LangSci Lunch Talk: Mike Fetters (LING)

Food and ideas bring people together.  Our weekly lunch talk series provides students and faculty with the opportunity to present their in-progress work to a supportive, interdisciplinary audience.

=====================================================

Sluicing as a Window into Knowledge of Islandhood in Children

Constraints on a certain type of long-distance dependency, called island constraints, have been a central focus of the syntax literature for several decades (Ross 1967, Chomsky 1973, Huang 1982, Chomsky 2001, inter alia). An illustration of an island constraint is given in (1) and (2), where what in (1) is interpreted as the instrument of the poking event, while in (2) no such reading is available. Several highly influential studies have investigated children’s knowledge of island constraints on extraction, with results ranging from mixed to adult-like performance (Otsu 1981, de Villiers, Roeper, and Vanikka 1990, de Villiers and Roeper 1995). Though these studies are often cited as evidence that preschoolers have acquired at least some of the relevant features of some island constraints, the experimental materials in the studies may be introducing a particular bias towards island-obeying responses independent of any feature of the child’s grammatical knowledge. Sluiced questions provide an excellent diagnostic of extralinguistic bias in studies concerned with the acquisition of islands, as these questions ameliorate island effects while retaining the semantic content of the analogous non-sluiced question (Ross 1969, Merchant 2001). In (3), the sluiced counterpart (2), it is possible to interpret that wh-item as the instrument of either the finding event of the poking event. The ambiguity of (3) allows for us to more directly observe the influence of discourse and processing features in the generation of a particular interpretation. Using a Question After Story task (de Villiers, Roeper, and Vanikka 1990) with questions like as (2) and (3) reveals that adults have a strong bias towards associating the wh-word with the matrix clause event (in this example, the finding event), even though the island constraint is not present in (3). Though data from children is still being collected, comparison of the child and adult responses will allow us to more carefully identify the joint contributions of the grammar and other interpretive mechanisms in how children understand long-distance dependencies.

(1) What did the cat poke the mouse with       ?

(2) What did Daisy find the cat that poked the mouse with       ?

(3) Daisy found the cat that poked the mouse with something. Can you tell me what?