sense that it takes upon itself the right to rebuke, to criticize, or to protest.”23 Holoka’s and Cairns’ ob- servations also square well with the use of the formula in the Odyssey. In two places ὑπόδρα ἰδών is linked to a suitor of Penelope and thus local aristocrat, who rebukes Odysseus aggressively, taking him to be a wandering beggar (Od. 17.458– 459, 18.387–388). By contrast, the other seven attestations of the formula in the Odyssey show Odysseus as aggressor, who in reality is the king of Ithaca and thus justified to rebuke the suitors besieging his palace. The use of the ὑπόδρα ἰδών for- mula is consistent in the Homeric epics as well as in its two appearances elsewhere (Hom.Hymn Bacch. 48, Scutum 445). Note that like upadraṣṭár- in Vedic literature, ὑπόδρα ἰδών can also be used of gods—unsurprisingly Zeus to whom a furious gaze is ascribed (Il. 5.888 and 15.13, δεινὰ δ’ ὑπόδρα ἰδὼν Ἥρην πρὸς µῦθον ἔειπεν).24 Some conclusions may be drawn as our evidence allows. In addition to the etymologically unmistakable correspondences between Homeric ὑπόδρα (ἰδών) and Vedic upadraṣṭár- as to the verbal root and the prepositional prefix, their semantic similarities are also too striking to be coincidences: both are applied to the authority of a superior onlooker who is justified to rebuke or criticize the one who violates social or religious decorum. An Indo-European (or at least Graeco-Indo-Iranian) pre-form *(h1)upo-derk ̑ might be reconstructed which underlies both Homeric ὑπόδρα and Vedic upadraṣṭár-. While the survival 22 TAPA 113 (1983) 4; cf. also Rakoczy, Böser Blick 44. 23 D. L. Cairns, “Ethics, Ethology, Terminology: Iliadic Anger and the Cross-cultural Study of Emotion,” YCS 32 (2003) 11–49, at 44. 24 See Holoka, TAPA 113 (1983) 10–11. RUOBING XIAN 267 ————— Greek, Roman, and Byzantine Studies 57 (2017) 261–267 of the archaisms in the Vedic Nomen agentis upadraṣṭár- is likely due to its exclusive use in ritual context, one may suppose that the archaisms rooted in the strictly fixed Homeric formula ὑπόδρα ἰδών25 are inherited from “a religious literature com- parable to the archaic and highly conservative poetic traditions of India and Iran [which] existed in second millennium Greece, transmitted by a sacerdotal élite—a sort of Mycenaean counterpart of the Vedas, which vanished along with that civilization.”26 March, 2017 Fudan University Shanghai, China ruobingxian@fudan.edu.cn 25 Two other linguistic features of this formula stand out. On the one hand, ὑπόδρα differs from other adverbial compounds of the same muster in omitting final sigma, which marks the adverbial function. See Chan- traine, La formation 5, “Noter la forme sans s”; cf. Risch, Wortbildung 364– 365. ὑποδράξ (Callim., Nic.) is a late form, built after ἐπιµίξ, ὀδάξ, etc. While Frisk’s thesis cited above that the form ὑπόδρα goes back to a neuter in adverbial function would account for the absence of final -s, I attempt to illuminate the prehistory of ὑπόδρα in light of the fact that the form is at- tested only in the fixed Homeric formula ὑπόδρα ἰδών, which is the seman- tically equivalent to Vedic upadraṣṭár-. On the other hand, ὑπόδρα ἰδών is also noteworthy in displaying the so-called ‘pléonasme impliqué’, since the phrase incorporates the verbal roots *derk ̑- and *u ̭ei ̯d-, both of which are verba videndi; cf. C. J. Ruijgh, Autour de “te épique”: études sur la syntaxe grecque (Am- sterdam 1971) 66, with special reference to Apollonius Dyscolus. 26 T. G. Barnes, “Homeric ΑΝΔΡΟΤΗΤΑ ΚΑΙ ΗΒΗΝ,” JHS 131 (2011) 1– 13, at 13, who continues: “Only bits and pieces of it resurface in the newly shaped literary genres of the first millennium. The project of identifying such bits and pieces would likely have important implications for our under- standing of the early development of epic.” Such “bits and pieces” seem not to be confined to epic composition. See Barnes, HSCP 107 (2013) 73–98; A. Nikolaev, “Showing Praise in Greek Choral Lyric and Beyond,” AJP 133 (2012) 543–572; J. M. Macedo, “Two Divine Epithets in Stesichorus: Poseidon ΙΠΠΟΚΕΛΕΥΘΟΣ and Aphrodite ΗΠΙΟΔΩΡΟΣ,” CP 111 (2016) 1– 18. I wish to thank GRBS's anonymous referee for his/her constructive com- ments and the editorial board for helpful suggestions