Syntactic Ergativity in Tongan: Resumptive Pronouns Revisited

Introduction It is well known that so-called ergative languages are classified roughly into two groups: those which show an ergative pattern only at the morphological level and those which exhibit an ergative pattern also at the syntactic level. Syntactic ergativity is a curious phenomenon in the following respects. First, while there is apparently some correlation between morphological and syntactic ergativity, the link between the two appears to be rather weak. On the one hand, no language with accusative morphology shows syntactic ergativity. On the other, morphological ergativity does not always entail syntactic ergativity. Evidence from languages such as Warlpiri and Niuean, which are morphologically ergative, but does not show syntactic ergativity, suggests that morphological ergativity does not necessarily induce syntactic ergativity. Secondly, even when a language does show syntactic ergativity, an ergative pattern is sometimes restricted to certain construction types. Two questions arise. What gives rise to syntactic ergativity? How is syntactic ergativity related to morphological ergativity? Among the phenomena concerning syntactic ergativity, relativisation seems to show a fairly consistent pattern crosslinguistically. That is, syntactically ergative languages generally show an ergative pattern with respect to relativisation: while absolutive (ABS) arguments can undergo normal relativisation (i.e., the gap strategy), ergative (ERG) arguments cannot. This restriction is manifested in two ways: a) relativization of ERG arguments is strictly prohibited and therefore, the structure must be