Search

This paper presents a description and analysis of Algonquian verbal classifiers, morphemes that index a salient superordinate feature (such as substance, consistency, or shape) of an argument or no...

The Central Numic (Uto-Aztecan) dual number marking system on nouns and pronouns is of interest because even though most of the component morphemes involved in the system are reconstructible to Pro...

This paper examines the status of two languages spoken in the Andean highlands of Ecuador: Imbabura Quichua and Media Lengua. The latter is a mixed language consisting of Quichua morphosyntax, including all system morphemes, and Spanish-derived roots for all lexical items. Imbabura Media Lengua is currently spoken in three small indigenous communities, surrounded by other communities where only Quichua and Spanish are spoken. Previous descriptions of Media Lengua suggest that it may not be systematically different from Quichua but rather a last-ditch attempt by Quichua speakers to retain their language in the face of growing Hispanization. This paper provides the results of several interactive experiments conducted with Quichua–Media Lengua bilinguals, which demonstrate that the psycholinguistic boundaries between Quichua and Media Lengua are clearly delimited for most speakers. Only among the youngest speakers—most of whom have received Quichua classes in school—is there some blurring of the distinction ...

Boas (1917) mentioned several phenomena found in the morphosyntax of Amerindian languages (including nominal incorporation and lexical affixation) that posed challenges to accepted views of synchronic and typological issues raised by genealogical relatedness and areal convergence on the continent, morpheme classes like roots, stems, and affixes in languages in general, and whether the fundamental unit of human speech is the word or the sentence. A century later, and despite the increased amount of data available and a substantial amount of theorizing on morphosyntactic issues, these challenges have proved considerably recalcitrant. Framed in terms of morpheme types, word domains, and word classes, this article surveys the progress since Boas’s introductory paper and examines open questions in the study of the morphosyntax of indigenous languages of the Americas.

This paper describes a morphological change which is currently taking place in Kaqchikel, a Mayan language of Guatemala. Historically, verbal movement morphemes appear between the ergative and the absolutive pronominal prefixes (Abs—Mvmt—Erg—Verb Root). However, this paper documents a change in the movement construction wherein the movement morpheme appears next to the main verb root rather than between the pronominal prefixes (Abs—Erg—Mvmt—Verb Root). The use and distribution of these forms were examined in terms of sociological and geographic variables, as well as phonological ones. Results show that the current distribution of the new formation where the movement morpheme follows the ergative prefix is largely governed by phonological factors, namely, whether or not the morpheme break between the ergative and the absolutive aligns with a syllable boundary.

This paper uses synchronic and diachronic typological data to trace the origin and development of the Zapotec causative marker *k-. It is suggested that the causative marker *k- arose as a language-internal innovation after Zapotec split away from its sister branch, Chatino. The paper relies on Song’s (1996) typology of causative marker formation to show that the causative prefix *k- has developed out of the marker of the potential mood following de-subordination of the subordinate clauses of purpose. This work contributes to the field of Zapotecan linguistics, by tracing the evolution of an important pan-Zapotec morpheme, and to that of diachronic typology (e.g., Bickel 2007), by validating a proposed developmental sequence in the area of valence-related morphology.

Most Panoan languages have closed sets of about 30 monosyllabic forms that attach phonologically to the front of nouns, adjectives, and verbs. These forms, mainly designating body-part notions and semantic extensions of these, appear at first glance to be synchronically derived from polysyllabic body-part noun roots. This paper offers the first detailed study of the phonological, morphological, syntactic, and semantic properties of prefixation in Kashibo-Kakataibo. In contrast to other authors’ analyses of Panoan prefixes, we present evidence to show that prefixes in Kashibo-Kakataibo are synchronically independent morphemes, rather than allomorphs of body-part nouns. Subsequently, we present the diachronic scenarios that can account for the perplexing formal and semantic similarities between body-part prefixes and body-part nouns in Kashibo-Kakataibo

This paper addresses two verbal affixes in Kari’nja, the ky- prefix and the -ng suffix, and the constructions in which they appear. -ng appears in several constructions, while ky- appears only in what is called here “the ky- construction.” Hoff (1986) assigned the two affixes to a greater interacting system of evidentiality in the language that includes these two morphemes as well as a closed class of 17 evidential particles. The present account posits that ky- was, historically, a marker of remote deixis and -ng an indicator of “uncertainty” modality. However, in the ky- construction, those etymological semantic values have been bleached and the ky- construction has become conventionalized as the unmarked way of referring to non-Speech-Act Participant events in all but recent past tense. This analysis better reflects speaker insights, is more consistent with analyses of cognate morphemes of other languages in the family, and gives adult learners a more concrete description, aiding acquisition of this form.

In this paper, we undertake the first systematic exploration of pan-Cariban ablaut, a pattern of stem-initial vowel alternation in the paradigm of possessed nouns, inflected verbs, and inflected postpositions. We identify several related, but distinct, ablaut patterns and reconstruct them to sources in two Proto-Carib morphemes, *i- ‘3’ and *j- ‘Relational Prefix’. This then leads us to ervise prior reconstructions of certain personmarking prefixes in Proto-Carib, as well as many vowel-initial lexical items, a few *j- initial lexical items, and two Proto-Carib derivational morphemes. A number of these reconstructions appear to support prior claims of distant genetic relationship between the Cariban and Tupian families.

The Northern Je languages (Macro-Je stock) do not show clear patterns of splitergativity; however, a better understanding of their alignment pattern is possible, if we examine the earlier stages of these languages. This study reconstructs the ergatively organized nominalization system found in subordinate clauses in Proto-Northern Je, which occurred as the argument of intransitive verbs of modality. These nominalized clauses were reanalyzed in pre-Timbira as main clauses and also underwent a semantic change to finite constructions, expressing recent past tense. Synchronically, Timbira displays an ergative–absolutive alignment conditioned by the expression of recent past tense, whereas Apinaje, Me$bengokre, and Suya present the ergative alignment conditioned by the categories of tense, aspect, and modality. The former two categories are coded by certain words or morphemes in predicate position, while the latter is coded by intransitive verbs of modality, as in pre-Timbira.

Focus subtypes can be characterized according to scope and pragmatic (e.g., [non]‐contrastive) information. While the existence of strategies dedicated to code scope seems clear, the existence of grammar dedicated to pragmatic contrast is in dispute (Lambrecht 1994 and Myhill and Xing 1996). Some models propose that contrast is expressed in sentence grammar (Dik et al. 1981 and Watters 1979); others argue that contrast belongs to conversational implicature (Lambrecht 1994). Kokama‐Kokamilla, an Amazonian language, has constructions that clearly distinguish focus subtypes according to both parameters. More specifically, this study examines the distribution of the clitic =pura. While this morpheme is an important cue to identifying pragmatically marked sentences, by itself it cannot be associated with specific subtypes of focus. It is the interaction of =pura with constituent order and prosody that produces formal distinctions for coding narrow vs. broad and contrastive vs. noncontrastive focus.

The aim of this study is to enhance the documentation of laryngeal segments in Capanahua and to contribute to studies of the interaction between prosodic structure and segments. Laryngeal segments in Capanahua show a number of positional effects. They can only surface within morpheme‐initial syllables. This study shows that within that position, the distinction between head and nonhead syllables plays a major role in governing the alternations shown by glottal stops. Coda glottal stops can only surface in head‐syllables since they inhibit glottal coalescence, a process triggered elsewhere. In contrast, in nonhead syllables, glottal stops fuse with the preceding vowel, resulting in glottalized vowels. The underlying representations of morphemes can only contain one glottal feature; that is, either [constricted glottis] or [spread glottis] but not both. However, in Capanahua, this restriction does not hold at the level of phonetic representation since glottal stops and glottal fricatives co‐occur when the g...

The Yagua (Peba‐Yaguan) classifier system shows a number of structural isomorphisms with Bora‐Witotoan classifier systems. This is seen in the use of classifiers to derive new nouns from a variety of root types, and in what at first glance may appear to be agreement functions in some contexts. In both families, classifiers are required in numeral and demonstrative structures, and have similar discourse functions. However, there is actually little or no evidence that Yagua or Bora‐Witotoan languages have borrowed classifier morphemes from each other. Internal to Yagua, it is possible that classifiers arose via a compounding process. The available historical records show that Yagua and Bora‐Witotoan speakers have traversed the same geographic territory, leading us to conclude that there has been at least structural influence across language families.

This paper investigates the phonological status of the glottal stop that occurs in root morphemes in Wanano and in other Eastern Tukanoan (ET) languages spoken in the Vaupes region of Brazil and Colombia (Piratapuyo, Tukano, Desano, and Siriano). Roots in these languages are lexically marked for the suprasegmental features of nasalization and tone, features I describe, analyze, and contrast with what I propose to be a third suprasegmental feature of roots: glottalization. The glottal stops that occur regularly in a subset of roots have been alternately analyzed in the ET literature as a secondary feature of vowels, as a full consonant segment, or as an articulatory phenomenon linked to tone. I review these analyses and contrast them with a presentation of glottalization as a third, independent suprasegmental feature of roots. I argue that the suprasegmental approach is applicable to a wider range of data and therefore demonstrates clear explanatory advantages over previous analyses.

Matses, a Panoan language spoken in Amazonian Peru and Brazil, has a set of 28 monosyllabic forms, representing mostly body parts, which are phonologically attached to the front of verbs, adjectives, and nouns. All Panoan languages seem to have a similar set of morphemes, and there is some controversy about whether these are prefixes or incorporated nouns. Here, this phenomenon is exemplified in detail for Matses, and I argue that while this process might have evolved from noun incorporation, synchronically it should be considered prefixation. This distinction is important for the synchronic interpretation of Matses prefixation as product of grammaticalization and for a comparative‐historical perspective of these Panoan forms. Among the characteristics of Matses prefixation that are not consistent with noun incorporation is an applicative‐like function, where the prefix allows into the clause an extra constituent that is in a Figure–Ground relation with absolutive arguments.

In this article, I claim that Otomi has a nasal morpheme that serves as an exponent of middle voice. My data come from the dialect of San Ildefonso Tultepec, but a similar phenomenon can be observed in other dialects, so that the claims made here could be extended to Otomi as a diasystem. The study is based on a sample of 72 verbs that display the middle morpheme, a sample intended as a quantitative reference for the qualitative patterns found in the language. About two‐thirds of the verbs in this sample still show traces of lexical derivation. However, the derivative process has become unproductive and most middle verbs in Otomi function as independent units in the lexicon, showing just a morphological connection to their source verbs. My sample also contains cases where the source of the derivation was difficult, if not impossible, to reconstruct.

In Totonacan languages, the verbal prefix /laa‐/ ‘recip’ occurs on transitive and ditransitive verbs to mark a reciprocal relation between a subject and a direct or indirect object. In many of the languages, this prefix also occurs on regular, nonreciprocal verbs to mark subject/object combinations involving only local (first‐ or second‐person) participants. This paper describes the distribution of the prefix /laa‐/ ‘recip’ in Totonacan languages and explains this distribution in both reciprocal and nonreciprocal contexts on the basis of its semantics. We argue that /laa‐/ is historically a reciprocal morpheme which signals that the individual members of a group perform more than one thematic role in a transitive or ditransitive relation. It is this common reciprocal function that unites the diverse uses of /laa‐/ ‘recip’ in Totonacan and provides the basis for the extension of this prefix into other, nonreciprocal domains.

A partir du modele developpe par Smith (1991) pour l'analyse de l'aspect dans les langues du monde, l'A. examine un morpheme en salish spokane qui joue le double role de marqueur d'aspect et de controle. Ce morpheme, qui indique un accomplissement apres un effort supplementaire, est appele morpheme de reussite. En examinant les constructions transitives, intransitives et moyennes ou ce morpheme apparait, l'A. montre qu'il a une nature aspectuelle definissant un evenement duratif aussi bien que telic c'est-a-dire, entrainant les etapes anterieures qui conduisent au resultat naturel. L'A. etend ainsi la portee de la structure de l'evenement developpee par Smith et esquisse les relations entre aspect et controle

1. Peculiarities of the polysynthetic language type. The term "polysynthesis" generally refers to the phenomenon of high average number of morphemes per word, especially in verbs. In the recent literature a number of new quasi-synonyms of this term have appeared. The traditional notion of polysynthesis is usually reserved for characterizing the purely morphological type (Mithun 1986b). The concept of head marking coined by Nichols (1986) defines the morphosyntactic pattern of grammatical relations as a type of case marking. The term "pronominal argument languages" invented by Jelinek (1985) within the GB framework, but allowing a general usage as well, provides the syntactic characterization of the basic clause (earlier accounts of the same phenomenon include Boas 1911 and Van Valin 1977; cf. also the related concepts of nonconfigurationality [Hale 1983] and lexically projected languages [Saxon 1989]). It is not my purpose here to compare these different terms and the approaches that underlie them. It is equally clear that although the contents of all mentioned terms are not identical, the objects (languages) they characterize coincide to a significant extent. For this reason I have loosely used the term "polysynthesis" in the title of this paper as an overall label of language type having complex verb morphology, case marking on the verb, and verbal pronominal affixes. Thus the most crucial trait of polysynthetic languages is that all (at least all core) arguments are marked on the verb. According to the typology of Nichols (1986), which distinguishes between dependent-marking and head-marking languages, on the clause level strictly dependent-marking languages mark roles only on the arguments of the verb by means of case inflection:1

Introduction. The Teguima language, more commonly known as Opata, was formerly spoken in northern Mexico, in the modern states of Sonora and Chihuahua. With the Dohema (also known as Eudeve) language, it formed the Opatan linguistic grouping. Opatan is further grouped along with Tarahumaran (Tarahumara, Guarijio) and Cahitan (Yaqui, Mayo) into the Taracahitic subfamily of Uto-Aztecan on the basis of lexical (Miller 1984) and phonological (Lionnet 1985) evidence. Eventual classification of Opatan must, of course, take morphological and syntactic data into account. This paper presents the major features of Opata inflectional morphology.' Our knowledge of Opata stems largely from two related sources, both from the same author. Lombardo (1702) is the published redaction, made two years before the author's death, of a manuscript now in the Newberry Library (Lombardo, n.d.). Because a large portion of the information on verb morphology (Books II and III) survives only in the printed version, it is necessary to use this portion of the published version. Both sources are foliated; I refer to the printed version with the abbreviation p (= printed folio) and to the manuscript with the abbreviationf(= manuscriptfolio). Thus, the reference "plO4v" means "printed folio 104, verso." Note that the hyphens that separate morpheme boundaries do not appear in the original. I have added interlinear glosses and an English translation below each example sentence. For single morphological words, only an English gloss is given. Where there may be doubt about the reference of an Opata word or sentence, the Spanish gloss is given in parentheses. When Spanish is quoted, it is reproduced exactly as in the source; no attempt has been made to modernize the spellings found in the sources, but this should not

morpheme which marks past tense also marks irrealis in sentences which describe unreal situations. Consider the following, randomly chosen examples. About Garo, Burling notes, "The rather elusive uses of this suffix [cim] fall mainly into two sets, one to indicate action that is past and done, the other as the coordinate in 'if' sentences."2 In Chipewyan, Li notes that the past tense occurs "...sometimes with the meaning of contrary to fact when used with a conditional clause."3 Of Old Marathi, Master says, "The past tense is also used as a conditional...."4 Nonexotic English, too, partakes of the phenomenon. The English morpheme -ed indicates past tense: (1) Mary liked to roll drunks. But -ed can also indicate irrealis.

Apalai is a language of the Carib family, spoken in the northern regions of the Maicuru, Parf, and Jari Rivers of Paria, Brazil. In the last three decades, the tribe has been decimated by various diseases contracted through their contact with civilization. At present, there are between 150 to 175 speakers of Apalai. Extended contact with Apalai speakers was made possible through contracts held between the Summer Institute of Linguistics and the Ministry of the Interior, National Indian Foundation (FUNAI) and the National Museum, Rio de Janeiro. The analysis was facilitated by reference to a concordance of 28,000 morphemes from Apalai texts. The concordance was prepared at the University of Oklahoma Computer Laboratory under the Project for Computer Support for Linguistic Field Research which was partially supported by National Science Foundation grant GS-1605. I have collected data on the Apalai language during my ten years of residence in the tribal location. The present article was written as part of a seminar held under the auspices of the Summer Institute of Linguistics workshop in 1974 at Belem, Para, Brazil, under the direction of Ivan Lowe. Azuma (Az) and Coboclo Velho (CV), Apalai Indians from the Paru River, were invaluable language helpers during the gathering of data for this article. I wish to thank Ivan Lowe

0. Introduction. In recent work, Athapaskanists (e.g., Kari 1979 and Rice 1983) have proposed a system in which morphemes called conjugation markers occur in the imperfective, perfective, and future modes of the verb. In this model, the optative is the only mode that occurs independent of the conjugation markers. The question to be explored in this article is whether this treatment of the optative as different from the other modes is indeed the correct one. It will be shown that in Slave,2 an Athapaskan language of Canada, the optative patterns like the imperfective and perfective modes in that it too combines with the different conjugation markers. In 1, an overview of the Slave verb is given. The form of two of the conjucation markers, *nand *s-, in the imperfective and perfective is discussed in 2. In 3, it is shown that if the optative is treated as containing conjugation marking, problems in accounting for variation in the form of the optative are easily solved. Finally, in 4, the implications for a model of the verb in Athapaskan in general are explored.

0. Introduction. In two recent papers, Goddard (1981b; 1983) attempts to refute the model of Algonquian linguistic prehistory I began presenting earlier (Proulx 1980a; 1980b) and to defend his own more traditional views. The points he raises are many and varied, and counterarguments will be presented (Proulx mss.). However, there is one key point that turns up in both papers, and which is sufficiently complex and important to warrant a full discussion of its own: the reconstruction of PA *aye.' My position is that there was a PA sequence *aye, reconstructible chiefly across morpheme boundaries, whose reflexes coincide with those of *a: in most of the non-Eastern languages and with those of *e: in some cases in the Eastern ones-but which gives underlying i: in Menominee and remains uncontracted in some cases in (some of) the Eastern languages. The sequence is crucial because the reconstruction of the PA subordinative order depends on it, and this reconstruction in turn weakens the theory that the Eastern Algonquian languages form a genetic subgrouping within Algonquian. To defend his PEA hypothesis, Goddard tries to disprove the reconstruction of the subordinative order by refuting the reconstruction of PA *aye. He procedes by (1) trying to discredit the etymologies supporting *aye, and (2) claiming that a PA sequence *aye could not exist because allegedly *ay-e contracts to PA *e:.

also to trace the history of the innovative changes that have taken place in the daughter languages. The innovations are treated under two rubrics: replacements, the more radical changes in which new stems have been substituted for those derived from the parent language; and accretions, the secondary changes in which new morphemes have been added to the stems. The independent pronoun paradigm offers a

0. In this article, I present an interesting morphophonemic phenomenon in Colville, an interior Salishan language of eastern Washington. Then I discuss similar phenomena in related languages and draw some diachronic inferences. The phenomenon in Colville is one I call "long-distance conditioning," where some element in the root interacts with an element in the suffix at a distance of several phonemes and often one or more morphemes.

0. Introduction. Tuyuca evidentials are a mandatory feature of the independent verb, combining information about person and tense with statements about the evidence on which the report is based. Tuyuca speakers must indicate that they obtained their information in one of the following ways: visually; through a sense other than the visual; through evidence of the state or event; by being told about the state or event; or by assuming what happened. Moreover, these five evidential categories are arranged hierarchically according to speaker preference. The following discussion demonstrates the use of these evidential categories, as well as describing a feature analysis of them. First, however, the reader deserves a brief introduction to the nature of the Tuyuca language. Tuyuca is the mother tongue of approximately 700 people who live along the Papuri and the Tiquie rivers and their tributaries in Colombia and Brazil. It is classified by Castellvi (1962:241-43) as a Tucanoan language and by Waltz and Wheeler (1972:128) as a member of the Central Eastern Tucanoan branch of that language family. The Tuyuca phonemes are /p t k b d g s r w y h ii u e a o/ plus /~/ (nasalization) and / '/ (stress). Syllable types are V and CV. Nasalization in the Tucanoan languages does not operate solely on the vowel, but on the syllable or morpheme, with the result that [m] is an allophone of /b/, [n] of /d/, [j] of /g/, and [f] of /y/. In Tuyuca, nasalization is a feature of the morpheme, so that every morpheme is either entirely nasalized or entirely oral. Stress is concurrent with high pitch and occurs once per word. The basic word order in Tuyuca is Subject-Object-Verb, but other orderings occur as discourse considerations demand. Short sentences of

I propose to summarize a number of findings from recent fieldwork and analysis which suggest that the evidence offered in support of the "classical" Na-Dene hypothesis (i.e., as set up by Sapir in his 1915 statement) is spurious, and that there is currently no empirical basis for including Haida in the Na-Dene grouping.' Writers on the Na-Dene hypothesis have in the past relied heavily on the significance of supposedly shared grammatical features; hence much of the debate, so far as Haida is concerned, hinges on the accuracy of Swanton's 1911 grammar for the Handbook of American Indian Languages.2 This is unfortunate, because much of Swanton's analysis contains major flaws, for example, inaccurate segmentation of individual morphemes and confusion of homophonous suffixes. The principal short

phonological analysis whereby v can be accounted for as a variant of p, even synchronically, so that the morpheme -vu can be recognized as a reduced unstressed variant of the demonstrative pu' that one, he. This can only be done by recognizing the unique syntactic environment in which v occurs, since there are two other suffixes in the language with underlying -p which always surface as p.25 This is stated in rule 3 by the restriction that the shape v occurs only at the boundary between a stem and the remaining suffixal material, obviously a boundary different from the ordinary stem-forming boundary or, for that matter, from a root boundary. The following set of contrasting sentences will illustrate the facts. ?oxomay I look for him. ( first-person subject, xomay look for [verb stem]); ?oxu'mayp We look for him. (? first-person subject, xu'mayp [plural verb stem]); ?3xamay-vu mowu'w You saw the one I was looking for. (I-look-for-him-specific-(object case) yousee); 2axu'mayp-vu mowu'w You saw the one we were looking for. Note also that the boundary between the stem and the suffix must be different from word boundary as is clear from the minimal pair: ?i'kwic pu'i that man (man that-subject); ?i'kwicvoc the man (man-specific-subject). 1.2.2. The above discussion has established the need for the additional stem boundary in Dieguefio, as a prelude to a presentation of another aspect of the phonological history of Yuman languages, that is, some peculiarities of the phonetic representation of some of the case morphemes, in particular the subject case marker, for which I propose the reconstruction *c. As described in the Yuman 25 Langdon, A Grammar of Diegueno, pp. 97, 115-17. literature,26 the shape of the subject marker is as follows in the various lan

1.0. The task proposed for this article involves a statement of certain methodological principles for the analysis of Aleut morphology, and an illustration of these principles by the presentation of Aleut sequences according to them. The methodological interest is primary; no definitive statement of Aleut is attempted. All the linguistic data on which this discussion is based is contained in a grammar of the Aleut language published in 1944.1 Materials for restatement of Aleut morphology were extracted principally from the extensive paradigmatic lists of verb and noun forms as printed in this grammar. In it utterances longer than single words are limited to a version of the Lord's Prayer, two songs, and a few random sentences. This makes syntactic analysis as discussed below impracticable here. The kind of analysis followed is one based entirely on distributional criteria. Meaning was not considered either in determining morpheme boundaries or in defining morpheme classes.2 Morphemes were identified by means of investigating the occurrence of sequences longer than one phoneme as they

0. We group Nahuatl words into three categories: nouns, verbs, and particles. Words of each category contain a word core, consisting of one or more morphemes, which we designate as a stem: noun stem (N); verb stems-intransitive (I), transitive (T), benefactive (B); particle stem (P) Some I and all P occur respectively as I and P. N, T, B, and other I occur only in sequence with one or more affixes (prefixes and suffixes).

in 4.2 and shown in detail in 8.5. Analysis of attributive stems is given in 8.1-8.4. Attributive stems (designated At) constitute a minor stem class related to, but different from, noun stems (N); attributives (designated Atr) constitute a minor word class related to, but different from, nouns (Nn). 8.1. Attributive stems may consist of one morpheme (Atl), two morphemes (At2), or three morphemes (At3). 8..1. Attributive stems consisting of one morpheme (At') may be monosyllabic, dissyllabic, or polysyllabic. Monosyllabic At' stems have the form CVC or CVCC. Examples show stems with form CVC ending in y: kay narrow, pay long. Other examples show stems with form CVCC ending in a cluster ?c, ks, hk: tA?c dry, ciks little, yAhk(-) black. Dissyllabic At' stems may end in a vowel in the form CVCV, CVCCV, or CCVCV. Examples are: home new, huyu round, popo white, kacu sour, malo bad, suni pretty, pA?a humid, ?oye good; ho?ma gray, cimbu short, sa?sa beautiful; klaru clear. Dissyllabic At' stems may also end in a consonant in the form CVCVC or CVCCVC. Examples show stems ending in c, h, y, n, j: hemec heavy, hAnAh light or quick, ponih slow, takay bitter, cikin spotted, ?usaB little; cuhcuh green, mAskuy ugly. Dissyllabic At' stems may also end in a cluster of two consonants in the form CVCVCC. Examples show stems ending in clusters ks, yh, hc, hs: 6AlAks little, 5okoyh slowly, panahc(-) slippery, capahs(-) red. Polysyllabic At' stems (three syllables or more) include mostly loan words. Examples: ?onrado honest, moderno modern. 8.1.2. Attributive stems consisting of two morphemes (At2) are of various types. Examples are shown here. Stems in which 50-70 + At' or N1 = At2: ?a?Puma tongue-tied (?ai= 51; Puma dumb = At'); ko?sawa crazy (ko?= 52; sawa air = N1); ni?oye pretty good (ni= 61; ?oye good = At'); saiyaci very bad (sa)= 71; yaci bad = At'). Stems in which VI + 300 = At2: sose cooked (sohsto cook = V1; -e = 314). Stems in which At' + At' = At2: yawapu?CA rose-colored (yawa tender = At'; pu?cA yellow = At'). Stems in which At' reduplicated = At2: ponihponih very slowly (ponih slowly = At'). 8.1.3. Attributive stems consisting of three morphemes (At3) are of various types. Examples are shown here. Stems in which 70 + At2 = At3: sagdome very near (san= 71; tome near = At2). Stems in which N2 + 300 = At3: na?Pat frightful (na?cifear = N2; -?ai = 313). 8.2. Prefixes included in attributive stems are numbered 51, 52, 61, 71. Prefix 51 ?ao(mouth, etc.) may precede the At' stem ?uma dumb to give the At2 stem ?a?Puma tongue-tied. In formula: 51 + At' = At2.

s in -va. *ay6m to go away > ?ay6mxava ?a?mani'k (object-of-my-future-going-away I-do not know) I do not know where I am to go? d. Type (d) contains only three examples: a566y his character, nature (as judged by appearance) acma?6y your state of mind, your record (of things you have said). 66. Interjectional themes are of exclamatory or adverbial meaning. Exclamatory: Am well! oh, well!; 6?e alas!; m6well; x6 oh!; xw6't disgust, astonishment; y6'xa come on! let's go! courage!; ?anY6p6? lucky me!; ?aya?aya. fright, disgust (used by women); ?awa?aw anger, disgust (used by women). Adverbial: tanay yesterday; mapisa now, at present; n4i-ri a?6tnyak, mapisa ?apAlytam ?ayd'k There used to be none (but), now I see that they are numerous; nyi?ak6r long ago; nyi?ak6r saklinyk vi-y6mak Long ago he ran away. 166 VOL. XIII This content downloaded from 207.46.13.85 on Fri, 18 Nov 2016 04:14:22 UTC All use subject to http://about.jstor.org/terms

0. A componential approach to the semantics of morpheme classes has proved useful in handling some elusive semantic problems of Tarascan morphology, particularly in the assignment of referential meanings to verbal thematic suffixes.2 The theoretical relevance of componential analysis has elsewhere been demonstrated chiefly with lexical material, particularly in the domains of kinship3 and folk science.4 That

2..1.1 Walapai has three major morpheme classes: verbs, nouns, and particles. Divisive for verbs are the closing verb suffixes /-wi/ and /-yu/, e.g., /makwauTiwi/, you are speaking, is marked as a verb by the suffix /-wi/, and /kwiJiOikyu/, it is soft, is marked as a verb by the suffix /-yu/. /-wi/ and /-yu/ are weak-stressed by-forms of /wi/, do, and /yu/, be, respectively. These weakstressed by-forms occur only as verb suffixes. 2.1.2. A noun may be formally defined as a word which may occur with the oblique case endings plus or minus the definite suffix /-a/, but with no other affixes, e.g., in /wilak/, by/toward the bush, /wilal/, in(to) the bush, and /wilam/, (away) from the bush, /wil/, bush, plant, is marked as a noun by the definite suffix /-a/ and the oblique case endings /-k/, /-1/, and /-m/ since there are no other suffixes. Verbs also have suffixes

0. It seems to be generally agreed that the translation of this or of any other list of isolated words or glosses into another language is almost sure to present some problems, the difficulty of whose solution should be neither underestimated nor exaggerated. And since the nature and number of such problems differ from one target language to another, the exposition of those that are encountered in any particular case may have a more general interest. On the other hand, comparison of basic vocabulary from several Arawakan languages reveals a disconcerting number of partly cognate equivalents; under which term I shall include not only pairs of compounds containing both cognate and noncognate members, but also any compound or one-time compound and such of its simple equivalents as are cognate with one member -cf., meaning tail: Bar6 ihibi, Lokono ihi and Guinao -ibi. Clearly, in such cases, imperfect understanding of the morphology, uncertainty as to a particular morpheme's relevance to the comparison, or even ignorance of a word's history may lead to distorting the count of 'glottochronologic sames.'

1 Araucanian is a native Southamerican language spoken by about 200,000 people in Chili, and by an only roughly calculated but far lesser number of speakers in Argentina (more or less 8,000). They are concentrated chiefly in the provinces of Bio-Bio, Malleco, and Cautin (Chile), and in the province of Neuqu6n in Argentina. 2 Rodolfo Lenz, Estudios Araucanos, Santiago de Chile, 1895-1897 (reprint with different paging from Anales de la Universidad de Chile, vols. 9098, 1895-1897. I quote from the reprint). I have made use of Lenz's other studies on the Araucanian language, especially the one appeared in Biblioteca de Dialectologia Hispanoamericana 6.234244 (Buenos Aires, 1940). Lenz was a distinguished phonetician; for data on his scholarship see Amado Alonso, Biblioteca de Dialectologia Hispanoamericana 6.269-278. 3 Lenz recognizes among the Araucanians from Chili four linguistic groups: Picunche, Huiliche, Pehuenche, and Moluche. This division agrees with the tribal classification by John M. Cooper in Handbook of Southamerican Indians 2.690 (1946), where the Moluche is called Mapuche. But since the differences between Pehuenche and Moluche are small, and only lexical, Lenz seems to consider them as two subdivisions of the same dialect (cf. 0.2. The corpus consists of 345 rather brief utterances in conversational form, 37 traditional tales, 7 descriptive narratives, and 15 songs. The transcription is phonetic. Each symbol is in general accurately and unambiguously explained, except perhaps a couple of them that need some interpretation as to their exact phonetic value.4 Lenz was consciously aware, in most cases, of free and conditioned variants-physiological, and phonetically conditioned variants, in his own terms-so that an approximate phonemic analysis of segmental phonemes is already implicit in his remarks. Nevertheless, the phonemic status of certain sounds remains somewhat dubious owing chiefly, I believe, to two circumstances: (1) sounds of low frequency, and morphemes of few occurrences seem not to have been sufficiently rechecked; (2) the transcription reflects, rather undiscriminately, incidental forms of careless speech, and divergencies in pronunciation probably due exclusively to differences in tempo. The latter will probably have to be treated as morphophonemic alternations, so that ultimately they cannot affect the phonemic analysis. More regrettable is the fact that features of stress, intonation, and juncture are either entirely disregarded or not satisfactorily described. 0.3. The present paper attempts to supply an explicit and more consistent phonemic formulation of Lenz's materials. Given the

1. Underlying pattern. Many morphemes in Macushi occur in alternate forms, the alternation depending upon the occurrence of certain vowels in some alternants which are absent in others. For example, the utterance wnamri mirror contains one morpheme in isolation. When the same morpheme occurs in an utterance with the prefix and suffix indicating first person, singular, alienable possession, it becomes wanmarri my mirror. Thus the morpheme

0. There has been considerable interest in the application of statistical techniques to the study of language vocabulary, variously called glottochronology and lexico-statistical dating. It was first established on the basis of counts from 13 languages that 'on the average about 81% of the basic-rootmorphemes of a language will survive as cognates after 1000 years." This implies that two related languages, 1000 years after they have split off from the parent language, will have 66 % of cognates from the basic morpheme list, that is, each will retain 81 % of the basic list and the overlapping of cognates will be 81 % of 81 %, or 66 %. It is this principle that is used by Swadesh in dating Salish relationships,2 although, since he does not use the same list that was used to estab

6.0. Arapaho verb forms are underlain by roots and/or by stems generated from the roots and are affixed by one or several of a large number of minor morphemes associating with the verb stems, e.g., hiineniinin66 I am a man (cf. hin6n man), b6neesei06noo I am big, bensesifnenininoo I am a big man; nenfiiehfinin you are a bird (cf. nii?6hii bird); neniit6woot6woo I hear it or them (inanimate goal), neniftono? I hear

positional slot filled by stems and affix-stem sequences which make up, by virtue of their occurrence in this slot, the class of themes. This makes it possible to avoid such statements as: a (noun) stem plus a (verb) stem equals a (noun) stem, and to say, rather, that a (noun) stem plus a (verb) stem equals a (noun) theme. Instead of saying that a (verb) stem in association with a prefix equals a (noun) stem, we say that a (verb) stem plus the prefix equals a (noun) theme. Morphemes thus remain uniquely classified as noun stem, verb stem, particle stem, prefix, or suffix. Certain combinations of these only enter into the various kinds of thematic formation indicated. 3.1. The positional slot of a theme may be identified in terms of its boundaries, the immediately preceding or immediately following morphemes or phonemes, which

it (Perfective) verb stems, like other The classifier which occurs immediately stems, may be immepreceding the verb stem assimilates to the )y one of the four morpreceding vowel as in (b) and (c) or is lled 'classifiers'. What deleted after a consonant or a geminate lis paper is the phonovowel sequence as in (a) and (d) respecn of two Sarcee classitively. (Tone sandhi is not included in this flexes of the Athapaskan discussion.) The morphophonemic alternations observed above are quite straight1 corresponds in Sarcee forward. Within the traditional phonemic pending on the environframework, either morph (i or a) may be Lg data illustrate the dischosen as the base form from which alloomorphs: morphs are derivable. Alternatively, one dal I will eat might prefer to represent the morpheme by mat (berries) berries a morphophoneme (e.g. I) which may be dal, you interpreted as a set of allomorphs, each dat you will eat er) eat berriles of which is to be chosen in a proper context er) eat berries by a morphophonemic rule. dal someone will Notice that the choice of either i or a as eat berries the base form is arbitrary. Assuming that i is chosen as the base form, we may conaadAl we have eaten struct the following three rules in order to b erries account for the morphophonemic alternawe eat (Perfect)ions: tions'

18. Before attempting to present in outline what appear from my texts as the salient features of this dialect's syntax, I may state briefly how I intend to use the terms WORD, PHRASE, CLAUSE and SENTENCE. By WORD will be meant, in general, a morphemic word, whether completely free (one which may, together with an appropriate intonation, constitute an utterance), such as naibuga I'm going or aueretu (served right is she) it serves her right, or clausally or phrasally bound, such as ma (a negative particle usually replacing mama in questions and rhetorical questions when the affirmative is assumed) in lubuidi barana ma isn't the sea beautiful? or ma-buga n' -bn didn't I tell thee? Those of the former type are taken to be single words when analysis shows them to contain, besides a root morpheme, bound or free, only recognized affixes in regular distribution with this morpheme. Those of the latter type are assumed to be words (rather than affixes or bound roots) when, as in the case of ma, they show both considerable mobility within the clause or sentence and freedom in their association with other forms belonging to different classes. Nor shall I hesitate to call such an enclitic as bai in fi-bai blow it (b2nd person sg., -aconjunctive auxiliary verb theme, -i 3rd person sg. masculine) a word; since it should be clear from what has been said under section 3 that person-marker prefixes always, and person-marker suffixes

0.1. The following text, recorded in the summer of 1961 from a bilingual Hopi woman from Shipawlavi, contains a large number of English stems as loan words utilized with Hopi morpheme suffixes. The informant, who is approximately fifty-six years old, has lived in Shipawlavi all her life except for a four year period in the 1920s when she was a student at the Sherman Institute, Riverside, California and for a period of three years during World War II when she lived in Winslow, Arizona and Barstow, California. Other texts recorded from the same informant in 1934 and 1938 show none of

We wish to credit Brent Berlin with getting us interested in problems of nomenclature which until now we have largely ignored in favor of lexical/semantic fields. The two interests, we find, composed of single morphemes. Extension of anatomical terms appears to procede always from the body to other cultural domains. It is not our intention to investigate in any detail the anatomical domain of English. However, similarly casual investigation of the Navajo anatomical terminology indicates a different situation. Therefore, (1) looking at the domain of the origin and the extensions of Navajo anatomical terms restricted in this paper to the 47 Navajo customary terms of the foot,2 is our first goal. This task is preparatory to the second, main goal, (2) to refine the universal propositions of principles of nomenclature by