(Non-)factive (non-)islands and meaning-based approaches

A key question in the literature on factive Weak Islands has been whether the effect is syntactic or semantic. Since Szabolcsi & Zwarts (1993), a key argument for the semantic nature of Weak Islands is the observation that the effect requires not just factivity, but also that the property described by the embedded clause is non-iterable with respect to the extracted argument (uniqueness). We present twocaveats concerning the notion of factivity needed in meaning-based approaches. First, we present novel data on factive non-islands showing that certain lexically factive verbs do not (always) lead to islandhood when combined with uniqueness. Second, recalling data from Cattell (1978), we argue that certain non-factive islands can be captured by the same meaning-based explanation. The emerging picture is that lexical factivity of the embedding verb is neither necessary nor sufficient to induce weak islands in combination with uniqueness; rather, what matters is whether or not there is a contextual entailment, pragmatic or lexical, that the complement proposition is true.

The Scrambling Paradox

It has been commonly observed that scrambling and wh-movement share sensitivity to strong movement constraints ( Webelhuth 1989 , Saito 1992 , Bailyn 1995 ). At the same time, the two processes clearly differ in certain other respects, such as wh-island sensitivity, a finding that has inspired a range of analyses of scrambling as entirely distinct from better-understood movement processes ( Müller and Sternefeld 1993 , Bošković and Takahashi 1998 , among many others). Careful comparison of Ā-scrambling and overt wh-movement in a language that shows both (Russian) reveals that this seemingly paradoxical behavior can be captured effectively in a probe-goal theory of scrambling that obeys a form of Relativized Minimality defined across feature classes, following Rizzi 2004 . The resulting analysis exposes the distinct nature of strong and weak islands, with consequences for our understanding of the core architecture of syntactic movement.

Intervention effects in grammar and language acquisition

Intervention effects have been thoroughly studied in formal syntax in the domain of weak islands. They also have recently been appealed to in the study of language acquisition, to capture certain difficulties that young children manifest in the mastery of some object A’-bar dependencies (relatives, questions, topicalizations). Can one unify such distinct utilizations of the concept of intervention under a single formal locality principle? This paper explores the possibility of a unitary approach by proposing solutions for observed discrepancies between the effects in adults and children, and more generally between the different utilizations of the concept of intervention in recent work on adult grammar and language acquisition. Relativized Minimality (RM) is seen as a formal principle penalizing configurations as a function of the distinctness between target and intervener in local relations, where distinctness is precisely expressed as a grammar-based notion. A unitary system consisting of RM and an explicit distinctness hierarchy is argued to be operative in intervention effects in grammar and language acquisition.

Structural and semantic ambiguity of why-questions: An overlooked case of weak islands in English

We argue that English why-questions are systematically ambiguous between a purpose and a reason interpretation, similarly to Mandarin, Russian, and Polish (contra Stepanov & Tsai 2008). We argue that the distinct semantic interpretations correspond to two distinct base-generated positions of why. While reason why is base-generated within CP (Rizzi 2001, Ko 2005), purpose why is adjoined to vP (Stepanov & Tsai 2008). Furthermore, we show that English purpose why, similarly to previously reported data from Mandarin, is only compatible with dynamic predicates with agentive subjects. We argue that this selectional restriction follows from two properties: (i) why semantically requires a proposition as its argument, and (ii) only dynamic predicates with agentive subjects have a syntactic structure that accommodates two adjunction sites of the relevant semantic type, i.e., they contain two distinct propositional levels (Bale 2007) and therefore two attachment sites for why. In contrast, propositionally simple predicates only have one propositional level and hence only one possible attachment site, which corresponds to the reason interpretation of why. Evidence for this proposal comes from the observation that only the lower why - associated with the purpose reading - is sensitive to negative islands, which suggests that its attachment site is below negation (vP), whereas the higher why is insensitive to island effects of this sort, which suggests that its base generated position is above negation (CP).

Swedish relative clauses as weak island

<p>Some recent accounts of relative clause extraction (RCE) in Swedish assume that clauses that allow extraction do not themselves involve A-bar dependencies, and that RCE is possible only from subject relatives (e.g. Kush et al. 2013). I present evidence that Swedish allows A-bar movement from non-subject RCs as well. But not just any type of phrase can be extracted. For example, certain non-argument wh-phrases cannot move out. This means that Swedish RCs are weak, rather than strong islands (cf. Szabolcsi 2006). Szabolcsi takes an algebraic approach to weak islands where phrases that denote individuals, which can be collected into sets forming Boolean algebras, can be extracted, whereas phrases that denote non-individuals, which cannot be collected into such sets, cannot. However, it is not obvious how to extend such an approach to Swedish RCs, since they allow extraction of some phrases that denote non-individuals, like ‘how late’, as long as they are linked to the discourse. Instead, I propose that the phrases that can move out of relative clauses carry discourse-related features (DR), and that the C-heads in Swedish RCs attract DR-marked phrases, making them available in later stages of the derivation. </p>

Quantificational and modal interveners in degree constructions

Heim (2001) points out that the relational semantics for degrees predicts ambiguities in sentences with comparatives and quantifiers such as _every girl_ that are not observed. She also notes that the same ambiguities do appear with strong modals such as _must_ and _have to_, but not with weaker modals such as _should_, _ought_, and _want_. The problem is to explain why these classes of expressions would behave differently, given that they are all standardly treated as universal quantifiers. I present several counter-examples to Heim's account of this data and then argue that the puzzle involving universal DPs is the same as the puzzle of weak islands in amount wh-expressions, and that it yields to the analysis of weak islands due to Szabolcsi & Zwarts (1993), who argue that degree expressions are restricted in their interaction with the semantic operations meet. This accounts for universal DPs but leaves to be explained the possibility of modal intervention with strong modals. I argue that the split between universal DPs and strong modals supports recent work proposing that modals are not quantifiers over worlds but scalar expressions. An independently motivated scalar semantics for strong modals generates the ambiguity in a way that is compatible with Szabolcsi & Zwarts' theory, and that the predicted truth-conditions are correct for both readings with strong modals. The corresponding account of mid-strength modals explains their lack of ambiguity as merely apparent, due to the fact that the truth-conditions of the two readings are virtually indistinguishable, and neither embodies the missing reading that the quantificational theory leads us to look for. These results support both the scalar semantics for modality and Szabolcsi & Zwarts' semantic approach to intervention constraints.