CC BY
15
D01:10.30842/ielc230690152555
Alexander Nikolaev
(Boston University, USA)
ETYMA GRAECA II
This paper proposes new Indo-European etymologies for the following Ancient Greek words: aXsn 'heat' (*h2leuhx- 'to burn', cf. Ved. rura- 'hot', Olr. loscaid 'burns', etc.); apsivrov 'braver, better' (*h2mei- 'to be strong', cf. Toch. B maiwe 'young', maiyya 'strength'); биаяерфеХос; 'dangerous, turbulent (of sea)' (*gwhembhh1- 'deep' cf. Ved. gabhira-, YAv. jafra-); sOsipa 'hair' ( < *h1(e)gh-uer-ih2-, cf. Toch. AB yok 'a single hair', pl. yakwa < *h1ighu-); Эфрос; 'biting, stinging' ( < *dhih2gw-ro-, cf. Lith. diegti / diëgti 'to poke, sting', Lat. figere 'to insert, pierce', Toch. AB tsaka- 'to bite; to pierce', etc.); i'Xiov 'female pubis' ( < *h2ih2/3-, cf. Pal. ihha-'flank', Lat. ilia 'groin'); рарятю 'seize' (*merkw-, cf. Toch. A mark- 'take away', Lat. merx 'commodity' < *'what is taken/bought'); ргрос; 'mime' (*meh2-i- '(to reveal by) making motion by hand', cf. Lith. moju 'wave hand', Proto-Slavic *majati 'to make signs with a hand', Yazgulyam amaw- 'to motion with a hand'); ютц 'bustard' ( < *uoti- 'bird-like', 'a sort of bird' ^ *uoti-, cf. Hitt. wattai- 'bird').
Keywords: etymology, Indo-European reconstruction, Ancient Greek, Anatolian, Tocharian, Indo-Iranian, Balto-Slavic, Latin, Celtic.
А. С. Николаев (Бостонский университет, США)
Etyma graeca II
В статье предлагается ряд новых этимологий для древнегреческих слов: aXsn 'жара' (*h2leuhx- 'гореть', ср. вед. rUra- 'горячий, пылающий', др.-ирл. loscaid 'горит', и т. д.); apsivrov 'лучше, сильнее' (*h2mei- 'быть сильным', ср. тох. B maiwe 'молодой', maiyya 'сила'); биаяерфеХо; 'опасный, бурный (о море)' (* gwhembhh1- 'глубокий', ср. вед. gabhira-, мл.-ав. jafra-); sOsipai 'волосы, грива' ( < *hi(e)gh-uer-ih2-, ср. тох. AB yok 'волос', pl. yakwa < *h1ighu-); Эфрос 'жгучий, острый' ( < *dhih2gw-ro-, ср. лит. diegti / diëgti 'колоть, жалить', лат. figere 'протыкать, вставлять', тох. AB tsaka- 'кусать, протыкать', и т.д.); i'^iov 'женский лобок' ( < *h1ih2/3-, ср. пал. ihha- 'пах', Lat. ilia 'брюшная полость'); рарлхю 'хватать' (*merkw-, ср. тох. A mark- 'забирать', лат. merx 'товар', т. е. 'то, что берут'); ргро; 'мим' (*meh2-i- 'показывать руками, жестикулировать', ср. лит. moju 'машу', праслав. *majati 'махать', язгулям. amaw- 'делать жест рукой'); ютц 'дрофа' ( < *uoti-^ *uoti-, ср. хетт. wattai- 'птица').
Ключевые слова: этимология, праиндоевропейская реконструкция, анатолийские, тохарские, индоиранские, балтославянские языки.
9. akm
Att. d^sa, Ion. d^sn 'heat, warmth' (Hom. +) is traditionally compared to the root of OE swelan 'to burn slowly', Lith. svilti 'to smoulder', Gk. si^n 'the sun's heat' (PIE *suelhi- 'singe, burn')1, but the complete absence of rough breathing (which would have been expected if the word continued a protoform with initial *su-) makes this etymology very problematic: for instance, d^saivoi^i and sv d^sa in Aristophanes (Ec. 540-41) are both transmitted with spiritus lenis in all of the seven manuscripts and no manuscripts aspirate d^saivsi (Men. fr. 892 K.-A.) either . The proponents of the etymological connection with *suelhi- have followed LfgrE 1.464 in supposing that psilotic Ionic d^sn was adopted as Attic d^sa from Ionian medical and scientific writings. However, in Greek literature the word refers to the warmth of the sun or heat in general and does not carry specifically medical connotations: in Od. 17.23 the
For the first instalment see Nikolaev 2014a from which the numeration in the present paper is continued. Greek texts are abbreviated after LiddellScott-Jones; Latin texts are abbreviated after Oxford Latin Dictionary. Translation of Homeric passages is by Lattimore, translations of other Greek and Latin texts are from Loeb Classical Library, unless otherwise indicated. Gk. = Ancient Greek, Goth. = Gothic, Lat. = Latin, Latv. = Latvian, Lith. = Lithuanian, Luw. = Luwian, Lyc. = Lycian, MCorn. = Middle Cornish, OAv. = Older Avestan (Gathic), OBret. = Old Breton, OE = Old English, Olr. = Old Irish, ON = Old Norse, PIE = Proto-Indo-European, Toch. = Tocharian, Umbr. = Umbrian, Ved. = Vedic Sanskrit, YAv. = Younger Avestan. ">" denotes regular phonological development, denotes morphological derivation.
I would like to thank Hannes Fellner, Jeffrey Henderson, Stefan Hofler, Boris Maslov, Craig Melchert, Sergio Neri, Alan Nussbaum, Martin Peters, Georges Pinault, David Sasseville, and Michael Weiss who kindly answered my questions and provided helpful feedback on individual ideas presented in this paper.
1 See GEW 65-66, EDG 62-63, DELG 53. This traditional etymology was recently advocated with customary learning and ingenuity by Hofler 20162017: 181-184, 190-191 who posited a derivational chain *suelhres-->
*sulhj-s-6- 'having glowing heat' ^ *sulh1-s-eh2 'glowing heat'.
2 2nd cent. CE grammarian Aelius Dionysius states that aXeaiv© was pronounced with a rough breathing (a 81 Erbse: Kai yap to 'aXeaiveaOai' Saativouaiv oi Attikoi), but this claim does not carry much weight since Aelius could easily have relied on analogy to genuine cases where an Ionic psilotic form stood next to an aspirated form in Attic.
reference is to the heat of the sun later in the day , while in Arist. Ec. 541 sv aXsa ... Kai oxpropaoiv "in warmth and coverings" is a hendyadis for "in warm coverings"4. The word aXsn and its derivatives are frequently used in the Hippocratic corpus, but again they never denote fever or bodily heat5. Since the word does not have a specifically medical or scientific meaning which could have prompted its adoption into Attic from Ionian prose, the lack of rough breathing in Att. aXsa presents an insurmountable obstacle for the derivation of the word from PIE *suelh/-.
A search for an alternative etymology is therefore warranted. The lack of contraction in Att. aXsa suggests that the hiatus is due to the loss of intervocalic *u, making the reconstruction of Proto-Gk. *aleu- plausible: the glosses aXsov Bsppov ^ x^iapov and aXso<^ §ianupo^ (Hsch. a 2281 and 2282 Cunningham) can be un-problematically taken back to *aleu-o-, while a zero-grade root form *alu- may be continued in aXuKpo<; 'lukewarm, tepid'6 and perhaps in Hsch. a 3296 aXnvsr ^nsi. Another formation made from the same Proto-Gk. root *aleu- is an s-stem *aleu-eh- reflected in the adj. aXssivo<; 'warm' and in the verb aXsaivro 'to warm up' formed from *aXs(s)h- with hyphaeresis7. Direct evidence for this s-stem is
3 aXX' epxeo^ epe 5' a^ei av^p o5e, xov ati KeXetiei; / arniK' enei K£ nupo; Oepe© aXen xe yev^xai "So go. This man whom you ordered will lead me, right after I get warm from the fire and there's heat".
4 os 5' ev aXea KaxaKeipevov Kai oxpropaoiv / KaxeXinov, &vep "I left you lying in warm spot, snugly blanketed, husband".
5 Cf. Vict. 3.65 nepinaxoioiv [...] Ppa5eaiv ev aXe^ "short walks in the sun"; Morb. Sacr. 13 ^ Kai ¿k yoxeo; ei; aXenv eXOfl Kai napa nop noXu "if he leaves the cold for warmth and a large fire"; Vict. 3.68 ^uXaooopevov xa; pexaPoXa; x©v yuxerav Kai xfl; aXen; eaOflxi naxevp "with the use of a thick garment to guard against sudden changes of heat and cold"; Aer. 19 o yap xei^rav KroXtiei Kai xfl; yn; ^ yiXoxn;, oxi ouk eoxiv aXen ou5e oKenn "they are stunted owing to the severe climate and the bareness of the land, where there is neither warmth nor shelter". The denominative verb aXeaiv© is used in the meaning 'apply warmth to, keep warm' in Aff. 26, Mul. 1.58, 2.17, Epid. 5.57, 7.76a.
Call. fr. 270 Pfeiffer yevxo 5' aXuKpa "they became warm", Nic. Al. 386 aXuKpoxepov 'slightly warmed', Suid. aXuKpa^ xXiapa 'warm', Hsch. a 3295 Cunningham aXuKpov eu5ivov.
7 Cf. peveaiv© 'to rage, to desire' derived from the s-stem found in pevo; 'rage, strength', apev^; 'feeble'.
found in the hapax 'warm'8 (Hes. Op. 49) which in my
opinion is best explained as an endocentric augmentation of the same type that we find in Homeric siks^o^ 'similar' beside sni-s(ks^o<; 'similar-ish', 'yellow, brown' beside srci^av0o<;
'brownish', etc.9. The meaning of is therefore 'hot-ish,
warm' which suits Hesiod's description of the smithy after hours quite well10. The stem *src-a^ss- can be analyzed as a form of an s-stem adjective -a^ssh- which matches the adj. a^ssivo^ < *aleehno-.
Proto-Gk. *aleu- can be compared with the following forms none of which have a satisfactory etymology: (a) Ved.
rura- 'deliriously hot', used to denote a kind of takman- 'fever'11 ( <
12
*h2luhx-ro-); (b) Proto-Iranian *rau- 'to burn' ( < *h2leuhx-) ; (c) Olr. loscaid 'burns, consumes by fire' ( < *luska-, a remodeling of thematic *luske/o- < Insular Celtic *luske/o- < *h2luhxske/o-) , and
8 The s-stem may also be attested at S. Ph. 859 aXe^; utcvo; eaOXoc; "sleep in the heat is sound" (trans. Jebb), although most editors (Lloyd-Jones and Wilson, following Radermacher and Kraus) adopt Reiske's conjecture
("good sleep has no fears").
9 See Stromberg 1946: 101-8.
10 nap 5' i'Oi xaAxeiov 0©kov Kai enaXea Xeaxnv / rapfl onoxe Kptioc; avepa; epyrov / iaxaver ev0a k' aoKvo; av^p peya oIkov o^eXXoi "pass by the bronze-worker's bench and his warm lounge in the wintry season, when the cold holds men back from fieldwork, but an unhesitating man could greatly foster his household".
11 AVS 1.25.4+. EWAia 2.456 "Nicht klar".
12 - —
Chorasmian caus. pcr'wy- 'to heat up' < *pati-rauaia-, Ossetic (Iron.)
arawyn / (Dig.) arawun 'to burn in fire', see Cheung 2007: 19.
13 British Celtic forms (MWelsh llosci 'to be on fire, to burn', MCorn. iesky 'to burn', OBret. lescsit gl. extorruit 'parched') cannot be reconciled with the Goidelic form without a host of ancillary assumptions: the only way of aligning the root vowel in Proto-British *loski- with that in Olr. loscaid is to reconstruct Proto-Celtic *loT-ske/o- where an o-grade would be extremely peculiar. But rather than following the contrived scenario of Klingenschmitt (1982: 194 n. 40) who reconstructed Proto-Celtic *lo(p)sk-i-, viz. as a causative made from a secondary root based on a reanalysis of *lh2p-ske/o- 'to be light' (Hitt. lap-, Gk. Xapn©), one may simply assume that MWelsh llosci, etc. are early Goidelic loanwords: this is a well-known and well-studied phenomenon, cf. MW byth 'always', MBret. bez, Corn. byth 'ever' which cannot be derived from Proto-Celtic *bitu- (irregular [0] in Welsh and Cornish, [z] in Breton) and are best explained as LW from Olr. bith 'perpetual'; cf. also MW kerbyt, OBret. cerpit 'chariot' vs. Olr. carpat 'id.' < Proto-Celtic *karbanto, or MW llwch (OW lichou) and
?
perhaps (d) Luw. (:)luha- 'warmth ' < *h2louh2/3-eh2 (with loss of initial *h2 by the Saussure's effect)14 and (e) Lat. lustrum 'purification ceremony' < *h2luhx-s-tro- 'instrument of burning' (and hence of purification)15. Based on these forms, we can reconstruct a new PIE root *h2/euhx- 'to burn; to be hot'16 and its derivatives, *h2leuhx-es- > Gk. *aXssh- and *h2leuhx-o- 'burning, hot' (cf. aXso<;) substantivized as *h2leuhx-eh2 > Ion. aXsn, Att. aXsa.
10. a^sivwv
a^sivrov 'better, braver' does not have a universally accepted
1 "7 •
etymology : this word (for which an original *ei in the root is made certain by Lesbian and Thessalian personal names in A^siv-) may be compared to Toch. B maiwe 'young', maiyyau (adj.) 'powerful, strong', maiyya ~ maiyyo 'strength, power' under the assumption that the original meaning of IE *h2moi-uo- was 'strong' and it could be used to refer to youthful vigor, hence the meaning of Toch. B maiwe18. Under this analysis, a^sivrov would have to be analyzed as a faux comparative, viz. an adj. *a^sivo^ 'strong' ( < *h2mei-no-) secondarily pressed into service as a suppletive comparative to
OBret. loch 'lake, pond' vs. Olr. loch ( < Proto-Celtic *loku-), see Rhys 1895; Pedersen 1909: 24; Bauer 2019.
14 It is unclear at present whether the same preform * (h2)louh2/3-eh2 > Luwic *luHa- is the source of Lyc. B 3 pl. ipv. lugatu (TL 44d.60; see GLyk 187) and Luw. (FLAMMAE)lax-ha-nu-wa/i-tax 'he burned down'.
15 If one is willing to believe that the purifying agent in the original version of the lustrum was fire which is an assumption that our present knowledge of the ritual does not allow to either confirm or deny. As a parallel, compare purgo / purigd 'purify ceremoniously' < *pur-ago < *puh2r-h2ge/o- 'to lead the fire about' with the PIE word for 'fire' (Umbr. pir, Gk. nop) as the first compound member (Thurneysen 1912-1913: 278) or februa 'means of purification' usually taken from *fexro-, PIE *dhegwh- 'to burn' (Pisani 1979), but recently argued to go back to *fesro- < *d(u)h2es-ro-, PIE *dhueh2- ' to smoke' (Lat. suffire, Gk. Oti©) by Vine 2020.
16 Depending on the assessment of Luw. (:)luha-, the root-final laryngeal may be specified as *h2/3-.
17
EDG 91, DELG 71, EDG 86. For an overview of usage and semantics as well as for further bibliography see Dieu 2011: 29-47.
18 A competing school of thought assumes that the original meaning of the Tocharian words was 'young' < 'small', and compares ON mjor 'slender, delicate' < *moiuo- 'little', following Lane 1938: 24, see DoT2 509.
ayaBôç19 and remodeled as an -on-stem by analogy to other n-stem comparatives20.
11.
The rare epic word ôuorcé^s^oç occurs once in the Iliad and three times in Hesiod. Its exact meaning is uncertain: in II. 16.748, Hes. Th. 440 and Op. 618 ôuorcé^s^oç is applied to turbulent sea or the
dangers of seafaring21, while in Op. 722 it is used of a person,
22
presumably in the sense 'disturbing' or 'troublous' . In later Greek ôuorcé^s^oç occurs in the epigram composed by a certain Herodi-cus the Babylonian (FGE 1) possibly active in the second or first century BCE: there is no consensus about the text and the
interpretation, but the meaning 'stormy, troublesome' is certain
01
enough . Finally, the word undergoes a large-scale revival in Nonnus where it is used as a recherché epithet of winds, torrents,
24
and Hesiod's native town notorious for its miserable weather .
19 Cf. Il. 15.641-2 apsivrov / navToia; apsm;, npsv n65a; paxsaOai "stronger in every faculty, whether running or fighting".
20 nXsov (n.) : nXsrov (m./f.) = apsivov (n.) : x, where x is resolved as apsivrov.
21 Il. 16.746-8: si б'л nou каг novx® ev ixOuosvTi yevoixo / noXXou; av Kopsasisv av^p 65s rpOsa / vno; anoOproaKrov, si каг биаперфеХо; sin "If only he were somewhere on the sea, where the fish swarm, he could fill the hunger of many men, by diving for oysters; he could go overboard from a boat even if [the sea] is Svansp^slog"; Hes. Op. 618 si 5s as vauxi^in; 5иал£рф£Хои ipspo; aipsi "if a desire for Svone/upelog seafaring grips you"; Th. 440 oi y^auK^v 5uaп£pфsХov spyaZovrni "those who work hard on ёиаже^феХод sea" (but see Fusillo 1981 who argued that in this verse the kenning for 'sea' is то 5uап£pфsХov: by using a substantivization with a literal meaning 'the turbulent one' to refer to the sea, Hesiod would demonstrate, yet again, his notorious aversion to the marine world).
22 pn5s no^u^sivou 5aixo; 5uaп£pфsХo; sivai "and do not be дшпе/иреХод in your mood at a dinner with many guests".
23 тойО' upiv sin, 5uап£pфsХol "let these things be yours, you SvanejupeAoi men". The editors, including Page, print voc.pl. assuming that the epithet refers to the rival grammarians of the school of Aristarchus ("you, turbulent people"), but the manuscript tradition of Athenaeus 5.222A, where the epigram is transmitted, has 5uап£pфsХov which could be construed with preceding тошо ("may this stormy (fate) be yours"). As modern commentators have correctly observed, the epithet resumes the nautical theme of the first verse фЕ^Ет', Apiarnpxsioi, en' supsa vrorn OaXaaan; / EXXa5a.
24 Dion. 2.550 (aupn); 13.75 (AaKpn); 22.171 (npoxsrov nornpo; 5uaп£pфsХov aAx^v); 24.64 (aupn).
There can be little doubt that the original sense of Suons^^sXo^ has to be sought in its application to the sea and its dangers .
The isolated26 Greek word is in my opinion related to Indo-Iranian *gabhHra- 'deep' > Ved. gabhira-, YAv. jafra-27: as a danger-conveying epithet of the sea, Gk. *dus-kwemphelo- is analyz-able as a bahuvrlhi-compound meaning 'having depth, which is bad' or possibly 'having too much depth, which is bad'. The initial voiced aspirate in PIE *gwhembhH- would undergo dissimilation in both branches (Grassmann's Law): *(-)gwhembh- > *(-)kwhemph- > *(-)kwemph-29 > Gk. (-)rcs^-30, *gwhmbh- > Indo-Iranian *gabh-31.
Since the -i- in Ved. gabhira- seems to point to a root-final laryngeal, the most economical solution would be to analyze the second member of Proto-Gk. *dus-kw(h)emphelo- 'having depth, which is bad' as a reflex of an adjective abstract *gwhembhhrlo-m
It is uncertain how much trust may be placed in Hsch. n 1385 nep^eXa^ StiaKoXa, xpaxsa, ßaöea 'difficult, rough, deep': Hansen is justifiably skeptical and suggests in his app. crit. that this is a glossa decurtata (viz. <5ua>ne^98Xa).
26 The etymological dictionaries are unanimous in calling Suanep^eXo; an "expressive word without etymology" (GEW 426-7; DELG 290; EDG 361).
27
The initial palatalization in the Avestan form is due to a contamination of full-grade root *jambhH- with zero-grade *gabhH: the initial consonant was leveled out in both branches, as *g- in Indo-Aryan and as *j in Iranian (jafra- < *jabra- < *gabra- X *jamban-).
8 Not simply 'having bad depth(s)'; compare Suanxn; 'bringing woe (nxn), which is bad' (noXepoio Suanxso; 7x Il., not tautological 'bringing bad woe') or Suaa^; 'having too much gust, which is bad' (avepoio Suaaeoc Il. 5.865).
OQ Lt
I accept the rule according to which intervocalic ND > ND in posttonic syllables on the way to Greek (see Miller 1977: 37; Hajnal 2005: 196-8;
Kümmel 2013: 160, 168-70, Miller 2010: 234-7), but even though this rule
h h
predicts that *(-)gw emb H(e)lo- should develop into Gk. |(-)rcepßeAß;, there would be plenty of ways for analogy to eliminate the allomorphy, especially since the derivational basis of
*gwhembhH(e)lo-m 'depth' would have been an oxytone adjectival stem in *-lo- which would not be subject to Miller's rule.
30 The "Aeolic" development of initial *kw > p before *e is fully consonantal with the fact that in early Greek literature Suanep^eXo; occurs only in epic and may therefore be at home in epic Aeolic dialect.
31 See Mayrhofer 1986: 112-5.
'depth' derived from the adjective *gwhmb%-l6- 'deep'32 directly reflected in Ved. gabhira-, YAv. jafra- and, with a different suffix,
-5 -5
in Toch. A kupar 'deep, deeply, depth' .
It remains to clarify the meaning of the Greek word. The passage from the Iliad cited above (16.748: si 8^ пои ка! novxro sv ixBuosvxt ysvotxo [...] si ка! 8иопг^фв^о<; sin) is interesting in two ways: its syntactic structure shows that for the poet 8иопг^фв^о<; was not an obsolete epithet fossilized in a formula, while the use of the epithet in the context of oyster farming makes it clear that the poet did not associate 8иопг^фв^о<; with depth, since oysters are harvested in shallow waters34. The most adequate translation would be 'troublesome, dangerous' which suits the Hesiodic usage perfectly and presents an unobjectionable semantic development from the original meaning 'having (too much) depth, which is bad'.
12. £9£ipa(i)
In Homeric epics the word гВвгра is used twice to refer to the golden mane of horses belonging to Zeus and Poseidon and in another three instances the word is used of the horse hair crest on Achilles' helmet
-5 С
(including when it was donned by Patroclus) . In the Hymn to Apollo (5.228) 80stpa denotes Tithonus' hair which has just begun turning grey, while the author of the Hymn to Dionysus chose to recycle the Homeric verse-end phrase ка^а! 8г nsptoosiovxo sBstpat "the lovely fringes were shaken" (scil. the plume on Achilles' helmet) to apply it to the god's locks. The word is only used in the plural in early hexameter poetry where nom.pl. гВвграг is
For the mechanism of producing adjective abstracts with an accent-shift and a concomitant insertion of a new full-grade in the root see Rau 2007. Compare the following pairs (back-reconstructed for PIE): *hxrno- 'guilty' (Ved. rna-) ^ *hxerno-n 'guilt' (YAv. arsna- 'injustice' attested in ardnatcaesa- 'punishing injustice') or *gnh1to- 'born' (Ved. jata-) ^ *genh1to-n 'the born one, child' (OHG kind).
33 See Pinault 2019: 499 (kupar < *kwd^ar(B < CToch. *kwamfiar/z < *gwmbhH-ro-). Of course, the Indo-Iranian forms may also go back to *gwhmbhhrro- 'deep' with the same suffix as in Toch. A k^par.
34 Cf. also Hes. Th. 440: those who yXauK^v Suanep^eXov epy&Zovxai are fishermen, not sailors, so this poet is not talking about deep sea either. I thank B. Maslov for drawing my attention to this.
35 Cf. also ¿OeipaSec; (Didymus' reading for yeveiaSec; of the paradosis) of Odysseus' beard at Od.16.176.
consistently verse-final. The word appears to have been exclusively poetic and does not appear in inscriptions until the Imperial times.
There is no satisfactory etymology for 808ipa on the books: "nicht sicher erklärt" (GEW 446), "uncertain" (EDG 376), while DELG 300 and LfgrE 2.410 endorse Frisk's earlier comparison with 'push, thrust' and with 80rov paraphrased in the scholia as 'ßXdnxrov'36. While in theory hair may be described as 'that which
-5 -7
one can shake back and forth' , methodologically it appears more prudent to look for potential cognates referring to hair.
808ipa looks like a substantivized feminine adjective made with the same possessive suffix as ni8ipa 'fatty, rich' ( = Ved. pivari 'id.') standing next to masc. rcirov 'fatty' ( = Ved. pivan-) and
"3 o
ultimately derived from *pihxur (Gk. rciap 'fat')38. If we apply this morphological analysis to 808ipa, one of the possible ways in which the form can back-reconstructed is *hj(e)ghuerih2-39, a derivative of *kiegh-ur /-uen- (cf. *pihxur ^ *pihxuerih2 > rci8ipa). The hetero-clitic stem *hiegh-ur /-uen- with a complex suffix may have been derived from an u-stem, cf. *seh2ul / -uen- 'sun' ^ *seh2u-'burning' or *dhemur / -uen- (Lat. femur, feminis 'thigh') ^ *dh(o)/emu- (Gk. 'thick')40. Precisely such an u-stem is in my
opinion directly continued in Toch. AB yok 'a single hair' < *h/!gV41. The plural yäkwa (PK NS 18 A al; M 3b7) with a-umlaut
36 See Frisk 1930.
37
As a parallel, Frisk adduced Lat. crinis which he compared to Goth. af-hrisjan 'to shake off; de Vaan calls this analysis "possible, but not very appealing" (EDLIL 145).
3 For this well-known derivational mechanism see Nikolaev 2019: 793.
39 For phonological development cf. *ghuer (Russ. zver') > 0^p).
40 As a parallel for word meaning 'wide, thick, fat' giving rise to a word for 'thigh' cf. Mod. English thigh < Gmc. *peuho- < PIE *teuk- (~ *teuhx-), Lith. tukti 'to be(come) fat'. PIE *dhemu- may either have been substantivized as 'the thick one' or have acquired the meaning 'thigh' via the mechanism known as "transferred epithet", assuming that the word for 'thick', 'thickness' was used in a fixed expression with a word for 'leg' (in Nikolaev 2010: 58-63 I used an exempli gratia reconstruction emu-*sreno- 'thick upper leg, thickness of the upper leg'; cf. Swed. (dial.) bus-lagg 'thigh' lit. 'swollen leg' [see Rietz 1862-7: 67] or Wakhi bajldng 'thigh' lit. 'thick leg', see Steblin-Kamenskij 1999: 92). On the difference between these two mechanisms, namely, ellipsis-based substantivization and direct substantivization, see Hofler 2020.
41 DoT2 550-1 compares yok with Arm. asr 'fleece' and Ved.
confirms the u-stem, while sg. yok shows u-umlaut: yok < * '(ku < *h1eghu- 'hair'. Toch. B or : pl. arwa 'wood' ( < *doru-) provides a perfect phonological parallel for both sound changes, as Hilmarsson 1985 already recognized42. Further descendants of *hjegh- 'hair' are unclear at the moment43.
Gippo^
In Hellenistic poetry we find adj. BtPpo^ of unknown meaning and etymology44. In Nic. Th. 35 the word is used of painful condition caused by a snake bite45, while in Al. 554-6 the word is used to
yasu- tentatively translated as 'pubic hair', but the latter gloss is entirely conjectural: yasu- 'semen discharge' can be unproblematically analyzed with Karl Hoffmann (apud Witzel & Goto 2007: 685) as a derivative of *hxiehl-(k)- 'to throw, to send forth' (LIV2 225), cf. Lat. eiaculari; as to Arm. asr, it has been traditionally and plausibly analyzed as a reflex of *poku- (cf. Lat. pecu, Ved. pasu-, Myc. po-ku-ta, etc., see Pinault 2003: 169-173 and Martirosyan 2010: 122-124). Another etymology of Toch. AB yok, not registered by the etymological dictionaries, is due to Paul Widmer apud Balles 1999: 11 n. 21: Widmer reconstructs an acrostatic i-stem *h1egwi- from the root *h1egw- 'to gleam, to blaze' (*h1ogw-ni- 'fire') further identified by Balles in Lat. sanguis (< *h1sh2n-h1gw-o/i- 'hell leuchtend wie Blut') and by Nussbaum apud Balles in Gk. epuoiPn and aKpTPnc (< *-h1gw-); the downsides of this analysis of Toch. yok are that pl. yakwa must be treated as analogical and that the meaning 'color' ( < *'gleaming') must be primary and the meaning 'hair' would be secondary, while the reverse ('hair' > '(any) color') is a much better-attested semantic development (see DoT2 550).
42 See Malzahn 2010: 10-12 for these sound changes.
43 In principle, Hitt. auwawa(i)- (c.) 'spider' may be a Luwianism and continue *hlegh-uo- 'hairy', hence 'a hairy spider' with a characteristic Luwian loss of a voiced velar and change of Proto-Anat. *e > a, but this etymology is not necessarily compelling from the zoological perspective and the meaning 'spider' is not certain, see Boddy 2021: 337 with references to older literature. Leaving this doubtful connection aside, for Greek and Tocharian one may reconstruct either *gh or *gh and possibly even *gw(h). The similarity with PIE *(h1)egh- 'out(side)' is potentially intriguing.
44 GEW 674, DELG 420, EDG 549. We also have personal names (©ippo;, ©iPprov, ©iPpaxo;), see Bechtel 1917: 508. The earliest onomastic attestation is LePpov (viz. ©ePpov) Alcm. fr. 1.3 PMGF.
45 ©iPp^v 5' e^eXaoeic; o^irov eniXroPea K^pa "you may expel the Oifiptfv and harmful doom that snakes bring" (ed. Jacques 2002, trans. Gow and Schofield 1953).
describe turtle eggs cooked on the coals46. A scholion on Nic. Ther. 35a (Crugnola) translates 0iPpo^ as 'hot', but there is no reliance on this ancient tradition: the paraphrase 'hot' may simply be due to the fact that the word 0iPpo^ bears some similarity both to 0sp^6<; and to xc^pa. I propose that the meaning of 0iPpo^ may have been 'stinging, biting, mordant' which is an epithet equally well-suited for sharp pain caused by snake's venom and for slightly sharp,
An
piquant taste of turtle eggs47.
If the word 0iPpo^ belonged to the same semantic field as o^u<;, 8pi^u<;, or niKpo^ and referred to various unpleasant sensory experiences, the use of the word in two more Hellenistic fragments becomes clear. The same scholion on Nic. Ther. 35a cites Call. fr. 654 (Pfeiffer) 0i0p^ Kurcpi8o<; ap^ovin^ (Ap^ovin^?) and Euphorion fr. 115 (Lightfoot) 0iPp^v ts Ss^ipa^iv48. Arena 1970 aptly compared the Callimachean fragment with Emp. fr. 122.2 DK Ap^ovin and marshalled other arguments in favor of the
view that for Hellenistic poets the word 0iPpo^ was confused with ("contaminato") and influenced by another rare word 08^8po^ 'solemn, august, venerable'. This is in principle possible, but a more economical solution would be to accept that in these two fragments Aphrodite and Semiramis are described as 'stinging', 'biting', 'tart' or 'bitter'. Both the goddess and the queen are emblematic of Eros49, and it requires no detailed substantiation that for Greeks Eros could be stinging or bitter: one only needs to think of Sappho's "Epo<; ... yAurcurciKpov (fr. 130 Voigt) or Aeschylus' §n^i0u^ov sproxo^ av0o^ "heart-stinging flower of love" (Ag. 743). It may even be possible to go a bit further and surmise that the word 0iPpo^ carried the specific
46 Nal p^v pnxivn xs Kai iepa epya pelican; / piZa xe xaApavoeaaa Kai ©ea Oi^pa xeAwn; / aAOaivei xoxe vepOe ropo; Za^eAoio Keparp; "again, pine-resin and the sacred produce of the bee and the root of all-heal and the Oifipd eggs of the tortoise are curative when you mix then on a hot fire".
47 An excellent parallel in English, suggested to me by B. Maslov, may be seen in the word tart whose meanings, at least through its history, have ranged from 'sharp, severe, painful' to 'sharp to the sense of taste, pungent' to 'acrimonious' (of a person).
48 Following the scholia, students of these fragments have translated Oi^p^ with 'hot', 'burning', or 'sultry', e.g. D'Alessio 1996: 2.761: "della bruciante Cipride"; Lightfoot 2010: 357: "sultry Semiramis"; Acosta-Hughes & Cusset 2012: 170: "l'ardente Semiramis".
49 For Semiramis' excessive lust cf. e.g. Diod. Sic. 2.13.4.
connotation of bee sting: for Eros imagined as a honey-bee flitting about the flowers cf. пот' snpov / sv тоц po§ot^ "Ерюта (Anac. 6.12 West), and a specific association between sting of love and sting of a bee is a common topos in post-Classical Greek poetry.
If 0фр6<; indeed meant 'biting, stinging'50, the etymology presents itself virtually unbidden: I submit that that the adjective goes back to Proto-Gk. *thTgwro- ( < *dhih2gw-ro-) made from the same root as Lith. diegti / diegti 'to poke, sting, hurt, prick' ( = Latv.
с 1 ГЛ
diegt 'to stab'), ¡diegti 'to sting' , dygus 'prickly' , Latv. daiga2
ГЛ
'kind of fly'53, OE dic 'ditch, pit' (and perhaps Mod. English dig), OHG dih 'moles, gurges'54, Lat. fTgere 'to insert, fix, pierce'55, and Toch. AB tsaka- 'to bite; to pierce'.56
Formally, *d4h2gw- ( < *d^h2igw- with a laryngeal metathesis) is best seen as a zero grade of PIE *dheh2igw-57 of the same complex
50 Hesychius' lexicon offers a long series of interpretamenta for OiPpo;, but one of them — specifically, the one that does not seem to come from ancient exegesis on Callimachus, Euphorion, or Nicander, — offers additional support for the hypothesis put forth in this paragraph: OiPp^v-[...] xive; 5e xaXerc^v (Hsch. 0 579 Cunningham).
51 The LKZ illustrates this meaning with a proverb that amusingly leads us back to Greek: bek nuo grieko kaip nuo zalcio, nes, jei tu prisiartinsi, jdiegs tave "run away from a Greek as if from a snake: if they get closer, they will sting you."
52 See Smoczynski 2007: 109.
53 See Muhlenbach & Endzelin 1923-1936: 2.430.
54 See EWAhd 2: 630-4.
55 The inscriptional form FIGIER (inf. pass.) in the Senatus consultum de Bacchanalibus (186 BCE, CIL I2 581.27) indicates that the long -T- in the forms of the Latin verb cannot continue an old diphthong (contrast EXDEICATIS in the same inscription, l. 22, from dTcere < PIE *deik-), while Old Lat. fTuere 'id.' (Cat. via Paul. Fest p. 92 M.) and the nounfTbula 'pin' ( < *fiuibula < *fiuedhla) show that the final consonant of the root goes back to PIE *-gw- (word-medial -g- in Classical Latin stem fTgere was introduced by analogy to the perfect, see Bock 2008: 239-240; Garnier 2010: 398.
56 See Ringe 1988-1990: 71; DoT2 800.
The disyllabic structure of *cteh2igw- suggests a secondary root. Such an extended root would not be unparalleled: cf. Gk. pai^o; 'bent inward', Goth. wraiqs 'crooked' (< *ureh2igw-) or Lat. saepes 'hedge' and Gk. aiya 'quickly' (from 'closely packed'?) on the basis of which de Vaan (EDLIL 533) reconstructed *seh2ip- clearly relatable to *seh2i- 'bind'; see also Janda 2001-2002.
structure as e.g. *gweh2id- > Gk. ^aiSpo^ 'bright', Lith. gaidrus 'id.'58. A reduplicated present stem *dhi-dheh2igw- 'to bite'59 ( > Proto-Gk. *t(h)ithaigw- with Grassmann's Law) made from this root would make a plausible point of origin for a nominal stem *t(h)ithaigw-o- 'act of biting; a bite, a sting' . From this reduplicated noun we can in turn posit a denominative verb *xi0aiP6ro 'to provide with a bite', hence 'to sting' (cf. nnpyo^ 'turret' ^ rcupy6ro 'equip with fortifications')61 from which first an agent noun *Ti0aiProxn^ 'stinger' (cf. 8§v6ro 'betroth' ^ 'matchmaker') and then a
verb in -roooro 'to sting' was derived by familiar rules of word formation (cf. ayproxn^ 'hunter; hunting' ^ ayproooro 'catch by
58 In LIV2 142 the root is reconstructed as *dheiHgw- 'to stick in', but this phonologically suspect reconstruction is proposed solely on the evidence of Lith. diegti, Latv. diegt and other forms in Baltic languages that point to Proto-Baltic *deig- / *daig-. As is acknowledged in the first footnote to the LIV lemma, the Proto-Baltic full grade *deig- may be secondary and in fact, there are many examples of such secondary full grades in Baltic languages. For instance, Lith. seilas 'noose, rope' ( < *sei-) can only be explained as a derivative of the root *seh2i- 'bind' discussed above in n. 57 under assumption that a new full grade *sei- was created in Proto-Baltic to match *si- < *sih2- < *sh2i- ~ *seh2i-. Similarly, the only way to align Lith. rieti 'to scold', Latv. riet 'to bark' ( < *rei-) with Russ. rajat' 'to make a noise' and Latv. rat 'to rebuke, scold' (< PIE *reh2i-) is to posit a secondary full grade *rei-. Examples could be multiplied; the point is that Baltic evidence for *deig- (to which one should add some Slavic forms overlooked in etymological dictionaries, namely, Sloven. and Croat. degati se 'to argue' < Proto-Slavic * degati, see Г. А. Ильинский apud Trubachev 1957: 95) is not incompatible with the reconstruction eh2igw- marshalled above.
59 Remodelled either from athematic *dhe-dhek2igw- (compare *dhe-dhek1->> *dhi-dhekj- > xiOnpi) or from thematic i-reduplicated *dhi- dhik2gw-e/o-(compare *s(t)i-sth2-e/o- >> *s(t)i-steh2- > iax^pi).
0 Cf. auvsoxpoc; 'joining, joint' ( < *-hehokhmo-) from the root of auvsx® (see Solmsen 1901: 256).
1 As A. J. Nussbaum reminds me, there is more than one way of arriving at *xi0aiPo®, and while it is possible that the hypothetical nominal stem *х19афо; / *xi9aiPn 'biting' was concretized to 'a bite/sting' and served as a derivational basis for a factitive verb *х19аф6ю 'to make a bite/sting', a different approach is just as possible: one could posit a nominal stem *xi0aiP6; with a passive meaning 'bitten, stung' that would make an -6ю present stem meaning 'to render bitten / stung', hence 'to bite, to sting' (compare e.g. Goo; 'sharp, sharpened, whetted' vs. 9о6ю 'to make sharp' on which see Nikolaev 2018).
hunting' or nrcvoro 'sleep' ^ ^ nrcvroooro 'be sleepy').
At this point our reconstruction forward is no longer an idle string of hypothesis, since the verb xiBaiProooro is attested in the Odyssey in the description of bees in the cave of the Nymphs: Od. 13.103-106 (ed. West):
? r /\ C» 5 9 V 1 r 9 C> f
ayxoOi o anx^ avxpov snqpaxov qsposios^, ipov Nu^arov ai vn'ia§8^ KaXsovxai.
? c» "\ <*v r **
sv os Kpnxqpe^ 18 Kat a^^i^opqs^ saoiv Xaivor svBa §' srccixa xiBaiProooouoi ^sXiooai. And near it a pleasant dusky cave, sacred to the nymphs who are called Naiads. In it are mixing bowls and amphoras of stone, and bees TiOaifiwaaovai there. The context gives no clear indication what the precise activity of the bees described by the verb xiBaiProooro might have been and modern scholarship has been agnostic about the meaning of the verb62. The ancient commentary offers the translation 'to store up (food / honey)' (xiBsvai Pooiv), but this is doubtlessly a folk etymology prompted by the similarity of first two syllables of xiBaiProooro to the stem of xiBq^i63. I propose that the verb here refers to one of the prototypical activity of the bees and should be translated as 'sting'.
It is entirely plausible to assume that the passage describes bees stinging mortals who dare enter the cave via the way of the gods. But it is also possible that the verse-final phrase xiBaiProooouoi ^sXiooai was no longer transparent to the singer, since because the verb had become obsolete: it was extracted from its original context (in which it meant 'bees sting') and used in the description of the Cave of the Nymphs simply because the latter featured bees64.
62 Skoda 1982: 214 ("obscur"); Hoekstra 1989: 171 ("exact sense and etym. unknown"); Rengakos 1994: 120 ("das immer noch unerklärte Hapax"); EDG 1482 ("origin ?"); Bowie 2013: 116: "a very rare word, of unknown meaning and etymology".
63 This folk etymology is responsible for the usage of the verb in later poetry: Antim. fr. 108 (Matthews) 'put away', [Lyc.] 622 ('irrigate'), Nic. Th. 199 ('nourish'). The famous image of a bee depositing honeycombs into Pindar's mouth in his metrical vita (peXiooa xi^ © eni aipßX© / XeiXeai v^niaxoiai xiöaißroaaouaa noxaxo, ed. Drachmann) is certainly due to a conscious allusion to the Odyssey passage.
64 For other similar cases see Leumann 1950: 208-235.
iXiov
Hsch. i 563 tXiov to Tq<; yuvaiKo^ s^Paiov Kai koo^ov
yuvaiKsiov napa Kroio^ (Cunningham) seems to contain two different words: the Coan word for a woman's ornament (koo^ov yuvaiKstov) is in all probability a derivative of dXsro 'to turn, to wind'65, while the word for female pubis (s^Paiov)66 looks very similar to Lat. (neut. pl.) ilia 'loin, groin, flank' to which it has been compared since Stokes 1894: 4667. The length of the first vowel in tXiov cannot be ascertained, but the length of the initial [i] in ilia is metrically assured (e.g. Verg. G. 3.507). If the Greek word is cognate with the Latin one68, we must reconstruct either *hxeiliio-(assuming itacism in Gk. iXiov) or *hxihxliio-69. That the latter option should be preferred is shown by Proto-Gmc. *ilip- > ON il 'foot sole' which is best derived from *hxihx-l-et- ^ *hxihx-lo- with pretonic "Dybo-shortening" . Under this analysis, the consonant *l is suffixal and not a part of the root.
A new comparandum may be available in Palaic where we find a mysterious word ihha- referring to a body part manipulated during the dissection of a sacrificial animal: the sacrificer is instructed to
• •"71
cut off the leg of the animal and hold the ihha in their hand . The meaning 'flank' suits the context admirably; if it is correct, Palaic ihha- may go back to *hjih2/3-o- 'flank' which may in turn be easily aligned with *hjih2/3-lo- / *hjih2/3-liio- posited above as the hypothetical source of Latin, Greek, and Germanic forms. Further
65 See GEW722, DELG 445, EDG 588.
66 The etymological dictionaries usually add Hsch. i 549 i'Xia^ 5©pa yuvaiKeia with Meineke's change of the interpretamentum into popia which is not obvious.
67 Comparison with Gaulish PNs Iliatus, Ilio-uico, Illio, etc. proposed ibid. is justifiably deemed implausible by Delamarre 2003: 189.
68 It is possible that i'Xiov 'pubis' is a loanword from Lat. ilium 'loin' (DELG 445), although the meanings are not identical.
9 The word is usually cited in works of reference with gen. -ium which, however, is not attested before Cassius Felix (5th cent. CE). Cat. 63.5 has gen. sg. ili (if B. Schmidt's conjecture is accepted), while dat.-loc. pl. ilibus is first attested in Celsus (2.7.1.1), who also has iliis, and Juvenal (5.136). There is thus no certainty that ilia is an i-stem rather than an io-stem.
70 Next to *hxih2-liio- > Gmc. *ilia-> Far. ill 'id.'. See Magnusson 1989: 418; Kroonen 2013^ 269.
71
See eDiAna s.v. The text (CTH 753.1.A) is currently prepared for publication by S. Gorke and D. Sasseville.
evidence for PIE *h1ieh2/3- / *h1eih2/3- in designations of body parts remains to be discovered.
"7 "7
Gk. ^aprcxro 'seize, take hold of, overtake, strike' is a well-attested
"73 7A
verb which has no etymology . The root ^aprc- can be straightforwardly back-reconstructed as either *merkw- or *merp-15. Since the root is not attested in Mycenaean16, the choice between root-final *kw and *p cannot be made on Greek-internal grounds, but the former option allows discovering plausible comparanda for Proto-Gk. *merkw- / *mrkw- in Tocharian and in Latin.
o
nn
Toch. A mark- means 'to take away' , cf. the privative compound sne-marklune which translates Sanskrit aharyo 'not to be taken away' (A 359.15). This meaning squares nicely with that of the Greek verb, e.g. Od. 10.116 anxix' sva sxaprov
rorcAiooaxo Sdnvov 'at once he [scil. giant Antiphates] seized one of my comrades and prepared him for dinner' or Sapph. 58.11-12 aAA'
77
Ruijgh (1957: 166) assigned this poetic word to the "Achaean" lexical stock based on the уАюаааг каха noAsic; that attribute рарлхю to the Cypriot dialect (Kunptov. spapysv sAaPsv), see Bowra 1970 [1934]: 43. Egetmeyer (2010: 488) very plausibly hypothesized that the verb AapPav® is not attested in Cypriot inscriptions precisely because it was replaced by раряхю in everyday use as the default verb 'to take'.
73
The nominal derivatives include раряхц 'kidnapper' (Aesch. Suppl. 826), Марфао;, a speaking name of a Centaur on a Chalcidian vase (CHA 23 in Wachter 2001: 188), and Kappapyi; psxpov aixiKov, то ^pipsSipvov. AioAsi; (Hsch. к 596 Cunningham).
74 GEW 2.178: "ohne aussergriechische Entsprechung" ( = EDG 907-8: "system without outer-Greek cognates"); DELG 643 is similarly agnostic. The comparison with papvapai (PIE *merh2— *merp-/-kw-?) hesitantly proposed by Egetmeyer (2010: 488) is a counsel of despair.
75 Original *r in the root is evinced by the variant врал- showing a different vocalization of *r: Hsch. в 1083 Ррауаг auAAaPsiv, avaAroaai, Kpoyai, Onpetiaai; в 1049 ep&nxsw saOisiv, Kponxsiv, ^avi^siv; s 94 ёвраяхе^ SKpunxev. ¿Аафи^; s 97 sepaysv sKpuysv. snisv. Kax^aysv (Cunningham).
6 The much-discussed theonym ma-qe (dat. sg.) attested in KN F 51 v. 2 in the context of grain offerings is unlikely to be related to the root of раряхю (the Mycenaean form may have to be read as ma-ka, given the possibility of confusion between signs *77 and *78).
77 A homophonous root mark- 'to besmirch, to smudge' in Tocharian B is probably unrelated, see Malzahn 2010: 755-6; DoT2 487.
auTov s^apys / xpovroi npAioy Ynpa; 'yet still grey old age in
-70
time did seize him [soil. Tithonus]' . Phonologically, there is no problem with the proposed equation either79: PIE *merkw- would give Proto-Tocharian *markw- > A mark- with elimination of labial coarticulation before another consonant: 3 sg. aor. *merkw-s(t) > Toch. A pret. III markas, cf. *h36kw-s 'eye' > Common Tocharian *(k > Toch. A ak, B ek80.
A tertium oomparationis for PIE *merkw- 'to seize, to take away' may be found in Lat. merx -ois (f.) 'commodity, merchandise' and its derivatives81. This etymological comparison requires the following ancillary hypothesis: on the way to Italic, the root *merkw-developed the meaning 'to buy'. There are ample parallels for a semantic development from 'take' to 'buy': for instance, one might compare Lat. emd 'take' > 'buy' or *ad-oaptd 'take' (REW 63) > Italian aooattare, French aoheter 'to buy'. Nomen aotionis *merkw-s 'buying, commerce' would regularly give Proto-Italic nom. sg. *merks with a concretized meaning 'commodity being bought (and sold)'82.
Gk. ^i^o^ denotes both an actor ('mime') and a kind of scenic sketch, founded by the Syracusan Sophron ('mimus'). This word
78 The text is by West 2005: 5.
79
Comparisons with Tocharian always have a degree of uncertainty, since PIE voiced, voiced aspirated and voiceless stops have nearly all fell together in Tocharian. While there is no alternative etymology for Toch. A mark- on the books, M. Weiss points out to me that the underlying PIE root could in theory also be *merg- 'divide' reflected in Hittite mark-abbi 'I divide, I cut up', as well as in the noun for 'border': YAv. marsza-'border, district', Lat. margd 'border', OIr. mruig 'territory' (< *mrogi-) and Proto-Gmc. *markd- 'boundary, region' (Goth. marka, etc.).
80 See Pinault 2008: 456-7.
81 No Indo-European etymology is registered for merx in EDLIL 376. The god's name Mercurius may or may not be a loanword from Etruscan: in the former case, it would have nothing to do with merx (for which, on the contrary, there are no specific indications of Etruscan origin, see Breyer
1993: 92-3 n. 313).
82
From nom.sg. the allomorph merk- was generalized throughout the paradigm (at the expense of *merqu-), just as the stem voc- was generalized from nom.sg. vox < *uokw-s. After nom. *merks had become mers (Plaut.), the velar stop was reintroduced into nom. sg. (so Leumann 1977: 221).
0-3
(for which the dictionaries offer no etymology ) and its derivative ^T^so^ai 'imitate' may be connected with the PIE root *meh2-i- for which a rather specific meaning '(to reveal by) making motion by hand, to indicate (by gesture)' can be reconstructed on the basis of the following cognates: Lith. moju 'wave hand', Proto-Slav. *majati 'to make signs with a hand', Yazgulyam amaw- 'to beckon, motion with a hand' and other Mod. Iranian reflexes of *fra-maia- (ESIYa 5.331)84. Formally and semantically, Gk. ^i^o^ can be unproblema-tically aligned with this root: zero grade *mh2i- > *mih2- (laryngeal metathesis before consonant), *mih2-mo- 'making signs by hands'85 ^ *mih2-mo- 'a person who makes signs by hands' > ^i^o^.
Gk. roxi^ 'bustard, Otis tarda, дрофа' is usually explained as a derivative from on^ 'ear'86, but there is not much about bustard's appearance that would suggest naming it an "owner of conspicuous ears": while bustards have ear tufts, these are not nearly as large as those of most owls whose name in Greek (roxo^) reflects this
0-7
physical feature . Instead, I would like to suggest a direct comparison between roxi^ and Hitt. wattai- 'bird' (KBo 4.2 ii 3 2 88) < *uot-oi- ^ *uot-i- 'one that moves / flies'89 made from PIE *uet-
83 GEW241, DELG 677, EDG 955.
84 *meh2-i- is an extended form of the root *meh2- continued e.g. in Gk.
'show, reveal' (LIV2 425). Ved. maya- 'magical power' (often mentioned in connection with pipo;) may or may not be related to the latter root.
85 Type oxpo; 'fortress' ('holding'), nxappo; 'sneezing', aKuApo; 'rending', etc., see Chantraine 1933: 132, 151; Probert 2006: 238-257.
86 See e.g. Robert 1911: 103, GEW 1153, DELG 1259.
87
This skepticism is shared by specialists in Greek ornithology: according to Thompson 1895: 200 "the etymology (soil. from otic; — A. N.) is doubtful"; Arnott 2007: 240 mentions "bristles across the ears of breeding males which probably (emphasis mine — A. N.) accounted for the bird's name".
88 For the passage see Bawanypeck 2005: 22, 30.
89 Parallels for words for birds derived from motion verbs may be found in MIr. ethait 'bird, fowl' (< *peth2ontih2) from the root *peth2- 'to fly' (LIV2 479) or the words for 'eagle' (Hitt. haras, Lith. erelis, Goth. ara, Gk. opvi;, opveov 'bird') derived from the root *h3er- 'to set (oneself) in motion' (LIV2 299).
(cf. MIr. fethid2 'makes one's way', Gmc. *uep- / *uap- 'to go around')90.
Under this hypothesis, Gk. *èxi- (later remade as èxiô-)91 goes back to *uoti-, a genitival vrddhi derivative made from the acrostatic stem *uot-i- that underlies the Hittite word for 'bird' . For genitival (= Zugehorigkeits-) derivatives of this type one may compare Av. ahura- 'lord' ^ ahuiri- 'divine', Lat. sacro- 'sacred', sacra 'rites' ^ sacri- 'of the rites, for sacrificing', or Gmc. *dala- 'valley' ^ *doli- 'of the valley, valley dweller' ( > Olcel. dœll)93. As A. J. Nussbaum has shown, across Indo-European languages, genitival derivatives with various formants routinely have the meaning 'X-like' vis-à-vis their base noun with the meaning 'X': Ved. vrsan-'bull' : vrsnya- 'bull-like, mighty', Gk. 0sôç 'god' : 0sïoç 'god-like', Lat. deus 'god' : dlvlnus 'god-like'. These genitival formations are also frequently substantivized in the meaning 'a kind of X': Hitt. pittula- 'loop, noose' : pittuliya- 'a noose of sorts' > 'constraint, anxiety'; *h2urgi- 'a turning' ( > Hitt. hurki- 'wheel') : *h2urgi-lo-'a turn of sorts' > 'deviation' ( > Hitt. hurkil 'perversion'); Gk. T8Ï%oç '(city) wall' : xci^iov 'wall (of a building)'94.
The meaning of the genitival vrddhi derivative *u5ti- would then be 'bird-like', 'a sort of bird', 'quasi-bird', which would not be an inappropriate way of referring to bustards, large birds who make their nests on the ground and prefer to run over flying.
References and abbreviations
Acosta-Hughes B. and Cusset, C. 2012: Euphorion. Oeuvre poétique et
autres fragments. Paris: Les Belles Lettres. Arena, R. 1970: Considerationi intorno agli aggettivi OiPpoç e Oepepoç. RIL 104, 307-314.
90 See Schaffner 2004 and for the Hittite word Nikolaev 2014b.
91 The folk-etymological connection with oti^ may have been responsible for the supposed remodelling of obl. *©xi- as ©xi5- (Xen. An. 1.5.2+).
92
An alternative derivational basis is an agent noun *uoto- 'moving/flying (around)'.
93 See Darms 1978: 208.
94 See Nussbaum 2009. According to Nussbaum, a genitival vrddhi derivative made from the athematic word for 'snake' provides an explanation for the ever-problematic Armenian cognate: *h1o/egwhi- 'snake' (> Gk. 091^) ^ genitival *h1igwhi- 'a sort of snake, herpetoid' (> Arm. iz 'viper').
Arnott, W. G. 2007: Birds in the Ancient World from A to Z. London / New York: Routledge.
Balles, I. 1999: Lateinisch sanguis "Blut". Compositiones indogermanicae in memoriam Jochem Schindler, ed. by H. Eichner, H.-Chr. Luschützky und V. Sadovski, 3-18. Praha: enigma corporation.
Bauer, B. 2019: Irish Loanwords in the Southwest British Celtic Languages. Eastern European Perspectives on Celtic Studies, ed. by M. Hornsby and K. Rosiak, 23-33. Cambridge Scholars Publishing.
Bawanypeck, D. 2005: Die Rituale der Auguren. Heidelberg: Winter.
Bechtel, F. 1917: Die historischen Personennamen des Griechischen bis zur Kaiserzeit. Halle: Niemeyer.
Bock, B. 2008: Die einfach thematischen Präsentien in der dritten Konjugation des Lateinischen. Graz: Leykam.
Boddy, K. 2021: The Composition and Tradition of Erimhus. Leiden; Boston: Brill.
Bowie, A. M. 2013: Homer, Odyssey, Books XIII and XIV. Cambridge: Cambridge University Press.
Breyer, G. 1993 : Etruskisches Sprachgut im Lateinischen unter Ausschluss des spezifisch onomastischen Bereiches. Leuven: Peeters.
Chantraine, P. 1933: La formation des noms en grec ancien. Paris: Klincksieck.
Cheung, J. 2007: Etymological Dictionary of the Iranian Verb. Leiden: Brill.
D'Alessio, G. 1996: Callimaco. 2 vols. Milan: Biblioteca universale Rizzoli.
Darms, G. 1978: Schwäher und Schwager, Hahn und Huhn: die Vrddhi-Ableitung im Germanischen. München: Kitzinger.
Delamarre, X. 2003: Dictionnaire de la langue gauloise: une approche linguistique du vieux-celtique continental. 2nd ed. Paris: Errance.
DELG = P. Chantraine. Dictionnaire étymologique de la langue grec. Histoire des mots, achevé par J. Taillardat, O. Masson et J.-L. Perpillou, avec en supplément, les Chroniques d'étymologie grecque (1-10) rassemblées par A. Blanc, Ch. de Lamberterie et J.-L. Perpillou. Paris: Klincksieck, 2009.
Dieu, É. 2011: Le supplétisme dans les formes de gradation en grec ancien et dans les langues indo-européennes. Genève: Droz, 2011.
DoT2 = Adams, D. Q. A Dictionary of Tocharian B: Revised and Greatly Enlarged. Amsterdam; Atlanta, GA: Rodopi, 2013.
EDG = R. S. P. Beekes. Etymological Dictionary of Greek. 2 vols. Leiden; Boston: Brill, 2010.
eDiAna = Digital Philological-Etymological Dictionary of the Minor Ancient Anatolian Corpus Languages (https://www.ediana.gwi.uni-muenchen.de/)
EDLIL = M. de Vaan. Etymological Dictionary of Latin and other Italic Languages. Leiden; Boston: Brill, 2008.
Egetmeyer, M. 2010: Le dialecte grec ancien de Chypre. Berlin; New York: de Gruyter.
ESIYa = V. S. Rastorgueva, D. I. Èdelman. [Etymological Dictionary of the Iranian Languages]. Moscow: Vostocnaja literatura, 2000-. В. С. Расторгуева, Д. И. Эдельман. Этимологический словарь иранских языков. М.: Восточная литература, 2000-.
EWAhd = A. L. Lloyd, O. Springer, and R. Lühr. Etymologisches Wörterbuch des Althochdeutschen. Göttingen: Vandenhoeck & Ruprecht, 1988—.
EWAia = M. Mayrhofer. Etymologisches Wörterbuch des Altindoarischen. 3 vols. Heidelberg: Winter, 1986-2001.
Frisk, H. 1930: Griechisch sösipa, öön. Göteborgs högskolas arsskrift 36(3), 1-5 (reprinted in Kleine Schriften zur Indogermanistik und griechischen Wortkunde, 281-286. Göteborg: Elanders Boktryckeri, 1966).
Fusillo, M. 1981: ГАаикп биаяерфеАос;. Una designazione del mare in Esiodo Th. 440. Studi di letteratura greca ( = Ricerche di filologia classica I), 9-23. Pisa: Giardini.
Garnier, R. 2010: Sur le vocalisme du verbe latin: étude synchronique et diachronique. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck.
GEW = H. Frisk. Griechisches etymologisches Wörterbuch. Heidelberg: Winter, 1960-1972.
GLyk = G. Neumann. Glossar des Lykischen. Wiesbaden: Harrassowitz, 2007.
Gow A. S. F., Schofield, A. F. 1953: Nicander. Cambridge: Cambridge University Press.
Hajnal, I. 2005: Das Frühgriechische zwischen Balkan und Ägäis - Einheit oder Vielfalt? Sprachkontakt und Sprachwandel: Akten der XI. Fachtagung der Indogermanischen Gesellschaft, ed. by G. Meiser and O. Hackstein, 185-214. Wiesbaden: Reichert.
Hilmarsson, J. 1985: The Vocalism of Tocharian AB yok 'hair; colour'. Indogermanische Forschungen 90, 83-87.
Hoekstra, A. 1989: A Commentary on Homer's Odyssey. Vol. 2. Oxford: Oxford University Press.
Höfler, S. 2016/2017[2019]: "La belle Hélène", a Generic Brothel, and the Development of *CRHC Sequences in Ancient Greek. Die Sprache 52, 177-201.
Höfler, S. 2020: Substantivization of Adjectives. Indo-European Linguistics 8, 181-204.
Jacques, J.-M. 2002 : Nicandre: Œuvres. T. 2: Les Thériaques. Paris: Les Belles Lettres.
Janda, M. 2001-2002: Etymologie von altgriechisch aintiç. Sbornik Praci Filozofické Fakulty Brnenské Univerzity ( = Studia minora Facultatis Philosophicae Universitatis Brunensis). Rada klasicka:
Graeco-Latina Brunensia 6-7 ( = Festschrift Antonin Bartonäk, ed. by I. Radova and K. Vaclavkova-Petrovicova), 123-134.
Klingenschmitt, G. 1982: Das altarmenische Verbum. Wiesbaden: Reichert.
Kroonen, G. 2013: Etymological Dictionary of Proto-Germanic. Leiden; Boston: Brill.
Kümmel, M. J. 2013: The Distribution of Roots Ending in IE *ND. The Sound of Indo-European 2, ed. by R. Sukac and O. Sefcik, 159-176. Munich: Lincom.
Lane, G. S. 1938: Problems of Tocharian Phonology. Language 14, 20-38.
Leumann, M. 1950: Homerische Wörter. Basel: F. Reinhardt.
Leumann, M. 1977: Lateinische Grammatik. Vol. 1: Laut- und Formenlehre. München: Beck.
LKZ = Lietuviy, kalbos zodynas, ed. by J. Balcikonis, J. Kruopas et al. 20 vols. Vilnius: Valstybine encyklopedj zodyn^ ir mokslo literatüros leidykla, 1941-2002 (online version: http://www.lkz.lt/).
Lightfoot, J. L. 2010: Hellenistic Collection. Cambridge, MA: Harvard University Press.
LIV2 = Lexikon der indogermanischen Verben. Die Wurzeln und ihre Primärstammbildungen, ed. by H. Rix, M. Kümmel, et al. 2nd ed. Wiesbaden: Reichert, 2001.
Magnusson, A. B. 1989: Islensk ordsifjabök. [Reykjavik]: Oröabök Haskölans.
Malzahn, M. 2010: The Tocharian Verbal System. Leiden; Boston: Brill.
Martirosyan, H. K. 2010: Etymological Dictionary of the Armenian Inherited Lexicon. Leiden / Boston: Brill.
Mayrhofer, M. 1986: Indogermanische Grammatik. Vol. I/2: Lautlehre: segmentale Phonologie des Indogermanischen. Heidelberg: Winter.
Miller, D. G. 1977: Some Theoretical and Typological Implications of an IE Root Structure constraint. JIES 5, 31-40.
Miller, D. G. 2010: Language Change and Linguistic Theory, 2 vols. Oxford University Press.
Mühlenbach, K., Endzelin J. 1923-1936: Lettisch-deutsches Wörterbuch. 6 vols. Riga: Lettisches Bildungsministerium.
Mülenbach, K., Endzellns, J. 1923-1936: Latviesu valodas värdnica. Riga: Izdevusi izglltlbas ministrija.
Nikolaev, A. 2010: Issledovanija po praindoevropejskoj imennoj morphologii. St. Petersburg: Nauka.
Николаев А. C. 2010: Исследования по праинд о европейской именной морфологии. СПб.: Наука.
Nikolaev, A. 2014a: An Etymological Miscellany. Indo-European Linguistics and Classical Philology 18, 708-717.
Nikolaev, A. 2014b: Hittite wattaes 'birds'. Münchener Studien zur Sprachwissenschaft 69, 257-267.
Nikolaev, A. 2018: Greek Oooç 'sharp', Hittite tuhs- 'to cut'. Vina Diem Celebrent: Studies in Linguistics and Philology in Honor of Brent Vine, ed. by D. Gunkel et al., 267-275. Ann Arbor: Beech Stave. Nikolaev, A. 2019: Aaeipa, the Lady of the Waters. Indo-European
Linguistics and Classical Philology 23, 790-795. Nussbaum, A. J. 2009: So How's the Little Woman?: Genitivals, Diminutivity, and PIE *gwen-h2, etc.. Talk given at the GSAS Workshop for Indo-European Linguistics, Harvard University (April 3, 2009).
Pedersen, H. 1909: Vergleichende Grammatik der keltischen Sprachen.
Vol. 1. Göttingen: Vandenhoeck & Ruprecht. Pinault, G.-J. 2003: Sur les thèmes indo-européens en *-u-: dérivation et étymologie. Indogermanisches Nomen: Derivation, Flexion, Ablaut. Akten der Arbeitstagung der Indogermanischen Gesellschaft (Freiburg, 19.-22. September 2001), ed. by E. Tichy, D. Wodtko, B. Irslinger, 153-188. Bremen: Ute Hempen. Pinault, G.-J. 2008: Chrestomathie tokharienne: textes et grammaire. Leuven: Peeters.
Pinault, G.-J. 2019: Hittito-Tocharica: Tracking the Bear Once More. "And I Knew Twelve Languages ": A Tribute to Massimo Poetto on the Occasion of His 70th Birthday, ed. by N. Bolatti Guzzo and P. Taracha, 496-509. Warsaw: Agade Bis; University of Warsaw, Faculty of Oriental Studies. Pisani, V. 1979: feber, februm, februärius e le continuazioni latini di sr.
Archivio Glottologico Italiano 64, 102-106. Probert, P. 2006: Ancient Greek Accentuation: Synchronic Patterns, Frequency Effects, and Prehistory. Oxford; New York: Oxford University Press.
Rengakos, A. 1994: Lykophron als Homererklärer. ZPE 102, 111-130. REW = W. Meyer-Lübke. Romanisches etymologisches Wörterbuch. 3rd ed.
Heidelberg: Winter, 1935. Rhys, J. 1895: Goidelic Words in Brythonic. Archaeologia Cambrensis 12, 264-302.
Rietz, J. E. 1862-1867: Svenskt dialektlexikon: Ordbok öfver svenska
allmogesprâket. Malmö: Cronholm. Ringe, D. 1988-1990: Evidence for the position of Tocharian in the Indo-
European family? Die Sprache 34, 59-123. Robert, F. 1911: Les noms des oiseaux en grec ancien: Étude sémantique.
Neuchâtel: Attinger frères. Ruijgh, C. J. 1957: L'élément achéen dans la langue épique. Assen: van Gorcum.
Schaffner, S. 2004: Mittelirisch fethid "geht, macht seinen Weg", althochdeutsch wadalön, wallon "umhergehen, wandern; umherwogen", altenglisch waduma "Woge, Welle", waöol "Vollmond" und Verwandtes. Die Indogermanistik und ihre Anrainer: Dritte Tagung der Vergleichenden Sprachwissenschaftler der Neuen Länder, ed.
by Th. Poschenrieder, 277-314. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck. Skoda, F. 1982: Le redoublement expressif: un universal linguistique. Paris: SELAF.
Smoczynski, W. 2007: Slownik etymologiczny j§zyka litewskiego. Vilnius:
Uniwersytet Wilenski. Solmsen, F. 1901: Untersuchungen zur griechischen Laut- und Verslehre.
Strassburg: Trübner. Steblin-Kamenskij, I. M. 1999: Etimologiceskij slovar' vahanskogo jazyka.
St. Petersburg: St. Petersburg State University. Stokes, W. 1894: Urkeltischer Sprachschatz. Übersetzt, überarbeitet und herausgegeben von Adalbert Bezzenberger. (= Vergleichendes Wörterbuch der indogermanischen Sprachen, ed. by August Fick, 4th ed., part 2). Göttingen: Vandenhoeck & Ruprecht. Strömberg, R. 1946: Greek Prefix Studies on the Use of Adjective Particles. (Göteborgs högskolas arsskrift 52:3). Göteborg: Elanders boktryckeri.
Thompson, D'A. W. 1895: A Glossary of Greek Birds. Oxford: Clarendon Press.
Thurneysen, R. 1912-1913: Zur Wortschöpfung im Lateinischen. I.
purgare. II läuigare -leuigare. IF 31, 276-281. Trubachev, O. N. 1957: [G. A. Il'inskij's Etymological Dictionary]. Voprosy jazykoznanija 1957/6, 91-96.
Трубачев, О. Н. 1957: Этимологический словарь Г. А. Ильинского. Вопросы языкознания 1957/6, 91-96 Vine, B. 2020: Latin febru(u)m and the Root Vocalism of Indo-European s-Stems. Talk given at the 39th East Coast Indo-European Conference, Virginia Polytechnic Institute and State University (12 June 2020).
Wachter, R. 2001: Non-Attic Greek Vase Inscriptions. Oxford University Press.
West, M. L. 2005: The New Sappho. Zeitschrift für Papyrologie und
Epigraphik 151, 1-9. Witzel, M., Goto, T. 2007: Rigveda: Das heilige Wissen. Erster und zweiter Liederkreis, unter Mitarbeit von E. Doyama und M. Jezic. Frankfurt a.M.: Verlag der Weltreligionen.
