This paper presents a fresh take on the origins and nature of Sri Lanka Malay (SLM), based on fieldwork data collected in 2003–2005 in Kirinda, in the south-east of Sri Lanka. It departs from previous studies of SLM in that it is based on substantial recordings of spoken data in natural settings as well as coverage of oral and written history. Work on SLM so far has offered significant insights into the nature of these varieties; due to limited data available, however, some aspects have failed to emerge which are important for our current understanding of SLM. In particular, I aim to show the value of first-hand historical research and natural linguistic data in order to achieve plausible accounts of genesis and accurate classifications of SLM varieties. Based on the combination of these approaches, this paper argues that SLM is the result of trilingual admixture, in which a typological shift from Malay to Lankan grammar occurs.

There is no unified policy for preservation and/or documentation of endangered languages. Each case engenders its own issues and constraints due to its particular cultural and historical situation. Documentation of the Amazonian Trumai language and culture illustrates the intricacy of the parameters that run the gamut from globalisation to local and punctual conflicts, and from the state of world affairs to quarrels over chiefdom inheritance.

This paper describes the verbal Tense-Aspect-Mood system of Awetí (Tupian, Central Brazil) in a Word-and-Paradigm approach.

One classification of Awetí verb forms contains clear aspect categories. A second set of independent classifications renders at least four moods and contains a third major TAM classification, factuality, that has one mainly temporal category Future, while others are partially or wholly modal.

Structural categories reflect the formal composition of the forms. Some forms are synthetic, ‘marked’ only by means of affixes, but many are analytic, containing auxiliary particles.

With selected sample forms we demonstrate in detail the interplay of structural and functional categories in Awetí verb paradigms.

As the result of language contact in the northern Tibetan region, one variety of the Mongolic language Monguor (ISO 639-3: MJG) realizes prosodic accent as a rising pitch contour. Furthermore, a small number of homophones have come to be distinguished by tonal contour. Although at least two Turkic and Mongolic languages have occasionally copied the most salient tonal features of some Chinese loanwords, this is the first known example of both distinctive pitch contrasts in native lexemes, as well as default prosodic accent at the utterance level. Such an incipient tonal system offers insight into the relationship between often-contested types of prosodic accent as well as the effects of intensive language contact.

In this article we present results from interdisciplinary research among the Kuikuro of the Upper Xingu (Brazil). The project integrates linguistic, ethnographic and archaeological data as a means to reconstruct the processes through which peoples speaking languages of the three largest South American linguistic groupings (Arawak, Carib, and Tupi), as well as a language isolate (Trumai), came to create a unique social system: the Upper Xingu sociocultural complex. We address the following questions: how did this system – spanning from the 9th century AD until the present and formed by peoples with distinct cultures and origins – come into being? Which cultural bases and historical circumstances led to its structuring? What role did language and multilingualism play in this process?

The paper discusses several linguistic peculiarities of three endangered Caucasian languages of Georgia that are currently being documented within the DoBeS “ECLinG” project, viz. Svan, Tsova-Tush (Batsbi), and Udi. The main focus lies on questions of the phonology (vowel systems, pharyngeal and laryngeal subsystems) and the morphosyntax (verbal agreement, ergativity) of these languages. The second part of the paper is devoted to the question whether the audiovisual material collected in the course of the project admits of establishing linguistic parameters of language endangerment with respect to the languages in question.

This paper focuses on Tapiete (Tupi-Guarani) and Vilela (Lule-Vilela), the two most severely endangered languages in the Argentine Chaco. Both show the results of linguistic attrition without obsolescence. However, the state of each one and chances for revitalization differ radically. While multigenerational Tapiete communities exist and are strengthening ties among themselves, the extreme paucity of Vilela speakers and the lack of a speech community have proven to be critical threats. The paper examines two particular aspects of language shift: borrowing in Tapiete and attrition in Vilela. Creative processes in the adaptation of loanwords have revealed linguistic resistance in Tapiete. In the case of Vilela, despite documented phonological and grammatical reduction, core linguistic structures have been activated through language remembering strategies. Although the pressure of Spanish has long been present, its structure has not been decisive to vernacular language loss. Rather, the abandonment of Tapiete and Vilela is rooted in the speakers' history of socio-cultural subordination.

Small and moribund languages seem to behave in some ways as if they were going to continue living forever. Their speakers – including those in the very terminal generation – may continue to introduce changes and innovations, including changes resulting in both simplification and in greater complexity. It is often difficult to disentangle whether a particular change is driven by internal restructuring, contact induced change, obsolescence effects, or some combination of these. We also find in moribund languages an unusually high incidence of variation both across and within speakers, variation that cannot be correlated to social or demographic factors given the very small size of the speech community. We present new data from a salvage documentation of Tofa, an endangered Turkic language of Siberia, to argue that moribund languages may provide an ideal laboratory to study the interaction of domains in language change.

Hočank is a highly endangered Siouan language of Wisconsin and Nebraska which currently is the object of an extensive documentation project at the University of Erfurt, Germany. The paper presents a descriptive investigation of parts of Hočank verb morphology and its implications for morphological theory. Hočank verb morphology – in particular the left side of the verbal complex – reveals cross-linguistically highly unusual and dispreferred patterns which pose a challenge to traditional and contemporaneous morphological theory. Hočank verbs show to some degree systematically a) discontinuous stems, b) stem-internal inflection, and c) inflectional morphology which is morphotactically closer to the verb root than derivational morphology. Diachronically, these patterns derive from the lexicalization of mostly derivational morphology or compounding which eventually led to the entrapment of inflectional prefixes, hence creating interfixes. The traditional notions of infixation and/or interfixation cannot account for these patterns in a satisfying way. Therefore, a partly new and systematized typology of affix types is proposed which takes into consideration the results of diachronic linguistics as well as grammaticalization theory. A grammaticalization path from interfixes to infixes is proposed with regard to the Hočank data but with relevance beyond this individual case.

The main goal of this study is to explore some of the phonetic, morphological, and syntactic resources of same-turn self-repair initiation that are available to speakers of Wichita, an indigenous North American language from the Caddoan family. A 28-minute natural conversation between three Wichita speakers is coded and analyzed within a conversation analytic approach. Within this approach the study focuses on same-turn self-repair, a subtype of natural conversational repair process in which an emerging utterance is stopped by the current speaker in some way and is then aborted, recast, continued, or redone by the speaker within the same turn.

Wichita is a polysynthetic language with complex morphology and morphophonemic alternations that accompany the formation of phrases and sentences (Rood 1976, 1996). Conversational analysis research in Wichita – and in structurally similar languages in general – is at a very early stage. This pioneering study thus focuses on a form-based analysis that illuminates the possible means for self-repair initiation in Wichita conversation, as well as giving insight into some of the phonetic and prosodic aspects that accompany the self-repair initiations. The study also touches on a few fundamental morphosyntactic issues by considering the nodes within complex Wichita “words” where self-repairs are generated.

This detailed form-based analysis yields some interesting conclusions about morphosyntactic constraints on self-repair in a polysynthetic language. In particular, the study uncovers what the relationship between conversational repair, syntax, and morphology is and how these facets of interaction and grammar are interdependent (Schegloff 1979). Furthermore, the results of the study help enrich our general understanding of the function of self-repair in conversational interaction, either in terms of the types of variations that we can expect cross-linguistically or in terms of the generalizations that we can make about natural discourse structure. The discussion at the end of the paper sketches how some of the findings are, or are not, compatible with general results from previous work on same-turn self-repair as outlined by Schegloff, Jefferson, and Sacks (1977) and Jasperson (1998, 2002).

This contribution emphasizes the role of multimedia data not only for archiving languages but also for creating opportunities for innovative analyses. In the case at hand, video material was collected as part of the documentation of Akhoe Haiom, a Khoisan language spoken in northern Namibia. The multimedia documentation project brought together linguistic and anthropological work to highlight connections between specialized domains, namely kinship terminology, interrogatives and reciprocals. These connections would have gone unnoticed or undocumented in more conventional modes of language description. It is suggested that such an approach may be particularly appropriate for the documentation of endangered languages since it directs the focus of attention away from isolated traits of languages towards more complex practices of communication that are also frequently threatened with extinction.

This paper presents a fresh take on the origins and nature of Sri Lanka Malay (SLM), based on fieldwork data collected in 2003–2005 in Kirinda, in the south-east of Sri Lanka. It departs from previous studies of SLM in that it is based on substantial recordings of spoken data in natural settings as well as coverage of oral and written history. Work on SLM so far has offered significant insights into the nature of these varieties; due to limited data available, however, some aspects have failed to emerge which are important for our current understanding of SLM. In particular, I aim to show the value of first-hand historical research and natural linguistic data in order to achieve plausible accounts of genesis and accurate classifications of SLM varieties. Based on the combination of these approaches, this paper argues that SLM is the result of trilingual admixture, in which a typological shift from Malay to Lankan grammar occurs.

There is no unified policy for preservation and/or documentation of endangered languages. Each case engenders its own issues and constraints due to its particular cultural and historical situation. Documentation of the Amazonian Trumai language and culture illustrates the intricacy of the parameters that run the gamut from globalisation to local and punctual conflicts, and from the state of world affairs to quarrels over chiefdom inheritance.

This paper describes the verbal Tense-Aspect-Mood system of Awetí (Tupian, Central Brazil) in a Word-and-Paradigm approach.

One classification of Awetí verb forms contains clear aspect categories. A second set of independent classifications renders at least four moods and contains a third major TAM classification, factuality, that has one mainly temporal category Future, while others are partially or wholly modal.

Structural categories reflect the formal composition of the forms. Some forms are synthetic, ‘marked’ only by means of affixes, but many are analytic, containing auxiliary particles.

With selected sample forms we demonstrate in detail the interplay of structural and functional categories in Awetí verb paradigms.

As the result of language contact in the northern Tibetan region, one variety of the Mongolic language Monguor (ISO 639-3: MJG) realizes prosodic accent as a rising pitch contour. Furthermore, a small number of homophones have come to be distinguished by tonal contour. Although at least two Turkic and Mongolic languages have occasionally copied the most salient tonal features of some Chinese loanwords, this is the first known example of both distinctive pitch contrasts in native lexemes, as well as default prosodic accent at the utterance level. Such an incipient tonal system offers insight into the relationship between often-contested types of prosodic accent as well as the effects of intensive language contact.

In this article we present results from interdisciplinary research among the Kuikuro of the Upper Xingu (Brazil). The project integrates linguistic, ethnographic and archaeological data as a means to reconstruct the processes through which peoples speaking languages of the three largest South American linguistic groupings (Arawak, Carib, and Tupi), as well as a language isolate (Trumai), came to create a unique social system: the Upper Xingu sociocultural complex. We address the following questions: how did this system – spanning from the 9th century AD until the present and formed by peoples with distinct cultures and origins – come into being? Which cultural bases and historical circumstances led to its structuring? What role did language and multilingualism play in this process?

The paper discusses several linguistic peculiarities of three endangered Caucasian languages of Georgia that are currently being documented within the DoBeS “ECLinG” project, viz. Svan, Tsova-Tush (Batsbi), and Udi. The main focus lies on questions of the phonology (vowel systems, pharyngeal and laryngeal subsystems) and the morphosyntax (verbal agreement, ergativity) of these languages. The second part of the paper is devoted to the question whether the audiovisual material collected in the course of the project admits of establishing linguistic parameters of language endangerment with respect to the languages in question.

This paper focuses on Tapiete (Tupi-Guarani) and Vilela (Lule-Vilela), the two most severely endangered languages in the Argentine Chaco. Both show the results of linguistic attrition without obsolescence. However, the state of each one and chances for revitalization differ radically. While multigenerational Tapiete communities exist and are strengthening ties among themselves, the extreme paucity of Vilela speakers and the lack of a speech community have proven to be critical threats. The paper examines two particular aspects of language shift: borrowing in Tapiete and attrition in Vilela. Creative processes in the adaptation of loanwords have revealed linguistic resistance in Tapiete. In the case of Vilela, despite documented phonological and grammatical reduction, core linguistic structures have been activated through language remembering strategies. Although the pressure of Spanish has long been present, its structure has not been decisive to vernacular language loss. Rather, the abandonment of Tapiete and Vilela is rooted in the speakers' history of socio-cultural subordination.

Small and moribund languages seem to behave in some ways as if they were going to continue living forever. Their speakers – including those in the very terminal generation – may continue to introduce changes and innovations, including changes resulting in both simplification and in greater complexity. It is often difficult to disentangle whether a particular change is driven by internal restructuring, contact induced change, obsolescence effects, or some combination of these. We also find in moribund languages an unusually high incidence of variation both across and within speakers, variation that cannot be correlated to social or demographic factors given the very small size of the speech community. We present new data from a salvage documentation of Tofa, an endangered Turkic language of Siberia, to argue that moribund languages may provide an ideal laboratory to study the interaction of domains in language change.

Hočank is a highly endangered Siouan language of Wisconsin and Nebraska which currently is the object of an extensive documentation project at the University of Erfurt, Germany. The paper presents a descriptive investigation of parts of Hočank verb morphology and its implications for morphological theory. Hočank verb morphology – in particular the left side of the verbal complex – reveals cross-linguistically highly unusual and dispreferred patterns which pose a challenge to traditional and contemporaneous morphological theory. Hočank verbs show to some degree systematically a) discontinuous stems, b) stem-internal inflection, and c) inflectional morphology which is morphotactically closer to the verb root than derivational morphology. Diachronically, these patterns derive from the lexicalization of mostly derivational morphology or compounding which eventually led to the entrapment of inflectional prefixes, hence creating interfixes. The traditional notions of infixation and/or interfixation cannot account for these patterns in a satisfying way. Therefore, a partly new and systematized typology of affix types is proposed which takes into consideration the results of diachronic linguistics as well as grammaticalization theory. A grammaticalization path from interfixes to infixes is proposed with regard to the Hočank data but with relevance beyond this individual case.

The main goal of this study is to explore some of the phonetic, morphological, and syntactic resources of same-turn self-repair initiation that are available to speakers of Wichita, an indigenous North American language from the Caddoan family. A 28-minute natural conversation between three Wichita speakers is coded and analyzed within a conversation analytic approach. Within this approach the study focuses on same-turn self-repair, a subtype of natural conversational repair process in which an emerging utterance is stopped by the current speaker in some way and is then aborted, recast, continued, or redone by the speaker within the same turn.

Wichita is a polysynthetic language with complex morphology and morphophonemic alternations that accompany the formation of phrases and sentences (Rood 1976, 1996). Conversational analysis research in Wichita – and in structurally similar languages in general – is at a very early stage. This pioneering study thus focuses on a form-based analysis that illuminates the possible means for self-repair initiation in Wichita conversation, as well as giving insight into some of the phonetic and prosodic aspects that accompany the self-repair initiations. The study also touches on a few fundamental morphosyntactic issues by considering the nodes within complex Wichita “words” where self-repairs are generated.

This detailed form-based analysis yields some interesting conclusions about morphosyntactic constraints on self-repair in a polysynthetic language. In particular, the study uncovers what the relationship between conversational repair, syntax, and morphology is and how these facets of interaction and grammar are interdependent (Schegloff 1979). Furthermore, the results of the study help enrich our general understanding of the function of self-repair in conversational interaction, either in terms of the types of variations that we can expect cross-linguistically or in terms of the generalizations that we can make about natural discourse structure. The discussion at the end of the paper sketches how some of the findings are, or are not, compatible with general results from previous work on same-turn self-repair as outlined by Schegloff, Jefferson, and Sacks (1977) and Jasperson (1998, 2002).

This contribution emphasizes the role of multimedia data not only for archiving languages but also for creating opportunities for innovative analyses. In the case at hand, video material was collected as part of the documentation of Akhoe Haiom, a Khoisan language spoken in northern Namibia. The multimedia documentation project brought together linguistic and anthropological work to highlight connections between specialized domains, namely kinship terminology, interrogatives and reciprocals. These connections would have gone unnoticed or undocumented in more conventional modes of language description. It is suggested that such an approach may be particularly appropriate for the documentation of endangered languages since it directs the focus of attention away from isolated traits of languages towards more complex practices of communication that are also frequently threatened with extinction.