In a seminal study, Stager and Werker (1997) reported findings showing that 14-month-old English-learning infants have the

This paper presents the results of a novel experimental approach to relative quantifier scope in German that elicits data in a less direct manner. Applying the covered-box method (Huang et al. 2013; Sayeed et al. 2019) to scope phenomena, we show that inverse scope is available to some extent in the free constituent order language German, thereby validating both earlier findings on other syntactic configurations in German (Radó & Bott 2018) and empirical claims on other free constituent languages (Japanese, Russian, Hindi), as well as recent corpus findings in Webelhuth (2020). Moreover, the results of the indirect covered-box experiment replicate findings from an earlier direct-query experiment on parallel target items, in which participants were asked directly about the availability of surface scope and inverse scope readings. The configuration of interest was constituted by canonical transitive clauses with deaccented existential subject and universal object QPs in which the restriction of the universal QP is provided by the context.

We present two off-line experiments on the interpretation of doubly quantified sentences with existential subject and universal object in German and English, which have been reported to allow for inverse readings only in English. We show for this specific syntactic configuration that there is no categorical cross-linguistic difference, but only a gradual one, with English more readily allowing for inverse scope than German. This supports a cross-linguistically unified analysis of inverse scope on which gradual differences between languages follow from language-specific properties and exposure effects. The results moreover suggest that relative clauses with indefinite head NPs allow for inverse readings, thereby challenging their status as scope islands (May 1977, Huang 1995), while being in line with introspective claims in semantic accounts (Barker 2012, 2019). Finally, our results suggest a high between-speaker variability and a strong impact of pragmatics.

No abstract