Ergatives Move Too Early: On an Instance of Opacity in Syntax

Abstract

Abstract. In this paper we examine the ban onĀ-movement of the external argument of a transitive verb that holds in many morphologically ergative languages. We argue that the prohibition against movement of the ergative subject should not be derived from restrictions on the movement of the ergative DP. Rather, we suggest that movement of the ergative argument is per se unproblematic, but if it applies, it applies too early, and thereby creates problems for its absolutive co-argument, which does not receive structural case. In morphologically accusative languages, no such movement asymmetry arises because arguments move too late to trigger the fatal consequences that moving ergatives cause. We present a relational, co-argument-based analysis that implies a strictly derivational syntax in which the order of operations plays an important role in deriving properties of the grammar.

The role of case in A-bar extraction asymmetries

“…Since only some morphologically ergative languages exhibit extraction asymmetries, tying this point of variation to an independently motivated parameter -the locus of case for the absolutive object, as illustrated in (3) -is a welcome result. The parameterization in (3) thus represents an advantage over proposals like Assmann et al 2013, which connects the ban on extracting ergatives to the same parameter which governs whether a language is morphologically ergative or morphologically accusative (via a general parameterization of the ordering of Merge and Agree operations). The present work shares with Assmann et al 2013 the idea that extraction asymmetries come about as a locality problem in clausal case assignment, but differs in that it does not make syntactic ergativity an expected consequence of morphological ergativity.…”

“…The parameterization in (3) thus represents an advantage over proposals like Assmann et al 2013, which connects the ban on extracting ergatives to the same parameter which governs whether a language is morphologically ergative or morphologically accusative (via a general parameterization of the ordering of Merge and Agree operations). The present work shares with Assmann et al 2013 the idea that extraction asymmetries come about as a locality problem in clausal case assignment, but differs in that it does not make syntactic ergativity an expected consequence of morphological ergativity. Within the subset of morphologically ergative languages which display extraction asymmetries (at the top right in (3)), we recognize a further possibility for variation in the mechanisms available to circumvent the ban on ergative extraction.…”

“…Instead, we argue that syntactic ergativity -at least in these languages -is the result of properties of case-assignment to absolutive arguments (see also Aldridge 2004;Assmann et al 2013;Bittner & Hale 1996a;Campana 1992). Accounts which reduce the ergative extraction ban to properties of the ergative noun-phrase itself face serious problems in Mayan.…”

Remnant movement in local derivational grammar

“…In contrast, if an intermediate step of α occurs after absolutive/nominative assignment by T, it comes too late to create problems, and there is counter-bleeding. In Assmann et al (2012), it is argued that there is independent evidence that ergative DPs move early (before Agree operations have taken place), whereas accusative DPs move late (after Agree operations have taken place; see Müller (2009)), yielding a bleeding effect in the first case, and a counterbleeding effect in the second. This latter consequence poses a problem from a declarative infinitives, and this is deleted by Equi NP deletion (or some other rule), then this latter rule must apply early, so that it can feed wanna-contraction.…”

“…4 As a third case exhibiting opaque interaction in syntax, consider the analysis of the prohibition against movement of ergative DPs given in Assmann et al (2012). The generalization to be derived is that in many ergative systems, an ergative DP cannot undergo A-bar movement (wh-movement, topicaliztion, relativization, focus movement), in contrast to an absolutive DP.…”