Bomhard - A Sketch of Proto-Indo-Anatolian Phonology (2024)

A Sketch of Proto-Indo-Anatolian Phonology Allan R. Bomhard Florence, SC USA 2024 A Sketch of Proto-Indo-Anatolian Phonology By Allan R. Bomhard Florence, SC 2024 Bomhard, Allan R. (1943— ) A Sketch of Proto-Indo-Anatolian Phonology. By Allan R. Bomhard (1) Anatolian languages; (2) Indo-European languages; (3) Comparative Linguistics. 2024. © 2024 by Allan R. Bomhard (see below). (Revised 27 August 2024.) COPYRIGHT NOTICE: This open-access work is published under a Creative Commons license. This means that you are free to copy, distribute, or display this work as long as you clearly attribute the work to its author and publisher, that you do not use this work for any commercial gain in any form, and that you in no way alter, transform, or build on this work outside of its use in normal academic scholarship without express permission. If you want to reuse or distribute this work, you must inform its new audience of its license terms. For more information, see details of the Creative Commons license at: http://creativecommons.org/licenses/by-nc-nd/4.0/ UNDER THE CREATIVE COMMONS LICENSE, YOU MAY: • Read this work free of charge. • Download and store this work free of charge; • Distribute this work for personal use free of charge; • Print this work, in whole or in part, for personal use; • Share this work in a context where no financial gain occurs. UNDER THE CREATIVE COMMONS LICENSE, YOU MAY NOT: • Gain financially from the work in any way; • Sell the work or seek monies in relation to the distribution of the work; • Use the work in any commercial activity of any kind; • Profit a third party indirectly through use or distribution of the work; • Distribute in or through a commercial body (with the exception of academic usage within educational institutions such as schools and universities); • Reproduce, distribute, or store the cover image outside of its function as a cover of this work; • Alter or build on the work outside of normal academic scholarship. Table of Contents Table of Contents ……………………………………………………….…… i 1. General Introduction …………………………………………………………. 1 2. The Phonological Systems of the Anatolian Daughter Languages ….………. 5 3. Anatolian and the Laryngeal Theory …………………………………….…... 21 4. The Reconstruction of the Proto-Anatolian Consonant System ….………….. 53 5. The Reconstruction of the Proto-Indo-Anatolian Vowel System ….………… 67 6. The Non-Anatolian Daughter Languages ……………………………………. 99 References ….………………………………………………………………… 125  Chapter 1 General Introduction 1.1. Indo-Anatolian Following a suggestion made by Emil Forrer, Edgar Sturtevant championed the theory that Proto-Anatolian and Proto-Indo-European were sister languages descended from an earlier common ancestor, which he called “Proto-Indo-Hittite”. According to Sturtevant, this theory was based upon the observation that the Anatolian branch (mainly Hittite at the time) contained a number of very archaic features that set it apart from the other daughter languages as a group. Consequently, Sturtevant concluded that the Anatolian branch must have become separated from the main speech community before the other branches came into being. Except for a small number of supporters, such as Warren Cowgill (1965 and 1971), this theory was not embraced by the majority of Indo-Europeanists, who regarded Anatolian as simply another branch of Indo-European, comparable to all the others. Gradually, as more work was done on Hittite and, as the remaining Anatolian languages became better understood, it became increasingly clear that Sturtevant’s views were not so easily dismissed. The turning point seems to have been reached in the late 1990s, when work (namely, the application of computational cladistics to the problem of subgrouping) by Donald Ringe, Tandy Warnow, and Ann Taylor demonstrated that the Anatolian branch was, indeed, the first to become separated from the main speech community, followed, in due course, first by Tocharian (cf. Peyrot 2019), and then by the other branches. As a result, this view has now become more widely accepted, and “Indo-Hittite” has been renamed “Indo-Anatolian” to incorporate all of the Anatolian daughter languages, not just Hittite. According to Peyrot (2019:188), the Indo-Anatolian family tree may be represented as follows (this is modified from the chart given in Ringe—Warnow—Taylor 2002:87): 2 Chapter 1 1.2. The Reconstruction of the Proto-Indo-Anatolian Phonological System One of the things that is axiomatic about all languages is that they change over time. The rate of change may differ among languages and even within languages, but change is inevitable. And yet, this fundamental axiom is embarrassingly ignored when it comes to the reconstruction of the Proto-Indo-Anatolian phonological system. The reconstructed forms given in the standard literature as the alleged ancestors of what is found in Hittite and the other Anatolian daughter languages are based almost always on the phonological system of Proto-Indo-European as reconstructed on the basis of a comparison of the nonAnatolian daughter languages, to wit (cf. Byrd 2018:2056—2057): 1 A. Consonants: Voiceless stop Voiced stop Voiced aspirate Fricative Nasal Liquid Glide Labial *p *b *bº *m Dental *t *d *dº *s *n *r, *l Palatal *% *“ *“º Velar *k *g *gº Labiovelar Postvelar *k¦ *g¦ *g¦º *h÷, *hø, *hù *u̯ *i̯ Note: The only real concession that has been made to the Anatolian data is the inclusion of laryngeals. B. Vowels: *i, *ī *e, *ē *ə *a, *ā *o, *ō *u, *ū Byrd notes: For the vowels, one usually reconstructs the typologically common five-vowel inventory with a correlation of length (*ā̆, *ē̆, *ī̆, *ō̆, *ū̆). To his credit and in contrast to most other scholars, Byrd (2018:2057) states: In the pages that follow, these traditional views will be maintained as the most likely state of affairs for a late stage of PIE, though it is probable that the system looked quite different at an earlier point in time. 1 Even though I have chosen the exposition given by Byrd (2018) as one of the more modern treatments of the subject, any one of a number of other sources could have been chosen with the same result. General Introduction 3 To my knowledge, except for Sturtevant, there has never been a systematic attempt to reconstruct the phonological system of the parent language of both the Anatolian daughter languages and the non-Anatolian daughter languages, namely, Proto-Indo-Anatolian. And yet, as astutely observed by Byrd, it is “probable that the system looked quite different at an earlier point in time”. The purpose of this book is to examine the evidence of the Anatolian languages, and, on that basis, to attempt to reconstruct the phonological systems of their immediate ancestors, namely, Proto-Anatolian and Proto-Indo-Anatolian. For the most part, inner Anatolian developments will not be considered — for information on these developments, the references listed at the end of this book should be consulted. Both Proto-Indo-Anatolian and Proto-Indo-European must have been fully natural languages, like any other language. Thus, it follows that we must strive for realism in our reconstructions — they must be based upon the concrete evidence presented by the attested daughter languages, and they must also be consistent with typological expectations. That is to say, our reconstructions must be driven exclusively by the available evidence, not by theory. The theory must be derived from the evidence — the evidence must not be distorted or cherry-picked to agree with or support a particular theory. (This point is repeated elsewhere in this book.) That is why, for example, I reject the traditional reconstruction of the Proto-Indo-European phonological system (see above), including the reconstruction of only three laryngeals (*h₁, *h₂, *h₃ [*H₁, *H₂, *H₃; *ə̯₁, *ə̯₂, *ə̯₃]) and the relegation of the vowel *a to a marginal status or even the denial of its existence altogether. As I have emphasized over and over again in previous publications, the traditional reconstruction of the Proto-Indo-European phonological system is, in several aspects, unquestionably not a typologically plausible system (cf. Bomhard 1981 and 2016; Gamkrelidze—Ivanov 1995; Hopper 1973; Salmons 1993). Like all natural languages, Proto-Indo-Anatolian and Proto-Indo-European must have had internal variation, must have been in contact with other languages, and must have changed over time. Some of the internal variation is apparent in the regional vocabulary items that have been recovered as well as in the shared innovations and shared archaisms found among various daughter languages. Indications of language contact have long been suggested, most recently by Bomhard (2019a) for Pre-Indo-European and Pre-Northwest Caucasian (see the Appendices to Chapter 4). Finally, as noted above, this book deals with the recovery of possible earlier stages of development. 1.3. Contents of This Book This book is divided into six chapters, plus references: 1. 2. 3. 4. 5. 6. General Introduction The Phonological Systems of the Anatolian Daughter Languages Anatolian and the Laryngeal Theory The Reconstruction of the Proto-Anatolian Consonant System The Reconstruction of the Proto-Indo-Anatolian Vowel System The Non-Anatolian Daughter Languages References 4 Chapter 1 Much of what is contained in this book has been taken from previous papers and books that I have prepared — these papers and books are cited throughout the current book and are listed in the references. I have updated and enhanced this material to reflect the most recent research in the field. Needless to say, it has been quite a challenge to weed through the many conflicting viewpoints found in the literature. In so doing, I have tried to select the theories that seem to me to be most consistent with the available evidence.  Chapter 2 The Phonological Systems of The Anatolian Daughter Languages 2.1. Introduction The first Anatolian daughter language to be discovered and described was Hittite. For many years, the remaining Anatolian daughter languages (Cuneiform and Hieroglyphic Luwian, Palaic, Kalasmaic, Kanishite,2 Lycian, Milyan, Lydian, Carian, Pisidian, and Sidetic3) were not sufficiently known to be included in the discussion. That situation has since been rectified, though not all questions have been answered. In this chapter, I would like to explore the phonological systems of the Anatolian daughter languages. In the following sections, the evidence from the major Anatolian daughter languages will be included. I would like to emphasize, however, that not every issue will be addressed, and not every point of view or proposed explanation will be considered. Instead, I will be stressing the developments in the individual daughter languages that seem to me to be best supported by the available evidence. For other points of view, the relevant entries listed in the references should be consulted. 2.2.0. Hittite Hittite is the best attested Anatolian daughter language. Hittite texts date from the 16th to the 13th centuries BCE and number well over 30,000 tablets. During that period of time, Hittite went through several stages of development (Old Hittite, Middle Hittite, and New or Neo-Hittite) and changes in its writing system (old ductus, new ductus). New texts (cuneiform tablets) are still being discovered, while the decipherment of existing texts is ongoing. For a brief discussion of the rediscovery, together with a grammatical sketch, of the Hittite language, cf. Beckman 2011. See also Genz—Mielke (eds.) 2011. Throughout its recorded history, Hittite was written in a form of cuneiform syllabary (cf. Archi 2010; Popova 2015),4 which was ultimately created to record Sumerian. Indeed, Sumerian logograms (so-called “Sumerograms”) regularly occur in Hittite texts (for a detailed description of Hittite writing conventions, cf. Hoffner—Melchert 2008:9—24; Sturtevant 1951:10—28). Due to the writing of Sumerograms for several common words, the underlying Hittite words are unknown. Finally, the cuneiform syllabary was an imperfect medium for writing Hittite, and, as a result, uncertainties remain concerning important aspects of Hittite phonology. For information on the origin of cuneiform writing in general, cf. Walker 1998; Woods 2020. Kanishite (Kanišite) is a dialect of Hittite (cf. Kloekhorst 2019). Pisidian, Sidetic, Kalasmaic, and Kanishite will not be considered in this book due to the paucity of evidence. The language of Troy was also likely to have been an Anatolian daughter language, though this is still quite uncertain. 4 A small number of texts are written in hieroglyphs. 2 3 6 Chapter 2 2.2.1. Sturtevant’s Law “Sturtevant’s Law” is the name given to the Hittite scribal convention according to which double writing of medial stops (though only when the cuneiform syllabary makes this possible, and even then not consistently [cf. Melchert 1994a:14]) in certain words contrasts with single writing of medial stops in certain other words. This writing convention is interpreted under Sturtevant’s Law to be the method by which the Hittite scribes indicated some sort of phonemic contrast, usually taken to be a contrast between medial voiceless stops on the one hand and medial voiced stops on the other (cf. Sturtevant 1951:26—28, §53). This interpretation is based upon the observation that words exhibiting medial double writing of stops generally correspond etymologically to words in other IndoEuropean languages with medial voiceless stops (or their equivalents), while words exhibiting medial single writing of stops generally correspond etymologically to words in other Indo-European languages with medial voiced stops (or their equivalents), the latter being derived from what has traditionally been reconstructed as either plain voiced stops or as voiced aspirated stops at the Proto-Indo-European level. The following examples illustrate the general patterning (the Proto-Indo-European reconstructions represent the Phonemic Pitch Stage of Proto-Indo-European [Pre-Anatolian Proto-Indo-European or Proto-Indo-Anatolian], which was the stage of development just before the separation of the Anatolian daughter languages from the main speech community). Note: The ProtoIndo-European reconstructions are according to the Glottalic Model of Indo-European consonantism): Medial Double Writing: A. Hittite (3 pl. pres.) li-ip-pa-an-zi ‘they smear’ (also written li-pa-a-an-zi) ~ Sanskrit liptá-ḥ ‘smeared, anointed’; Greek λιπαρός ‘oily, greasy’, λίπος ‘fat, oil’ < Proto-IndoEuropean *lipº-. B. Hittite a-ap-pa ‘afterwards, back, again’ ~ Sanskrit ápa ‘away, from, off’; Greek ἄπο, ἀπό ‘from, away from, far from, apart from, away, off, back again’ < Proto-IndoEuropean *hepºa- [*hapºa-] (later *hepºo- [*hapºo-]). C. Hittite (3 sg. mid.) ki-it-ta(-ri) ‘lies’ ~ Sanskrit (3 sg. mid. impf.) á-śeta ‘lay’, (3 sg. mid. pres.) śéte ‘lies’; Avestan saēte ‘lies’; Greek (3 sg. impf.) ἔ-κειτο ‘lay’, (3 sg. pres.) κεῖται ‘lies’ < Proto-Indo-European 3 sg. mid. ending *-tºa- (later *-tºo-). D. Hittite (acc. sg.) ú-it-ta-an ‘year’ ~ Greek ἔτος ‘year’; Latin vetus ‘old’ < Proto-IndoEuropean *wetº-. E. Hittite (3 sg. pres.) lu-uk-ki-iz-zi ‘kindles, grows light’ (also written lu-uk-zi) ~ Greek λευκός ‘light, bright, brilliant, white’; Latin lūceō ‘to shine’ < Proto-Indo-European *lukº-, *lewkº-. Medial Single Writing: A. Hittite (nom.-acc. sg.) ne-pí-iš ‘heaven, sky’ ~ Sanskrit nábhas- ‘sky, cloud, mist’; Greek νέφος ‘cloud’; Old Church Slavic nebo ‘sky’ < Proto-Indo-European *nebas(later *nebºos-). The Phonological Systems of the Anatolian Daughter Languages 7 B. Hittite (nom.-acc. sg.) pí-e-da-an ‘place’ ~ Sanskrit padám ‘step, footstep, position, site’; Greek πέδον ‘the ground, earth’ < Proto-Indo-European *pºet’am (later *pºet’om). C. Hittite (nom.-acc. sg.) wa-a-tar ‘water’ ~ Sanskrit udán- ‘water’; Greek ὕδωρ ‘water’; Gothic watō ‘water’; Old Church Slavic voda ‘water’ < Proto-Indo-European *wet’/*wat’-/*ut’- (later *wet’-/*wot’-/*ut’-). D. Hittite (1 sg. pres.) e-it-mi ‘I eat’ ~ Sanskrit ádmi ‘I eat’; Greek ἔδομαι ‘I eat’; Latin edō ‘I eat’ < Proto-Indo-European *ʔet’-. E. Hittite (nom.-acc. sg.) i-ú-kán, i-ú-ga-an ‘yoke’ ~ Sanskrit yugám ‘yoke’; Greek ζυγόν ‘yoke’; Latin iugum ‘yoke’; Gothic juk ‘yoke’; Old Church Slavic igo (< *jъgo) ‘yoke’ < Proto-Indo-European *yuk’am (later *yuk’om). F. Hittite (nom. sg.) ḫar-ki-iš ‘white’ ~ Sanskrit árjuna-ḥ ‘white, bright’; Greek ἀργός ‘shining, bright, glistening’; Latin argentum ‘silver’ < Proto-Indo-European *‿ ħherk’[*‿ ħhark’-]. G. Hittite (nom. sg.) pár-ku-uš ‘high’ ~ Armenian barjr ‘high’; Sanskrit bṛhánt- ‘high’ < Proto-Indo-European *br̥ g- (later *bºr̥ gº-). H. Hittite (nom.-acc. sg.) ḫé-kur, ḫé-gur ‘summit, peak’ ~ Sanskrit ágram ‘point, tip, summit’ < Proto-Indo-European *Hek’¦r-. There also exist several well-known exceptions to Sturtevant’s Law, in which words exhibiting medial double writing of stops in Hittite correspond etymologically to words in other Indo-European daughter languages with medial voiced stops. Examples include (cf. Kronasser 1966:14; Bomhard 1984b:116): A. Hittite ú-uk-ga ‘I’ (also written ú-uk, ú-ga; Melchert [1994a:7] considers the u to be analogical after the 2 sg. personal pronoun tu-uk, tu-ga ‘you’, while Kloekhorst [2008b:112—114] considers it to be from the Proto-Anatolian oblique form *ʔMúg) ~ Latin egō, egŏ ‘I’; Greek ἐγώ(ν) ‘I’ < Proto-Indo-European *ʔek’-aH (later *ʔek’-oH). B. Hittite 2 pl. mediopassive primary ending -dduma in, for example, i-ya-at-du-ma ‘you go’ ~ Sanskrit 2 pl. mid. secondary ending -dhvam; Avestan 2 pl. mid. secondary ending -δwəm; Greek 2 dual mid. primary and secondary ending -σθον (< *-zdºwom), 2 pl. mid. ending -σθε (< *-zdºwe) < Proto-Indo-European *-dwem/*-dwam/*-dum (later *-dºwem/*-dºwom/*-dºum). C. Hittite (3 sg. pres.) píd-da-i, pád-da-i ‘to dig’ ~ Latin fodiō ‘to dig’; Lithuanian bedù, bèsti ‘to dig, to bury’; Gaulish bedo- ‘canal, ditch’; Old Church Slavic bodǫ, bosti ‘to stick, to prick’ < Proto-Indo-European *bed-/*bad- (later *bºedº-/*bºodº-). D. Hittite (acc.-dat. sg.) am-mu-uk-ga ‘to me’ (also written am-mu-uk, am-mu-ug-ga, ammu-uq-qa, am-mu-uk-qa) ~ Greek (acc. sg.) ἔμε-γε ‘me’ < Proto-Indo-European *-k’e. E. Hittite (nom. sg.) me-ik-ki-iš ‘large’ ~ Greek μέγας ‘great’ < Proto-Indo-European *mek’-. 2.2.2. The Glottalic Model of Proto-Indo-European Consonantism At the beginning of the twentieth century, the Neogrammarian reconstruction of the Proto-Indo-European phonological system, which was based upon the principle that sound 8 Chapter 2 laws admit no exceptions, was widely accepted as being a fairly accurate representation of what had existed in the Indo-European parent language. To this day, the Neogrammarian reconstruction (or variations of that reconstruction) enjoys widespread support among Indo-Europeanists. The Neogrammarian reconstruction of the Proto-Indo-European stop system, which was modeled after the phonological system found in Old Indo-Aryan (represented by Vedic and Classical Sanskrit) consisted of a four-way contrast of (1) plain voiceless stops, (2) voiceless aspirated stops, (3) plain voiced stops, and (4) voiced aspirated stops, thus (cf. Brugmann 1904:52; see also Szemerényi 1996:54—69; Burrow 1973:67): 1 2 3 4 p t % q qß ph th %h qh qßh b d “ œ œß bh dh “h œh œßh (bilabial) (dental) (palatal) (pure velar) (labiovelar) During the last century, it became widely accepted that the traditional voiceless aspirates (column 2) should be removed from the Proto-Indo-European phonological inventory (cf. Bomhard 1986:69—71 for details). The problem with removing the voiceless aspirates, however, is that the resulting system has no typological parallels among the known languages of the world (cf. Jakobson 1971[1957]:528; Martinet 1970:115). And yet, on structural grounds, positing a three-way contrast (without the voiceless aspirates) for Proto-Indo-European instead of the four-way contrast (with the voiceless aspirates) posited by the Neogrammarians seems fully justified. There are also problems involving the traditional plain voiced stops (column 3). One such problem, which is usually mentioned in the standard handbooks, is the unexpected statistically low frequency of occurrence of the traditional plain voiced bilabial stop *b. Such a frequency distribution is not at all characteristic of /b/ in natural languages having a voicing contrast in stops (for details, cf. Gamkrelidze 1978:9—46). Rather, the frequency distribution points to the original non-voiced character of this sound in Proto-IndoEuropean. Indeed, the frequency distribution of all of the traditional plain voiced stops (*b, *d, *“, *œ, *œß) points to the non-voiced character of the entire series when viewed from a typological perspective. Moreover, the traditional plain voiced stops are rarely found in pronouns and in inflectional affixes. Finally, there is the problem of the root structure constraint that prohibits the co-occurrence of two plain voiced stops in a given root. It was in trying to find solutions to these problems in particular that the Georgian scholar Thomas V. Gamkrelidze and the Russian scholar Vjačeslav V. Ivanov, on the one hand, and the British-born American scholar Paul J. Hopper, on the other, working independently, were led to propose, in the early 1970’s, a radical revision of the ProtoIndo-European stop system. Observing that the traditional plain voiced stops seemed to exhibit many of the typological characteristics of glottalized stops (ejectives), they proposed reinterpreting this series as ejectives. In their version of what has now come to be known as the “Glottalic Theory”, Gamkrelidze and Ivanov made no changes to the The Phonological Systems of the Anatolian Daughter Languages 9 traditional voiced aspirates, but they reinterpreted the traditional plain voiceless stops as voiceless aspirates. In this revised interpretation, aspiration is viewed as a redundant feature, and the phonemes in question could be realized as allophonic variants with or without aspiration depending upon the paradigmatic alternation of root phonemes. The system of Gamkrelidze and Ivanov may be represented as follows (cf. Gamkrelidze 1976:403 and 2001a:84): 1 2 3 4 ph/p th/t $h/$ kh/k kßh/kß - (p’) t’ $’ k’ k’ß bh/b dh/d ĝh/ĝ gh/g gßh/gß (bilabial) (dental) (palatal) (pure velar) (labiovelar) The revisions proposed by Gamkrelidze, Hopper, and Ivanov provide typologically natural explanations for the problems mentioned above, specifically: A. By reinterpreting the traditional plain voiceless stops (column 1) as voiceless aspirates, there is no longer a problem, from a typological point of view, with positing a series of voiced aspirates (column 4) for Proto-Indo-European, since the imbalance caused by the removal of the traditional voiceless aspirates (column 2) is eliminated. B. Reinterpretation of the traditional plain voiced stops (column 3) as glottalics makes it easy to account for the statistically low frequency of occurrence of the traditional plain voiced bilabial stop *b (which becomes a bilabial ejective *p’ in the revised system), since the glottalic member is always characterized by a low frequency of occurrence (there often being a total absence at this point of articulation) in the bilabial series in attested languages having ejectives. C. In languages having ejectives, it is common for ejectives to be either excluded from or underrepresented in inflectional affixes and pronouns. D. Several languages with ejectives have a constraint against the co-occurrence of two ejectives in a root. Thus, reinterpretation of the traditional plain voiced stops as glottalics provides a typologically natural explanation for the root structure constraint prohibiting the co-occurrence of two (traditional) plain voiced stops in a given root. Moreover, the revisions proposed by Gamkrelidze, Hopper, and Ivanov provide new insights into the underlying principles governing Grassmann’s Law and Barthomomae’s Law. Finally, it may be noted that strong support for the changes proposed by Gamkrelidze, Hopper, and Ivanov is to be found in Germanic, Armenian, and (the poorly-attested) Phrygian (cf. Diakonoff—Neroznak 1985:5). According to the traditional interpretation, Germanic, Armenian, and Phrygian had been thought to have undergone “sound shifts” (in German, Lautverschiebungen). Under the revised interpretation, however, they are rightly seen as relic areas. The Proto-Indo-European stop system reconstructed above may be viewed as reflecting a late stage of development. For Pre-Anatolian Proto-Indo-European (= Proto-Indo- 10 Chapter 2 Anatolian), I have argued elsewhere that the traditional voiced aspirates are to be reconstructed as plain voiced stops and that the development of this series into voiced aspirates is a later development (cf. Bomhard 2023.1:114—115). That this series was not aspirated in Pre-Anatolian Proto-Indo-European will be demonstrated below. 2.2.3. From Proto-Indo-European to Proto-Anatolian The three series reconstructed above for Proto-Indo-European by Gamkrelidze and Ivanov were preserved as separate series in Proto-Anatolian (cf. Bomhard 2019b). This is clear, for instance, from the different treatment of the voiced and voiceless velar stops before high front vowels in the Luwian branch of Anatolian (Hieroglyphic and Cuneiform Luwian along with the later Lycian). Here, the voiceless members are preserved, while the voiced member is lost; for example: A. Cuneiform Luwian kiša- ‘to comb, to card’ ~ Hittite kišai- ‘to comb’ < Proto-Anatolian *kºes- ‘to comb, to card’ < Pre-Anatolian Proto-Indo-European *kºes-. Probable nonAnatolian cognates include Greek κέσκεον (< *kºes-kºes- [reduplicated]) ‘tow, oakum’; Old Church Slavic češǫ, česati ‘to comb, to pull off’; Russian česátʹ [чеѕать] ‘to comb, to card’; Lithuanian kasù, kàsti ‘to dig, to rake’. B. Cuneiform Luwian (nom. sg.) (i-)iš-ša-ri-iš ‘hand’; Hieroglyphic Luwian (dat. sg.) istri ‘hand’; Lycian izri- ‘hand’ (all with loss of an earlier initial voiced velar before high front vowel) ~ Hittite (nom.-acc. sg.) ki-eš-šar ‘hand’ < Proto-Anatolian *gēsar ‘hand’ < Pre-Anatolian Proto-Indo-European *gēsr̥ . Non-Anatolian cognates include Sanskrit hásta-ḥ ‘hand’; Old Persian dasta- ‘hand’; Avestan zasta- ‘hand’; Latin praestō (< *prae-hestōd) ‘at hand, ready’. Note: The Hieroglyphic Luwian form contains an epenthetic t. More evidence is possibly to be found in the treatment of dentals initially before high front vowels and *y in Hittite. In this case, the voiceless aspirated and plain voiced members are preserved (though *tº- later becomes z- (= /ˆ/) in this environment in Hittite, but not in the other older Anatolian languages), while the plain voiceless (from earlier glottalized) member becomes š, as shown in the following examples (cf. Melchert 1994a:118): A. Hittite (dat.-loc. sg.) šiwatti ‘day’ ~ Palaic (nom. sg.) Ti-ya-az(-) name of the sun-god; Cuneiform Luwian (nom. sg.) Ti-wa-az name of the sun-god; Hieroglyphic Luwian Tiwat- name of the sun-god, (adj.) tiwatami- ‘bright, sunny’ < Proto-Anatolian *tyēwat< Pre-Anatolian Proto-Indo-European *t’yēw-. Non-Anatolian cognates include: Sanskrit dyáu-ḥ ‘heaven, sky, day’; Armenian tiv ‘day’; Latin diēs ‘day’; Old Irish dïe, dïa ‘day’. B. Hittite (gen. sg.) ši-(i-)ú-na-aš ‘god’ < Proto-Anatolian *tyū́ - < Pre-Anatolian ProtoIndo-European *t’y»w- (cf. Melchert 1994a:150). Non-Anatolian cognates include Greek Ζεύς ‘Zeus’, δῖος ‘god-like, divine’; Sanskrit devá-ḥ ‘god’; Latin deus ‘god’. The Phonological Systems of the Anatolian Daughter Languages 11 There may be additional evidence from the later Lycian and Lydian, as Shevoroshkin (1988b) has tried to show. Shevoroshkin claims, for instance, that the (traditional) ProtoIndo-European stop system developed as follows in Lycian: ProtoIndoEuropean t d dh Milyan Initially Milyan Medially Lycian Initially Lycian Medially tdt- -t-d-d- tddt- -t-d-d- Some of the evidence that Shevoroshkin adduces to support his views, however, is questionable and is to be treated with the utmost caution. Moreover, there is other Lycian evidence, not cited by Shevoroshkin, that points to alternative interpretations (note especially Melchert’s [1994a:53—54] critical assessment of Shevoroshkin’s views). There are enough clues within the Anatolian daughter languages to support the contention that the three series of stops reconstructed for the phonological system of the IndoEuropean parent language maintained their separate identity in Proto-Anatolian. It is not possible to tell, however, whether or not series 3 was glottalized at the Proto-Anatolian level, though there is nothing to indicate otherwise. Consequently, I assume that series 3 was glottalized in Proto-Anatolian. The most important point to bear in mind is that it is series 3 and 4 that are represented by medial single writing in Hittite and that it is series 1 that is represented by medial double writing. Thus, the Proto-Anatolian stop system is probably to be reconstructed as follows: 1 2 3 4 pº tº kº k¦º - p’ t’ k’ k’¦ b d g g¦ Notes: 1. Melchert (1994a:53) assumes that series 3 and 4 merged in Proto-Anatolian. He further assumes (1994a:21) that the earlier voicing contrast was replaced by a fortis ~ lenis opposition in the older Anatolian daughter languages. 2. According to Kloekhorst (2016:226—228), series 3 was pre-glottalized in Proto-Anatolian. 2.2.4. Hittite As mentioned above (§2.2.0), Hittite was written in a cuneiform syllabary derived from a form of Old Akkadian cuneiform in use in Northern Syria in the beginning of the second millennium BCE (cf. Gamkrelidze 1968:91—92). Now, the older cuneiform writing system, which was developed by the Sumerians, was not suited to rendering Akkadian, 12 Chapter 2 much less Hittite. In Old Akkadian, voiceless, voiced, and emphatic consonants were not differentiated in the writing system, though methods were gradually developed to represent most of the Akkadian phonological distinctions. This is important, for no attempt was ever made, even after Akkadian had introduced separate syllabograms to differentiate voiceless, voiced, and emphatic consonants, to modify the Hittite writing practices to make use of the same methods to note a voicing contrast in stops. We must conclude, therefore, that the Hittite scribes did not feel that it was worthy of noting such a contrast, regardless of what the underlying phonetics may have been. What then, if anything, does medial double writing of stops indicate if not a voicing contrast? The answer to this question can be ascertained by looking closely at the ProtoAnatolian stop system reconstructed above. Series 1 is differentiated from series 3 by the presence of aspiration in series 1 and by glottalization in series 3 and from series 4 by the absence of voicing, while series 3 and 4 are differentiated from each other by a contrast in voicing. Since it is only series 1 that is represented by medial double writing, it must have been the feature of aspiration that was considered significant by the Hittite scribes. This means that series 4 cannot have been aspirated since it, too, would have been represented by medial double writing. It also means that the opposition of medial double writing and medial single writing cannot have indicated a voicing contrast, since, if that had been the case, then series 3 would also have been represented by medial double writing, which is clearly not the case, both series 3 and 4 being represented by medial single writing. It should be noted here that Gamkrelidze (1968:94) was the first to suggest that medial double writing of stops in Hittite was used as a means to indicate the presence of aspiration (Patri 2019 reaches the same conclusion): The aspirated stops were rendered in Hittite cuneiform by double writing of consonants, whereas single writing was used to represent plain stops. Gamkrelidze devotes two later articles (1982 and 2008) to a detailed analysis of Hittite consonantism, noting specifically in the first article (1982:78—79): In light of these facts, Sturtevant’s rule acquires a completely different significance: The graphic reduplication of plosives is used to denote not the simple voiceless plosives but the corresponding aspirated phonemes, while their single writing was used for non-aspirated consonants. Thus we can reach the conclusion that the Hittite phonological system was characterized by two series of plosives: aspirated ones denoted by the graphic reduplication of the relevant consonant on the one hand, and non-aspirated ones on the other, denoted by single writing of the corresponding consonant. Three series of Proto-Indo-European plosives: 1) glottalized, 2) voiced (aspirated), and 3) voiceless (aspirated) were reduced in the Hittite phonological system into two series opposed to each other by virtue of aspiration. The differentiating feature for the phonological opposition of plosives is only the factor of aspiration (tenseness), regardless of the original voiced/unvoiced opposition of the plosives, which had phonemic significance in the ProtoIndo-European system. The correlation of Proto-Indo-European plosives depending upon whether they were voiced, voiceless or glottalized was replaced in the Hittite phonological system by the correlation on the basis of “aspiration” (tenseness). The Phonological Systems of the Anatolian Daughter Languages 13 The feature of aspiration, which had been phonologically irrelevant with the phonemes of series 2) and 3) in Proto-Indo-European, became a phonologically significant feature in the Hittite system of plosives. In the process, the Proto-Indo-European series 1) and 2) merged into a general series of non-aspirated plosives as opposed to the series of aspirated ones, which derives from the Proto-Indo-European series 3) of voiceless (aspirated) plosives. With the adoption of the Akkadian cuneiform writing, the two series of Hittite plosives — the simple and the aspirated — were written not by the signs for voiced and voiceless plosives, as these were not differentiated in the early Akkadian writing system, but with the single and double writing of the respective consonants. Accordingly, the single writing of a consonant was used to express simple plosives, while for the Hittite aspirated (tense) plosives a new means of denotation was found, that is the reduplication of the consonant in question, by which was solved the problem of how to differentiate graphically between simple plosive and the corresponding aspirated consonant. This does not explain the whole picture, however, for we must still account for the exceptions to Sturtevant’s Law. Since the exceptions exhibit medial double writing of stops in Hittite words which correspond etymologically to words in other Indo-European languages with medial voiced stops (or their equivalents, these being derived from either earlier glottalized stops or earlier voiced aspirates at the Proto-Indo-European level), the distinguishing characteristic cannot have been aspiration. Let us take a look at each of the exceptions listed previously (as above, the Proto-Indo-European reconstructions represent the Phonemic Pitch Stage of Proto-Indo-European [“Pre-Anatolian Proto-Indo-European”, that is, Proto-Indo-Anatolian]): A. Hittite ú-uk-ga ‘I’: in this case, we are dealing with a particle added to the stem (cf. Sturtevant 1951:103, §170a), so that, in fact, we have gemination: Proto-IndoEuropean *ʔek’-+-k’e/a > Proto-Anatolian *ek+ka > Hittite (with analogical u) *uk+ka. The particle appears in Greek as -γε (dialectal -γα). B. Hittite 2 pl. mediopassive ending -dduma in, for example, i-ya-at-du-ma ‘you go’: here, the verb stem is probably to be reconstructed as *ʔy-eh- [*ʔy-ah-] (so Sturtevant 1951:34, §61; Puhvel 1984— .1/2:334—335, however, considers iya- to be a thematic stem comparable to Vedic 3 sg. pres. ayate and derives it from earlier *eyo-), to which the ending *-dum- has been added. Most likely, the second laryngeal (*h), which was lost as an independent phoneme in Hittite, has merged with the following dental, producing a geminate. Thus, double writing here indicates former presence of a laryngeal, which has left a trace in the gemination of the following stop. C. Hittite (3 sg. pres.) píd-da-i, pád-da-i ‘to dig’: here, we are dealing with a Proto-IndoEuropean stem *bed-/*bad-, to which a laryngeal suffix has been added: *bed-+-H-. In this case, the laryngeal has merged with the preceding stop, producing a geminate. D. Hittite (acc.-dat. sg.) am-mu-uk-ga ‘to me’ (also written am-mu-uk, am-mu-ug-ga, ammu-uq-qa, am-mu-uk-qa): as in the first example, we are dealing with a particle that has been added to the stem, thus producing gemination. E. Hittite (nom. sg.) me-ik-ki-iš ‘large’: this is similar to the third example in that a laryngeal suffix has merged with a preceding stop, producing a geminate: Proto-IndoEuropean *mek’-+-Hi- > Hittite *mekkis (cf. Kimball 1999:261). 14 Chapter 2 2.2.5. Reinterpretation of Sturtevant’s Law Thus, Sturtevant’s Law is to be reinterpreted as follows: double writing of medial stops indicates stop plus something additional, that is, either aspiration or gemination, while single writing of medial stops indicates a plain stop pure and simple. It does not indicate a voicing contrast as traditionally assumed. Kloekhorst (2014:544—547) also interprets the contrast as one of length. According to Melchert (1994a:21), medial double writing of stops in Hittite indicates fortis (= long) articulation, while single writing indicates lenis (= short) articulation. Similar views are expressed by Kimball (1999:94—95); see also Jäntti 2017, Sturtevant 1932, and Yates 2019a. Specifically, Melchert (1994a:117) notes: I assume that the PA [= Proto-Anatolian] contrast of voiceless/voiced stops has been reanalyzed in Hittite as one of fortis/lenis, with the realization in medial position being that of long/short. I retain the standard symbols for voiceless and voiced stops for convenience. One important contributing factor in this reanalysis was the devoicing of voiced stops in wordinitial position. Melchert’s views are not necessarily incompatible with the conclusions reached here. On fortis/lenis articulation, cf. Ladefoged—Maddieson (1995:95—990, Laver (1994:344). Both Ladefoged—Maddieson and Laver caution against the careless use of these terms. 2.2.6. The Hittite Phonological System According to Yakubovich (2020:227), the Hittite phonological system was most likely as follows (see also Melchert 1984 and 1994a: Kimball 1999; Watkins 2004:556): Stops:5 Affricate: Fricatives: /p(p)/ /b/ Nasals: Liquids: Glides: /m/ Vowels: /i/, /i:/ /w/ /t(t)/ /d/ /͜ts/ /s/ /n/ /r/, /l/ /e/, /e:/ /a/, /a:/ /k(k)/ /g/ /k(k)ʷ/ /gʷ/ /x/ /ɣ/ /xʷ/ /ɣʷ/ /j/ /o/, /o:/6 /u/, /u:/ According to Bomhard (2019b), the voiceless stops (/p(p)/, /t(t)/, /k(k)/, /k(k)ʷ/) were actually aspirated voiceless stops (/pʰ/, /tʰ/, /kʰ/, /kʷʰ/). Bomhard interprets medial double writing of stops in Hittite as the means used to indicate the presence of aspiration, on the one hand, and gemination, on the other hand. This interpretation was first suggested by Thomas Gamkrelidze (1968 and 1982) ⸺ see also Patri 2019:204. 6 The existence of /o/ and /o:/ in Hittite is controversial (for discussion, cf. Kloekhorst 2008a:35—60). 5 The Phonological Systems of the Anatolian Daughter Languages 15 Regarding the laryngeals, Yakubovich (2020:227—229) states: A peculiarity that sets Hittite phonologically apart from the non-Anatolian Indo-European languages is the presence of so-called “laryngeals”, namely, /x/, /ɣ/, /xʷ/, and /ɣʷ/. They are rendered in Hittite by the same signs <ḫa>, <ḫi>, <ḫu>, etc. as those used for combinations involving the Akkadian fricative /ḫ/. The system of four contrasting “laryngeals” proposed above is based on Kloekhorst (2008, 27), while certain other reconstructions assume fewer elements belonging to this class. The precise phonetic realization of Hittite laryngeals is subject to much debate, but the renderings of the phoneme /x/ in Egyptian and Ugaritic converge in the interpretation of its counterparts in the first millennium Anatolian languages leading to the conclusion that its most likely articulation in Hittite was a velar or uvular fricative (cf. Weiss 2016). This appears to agree with the assumed value of /ḫ/ in Akkadian (cf. Chapter 7). The ancient Indo-European languages outside Anatolian are commonly assumed to have lost their “laryngeals” with secondary phonological effects, such as compensatory lengthening and change in vowel quality. For example, *peh₂s- ‘to pasture, protect’ is thought to have yielded Hittite paḫš- and *pās- in many Indo-European groups outside Anatolia, as in Latin pās-tor ‘shepherd’ (Kimball 1999, 402). Nevertheless, the hypothesis that the loss of “laryngeals” represents a classifying innovation of non-Anatolian Indo-European, although vehemently defended in (Lehrman 1998), is now a minority view. Most Indo-Europeanists are convinced that the “laryngeals” /h₁/, /h₂/, and /h₃/ are independently required in order to explain non-trivial correspondences between non-Anatolian Indo-European languages, e.g. Sanskrit guru- ~ Greek βαρύς ~ Latin grāvis ‘heavy’, which are commonly reconstructed as PIE *gʷr̥ h₂u- ‘heavy’. The set of such beliefs, which had begun to develop long before the decipherment of Hittite, is known as the Laryngeal Theory. At the same time, the frequently advanced claim that Hittite data provided a definite confirmation to the validity of the Laryngeal Theory is inaccurate, because the number and distribution of “laryngeals” in Hittite is different from what is typically reconstructed for ProtoIndo-European. In fact, the precise pattern of correspondence between the reconstructed IndoEuropean and Anatolian “laryngeals” represents one of the most controversial aspects of Anatolian historical phonology. In order to appreciate the extent of existing differences of opinion, it should suffice to compare the accounts of Melchert (1994, 64—74, 122), on the one hand, and Kloekhorst (2006, 77—95), on the other hand. For our purposes, the following transliterated cuneiform signs were available in Hittite to write laryngeals (cf. Sturtevant 1951:13—14): aḫ iḫ uḫ daḫ duḫ ḫa ḫal ḫar ḫur ḫaš ḫat ḫé ḫi ḫu maḫ saḫ taḫ túḫ These signs were ultimately taken over from Akkadian. As noted in the quotation above from Yakubovich, ḫ represents a voiceless velar fricative (IPA [x]) in Akkadian (cf. von Soden 1995:31). As with the writing of stops, medial single as opposed to medial double writing of laryngeals is usually taken to indicate some sort of phonemic contrast. The laryngeals *H₂ and *H₃ are assumed to have been preserved in Hittite (and Anatolian in general), while *H₁ and *H₄ are assumed to have been lost. 16 Chapter 2 2.3. Palaic Palaic is very poorly documented. It is only found in a small number of cuneiform texts preserved in the Hittite national archives at Hattusas. The texts deal with cultic/ritualistic matters. Palaic had ceased to be a spoken language by the Neo-Hittite period (14th—13th centuries BCE), perhaps even earlier. According to Melchert (2004b:586), the Palaic phonological system included at least the following consonants: /p/ /b/ /f/8 /m/ /t/ /d/ /͜ts/ /s/ /n/ /l/ /w/ /kʷ/7 /gʷ/ /k/ /g/ /ʒ/9 /x/10 /ɣ/ /r/ /y/ According to Melchert (2004b:587), Palaic had the following vowels: /i/, /i:/ /e/, /e:/ /u/, /u:/ /a/, /a:/ 2.4. Cuneiform Luwian Cuneiform Luwian is much better documented than Palaic. As the name suggests, it is the form of Luwian written in the cuneiform syllabary. It is found first in an extensive body of texts preserved in the Hittite national archives at Hattusas. The texts are primarily ritualistic in nature, and most date from the Neo-Hittite period (14th—13th centuries BCE). In addition to the specifically Cuneiform Luwian documents, a number of Luwian words are scattered here and there throughout Hittite texts, where they are indicated as such by so-called “Glossenkeil” — distinguishing wedges placed before the words in question. Lastly, there are a number of Luwian loanwords in Hittite. Cf. Yakubovich 2011:539— 541 for details on the Luwian corpus. According to Melchert (2020a:247—248), the Cuneiform Luwian phonological system was most likely as follows (see also Melchert 2004a:579—580; Kloekhorst 2008b:118): I have added the voiceless and voiced labioverlars /kw/ and /gw/ to the table. They are missing in the original, though Melchert (2004b:586) does mention that there is “a good chance” that they should be included. 8 /f/ is found in Hattic loanwords. 9 /ʒ/ may represent a weak palatal fricative, though this is not entirely certain. 10 Melchert interprets the laryngeals as voiceless and voiced pharyngeal fricatives (/ħ/ [= *H₂] and /ʕ/ [= *H₃]). However, he notes that they could have been voiceless and voiced velar fricatives (/x/ and /ɣ/) instead. The interpretation of the Palaic laryngeals as velar fricatives is now the prevailing view. 7 The Phonological Systems of the Anatolian Daughter Languages Stops: Affricate: Fricatives: /p/ /b/ Nasals: Liquids: Glides: /m/ Vowels: /i/, /i:/ /t/ /d/ /͜ts/ /s/ /n/ /r/, /l/ /w/ /k/ /g/ /kʷ/ /gʷ/ /x/ /ɣ/ /xʷ/ /ɣʷ/ /j/ 11 /a/, /a:/ 17 /o/, /o:/ /u/, /u:/ 2.5. Hieroglyphic Luwian As the name implies, Hieroglyphic Luwian is the form of Luwian written in a native hieroglyphic script. The script was used to inscribe writings on stone monuments and seals. The hieroglyphic inscriptions begin to appear from the 13th century BCE, though some writings may actually be older. The script contains over 500 signs (cf. Laroche 1960), some of which have more than one value.12 The ultimate origin of the script is unknown. No doubt, the phonological system given above for Cuneiform Luwian came close to what existed in Hieroglyphic Luwian. However, due to the limitations of the Hieroglyphic script only the following vowels and consonants were represented in the writing system — these are rendered in accordance with the conventional transcription: Vowels: a i u Consonants: p t z s n r, l k m w h y According to Sasseville (2024), the dental consonants d, r, l, and n converged into a flap /ɾ/ in Hieroglyphic Luwian under certain conditions. This is commonly referred to in the literature as “rhotacism” (cf. Arbeitman—Ayala 1981). Sasseville points out that a similar change is found in Lydian. Whether Luwian (and Hittite) had the vowels /o/ and /o:/ is quite uncertain. Two signs exhibit dual vocalization (ra/i and wa/i), while one sign exhibits triple vocalization (la/i/u). Moreover, some sounds could be written with several different signs. 11 12 18 Chapter 2 2.6. Lycian and Milyan Lycian and Milyan are closely-related dialects. They are written in an alphabet derived from or close to Greek. Both are considered to belong to the Luwian branch of Anatolian (cf. Adiego 2016:51; Kloekhorst 2022:67—68). The Lycian corpus is represented by a little over 150 stone inscriptions as well as over 200 coins. Only two of the extant texts are written in Milyan. In certain aspects, Milyan appears to be more archaic than Lycian. According to Melchert (2004c:592), the Lycian phonological system included the following consonants (see also Kloekhorst 2008b:123—126 and 128):13 Stops:14 Affricate: Fricatives: Nasals: Liquids: Glides: /p/ 21 /β/ /m/ /θ/ /w/ /t/15 /͜ts/18 /s/ /h/19 /ð/ /n/ /r/, /l/ /y/ /c/16 /k/ /kʷ/17 /x/20 /ɣ/ Lycian had the following vowels (cf. Melchert 2004c:595—596) (see also Kloekhorst 2008b:119—123 and 128): /i/ /e/ /u/ /a/ Each vowel had contrasting nasalized varieties, though separate letters exist for only /ã/ and /ẽ/. Lycian vowels were subject to a widespread vowel assimilation rule — Melchert (2004c:595) renders this rule as follows: V [-high] > V [α back] /__C0V [α back]. Melchert points out that there are many exceptions to this rule. Finally, it should be mentioned that prehistoric syncope was widespread in Lycian (cf. Melchert 2004c:596). 2.7. Lydian Lydian is rather poorly documented and still not completely understood. There are only a little over extant 100 texts, less than 30 of which contain more than just a few words. The transcription given here is slightly different from that used by Melchert. The stops have voiceless and voiced allophones. According to Melchert (2004c:593), the voiced allophones occur after nasals and nasalized vowels, while the voiceless allophones occur elsewhere. 15 Proto-Anatolian *kw becomes t in Lycian before i and possibly e. 16 /c/, /θ/, and /h/ are only found in Lycian, not in Milyan. 17 /kw/ is found only in Milyan in personal names. 18 This sound is traditionally transcribed as z. 19 /h/ is due to a change of /s/ to /h/ in Lycian — this change did not occur in Milyan. 20 /x/ (traditionally transcribed as χ) appears as q in the name of the Storm-God trqqñt-. It probably represents a labialized /x/, that is, /xw/. 21 The phonemes shown in the table as /β/, /ð/, and /ɣ/ are traditionally transcribed as b, d, and g. 13 14 The Phonological Systems of the Anatolian Daughter Languages 19 Lydian is written in an alphabet derived from or close to Greek. According to Melchert (2004d:603), the Lydian phonological system included the following consonants (see also Gusmani 2000—2002:21—23; Kloekhorst 2023a, summary pp. 130—131; Oreshko 2019): Stops:22 Affricates: /p/ Fricatives: /f/ /v/ /m/ Nasals: Liquids: /ð/ (?) /t/ /͜ts/ (?) /‿ dz/ (?) /s/23 /ç/24 /k/ /kw/ /n/, /ν/25 /l/ /r/, /λ/26 Lydian probably had the following vowels (cf. Melchert 2004d:604): /i/ /e/ /a/ /u/ /o/ There were also two nasalized vowels, though their phonetic values are uncertain. 2.8. Carian The Carian corpus is extremely limited. The extant texts can be dated approximately to the fourth to the third centuries BCE. Though there has been substantial progress over the past forty years in the interpretation of the values of the Carian letters, there are still unresolved problems and disagreements (cf. Kloekhorst 2008b:137—139). As noted by Melchert (2004e:609—610) concerning the decipherment and interpretation of the values of the Carian letters: A new era began in 1981 when John Ray first successfully exploited the evidence of the Carian-Egyptian bilingual tomb inscriptions to establish radically new values for several Carian letters, as well as to confirm the values of others. Additional investigation, notably by Ray, Ignacio Adiego, and Dieter Schürr, has led to further revisions and refinements of the new system. The basic validity of this approach was shown by its correct prediction of Carian personal names which have subsequently appeared in Greek sources. Nevertheless, many uncertainties and unsolved problems remained, and several reputable experts were skeptical of the new interpretation of the Carian alphabet. Melchert (2004e:610) lists the following subset of characters of the Carian alphabet: Though underlyingly voiceless, Lydian stops may have had voiced allophones in certain environments, though this is not normally indicated in the writing. 23 /s/ is traditionally transliterated as ś. 24 /ç/ is traditionally transliterated as s. 25 The phonetic value of the letter transcribed as ν is uncertain. 26 /λ/ is probably the result of the prehistoric palatalization of /l/. 22 20 Chapter 2 Three of the letters in this table are of particular importance to our understanding of the development of laryngeals in Anatolian under investigation in this paper, namely, the letters transcribed by Melchert as /q/, /x/, and /k/. Adiego (2004:242—245) assigns slightly different values for two of the letters, namely, /k/ = Melchert’s /x/ and /k̑/ = Melchert’s /k/. Both Adiego and Melchert agree on /q/. Several scholars (Kloekhorst, Schürr, and Simon) have adopted the new values in recent work on Carian and Anatolian, in general — Brosch (no date) provides an excellent summary of the views of these scholars, while Simon (2023) lists the lexical evidence. However, Adiego’s arguments in favor of the values he assigns to these letters are highly conjectural and based upon conflicting evidence. When one takes into consideration etymological factors, it can be observed that Carian /k/ (Melchert) ~ /k̑/ (Adiego) corresponds etymologically to /k/ in older Anatolian languages (Hittite, Palaic, and Cuneiform and Hieroglyphic Luwian), while /x/ (Melchert) ~ /k/ (Adiego) corresponds etymologically to /ḫ/ in older Anatolian languages. Finally, /q/ appears to represent /x¦/, just as in Lycian. In my opinion, this speaks in favor of the values given by Melchert in the above table and against the values assigned by Adiego.  Chapter 3 Anatolian and the Laryngeal Theory 3.1. Introduction Let us start by reiterating the basic premises of the “Laryngeal Theory”. Beginning with an article published in 1927 and culminating in the theory presented in Chapter 2 in his 1935 book Études indoeuropéennes I, the Polish Indo-Europeanist Jerzy Kuryłowicz (1895—1978), attempted to demonstrate that the so-called “original long vowels” reconstructed by the Neogrammarians for Proto-Indo-European were the result of a contraction of short vowels with certain consonantal elements, upon their loss. Kuryłowicz (1935:28) called these elements “consonantal ə (schwa)”, which he wrote *ə̯ (= *H) Kuryłowicz made additional discoveries as a result of his investigations, eventually developing the core tenets of the Laryngeal Theory, which remain to the present day. In his 1935 book (1935:28—30), Kuryłowicz summarized his findings as follows: 1. Every Proto-Indo-European original long vowel (that is to say, not arising through analogical lengthening nor through the contraction of two short vowels) is a contraction of a short vowel with one of three following consonantal elements (*ə̯₁, *ə̯₂, *ə̯₃), thus: *e + *ə̯₁ > *ē, *e + *ə̯₂ > *ā, *e + *ə̯₃ > *ō; *o (derived from *e through qualitative Ablaut) + *ə̯₁ > *ō. Notes: (1) Kuryłowicz was uncertain about whether *o + * ə̯₂ > *ō, due to the lack of positive examples. He was also unable to ascertain whether *ō was the result of *o + *ə̯₁ or of *e + *ə̯₃. (2) According to Sturtevant (1942:27 and 1951:49—51), *ə̯₃ did not change a contiguous *e to *o. 2. When *ə̯ was found between vowels, it disappeared. The resulting vowels were then contracted, thus: *axə̯ax > *axax > *āx. 3. When *ə̯ was found between two consonants, it disappeared without a trace, except in Greek. 4. When *ə̯ was found between a consonant and an immediately following vowel, it also disappeared. However, in Indo-Iranian, *k, *t, *p + prevocalic *ə̯ > *kh, *th, *ph. The sound which the Neogrammarians designated as vocalic *ə is the reduced vowel *e, which remained after the disappearance of the consonant *ə̯. 5. Every Proto-Indo-European word beginning with a vowel has lost an initial *ə̯. The character of the lost *ə̯ is revealed by the quality of the remaining vowel, thus: *e< *ə̯₁e-, *a- < *ə̯₂e-, *o- < *ə̯₃e- (provided that “original *o” is involved and not an Ablaut grade of *e). On the other hand, there were also certain roots beginning with a consonant that were derived from an earlier sequence of *ə̯ + consonant. Such roots were simplified by the simple loss of *ə̯, except in Greek and Armenian. In composition (and also in sandhi), after a preceding short vowel, the loss of initial preconsonantal *ə̯ resulted in the lengthening of that vowel: *E + *ə̯T > long vowel + *T (*E = vowel; *T = consonant). 6. In Hittite words of Indo-European origin, ḫ seems to continue *ə̯₂. However, there is also a group of Indo-European words with *ā̆ in which ḫ is missing in Hittite. 22 Chapter 3 These seem to be due to a different cause than a neighboring *ə̯₂. These can be attributed to the presence of a fourth type of *ə̯ in Proto-Indo-European, namely, *ə̯₄, which was lost in Hittite. To recapitulate: 1. Proto-Indo-European had at least three, possibly four “laryngeals” (= Kuryłowicz’s “consonantal ə [schwa]”). (Note: The term “laryngeals” refers to these sounds as a class and says nothing about their phonetic make-up (cf. Laroche 1986:134). The term was introduced by Hermann Möller [1917].) 2. Two of these laryngeals are preserved in Hittite, where they are written (initially) ḫ-, (medially) -ḫ(ḫ)-. 3. The loss of laryngeals between a short vowel and a following consonant resulted in the compensatory lengthening of these vowels. 4. Several of these laryngeals had “vowel coloring” effects — that is to say that they changed the quality of contiguous vowels. 5. Every Proto-Indo-European root began with a consonant. The roots traditionally reconstructed with an initial vowel have lost a preceding laryngeal. 6. Though not directly preserved as such, the laryngeals left discernable traces in the non-Anatolian Indo-European daughter languages. Note: *H (*ə̯₂) is now known to have been partially preserved in Classical Armenian. Kuryłowicz correlated his theory of “consonantal ə (schwa)” with the “coéfficients sonantiques” posited in 1878 by the young Ferdinand de Saussure and with the “laryngeals” posited by Hermann Möller (1917) on the basis of his comparative work on Indo-European and Semitic. Möller’s term has stuck. Though there have been many refinements in the Laryngeal Theory since 1935, Kuryłowicz’s presentation remains the foundation of the theory. Nevertheless, though the Laryngeal Theory has now gained universal acceptance, the exact number and phonetic character of the laryngeals are still being debated. Finally, the Proto-Indo-European reconstructions I am proposing in what follows are informed by my work on distant linguistic relationship (cf. Bomhard 2023). Hence, where they exist, references are given to the pertinent Nostratic etymologies. Moreover, the notes accompanying the individual examples are particularly important. 3.2. The Anatolian Evidence At the time when Kuryłowicz wrote (1935), the remaining Anatolian daughter languages (Cuneiform and Hieroglyphic Luwian, Palaic, Kalasmaic, Lycian, Milyan, Lydian, Carian, Pisidian, and Sidetic27) were not sufficiently known to be included in the discussion. That situation has since been rectified, though not all questions have been answered. In the following sections, the evidence from the Anatolian daughter languages will be included. The following symbols: *H₁, *H₂, *H₃, *H₄ will be used in this part of the 27 Pisidian, Sidetic, and Kalasmaic will not be considered in this book due to the paucity of evidence. Anatolian and the Laryngeal Theory 23 discussion when referring to the Proto-Indo-European laryngeals (other notational conventions are found in the relevant literature: *ə̯₁, *ə̯₂, *ə̯₃, *ə̯₄; *h₁, *h₂, *h₃, *h₄; etc.). Also, the Proto-Indo-European reconstructions will be given in accordance with the Glottalic Model of Proto-Indo-European consonantism (cf. Bomhard 2016; Salmons 1993) — when different, they will be immediately followed, in brackets, by the traditional reconstructions found in the standard etymological dictionaries and comparative grammars. Finally, at this point, I will not address the question of subgrouping, though I have long supported the view that the Anatolian branch was the first to split off from the main Indo-European speech community — this view is now gaining wide acceptance. I would like to emphasize that not every issue will be addressed, and not every point of view or proposed explanation will be considered. Instead, I will be stressing what seem to me to be the principal developments of the laryngeals in the Anatolian daughter languages — that is to say, the developments that seem to be best supported by the available evidence. 3.3. Examples The following select examples illustrate the representation of laryngeals in Anatolian (the forms, meanings, and etymologies are taken from the following: Kimball 1999; Kloekhorst 2008a; Puhvel 1984— ; Tischler 1977— ; Sturtevant 1951; Melchert 1984 and 1994a — these works, as well as the other etymological dictionaries listed in the references at the end of this paper, must be consulted for additional information). The Hittite, Palaic, Cuneiform Luwian, and Hieroglyphic Luwian forms are given in plene writing: 1. *H₁ > Ø in Anatolian: Hittite (1st sg. pres. act.) e-eš-mi ‘to be’ (< *H₁es-mi); Cuneiform Luwian (3rd sg. pres. act.) a-aš-ti ‘to be’; Hieroglyphic Luwian (3rd sg. pres. act.) a-sa-ti, á-sa-ti ‘to be’; Palaic (3rd sg. imptv. act.) a-aš-du ‘to be’; Lycian (3rd sg. pres. act.) esi ‘to be’ ~ Sanskrit ásti ‘to be’; Greek ἐστί ‘to be’; Latin est ‘to be’; Gothic ist ‘to be’; Old Lithuanian ẽsti ‘to be’; Old Church Slavic jestь ‘to be’. Hittite (1st sg. pres. act.) e-id-mi ‘to eat’ (< *H₁et’-mi [*H₁ed-mi]); Palaic (3rd pl. pres. act.) a-ta-a-an-ti, a-da-a-an[-ti] ‘to eat’; Cuneiform Luwian (3rd pl. imptv. act.) a-da-an-du ‘to eat’; Hieroglyphic Luwian (3rd pl. imptv. act.) á-tà-tu-u ‘to eat’ ~ Sanskrit ádmi ‘to eat, to consume, to devour’; Greek ἔδω, ἔδομαι ‘to eat, to devour; (of worms) to gnaw’; Armenian utem ‘to eat’; Latin edō ‘to eat’; Gothic itan ‘to eat’; Old Icelandic eta ‘to eat’; Norwegian eta ‘to eat’; Swedish äta ‘to eat’; Old English etan ‘to eat’; Old Frisian eta, īta ‘to eat’; Old Saxon etan ‘to eat’; Dutch eten ‘to eat’; Old High German ezzan ‘to eat’ (New High German essen); Lithuanian jdu, jsti ‘to eat’, jda ‘food’; Latvian êst ‘to eat’; Old Prussian īst ‘to eat’; Old Church Slavic jasti ‘to eat’; Russian jestʹ [есть] ‘to eat’; Polish jeść ‘to eat’; Czech jísti ‘to eat’. [Cf. Bomhard 2023.3:814—815, no. 697, Proto-Nostratic root *ʔit’- (~ *ʔet’-): (vb.) *ʔit’- ‘to chew, to bite, to eat, to consume’; (n.) *ʔit’-a ‘the act of eating; that which is eaten: food, nourishment’.] 24 Chapter 3 Hittite (1st sg. pres. mid.) e-eš-ḫa-ḫa-ri ‘to sit down, to seat oneself; to sit; (act.) to sit, to reside; (trans.) to settle’ (< *H₁eH₁s-); Hieroglyphic Luwian (3rd pl. pret. act.) SOLIUM á-sa-ta ‘to be seated, to dwell’, i-sà-nu-wa/i- ‘to seat, to cause to sit’, i-sàtara/i-tá- ‘throne’ ~ Sanskrit ā́ smi ‘to sit’ (< *H₁eH₁s-mi). [Cf. Bomhard 2023. 3:785—786, no. 659, Proto-Nostratic root *ʔasʸ- (~ *ʔəsʸ-): (vb.) *ʔasʸ- ‘to put, to place, to set; to sit, to be seated’; (n.) *ʔasʸ-a ‘place, seat’; (adj.) ‘put, placed, set, established’.] Hittite (nom.-acc. sg.) e-eš-ḫar, iš-ḫar ‘blood; bloodshed’ (< *H₁esH₂r̥ ); Cuneiform Luwian (nom.-acc. sg.) a-aš-ḫar-ša ‘blood’; Hieroglyphic Luwian á-sa-ha+ra/imi-sà (n.) ‘blood offering’ ~ Sanskrit ásṛk ‘blood’ (the Sanskrit form contains an epenthetic k); Tocharian A ysār ‘blood’; Greek ἔαρ ‘blood’; Latin assyr ‘blood’. Note: ḫ is sometimes missing in Hittite, as in (nom.-acc. sg.) e-eš-šar (= e-eš-ḫar) and (gen. sg.) e-eš-na-aš (= e-eš-ḫa-na-aš). According to Kloekhorst (2008a:258), these forms represent scribal errors, while Kimball (1999:379—380) considers the loss of ḫ in these forms to be an archaism. Hittite (nom. sg.) at-ta-aš, ad-da-aš ‘father’ (< *H₁atºtºa [*H₁atta]) ~ Greek ἄττα ‘daddy’; Latin atta ‘father’; Gothic atta ‘father’; Old Frisian aththa ‘father’; Old High German atto ‘father’ (Middle High German atte, ätte ‘father’); Albanian atë ‘father’; Old Church Slavic otьcь ‘father’; Russian otéc [отец] ‘father’; Sanskrit (f.) attā ‘mother’ (*atta- ‘father’ is unattested, but note the following: Assamese ātā form of address to a respectable older man; Gujarati ātāji ‘grandfather’; Sinhalese ātā ‘grandfather’; Sindhi ado ‘brother’; Lahndi addā ‘father’; etc.). Note: This is a nursery word. Assuming that there was an initial laryngeal here, the only acceptable candidate is *H₁ (cf. Hamp 1965a:136 *ʔat(t)- ??) — and original *a. [Cf. Bomhard 2023.3:786—787, no. 680, Proto-Nostratic (n.) *ʔatºtºa ‘older male relative, father’ (nursery word).] Hittite (nom.-acc. sg. n.) a-iš, a-i-iš ‘mouth’ (< *H₁oH₁-es-); Cuneiform Luwian (nom.acc. sg.) a-a-aš-ša ‘mouth’ and, perhaps, (3rd sg. pret.) a-aš-ša-at-ta ‘to say’; Hieroglyphic Luwian (3rd sg. pres. act.) á-sa5-za-ta ‘to speak’ ~ Sanskrit ā́ s‘mouth’; Latin ōs ‘mouth’. Notes: (1) The reconstruction of the Proto-IndoEuropean form as either *h₃oh₁-os-, -es- (cf. De Vaan 2008:436; Mayrhofer 1986— 2001.I:81—82) or *h₁oh₃-es- (cf. Kloekhorst 2008a:166—167), with *H₃ in either initial or medial position, is impossible since *H₃ is preserved in Hittite. (2) Sturtevant (1951: 36, §61c) reconstructs Proto-Indo-Hittite *Aōys, *Aysos ‘mouth’, while Puhvel (1984— .1/2:15—17) reconstructs Proto-Indo-European (nom.-acc. sg.) *A₁ʷéE₁-es (> *ōyes > *āyes > *āyis > Hittite a-iš, a-i-iš), etc. Hittite (nom. sg. c.) a-aš-šu-uš ‘good; dear; favorable’ (< *H₁osu-s [cf. Melchert 1994:63 *ósu-]), (n.) (nom.-acc. sg.) a-aš-šu ‘good(ness), good things; goods, possessions’ ~ Greek ἐύς ‘good, brave, noble’; Sanskrit sú, su- ‘good’. Note: Kloekhorst (2008a:223—225) reconstructs Proto-Indo-European *h₁oh₁s-u- (?), in part to account for the Greek variant form ἠύς ‘good, brave’ as if from Proto-Indo- Anatolian and the Laryngeal Theory 25 European *H₁eH₁s-u-. However, the initial ἠ- in this form is usually attributed to metrical lengthening (cf. Beekes 2010.I:484—485; Chantraine 1968—1980.I:388; Frisk 1970—1973.I: 594—595; Wodtko—Irslinger—Schneider 2008:246, note 6). Hittite (3rd sg. pres. act.) pa-a-ši, pa-aš-zi ‘to swallow, to gulp down’ (< *pºoH₁-s[*poH₃s-]); Cuneiform Luwian (3rd sg. pret. act.) pa-aš-ta ‘to swallow’ ~ Sanskrit (reduplicated) píbati ‘to drink’; Greek πῑ́νω ‘to drink’; Latin (reduplicated) bibō ‘to drink’; Old Church Slavic piti ‘to drink’. Note: The Proto-Indo-European root is typically reconstructed as *poH₃- with *H₃, the thinking here being that, since *H₃ was a voiced laryngeal, it must be reconstructed here to account for the voicing of medial *p to *b in, for example, the Sanskrit and Latin forms cited above. However, *H₃ is preserved in Anatolian, and a laryngeal reflex is missing in both Hittite and Cuneiform Luwian. Now, according to the Glottalic Model of Proto-Indo-European consonantism, the laryngeal in question in this example is not *H₃ but *H₁, thus avoiding having to reconstruct a laryngeal (*H₃) for which there is no evidence in either Hittite or Cuneiform Luwian. The Proto-Indo-European form leading to the Sanskrit and Latin derivatives would thus have been (reduplicated) *pºi-pºH₁-e-ti > *pi-be-ti, assuming here that *H₁ was a glottal stop (*ʔ), which is now the common view (see below), and also assuming, consistent with the tenets of the Glottalic Model, that *-pºʔ- > *-b- (cf. Hopper 1977b:70). [Cf. Bomhard 2023.2:162—164, no. 134, Proto-Nostratic root *pºuʔ- (~ *pºoʔ-): (vb.) *pºuʔ- ‘to swell, to fatten’; (n.) *pºuʔ-a ‘swelling, fullness, fat(ness)’.] Hittite enclitic demonstrative particle (nom. sg.) -aš, (acc. sg.) -an, (n. sg.) -at ‘he, she, it’; (nom. sg. c.) a-ši, a-ši-iš ‘that (one)’, (dat. sg.) e-di, i-di, e-da-ni ‘to or for him, her, it’ (< *H₁e-) ~ Sanskrit ayám ‘this’ (gen. sg. m./n. a-syá, á-sya; f. a-syáḥ), idám ‘this’, (f.) iyám ‘she, this’, á-taḥ ‘from this, hence’ (< *H₁e-tºo-s [*H1e-to-s]), (n.) e-tát ‘this, this here’, ihá ‘here’, e-ṣá (f. e-ṣā) ‘this’; Old Persian a- ‘this’, aita‘this’, ima- ‘this’, iyam this’, idā ‘here’; Avestan a- ‘this’, aētat̰ ‘this’, ima- ‘this’, iδa ‘here’; Latin is, ea, id ‘he, she, it; this or that person or thing’; Oscan eiso- ‘this’; Umbrian (dat. sg.) esmei ‘to this, to it’; Old Irish é ‘he, they’, ed ‘it’; Gothic anaphoric pronoun is ‘he’, ita ‘it’; Old Icelandic relative particle es (later er) ‘who, which, what’; Old Saxon et, it ‘it’; Old High German er, ir ‘he’, ez, iz ‘it’; Lithuanian jìs (< *is) ‘he’. Note: Various extended forms must be reconstructed in Proto-Indo-European to account for developments in the daughter languages: *H₁e/*H₁o-+-y/i- > *H₁ey-/*H₁oy-/*H₁i-; *H₁e-/*H₁o-+-tºo-; etc. [Cf. Bomhard 2023.3: 746—751, no. 627, Proto-Nostratic demonstrative stems (originally deictic particles): proximate *ʔi- (~ *ʔe-) ‘this’; intermediate *ʔu- (~ *ʔo-) ‘that’; distant *ʔa- (~ *ʔə-) ‘that yonder, that over there’. Note: These stems regularly combined with other deictic particles: *ʔa/i/u+na-, *ʔa/i/u+ša-, *ʔa/i/u+ma-, *ʔa/i/u+tºa-, *ʔa/i/u+kºa-, *ʔa/i/u+ya-, etc. already in Proto-Nostratic.] Hittite (nom. sg.) an-na-aš ‘mother’ (< *H₁an(n)-o-s); Palaic (nom. sg.) an-na-aš ‘mother’; Cuneiform Luwian (nom. sg.) an-ni-iš, a-an-ni-iš ‘mother’; Lycian (nom. sg.) ẽni ‘mother’; Lydian (nom. sg.) ẽnaś ‘mother’ ~ Latin anna ‘foster-mother’; 26 Chapter 3 (?) Greek (Hesychius) ἀννίς· ‘grandmother’. Notes: (1) Hieroglyphic Luwian MATER-nat/i ‘mother’ may be read as /anna(i)-/ (cf. Kloekhorst 2008a:174). (2) The Proto-Indo-European ancestor of these forms is regularly reconstructed as *H₂en(n)- or the like to account for the initial a- (cf., for example, Beekes 2010.I:107 *h₂en- — Kloekhorst 2008a:174 reconstructs Proto-Anatolian *Honno-). However, this is a nursery word, for which *H₁an(n)-, with initial *H₁and original *a, seems to be the most plausible reconstruction. [Cf. Bomhard 2023.3:775—776, no. 649, Proto-Nostratic (n.) *ʔanʸa ‘mother, aunt’ (nursery word).] Hittite (3rd sg. pres. active) i-ya-(az-)zi, i-e-iz-zi ‘to do, to make, to treat, to beget, to perform (duty, ritual), to celebrate (deity, feast)’ (< Proto-Anatolian *iya- ~ *aya~ *ya-/*yē- ‘to do, to make, to perform, etc.’ < *H₁(e)yo- originally an interrogative verb stem meaning ‘to do what?, to act in what manner?’, later simply ‘to do, to make, to perform’); Cuneiform Luwian (3rd sg. pres. pass.) a-a-ya-ri ‘to make’; Hieroglyphic Luwian a(i)a- ‘to make’; Lycian (3rd sg. pres.) ati (< *ayati) ‘to make’; Lydian i- ‘to make’ ~ Tocharian A/B yām- ‘to do, to make, to commit, to effect’. [Cf. Bomhard 2023.3:789—790, no. 663, Proto-Nostratic root *ʔay- (~ *ʔəy-): interrogative verb stem *ʔay- ‘to do what?, to act in what matter?’] Hittite (imptv.) i-it ‘go!’ (< *H₁ey-/*H₁oy-/*H₁i- ‘to go’); Cuneiform Luwian (3rd sg. pres.) i-ti ‘goes’ ~ Greek (1st sg. pres.) εἶμι ‘I go’, (1st pl. pres.) ἴμεν ‘we go’; Sanskrit (1st sg. pres.) émi ‘I go’, (3rd sg. pres.) éti ‘goes’, (1st pl. pres.) imáḥ ‘we go’, (3rd pl. pres.) yánti ‘they go’; Avestan (3rd sg. pres.) aēiti ‘goes’; Old Persian (3rd sg. pres.) aitiy ‘goes’; Latin (1st sg. pres.) eō ‘I go’; Old Lithuanian (1st sg. pres.) eĩmi ‘I go’, (3rd sg. pres.) eĩti ‘goes’; Old Church Slavic ido˛, iti ‘to go’; Tocharian A (1st pl.) ymäs ‘we go’, B (1st sg.) yam, yaṁ ‘I go’. [Cf. Bomhard 2023.3:792—794, no. 665, Proto-Nostratic root *ʔay- (~ *ʔəy-): (vb.) *ʔay- ‘to go, to proceed’, (n.) *ʔay-a ‘journey’; also Bomhard 2023.3:815—817, no. 688, ProtoNostratic root *ʔiy- (~ *ʔey-): (vb.) *ʔiy- ‘to come, to go’, (n.) *ʔiy-a ‘approach, arrival; path, way’.] Hittite /*ekku-/ ‘horse’ (< *H₁ekºu- [in Anatolian]; *H₁ekºw-o-s ‘horse’ [*H₁é%u̯ o-] [in the non-Anatolian daughter languages] [literally, ‘the spirited, violent, fiery, or wild one’]); Cuneiform Luwian /*āššu-/ ‘horse’; Hieroglyphic Luwian á-sù-wa- /ásu-/ ‘horse’; Lycian esbe- ‘horse’ ~ Sanskrit áśva-ḥ ‘horse’; Avestan aspa- ‘horse’; Mycenaean i-qo (hiqqu̯ o-) ‘horse’; Greek ἵππος ‘horse’; Latin equus ‘horse’; Old Irish ech ‘horse’; Gothic *aiƕa- ‘horse’ in *aiƕatundi ‘bramble, prickly bush’ (literally, ‘horse-thorn’); Old Icelandic jór (< *eχwaʀ < *eχwaz) ‘stallion, steed’; Old English eoh ‘horse’; Old Saxon ehu- horse’ in ehu-skalk ‘horse-servant’; Lithuanian (f.) ašvà (Old Lithuanian ešva) ‘mare’; Tocharian A yuk, B yakwe ‘horse’, B yäkwaṣke ‘little horse’. Notes: (1) Kloekhorst (2008a:237—239) reconstructs Proto-Anatolian *ʔeḱu- ‘horse’. (2) There are no attested o-grade forms. (3) The Proto-Indo-European word for ‘horse’ is not in any way, shape, or form related to the Proto-Indo-European word for ‘swift’ (*ōku-, according to Anatolian and the Laryngeal Theory 27 Watkins 1985:45). [Cf. Bomhard 2023.3:798—799, no. 671, Proto-Nostratic root (Eurasiatic only) *ʔekº-: (vg.) *ʔekº- ‘to move quickly, to rage; to be furious, raging, violent, spirited, fiery, wild’; (n.) *ʔekº-a ‘rapid or violent movement, fury, rage’.] Hittite (3rd sg. pres.) a-ri ‘to arrive, to come’, (3rd sg. pres.) a-ra-(a-)i ‘to (a)rise, to lift, to raise; to (a)rouse’, (3rd sg. pres.) a-ar-aš-ki-iz-zi ‘to be arriving’, (3rd sg. pres.) ar-nu-(uz-)zi ‘to move along, to make go; to stir, to raise; to transport, to deport, to remove; to bring, to transmit, to deliver, to produce; to further, to promote’, (3rd sg. pres.) (a-)ar-aš-zi ‘to flow’ (< *H₁or-/*H₁r̥ - ‘to move, to set in motion; to arise, to rise; to raise’) ~ Sanskrit árṣati ‘to flow’, árṇa-ḥ ‘undulating, surging; wave’, ṛccháti ‘to go, to move, to send’, ṛṇóti ‘to go, to move, to arise’; Avestan ar- ‘to go, to move, to come’, aurva-, aurvant- ‘rapid, quick’, ərənaoiti ‘to set in motion’; Old Persian ar- ‘to move, to go or come toward’, aruvā ‘action’, aruva- ‘rapid, quick’; Greek ὄρνῡμι ‘to urge on, to incite, to move, to stir oneself, to make to arise’; Latin orior ‘to rise, to arise’. [Cf. Bomhard 2023.3:820—822, no. 691, Proto-Nostratic root *ʔor-: (vb.) *ʔor- ‘to move rapidly, quickly, hastily; to set in motion’; (n.) *ʔor-a ‘any rapid motion: running, flowing, pouring, etc.’; (adj.) ‘rapid, quick, hasty’; also Bomhard 2023.3:822—823, no. 692, Proto-Nostratic root *ʔorʸ-: (vb.) *ʔorʸ- ‘to rise (up)’, (n.) *ʔorʸ-a ‘rising movement or motion’. Note: Two separate Proto-Nostratic stems have fallen together in Proto-Indo-European: (A) *ʔor- ‘to move rapidly, quickly, hastily; to set in motion’ and (B) *ʔorʸ- ‘to rise (up)’.] Hittite (3rd sg. pres. act.) a-ar-ki ‘to mount, to copulate (with)’, (nom. pl.) ar-ki-i-e-eš ‘testicles’ (< *H₁orgº-/*H₁r̥ gº- ‘to mount, to copulate (with)’, *H₁orgºi-s ‘testicle’) ~ Avestan ərəzi ‘scrotum’; Greek ὄρχις ‘testicle’; Armenian orjikº ‘testicles’, orj ‘male’; Old Irish uirge ‘testicle’; Old Icelandic argr ‘unmanly, effeminate, cowardly; passive homosexual’, ergi ‘lust, lewdness’; Old English earg ‘cowardly; bad, depraved’; Old Frisian erch (also erg, arch) ‘angry, evil; wrong, bad, disgraceful; severe (wounds)’, erg ‘mean, cowardly’; Old Saxon arug ‘mean, cowardly’; Old High German arg, arag ‘mean, cowardly’; Lithuanian aržùs ‘lusty’, er͂ žilas (dial. ar͂ žilas) ‘stallion’; Albanian herdhë ‘testicle’. Note: Kloekhorst (2008a:203—204) reconstructs Proto-Indo-European *h₃rǵº-o, *h₃órǵºei, with initial *H₃. However, the Hittite evidence does not support such a reconstruction. [Cf. Bomhard 2023.3:822—823, no. 692, Proto-Nostratic (vb.) *ʔorʸ- ‘to rise (up)’: extended forms (cf. Bomhard 2023.3:823—824, no. 693): (vb.) *ʔorʸ-V-g- ‘to climb on, to mount, to copulate (with)’; (n.) *ʔorʸ-g-a ‘mounting, copulation’.] Comments: 1. One of the things that I see quite often in the literature involving laryngeals is that theory is allowed to take precedence over evidence, acting as a kind of intellectual “straitjacket” (cf. Pulleyblank 1993:65) — critical thinking requires that we free ourselves from any and all preconceived notions. This does not mean, however, 28 Chapter 3 that every proposal warrants equal consideration, nor does it mean that we suspend sound judgment. Ideas that are patently crackpot should be unequivocally rejected. An example of the approach under discussion here is the Hittite word for ‘mouth’ (a-iš, a-i-iš) cited above. The thinking here seems to be that non-apophonic *o in Indo-European always implies the presence of *H₃, even when corroborating Hittite evidence is lacking. Other, more frequent examples involve the reconstruction of *H₂ to indicate the “coloring” of *e to *a, even when corroborating Hittite evidence is lacking. (To his credit, Kuryłowicz got around this latter conundrum quite nicely by positing *H₄ — Sturtevant, Lehmann, Mallory—Adams, and Bomhard, among others, accept Kuryłowicz’s view in this matter.) This approach places the scholars reconstructing these laryngeals in the awkward position of having to explain why the laryngeals in question (*H₂ and *H₃) are sometimes present in the Hittite data and sometimes not — in other words, of having to specify the conditioning factors leading either to the retention or to the loss of these laryngeals in identical environments. This is rarely, if ever, done, and when explanations are offered, they often seem rather ad hoc. That is not to say that these laryngeals are never lost in Hittite — that is quite a different thing than reconstructing these laryngeals without sufficient evidence. I reject this approach — in my opinion, the evidence should always take precedence over theory and should never be altered to fit the theory. When anomalies occur, they require careful analysis and viable explanations, even if it means amending or abandoning the theory or the faulty reconstructions resulting from the erroneous generalization/application of the theory — as noted by Kimball (1999:386): “Commonly cited examples of loss can usually be explained in other ways”, and she gives several examples to illustrate this point. It should thus be perfectly clear from the examples listed above, as well as the examples listed by Kimball, that credible alternative explanations are almost always available to the questionable reconstructions found in the relevant literature. 2. Returning to the laryngeal under discussion in this section, as we have seen from the examples given above, *H₁ was found in the vicinity of all vowels, *e, *a, *o traditionally reconstructed for Proto-Indo-European. Moreover, *H₁ did not change the quality of contiguous vowels in any way, shape, or form at the Proto-IndoEuropean level. 3. *o and *a of traditional Proto-Indo-European appear as a in Hittite and Palaic, while *e, *o, and *a of traditional Proto-Indo-European appear as a in Cuneiform and Hieroglyphic Luwian. Further afield, Common Luwian a appears mostly as e in Lycian. 2. *H₂ > (1) ḫ- (initially), -ḫ(ḫ)- (medially) in Hittite, Palaic, Cuneiform Luwian, and Hieroglyphic Luwian (written h-, -h(h)-); > (2) χ- (initially), -χ-/-g- (medially), -q- (< */x¦/) in Lycian: Hittite (nom. sg.) ḫu-uḫ-ḫa-aš ‘grandfather’ (< *H₂ewH₂os); Cuneiform Luwian (abl.instr.) ḫu-u-ḫa-ti ‘grandfather’, also found in the anthroponyms Ḫuḫḫazitis, Ḫūḫananis (not in Kloekhorst, but cf. Puhvel 1984— .3:355—358); Hieroglyphic Luwian (nom. pl.) AVUShu-ha-zi ‘grandfather’; Lycian χuga- ‘grandfather’, also Anatolian and the Laryngeal Theory 29 found in the anthroponyms Epñχuχa and Κουγας — Melchert (1994a:289) considers the second χ in the name Epñχuχa to be secondary; perhaps Carian quq‘grandfather’ ~ Latin avus ‘grandfather’; Old Irish áue ‘grandson’; Armenian haw ‘grandfather’; Gothic awō (f.) ‘grandmother’. Note: According to Kloekhorst (2008a:353): Since Sturtevant (1928c: 163), these words are generally connected with Lat. avus, Arm. haw, etc. ‘grandfather’. It is clear that Lat. a- and Arm. ha- must reflect *h₂e-, which corresponds to Hitt. ḫ-. The second -ḫḫ- in Hittite corresponds to the acute intonation in SCr. ȕjāk which points to a laryngeal. Since *h₃ was lost intervocalically (cf. Melchert 1987b: 23f), it is likely that we must reconstruct *h₂ here as well. Thus, we arrive at *h₂euh₂-. The question remains why Hittite shows geminated -ḫḫwhere the Luwian languages show single -ḫ-. In my view, this problem can only be solved by assuming that this word originally was a root noun. If we reconstruct *h₂éuh₂-s, *h₂éuh₂-m, *h₂uh₂-ós, we can explain that on the one hand we find the thematicized stem *h₂éuh₂-o- in CLuw. ḫūḫa-, Lyc. χuge-, and also Lat. avus, Arm. haw, Goth. awo, etc., but on the other a thematicized stem *h₂uh₂-ó- which regularly yields Hitt. ḫuḫḫa- without lenition of *-h₂-. Compare šūḫḫ-, šuḫḫa- for a similar thematicization. It should be noted, however, that medial single writing of ḫ is also found in Hittite in the derivatives (nom. pl.) ḫu-u-ḫa-an-te-iš (alongside ḫu-uḫ-ḫa-an-te-iš, with medial double writing) ‘(great)grandfather’ and (nom.-acc. pl. n.) ḫu-u-ḫa-da-al-la ‘grandfatherly’ (this is probably a Luwian form, cf. Melchert 1993b:71). In view of these Hittite forms, as well as the Luwian anthroponyms cited above, Kloekhorst’s conjecture cannot be considered the final word on this matter. [Cf. Bomhard 2023. 3:884—885, no. 745, Proto-Nostratic (n.) *ħaw-a ‘a relative on the mother’s side’.] Hittite (nom.-acc. sg.) pa-aḫ-ḫur, pa-aḫ-ḫu-ur, pa-a-aḫ-ḫu-ur ‘fire’ (< *pºeH₂-ur [*peh₂-wr̥ : Adams, Kimball, Melchert, Yates], [*peh₂-ur: Kloekhorst], [*péxwr: Sturtevant]); Cuneiform Luwian (nom.-acc. sg.) pa-a-ḫu-u-ur ‘fire’ and, perhaps, (3rd sg. pret.) pa-wa-ar-it-ta ‘to light a fire’, without -ḫ- ~ Greek πῦρ ‘fire’; Armenian hur ‘fire’; Old Icelandic fýrr, fúrr ‘fire’; Old English fȳr ‘fire’; Tocharian A por, B pūwar ‘fire’. [Cf. Bomhard 2023.3:170—171, no. 140, Proto-Nostratic root *pºaħ- (~ *pºəħ-): (vb.) *pºaħ- ‘to warm, to heat, to burn’; (n.) *pºaħ-a ‘fire, flame, spark’; extended forms: (vg.) *pºaħ-V-w- ‘to warm, to heat, to burn’; (n.) *pºaħ-w-a ‘fire, flame, spark’.] Notes: (1) Kloekhorst (2008a:613—614) speculates that a labialized laryngeal may be involved here: *páHʷr, *paHʷénas. (2) The Luwian verb pa-wa-ar-it-ta ‘to light a fire’ may not be derived from or in any way related to pa-a-ḫu-u-ur ‘fire’. It may tentatively be compared with Greek φαύζειν ‘to roast, to fry, to parch’ and φαῦσιγξ ‘blister from burning, any blister’, provided these are not Pre-Greek loans (cf. Beekes 2010.II:1559). Boisacq (1950:1018), on the other hand, assumes Indo-European origin for the Greek forms cited here and compares φωΐς ‘blister on the skin, caused by a burn’ (< *bhōu-). Hofmann (1966:393) agrees with Boisacq. However, this etymology is rejected by Chantraine (1968—1980.II:1183), but later, Chantraine (1968—1980.II:1036) reconstructs *bhō-w- as the source of Greek (f. pl.) φωΐδες ‘blisters’. Frisk (1970— 30 Chapter 3 1973.II:998 and II:1057) does not really clarify the situation. Finally, Mann (1984—1987:68) brings in Low German bäuten ‘to make fire’ (pt. bödd, pp. bött) (cf. Middle Low German boten, buten ‘to make fire’; East Frisian böten ‘to make fire, to heat’) and reconstructs Proto-Indo-European *bhaudō, -i̯ ō ‘(to make) fire, (to) burn’ (> Proto-Germanic *ƀautan ‘to make fire’). On the basis of the above forms from Cuneiform Luwian, Greek, and West Germanic, we can cautiously reconstruct a Proto-Indo-European stem *bºeH₄-w/u-/*bºoH₄-w/u- ‘to light a fire’— an extended form of the root *bºeH₄-/*bºoH₄- (> *bºā-/*bºō-) (not *H₂) ‘to be bright, shining; to bring to light, to cause to appear; to make clear’ found in: Sanskrit bhā́ ti ‘to shine, to be bright, to be luminous; to be splendid or beautiful; to be conspicuous or eminent; to appear, to seem; to show one’s self, to manifest any feeling; to be, to exist’; Avestan bānu- ‘splendor’; Greek φάω ‘to give light, to shine’, φᾱνός ‘light, bright, joyous’, φαίνω ‘to bring to light, to cause to appear; to make known, to reveal, to disclose; to make clear; to show forth, to display; to set forth, to expound; to inform against one, to denounce; to give light, to shine; to come to light, to be visible, to appear; to come into being; to come about; to appear to be’, φάος, φῶς ‘light, daylight; light of the eyes’ (pl. φάεα ‘eyes’); Old Irish bán ‘white’; Old English bōnian ‘to polish’; New High German bohnen ‘to polish, to wax (floor)’ (cf. Rix 2001:68—69 *bºeh₂-/*bºh₂- ‘to glisten, to shine’; Pokorny 1959:104—105 *bhā-, *bhō-, *bhǝ- ‘to glisten’; Walde 1927—1932.II:122—123 *bhā-; Mallory—Adams 1997:513 *bheh₂- ‘to shine’; Wodtko—Irslinger— Schneider 2008:7—11 *bºeh₂-; Watkins (ed.) 2000:7 *bhā- ‘to shine’ (oldest form *bheH₂-, colored to bhaH₂-, contracted to *bhā-); Chantraine 1968— 1980.II:1168—1170 *bhā- (= *bheH₂-) and II:1170—1172; Boisacq 1950:1010— 1011 *bhā- and 1014—1015; Hofmann 1966:389—390 *bhā-; Frisk 1970— 1973.II:992—994 and II:989—991; Beekes 2010.II:1545—1546 *bheh₂- ‘to light, to shine’, II:1551—1552; Mayrhofer 1956—1980.II:493—494 *bhā-). Needless to say, this is all quite speculative. [Cf. Bomhard 2023.2:18—19, no. 13, ProtoNostratic root *bah- (~ *bəh-): (vb.) *bah- ‘to shine’; (n.) *bah-a ‘brilliance, brightness, splendor, beauty; light’; (adj.) ‘shining, bright, radiant’.] Hittite (1st sg. pres. act.) pa-aḫ-ḫa-aš-ḫi, (1st sg. pres. mid.) pa-aḫ-ḫa-aš-ḫa ‘to protect, to guard, to defend; to observe (agreements), to keep (oaths), to obey (commands), to keep (a secret)’ (< *pºeH₂s- [*peH₂s-]); (with medial single writing) (1st sg. pres. act.) pa-aḫ-ša-nu-mi (causative); (?) Cuneiform Luwian (3rd sg. imptv.) pa-ad-du ‘to protect’ (meaning uncertain), without a laryngeal ~ Sanskrit (Vedic) pā́ ti ‘to protect, to preserve, to keep’; Tocharian A pās-, B pāsk- ‘to guard, to protect; to practice (moral behavior), to obey (rules)’, B -pāṣṣe ‘behavior’. Notes: (1) The Anatolian forms are also commonly compared with the following: Latin pāscō ‘to cause to eat, to feed, to pasture, to drive to pasture’; Old Church Slavic pasti ‘to pasture, to feed, to herd’; Serbo-Croatian pȁsti ‘to pasture, to look after’. (2) Kloekhorst points out that the form (1st pres. act.) pa-aḫ-ḫa-aš-mi occurs only once. He assumes that the active verb originally belonged to the ḫi-conjugation. [Cf. Bomhard 2023.2:106—107, no. 92, Proto-Nostratic root *pºaħ- (~ *pºəħ-): (vb.) *pºaħ- ‘to eat’; (n.) *pºaħ-a ‘food, nourishment’.] Anatolian and the Laryngeal Theory 31 Hittite (3rd sg. pres. act.) la-a-ḫu-i, la-a-ḫu-u-i, la-ḫu-i, la-a-ḫu-wa-i, etc.; also la-aḫḫu-uz-zi, la-ḫu-uz-zi, etc. ‘(tr.) to pour, to cast (objects from metal); (intr.) ‘to (over)flow’; (nom. sg.) la-aḫ-ḫu-uš ‘containter’, (instr. sg.) la-aḫ-ḫu-e-eš-ni-it ‘pouring cup’ (< *leH₂-w/u-/*loH₂-w/u-); Cuneiform Luwian (1st sg. pret. act.) laḫu-ni-i-ḫa ‘to wash away’; (without ḫ) (part.) la-a-ú-na-i-mi-iš(), la-ú-na-i-[mi-š()] ‘poured’, (3rd pl. pret.) lu-u-wa-an-da ‘to pour’, (2nd sg. impv.) li-lu-u-wa(-a) ‘pour!’ ~ Greek ληνός (Doric λᾱνός) ‘anything shaped like a tub or a trough: a wine-vat, a trough (for watering cattle), a watering place’ (< *lā-no-s < *leH₂-nos). Note: The Anatolian forms are not related to Greek λούω ‘to wash, to bathe’, Latin lavō ‘to wash, to bathe’, etc., which must be derived from Proto-IndoEuropean *lewH₂-/*lowH₂- ‘to wash, to bathe’ (cf. Kloekhorst 2008a:512—513). [Cf. Bomhard 2023.3:1173, no. 965, Proto-Nostratic root *laħ- (~ *ləħ-): (vb.) *laħ- ‘to make flow, to pour, to moisten, to wet’; (n.) *laħ-a ‘flowing, pouring; moistness, wetness’.) Hittite (nom. sg.) ḫa-at-ta-an-za (< *ḫakt-ant-) ‘intelligent, clever, wise’, ḫattaḫḫ- ‘to make clever, to instruct’, (nom. sg.) ḫa-at-ta-a-tar ‘intelligence, (wise) counsel, wisdom’ (< *H₂ekº-tº- [*H₂ek-t-]) ~ Gothic aha ‘mind, understanding’, ahjan ‘to think’, ahma ‘spirit’, *ahmateins ‘inspiration’, *ahmeins ‘spiritual’; Old Icelandic Ktla (< *aχtilōn) ‘to think, to mean, to suppose’, Ktlan ‘thought, meaning, opinion’; Old English eaht ‘council, deliberation, consideration’, eahtian ‘to watch over, to hold council, to deliberate, to consider’; Old Frisian achte ‘consideration’, achtia ‘to consider’; Old High German ahta ‘consideration’ (New High German Acht), ahtōn ‘to consider’ (New High German achten). Notes: This etymology is taken from Puhvel 1984— .3:260—263. [Cf. Bomhard 2023.3:855—856, no. 721, ProtoNostratic root *ħakº- (~ *ħəkº-): (vb.) *ħakº- ‘to be mentally sharp, keen’; (n.) *ħakº-a ‘wisdom, sound judgment, understanding’.] Hittite (gen. sg.) ḫal-lu-wa-aš ‘hollow, pit’, (gen. sg.) ḫal-lu-u-wa-aš ‘hollow, deep’, (denominative verb, 3rd sg. pret. act.) ḫal-lu-wa-nu-ut ‘to put down (deep), to lower, to let deteriorate’ (< *H₂el-wo-) ~ Latin alvus ‘belly, womb’, alveus ‘a hollow, cavity’. Note: This etymology is taken from Puhvel 1984— .3:47—49. [Cf. Bomhard 2023.3:863, no. 728, Proto-Nostratic (n.) *ħalʸ-a ‘hole, hollow, cavity’.] Hittite (reduplicated) (1st sg. pres. act.) [ḫ]a-ma-an-ga-aḫ-ḫi, ḫa-ma-an-ga-mi ‘to tie, to bind, to betroth’ (< *ḫam-ang- < *ḫan-ang-, through dissimilation [cf. Greek ἀνάγκη ‘force, constraint’, ἀναγκάζω ‘to force, to compel; to constrain’]) (< *H₂engº-) ~ Sanskrit aṁhú-ḥ ‘narrow’; Greek ἄγχω ‘to compress, to press tight; to strangle’; Latin angō ‘to press tightly; to strangle, to throttle; to hurt, to distress’, angor ‘mental distress, anguish, trouble’; Gothic aggwus ‘narrow’; Old Icelandic öngr ‘narrow’; Old English enge ‘narrow; causing anxiety, painful, severe’; Old Saxon engi ‘narrow’; Dutch eng ‘narrow’; Old High German angi, engi ‘narrow’ (New High German eng ‘narrow, cramped, tight, confined’); Old Church Slavic o˛zъ-kъ ‘narrow’; Lithuanian añkštas ‘narrow, cramped, tight’. [Cf. Bomhard 2023. 32 Chapter 3 3:869—871, no. 733, Proto-Nostratic root *ħan- (~ *ħən-): extended form: (vb.) *ħan-V-g- ‘to tie tightly, to constrict, to make narrow; to choke, to strangle’; (n.) *ħan-g-a ‘throat’; (adj.) ‘narrow, constricted’.] Hittite (n.) (nom. sg.) ḫa-ap-pí-na-az ‘wealth’; (adj.) (nom. sg.) ḫa-ap-pí-na-an-za ‘wealthy, rich’ (< *H₂opº-en-o- [*H₃ep-en-o-]); Cuneiform Luwian ḫa-ap-pí-naat-ta-an-za ‘wealth, riches’ ~ Sanskrit ápnas- ‘possession, property’; Avestan afnah-vant- ‘rich in possessions’; Latin opulens (< *open-ont-) ‘rich, wealthy’, ops ‘might, power’, opēs ‘resources, means, wealth’, Ops ‘the goddess of abundance’. Note: Proto-Indo-European *H₂opº- probably originally meant ‘to gather, to collect’, specifically, ‘to gather wealth’. [Cf. Bomhard 2023.3:872—874, no. 735, Proto-Nostratic root *ħapº- (~ *ħəpº-): (vb.) *ħapº- ‘to take, gather, or collect (with the hands or arms)’; (n.) *ħapº-a ‘that which has been gathered or collected: plenty, fullness, abundance, wealth, possessions, property; embrace, armful, handful’.] Hittite (3rd sg. pres. act.) ḫar-aš-zi ‘to till (the soil)’, (nom.-acc. sg.) ḫar-ša-u-wa-ar, ḫar-ša-a-u-ar “tilled land’ (< *H₂er(H)-) ~ Greek ἀρόω ‘to plow’; Latin arō ‘to plow’; Old Irish airim ‘to plow’; Gothic arjan ‘to plow’; Old Icelandic erja ‘to plow’; Old English erian ‘to plow’, ierþ ‘plowing’; Old High German erran ‘to plow’; Lithuanian ariù, árti ‘to plow, to till’; Old Church Slavic ralu ‘a plow’, orjǫ, orati ‘to plow’; Tocharian A āre ‘a plow’. Note: Kloekhorst (2008a:312—314) reconstructs Proto-Indo-European *h₂órh₃-s-ei / *h₂rh₃-s-énti. [Cf. Bomhard 2023. 3:879—880, no. 740, Proto-Nostratic root *ħar- (~ *ħər-): (vb.) *ħar- ‘to scratch, to scrape’ (> ‘to plow’ in the daughter languages); (n.) *ħar-a ‘scraping, scratching’.] Hittite (acc. sg.) ḫa-aš-ša-an ‘hearth’, (nom. sg.) ḫa-a-aš, ḫa-aš-ša-aš ‘ash(es); soda ash, potash, soap’ (< *H₂es-) ~ (?) Greek ἄζω ‘to be dry’; Latin āra ‘altar’ (Old Latin āsa); Umbrian (dat. sg.) ase ‘altar’; Gothic azgō ‘cinder, ashes’; Old Icelandic aska ‘ashes’; Swedish aska ‘ashes’; Danish aske ‘ashes’; Old English asce, Ksce ‘ashes’; Dutch asch ‘ashes’; Old High German asca ‘ashes’ (New High German Asche); Czech ozd ‘parched malt’, ozdíti ‘to dry malt’; Tocharian B ās- ‘to become dry, to dry out, to dry up, to parch’, asāre ‘dry’. Notes: (1) Kloekhorst (2008a: 318—319 and 322—323) reconstructs Proto-Indo-European *h₂éh₁-s- to account for the long initial ā in Latin (and Hittite), while acknowledging that a short initial ă is found in the Germanic cognates. However, Lindeman (1997:57) points out that lengthened-grade is also possible (*H₂ēs- [phonetically *H₂ās-]). (2) Boisacq (1950: 16) and Frisk (1970—1973.I:25—26) derive Greek ἄζω ‘to be dry’ from *azd-, extended form of *ā̆s-, while Beekes (2010.I:26—27), among others, derives it from *h₂ed-. [Cf. Bomhard 2023.3:881—882, no. 742, Proto-Nostratic root *ħas(~ *ħəs-): (vb.) *ħas- ‘to burn, to be hot’; (n.) *ħas-a ‘cinder, ember, ashes; heat’.] Hittite (nom. sg.) (GIŠ)ḫa-aš-ši-ka₄-aš, (GIŠ)ḫa-ši-ik-ka₄-aš ‘a tree and its fruit’ (< *H₂es-, *H₂ō̆s-) ~ Greek ὀξύη (< *ὀσκ[ε]σ- ?) ‘a kind of beech-tree’; Armenian hac ̣i ‘ash-tree’; Albanian ah (< *oskā) ‘beech-tree’, ashe ‘holly’; Ligurian ‘Οσκίλα Anatolian and the Laryngeal Theory 33 ‘ash forest’; Latin ornus (< *os-en-os) ‘mountain-ash’; Old Irish (h)uinn-ius ‘ashtree’; Welsh onn-en ‘ash-tree’; Breton ounn-enn ‘ash-tree’; Old Icelandic askr ‘ashtree’, eski ‘ashen box’; Swedish ask ‘ash-tree’; Old English Ksc ‘ash-tree’; North Frisian esk ‘ash-tree’; Dutch esch ‘ash-tree’; Old High German ask ‘ash-tree’ (New High German Esche); Old Prussian woasis ‘ash-tree’; Lithuanian úosis (< *ōs-) ‘ash-tree’; Russian jásenʹ [ясень] ‘ash-tree’. [Cf. Bomhard 2023.3:882—883, no. 743, Proto-Nostratic root *ħasʸ- (~ *ħəsʸ-) (used as the base to designate various tree names): (n.) *ħasʸ-a ‘a tree and its fruit’.] Hittite (nom. sg.) ḫa-tu-ga-aš ‘terrible, baleful, fearsome, awesome’, (nom.-acc. sg.) ḫa-tu-ga-tar ‘terror, awesomeness’, (3rd sg. pres. act.) ḫa-tu-ki-iš-zi ‘to become terrible’, (3rd pl. pres. act.) (?) ḫa-tu-ga-nu-wa-an-[zi] ‘to terrify’ (< *H₂(e)t’-, *H₂(e)t’- [*H₂(e)d-, *H₂(e)d-]) ~ Greek ὀδύσσομαι ‘to be wroth against, to be angry with, to hate’, ʼΟδυσσεύς ‘Ulysses, Odysseus’ (< ‘Fearsome’); Latin ōdī ‘to hate’, ŏdium ‘hatred, grudge, ill will, animosity, enmity, aversion’, odiōsus ‘hateful, odious, vexatious, offensive, unpleasant, disagreeable, annoying, troublesome’; Armenian ateam ‘to hate’, ateli ‘hated, hostile’; Crimean Gothic atochta ‘bad’; Old Icelandic atall ‘fierce’; Old English atol ‘terrible, dire, loathsome, horrid’; Breton œz ‘horror’, œzi ‘to be terrified’. Note: Kloekhorst (2008a:336—337) compares Greek ἀτύζομαι ‘to be distraught from fear, bewildered; to be distraught with grief; to be amazed at; to strike with terror or amazement’ instead. Beekes (2010.1:167) supports Kloekhorst’s etymology. However, as noted by Kloekhorst, -t- is never written double in Hittite. In my opinion, this speaks in favor of the alternative etymology given above and supported by Puhvel (1984— .3:274—277) and suggests that Greek ἀτύζομαι may be a later borrowing and not inherited from Proto-Indo-European. [Cf. Bomhard 2023.3:883—884, no. 744, Proto-Nostratic root *ħat’- (~ *ħət’-): (v.) (vb.) *ħat’- ‘to shake, to tremble; to be shaken, startled, frightened, terrified, afraid’; (n.) (n.) *ħat’-a ‘trembling, shaking’.] Hittite (3rd sg. pres.) ḫi-in-ik-zi, ḫi-in-ga-zi, ḫi-ni-ik-zi ‘to present, to deliver, to offer, to allot’ (< *H₂inkº- [*H₂nek̑ -]) ~ Sanskrit aśnóti ‘to reach, to come to, to arrive at, to get, to obtain; to master, to become master of; to offer’; Latin nancior ‘to get, to obtain’, nanciscor ‘to get, to gain, to receive, to meet’; Tocharian A ents-, B enk‘to seize, to take’, B enkalñe ‘grasping or clinging to existence; assumption, taking to oneself’, B enkäl ‘feeling, passion’. The following may belong here as well: Hittite (nom.-acc. sg.) ḫe-en-ka-an, ḫi-in-kán, ḫi-in-ga-an, ḫe-e-en-kán ‘death, doom, deadly disease, plague’. Note: I assume that, not only did *H₂ lower a contiguous *e to *a in Proto-Indo-European, it also lowered a contiguous *i to *e and a contiguous *u to *o (cf. Bomhard 2023.1:75—77). This explains examples of ḫe- and -eḫ(ḫ)- in Hittite, where ḫ < *H₂. It appears that these changes were still in progress at the time when the Anatolian branch separated from the main speech community, as evident in the fluctuation between ḫi- and -iḫ(ḫ)- and ḫe- and -eḫ(ḫ)in Hittite (cf. Kloekhorst 2008a:339—340: “Already in the oldest texts (OS and OH/MS) we find spellings ḫi-in-k°, ḫe-en-k°, ḫé-en-k° besides each other”; Puhvel 1984— .3:296—300). The contrary theory, according to which e became i in this 34 Chapter 3 environment, is phonetically improbable, regardless of what may have happened elsewhere. An important point needs to be made here: *i and *u had more than one origin in Proto-Indo-European. In some cases, *i and *u were original (that is to say, they were inherited from Proto-Nostratic), while, in other cases, they resulted from the stress-conditioned weakening of *ey/*oy (or *ye/*yo) and *ew/*ow (or *we/*wo), respectively. Only original *i and *u were lowered to *e and *o, respectively, when contiguous with *H₂. When *i and *u resulted from the stressconditioned weakening of *ey/*oy (or *ye/*yo) and *ew/*ow (or *we/*wo), however, they were not lowered to *e and *o, respectively, under the influence of *H₂, since such a change would have disrupted the integrity of the ablaut relationship. On the other hand, it is possible to envision a scenario in which *H₂ originally did have an assimilatory effect on *i and *u resulting from the stressconditioned weakening of *ey/*oy (or *ye/*yo) and *ew/*ow (or *we/*wo) as well, but where *i and *u were later analogically restored. No doubt, we are dealing with chronologically distinct developments here, with ablaut being older. [Cf. Bomhard 2023.3:894—895, no. 752, Proto-Nostratic root *ħin- (~ *ħen-): extended form: (vb.) (vb.) *ħin-V-kº- ‘to reach, to come to, to arrive at, to gain; to offer, to present’; (n.) *ħin-kº-a ‘gain, mastery, experience; offering, present’.] Hittite (nom.-sg.) ḫa-an-za ‘front, front part’, (nom. sg.) ḫa-an-te-iz-zi-iš ‘forward, front, first; first-born, earliest; foremost’, ḫa-an-ti ‘in front of, before’ (< *H₂entº[*H₂ent-]); Cuneiform Luwian (nom. sg.) ḫa-an-te-le-eš ‘first, foremost’, (acc. sg.) ḫa-an-da-wa-te-en ‘leader, chief’; Hieroglyphic Luwian hant- ‘face, forehead’, hantil(i)- ‘first, former’, hanti ‘against’; Lycian (3rd sg. pret. act.) χñtawate, χñtewete ‘to lead, to direct, to rule’, χñtawata- ‘leader, chief’ ~ Sanskrit ánti ‘before’, ántya-ḥ ‘last (in time, place, or order)’, ánta-ḥ ‘end, limit, boundary’; Greek ἀντί ‘opposite’, ἄντα ‘over, against, face to face’; Latin ante ‘before’; Oscan ant ‘till’; Gothic and ‘along, throughout, towards’, andeis ‘end’; Old Icelandic (prefix) and- ‘opposite, against, towards’, endi, endir ‘end’, endr ‘in times past, formerly’, enda ‘to end, to bring to an end’; Old English (prefix) and-, ond‘opposite, against, towards’, ende ‘end, limit, border’; Old Frisian enda ‘end’; Old Saxon (prefix) and-, ant- ‘opposite, against, towards’, endi ‘end’; Dutch einde ‘end’; Old High German (prefix) ant-, int-, ent- ‘opposite, against, towards’ (New High German ant-, ent-), anti, enti ‘end’ (New High German Ende); Lithuanian añt (earlier antà) ‘on, upon’; Tocharian A ānt, B ānte ‘surface, forehead’. [Cf. Bomhard 2023.3:938—940, no. 791, Proto-Nostratic (n.) *xaŋ-tº-a ‘the most prominent or foremost (person or thing), front, front part’.] Hittite (3rd sg. pres.) ḫu-ul-la-a-i ‘to smite, to destroy’, (ptc.) ḫu-ul-ḫu-li-ya-an-te-eš ‘smitten’, ḫu-ul-la-an-za-iš ‘battle’ (< *H₂ul-) ~ Greek ὄλλῡμι ‘to destroy, to make an end of’, ὄλεθρος ‘ruin, destruction, death’; Latin ab-oleō ‘to destroy’. Note: So far as I can determine, this etymology was first suggested by Couvreur (1937: 144—146), but it was subsequently rejected my most other scholars on the basis of the difference between the stem vowels of Hittite, on the one hand, and Greek and Latin, on the other hand — Cowgill (1965:146—147 and 157), for one, accepts the Anatolian and the Laryngeal Theory 35 comparison of the Hittite and Greek and Latin forms. However, this comparison can be revived if we consider the original form to have been *H₂ul-, which later became *H₂ol-, with *-o- from earlier *-u- under the influence of the preceding laryngeal. [Cf. Bomhard 2023.1:76 and 3:900—901, no. 757, Proto-Nostratic root *ħul- (~ *ħol-): (vb.) *ħul- ‘to destroy, to lay waste, to cause to perish’; (n.) *ħula ‘ruin, destruction; end, death’.] Hittite (3rd sg. pres.) ḫa-at-zi, ḫa-at-ta-i, ḫa-at-ta-a-i, ḫa-ad-da-i; ḫa-az-zi-zi, ḫa-az-ziaz-zi ‘to make a hole (in), to pierce, to prick, to stab, to slash, to perforate, to penetrate, to stick (as a means of killing), to hit (a target), to strike (especially a musical instrument), to engrave (a tablet)’, (1st sg. pres.) ḫa-at-ta-ra-a-mi ‘to prick, to incise, to engrave, to inscribe’, (nom.-acc. sg.) ḫa-at-tal-la-an ‘club, mace’, (nom.-acc. sg.) ḫa-at-ta-ra-a[n] ‘prick, awl’, (nom. sg.) ḫa-at-tal-ki-iš-na-aš ‘thorn-bush’ (< *H₂et’-H- [*H₂ed-H-]); (3rd sg. pres.) ḫa-at-ra-a-iz-zi ‘to write, to send written word (about), to report, to order, to dispatch’, (nom.-acc. sg.) ḫa-at-rieš-šar ‘written message, decree’ (< *H₂et’-ro- [*H₂ed-ro-]); Cuneiform Luwian (3rd sg. pret.) ḫa-at-ta-ri-it-ta ‘to prick, to pierce’, (acc. sg.) ḫa-at-ta-ra-an ‘prick’; Hieroglyphic Luwian ha-tu+ra/i-à-s ‘letter’, (imptv.) ha-tu+ra/i+à ‘write!’; Lycian χttadi ‘to hurt, to damage’, χdrñna (?) ‘inscription’ (?) ~ Armenian hatanem ‘to cut’, hat ‘piece, cut, slice’; Avestan aδu ‘water-course, brook, canal’. Note: Though the comparison of Armenian hatanem with the Anatolian forms is semantically flawless, there are problems with the phonology, since double writing of the dental stop in Hittite points to original *-tº- [*-t-], while the Armenian form points to original *-t’- [*-d-]. However, double writing of medial stops in Hittite can also indicate the former presence of a laryngeal as in (nom. sg.) me-ik-ki-iš ‘large’, which is to be derived from earlier *mek’- plus the suffix *-Hi- > *mek’Hi> Hittite me-ik-ki-iš. Thus, comparison of Armenian hatanem with the Anatolian forms having medial double writing can be maintained if we derive the Anatolian forms from earlier *H₂et’-H-, which would yield Hittite ḫatta- as the regular outcome. Support for this interpretation may be found in Hittite ḫatrai-, which has consistent single writing. Thus, it is possible to envision a Pre-Anatolian root *H₂et’-, which was then extended by two separate suffixes in Proto-Anatolian proper: (A) *H₂et’-Ho-, yielding Hittite ḫatta- upon loss of the medial laryngeal, and (B) *H₂et’-ro-, yielding Hittite ḫatra-. Stem (A) was further extended by a suffix -ra-, giving the attested agent noun ḫattara- ‘prick, awl’, which, in turn, served as the basis of the denominative verb ḫattarai-. Other derivatives of stem (A) are ḫattatta- ‘club, mace’ and ḫattalkešna- ‘thorn-bush’. The agent noun *ḫatra-, from stem (B) and from which the denominative verb ḫatrai- is derived, is unattested in Hittite. [Cf. Bomhard 2023.3:940—942, no. 792, Proto-Nostratic root *xat’- (~ *xət’-): (vb.) *xat’- ‘to cut into, to hollow out, to engrave, to prick, to pierce’; (n.) *xat’-a ‘slice, carving, engraving, engraved line, incision’.] Hittite (nom. sg.) ḫu-u-ma-an-za ‘all, whole’ (< *H₂um-) ~ Latin omnis ‘all, every, whole’ (cf. Couvreur 1937:144—146; Kronasser 1956:41; Pedersen 1938:165). Note: Both Polomé (1965:18) and Puhvel (1984— .3:380) reject this etymology — 36 Chapter 3 Puhvel derives Latin omnis from *opnis. On the other hand, Walde—Hofmann (1965—1972.II:209—210) mention Oscan úmbn, which points to earlier *omb-nis and not *opnis as the source of both the Oscan form and Latin omnis. *omb-nis may contain an epenthetic b, in which case the original form would have been *omni-s. Here, -ni- is a suffix. Likewise, in Hittite ḫu-u-ma-an-za, the stem is *ḫum-, and the -anz(a) is a suffix. Thus, this comparison can be revived if we consider the original form to have been *H₂um-, which later became *H₂om-, with *-o- from earlier *-u- under the influence of the preceding laryngeal. Such an explanation overcomes the objections raised against this etymology based upon the irregular correspondence of Hittite u and Latin o. [Cf. Bomhard 2023.1:76.] Hittite (1st sg. pres. act.) na-aḫ-mi, (3rd sg. pres. act.) na-aḫ-ša-ri-ya-az-zi, (1st sg. pret. act.) na-aḫ-ḫu-un, na-a-ḫu-un ‘to fear, to be or become afraid; to be respectful, to be careful’, (nom. sg.) na-aḫ-ša-ra-az ‘fear, fright; respect, reverence, awe; frightfulness’ (< *neH₂-); Cuneiform Luwian (nom. pl.) na-aḫ-ḫu-wa-aš-ši-en<zi> ‘fearful’ or ‘fearsome’ (?), (3rd sg. pret. act.) na-aḫ-ḫu-u-wa-i ‘to be afraid, worried’ (impersonal) ~ Old Irish nár (< *nāsros) ‘modest, bashful’. [Cf. Bomhard 2023.3:1143—1144, no. 941, Proto-Nostratic root *naħ- (~ *nəħ-): (vb.) *naħ- ‘to tremble, to shake; to fear, to be afraid’; (n.) *naħ-a ‘fear’.] Hittite (3rd sg. pres. act.) ḫu-iš-zi ‘to live, to survive’ (< *H₂w-es-); Hieroglyphic Luwian (nom.-acc. sg.) BESTIAHWI-sà+ra/i-sa, BESTIAHWI-sa5+ra/i /hwisar/ ‘game, wild beasts’ ~ Sanskrit vásati ‘to dwell, to live, to inhabit; to dwell in, to abide in; to dwell or live near’; Greek (aor.) ἄεσα ‘to spend the night’; Middle Irish fóaid ‘to pass the night, to dwell’; Gothic wisan ‘to be, to remain’; Old Icelandic vesa ‘to be’; Old English wesan ‘to be, to happen’; Old High German wesan ‘to be’; Tocharian B wäs- ‘to dwell, to abide, to remain, to lie (on)’. Note: Curiously, Cuneiform Luwian has (nom.-acc. sg.) ḫu-u-i-tar-ša ‘wild animal’. Hittite (nom.-acc. sg.) me-e-ḫu-ur, me-e-ḫur, me-ḫur ‘time’ (< *meH₂- < *miH₂-) ~ Sanskrit māti-ḥ ‘measure, accurate knowledge’, mā́ ti, mímāti ‘to measure, to mete out, to mark off’; Latin mētior ‘to measure’; Gothic mēl ‘time’; Old Icelandic mál ‘measure; time, high time; meal’; Old English mbþ ‘measure, degree, proportion’, mbl ‘measure; (appointed) time, occasion; time for eating, meal’; Old Frisian mēl ‘time, mealtime’; Dutch maal ‘(n.) meal; (m.) time’; Old High German māl ‘time’ (New High German Mal). Note: In spite of consistent single writing of ḫ in Hittite, the laryngeal involved here is *H₂, as reconstructed, for example, by Kloekhorst (2008a:567—568). I consider this to be another example of the change of original *i to *e under the influence of *H₂. Puhvel (1984— .6:108—112) sardonically notes: The enormous, aporia-studded amount of attention expended on the etymology of mehur (see the account by Tischler, Glossar L—M 171—4; cf. Neu, IBS 52:184 [1987]) is a prime example of preconceptions and “theory” dragooning and hamstringing data. Derivation from IE *mē- has been around since Hrozný (SH 70) but Anatolian and the Laryngeal Theory 37 has typically stumbled on the doctrines of “trilaryngealism” (e-vocalism incompatible with Hittite h), so that Rieken (Stammbildung 340) could still claim in 1999 that “all attempts to connect mēhur with IE *mē- founder thereon.” Puhvel supports derivation from (traditional) *mē- ‘to measure’, as do I. [Cf. Bomhard 2023.3:1101—1102, no. 913, Proto-Nostratic root *miħ- (~ *meħ-): (vb.) *miħ- ‘to measure, to mark off’; (n.) *miħ-a ‘measure, measurement’.] Hittite (3rd sg. pres. act.) ta-ru-uḫ-zi, tar-uḫ-zi, tar-ru-uḫ-zi, tar-ḫu-uz-zi, etc. ‘to prevail, to conquer, to be powerful, to be able’, (nom.-acc. sg.) tar-ḫu-u-i-li ‘strong, powerful’, *dTarḫunna- name of the Storm God (< *tºerH₂-w/u- [*terH₂-w/u-]); Cuneiform Luwian dTarḫunt- / dTarḫuwant- name of the Storm God; Hieroglyphic Luwian Tarhunt-, Tarhunza- name of the Storm God; Lycian / Milyan Trqqñtname of the Storm God; Lydian (?) tarvτalli- ‘of Tarvra’ (nom. sg. tarvτallis); Carian trq(u)δ- name of the Storm God ~ Sanskrit tū́ rvati ‘to overpower, to overcome, to excel’. Note: Kloekhorst (2008a:838) derives Lycian Trqqñt- first from Proto-Anatolian *trHwent- but later, on the same page, from (Proto-IndoEuropean) *trh2uent-. Hittite (nom. sg.) ḫa-a-ra-aš, (gen. sg.) ḫa-ra-na-aš) ‘eagle’; Palaic ḫa-ra-a-aš ‘eagle’ (< *H₂or-/*H₂r̥ -) ~ Greek ὄρνις ‘bird’; Armenian oror ‘kite, gull’; Welsh eryr ‘eagle’; Gothic ara ‘eagle’; Old Icelandic (poet.) ari, örn (< *arnu-) (gen. sg. arnar, acc. örnu, pl. ernir) ‘eagle’; Old English earn ‘eagle’ (Middle English ern(e), earn); Old High German aro, arn ‘eagle’ (New High German [poetic] Aar); Lithuanian erẽlis (dial. arẽlis) ‘eagle’; Latvian èrglis ‘eagle’; Old Prussian arelie ‘eagle’; Old Church Slavic orьlъ ‘eagle’; Russian orël [орëл] ‘eagle’; Czech orel ‘eagle’; Polish orzeł ‘eagle’; Upper Sorbian worjoł ‘eagle’; Lower Sorbian jerjoł, jerjeł ‘eagle’; Bulgarian orél ‘eagle’; Serbo-Croatian órao ‘eagle’. Notes: (1) Pokorny (1959: 325—326) reconstructs Proto-Indo-European *er- on the basis of Lithuanian erẽlis, but Cowgill (1965:146, fn. 2) questions the validity of this reconstruction since he takes Lithuanian erẽlis to be assimilated from the dialectal form arẽlis. Cowgill points out that the relative antiquity of the Lithuanian dialectal form is confirmed by Old Prussian arelie. Finally, he points out that Latvian èrglis has undergone even more remodeling. (2) This is another example of the change of *u to *o under the influence of *H₂. [Bomhard 2023.3:901—902, no. 758, Proto-Nostratic (n.) *ħur-a (and/or *ħer-a ?) ‘hawk-like bird: falcon, hawk, eagle, kite’.] 3. *H₃ > (1) ḫ- (initially), -ḫ- (medially) in Hittite, Palaic, Cuneiform Luwian, and Hieroglyphic Luwian (written h-, -h-); > (2) χ- (initially), -χ-/-g- (medially), -q- (< */ɣ¦/) in Lycian: As noted by Kimball (1987), the reconstruction of *H₃ is one of the most challenging problems in Indo-European comparative linguistics, due to the ambiguity of the available evidence (see also Melchert 1994a:71, §4.1.3.3). I agree with Melchert’s (1994a:72) statement: “I share the view of Normier (1980a: 58), Watkins 38 Chapter 3 (1982c: 457), Bernabé (1983: 39ff), Kimball (1983 & 1987a) and others that */h3/ is preserved initially as ḫ- in Hittite, Palaic and Cuneiform Luvian.” Beyond that, scholars differ greatly in their opinions regarding which words are to be reconstruct with *H₃. The one thing that seems certain, though, is that *H₂ and *H₃ were phonetically distinct. *H₃ is usually interpreted as the voiced counterpart of *H₂. I have only included a few examples below — those that seem certain to me based mostly upon my work on distant linguistic relationship. Hittite (nom. pl. ?) ḫa-a-u-e-eš ‘sheep’ (< *H₃owi-s); Hieroglyphic Luwian (nom. sg.) OVIS.ANIMAL há-wá/í-i-sá /hawis/ ‘sheep’; Cuneiform Luwian (nom. sg.) ḫa-a-ú-i-iš ‘sheep’; Lycian (acc. sg.) χawã ‘sheep’ ~ Sanskrit ávi-ḥ ‘sheep’; Greek ὄɩ̈ ς, οἶς ‘sheep’; Latin ovis ‘sheep’; Armenian hov-iw ‘shepherd’; Old Irish oí ‘sheep’; Gothic awēþi ‘herd of sheep’; Old English ēow, ēaw, ēw ‘sheep’, ēowu, ēowe ‘ewe’, ēowd, ēowde ‘herd of sheep’; Old Frisian ei ‘ewe’; Old Saxon ewwi ‘ewe’; Dutch ooi ‘ewe’; Old High German ouwi, ou ‘ewe’, ewit, owiti ‘herd of sheep’; Lithuanian avìs ‘sheep’; Latvian avs ‘sheep’; Old Church Slavic ovьca (< *owi-kā) ‘sheep’; Tocharian B eye ‘sheep’, ā(u)w ‘ewe’, aiyye ‘ovine, pertaining to sheep’. Notes: (1) Kimball (1999:142) reconstructs initial *h2- here, but this interpretation is rejected by Kortlandt (2001:2). Kloekhorst (2008a:337—338) reconstructs *h3eu̯ i- with initial *H₃- as do Beekes (2010.II:1060—1061) and Derksen (2008:384 and 2015:74). (2) In my opinion, the *-o- in the initial syllable is original here, that is to say, it is inherited from Proto-Nostratic. (3) Lycian (acc. sg.) χavã ‘sheep’ shows that initial *H₃- became χ- in Lycian. However, according to Melchert (1994a:72) and Kimball (1987), initial *H₃- was lost in Lycian. [Cf. Bomhard 2023.3:930— 931, no. 785, Proto-Nostratic (n.) *ʕuw-a (~ *ʕow-a) ‘herd of small animals, sheep and goats’.] Hittite (nom.-acc. sg.) ḫa-aš-ta-a-i, ḫa-aš-ta-i, ḫa-aš-da-i, [ḫ]a-aš-da-a-i ‘bone(s)’ (< *H₃ostº- [*H3ost-]); Cuneiform Luwian (nom.-acc. sg.) ḫa-a-aš-ša, ḫa-aš-ša ‘bone’ ~ Sanskrit ásthi, (gen. sg.) asthnáḥ ‘bone’; Greek ὀστέον ‘bone’; Latin os ‘bone’; Albanian asht, ashti ‘bone’. Notes: (1) The following is also found in Hittite: (nom.acc. sg.) É ḥé-eš-ta-a, (gen. sg.) É ḥi-iš-ta-a-aš, É ḥi-iš-ta-aš, É ḥé-eš-ta-a-aš, É ḥé-eš-ta-aš ‘mausoleum ?’. Most scholars connect this form with ḫa-aš-ta-a-i ‘bone(s)’ as ‘bone-house’ > ‘sepulcher, mausoleum’, but Kloekhorst (2008a:346— 347) prefers to see it as a borrowing (“foreignism”). Kloekhorst does not identify the source of the borrowing, but, given the fact that the paradigm “hardly shows any inflected forms” and given the specialized cultic/ritualistic meaning, I am inclined to accept Kloekhorst’s views. (2) The Sanskrit form requires a laryngeal suffix to account for the aspiration: ásthi < *H₃ostº-H- [*H3ost-H-]. Hittite (3rd sg. pres. act.) ḫar-ap-zi ‘to separate oneself and (re)associate oneself elsewhere’ (< *H₃or-bº-) ~ Sanskrit árbha-ḥ ‘little, small; child’; Armenian orb ‘orphan’; Greek ὀρφανός ‘orphan, without parents, fatherless; (metaph.) abandoned, bereft’; Latin orbus ‘bereft, deprived by death of a relative or other dear one; bereaved (of); childless; an orphan’; Old Irish orb ‘heir’, orb(b)e, orpe Anatolian and the Laryngeal Theory 39 ‘inheritance’; Gothic arbi ‘inheritance’, arbja ‘heir’ (f. arbjō ‘heiress’); Old Icelandic arfi ‘heir, heiress’, arfr ‘inheritance, patrimony’, erfa ‘to inherit’, erfð ‘inheritance’; Old Swedish arve, arver ‘heir’; Danish arv ‘heir’; Norwegian arv ‘heir’; Old English ierfa, irfa ‘heir’, ierfe ‘inheritance, bequest, property’, erfe, irfe, yrfe ‘inheritance, (inerited) property’, irfan, yrfan ‘to inherit’; Old Frisian erva ‘heir’, erve ‘inheritance, inherited land, landed property’; Old Saxon erƀi ‘inheritance’; Middle Dutch erve ‘heir’; Old High German arbi, erbi ‘inheritance’, arbeo, erbo ‘heir’ (New High German Erbe ‘inheritance; heir’); Old Church Slavic rabъ ‘servant, slave’; Russian rab [раб] ‘slave, serf, bondsman’ (f. rabá [раба] ‘slave, serf, bondmaid’). [Cf. Bomhard 2023.3:950—952 no. 801, Proto-Nostratic root *¦or-: (vb.) *¦or- ‘to leave, to go away, to depart, to separate, to abandon’; (n.) *¦or-a ‘leaving, departure; separation; abandonment’; extended form: (vb.) *¦orV-b- ‘to leave, to go away, to depart, to separate, to abandon’; (n.) *¦or-b-a ‘leaving, departure; separation; abandonment’.] Hittite (nom.-acc. sg.) še-e-ḫur, še-e-ḫu-ur, še-e-ḫu-wa-ar ‘urine’, (3rd sg. pret. act.) še-e-ḫu-ri-ya-[a]t ‘to urinate’, (nom. sg.) še-e-ḫu-ga-ni-ya-u-wa-an-za ‘besmeared with urine’ (< *seH3-ur < *siH3-ur). Notes: (1) There are no known cognates, either in the other Anatolian languages or in the non-Anatolian Indo-European daughter languages. (2) Later in this chapter, I am going to explain why I believe that it is extremely probable that *H₂ and *H3 had exactly the same vowel-coloring effects, in this case, *i > *e under the influence of *H3. (3) Given the ambiguities involved, derivation of Hittite (nom.-acc. sg.) še-e-ḫur, še-e-ḫu-ur, še-e-ḫu-wa-ar ‘urine’ from *seH2-ur (< *siH2-ur), with *H2 instead of *H3, is also a possibility. (4) As observed by Kloekhorst (2008a:741—742): “The noun shows the same inflection as mēḫur / mēḫun- ‘time’.” (5) Kloekhorst’s conjecture that Hittite še-e-ḫur may be a loan from Palaic cannot be proven. (6) Sturtevant (1951:50, §75) also reconstructs *H3 here (Indo-Hittite *se¦wr), but for different reasons. 4. *H4 > Ø in Anatolian: Hittite (nom. sg.) al-pa-aš, al-pa-a-aš ‘(storm) cloud’ (< *H4el-bºo-s) ~ (?) Greek ἄλφος ‘whiteness, white leprosy’; Latin albus ‘white’; Umbrian alfu ‘white’; Old Icelandic elptr ‘swan’ (named for its white color); Old English ielfetu ‘swan’; Old High German albiz ‘swan’; Old Church Slavic lebedь ‘swan’ (< Proto-Slavic *olbǭdь; *oldǭtь; *elbedь; *elbǫtь ‘swan’ [cf. Derksen 2008:365—366]); Polish łabędź ‘swan’; Czech labud ‘swan’; Russian lébedʹ [лебедь] ‘swan’. Note: This etymology is rejected by Kloekhorst (2008a:169), mainly on semantic grounds. However, he also points out that he has “no better IE etymology for this word.” See also Puhvel (1984— .1/2:37—38), who also questions this etymology on semantic grounds. However, the proposed alternative etymologies mentioned by Puhvel fare even worse (cf. Woodhouse 2012:226—227). [Cf. Bomhard 2023.3:837—838, no. 705, Proto-Nostratic root *hal- (~ *həl-): (vb.) *hal- ‘to light up, to shine, to radiate, to brighten up, to beam forth’; (n.) *hal-a ‘clearness, brightness, radiance, purity’; (adj.) ‘clear, pure, bright, shining, radiant’.] 40 Chapter 3 Hittite (adv.; postpos.) a-ap-pa ‘behind, afterwards; back, again, further’, a-ap-pa-an (adv.) behind, after(wards)’ (< *H4epºo [*H4epo]); Cuneiform Luwian a-ap-pa ‘back, again, after’; Hieroglyphic Luwian á-pa-na ‘after, behind, again’; Lycian (adv.) epñ ‘afterwards’, epñte (adv.) ‘thereafter’, epre/i- (adj.) ‘back-, rear-’ ~ Sanskrit ápa ‘away, forth, back’; Old Persian (prefix) apa- ‘away’; Greek ἄπο, άπό ‘off, away, back’; Latin ab ‘away from’; Gothic af ‘of, from, by, away from’; Old Icelandic af ‘off, from’; Old English of ‘from, away from’; Old Frisian af, of ‘off, from, away from’; Old Saxon af ‘off, from, away from’; Dutch af ‘off, down’; Old High German ab, aba ‘off, from, away from’ (New High German ab). [Cf. Bomhard 2023.3:847, no. 712, Proto-Nostratic root *hapº- (~ *həpº-): (vb.) *hapº‘to turn, to turn away, to turn back’; (n.) *hapº-a ‘the act of turning away, turning back, overturning’; (adj.) ‘turned away from, turned back, overturned’.] Hittite (nom.-acc. sg.) a-ra-a-u-(wa-)aš ‘free’, (1st sg. pres.) a-ra-wa-aḫ-ḫi ‘to set free, to make free’ (< *h4er-wo-/*h4or-wo-); Lycian arawa ‘free’, arawã ‘exempt from tax’, ʼΕρεύαϛ /*erewa-/ ‘free(city)’ ~ Lithuanian árvas ‘free’. Notes: (1) Puhvel’s (1984— .1/2:119—121) rejection notwithstanding, the most convincing IndoEuropean cognate remains Lithuanian árvas ‘free’. See also Tischler 1977— :53— 55. (2) This etymology is accepted by Gamkrelidze—Ivanov (1995.I:397—398 and I:781), who reconstruct *arw- (I:397) and *arwo- ‘free agriculturalist’ (I:781). However, the putative Latin, Middle Irish, Greek, and Armenian cognates adduced by Gamkrelidze—Ivanov do not belong here. [Cf. Bomhard 2023.3:851, no. 717, Proto-Nostratic root *her- and/or *hor-: (vb.) *her- and/or *hor- ‘to escape, to flee, to run away’; (n.) *her-a and/or *hor-a ‘escape, flight’; (adj.) ‘escaped, liberated, free’.] Hittite (nom. sg.) ta-ya-az-zi-il ‘theft’, (3rd sg. pres. act.) ta-a-i-ez-zi, ta-i-ez-zi, etc. ‘to steal (from)’ (< *tºeH4-ye/o- [*(s)teH2-ye/o-]) ~ Sanskrit tāyú-ḥ ‘theft’, stená-ḥ ‘thief, robber’, stāyát ‘in secret’; Avestan tāyuš ‘thief’; Greek τητάω ‘to rob’; Old Irish táid ‘thief’; Old Church Slavic tatь ‘thief’. Note: The Sanskrit forms show socalled “movable s” or “mobile s”. Hittite (acc. sg.) ma-ak-la-an-ta-an ‘thin, meager, slim (of animals)’ (< *meH4kº-lontº- [*meH2k̂ -lo-nt-]) ~ Greek μῆκος (Doric μᾶκος) ‘length’, μακρός ‘long, tall’; Latin macer ‘thin’; Old Icelandic magr ‘lean’; Old English mKger ‘lean, meager’; Old High German magar ‘thin, meager’ (New High German mager). Hittite (1st sg. pres. act.) ti-ya-mi ‘to step, to go stand, to place oneself, to set in’ (< *(s)tº(e)H4-ye/o- [*(s)t(e)H2-ye/o-]; Cuneiform Luwian (3rd sg. pres. act.) ta-a-i ‘to come to stand’; Hieroglyphic Luwian (3rd sg. pres. act.) ta-i ‘to come to stand’ ~ Sanskrit (reduplicated) tíṣṭhati ‘to stand’; Greek (reduplicated) ἵστημι ‘to make to stand; to stand’, στατός ‘placed, standing’; Latin (reduplicated) sistō ‘to cause to stand, to set, to place’; Gothic standan ‘to stand’; Old Icelandic standa ‘to stand’; Old English standan ‘to stand’; Old Frisian stonda, stān ‘to stand’; Old Saxon standan, stān, stēn ‘to stand’; Old High German stantan, stān, stēn ‘to stand’ (New Anatolian and the Laryngeal Theory 41 High German stehen); Lithuanian stóti ‘to stand’; Old Church Slavic stati ‘to stand, to become’. Notes: (1) The Proto-Indo-European root contains so-called “movable s” or “mobile s”. (2) As noted by Kloekhorst (2008a:879—880): From the beginning of Hittite studies it has been in debate whether tii̯ e/a-zi goes back to the PIE root *dºeh1- ‘to put’ or *steh2- ‘to stand’. The former root would be possible in view of the meaning ‘to place oneself’ and the NH merger of tii̯ e/a-zi with dai-i / ti-, which clearly must reflect *dºeh1-. An etymological connection with *steh2- would much better fit the meaning ‘to step, to go stand’, however, which cannot easily be derived from an original meaning ‘to put, to place’. Kloekhorst reconstructs Proto-Indo-European *(s)th2-i̯ e/o-. 3.4. Labialized Laryngeals Adrados, Kloekhorst, Martinet, and Puhvel, among others, have proposed that one or more labialized laryngeals should be reconstructed for Proto-Indo-European, and, indeed, there is some evidence to support such a view. The following are taken from Bomhard (2023.3:943—947, nos. 794—798): Proto-Indo-European *H2¦el-/*H2¦ol-/*H2¦l̥ - ‘to draw, to pull, to tear out’: Avestan (in compounds) varək- ‘to draw’; Latin vellō ‘to pluck, to pull, to tear out’; Lithuanian velkù, vil͂ kti ‘to drag, to pull’; Old Church Slavic vlěkǫ, vlěšti ‘to draw, to drag’; Gothic wilwan ‘to rob, to plunder’, wilwa ‘robber’. Note: There may be a connection here with the words for ‘wool’ in the sense ‘to pluck (wool)’, in which case, we can add the following: Hittite (dat.-loc. sg.) ḫu-u-la[n(i)] ‘wool’ (< *H2w(e)lH1-n-); Cuneiform Luwian *ḫulana/i- ‘wool’ ~ Sanskrit ū́ rṇa-ḥ ‘wool’; Greek λῆνος ‘wool’; Latin lāna ‘wool’; Welsh gwlan ‘wool’; Gothic wulla ‘wool’; Old Icelandic ull ‘wool’; Old English wull ‘wool’; Old High German wolla ‘wool’ (New High German Wolle); Russian vólna [волна] ‘fleece, wool’; Lithuanian vìlna ‘wool’. For the semantics, cf. Buck 1949:400, no. 6.22 wool: “… prob. the same as Lat. vellere, etc. ‘tear, pluck’.” [Cf. Bomhard 2023.3:943—944, no 794, ProtoNostratic root *x¦al-/*x¦əl-: (vb.) *x¦al- ‘to pull (off, out), to tear (off, out)’; (n.) *xwal-a ‘the act of pulling or tearing (off, out)’.] Proto-Indo-European *H2¦et’- [*H2¦ed-] ‘to say, to speak’: Sanskrit vāda-ḥ ‘speech, discourse, talk, utterance, statement’, vádati ‘to speak, to say, to utter, to tell, to report, to speak to, to talk with, to address’; Greek (?) ἀείδω (< *ἀ+είδω < *aweud-) ‘to sing’, αὐδάω ‘to utter sounds, to speak’, αὐδή (Doric αὐδά) ‘the human voice, speech’, (?) ἀηδώ, ἀηδών ‘nightingale’; Lithuanian vadinù, vadìnti ‘to call, to name’. Note: There are no known Anatolian cognates. [Cf. Bomhard 2023.3: 945—946, no. 796, Proto-Nostratic root *x¦at’- (~ *x¦ət’-): (vb.) *x¦at’- ‘to chatter, to speak’; (n.) *x¦at’-a ‘chatter, talk’. 42 Chapter 3 Proto-Indo-European *H2¦er-kº- [*H2¦er-k-] ‘to cry, to squeal’: Old Church Slavic vrěštǫ, vrěštati ‘to cry, to squeal’; Czech vřískat, vřeštět ‘to cry, to whimper’; Lithuanian verkiù, ver͂ kti ‘to weep, to cry’, verkšnà ‘cry-baby’, ver͂ ksmas ‘weeping, crying’. Note: There are no known Anatolian cognates. [Cf. Bomhard 2023.3:947, no. 798, Proto-Nostratic root *x¦ir- (~ *x¦er-): (vb.) *x¦ir- ‘to make a loud noise, to make a shrill sound’; (n.) *x¦ir-a ‘loud noise’.] 3.5. The Phonetic Values of the Laryngeals The literature listed at the end of this book really needs to be consulted to get an idea of how much opinions differ regarding the phonetic values of the various laryngeals. Not only that, but also on how widely different interpretations of the same data lead to positing different laryngeals at the Proto-Indo-European level. I am hoping that my own endeavors lead to clarification and not more confusion. To begin, it might be helpful to repeat what I had previously written on the question of the phonetic values assigned to the laryngeals (cf. Bomhard 2023.1:72—79):28 One of the most difficult riddles to solve has been and continues to be the determination of the probable phonetic values of the various laryngeals (cf. Kessler no date). Some attempts include the following: 1. Sturtevant (1942:19), following Sapir (1938), assigns the following phonetic values to the laryngeals: *H1 (*ə̯1) = a glottal stop with frontal timbre (Sturtevant writes *’); *H4 (*ə̯4) = a glottal stop with velar timbre (Sturtevant writes *' [in later works, Sturtevant writes *h]); *H2 (*ə̯2) = a voiceless velar spirant (Sturtevant writes *x); *H3 (*ə̯3) = a voiced velar spirant (Sturtevant writes *¦). See also Pulleyblank 1993:65 and 90. 2. According to Lehmann (1952:103—108), *H1 (*ə̯1) was either a weakly aspirated glottal fricative (Lehmann writes *") or a pharyngeal fricative; *H4 (*ə̯4) was apparently a glottal aspirated fricative (Lehmann writes *h); *H2 (*ə̯2) was a voiceless velar fricative (Lehmann writes *x); and *H3 (*ə̯3) was a rounded voiced velar fricative *[ɣw] (Lehmann writes *¦). 3. Keiler (1970:68) posits the following values: *H1 (*ə̯1) = a voiceless glottal fricative */h/; *H2 (*ə̯2) = a voiceless pharyngeal fricative */ħ/; and *H3 (*ə̯3) = a voiced pharyngeal fricative */ʕ/. Couvreur (1937) arrived at the same conclusion, while Pulleyblank (1993:90) proposes that a pharyngeal glide may have been at least partially involved for *H3, which he writes /ă/. 4. Finally, Colarusso (1981:550) assigns the following values: *H1 (*ə̯1) = either a glottal stop */ʔ/ or voiceless and voiced pharyngealized velar fricatives,29 */xˁ/ and */ɣˁ/ (Colarusso writes */x̄/ and */ɣ̄/), respectively; *H2 (*ə̯2) = voiceless and voiced This is modified (enhanced, reformatted, and updated) from the original. Colarusso calls them “pharyngealized uvulars” and compares them to similar sounds in Semitic, where he states that they are “actually pharyngealized velars underlyingly”, and he writes them in the chart on page 550 as pharyngealized velars (*/x̄/ and */ɣ̄/). Consequently, I have labeled them “voiceless and voiced pharyngealized velar fricatives”. 28 29 Anatolian and the Laryngeal Theory 43 pharyngeal fricatives, */ħ/ (Colarusso writes */ḥ/) and */ʕ/, respectively; *H3 (*ə̯3) = either labialized voiceless and voiced pharyngeal fricatives, */ħw/ (Colarusso writes */ḥ¦/) and */ʕ¦/, respectively, or a labialized glottal stop */ʔ¦/; and *H4 (*ə̯4) = a voiceless glottal fricative */h/. According to Colarusso (1981:512), Couvreur (1937:264), Fortson (2010:64), Messing (1947:223—225), Sturtevant (1942:19 and 1951:54), Kloekhorst (2008a:62), Pulleyblank (1993:65 and 87—90), and Pooth (2015:11), *H1 (*ə̯1) was a glottal stop */ʔ/. The interpretation of *H1 (*ə̯1) as a glottal stop explains why this laryngeal did not color contiguous vowels. As noted by Catford (1977:105): “simple glottal stop has no influence on the quality of contiguous vowels”. This is verifiable from both Northwest Caucasian and Arabic, where glottal stops have no effect on vowel quality (cf. Colarusso 1981:511 for Northwest Caucasian and Al-Ani 1970:60—62 for Arabic). Moreover, loss of a glottal stop between an immediately preceding short vowel and an immediately following nonsyllabic causes compensatory lengthening of the vowel in Akkadian and Arabic (cf. Cantineau 1960:79; Couvreur 1937:288—289; Moscati [ed.] 1964:61—64; J. Watson 2002:18—19). Note the following examples from Akkadian (these examples are taken from Couvreur 1937:288—289): 1. Akkadian *ra"šu > rāšu (later rēšu) ‘head’; Hebrew rō"š [var)] ‘head’; Aramaic rēšā ‘head’; Phoenician r"š ‘head’; Arabic ra"s ‘head’; Epigraphic South Arabian r"s ‘head’; Śḥeri / Jibbāli rέš/réš ‘head’; Soqoṭri riy ‘head’; Ugaritic rÕs ‘head’; Geez / Ethiopic rə"əs ‘head’ [ርእስ]; Tigrinya rə"si ‘head’; Tigre rä"as ‘head’; Amharic ras ‘head’. Cf. Militarëv 2011:75, no. 38. 2. Akkadian *raḥmu > *reḥmu > *re"mu > rēmu ‘grace, mercy’; Hebrew raḥūm [<Wjr^] ‘compassionate’; Arabic raḥima ‘to have mercy, compassion’, raḥma ‘pity, compassion’; Śḥeri / Jibbāli raḥám ‘to be kind’; Mehri rəḥām ‘to be kind to someone’; Ḥarsūsi reḥam ‘to pity’; Ugaritic rḥm ‘to be kind’; Tigre räḥama ‘to have pity on’ (Arabic loan). 3. Akkadian *ba«lu > *be«lu > *be"lu > bēlu ‘owner, lord’; Hebrew ba«al [lu^B]^ ‘lord, owner’; Ugaritic b«l ‘owner of the house’; Arabic ba«l ‘husband, master, owner’; Epigraphic South Arabian b«l ‘master, owner’; Ḥarsūsi bāl ‘master, lord’; Mehri bāl ‘owner, possessor’; Śḥeri / Jibbāli bá«al ‘person owning’; Soqoṭri ba«l ‘master, lord’; Geez / Ethiopic ba«āl [በዓል] ‘owner, master’; Tigre bä«al ‘master’; Tigrinya bä«al, ba«al ‘master’; Amharic bal ‘master’. Identical developments are assumed for *H1 (*ə̯1) in Proto-Indo-European. This laryngeal is not directly attested in any of the Indo-European daughter languages, including Hittite and the other Anatolian daughter languages (cf. Bomhard 1976:230; Sturtevant 1942:53 and 1951:154). Additional confirmation that *H1 (*ə̯1) was a glottal stop is provided by the following forms (discussed above): Sanskrit (3rd sg.) píbati ‘drinks’, Latin bibit ‘drinks’, Old Irish ibid ‘drinks’. The Proto-Indo-European antecedent would have been the reduplicated verbal form (3rd sg. pres.) *pºi-pºH1-etºi [*pi-pH1-eti] ‘drinks’, that is, *pºi-pºʔ-etºi. Now, according to Gamkrelidze, Hopper, and Ivanov, glottalized stops become voiced stops in 44 Chapter 3 Sanskrit, Latin, and Old Irish. Likewise, we would expect the cluster *-pºʔ- to become /b/ in these languages, and this is exactly what we do in fact find. The following developments may be assumed (cf. Gamkrelidze—Ivanov 1995.I:856): *pºi-pºʔ-etºi > (with deaspiration of *-pº- in the cluster *-pºʔ-) *pºi-pʔ-etºi > (with reanalysis of *-pʔ- as *-p’-) *pºi-p’-etºi > (with deglottalization) *pºi-p-etºi > (with voicing of medial *p) *pºi-b-etºi > (with deaspiration of voiceless aspirates) *pi-b-eti > Sanskrit píbati ‘drinks’, Latin bibit ‘drinks’, Old Irish ibid ‘drinks’. Kuryłowicz (1935:29—30) sets up *H4 (*ə̯4) to account for those cases in which an a in the non-Anatolian daughter languages corresponds to an a in Hittite, and Hittite lacks a contiguous laryngeal reflex. That is to say that *H4 (*ə̯4) is not directly attested in Hittite or in any of the other daughter languages (cf. Bomhard 1976:230; Sturtevant 1942:42 and 1951:51—52), though its former presence can be determined by the fact that it changed a contiguous *e to *a and by the fact that it caused compensatory vowel lengthening when lost between an immediately preceding short vowel and an immediately following nonsyllabic. According to Hopper (1977a:49—50), typological evidence implies that the voiceless laryngeal fricative /h/ should be added to the Proto-Indo-European phonemic inventory, and this coincides with the phonetic value assigned to *H4 (*ə̯4) by Colarusso (1981:512), Lehmann (1952:108), and (apparently) Sturtevant (1951:52). In terms of distinctive feature theory, /h/ is [+cons, +low, -voice, +cont, +grave]. As far as we are concerned, the most important feature is [low]. According to Chomsky—Halle (1968:305), the articulatory gesture behind the feature [low] is a “lowering [of] the body of the tongue below the level it occupies in the neutral position”, while Colarusso (1981:509) defines it as “an opening of the oral cavity to enhance resonance”. It was the presence of this feature that was responsible for the lowering of a contiguous *e to *a. Finally, we may note that developments similar to those assumed for *H4 (*ə̯4) in Proto-Indo-European are found in Ubykh and in the Circassian languages, where /h/ (and /h¦/) lowers and colors contiguous vowels and also causes compensatory vowel lengthening when lost (cf. Colarusso 1975: 396). Reflexes of *H2 (*ə̯2) are found in Hittite and the other older Anatolian languages (that is, Palaic and Cuneiform and Hieroglyphic Luwian), where they are written (ḫ)ḫ (cf. Fortson 2010:178; Sturtevant 1942:35 and 1951:47). This laryngeal also survives in Lycian, where it is written χ. Like *H4 (*ə̯4), *H2 (*ə̯2) lowers a contiguous *e to *a. On this basis, we would expect *H2 (*ə̯2) also to be characterized by the presence of the feature [low]. A good candidate to assign as the phonetic values of *H2 (*ə̯2) would be the voiceless multiply-articulated pharyngeal/laryngeal /¸/ (it could also have been an adytal [+CP, +low]). Not only is this sound marked by the presence of the feature [low], which accounts for the lowering of adjacent vowels, but it also makes it easy to account for the fact that *H2 (*ə̯2) appears as h in Armenian before full-grade vowels. We can envision a change of *‿ ħh into *h similar to what is found in the Ashkharwa dialect of Abkhaz (cf. Colarusso 1981:516). The resulting *h would have subsequently been lost in all of the non-Anatolian daughter languages except Pre-Armenian. As in Ashkharwa, we may venture a guess that *‿ ħh developed from the earlier pharyngeal *ħ in Pre-Indo-Anatolian. Indeed, support for such an assumption comes from the lexical parallels between Proto-Indo-European and Proto-Afrasian, where Proto-Indo-European *H2 (*ə̯2) corresponds to Proto-Afrasian *ħ. Anatolian and the Laryngeal Theory 45 Finally, we should take note of Jakobson’s (1971[1956]:518—520) description of similar sounds in Arabic (see also J. Watson 2002:44—45): … /ḥ/ is essentially a pharyngealized laryngeal. Of the two phonemes of this type, /ḥ/ is usually produced without voice and /‛/ with voice. Since a considerable part of the air used with /‛/ is consumed by voicing alone, this phoneme is a lenis, in contradistinction to the fortis /ḥ/. Thanks to the pharyngeal contraction, the voice-pitch in /‛/ and the whisper-pitch in /ḥ/ are very low: “In passing to /‛/ from a preceding vowel the voice has to descend rapidly, often through more than an octave, and is cut off at its lowest pitch. If a vowel follows, the pitch begins at its lowest level and rises quickly, through a similar interval, to normal vowel pitch.” (1971[1956]:518— 519) As to the influence upon the adjacent vowels, the componential analysis of a phoneme cannot proceed from the contextual variants of neighboring phonemes: often the variation is due not to a single feature but to a combination of concurrent features. Furthermore, in many instances the pharyngeals modify adjacent vowels in the same direction as pharyngealized buccals. In colloquial Egyptian both the pharyngealized buccals and the pharyngeals appear to exert a modifying retracting influence on preceding and following a-vowels (Gairdner, p. 46f.). In the dialect of El-Hamma, Cantineau observes that the /a:/ is pronounced “entre a et o ouvert” in contact with pharyngealized dentals, while in contact with pharyngeals it is realized as “a moyen français”, in contact with velars it oscillates between the two positions mentioned, and in other contexts it is a front vowel. In the same dialect the phoneme /u:/ is shifted towards the closed o in the neighborhood of pharyngealized dentals, velars, and pharyngeals (1951, p. 78f.). (1971[1956]:520) It is more difficult to determine the phonetic value of *H3 (*ə̯3) than of any of the other laryngeals. Reflexes of *H3 (*ə̯3) are also found in the older Anatolian languages (cf. Bomhard 1976:228—230; Fortson 2010:178; Sturtevant 1942:44 and 1951:49—51). Kuryłowicz (1935:28—30) tried to show that *H3 (*ə̯3) changed a contiguous *e to *o, but Sturtevant (1938:104—111 and 1942:20) has argued against such an assumption. The majority of scholars are inclined to accept Kuryłowicz’s interpretation. Now, several scholars (Sturtevant, Lehmann, Keiler, and others) have suggested that *H3 (*ə̯3) was the voiced counterpart of *H2 (*ə̯2). Consequently, we can speculate that *H3 (*ə̯3) was a voiced multiply-articulated pharyngeal/laryngeal */‿ ʕɦ/ (as with */‿ ħh/, it could also have been an adytal [+CP, +low]). Now, a more careful examination indicates that *H3 and *H2 may actually have had the same vowel-coloring effects — Pulleyblank (1993:90) arrives at the same conclusion: … the effect of the O laryngeal is seen to be the same as that of the A laryngeal, only earlier in taking effect. We know from the examples given above that *H2 lowered and colored a contiguous *e to *a, original *i to e, and original *u to *o. As in the Arabic case discussed by Jakobson above, we would expect *H3 to have had a similar effect on these vowels in early ProtoIndo-European. That is to say that we would expect *H3 to have lowered and colored a contiguous *e to *a, original *i to e, and original *u to *o. In fact, there is some evidence — albeit controversial — within Indo-European itself to support this, as the following 46 Chapter 3 examples involving *H2 illustrate (these are also listed above, together with additional examples): 1. Early Proto-Indo-European *H2inkº- > later Proto-Indo-European *H2enkº- ‘to reach, to come to, to arrive at’ (Pokorny 1959:316—318 reconstructs *enek̂ -, *nek̂ -, *enk̂ -, *n̥ k̂ - here): Hittite (3rd sg.) ḫi-in-ik-zi ‘to present, to deliver, to offer, to allot’; Sanskrit aśnóti ‘to reach, to come to, to arrive at, to get, to obtain; to master; to offer’; Latin nancior ‘to get, to gain, to obtain’, nanciscor ‘to get, to gain, to receive, to meet’; Tocharian A ents-, B eṅk- ‘to seize, to take’. Cf. Puhvel 1984— .3:289—292; Melchert 1994a:143—144. The Hittite form directly attests *H2inkº-. Note: That the transition from *i to *e was already taking place as early as Hittite is shown by forms such as (nom.-acc. sg.) ḫé-en-gur ‘consignment, offering, oblation, gift, tribute’ beside (nom.acc. sg.) ḫi-in-ku-wa-ar. The same variation occurs in (nom.-acc. sg.) ḫé-en-kan ‘death, doom, deadly, disease, plague’ alongside (nom.-acc. sg.) ḫi-in-kán. 2. Early Proto-Indo-European *H2ul- > later Proto-Indo-European *H2ol- ‘to destroy’ (Pokorny 1959:777 reconstructs *ol-[e]- here): Hittite (3rd sg.) ḫu-ul-la-a-i ‘to smite, to destroy, to defeat’; Latin ab-oleō ‘to destroy’; Greek ὄλλῡμι ‘to destroy’. Cf. Couvreur 1937:143—144; Cowgill 1965:146—147 and 157 (Cowgill derives the Greek form from *Ol̥ -ne-O-mi and considers the ο to be a replacement for original α — nonetheless, Cowgill accepts the comparison with Hittite ḫu-ul-la-a-i). The Hittite form directly attests *H2ul-. Note: Kloekhorst (2008b:358—360) derives the Hittite form in question from Proto-Indo-European *h2uelh1-: *h2ul-né-h1-ti, *h2ul-n-h1-énti and compares Old Irish follnadar ‘to rule’; Latin valeō ‘to be powerful’; Lithuanian véldu ‘to own’; Gothic waldan ‘to rule’; and Old Church Slavic vladǫ ‘to rule’. The semantics really do not match, however. Consequently, I reject this etymology. Cf. also Melchert 1994a:55—56 (*h2/3wl̥ -ne-h1- > Hittite ḫulle- ‘fight’), 66, and 82. Finally, Polomé (1965:18) and Puhvel (1984— .3:368) reject this etymology. 3. Early Proto-Indo-European *H2um- > later Proto-Indo-European *H2om- ‘all, whole’: Hittite (nom. sg.) ḫu-u-ma-an-za ‘all, whole’; Latin omnis ‘all, every, whole’. Cf. Couvreur 1937:144—146; Kronasser 1956:41; Pedersen 1938:165. The Hittite form directly attests *H2um-. Note: Polomé (1965:18) and Puhvel (1984— .3:380) reject this etymology — Puhvel derives Latin omnis from *opnis (as does de Vaan 2008:428). Ernout—Meillet (1979:461—462) state that there is “nothing similar elsewhere” (“Aucun mot pareil ailleurs”) to Latin omnis. On the other hand, Walde—Hofmann (1965—1972.II:209—210) mention Oscan úmbn, which points to earlier *omb-nis and not *opnis as the source of both the Oscan form and Latin omnis. *omb-nis may contain an epenthetic b, in which case the original form would have been *om-ni-s. Here, -niis a suffix. Likewise, in Hittite ḫu-u-ma-an-za, the stem is *ḫum-, and the -anz(a) is a suffix (< *-onts or *-n̥ ts). Thus, this etymology can be revived if we consider the original form to have been *H2um-, which later became *H2om-, with *-o- from earlier *-u- under the influence of the preceding laryngeal. Such an explanation overcomes the objections raised against this etymology based upon the irregular correspondence of Hittite u and Latin o. Anatolian and the Laryngeal Theory 47 As for *H3, there is a strong probability that the change of *i to *e under the influence of *H3 can be observed in Hittite (nom.-acc. sg.) še-e-ḫur, še-e-ḫu-ur, še-e-ḫu-wa-ar ‘urine’, (nom. sg.) še-e-ḫu-ga-ni-ya-u-wa-an-za ‘besmeared with urine’, and (3rd sg. pret. act.) še-e-ḫu-ri-ya-[a]t ‘to urinate’ (< *seH3-ur < *siH3-ur), discussed above. We may summarize our findings by setting up the following matrix: Traditional *H1 (*ə̯1) Traditional *H4 (*ə̯4) Traditional *H2 (*ə̯2) Traditional *H3 (*ə̯3) e lowered and colored to a i lowered and colored to e u lowered and colored to o Preserved in Anatolian Partially preserved in Armenian *ʔ + - *h + + + + - *‿ ħh + + + + + + *‿ ʕɦ + + + + + - That completes (and updates) what I had previously written — but there is more. It is hard to envision how *‿ ħh (= *H2) and *‿ ʕɦ (= *H3) could have become *x and *ɣ, which are the phonetic values commonly assigned to the reflexes of these laryngeals in the older Anatolian languages by most specialists.30 This seems to indicate that things were more complicated than previously thought. Now, my work on distant linguistic comparison indicates that, at an early stage of development, Pre-Indo-Anatolian had, at the very least, the following relevant phonemes (cf. Bomhard 2023.1:78): Glottal stops: Glottal fricative: Velar fricatives: voiceless: voiced: Pharyngeal fricatives: voiceless: voiced: Plain */ʔ/ */h/ */x/ */ɣ/ */ħ/ */ʕ/ Labialized */ʔ¦/ */x¦/ */ħ¦/ These phonemes remained intact at least up to what I call the “Phonemic Stress Stage” (= Proto-Indo-Anatolian) of Proto-Indo-European (cf. Bomhard 2023.1:78).31 Contrary to what I had assumed when I prepared my 2023 book, it now seems likely that these phonemes also remained into the “Phonemic Pitch Stage” of Proto-Indo-European, except Kümmel 2007 and Weiss 2016 argue that *H2 and *H3 may have been uvular fricatives, while Kümmel 2022 argues in favor of uvular stops. 31 Recently, new terminology has been proposed to describe the various stages of development. “IndoAnatolian” has replaced “Indo-Hittite” to describe the stage of development prior to the separation of the Anatolian branch from the main speech community. Next, “Indo-Tocharian” has been coined to describe the stage of development following the separation of Anatolian but prior to the separation of Tocharian. Finally, “Indo-European” has been retained to describe the stage following the separation of Tocharian and prior to the emergence of the remaining branches. 30 48 Chapter 3 that */ħ/ and */ʕ/ became the multiply-articulated voiceless and voiced pharyngeal/ laryngeal fricatives */‿ ħh/ and */‿ ʕɦ/, respectively. Prior to these changes, these laryngeals had no “vowel coloring” effects on contiguous vowels. After these changes, however, these sounds lowered contiguous vowels: *e > *a, original *i > *e and original *u > *o. This is extremely important, since I have maintained (Bomhard 2023.1:78) that the Anatolian branch became separated from the main speech community at the end of the “Phonemic Pitch Stage” of Proto-Indo-European. In Pre-Anatolian, */‿ ħh/ and */x/ merged into */x/, while */‿ ʕɦ/ and */ɣ/ merged into */ɣ/. (At the same time, */ʔ/ [= *H1] and */h/ [= *H4] were lost.) This accounts for the reflexes found in the older Anatolian daughter languages. Things were different, however, in the Indo-European ancestor of the non-Anatolian daughter languages. Here, */‿ ħh/ and */x/ merged into */‿ ħh/, while */‿ ʕɦ/ and */ɣ/ merged into */‿ ʕɦ/. (As a typological parallel, it may be mentioned that a similar change has taken place in Hebrew, where Proto-Semitic */ʕ/ and */ɣ/ have merged into /ʕ/ [u], and */ħ/ and */x/ have merged into /ħ/ [j] ⸺ the change was complete by the Middle Hebrew period [cf. Lipiński 2001:152—153].) Then, in the stage of development which I have called “Disintegrating Indo-European”,32 the laryngeals were mostly lost. First, the laryngeals */ʔ/ and */h/ were lost initially before vowels. In all other environments, */ʔ/ and */h/ merged into */h/. Then, the laryngeals */‿ ħh/ and */‿ ʕɦ/ became */h/ (*/‿ ħh/ > */h/; */‿ ʕɦ/ > */ɦ/ > */h/). At this time, the single remaining laryngeal */h/ had no vowel-coloring effects on contiguous vowels. Finally, this */h/ was lost initially before vowels (except in PreArmenian) and medially between an immediately preceding vowel and a following nonsyllabic. This latter change caused compensatory lengthening of preceding short vowels (*eHC, *oHC, *aHC, *iHC, *uHC > *ēC, *ōC, *āC, *īC, *ūC). Note: */h/ may have been simply lost without a trace in certain contexts (cf. Byrd 2010). 3.6. Old Letters, New Values Earlier in this chapter, under the discussion of Carian phonology, I mentioned that new values have been assigned to several Carian letters. Let me repeat what I said (I am referring to the table prepared by Melchert 2004e:610 and given above): Three of the letters in this table are of particular importance to our understanding of the development of laryngeals in Anatolian under investigation in this paper, namely, the letters transcribed by Melchert as /q/, /x/, and /k/. Adiego (2004:242—245) assigns slightly different values for two of the letters, namely, /k/ = Melchert’s /x/ and /k̑/ = Melchert’s /k/. Both Adiego and Melchert agree on /q/. Several scholars (Kloekhorst, Schürr, and Simon) have adopted the new values in recent work on Carian and Anatolian, in general — Brosch (no date) provides an excellent summary of the views of these scholars, while Simon (2021) lists the lexical evidence. However, Adiego’s arguments in favor of the values he assigns to these letters are highly conjectural and based upon conflicting evidence. When one takes into consideration etymological factors, it can be observed that Carian /k/ (Melchert) ~ /k̑/ (Adiego) corresponds etymologically to /k/ in older Anatolian languages (Hittite, Palaic, and Cuneiform and Hieroglyphic Luwian), while /x/ (Melchert) ~ /k/ (Adiego) corresponds etymologically to /ḫ/ in older Anatolian languages. /q/, on the other hand, appears to represent /x¦/, just as in Lycian (see footnote 17). This is the stage of development between the separation of the Anatolian branch from the main speech community and the emergence of the various non-Anatolian Indo-European daughter languages. 32 Anatolian and the Laryngeal Theory 49 In my opinion, this speaks in favor of the values given by Melchert in the above table and against the values assigned by Adiego. I want to stress that the evidence upon which the new values are based has not changed. What is new is the interpretation. Brosch (no date) lists the Carian vocabulary typically interpreted as containing reflexes of the laryngeals — remember that the k in these forms represents the Carian letter traditionally interpreted as χ (= Melchert’s x): 1. The element kδ° ‘former, first’ in the name kδ-uśolš and the controversial kδou‘king’ (?), which have been compared with Lycian χñtawat(i)- ‘ruler, sovereign’ (< *h2ent-). 2. p/bik° (mostly Greek Pig°) ‘light’ in various proper names (< *bºēh2-). 3. kb- ‘river, stream’ in the place names kbid- ‘Kaunos’ and kbo- ‘Keramos’ (< *h2eb(h)-o/n-). 4. ksbo ‘ankle’ in personal names (< *h2emsu-+-ā) (cf. Lycian χahba). 5. Place name kiδb- (< Hinduwa). 6. The Greek gloss κόον ‘sheep’, perhaps also in the title koíoλ (< *h3eu̯ i-) (cf. Lycian χawã). 7. Proper name quq- (Greek Gýgēs, Gugos), also in proper names dquq (Greek Idagygos) (< *h2eu̯ h2o/eh2- ‘grandfather’) (cf. Lycian χuge). 8. Name of the Storm God trq(u)δ- (< *tr̻ h2unt-) (cf. Lycian Trqqñt-). 9. qdarŕou- ‘servant’ in personal names (cf. Cuneiform Luwian ḫutarlā-). 10. Conjunction =q ‘and’ (< *=h3e or *=h2o or *=kwe ?). 11. Probably the pronoun χiχ (= Lycian tike, Milyan kike; Cuneiform Luwian kuišḫa). Brosch mentions that there are several other items without etymologies. Were this all, it would not be that big an issue, given the extremely small size of the Carian corpus and the uncertainty surrounding the alleged reflexes of the laryngeals, and the debate on the values of these letters could have been confined to Carian, without too much damage. However, the new interpretation has spread to Lycian as well, and this is important. Brosch summarizes the new interpretation in the following table — he gives the names of the major scholars involved and references to the relevant publications: ProtoAnatolian *H*-H*-h*(-)Hw- Hittite, Palaic, Cun. Luw. ḫ-ḫḫ-ḫ(-ḫ)ḫu- Lycian χ [k] χ [k] g [γ] q [kw?] Simon (2011) k [k], q [q]/__u k [k] q [q] qu [qu] < *h2u Kloekhorst (2008b) k [k] ? q [kw] Carian Adiego (2007) Schürr (2001) Adiego (1995) ? k [k], q [q] k [k], q [q] k [k] q /kw/k] The final step in the evolution of these ideas is the proposal that the laryngeals preserved in Anatolian (namely, *H2 and *H3) were actually uvular stops in Pre-Anatolian 50 Chapter 3 (cf. Kloekhorst 2018).33 The thinking is that they would have been preserved as stops in Lycian and Carian but changed to uvular fricatives in Hittite, Palaic, and Cuneiform and Hieroglyphic Luwian. Kloekhorst (2018:79) concludes regarding *H2: We have seen that, typologically, it is difficult to understand how Lycian and Carian [k] could have been the outcome of *h2 if the latter sound originally was a uvular fricative. Instead, Lyc./Car. [k] is much better explained from a uvular stop, which can also account for the uvular fricative as found in Hittite and Luwian. Moreover, there are additional arguments to be given in favour of such a reconstruction, especially the fact that *h2 yields a fortis (long) consonant in Hittite and CLuwian. All in all, I want to propose the following values for *h2 in the different language stages: PIA *h2 = *[q:] Pre-PIE *[χ] PAnat. *[q:] PLuw. *[q:] Cl. PIE *[ħ] or *[ʕ] Proto-Caro-Lycian *[k:] Other IE languages Car. [k] Lyc. [k] Luw. [χ:] Hitt. [χ:] The interpretation of the laryngeals *H2 and *H3 as uvular stops in Pre- or Proto-IndoAnatolian proper has major implications for the basic tenets of the Laryngeal Theory listed at the beginning of this chapter. First, there is the question of the vowel-coloring effects traditionally attributed to the laryngeals. The vowel-coloring effects must have already begun in Pre- or Proto-Indo-Anatolian, since they can clearly be observed in the Anatolian daughter languages. As noted by Colarusso (1981), plain uvular stops and fricatives cannot have produced these effects. Colarusso discusses, in detail, typological parallels with Northwest Caucasian, Northeast Caucasian, and Semitic to illustrate this point. Consequently, plain uvular stops and fricatives can be ruled out at the Proto-Indo-Anatolian level on this basis alone. Next, the plain pharyngeal fricatives reconstructed by Kloekhorst for the Proto-Indo-European precursor of the non-Anatolian daughter languages also cannot have produced the vowel-coloring effects traditionally attributed to the laryngeals — additional phonetic features and developments are required. (Here, it may be noted that the loss of pharyngeal fricatives, as well as other “laryngeals”, in Akkadian resulted in the change of contiguous /a/ to /e/ [cf. Lipiński 2001:149] ⸺ this is the exact opposite of the vowel-coloring effects traditionally attributed to “a-coloring laryngeals” in Proto-IndoEuropean.) This was illustrated above in the table of Semitic developments originally prepared by Couvreur (1937:288—289) and is also discussed by Colarusso (1981) — instructive here as well are the typological parallels between the development of the 33 Similar ideas were expressed by Kaiser—Shevoroshkin (1985), regarding Proto-Nostratic. Anatolian and the Laryngeal Theory 51 “laryngeals” in Coptic, on the one hand, and in Proto-Indo-European, on the other hand, discussed by Greenberg (1969:175—184). It may also be noted that the laryngeals do not pattern as stops during any period of development. Finally, my work on distant linguistic comparison (cf. Bomhard 2023) does not support these views. I have given references to that monograph throughout the present chapter. Moreover, my 2019 Journal of IndoEuropean Studies paper entitled “The Origins of Proto-Indo-European: The Caucasian Substrate Hypothesis”,34 in which I present evidence of prehistoric language contact between the precursors of Proto-Indo-European and Proto-Northwest Caucasian, does not support these views. In sum, these theories simply are not supported by the cumulative evidence. It seems to me that the initial mistake began with the interpretation of the Carian reflexes of the laryngeals preserved in Anatolian as stops. This simply cannot be correct. Likewise, the interpretation of the Lycian reflexes of these laryngeals as stops cannot be correct. Consequently, I urge that these theories be abandoned and that the earlier values assigned to the letters in question in Carian and Lycian be reinstated.  This paper was substantially expanded and reissued in book format under a Creative Commons license in 2023 under the title Prehistoric Language Contact on the Steppes: The Case of Indo-European and Northwest Caucasian. 34 Chapter 4 The Reconstruction of the Proto-Anatolian Consonant System 4.1. Introduction Before the discovery of Hittite and the other Anatolian Indo-European daughter languages, the consonant system of Sanskrit, with its four-way contrast of (1) plain voiceless ~ (2) voiceless aspirated ~ (3) plain voiced ~ (4) voiced aspirated stops and affricates, was thought to preserve the consonant system of the Indo-European parent language better than any of the other older Indo-European daughter languages: (1) k c ṭ t p Velar Palatal (affricates) Retroflex Dental Bilabial (2) kh ch ṭh th ph (3) g j ḍ d b (4) gh jh ḍh dh bh Indeed, the prestige of Sanskrit is such that it has served as the model for the reconstruction of the Proto-Indo-European consonant system from the time of Schleicher to the present day. Most of the current handbooks/comparative grammars reconstruct a modified version of the Sanskrit consonant system for Proto-Indo-European, eliminating, notably, the voiceless aspirates (column 2 above) as well as the palatal and retroflex sounds of Sanskrit, while adding labiovelars, thus (cf. Adrados 1975.I:108; Beekes 2011:119— 120 [traditional transcription]; Byrd 2018:2056—2079, table pp. 2056—2057; Cowgill— Mayrhofer 1986:91—118; Clackson 2007:34; Fortson 2010:56; Fulk 2018:99; Lehmann 1952:7—8; Mallory—Adams 1997:459 and 2006:55; Meier—Brügger 2003:96—139, table p. 126, and 2010:226—270, table p. 257; Meillet 1964:82—98; Shevelov 1964:25— 26; Sihler 1995:135; Watkins 1998:34; Wodtko—Irslinger—Schneider 2008): Voiceless Voiced Voiced aspirated Labial *p (*b) *bh Dental Palatal *t *$ *d *’ *dh *’h Velar *k *g *gh Labiovelar *k¦ *g¦ *g¦h Note: Each author uses a slightly different transcription, especially for the palatals. There are well-known and often-mentioned problems with this reconstruction (cf. Mallory—Adams 2006:50—53), especially the typological implausibility of such a system. Since I have discussed these problems in a previous paper (Bomhard 2016), I will not go into details here. Suffice it to say that the traditional reconstruction of the Proto- 54 Chapter 4 Indo-European system of stops, as represented above, is in desperate need of a major overhaul. One of the most important axioms of the comparative method is that the oldest attested (and highest quality) data should play a pivotal role in the reconstruction of protolanguages. In the case of Indo-European, the oldest attested data come from the Anatolian branch, the consensus now being that this branch was the first to split off from the rest of the Indo-European speech community (cf. Jasanoff 2017:233—234). As it happens, the Anatolian branch contains several archaic features in the phonological systems of its member languages (namely, Hittite, Palaic, Cuneiform and Hieroglyphic Luwian, Lycian, Lydian, and several other poorly-attested languages), such as the preservation of laryngeals, not found in any of the other Indo-European daughter languages (cf. Jasanoff 2017:221—225) — a trace of an earlier laryngeal most likely exists in Armenian (cf. Winter 1965), while Kümmel (2018) tries to show that laryngeals may have been preserved in Iranian as well. In my opinion, the significance of the Anatolian phonological data, except for the laryngeals, has not been fully appreciated. From the very start, it must be made clear that there are still many unresolved questions regarding Hittite/Anatolian phonology. First, there is often disagreement among specialists concerning the interpretation of different writing conventions (cf. Jäntti 2017), though it is now generally accepted that “Sturtevant’s Law” (see Chapter 2 for details), that is, medial double writing versus medial single writing of stops, represents a genuine phonemic distinction in Hittite (cf. Kimball 1999:90—95), either a voicing contrast or a fortis ~ lenis contrast or even something else altogether — Kloekhorst (2016:213—217) and Jäntti (2017), for example, consider the distinction between medial single writing versus medial double writing of stops in Hittite to be one of length, while Melchert (1994) and Yates (2019a) maintain that medial double writing indicates simple gemination of earlier voiceless obstruents. Specifically, Yates (2019a) maintains that: … Sturtevant’s Law was a conditioned pre-Hittite sound change whereby (i) contrastively voiceless word-medial obstruents regularly underwent gemination (cf. Melchert 1994), but gemination was blocked for stops in pre-stop position; and (ii) the inherited [±voice] contrast was then lost, replaced by the [±long] opposition observed in Hittite (cf. Blevins 2004). I provide empirical and typological support for this novel restriction, which is shown not only to account straight-forwardly for data that is problematic under previous analyses, but also to be phonetically motivated, a natural consequence of the poorly cued durational contrast between voiceless and voiced stops in pre-stop environments. I develop an optimality-theoretic analysis of this gemination pattern in pre-Hittite, and discuss how this grammar gave rise to synchronic Hittite via “transphonologization” (Hyman 1976, 2013). Finally, it is argued that this analysis supports deriving the Hittite stop system from the Proto-Indo-European system as traditionally reconstructed with an opposition between voiceless, voiced, and breathy voiced stops (contra Kloekhorst 2016, Jäntti 2017). Though the recent paper by Simon (2019) is directed specifically against the views of Kloekhorst and Jäntti, his conclusions apply equally to the views advanced by Yates. It is thus worth quoting Simon’s concluding remarks in full: Kloekhorst’s arguments against the traditional voiced/voiceless contrast in Anatolian stops is not probative. None of his arguments necessarily require a contrast in length. Moreover, The Reconstruction of the Proto-Anatolian Consonant System 55 transcriptions and loanwords from half a dozen languages equivocally and unambiguously show that Hittite and Luwian stops were always perceived as voiceless and voiced stops and never as geminates, pace Melchert and Kloekhorst. One cannot reject this evidence since the perception was consistent across many languages and many centuries (almost one millennium) involving both Hittite and Luwian. In other words, there is no reason to assume that the contrast in Anatolian stops was in length; and consequently the contrast in voice is neither a shared innovation nor a defining feature of the non-Anatolian Indo-European languages. Next, there is still no consensus on the significance of plene writing of vowels (cf. Makarova 2001). These unresolved questions notwithstanding, the Anatolian data can still tell us a great deal about the phonological system of their immediate Proto-Indo-Anatolian ancestor. As we shall see, it is nothing at all like the Sanskrit-based reconstruction mentioned above. 4.2. Stops It is clear that Proto-Anatolian had a voicing contrast in stops in all positions, initial, medial, and final (cf. Kimball 2017:252—253; Melchert 2017:177; Luraghi 1998:175), though this may have developed into a fortis ~ lenis contrast independently in the various Anatolian daughter languages. From the point of view of the Proto-Indo-Anatolian ancestor of Proto-Anatolian, it is the preservation of a voicing contrast that is of paramount importance and not the inner-Anatolian developments. Thus, at a minimum, the following stops can be confidently reconstructed for Proto-Anatolian: Bilabial Dental Velar Labiovelar Voiceless p t k k¦ Voiced b d g g¦ Now, it should be noted that there is absolutely no evidence whatsoever from any of the Anatolian daughter languages that a series of voiced aspirates existed in ProtoAnatolian (cf. Kimball 2017:252). This is an important point. It implies that the voiced aspirates traditionally reconstructed for Proto-Indo-European must have developed after the separation of the Anatolian branch from the main speech community. In fact, the voiced aspirates are only really needed to account for developments in Indo-Aryan (but not Iranian), Greek, Armenian, and Italic. As noted above, the traditional reconstruction of the Proto-Indo-European stop system posits a three-way contrast of (1) plain voiceless stops, (2) plain voiced stops, and (3) voiced aspirated stops. While the voiceless stops posited for Proto-Anatolian above unambiguously correspond to the plain voiceless stops (*p, *t, *$, *k, *k¦) of traditional Proto-Indo-European, the voiced stops posited for Proto-Anatolian above are usually taken to correspond to both the plain voiced stops (*b, *d, *’, *g, *g¦) and the voiced aspirated stops (*bº, *dº, *’º, *gº, *g¦º) of traditional Proto-Indo-European. This is interpreted to mean that the plain voiced stops and voiced aspirates of traditional Proto-Indo-European 56 Chapter 4 have merged into a single series of plain voiced stops in Proto-Anatolian (cf., for example, Kloekhorst 2008a:17). Though on the surface, this appears to account quite nicely for the Anatolian developments, there are indications from within Anatolian itself that a three-way contrast must be reconstructed in the system of stops for Proto-Anatolian instead of a simple two-way contrast. The evidence, though sparse, comes mainly from Luwian and, to a lesser extent, Hittite. In Luwian, traditional Proto-Indo-European *k was preserved before high front vowels, as in the following example: Cuneiform Luwian kiš- ‘to comb, to card’ = Hittite (3rd sg. pres. act.) ki-iš-zi ‘to comb’ < Proto-Anatolian *kes- ‘to comb, to card’ < traditional Pre-Anatolian Proto-IndoEuropean *kes- ‘to comb, to card’. Probable non-Anatolian cognates include Greek κέσκεον (< *kes-kes-) ‘tow, oakum’; Old Church Slavic češǫ, česati ‘to comb, to pull off’. On the other hand, traditional Proto-Indo-European *’h was lost in the same environment in Luwian, as shown by the following examples: 1. Cuneiform Luwian (nom. sg.) (i-)iš-ša-ri-iš, i-iš-ri-iš, etc. ‘hand’; Hieroglyphic Luwian (acc. sg.) i-sà-tara/i-na ‘hand’; Lycian izri- ‘hand’ (all with loss of an earlier initial voiced velar before high front vowel) = Hittite (nom.-acc. sg.) ki-eš-šar ‘hand’ < ProtoAnatolian *gēsar ‘hand’ < traditional Pre-Anatolian Proto-Indo-European *’hēsr̥ ‘hand’. Non-Anatolian cognates (from the same root but with a different suffix) include the following: Sanskrit hásta-ḥ ‘hand’; Old Persian dasta- ‘hand’; Avestan zasta‘hand’; Latin praestō (< *prae-hestōd) ‘at hand, ready’. Note: The Hieroglyphic Luwian form contains an epenthetic t. 2. Cuneiform Luwian (gen. sg.) im-ma-ra-aš-ša ‘open country’ = Hittite (nom. sg.) gi-imra-aš ‘the outdoors, countryside, field; military campaign’ < Proto-Anatolian *gim-ra‘open country, countryside, field’ < traditional Pre-Anatolian Proto-Indo-European *’him-ro- ‘open country, countryside’ (< ‘the wintry steppe, the inhospitable outdoors’ [cf. Kloekhorst 2008:476—477; but see Garnier 2018 for a different interpretation]). Non-Anatolian cognates include Greek χειμερινός ‘of or in winter; stormy’; Armenian jmeṙn ‘winter, snow-storm’. Now, what about traditional *’. Curiously, it appears to have been preserved before high front vowels in Luwian, provided, of course, that the following is a valid etymology: Cuneiform Luwian ([adj.] gen.) ki-in-za-al-pa-aš-ši-iš ‘kind, gentle (?)’ = Hittite (2nd sg. present act.) ge-en-zu-wa-i-ši ‘to treat gently, to be compassionate (towards), to be kind’, ([adj.] nom. sg.) ge-en-zu-wa-la-aš, gi-in-zu-wa-la-aš ‘kindhearted, gentle’ < Proto-Anatolian *gen-s- ‘to treat gently, to be kind’ < traditional Pre-Anatolian ProtoIndo-European *’enH₁-(s-) ‘to treat gently, to be kind’. Non-Anatolian cognates include Latin gentīlis ‘belonging to one’s family, clan, tribe, or nationality’ (> Modern English gentle in the sense ‘considerate or kindly in disposition; amiable and tender’); The Reconstruction of the Proto-Anatolian Consonant System 57 Old English gecynd ‘nature, quality, character; race, species; origin, generation, birth’, gecynde ‘natural, native, innate; proper, fitting, lawful’, gecyndelic ‘kindly; natural, innate; proper’. Note: The meaning of the Cuneiform Luwian form is uncertain. The interpretation given here is that of the author, based upon the putative Hittite cognate. Thus, we can clearly see that the three series have different treatments before high front vowels in Luwian. Traditional Pre-Anatolian Proto-Indo-European *k and *’ were preserved, while *’h was lost. If, as is often claimed, the traditional Pre-Anatolian ProtoIndo-European plain voiced stops and voiced aspirates had merged in Proto-Anatolian, we would expect *’ to have been lost before high front vowels in Luwian as well, which does not appear to be the case. More evidence is to be found in the treatment of dentals initially before high front vowels and y in Hittite. In this case, the voiceless dental stop (*t) of traditional Proto-IndoEuropean became an affricate z (= ͜/ts/) in Hittite (cf. Sturtevant 1951:60—61; Melchert 1994:117; Kimball 1999:287—288), while the voiced aspirated dental stop (*dh) of traditional Proto-Indo-European was preserved (cf. Kimball 1999:292), as in (3 sg. pres. act. ?) ti-it-ta-i ‘to install, to assign’ < *dhi-dhH₁-oi-ei (cf. Kloekhorst 2008:881—882 — other reconstructions are possible). The traditional plain voiced dental stop (*d), on the other hand, became š in Hittite (cf. Kimball 1999:291—292) (but not in Cuneiform and Hieroglyphic Luwian or Palaic), as shown in the following example: Hittite (nom. sg.) ši-i-wa-az ‘day’ and (gen. sg.) ši-(i-)ú-na-aš ‘god’ = Palaic (nom. sg.) Ti-ya-az(-) name of the sun-god, (nom. sg.) ti-ú-na-aš ‘god’; Hieroglyphic Luwian Tiwat- name of the sun-god, (adj.) tiwatami- ‘bright, sunny’; Cuneiform Luwian (nom. sg.) Ti-wa-az name of the sun-god < Proto-Anatolian *tyēwat- < Pre-Anatolian ProtoIndo-European *dyēw- (cf. Benveniste 1962:8; Kloekhorst 2008:763—764). NonAnatolian cognates include: Sanskrit dyáu-ḥ ‘heaven, sky, day’, devá-ḥ ‘god’; Greek Ζεύς ‘Zeus’, δῖος ‘god-like, divine’; Armenian tiv ‘day’; Latin diēs ‘day’, deus ‘god’; etc. Here once again, had the traditional Pre-Anatolian Proto-Indo-European plain voiced stops and voiced aspirates merged in Proto-Anatolian, we would expect *d and *dh to have the same treatment before high front vowels and y in Hittite, which is not the case. Though the cumulative evidence from the Anatolian daughter languages, such as it is, points to three series of stops in Proto-Anatolian instead of two, we run into a potential problem, if, as stated above, there is no evidence for earlier voiced aspirated stops in either Proto-Anatolian or Pre-Anatolian Proto-Indo-European. It is wrong, lacking corroborating evidence, to project the Sanskrit-based reconstruction of the Proto-Indo-European consonant system back in time to Pre-Anatolian Proto-Indo-European. In other words, the Anatolian evidence should be taken at face-value. Consequently, given that Pre-Anatolian Proto-Indo-European most likely did not have voiced aspirated stops as part of its consonant inventory, then a solution to this problem is really not possible within the context of the traditional reconstruction of the Proto-Indo-European consonant system. However, a viable solution was proposed in three articles by Gamkrelidze. Essentially, based upon a 58 Chapter 4 careful examination and analysis of Hittite scribal conventions, Gamkrelidze proposed a radical new interpretation of Sturtevant’s Law. This new interpretation has major implications not only for Proto-Anatolian but also for Pre-Anatolian Proto-Indo-European. In the first article, Gamkrelidze (1968:91—92) maintains that Hittite was written in a cuneiform syllabary derived from a form of Old Akkadian cuneiform in use in Northern Syria in the beginning of the second millennium BCE. Now, the older cuneiform writing system, which was developed by the Sumerians, was not suited to rendering Akkadian, much less Hittite. In Old Akkadian, voiceless, voiced, and emphatic consonants were not differentiated in the writing system, though methods were gradually developed to represent most of the Akkadian phonological distinctions. This is important, for no attempt was ever made, even after Akkadian had introduced separate syllabograms to differentiate voiceless, voiced, and emphatic consonants, to modify the Hittite writing practices to make use of the same methods to note a voicing contrast in stops (cf. Melchert 2017:177; Kloekhorst 2008:21 and 2016:214—215). We must conclude, therefore, that the Hittite scribes did not feel that it was worthy of noting such a contrast, regardless of what the underlying phonetics may have been. What then, if anything, does medial double writing of stops indicate if not a voicing contrast? According to Gamkrelidze (1968:94), medial double writing of stops in Hittite was used as a means to indicate the presence of aspiration (Patri 2009 and 2019 reaches the same conclusion). In the two subsequent articles (1982 and 2008), Gamkrelidze elaborates further on this proposal. But, there is more. There are several troubling exceptions to Sturtevant’s Law in which words exhibiting medial double writing of stops in Hittite correspond etymologically to words in other Indo-European daughter languages with medial voiced stops. These alleged exceptions can best be explained as geminates (cf. Bomhard 2000 for discussion and examples). Taking into consideration Gamkrelidze’s findings as well as the apparent exceptions, Sturtevant’s Law is to be reinterpreted as follows: double writing of medial stops indicates stop plus something additional, that is, either aspiration or gemination, while single writing of medial stops indicates a plain stop pure and simple. It does not indicate either a voicing contrast or a fortis ~ lenis contrast as traditionally assumed. There is another, very important implication of Gamkrelidze’s theories. As stated at the beginning of this section, “there is absolutely no evidence whatsoever from any of the Anatolian daughter languages that a series of voiced aspirates existed in Proto-Anatolian”. Since it was clearly the feature of aspiration that was important to the Hittite scribes, this means that, had there been a series of voiced aspirates in Hittite, we would expect that they, too, would have been written double medially. However, this is not the case. The sounds in question (series 3 below) are consistently written single medially — no aspiration, no double writing; hence, no voiced aspirates. Given that three series of stops must be reconstructed for Proto-Anatolian, we can now add a third series to its consonant inventory on the basis of the new interpretation of Sturtevant’s Law by Gamkrelidze, namely, voiceless aspirates, thus: The Reconstruction of the Proto-Anatolian Consonant System Bilabial Dental Velar Labiovelar (1) Voiceless Aspirated pº tº kº k¦º (2) Plain Voiceless p t k k¦ 59 (3) Plain Voiced b d g g¦ The voiceless aspirated stops (column 1) correspond to the plain voiceless stops (*p, *t, *$, *k, *k¦) of traditional Proto-Indo-European, while the plain voiceless stops (column 2) correspond to the plain voiced stops (*b, *d, *’, *g, *g¦), and the plain voiced stops (column 3) correspond to the voiced aspirated stops (*bº, *dº, *’º, *gº, *g¦º). On the surface, it would appear that a sound-shift similar to what is assumed to have taken place in early Proto-Germanic — so-called “Grimm’s Law” (cf. Fulk 2018:102— 105) — is to be set up to account for the development of the above Proto-Anatolian consonant system from traditional Proto-Indo-European. But is this really warranted? No — not if the glottalic model of Proto-Indo-European consonantism (on which, see Bomhard 2016 and 2023.1:57—67; Gamkrelidze 1987 and 2003; Gamkrelidze—Ivanov 1972, 1973, and 1995.I:5—16; Hopper 1973, 1977a, 1977b, 1981, and 1982; Salmons 1993) is taken as the starting point. Then, all that is needed is the deglottalization of the glottalics to arrive at the Proto-Anatolian consonant system proposed above (so also Kloekhorst 2008a:25, but see below). Such a change would eliminate the need to assume a sound-shift. Thus, under the glottalic model of Proto-Indo-European consonantism (specifically, as formulated by Bomhard 2016), the voiceless aspirated stops (column 1) correspond to voiceless aspirated stops (*pº, *tº, *$º, *kº, *k¦º) in the glottalic model, while the plain voiceless stops (column 2) correspond to glottalized stops (*p’, *t’, *$’, *k’, *k’¦), and the plain voiced stops (column 3) correspond to plain voiced stops (*b, *d, *’, *g, *g¦). Viewed in this light, the Proto-Anatolian stop system reconstructed above provides strong support for the glottalic model of Proto-Indo-European consonantism as opposed to the traditional reconstruction. Furthermore, it eliminates the need to posit a series of voiced aspirates in its immediate Proto-Indo-Anatolian ancestor. Finally, it should be noted that this new interpretation is completely natural from a typological perspective, thus overcoming the problem of the typological implausibility of the traditional reconstruction of the Proto-Indo-European stop system. At this point, it is necessary to discuss a new theory proposed by Kloekhorst, which affects the reconstruction of series 2 above. In an unpublished paper (2014c), Kloekhorst argues that Luwian provides evidence for the preservation of glottalization, explicitly, preglottalization, in series 2 above. He presents more evidence in his 2016 article (Kloekhorst 2016:226—228), concluding: All in all, my reconstruction of the pre-Proto-Anatolian stop system is one of three series, namely one of voiceless long stops, one of pre-glottalized voiceless short stops, and one of plain voiceless short stops, corresponding to the PIE stops in the way shown in Tab. 2 (exemplified by the row of dental stops). 60 Chapter 4 Thus, for Pre-Proto-Anatolian, Kloekhorst reconstructs */t:/, */ˀt/, */t/ (= traditional ProtoIndo-European *t, *d, *dh, respectively), using the dentals for illustration (however, Simon 2019 presents compelling evidence against Kloekhorst’s interpretation — see above for details). Now, the cumulative evidence from all of the Indo-European daughter languages seems to indicate that the glottalized stops (that is, *p’, *t’, *$’, *k’, *k’¦) were originally postglottalized in all positions in Proto-Indo-European, including Pre-Anatolian Proto-IndoEuropean. In my 2016 paper entitled “The Glottalic Model of Proto-Indo-European Consonantism: Re-igniting the Dialog”, I proposed that there may have been dialectal variation in the timing of glottalization in the glottalic series before their ultimate loss, and I cited the example of the American Indian language Sm̓algyax (also called Coast Tsimshian) spoken in northwestern British Columbia and southeastern Alaska as a possible typological parallel (Bomhard 2016:387). The following rules for the timing of glottalization may tentatively be postulated for at least some dialects of Proto-IndoEuropean: 1. If a glottalized segment occurs before a vowel, it is post-glottalized: /C̓/ → [Cˀ]/__V. 2. If a glottalized segment occurs after a vowel, it is pre-glottalized: /C̓/ → [ˀC]/V__. 3. If a glottalized segment occurs word final before pause, it is deglottalized and unreleased: /C̓/ → [C˺]/__#. 4. If a glottalized segment occurs between vowels, it is post-glottalized if the accent falls on the second vowel: /C̓/ → [Cˀ]/V__V́ . 5. If a glottalized segment occurs between vowels, it is pre-glottalized if the accent falls on the first vowel: /C̓/ → [ˀC]/V́__V. Notes: 1. C̓ = any glottalic (/p’/, /t’/, /k’/, /k’¦/). 2. Rule no. 2 would account for Winter’s Law in Balto-Slavic. 3. Rule no. 2 would account for the “vestjysk stød” in the western dialects of Danish and preaspiration in West Scandinavian, if these are not recent developments, as some have maintained. 4. Kloekhorst has recently (2016:226—228) proposed that the glottalics became preglottalized in Proto-Anatolian and that the glottalization was eventually lost: *t’, *k’, *k’¦ → *’t, *’k, *’k¦ → *t, *k, *k¦. The above rules would account for Kloekhorst’s views. Setting up the above rules would account for Kloekhorst’s views on pre-glottalization. It should also be noted that such a scenario is hinted at by Salmons (1993:24) but not elaborated upon. Thus, incorporating Kloekhorst’s views, the Proto-Anatolian system of stops may now be fully reconstructed as follows: The Reconstruction of the Proto-Anatolian Consonant System Voiceless Aspirated pº tº kº k¦º Bilabial Dental Velar Labiovelar Glottalized p’ t’ k’ k’¦ 61 Plain Voiced b d g g¦ Note: The velar series may be assumed to have had non-phonemic palatalized allophones in certain environments (cf. Bomhard 2023.1:119—122). These allophones became phonemic in Luwian (cf. Melchert 2017:176) as well as in several of the nonAnatolian Indo-European daughter languages — the so-called “satǝm” languages. 4.3. Proto-Anatolian Consonant System In a paper published in Wekwos in 2019 (Bomhard 2019b), I proposed that the ProtoAnatolian stop system be reconstructed as follows, exactly as established in this chapter: Stops: Bilabial Voiceless aspirated: /pº/ 36 Glottalized (ejectives): /p’/ Plain voiced: /b/ Dental Velar35 Labiovelar /tº/ /t’/ /d/ /kº/ /k’/ /g/ /k¦º/ /k’¦/ /g¦/ In addition, the following phonemes must be reconstructed for Proto-Anatolian: Affricate: Fricatives: Nasals: Liquids: Glides: /͜tsº/ /s/ /m/ /w/ /n/ /r/, /l/ /x/ (= H2) /ɣ/ (= H3) /x¦/ (= H2¦) /ɣ¦/ (= H3¦) /y/ 4.4. Concluding Remarks We have now come to the end of our study of Proto-Anatolian consonantism. As noted in Chapter 1: General Introduction, I intentionally did not discuss all of the issues regarding Anatolian phonology — the standard handbooks given in the references should be consulted for more information regarding inner-Anatolian developments. I also did not The velar series may be assumed to have had non-phonemic palatalized allophones in certain environments (cf. Bomhard 2023.1:119—122). These allophones became phonemic in Luwian (cf. Melchert 2017:176) as well as in several of the non-Anatolian Indo-European daughter languages — the so-called “satǝm” languages. 36 As previously noted, Kloekhorst (2016:226—228) considers this series to have been pre-glottalized in Proto-Anatolian. 35 62 Chapter 4 discuss the glides or the nasals and liquids. Rather, to repeat, I have explored the important, though generally ignored, role that the older Anatolian daughter languages have to play in the reconstruction of the Proto-Indo-Anatolian consonant system, concentrating on the system of stops in this chapter. The laryngeals are not treated here — see Chapter 3 for details on the laryngeals. In this chapter, I have tried to show that the consonant system reconstructed for ProtoAnatolian can best be explained on the basis the glottalic model of Proto-Indo-European (more accurately, Proto-Indo-Anatolian) consonantism rather than the traditional, Sanskritbased reconstruction. ■ The Reconstruction of the Proto-Anatolian Consonant System 63 APPENDIX1: CORROBORATING EVIDENCE: LINGUISTICS As is well known, the glottalic model of Proto-Indo-European consonantism (the “Glottalic Theory”) has not escaped criticism. Though it generated strong interest in the scholarly community when it was first proposed, the Glottalic Theory is now mostly ignored, except among the Indo-European scholars working in Leiden, The Netherlands. It is usually only mentioned in passing in the standard handbooks. In 2016, I published a paper in Slovo a slovesnost in which I addressed all of the criticisms that have been lodged to date against the Glottalic Theory. Most of the criticisms were fairly easy to refute, though a minority were a bit more challenging. In the end, not a single criticism was able to withstand scrutiny — all were rejected. The glottalic model of Proto-Indo-European consonantism was initially proposed to rectify the typological implausibility of the traditional, Sanskrit-based reconstruction, and it was based strictly on the analysis of Indo-European evidence, though typological data were used in conjunction with this analysis to demonstrate that the glottalic model represented a more natural reconstruction, with numerous parallels among attested languages. Over the past three decades or so, especially after the collapse of the Soviet Union and the opening of new opportunities for archeologists, enormous progress has been made in substantiating the Kurgan Theory, propounded especially by the late Marija Gimbutas, according to which the homeland of the speakers of Proto-Indo-European was located on the European steppes to the north of and between the Black and Caspian Seas. So much progress has been made recently that all other theories regarding the Indo-European homeland have now been largely abandoned. At the time the Glottalic Theory was proposed, corroborating evidence from neighboring or distantly-related languages was not brought into the picture, and such evidence has not figured prominently in discussions about Proto-Indo-European, except for some tantalizing hints in the work of Frederik Kortlandt. That gap has now been filled — on Friday, 10 July 2015, I presented a paper entitled “The Origins of Proto-IndoEuropean: The Caucasian Substrate Hypothesis” at a conference held at Leiden University, The Netherlands, named “The Precursors of Proto-Indo-European: The Indo-Hittite and Indo-Uralic Hypotheses”. The basic premise of this paper is that, around 5,000 BCE, PreIndo-Europeans migrated southwestward from the area north of the Caspian and Aral Seas to the northeastern shores of the Black Sea, where they encountered and mingled with speakers of primordial Northwest Caucasian languages. The prolonged period of contact that took place between these two language groups gave rise to the form of Proto-IndoEuropean traditionally reconstructed in the standard handbooks. A modified and greatly expanded version of that paper subsequently appeared (2019) in the Journal of IndoEuropean Studies. In that paper, among other things, I presented lexical evidence for language contact between Northwest Caucasian and Proto-Indo-European, proposing nearly 200 lexical parallels (as of 2023, 305 lexical parallels), arranged into semantic fields. These lexical parallels represent borrowings either from Proto-Indo-European into Northwest Caucasian or from Northwest Caucasian into Proto-Indo-European. 64 Chapter 4 The most important finding from that paper, as it relates to the topics under discussion here, is that corroborating evidence providing strong support for the glottalic model of Proto-Indo-European consonantism has now been uncovered on the basis of prehistoric language contact between Proto-Indo-European and Northwest Caucasian. The following sound correspondences can be provisionally established between ProtoIndo-European, Common Abkhaz, and Proto-Circassian (these sound correspondences are based upon the examples given in Bomhard 2019 [expanded version 2023]): Indo-European *pº *tº *kº *k¦º *pºVs*pºV͜H₂*p’ *t’ *k’ *k’¦ *bº *bºVs*dº *gº *g¦º *s *H₁ (= *ʔ) *H₄ (= *h) *H₂ (= *͜ħh) (< *ħ) *H₃ (= *͜ʕɦ) (< *ʕ) *w *y *m *n *n̥ *l *r *a, *e, *o *i, *u Common Abkhaz *p *t, *c, *ć, *č, *čʹ *k, *kʹ, *q *kº, *qº *psV*px̌V*p’ *t’, *c’, *ć’, *č’, *č’ʹ *k’, *k’ʹ, *q’, *q’ʹ *k’º, *q’º *b *bzV-, bžV*d, *ʒ, *ʒ́, *ǯ, *ǯʹ *g, *gʹ, *ɣ, *ɣʹ *gº, *ɣº *s, *ś, *š, *šʹ, *z, *ź, *ž, *žʹ *Ø *Ø *ħ, *x̌, *x̌ʹ *ɦ (< *͜ʕɦ < *ʕ) *w *j *m *n *a *l *r *a, *ǝ *ǝ Proto-Circassian *pº, *p: *tº, *t:, *cº, *c:, *čº, *č:, *čºʹ, *č:ʹ *kº, *k:, *qº, *q: *kºº, *k:º, *qºº, *q:º *PsV*p’ *t’, *c’, *č’, *č’ʹ *k’, *q’ *k’º, *q’º *b *PzV*d, *ʒ, *ǯ, *ǯʹ *g, *ǧ, *ɣ *gº, *ǧº *s, *ś, *š, *šº, *šºʹ, *š:, *š:ʹ, *z, *ž, *žʹ *Ø, *h *Ø *ħ, *x, *x̌ *ĝ (< *ʕ) *w *y *m *n *a *l, *λ *r *a, *ǝ *ǝ The Reconstruction of the Proto-Anatolian Consonant System 65 By way of comparison, it may be noted that Chirikba (2016: 9—11) reconstructs the early Proto-Northwest Caucasian phonological system as follows (his transcription has been retained): Consonants: b d pº tº p’ t’ g ɢ kº qº k’ q’ ʔ ʒ ǯ Ł c č ƛ i e ü ö c’ č’ ƛ’ z ž L ĝ γ ʕ s š λ x χ H m n r l w j Vowels: ə a u o Notes: 1. Cf. Colarusso (1989:28) for a slightly different reconstruction of the Proto-Northwest Caucasian phonological system. 2. Cf. Colarusso (1975 and 2014) for a comprehensive description and analysis of the phonological systems of the various Northwest Caucasian daughter languages. 3. Cf. Colarusso (1992) for a comparison between Proto-Indo-European and Northwest Caucasian. See also Bomhard 2019a for a positive assessment of and expansion upon Colarusso’s views. It may be noted here that I strongly suspect that early North Caucasian languages were once more widely spread geographically. As a consequence of Indo-European expansions (and later Turkic incursions), the North Caucasian languages have been replaced over much of that area and have been slowly pushed back to their mountainous retreat in the Caucasus Mountains, where, thankfully, they survive until the present day. ■ APPENDIX 2: CORROBORATING EVIDENCE: ARCHEOLOGY AND GENETICS  There has been a tremendous amount of research appearing in print recently on the continuing advances in the gathering and analysis of archeological and genetic (aDNA) evidence from the areas surrounding the Black and Caspian Seas. For example, here is the introductory synopsis of one recent paper by Alexey Nikitin and Svetlana Ivanova (2022): 66 Chapter 4 The appearance of steppe genetic ancestry in Europe in the 3rd millennium BCE coincided with the beginning of a new cultural and economic era dominated by pastoralist economy and progressively more centralized social institutions. These genetic and socio-cultural elements were brought to Europe by the descendants of Eneolithic inhabitants of the Ponto-Caspian steppe. The details of the formation of the steppe genetic package and the identity of the contributors of its genetic components remain poorly understood. We propose that steppe genetic ancestry, as well as the cultural attributes that characterize the Early Bronze Age steppe pastoralists such as the Yamna(ya) (Pit Grave) culture complex, formed as the result of activities associated with the function of the circum-Pontic trade network. A millennium-long association among the Eneolithic cultures of the Ponto-Caspian steppe and forest-steppe, the eastern Balkan cultures of west Pontic, and populations of the Caucasus and northeast Anatolia, led to the integration of the elements of genetics, subsistence strategies, material culture, and worldview, to produce the foundation of a novel genetic and socio-cultural phenomenon by the last third of the 4th millennium BCE. Nikitin—Ivanovna (2022) conclude: These trade exchanges also promoted the formation of the PCS ancestry during the second half of the 5th millennium BCE, through interactions between the steppe inhabitants and their partners in the trade of copper and other material goods. This genetic ancestry, combining EHG, CHG and, to varying extent, Neolithic Anatolian/Levantine farmer components, became the genetic signature of the Yamna people in the EBA, which spread throughout Europe by the middle of the 3rd millennium BCE. It is also conceivable that the Indo-European language group was formed within this circum-Pontic trade network and involved linguistic elements of the participant groups and cultures of Anatolia, eastern Balkans, the Ponto-Caspian steppe and adjacent forest steppe, and the Caucasus. Other researchers include David Reich, David W. Anthony, Iosef Lazarides, and many others, especially several prominent Ukrainian archeologists and geneticists — and not to forget, once again, the late Marija Gimbutas. As noted above, in 2019, I published a rather long paper in the Journal of Indo-European Studies (vol. 47, nos. 1/2, pp. 9-124) entitled “The Origins of Proto-Indo-European: The Caucasian Substrate Hypothesis”. I subsequently (2023) expanded that paper into a book (146 pp.) under the title Prehistoric Language Contact on the Steppes: The Case of Indo-European and Northwest Caucasian. Both the paper and the book are available on academia.edu for free download. These works provide the linguistic corroboration for the growing body of archeological and genetic evidence pointing to the Pontic-Caspian steppes as the homeland of the Indo-Europeans. That is to say, this is where Proto-Indo-European (and Proto-Indo-Anatolian) formed and from which Indo-European-speaking tribes later migrated westward into Europe proper, eastward into Western China (Xinjiang), and Southeastward into Iran and India.   Chapter 5 The Reconstruction of the Proto-Indo-Anatolian Vowel System 5.1. Proto-Anatolian Vowel System According to Kimball (2017:249—251), the following vowels are to be reconstructed for Proto-Anatolian (see also Melchert 2017:176 2020b:265; Kloekhorst 2008a:17 — Kloekhorst does not reconstruct Proto-Anatolian */æ:/): /i/, /i:/ /e/, /e:/ /æ:/ /u/, /u:/ /o/, /o:/ /a/, /a:/ I would reconstruct a similar set of vowels for Proto-Anatolian, except for */æ:/ and */o/ ~ */o:/ (see below). Kimball maintains that */æ:/ developed from earlier *eH₁ (see also Melchert 2017:176 and 2020b:265). Kimball does not provide justification for reconstructing */o/ ~ */o:/ for Proto-Anatolian. However, Melchert (1992:186, 1994:291—294, and 2017:176) justifies the reconstruction of Proto-Anatolian */o/ ~ */o:/ on the basis of developments in Lycian. Melchert claims that Lycian /e/ comes, in part, from earlier */o/ and provides several examples. In my opinion, we are dealing with specific Lycian developments here, inasmuch as there is absolutely no evidence from Cuneiform and Hieroglyphic Luwian for /o/ distinct from /a/, claims to the contrary notwithstanding — the much later Lycian belongs to the Luwian branch of Anatolian. The vowels */a/ and */o/ of traditional Proto-Indo-European merged in Hittite and Palaic (*a, *o > a), while the */e/, */a/, and */o/ of traditional Proto-Indo-European merged in Cuneiform and Hieroglyphic Luwian (*e, *a, *o > a). It should be noted that Kloekhorst (2008a:35—60) provides plausible evidence for the existence of /o/ in Hittite.2 A careful review of the evidence provided by Kloekhorst, however, indicates that this was strictly a Hittite development and should not be projected back to Proto-Anatolian — it is found in specific environments and under specific conditions (see table below; Kloekhorst also gives a comprehensive list of supporting examples). It is definitely not descended from the */o/ of traditional Proto-Indo-European, which consistently appears as /a/ in Hittite, as well as in Palaic and Cuneiform and Hieroglyphic Luwian. Kloekhorst (2008a:58—59) summarizes his findings as follows: On the basis of the treatments above, we arrive at the following overview of the interconsonantal reflexes of PIE *ou, *eu and *u. Note that when -ḫ- is the preceding consonant, the reflexes of these are always /o/. 2 The same claim is made by Held—Schmalstieg 1969:95 (Hittite /o/ < /*au, *ou, *eu/) and Held—Schmalstieg—Gertz 1988:6. See also Melchert 2020b:263—264. According to Melchert, “…the contrast is also valid for Palaic and Luwian with differences only in detail…” 68 Chapter 5 C_ḫ C_i C_k C_l C_m C_n OH NH C_p C_r C_s OH *ou /o/ /o/ /o/ /au/ *eu /o/ *u /o/ /ū/ /ū/ /u/ /u/ /au/ ↓ /ao/ /o/ /au/ /au/ /ū/ ↓ /ū/ /o/ /o/ /ū/ /u/ ↓ /o/ /o/ /o/ /u/ /au/ /o/ /ū/ /ū/ NH C_t C_? other /o/ < *CuHn ? /o/ < *Cóm ↓ ↓ /o/ /o/ /u/ < *ºC(o)ms /o/ < *Cóms ↓ ↓ /o/ /o/ Thus, eliminating the */æ:/ proposed by Kimball and the */o/ ~ */o:/ proposed by Kimball and Melchert, the Proto-Anatolian vowel system may be reconstructed as follows: /i/, /i:/ /e/, /e:/ /u/, /u:/ /a/, /a: / To complicate the situation, Proto-Indo-Anatolian must also have had non-apophonic */e/ and */o/ inherited from Proto-Nostratic. On the basis of the available lexical evidence, it appears that non-apophonic */e/ remained in Proto-Anatolian, while non-apophonic */o/ merged with */a/. In contrast, non-apophonic */e/ and */o/ remained distinct in the IndoEuropean ancestor of the non-Anatolian daughter languages. The development of nonapophonic */o/ in both the Anatolian and non-Anatolian daughter languages may be illustrated by the following example: A. Anatolian: Hittite (nom. pl. ?) ḫa-a-u-e-eš ‘sheep’ (< [nom. sg.] *H3owi-s, with nonapophonic */o/ in the first syllable); Cuneiform Luwian (nom. sg.) ḫa-a-ú-i-iš ‘sheep’; Hieroglyphic Luwian (nom. sg.) OVIS.ANIMALhá-wá/í-i-sá /hawis/ ‘sheep’; Lycian (acc. sg.) χawã ‘sheep’. B. Other: Sanskrit ávi-ḥ ‘sheep, ram’; Greek ὄɩ̈ ς, οἶς ‘sheep’; Armenian hov-iw ‘shepherd’; Latin ovis ‘sheep’; Old Irish oí ‘sheep’; Gothic awēþi ‘herd of sheep’; Old English ēow, ēaw, ēw ‘sheep’, ēowu, ēowe ‘ewe’, ēowd, ēowde ‘herd of sheep’; Old Frisian ei ‘ewe’; Old Saxon ewwi ‘ewe’; Dutch ooi ‘ewe’; Old High German ouwi, ou ‘ewe’, ewit, owiti ‘herd of sheep’; Lithuanian avìs ‘sheep’, ãvinas ‘ram’; Latvian avs ‘sheep’; Old Church Slavic ovьca (< *owi-kā) ‘sheep’, ovьnъ ‘ram’; Russian ovcá [овца] ‘sheep, ewe’; The Reconstruction of the Proto-Anatolian Vowel System 69 Czech ovce ‘sheep’; Polish owca ‘sheep’; Tocharian B eye ‘sheep’, ā(u)w ‘ewe’, aiyye ‘ovine, pertaining to sheep’. Notes: 1. Kimball (1999:142) reconstructs initial *h2- here, but this interpretation is rejected by Kortlandt (2001:2). Kloekhorst (2008a:337—338) reconstructs *h3eu̯ i- with initial *H3- as do Beekes (2010.II:1060—1061) and Derksen (2008:384 and 2015:74). 2. In my opinion, the */o/ in the initial syllable is original here, that is to say, it is inherited from Proto-Nostratic (cf. Bomhard 2023.3:930—931, no. 785, Proto-Nostratic (n.) *ʕuw-a (~ *ʕow-a) ‘herd of small animals, sheep and goats’). 3. Lycian (acc. sg.) χavã ‘sheep’ shows that initial *H3- became χ- in Lycian. However, according to Kimball (1987) and Melchert (1994a:72), initial *H3- was lost in Lycian. (Cf. Bomhard 2023.3:930—931, no. 785, Proto-Nostratic (n.) *ʕuw-a (~ *ʕow-a) ‘herd of small animals, sheep and goats’.) 4. This example shows that initial *H3- (Kuryłowicz’s *›; Sturtevant’s *¦) became h- in Armenian, as did initial *H2- (Kuryłowicz’s *š; Sturtevant’s *x). 5.2. History of the Reconstruction of the Proto-Indo-European Vowel System There is a strong temptation to project the Proto-Anatolian vowel system reconstructed above back into its Proto-Indo-Anatolian ancestor. However, is this really justified? Is it not possible instead to see hints of an even earlier system? Indeed, it is. Let us review the evidence (this is a slightly modified version of Bomhard 2023.1:123—134). The first attempt to reconstruct the Proto-Indo-European vowel system was made by August Schleicher. Schleicher’s system was as follows (1876:11): Original Vowel a-grade i-grade u-grade a i u First Increment Second Increment a + a = aa a + i = ai a + u = au a + aa = āa a + ai = āi a + au = āu Even though Schleicher’s system, which was modeled after that of Old Indic, was able to account for many of the developments found in the Indo-European daughter languages, there remained many unsolved problems, and his system did not endure the onslaughts of a series of brilliant discoveries made in the seventies of the nineteenth century by a younger generation of scholars, the so-called “Neogrammarians” (Junggrammatiker). Perhaps the most important discovery of the Neogrammarian period was the “Law of Palatals” (cf. Collinge 1985:133—142), according to which an original *k, for example, developed into c in Old Indic under the influence of a following *ē̆, *ī̆, or *y. This discovery firmly established the primacy of the vowel systems found in the European daughter languages and proved that the Indo-Iranian system had resulted from an innovation in which original *ē̆, *ō̆, and *ā̆ had merged into *ā̆. Also important was the demonstration by the Neogrammarians that the Indo-European parent language had syllabic liquids and nasals. 70 Chapter 5 According to the Neogrammarians (cf. Hübschmann 1885; Brugmann 1904:52 and 66—99), Proto-Indo-European had the following vowel system: Monophthongs: e ē o ō a ā i ī u ū ə Diphthongs: ei̯ ēi̯ oi̯ ōi̯ ai̯ āi̯ əi̯ eu̯ ēu̯ ou̯ ōu̯ Semivowels: i̯ u̯ (j ?) au̯ āu̯ əu̯ Brugmann (1904:52) also reconstructs the following syllabic liquids and nasals: l̥ l̥̄ r̥ r̥ ̄ m̥ m̥̄ n̥ n̥̄ n̥̑ n̥̑̄ ™ › Throughout the greater part of the twentieth century, the Neogrammarian view was steadily attacked. It was dealt its first major blow in 1927 with Kuryłowicz’s demonstration that one of de Saussure’s “coefficients sonantiques” was preserved in Hittite. In one fell swoop, the so-called “original” long vowels (as well as the long syllabic liquids and nasals) were eliminated as was *a, which was taken to result from *e when next to an “a-coloring” laryngeal. The next to go were the diphthongs, which were reanalyzed as clusters of vowel plus non-syllabic resonant and non-syllabic resonant plus vowel (cf. Lehmann 1952:10— 14). The independent status of *i and *u had early been questioned by Meillet (1964:118— 122), who regarded them as the syllabic forms of *y (*i̯ ) and *w (*u̯ ), respectively. Finally, a strict adherence to Hirt’s Ablaut and accentuation theories made it possible to eliminate apophonic *o, which was taken to result from an earlier *e when the accent was shifted from the *e to another syllable (cf. Burrow 1973:112—113; Hirt 1921—1927.II:173—179; Lehmann 1952:109—110). By applying all of these theories, it became possible to reduce the Proto-Indo-European vowel system to a single member: *e. It should be made clear that this extreme view was never universally accepted. In fact, it was vigorously attacked by several scholars, including Roman Jakobson (1971[1957]:528), who soberly noted: “The one-vowel picture of Proto-Indo-European finds no support in the recorded languages of the world.” See also Trubetzkoy 1969:96. In 1967, Szemerényi, relying heavily on typological data to support his arguments, reinstated all of the vowels reconstructed by the Neogrammarians: e ē o ō a ā i ī u ū ə Szemerényi (1967:97, fn. 91), however, ignores the diphthongs, “whose phonemic status is disputed”. I fully support Szemerényi’s views on the vowels and would reconstruct an identical system for the Proto-Indo-European antecedent of the non-Anatolian daughter languages. The Reconstruction of the Proto-Anatolian Vowel System 71 Proto-Indo-European, as also, for example, Proto-Kartvelian, Northwest Caucasian, and Proto-Semitic, was characterized by an interchange of vocalic elements that could occur in any syllable. This interchange, which is commonly called “Ablaut”, “apophony”, or “vowel gradation”, was partially correlated with the position of the accent and with distinctions between grammatical relationships (cf. Burrow 1973:108—117). According to the traditional reconstruction, the fundamental vowel was *e, which could be changed to *o under certain conditions. Under other conditions, however, the vowel could either be reduced or even lost altogether. Finally, the position of the fundamental vowel could change — this type of alternation is known as “Schwebeablaut” (for details, cf. Anttila 1969). An example here would be *k’en-u ‘knee’ (cf. Hittite gi-e-nu ‘knee’; Latin genu ‘knee’), as opposed to *k’n-ew- also ‘knee’ (cf. Gothic kniu ‘knee’; Old Icelandic kné ‘knee’; Old English cnēo ‘knee’; Old Frisian kniu, knē, knī ‘knee’; Old Saxon knio ‘knee’; Old High German kneo, knio ‘knee’). Several gradation series are traditionally distinguished, and the general scheme may be represented as follows (cf. Beekes 1995:164—167; Brugmann 1904:138—150; Buck 1933:106—117; Clackson 2007:71—75; Fortson 2010:79—83; Hirt 1900 and 1921— 1927, vol. II; Hübschmann 1885:71—180; Kuryłowicz 1956 and 1968:199—333; MeierBrügger 2003:144—152; Meillet 1964:153—168; Sihler 1995:108—135; Szemerényi 1996:83—93; Watkins 1998:51—53): I. Short Vowel Gradation: Lengthened-Grade Normal-Grade Reduced-Grade Zero-Grade A. ē~ō e~o Ø B. ēy ~ ōy ēw ~ ōw ēm ~ ōm ēn ~ ōn ēl ~ ōl ēr ~ ōr ey ~ oy ew ~ ow em ~ om en ~ on el ~ ol er ~ or a~o ay aw ə (traditional schwa secundum, *ь or *e) i, əyV (> iyV) u, əwV (> uwV) m̥ , əmV (m̥ mV) n̥ , ənV (n̥ nV) l̥ , əlV (l̥ lV) r̥ , ərV (r̥ rV) ə i, əyV (> iyV) u, əwV (> uwV) C. D. y w m n l r Ø y w II. Long Vowel Gradation: E. F. G. ē~ō ō ā~ō h̥ (traditional schwa primum, *ə) h̥ h̥ In this scheme, the most common vowel was *e, and the most common gradation pattern was the *e ~ *o contrast. The vowel *a was of relatively low statistical frequency and, at least according to Meillet (1964:154), did not take part in the regular gradation 72 Chapter 5 patterning. It should be pointed out, however, that several rare examples of an *a ~ *o contrast are attested in the non-Anatolian daughter languages, one probable example being: Greek ἄγω ‘to lead, to carry, to convey, to bring’ ὄγμος ‘any straight line: a furrow, path, etc.’ Colarusso (1981:499) has astutely observed: “…the PIE vowel system *e ~ *o is typologically utterly bizarre. Even adding *a to this system does not change this fact.” Perhaps the most typologically unusual thing about the Proto-Indo-European vowel system as traditionally reconstructed is indeed the great importance of the *e ~ *o Ablaut and the concomitant marginality of *a. Adding laryngeals only makes the system even more unusual since *a then becomes mostly (but not in every case!) a positional variant of *e. Rather, we would expect the relationship to be reversed. All languages surveyed by Crothers (1978:93—152) have the vowel /a/, and this vowel is consistently characterized by a high frequency of occurrence (cf., for example, the frequency counts given in Greenberg 1966:18—19). Moreover, in the Kartvelian languages, Northwest Caucasian languages, and Semitic languages, which also exhibit Ablaut either as an active process or as a relic of an earlier, fully functioning Ablaut process, the vowel /a/ is an integral part of the Ablaut system (cf. Gamkrelidze 1966:80—81 for Kartvelian; Colarusso 1981:499— 502 for Northwest Caucasian; and Kuryłowicz 1962 for Semitic). Clearly, if typological evidence is to have any meaning, there is something wrong with the traditional reconstruction of the Proto-Indo-European vowel system. Yet, if the Comparative Method is to have any validity, there must be some truth to that reconstruction. This seeming conflict can be resolved quite easily, I believe. We can consider the traditional reconstruction to be mainly correct, but only for that form of Proto-IndoEuropean spoken immediately prior to the emergence of the non-Anatolian daughter languages, that is, what I call “Disintegrating Indo-European”. The vowel system of this form of Proto-Indo-European is by no means ancient and is the end product of a long, complicated evolution. The earliest Proto-Indo-European (Proto-Indo-Anatolian) vowel system may thus be reconstructed as follows: Vowels: Also the sequences: i e iy iw ə~a uy uw u o ey ew oy ow əy ~ ay əw ~ aw Notes: 1. */e/ and */o/ represent so-called “non-apophonic” */e/ and */o/. These vowels were not involved in the qualitative Ablaut traditionally reconstructed (see below). 2. There may have been corresponding long vowels as well, but this is uncertain. Some words, such as *mā- ‘mother’ (no laryngeal! — not *meA-!), must always have had a long vowel (cf. Jakobson 1971[1960]). The Reconstruction of the Proto-Anatolian Vowel System 73 I follow Pulleyblank (1965:86—101) in his reinterpretation of the *e ~ *o Ablaut of traditional Proto-Indo-European as an *ə ~ *a Ablaut. Pulleyblank was not the first to suggest such a reinterpretation. Pulleyblank mentions that a similar Ablaut pattern exists in Kabardian. Colarusso (1981:499—501) proposes a similar reinterpretation and also discusses possible typological parallels with the Northwest Caucasian languages. See also Matasović 2003. According to Hirt (1921:172—199) and those who follow his theories (Burrow and Lehmann, for example), the oldest Ablaut alternation was the full-grade ~ zero-grade contrast. This alternation is assumed to have arisen at a time when the Proto-Indo-European (better, Proto-Indo-Anatolian) phonological system was characterized by a strong stress accent. This accent caused the weakening and loss of the vowels of unstressed syllables. This period may be called the Phonemic Stress Stage of Proto-Indo-European (cf. Lehmann 1952:111—112). At a later date, stress became phonemically non-distinctive, and ProtoIndo-European was characterized by an accent system based on pitch. This period may be called the Phonemic Pitch Stage of Proto-Indo-European (cf. Lehmann 1952:109—110). It was supposedly during this period that the *e ~ *o contrast came into being. *e is assumed to have been changed to *o when the accent was shifted from the *e to another syllable. These theories find support in the fact that the position of the accent is partially correlated with Ablaut patterning in both Greek and Old Indic. Counter-examples are usually explained as due to analogical developments or as later forms that came into being after the accent lost its ability to influence the vowels (cf. Burrow 1973:112). Though Kuryłowicz originally adhered to Hirt’s theories as well, he later (1956:36— 96) tried to show that the *e ~ *o contrast existed prior to the development of the full-grade ~ zero-grade contrast. Kuryłowicz argues that the numerous counter-examples with accented *o indicate that qualitative Ablaut was a morphological device in its own right and only superficially connected with the positioning of the accent. Moreover, he notes that, while vowel weakening and loss are closely tied to the accent, a change in vowel quality is primarily due to the environment — in other words, there is no cause-and-effect relationship between qualitative Ablaut and accentuation. These are convincing arguments and are the primary basis for my belief that qualitative Ablaut existed at the earliest reconstructable period of Proto-Indo-European. Moreover, Kuryłowicz’s views find support in the data from the other Nostratic languages (note here especially Greenberg 1990:125: “… the Indo-European e:o (i.e. e:a) Ablaut is very old and is part of a larger system of alternations which has correspondences in a number of other branches of Eurasiatic”). The development of *ə into *e, which must have occurred fairly early since it is already found in Hittite, is relatively easy to explain: *e was the normal allophone of *ə under stress. John Colarusso (personal correspondence) has informed me of a similar development in Ubykh and Circassian, where accented /ə/ > [e]. We may assume that *a had a rounded allophone in certain phonetic environments (cf. Colarusso 1981:500), perhaps next to labiovelars as well as when next to *w. In late Disintegrating Indo-European, these allophones were reapportioned, and apophonic *a was rephonemicized as *o. That this is an extremely late development is shown (A) by the fact that it had not yet occurred in the Anatolian languages and (B) by the widespread tendency of *a and *o to have identical reflexes in several of the non-Anatolian daughter languages. 74 Chapter 5 No doubt, the phonemicization of apophonic *o was facilitated by the presence of nonapophonic *o in the system. This does not mean, however, that *a was totally eliminated. In fact, the vowel *a must be reconstructed as an independent phoneme distinct form *o for Disintegrating Indo-European (cf. Szemerényi 1964a:2—9, 1967:83—84, and 1996:135—136). Disintegrating Indo-European *a had several sources: First, it arose from the lowering and coloring of *e (< *ə́) when next to a-coloring laryngeals. Next, it was found in a small number of relic forms that somehow escaped conversion to the regular *e ~ *o Ablaut patterning as traditionally reconstructed. Finally, in was retained in loanwords from other languages. (Cf. Kümmel 2012:306—310 for a somewhat different approach; see also Hovers 2021.) The origin of the Proto-Indo-European long vowels has always been enigmatic. Many theories have been proposed, none of which has been completely convincing. One thing seems certain, though: the long vowels developed over a long period of time and had several causes. The evidence for the existence of original long vowels is meager at best, and there seems little reason to suppose that long vowels existed in Pre-Proto-Indo-European. Rather, long vowels may be assumed to have arisen solely in Proto-Indo-European proper. First, long vowels arose from the contraction of two short vowels. Though probably not frequent in the earlier stages of development, contraction became increasingly important, especially in the later stages of the Indo-European parent language and the earlier stages of the nonAnatolian daughter languages, when the upheavals caused by the loss of whole classes of phonemes — such as the laryngeals, for example — often brought two or more previously separated vowels into contact. Long vowels also arose from the monophthongization of diphthongs and from the lengthening of short vowels to compensate for the loss of a following phoneme. The most important cause of compensatory lengthening was the loss of preconsonantal laryngeals after short vowels in Disintegrating Indo-European. Finally, long vowels arose by means of the analogical process known as “vṛddhi” (cf. Burrow 1973:199—291; Kuryłowicz 1968:298—307). In reconstructing the Proto-Indo-European phonological system, the vowels *i and *u are usually treated as allophones of *y (*i̯ ) and *w (*u̯ ) respectively and are classed with the resonants *m/*m̥ , *n/*n̥ , *l/*l̥ , *r/*r̥ (cf. Lehmann 1952:10—14; Meillet 1964:105— 126). However, as pointed out by Szemerényi (1967:82), the patterning of these sounds is not entirely parallel. For the earliest form of Proto-Indo-European (Proto-Indo-Anatolian), *i and *u should, in fact, be considered as independent phonemic entities and should be classed with the vowels rather than the resonants. The glides *y and *w should also be considered as independent phonemes during the early stages of development within ProtoIndo-European. At a later date, however, after various sound changes had taken place, the patterning had been modified in such a way that *i ~ *y and *u ~ *w were mostly in complementary distribution (cf. Gamkrelidze—Ivanov 1995.I: 137—138; Schmitt-Brandt 1973:79—91; Szemerényi 1996:136. As further noted by Gamkrelidze—Ivanov (1995.I: 141—142), the changes alluded to above brought about a major restructuring of the functional role of the high vowels *i and *u (this version differs slightly from the English translation made by Johanna Nichols — here, the term “resonant” has been substituted for “sonant” [Russian сонант]): The Reconstruction of the Proto-Anatolian Vowel System 75 These phonetic and phonemic changes inevitably led to a split of the consonants into consonants proper and resonants (i̯ , u̯ , r, l, m, n). The resonants had syllabic and non-syllabic allophones depending on context, while the consonants proper had only non-syllabic allophones (and the vowels were exclusively syllabic in all positions). Consequently, resonants had the feature value [±syllabic], in contrast to consonants proper [-syllabic] and vowels [+syllabic]. These changes produced a major restructuring of the vowel system. The original high vowels i and u became identified with the syllabic allophones of i̯ and u̯ . Hence, they acquired the status of resonants, specifically, the syllabic allophones of resonants. These vowels had not formerly been in alternation with non-syllabic elements but were now pulled into the system of resonant alternations, which severed their connection to the class of vowels. This reanalysis of original i and u led, in some cases, to secondary full-grade formations, with V being inserted into roots in which i or u had originally been root vowels but which were now interpreted as zero-grade resonants. This apparently took place in the parallel markers of the Indo-European locative *-i/*-ei̯ . Of the two locative forms, the stressed *-i is clearly the older and reflects the ancient full-grade vowel *i: Gk. ποδ-ί, Skt. pad-í ‘in the foot’, dat-í ‘in the tooth’, janas-í ‘in birth’, Hitt. nepiš-i ‘in the sky’, beside the later Slavic nebes-i ‘in the sky’. Other full-grade forms in which the diphthongs ei̯ and eu̯ correspond to i and u in the zero-grade, often adduced as illustration of the full-grade ~ zero-grade alternation, may well represent later formations from roots with original *i and *u which were secondarily identified with syllabic resonants and reinterpreted as zero-grade. According to the traditional reconstruction, Proto-Indo-European is assumed to have had the following short diphthongs: ei̯ eu̯ oi̯ ou̯ ai̯ au̯ əi̯ əu̯ In the reduced-grade, the semivowels alone appear: i u Szemerényi (1996:141) notes that, while this looks good on paper, it is difficult to imagine the process that would have led to *i and *u in the reduced-grade. He points out that it most certainly could not have been due to a simple loss of *e, *o, and *a. The actual process leading to the appearance of *i and *u in the reduced-grade was probably along the following lines: A. After phonemicization of a strong stress accent, stress-conditioned weakening of the vowel to *ə (= traditional schwa secumdum, *ь or *e) in unstressed syllables. B. Assimilation of *ə to *i before *y and to *u before *w. C. Passage of *iy to *ī and of *uw to *ū. D. Shortening of *ī to *i and of *ū to *u. A typological parallel exists in Riffian Berber, where itawi-d ‘he brings’ developed from earlier *yəttawəy-dd, with both *yə and *əy > i (cf. Kossmann 2012:28). The same development may be observed in Kabardian (cf. Chirikba 1996a:52). 76 Chapter 5 This is only part of the story, however, since it focuses primarily on the developments affecting the Pre-Proto-Indo-European *əy ~ *ay and *əw ~ *aw. Pre-Proto-Indo-European also had the following sequences: *iy ~ *ey, *uy ~ *oy, *iw ~ *ew, and *uw ~ *ow, and these need to be considered as well. A summary of the developments is given below (only the beginning and end points are shown): əyC əyC əyV əyV əyC əyV > > > > > > eyC ayC eyV ayV iC iyV when stressed when stressed and preceded by an a-coloring laryngeal when stressed when stressed and preceded by an a-coloring laryngeal when unstressed when unstressed ayC ayV ayC ayV > > > > oyC oyV iC iyV when stressed when stressed when unstressed when unstressed iyC iyC iyV iyV iyC iyV > > > > > > īC ēC iyV eyV iC iyV when stressed when stressed and preceded by an a-coloring laryngeal when stressed when stressed and preceded by an a-coloring laryngeal when unstressed when unstressed eyC eyV eyC eyV > > > > ēC eyV iC iyV when stressed when stressed when unstressed when unstressed uyC uyC uyV uyV uyC uyV > > > > > > īC ? ēC ? iyV ? eyV ? iC iyV when stressed when stressed and preceded by an a-coloring laryngeal when stressed when stressed and preceded by an a-coloring laryngeal when unstressed when unstressed oyC oyV oyC oyV > > > > oyC oyV iC iyV when stressed when stressed when unstressed when unstressed əwC əwC əwV > > > ewC awC ewV when stressed when stressed and preceded by an a-coloring laryngeal when stressed The Reconstruction of the Proto-Anatolian Vowel System əwV əwC əwV > > > awV uC uwV when stressed and preceded by an a-coloring laryngeal when unstressed when unstressed awC awV awC awV > > > > owC owV uC uwV when stressed when stressed when unstressed when unstressed uwC uwC uwV uwV uwC uwV > > > > > > ūC ōC uwV owV uC uwV when stressed when stressed and preceded by an a-coloring laryngeal when stressed when stressed and preceded by an a-coloring laryngeal when unstressed when unstressed owC owV owC owV > > > > ōC owV uC uwV when stressed when stressed when unstressed when unstressed iwC iwC iwV iwV iwC iwV > > > > > > ūC ? ōC ? uwV ? owV ? uC uwV when stressed when stressed and preceded by an a-coloring laryngeal when stressed when stressed and preceded by an a-coloring laryngeal when unstressed when unstressed ewC ewV ewC ewV > > > > ewC ewV uC uwV when stressed when stressed when unstressed when unstressed 77 In addition to the sequences of vowel plus *y and *w, the earliest form of Proto-IndoEuropean also had sequences of *y and *w plus vowel. In unstressed positions, the vowel was first reduced to *ə. *ə was then assimilated to *i after *y, and the preceding *y was lost, while after *w, it was assimilated to *u, and the preceding *w was lost: yəC wəC > > yiC wuC > > iC uC The vowels *i and *u were converted into the corresponding glides *y and *w respectively when directly followed by another vowel: iV > yV 78 Chapter 5 uV > wV In a 1967 book devoted to the study of the Indo-European vowel system, Robert Schmitt-Brandt (1967:8—31 [2nd edition 1973]) concludes that it is necessary to assume a fundamental form in *i and *u for certain kinds of roots in Proto-Indo-European and to view the full-grade forms as secondary. This conclusion is reached on the basis of the observation that, as a general rule, when diphthongs are “reduced”, long monophthongs result and not, as traditionally assumed, short monophthongs. Support for this conclusion is to be found in root-nouns, which appear in the reduced-grade (Schmitt-Brandt cites *di$-, *wi$-, and *duk- [I would write *t’ikº-, *wikº-, and *t’ukº-, following the glottalic model of Proto-Indo-European consonantism] as examples), this being their original form. To have it the other way around, with *ey, *ew, etc. as the original forms, would lead, in Schmitt-Brandt’s opinion, to reduced-grade forms with *ī and *ū: *dey$- > *dī$-, *wey$> *wī$-, and *dewk- > *dūk-, etc. Schmitt-Brandt thus posits *i and *u as independent vowels in Proto-Indo-European and explains the full-grade forms in *ey, *ew, etc. as due to analogy. Finally, Schmitt-Brandt (1967:79—91) explains that, in an earlier period of Proto-Indo-European, *y and *w (he writes *i̯ and *u̯ ) were consonants in their own right and were not connected with the independent vowels *i and *u. Somewhat similar views are expressed by William F. Wyatt (1970:58 and fn. 24). The parts of Schmitt-Brandt’s theories outlined in the preceding paragraph make a lot of sense, at least on the surface. Other parts of his theories, however, have purposely been left out of the discussion since, at least in my opinion, they are less convincing (see here the review of Schmitt-Brandt’s book by Kuryłowicz 1969:41—49). What Schmitt-Brandt has correctly identified is the fact that, in certain specific instances, it is necessary to assume secondary full-grade forms. Schmitt-Brandt is also correct in seeing the vowels *i and *u as independent vowels at an early stage of development within Proto-IndoEuropean. Where his theories are mistaken, however, is in the assumption that the reduction of diphthongs can only lead to long monophthongs. While this is indeed a very common development, it is not the only possible outcome. Here, we can cite developments in the Romance languages: Classical Latin had both long vowels and short vowels along with three diphthongs, namely, ae, oe and au. In Vulgar Latin, length distinctions were lost, and the earlier long vowels were realized as closed vowels, while the earlier short vowels were realized as open vowels. At the same time, ae > open e [ę], oe > closed e [ẹ], and au > closed o [ọ] (though there is actually a great deal of variation in the development of au). In Balkan Romance, unstressed front vowels merged into [e], and unstressed back vowels merged into [u], except for [a] and [ị] (closed [i] < Classical Latin ī), which remained intact. (Cf. Mendeloff 1969:4—16 for details about the development of the vowels and diphthongs in the Romance languages.) Another problem with SchmittBrandt’s theories concerns the failure to recognize the fact that the latest period of ProtoIndo-European contained the remnants of multiple successive earlier periods of development. The reduction of diphthongs in unaccented syllables had a different outcome in the earliest period than in later periods — in the earliest period, short monophthongs resulted from the stress-conditioned weakening of diphthongs in unstressed syllables, while in later periods, when stress was no longer phonemically distinctive, long monophthongs resulted. The Reconstruction of the Proto-Anatolian Vowel System 79 In Post-Anatolian Proto-Indo-European, secondary diphthongs arose as follows: By this time, the *e ~ *o Ablaut patterning had assumed an important role in the emerging morphological system of Proto-Indo-European. In order to bring stems such as *t’ikº-, *wikº-, and *t’ukº- (traditional *di$-, *wi$-, and *duk-), and the like into line with the regular *e ~ *o Ablaut patterning as traditionally reconstructed, *e and *o were inserted before accented *i and *u, thus creating secondary diphthongs: Ø > e and o /___ { } in *t’íkº-, *wíkº-, *t’úkº-, yielding *t’éykº-, *wéykº-, *t’éwkº-, etc. í ú The development of secondary diphthongs was restricted to certain specific grammatical environments (such as the singular indicative verbal forms) — that is to say, not every accented *í and *ú was affected (cf., for example, forms such as Sanskrit nom. sg. agní-ḥ ‘fire’ and sūnú-ḥ ‘son’ or Hittite nom. sg. šal-li-iš ‘glorious’ and a-aš-šu-uš ‘good’, which must always have had *í and *ú). We may note at this point that secondary full-grade forms could also be created from syllabic resonant stems when the accent was shifted to the stem from another syllable (*Cm̥ ́ - > *Cém-, *Cn̥ ́ - > *Cén-, *Cl̥ ́- > *Cél-, *Cr̥ ́ > *Cér-). The picture is still not complete, though, for we must also consider how laryngeals fit into the picture: The loss of laryngeals in sequences such as *eHiC, *eHuC, and the like resulted in short diphthongs when accented (*éHiC > *eyC and *éHuC > *ewC, etc.) — the preceding vowel was definitely not lengthened — but long monophthongs when unaccented (*əHiC > *əyC > *iyC > *īC and *əHuC > *əwC > *uwC > *ūC, etc.). Since these changes were later than the changes previously described and since stress was no longer phonemically distinctive, the resulting long monophthongs were not shortened to *i and *u respectively. In sequences such as *eHaxC and the like, the loss of the intervocalic laryngeal first produced a sequence of two short vowels. These vowels were then contracted to form a long vowel: *eHaxC > *eaxC > *ēC, *oHaxC > *oaxC > *ōC, *aHaxC > *aaxC > *āC. We may close this section by mentioning one last point. The numerous cases of vṛddhied stems in the non-Anatolian daughter languages — especially Old Indic — appear to represent a contamination of types *CēC and *CōC with types *CeyV-/ *CoyV- and *CewV-/*CowV-, producing the new types *CēyV-/*CōyV- and *CēwV-/*CōwV- (cf. Schmalstieg 1973:108). 5.3. Examples Selected Hittite declensions: 1. at-ta-aš ‘father’ (common gender a-stem; mobile accent; a-grade generalized throughout the paradigm; nursery word): Singular: Nominative Hittite at-ta-aš, ad-da-aš Indo-Anatolian *H₁átºtºas Indo-European *H₁átºtºos 80 Chapter 5 Accusative at-ta-an, ad-da-an Vocative (no accent) at-ta Genitive at-ta-aš, ad-da-aš Dative-Locative at-ti *H₁átºtºam *H₁átºtºom *H₁atºtºa *H₁atºtºás *H₁atºtºo *H₁atºtºós *H₁atºtºí Ablative (sg./pl.) at-ta-az Instrumental (sg./pl.) Allative / Directive *H₁atºtºátºs *H₁atºtºí ~ *H₁atºtºéy / *H₁atºtóy *H₁atºtºótºs *H₁átºtºes *H₁átºtºm̥ s *H₁átºtºes *H₁átºtºm̥ s *H₁atºtºás *H₁atºtºás *H₁atºtºós *H₁atºtºós Plural: Nominative Accusative Genitive Dative-Locative at-ti-e-eš at-tu-uš, ad-du-uš ad-da-aš ad-da-aš Cognates: A. Anatolian: Cuneiform Luwian (nom. sg.) ta-a-ti-iš ‘father’; Hieroglyphic Luwian (nom. sg.) tá-ti-sa ‘father’; Lycian (nom. sg.) tedi ‘father’; Lydian taada- ‘father’. All from Proto-Indo-Anatolian *tºat’a/i- (traditional *todo-) ‘father’. B. Other: Assamese ātā term of respect when addressing an older male; Sinhalese ātā ‘grandfather’; Sindhi ado ‘brother’; Greek ἄττα ‘daddy’; Latin atta term of respect when addressing an older male, ‘father, grandfather’; Old Irish aite ‘father’; Gothic atta ‘father’; Old Icelandic atti ‘father’; Old High German atto ‘forefather’; East Frisian atte, ette ‘father’;Old Church Slavic otьcь ‘father’; Russian otéc [отец] ‘father’; Czech, Slovak otec ‘father’; Serbo-Croatian òtac ‘father’; Albanian at, ati, atë ‘father’. Kloekhorst 2008a:225—226; Puhvel 1984— .1/2:224—226; Tischler 1977— .I:92—93; Friedrich 1991:38; Sturtevant 1936:32 and 1951:§58, §70, §132a; Gusmani 1980— 1982:98; Melchert 1993a:69 and 1993b:221; Pokorny 1959:71 *ā̆tos, *atta; Walde 1927— 1932.I:44 *atta; Mann 1948:14 and 1984—1987:39 *atā (*attā, -os, -i̯ os) ‘daddy’; Watkins 2000:6 *atto- ‘father’; Mallory—Adams 1997:195 *at- (or *haet- or *h₁at- ?) ‘father’; Buck 1949:§2.35 father; Turner 1966—1969.I:12, no. 221, *atta- ‘father, grandfather’; Walde—Hofmann 1965—1972.I:77 *ā̆to-lo-; Ernout—Meillet 2001:54; De Vaan 2008:60 *h₂et-o- a nursery word for ‘father’; Boisacq 1950:98; Beekes 2010.1:165; Hofmann 1966:27; Frisk 1970—1973.I:182; Chantraine 1968—1980.I:135; Matasović 2009:46 Proto-Celtic *attyo- ‘father, foster father’, Proto-Indo-European *atta ‘father’; Orël 1998:14 and 2003:27 Proto-Germanic *attōn ‘father’; Kroonen 2013:39 ProtoGermanic *attan- ‘father’; Feist 1939:62; Lehmann 1982:46; Schuhmann 2024:28, no. 85, Proto-Germanic *attan-; Uhlenbeck 1900:18; De Vries 1977:17; Derksen 2008:383 ProtoSlavic *otьcь ‘father’; Huld 1984:39; Bomhard 2023.3:786—787, no. 660, Proto-Nostratic *ʔatºtºa ‘older male relative, father’ (nursery word). The Reconstruction of the Proto-Anatolian Vowel System 81 2. an-na-aš ‘mother’ (common gender a-stem; mobile accent; a-grade generalized throughout the paradigm; nursery word): Singular: Nominative Accusative Vocative (no accent) Genitive Dative-Locative Ablative (sg./pl.) Hittite an-na-aš an-na-an Indo-Anatolian *H₁ánnas *H₁ánnam Indo-European *H₁ánnos *H₁ánnom an-na-aš an-ni *H₁annás *H₁anní an-na-az, an-na-za *H₁annátºs *H₁annós *H₁anní ~ *H₁annéy / *H₄annóy *H₁annótºs Instrumental (sg./pl.) Allative / Directive an-na *H₁anná Plural Nominative Accusative *H₁ánnes *H₁ánnm̥ s an-ni-iš an-nu-uš, an-ni-uš *H₁ánnes *H₁ánnm̥ s Genitive Dative-Locative Cognates: A. Anatolian: Palaic (nom. sg.) an-na-aš ‘mother’; Cuneiform Luwian (nom. sg.) a-an-niiš ‘mother’, (nom. sg.) a-an-na-wa-an-n[i-iš] ‘stepmother’; Hieroglyphic Luwian MATER-nati- (= /anant(i)-/ ?) ‘mother’; Lydian ẽna-, ẽnar- ‘mother’; Lycian ẽne/i‘mother’; Milyan ẽne/i- ‘mother’; Carian en ‘mother’. Note: Not related to Hittite (nom. sg.) ḫa-an-na-aš ‘grandmother’ < *H₂anna- (Proto-Indo-European *H₂anno-). B. Other: Latin anna ‘foster-mother’; (?) Greek (Hesychius) ἀννίς· ‘grandmother’. Kloekhorst 2008a:174 Proto-Anatolian *Honno-; Puhvel 1984— .1/2:55—57; Tischler 1977— .I:24—25; Friedrich 1991:21; Sturtevant 1936:21; Gusmani 1980—1982:51; Melchert 1993a:23 and 1993b:12; Simon 2023, Carian no. 30; Walde 1927—1932.I:55— 56 *an-; Pokorny 1959:36—37 *an-; Mann 1984—1987:21 *anā (hypocoristic) ‘mother, grandmother’; Watkins 2000:4 *an- ‘old woman, ancestor’; Mallory—Adams 1997:385— 386 *h₄en- (or *an-) ‘(old) woman, mother’; Buck 1949:§2.36 mother; Walde—Hofmann 1965—1972.I:50; Beekes 2010.1:107; Frisk 1970—1973.I:112; Hofmann 1966:19; Chantraine 1968—1980.I:91; Bomhard 2023.3:775—776, no. 649, Proto-Nostratic *ʔan¨a ‘mother, aunt’ (nursery word). 3. e-eš-ḫar ‘blood’ (neuter gender r/n-stem; mobile accent; e-grade generalized throughout the paradigm): Singular: Nom.-Acc. Hitttite e-eš-ḫar, Indo-Anatolian *H₁ésH₂r̥ Indo-European *H₁ésH₂r̥ 82 Chapter 5 iš-ḫar Vocative (no accent) Genitive iš-ḫa-na-a-aš, *H₁əsH₂nás iš-ḫa-a-na-aš, iš-ḫa-na-aš, e-eš-ḫa-na-aš, e-eš-na-aš Dative-Locative e-eš-ḫa-ni-i, *H₁əsH₂ní iš-ḫa-ni(-i) Ablative (sg./pl.) e-eš-ḫa-na-az, *H₁əsH₂nátºs iš-ḫa-na-az, e-eš-ḫa-na-az, iš-ḫa-na-an-za Instrumental (sg./pl.) e-eš-ḫa-an-ta, *H₁əsH₂án(tº) iš-ḫa-an-da, e-eš-ḫa-ni-it *H₁əsH₂nét *H₁əsH₂nós *H₁əsH₂ní ~ *H₁əsH₂néy / *H₁esH₂nóy *H₁əsH₂nótºs *H₁əsH₂nóntºs *H₁əsH₂ón(tº) *H₁əsH₂nétº Plural (unattested): Cognates: A. Anatolian: Cuneiform Luwian (nom.-acc. sg.) a-aš-ḫar-ša, [a]aš-ḫa-ar ‘blood’. (adj.; nom. sg. c.) a-aš-ḫa-nu-wa-an-ti-iš ‘bloody’; Hieroglyphic Luwian (n.; nom.-acc. sg.) á-sa-ha-na-ti-sa-za ‘blood-offering’ B. Other: Sanskrit (nom.-acc. sg.) ásṛk ‘blood’, (gen. sg.) asnás; Tocharian A ysār ‘blood’, B yasar ‘blood’; Greek ἔαρ ‘blood’; Latin assyr, aser ‘blood’; Latvian asins ‘blood’; Armenian ariwn ‘blood’. Kloekhorst 2008a:256—260 Proto-Indo-European (nom.-acc. sg.) *h₁ésh₂r, (gen.sg.) *h₁sh₂éns (> Pre-Hittite *h₁esh₂enós) ― Kloekhorst argues convincingly that a reconstruction h₁ḗsh₂r, with a long vowel in the first syllable, is not justified; Puhvel 1984— .1/2:305—313 Proto-Indo-European (nom.-acc. sg.) *ḗsH₁r̥ , (gen.sg.) *ésH₁-n̥ -s; Tischler 1977— .I:112—115; Friedrich 1991:43; Sturtevant 1936:37, 1942:§47b, §59b, §66b, and 1951:§56, §65, §75, §87 Proto-Indo-Hittite *Eés¦ьr, §141 ― according to Sturtevant, the instrumental plural in -an-ta is an analogical reformation, while “the commoner instrumental ending -it must be interpreted as standing for -et”; Pokorny 1959:343 *ē̆s-r̥ (gß). ‘blood’ (gen.) *es-n-és; Walde 1927—1932.I:162 *ē̆s-r̥ (gß) blood’, (gen.) *es-n-és, *senés; Mann 1984—1987:249—250 *es"ər ‘blood; tears; sweat; lymph’, (gen. sg. or poss.) *es"nos ― Mann notes: “The ultimate phonology of this word-type is entirely obscure.”; Buck 1949:§4.15 blood; Mallory—Adams 1997:71 *h₁ésh₂r̥ - (nom. *h₁ésh₂ōr) ‘(flowing) blood’, (gen.) *h₁esh₂nós ― Mallory—Adams note: “Its geographical spread (including Anatolian) and archaic morphology ensure PIE status.”; Mayrhofer 1956—1980.I:66 “old heteroclitic”; Uhlenbeck 1898—1899.1:18; Burrow 1973:221; Martirosyan 2010:138 *h₁esh₂r > *ehar > *ar-+-iwn; Beekes 2010.I:366 *h₁esh₂-r; Frisk 1970—1973.I:432; Hofmann 1966:66; Chantraine 1968—1980.I:308; De Vaan 2008:58 Proto-Indo-European (nom.-acc. sg.) *h₁ésh₂-r, (gen. sg.) *h₁sh₂-én-s; The Reconstruction of the Proto-Anatolian Vowel System 83 Walde—Hofmann 1965—1972.I:72 *ē̆sr̥ g, (gen.) *esnés; Ernout—Meillet 2001:52; Adams 2013:525 *h₁ésh₂ōr; Van Windekens 1976—1982.I:607 *esōr; Bomhard 2023.1:832 (comparison with Northwest Caucasian) Proto-Indo-European (nom.-acc. sg.) *ʔés-‿ ħh-r̥ ; (gen. sg.) *ʔs-‿ ħh-én-s, *ʔs-‿ ħhn-és. 4. iš-ḫa-aš ‘master, lord, owner’ (common gender a-stem; mobile accent; with initial i(< e-) generalized throughout the paradigm. According to Kloekhorst 2008a:390: “The rare spellings with initial e-eš-ḫ° or eš-ḫ° are all NS…”): Singular: Nominative Hitttite iš-ḫa-a-aš, iš-ḫa-aš Accusative iš-ḫa-a-an Vocative (no accent) iš-ḫa-a, iš-ḫa Genitive iš-ḫa-a-aš Dative-Locative iš-ḫi-i=š-ši, iš-ḫi-e=š-ši, iš-ḫi-i, eš-ḫé Ablative (sg./pl.) Instrumental (sg./pl.) Allative / Directive Plural: Nominative Accusative Genitive Dative-Locative iš-ḫe-e-eš iš-ḫé-eš EN-MEŠ-uš iš-ḫa-a-aš iš-ḫa-aš Indo-Anatolian *H₁ésH₂as Indo-European *H₁ésH₂os *H₁ésH₂am *H₁ésH₂om *H₁esH₂a *H₁esH₂o *H₁əsH₂ás *H₁əsH₂ós *H₁əsH₂í ~ *H₁əsH₂éy *H₁əsH₂í ~ *H₁əsH₂éy / *H₁əsH₂óy *H₁ésH₂es *H₁ésH₂es *H₁ésH₂m̥ s *H₁əsH₂ás *H₁əsH₂ás *H₁ésH₂m̥ s *H₁əsH₂ós *H₁əsH₂ós Cognates: A. Anatolian: No known Anatolian cognates. B. Other: Latin erus ‘the master of a house or family; master, lord, owner, proprietor’. Note: This etymology is rejected by Puhvel 1984— .1/2:385—390. He cautiously favors comparison with Armenian išxan ‘ruler, prince’, išxel ‘to rule’ instead, as “[e]ither a borrowing from Hittite into Armenian or into both from some common ‘culture word’ source…” Kloekhorst 2008a:390 Proto-Indo-European *h₁esh₂-ó-; Puhvel 1984— .1/2:385—390; Tischler 1977— .I:372—377; Friedrich 1991:85; Sturtevant 1936:37, under eshās, 1942:§56a, Proto-Indo-Hittite *’esḪos, and 1951:§47, §58, §87, §121; Pokorny 1959:342; Walde 1927—1932.I:161; Buck 1949:1329—1332, §19.41 master; Mallory—Adams 1997:371—372 Proto-Indo-European ? *h₁esh₂ós ‘master’; De Vaan 2008:194—195 Proto-Indo-European *h₁esh₂-ó- ‘master’; Walde—Hofmann 1965—1972.I:419; Ernout— Meillet 2001:201—202. 84 Chapter 5 The Old Hittite case endings may be summarized as follows (cf. Hoffner—Melchert 2008:69—78, noun and adjective declensions 79—131; Beckman 2011:524; Van den Hout 2011:17; Sturtevant 1951:84—91; Kloekhorst 2008a:104—105; Friedrich 1960:42—45; Lauffenburger 2008:23; Luraghi 1997:15; Rieken 2011:45; Yakubovich 2020:229; Watkins 2004:560—561; Neu 1979:178): Singular Plural -š, -Ø -eš -n, -an -uš -Ø, -an, -n -Ø, -a -i, -Ø -anza (/-ants/) -anteš -aš -an, -aš -i, -ya, -Ø -aš -a Singular/Plural -az, -za (/-ts/) V-t, C-it, -d/ta Nominative, common Accusative Nominative-accusative, neuter Vocative Ergative Genitive Dative-locative Allative Ablative Instrumental As noted by Kloekhorst (2013), Late Proto-Indo-European noun stems could be classified into four basic types according to their accent-Ablaut patterns: (1) acrostatic, (2) proterokinetic, (3) hysterokinetic, and (4) amphikinetic, as follows (R = Root; S = Suffix; E = Ending): 1. Acrostatic Strong Weak Loc. R ḗ/ó é é 3. Hysterokinetic R Strong Weak Loc. - S - S é é E - E é (-i) 2. Proterokinetic R Strong é Weak Loc. - S é ḗ E - 4. Amphikinetic R Strong é Weak Loc. - S o é E é (-i) Szemerényi (1996:162) adds a fifth type: 5. Mesostatic: the accent is on the suffix throughout the paradigm. An even more elaborate system is set up by Meier-Brügger (2003:205—218) — see also Fortson 2010:§§6.19—6.28; Clackson 2007:79—86. Kloekhorst (2013:124) concludes that this model does not accurately reflect the noun stem patterning found in an earlier stage of development. Rather, Kloekhorst posits a model The Reconstruction of the Proto-Anatolian Vowel System 85 consisting of only three accent-Ablaut patterns for the earlier stage: (1) static, (2) proterodynamic, and (3) hysterodynamic. He posits the following original hysterodynamic stem patterning: *CéC-C ~ *CC-éC-m ~ *CC-C-és He derives the later hysterokinetic and amphikinetic stem patterning from the earlier hysterodynamic stem patterning. In this theory, it is assumed that the morphologically significant vowel was always *e. However, as rightly pointed out by Clackson (2007:86): It is possible that both Greek and Sanskrit made the same innovatory change in the accent pattern of the word for ‘foot’ when they replaced the inherited paradigm, following their separation from PIE. However, other comparative evidence makes it likely that, at the later stages of PIE, the hypothesis that each word has only one *e vowel which was positioned in the locus of the accent cannot be true. We have already seen cases where the vowel *o can also stand in accented syllables in opposition to *e, as in the very same paradigm, nominative *pód-s and genitive *péd-s. There are also roots and suffixes which always show *o and never *e in any language, such as *póti- ‘master’; reconstructed words which show two e-grades, such as *pénkwe ‘five’; and words where the comparative evidence suggests that the location of the accent was other than where the vowel *e stood, as *septḿ ‘seven’ and *wóid-e ‘he/she knows’. Building upon Kloekhorst’s views, it appears that the earliest noun paradigm patterning for neuter action nouns was based upon the following dynamic relationship between accent-Ablaut and case endings: in neuter action nouns, the accent was placed on the root in the so-called “strong” cases, which also had a full-grade vowel (*CV́ C-C-), while the vowels of the suffix and inflectional ending were either reduced (= reduced-grade) or eliminated altogether (= zero-grade), unless the loss of the vowel would have resulted in unpronounceable consonant clusters or would have obscured the form and meaning of the case ending. In the so-called “weak” cases (= “oblique” cases), the accent was shifted from the root to either the suffix or the inflectional ending, which had a full-grade vowel, while the vowel of the root was either reduced or eliminated altogether, unless the loss of the vowel would have resulted in unpronounceable consonant clusters. The accent was always shifted to the inflectional ending in the weak cases when one existed. When no inflectional ending was present, the accent was shifted to the suffix in the weak cases instead, thus: 1. Strong cases: *CV́ C-C-; 2. Weak cases, with case ending: *CC-C-V́ (C) / *CC-C-CV́ (C); 3. Weak cases, without case ending: *CC-V́ C-Ø. It was the interplay of this patterning with the developing system of indicating grammatical relationships mainly by means of case endings that resulted in the accent-Ablaut patterns reconstructed above. Earlier, there were fewer case endings, and the morphological relationships at issue were indicated by means of postpositions and word order (SOV) combined with the accent-Ablaut rules just mentioned. It should be noted that the earlier 86 Chapter 5 accent-Ablaut patterns became somewhat obscured in the daughter languages by the leveling out of the patterning in favor of fixed position throughout the paradigm. For comparison, the Disintegrating Indo-European system of case endings has been reconstructed as follows (cf. Bomhard 2023.I:521—522; Adrados—Bernabé—Mendoza 1995—1998.II:45—94; Beekes 2011:185—217; Brugmann 1904:373—399; Clackson 2007:92—100; Fortson 2010:113—139; Fulk 2018:141—180; Gray 1932; Hirt 1921— 1927.3:33—81; Meier-Brügger 2003:195—199; Meillet 1964:292—300; Lundquist— Yates 2018:2083; Pooth 2014; Schmalstieg 1980:46—87; Sihler 1995:248—256; SchmittBrandt 1998:180—220; Shields 1982; Szemerényi 1996:157—192; Watkins 1998:65— 66) (the following table is a composite drawn from multiple sources and aims to be as comprehensive as possible; some of the reconstructions are more certain than others — the dual endings are particularly uncertain): Case Singular Plural Nominative *-s, *-Ø *-es Vocative *-Ø *-es Accusative *-m/*-m̥ Genitive *-es/*-os/*-s *-m̥ s/*-ms or *-n̥ s/*-ns *-om/*-ōm Ablative Dative *-es/*-os/*-s; *-bº(y)os, *-mos *-ētº/*-ōtº (< *-e/o-H÷(e)tº) *-ey *-bº(y)os, *-mos *-bºyō (?), *-mō (?) Locative *-i, *-Ø *-su/*-si *-ow (?) *-bºis, *-mis *-bºyō (?), *-mō (?) Instrumental *-(e)H÷ Dual } *-H÷(e) *-ows (?), *-oH÷s (?) *-bºyō (?), *-mō (?) Missing from this table is the thematic nominative-accusative neuter singular ending *-m — this form is to be derived from the accusative singular ending. The *-bº- and *-mendings found in several of the concrete cases are usually considered to be late additions, and some have even questioned whether they should even be posited for the Indo-European parent language. They are not found in Hittite or the other Anatolian daughter languages. No doubt, these endings were originally postpositions that were gradually incorporated into the case system, with some daughter languages choosing *-bº- and others choosing *-m-. They probably should not be reconstructed as case endings at the Proto-Indo-European level. In like manner, the genitive plural probably arose from the accusative singular, while the genitive singular and nominative singular endings in *-s must have had a common origin — these endings later spread from the genitive singular to the ablative singular. The dual was a late addition, while the plural originally had a reduced set of endings compared to what was found in the singular — this is the picture that emerges when the Hittite and other Anatolian data are brought into consideration (see above for details). We may note here that the Proto-Uralic ablative ending *-ta and the Proto-Elamo-Dravidian oblique/ locative ending *-tə are most likely related to the Anatolian instrumental singular endings The Reconstruction of the Proto-Anatolian Vowel System 87 within Indo-Anatolian: Hittite -it, -et, (rare) -d/ta; Palaic -az; Luwian -ati; Lycian -adi, -edi; Lydian -ad (cf. Bomhard 2023.I:438—440). The rules governing the position of the accent in Proto-Indo-Anatolian may be stated rather simply: 1. As indicated above, neuter action nouns were accented on the stem in the so-called “strong” cases but on the case ending in the so-called “weak” cases (cf. Burrow 1973: 220—226). 2. Common gender agent noun/adjectives were accented on the suffix throughout the paradigm (cf. Burrow 1973:119). 3. Athematic verbs were accented on the stem in the singular but on the ending in the plural (and dual) in the indicative but on the ending throughout the middle (cf. Burrow 1973:303). Note: The dual was a post-Anatolian development. The thematic formations require special comment. It seems that thematic agent noun/adjectives were originally accented on the ending in the strong cases and on the stem in the weak cases. This pattern is the exact opposite of what is found in the neuter action nouns. The original form of the nominative singular consisted of the accented thematic vowel alone. It is this ending that is still found in the vocative singular in the daughter languages and in relic forms such as the word for the number ‘five’, *pºenk¦ºe (*pe•qße in Brugmann’s transcription). The nominative singular in *-os is a later formation and has the same origin as the genitive singular (cf. Szemerényi 1972a:156; Van Wijk 1902). The system of accentuation found in Disintegrating Indo-European was by no means ancient. The earliest period of Proto-Indo-European that can be reconstructed appears to have been characterized by a strong stress accent (cf. Burrow 1973:108—112; Lehmann 1952:111—112, §15.4, and 1993:131—132; Szemerényi 1996:111—113) — following Lehmann, this period may be called the Phonemic Stress Stage, or, in more current terms, Proto-Indo-Anatolian. This accent caused the weakening and/or loss of the vowels of unaccented syllables. There was a contrast between those syllables with stress and those syllables without stress. Stress was used as an internal grammatical morpheme, the stressed syllable being the morphologically distinctive syllable. The phonemicization of a strong stress accent in Pre-Indo-Anatolian caused a major restructuring of the inherited vowel system and brought about the development of syllabic liquids and nasals (cf. Lehmann 1993:138). In the latest period of Proto-Indo-European, quantitative Ablaut was no longer a productive process. Had there been a strong stress accent at this time, each Proto-IndoEuropean word could have had only one syllable with full-grade vowel, the vowels of the unstressed syllables having all been reduced or eliminated. However, since the majority of reconstructed Proto-Indo-European words have more than one full-grade vowel, the stress accent must have become non-distinctive at some point prior to the latest stage of development. In closing, we may quote Pooth 2020:24: …the idea that Proto-Indo-European morphology was completely identical to the one of Ancient Greek and Vedic Sanskrit is rather unlikely in my view… 88 Chapter 5 The evidence from the Anatolian daughter languages requires a thorough reassessment of the traditional reconstruction, especially for the earlier stages of development. Selected Hittite Verb Conjugations: A. -mi Conjugation: 1. e-eš- ‘to be’: Present Active Singular Indo-Anatolian Indo-European 1. 2. 3. e-eš-mi e-eš-ši, e-eš-ti (-ti = -te) e-eš-zi, i-eš-zi *H₁ə́s-m-i *H₁és-m-i *H₁ə́s-s-i, *H₁ə́s-tº-ə *H₁és-s-i *H₁ə́s-tº-i *H₁és-tº-i e-šu-wa-ni *H₁əs-wá-n-i [*H₁(ə)s-mé-n/s-i] a-ša-an-zi *H₁əs-án-tº-i [*H₁(ə)s-én-tº-i, *H₁(ə)s-ón-tº-i] e-šu-un e-eš-ta e-eš-ta *H₁ə́s-m̥ *H₁ə́s-tº *H₁ə́s-tº *H₁és-m̥ [*H₁és-s] *H₁és-tº e-šu-u-en, e-šu-en, e-eš-šu-u-en e-eš-te-en e-še-er, e-šer, e-še-ir *H₁əs-wə́-n [*H₁(ə)s-mé-n/s] *H₁əs-tºə́-n *H₁əs-ə́r *H₁(ə)s-tºé-n/s [*H₁(ə)s-én-tº, *H₁(ə)s-ón-tº] Plural 1. 2. 3. Preterite Active Singular 1. 2. 3. Plural 1. 2. 3. Notes: 1. The 2nd sg. pres. act. (OH?/NS) e-eš-ti can only represent an archaic form in *-te (cf. 2nd pl. pret. act. e-eš-te-en). As noted by Kloekhorst (2008a:251—252), had this ending been *-ti, with *i, it would have yielded Hittite -zi, with affricativization of t before i as in the 3rd sg. pres. act. e-eš-zi. The Reconstruction of the Proto-Anatolian Vowel System 89 2. There was a change of accent and ablaut in this and other athematic stems in ProtoIndo-Anatolian — in the singular, the stem had full-grade vowel and was accented, while, in the plural, the stem had zero-grade vowel, and the accent was shifted to the ending. 3. The e-grade in the plural in Hittite is due to analogy. Cognates: A. Anatolian: Palaic (2nd sg. imp. act.) a-aš, (3rd sg. imp. act.) a-aš-du, (3rd pl. imp. act. a-ša-an-du, a-še-en-du; Cuneiform Luwian (3rd sg. pres. act.) a-aš-ti, (1st sg. pret. act.) aš-ḫa (?), (3rd sg. pret. act.) a-aš-ta, (3rd sg. imp.) a-aš-du, (3rd pl. imp.) a-ša-an-du; Hieroglyphic Luwian (3rd sg. pres. act.) a-sa-ti, (2nd pl. pres. act.) a-sa-ta-ni, (3rd pl. pres. act.) á-sa-ti; (1st sg. pret. act.) á-sa-ha, á-sá-ha, á-sa-ha-ʹ, á-sa₈-ha, (3rd sg. pret. act.) á-sa-ta, á-sa-tá, á-sa-ta-ʹ, sa-ta, a-sa-tá, sa-tá-ʹ, sa-ta; Lydian (1st sg. pres.) -im; Lycian (3rd sg. pres. act.) esi, (3rd sg. imp. act.) esu. B. Other: Sanskrit (1st sg. pres.) ásmi, (2nd sg. pres.) ási (instead of *ássi), (3rd sg. pres.) ásti, (1st pl. pres.) smás, (2nd pl. pres.) sthá, (3rd pl. pres.) sánti; Greek (1st sg. pres) εἰμί, (2nd sg. pres.) εἶ, (3rd sg. pres.) ἐστί; Latin (3rd sg. pres.) est, (3rd pl. pres.) sunt; Gothic (1st sg. pres.) im, (2nd sg. pres.) is, (3rd sg. pres.) ist, (1st pl. pres.) sijum, (2nd pl. pres.) sijuþ, (3rd pl. pres.) sind; Old Icelandic (1st sg. pres.) em, (2nd sg. pres.) ert, (3rd sg. pres.) er, (1st pl. pres.) erum, (2nd pl. pres.) eruð, (3rd pl. pres.) eru; Old Swedish (1st sg. pres.) Km, (2nd sg. pres.) Kst, (3rd sg. pres.) is; Old English (West Saxon) bēon ‘to be’: (1st sg. pres.) eom, (2nd sg. pres.) eart, (3rd sg. pres.) is, (pl.) sint, sindon; Old Frisian (3rd sg. pres.) is; Old Saxon (3rd sg. pres.) ist; Dutch (3rd sg. pres.) is; Old High German (3rd sg. pres.) ist; Old Irish (1st sg. pres.) am (< *esmi), (2nd sg. pres.) at, (3rd sg. pres.) is (< *esti), (1st pl. pres.) ammi (< *esm-), (2nd pl. pres.) adib, (3rd pl. pres.) it (< *senti); Old Welsh (3rd sg. pres.) is; Old Breton (3rd sg. pres.) is; Gaulish (1st sg. pres.) immi; Lithuanian (1st sg. pres.) esmì, (2nd sg. pres.) esì, (3rd sg. pres.) ẽsti (èsti), (1st pl. pres.) ẽsme (èsme), (2nd pl. pres.) ẽste (èste); Old Prussian (1st sg. pres.) asmai, asmu, asmau, (2nd sg. pres.) assai, assei, asse, (3rd sg. pres.) ast, (1st pl. pres.) asmai, (2nd pl. pres.) asti, estei, astei; Old Church Slavic (1st sg. pres.) jesmь, (2nd sg. pres.) jesi, (3rd sg. pres.) jestъ, (1st pl. pres.) jesmъ, (2nd pl. pres.) jeste, (3rd pl. pres.) sątъ; Armenian (1st sg. pres.) em, (2nd sg. pres.) es, (3rd sg. pres.) ē; Albanian (1st sg. pres.) jam, (2nd sg. pres.) je, (3rd sg. pres.) është. Kloekhorst 2008a:250—252 Proto-Indo-European *h₁és-ti, *h₁s-énti — according to Kloekhorst, the spelling (3rd sg. pres. act.) i-eš-zi is a scribal error; Friedrich 1991:42; Puhvel 1984— .1/2:285—291; Tischler 1977— .I:109—110; Sturtevant 1936:36 and 1951:149, §299; Gusmani 1980—1982:58 -im < *esmi; Melchert 1993a:20 and 1993b:33; Laroche 1959:32—33; Carruba 1970:52; Pokorny 1959:340—342 *es-; Walde 1927— 1932.I:160—161 *es-; Mann 1984—1987:250 *esmi (*es-si, *es-ti) ‘am’ and 253 *esti ‘is’; Rix 2001:241—242 *h₁es-; Buck 1949:§9.91 ― according to Buck, “Words for ‘be’, denoting existence and serving as the copula, are mostly derived from two roots, of which one (*es-) was the most colorless, while the other (*bheu-, *bhū-) evidently had the primary sense of ‘come into being, become’.”; Mallory—Adams 1997:53 *h₁es- ‘be’; Mayrhofer 1956—1980.I:67; Uhlenbeck 1898—1899.1:18—19; Martirosyan 2010:255 90 Chapter 5 *h₁es-mi, *h₁es-si, *h₁es-ti, (3rd pl. pres.) *h₁s-énti; Beekes 2010.I:389 *h₁es-; Frisk 1970—1973.I:463—464; Hofmann 1966:73; Chantraine 1968—1980.I:322—323 root *es- “old athematic present root, with vowel alternation”; De Vaan 2008:599 *h₁es-/*h₁s‘to be’; Ernout—Meillet 2001:665—666; Walde—Hofmann 1965—1972.II:628—629; Kroonen 2013:582 *h₁es- ‘to be’; Orël 2003:458; Schuhmann 2024:161, no. 462, *h₁es-ti; Feist 1939:292 *esmi, *es(s)i, *esti; Lehmann 1986:205 *es- ← *"es-; Uhlenbeck 1900:89 and 142 under sunja ‘true’: *es-; De Vries 1977:101; Weekley 1921:129; Klein 1971:58 *esmi and 816 *es-ti; Onions 1966:81 *es-, *s- and 846 under sooth; Boutkan—Siebinga 2005:195 *h₁esti; Vercoullie 1898:121; Kluge—Seebold 2011:451 *esti and 840 under sein; Walshe 1951:208 under sein; Smoczyński 2007:149 under esù: *h₁es-mi, *-si, *-ti; Fraenkel 1962—1965.I:124; Derksen 2008:146 *h₁es-mi, *h₁es-ti and 2015:157 *h₁es-mi, *h₁es-ti; Orël 1998:156 *es-; Huld 1984:76—77 *E₁esmi; Meyer 1891:160—161 *ésmi; Matasović 2009:118 **h₁es-. 2. ku-en- ‘to strike, to smite, to slay, to kill’: Present Active Singular Indo-Anatolian Indo-European 1. 2. ku-e-mi ku-e-ši, ku-en-ti (-ti = -te) ku-e-ti ku-(e-)en-zi *g¦ə́n-m-i *g¦ə́n-s-i, *g¦ə́n-tº-ə *g¦ə́n-tº-i *g¦ºén-m-i *g¦ºén-s-i ku-na-an-zi *g¦n-án-tº-i *g¦ºn-ón-tº-i ku-e-nu-un ku-en-ta ku-(e-)en-ta *g¦ə́n-m̥ *g¦ə́n-tº *g¦ə́n-tº *g¦ºén-m̥ [*g¦ºén-s] *g¦ºén-tº ku-e-u-en ku-en-tin ku-e-nir, ku-e-ni-ir *g¦n̥ -wə́-n *g¦n̥ -tºə́-n *g¦n-ə́r [*g¦ºn̥ -mé-n/s] *g¦ºn̥ -tºé-n/s [*g¦ºn-én-tº, *g¦ºn-ón-tº] 3. *g¦ºén-tº-i Plural 1. 2. 3. Preterite Active Singular 1. 2. 3. Plural 1. 2. 3. The Reconstruction of the Proto-Anatolian Vowel System 91 Cognates: A. Anatolian: (?) Lydian, qēn- (qān-) verb root in ucbaqēn- and fisqān-. B. Other: Sanskrit (3rd sg. pres.) hánti ‘to strike, to kill, to slay’, (3rd pl. pres.) ghnánti; Avestan ǰaiṇti ‘to strike, to slay, to kill’; Greek θείνω ‘to strike; to slay, to kill’, φατός ‘slain’, φόνος ‘murder, manslaughter’; Armenian gan ‘beating, blow’; Latin -fendō in dē-fendō ‘to repel, to repulse, to ward off, to drive away’, of-fendō ‘to strike, to knock, to dash against’; also īn-fēnsus ‘hostile, aggressive’; Old Irish gon(a)id ‘to wound, to slay’; Burgundian *gunþs ‘fight’; Old Icelandic gunnr, guðr ‘war, battle’; Old Saxon gūdea ‘battle’, gūth- ‘battle’ in: gūthfano ‘regimental colors’, gūthhamo ‘armor’; Old English gūð ‘combat, battle, war’; Old High German gund- ‘battle’; Lithuanian genù, giñti ‘to drive, to propel’, geniù, genjti ‘to lop, to trim, to prune (tree, branches)’, ginù, gìnti ‘to defend, to protect; to forbid, to prohibit’, ganaũ, ganýti ‘to graze, to pasture; to herd, to tend (the cattle)’; Latvian dzenēju, dzenêt ‘to trim, to prune’, dzęnu, dzìt ‘to chase, to drive, to persecute’; Old Prussian guntwei ‘to chase, to drive’; Old Church Slavic ženǫ, gъnati ‘to chase, to persecute’, gonjǫ, goniti ‘to chase, to drive’; Czech hnáti ‘to chase, to drive, to compel’, hon ‘hunt’, honiti ‘to chase, to hunt, to persecute’; Polish gonić ‘to chase, to hunt, to persecute’; Slovakian honitʹ ‘to chase, to hunt, to persecute’; Serbo-Croatian gȍnim, gòniti ‘to chase, to persecute’; Bulgarian gónja ‘to chase, to hunt, to persecute’; Slovenian gǫ́nim, góniti ‘to drive repeatedly’; Russian gnatʹ [гнать] ‘to drive, to drive away, to turn out, to chase, to chase away; to urge, to thrust, to jostle, to hustle, to shove, to jog, to jolt’, (dial.) gonítʹ [гонить] ‘to persecute, to oppress’. Kloekhorst 2008a:485—486 Proto-Indo-European *g¦ºén-ti/*g¦ºn-énti, *g¦ºn-s#é/ó-; Puhvel 1984— .4:206—212 *gh¦en-; Friedrich 1991:112—113 Proto-Indo-European *gßhen-; Tischler 1977— .I/4:604—606; Sturtevant 1936:87 and 1951:38, §62a, IndoHittite *gꜤwénty, *gꜤwnénty and 58, §81; Gusmani 1980—1982:83; Pokorny 1959:491— 493 *gßhen-(ə)- ‘to strike, to hit’; Walde 1927—1932.I:679—681 *gßhen-, also as a seṭbase *gßhenē-; Mann 1984—1987:379—380 *gu̯ hen- ‘to drive, to beat, to kill’; Mallory— Adams 1997:548 *g¦hen- ‘to strike’ (pres. *g¦hénti): “Practically universally attested and clearly old in IE, this is the archetypal verb for ‘strike’ in PIE.”; Rix 2001:218—219 *gߺen-; Buck 1949:§9.21 strike (hit, beat) *g¦hen-; Watkins 2000:35 *g¦hen- ‘to strike, to kill’; Mayrhofer 1956—1980.III:575—577; Uhlenbeck 1898—1899.2:357; Boisacq 1950:336 *œßhen-i̯ ō, *œßhono-s, *œßhn̥ -; Chantraine 1968—1980.I:425—426 *gh¦en-; Frisk 1970—1973.I:657—658 *œßhen-; Beekes 2010.I:536—537 *g¦ºen- and II:1586 *g¦ºon-o-; Hofmann 1966:112 *gßhen-i̯ ō; Martirosyan 2010:198; Hübschman 1883:24; De Vaan 2008:210—211 *g¦ºen-dº- ‘to hit, to strike’; Ernout—Meillet 2001:224—225 *gh¦en-; Walde—Hofmann 1965—1972.I:332—333 *gßhen- and I:478; Matasović 2009:144 *g¦ºen-; Kroonen 2013:196 *g¦ºén-ti; Orël 2003:146 Proto-Germanic *ᵹunþz; De Vries 1977:195 Proto-Norse *gunþʀ; Smoczyński 2007:156—157, 169, 180—181, 181; Derksen 2008:177, 197 *g¦ºon- and 2015:170—171*g¦ºen-, 177, 177—178; Fraenkel 1962—1965.I:152—153. 92 Chapter 5 B. -ḫi Conjugation: 3. a-ar- ‘to come (to), to arrive (at)’: Present Active Singular Indo-Anatolian Indo-European 1. 2. 3. *H₁ar-H₂á *H₁ar-tºə́ *H₁ar-ə́ *H₁or-H₂á *H₁or-tº(e)+(H₂á) *H₁or-é a-ar-ḫi (-ḫi < *-H₂á+i) a-ar-ti (-ti < *-tºə́+i) a-ri (-i < *-ə́+i) Note: As pointed out by Sturtevant (1951:131—132, §231), the Hittite 1st singular ending in -(ḫ)ḫi resulted from the spread of final -i to the present active ending. The original ending here was -(ḫ)ḫa, which is still preserved in Luwian and Lycian: cf. Cuneiform Luwian (1st sg. pret.) ta-pár-ḫa, da-pár-ḫa ‘I ruled, I governed’, a-ḫa ‘I made, I did’, ú-i-ši-ḫa ‘I pressed, I squeezed’; Hieroglyphic Luwian (1st sg. pret.) á-ha ‘I made, I did’, á-sa-ha, á-sá-ha ‘I sat’, i-zi-i-ha ‘I made’, pi-ia-ha ‘I gave’, (X)suwa/i-ha ‘I filled’, ta-ma-ha ‘I built’; Lycian (1st sg. pret.) aχa, aχã, agã ‘I made, I did’, (1st sg. pret.) pijaχa, pijaχã ‘I gave’, prñnawaχã ‘I built’. The final -i has spread to the other -ḫi conjugation present active endings in Hittite as well. Plural 1. 2. 3. e-ir-u-e-ni, e-ru-(u-)e-ni (a-)ar-te-ni, e-ir-te-ni a-ra-an-zi [*H₁ə́r-wə-n-i] [*H₁ə́r-tºə-n-i] [*H₁ár-an-tº-i] Notes: 1. In Hittite, the patterning here appears to have been accent on the ending in the singular, and with full-grade vowel in the root, but on the stem in the plural (cf. Hart 1980:57— 58) — this was the exact opposite of the patterning found in the -mi conjugation. Exceptions to this patterning can be explained as the beginnings of the shift in accent to conform with the patterning found in the -mi conjugation. This shift was completed in the non-Anatolian daughter languages, where the accent was the same as in the -mi conjugation, that is, on the stem, which was in o-grade, in the singular but on the ending, with stem in reduced-grade, in the plural. The non-Anatolian daughter languages also have introduced reduplication into the majority of stems here. The original patterning seems to have been accent on the ending throughout the paradigm (both singular and plural). 2. All of the present active plural personal endings of the Hittite -ḫi conjugation were taken over from the -mi conjugation. The Reconstruction of the Proto-Anatolian Vowel System 93 Preterite Active Singular 1. 2. 3. (a-)-ar-aḫ-ḫu-un, (a-)ar-ḫu-un a-ar-aš, a-ar-ta Plural 1. 2. 3. e-ru-u-en, e-ir-u-en a-ar-tin, ir-te-in e-ri-(e-)ir, i-e-ri-ir [*H₁ə́r-ər] (< *H₁ər-ə́r) *H₁er-ér Notes: 1. As noted above, Hittite has replaced all of the present active plural endings with those of the -mi conjugation. The same has happened for all of the singular preterite active personal endings and for the 1st and 2nd person plural preterite active endings. Here, only the 3rd plural preterite active personal ending is original. 2. Hittite has repurposed the -ḫi conjugation into an alternative to the -mi conjugation (cf. Patri 2020:209) — that is, same function, slightly different endings (see note 1 above). This represents a Hittite-specific innovation. 3. Primarily on the basis of a comparison of the non-Anatolian daughter languages, the earliest stative personal endings (> Hittite -ḫi conjugation singular personal endings, see above, note 1 above) were most likely as follows (cf. Lehmann 2002:171; Kortlandt 2007b:22 — Kortlandt writes *q₂ for *H₂; Bomhard 2023.1:643 and 1:648): Person 1st sg. 2nd sg. 3rd sg. 1st pl. 2nd pl. 3rd pl. Endings *-H₂a *-tºe; later *-tºH₂a (< *-tº(e)+*-H₂a) *-e *-me- (?) *-e *-ēr, *-r̥ s 4. The *-H₂a found in the 2nd singular personal ending (*-tºH₂a < *-tº(e)+*-H₂a) is taken from the 1st singular and appears to be a post-Anatolian development. Cognates: Anatolian: No known Anatolian cognates. But cf. Hittite (3rd sg. pres. act.) a-ra-a-i (old script a-ra-i) ‘to (a)rise, to lift, to raise’ (originally ḫi-conjugation); Cuneiform Luwian (3rd sg. pres. act.) a-ri-it-ta ‘to raise’, (nom.-acc. pl.) GIŠa-ri-ya-la ‘carrying basket’; Lycian (3rd sg. pres. act.) erije ‘to raise’. 94 Chapter 5 Other: Sanskrit árṣati ‘to flow’, árṇa-ḥ ‘undulating, surging; wave’, ṛccháti ‘to go, to move, to send’, ṛṇóti ‘to go, to move, to arise’; Avestan ar- ‘to go, to move, to come’, aurva-, aurvant- ‘rapid, quick’, ərənaoiti ‘to set in motion’; Old Persian ar- ‘to move, to go or come toward’, aruvā ‘action’, aruva- ‘rapid, quick’; Greek ὄρνῡμι ‘to urge on, to incite, to move, to stir oneself, to make to arise’; Latin orior ‘to rise, to arise’. Kloekhorst 2008a:196—197 *h÷ór-ei / *h÷r-énti, *h÷r-s#é/ó-; Puhvel 1984— .1/2:108— 111 *er- ‘to move, to stir, to raise’, 1/2:123—127 *E̥ ÷róy(Hø)-A÷ey, 1/2:162—167 *r̥ -new/ *r̥ -nu-, and 1/2:170—172 *Høér-s-, *Hør-és-; Friedrich 1991:27; Sturtevant 1936:25, 26, 28 and 1951:30, §56, Indo-Hittite *ʼers-, 32, §58, Indo-Hittite *ʼore; Rix 2001:241 *h÷ers‘to flow’, 252 *h÷rei̯ - ‘to rise, to rise up’; Pokorny 1959:326—332 *er-, *or-, *r- ‘to set in motion; to incite, to stir up, to arouse; to arise’; Walde 1927—1932.I:136—142 *er-; Mann 1984—1987:249 *ersō ‘to go, to glide, to wander, to creep, to dawdle’, 884 *or- ‘to start, to start up, to rise’, 889 *orneu-, *or-nū-mi ‘to move, to rouse; to dash, to fly’, 891 *orsō, -i̯ ō ‘to go, to proceed, to flow, to slide, to glide, to creep’; Watkins 2000:23 *er- ‘to move, to set in motion’ (oldest form *™er-); Gamkrelidze—Ivanov 1995.I:187, I:194 *er-/*r̥ -, *or- ‘to rise, to get up; to come into motion; to attain’, I:172 *r̥ neu-, *ornu- ‘to move’; Mallory—Adams 1977:506 *h÷er- ‘to set in motion’; Mayrhofer 1956—1980.I:51, I:53, I:119 *er-, and I:122; Boisacq 1950:714—716 *er-, *ere-; *erei-, *ereu-; Hofmann 1966:238—239 *er-; *ereu-, *erei-; Chantraine 1968—1980.II:823—824 *er-, *r̥ -; Frisk 1970—1973.II:422—424 *er-; Beekes 2010.II:1107 *h÷er- ‘to (a)rise’; Walde—Hofmann 1965—1972.II:222—223 *er- (*ere-, *erē-); Ernout—Meillet 1979:468; De Vaan 2008:434—435. Note: Two separate Proto-Nostratic stems have fallen together in ProtoIndo-European: (A) *ʔor- ‘to move rapidly, quickly, hastily; to set in motion’ (cf. Bomhard 2023.3:820—822, no. 691) and (B) *ʔor¨- ‘to rise (up)’ (cf. Bomhard 2023.3:822—823, no. 692). *woyt’- ‘to know’: Proto-IndoEuropean Old Sanskrit Avestan Homeric Greek Gothic Latin Singular 1 *wóyt’-Høa 2 *wóyt’-tºH₂a 3 *wóyt’-e véda véttha véda (+)οἶδα (+)οἶσθα (+)οἶδε wait waist wait vīdī vīdistī vīdit Plural 1 *wit’-mé 2 *wit’-é 3 *wit’-ḗr vidmá vidá vidúr (+)ἴδμεν (+)ἴστε (+)ἴσ(σ)ᾱσι witum wituþ witun vīdimus vīdistis vīdēre, -ē̆runt vaēdā vōistā vaēdā The Hittite personal endings may be summarized as follows (cf. Beckman 2011:527— 528; Watkins 2004:566): The Reconstruction of the Proto-Anatolian Vowel System 1st person 2nd person 3rd person Active Present Singular mi-conjugation ḫi-conjugation -mi -ḫi -ši -ti -zi -i 1st person 2nd person 3rd person Active Preterite Singular mi-conjugation ḫi-conjugation -un, -nun -ḫun -š -ta, t -t -š 1st person 2nd person 3rd person Medio-Passive Present Singular mi-conjugation ḫi-conjugation -(ḫ)ḫa(ri) -ta(ri) -ta(ri) -a(ri) 1st person 2nd person 3rd person Medio-Passive Preterite Singular mi-conjugation ḫi-conjugation -(ḫ)ḫat(i) -tat(i) -ta(ti) -a(ti) 1st person 2nd person 3rd person Active Imperative Singular mi-conjugation ḫi-conjugation -(a)llu -(a)llu -Ø, -i, -t -Ø, -i -tu -u 1st person 2nd person 3rd person Medio-Passive Imperative Singular mi-conjugation ḫi-conjugation -(ḫ)ḫaru -ḫuti -taru -aru Plural -weni, -meni -teni -anzi Plural -wen, -men -ten -ir Plural -wašta(ti) -tuma(ri) -anta(ri) Plural -waštati -tumat(i) -antat(i) Plural -weni -ten -antu Plural -waštati -tumat(i) -antaru 95 96 Chapter 5 5.4. Proto-Indo-Anatolian Root Structure Patterning A careful analysis of the root structure patterning led Benveniste to the discovery of the basic laws governing that patterning. According to Benveniste (1935:170—171), these laws may be stated as follows (see also Lehmann 1952:17—18) — this is a revised version, based upon the findings presented in this book: 1. The Proto-Indo-Anatolian root was monosyllabic, composed of the fundamental vowel (*ə ~ *a; *i; *u; probably also “original”, that is, “non-apophonic”, *ĕ and *ŏ) between two different consonants. 2. In this constant scheme: consonant plus vowel plus consonant (*CVC-), the consonants could be of any order provided that they were different: however, the cooccurrence of both a voiceless stop and an aspirated voiced stop was forbidden. 3. The addition of a suffix to the root gave rise to two alternating stem types: Type 1: root in full-grade and accented, suffix in zero-grade (or reduced-grade); Type 2: root in zero-grade (or reduced-grade), suffix in full-grade and accented. 4. A single determinative could be added to the suffix, either after the suffix of stem Type 2, or, if n, inserted between the root element and the suffix of stem Type 2. 5. Further addition of determinatives or suffixes pointed to a nominal stem. In the earliest form of Proto-Indo-Anatolian, there were three fundamental stem types: (A) verbal stems, (B) nominal and adjectival stems, and (C) pronominal and indeclinable stems. The phonemicization of a strong stress accent in Early Proto-Indo-Anatolian disrupted the patterning outlined above. The positioning of the stress was morphologically distinctive, serving as a means to differentiate grammatical relationships. All vowels were retained when stressed but were either weakened (= “reduced-grade”) or totally eliminated altogether (= “zero-grade”) when unstressed: the choice between the reduced-grade versus the zero-grade depended upon the position of the unstressed syllable relative to the stressed syllable as well as upon the laws of syllabicity in effect at that time. Finally, it was at this stage of development that the syllabic allophones of the resonants came into being. Thus, the stress-conditioned Ablaut alternations gave rise to two distinct forms of extended stems: Type 1: Root in full-grade and accented, suffix in zero-grade: *CV́ CC-. Type 2: Root in zero-grade, suffix in full-grade and accented: *CCV́ C-. When used as a verbal stem, Type 1 could undergo no further extension. However, Type 2 could be further extended by means of a “determinative”. Further addition of a determinative or suffixes pointed to a nominal stem (cf. Benveniste 1935:171; Lehmann 1952:17). According to Benveniste (1935:148), a “suffix” was characterized by two alternating forms (*-et-/*-t-, *-en-/*-n-, *-ek-/*-k-, etc.), while a “determinative” was characterized by a fixed consonantal form (*-t-, *-n-, *-k-, etc.). Finally, Benveniste (1935:164) notes: The Reconstruction of the Proto-Anatolian Vowel System 97 … in the numerous cases where the initial [consonant group has been reconstructed in the shape] *(s)k-, *(s)t-, *(s)p-, etc., with unstable sibilant, it is generally a question of prefixation, and it may be observed that the root begins with the [plain] consonant [alone excluding the sibilant]. In Pre-Indo-Anatolian, qualitative Ablaut most likely began as a purely phonological alternation. During the course of its development, however, these Ablaut alternations gradually became grammaticalized. We may venture a guess that prehistoric language contact with what was to become Proto-Northwest Caucasian was the impetus for the development of qualitative Ablaut (cf. Bomhard 2019a). 5.5. Concluding Remarks In Chapter 2, we surveyed the phonological systems of the principal Anatolian daughter languages. In Chapter 3, we discussed the Proto-Indo-Anatolian laryngeals. In Chapter 4, we discussed the reconstruction of the Proto-Indo-Anatolian consonant system. Finally, in Chapter 5, we discussed the reconstruction of the Proto-Indo-Anatolian vowels. We are thus now in a position to present a complete reconstruction of the Proto-Indo-Anatolian phonological system: Stops: Voiceless aspirated: Glottalized (ejectives):4 Plain voiced: Bilabial Dental Velar3 Labiovelar pº p’ b tº t’ d kº k’ g k¦º k’¦ g¦ ʔ h x ɣ ħh ‿ ʕɦ ‿ ʔ¦ Affricate: Fricative: Laryngeals: ˆº s Glottal stops: Glottal fricative: Velar fricatives: voiceless: voiced: Pharyngeal/laryngeals: voiceless: voiced: Nasals and Liquids: Glides: 3 m/m̥ w n/n̥ l/l̥ x¦ ħh¦ ‿ r/r̥ y The velar series may be assumed to have had non-phonemic palatalized allophones in certain environments (cf. Bomhard 2023.1:119—122). These allophones became phonemic in Luwian (cf. Melchert 2017:176) as well as in several of the non-Anatolian daughter languages — the so-called “satǝm” languages. 4 As previously noted, Kloekhorst (2016:226—228) considers this series to have been pre-glottalized in Proto-Anatolian. 98 Chapter 5 Vowels: i e Also the sequences: iy iw ə~a uy uw u o ey ew oy ow əy ~ ay əw ~ aw Note: According to Kloekhorst (2022:75), “Proto-Anatolian started to diverge sometime around the thirty-first century BCE” (note also the chart on p. 76). For comparison, consider the following “Fortified Proto-Indo-European” phonological system reconstructed by Colarusso (1992:23) on the basis of a comparison with Northwest Caucasian: Consonants: Vowels: pº tº kº¨ (kº kº¦ qº qº¦ b d g¨ g g¦ - t’ k’¨ k’) k’¦ q’ q’¦ ʔ ʔ¦ h s m n x x¦ ḥ ḥ¦ ɣ ɣ¦ ʕ ʕ¦ w r l ə ~ a (plus tonal stress) Note: Colarusso assumes that the laryngeals were lost in stages. The earliest laryngeals to be lost were *ʔ, *h, and *ʔ¦. The loss of these laryngeals between preceding short vowels and a following obstruent gave rise to “inherently” long vowels. The remaining laryngeals underwent various changes and were eventually lost altogether prior to the emergence of the non-Anatolian daughter languages. Some laryngeal reflexes persisted in Anatolian.  Chapter 6 The Non-Anatolian Daughter Languages 6.1. Introduction The present chapter discusses the development of the revised Proto-Indo-Anatolian phonological system down to so-called “Disintegrating Indo-European”, the last stage of Proto-Indo-European proper, and ending with the various non-Anatolian Indo-European daughter languages, concentrating, especially, on Armenian, Old Indic (Old Indo-Aryan), Greek, and Italic. Here, Armenian has a crucial role to play in understanding the developments in the early prehistory of these daughter languages. We will begin by looking at Disintegrating Indo-European and then discuss Tocharian, Germanic, Celtic, Slavic, Baltic, and Albanian, after which we will move on to Armenian and finish with Indo-Iranian, Greek, Italic, and Phrygian. For the sake of simplicity and continuity, I will repeat (and expand upon) some of what I have previously written. 6.2. Disintegrating Indo-European At the end of the last chapter, I proposed that the Proto-Indo-Anatolian phonological system be reconstructed as follows: Stops: Voiceless aspirated: Glottalized (ejectives): Plain voiced: Bilabial Dental Velar Labiovelar pº p’ b tº t’ d kº k’ g k¦º k’¦ g¦ ʔ h x ɣ ħ‿h ‿ ʕɦ ʔ¦ Affricate: Fricative: Laryngeals: ˆº s Glottal stops: Glottal fricative: Velar fricatives: voiceless: voiced: Pharyngeal/laryngeals: voiceless: voiced: Nasals and Liquids: Glides: m/m̥ w n/n̥ l/l̥ x¦ ħh¦ ‿ r/r̥ y 100 Chapter 6 Vowels: Also the sequences: i e iy iw ə~a uy uw u o ey ew oy ow əy ~ ay əw ~ aw The following series of phonological changes may be assumed to have taken place in the Indo-European parent language after the separation of the Anatolian branch and before the emergence of the individual non-Anatolian Indo-European daughter languages: 1. 2. 3. 4. The laryngeals *ʔ and *h were lost initially before vowels. *x and *‿ ħh merged into *‿ ħh, while *ɣ and *‿ ʕɦ merged into *‿ ʕɦ. Following that, *‿ ħh > *h and *‿ ʕɦ > *ɦ > *h initially before vowels. Next, all medial and final laryngeals merged into *h. This laryngeal no longer had any vowel-coloring effects. 5. Finally, the single remaining laryngeal *h was lost initially before vowels (except in Pre-Armenian) and medially between an immediately preceding vowel and a following non-syllabic. This latter change caused the compensatory lengthening of preceding short vowels, thus: eHC oHC aHC iHC uHC > > > > > ēC ōC āC īC ūC 6. *h was preserved in all other positions. *h had a syllabic allophone, *h̥ , when between two non-syllabics. This is the traditional schwa primum (*ə). 7. Glottalization was probably lost in late Disintegrating Indo-European itself just as the individual non-Anatolian daughter languages were beginning to emerge, though vestiges may have remained here and there in the daughter languages. 8. The earlier plain voiced stops developed into voiced aspirates, at least in some dialects of Disintegrating Indo-European. 9. The *e ~ *a (< *ə ~ *a) qualitative Ablaut of Proto-Indo-Anatolian developed into an *e ~ *o Ablaut. This change may still have been in progress when the individual non-Anatolian daughter languages began to emerge. 10. New Ablaut relationships developed as a result of the loss of laryngeals and the growth of secondary full-grade forms. 11. Proto-Indo-Anatolian was a stress-accent language, whereas Proto-Indo-European became a pitch-accent language (for a general discussion of stress and stress-accent systems, cf. Hyman 1975:204—212 and, for pitch-accent systems, pp. 230—233). For the latest period of development of the Indo-European parent language, the stage I call “Disintegrating Indo-European” — after the separation of the Anatolian branch from The Non-Anatolian Daughter Languages 101 the rest of the Indo-European speech community and before the emergence of the individual non-Anatolian Indo-European daughter languages —, the Proto-Indo-European antecedent of the satəm daughter languages may be reconstructed as follows (column 1 is voiceless aspirated, column 2 is glottalized [ejectives], and column 3 is voiced aspirated): Obstruents: (1) pº tº k¨º kº k¦º Laryngeals: (2) p’ t’ k’¨ k’ k’¦ s (3) bº dº g¨º gº g¦º (bilabial) (dental) (palatovelar) (velar) (labiovelar) h/h̥ Resonants: m/m̥ n/n̥ l/l̥ r/r̥ w/u y/i Vowels: e ē o ō a ā (i) ī (u) ū ə The most significant difference between the phonological system of the Disintegrating Indo-European antecedent of the satəm dialects and that of the centum dialects was in the treatment of the gutturals. In the centum dialects, the labiovelars did not become delabialized, and the palatovelars remained subphonemic. The phonological system of the Disintegrating Indo-European antecedent of the centum daughter languages, on the other hand, may be reconstructed thus (column 1 is voiceless aspirated, column 2 is glottalized, and column 3 is voiced aspirated: Obstruents: (1) pº tº kº k¦º Laryngeals: (2) p’ t’ k’ k’¦ s (3) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) h/h̥ Resonants: m/m̥ n/n̥ l/l̥ r/r̥ w/u y/i Vowels: e ē o ō a ā (i) ī (u) ū ə Note: To repeat, even though I have reconstructed a series of voiced aspirates above (column 3), such sounds are really only needed to explain developments in Armenian, Old Indic (Old Indo-Aryan), Greek, and Italic, as we shall see below. 102 Chapter 6 6.3. Tocharian In Tocharian, the distinction between voiceless, glottalized, and voiced (traditional voiceless, voiced, and voiced aspirated) stops was eliminated. However, Tocharian originally preserved the older contrast. While this contrast still existed, *t’ was lost before non-syllabic resonants (cf. Van Windekens 1976—1982.I:82—83, §241), while *tº and *d remained. The elimination of the older contrast must, therefore, have taken place after the loss of *t’ before non-syllabic resonants. 1. No doubt, the first step involved the deaspiration of the voiceless aspirates. 2. This was followed by the deglottalization of *p’, *t’, *k’, and *k’¦ and their merger with the voiceless stops *p, *t, *k, and *k¦, respectively. This is shown by the fact that *mp remained mp, while *mb became m (cf. Van Windekens 1976—1982.I:79), and by the fact that *t and *t’ had the same treatment before front vowels, namely, palatalization to c, while *d went its own way under the same conditions — palatalization to *dz > ts (cf. Van Windekens 1976—1982.I:83—84). 3. Last, the voiced stops were devoiced and merged with the plain voiceless stops. These developments may be summarized as follows: I Pre-Tocharian pº, p’, b tº, t’, d kº, k’, g k¦º, k’¦, g¦ II deaspiration > > > > p, p’, b t, t’, d k, k’, g k¦, k’¦, g¦ > > > > III deglottalization IV devoicing p, b t, d k, g k¦, g¦ p (w) t (c, ts) k (ç) k(w/u) (k, ç) > > > > *s usually remained but was palatalized to ṣ before front vowels. The non-syllabic resonants generally remained. The Disintegrating Indo-European vowels and diphthongs were greatly modified. For details on the Tocharian developments, cf.: Adams 1988:36—42 and 2017:458— 461; Gamkrelidze—Ivanov 1995.I:43—44; Hackstein 2017; Ringe 1996; Krause 1952 and 1955; Krause—Thomas 1960.I:61—68; Van Windekens 1976—1982.I:76—94. 6.4. Germanic To begin, I would like to address a statement made by Fulk in his recent book A Comparative Grammar of the Early Germanic Languages (Fulk 2018:100): The chief implication of the glottalic theory for Germanic linguistics is that it permits Germanic (along with Armenian) to be regarded not as a highly innovative branch in its consonantism but as an exceptionally conservative one, whereas the IE languages usually regarded as hewing closest to the PIE consonant system, especially Sanskrit and Greek, turn out to do nothing of The Non-Anatolian Daughter Languages 103 the sort. That Germanic should have remained so conservative while the European languages in closest proximity to it in prehistoric times all altered the inherited obstruents in similar ways is difficult to credit. And yet although the glottalic theory is not now widely supported, there is a considerable degree of concurrence that the reconstruction of PIE obstruents represented in §6.1 is implausible and awaits replacement by a creditable reconstruction. Nonetheless, it need not be the case that such an alternative reconstruction is what must be assumed for the latest stages of PIE, since it is of course possible that the typological peculiarities of PIE mentioned above are the consequence of an earlier obstruent system that had already changed before any of the extant IE families had developed individuating characteristics. That is to say, it is not a given that any IE language should directly reflect that earlier state of affairs rather than a laterdeveloped obstruent system similar to that arrived at (in §6.1) by the comparative method. The supposition that Germanic is an especially archaic branch of IE is at all events unsupported by its verb system, which appears to be a simplification of that reconstructed for late PIE (§12.9), showing no marked resemblance to the Hittite verb system. I do not understand the logic here. I see no problem whatsoever in viewing a particular language or branch as conservative in one area and innovative in another. This means that the innovations in the Germanic verb system do not preclude the retention of archaic features in the Germanic phonological system. (Note: This does not imply that there is any fault with Fulk’s book as a whole — it is an excellent monograph and a valuable resource, comprehensive in scope and current in its coverage of the field.) This same point is made by Hejná—Walkden (2022:262) in their discussion of the difference in the rates of change of noun morphology as opposed to verb morphology between Old English and Middle English: If you pause at this point to compare the verbal endings in Old English with the ones presented for Middle English in the last chapter (§5.3.2), you’ll see that on the whole there’s not a huge amount of difference: the big changes in verbal morphology in English take place between the Middle and Modern periods. This is different for nominal morphology, which (as you’ll soon see) is considerably more complex in Old English than in Middle English. This kind of fluctuation in rates of change is not unusual! It’s not the case that all aspects of a language have to change at the same speed or at the same time, … Germanic, like Armenian, is extremely conservative in its phonology — the Disintegrating Indo-European consonant system is preserved better in these two branches than in any of the other daughter languages (cf. Polomé 1982 on the archaic nature of Germanic). Unlike Armenian, however, Germanic preserves the older contrast between velars and labiovelars, though, in the course of development, they first became voiceless fricatives and then, at a later date and under certain specific conditions, voiced fricatives (see below for details). Armenian, on the other hand, belongs to the satəm group of languages and is, therefore, descended from that form of Disintegrating Indo-European in which this contrast was replaced by a contrast between palatovelars and plain velars. In the Pre-Germanic dialect of Disintegrating Indo-European, the glottalics were deglottalized, resulting in the following system, with the three-way contrast (1) voiceless aspirated ~ (2) plain (unaspirated) voiceless ~ (3) plain voiced (note: voiced aspirates are not needed in order to account for the Germanic developments): 104 Chapter 6 Bilabial: Dental: Velar: Labiovelar: (1) pº tº kº k¦º (2) p t k k¦ (3) b d g g¦ Note: Glottalization may have been preserved in Pre-Germanic in series 2 above. Glottalization has been proposed to account for the vestjysk stød in Danish and Icelandic preaspiration (cf. Kortlandt 1988b). The following series of changes may be postulated to account for the development of the Disintegrating Indo-European obstruent system into that found in Proto-Germanic: 1. The voiceless aspirates (series 1) become voiceless fricatives: *pº, *tº, *kº, *k¦º > *f, *þ, *χ, *χw, except after *s-. 2. Later, the resulting voiceless fricatives became the voiced fricatives *ƀ, *ð, *ᵹ, and *ᵹw, respectively, except (A) initially and (B) medially between vowels when the accent fell on the contiguous preceding syllable (Verner’s Law). *s was also changed to *z under the same conditions. Cf. Fulk 2018:107—110. 3. *b remained initially, in gemination, and after nasals; *d initially, in gemination, and after nasals, *l, *z, and *g; and *g only in gemination and after nasals. In other positions, however, *b, *d, *g were changed into the voiced fricatives *ƀ, *ð, *ᵹ, respectively. *g¦ became *ᵹ initially and *w medially (cf. Wright—Wright 1925:131). The resulting Proto-Germanic consonant system may thus be reconstructed as follows (cf. Fulk 2018:102—112; Moulton 1972): Stops Bilabial: Dental: Velar: Labiovelar: p t k kw Fricatives b d g (gw) f þ χ χw ƀ ð ᵹ (ᵹw) In Germanic, *a and *o merged into *a, and *ā and *ō merged into *ō. *e become *i (A) before a nasal plus consonant (*eNC > *iNC) and (B) when *i, *ī, or *y followed. *ey became *ī. *i was changed to *e and *u to *o when *a, *o, or *e appeared in the following syllable except when a nasal plus consonant intervened. In the sequences *anχ, *inχ, and *unχ, the n was lost, and the vowels were lengthened. *m̥ , *n̥ , *l̥ , and *r̥ developed into *um, *un, *ul, and *ur, respectively. The Proto-Germanic vowels and diphthongs may be reconstructed as follows: Vowels: i e u a ī ē ū ō The Non-Anatolian Daughter Languages Diphthongs: ay aw 105 ew The consonantal resonants remained unchanged except that final *m became *n. This change is also found in Anatolian, Greek, Celtic, and probably Balto-Slavic. The inner-Germanic developments are quite complicated, and Fulk’s 2018 book mentioned above should be consulted for details. See also Stiles 2017; Bousquette— Salmons 2017:391—398. 6.5. Celtic In the Pre-Celtic dialect of Disintegrating Indo-European, the glottalics were deglottalized, resulting in the following system, with the three-way contrast (1) voiceless aspirated ~ (2) plain (unaspirated) voiceless ~ (3) plain voiced (note: voiced aspirates are not needed in order to account for the Celtic developments): Bilabial: Dental: Velar: Labiovelar: (1) pº tº kº k¦º (2) p t k k¦ (3) b d g g¦ The following discussion will be confined to Old Irish; only the major developments will be discussed: 1. The earlier dental and velar ejectives (*t’ and *k’) merged completely with their plain voiced counterparts (*d and *g) in Pre-Celtic. The developments may be assumed to have been ejective > plain voiceless stop (through deglottalization) > voiced stop (through voicing): *t’ > *t > *d and *k’ > *k > *g. There is no evidence in Proto-Celtic for an earlier bilabial ejective *p’. 2. Next, the voiced labiovelar *g¦ was delabialized and merged with *g. 3. Then, the glottalized labiovelar *k’¦ developed (A) into *b initially and medially after consonants and (B) into *g initially before *u and medially between vowels and before consonants. 4. Original *pº was lost in all of the Celtic languages: *pº > *h > *Ø. However, p has been reintroduced into Old Irish through loanwords. The consonants developed positional allophones under various conditions: 1. Palatal allophones developed in the vicinity of original *i, *ī, *e, and *ē. 2. Velar allophones arose in the vicinity of original *u and *ū. 3. Neutral allophones were found in the vicinity of original *a, *ā, *o, and *ō. In Old Irish, the palatal and velar allophones were indicated as such in writing by surrounding vowels. Unpronounced vowels were often introduced to indicate the quality of the following consonant. /p, t, c, b, d, g/ became the fricatives /f, θ, χ, v, ð, ɣ/ (written 106 Chapter 6 ph, th, ch, b, d, g), respectively, initially after words that end or that formerly ended in a vowel and medially between vowels. /m, n, l, r/ became /μ, ν, λ, ρ/ (written m, n, l, r), respectively, and /s/ became /h/ under the same conditions. /μ/ was probably a nasalized /v/, while /ν, λ, ρ/ were lax variants of /n, l, r/. Consonants were changed as follows initially when the preceding word ended or formerly ended in a nasal: 1. 2. 3. 4. 5. /p, t, c/ became /b, d, g/ (written p, t, c) /b, d/ first became /mb, nd/ and then /mm, nn/ /f/ became /v/ (written b) /n/ was written before vowels /s, r, l, m, n/ were doubled after proclitic vowels Old Irish thus had the following system of consonants (the written form is given first followed by the allophones in slashes): p ph f b m /p, b/ /f/ /f/ /b, v/ /m, μ/ l h t th s d n /l, λ/ /h/ /t, d/ /θ/ /s/ /d, ð/ /n, ν/ r c ch /k, g/ /χ/ g [n] /r, ρ/ /g, ɣ/ /ŋ/ Except for the merger of *ō and *ā into á and of *ī and *ē into í, the long and short vowels were mostly preserved in accented syllables. In unaccented syllables, vowels were either lost or subject to various modifications governed by a complicated set of rules. *i and *u became e and o, respectively, under the influence of a or o in the following syllable. *ew and *ow merged into ó/úa, *ey became é/ía, *oy became óe/oí, and *ay became aí/áe in accented syllables. The Old Irish vowel system was as follows: Vowels: i í e é Diphthongs: íu ía éu/éo áu a á o ó u ú úa uí oí/óe ái/áe *y was lost. *w became f initially and b /v/ after r, l, d. *m, *n, *l, *r were preserved except that final *m became n. In the sequences *Vnt, *Vnc(h), and *Vns, the *n was lost, and the preceding vowel was lengthened. The developments of the syllabic nasals and liquids were complicated, though, in general, *m̥ , *n̥ , *l̥ , *r̥ became am, an, al, ar, respectively, before vowels and em, en, li (le), ri (re), respectively, elsewhere. For details on the inner-Celtic developments, cf. Lewis—Pedersen 1937:1—157; Pedersen 1909 (vol. I); Sims—Williams 2017:361—367; Stifter 2017. The Non-Anatolian Daughter Languages 107 6.6. Slavic In the Pre-Slavic dialect of Disintegrating Indo-European, the glottalics were deglottalized, resulting in the following system, with the three-way contrast (1) voiceless aspirated ~ (2) plain (unaspirated) voiceless ~ (3) plain voiced (note: voiced aspirates are not needed in order to account for the Slavic developments): Bilabial: Dental: Palatovelar: Velar: (Labiovelar: (1) pº tº k¨º kº k¦º (2) p t k¨ k k¦ (3) b d g¨º g g¦) Note: Glottalization may have remained in series 2 into Pre-Balto-Slavic. Glottalization has been proposed as an explanation for Winter’s Law (on which, cf. Collinge 1985: 225—227). In Pre-Slavic, Pre-Baltic, Pre-Indo-Iranian, Pre-Armenian, and Pre-Albanian (the socalled “satəm” languages), the velars developed palatalized allophones when contiguous with front vowels, apophonic *o, and *y. In the early prehistory of these branches, the labiovelars were (perhaps only partially at first) delabialized. The newly delabialized (labio)velars merged with the unpalatalized allophones of the velars. This change brought about the phonemicization of the palatals since both palatalized velars (from earlier plain velars) and unpalatalized velars (from earlier labiovelars) were now found in the vicinity of front vowels, apophonic *o, and *y. The following series of changes may be postulated for the development of the Disintegrating Indo-European system of obstruents into the system found in Proto-Slavic: 1. The ejectives merged completely with the plain voiced stops (*b, *d, *g¨, and *g) in Pre-Slavic. The development may be assumed to have been ejective > plain voiceless stop (through deglottalization) > voiced stop (through voicing): *p’ > *p > *b, *t’ > *t > *d, *k’¨ > *k¨ > *g¨, and *k’ > *k > *g. The loss of glottalization caused lengthening of preceding contiguous short vowels (Winter’s Law). 2. Then, the voiceless aspirates were deaspirated: *pº, *tº, *k¨º, *kº > *p, *t, *k¨, *k. Note: there are a small number of examples in which *kº appears to become *x in ProtoSlavic. These are best explained as borrowings, probably from Iranian (cf. Carlton 1991:95). 3. After *r, *u, *k, *i, *s became *x (> *š before front vowels) (this is the so-called “rukirule”). A similar change is found in Indo-Iranian (see below). 4. *k¨ and *g¨ became *s and *z, respectively. No doubt, the developments were as follows: *k¨ > *t¨ > *͜ts > *s and *g¨ > *d¨ > *͜dz > *z. 5. *k and *g were palatalized to *č and *ž, respectively, before front vowels and *y. 6. The syllabic resonants *m̥ , *n̥ , *l̥ , *r̥ developed into *i (or *u) plus *m, *n, *l, *r, thus: *m̥ , *n̥ , *l̥ , *r̥ > *im, *in, *il, *ir. 108 Chapter 6 7. At a later date, *k and *g were palatalized to *c and *dz, respectively, before *ě (< *oy). *t, *d, *n, *l, *r plus the semivowel *y became *t¨, *d¨, *n¨, *l¨, *r¨, respectively, while *s became *š under the same conditions. 8. *p, *b, *m, *v plus *y became *pl¨, *bl¨, *ml¨, *vl¨, respectively. 9. *a and *o merged into *o, and *ā and *ō merged into *a. *ey and *ī both became *i, and *oy (< *ay and *oy) and *ē became *ě. *ū became *y, *i became *ь, and *u became *ъ. *e plus a nasal became *ę and *o plus a nasal became *ǫ. *ow (< *aw and *ow) became *u. The Common Slavic phonological system may be reconstructed as follows: Stops: Fricatives: Affricates: Nasals: Liquids: p b f t d s z c n r l¨ m l Semivowels: v Vowels: ь e o ъ t¨ d¨ š ž k g x (γ) n¨ r¨ j i y ě u a ę ǫ Note: The palatalized consonants may also be written as follows: *tʹ, *dʹ, *ń, *ŕ, *ĺ. For details on the inner-Slavic developments, cf. Collins 2018; Marc L. Greenberg 2017:522—533; Kim 2018; Shevelov 1964. 6.7. Baltic In the Pre-Baltic dialect of Disintegrating Indo-European, the glottalics were deglottalized, resulting in the following system, with the three-way contrast (1) voiceless aspirated ~ (2) plain (unaspirated) voiceless ~ (3) plain voiced (note: voiced aspirates are not needed in order to account for the Slavic developments): Bilabial: Dental: Palatovelar: Velar: (Labiovelar: (1) pº tº k¨º kº k¦º (2) p t k¨ k k¦ (3) b d g¨º g g¦) The Baltic developments were fairly similar to the early Slavic developments (see above), except that *k¨ and *g¨ became *š and *ž, respectively. As in Pre-Slavic, the The Non-Anatolian Daughter Languages 109 ejectives merged completely with the plain voiced stops in Pre-Baltic. Lithuanian shows the change of *s to *š after *k and *r but not after *i and *u as in Slavic and Indo-Iranian. The syllabic resonants *m̥ , *n̥ , *l̥ , *r̥ developed into *i (or *u) plus *m, *n, *l, *r, thus: *m̥ , *n̥ , *l̥ , *r̥ > *im, *in, *il, *ir. In Lithuanian, t plus j (= y) and d plus j (= y) became či and dži, respectively; t plus l and d plus l became kl and gl, respectively. Except for the merger of *a and *o into *a, *ay and *oy into *ai, and *aw and *ow into *au, the vowel system remained reasonably faithful to that of Disintegrating IndoEuropean. Unlike Slavic and Germanic, Baltic did not merge Disintegrating Indo-European *ā and *ō. The Common Baltic consonant system may be reconstructed as follows: p t k k¨ (?) b d g g¨ (?) š s ž (z) r l m n [ŋ] [n¨] (?) y w For details on the inner-Baltic developments, cf. Kim 2018; Petit 2018; Young 2017: 489—499. 6.8. Albanian Though the Albanian developments are still not completely understood, some tentative conclusions are possible. 1. In Pre-Albanian (as in Pre-Slavic, Pre-Baltic, Pre-Indo-Iranian, and Pre-Armenian), the velars developed palatalized allophones when contiguous with front vowels, apophonic *o, and *y. In the early prehistory of these branches, the labiovelars were (perhaps only partially at first) delabialized. The newly delabialized (labio)velars merged with the unpalatalized allophones of the velars. This change brought about the phonemicization of the palatals since both palatalized velars (from earlier plain velars) and unpalatalized velars (from earlier labiovelars) were now found in the vicinity of front vowels, apophonic *o, and *y. Note: Albanian provides the strongest evidence for the existence of three distinct guttural series in its Disintegrating Indo-European ancestor: the labiovelars are distinguished from the plain velars by the fact that the former are palatalized to sibilants before front vowels, while the latter are not (cf. Mann 1977:24— 25 and 34—35). 2. The ejectives were deglottalized: *p’, *t’, *k’¨, *k’, *k’¦ > *p, *t, *k¨, *k, *k¦. 3. Then, the palatals became palatalized alveolars: *k¨º, *k¨, *g¨ > *t¨º, *t¨, *d¨. These later developed into voiceless and voiced interdental fricatives (*t¨º, *t¨, *d¨ > *θº, *θ, *δ). 110 Chapter 6 4. Next, the plain voiceless stops (from earlier ejectives) became plain voiced stops: *p, *t, *k¨, *k, *k¦ > *b, *d, *g¨, *g, *g¦. In general, the developments of the plain voiced stops and the former ejectives are identical, though initial *g¨ (> *d¨) appears as d, while initial *k’¨ appears as dh (cf. Mann 1977:33). This seems to indicate that the bilabial and dental stops may have developed ahead of and slightly differently from the palatal, velar, and labiovelar stops. 5. Finally, the voiceless aspirates were deaspirated: *pº, *tº, *t¨º, *kº > *p, *t, *t¨, *k. The Albanian developments may be summarized as follows: I palatalization of velars and (partial) delabialization of labiovelars Bilabial: Dental: Palatal: Velar: Labiovelar: pº, (p’), b tº, t’, d k¨º, k’¨, g¨ kº, k’, g k¦º, k’¦, g¦ IV voicing of voiceless stops pº, b tº, d t¨º, d¨÷, d¨ø kº, g k¦º, g¦ > > > > > II deglottalization of ejectives III palatals become palatalized alveolars pº, p, b tº, t, d k¨º, k¨, g¨ kº, k, g k¦º, k¦, g¦ pº, p, b tº, t, d t¨º, t¨, d¨ kº, k, g k¦º, k¦, g¦ V deaspiration of voiceless aspirates > > > > > p, b t, d t¨, d¨÷, d¨ø k, g k¦, g¦ > > > > > > > > > > VI Albanian > > > > > p, b t, d th, d (dh), dh k (q), g (gj) k (q, s), g (gj, z) For details on the inner Albanian developments, cf.: Camaj 1984:1—8; De Vaan 2018; Hamp 1965a; Huld 1984:138—157; Mann 1977:24—25 and 32—36; Orël 2000:1—122; Rusakov 2017:560—572. 6.9. Armenian In the early prehistory of Pre-Armenian, Pre-Indo-Iranian, Pre-Greek, and Pre-Italic, the glottalics first became plain voiceless stops (through deglottalization), and the voiced stops then became voiced aspirates. These were context-free developments. Next, at a later date, in Pre-Indo-Iranian, Pre-Greek, and Pre-Italic, but not in Pre-Armenian, the plain voiceless stops became voiced stops. Again, this was a context-free development in these The Non-Anatolian Daughter Languages 111 daughter languages. Armenian, however, preserves the first stage of this shift — that is to say, the plain voiceless stops remained as such and were not changed to voiced stops. Thus, the Classical Armenian phonological system directly attests the three-way contrast (1) voiceless aspirated ~ (2) plain voiceless ~ (3) voiced aspirated in its occlusive system (on the interpretation of the Classical Armenian sounds traditionally transcribed as /b/, /d/, /g/, /j/, and /ǰ/ as voiced aspirates, cf. Godel 1975:9—10; Garrett 1998; Schirru 2012; Seyfarth—Garellek 2018 [Yerevan dialect]). This is not the whole story, however. There is a tremendous amount of variation in Modern Armenian dialects, in some cases even pointing to the retention of glottalization in series 2. On the other hand, in some dialects, plain voiceless stops correspond to the glottalics, while in still others, plain voiced stops are found (for details, cf. Fleming 2000; Martirosyan 2019:51—60; Vaux 1998). Thus, we have here, in a single, attested IndoEuropean branch, all of the changes that the glottalics are thought to have undergone in the various other non-Anatolian Indo-European daughter languages (retention, voicing, deglottalization)! There is no need to invoke typological parallels nor to formulate elaborate hypotheses — Armenian has it all! For the Pre-Armenian dialect of Disintegrating Indo-European, the obstruent system may be reconstructed as follows, taking into consideration all of the evidence from the Modern Armenian dialects, as well as Classical Armenian (column 1 is voiceless aspirated, column 2 is glottalized, and column 3 is voiced aspirated): Obstruents: (1) pº tº k¨º kº (2) p’ t’ k’¨ k’ (3) bº dº g¨º gº (bilabial) (dental) (palatovelar) (velar) The following series of changes may be postulated for the development of the Disintegrating Indo-European system of obstruents into the system found in Classical Armenian: 1. In Pre-Armenian (as in Pre-Slavic, Pre-Baltic, Pre-Albanian, and Pre-Indo-Iranian), the velars developed palatalized allophones when contiguous with front vowels, apophonic *o, and *y. Next, the labiovelars were (perhaps only partially at first) delabialized. The newly delabialized (labio)velars then merged with the unpalatalized allophones of the velars. This change brought about the phonemicization of the palatals since both palatalized velars (from earlier plain velars) and unpalatalized velars (from earlier labiovelars) were now found in the vicinity of front vowels, apophonic *o, and *y. 2. Next, the glottalics were deglottalized: *p’, *t’, *k’¨, *k’ > *p, *t, *k¨, *k. Note: There are no examples of *p’ in Armenian. 3. Then, the plain voiced stops became voiced aspirates: *b, *d, *g¨, *g > *bº, *dº, *g¨º, *gº. As mentioned above, this was a context-free development. It should be noted that Grassmann’s Law did not operate in Armenian (cf. Vennemann 1989:239). 112 Chapter 6 4. The Pre-Armenian voiced aspirates remained except that, medially between vowels, *bº > w, *g¨º > *jº /͜dzº/ > z, and *gº > ž, while *gº remained initially before back vowels but was changed to ǰ /͜džº/ before front vowels. 5. The syllabic resonants *m̥ , *n̥ , *l̥ , *r̥ developed into *a plus *m, *n, *l, *r: *m̥ , *n̥ , *l̥ , *r̥ > am, an, al, ar (aṙ before n). 6. l became ł before consonants. 7. *w became g or v. 8. *s became h or Ø initially before vowels. 9. As in Indo-Iranian, Slavic, and Lithuanian, *s became š after r. 10. *sk and *ks became c ̣. At a later date, earlier clusters of voiceless stop plus laryngeal developed as follows (cf. Martirosyan 2010:716—717): pH tH kH > > > pº tº x In Armenian, some of the reflexes of the original voiceless aspirates merged with the reflexes of the new voiceless aspirates. This happened in the case of certain onomatopoeic terms, where, for example, original *pº and *kº appear as pº and x, respectively, as if they were from earlier *pH and *kH. In like manner, the aspiration of the original voiceless aspirates was preserved in Armenian after initial *s- (a similar development took place in Indo-Iranian). Finally, *tº and *tH have mostly merged in Armenian, though earlier *rtº has become rd, while *rtH has become rtº (cf. Meillet 1967:104—105 and 1984:78—79). The Armenian developments may be summarized as follows: I Palatalization of velars and delabialization of labiovelars II III Deglottalization Development of ejectives of voiced aspirates pº, (p’), b > pº, (p), b tº, t’, d k¨º, k’¨, g¨ kº, k’, g > tº, t, d > k¨º, k¨, g¨ > kº, k, g IV Classical Armenian (traditional transcription) > pº, (p), bº > h (w, Ø), -, b (w) > tº, t, dº > tº, t, d > k¨º, k¨, g¨º > s, c, j (z) > kº, k, gº > kº, k, g (ǰ, ž) Note: As noted above, glottalization is preserved in some Modern Armenian dialects. The Classical Armenian (Grabar) consonant system is as follows: The Non-Anatolian Daughter Languages Stops Affricates Fricatives Liquids Glides Voiceless Aspirated Voiceless Unaspirated Voiced Nasals pº tº kº cº čº p t k c č s š h x l r v w b d g j ȷ̌ z ž m n 113 ł ṙ y Notes: 1. The voiceless aspirated series is more often transcribed as follows: pꜤ, tꜤ, kꜤ, cꜤ, čꜤ. 2. As noted above, /b/, /d/, /g/, /j/, and /ǰ/ were voiced aspirates. Armenian is the only non-Anatolian daughter language that has preserved a trace of a consonantal laryngeal. Kuryłowicz’s *š and *› (Sturtevant’s *x and *¦) appear as h initially before full-grade vowels in a small number of words (cf. Winter 1965:102). The following examples have cognates in the Anatolian languages: A. Armenian hav ‘grandfather’ (< Pre-Armenian *hawhos): Hittite ḫu-uḫ-ḫa-aš ‘grandfather’; Hieroglyphic Luwian huhas ‘grandfather’; Lycian χuga- ‘grandfather’. Cf. Latin avus ‘grandfather’; Gothic awō (f.) ‘grandmother’; Old Irish áue ‘grandson’; Lithuanian avýnas ‘uncle’. Puhvel 1984— .3:355—358; Kloekhorst 2008a:352—353. B. Armenian hoviw ‘shepherd’ (< Pre-Armenian *howi-pā-): Hittite (nom. sg. or pl. ?) ḫa-a-u-e-eš ‘sheep’; Luwian ḫa-a-ú-i-iš ‘sheep’; Hieroglyphic Luwian hawis ‘sheep’; Lycian χava ‘sheep’. Cf. Sanskrit ávi-ḥ ‘sheep’; Greek ὄϊς, οἶς ‘sheep’; Latin ovis ‘sheep’; Lithuanian avìs ‘sheep’. Kloekhorst 2008a:337—338 *h3e„i-; Puhvel 1984— .3:279—280; Wodtko—Irslinger—Schneider 2008:335—339. C. Armenian haravunkº ‘arable land’ (< Pre-Armenian *har- ‘to plow’): Hittite ḫaraš-zi ‘to plow’. Cf. Greek ἀρόω ‘to plow, to till’; Latin arō ‘to plow, to till’; Gothic arjan ‘to plow’; Lithuanian ariù ‘to plow, to till’; Tocharian B āre ‘plow’. But note Armenian arawr ‘plow’ without initial h. Puhvel (1984— .3:184—185), on the other hand, derives the Hittite form from Akkadian ḫarāšu ‘to plant’ or ḫarāṣu ‘to dig a furrow’; but cf. Tischler 1977— :182—183; Kloekhorst 2008a:312—314. D. Armenian hogi ‘wind, spirit’ (< Pre-Armenian *howyo-), hov ‘wind’, hovem ‘to let air in’: Hittite ḫu-wa-an-za ‘wind’. Cf. Sanskrit vā́ ti ‘to blow’; Greek ἄημι ‘to blow, to breathe’; Latin ventus ‘wind’; Gothic winds ‘wind’; Tocharian A want ‘wind’; Lithuanian vjjas ‘wind’. Puhvel 1984— .3:428—429; Kloekhorst 2008a:368. 114 Chapter 6 E. Armenian han ‘grandmother’ (< Pre-Armenian *hano-s): Hittite (gen. pl.) ḫa-anna-aš ‘grandmother’; Lycian χñna- or χñni- ‘grandmother’. Cf. Old High German ana ‘grandmother’; Latin anus ‘old woman’; Old Prussian ane ‘grandmother’. Puhvel 1984— .3:84—86; Kloekhorst 2008a:285—286. F. Armenian harkanem ‘to split, to fell’ (< Pre-Armenian *hark’-): Hittite ḫar-ga-aš ‘destruction, ruin’, ḫa-ar-ak-zi ‘to be destroyed’. Cf. Old Irish orgaim ‘to strike, to destroy’. This etymology is rejected by Puhvel 1984— .3:157—168; but cf. Kloekhorst 2008a:306—307; Benveniste 1935:162. G. Armenian hac ̣i ‘ash-tree’ (< Pre-Armenian *hask¨o-): Hittite GIŠḫa-aš-ši-ka4-aš ‘a tree and its fruit (?)’. Cf. Old Icelandic askr ‘ash-tree’; Old High German ask ‘ashtree’. This comparison is not mentioned in Puhvel 1984— .3:232; but cf. Tischler 1977— :200—201; Kloekhorst 2008a:324. H. Armenian Hay ‘Armenian’: Hittite Ḫayaša the name of a region (cf. Meillet 1936: 9). No doubt this term has been borrowed by Armenian. The following examples have no known Anatolian cognates: I. Armenian hav ‘bird’ (< Pre-Armenian *hawi-s): Latin avis ‘bird’; Sanskrit ví-ḥ ‘bird’. J. Armenian hot ‘smell’ (< Pre-Armenian *hot’os-): Latin odor ‘smell’; Greek ὄζω ‘to smell’. K. Armenian hum ‘raw’ (< Pre-Armenian *hōmo-s): Sanskrit ām#-ḥ ‘raw’; Greek ὠμός ‘raw’. The Armenian material is not without problems, however. Both Meillet (1936:38) and Winter (1965b:102) point out that initial h is unstable. This means that the same word sometimes has two alternates, one with h- and one without — Meillet’s example is hogi ‘wind, spirit’ beside ogi. Furthermore, h- is sometimes missing where the Hittite cognate unequivocally points to original *H2 such as in Armenian arcatº ‘silver’ beside Hittite ḫarkiš ‘white’ (other cognates include Greek ἀργός ‘bright, white’ and Latin argentum ‘silver’). Consequently, the Armenian material, though extremely valuable, must be used with caution. The Neogrammarians and their followers — with the exception of Ferdinand de Saussure — did not reconstruct laryngeals as part of the Proto-Indo-European phonological system. However, they had all of the tools at their disposal to do so. First of all, as early as 1878, de Saussure had posited his now famous “coefficients sonantiques” solely on the basis of an analysis of the patterns of vowel gradation. Secondly, Armenian has a clear reflex of one of de Saussure’s “coefficients”. Unfortunately, the Armenian evidence escaped detection until after the discovery in 1927 by Kuryłowicz that one of de Saussure’s “coefficients” was preserved in Hittite. It was only then that the Armenian material was reexamined by Austin (1942:22—25) and the laryngeal reflex found. It should be noted that Albert Cuny made the same discovery at the same time (1927) as Kuryłowicz. For more information on the inner-Armenian developments, cf. Fleming 2000; Godel 1975:69—91; Macak 2017; Meillet 1936: 23—59; Vaux 1998; Martirosyan 2010:705— 748. The Non-Anatolian Daughter Languages 115 6.10. Indo-Iranian For the Pre-Indo-Iranian dialect of Disintegrating Indo-European, the obstruent system may be reconstructed as follows (column 1 is voiceless aspirated, column 2 is plain voiceless, and column 3 is voiced aspirated) (note: glottalization is not needed in order to account for the Indo-Iranian developments of series 2): Obstruents: (1) pº tº k¨º kº k¦º (2) p t k¨ k k¦ (3) bº dº g¨º gº g¦º (bilabial) (dental) (palatovelar) (velar) (labiovelar) The changes leading from the Pre-Indo-Iranian to Proto-Indo-Iranian are particularly complicated. The first three steps are identical to what is assumed to have happened in PreArmenian (and also Pre-Greek and Pre-Italic). 1. In Pre-Indo-Iranian (as in Pre-Slavic, Pre-Baltic, Pre-Albanian, and Pre-Armenian), the velars developed palatalized allophones when contiguous with front vowels, apophonic *o, and *y. Next, the labiovelars were (perhaps only partially at first) delabialized. The newly delabialized (labio)velars then merged with the unpalatalized allophones of the velars. This change brought about the phonemicization of the palatals since both palatalized velars (from earlier plain velars) and unpalatalized velars (from earlier labiovelars) were now found in the vicinity of front vowels, apophonic *o, and *y. 2. Next, the glottalics were deglottalized: *p’, *t’, *k’¨, *k’ > *p, *t, *k¨, *k. 3. Then, the plain voiced stops became voiced aspirates: *b, *d, *g¨, *g > *bº, *dº, *g¨º, *gº. This was a context-free development. This was the stage reached by Armenian. 4. When two voiced aspirates cooccurred in a root, the first was deaspirated (Grassmann’s Law). It should be noted that Grassmann’s Law only appears in Old Indic. In Iranian (Old Persian and Avestan), the plain voiced stops and the voiced aspirates have the same treatment (cf. Kent 1953:29). 5. In Pre-Indo-Iranian (and in Pre-Greek and Pre-Italic), but unlike Pre-Armenian, the plain (unaspirated) voiceless stops (from earlier glottalics) developed into plain (unaspirated) voiced stops: *p, *t, *k¨, *k > *b, *d, *g¨, *g. This was a context-free development. (As a typological parallel, it may be noted that an identical change has taken place in the Northwest Caucasian languages Šapsegh and Kabardian.) Obstruents: (1) pº tº k¨º kº k¦º (2) b d k¨ g g¦ (3) bº dº g¨º gº g¦º (bilabial) (dental) (palatovelar) (velar) (labiovelar) 116 Chapter 6 6. The imbalance caused by the voicing of the plain voiceless stops caused the voiceless aspirates to be partially deaspirated: Obstruents: (1) p t k¨ k k¦ (2) pº tº k¨º kº k¦º (3) b d k¨ g g¦ (4) bº dº g¨º gº g¦º (bilabial) (dental) (palatovelar) (velar) (labiovelar) The deaspiration took place everywhere except (A) after initial *s- and (B) in onomatopoeia. However, aspiration was lost in the clusters *spº-, *stº-, *skº- when an earlier laryngeal followed in the stem or when another aspirated stop followed in the stem: *(s)tºeHy- > *(s)teHy- > *(s)tāy- (cf. Sanskrit stāyati ‘he, she steals’, stāyú-ḥ, tāyú-ḥ ‘thief, robber’); *(s)tºeHi- > *(s)teHi- > *(s)tai- (cf. Sanskrit stená-ḥ ‘thief’, stéya-ḥ ‘theft, robbery’). *(s)tºenH- > *(s)tenH- > *(s)ten- (cf. Sanskrit stanati ‘resounds, reverberates’). Note: Apparent exceptions to these rules appear to be due to the generalization of variant forms of the stems in question, or, in some cases, they are due to borrowing. 7. Additional voiceless aspirates arose from earlier clusters of voiceless stop plus laryngeal: *pH, *tH, *kH > *pº, *tº, *kº, respectively. 8. *s was changed into *š after *r, *u, *k, *i (this is the so-called “ruki-rule”). A similar change is also found in Slavic. 9. *k¨, *g¨, *g¨º were affricated to *͜ts, *͜dz, *͜dzº, respectively (cf. Burrow 1973:74). 10. Following that, the velars *k, *g, *gº were palatalized to *k¨, *g¨, *g¨º, respectively, before *ē̆, *ī̆, and *y (cf. Mayrhofer 1972:24). Note: *kº was not palatalized. 11. After the palatalization of the velars had taken place, the short vowels merged into *a, and the long vowels merged into *ā. Earlier *o became ā in open syllables (Brugmann’s Law). 12. The syllabic nasals became a, and the syllabic laryngeal (*h̥ ) partially merged with i. 13. *h was then lost after a (< *m̥ and *n̥ ) with compensatory lengthening. 14. *r and *l merged into r, and *r̥ and *l̥ merged into r̥ . In Avestan and Old Persian, the plain and aspirated voiced stops merged. The voiceless aspirates became fricatives except after a sibilant, where they were deaspirated. The plain voiceless stops developed into fricatives when immediately followed by a consonant unless a sibilant preceded. In Old Indic (Vedic and Classical Sanskrit), *͜dz and *g¨ merged into j, and *͜dzº and *g¨º merged into h. The Old Indic phonological system was as follows (column 1 is plain voiceless, column 2 is voiceless aspirated, column 3 is plain voiced, column 4 is voiced aspirated, and column 5 is nasal (cf. Burrow 1973:67—117; Kobayashi 2017:231; Mayrhofer 1972:17): The Non-Anatolian Daughter Languages (1) k c ṭ t p y ś h ḥ ṁ (2) kh ch ṭh th ph r ṣ (3) g j ḍ d b l s (4) gh jh ḍh dh bh v (5) ṅ ñ ṇ n m Vowels: a ā i ī u ū ṛ ṝ ḷ Diphthongs: ai au Velar: Palatal: Retroflex: Dental: Bilabial: Semivowels: Sibilants: Aspirate: Visarga: Anusvāra: e 117 o Once the above system was established, it remained remarkably stable for well over three thousand years — the phonological systems of the modern Indo-Aryan languages remain to this day similar in structure to the phonological system of Old Indic (cf. Bloch 1965:96—97; Kobayashi 2004; see Ghatage 1962 for examples). This fact raises an interesting question about the phonological system reconstructed for the Indo-European parent language by the Neogrammarians: The Neogrammarian reconstruction is extremely close to the phonological system of Old Indic. If the Neogrammarian system were in fact an accurate representation of what had existed in Proto-Indo-European, one may legitimately ask why it, too, did not remain stable in the majority, if not all, of the IndoEuropean daughter languages. It thus seems to be a fair conclusion that the Proto-IndoEuropean phonological system was not in fact similar to that of Old Indic and that the Old Indic system was an innovation. 6.11. Greek For the Pre-Greek dialect of Disintegrating Indo-European, the obstruent system may be reconstucted as follows (column 1 is voiceless aspirated, column 2 is plain voiceless, and column 3 is voiced aspirated) (note: glottalization in is not needed in order to account for the Greek developments of series 2): Obstruents: (1) pº tº kº k¦º (2) p t k k¦ (3) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) Many of the early Pre-Greek developments were similar to what is assumed to have happened in Pre-Armenian and Pre-Indo-Iranian. However, Greek is a so-called “centum” 118 Chapter 6 language, which means that it initially preserved the original contrast between velars and labiovelars. Unlike Pre-Armenian and Pre-Indo-Iranian, but similar to Italic, Greek changed the voiced aspirates into voiceless aspirates. The following series of changes may be postulated for the development of the Disintegrating Indo-European system of obstruents into the system found in Proto-Greek: 1. First, the glottalics were deglottalized: *p’, *t’, *k’, *k’¦ > *p, *t, *k, *k¦. 2. Then, the plain voiced stops became voiced aspirates: *b, *d, *g, *g¦ > *bº, *dº, *gº, *g¦º. This was a context-free development. 3. As in Old Indic, when two voiced aspirates cooccurred in a root, the first was deaspirated (Grassmann’s Law). 4. In Pre-Greek (and in Pre-Indo-Iranian and Pre-Italic), but unlike Pre-Armenian, the plain (unaspirated) voiceless stops (from earlier glottalics) developed into plain (unaspirated) voiced stops: *p, *t, *k, *k¦ > *b, *d, *g, *g¦ (cf. Gamkrelidze—Ivanov 1995.I:52—57). This was a context-free development. Obstruents: (1) pº tº kº k¦º (2) b d g g¦ (3) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) 5. The imbalance caused by the voicing of the plain voiceless stops caused the voiceless aspirates to be partially deaspirated. Obstruents: (1) p t k k¦ (2) pº tº kº k¦º (3) b d g g¦ (4) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) Note: Emonds (1972:120) also assumes that some of the examples of voiceless aspirates found in Indo-Iranian, Greek, and Armenian are derived from the original voiceless aspirates, that is to say, they failed to undergo the expected deaspiration. Edmonds accounts for this by “reintroduction from a dialect that did not undergo Z2 [deaspiration]”. In other words, he sees them as borrowings. While this may be true in some cases, I prefer to see them mostly as the natural result of developments within these branches themselves. 6. Additional voiceless aspirates arose from earlier clusters of voiceless stop plus laryngeal: *pH, *tH, *kH > *pº, *tº, *kº, respectively. 7. At a later date, the voiced aspirates were devoiced — the unaspirated allophones became plain (unaspirated) voiceless stops, and the aspirated allophones became voiceless aspirates: *b ~ *bº, *d ~ *dº, *g ~ *gº, *g¦ ~ *g¦º > *p ~ *pº, *t ~ *tº, *k ~ *kº, *k¦ ~ *k¦º. The newly-formed plain and aspirated voiceless stops merged completely with the previously-existing plain and aspirated voiceless stops. As a The Non-Anatolian Daughter Languages 119 typological parallel, it may be noted that similar devoicing of earlier voiced aspirates took place in Romany (cf. Meillet 1967:100 and 1984:76). The labiovelars were eliminated in Greek in historic times. The process of elimination probably occurred in several stages. Since the labiovelars mostly remain in Mycenaean, their elimination can reasonably be placed between the Mycenaean period and the beginning of the alphabetic period, that is, between about 1400—900 BCE (cf. Lejeune 1972:43—53). The developments were as follows: 1. Before or after u, *k¦, *k¦º, and *g¦ were delabialized, and the resulting phonemes merged with k, kº, and g (written κ, χ, and γ), respectively. 2. Next, *k¦, *k¦º, and *g¦ were palatalized before ē̆ and ī̆. The resulting sounds then merged with t, tº, and d (written τ, θ, and δ), respectively, in the majority of Greek dialects. 3. Finally, all remaining labiovelars became bilabials: *k¦, *k¦º, and *g¦ > p, pº, and b (written π, φ, and β). *m, *n, *l, *r generally remained in Greek except that final *-m became -n (written ν) as in Anatolian, Germanic, Celtic, and probably Baltic and Slavic. *m̥ , *n̥ , *l̥ , *r̥ developed into αμ, αν, αλ, αρ, respectively, before vowels. Before consonants, *m̥ and *n̥ merged into α, while *l̥ and *r̥ became αλ/λα and αρ/ρα, respectively. *s, *y, and *w were lost medially between vowels. Initially before vowels, *s became h (written ʽ), *y became either h or z (written ʽ and ζ, respectively), while *w was lost in Attic-Ionic. *s remained when final and when before or after voiceless stops. The vowels and diphthongs were well-preserved in all of the Greek dialects. The most important change was that of ᾱ to η in Attic-Ionic. Additional changes worth mentioning include the compensatory lengthening of short vowels, the shortening of long vowels, and the development of new long vowels through contraction. For more information on the Greek developments, cf. Lejeune 1972:187—263. 6.12. Italic Italic is divided into two distinct branches, namely, Oscan-Umbrian (also called Sabellian or Sabellic) and Latin-Faliscan. The Oscan-Umbrian branch includes a number of poorly-attested languages besides Oscan and Umbrian — these include Aequian, Marrucinian, Marsian, Paelignian, Sabinian, Southern Picenian, Vestinian, and Volscian (cf. Sihler 1995:14). The differences between Oscan-Umbrian, on the one hand, and LatinFaliscan, on the other, are extremely pronounced, so much so that some scholars deny any special relationship between these two groups and see them instead as two separate branches of Indo-European (for details about the issues involved, cf. Beeler 1966:51—58). For the Pre-Italic dialect of Disintegrating Indo-European, the obstruent system may be reconstructed as follows (column 1 is voiceless aspirated, column 2 is plain voiceless, and column 3 is voiced aspirated) (note: glottalization is not needed in order to account for the Italic developments of series 2): 120 Chapter 6 Obstruents: (1) pº tº kº k¦º (2) p t k k¦ (3) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) Many of the early Pre-Italic developments were similar to what is assumed to have happened in Pre-Greek. Like Greek, Italic belonged to the so-called “centum” languages, which means that it initially preserved the original contrast between velars and labiovelars. The following series of changes may be postulated for the development of the Disintegrating Indo-European system of obstruents into the system found in Proto-Italic: 1. First, the glottalics were deglottalized: *p’, *t’, *k’, *k’¦ > *p, *t, *k, *k¦. 2. Then, the plain voiced stops became voiced aspirates: *b, *d, *g, *g¦ > *bº, *dº, *gº, *g¦º. This was a context-free development. Note: Grassmann’s Law did not operate in Italic. 3. In Pre-Italic (and in Pre-Indo-Iranian and Pre-Greek), but unlike Pre-Armenian, the plain (unaspirated) voiceless stops (from earlier glottalics) developed into plain (unaspirated) voiced stops: *p, *t, *k, *k¦ > *b, *d, *g, *g¦ (cf. Gamkrelidze—Ivanov 1995.I:57—65). This was a context-free development. Obstruents: (1) pº tº kº k¦º (2) b d g g¦ (3) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) 4. The imbalance caused by the voicing of the plain voiceless stops caused the voiceless aspirates to be partially deaspirated. Obstruents: (1) p t k k¦ (2) pº tº kº k¦º (3) b d g g¦ (4) bº dº gº g¦º (bilabial) (dental) (velar) (labiovelar) 5. Additional voiceless aspirates arose from earlier clusters of voiceless stop plus laryngeal: *pH, *tH, *kH > *pº, *tº, *kº, respectively. 6. At a later date, the voiced aspirates were devoiced: *bº, *dº, *gº, *g¦º > *pº, *tº, *kº, *k¦º. The newly-formed aspirated voiceless stops merged completely with the previously-existing aspirated voiceless stops. 7. Finally, the voiceless aspirates (from earlier voiced aspirates as well as from clusters of voiceless stop plus laryngeal) became voiceless fricatives. The Non-Anatolian Daughter Languages bº dº gº g¦º > > > > pº tº kº k¦º > > > > φ θ χ χ¦ > > > > 121 f f h f In Oscan and Umbrian, *φ, *θ, and *χ¦ merged into f, while *χ became h. In Latin, the merger of *φ, *θ, and *χ¦ into f only took place initially. *φ became b medially; *θ became (A) d medially but (B) b before or after r, before l, or after u; and *χ¦ became (A) v between vowels, (B) gu after n, but (C) g before consonants or u. *χ became (A) h initially in Latin but (B) g when before or after consonants and (C) f when before u. *m, *n, *l, *r were preserved. *y remained initially in Latin (written i) but was lost between vowels, while *w (written v) was unchanged. *m̥ , *n̥ , *l̥ , *r̥ developed into a plus m, n, l, r, respectively, before vowels. Elsewhere, *l̥ and *r̥ became ol and or, respectively, and *m̥ and *n̥ became em and en, respectively. *s generally remained, though it was voiced to z between vowels. The z was retained in Oscan but was changed to r in Umbrian and Latin. The vowels generally remained in accented syllables but were weakened or lost in unaccented syllables. The vowels underwent the following modifications in Latin. Final i became e. e became i before ng, gn, nc, and ngu. e became o before or after w and before l. o became u (1) before nc, ngu, mb, and before l plus a consonant, (2) in final syllables ending in a consonant, and (3) medially before l or before two consonants. vo became ve before r plus a consonant, before s plus a consonant, and before t. ov became av. The diphthongs were preserved in Oscan but underwent various changes in Umbrian and Latin. ei became ī, and oi, eu, and ou became ū in Latin. For details on the inner-Italic developments, cf. Buck 1933:78—167 (Latin and Greek); Lindsay 1894:219—315; Meisser 2017:743—751; Sihler 1995: 35—242 (Latin and Greek); Stuart-Smith 2004; Wallace 2017: 325—329; Weiss 2009:31—193. 6.13. Phrygian and Thracian Like Germanic and Armenian, Phrygian is usually considered to be a relic area in which the Proto-Indo-European stop system is better preserved than it is in the remaining daughter languages (cf. Gamkrelidze—Ivanov 1995.I:804—805). Unfortunately, the Phrygian corpus is so small that it is not possible to trace all of the developments with certainty. However, the following developments are clear (except as noted) (cf. Diakonoff— Neroznak 1985:5—6; Neroznak 1992:272—274; Ligorio—Lubotsky 2013:184—187 and 2018:1821—1824; Fortson 2010:461—462; Georgiev 1981:131—132; Obrador-Cursach 2020:62—76; R. Woodhouse 2006 and 2010): Proto-Indo-European b pº d tº Phrygian > > > > b p (also ph) d t (also th) 122 Chapter 6 t’ g, g¦ kº, k¦º k’, k’¦ g¨ k¨ k’¨ > > > > > > > t g k (also kh) k z (?) s (?) z (?) Note: The reflexes of the palatovelars are unclear. According to Fortson (2010: 461) and Obrador-Cursach (2020:70), Phrygian was a centum language, as was most certainly Thraco-Dacian (cf. Vinereanu 2023). As can be seen, the voiced stops remained unchanged. The voiceless aspirates also remained unchanged, though the aspiration is usually not indicated in the writing. Finally, the glottalics were simply deglottalized. It should be mentioned, however, that this interpretation is challenged by Brixhe (1994:171—172 and 2004:782). Phrygian had five short vowels (a, e, i, o, u) and at least four long vowels (ā, ī, ō, ū), though the long vowels were not indicated in the writing. Proto-Indo-European *ē and *ā merged into ā in Phrygian. Cf. Obrador-Cursach 2020:62. The Thracian developments appear to be similar to those given above for Phrygian (cf. Georgiev 1981:118—119; see also Brixhe—Panayotou 1994a:198—199; Katičić 1976.I: 128—153), though this interpretation has recently been called into question by the work of Svetlana Yanakijeva. 6.14. Concluding Remarks The phonological developments in each Indo-European branch are far more complicated than indicated in this chapter. Only the main lines of development have been traced here, the principal purpose being to show how the Glottalic Model of Proto-IndoEuropean consonantism can account for the phonological developments in the main nonAnatolian Indo-European daughter languages in a completely natural, straightforward, and plausible manner. To that end, trajectories of the developments in each branch have been provided. The various comparative grammars and handbooks listed in the references at the end of this book should be consulted for details about further developments in the various daughter languages. Particular emphasis has been placed in this chapter on Armenian as the key to understanding the developments in Old Indic, Greek, and Italic. It may be noted that the trajectories for each of these branches leads directly to the traditional, Neogrammarian system of stops at a certain period in their prehistory (cf. Joseph—Wallace 1994) (column 1 is plain voiceless, column 2 is voiceless aspirated, column 3 is plain voiced, and column 4 is voiced aspirated): The Non-Anatolian Daughter Languages Obstruents: (1) p t k¨ k k¦ (2) pº tº k¨º kº k¦º (3) b d k¨ g g¦ (4) bº dº g¨º gº g¦º 123 (bilabial) (dental) (palatovelar) (velar) (labiovelar) This is identical to Brugmann’s reconstruction (1904:52), though Brugmann uses a different transcription from that given above, namely: Occlusives: (1) p t % q qß (2) ph th %h qh qßh (3) b d “ œ œß (4) bh dh “h œh œßh (bilabial) (dental) (palatal) (pure velar) (labiovelar) However — and this must be strongly emphasized —, such a reconstruction is only needed to account for developments in these branches. This is clearly a late development in the Disintegrating Indo-European antecedents of Old Indic, Greek, and Italic and is not in any way representative of earlier periods of development within the Indo-European parent language.  References Adams, Douglas Q. 1988 Tocharian Historical Phonology and Morphology. New Haven, CT: American Oriental Society. 2017 “Tocharian”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 452—475. 2013 A Dictionary of Tocharian B. Revised and greatly enlarged 2nd edition. Amsterdam and New York, NY: Rodopi. Adiego, Ignacio J. 1995 “Contributiones al descriframiento del Cario” [Contributions to the Decipherment of Carian], Kadmos 34:18—34. 2007 The Carian Language. With an appendix by Koray Konuk. Leiden and Boston, MA: E. J. Brill. 2010 “Recent Developments in the Decipherment of Carian”, in: Riet van Bremen and Jan-Mathieu Carbon (eds.), Hellenistic Karia: Proceedings of the First International Conference on Helenistic Karia, Oxford 29 June — 2 July 2006. Bordeaux: De Boccard, pp. 147—176. 2016 “Anatolian Languages and Proto-Indo-European”, Veleia 33:49—64. 2019 “Archaic Carian”, in: Olivier Henry and Koray Konuk (eds.), Karia Arkhaia: La Carie, des origines à la période pré-hékatomnide [Archaic Caria: Caria from Its Origins to the Pre-Hecatomnid Period]. 4èmes Rencontres d’Archéologie de l’Iféa, Istanbul, 14—16 November 2013. Istanbul: Institut Français d’Études Anatoliennes Georges Dumézil, pp. 23—41. Adiego, Ignasi Xavier, José Virgilio García Trabazo, Mariona Vernet, Bartomen Obrador Cursach, Elenaa Martínez Rodrigues (eds.) 2019 Luwic Dialects and Anatolian: Inheritance and Diffusion. Barcelona: Universitat de Barcelona. Adrados, Francisco R[odríguez] 1961 Estudios sobre las laringales indoeuropeas [Studies on the IndoEuropean Laryngeals]. (= Manuales y anejos de “Emerita” 19.) Madrid: Consejo Superior de Investigaciones Cientificas (C.S.I.C.). 1975 Lingüística indoeuropea [Indo-European Linguistics]. 2 vols. Madrid: Gredos. 1981a “Further Considerations on the Phonetics and Morphologization of Hü and Hß in Indo-European”, Emerita XLIX.1:231—271. 1981b “More on Laryngeals with Labial and Palatal Appendices”, Folia Linguistica Historica 2.2:191—235. 1988 Nuevos estudios de lingüística indoeuropea [New Studies in IndoEuropean Linguistics]. Madrid: Instituto de Filología: Manuales y Anejos de “Emerita” XXXVII. 126 References Adrados, Francisco R[odríguez], Alberto Bernabé, and Julia Mendoza 1995—1998 Manual de lingüística indoeuropea [Manual of Indo-European Linguistics]. 3 vols. Madrid: Ediciones Clásicas. [2010] [English translation. Vol. I. Louvain: Peeters.] Al-Ani, Salman 1970 Arabic Phonology. The Hague: Mouton. Allen, W[illiam] Sydney 1956 “Structure and System in the Abaza Verbal Complex”, Transactions of the Philological Society 55.1:127—176. 1965 “On One Vowel Systems”, Lingua 13:111—124. 1973 Accent and Rhythm, Cambridge: Cambridge University Press. 1976 “The PIE Aspirates: Phonetic and Typological Factors in Reconstruction”, in: Aphonse Juilland (ed.), Linguistic Studies Offered to Joseph Greenberg. Saratoga, CA: Anma Libri, vol. 2, pp. 237—247. Anderson, Stephen R. 1978 “Syllables, Segments, and Northwest Caucasian Languages”, in: Alan Bell and Joan B. Hooper (eds.), Syllables and Segments. (= NorthHolland Linguistic Series 40. Papers from the Symposium on Segment Organization and the Syllable, Boulder, Colorado, October 21—23, 1977.) Amsterdam: North Holland Publishing Company, pp. 47—58. Anthony, David W. 2013 “Two IE Phylogenies, Three PIE Migrations, and Four Kinds of Steppe Pastoralism”, Journal of Language Relationship 9:1—21. 2019 “Archaeology, Genetics, and Language in the Steppes: A Comment on Bomhard”, Journal of Indo-European Studies 47.1/2:175—198. Anttila, Raimo 1969 Proto-Indo-European Schwebeablaut. Berkeley and Los Angeles, CA: University of California Press. Arbeitman, Yoël L. (ed.) 1988 A Linguistic Happening in Memory of Ben Schwartz: Studies in Anatolian, Italic, and Other Indo-European Languages. Louvain-laNeuve: Peeters. 2000 The Asia Minor Connection: Studies on the Pre-Greek Languages in Memory of Charles Carter. Leuven and Paris: Peeters. Arbeitman, Yoël L., and Allan R. Bomhard (eds.) 1981 Bono Homini Donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns. (= Current Issues in Linguistic Theory, vol. 16, parts I and II.) Amsterdam: John Benjamins. Arbeitman, Yoël L.. and Gilbert-James Ayala 1981 “Rhotacism in Hieroglyphic Luwian”, in: Yoël L. Arbeitman and Allan R. Bomhard (eds.), Bono Homini Donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns. 2 vols. Amsterdam and Philadelphia, PA: John Benjamins, vol. 1, pp. 25—28. References Archi, Alfonso 2010 127 “When Did the Hittites Begin to Write in Hittite?”, in: Yoram Cohen, Amir Gilan, and Jared L. Miller (eds.), Pax Hethitica: Studies on the Hittites and Their Neighbors in Honor of Itamar Singer. Wiesbaden: Harrassowitz Verlag, pp. 37—46. Austin, William A. 1942 “Is Armenian an Anatolian Language?”, Language 18:22—25. Bader, Françoise 1988 “Noms de parenté anatoliens et formations à laryngale” [Anatolian Names of Relationship and Laryngeal Formations], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 17—48. 1990 “Traitements de laryngales en groupe: allongement compensatoire, assimilation, anaptyxe” [The Treatments of Laryngeals in Group: Compensatory Lengthening, Assimilation, Anaptyxis], in: Jean Kellens (ed.), La reconstruction des laryngales [The Reconstruction of the Laryngeals]. Paris: Société d’Édition “Les Belles Lettres”, pp. 1—47. Ballester, Xaverio 2023 “On Indo-European Triliteralism”, International Journal of Human Cultural Studies 33:599—605. Bammesberger, Alfred 1984 Studien zur Laryngaltheorie [Studies on the Laryngeal Theory]. Göttingen: Vandenhoeck & Ruprecht. 1989 “The Laryngeal Theory and the Phonology of Prehistoric Greek”, in: Theo Vennemann (ed.), The New Sound of Indo-European: Essays in Phonological Reconstruction. Berlin and New York, NY: Mouton de Gruyter, pp. 35—41. Bammesberger, Alfred (ed.) 1988 Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter. Barton, Charles R. 1988 “Gk. ā̆esa, Arm. agay and PIE *h÷”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Lautund Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 49—58. Beckman, Gary 1996 “The Hittite Language and Its Decipherment”, Bulletin of the Canadian Society for Mesopotamian Studies 31:23—30. 2011 “The Hittite Language: Recovery and Grammatical Sketch”, in: Sharon R. Steadman and Gregory McMahon (eds.), The Oxford Handbook of 128 References Ancient Anatolia, 10,000—323 B.C.E. Oxford and New York, NY: Oxford University Press, pp. 517—533. Beckman, Gary, Richard Beal, and Gregory McMahon (eds.) 2003 Hittite Studies in Honor of Harry A. Hoffner Jr. on the Occasion of His 65th Birthday. Winona Lakes, IN: Eisenbrauns. Beekes, Robert S[tephen] P[aul] 1969 The Development of the Proto-Indo-European Laryngeals in Greek. The Hague: Mouton. 1981 “Intervocalic Laryngeal in Gatha-Avestan”, in: Yoël L. Arbeitman and Allan R. Bomhard (eds.), Bono Homini Donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns. Amsterdam: John Benjamins, part I, pp. 47—64. 1985 The Origins of the Indo-European Nominal Inflection. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck. 1988 “Laryngeal Developments: A Survey”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 59—105. 1989 “The Nature of the Proto-Indo-European Laryngeals”, in: Theo Vennemann (ed.), The New Sound of Indo-European: Essays in Phonological Reconstruction. Berlin and New York, NY: Mouton de Gruyter, pp. 23—33. 1995 Comparative Indo-European Linguistics: An Introduction. Amsterdam and Philadelphia, PA: John Benjamins. [2011] [2nd edition, revised and corrected by Michiel de Vaan.] 2003 “Luwians and Lydians”, Kadmos 42:47—49. 2010 Etymological Dictionary of Greek. With the assistance of Lucien van Beek. 2 vols. Leiden and Boston, MA: E. J. Brill. Beeler, Madison S. 1966 “Interrelationships with Italic”, in: Henrik Birnbaum and Jaan Puhvel (eds.), Ancient Indo-European Dialects. Berkeley and Los Angeles, CA: University of California Press, pp. 51—58. Benveniste, Émile 1935 Origines de la formation des noms en indo-européen [Origins of the Formation of Nouns in Indo-European]. Paris: Adrien-Maisonneuve. 1948 Noms d’agent et noms d’action en indo-européen [Agent Nouns and Action Nouns in Indo-European]. Paris: Adrien-Maison-neuve. 1952 “Lycien” [Lycian], in: Antoine Meillet and Marcel Cohen (eds.), Les langues du monde [The Languages of the World]. Paris: Honoré Champion, vol. I, pp. 205—208. 1954 “Études hittites et indo-européennes” [Hittite and Indo-European Studies], Bulletin de la Société Linguistique 50:29—43. 1962 Hittite et indo-européen [Hittite and Indo-European]. Paris: AdrienMaisonneuve. References Blažek, Václav 2011 Bloch, Jules 1965 Boisacq, Émile 1950 129 “Indo-European Laryngeals in Afroasiatic Perspective”, Journal of Language Relationship 5:1—22. Indo-Aryan from the Vedas to Modern Times. English edition largely revised by the author and translated by Alfred Master. Paris: AdrienMaisonneuve. Dictionnaire étymologique de la langue grecque: étudiée dans ses rapports avec les autres langues indo-européennes [Etymological Dictionary of the Greek Language: Studied in Its Relationships with the Other Indo-European Languages]. 4th edition, with an index by Helmut Rix (1st edition 1916). Heidelberg: Carl Winter. Bomhard, Allan R. 1973 “Some Anatolian Etymologies”, Revue Hittite et Asianique XXXI: 111—113. 1976 “The Placing of the Anatolian Languages”, Orbis XXV.2:199—239. 1981 “A New Look at Indo-European”, Journal of Indo-European Studies 9.3/4:332—337. 1986 “The Aspirated Stops of Proto-Indo-European”, Diachronica III.1:67— 80. 2000 “Sturtevant’s Law in Hittite: A Reassessment”, in: Yoël L. Arbeitman (ed.), Studies on the Pre-Greek Languages in Memory of Charles Carter. Leuven and Paris: Peeters, pp. 35—46. 2004 “The Proto-Indo-European Laryngeals”, in: Adam Hyllested, Anders Richardt Jørgensen, Jenny Helena Larsson, and Thomas Olander (eds.), Per Aspera ad Asteriscos. Studia Indogermanica in honorem Jens Elmegård Rasmussen sexagenarii Idibus Martiis anno MMIV [Through Hardship to the Stars: Indo-European Studies in Honor of Jens Elmegård Rasmussen on His Sixtieth Birthday, the Ides of March 2004]. Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft, pp. 69—79. 2016 “The Glottalic Model of Proto-Indo-European Consonantism: Reigniting the Dialog”, Slovo a slovesnost 77:371—391. 2019a “The Origins of Proto-Indo-European: The Caucasian Substrate Hypothesis”. Journal of Indo-European Studies 47.1/2:9—124. Note: This paper was substantially expanded and reissued in book format under a Creative Commons license in 2023 under the title Prehistoric Language Contact on the Steppes: The Case of Indo-European and Northwest Caucasian. 2019b “The Importance of Hittite and the Other Anatolian Daughter Languages for the Reconstruction of the Proto-Indo-European Phonological System”, Wékwos 5. 2023 A Comprehensive Introduction to Nostratic Comparative Linguistics: With Special Reference to Indo-European. 5th edition, 5 vols., 3,239 pp. 130 References (Revised 24 February 2024 — this work is regularly updated.) OpenAccess publication. 2024 “The Importance of Armenian for Understanding the Development of the Proto-Indo-European Phonological System in Old Indic, Greek, and Italic”. Originally prepared in 2019, revised 4 January 2024. OpenAccess Publication. Borgstrøm, Carl Hj. 1949 “Thoughts about Indo-European Vowel Gradation”, Nordsk Tidsskrift for Sprogvidenskap 15:137—187. 1952 “Additional Remarks about Vowel Gradation”, Nordsk Tidsskrift for Sprogvidenskap 16:136—147. 1954 “Internal Reconstruction of Pre-Indo-European Word-Forms”, Word X:275—287. Bousquette, Joshua, and Joseph Salmons 2017 “Germanic”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 387—420. Boutkan, Dirk, and Sjoerd Michiel Siebinga 2005 Old Frisian Etymological Dictionary. Leiden and Boston, MA: E. J. Brill. Brixhe, Claude 1990 “Comparaison et langues faiblement documentées: l’exemple du phrygien et ses voyelles longues” [Comparison and Poorly Documented Languages: The Example of Phrygian and Its Long Vowels], in: Jean Kellens (ed.), La reconstruction des laryngales [The Reconstruction of the Laryngeals]. Paris: Société d’Édition “Les Belles Lettres”, pp. 59— 99. 1994 “Le phrygien” [Phrygian], in: Françoise Bader (ed.), Langues indoeuropéennes [Indo-European Languages]. Paris: CNRS Éditions, pp. 165—178. 2004 “Phrygian”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 777—788. 2018a “Thracian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical IndoEuropean Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1850—1854. 2018b “Macedonian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical IndoEuropean Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1862—1867. Brixhe, Claude, and Anna Panayotou 1994a “Le thrace” [Thracian], in: Françoise Bader (ed.), Langues indoeuropéennes [Indo-European Languages]. Paris: CNRS Éditions, pp. 179—203. References 1994b Brosch, Cyril No date Brugmann, Karl 1904 131 “Le macédonien” [Macedonian], in: Françoise Bader (ed.), Langues indo-européennes [Indo-European Languages]. Paris: CNRS Éditions, pp. 205—220. “Zur Vertretung der Laryngale im Karischen’ [On the Treatment of the Laryngeals in Carian]. Published on-line. Kurze vergleichende Grammatik der indogermanischen Sprachen. Auf Grund des fünfbändigen “Grundriß der vergleichenden Grammatik der indogermanischen Sprachen” von K. Brugmann und B. Delbrück [Concise Comparative Grammar of the Indo-European Languages. Based upon the Five-Volume “Elements of the Comparative Grammar of the Indo-European Languages” by K. Brugmann and B. Delbrück]. Berlin: Walter de Gruyter. Reprinted 1970. 1905 Abrégé de grammaire comparée des langues indo-européennes, d’après le précis de grammaire comparée de K. Brugmann et B. Delbrück [Concise Comparative Grammar of the Indo-European Languages, based upon the “Elements of the Comparative Grammar of the IndoEuropean Languages” by K. Brugmann and B. Delbrück]. Translated by J. Bloch, A. Cuny, and A. Ernout under the direction of A. Meillet and R. Gauthiot. Paris: Librairie C. Klincksieck. Brugmann, Karl, and Berthold Delbrück 1897—1916 Grundriß der vergleichenden Grammatik der indogermanischen Sprachen [Elements of the Comparative Grammar of the IndoEuropean Languages]. 2nd edition. Berlin: Walter de Gruyter. Reprinted 1967. Buck, Carl D[arling] 1928 A Grammar of Oscan and Umbrian. 2nd edition reprinted 1974. Hildesheim: Georg Olms. 1933 Comparative Grammar of Greek and Latin. 10th impression 1966. Chicago, IL, and London: University of Chicago Press. 1949 A Dictionary of Selected Synonyms in the Principal Indo-European Languages. Chicago, IL: University of Chicago Press. [1988] [Paperback edition.] Burrow, Thomas 1973 The Sanskrit Language. 3rd edition. London: Faber & Faber. 1979 The Problem of Shwa in Sanskrit. Oxford: Oxford University Press. Byrd, Andrew 2010 Reconstructing Indo-European Syllabification. Ph.D. dissertation, University of California, Los Angeles, CA. 2015 The Indo-European Syllable. Leiden and Boston, MA: E. J. Brill. 2018 “The Phonology of Proto-Indo-European”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 2056—2079. 132 References Calin, Didier 2019 Camaj, Martin 1984 A Short English Lycian/Milyan Lexicon. Riga. Albanian Grammar with Exercises, Chrestomathy and Glossaries. Collaborated on and translated by Leonard Fox. Wiesbaden: Otto Harrassowitz. Campbell, Lyle, and Andrew Garrett 1993 Review of Jorma Koivulehto (1991), Uralische Evidenz für die Laryngaltheorie [Uralic Evidence for the Laryngeal Theory], Language 69:832—836. Cantineau, Jean 1960 Cours de phonétique arabe [Course in Arabic Phonetics]. Reprint. Paris: Librairie C. Klincksieck. Carlton, Terence R. 1991 Introduction to the Phonological History of the Slavic Languages. Columbus, OH: Slavica Publishers, Inc. Carruba, Onofrio 1970 Das Paläische: Texte, Grammatik, Lexikon [Palaic: Texts, Grammar, Lexicon]. Wiesbaden: Otto Harrassowitz. 2009 “Indo-European Vowel Alternations (Ablaut/Apophony)”, in Angela Marcantonio (ed.), The Indo-European Language Family: Questions about Its Status. Washington, DC: Institute for the Study of Man, pp. 5:1—42. Carruba, Onofrio (ed.) 1992 Per una grammatica ittita / Towards a Hittite Grammar. Pavia: Gianni Iuculano Editore. Carter, Charles 1988 “Indo-Hittite Again”, in: Yoël L. Arbeitman, A Linguistic Happening in Memory of Ben Schwartz. Louvain-la-Neuve: Peeters, pp. 157—161. Catford, John C. 1977 Fundamental Problems in Phonetics. Bloomington, IN: Indiana University Press. Chantraine, Pierre 1968—1980 Dictionnaire étymologique de la langue grecque. Histoire des mots. [Etymological Dictionary of the Greek Language. History of Words]. 2 vols. Paris: Librairie C. Klincksieck. Chirikba, Vjacheslav A. 1996a Common West Caucasian: The Reconstruction of Its Phonological System and Parts of Its Lexicon and Morphology. Leiden: Research School CNWS. 1996b A Dictionary of Common Abkhaz. Leiden. The Author. 1999 “The West Caucasian Material in ‘The North Caucasian Etymological Dictionary’ by S. A. Nikolayev and S. A. Starostin”. Leiden: CNWS Publications. 2003 Abkhaz. Munich: LINCOM Europa. References 133 “The Problem of the Caucasian Sprachbund”, in: Pieter Muysken (ed.), From Linguistic Areas to Areal Linguistics. Amsterdam and Philadelphia, PA: John Benjamins, pp. 25—93. 2016 “From North to North West: How North-West Caucasian Evolved from North Caucasian”, Mother Tongue XXI:1—28. Chomsky, Noam, and Morris Halle 1968 The Sound Pattern of English. New York, NY: Harper & Row. Christol, Alain 1990 “Les laryngales entre phonétique et phonologie” [The Laryngeals between Phonetics and Phonology], in: Jean Kellens (ed.), La reconstruction des laryngales [The Reconstruction of the Laryngeals]. Paris: Société d’Édition “Les Belles Lettres”, pp. 101—127. Clackson, James P. T. 2000 “Time Depth in Indo-European”, in: Colin Renfrew, April McMahon, and R. Larry Trask (eds.), Time Depth in Historical Linguistics. Cambridge: McDonald Institute for Archaeological Research, pp. 441—454. 2007 Indo-European Linguistics: An Introduction. Cambridge: Cambridge University Press. 2013 “The Origins of the Indic Languages: The Indo-European Model”, in: Angela Marcantonio and Girish Nath Jha (eds.), Perspectives on the Origin of Indian Civilization. New Delhi: D. K. Printworld, pp. 259— 287. Cohen, David 1968 “Langues chamito-sémitiques” [Hamito-Semitic Languages], in: André Martinet (ed.), Le langage [Language]. Paris: Éditions Gallimard, pp. 1288—1330. Cohen, David (ed.) 1988 Les langues dans le monde ancien et moderne: langues chamitosémitiques [Languages in the Ancient and Modern World: HamitoSemitic Languages]. Paris: Centre National de la Recherche Scientifique. Colarusso, John 1975 The Northwest Caucasian Languages: A Phonological Survey. Ph.D. dissertation, Harvard University. 1978 “The Typology of Pharyngeals and Pharyngealization: Caucasian Examples”. Paper presented at the Sixth Meeting (April 9—10, 1978) of the Toronto North American Conference on Afro-Asiatic Linguistics. 1981 “Typological Parallels between Proto-Indo-European and the Northwest Caucasian Languages”, in: Yoël L. Arbeitman and Allan R. Bomhard (eds.), Bono Homini Donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns. Amsterdam: John Benjamins, part I, pp. 475—557. 1989 “Proto-Northwest Caucasian (or How to Crack a Very Hard Nut)”, in: Howard Aronson (ed.), The Non-Slavic Languages of the USSR, 2008 134 References 1992 2014 Linguistic Studies. Chicago, IL: Chicago Linguistic Society, pp. 20— 55. “Phyletic Links between Proto-Indo-European and Proto-Northwest Caucasian”, in: Howard I. Aronson (ed.), The Non-Slavic Languages of the USSR: Linguistic Studies (Second Series). Chicago, IL: Chicago Linguistic Society, pp. 19—54. Reprinted in Mother Tongue 21:8—20 (1994). The Northwest Caucasian Languages: A Phonological Survey. London and New York, NY: Routledge. (Same as Colarusso 1975 but with emendations and additional references.) Collinder, Björn 1967 “Die indouralische Sprachvergleichung und die Laryngaltheorie” [The Indo-Uralic Linguistic Comparison and the Laryngeal Theory], Die Sprache 13:179—190. 1970 “Nachtrag zum Aufsatz ‘Die indouralische Sprachvergleichung und die Laryngaltheorie’” [Addenda to the Article “The Indo-Uralic Linguistic Comparison and the Laryngeal Theory”], Die Sprache 16:174—175. Collinge, N[eville] E[dgar] 1970 “The Indo-European Laryngeal”, in: N. E. Collinge, Collectanea Linguistica: Essays in General and Genetic Linguistics. The Hague: Mouton, pp. 67—101. 1985 The Laws of Indo-European. Amsterdam: John Benjamins. Collins, Daniel 2018 “The Phonology of Slavic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1313—1538. Connolly, Leo A. 1989 “Laryngeal Metathesis: An Aryan Peculiarity?”, in: Theo Vennemann (ed.), The New Sound of Indo-European: Essays in Phonological Reconstruction. Berlin and New York, NY: Mouton de Gruyter, pp. 43—51. Čop, Bojan 1970 “Die indouralische Sprachverwandtschaft und die indogermanische Laryngaltheorie” [The Indo-Uralic Linguistic Comparison and the IndoEuropean Laryngeal Theory], Slovenska Akademija Znanosti in Umetnosti VII.5:185—229. Cotticelli Kurras, Paola 2009 “La ricostruzione della protolingua indoeuropea alla luce dei datti anatolici” [The Reconstruction of the Indo-European Proto-Language in Light of the Anatolian Data], Incontri Linguistici 32:117—162. Cotticelli-Kurras, Paola, and Federico Giusfredi 2018 “Ancient Anatolian Languages and Cultures in Contact: Some Methodological Observations”, Journal of Language Relationship 16.3:172— 193. References 135 Couvreur, Walter 1935 De Hittitische ḫ [Hittite ḫ]. Leuven: Katholieke Universiteit te Leuven. 1937 De Hettitische Ḫ: Een Bijdrage tot de Studie van het Indo-Europeesche Vocalisme [Hittite Ḫ: A Contribution to the Study of Indo-European Vocalism]. Louvain: Bibliothèque du Muséon. Cowgill, Warren 1965 “Greek Evidence”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 142—180. 1979 “Anatolian ḫi-Conjugation and Indo-European Perfect: Installment II”, in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and Indo-European: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the IndoEuropean Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 25—39. Cowgill, Warren, and Manfred Mayrhofer 1986 Indogermanische Grammatik I: 1. Einleitung. 2. Lautlehre [IndoEuropean Grammar I: 1. Introduction. 2. Phonology]. Heidelberg: Carl Winter. Crothers, John 1978 “Typology and Universals of Vowel Systems”, in: Joseph H. Greenberg (ed.), Universals of Human Language. Volume 2: Phonology. Stanford, CA: Stanford University Press, pp. 93—152. Cuny, Albert 1927 “Réflexions sur le type χρῆ (ζῆν) et le type ἠχώ”, in Symbolae Grammaticae in honorem Joannis Rozwadowski 1. Cracow: Gebethner & Wolff, pp. 85—94. Darden, Bill J. 2002 “On the Question of the Archaism of the Hittite Verb”, in: Fabrice Cavoto (ed.), The Linguist’s Linguist: A Collection of Papers in Honour of Alexis Manaster Ramer. Munich: LINCOM Europa, pp. 127—134. De Saussure, Ferdinand 1879 Mémoire sur le système primitif des voyelles dans les langues indoeuropéennes [An Essay on the Primitive System of Vowels in the IndoEuropean Languages]. Reprinted 1978. Hildesheim: Georg Olms. (Reprinted 2009 by Cambridge University Press.) De Vaan, Michiel 2008 Etymological Dictionary of Latin and the Other Italic Languages. Leiden and Boston, MA: E. J. Brill. 2018 “The Phonology of Albanian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1732—1749. 136 References De Vries, Jan 1962 Altnordisches etymologisches Wörterbuch [Old Norse Etymological Dictionary]. 2nd edition. Leiden: E. J. Brill. [1977] [Reprinted.] Dempsey, Timothy R. 2015 Verbal Reduplication in Anatolian. Ph.D. dissertation, UCLA. Derksen, Rick 2008 Etymological Dictionary of the Slavic Inherited Lexicon. Leiden and Boston, MA: E. J. Brill. 2015 Etymological Dictionary of the Baltic Inherited Lexicon. Leiden and Boston, MA: E. J. Brill. Diakonoff, Igor M., and V[ladimir] P. Neroznak 1985 Phrygian. Delmar, NY: Caravan Books. Dieu, Éric 2020 Review of Sylvain Patri (2019), Phonologie hittite, Bulletin de la Société de Linguistique de Paris 115.2:63—69. Dunkel, George E. 1988 “Laryngeals and Particles: *høu, *u, and *awo”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 107—121. Eichner, Heiner 1987 “Die Entdeckung des lydischen Akzents” [The Discovery of the Lydian Accent], Bibliotheca Orientalis XLIV.2:80—88. 1988 “Anatolisch und Trilaryngalismus” [Anatolian and Three Laryngeals], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 123—151. Emonds, Joseph Embley 1972 “A Reformulation of Grimm’s Law”, in: Michael K. Brame (ed.), Contributions to Generative Phonology. Austin, TX: University of Texas Press, pp. 108—122. Ernout, Alfred, and Antoine Meillet 1979 Dictionnaire étymologique de la langue latine: Histoire des mots [Etymological Dictionary of the Latin Language: History of Words]. 4th edition. Paris: Librairie C. Klincksieck. Falileyev, Alexander 2000 Etymological Glossary of Old Welsh. Tübingen: Max Niemeyer. Fallon, Paul D. 2002 The Synchronic and Diachronic Phonology of Ejectives. New York, NY: Routledge. References Feist, Sigmund 1939 137 Vergleichendes Wörterbuch der gotischen Sprache [Comparative Dictionary of the Gothic Language]. 3rd edition. Leiden: E. J. Brill. Finkelberg, Margalit 1997 “Anatolian Languages and Indo-European Migrations to Greece”, The Classical World 91.1:3—20. Fleming, Harold C. 2000 “Glottalization in Eastern Armenian”, Journal of Indo-European Studies 28.1/2:155—196. Fordyce, James F. 1980 “On the Nature of Glottalic and Laryngealized Consonant and Vowel Systems”, in: UCLA Working Papers in Phonetics, no. 50. Los Angeles, CA: UCLA Department of Linguistics, pp. 122—154. Fortson, Benjamin W., IV 2010 Indo-European Language and Culture: An Introduction. 2nd edition. Chichester: Wiley-Blackwell. Fraenkel, Ernst 1962—1965 Litauisches etymologisches Wörterbuch [Lithuanian Etymological Dictionary]. 2 vols. Heidelberg: Carl Winter. Friedrich, Johannes 1960 Hethitisches Elementarbuch [Hittite Primer]. Vol. I. 2nd edition. Heidelberg: Carl Winter. 1991 Kurzgefaßtes Hethitisches Wörterbuch [A Concise Hittite Dictionary]. Reprint of the 1952 edition along with the supplements. Heidelberg: Carl Winter. Friedrich, Johannes, Erica Reiner, Annelies Kammenhuber, Günter Neumann, and Alfred Heubeck (contributors) 1969 Altkleinasiatische Sprachen [Old Near Eastern Languages]. Leiden and Köln: E. J. Brill. Frisk, Hjalmar 1970—1973 Griechisches etymologisches Wörterbuch [Greek Etymological Dictionary]. 3 vols. Heidelberg: Carl Winter. Frotscher, Michael 2012 “Die luwischen Entsprechungen der hethitischen Verben des Typs dāi/ tii̯ anzi und ein neues Lautgesetz urindogermanisch *oi̯ > urluwisch *u̯e” [The Luwian Correspondence of the Hittite Verb of the Type dāi/tii̯ anzi and a New Proto-Indo-European Sound Law *oi̯ > Proto-Luwian *u̯e], International Journal of Diachronic Linguistics and Linguistic Reconstruction 9:137—194. Fulk, R[obert] D[ennis] 2018 A Comparative Grammar of the Early Germanic Languages. Amsterdam and Philadelphia, PA: John Benjamins. Gamkrelidze, Thomas V[alerʹanovič] 1966 “A Typology of Common Kartvelian”, Language 42:69—83. 138 References 1967 “Kartvelian and Indo-European: A Typological Comparison of Reconstructed Linguistic Systems”, in: To Honor Roman Jakobson. The Hague: Mouton, vol. I, pp. 707—717. 1968 “Hittite and the Laryngeal Theory”, in: J. C. Heesterman, G. H. Schokker, and V. I. Subramoniam (eds.), Pratidānam: Indian, Iranian and Indo-European Studies Presented to Franciscus Bernardus Kuipers on his Sixtieth Birthday. The Hague: Mouton, pp. 89—97. 1970 “‘Anatolian Languages’ and the Problem of Indo-European Migration to Asia Minor”, in: Roman Jakobson and Shigeo Kawamoto (eds.), Studies in General and Oriental Linguistics Presented to Shirô Hattori on the Occasion of His Sixtieth Birthday. Tokyo: TEC Company, Ltd., pp. 138—143. 1976 “Linguistic Typology and Indo-European Reconstruction”, in: Alphonse Juilland (ed.), Linguistic Studies Offered to Joseph Greenberg. Saratoga, CA: Anma Libri, vol. 2, pp. 399—406. 1978 “On the Correlation of Stops and Fricatives in a Phonological System”, in: Joseph H. Greenberg (ed.), Universals of Human Language, vol. 2, Phonology. Stanford, CA: Stanford University Press, pp. 9—46. 1981 “Language Typology and Language Universals and Their Implications for the Reconstruction of the Indo-European Stop System”, in: Yoël L. Arbeitman and Allan R. Bomhard (eds.), Bono Homini Donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns. Amsterdam: John Benjamins, part II, pp. 571—609. 1982 “Problems of Consonantism of the Cuneiform Hittite Language”, in: J. N. Postgate (ed.), Societies and Languages of the Ancient Near East: Studies in Honour of I. M. Diakonoff. Warminster: Aris and Phillips, pp. 76—80. 2001a “André Martinet et la ‘théorie glottalique indo-européenne’” [André Martinet and the ‘Indo-European Glottalic Theory’], La Linguistique 37.1:81—88. 2001b “Anatolian in Light of the Glottalic Theory”, in: Otto Carruba and Wolfgang Meid (eds.), Anatolisch und Indogermanisch / Anatolico e Indoeuropeo. Akten des Kolloquiums der Indogermanischen Gesellschaft, Pavia, 22—25. September 1998 [Anatolian and Indo-European. Acts of the Colloquium of the Indo-European Society, Pavia, 22—25 September 1998]. Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft, pp. 119—127. 2008 “The Problem of the Origin of the Hittite Cuneiform”, Bulletin of the Georgian National Academy of Sciences 2.3:169—174. Gamkrelidze, Thomas V., and Vjačeslav V. Ivanov 1995 Indo-European and the Indo-Europeans: A Reconstruction and Historical Typological Analysis of a Protolanguage and a ProtoCulture. 2 vols. English translation by Johanna Nichols. Berlin, New York, NY, and Amsterdam: Mouton de Gruyter. References 139 Garnier, Romain 2018 “Report of Mark Weeden and Lee Z. Ullmann (eds.), Hittite Landscape and Geography, with maps by Zenobia Homan. (= Handbook of Oriental Studies, vol. 121.) Leiden and Boston, MA: E. J. Brill”, Journal Asiatique 306/2:338—341. Garrett, Andrew 1998 “Adjarian’s Law, the Glottalic Theory, and the Position of Armenian”, Proceedings of the Annual Meeting of the Berkeley Linguistics Society 24.2:12—23. Gelb, Ignace J. 1931 Hittite Hieroglyphs I. Chicago, IL: University of Chicago Press. Genz, Hermann, and Dirk Paul Mielke (eds.) 2011 Insights into Hittite History and Archaeology. Leuven, Paris, Walpole, MA: Peeters. Georgiev, Vladimir 1964 “On the Present State of Indo-European Linguistics”, in: Horace G. Lunt (ed.), Proceedings of the Ninth International Congress of Linguists. The Hague: Mouton, pp. 738—742. 1966 Introduzione alla storia delle lingue indeuropee [Introduction to the History of the Indo-European Languages]. 2nd edition. Rome: Edizioni dell’Ateneo. 1981 Introduction to the History of the Indo-European Languages. 3rd edition. Sofia: Publishing House of the Bulgarian Academy of Sciences. 1984 “Die drei Hauptperioden des Indoeuropäischen” [The Three Main Periods of Indo-European], Diachronica 1.1:65—78. Ghatage, A[mrit] M[adhav] 1962 Historical Linguistics and Indo-Aryan Languages. Bombay: University of Bombay Press. Giusfredi, Federico, Alvise Matessi, and Valerio Pisaniello 2023 Contacts of Languages and Peoples in the Hittite and Post-Hittite World. Vol. 1: The Bronze Age and Hatti. With contributions by Paola Cotticelli-Kurras, Alfredo Rizza, Maurizio Viano, and Ilya Yakubovich. Leiden and Boston, MA: E. J. Brill. Godel, Robert 1975 An Introduction to the Study of Classical Armenian. Wiesbaden: Dr. Ludwig Reichert Verlag. Goedegebuure, Petra 2013 “Split Ergativity in Hittite”, Zeitschrift für Assyriologie und vorderasiatische Archäologie 102.2:270—303. Goldfeder, Sean B[enjamin] 2016 A Reconstruction of Pre-Proto-Indo-Hittite. Manuscript. Downloaded from academia.edu. Gray, Louis H. 1932 “On Indo-European Noun-Declension: Especially of -o- and -ā-Stems”, Language 8.3:183—199. 140 References Greenberg, Joseph H. 1966 Language Universals. The Hague: Mouton. 1969 “Some Methods of Dynamic Comparison in Linguistics”, in: Jaan Puhvel (ed.), Substance and Structure of Language. Berkeley and Los Angeles, CA: University of California Press, pp. 147—203. 1970 “Some Generalizations Concerning Glottalic Consonants”, International Journal of American Linguistics 36:123—145. 1990 “The Prehistory of the Indo-European Vowel System in Comparative and Typological Perspective”, in: Vitaly Shevoroshkin (ed.), ProtoLanguages and Proto-Cultures. Bochum: Brockmeyer, pp. 77—136. Greenberg, Marc L. 2017 “Slavic”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 519—551. Greppin, John A. C. 1988 “Laryngeal Residue in Armenian: Genetic and Loan Evidence”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 179—193. Grestenberger, Laura 2019 “On Hittite iškallāri and the PIE ‘Stative’,” in: Adam Alvah Catt, Ronald I. Kim, and Brent Vine (eds.), Anatolian and Indo-European Studies in Honor of Kazuhiko Yoshida. Ann Arbor, MI, and New York, NY: Beech Stave Press, pp. 91—105. Gurney, O[liver] R[obert] 1990 The Hittites. 2nd edition, reprinted with revisions. London: Penguin Books. Gusmani, Roberto 1964 Lydisches Wörterbuch [Lydian Dictionary]. Heidelberg: Carl Winter. [1980—1982] [2nd edition.] 1979 “Ittito, teoria laringalistica e recostruzione” [Hittite, Laryngeal Theory, and Reconstruction], in Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and IndoEuropean: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the Indo-European Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 63—71. Güterbock, Hans, and Harry A. Hoffner, Jr. (eds.) 1980— The Hittite Dictionary of the University of Chicago. Chicago, IL: The Oriental Institute of the University of Chicago. Hackstein, Olav 2017 “The Phonology of Tocharian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and References 141 Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. II, pp. 1304—1335. Hajnal, Ivo I. 1995 Der lykische Vokalismus [The Lycian Vowels]. Zürich: Universität Zürich. 2001 “Lydian: Late Hittite or Neo-Luwian?” Manuscript. Hajnal, Ivo, and Katharina Zipser 2016 “Phonological Movement in Luwian: The Case of the Hieroglyphic Luwian Verbal Endings in -si”, in: Stephanie W. Jamison, H. Craig Melchert, and Brent Vine (eds), Proceedings of the 26th Annual UCLA Indo-European Conference, October 24—25, 2014. Bremen: Hempen Verlag, pp. 47—62. Hammerich, Louis 1948 Laryngeal before Sonant. Copenhagen: Munksgaard. Hamp, Eric P. 1965a “Evidence in Albanian”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 123—141. 1965b “Evidence in Keltic”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 224—235. Harðarson, Jón Axel 2015 “The Proto-Indo-European Collective and the Hittite Neuter Plural”. Paper delivered at the conference “Hrozný and Hittite: The First Hundred Years”, Charles University, Prague, 11—14 November 2015. Hart, Gillian 1980 “The Ablaut Present and Preterites in Hittite Radical Verbs”, in: Anatolian Studies, Vol. XXX. Journal of the British Institute of Archaeology at Ankara. Special Number in Honour of the Seventieth Birthday of Professor O. R. Gurney, 28 January 1981. London: The British Institute of Archaeology at Ankara, pp. 51—61. 1983 “Problems of Writing and Phonology in Cuneiform Hittite”, Transactions and Proceedings of the American Philological Association 81.1:100—154. Hasselbach-Andee, Rebecca 2020 “Akkadian”, in: Rebecca Hasselbach-Andee (ed.), A Companion to Ancient Near Eastern Languages. Hoboken, NJ: John Wiley & Sons, Inc., pp. 129—147. Hasselbach-Andee, Rebecca (ed.) 2020 A Companion to Ancient Near Eastern Languages. Hoboken, NJ: John Wiley & Sons, Inc. Hawkins, David, Anna Murpurgo Davies, and Günter Neumann 1974 Hittite Hieroglyphs and Luwian: New Evidence for the Connection. Göttingen: Vandenhoeck and Ruprecht. Hejná, Míša, and George Walkden 2022 A History of English. Berlin: Language Science Press. 142 References Held, Warren H., Jr., and William R. Schmalstieg 1969 “Some Comments on the Hittite Phonemic System”, General Linguistics 9:93—100. Held, Warren H., Jr., William R. Schmalstieg, and Janet E. Gertz 1988 Beginning Hittite. Columbus, OH: Slavica Publishers, Inc. Hirt, Hermann Alfred 1895 Die indogermanische Akzent. Ein Handbuch [The Indo-European Accent: A Manual]. Straßburg: Karl J. Trübner. 1900 Die indogermanische Ablaut, vornehmlich in seinem Verhältnis zur Betonung [The Indo-European Ablaut, Especially in its Relationship to Tone]. Straßburg: Karl J. Trübner. 1921—1927 Indogermanische Grammatik [Indo-European Grammar]. 7 vols. Heidelberg: Carl Winter. (Reprinted 2009 by Cambridge University Press.) Hockett, Charles F. 1955 A Manual of Phonology. Chicago, IL: University of Chicago Press. Hoenigswald, Henry M. 1965 “Indo-Iranian Evidence”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 93—99. 1988 “A Note on Semivowel Behavior and its Implications for the Laryngeals”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 59—105. Hoffner, Harry A., Jr. 1967 An English-Hittite Glossary. (= Revue Hittite et Asianique XXV.) Paris: Librairie C. Klincksieck. Hoffner, Harry A., Jr., and Gary Beckman (eds.) 1986 Kaniššuwar: A Tribute to Hans G. Güterbock on His Seventy-fifth Birthday, May 27, 1983. Chicago, IL: The Oriental Institute of the University of Chicago. Hoffner, Harry A., Jr., and H. Craig Melchert 2008 A Grammar of the Hittite Language. 2 volumes. Winona Lake, IN: Eisenbrauns. Hofmann, J[ohann] B[aptist] 1966 Etymologisches Wörterbuch des Griechischen [Etymological Dictionary of Greek]. Munich: R. Oldenbourg Verlag. Hopper, Paul J. 1973 “Glottalized and Murmured Occlusives in Indo-European”, Glossa 7:141—166. 1977a “The Typology of the Proto-Indo-European Segmental Inventory”, Journal of Indo-European Studies 5.1:41—53. 1977b “Indo-European Consonantism and the ‘New Look’,” Orbis XXVI.1:57—72. References 1981 1982 1990 Hovers, Onno 2021 143 “‘Decem’ and ‘Taihun’ Languages: An Indo-European Isogloss”, in: Yoël L. Arbeitman and Allan R. Bomhard (eds.), Bono Homini Donum: Essays in Historical Linguistics in Memory of J. Alexander Kerns. Amsterdam: John Benjamins, part I, pp. 133—142. “Areal Typology and the Early Indo-European Consonant System”, in: Edgar C. Polomé (ed.), The Indo-Europeans in the Fourth and Third Millennia. Ann Arbor, MI: Karoma Publishers, pp. 121—139. “A Glottalic Interpretation of the Germanic EXPRESSIVE GEMINATION”, in: Jerold A. Edmondson, Crawford Feagin, and Peter Mühlhäusler (eds.), Development and Diversity: Language Variation across Time and Space. A Festschrift for Charles-James N. Bailey. Arlington, TX: The Summer Institute of Linguistics and the University of Texas at Arlington, pp. 85—93. “The Two Short o’s of Indo-European”. Manuscript (draft version July 2021). Hübschmann, Heinrich 1883 Armenische Studien [Armenian Studies]. Leipzig: Druck und Verlag Von Breitkopf & Härtel. 1885 Die indogermanische Vocalsystem [The Indo-European Vowel System]. Reprinted 1975. Amsterdam: Oriental Press. Huld, Martin E. 1984 Basic Albanian Etymologies. Columbus, OH: Slavica Publishers, Inc. Hyllested, Adam 2009 “Internal Reconstruction vs. External Comparison: The Case of the Indo-Uralic Laryngeals”, in: Jens Elmegård Rasmussen and Thomas Olander (eds.), Internal Reconstruction in Indo-European: Methods, Results, and Problems. Section Papers from the XVI International Conference on Historical Linguistics, University of Copenhagen, 11— 15 August 2003. Copenhagen: Museum Tusculanum Press, pp. 111— 136. Hyman, Larry 1975 Phonology: Theory and Analysis. New York, NY: Holt, Rinehart and Winston Jakobson, Roman 1971(1956) “Mufaxxama: The ‘Emphatic’ Phonemes in Arabic”, in: Roman Jakobson, Selected Writings. 2nd edition. The Hague: Mouton, vol. I, pp. 510—522. (Written in Cambridge, MA, in 1956 and published in: Studies Presented to Joshua Whatmough [The Hague: Mouton, 1957].) 1971(1957) “Typological Studies and their Contribution to Historical Comparative Linguistics”, in: Roman Jakobson, Selected Writings. 2nd edition. The Hague: Mouton, vol. I, pp. 523—532. (Originally presented as a report at the first plenary session of the Eighth International Congress of Linguists, Oslo, August 1957.) 144 References 1971(1960) “Why ‘Mama’ and ‘Papa’,” in: Roman Jakobson, Selected Writings. 2nd edition. The Hague: Mouton, vol. I, pp. 538—545. (Written in Stanford, CA, 1959 for: Perspectives in Psychological Theory, Dedicated to Heinz Werner [New York, NY, 1960].) Jamison, Stephanie W. 1988 “The Quantity and Outcome of Vocalized Laryngeals in Indic”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 213—226. Jäntti, Selim Axel Oscar 2017 Geminate Stops in Anatolian: Evidence and Typological Implications. Master Thesis, Leiden University. No date “The Position of Anatolian in Indo-European: An Overview”. Manuscript. Downloaded from academia.edu. Jasanoff, Jay 1979 “The Position of the ḫi-Conjugation”, in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and IndoEuropean: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the Indo-European Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 79—90. 2003 Hittite and the Indo-European Verb. Oxford: Oxford University Press. 2017 “The Impact of Hittite and Tocharian: Rethinking Indo-European in the 20th Century and Beyond”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. I, pp. 220—238. Jonsson, Hans 1978 The Laryngeal Theory: A Critical Survey. Lund: Gleerup. Joseph, Brian, and Rex E. Wallace 1994 “Proto-Indo-European Voiced Aspirates in Italic: A Test for the ‘Glottalic Theory’,” Historische Sprachforschung / Historical Linguistics 107:244—261. Josephson, Folke 1979 “Assibilation in Anatolian”, in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and IndoEuropean: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the Indo-European Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 92—103. References 2011 145 “Allative in Indo-European”, in: Michèle Fruyt, Michel Mazoyer, and Dennis Pardee (eds.), Grammatical Case in the Languages of the Middle East and Europe. Acts of the International Colloquium “Variations, concurrence et evolution des cas dans divers domaines linguistiques”, Paris, 2—4 April 2007. Chicago, IL: The Oriental Institute of the University of Chicago, pp. 143—150. Kaiser, Mark, and Vitalij Shevoroshkin 1985 “On Indo-European Laryngeals and Vowels”, Journal of IndoEuropean Studies 13.3/4:377—413. Kassian, Alexei S. 2002 “Indo-European Accentology and Hittite Data. Nomina”, in: Vitaly Shevoroshkin and Paul Sidwell (eds.), Anatolian Languages. (AHL Studies in the Science and History of Language 6.) Canberra: Association for the History of Language, pp. 49—60. Katičić, Radoslav 1976 Ancient Languages of the Balkans. 2 vols. The Hague: Mouton. Keiler, Allan R. 1970 A Phonological Study of the Indo-European Laryngeals. The Hague: Mouton. Kellens, Jean (ed.) 1990 La reconstruction des laryngales [The Reconstruction of the Laryngeals]. Paris: Société d’Édition “Les Belles Lettres”. Kent, Roland G. 1945 The Sounds of Latin. Baltimore: Linguistic Society of America. 1953 Old Persian. 2nd edition. New Haven, CT: American Oriental Society. Kerns, J. Alexander, and Benjamin Schwartz 1940 “The Laryngeal Hypothesis and Indo-Hittite, Indo-European Vocalism”, Journal of the American Oriental Society 60:181—192. 1963a “Initial Laryngeals in Tocharian”, Journal of the American Oriental Society 83.2:361—362. Kessler, Brett No Date “On the Phonological Nature of the Proto-Indo-European Laryngeals” (download: http://spell.psychology.wustl.edu/PIE-laryngeals/). Kim, Ronald I. 2018 “The Phonology of Balto-Slavic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1974—1985. Kim, Ronald, Norbert Oettinger, Elisabeth Rieken, and Michael Weiss (eds.) 2010 Ex Anatolia Lux: Anatolian and Indo-European Studies in Honor of H. Craig Melchert on the Occasion of His Sixty-fifth Birthday. Ann Arbor, MI: Beech Stave Press. Kimball, Sara E. 1983 Hittite Plene Writing. Philadelphia, PA: University of Pennsylvania Press. 146 References “Hù in Anatolian”, in: George Cardona and Norman H. Zide (eds.), Festschrift for Henry Hoenigswald on the Occasion of His Seventieth Birthday. Tübingen: Gunter Narr, pp. 185—192. 1988 “Analogy, Secondary Ablaut and *OHø in Common Greek”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 241—256. 1999 Hittite Historical Phonology. Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft. 2017 “The Phonology of Anatolian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. I, pp. 249—256. Klein, Ernest David 1971 A Comprehensive Etymological Dictionary of the English Language. Amsterdam, London, New York, NY: Elsevier. Klein, Jared S. 1988 “Proto-Indo-European *g¦iHù- ‘live’ and Related Problems of Laryngeals in Greek”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Lautund Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 257—279. Klein, Jared S., Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.) 2017—2018 Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton. Klimov, G[eorgij] A[ndrejevič] 1989 “Рефлекс индоевропейского ларингального картвельских языках?” [A Reflex of an Indo-European Laryngeal in Kartvelian Languages?], Вопросы Языкознания (Voprosy Jazykoznanija) 1989.6:23—28. Kloekhorst, Alwin 2005 “The Fate of PIE *h3- in Hittite”. Manuscript. Downloaded from academia.edu. 2006a “Initial Laryngeals in Anatolian”, Historische Sprachforschung / Historical Linguistics, 119:77—108. 2006b “Čop’s Law in Luwian Revisited”, Die Sprache 46.2:131—136. 2008a Etymological Dictionary of the Hittite Inherited Lexicon. Leiden and Boston, MA: E. J. Brill. 2008b “Studies in Lycian and Carian Phonology and Morphology”, Kadmos 47:137—145. 2013 “Indo-European Nominal Ablaut Patterns: The Anatolian Evidence”, in: Götz Keydana, Paul Widmer, and Thomas Olander (eds.), IndoEuropean Accent and Ablaut. Copenhagen: Museum Tuscalanum Press, University of Copenhagen, pp. 107—128. 1987 References 2014a 147 Accent in Hittite: A Study in Plene Spelling, Consonant Gradation, Clitics, and Metrics. Wiesbaden: Otto Harrassowitz. 2014b “Once More on Hittite ā/e-Ablauting ḫi-Verbs”, Indogermanische Forschungen 119:55—77. 2014c “Luwian and the Glottalic Theory”. Manuscript. 2016 “The Anatolian Stop System and the Indo-Hittite Hypothesis”, Indogermanische Forschungen 121:213—247. 2017 “The Hittite Genitive Ending -ā̆n”, in: Bjarne Simmelkjær, Sandgaard Hansen, Adam Hyllested, Anders Richardt Jørgensen, Guus Kroonen, Jenny Helena Larson, Benedicte Nielsen Whitehead, Thomas Olander, and Tobias Mosbæk Søborg (eds.), Usque ad Radices: Indo-European Studies in Honour of Birgit Annette Olsen. Copenhagen: Museum Tusculanum Press, pp. 385—400. 2018 “Anatolian Evidence Suggests that the Indo-European Laryngeals Were Uvular Stops”, Indo-European Linguistics 6:69—94. 2019 Kanišite Hittite: The Earliest Attested Record of Indo-European. Leiden: E. J. Brill. 2022 “Anatolian”, in: Thomas Olander (ed.), The Indo-European Language Family: A Phylogenetic Perspective. Cambridge: Cambridge University Press, pp. 63—82. 2023a “New Interpretations in Lydian Phonology”, in: José-Virgilio García Trabazo, Ignasi-Xavier Adiego, Mariona Vernet, Bartomeo ObradorCursach, Susana Soler (eds.), New Approaches on Anatolian Linguistics. Barcino Monographica Orientalia, Volume 22 — Series Anatolica et Indogermanica 4, pp. 115—133. 2023b “Proto-Indo-Anatolian, the ‘Anatolian Split’ and the ‘Anatolian Trek’: A Comparative Linguistic Perspective”, in: Kristian Kristiansen, Guus Kroonen, and Eske Willerslev (eds.), The Indo-European Puzzle Revisited: Integrating Archaeology, Genetics, and Linguistics. Cambridge: Cambridge University Press, pp. 42—59. Kloekhorst, Alwin, and Tijmen Pronk 2019 “Introduction: Reconstructing Proto-Indo-Anatolian and Proto-IndoUralic”, in Alwin Kloekhorst and Tijmen Pronk (eds.), The Precursors of Proto-Indo-Anatolian: The Indo-Anatolian and Indo-Uralic Hypotheses. Leiden: E. J. Brill, pp. 1—14. Kluge, Friedrich, and Frederick Lutz 1898 English Etymology. A Select Glossary Serving as an Introduction to the History of the English Language. Boston, MA: D. C. Heath & Co., Publishers. Kluge, Friedrich, and Walther Mitzka 1967 Etymologisches Wörterbuch der deutschen Sprache [Etymological Dictionary of the German Language]. 20th edition. Berlin and New York, NY: Walter de Gruyter. 148 References Kluge, Friedrich, and Elmar Seebold 1989 Etymologisches Wörterbuch der deutschen Sprache [Etymological Dictionary of the German Language]. 22nd edition. Berlin and New York, NY: Walter de Gruyter. [2011] [25th edition.] Kobayashi, Masato 2004 Historical Phonology of Old Indo-Aryan Consonants. (= Study of Languages and Cultures of Asia and Africa Monograph Series 42.) Tokyo: Research Institute for Languages and Cultures of Asia and Africa, Tokyo University of Foreign Studies. 2017 “The Phonology of Indic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. I, pp. 325—344. Koivulehto, Jorma 1988 “Idg. Laryngale und die finnisch-ugrische Evidenz” [Indo-European Laryngeals and the Finno-Ugrian Evidence], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 281—297. 1992 Uralic Evidence for the Laryngeal Theory. Vienna: Publishing House of the Austrian Academy of Sciences. Kortlandt, Frederik [Herman Henri] 1981 “Glottalic Consonants in Sindhi and Proto-Indo-European”, IndoIranian Journal 21:15—19. 1988a “The Laryngeal Theory and Slavic Accentuation”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 299—311. 1988b “Vestjysk stød, Icelandic Preaspiration, and Proto-Indo-European Glottalic Stops”, in: Mohammad Ali Jazayery and Werner Winter (eds.), Languages and Cultures: Studies in Honor of Edgar C. Polomé. Berlin: Mouton de Gruyter, pp. 353—359. 2001 “Initial Laryngeals in Anatolian”. Manuscript. Downloaded from academia.edu. 2007a “Hittite ḫi-Verbs and the Indo-European Perfect”. Manuscript. Downloaded from academia.edu. 2007b “Stative and Middle in Hittite and Indo-European”. Manuscript. Downloaded from academia.edu. 2009 “C. C. Uhlenbeck on Indo-European, Uralic, and Caucasian”, Historische Sprachforschung 122:39—47. References 149 Kossmann, Maarten G. 2012 “Berber”, in: Zygmunt Frajzyngier and Erin Shay (eds.), The Afroasiatic Languages. Cambridge: Cambridge University Press, pp. 18—101. Krause, Wolfgang 1952 Westtocharische Grammatik [West Tocharian Grammar]. Heidelberg: Carl Winter. 1955 Tocharisch [Tocharian]. Leiden: E. J. Brill. Krause, Wolfgang, and Werner Thomas 1960—1964 Tocharisches Elementarbuch [Tocharian Primer]. 2 vols. Heidelberg: Carl Winter. Kronasser, Heinz 1956 Vergleichende Laut- und Formenlehre des Hethitischen [Comparative Phonology and Morphology of Hittite]. Heidelberg: Carl Winter. 1966 Etymologie der hethitischen Sprache [Etymology of the Hittite Language]. Vol. 1. Wiesbaden: Otto Harrassowitz. Kroonen, Guus 2013 Etymological Dictionary of Proto-Germanic. Leiden and Boston, MA: E. J. Brill. Kuipers, Aert 1960 Phoneme and Morpheme in Kabardian (Eastern Adyghe). The Hague: Mouton. 1975 A Dictionary of Proto-Circassian Roots. Lisse: Peter de Ridder Press. Kümmel, Martin J. 2007 Konsonantenwandel. Wiesbaden: Reichert. 2012 “Typology and Reconstruction: The Consonants and Vowels of ProtoIndo-European”, in: Benedicte Nielsen Whitehead, Thomas Olander, Birgit Anette Olsen, and Jens Elmegård Rasmussen (eds.), The Sound of Indo-European: Phonetics, Phonemics, and Morphophonemics. Copenhagen: Museum Tusculanum Press, pp. 291—329. 2014 “The Conditioning for Secondary ḫ in Hittite”, in: Piotr Taracha (ed.) (with the assistance of Magdalena Kapelus), Proceedings of the 8th International Congress of Hittitology, Warsaw, 5—9 September 2011. Warsaw: Agade, pp. 431—436. 2018 “The Survival of Laryngeals in Iranian”, in: Lucien Beek, Alwin Kloekhorst, Guus Kroonen, Michaël Peyrot, and Tijmen Pronk (eds.), Farnah: Indo-Iranian and Indo-European Studies in Honor of Sasha Lubotsky. Ann Arbor, MI, and New York, NY: Beech Stave Press, pp. 167—172. 2022 “On New Reconstructions of PIE “laryngeals”, Especially as Uvular Stops”, Acta Linguistica Petropolitana 18.1:199—215. Kuryłowicz, Jerzy 1935 Études indoeuropéennes I [Indo-European Studies I]. Kraków: Polish Academy. 1956 L’apophonie en indo-européen [Apophony in Indo-European]. Wrocław: Polish Academy. 150 References 1958 L’accentuation des langues indo-européennes [The Accentuation of the Indo-European Languages]. 2nd edition. Wrocław: Polish Academy. 1962 L’apophonie en sémitique [Apophony in Semitic]. Wrocław: Polish Academy. 1964 The Inflectional Categories of Indo-European. Heidelberg: Carl Winter. 1968 Indogermanische Grammatik. Band II: Akzent/Ablaut [Indo-European Grammar. Vol. II: Accentuation/Ablaut]. Heidelberg: Carl Winter. 1969 Review of Robert Schmitt-Brandt (1967), Die Entwicklung des indogermanischen Vokalsystems [The Development of the Indo-European Vowel System], Bulletin de la Société de Linguistique 63:41—49. 1979 “Die hethitische ḫi-Konjugation” [The Hittite ḫi-Conjugation], in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and Indo-European: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the IndoEuropean Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 143—146. Ladefoged, Peter, and Ian Maddieson 1995 The Sounds of the World’s Languages. Oxford and Cambridge, MA: Blackwell. Laroche, Emmanel 1954 “Études sur les hiéroglyphes hittites” [Studies on the Hittite Hieroglyphs], Syria 31.1/2:99—117. 1959 Dictionnaire de la langue louvite [Dictionary of the Luwian Language]. Paris: Adrien-Maisonneuve. 1960 Les hiéroglyphes hittites [The Hittite Hieroglyphs]. Vol. I. Paris: Centre National de la Recherche Scientifique. 1986 “Les laryngales de l’anatolien: état des questions” [The Laryngeals of Anatolian: State of the Matter], Comptes-rendus des séances de l’Académie des Belles-Lettres, 130e année, no. 1 (1986), pp. 134—140. Lauffenberger, Olivier 2008 Hittite Grammar. Manuscript. Laver, John 1994 Principles of Phonetics. Cambridge: Cambridge University Press. Lazarides, Iosif, etal. 2024 “The Genetic Origin of the Indo-Europeans”. This version posted 18 April 2024. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 international license: bioRxiv preprint doi: https://doi.org/10.1101/2024.04.17.589597. Lehmann, Winfred P[aul] 1952 Proto-Indo-European Phonology. Austin, TX: University of Texas Press. References 151 “Germanic Evidence”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 212—223. 1986 A Gothic Etymological Dictionary. Leiden: E. J. Brill. 1993a Theoretical Bases of Indo-European Linguistics. London, Boston, MA, Melbourne, and Henley: Routledge. 1993b Comment on “The Typology of Indo-European” by Edwin Pulleyblank, Journal of Indo-European Studies 21.1/2:119—121. 2002 Pre-Indo-European. (= Journal of Indo-European Studies Monograph 41.) Washington, DC: Institute for the Study of Man. Lehrman, Alexander 1996 “Indo-Hittite Revisited”, Indogermanische Forschungen 101:73—88. 1997 “Hittite ga-ne-eš-+ and the Laryngeal Theory”, Indogermanische Forschungen 102:151—155. 1998 Indo-Hittite Redux. Studies in Anatolian and Indo-European Verb Morphology. Moscow: Paleograph. 2001 “Reconstructing Proto-Indo-Hittite”, in: Robert Drews (ed.), Greater Anatolia and the Indo-Hittite Language Family. Papers Presented at a Colloquium Hosted by the University of Richmond, March 18—19, 2000. Washington, DC: Institute for the Study of Man, pp. 106—130. Lejeune, Michel 1972 Phonétique historique du mycénien et du grec ancien [Historical Phonology of Mycenaean and Ancient Greek]. Paris: Librairie C. Klincksieck. Lewis, Henry, and Holger Pedersen 1937 A Concise Comparative Celtic Grammar. Reprinted with corrections and a supplement, 1961. Göttingen: Vandenhoeck & Ruprecht. Ligorio, Orsat, and Alexander Lubotsky 2013 “Фригийский Язык” [The Phrygian Language], in: Yuri B. Koryakov and Andrej A. Kibrik (eds.), Языки мира: Релик-товые индоевропейские языки Передней и Центральной Азии [Languages of the World: Relict Indo-European Languages of Western and Central Asia]. Moscow: Academia, pp. 180—195. 2018 “Phrygian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical IndoEuropean Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1816—1831. Lindeman, Frederik Otto 1970 Einführung in die Laryngaltheorie [Introduction to the Laryngeal Theory]. Berlin: Walter de Gruyter. 1987 Introduction to the “Laryngeal Theory”. Oslo: Norwegian University Press. [1997] [Revised edition. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck.] Lindsay, W[allace] M[artin] 1894 The Latin Language. Reprinted 1963. New York, NY: Hafner. 1965 152 References Lipiński, Edward 2001 Semitic Languages: Outline of a Comparative Grammar. 2nd edition. Louvain / Leuven: Peeters. Lubotsky, A[lexander] M. 1988 The System of Nominal Accentuation in Sanskrit and Proto-Indo-European. Leiden, New York, NY, København, Köln: E. J. Brill. 1989 “Against a Proto-Indo-European Phoneme *a”, in: Theo Vennemann (ed.), The New Sound of Indo-European: Essays in Phonological Reconstruction. Berlin and New York, NY: Mouton de Gruyter, pp. 53—66. 1990 “La loi de Brugmann et *Hùe-” [Brugmann’s Law and *Hùe-], in: Jean Kellens (ed.), La reconstruction des laryngales [The Reconstruction of the Laryngeals]. Paris: Société d’Édition “Les Belles Lettres”, pp. 129—136. Lundquist, Jesse 2016 “On the Origin of the Anatolian 1st Person Plural Ending”. Paper presented at the 226th Annual Meeting of the American Oriental Society, Boston, MA. Lundquist, Jesse, and Anthony D. Yates 2018 “The Morphology of Proto-Indo-European”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 2079—2195. Luraghi, Sylvia 1997 Hitttite. Munich: LINCOM Europa. 1998 “The Anatolian Languages”, in: Anna Giacalone Ramat and Paolo Ramat (eds.), The Indo-European Languages. London and New York, NY: Routledge, pp. 169—196. Macak, Martin 2017 “The Phonology of Classical Armenian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. II, pp. 1037—1079. Machek, Václav 1997 Etymologický Slovník Jazyka Českého [Etymological Dictionary of the Czech Language]. Prague: Nakladatelství Lidové Noviny. Makarova, Aleksandra 2001 Plene-Writing in Neo-Hittite. Senior Thesis in the Department of Linguistics, University of California, Berkeley. Mallory, James P., and Douglas Q. Adams 2006 The Oxford Introduction to Proto-Indo-European and the Proto-IndoEuropean World. Oxford and New York, NY: Oxford University Press. Mallory, James P., and Douglas Q. Adams (eds.) 1997 Encyclopedia of Indo-European Culture. London and Chicago, IL: Fitzroy Dearborn Publishers. References 153 Mańczak, Witold 1960 “L’origine de l’apophonie e/o en indo-européen”, Lingua 9:277—287. 1979 “L’apophonie e/o en grec”, in: Bela Brogyanyí (ed.), Studies in Diachronic, Synchronic and Typological Linguistics: Fest-schrift for Oswald Szemerényi. Amsterdam: John Benjamins, part II, pp. 529— 535. Mann, Stuart E. 1977 An Albanian Historical Grammar. Hamburg: Helmut Buske. 1984—1987 An Indo-European Comparative Dictionary. Hamburg: Helmut Buske. Markey, Thomas L. 1988 “The Laryngeal Theory and Aspects of Germanic Phonology”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 313—326. Martinet, André 1953 “Non-apophonic o-Vocalism in IE”, Word IX:253—267. 1970 Économie des changements phonétiques [Economy of Phonetic Changes]. 3rd edition. Bern: Francke Verlag. 1975 Évolution des langues et reconstruction [Evolution of Languages and Reconstruction]. Vendôme: Presses Univer-sitaires de France. 1975(1953) “Remarques sur le consonantisme sémitique” [Remarks on Semitic Consonantism], in: André Martinet, Évolution des langues et reconstruction [Evolution of Languages and Reconstruction]. Vendôme: Presses Universitaires de France, pp. 248—261. (Originally published in: Bulletin de la Société de Linguistique 49 [1953].) 1975(1967) “Les laryngales” [The Laryngeals], in: André Martinet, Évolution des langues et reconstruction [Evolution of Languages and Reconstruction]. Vendôme: Presses Universitaires de France, pp. 114—168. (Originally published in: Phonetica I [1967].) 1975(1972) “Réflexion sur le vocalisme de l’indo-européen commun” [Thoughts on the Vocalism of Common Indo-European], in: André Martinet, Évolution des langues et reconstruction [Evolution of Languages and Reconstruction]. Vendôme: Presses Universitaires de France, pp. 108— 113. (Originally published in: Homenaje a Antonio Tovar [Homage to Antonio Tovar], Madrid: Gredos [1972].) 1986 Des steppes aux océans: l’indo-européen et les “Indo-Européens” [From the Steppes to the Oceans: Indo-European and the “IndoEuropeans”]. Paris: Payot. Martirosyan, Hrach 2008 Studies in Armenian Etymology, with Special Emphasis on Dialects and Culture. Indo-European Heritage. Ph.D. dissertation, Leiden University. 2010 Etymological Dictionary of the Armenian Inherited Lexicon. Leiden and Boston, MA: E. J. Brill. 154 References 2013 2019 “The Place of Armenian in the Indo-European Language Family: The Relationship with Greek and Indo-Iranian”, Journal of Language Relationship 10:75—137. “The Armenian Dialects”, in: Geoffrey Haig and Geoffrey Khan (eds.), The Languages and Linguistics of Western Asia: An Areal Perspective. Berlin and Boston, MA: De Gruyter Mouton, pp. 46—105. Matasović, Ranko 2003 “The Proto-Indo-European Vowel System from the Typological Point of View”, GOVOR XX.1/2:249—262. 2009 Etymological Dictionary of Proto-Celtic. Leiden and Boston, MA: E. J. Brill. Mayrhofer, Manfred 1956—1980 Kurzegefaßtes etymologisches Wörterbuch des Altindischen [A Concise Etymological Dictionary of Old Indic]. 4 vols. Heidelberg: Carl Winter. 1972 A Sanskrit Grammar. English translation by Gordon B. Ford, Jr. University, AL: University of Alabama Press. 1986—2001 Etymologisches Wörterbuch des Altindoarischen [Etymological Dictionary of Old Indo-Aryan]. 3 vols. Heidelberg: Carl Winter. 1983 Sanskrit und die Sprachen Alteuropas: Zwei Jahrhunderte des Widerspiels von Entdeckungen und Irrtumern [Sanskrit and the Languages of Ancient Europe: Two Hundred Years of Contradictions of Discoveries and Misunderstandings]. Göttingen: Vandenhoeck & Ruprecht. 1988 “Die Laryngaltheorie im phonologischen Halbband der ‘Indogermanischen Grammatik’” [The Laryngeal Theory in the Phonological Half of the “Indo-European Grammar”], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 327—332. 2004 Die Hauptprobleme der indogermanischen Lautlehre seit Bechtel [The Major Issues in Indo-European Phonology Since Bechtel]. Vienna: Österreichische Akadamie der Wissenschaften. Meid, Wolfgang 1979 “Der Archaismus des Hethitischen” [The Archaic Nature of Hittite], in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and Indo-European: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the IndoEuropean Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 159—176. 1988 “Einige persönliche und sachliche Bemerkungen zur Laryngaltheorie” [Some Personal and Pertinent Remarks on the Laryngeal Theory], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The References 155 Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 333—353. Meier-Brügger, Michael 2003 Indo-European Linguistics. In cooperation with Matthias Fritz and Manfred Mayrhofer. Translated by Charles Gertmenian. Berlin: Walter de Gruyter. 2010 Indogermanische Sprachwissenschaft [Indo-European Linguistics]. 9th edition. Berlin: Walter de Gruyter. Meillet, [Paul Jules] Antoine 1936 Esquisse d’une grammaire comparée de l’arménien classique [Sketch of a Comparative Grammar of Classical Armenian]. 2nd edition. Vienna: Imprimérie des PP Mekhitharistes. 1964 Introduction à l’étude comparative des langues indo-européennes [Introduction to the Comparative Study of the Indo-European Languages]. University, AL: University of Alabama Press. Reprint of 8th edition (1937). 1967 The Indo-European Dialects. English translation by N. Rosenberg. University, AL: University of Alabama Press. 1984 Les dialectes indo-européens [The Indo-European Dialects]. Reprint of 2nd edition (1922). Geneva: Éditions Slatkine. Meiser, Gerhard 1998 Historische Laut- und Formenlehre der lateinischen Sprache [Historical Phonology and Morphology of the Latin Language]. Darmstadt: Wissenschaftliche Buchgesellschaft. 2017 “The Phonology of Italic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. II, pp. 743—751. Melchert, H. Craig 1984 Studies in Hittite Historical Phonology. Göttingen: Vandenhoeck & Ruprecht. 1987 “Reflexes of *h3 in Anatolian”, Die Sprache 33:19—28. 1989 Lycian Lexicon. Chapel Hill, NC: The Author. 1992a “Hittite Vocalism”, in: Onofrio Carruba (ed.), Per una Grammatica Ittita / Towards a Hittite Grammar. Pavia: Gianni Iuculano Editore, pp. 181—196. 1992b “The Middle Voice in Lycian”, Historische Sprachforschung / Historical Linguistics 105.2: 189—199. [1993a] [2nd fully revised edition.] 1993b Cuneiform Luvian Lexicon. Chapel Hill, NC: The Author. 1994a Anatolian Historical Phonology. Amsterdam and Atlanta, GA: Rodopi. 1994b “Anatolian”, in: Françoise Bader (ed.), Langues indo-européennes [Indo-European Languages]. Paris: CNRS Éditions, pp. 121—136. 1997 “Hittite Phonology”, in: Alan S. Kaye (ed.), Phonologies of Asia and Africa. Winona Lake, IN: Eisenbrauns, vol. 2, pp. 555—567. 156 References “Language”, in: H. Craig Melchert (ed.), The Luwians. Leiden and Boston, MA: E. J. Brill. 2004a “Luvian”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 576—584. 2004b “Palaic”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 585—590. 2004c “Lycian”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 591—600. 2004d “Lydian”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 601—608. 2004e “Carian”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 609—613. 2004f A Dictionary of the Lycian Language. Ann Arbor, MI, and New York, NY: Beech Stave Press. 2011 “The Problem of the Ergative Case in Hittite”, in: Michèle Fruyt, Michel Mazoyer, and Dennis Pardee (eds.), Grammatical Case in the Languages of the Middle East and Europe. Acts of the International Colloquium “Variations, concurrence et evolution des cas dans divers domaines linguistiques”, Paris, 2—4 April 2007. Chicago, IL: The Oriental Institute of the University of Chicago, pp. 161—167. 2015 “Hittite and Indo-European: Revolution and Counterrevolution”. Manuscript. 2017 “Anatolian”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 171—201. 2020a “Luwian”, in: Rebecca Hasselbach-Andee (ed.), A Companion to Ancient Near Eastern Languages. Hoboken, NJ: John Wiley & Sons, Inc., pp. 239—256. 2020b “Hitttite Historical Phonology after 100 Years (and after 20 Years)”, in: Ronald I. Kim, Jana Mynářová, and Peter Pavúk (eds.), Hrozný and Hittite: The First Hundred Years. Proceedings of the International Conference Held at Charles University, Prague, 11—14 November 2015. Leiden and Boston, MA: E. J. Brill, pp. 258—276. Melchert, H. Craig (ed.) 2003 The Luwians. Leiden and Boston, MA: E. J. Brill. Melchert, H. Craig, and Ilya Yakubovich No date “New Luwian Verbal Endings of the First Person Plural”. Manuscript. Downloaded from academia.edu. 2003 References 157 Mendeloff, Henry 1969 A Manual of Comparative Romance Linguistics: Phonology and Morphology. Washington, DC: The Catholic University of America Press. Meriggi, Piero 1962 Hieroglyphisch-Hethitisches Glossar [Hieroglyphic Hittite Glossary]. 2nd expanded edition. Wiesbaden: Otto Harrassowitz. 1966—1975 Manuale di eteo geroglyfico [Manual of Hieroglyphic Hittite]. 3 vols. Rome: Edizioni dell’Ateneo. 1980 Schizzo grammaticale dell’anatolico. Rome: Accademia Nazionale dei Lincei. Messing, Gordon M[yron] 1947 “Selected Studies in Indo-European Phonology”, Harvard Studies in Classical Philology 56/57:161—232. Meyer, Gustav 1891 Etymologisches Wörterbuch der albanischen Sprache [Etymological Dictionary of the Albanian Language]. Reprinted 1962. Leipzig: Zentralantiquariat der DDR. Militarëv, Alexander Y. 2011 “A Complete Etymology-Based Hundred Wordlist of Semitic Updated: Items 35—54”, Journal of Language Relationship 5:69—95. Möller, Hermann 1917 Die semitisch-vorindogermanisch laryngalen Konsonanten [The Semitic / Pre-Indo-European Laryngeal Consonants]. Copenhagen: Andr. Fred. Host and Son. Moscati, Sabatino (ed.) 1964 An Introduction to the Comparative Grammar of the Semitic Languages. Wiesbaden: Otto Harrassowitz. Moulton, William G. 1972 “The Proto-Germanic Non-Syllabics (Consonants)”, in: Franz van Coetsem and Herbert L. Kufner (eds.), Toward a Grammar of ProtoGermanic. Tübingen: Max Niemeyer, pp. 141—173. Mouton, Alice 2007 “Cunéiformes et hieroglyphs chez les Hittites”, in: Brigitte Lion and Cécile Michel (eds.), Histoires de déchiffrements: Les écritures du Proche-Orient à l’Égée [History of Decipherments: The Writings from the Near East to the Aegean]. Paris: Éditions Errance, pp. 141—151. Mouton, Alice (ed.) 2017 Hittitology Today: Studies on Hittite and Neo-Hittite Anatolia in Honor of Emmanuel Laroche’s 100th Birthday / L’Hittitologie Aujourd’hui: Études sur l’Anatolie hittite et néo-hittite à l’occasion du centenaire de la naissance d’Emmanuel Laroche. Istanbul: Institut Français d’Études Anatoliennes Georges Dumézil. 158 References Neroznak, Vladimir P[etrovič]. 1992 “Phrygian”, in: Edgar C. Polomé and Werner Winter (eds.), Reconstructing Languages and Cultures. Berlin and New York, NY: Mouton de Gruyter, pp. 271—278. Neu, Erich 1968 Das hethitische Mediopassiv und seine indogermanischen Grundlagen [The Hittite Mediopassive and Its Indo-European Basis]. Wiesbaden: Otto Harrassowitz. 1979 “Einige Überlegungen zu den hethitischen Kasusendungen” [Some Reflections on the Hittite Case Endings], in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and IndoEuropean: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the Indo-European Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 177—196. Neu, Erich (ed.) 1982 Investigationes Philologicae et Comparativae: Gedenkschrift für Heinz Kronasser [Philological and Comparative Investigations: Memorial Volume for Heinz Kronasser]. Wiesbaden: Otto Harrassowitz. Neu, Erich, and Wolfgang Meid (eds.) 1979 Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and Indo-European: Comparative Studies on the Historical Grammar and DialectalGeographic Position of the Indo-European Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck. Neumann, Günter 2007 Glossar des Lykischen [Lycian Glossary]. Wiesbaden: Otto Harrassowitz. Nikitin, Alexey, and Svetlana Ivanova 2022 “Long-Distance Exchanges along the Black Sea Coast in the Eneolithic and the Steppe Genetic Ancestry Problem”. Manuscript, version 1, 28 November 2022. Cambridge open access. Nikitin, Alexey G, etal. 2024 “A Genomic History of the North Pontic Region from the Neolithic to the Bronze Age”. This version posted 18 April 2024. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 license: bioRxiv preprint doi: https://doi.org/10.1101/2024.04.17.589600. References 159 Norbruis, Stefan 2021 Indo-European Origins of Anatolian Morphology and Semantics. Innovations and Archaisms in Hittite, Luwian and Lycian. Amsterdam: LOT. Obrador-Cursach, Bartomeu 2020 The Phrygian Language. Leiden and Boston, MA: E. J. Brill. Oettinger, Norbert 1979 Die Stammbildung des hethitischen Verbums [The Formation of Hittite Verbs]. Nürnberg: Hans Carl. 1988 “Der indogermanische Nominativ Dual aus laryngalistische Sicht” [The Indo-European Nominative Dual from a Laryngeal Point of View], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 355—359. 2022 “Hethitisch menu- ‘Mensch, männlich’” [Hittite menu- ‘man, person; male, manly’], Historische Sprachforschung / Historical Linguistics 135:255—263. 2023 “ḫi-Konjugation und imperfektiver Aspekt im Luwischen und Hethitischen” [ḫi-Conjugation and Imperfective Aspect in Luwian and Hittite], in: José-Virgilio García Trabazo, Ignasi-Xavier Adiego, Mariona Vernet, Bartomeo Obrador-Cursach, Susana Soler (eds.), New Approaches on Anatolian Linguistics. Barcino Monographica Orientalia, Volume 22 — Series Anatolica et Indogermanica 4, pp. 175—182. Olander, Thomas (ed.) 2022 The Indo-European Language Family: A Phylogenetic Perspective. Cambridge: Cambridge University Press. O’Leary, De Lacy 1923 Comparative Grammar of the Semitic Languages. Reprinted 1969. Amsterdam: Philo Press. Onions, C[harles] T[albot] (ed.) 1966 The Oxford Dictionary of English Etymology. Oxford: Clarendon Press. Orël, Vladimir 1998 Albanian Etymological Dictionary. Leiden: E. J. Brill. 2000 A Concise Historical Grammar of the Albanian Language. Leiden: E. J. Brill. 2003 A Handbook of Germanic Etymology. Leiden: E. J. Brill. Oreshko, Rostislav 2019 “Phonetic Value of Lydian Letter <d> Revisited and Development of PIE Dentals in Lydian”, Wekwos 4:191—262. Patri, Sylvain 2009 “La perception des consonnes hittites dans les langues étrangères du XIIIe siècle”, Zeitschrift für Assyriologie und vorderasiatische Archäologie 99.1:87—126. 2019 Phonologie Hittite [Hitttite Phonology]. Leiden: E. J. Brill. 160 References 2020 2023 Payne, Annick 2010a 2010b 2015 “L’emplois de la flexion en -hhe en hittite” [The Uses of the Inflection in -hhe in Hittite], Historische Sprachforschung / Historical Linguistics 133:209—246. “Identification du parfait périphrastique en hittite”, Orientalia 2023.1: 44—70. “‘Writing’ in Hieroglyphic Luwian”, in: Itamar Singer (ed.), Luwian and Hittite Studies Presented to J. David Hawkins on the Occasion of His 70th Birthday. Tel Aviv: Emery and Claire Yass Publications in Archaeology — Institute of Archaeology, Tel Aviv University, pp. 182—187. Hieroglyphic Luwian: An Introduction with Original Texts. 2nd revised edition. Wiesbaden: Otto Harrassowitz. Schrift und Schriftlichkeit: Die anatolische Hieroglyphenschrift [Writing and Literacy: The Anatolian Hieroglyphic Script]. Wiesbaden: Otto Harrassowitz. Pedersen, Holger 1909—1913 Vergleichende Grammatik der keltischen Sprachen [Comparative Grammar of the Celtic Languages]. 2 vols. Göttingen: Vandenhoeck & Ruprecht. 1938 Hittitisch und die anderen indoeuropäischen Sprachen [Hittite and the Other Indo-European Languages]. Copenhagen: Levin and Munksgaard. 1945 Lykisch und Hittitisch [Lycian and Hittite]. Copenhagen: Ejnar Munksgaard. Penney, J[ohn] H. W. 1988 “Laryngeals and the Indo-European Root”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 361—372. Peters, Martin 1980 Untersuchungen zur Vertretung der indogermanischen Laryngale im Griechischen [Investigations into the Treatment of the Indo-European Laryngeals in Greek]. Vienna: Österreichischen Akademie der Wissenschaften. 1988 “Zur Frage strukturell uneinheitlicher Laryngalreflexe in idg. Einzelsprachen” [On the Question of Structurally Irregular Laryngeal Reflexes in Individual Indo-European Languages], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 373—381. References Petit, Daniel 2018 Petráček, Karel 1981 Peyrot, Michaël 2019 Pirart, Eric 1990 161 “The Phonology of Baltic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1640—1651. “K teorie laryngál” [On the Laryngeal Theory], Slovo a Slovesnost 42:262—268. “Indo-Uralic, Indo-Anatolian, Indo-Tocharian”, in: Alwin Kloekhorst and Tijmen Pronk (eds.), The Precursors of Proto-Indo-Anatolian: The Indo-Anatolian and Indo-Uralic Hypotheses. Leiden: E. J. Brill, pp. 186—202, “Le traitement des laryngales finales en védique ancien” [The Treatment of Final Laryngeals in Ancient Vedic], in: Jean Kellens (ed.), La reconstruction des laryngales [The Reconstruction of the Laryngeals]. Paris: Société d’Édition “Les Belles Lettres”, pp. 137—147. Plöchl, Reinhold 2003 Einführung ins Hieroglyphen-Luwische [Introduction to Hieroglyphic Luwian]. (= Dresdner Beiträge zur Hethitologie 8.) Dresden: Verlag der TU Dresden. Pokorny, Julius 1959 Indogermanisches etymologisches Wörterbuch [Indo-European Etymological Dictionary]. Vol. 1. Bern: Francke Verlag. Polomé, Edgar C. 1950 “Reflexes de laryngales en arménien” [Reflexes of Laryngeals in Armenian], Annuaire de l’Institut de Philologie et d’Histoire Orientales et Slaves 10:539—569. 1952 “On the Origin of Hittite ḫ”, Language 28:444—446. 1965 “The Laryngeal Theory So Far”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 9—78. 1971 “Reflexes of Laryngeals in Indo-Iranian with Special Reference to the Problem of the Voiceless Aspirates”, in: J. Weinstock (ed.), Saga og Språk: Studies in Language and Literature in Honor of Lee Hollander. Austin, TX: Pemberton Press, pp. 233—250. 1980 “Armenian and the Proto-Indo-European Laryngeals”, in: John A. C. Greppin (ed.), First International Conference on Armenian Linguistics, Proceedings. Delmar, NY: Caravan Books, pp. 17—33. 1982 “Germanic as an Archaic Indo-European Language”, in: Ernst S. Dick and Kurt R. Jankowsky (eds.), Festschrift für Karl Schneider [Commemorative Volume for Karl Schneider]. Amsterdam: John Benjamins, pp. 51—59. 1987 “Recent Developments in the Laryngeal Theory”, Journal of IndoEuropean Studies 15.1/2:159—167. 162 References 1988 “Are There Traces of Laryngeals in Germanic?”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 383—414. Pooth, Roland A. 2014 “Voice, Direction, Case, and Alignment in Proto-Indo-European IV. The Proto-Indo-European Case System”. Manuscript. Available for free download from academia.edu. 2015 “A Typological Overview of Proto-Indo-European.” Manuscript. Available for free download from academia.edu. 2020 “The Original Functions of Indo-European Ablaut: Some Remarks on Proto-Indo-European Noun Gradation and Ablaut”, New Trends in Indo-European Linguistics 1:1—26. Popova, Olga 2015 “L’introduction de l’écriture cunéiforme chez les Hittites au IIe millénaire av. J.-C” [The Introduction of Cuneiform Writing among the Hittites during the IInd Millennium BCE], in: Dominique Briquel and Françoise Briquel (eds.), Écriture et communication [Writing and Communication]. Electronic edition of CTHS (Actes des congrès des sociétés historiques et scientifiques, 2015, pp. 35—45. Pronk, Tijmen 2021 “Indo-European Secondary Products Terminology and the Dating of Proto-Indo-Anatolian”, Journal of Indo-European Studies 49.1/2:141— 170. Puhvel, Jaan 1960 Laryngeals and the Indo-European Verb. Berkeley and Los Angeles, CA: University of California Press. 1965 “Evidence in Anatolian”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 79—92. 1966 “Dialectal Aspects of the Anatolian Branch of Indo-European”, in: Henrik Birnbaum and Jaan Puhvel (eds.), Ancient Indo-European Dialects. Berkeley and Los Angeles, CA: University of California Press, pp. 235—247. 1974 “On Labiovelars in Hittite”, Journal of the American Oriental Society 94.3:291—295. 1984— Hittite Etymological Dictionary. Berlin, New York, NY, and Amsterdam: Mouton de Gruyter. 1994 “Anatolian: Autochthon or Interloper?”, Journal of Indo-European Studies 22.3/4:251—263. Pulleyblank, Edwin G. 1965 “The IE Vowel System and Qualitative Ablaut”, Word 21:86—101. 1993 “The Typology of Indo-European”, Journal of Indo-European Studies 21.1/2:63—118. References Pyysalo, Jouno 2013 163 System PIE: The Primary Phoneme Inventory and Sound Law System for Proto-Indo-European. Ph.D. dissertation, University of Helsinki. (Publications of the Institute for Asian and African Studies, 15.) Helsinki: Unigrafia. 2015 “Does the Laryngeal Theory Have a Theory? On the Incompatibility of a Viable Three-Laryngeal Model with the Anatolian Data”, Wekwos 2. Radner, Karen, and Eleanor Robson 2011 The Oxford Handbook of Cuneiform Culture. Oxford: Oxford University Press. Raster, Manu 2007 Hieroglyphenluwisches Lexikon [Hieroglyphic Luwian Lexicon]. Master’s Thesis, Wilhelms-Universität Münster, Westfalen. Reynolds, Elinor, Paula West, and John Coleman 2000 “Proto-Indo-European ‘Laryngeals’ were Vocalic”, Diachronica XVII.2:351—387. Rieken, Elisabeth 1999 Untersuchungen zur nominalen Stammbildung des Hethitischen [Research into the Nominal Stem Formation of Hittite]. Wiesbaden: Otto Harrassowitz. 2009 “Der Archaismus des Hethitischen: eine Bestandsaufnahme” [The Archaic Nature of Hittite: An Inventory], Incontri Linguistici 32:37— 52. 2011 Einführung in die hethitische Sprache und Schrifte [Introduction to the Hittite Language and Writing]. Münster: Ugarit-Verlag. 2015 “Bermerkungen zum Ursprung einiger Merkmale der anatolischen Hieroglyphenschrift” [Remarks on several Theories on the Origin of the Anatolian Hieroglyphic Writing], Die Welt des Orients 45:216—231. 2017 “Word-internal Plene Spelling with <i> and <e> in Cuneiform Luwian Texts”, Journal of Language Relationship 15.1:19—30. Ringe, Donald 1988 “Laryngeal Isoglosses in the Western Indo-European Languages”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 415—441. 1996 On the Chronology of Sound Changes in Tocharian. Vol. 1. New Haven, CT: American Oriental Society. Ringe, Donald, Tandy Warnow, and Ann Taylor 2002 “Indo-European and Computational Cladistics”, Transactions of the Philological Society 100:59—129. Ringe, Donald, Tandy Warnow, Ann Taylor, Alexander Michailov, and Libby Levison 1998 “Computational Cladistics and the Position of Tocharian”, in: Victor H. Mair (ed.), The Bronze Age and Early Iron Age Peoples of Eastern 164 References Central Asia. Washington, DC/Philadelphia, PA: Institute for the Study of Man/University of Pennsylvania, vol. I, pp. 391—414. Rix, Helmut 2001 Lexikon der indogermanischen Verben: Die Wurzeln und ihre Primärstammbildungen [Lexicon of Indo-European Verbs: The Roots and Their Primary Stem Formations]. 2nd expanded and corrected edition, revised by Martin Kümmel and Helmut Rix. Wiesbaden: Dr. Ludwig Reichert Verlag Rosenkranz, Bernhard 1953 “Die hethitische ḫi-Konjugation” [The Hittite ḫi-Conjugation], Jahrbuch für kleinasiatische Forschung 2:239—249. 1978 Vergleichende Untersuchungen der altanatolischen Sprachen [Comparative Investigations on the Old Anatolian Languages]. The Hague: Mouton. 1979 “Archaismen im Hethitischen” [Archaic Features in Hittite], in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and Indo-European: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the IndoEuropean Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 219—229. Ruhlen, Merritt 1975 A Guide to the Languages of the World. Stanford, CA: Language Universals Project, Stanford University. 1987 A Guide to the World’s Languages. Stanford, CA: Stanford University Press. Ruijgh, C[ornelis] J[ord] 1988 “Observations sur les traitements des laryngales en grec préhistorique” [Observations on the Treatments of Laryngeals in Prehistoric Greek], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 443—469. Rusakov, Alexander 2017 “Albanian”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 552—608. 2018 “The Documentation of Albanian”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 volumes. Berlin and Boston, MA: De Gruyter Mouton, vol. III, pp. 1716—1731. Salmons, Joseph C. 1993 The Glottalic Theory: Survey and Synthesis. McLean, VA: Institute for the Study of Man. References Sapir, Eduard 1938 165 “Glottalized Continuants in Navaho, Nootka, and Kwakiutl (with a Note on Indo-European)”, Language 14:248—274. Sasseville, David 2024 “Rhotacism in 1st-millennium BC Anatolia”. Manuscript. Downloaded from academia.edu. Schirru, Giancarlo 2012 “Laryngeal Features of Armenian Dialects”, in: Benedicte Nielsen Whitehead, Thomas Olander, Birgit Anette Olsen, and Jens Elmegård Rasmussen (eds.), The Sound of Indo-European: Phonetics, Phonemics, and Morphophonemics. Copenhagen: Museum Tusculanum Press, University of Copenhagen, pp. 435—457. Schleicher, August 1876 Compendium der vergleichenden Grammatik der indogermanischen Sprachen [Compendium of the Comparative Grammar of the IndoEuropean Languages]. 4th edition. Reprinted 1974. Hildesheim: Georg Olms. Schmalstieg, William R. 1966 “Neutralization of /a/ and /e/ in Hittite and Baltic”, Istituto Orientale di Napoli, Annali, Sezione Lnguistica VII:53—59. 1973 “New Thoughts on Indo-European Phonology”, Zeitschrift für vergleichende Sprachforschung (KZ) 87:99—157. 1980 Indo-European Linguistics: A New Synthesis. University Park, PA, and London: The Pennsylvania State University Press. Schmidt, Karl-Horst 1993 Comments on “The Typology of Indo-European” by Edwin Pulleyblank, Journal of Indo-European Studies 21.1/2:123—133. Schmidt, Klaus T. 1988 “Stellungsbedingte Konsonantisierung von š im Tocharischen?” [Was There a Positionally Conditioned Occlusivization of š in Tocharian?], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 471—480. Schmitt, Rüdiger 1988 “Betrachtungen über Pragmatik und Systematik in der Laryngaltheorie” [Reflections on Pragmatics and Systematics in the Laryngeal Theory], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 481—495. Schmitt-Brandt, Robert 1967 Die Entwicklung des indogermanischen Vokalsystems [The Development of the Indo-European Vowel System]. Heidelberg: Julius Groos. [1973] [2nd edition.] 166 References 1998 Schrijver, Peter 1991 Einführung in die Indogermanistik [Introduction to Indo-European Studies]. Tübingen: Francke Verlag. The Reflexes of the Proto-Indo-European Laryngeals in Latin. Amsterdam and Atlanta, GA: Rodopi. Schuhmann, Roland 2024 Grundriss der gotischen Etymologie. Manuscript, dated 11 February 2024. Downloaded from academia.edu. Schürr. Diether 2001 “Karische und lykische Sibilanten” [Carian and Lycian Sibilants], Indogermanische Forschungen 106:94—121. 2018 “Gibt es ‘Ergative’ im Lykischen?” [Do ‘Ergatives’ Exist in Lycian?], Gephyra 15:1—9. 2019 “Urindogermanisch Wein und Met in den anatolischen Sprachen” [Proto-Indo-European Wine and Mead in the Anatolian Languages], Aramazd: Armenian Journal of Near Eastern Studies XIII,1:44—58. Schwartz, Benjamin 1970 “Laryngeals: A Brief Sketch of the Current Status of the Theory”, in: Robert C. Lugton and Milton Saltzer (eds.), Studies in Honor of J. Alexander Kerns. The Hague: Mouton, pp. 95—102. Seebold, Elmar 1988 “Wissenschaftsgeschichte und Theorieformulierung. Oder: Wie nötig ist die Laryngalhypothese?” [Scientific History and Theoretical Formulation. Or: How Necessary is the Laryngeal Theory?], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 497—525. Seyfarth, Scott, and Marc Garellek 2018 “Plosive Voicing Acoustics and Voice Quality in Yerevan Armenian”, Journal of Phonetics 71: 425—450. Shevelov, George Y. 1964 A Prehistory of Slavic: The Historical Phonology of Common Slavic. New York, NY: Columbia University Press. Shevoroshkin, Vitalij [Vitaly] 1964a “On Karian”, Revue Hittite et Asianique XXII.74:1—55. 1964b “Zur karischen Schrift und Sprache” [On Carian Writing and Language], Kadmos 3:72—87. 1968 “Karisch, Lydisch, Lykisch” [Carian, Lydian, Lycian], Klio 50:53—69. 1988a “On Laryngeals”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the IndoEuropean Sound and Form Systems]. Heidelberg: Carl Winter, pp. 527—546. References 1988b 1991 2000 2012 167 “Indo-European Consonants in Anatolian”, in: Yoël L. Arbeitman, A Linguistic Happening in Memory of Ben Schwartz. Louvain-la-Neuve: Peeters, pp. 283—303. “On Carian Language and Writing”, in: Roger Pearson (ed.), Perspectives on Indo-European Language, Culture and Religion: Festschrift for Edgar C. Polomé. McLean, VA: Institute for the Study of Man, vol. 1, pp. 117—135. “Notes on Lycian and Milyan”, in: Yoël L. Arbeitman (ed.), The Asia Minor Connection: Studies on the Pre-Greek Languages in Memory of Charles Carter. Leuven and Paris: Peeters, pp. 205—211. “Anatolian Laryngeals in Milyan”, in: Benedicte Nielsen Whitehead, Thomas Olander, Birgit Anette Olsen, and Jens Elmegård Rasmussen (eds.), The Sound of Indo-European: Phonetics, Phonemics, and Morphophonemics. Copenhagen: Museum Tusculanum Press, University of Copenhagen, pp. 459—484. Shields, Kenneth 1982 Indo-European Noun Inflection. University Park, PA: The Pennsylvania State University Press. 1992 A History of Indo-European Verb Morphology. Amsterdam and Philadelphia, PA: John Benjamins. Sideltsev, Andrei, and Ilya Yakubovich 2016 “The Origin of Lycian Indefinite Pronouns and Its Phonological Implications”, Münchener Studien zur Sprachwissenschaft 70.1:75—124. Sihler, Andrew L. 1988 “Greek Reflexes of Syllabic Laryngeals with a Postscript on PIE Kinship Terms in *-Høter”, in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Lautund Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 547—561. 1995 New Comparative Grammar of Greek and Latin. New York, NY, and Oxford: Oxford University Press. Simon, Zsolt 2007 Review of J. Marangozis (2003), A Short Grammar of Hieroglyphic Luwian, Acta Antiqua Scientiarum Hungaricae 47:139—142. 2008 Review of Ignacio J. Adiego (2007), The Carian Language, Acta Antiqua Scientiarum Hungaricae 48:457—475. 2011 “Die Fortsetzung der Laryngale im Karischen” [The Continuation of the Laryngeals in Carian], in: Thomas Krisch and Thomas Lindner (eds.), Indogermanistik und Linguistik im Dialog. Akten der XIII. Fachtagung der Indogermanischen Gesellschaft vom 21. bis 27. September 2008 in Salzburg [Indo-European Studies and Linguistics in Dialog. Proceedings of the 13th Meeting of the Indo-European Society, 21—27 September 2008, in Salzburg]. Wiesbaden: Reichert, pp. 538—547. 168 References 2014 “Die phonetische Wert der luwischen Laryngale” [The Phonetic Value of the Luwian Laryngeals], in: Piotr Taracha (ed.), with the assistance of Magdalena Kapełuś, Proceedings of the Eighth International Congress of Hittitology, Warsaw, 5—9 September 2011. Warsaw: Agade, pp. 873—894. 2017 “Selected Pisidian Problems and the Position of Pisidian within the Anatolian Languages”, Journal of Language Relationship 15.1:31—42. 2019 “The Anatolian Stop System and the Indo-Hittite Hypothesis — Revisited”, in: Matilde Serangeli and Thomas Olander (eds.), Linguistic and Archaeological Perspectives on the Early Stages of Indo-European. Leiden and Boston, MA: E. J. Brill, pp. 236—250. 2020a “The Anatolian Stop System and the Indo-Hittite Hypothesis — Revisited”, in: Matilde Serangeli and Thomas Olander (eds.), Dispersals and Diversification: Linguistic and Archaeological Perspectives on the Early Stages of Indo-European. Leiden and Boston, MA: E. J. Brill, pp. 236—250. 2020b “On the Origins of the Hieroglyphic Luwian Writing System: The Chronological Problem”, Chatreššar 2:42—56. Open-access publication. 2023 “Entries published in the Digital Philological-Etymological Dictionary of Minor Ancient Anatolian Corpus Languages (eDiAna)”. Published on-line, 25 April 2023. This list is regularly updated. Sims-Williams, Patrick 2017 “Celtic”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 352—386. Singer, Itamar (ed.) 2010 Luwian and Hittite Studies Presented to J. David Hawkins on the Occasion of His 70th Birthday. Tel Aviv: Emery and Claire Yass Publications in Archaeology — Institute of Archaeology, Tel Aviv University. Skydsbjerg Friis, Louise 2015—2016 The Position of Lydian within Anatolian. Bachelor’s Thesis. Smoczyński, Wojciech 2006 Laringalų teorija ir lietuvių kalba [The Laryngeal Theory and the Lithuanian Language]. Vilnius: Lithuanian Language Institute. 2007 Lietuvių Kalbos Etimologinis Žodynas / Słownik Etymologiczny Języka Litewskiego [Etymological Dictionary of the Lithuanian Language]. 2 vols. Vilnius: University of Vilnius, Faculty of Philosophy. Spuler, B[ertold] (ed.) 1969 Altkleinasiatische Sprachen [Ancient Near Eastern Languages]. Leiden: E. J. Brill. Steadman, Sharon R., and Gregory McMahon (eds.) 2011 The Oxford Handbook of Ancient Anatolia, 10,000—323 B.C.E. Oxford and New York, NY: Oxford University Press. References Stifter, David 2017 169 “The Phonology of Celtic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. II, pp. 1188—1202. Stiles, Patrick V. 2017 “The Phonology of Germanic”, in: Jared S. Klein, Brian D. Joseph, Matthias Fritz, and Mark Wenthe (eds.), Handbook of Comparative and Historical Indo-European Linguistics. 3 vols. Berlin and Boston, MA: De Gruyter Mouton, vol. II, pp. 888—912. Strunk, Klaus 1988 “Über Laryngale und einige reduplizierte Verbalstämme” [On Laryngeals and Several Reduplicated Verbal Stems], in: Alfred Bammesberger (ed.), Die Laryngaltheorie und die Rekonstruktion des indogermanischen Laut- und Formensystems [The Laryngeal Theory and the Reconstruction of the Indo-European Sound and Form Systems]. Heidelberg: Carl Winter, pp. 563—582. Stuart-Smith, Jane 2004 Phonetics and Philology: Sound Change in Italic. Oxford: Oxford University Press. Sturtevant, Edgar H[oward] 1932 “The Development of Stops in Hittite”, Journal of the American Oriental Society 52:1—12. 1933 A Comparative Grammar of the Hittite Language. Baltimore, MD: Linguistic Society of America. 1936 A Hittite Glossary. 2nd edition. Philadelphia, PA: Linguistic Society of America. [1951] [Revised edition. Vol. I. New Haven, CT: Yale University Press.] 1938 “Hittite Evidence against Full-Grade o”, Language 14:104—111. 1940 “The Hittite Language”, The Classical Weekly 33.12:134—137. 1942 The Indo-Hittite Laryngeals. Baltimore, MD: Linguistic Society of America. Sutter, Edgar 2014 “Das hethitische Phoneme /xw/” [The Hittite Phoneme /xw/]. Working paper no. 59. Institut für Linguistik, Abteilung Allgemeine Sprachwissenschaft, Universität zu Köln. Swiggers, Pierre 1989 “On (the Nature of) Proto-Indo-European Laryngeals”, in: Theo Vennemann (ed.), The New Sound of Indo-European: Essays in Phonological Reconstruction. Berlin and New York, NY: Mouton de Gruyter, pp. 77—79. Szemerényi, Oswald 1967 “The New Look of Indo-European: Reconstruction and Typology”, Phonetica 17:65—99. 170 References 1973 1979 1990 1996 Taracha, Piotr 2012 “La théorie des laryngales de Saussure à Kuryłowicz et à Benveniste” [The Laryngeal Theory from Saussure to Kuryłowicz and Benveniste], Bulletin de la Société de Linguistique LXVIII:1—25. “Palaic and the Indo-European Laryngeals”, in: Florilegium Anatolicum: Mélanges offerts à E. Laroche [Anatolian Cullings: Miscellanea Offered to E. Laroche]. Paris: Éditions E. de Boccard, pp. 315—319. Introduction to Indo-European Linguistics. Translated from the 4th edition (1990) of Einführung in die vergleichende Sprachwissenschaft [Introduction to Comparative Linguistics] by David Morgan Jones, with additional notes and references. Oxford: Clarendon Press. “Hittitology Up to Date: Issues and New Approaches”, Rocznik Orientalistyczny. T. LXV. Z., pp. 212—223. Tischler, Johann 1977— Hethitisches etymologisches Wörterbuch. Mit Beiträgen von Günter Neumann [Hittite Etymological Dictionary: With Contributions by Günter Neumann]. Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft. Trubetzkoy, N[ikolaj] S[ergejevič] 1967 Principes de phonologie [Principles of Phonology]. French translation of Grundzüge der Phonologie (Prague [1939]) by J. Cantineau. Paris: Librairie C. Klincksieck. 1969 Principles of Phonology. English translation of Grundzüge der Phonologie (Prague [1939]) by Christiane A. Baltaxe. Berkeley and Los Angeles, CA: University of California Press. Turner, Ralph L[illey] 1966—1969 A Comparative Dictionary of the Indo-Aryan Languages. 2 vols. Impression of 1973. Indices (1969) compiled by Dorothy Rivers Turner. Oxford: Oxford University Press. Uhlenbeck, C[hristianus] C[ornelius] 1898—1899 Kurzgefasstes etymologisches Wörterbuch der altindischen Sprache [A Concise Etymological Dictionary of the Old Indic Language]. 2 vols. Amsterdam: Johannes Müller. 1900 Kurzegefasstes etymologisches Wörterbuch der gotischen Sprache [A Concise Etymological Dictionary of the Gothic Language]. 2nd edition. Amsterdam: Johannes Müller. Van den Hout, Theo P. J. 1995 “Lycian Consonantal Orthography and Some of Its Consequences for Lycian Phonology”, in: Theo P. J. van den Hout and Johan de Roos (eds.), Studio Historiae Ardens. Ancient Near Eastern Studies Presented to Philo H. J. Houwink ten Cate on the Occasion of his 65th Birthday. Leiden: Nederlands Historisch-Archaeologisch Instituut te Istanbul, pp. 105—141. References 2009 171 “A Century of Hittite Text Dating and the Origins of the Hittite Cuneiform Script”, Incontri Linguistici 32:11—35. 2011 The Elements of Hittite. Cambridge: Cambridge University Press. Van Wijk, Nicolaas 1902 Der nominale Genetiv Singular im Indogermanischen in seinem Verhältnis zum Nominativ [The Nominal Genitive Singular in IndoEuropean and its Relation to the Nominative]. Zwolle: De Erven J. J. Tijl. Van Windekens, A[lbert] J[oris] 1976—1982 Le tokharien confronté avec les autres langues indo-européennes [Tocharian Compared with the Other Indo-European Languages]. 3 vols. Louvain: Centre International de Dialectologie Générale. Vaux, Bert 1998 The Phonology of Armenian. Reprinted 2002. Oxford: Clarendon Press. 2001 “A Note on Pharyngeal Features”. Manuscript. Vennemann, Theo 1989 “Indo-European Consonant Shifts — Algebra, Arithmetic, or Normal Science”, in: Theo Vennemann (ed.), The New Sound of Indo-European: Essays in Phonological Reconstruction. Berlin and New York, NY: Mouton de Gruyter, pp. 231—244. Vercoullie, J[ozef] 1898 Beknopt Etymologisch Woordenboek der Nederlandsche Taal [Concise Etymological Dictionary of the Dutch Language]. 2nd expanded and corrected edition. Gent: J. Vuylsteke; ’S-Gravenhage: Martinus Nijhoff. Vernet, Eulàlia, and Mariona Vernet 2019 “On the Hittite and Luwian Origin of Some Common Nouns in Cappadocian Old Assyrian Texts: A New Examination”, in: Yayına Hazırlayan (ed.), Acts of the IXth International Congress of Hittitology, Çorum, September 8—14, 2014. Çorum, Türkiye, vol. II, pp. 1123— 1165. Vinereanu, Mihai 2008 Dicţionar Etimologic al Limbii Române (Pe baza cercetărilor de IndoEuropenistică) [Etymological Dictionary of the Romanian Language: Based on a Survey of Indo-European]. Bucureşti: Alcor Edimpex. [2023] [2nd revised and enlarged edition. 2 vols. Bucureşti: Editura URANUS.] 2010 Nostratic Roots in Romanian Language / Rădăcini Nostratice în Limba Română. Edited by Corina Firuţă. Bucureşti: Alcor Edimplex. 2023 “On the Centum Features of Thraco-Dacian Language”. Manuscript. Von Soden, Wolfram 1995 Grundriß der akkadischen Grammatik [Elements of Akkadian Grammar]. 3rd edition. Rome: Pontificium Institutum Biblicum. Waal, Willemijn 2012 “Writing in Anatolia: The Origins of the Anatolian Hieroglyphs and the Introductions of the Cuneiform Script”, Altorientalische Forschungen 39.2:287—315. 172 References Walde, Alois 1927—1932 Vergleichendes Wörterbuch der indogermanischen Sprachen [Comparative Dictionary of the Indo-European Languages]. Revised and edited by Julius Pokorny. 3 vols. Reprinted 1973. Berlin: Walter de Gruyter. Walde, Alois, and J[ohann] B[aptist] Hofmann 1965—1972 Lateinisches etymologisches Wörterbuch [Latin Etymological Dictionary]. 3 vols. 5th edition. Heidelberg: Carl Winter. Walker, C. B. F. 1998 “Cuneiform”, in: J. T. Hooker (ed.), Reading the Past: Ancient Writing from Cuneiform to the Alphabet. Originally published in 1990 by the Trustees of the British Museum. Reprinted 1998 by Barnes & Noble, Inc., by arrangement with The British Museum Company, Limited. Wallace, Rex 2017 “Italic”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 317—351. Walshe, M[aurice] O’C[onnell] 1951 A Concise German Etymological Dictionary. With a supplement by Marianne Winder of some Middle High German words extinct in Modern German. London: Routledge and Kegan Paul, Ltd. Watkins, Calvert 1965a “Evidence in Balto-Slavic”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 116—122. 1965b “Evidence in Italic”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 181—189. 1969 Indogermanische Grammatik. Band III/1: Geschichte der indogermanischen Verbalflexion [Indo-European Grammar. Vol. III/1: History of Indo-European Verbal Inflection]. Heidelberg: Carl Winter. 1975 “Reflexes of Laryngeals in Certain Morphological Categories in the Indo-European Languages of Anatolia”, in: Calvert Watkins (ed.), IndoEuropean Studies II. Cambridge, MA: Harvard University Press, pp. 1—27. 1998 “Proto-Indo-European: Comparison and Reconstruction”, in: Anna Giacalone Ramat and Paolo Ramat (eds.), The Indo-European Languages. London and New York, NY: Routledge, pp. 25—73. 2004 “Hittite”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 551—575. 2004 “Hittite”, in: Roger D. Woodard (ed.), The Cambridge Encyclopedia of the World’s Ancient Languages. Cambridge: Cambridge University Press, pp. 551—575. 2003 “Hittite ku-ku-uš-zi, KUB 10.99 i 29”, in: Gary Beckman, Richard Beal, and Gregory McMahon (eds.), Hittite Studies in Honor of Harry A. Hoffner Jr. on the Occasion of His 65th Birthday. Winona Lakes, IN: Eisenbrauns, pp. 389—391. References 173 Watkins, Calvert (ed.) 1985 The American Heritage Dictionary of Indo-European Roots. Boston, MA: Houghton Mifflin Company. [1992] [Revised edition. Included as an Appendix to the 3rd edition of the American Heritage Dictionary of the English Language. Boston, MA: Houghton Mifflin Co., pp. 2090—2134.] [2000] [2nd edition.] [2011] [3rd edition.] Watson, Janet C. E. 2002 The Phonology and Morphology of Arabic. Oxford and New York, NY: Oxford University Press. Weeden, Mark 2011 “Spelling, Phonology and Etymology in Hittite Historical Linguistics”. Review of Alvin Kloekhorst (2008), Etymological Dictionary of the Hittite Inherited Lexicon. Bulletin of SOAS 74.1:59—76. Weekley, Ernest 1921 An Etymological Dictionary of Modern English. London: John Murray. Weeks, David Michael 1985 Hittite Vocabulary: An Anatolian Appendix to Buck’s “Dictionary of Selected Synonyms in the Principal Indo-European Languages”. Ph.D. dissertation, University of California, Los Angeles. Weiss, Michael 2009 Outline of the Historical and Comparative Grammar of Latin. Ann Arbor, MI: Beech Stave Press. 2016 “The Proto-Indo-European Laryngeals and the Name of Cilicia in the Iron Age”, in: Andrew M. Byrd, Jessica DeLisi, and Mark Wenthe (eds.), Tavet Tat Satyam. Studies in Honor of Jared S. Klein on the Occasion of His Seventieth Birthday. Ann Arbor, MI, and New York, NY: Beech Stave Press, pp. 331—340. Weitenberg, J[oseph] J. S. 1979 “Einige Bemerkungen zu den hethitischen Diphthong-Stämen” [Some Remarks on the Hittite Diphthong Stems], in: Erich Neu and Wolfgang Meid (eds.), Hethitisch und Indogermanisch: Vergleichende Studien zur historischen Grammatik und zur dialectgeographischen Stellung der indogermanischen Sprachgruppe Altkleinasiens [Hittite and IndoEuropean: Comparative Studies on the Historical Grammar and Dialectal-Geographic Position of the Indo-European Languages of the Ancient Near East]. Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, pp. 289—303. Werner, Rudolf 1991 Kleine Einführung ins Hieroglyphen-Luwische [A Brief Introduction to Hieroglyphic Luwian]. Göttingen: Vandenhoeck and Ruprecht. [1990] [English translation of Kleine Einführung ins Hieroglyphen-Luwische [A Brief Introduction to Hieroglyphic Luwian] by Rudolf Werner (in collaboration with Barbara Luscher). Manuscript.] 174 References Winter, Werner 1965a “Armenian Evidence”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 100—115. 1965b “Tocharian Evidence”, in: Werner Winter (ed.), Evidence for Laryngeals. The Hague: Mouton, pp. 190—211. Whitehead, Benedicte Nielsen, Thomas Olander, Birgit Anette Olsen, and Jens Elmegård Rasmussen (eds.) 2012 The Sound of Indo-European: Phonetics, Phonemics, and Morphophonemics. Copenhagen: Museum Tusculanum Press, University of Copenhagen. Wilhelm, Gernot 2010 “‘Remarks on the Hittite Cuneiform Script”, in: Itamar Singer (ed.), Luwian and Hittite Studies Presented to J. David Hawkins on the Occasion of His 70th Birthday. Tel Aviv: Emery and Claire Yass Publications in Archaeology — Institute of Archaeology, Tel Aviv University, pp. 256—262. Winter, Werner (ed.) 1965 Evidence for Laryngeals. The Hague: Mouton. Wodtko, Dagmar, Britta Irslinger, and Carolin Schneider 2008 Nomina im indogermanischen Lexikon [Nouns in the Indo-European Lexicon]. Heidelberg: Carl Winter. Woodhouse Robert 2006 “Conditioned Devoicing of Mediae in Phrygian”, Studia Etymologica Cracoviensia 11:158—191. 2010 “Devoicing of PIE Media in Phrygian”, Historische Sprachforschung 122:208—227. 2012 “Hittite Etymologies and Notes”, Studia Linguistica Universitatis Iagellonicae Crocoviensis 129:226—243. 2013 “On the Reality of the Laryngeal Theory: A Response to Witold Mańczak”, Historische Sprachforschung / Historical Linguistics 126: 3—32. Woods, Christopher 2020 “The Emergence of Cuneiform Writing”, in: Rebecca HasselbachAndee (ed.), A Companion to Ancient Near Eastern Languages. Hoboken, NJ: John Wiley & Sons, Inc., pp. 27—46. Woudhuizen, Fred C. 2010—2011 “Two Notes on Lydian”, TALANTA XLII—XLIII:207—213. 2015 Hieroglyphic Luwian: Texts, Grammar, Indices. Manuscript. 2019 “Origin of the Luwian Hieroglyphic Script”, in: Yayına Hazırlayan (ed.), Acts of the IXth International Congress of Hittitology, Çorum, September 8—14, 2014. Çorum, Türkiye, vol. II, pp. 1189—1214. Wright, Joseph, and Elizabeth Mary Wright 1925 Old English Grammar. 3rd edition. Oxford: Oxford University Press. Wyatt, William F. 1970 Indo-European /a/. Philadelphia, PA: University of Pennsylvania Press. References 175 Yakubovich, Ilya 2008 Sociolinguistics and the Luwian Language. Ph.D. dissertation, The University of Chicago. 2010 Review of Harry A. Hoffner and H. Craig Melchert (2008), A Grammar of the Hittite Language, Bibliotheca Orientalis LXVII.1/2:148—154. 2011 “Luwian and the Luwians”, in: Sharon R. Steadman and Gregory McMahon (eds.), The Oxford Handbook of Ancient Anatolia, 10,000— 323 B.C.E. Oxford and New York, NY: Oxford University Press, pp. 534—547. 2012 Review of Ignacio J. Adiego (2007), The Carian Language, Journal of Near Eastern Studies 71.1:131—133. 2020 “Hittite”, in: Rebecca Hasselbach-Andee (ed.), A Companion to Ancient Near Eastern Languages. Hoboken, NJ: John Wiley & Sons, Inc., pp. 221—237. 2022a Review of Sylvain Patri (2019), Phonologie hittite, Journal of the American Oriental Society 142.1:239—242. 2022b “The Place of Lydian in the Anatolian Family through the Lens of Recent Research”, Journal of Language Relationship 20.3:191—221. 2022c “Once Again on the Evolution of the Anatolian Hieroglyphic Script”, Kadmos 61.1/2:37—60. Yates, Anthony D. 2015 “Anatolian Default Accentuation and its Diachronic Consequences”, Indo-European Linguistics 3:145—187. 2017 “Hittite ‘fire’ and ‘proterokinesis’ as Epiphenomenon”. WeCIEC 2017 — 3/4 November 2017. Downloaded from academia.edu. 2019a “The Phonology, Phonetics, and Diachrony of Sturtevant’s Law”, IndoEuropean Linguistics 7:241—307. 2019b “Suffixal *o-vocalism without “Amphikinesis”: On Proto-Indo-European *oi-stems and Ablaut as a Diagnostic for Word Stress”, in: David M. Goldstein, Stephanie W. Jamison, and Brent Vine (eds.), Proceedings of the 30th Annual UCLA Indo-European Conference. Bremen: Hempen, pp. 199—221. Young, Steven 2017 “Baltic”, in: Mate Kapović (ed.), The Indo-European Languages. 2nd edition. London and New York, NY: Routledge, pp. 486—518. Zair, Nicholas 2012 The Reflexes of the Proto-Indo-European Laryngeals in Celtic. Leiden and Boston, MA: E. J. Brill. Zgusta, Ladislav 1951 “La théorie laryngale” [The Laryngeal Theory], Archív Orientální 19: 428—472. 