Volume 39, Issue 3 - Special Issue: Empirical Approaches to Elliptical Constructions; Guest Editor: Gabriela Bîlbîie

Ellipsis is a pervasive phenomenon across the world’s languages, and it is easy to see why: it allows speakers to omit certain parts of their utterances while nonetheless conveying their full meaning, which contributes to making linguistic communication highly efficient. While there is broad consensus that elliptical utterances depend on the context in some way, the nature of this dependency remains controversial. In this paper, I re-evaluate the merits of two classes of ellipsis theories: identity theories, which posit that material can be elided only if it is identical to a linguistic antecedent; and referential theories, which assume that ellipsis is enabled by the same underlying mechanism that governs other forms of discourse reference. I argue that both empirical and theoretical considerations favor referential theories in this comparison, and in doing so I outline new adequacy criteria for linguistic theories aimed at explaining the nature of the linguistic and non-linguistic context and how it interfaces with context-dependent linguistic devices.

Hungarian multiple sluicing has been claimed to only be allowed in contexts that set up a pair-list, but not a single-pair reading. This has been taken as evidence that multiple sluicing is derived from multiple wh-fronting questions, which only license pair-list, but not single-pair answers. Providing novel experimental evidence, we show that all three relevant constructions in Hungarian – multiple sluicing, single wh-fronting questions, and multiple wh-fronting questions – in fact pattern alike in their answerhood conditions: both pair-list and single-pair readings are allowed, with a modest preference for single-pair readings. Based on our results, we thus argue that answerhood conditions are not sufficient to determine the source of multiple sluicing.

This paper experimentally investigates the two generalizations for multiple sluicing (MS) recently presented by Klaus Abels and Veneeta Dayal: first, that wh-remnants must have clausemate correlates in the antecedent utterance and, second, that wh-remnants in MS can have correlates in the antecedent clause that are contained in a strong syntactic island. The fact that MS displays both of these properties is puzzling since island insensitivity under sluicing favors a non-sententialist approach to MS, while the clausemate requirement on MS is most straightforwardly explained by postulating a silent structure at the ellipsis site. Even though the clausemate condition has been reported in several languages, no experimental work has been conducted so far to examine its precise effects on the acceptability of MS constructions. In this paper, I will present the results of a series of experiments in German, English, and Spanish (employing both acceptability judgment tasks and a self-paced reading task), where the factors of clausemateness and islandhood have been examined systematically. The results provide solid cross-linguistic support for Abels and Dayal’s generalizations by showing that multiple sluices originating from islands and non-islands are equally acceptable and do not exhibit online processing differences. Furthermore, the acceptability judgment tasks show a significant degradation in acceptability when the correlates in the antecedent do not stem from the same finite clause, thus violating the clausemate condition. I will interpret these results as supporting a particular strand of sententialist research known as the island evasion approach and, in particular, defend that MS is derived from a non-isomorphic short source sluice.

Gapping elides a finite verb in the non-initial conjunct of a coordinate structure while VP ellipsis deletes a whole VP after an auxiliary. Unlike these two, pseudogapping elides most of the VP except one remnant. Pseudogapping additionally differs from gapping and VP ellipsis, in that it involves ellipsis of part of a non-finite VP. In this paper we provide a Construction Grammar account of pseudogapping that captures its similarities with as well as differences from other related elliptical constructions like VP ellipsis. Our construction-based analysis, which capitalizes on the inheritance network of constructions to capture broad similarities and unique differences among these constructions, allows us to account for the full range of extant data.

Additive fragments, comprising a nominal remnant and an additive adverb (e.g., too , either ), are a particular type of stripping. On the basis of new corpus data in English and French, we show that such fragments do not always have a verbal clause as their antecedent, and that when they do, different kinds of mismatch are possible between a verbal equivalent and the actual fragment. This challenges most approaches based on syntactic reconstruction. We also show that their interpretation is more flexible than previously thought, since they can be used for interrogative, exclamatory, or ordering purposes. We distinguish between their contrastive (non-coreferent) use ( A: John left. B: Me too .) and emphatic (coreferent) use (A: John left. B: HIM too!). We propose a direct interpretation analysis that resorts to no syntactic reconstruction of a verbal clause. The proposed analysis, developed within the framework of construction-based HPSG, allows us to capture not only their properties sharing with other fragments (short answers and negative stripping) but also their unique constructional properties.

Gapping in embedded environments may occur in two configurations: (i) the whole coordination containing both conjuncts is embedded (= embedded gapping, EG), (ii) the second (i.e. elliptical) clause is embedded within its own conjunct (= single conjunct embedded gapping, SCEG). Languages seem to differ in their restrictions on these two structures: EG in some languages does not allow for a complementizer in the elliptical conjunct, while it does in other languages. SCEG is outright unacceptable in some languages but acceptable in other languages. Overall, languages seem to fall into two groups such that one group allows a complementizer in the elliptical conjunct of EG and generally allows SCEG, whereas the other group allows neither. We present four experiments in Spanish on the acceptability of the complementizer que ‘that’ in the elliptical conjunct in EG. Our results suggest that que in Spanish EG is overall subject to similar restrictions as SCEG gapping: There are different degrees of degradation depending on the (type of) embedding verb without outright unacceptability. While the relevant property has been argued to be factivity for SCEG, we argue that it is not the factivity of the embedding verb as such that drives acceptability, but assertion embedding. We outline a theoretical proposal building on existing accounts of structural ambiguity in gapping, the truncation of complement CPs under some verbs including factives, and the general flexibility of the semantic/pragmatic categories factivity and assertion.

This paper empirically tests the embedding constraints on gapping in Persian. It has been suggested that gapping differs from other kinds of ellipsis in banning embedding. However, the first counter-examples in the literature come from Persian. Following up on previous experiments on embedded gapping in several languages, we report the results of two acceptability judgment tasks. Our results show that, while embedded gapping is overall acceptable in Persian, speakers’ acceptability judgements also vary depending on the semantic type of the embedding predicate, as well as the presence/absence of the complementizer. Data from Persian highlight that, despite the cross-linguistic variation observed with respect to the acceptability of embedded gapping, a general semantic constraint is at work across languages: non-factive verbs embed more easily than factive ones; inside factive verbs, semi-factive (cognitive) predicates embed more easily than true factive (emotive) ones. Moreover, whereas previous theoretical literature indicates no systematic preference for the absence or the presence of the complementizer in Persian, these new experimental data suggest a preference for complementizer drop. To account for the gradience observed in our experimental data, we propose an approach of gapping based on acceptability rather than grammaticality.