Geoffrey Caveney - Academia.edu (original) (raw)

Papers by Geoffrey Caveney

* simple consonant inventory * complex consonant inventory * complex vowel inventory * simple vow... more * simple consonant inventory * complex consonant inventory * complex vowel inventory * simple vowel inventory

The paper summarizes recent advances in explaining the origins of Finnic l-cases and western Ural... more The paper summarizes recent advances in explaining the origins of Finnic l-cases and western Uralic s-cases as deriving from the grammaticalization of postpositions. The paper presents a new hypothesis for the source of western Uralic s-cases and Samoyedic *ntə-cases: the Uralic noun root *sorśi 'span, quarter' (measurement). This root's reflexes and local / spatial meanings in the three Ugric branches and in Mari are presented. The significance of initial š in the Mari reflex and initial *s in the Khanty reflex, based on a known and regular Khanty sound law, are discussed. The paper then presents and discusses the postpositions Komi (Permic) šör-, Mansi tār(ä)-, Surgut (Eastern) Khanty surə-, and Shuryshkar (Northern) Khanty sur(ə)-, explaining their likely derivation from *sorśi and providing numerous examples of their functional usage. It is argued that the local and spatial functions of these postpositions are quite similar to those of the western Uralic s-cases. The paper thus proposes that these Permic and Ugric postpositions are cognate to the western Uralic s-cases and Samoyedic *ntə-cases. The conclusion also recognizes the alternate possibility that the previously proposed source *seCV-'inside, interior' may be the origin of the Saami, Finnic, and Mordvinic s-cases, while *sorśi may be the origin of the Mari "lative" case -(e)š, the Mari s-cases, and/or the Samoyedic *ntə-cases. The point is made that if the Proto-Finno-Volgaic stage did not exist, then the s-cases of the four western Uralic branches did not necessarily have to develop from the same source.

This paper provides evidence for a genetic relationship between the Sino-Tibetan and Na-Dene lang... more This paper provides evidence for a genetic relationship between the Sino-Tibetan and Na-Dene language families, by making a comparison between Na-Dene lexical forms with initial *kw- / *gw- / *xw- and Sino-Tibetan forms with initial ŋ-, which are the initial consonants of the 1st person singular pronominal forms in the respective families. The paper reviews some parallels in Sino-Tibetan and Na-Dene verbal morphology and describes each family’s 1st person singular pronominal forms. It then presents the results of a study of all Athabaskan and Na-Dene lexical roots that have initial consonants with the same place of articulation as the 1st person pronouns, and identifies 21 such roots as candidates for comparison with Sino-Tibetan. Out of these roots, five cognate sets with Sino-Tibetan roots with initial ŋ- are then presented. Finally, a statistical estimate is calculated of the probability of this amount of resemblances occurring by chance, which is found to be extremely small.

The Ramanujan Journal, 2012

Gronwall's function G is defined for n > 1 by G(n) = σ(n) n log log n where σ(n) is the sum of th... more Gronwall's function G is defined for n > 1 by G(n) = σ(n) n log log n where σ(n) is the sum of the divisors of n. We call an integer N > 1 a GA1 number if N is composite and G(N) ≥ G(N/p) for all prime factors p of N. We say that N is a GA2 number if G(N) ≥ G(aN) for all multiples aN of N. In arXiv 1110.5078, we used Robin's and Gronwall's theorems on G to prove that the Riemann Hypothesis (RH) is true if and only if 4 is the only number that is both GA1 and GA2. Here, we study GA1 numbers and GA2 numbers separately. We compare them with superabundant (SA) and colossally abundant (CA) numbers (first studied by Ramanujan). We give algorithms for computing GA1 numbers ; the smallest one with more than two prime factors is 183783600, while the smallest odd one is 1058462574572984015114271643676625. We find nineteen GA2 numbers ≤ 5040, and prove that a GA2 number N > 5040 exists if and only if RH is false, in which case N is even and > 10 8576 .

Drafts by Geoffrey Caveney

This paper discusses the very large number of noun roots in the Taulil (Papuan) language of East ... more This paper discusses the very large number of noun roots in the Taulil (Papuan) language of East New Britain, Papua New Guinea that begin with the first syllable kə-. This segment is synchronically opaque, but the paper identifies the definite article kə found in the nearby Baining language Mali and proposes that many of these Taulil words are loanwords from an earlier stage of a neighboring Baining language. It is proposed that noun phrases including the Baining article were borrowed as opaque noun root forms in Taulil, whose noun phrase syntax does not have articles or determiners. (Compare Arabic borrowings into European languages with the Arabic article al as part of the loanwords' noun root forms, e.g. algebra.) Of particular cultural significance are the names of the two Taulil moiety groups, Kəmətarə and Kəvuik, which are both names of birds and mythical moons. The paper identifies the Mali form vuik 'above' and proposes the etymology of Kəvuik as an ancient borrowing of a Baining noun phrase similar to Mali kə vuik=ki 'the feminine (one who is) above' ('moon' is a feminine noun in both Taulil and Baining languages). The paper also proposes Baining loanword etymologies for the Taulil nouns kətiləm 'prawn', kəguiŋ 'grass', kəl=e 'granddaughter', kəl=a 'grandson', kəlak 'house', and kəlum 'green sp.; aibika', all analyzed as borrowings of Baining noun phrases including the article kə. The paper mentions more speculative possible Baining loanword etymologies for the Taulil words kəledak 'small', kəmar 'war', kəbərək 'blood', and kəvənav 'rain'. Finally, the paper lists 34 additional Taulil noun root forms with initial kə- to illustrate the scope of this layer of nouns in Taulil and to stimulate further research into Taulil etymology by other scholars.

This paper presents a new reading and interpretation of the Issyk inscription of ancient Central ... more This paper presents a new reading and interpretation of the Issyk inscription of ancient Central Asia (circa 6 th-2 nd c. BCE), in the territory of modern Kazakhstan, which is written on a silver bowl found in the burial mound of a young deceased person from a prominent family, known as the "golden man". The analysis is based on the very recent partial decipherment of the Kushan script by Bonmann, Halfmann, Korobzow, and Bobomulloev (BHKB). The paper applies the probable phonetic values of the characters of the script, determined by BHKB, to the characters appearing in the Issyk inscription. Additionally, the phonetic value = /dʒ/ is hypothesized for character 11, in connection with BHKB's phonetic value = /j/ for the similar character 19. The paper observes, based on a statistical analysis of the frequencies of vowel diacritics, that the use of these diacritics appears to be more rudimentary and less developed in the Issyk inscription than it was in the inscriptions written centuries later that were deciphered by BHKB. In particular, it is proposed that the vowel length distinction /ā/ vs. /a/ may not be consistently distinguished by the characters in the Issyk inscription. Based on all of this preliminary analysis, the paper then proceeds to provide a reading and interpretation of line 1 of the Issyk inscription, identified to be written in an archaic form of the Saka language (Middle Iranian): ama cū juvā hvata ata naṣata jamākā, "For you, who was indeed (too/very) young yourself (and) perished, [this] vessel…." A detailed analysis of each word, and of the phonology, morphology, and syntax of the statement, is provided. The paper concludes with a more provisional reading and interpretation of line 2 of the Issyk inscription, which is much more difficult to decipher because both the first character and the last character of line 2 are difficult to identify with certainty.

This paper presents sound laws for strictly regular sound correspondences between both first vowe... more This paper presents sound laws for strictly regular sound correspondences between both first vowels and second vowels in Proto-Uralic and Proto-Eskimo roots. Uralic-Eskimo initial and intervocalic consonant sound laws are also summarized and stated. The paper presents the Uralic-Eskimo vowel pair correspondence classes that follow from these sound laws and correspondences, as well as correspondence sets of lexical examples for each such vowel pair correspondence class. A total of 51 such lexical examples of the Uralic-Eskimo vowel pair correspondences, with strong semantic links, are presented. An appendix presents an additional 44 possible lexical examples of the Uralic-Eskimo vowel pair correspondences, exhibiting greater semantic drift.

This document presents an ordered sequence of conditioned regular sound changes to account for th... more This document presents an ordered sequence of conditioned regular sound changes to account for the development of the Proto-Hungarian long and short back vowels *ū, *ŭ, *ā, *ă from Proto-Uralic *u, *o, *a, *e̮ _a. The sound changes are proposed to have occurred in an ordered sequence of stages, labeled as Pre-Hungarian I, II, III, IV, leading to Proto-Hungarian. Certain Post-Proto-Hungarian secondary developments, changing the length and height of certain back vowels, are also presented. A table of 101 lexical examples of the development of these Proto-Hungarian long and short back vowels from Proto-Uralic according to these sound laws is presented, with the proposed form of each root at each stage of development provided. Exceptional lexical examples are discussed, with explanations proposed for many of them.

This paper presents sound laws for strictly regular sound correspondences between both first vowe... more This paper presents sound laws for strictly regular sound correspondences between both first vowels and second vowels in Proto-Uralic and Proto-Eskimo roots. Uralic-Eskimo initial and intervocalic consonant sound laws are also summarized and stated. The paper presents the Uralic-Eskimo vowel pair correspondence classes that follow from these sound laws and correspondences, as well as correspondence sets of lexical examples for each such vowel pair correspondence class. A total of 49 such lexical examples of the Uralic-Eskimo vowel pair correspondences, with strong semantic links, are presented. An appendix presents an additional 45 possible lexical examples of the Uralic-Eskimo vowel pair correspondences, exhibiting greater semantic drift.

This document presents an ordered sequence of conditioned regular sound changes to account for th... more This document presents an ordered sequence of conditioned regular sound changes to account for the development of the Proto-Hungarian long and short back vowels *ū, *ŭ, *ā, *ă from Proto-Uralic *u, *o, *a, *e̮_a. The sound changes are proposed to have occurred in an ordered sequence of stages, labeled as Pre-Hungarian I, II, III, IV, leading to Proto-Hungarian. Certain Post-Proto-Hungarian secondary developments, changing the length and height of certain back vowels, are also presented. A table of 100 lexical examples of the development of these Proto-Hungarian long and short back vowels from Proto-Uralic according to these sound laws is presented, with the proposed form of each root at each stage of development provided. Exceptional lexical examples are discussed, with explanations proposed for many of them.

This document presents a "proof of concept" of the author's broader analysis of the historical de... more This document presents a "proof of concept" of the author's broader analysis of the historical development of Proto-Hungarian long and short back vowels from Proto-Uralic, by focusing on what the author considers to be the most difficult and complicated case: a sequence of ordered and conditioned regular sound laws to account for the historical development of the set of roots with Proto-Uralic first vowel *o before Proto-Hungarian *r, *l, *v. The document proposes a sequence of several stages of development: Pre-Hungarian I, II, III, leading to Proto-Hungarian. Certain Post-Proto-Hungarian conditioned secondary changes in first vowel lengths are also proposed and described, arriving at attested forms in Old Hungarian and modern Hungarian.

This document presents a provisional set of phonological conditioning factors that appear to dete... more This document presents a provisional set of phonological conditioning factors that appear to determine the development of the Hungarian back vowels /u/, /o/, /a/ from the Proto-Uralic back vowels *u and *o.

This paper proposes a set of conditioning factors to explain the split of Proto-Uralic first vowe... more This paper proposes a set of conditioning factors to explain the split of Proto-Uralic first vowel *e into Proto-Khanty *i in some roots and *ää in others, a known phenomenon that has previously remained unexplained. The paper quotes the 13 examples of PU *e > PKh *i and 5 examples of PU *e > PKh *ää cited by Luobbal Sámmol Sámmol Ánte (Ante Aikio) in his 2015 study of Ob-Ugric etymologies. The paper observes the pattern that PU *e > PKh *i occurs word-initially and after labials, while PU *e > PKh *ää occurs elsewhere, and notes the same conditioning environment for another Proto-Khanty vocalic development described in Mikhail Zhivlov's 2014 study of Uralic vocalism. (The paper suggests in passing that word-initially and after labials PU *e may have developed to **ɔ̈ and then directly to mid vowel reflexes in many Khanty dialects without proceeding through an actual stage of *i.) The paper then notes the analysis of Sammallahti 1988, cited in Aikio 2006, that PU *e > PKh *i where the PU second vowel is *i with a single intervocalic consonant between them. The paper applies this rule to all examples with PU second vowel *i, thus explaining PU *tejmi ‘soft; turn soft’ > PKh **tim-, PU *ćeki ‘catfish / burbot’ > PKh *siɣ, PU *ki/eči ‘illness’ > PKh *kičV, PU *šelki- ‘fly’ > PKh *Ligel-, PU *ńeri ‘twig’ > PKh *ńir(ii), and then proposes also to apply the rule before PU *j in an intervocalic cluster, thus explaining PU *neljä ‘four’ > PKh *ńilǝ.
The paper thus proposes a complete sound law defining all conditioning factors: PU *e > PKh *i in PU word-initial position, after PU initial labials *p, *m, *w, before PU second vowel *i, and/or before PU *j in intervocalic cluster; PU *e > PKh *ää when none of the above conditions apply, i.e., after a non-labial consonant and before PU second vowel *ä without intervocalic *j. The paper identifies 5 apparent exceptions to the sound law, casts doubt on a formally possible but unlikely additional sound rule to explain 1-3 of them, and instead proposes an expressive sound-symbolic preference for a long open low vowel to explain PU *pečä- ‘rip up’ > PKh *pääč- (not **pič-) and PU *peksä- ‘beat’ > PKh *pääɣəl- (not **piɣəl-) ‘forge’. The paper notes that PU *pi/ešä- ‘cook’ has an uncertain first vowel and contains the problematic PU phoneme *š. Finally, the paper proposes that contamination and confusion between the reflexes of the roots PU *keri ‘tree bark’ and PU *ke/ärä ‘bunch’ > PKh ‘bundle (of fish, skins, beads, sticks, bark?)’ led to irregular development of PKh *käär instead of **kir and *kerää instead of **käärää / **kärää.

This paper presents sets of lexical comparisons between Proto-Uralic and Proto-Eskimo forms exhib... more This paper presents sets of lexical comparisons between Proto-Uralic and Proto-Eskimo forms exhibiting regular sound correspondences between Uralic and Eskimo intervocalic consonants and clusters as well as between Uralic and Eskimo first vowels. The paper presents comparisons exhibiting the intervocalic correspondences Uralic *-δ-/*-δ’- ~ Eskimo *-l-/*-t (10 examples), Uralic *-(n/ń)č/ć- ~ Eskimo *-k/q- (18 examples), Uralic *-l- ~ Eskimo *-t / *-l- (7 examples), Uralic *-(C)k- ~ Eskimo *-k/q- (17 examples), Uralic *-n-/*-ŋ- ~ Eskimo *-n-/*-ŋ- (9 examples), Uralic *-m-/*-p- ~ Eskimo *-m- (12 examples), Uralic *-(C)t- ~ Eskimo *-t- / *-c- (6 examples), Uralic *-w- ~ Eskimo *-v- (3 examples), Uralic *-r- ~ Eskimo *-ł-/-l- (9 examples), and Uralic *-j- ~ Eskimo *-y- (2 examples). Significantly, Section 11 of the paper demonstrates that these lexical comparisons also exhibit strikingly consistent regular first vowel correspondences: PU *e ~ PEsk *ə (11 examples); PU *e̮ ~ PEsk *a (10 examples); PU *u ~ PEsk *i before PU 2nd vowel *a, except before PEsk-Al nasal (7 examples), PU *u ~ PEsk *a before PEsk-Al nasal and PU 2nd vowel *a (3 examples), or PU *u ~ PEsk *u elsewhere (10 examples); PU *ü ~ PEsk *i before PU 2nd vowel *ä (5 examples) or PU *ü ~ PEsk *u elsewhere (5 examples); PU *ä ~ PEsk *i in PU closed syllables or before any obstruent (7 examples) or PU *ä ~ PEsk *a elsewhere (3 examples); PU *a ~ PEsk *i in PU closed syllables or before any obstruent, where the PU 2nd vowel is *i (3 examples) or PU *a ~ PEsk *a elsewhere (10 examples); PU *o ~ PEsk *u (11 examples); PU *i ~ PEsk *i (5 examples). This analysis leaves only a pair of exceptional examples (PU *e ~ PEsk *a and PU *näki- ~ PEsk *nyaku-) and a comparison of derivational suffixes without first vowels, otherwise accounting for all 93 examples of the intervocalic consonant correspondences. At the conclusion of the paper, a provisional set of Uralic-Eskimo first vowel correspondence sound laws with specific conditions is proposed and concisely stated.

The Ramanujan Journal, 2012

This paper presents sound laws for strictly regular sound correspondences between both first vowe... more This paper presents sound laws for strictly regular sound correspondences between both first vowels and second vowels in Proto-Uralic and Proto-Eskimo roots. Uralic-Eskimo initial and intervocalic consonant sound laws are also summarized and stated. The paper presents the Uralic-Eskimo vowel pair correspondence classes that follow from these sound laws and correspondences, as well as correspondence sets of lexical examples for each such vowel pair correspondence class. A total of 49 such lexical examples of the Uralic-Eskimo vowel pair correspondences, with strong semantic links, are presented. An appendix presents an additional 45 possible lexical examples of the Uralic-Eskimo vowel pair correspondences, exhibiting greater semantic drift.

This document presents an ordered sequence of conditioned regular sound changes to account for th... more This document presents an ordered sequence of conditioned regular sound changes to account for the development of the Proto-Hungarian long and short back vowels *ū, *ŭ, *ā, *ă from Proto-Uralic *u, *o, *a, *e̮_a. The sound changes are proposed to have occurred in an ordered sequence of stages, labeled as Pre-Hungarian I, II, III, IV, leading to Proto-Hungarian. Certain Post-Proto-Hungarian secondary developments, changing the length and height of certain back vowels, are also presented. A table of 100 lexical examples of the development of these Proto-Hungarian long and short back vowels from Proto-Uralic according to these sound laws is presented, with the proposed form of each root at each stage of development provided. Exceptional lexical examples are discussed, with explanations proposed for many of them.

This document proposes a new sound correspondence between Proto-Uralic initial *ć- and Proto-Eski... more This document proposes a new sound correspondence between Proto-Uralic initial *ć- and Proto-Eskimo initial *k-, presenting a correspondence set of 18 lexical comparisons between such Uralic and Eskimo roots. It is noted that even the most extensive previous proposals of Uralic-Eskimo sound correspondences, such as Fortescue 1998, did not include this correspondence. The document describes the relatively recent proposal by Zhivlov (2014) to reconstruct the Proto-Uralic phoneme as *ć rather than the traditional standard *ś, the support for this proposal in Aikio 2019, and the adoption of this reconstruction in Ánte's draft of the Uralic Etymological Dictionary (2020). The document suggests that this more accurate phonological reconstruction of the Proto-Uralic phoneme motivated the author to consider the PEsk *k ~ PU *ć correspondence as a natural result of a typologically common palatalization sound change *k > *ć in Proto-Uralic, whereas previously it would have been much less natural for anyone to consider a correspondence between *k and the fricative sibilant phoneme *ś. The document notes the distinctive medial intervocalic consonant correspondences found in the possible Uralic-Eskimo comparisons presented in Aikio 2019. The document notes the lack of PU *ći- forms in the comparisons here, and indeed the rarity of PU *ći- in general, and suggests that PUEsk initial *ki- > *kj- > *k j- may have been the original conditioning environment of the palatalization > *ć- in PU, possibly explaining why *i rarely remained to appear after *ć- in PU.

This sketch presents a new analysis of the Proto-Eskaleut (Proto-Eskimo-Aleut) vowel system as an... more This sketch presents a new analysis of the Proto-Eskaleut (Proto-Eskimo-Aleut) vowel system as an 8-vowel harmonic system, rather than the traditional 3-vowel or 4-vowel system, in order to explain the otherwise unexplained and seemingly arbitrary correspondences between Aleut and Proto-Eskimoan coronal obstruent phonemes: Eskimoan *t may correspond with either Aleut t or ch, while Eskimoan *c may correspond with either Aleut ch or s, for which no regular phonetic conditioning environments have been demonstrated to date. This sketch proposes the thesis that the Aleut phoneme ch developed from both Proto-Eskaleut *t and *s before original Proto-Eskaleut front vowels. Then at a later stage, a drastic simplification of the Aleut vowel system (similar but not identical to a parallel simplification of the Eskimoan vowel system) eliminated the backness contrasts (front vs. back/central position) that had conditioned the development of Aleut ch, obscuring the origin of this phoneme and the basis of the Aleut-Eskimoan coronal obstruent correspondences. This hypothesis is motivated and justified by a discussion of attested vowel harmony in Aleut, and also of attested fronted vowel allophones of back/central vowel phonemes and mid vowel allophones of high vowel phonemes in Aleut as well.

This paper presents a new reading and interpretation of lines 2-3 of the Lydian language inscript... more This paper presents a new reading and interpretation of lines 2-3 of the Lydian language inscription on the stele LW 22: 〈dumms ebad fawnẽriš amẽd sfardẽtλ astrkoλ〉 / 〈mλatalad mλimns aλidad wiswid kaττirs〉 is interpreted as “While/As he [Mλimnas] is inscribing here, to/for the Sardian patron saint, / Mλimnas promises to prevent/forbid [literally “promises the preventing/forbidding of”] any alteration”. The crucial new piece of the analysis focuses on the words 〈fawnẽriš amẽd〉 in line 2, which are interpreted as a present progressive aspectual verb phrase meaning “is inscribing”. The analysis of the Lydian word 〈amẽd〉 as the helping verb “is” prompts a brief and preliminary commentary on a possible comparison with a similar root in another contemporary language of the broader region.

This document is a supplement to a more fully developed theory demonstrating strong evidence of I... more This document is a supplement to a more fully developed theory demonstrating strong evidence of Indo-Aryan words and grammar in 10 Minoan Linear A inscriptions. The author has recently posted two draft papers devoted to this topic on this website, “Evidence of Indo-Aryan words in the Minoan Linear A libation formula,” which provided mainly background information justifying the historical possibility of the argument, and “Evidence of Indo-Aryan words and grammar in 8 Minoan Linear A inscriptions,” which provided much more substantial and detailed evidence for the theory.

This paper substantially expands upon the author’s previous draft paper, “Evidence of Indo-Aryan ... more This paper substantially expands upon the author’s previous draft paper, “Evidence of Indo-Aryan words in the Minoan Linear A libation formula”. The paper begins with the argument that it is methodologically sound to compare certain Linear A words with certain very similar but not identical Indo-Aryan forms only attested in relatively later stages of Sanskrit (i.e., not in the Rigveda), because one cannot expect to find all the same words and expressions attested both in Minoan inscriptions and in a work of epic poetry, even if they indeed reflect dialects of the same language. The paper emphasizes the consistent use of the same basic phonetic values and the same basic principles of the representation of the language by the script in this author’s analysis of Linear A and in Ventris & Chadwick’s analysis of Linear B. The basic principles of the syllabic sign system of Linear B are summarized, and the phonemic reading rules specific to Linear A are presented. The main body of the paper comprises the reading and analysis of 8 Minoan Linear A inscriptions as grammatical Indo-Aryan statements. The paper concludes, not modestly but honestly, with the author’s opinion that the resemblances between the words in the Linear A inscriptions and the corresponding Sanskrit forms, words, and phrases presented and analyzed in this paper are much closer than could possibly be due to pure random chance coincidence alone, and thus that the language of the 8 Linear A inscriptions presented and analyzed in this paper must represent some dialect of an Indo-Aryan language.

This paper presents a detailed syntactical and grammatical analysis of the statement, written in ... more This paper presents a detailed syntactical and grammatical analysis of the statement, written in the Linear A script of the Minoan civilization on Crete, which is known as the libation formula. The hypothesis is proposed that certain words in this statement, and possibly the entire statement, have an Indo-Aryan origin, related to the ancient Vedic Sanskrit language as well as to the Indo-Aryan superstrate in the Mitanni language of Anatolia dating to the 16th century BCE. The paper refutes the argument that the language of Linear A cannot be Indo-European, and also rejects Hubert La Marle's Indo-Iranian theory of Linear A. The paper discusses the possible invasion of Crete from Anatolia circa 1700 BCE, and demonstrates that most Linear A writing is dated after 1600 BCE. The paper argues that the Linear A logogram *301 is likely to represent a Minoan deity, quite possibly the Minoan Great Goddess, and unlikely to be a part of a verb root, where it appears in the first word of the libation formula. The paper connects the alternation between the endings -SI and -TI in a certain word of the libation formula with the alternation between one or two names of dedicants in the formula, and analyzes these endings as Indo-Aryan 2nd person singular and plural verb endings respectively. Finally, a grammatical analysis is made of the complete statement of the Linear A libation formula, in the course of which certain Linear A words are analyzed to represent Indo-Aryan words, such as DU-PU2-RE = dyu-putre, (J)A-SA-SA-RA(-ME) = ācāra(m), I-PI-NA-MA = aipi nama, and SI-RU-TE = śrutyai.

This paper examines the question of how strongly head-marking languages such as Yana and Washo, w... more This paper examines the question of how strongly head-marking languages such as Yana and Washo, which have no noun case system and few or no prepositions or postpositions, can distinguish between statements expressing ablative or elative directional motion ("away from" or "out of" a location or point of reference) vs. allative or illative directional motion ("to/ toward" or "into" a location or point of reference). The introduction summarizes the few brief existing accounts of Yana grammar, reviews the distinction between head-marking and dependent-marking morphology, and surveys how various more familiar dependent-marking languages such as English, Latin, Russian, Finnish and other Uralic languages, Inuktitut (double marking), and Ichishkíin express the ablative/elative vs. allative/illative distinction: with prepositional or postpositional noun phrases, with noun case suffixes, or with a combination of both. The paper then focuses on primarily head-marking Yana and Washo, while noting a couple minor examples of emergent dependent-marking morphology in Yana, possibly a result of areal influence of the Plateau Penutian languages Klamath and/or Ichishkíin in the Bipartite Stem Belt of Oregon and northern California. The paper analyzes Washo directional verb suffixes, while noting the apparent limitations and difficulties in expressing certain types of statements of directional motion with respect to a reference point other than the speaker. The hypothesis is raised that perhaps strictly head-marking morphology makes the expression of statements with such meanings more difficult in general, and thus perhaps in certain situations of language contact and areal influence, dependent-marking features are more likely to spread to head-marking languages than the other way around. The paper then focuses on the expression of directional motion in Yana, reviewing Leanne Hinton's excellent summary in her 1988 sketch of Yana morphology. The eight cardinal directional motion suffixes, an allative and ablative pair for each of the four cardinal compass directions, are highlighted as the most basic and frequent directional suffixes in Yana. The paper reviews these forms, analyzes examples of verb forms containing them, and finally reviews and analyzes numerous examples of verb forms with such directional suffixes in the context of their occurrence in Yana texts. In the course of this detailed analysis of text examples, certain oversights or apparent errors in the Yana Dictionary are found, explained, and corrected.

This paper presents 34 Plateau Penutian loanword etymologies, 32 of them borrowings from the Ichi... more This paper presents 34 Plateau Penutian loanword etymologies, 32 of them borrowings from the Ichishkíin (Sahaptin) language, of morphologically opaque forms in the lexicon of Yana, a language isolate of northern California, proposed to belong to the hypothetical Hokan language family, that went extinct almost a century ago. Little research has been done on Yana since Sapir's essential documentation and analysis of it a century ago, and apart from the efforts to identify Hokan cognates, no research has ever been done into the possible etymologies of other forms in the attested Yana lexicon. The paper describes the linguistic area of central and eastern Oregon and northern California, the languages of which exhibit the distinctive areal morphological feature of bipartite verb stems, as analyzed by Jacobsen, DeLancey, and Jansen. Plateau Penutian and northern California Hokan languages belonged to the core of this Bipartite Verb Stem Belt linguistic area, including in particular Klamath, Ichishkíin, Yana, Atsugewi, Washo, and quite possibly other languages as well. The paper explains how this historical linguistic analysis of the area implies that these languages were likely in contact with each other as recently as 500 to 2,000 years before present. Thus the plausibility of the presence of a substantial number of Ichishkíin loanwords borrowed into the Yana lexicon during this time period is established. Finally, the paper presents each of the 34 individual loanword etymologies, providing supporting analysis of phonological, morphological, and semantic details where necessary.

This paper presents evidence from internal comparison of Yana lexical forms to show that in numer... more This paper presents evidence from internal comparison of Yana lexical forms to show that in numerous cases, forms with an initial stop b- may actually be derived from roots with an original initial nasal m- as a result of a sequence of phonological changes: 1. loss of an unstressed vowel in one allomorph (CVCV- > CCV-), 2. denasalization m > b before another consonant in the CCV- form, and 3. analogical extension of the initial b- from the CCV- allomorph to the CVCV- allomorph. The model for this analysis is the even more elaborate sequence of changes proposed by Jacobsen in one example in his important 1976 analysis of the historical development of the Yana stop series: madu > *bdu(i) > (*badi >) pdi > padi. The paper presents 9 examples of pairs of Yana lexical forms of the shape CVC(C)(V)-, one with initial m- and the other with initial b-, that exhibit striking parallels both phonologically and semantically. It is proposed that in these cases the initial m- form was original and the initial b- form arose as a result of the process described above. The paper further presents 7 additional examples of such pairs of Yana forms in which one or both of the forms is of the shape CV-; it is argued that the same process still may have occurred in such forms due to loss of the vowel when unstressed before a following suffix with an initial consonant, in which case the same phonological environments for the sequence of changes could have been present. Finally, the paper concludes with observations about the implications of this analysis for Hokan etymology: many Yana forms with initial b- may be derived from Hokan roots with an original initial nasal *m- rather than from roots with an original initial stop *p-. The possible illustrative example of Yana bui- and the Hokan prefix #ma- 'with the foot' is discussed, to demonstrate both the possibilities and the complications involved in analyzing the etymology of Yana forms that may have arisen as a result of the process described in this paper.