Grammaticalization

Demonstratives have been identified as the source items for various grammaticalization processes, including development into copulas. In this paper, I provide evidence of grammaticalization of demonstratives into copulas in Passamaquoddy (Algonquian), and, taking into account the facts of Passamaquoddy word order and information structure, I propose grammaticalization pathways different from both the topic-comment reanalysis and sentence-afterthought reanalysis accounts of Li and Thompson (1977) and Gildea (1993) respectively.

Functionally, demonstratives are commonly defined as deictic morphemes that orient the addressee's attention to something in the speech situation or in the linguistic discourse. In Passamaquoddy, such entity-referring demonstratives agree in number, animacy, obviation, and absentativity with the referent nominal, and must precede that nominal if it is overtly present. There are three deictic distances - near-speaker (nS), near-addressee (nA), and away-from-speaker-and-addressee (aSA).

In a number of verbless constructions with either one or two nominal terms, words with the forms of near-addressee demonstratives are obligatory, occurring after the term in one-term sentences and between the two terms in two-term sentences. In some constructions, as in [1], the demonstrative word may agree in number and animacy with the argument nominal; the animate singular form not agrees with the animate singular nil 'I', although the inanimate singular form nit does not. In other constructions, as shown in [2], the demonstrative word is always morphologically invariant - the inanimate singular form nit is used with animate singular nil 'I' and wot 'this [animate]' and with animate plural nilun 'we'. The grammatical properties of the demonstrative words in [1]-[2] suggest that they are no longer entity-referring demonstratives, but have undergone grammaticalization into copulas.

Given the data, I propose that the structures which initially resulted in copula reanalyses are those where the first term is focused information and the second term is the topic. For one-term constructions, this is consistent with the facts of sentences like [2], where the term is focused information and occurs clause-initially. For two-term constructions, the hypothesis is consistent with the position of the demonstrative word, as well as the fact that sentences like [1] can now be interpreted with an indefinite second term as long as the context is one where the first term is still focused information.

References:

Gildea, Spike. 1993. The development of tense markers from demonstrative pronouns in Panare (Cariban). Studies in Language 17: 53-73.
Li, Charles, and Sandra Thompson. 1977. A mechanism for the development of copula morphemes. In Li, Charles (ed.), Mechanisms of syntactic change, 414-444. Austin: University of Texas Press.