CC BY
25
IE. *gh¥ in GERMANIC.
A SUPPLEMENT TO VERNER’S LAW
Резюме: В статье рассматриваются два разных отражения индоевропейского лабиовелярного *ghw в древнегерманском, которые могут быть обусловлены местом индоевропейского и древнегерманского ударения в слове. Выявленная закономерность соотносится с известным в индоевропеистике законом Вернера.
Ключевые слова: Germanic reflexes of the labiovelar *ghw, Indo-European accent, Proto-Germanic, Verner’s Law.
I. Introduction.
As far as I know, there occur three reflexes of IE. *k in the Germanic languages1:
[1] IE. *k > Gmc. *hw in the initial position and medially after the stress, e.g. Gmc. *hwalaz m. ‘a big fish; whale’ > ON. hvalr, Icel. hvalur, OE. hwjl, E. whale, OHG. (h)wal m. ‘Walfisch’, walira f. ‘Wels’, late MHG. wels (s-stem) ‘id.’ (cf. Sevilla Rodriguez 1989: 177-180; Kluge, Seebold 1999: 872); IE. *ekwos m. (cf. Skt. asvah) ‘horse’ > Gmc. *Shwaz m. ‘id.’ > Goth. aiha-, OSax. ehu-, OE. eoh, ON. jor ‘horse’ (de Vries 1977: 293; Lehmann 1986: 15, 100; Blazek 1999: 23).
[2] IE. *k > Gmc. *gw > *w or *g in the medial position before accent, e.g. Gmc. *tegwaz ‘slave’ > Goth. piwos m. pl. ‘slaves’, ON. pewaR ‘Lehnsmann’, cf. OInd. takva- adj. ‘quick’ (Lubotsky 1988: 95); Gmc. *wulgwiz f. ‘she-wolf’ > ON. ylgr f. ‘id.’ (de Vries 1977: 677), cf. Skt. vrkfh f. ‘she-wolf, female jackal’.
1 Abbreviations used: adj. - adjective, cf. = confer (compare), e.g. - exempli gratia, f. - feminine, id. - idem, m. - masculine, n. - neuter; No. - number, pl. -plural, PN - personal name, Sect. - section.
Languages: Att. - Attic (Greek), Dan. - Danish, Far. - Faroese, G. - German, Gk. - Greek, Gmc. - Germanic, Goth. - Gothic, Gk. - Greek, E. - English, Icel. - Icelandic, IE. - Indo-European, Lat. - Latin, MHG. - Middle High German, MLG. - Middle Low German, Norw. - Norwegian, OE. - Old English, OFris. -Old Frisian, OHG. - Old High German, OIcel. - Old Icelandic, OInd. - Old Indic, OIr. - Old Irish, ON. - Old Nordic, OSax. - Old Saxon, OSwed. - Old Swedish, PSl. - Proto-Slavic, Russ. - Russian, Skt. - Sanskrit, Swed. -Swedish, Toch. A - Tocharian A or Kuchean.
[3] IE. *k > Gmc. *f in some irregular cases, e.g. Gmc. *fmfi ‘five’ beside Skt. panca ‘5’, Gk. Att. pevte ‘id.’ (< IE. *pSnk:e), *fimftaz ‘fifth’ beside Gk. pemptoj ‘id.’ (< IE. *penkutos), *wUlfaz m. ‘wolf’ beside Skt. vrkah ‘id.’ (< IE. *ulkuos).
о • v о /
It is possible to demonstrate that these three reflexes can be attested in one and the same word, cf. IE. *ukvnos m. ‘oven’ > [1] *Uhwnaz (Goth. auhns), [2] *ugwnaz (OSwed. oghn) and [3] *Ufnaz m. ‘id.’ (On. ofn, OE. ofen, OHG. ofan) (Зализняк 1964: 226, No. 39.7).
Ernst Seebold (1967; 1980) finds, by analogy, three continuations of IE. *gh in Common Germanic: *g- (e.g. *gUnpio f. ‘fight’), *w- (e.g. *warmaz adj. ‘warm’) and *b- (e.g. *bedjan ‘bitten’, as if from IE. *ghvedh- ‘bitten, begehren’)2. If *b-, as well as *f-, is the result of an irregular development, then the regular distribution of the reflexes *g-/*w- (both from IE. *gh) should be explained by analogy to the continuation of IE. *k. The alternation between West Germanic *baumaz m. ‘tree, trunk’ (OE. beam, OSax. bom, OHG. boum) and North Germanic *bagmaz m. ‘id.’ (Goth. bagms, OIcel. badmr, OSved. bagn; Зализняк 1964: 226, No. 39.1), analogical to *Uhwnaz / *ugwnaz / *ufnaz, clearly demonstrates that this distribution must be connected most likely with the position of the Indo-European and Proto-Germanic stress.
In the following survey of the material I shall attempt to include those examples, for which a reconstruction with *gh seems reasonably well founded and which possess some accentually valuable equivalents in other Indo-European languages, especially in Sanskrit, Greek and Balto-Slavic.
2 Speirs (1978: 1) notes that “the reflexes of PIE *gh in Germanic are still a matter of controversy”. Ringe (2006: 105-116) adopts some Seebold’s solutions, stressing that he has “tried to use only relatively certain examples as evidence”. According to him, PIE. *ghv became Gmc. *b in the initial position, expect “when it had already been delabialized by a following *u”. However, examples for this development are rare and highly uncertain, as stressed by Polome (1987: 306-307) and Ringe (2006: 106). In my opinion, Gmc. *b- appears to represent an irregular reflex of IE. *ghu, as well as Gmc. *f (or *b) seems an unexpected continuation of IE. *F, see also MaKaeB (1962: 26). Most recently Johnsen (2007) has followed Seebold’s way of thinking, which seems unfruitable.
1. Reflexes of IE. *gh in the initial position.
In my opinion, there are only two different reflexes of IE. *gh in the initial position, namely Gmc. *g and *w. These two reflexes occur in some related pairs of the Germanic words, e.g. *garmaz m. ‘fire dog’ vs. *warmaz adj. ‘warm, hot’ (both from IE. *ghvormos), OIcel. gandr adj. ‘monster, magic’, also m. ‘wand’ vs. Dan. dial. vann, E. wand ‘magic staff’, Goth. wandus ‘rod, twig’, OIcel. vgndr m. ‘id.’ (from Gmc. *gandaz adj. ‘monster, magic’ next to Gmc. *wandaz / *wanduz m. ‘wand, magic staff’). Taking into account these Germanic facts Markey (1980: 287) suggested “an apparently arbitrary alternation of g-/w- as reflexes initially before a”. In my opinion, the observed alternation is not “arbitrary”, but it is caused by a regular continuation of the Proto-Germanic variation of the original Indo-European accent. Examples discussed below seem relevant both from the etymological point of view and the accentological one.
2. Evidence for IE. *gh > Gmc. *g.
2.1. Gmc. *garmaz m. ‘fire dog’ (cf. OIcel. garmr m. ‘hound of Hell, fire dog’, only in skaldic usage; Far. garmur ‘dog’, etc.) < IE. *ghvormos m. ‘the fiery one’ (whence ‘dog of hell’). This term seems to reflect an oppositional accentuation to the Germanic adjective *warmaz < IE. *ghuormos (see No. 3.1.). The same accento-logical opposition between a proper name and a corresponding appelative may be found in most Indo-European languages, e.g. Gk. PN FalSpo" vs. adjective faiSpO" ‘clear, brilliant’, Gk. PN Fpl£,og vs. adjective Фрг£,од ‘standing on end, bristling’.
2.2. Gmc. *gadja n. ‘Sinn’ (cf. OIcel. ged n. ‘mind, mood, spirits’, also ‘wits, disposition’, Norw. gjed, Dan. dial. gied; de Vries 1977: 159; Polome 1987: 305, 308) < IE. *ghvodhiom < IE. *ghvedh-‘bitten, begehren’ (Pokorny 1959: 488), cf. Greek под-од m. ‘longing after, yearning, fond desire, regret’. See also OIr. guide f. ‘prayer’ (< Celtic *godya) and Gk. epi-poGla f. ‘a longing after’ (< IE. *ghvodhiya with a retraction of the stress). An oppositional oxytone accent is seen in Gk. под^ f. ‘fond desire, longing, want’.
2.3. Gmc. *gelda n. ‘payment’ (cf. Goth. gild n. ‘tax’, OIcel. gjald n. ‘payment, recompense, punishment’; OE. gyld, gield ‘payment, service, substitute, sacrifice’; OFris. jeld ‘Kaufpreis’; OSax. geld n. ‘Vergeltung, Lohn’; OHG. gelt m. or n. ‘payment, reward, sacrifice’, G. Geld n.; Зализняк 1963: 140, No. 4.24; Lehmann 1986: 155; Kluge-Seebold 1999: 309) < IE. *ghveldh-o-m n.
‘payment’ (see Pokorny 1959: 436), cf. Gk. zeXftoq n. (es-stem) ‘debt, payment due’ (< *dsXdoq and IE. *ghveldh-os- n.).
2.4. Gmc. *gunpiz f. ‘fight, combat’ (cf. OIcel. gunnr, gudr f. ‘fight, killing’, OE. gud ‘combat’, OHG. gund-fano ‘battle-flag’; de Vries 1977: 195; Polome 1987: 308; Ringe 2006: 92) < IE. *ghuntis f. (i-stem) ‘fight, killing’, cf. Lith. gintis f. ‘defence, fighting’ (2 AP appears in dialects, 4 AP in the literary Lithuanian). A Sanskrit equivalent is attested with no stress: hatih f. (i-stem) ‘striking, a stroke or blow; killing, destroying, destruction’ (Monier-Williams 1999: 1287), though Pokorny (1959 492) under the heading *ghven-‘schlagen’ gives a form with the oxytone stress hati'h f. ‘das Schlagen, Schlag, Multiplikation’. The oxytone stress is firmly reconstructed after the Germanic data (the preservation of *p after stress according to Verner’s law).
2.5. Gmc. *gunpjo f. ‘fight’ (cf. OSax. gudea f. ‘Kampf, Schlacht’; de Vries 1977: 195) < IE. *ghvntia f. (a-stem), cf. OLit. gincia f. ‘defence’, Lith. dial. gincia f. ‘Streit’ (Pokorny 1959: 498); Skt. hatya- f. ‘killing, slaying, slaughter’ (Monier-Williams 1999: 1287)3. The original oxytonesis can be reconstructed on the basis of the Germanic lexical material.
3. Possible evidence for IE. *gh > Gmc. *w.
3.1. Gmc. *warmaz adj. ‘warm’ (cf. OIcel. varmr adj. ‘warm’, Norw., Swed. and Dan. varm, OE. wearm, E. warm; OSax. warm, OHG. and G. warm ‘id.’; de Vries 1977: 646) < IE. *ghuormos adj. ‘hot, warm’, cf. Skt. gharmah m. ‘glot, heat’ (Lubotsky 1988: 89), Pashto yarma f. ‘noon, heat of the sun’, also ‘sun’ in some dialects (^w6o 1974: 81, 83; Morgenstierne 2003: 32); Greek dep^oq adj. ‘hot, warm’, dep^ov n. ‘heat’, dep^a n. pl. o-stem ‘hot baths’ (< IE. *ghvermos adj. ‘hot, warm’ < IE. *ghver- ‘to warm’). Other related terms (e.g. Lat. formus adj. ‘warm, hot’, Toch. B sarme ‘id.’) are accentologically irrelevant.
3.2. Gmc. *wada n. ‘ford’ (cf. OIcel. vad n. ‘ford, a shallow place in water’, Norw., Swed., Dan. wad; OE. wwd ‘water, sea’; OHG. wat ‘ford’; de Vries 1977: 637) < IE. *ghu‘dhom n. ‘ford’, cf.
3 Pokorny (1959: 492) gives the accented form hatya (spat!) ‘Totung’, which is not attested in the literary sources. See also Lubotsky (1988). However, Seebold (1967: 105) refers to the Vedic compound musti-hatya f. ‘Faustkampf’.
Skt. gadham n. ‘ford, shoal’. See also Lat. vadum n. ‘shallow, shoal, ford’.
3.3. Gmc. *wahsaz adj. ‘pointed’ (cf. OHG. wahs adj. ‘scharf, spitzig’) < IE. *ghroksos or *ghuoksos adj. ‘id.’, cf. Greek фо£,од adj. ‘pointed, peaked in head’ (so Seebold 1967: 111, 1980: 473-474 after Fick 1891: 348). All the Greek (and Indo-European) adjectives ending with -so- show the oxytone stress, e.g. Gk. ка^уод adj. ‘crooked, bent’, фрг^од adj. ‘standing on end, bristling’ (Chantraine 1968-1977: 1221-1222).
3.4. Gmc. *wiba n. ‘woman, wife’ (cf. OIcel. vif n. ‘woman, weif, lady’; OE. wf, E. wife; OSax. wf; OHG. wib, G. Weib n.; Зализняк 1963: 145, No. 4.78; de Vries 1977: 661; Kluge-Seebold 1999: 879) < IE. *ghuipom n. ‘shame’, cf. Toch. A kip, B kwipe ‘shame, pudenda’ (Hilmarsson 1996: 208-211). The original oxytone accent is confirmed by Verner’s law (IE. *p > Gmc. *b before stress).
3.5. Gmc. *wambo f. ‘belly’ (cf. Goth. wamba f. ‘womb, belly’; OIcel. vgmb; OE. wamb, E. womb; OFris., MLG. wamme; OHG. wambo; G. dial. Wamme, Wampe; Зализняк 1963: 148, No. 7.19; Seebold 1980: 470-471; Lehmann 1986: 393; Kluge-Seebold 1999: 874) < IE. *ghrombha f. ‘belly, womb, vulva’, cf. Skt. gabhah m. ‘vulva’ (Monier-Williams 1999: 346), as if from IE. *ghvmbhos m. ‘id.’.
Further examples for the development of IE. *ghv- to Gmc. *w-are disputable etymologically and uncertain accentologically. Gmc. *w6piaz adj. ‘pleasant, sweet’ (Goth. dauns wopi ‘sweet smell’, OE. wide adj. ‘sweet, pleasant’, OHG. wuodi adj. ‘sweet, mild’, OIcel. Sdri ‘better’; Seebold 1967: 110, Polome 1987: 309) appears to be related to the Greek gloss frotiov • pposfilej, (‘pleasant, sweat’), which is attested in the Hesychian glossary (VI cent. AD). The gloss in question is given without ethnic reference, thus it is uncertain whether initial f- represents a dialectal rendering of the Greek digamma (like in Pamphylian) or a regular Greek reflex of IE. *bh or perhaps IE. *ghr. Theoretically, the suggested Greek-Germanic isogloss, which seems unpeccable from the semantic and accentual point of view, could derive from IE. *ghv6tiios or *ghu6tiios, but I would prefer a reading of the letter f- as a Pamphylian rendering of Common Greek *#- (cf. Gk. Pamph. flkati ‘twenty’ < Gk. *#1каи < IE. _ *uikmti ‘20’). In this case an alternative derivation from IE. *u6tios adj. ‘pleasant, sweat’ is highly probable, whereas a root connection with OInd. (RV) vamah adj.
‘lovely, dear, pleasant, agreeable, fair, beautiful, splendid, noble’ (as if from IE. *ud-mo-s) cannot be excluded (cf. de Vries 1977: 684).
4. Possible evidence for -ghu- > Gmc. -g-.
4.1. Gmc. *dagaz m. ‘day’ (cf. Goth. dags, OIcel. dagr, E. day, OSax. dag, OHG. tag, G. Tag ‘day’; Зализняк 1963: 130, No. 1.12; de Vries 1977: 71-72; Lehmann 1986: 86) < IE. *dhoghvos m. ‘hot weather’, cf. Lith. dagas m. ‘heat of day, harvest’ 4 AP (Illich-Svitych 1979: 30, Sect. 10), Skt. ni-daghah m. ‘heat, warmth; the hot season (May and June), summer; internal heat’ (Monier-Williams 1999: 548). The Indo-European appellative is a derivative from the root *dheghv- ‘to burn, to warm’ (Pokorny 1959: 240; Watkins 1970: 1512).
4.2. Gmc. *sangaz m. (o-stem) ‘song’ (cf. OIcel. sQngr m. ‘song, music, melody, priest’s poem’, OE. sang, E. song, OSax. sang, OHG. sang, cf. Gothic saggws m. i-stem ‘Gesang’; Зализняк 1964: 188, No. 26.26; de Vries 1977: 578) < IE. *songhuos m. ‘voice, sound, song’, cf. Greek 6ф» f. ‘voice, sound’ (< IE. *songhva f. ‘id.’).
4.3. Gmc. *angaz m. (o-stem) ‘smell’ (cf. OIcel. angr m. ‘Duft, Geruch’; de Vries 1977: 10) < IE. *onghuos, cf. Gk. dial. 6ф» f. ‘smell’ (< IE. *onghua f. ‘id.’). The same root *onghu- may be found in Common Slavic *Qchb m. ‘smell’ (cf. Pol. wqch ‘sense of smell’) (< *onghv-so-) and *Qchati vb. ‘to smell’ (cf. Pol. wqchac ‘to smell’) (an extension on -s- from the Indo-European root *onghr-).
4.4. Gmc. *wagjaz m. ‘wedge’ (OIcel. veggr m. ‘Keil’; OE. wecg, E. wedge; OSax. veggi, OHG weggi, wecki m. ‘wedge’; Зализняк 1964: 214, No. 31.14; Seebold 1967: 128-129; de Vries 1977: 650) < IE. *uoghuios, cf. Lith. vagis m. ‘wedge’ (4 AP in the dialects, but a secondary 2 AP occurs in the standard language), Latvian vadzis m. ‘wedge’ (Fraenkel 1962-1965: 1179). The nominal root *ueghv- ~ *uoghu- is attested in Greek (Hesychian) 6fvlj f. ‘ploughshare; plough’ (< IE. *uogh-n(-s), Lat. vomer n. (es-stem) ‘ploughshare’ and perhaps in Old Irish fecc ‘tooth’ (as if from Celtic *vegna f. and IE. *ueghv-na-).
4.5. Gmc. *tungon f. (n-stem) ‘tongue’ (Goth. tuggo, OIcel. tunga, OE. tunge, E. tongue, OSax. tunga, OHG. zunga, G. Zunge f. ‘tongue’; Зализняк 1963: 156, No. 14.15) < PGmc. *tungwo f. (o stem, remodelled later as an on-stem) < IE. *dnghua f. ‘tongue’ (Ringe 2006: 81, 90-91). The oxytonesis is confirmed by OInd. jihva f. ‘tongue’ (Monier-Williams 1999: 422; Lubotsky 1988: 103). Other
related terms (e.g. OLat. dingua, Lat. lingua f. ‘tongue’, Avest. hizu-and hizuva- f. ‘id.’) are accentologically irrelevant.
4.6. Gmc. *lungraz adj. ‘swift’ (cf. OSax. lungar ‘powerful’, OE. lungre adv. ‘quickly, soon’; Ringe 2006: 91, 92) < PIE. *hjlnghuros adj. ‘light (in weight)’, cf. Gk. elafpoj adj. ‘light, nimble’. All the adjectives which end with the suffix *-ro- demonstrate the oxytone stress.
5. Possible evidence for -ghu- > Gmc. -w-.
5.1. Gmc. *snaiwaz m. ‘snow’ (cf. Gothic snaiws, ON. snaer, OE. snaw, E. snow; OSax. sneo, OHG. sne(o), G. Schnee; Зализняк 1964: 189, No. 29.31) < IE. *snoighvos m. ‘snow’ (Ringe 2006: 107), cf. PSl. *snigb (gen. *sniga) AP a ‘snow’ (cf. Pol. snieg; SC. снёг, cHisa; Slovenian sneg; Russ. снег, cnSaa; Derksen 2008: 457);
N/"
Latv. sniegs, Lith. sniigas ‘snow’ 2 AP in a north-western Zemaitis dialect, but 4 AP (of secondary origin) is more usual in the dialects and in the standard language (Illich-Svitych 1979: 102, Sect. 44).
5.2. Gmc. *sarwaz m. ‘roach, Rutilus rutilus L.’ (cf. OS wed. sarv, Swed. sarv, sarf) < IE. *sorghros m. ‘a kind of fish’, cf. Greek (Herodian) opfoj m. ‘a sea-fish’. See also Russ. сорога f. ‘roach’ (< PSl. *sorga f. and IE. *sorghva f. ‘a kind of fish’).
5.3. Gmc. *rawiz- n. (es-stem) ‘reed’ and *rawizaz m. (secondary o-stem) ‘id.’ (cf. Goth. raus n., ON. reyr n., OHG. ror n., MLG ror; also ON. reyrr m. a-stem ‘reed’, OHG. rorra f. jo-stem ‘id.’; de Vries 1977: 443, Lehmann 1986: 282) < PIE. *hsrdghves- n. (es-stem) ‘reed’, cf. Gk. opofoj n. (es-stem) ‘reed used for thatching houses’ (< *0p0f0j). A related form is seen in Slavic *rogozjb ‘reed’ (< IE. *H3roghr-).
5.4. Gmc. *newran- n. or m. (n-stem) ‘kidney’ (OIcel. nyra n., OSwed. niure, ME. nere, MLG. nere, OHG. nioro m.; Зализняк 1964: 217, No. 32.21; Seebold 1967: 117; de Vries 1977: 413) < IE. *rnghvron- (Pokorny 1959: 318), cf. Praeneste Latin nefrones m. pl., Lanuvian nebrundines ‘kidneys’ (Muller 1926: 285-286). The n-stem neuters are always barytone both in Sanskrit and Greek (Lubotsky 1988: 108, 147; Risch 1974: 59-62). Thus by an accento-logical analogy the Germanic appelative neuran- (representing a neuter, cf. OIcel. nyra n. ‘kidney’) had to possess originally the barytone stress (IE. *mghvron-). The o-stem appelative, attested in Gk. Увфрод m. ‘kidney’, usually in pl. ‘kidneys’ (< IE. *neghurds), presents an oppositional accentuation.
At present I can find only these 20 items. For other, but accentually or etymologically irrelevant, examples, see Seebold (1967: 105-130; 1980: 450-484) and Polome (1987: 303-310).
On the basis of the above-mentioned lexical material it should be concluded that the initial IE. *ghr- yields Gmc. *g- in the position before accent, *w- elsewhere. In the medial position the continuants of IE. *ghr show the same distribution (i.e. -g- before accent, -w-elsewhere).
The development of IE. *gh appear to be analogous to the development of the Indo-European voiceless consonants in Germanic (e.g. IE. *k yields Gmc. *h initially, but both *h and *g in the intervocal position). The last process is explained by the so called Verner’s law (see Collinge 1985: 203-216). It seems highly probable that the double reflexes of IE. *ghr were caused by a phonological rule similar to Verner’s law (G. das Vernersche Gesetz).
6. Conclusions.
The above analysis of the Germanic lexical material shows clearly that the double representation of IE. *ghr in Proto-Germanic is determined by position of the Indo-European and Proto-Germanic stress. Generally, if one compares the Germanic forms and their Indo-European prototypes from the view-point of Indo-European accentuation, it is easy to prove that IE. *ghr is reflected as Gmc. *g before stress (see No. 2.1-5; 4.1-6) and Gmc. *w elsewhere (see No. 3.1-5; 5.1-4).
References
Дыбо 1974 - Дыбо В. А. Афганское ударение и его значение для индоевропейской и балто-славянской акцентологии, часть I. Именная акцентуация // Балто-славянское исследования. М.: «Наука». С. 67-105.
Макаев 1962 - Макаев Е. А. Сравнительная грамматика германских языков, т. II: Фонология. М.: Издательство Академии Наук СССР. Зализняк 1963-1964 - Зализняк A. A. Материалы для изучения морфологической структуры древнегерманских существительных. Ч. I // Этимология 1963. С. 124-160; idem. Ч. II // Этимология 1964. С. 160-235.
Blazek 1999 - Blazek V. Is Indo-European *Hjekwo- ‘Horse’ Really of Indo-European Origin? // Studia Indogermanica Lodziensia 2, 1999. P. 21-32.
Chantraine 1968-1977 - Chantraine P. Dictionaire etymologique de la langue grecque. Histoire des mots. T. I-IV, Paris: Ed. Klincksieck. Collinge 1985 - Collinge N. E. The Laws of Indo-European. Amsterdam; Philadelphia: John Benjamins Publishing Company.
Derksen 2008 - Derksen R. Etymological Dictionary of the Slavic Inherited Lexicon. Leiden; Boston: E. J. Brill.
de Vries 1977 - de Vries J. Altnordisches etymologisches Worterbuch. 2. verbesserte Auflage. Leiden; E. J. Brill.
Fick 1891 - Fick A. Vergleichendes Worterbuch der indogermanischen Sprachen. 4. Auflage. Gottingen: Vandenhoeck und Ruprecht’s Verlag.
Hilmarsson 1996 - Hilmarsson J. Materials for a Tocharian Historical and Etymological Dictionary. Reykjavik: Malvisindastofnun Haskola Islands.
Illich-Svitych 1979 - Illich-Svitych V. M. Nominal Accentuation in Baltic and Slavic. Cambridge (Massachusetts); London (England): The MIT Press.
Johnsen 2007 - Johnsen S. The development of voiced labiovelars in Germanic // Harvard Working Papers in Linguistics 12, 2007. P. 15-26.
Kluge, Seebold 1999 - Kluge F., Seebold E.. Etymologisches Worterbuch der deutschen Sprache, 23. Erweiterte Auflage / bearbeitet von E. Seebold. Berlin; New York: Walter de Gruyter.
Lehmann 1986 - Lehmann W. A Gothic Etymological Dictionary. Leiden: E. J. Brill.
Lubotsky 1988 - Lubotsky A. M. The System of Nominal Accentuation in Sanskrit and Proto-Indo-European. Leiden; New York; K0benhavn; Koln: E. J. Brill.
Markey 1980 - Markey Th. L. Delabialization in Germanic // Folia Linguistica Historica 1(2). P. 285-293.
Monier-Williams 1999 - Monier-Williams M. A Sanskrit-English Dictionary. Delhi: Motilal Banarsidass Publishers.
Morgenstierne 2003 - Morgenstierne G. A New Etymological Vocabulary of Pashto. Wiesbaden: Reichert Verlag.
Muller 1926 - Muller F. Altitalisches Worterbuch. Gottingen: Vandenhoeck & Ruprecht.
Pokorny 1959 - Pokorny J. Indogermanisches etymologisches Worterbuch. Bern; Munich: Francke Verlag.
Polome 1987 - Polome E. C. Initial PIE *gwh- in Germanic // E. Cardona, N. H. Zide (eds.). Festschrift for Henry Hoenigswald. On the Occasion of his Seventieth Birthday. Tubingen: Gunter Narr Verlag. P. 303-313.
Ringe 2006 - Ringe D. From Proto-Indo-European to Proto-Germanic, Oxford; New York: Oxford University Press.
Risch 1974 - Risch E. Wortbildung der homerischen Sprache. Zweite Auflage. Berlin.
Seebold 1967 - Seebold E.. Die Vertretung von idg. *guh im Germanischen // Zeitschrift fur vergleichende Sprachforschung (KZ) 81. P. 104-133.
Seebold 1980 - Seebold E.. Etymologie und Lautgesetz // M. Mayrhofer, M. Peters, O. E. Pfeiffer (Hrsg.). Lautgeschichte und Etymologie. Akten der VII. Fachtagung der Indogermanischen Gesellschaft. Wien 24.-29. September 1978. Wien: Reichelt Verlag. P. 450-484.
Sevilla Rodriguez 1989 - Sevilla Rodriguez M.. Indo-European *(s)kwalo/is ‘sheat-fish’ // Journal of Indo-European Studies 17, No. 1-2. P. 177180.
Speirs 1978 - Speirs A. G. E. The Proto-Indo-European Labiovelars, Amsterdam.
Watkins 1970 - Watkins C.. Indo-European Roots // Morris W. (ed.). The American Dictionary of the English Language. Boston; New York; Atlanta; Geneva (Illinois); Dallas; Palo Alto: American Heritage Publishing Co., Inc. And Houghton Mifflin Company. P. 1505-1550.
K. T. Witczak. IE *ghw in Germanic. A Supplement to Verner’s Law
Two regular reflexes of IE. *ghv in Proto-Germanic (*w and *g) can be explained by the place of the Indo-European accent, e.g. Gmc. *garmaz m. ‘fire dog’ (< IE. *ghuormos) vs. Gmc. *warmaz adj. ‘warm, hot’ (< IE. *gh¥ormos); Gmc. *dagaz m. ‘day’ (< IE. *dhoghvos) vs. Gmc. *snaiwaz m. ‘snow’ (< *snoighvos m. ‘snow’). The double distribution of IE. *ghv in Proto-Germanic, which is motivated by the primitive accent, can be described as a rule similar to das Vernersche Gesetz.
