The knowledge of binding principles in early child grammar: Experimental evidence from 30-month-old toddlers

ABSTRACT

We investigated toddlers’ understanding of the hierarchical syntactic configurations that constrain the referential meanings of reflexives and pronouns. In particular, reflexives must co-refer with the c-commanding antecedent within the local domain (Principle A) (e.g., He_i washes himself_i. John_i knows that Bill_j likes himself_j.), and pronoun reference must be disjoint from its local c-commanding antecedent (Principle B) (e.g., He_i washes him_j. John_i knows that Bill_j likes him_i.). French-learning 30-month-olds participated in an eye-tracker task. In Experiment 1 each trial presented a simple sentence containing a reflexive or a pronoun and a target verb (e.g., Il_i se_i/le_j lave. “he_i washes himself_i/him_j”), while displaying side-by-side pictures depicting self- directed versus other-directed actions of the target verb. Experiment 2 tested the locality conditions of the principles using complex sentences, with the target verb and the reflexive/pronoun appearing in the embedded clause (e.g., Le bébé_i voit que le papa_j se_j/le_i lave. “The baby_i sees that the dad_j washes himself_j/him_i”); the pictures were those of Experiment 1, with the patient of the verb being the local antecedent in the embedded clause versus the non-local antecedent in the matrix-clause. Experiment 3 tested the c-commanding requirements of the principles with sentences containing a complex subject, a reflexive/pronoun and a target verb (e.g., Le bébé_i du papa_j se_i/le_j lave. “The baby_i of the dad_j washes himself_i/him_j”). In the pictures the patient was the linearly farther versus closer antecedent. Eye-gazes to action images showed adult-like interpretations across sentential structures, demonstrating that binding principles are in child grammar during age two.

Acknowledgments

We are grateful to Julie Franck for insightful discussions that inspired us to conduct this line of research. We thank all the parents and toddlers who participated in the study.

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The data that support the findings of this study are openly available on the OSF (Open Science Framework) at this link: https://osf.io/h5gcm/.

Notes

¹ Bound means c-commanded by and co-indexed with an antecedent. C-command means that node A c-commands node B if A and B do not dominate each other and the first branching node dominating A also dominates B.

² Roughly speaking, local domain can be viewed as referring to the minimal clause containing the anaphor or the pronoun.

³ The object pronouns le, la, les, and l’ in French are also determiners (‘the’ in different genders and numbers). Previous work showed that French-learning 18-month-olds can distinguish the different syntactic categories of these homophonous function words in sentential context (Brusini et al. 2017).

⁴ One of our pronoun/reflexive contrasts involved the contracted l’ versus s’ forms. These forms were syllabified with the following vowel-initial verb habiller (Elle l’habille versus Elle s’habille). Infants’ performance for this contrast resembled that for the uncontracted forms (le versus se). In the literature, similar forms of consonantal prefixes that were syllabified with the following vowel-initial verb (e.g., /l/arrive ‘3^rd-person-singular-prefix arrive’; /z/arrivent ‘3^rd-person-plural-prefix arrive’) were correctly processed for subject-verb agreement by French-learning infants in perceptual and comprehension experiments (14-month-olds in Culbertson et al. 2016, 24-month-olds in Koulaguina et al. 2019, 30-month-olds in Legendre et al. 2010).

⁵ We note that since the sentences in Experiment 3 were mono-clausal, they could not help clarifying how much the toddlers in Experiment 2 understood about the binding relation between the matrix NP and the pronoun/reflexive in the embedded clause. Children could have succeeded in Experiment 2 with the binding knowledge of the local clause alone, without knowing that the pronoun, but not the reflexive, can be bound by the matrix NP. This strategy would be compatible with Principles A and B of Binding Theory.

⁶ The graph seems to show two other windows earlier in the trials when we eye-inspect the time course, although they were not identified as significant by the cluster analysis. In the Pronoun trials the bump (i.e., increased looking) towards the other-directed events occurred around 3,000 msec, during the pre-window. In the Reflexive trials the bump, also towards the other-directed events, occurred at around 6,500 msec, which was after the onset of the reflexive (at 5,427 msec or before). It is unclear if these bumps indicate a general visual bias towards other-directed events in toddlers before they fully analyze the sentence. Such bumps were not present in Experiments 1 and 2, suggesting that Experiment 3 was perhaps harder for toddlers.

Additional information

Funding

This research was supported by funding from Natural Sciences and Engineering Research Council of Canada (NSERC) and Canadian Foundation for Innovation (CFI) to the second author.