Remerciez-le!

Remerciez @Admin pour avoir partagé cet document gratuitement, de la manière la plus simple, en partageant sur les réseaux sociaux.

THE SOLUTION TO THE PROTO-INDO-EUROPEAN LARYNGEAL PROBLEM: THE GLOTTAL FRICATIV

THE SOLUTION TO THE PROTO-INDO-EUROPEAN LARYNGEAL PROBLEM: THE GLOTTAL FRICATIVE THEORY OF HITTITE ḫ ≡ PIE *h/ɦ ABSTRACT. This paper honours BEDŘICH HROZNÝ’s discovery of the Indo-European character of Hittite by dedicating the solution to the Proto-Indo-European laryngeal problem to him: (i) Proto-Indo-European had a single laryngeal *ḫ (ZGUSTA 1951), a glottal fricative (SZÉMERENYI 1970) with voiceless and voiced variants PIE *h/ɦ (PYYSALO 2013 = SPIE). (ii) The proto-language also had a vowel PIE *ɑ (*ǝ = *A), not identical with Hitt. ḫ (versus KURYŁOWICZ 1927) but a low/open back unrounded vowel. (iii) PIE *h/ɦ and *ɑ always appeared together in diphonemic pairs PIE *hɑ/ɑh/ɦɑ/ɑɦ (strict phonotactic selection). (iv) These hypotheses suffice to solve the problems of the entire segmental Indo- European sound law system, completely revised in SPIE. The outcome is internally and externally consistent, and surpasses the earlier models in accuracy. (v) PIE *h/ɦ and *ɑ constitute the missing link of the proto-sound system, and imply the primary Proto-Indo-European phoneme inventory: PIE *o/ō *e/ē *ɑ/ɑ̄? *h/ɦ *i/i̯ *k/g *l/l̥ *m/m̥ *n/n̥ *p/b *r/r̥ *s/z *t/d *u/u̯ These revisions are now digitally tested and proven with the foma programming language at http://pielexicon.hum.helsinki.fi: The success rate in generating more than a hundred most ancient Indo-European languages exceeds 99%. 0. THE SOLUTION TO THE PIE LARYNGEAL PROBLEM AND ITS CONSEQUENCES 0.1. A hundred years after HROZNÝ’s (1915, 1917) discovery of the Indo-European origin of Hittite this article presents the solution to the Proto-Indo-European laryngeal problem implied by the comparative method, and outlines the primary Proto-Indo-European phoneme inventory and the revised Indo-European sound laws, which consistently generate the Indo-European data in a valid (sound and complete) system.1 0.2. Indo-European linguistics has produced four schools of reconstruction, the Paleogrammarians, the Neogrammarians, the laryngeal theory, and monolaryngealism. In SPIE it is shown that all theories contain correct and incorrect 1 Due to the space limit only the most vital features are presented here, and the reader should consult SPIE for the full discussion and details. solutions absent from others. SPIE pragmatically compiles a revised glottal fricative theory (GFT) on the chassis of all the correct (or correctable) solutions. §1. The nine Neogrammarian correspondence sets for vowels (including ‘schwa-loss’), the single *H (monolaryngealism), and the colouring rules for *h2 are combined into the comparative solution to the problem of *H and the vowel system. §2. The Neogrammarian Sanskrito-centric reconstruction of the PIE sonorant system is replaced with VERNER’s comparative one as detailed in §2. §3. With regard to the four series T/Th/D/Dh the tenues are taken as primary. The tenues aspiratae are segmented à la SAUSSURE and KURYŁOWICZ, adding *ɑ. The mediae are dealt with using MAGNUSSON and MEILLET’s root constraint and *ɦ (PYYSALO) for roots D with a single stop. The mediae aspiratae are analyzed à la CUNY and SZEMERÉNYI as D+*ɦ, adding *ɑ (PYYSALO). In the centum-satem isogloss the labiovelars and palatovelars are explained as K+*u/u̯ (REICHELT) and K+*i/i̯ (SZEMERÉNYI) as detailed in §3. 0.3. By 2015 the revisions of SPIE, have been tested with the foma programming language designed for the purpose:2 The rules are consistent and generate Indo- European data with an accuracy rate of more than 99%. This is hardly surprising since GFT synthesizes two centuries of top research, and I wish to congratulate all, starting from Sir WILLIAM JONES, who have contributed to this groundbreaking success of Indo-European linguistics. 1. THE GLOTTAL FRICATIVE PIE *h/ɦ AND THE PIE VOWEL SYSTEM 1.1. PIE *h/ɦ: THE VOICELESS/VOICED GLOTTAL FRICATIVE (SZ *h, LT *h2) 1.1.1. PIE *h/ɦ, the voiceless/voiced glottal fricative, is defined as (SPIE 75-97, 459-464, 467): Hitt. ḫ, Gr. Ø, RV. Ø/. PIE *h/*ɦ (SZ *h/LT *h2). The segmental *h/ɦ (= ḫ) is preserved in Old Anatolian as Hitt. ḫ (Pal. ḫ/CLu. ḫ/HLu. ḫ) but lost in other languages except for the hiatus in the oldest Indo-Iranian meter. (i) For *h, the voiceless fricative, see Hitt. ḫantei ‘vorne’ (HEG H:149, ḫa-an-ti-i): Lat. ante ‘vor(her)’ (WH 1:53), from *hɑent- (IEW 48). (ii) For *ɦ, the voiced fricative, see Hitt. ḫarga·nau- ‘Sohle?, Ferse?’ (HEG H:176, ḫar-ga-na-ú): Gr. ἀργό- ‘schnell beweglich’ and RV. 3j·īṣá- ‘gerade darauf los eilend’ (WbRV. 279), from √ɦɑrgi- (IEW 854f.). 2 For foma by Måns Huldén, see https://code.google.com/p/foma/. 1.1.2. The alternation *h:ɦ is reflected in the corresponding voice of the surrounding stop(s). Thus, the voiceless variant of *ɦɑergi- (ἀργό-: ḫarga·nau-) is *hɑerki- in Hes. ἀρκήσ- ‘ταχύς’ (LSJ. 242), ποδ·άρκησ- ‘having swift feet’ (Iliad 1:121): Lith. aršù- ‘celer’ (LiEtWb. 15-16). The condition(s) of the alternation remain unknown, but the places of articulation of *h/ɦ, voice and tone coincide in the larynx, making this the fundamental and most complex problem of Indo-European linguistics (SPIE 459-461). 1.2. PIE *ɑ/ɑ́: THE LOW/OPEN BACK UNROUNDED VOWEL (Neogr. *–/ǝ)3 1.2.1. PIE *ɑ/ɑ́, the low/open back unrounded vowel, the former schwa indogermanicum (*A), is split in two (SPIE 97-104): Hitt. Ø, Gr. Ø, RV. Ø PIE *ɑ (–)4 Hitt. a, Gr. α, RV. i PIE *ɑ́ (*ǝ) 1.2.2. *ɑ, if not originally accented (*ǝ), was lost in all languages (SPIE 102-103). 1.2.3. *ɑ́, if originally accented (*F), became /a/ in all languages (Gr. α) except for IIr. i. Recognizably the type OInd. kiṇa- ‘Schwiele’ (EWA 3:90): Lat. callo- ‘Schwiele’ (WH 1:139) from *kɑ́hlno- (SPIE 462, IEW 423-524) with *ɑ́ did not undergo the second palatalization. 1.3. PIE *hɑ/ɑh/ɦɑ/ɑɦ, THE DIPHONEMIC HYPOTHESIS OF PIE *H 1.3.1. PIE *h/ɦ and PIE *ɑ always appeared together in the diphonemic pairs *hɑ/ɑh/ɦɑ/ɑɦ.5 1.3.2. Against MØLLER (1879, 1880) and KURYŁOWICZ (1927) *ɑ (*A) is not identified with a laryngeal/Hitt. ḫ, but, as comparatively proven by the Neogrammarians, reconstructed as a vowel. Simultaneously, *h/ɦ is not considered unrelated to ‘a-colouring’ as in ZGUSTA 1951, but both *ɑ and *ḫ are reconstructed e.g. in PIE *pɑhter- ‘father’ instead of Neogr. *pǝter- or LT *ph2ter-. 1.3.3. The proof of the diphonemic hypothesis has already begun in PIE Lexicon where a value *hɑ/ɑh/ɦɑ/ɑɦ consistent with the data is reconstructed for every root implying *ḫ or *ɑ. With all roots reconstructed also the inductive hypothesis will be proven. 1.4. PIE *eɑ/ɑe: THE COLOURING RULES OF SAUSSURE AND MØLLER (Neogr. *a) 1.4.1. PIE *ɑ and adjacent *e assimilated according to SAUSSURE’s (1878) and MØLLER’S (1879, 1880) rules, splitting Neogr. *a in two (SPIE 104-110): 3 Due to technical (font) problems *a/á was used in SPIE instead of proper *ɑ/ɑ́. 4 The Neogrammarian reconstruction is provided in this column. 5 Zgusta’s *H is used as a cover symbol for PIE *hɑ/ɑh/ɦɑ/ɑɦ. Hitt. a(ḫ), Gr. α, RV. a PIE *eɑ(ḫ) (*a1) Hitt. (ḫ)a, Gr. α, RV. a PIE *(ḫ)ɑe (*a2) (i) For *eɑḫ, see Hitt. naḫ- ‘fürchten’ (HEG N:246, na-aḫ-mi): RV. nám·na- ‘sich beugen/neigen’ (WbRV. 711), and OGaul. nem·na·li- ‘célébrer’ (DLG. 234) from √neɑh- (IEW 754).6 (ii) For *ḫɑe, see Hitt. ḫarunai- ‘aufhellen’ (HEG H:190, ḫa-ru-na-iz-[zi]), RV. aruṇá- ‘rötlich’ (WbRV. 107), Arm. arev- ‘Sonne’ (ArmGr. 1:424) from √ḫɑeru-. 1.4.2. Against communis opinio SAUSSURE’s compensatory lengthening is erroneous: PIE *eɑḫ –› *aɑḫ –› Hitt. aḫ –› IE /a/ resulted in a short vowel (SPIE 107-110). 1.5. PIE *ēɑ/ɑē: THE COLOURING RULES FOR Neogr. *ā 1.5.1. PIE *ɑ and *ē assimilated into Indo-European /ā/, and Neogr. *ā is split in two (SPIE 111-114): Hitt. ā(ḫ), Do. ᾱ, RV. ā PIE *ēɑ(ḫ) (*ā1) Hitt. (ḫ)ā, Do. ᾱ, RV. ā PIE *(ḫ)ɑē (*ā2) (i) For *ēɑḫ, see Hitt. paḫš- ‘protect’ (CHD P:2f., pa-aḫ-ši): Lat. pāstōr- ‘Hirt’ (WH 2:260) from *pēɑḫs- (IEW 839). (ii) For *ḫɑē, see Hitt. ḫaša- ‘Feuerstelle’ (HEG H:196, ḫa-aš-ša-aš): OLat. āsā- ‘Altar’ (WH 1:61). 1.5.2. Until EICHNER (1973:53, 71f.) LT reconstructed no long vowels. After EICHNER *ē is often restored to explain the missing ‘a-colouring’ in Hitt. eḫ/ḫe with the rules *ēh2, *h2ē –› IE ē. Consequently, the correct rules were only formulated in SPIE. 1.6. PIE *ē/e/Ø: QUANTITATIVE ABLAUT WITH AND WITHOUT PIE *ɑ́/ɑ 1.6.1. The PIE *e-ABLAUT had three quantities *ē/e/Ø (SPIE 111-114), attested e.g. in √u̯eJh- ‘fahren’ (IEW 1118-1120)7: (i) For *ē, see Lat. uēx- ‘führen’ (WH 2:742), OCS. vĕs- ‘fahren’ (Sadnik √1063). (ii) For *e, see Lat. uehō (WH 2:742), Go. wig- ‘Weg’ (GoEtD. 402). (iii) For Ø, see RV. uh- (WbRV. 1243), LAv. vaoz- ‘worwärtzbringen’ (AIWb. 1329- 1330). 1.6.2. With *ɑ/ɑ́ the *e-ABLAUT yielded four correspondence sets (SPIE 114): Hitt. (ḫ)Ø/Ø(ḫ), Gr. Ø, RV. Ø PIE *(ḫ)ɑ/*ɑ(ḫ) (–) Hitt. (ḫ)a/a(ḫ), Gr. α, RV. i PIE *(ḫ)ɑ́/*ɑ́(ḫ) (*ǝ) Hitt. (ḫ)a/a(ḫ), Gr. α, RV. a PIE *(ḫ)ɑe/*eɑ(ḫ) (*a) 6 For OGaul. nemnali- ‘célébrer’ see PYYSALO (2015b: §2). 7 For the absence of *H, see HLu. PES2(-)wa/i-za=ha ‘carried’ (CHLu. 2.11.7). Hitt. (ḫ)ā/ā(ḫ), Do. ᾱ, RV. ā PIE *(ḫ)ɑē/*ēɑ(ḫ) (*ā) Thus PIE *ɑ/ɑ́+*e/ē equals Neogr. *Ø/ǝ/a/ā. 1.7. PIE *oḫɑ/oɑḫ: THE REVISIED BRUGMANN’S LAW (Neogr. *o) 1.7.1. Neogr. *o, the cause of BRUGMANN’s law, stands for *oH (SPIE 121-125): Gr. ο, RV. a/ā PIE *oḫɑ/*oɑḫ (*o) The Indo-Iranian lengthening appears in *oHCV, otherwise the outcome is short. (i) For *oHCV (BRUGMANN’s law), see Hitt. ḫu̯ara- ‘schmücken’ (HEG H:299-300, ḫu-u-ua-ra-an-zi): LAv. gaošāvara- ‘Ohrschmück’ (AIWb. 486) from *hɑu̯oro- (IEW 1150-1151). (ii) For *oHCV in Anatolian, see CLu. GKLlaḫba- ‘Elfenbein’ (DLL. 61), Hitt. uploads/Geographie/ the-solution-to-the-proto-indo-european.pdf