Coordination in Tongan: Is Syntactic Ergativity Real?

Yuko Otsuka

University of Hawai‘i at Manoa

Syntactic ergativity is a curious phenomenon in the following respects. On the one hand, there is apparently some correlation between morphological and syntactic ergativity: no language with accusative morphology shows syntactic ergativity. On the other, some morphologically ergative languages do not show syntactic ergativity (e.g., Warlpiri). Furthermore, even in those languages that show syntactic ergativity, an ergative pattern is often restricted to certain construction types. While some have proposed that syntactic ergativity is a parametric variation (Bittner and Hale 1996, Ura 2000), such a claim runs into difficulties when syntactic split is taken into consideration.

In this paper, I argue that syntactic ergativity is a consequence of the interaction of the following two factors: morphological ergativity and the fact that syntactic rules are either Case-sensitive or theta-role sensitive. An ergative pattern arises when the relevant operation is Case-sensitive. In other words, it is the property of a particular operation, rather than the language itself, that causes syntactic ergativity. The current study provides an analysis of syntactic split in the Minimalist framework (Chomsky 1995, 2000).

            Tongan shows the following syntactic split: an ergative pattern with relativization and one type of coordination (pea), and an accusative pattern with the other type of coordination (mo) (Dixon 1979, Otsuka 2000). With regard to relativization, Otsuka (2002) shows that it is a Case sensitive operation. The contrast between the two types of coordination is illustrated in (1) and (2). Note also that when a pronoun follows instead of a gap, the relevant constraint is no longer effective in pea-coordination but remains valid in mo-coordination. See (3). This suggests that it is a condition on deletion with pea, but with mo, it is a condition on coordination itself.

            Assuming that deletion under coodination is a PF operation, I propose that deletion in pea-coordination must be licensed by feature-matching, including Case features. This deviates from Chomsky’s original proposal, in which Case feature is considered to be deleted after checking. I argue, however, that Case features on NPs remain accessible after checking. This is also proposed on independent grounds in Otsuka’s (2002) analysis of resumptive pronouns. As for the gap in mo-coordination, I propose that licensing must take place in syntax rather than PF. Otsuka (2000) proposes that mo requires that theta-identification in the sense of Higginbotham (1985) apply obligatorily to [Spec, vP] of the two clauses it conjoins. This hypothesis is supported by the fact that raising verbs, which are unarguably unaccusative, cannot occur in mo-construction, as illustrated in (4). Obligatory theta-identification ensures that the two clauses conjoined by mo share the same subject. Otherwise, the derivation would be cancelled before reaching PF. Once the second NP is licensed in this way, the deletion rule applies subsequently in PF.

In summary, the current study shows that the syntactic split in Tongan results from the fact that pea-coordination is Case sensitive, while mo-coordination is theta-role sensitive. This argues against the view which assumes the existence of a parameter that distinguishes syntactically ergative languages from those which are not.

(1)        a.            Na’e tangi ‘a Hina_i pea taa’i ‘e Mele e_i.

                        Pst cry ABS Hina  and hit ERG Mele

                        “Hina cried and Mele hit (her).”

            b.            *Na’e tangi ‘a Hina_i pea taa’i e_i ‘a Mele.

                        Pst  cry ABS Hina and hit  ABS Mele

                        “Hina cried and (she) hit Mele.”

c.            Na’e taa’i ‘e Hina_i ‘a Mele_j pea tangi e_*i/j.

Pst hit ERG Hina ABS Mele and cry

“Hina hit Mele and (*H/M) cried.”

(2)        a.            *Na’e tangi ‘a Hina_i mo taa’i ‘e Mele e_i.

            Pst  cry ABS Hina and hit ERG Mele

            “Hina cried and Mele hit (her).”

b.            Na’e tangi ‘a Hina_i mo taa’i e_i ‘a Mele.

            Pst cry ABS Hina and hit ABS Mele

            “Hina cried and (she) hit Mele.”

c.            Na’e taa’i ‘e Hina_i ‘a Mele_j mo kata e_i/*j.

            Pst hit ERG Hina ABS Mele and laugh

            “Hina hit Mele and (H/*M) laughed.”

(3)        a.            Na’e tangi ‘a Hina_i pea ne_i taa’i ‘a Mele.

Pst cry ABS Hina and 3.s. hit ABS Mele

 “Hina cried and she hit Mele.”

b.            Na’e taa’i ‘e Hina_i ‘a Mele_j pea ne_i/j tangi.

Pst hit ERG Hina ABS Mele and 3.s. cry

 “Hina hit Mele and Mele/Hina cried.”

c.            *Na’e tangi ‘a Hina_i mo taa’i ia_i ‘e Mele.

                        Pst cry ABS Hina and hit 3.s. ERG Mele

            “Hina cried and Mele hit her.”

(4)        a.            *‘Oku puke ‘a Mele mo tapu pro ke mou hū ki hono loki.

            Prs sick ABS Mele and forbidden that 2.pl. enter to poss.3.s. room

            “Mele is sick and it is forbidden that you enter her room.”

b.            *Na’e ‘ikai pro ke ha’u ‘a Sione mo ‘ita lahi ‘a Mele.

            Pst not  ke come ABS Sione and angry much ABS Mele

            Lit. “It wasn’t that Sione come and Mele was angry.”

References:

Bittner, Maria and Hale, Ken (1996) Ergativity: Towards a theory of a heterogeneous class. Linguistic Inquiry 27: 531-604.

Chomsky, Noam (1995) The Minimalist Program. Cambridge, Mass.: MIT Press.

Chomsky, Noam (2000) Minimalist inquiries: The framework. In R. Martin, D. Michaels, and J. Uriagereka eds. Step by step, Cambridge, Mass.: MIT Press. pp.91-155.

Higginbotham, James (1985) On semantics. Linguistic Inquiry 16: 547-593.

Otsuka, Yuko (2000) Ergativity in Tongan. Doctoral thesis. Oxford University.

Otsuka, Yuko (2002) Syntactic ergativity in Tongan: Resumptive pronouns revisited. Paper presented at Workshop on ergativity, 2002. University of Toronto.

Ura, Hiroyuki (2000) Checking theory and grammatical functions in Universal Grammar. Oxford: Oxford University Press.