Nicholas Welch

Dene (Athapaskan) languages typically have a small inventory of semantically light verbs. This chapter demonstrates that their interpretations derive wholly from their syntactic context in predictable ways and proposes these verbs are spellouts of morphosyntactic structure with either semantically vacuous roots or none at all. They are shown to form a cline of structural complexity and it is suggested that some of the cross-linguistic semantic variability observed in light verbs may originate from this structural variation. Additionally, since these verbs serve as matrix verbs of full clauses, they cast doubt on claims that light verbs are syntactically dependent on main verbs. The extreme semantic impoverishment and configurational relatedness of these verbs suggests that they are a unified class. Two of them are commonly termed copulas; the data and analysis presented here, however, suggest that a principled distinction between copulas and light verbs may ultimately be illusory.

A widely accepted assumption in both the syntactic and semantic literature is that copulas lack semantic content. A consequent question is how to explain the existence in certain languages of two copular verbs that give rise to different interpretations. Such is the case in numerous languages of the Dene family (formerly known as Athapaskan). We explain this situation with the hypothesis that the copulas realize an underlying three-copula system differing in argument structure. Differences between the interpretations of copular clauses in these languages originate in the compositional semantics of these structures, not in any lexical semantic differences.This hypothesis successfully predicts the distributional differences between the surface forms of the Dene copulas, such as their compatibility with adjuncts of time and intentionality, interactions with accusative case, and semantic lifetime effects.

The languages of the Dene (aka Athapaskan) family in North America almost universally employ two copular verbs. In several languages of this family, copular forms are also employed as verbal auxiliaries: forms of one copula mark clausal focus while forms of the other mark TAM (tense/aspect/mode) categories. With reference to two Dene languages in particular, Tłı̨chǫ Yatıì and Tsúùt’ínà, I explain this difference by positing distinct grammaticalization paths and motivations for each copula: both focus and TAM markers originate in a uniclausal reanalysis of biclausal constructions, the former from constructions where the matrix clause asserts the truth of the embedded clause, and the latter from those where the matrix clause supplies extra TAM information to the embedded clause. Both grammaticalizations involve an upward reanalysis of copulas as functional heads.

In Tłıc̨hǫ Yatıì (Dene/Athapaskan; aka Dogrib), patterns of copula use appear to undermineMoro’s (1997: 248–261) claim that copulas’only function is to provide a site for the morphological realization of inflection. The aimof this note is to show that copulas in this language are obligatory with all nominal predicates, but occur with adjectival predicates only when either the subject has Φ-features requiring agreement or a marked tense/ aspect/mode (TAM) interpretation is intended. This asymmetry suggests that their obligatory occurrence with all nominal predicates must be motivated by other factors. Moro’s contention that copulas are only markers of inflection is a very old claim, going back to Aristotle (Moro 1997: 249). Though the view that copulas are enablers of predication continues to be supported by many syntacticians today (e.g., among others, Bowers 1993, den Dikken 2006, Baker and Vinokurova 2012), Moro’s book demonstrates that there is strong evidence for his claim in the Indo-European languages. The existence of languages like Mandarin, in which verbs do not show inflection, but copulas are required with nominal predicates (Zhan and Sun 2013: 762), suggests that this view may not be universally correct, but rather, the content of copulas may be parametrized. The present note shows that there exists at least one language where both views of copulas are supported, but by different lexical categories. This result implies that parametrization of the role of copulas must apply to individual non-verbal categories.

In Tłı̨chǫ Yatıì (Dogrib), spoken in the Northwest Territories, Canada, the periphrastic particle ı̨lè, which has been considered a past marker, is optional in the clause. As in other Dene (Athapaskan) languages, viewpoint aspect is encoded morphologically on the verb. These two facts can give the impression that aspect is the only obligatory temporal category in this language and that other temporal distinctions are peripheral. I argue, on the contrary, that a future/non-future distinction is a necessary element of well-formed clauses, that Future, rather than Past/Non-Past Tense or Aspect, serves as an anchor in the sense of Enç (1987), and that anchoring does not necessarily correlate with the prominence of a temporal category in the sense of Bhat (1999). I adduce evidence in support of these proposals from the contrasting obligatoriness of past and future marking in predicates, as well as from word order facts.

Tłı̨chǫ Yatıì, a Dene language of the Northwest Territories, Canada, has a number of post-verbal auxiliaries and particles indicating categories such as futurity, mode, negation, information structure, and evidentiality. The interaction of these elements reveals that they occur in a strict order, which in turn illuminates the structure of the clause in this language, with positions for future, mode, negation, and focus as functional categories at the right edge.

In Tłı̨chǫ Yatıì (Dene, aka Athapaskan), copulas appear obligatorily with adjectives predicated of animate subjects, but are barred from appearing with adjectives predicated of inanimates. I propose that this asymmetry arises from a requirement to realize grammatical agreement for person, and that animate nouns alone bear a person feature. Unlike verbs, adjectives in this language cannot inflect; hence copulas are inserted in adjectival predicates as a rescue strategy to avoid ungrammaticality.

In TłĮchǫ Yatıı̀ (Dene, aka Athapaskan), copulas appear obligatorily with adjectives predicated of animate subjects, but are barred from appearing with adjectives predicated of inanimates. I propose that this asymmetry arises from a requirement to realize grammatical agreement for person, and that animate nouns alone bear a person feature. Unlike verbs, adjectives in this language cannot inflect; hence copulas are inserted in adjectival predicates as a rescue strategy to avoid ungrammaticality.

The languages of the Dene (aka Athapaskan) family in North America almost universally employ two copular verbs. In several languages of this family, copular forms are also employed as verbal auxiliaries: forms of one copula appear as markers of clausal focus while forms of the other mark TAM (tense/aspect/mode) categories. With reference to two Dene languages in particular, Tłı̨ chǫ Yatıı̀ and Tsuú̀ t'ı́nàı́nà , I explain this difference by positing distinct grammaticalization paths and motivations for each copula: both focus and TAM markers originate in a uniclausal reanalysis of biclausal constructions, the former from constructions where the matrix clause asserts the truth of the embedded clause, and the latter from those where the matrix clause supplies extra TAM information to the embedded clause. Both grammaticalizations involve an upward reanalysis of copulas as functional heads. Résumé Les langues de la famille Déné(DénéDéné(Athapascane) en AmériqueAmérique du Nord emploient presque universellement deux verbes copulaires. Dans plusieurs langues déneésdéneés, les formes copulaires sont aussi utiliseés comme auxil-iaires verbaux: les formes d'une copule marquent le focus sentential tandis que les formes de l'autre copule marquent les catégoriescatégories temporelles (temps / aspect / mode). A par-tir de deux langues déneésdéneés en particulier, le tłı̨ chǫ-yatıı̀ et le tsuú̀ t'ı́nàı́nà , j'explique cette différencedifférence en proposant des processus de grammaticalisation distincts ainsi que des motivations spéciiiquesspéciiiques pour chaque copule. Je propose qu'à la fois les marqueurs de focus et les marqueurs temporels sont tous deux issus de reánalyses uni-phrastiques de constructions à l'origine bi-phrastiques. Les marqueurs de focus dériventdérivent de constructions où la phrase matrice indique la valeur de véritédevéritévéritéde la phrase subordonneé, tandis que les marqueurs de temporaux dériventdérivent de constructions où la phrase matrice fournit de l'information temporelle additionelle à la phrase subordonneé. Les deux processus de grammaticalisation impliquent une reánalyse des copules comme tê tes fonctionnelles situeés plus haut dans l'arbre syntaxique.

Tłchǫ Yatıı ̀ , a Dene language of the Northwest Territories, Canada, has a number of post-verbal auxiliaries and particles indicating categories such as futurity, mode, negation, information structure, and evidentiality. The interaction of these elements reveals that they occur in a strict order, which in turn illuminates the structure of the clause in this language, with positions for future, mode, negation, and focus as functional categories at the right edge.

In Tłchǫ Yatıı ̀ (Dogrib), spoken in the Northwest Territories, Canada, the periphrastic particle ı ̨ lè, which has been considered a past marker, is optional in the clause. As in other Dene (Athapaskan) languages, viewpoint aspect is encoded morphologically on the verb. These two facts can give the impression that aspect is the only obligatory temporal category in this language and that other temporal distinctions are peripheral. I argue, on the contrary, that a future/non-future distinction is a necessary element of well-formed clauses, that Future, rather than Past/Non-Past Tense or Aspect, serves as an anchor in the sense of Enç (1987), and that anchoring does not necessarily correlate with the prominence of a temporal category in the sense of Bhat (1999). I adduce evidence in support of these proposals from the contrasting obligatoriness of past and future marking in predicates, as well as from word order facts.