Grammatical feature dissimilarities make relative clauses easier: A comprehension study with Italian children

Flavia Adani,Heather K.J Van Der Lely,Matteo Forgiarini,Maria Teresa Guasti

doi:10.1016/j.lingua.2010.03.018

Flavia Adani, Heather K.J Van Der Lely + Show 2 more

Open Access

https://doi.org/10.1016/j.lingua.2010.03.018

Copy DOI

Abstract

The Relativized Minimality approach to A′-dependencies (Friedmann et al., 2009) predicts that headed object relative clauses (RCs) and which—questions are the most difficult, due to the presence of a lexical restriction on both the subject and the object DP which creates intervention. We investigated comprehension of center-embedded headed object RCs with Italian children, where Number and Gender feature values on subject and object DPs are manipulated. We found that, Number conditions are always more accurate than Gender ones, showing that intervention is sensitive to DP-internal structure. We propose a finer definition of the lexical restriction where external and syntactically active features (such as Number) reduce intervention whereas internal and (possibly) lexicalized features (such as Gender) do so to a lesser extent. Our results are also compatible with a memory interference approach in which the human parser is sensitive to highly specific properties of the linguistic input, such as the cue-based model (Van Dyke, 2007).

Highlights

It is a well documented fact in the psycholinguistic literature on head-initial languages that subject relative clauses (RCs) (1a) are generally unproblematic whereas object RCs (1b) are more difficult for children to comprehend and they are costlier for adults to process
Both subject and object RCs involve the interpretation of a constituent that is displaced with respect to the argument position where, in minimalist terms, it was merged: 1)
We addressed the question as to whether Number and Gender features modulate the comprehension of object RCs by Italian children and, if so, in what way

Summary

Introduction

It is a well documented fact in the psycholinguistic literature on head-initial languages that subject RCs (1a) are generally unproblematic whereas object RCs (1b) are more difficult for children to comprehend and they are costlier for adults to process. Both subject and object RCs involve the interpretation of a constituent (i.e. the RC head, the dog in 1) that is displaced with respect to the argument position where, in minimalist terms, it was merged (the position indicated in angled brackets): 1) a. Object RCs have been attested to be hard both in comprehension and production as well as across different populations: adults

Objectives

Methods

Results

Conclusion