Search

Murrinhpatha (non-Pama-Nyungan, Australia) is typologically unusual in having a single applicative construction with the semantics of source/malefactive, but never benefactive. In this paper I discuss the development of this applicative from an incorporated body part meaning ‘hand’. I show that the applicative developed from a reanalysis of the external possession construction; and that the applicative morphology developed from the incorporated body part, rather than from a verbal or adpositional source. This contributes to our understanding of the typology of applicative constructions and also highlights the value in exploring the complex verbal constructions of polysynthetic languages to inform our understanding of grammaticalisation possibilities.

Low-resource polysynthetic languages pose many challenges in NLP tasks, such as morphological analysis and Machine Translation, due to available resources and tools, and the morphologically complex languages. This research focuses on the morphological segmentation while adapting an unsupervised approach based on Adaptor Grammars in low-resource setting. Experiments and evaluations on Inuinnaqtun, one of Inuit language family in Northern Canada, considered a language that will be extinct in less than two generations, have shown promising results.

This paper reviews the definitions and operationalisations of the notion of “polysynthesis” proposed in the typological literature and applies them to Lithuanian (verbal) morphology. It is shown that while Lithuanian falls short of polysynthesis in terms of morphemes-to-words ratio and lacks such features as polypersonalism and incorporation, it still possesses certain properties associated with polysynthesis. These include the so-called lexical affixes (it is argued that Lithuanian verbal prefixes are an example of these) and, to a limited extent, “productive non-inflectional concatenation”, i.e. morphemes with a high combinatory potential and compositional meanings, including successively applying verbalisers and nominalisers. These observations not only shed novel light on some well-known facts of Lithuanian grammar, but also further underscore the multifactorial and vague nature of polysynthesis.

This paper describes the development of the first Universal Dependencies (UD) treebank for St. Lawrence Island Yupik, an endangered language spoken in the Bering Strait region. While the UD guidelines provided a general framework for our annotations, language-specific decisions were made necessary by the rich morphology of the polysynthetic language. Most notably, we annotated a corpus at the morpheme level as well as the word level. The morpheme level annotation was conducted using an existing morphological analyzer and manual disambiguation. By comparing the two resulting annotation schemes, we argue that morpheme-level annotation is essential for polysynthetic languages like St. Lawrence Island Yupik. Word-level annotation results in degenerate trees for some Yupik sentences and often fails to capture syntactic relations that can be manifested at the morpheme level. Dependency parsing experiments provide further support for morpheme-level annotation. Implications for UD annotation of other polysynthetic languages are discussed.

Definition modelling is the task of automatically generating a dictionary-style definition given a target word. In this paper, we consider cross-lingual definition generation. Specifically, we generate English definitions for Wolastoqey (Malecite-Passamaquoddy) words. Wolastoqey is an endangered, low-resource polysynthetic language. We hypothesize that sub-word representations based on byte pair encoding (Sennrich et al., 2016) can be leveraged to represent morphologically-complex Wolastoqey words and overcome the challenge of not having large corpora available for training. Our experimental results demonstrate that this approach outperforms baseline methods in terms of BLEU score.

St. Lawrence Island Yupik (ISO 639-3: ess) is an endangered polysynthetic language in the Inuit-Yupik language family indigenous to Alaska and Chukotka. This work presents a step-by-step pipeline for the digitization of written texts, and the first publicly available digital corpus for St. Lawrence Island Yupik, created using that pipeline. This corpus has great potential for future linguistic inquiry and research in NLP. It was also developed for use in Yupik language education and revitalization, with a primary goal of enabling easy access to Yupik texts by educators and by members of the Yupik community. A secondary goal is to support development of language technology such as spell-checkers, text-completion systems, interactive e-books, and language learning apps for use by the Yupik community., Comment: ComputEL-4

This paper describes a three-year project at the National Research Council of Canada aimed at developing software to assist Indigenous communities in their efforts to preserve their languages and extend their use. The project aimed to work within the empowerment paradigm, where the linguistic goals of communities have at least equal weight with those of the researchers, and where collaboration with communities is central. Because many of the technological directions we took were in response to community needs, the project ended up as a collection of diverse subprojects, including the creation of a sophisticated framework for building verb conjugators for highly inflectional polysynthetic languages (a verb conjugator for Kanyen’kéha, in the Iroquoian language family, was built in the framework), release of what is probably the largest available corpus of sentences in a polysynthetic language (Inuktut) aligned with English sentences and experiments with machine translation (MT) systems trained on this corpus, free online services based on automatic speech recognition (ASR) for easing the transcription bottleneck for recordings of speech in Indigenous languages (and other languages), limited-domain text-to-speech synthesis for some Indigenous languages, and several other subprojects., Ce rapport technique décrit un projet de trois ans au Conseil national de recherches du Canada pour le développement de logiciels visant à soutenir les communautés autochtones dans leurs efforts de préservation et de revitalisation de leurs langues. Ce projet s’efforce de respecter le principe de l’autonomisation : l’importance accordée aux objectifs linguistiques des communautés est supérieure ou égale à celle accordée à ceux des chercheurs et la collaboration avec les communautés est centrale. Puisque beaucoup des orientations technologiques ont été choisies en réponse aux besoins des communautés, le projet a donné lieu à divers sous-projets, notamment : la création d’une plateforme pour créer des conjugueurs pour les langues polysynthétiques hautement flexionnelles (un conjugueur pour la langue iroquoienne kanyen'kéha a été développé à l’aide de cette plateforme); la publication d’un corpus bilingue inuktut–anglais, qui est probablement le plus grand corpus de phrases disponible pour une langue polysynthétique; un projet d’expérimentation de la traduction automatique entrainée sur ce corpus; des services en ligne gratuits basés sur la reconnaissance de la parole automatique pour soulager le goulot d’étranglement de la transcription des enregistrements en langues autochtones; un projet sur l’utilisation de la synthèse vocale à domaine restreint pour certaines langues autochtones; et plusieurs autres sous-projets.

Polysynthetic languages have exceptionally large and sparse vocabularies, thanks to the number of morpheme slots and combinations in a word. This complexity, together with a general scarcity of written data, poses a challenge to the development of natural language technologies. To address this challenge, we offer linguistically-informed approaches for bootstrapping a neural morphological analyzer, and demonstrate its application to Kunwinjku, a polysynthetic Australian language. We generate data from a finite state transducer to train an encoder-decoder model. We improve the model by “hallucinating” missing linguistic structure into the training data, and by resampling from a Zipf distribution to simulate a more natural distribution of morphemes. The best model accounts for all instances of reduplication in the test set and achieves an accuracy of 94.7% overall, a 10 percentage point improvement over the FST baseline. This process demonstrates the feasibility of bootstrapping a neural morph analyzer from minimal resources.

Text input technologies for low-resource languages support literacy, content authoring, and language learning. However, tasks such as word completion pose a challenge for morphologically complex languages thanks to the combinatorial explosion of possible words. We have developed a method for morphologically-aware text input in Kunwinjku, a polysynthetic language of northern Australia. We modify an existing finite state recognizer to map input morph prefixes to morph completions, respecting the morphosyntax and morphophonology of the language. We demonstrate the portability of the method by applying it to Turkish. We show that the space of proximal morph completions is many orders of magnitude smaller than the space of full word completions for Kunwinjku. We provide a visualization of the morph completion space to enable the text completion parameters to be fine-tuned. Finally, we report on a web services deployment, along with a web interface which helps users enter morphologically complex words and which retrieves corresponding entries from the lexicon.

Indigenous languages have been very challenging when dealing with NLP tasks and applications because of multiple reasons. These languages, in linguistic typology, are polysynthetic and highly inflected with rich morphophonemics and variable dialectal-dependent spellings; which affected studies on any NLP task in the recent years. Moreover, Indigenous languages have been considered as low-resource and/or endangered; which poses a great challenge for research related to Artificial Intelligence and its fields, such as NLP and machine learning. In this paper, we propose a study on the Inuktitut language through pre-processing and neural machine translation, in order to revitalize the language which belongs to the Inuit family, a type of polysynthetic languages spoken in Northern Canada. Our focus is concentrated on: (1) the preprocessing phase, and (2) applications on specific NLP tasks such as morphological analysis and neural machine translation, both for Indigenous languages of Canada. Our evaluations in the context of lowresource Inuktitut-English Neural Machine Translation, showed significant improvements of the proposed approach compared to the state-of-the-art.

Morphological segmentation of words is the process of dividing a word into smaller units called morphemes; it is tricky especially when a morphologically rich or polysynthetic language is under question. In this work, we designed and evaluated several Recurrent Neural Network (RNN) based models as well as various other machine learning based approaches for the morphological segmentation task. We trained our models using annotated segmentation lexicons. To evaluate the effect of the training data size on our models, we decided to create a large hand-annotated morphologically segmented corpus of Persian words, which is, to the best of our knowledge, the first and the only segmentation lexicon for the Persian language. In the experimental phase, using the hand-annotated Persian lexicon and two smaller similar lexicons for Czech and Finnish languages, we evaluated the effect of the training data size, different hyper-parameters settings as well as different RNN-based models.

The adaptation of neural approaches to NLP is a landmark achievement that has called into question the utility of linguistics in the development of computational systems. This research proposal consequently explores this question in the context of a neural morphological analyzer for a polysynthetic language, St. Lawrence Island Yupik. It asks whether incorporating elements of Yupik linguistics into the implementation of the analyzer can improve performance, both in low-resource settings and in high-resource settings, where rich quantities of data are readily available.

Polysynthetic languages pose a challenge for morphological analysis due to the root-morpheme complexity and to the word class “squish”. In addition, many of these polysynthetic languages are low-resource. We propose unsupervised approaches for morphological segmentation of low-resource polysynthetic languages based on Adaptor Grammars (AG) (Eskander et al., 2016). We experiment with four languages from the Uto-Aztecan family. Our AG-based approaches outperform other unsupervised approaches and show promise when compared to supervised methods, outperforming them on two of the four languages.

Morphological analysis is a critical enabling technology for polysynthetic languages. We present a neural morphological analyzer for case-inflected nouns in St. Lawrence Island Yupik, an endangered polysythetic language in the Inuit-Yupik language family, treating morphological analysis as a recurrent neural sequence-to-sequence task. By utilizing an existing finite-state morphological analyzer to create training data, we improve analysis coverage on attested Yupik word types from approximately 75% for the existing finite-state analyzer to 100% for the neural analyzer. At the same time, we achieve a substantially higher level of accuracy on a held-out testing set, from 78.9% accuracy for the finite-state analyzer to 92.2% accuracy for our neural analyzer.

Morphological segmentation for polysynthetic languages is challenging, because a word may consist of many individual morphemes and training data can be extremely scarce. Since neural sequence-to-sequence (seq2seq) models define the state of the art for morphological segmentation in high-resource settings and for (mostly) European languages, we first show that they also obtain competitive performance for Mexican polysynthetic languages in minimal-resource settings. We then propose two novel multi-task training approaches -one with, one without need for external unlabeled resources-, and two corresponding data augmentation methods, improving over the neural baseline for all languages. Finally, we explore cross-lingual transfer as a third way to fortify our neural model and show that we can train one single multi-lingual model for related languages while maintaining comparable or even improved performance, thus reducing the amount of parameters by close to 75%. We provide our morphological segmentation datasets for Mexicanero, Nahuatl, Wixarika and Yorem Nokki for future research., Comment: Long Paper, 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

This chapter reports on initial findings of an ongoing large-scale research project into the acquisition of Murrinhpatha, a polysynthetic language of the Daly River region of the Northern Territory of Australia with complex morphology. The complex verbal structures in Murrinhpatha, which can contain a large number of morphemes and bipartite stem morphology discontinuously distributed throughout the verbal template, raise a multitude of questions for acquisition. In this chapter we focus particularly on the acquisition of the complex predicate system in the verb, and the acquisition of subject-marking categories and tense/aspect/mood. Our findings are based on the language development of five Murrinhpatha acquiring children aged from 2;7–4;11 years.

This chapter describes what happens at the structural level to polysynthetic languages during language obsolescence, attrition, and loss. The changes that take place in decaying polysynthetic languages should be distinguished from (a) those occurring in all obsolescent languages regardless of their type, and (b) changes in “healthy” polysynthetic languages. It is shown that the consequences of polysynthetic language decay are primarily manifested in the collapse of morphological complexity, involving the loss of morphological ‘slots’, the reduction in the number of bound morphemes and their substitution by free ones, the ‘fossilization’ of markers and their reanalysis, the deprivation of word formation productivity, the destruction of noun incorporation, and reduction of allomorphy.

Purepecha (language isolate, Mexico) has one relevant characteristic that leads to identifying it as a polysynthetic language: productive verbal morphology (in particular locative suffixes). Purepecha is a predominantly dependent-marking language, as its pronominal markers are enclitics, generally second position enclitics. But, in some contexts Purepecha shows head-marking characteristics. Today, pronominal enclitics exhibit variation, tending to move to the rightmost position in the clause; they may encliticize to the predicate itself, showing a head-attraction or polypersonalism strategy and making Purepecha more polysynthetic. But this language lacks noun incorporation. Purepecha has three types of non-finite clause: two subordinate clauses (non-finite complement clauses and purpose clauses) and a syntactically independent clause (the chain-medial clause). This seemingly inconsistent situation (characterized by a correlation of different properties, some of which have not been identified as polysynthetic) calls for addressing the typological classification of Purepecha among the polysynthetic languages.

One of the major challenges in acquiring a language is being able to use morphology as an adult would, and thus, a considerable amount of acquisition research has focused on morphological production and comprehension. Most of this research, however, has focused on the acquisition of morphology in isolating languages, or languages (such as English) with limited inflectional morphology. The nature of the learning task is different, and potentially more challenging, when the child is learning a polysynthetic language – a language in which words are highly morphologically complex, expressing in a single word what in English takes a multi-word clause. To date, there has been no cross-linguistic survey of how children approach thispuzzle and learnpolysynthetic languages.Thispaperaimstoprovide suchasurvey, including a discussion of some of the general findings in the literature regarding the acquisition of polysynthetic systems.

This paper argues that denominal verb formation in Ojibwe is similar to nominal incorporation: a complex nominal (containing inflectional and derivational affixes) incorporates into a verb. This process differs from denominalization in languages such as English where denominal verbs are the result of the concatenation of derivational affixes and nouns that are simple roots. This paper contributes to ongoing discussion on the composition of words in polysynthetic languages, the nature of nominal incorporation, and the relationship between morphology and syntax.[KEYWORDS: denominal verbs, nominal incorporation, Ojibwe, Algonquian]

Pronominal affixes in polysynthetic languages have an ambiguous status in the sense that they have characteristics normally associated with free pronouns as well as characteristics associated with agreement markers. This situation arises because pronominal affixes represent intermediate stages in a diachronic development from independent pronouns to agreement markers. Because this diachronic change is not abrupt, pronominal affixes can show different characteristics from language to language. By presenting an in-depth discussion of the pronominal affixes of Yurakare, an unclassified language from Bolivia, I argue that these so-called intermediate stages as typically attested in polysynthetic languages actually represent economical systems that combine advantages of agreement markers and of free pronouns. In terms of diachronic development, such ‘intermediate’ systems, being functionally well-adapted, appear to be rather stable, and it can even be reinforced by subsequent diachronic developments.

We propose for Kadiwéu, a polysynthetic language of Brazil, an extension of the POS annotation of the Tycho Brahe Annotated Corpus of Historical Portuguese (www.tycho.iel.unicamp.br/~tycho/corpus) – henceforth TBC, which consists in tagging both words and morphemes, yielding a two-level annotation. The tagging of words is necessary to generate the syntactic parsing that is missing from the current corpuses of Brazilian native languages. The morphological tagging is also crucial for polysynthetic languages since it allows searching for grammatical properties encoded by the morphemes. This is a pioneer proposal since it is the first time an American Indian language will be part of a Corpus allowing grammatical searches that include morphological and syntactic information.

The Cayuga (Northern Iroquoian) word has been identified as a domain (one relevant for pitch accent assignment [Chafe 1977:169]). The goal of this paper is to determine which linguistic constituent corresponds to the Cayuga word domain. This paper presents evidence that the Cayuga word domain does not correspond to the grammatical word, nor to the prosodic word; instead, it is coextensive with the Phonological Phrase domain. Several characteristics of words in polysynthetic languages fall out from this analysis.

Given the substantial variation in the nature of the grammatical word (GW) across languages, this paper addresses the question of whether the Phonological Word (PW) exhibits the same degree of variation or rather abstracts away from it due to the typically flatter nature of the phonological hierarchy. Various types of languages are examined, focusing on isolating and polysynthetic languages—opposite ends of a word structure continuum. It is demonstrated that, indeed, the PW exhibits substantially less variation across languages than might be expected on the basis of the differences in GW structure. Furthermore, it is shown that an additional constituent (i.e., the Clitic Group, renamed Composite Group) is required between the PW and the Phonological Phrase to fully account for the interface between morpho-syntactic and phonological structures.

There has been much debate about whether mixed languages arise from code-switching. This paper presents one clear example of this kind of genesis, Gurindji Kriol, and other probable examples, from recent language contact in Australia between traditional Australian languages and English-based pidgins/creoles. In particular the paper focuses on what has been called the Verbal-Nominal split in the genesis of these languages, which is parallel to other cases elswhete in the world, such as Michif. Here the Verbal-Nominal split is reanalysed as a split between INFL (Tense-Aspect-Mood) dominated elements and the rest of the clause. There are two classes of such INFL mixed languages with contrasting characteristics: those in which the new language takes over the INFL elements and the nominal morphology is still drawn from the old language, like Gurindji Kriol; and those in which the verb and its morphology is retained from the old language but other elements are drawn from the new language. This is explained in terms of the 'arrested turnover' hypothesis of Myers-Scotton. The original 'centre of gravity' hypothesis of McConvell related the two kinds of mixed language outcomes to the grammatical type of the old language: whether it was 'dependent-marking' or 'headmarking'. In this paper this hypothesis is modified by seeing the important causal factor in the second type as incorporation of INFL and pronouns in the verb in head-marking and polysynthetic languages. Finally some other examples of mixed languages of the INFL-split type are mentioned, and a research program outlined aiming to detect where this kind of language-mixing forms part of the history of other languages by looking at the current pattern of composition of elements from different language sources.

It has been claimed that all languages have complementation (Noonan 1985), but this assumption has been called into question (Thompson 2002, Englebretson 2003, and Dixon 1995). An examination of Chimariko texts reveals no syntactic evidence for complementation. Chimariko is a polysynthetic language with a complex verb morphology which follows a hierarchical system, as well as an agent–patient distinction, in the marking of core arguments. There are no obvious complementizers, and putative complements are never marked as arguments on a predicate. The semantic concepts expressed by complements in some languages are coded in Chimariko (1) as separate clauses with no reduction of the predicates or any other restrictions or (2) as an integral part of the verbal morphology. By applying Noonan’s analysis of semantic classes of complement‐taking predicates to Chimariko, I demonstrate that this language represents a counterexample to the universality claim that all languages have complementation.

This paper describes the behavior of a subordination marker -n in modal necessity constructions in West Circassian, a polysynthetic language belonging to the Northwest Caucasian family. We show that -n functions as a simple suffix in the non-epistemic construction and as a phrasal affix in the epistemic construction. Hence, this morpheme violates the principle according to which the formal characteristics of a linguistic element should remain the same in different contexts of its use. This violation is explained by the difference in the semantic contribution of the suffix under discussion in different patterns and by the typological specifics of West Circassian, which allows its speakers to manipulate with morphemes more freely.

Although there exist comprehensive morphologically annotated corpora for many morphologically rich languages, there have been no such corpora for any polysynthetic language so far. Polysynthetic languages raise a variety of theoretical and practical challenges for corpus linguistics. Some of these challenges have been partly addressed when developing corpora for e. g. Turkic or Uralic languages, while others are unique for this kind of languages. Our paper identifies the most prominent challenges that we are facing in the course of development of West Circassian (Adyghe) corpus, and offer possible solutions. These include the tokenization problem, which involves delimiting morphology from syntax, the problem with lemmatization and part-of-speech tagging, and a number of glossing and search problems.

The long chain of verbal prefixes which is common in the Old Irish glosses is a phenomenon which has parallels in Slavic. The normal order of Old Irish prefixes is known. Violations of that order can be observed. The prefixes are homonymous with prepositions and thus have a local meaning. The exact function of the prefixes in the chain is, however, not always obvious. Some prefixes are telic and have a weakened local meaning. The reason for the order in which the prefixes occur has not been well explained. It is often difficult to determine where a certain prefix was placed when it was added to a preexisting shorter group of prefixes. Ro, con and ad can be infixed (in some cases prefixed) to indicate that an act or a state is seen as completed (perfective or perfect) or where ro and con indicate possibility or ability. The question can arise whether a certain prefix is of local or ‘aspectual’ nature. Doubling of prefixes occurs as in comtherchomracc ‘assembly’. In do·é-com-nacht ‘has bestowed’ ‘perfectivizing’ con- has been infixed into a pre-existing do·ind-naig. In·r-úa-lad ‘I have entered’ shows a similar process but with the ‘perfectivizing’ ro- in a different position. -r-ind·úa-lad (Ml 93cl4) ‘thou hast entered’ shows that this is not the only possible order. Multiple prefixes in Slavic were recently discussed by Filip (2003) who drew attention to the problematic nature of the function of prefixes as grammatical (inflexional) markers of perfective aspect. She also discussed their behaviour as directional prefixes. Most of her examples are from Czech and contain two (or three) prefixes. We shall draw attention to the prefixed verbs of Bulgarian which has longer chains of prefixes. It will therefore be interesting to compare Bulgarian with Old Irish. Many questions concerning these chains are similar to those that arise in the analysis of the OI chains of prefixes both in regard to function and position in the chain. The different position of 'aspectual' prefixes in Slavic and Irish should be observed. In this paper I shall mainly treat the functional properties of the prefixes. Bulgarian pridobivam is not too difficult to analyze. Izpoprebivam is more complicated as it shows actional iz- and the likewise actional po- in the initial position which is natural for such prefixes in Slavic languages. Izponaprikàzvam shows four prefixes though the very common combined prefix izpo- has been analyzed as being one entity. Nevertheless, po- seems to have kept its distributive meaning. Doubling of prefixes is found in Bulgarian. Filip (2003) considers the prefixes to be derivational morphemes, but does not see them as formal markers of perfectivity. The action of a prefixed verb may be quantized, telic or bounded. Böttger (2004) discusses grammatical and lexical derivation, grammaticalization and the Russian prefixes po-, zu- and ot- in a diachronical perspective. We shall discuss these theories and their applicability to OI and Bulgarian. The “directional and measurement usages” of Slavic and Irish prefixes will finally be compared with the functions of the Hittite enclitic directional particles.

International audience; In this paper, we review the documented diachronic pathways leading to antipassive markers in the world's languages and show that Japhug Rgyalrong, a polysynthetic language belonging to the Sino-Tibetan family, attests a previously unreported source of antipassives. In Japhug, the two antipassive constructions (human and non-human antipassive) are built from the base verb through a two-step process: first nominalization into an action nominal, and second denominal verbalizing derivation of the action noun into an intransitive verb. Nominalization neutralizes the verb's transitivity, and a new transitivity value is allocated by the denominal prefix. A similar pathway is proposed for other derivations, in particular the applicative.

Data from the transcript of a public meeting are used to examine patterns of code switching in Tiwi, a head-marking polysynthetic language of northern Australia where an N-V mixed language has emerged among younger speakers (Lee 1987, McConvell 2008). McConvell’s theory predicts that code switching behavior should reflect the same center of gravity principles as found in the resulting mixed language. In particular, a head-marking language like Tiwi should most commonly provide the verbal part of code-switched utterances. Although the corpus used here is too small to draw definitive conclusions, the data do not support the supposed connections between morphological typology, code switching, and mixed language genesis. Reasonable interpretations of the data show no preference for code switching that retains Tiwi’s verbal subsystem, as predicted by McConvell’s center of gravity model.

This paper presents aspects of a computational model of the morphology of Plains Cree based on the technology of finite state transducers (FST). The paper focuses in particular on the modeling of nominal morphology. Plains Cree is a polysynthetic language whose nominal morphology relies on prefixes, suffixes and circumfixes. The model of Plains Cree morphology is capable of handling these complex affixation patterns and the morphophonological alternations that they engender. Plains Cree is an endangered Algonquian language spoken in numerous communities across Canada. The language has no agreed upon standard orthography, and exhibits widespread variation. We describe problems encountered and solutions found, while contextualizing the endeavor in the description, documentation and revitalization of First Nations Languages in Canada.

The Online Dictionary of New Zealand Sign Language (ODNZSL), launched in 2011, is an example of a contemporary sign language dictionary that leverages the 21st century advan­tages of a digital medium and an existing body of descriptive research on the language, including a small electronic corpus of New Zealand Sign Language. Innovations in recent online dictionaries of other signed lan­guages informed development of this bilingual, bi-directional, multi­media dic­tionary. Video content and search capacities in an online medium are a huge advance in more directly repre­senting a signed lexicon and enabling users to access content in versatile ways, yet do not resolve all of the theoretical challenges that face sign language dictionary makers. Con­sidera­tions in the editing and production of the ODNZSL are discussed in this article, including issues of determin­ing lexemes and word class in a polysynthetic language, deriving usage exam­ples from a small corpus, and dealing with sociolinguistic variation in the selection and perform­ance of con­tent. Keywords: sign language lexicography, online dictionaries, multimedia dictionaries, bilingual dictionaries, learner dictionaries, new zealand sign language, video content, sign language corpus, polysynthetic mor­phology, polysemy, sociolinguistic variation, sign language lin­guistics, user profile

The American scholar Edward Sapir proposed that the Na-Dene languages of North America (Haida, Tlingit, Eyak, Athabaskan) are genetically related to the Sino-Tibetan languages of Asia (Sinitic, Karenic, Tibeto-Burman). Sapir and other linguists have found correspondences between the phonetic systems, and more importantly, the grammar and basic vocabulary of both families. It is maintained here that Spair was essentially correct about this genetic connection (“Sino-Dene”), and that there is no evidence that he ever retracted it. Recent research corroborates Sapir's hypothesis and indicates the high probability of a widespread Dene-Caucasian family, which dispersed some 11, 000 years ago. Some new linguistic evidence for this family is also presented here for the first time.

Eskimo languages are commonly characterized as displaying rather ‘free’ word order as compared to the major western European languages. Nevertheless, there is in West Greenlandic at least a clearly dominant, pragmatically neutral ordering pattern. Deviation from this – when possible at all – results in specifiable contextual marking (the factors involved will be discussed and illustrated in section 2). In fact, the degree of ‘freedom’ involved may vary considerably from dialect to dialect (and from language to language), also through time and according to register/medium. Specifically I shall be claiming thatnoEskimo dialect is of the purely pragmatically based word order type (lacking a syntactic ‘basic order’) which Mithun claims is typical for polysynthetic languages with inflected verbs that can stand as independent sentences (Mithun, 1987: 323). Unlike the type of language that Mithun describes, which includes (Iroquoian) Cayuga and (‘Penutian’) Coos, for example, I shall argue that West Greenlandic (WG), a highly polysynthetic language, behaves more like Slavic languages in this respect, though the ‘neutral’ pattern there is of course SVO rather than SOV. Much as described for Czech and Russian by the Prague School functionalists, word order in WG seems to reflect the common ‘functional sentence perspective’ whereby – ignoring postposed ‘afterthought/clarificatory’ material – early position in the sentence is associated with given material of low communicative dynamism, whereas later position is associated with new or important material of high communicative dynamism (see Firbas, 1974). This is the reverse of the situation described by Mithun.

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

The aim of this paper is to provide a more accurate typological classification of Modern Greek. The verb in MG shows many polysynthetic traits, such as noun and adverb incorporation into the verbal complex, a large inventory of bound morphemes, pronominal marking of objects, many potential slots before the verbal head, nonconfigurational syntax, etc. On the basis of these traits, MG has similarities with polysynthetic languages such as Abkhaz, Cayuga, Chukchi, Mohawk, Nahuatl, a.o. I will show that the abundance of similar patterns between MG and polysynthesis point to the evolution of a new system away from the traditional dependent-marking strategy and simple synthesis towards head-marking and polysynthesis. Finally, I will point to the risk of undertaking a direct comparison of different language systems by discussing the pronominal head-marking strategies in MG and the North American languages.

This article uses instrumental data from natural speech to examine the phenomenon of pause placement within the verbal word in Dalabon, a polysynthetic Australian language of Arnhem Land. Though the phenomenon is incipient and in two sample texts occurs in only around 4% of verbs, there are clear possibilities for interrupting the grammatical word by pause after the pronominal prefix and some associated material at the left edge, though these within-word pauses are significantly shorter, on average, than those between words. Within-word pause placement is not random, but is restricted to certain a‰x boundaries; it requires that the paused-after material be at least dimoraic, and that the remaining material in the verbal word be at least disyllabic. Bininj Gun-wok, another polysynthetic language closely related to Dalabon, does not allow pauses to interrupt the verbal word, and the Dalabon development appears to be tied up with certain morphological innovations that have increased the proportion of closed syllables in the pronominal prefix zone of the verb. Though only incipient and not yet phonologized, pause placement in Dalabon verbs suggests a phonologydriven route by which polysynthetic languages may ultimately become less morphologically complex by fracturing into smaller units.

It is the aim of this paper to evaluate the various types of sentential complementation available in terms of complement control cross-linguistically. I will propose a lexical classification of control classes on the basis of the instantiated subordination patterns. I want to focus on an important distinction, namely that of structural vs. inherent control. Structural control is found with predicates that select a clausal complement whose structure requires argument identification and thus 'induces' control. Infinitival complements are prototypical cases for this kind of control because in most languages infinitival complements can only 'survive' in structures of control or raising. The interesting question is which predicates license structural control and which cross-linguistic differences emerge between potential licensors. Inherent control is found with predicates that require control readings independent of the instantiated structure of sentential complementation (e.g. a directive predicate such as zwingen 'force'). In addition, I will recapitulate and add arguments for the dual lexical-syntactic nature of complement control., ZAS Papers in Linguistics, Bd. 47 (2007): Studies in complement control

In Proto-Algic, a primary noun stem often ends in a derivational final, a secondary one in a nominalizer. Both usually resemble demonstrative pronouns in shape. A dependent noun stem lacks a root, and generally begins in **i:, **o:, or in a consonant other than **2 or **k(w). In those cases where it is paired with a medial beginning in **2 or **k(w), a prefix receptor (**-c, **--I, **-s, **-S, or **-t) is prefixed. Inflection for allocation differs for kin, human body parts, parts of nonhumans, and general nouns.

1 This article began with the problem and hypothesis presented in 5, while I was working on Wasco-Wishram with Hiram Smith and Philip Kahclamet at Warm Springs Reservation, Oregon in the summer of 1956. The idea was developed into the initial substantive section (2) and was the springboard for a paper written while I was a fellow of the Center for Advanced Study in the Behavioral Sciences in 1957-58 (Dell Hymes, "On Chinookan Cognitive Style" [1958]). The general account of a cognitive style in that paper, including mention of the analysis of the tense system, was summarized in section 5 of Dell Hymes, "On Topology of Cognitive Styles in Language (with examples from Chinookan)," AnthropologicalLinguistics 3, no. 1 (1961): 38-41; but the paper as a whole has remained unpublished. It is fortunate that this contribution of mine to Chinookan grammar should be published to honor a scholar who knew Sapir, Walter Dyk, and Philip Kahclamet and whose contribution to the tradition in which they worked is so universally esteemed. 2 E. H. Sturtevant, An Introduction to Linguistic Science (New Haven, Conn.: Yale University Press, 1947), p. 58; and Harry Hoijer, "Some Problems of American Indian Linguistic Research," University of California Publications in Linguistics 10 (1954): 10. 3 Mary R. Haas, "Ablaut and Its Function in Muskogee," Language 16 (1940): 141-50.