Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Languages and Cultures

First Advisor

Robert M. Hammond

Committee Chair

Robert M. Hammond

Committee Member 1

Lori A. Czerwionka

Committee Member 2

Elaine J. Francis

Committee Member 3

Daniel J. Olson


This study explores the potential effect of a second language (L2) on first language (L1) encoding of motion events. The domain of interest is MANNER and the goal is to investigate if the degree of manner salience can be restructured under the effect of a L2. Slobin (2004, 2006) proposes an expansion of Talmy’s (1985, 1991, 2000) binary typology and observes that the degree of manner saliencevaries cross-linguistically. The two languages investigated in this study, Spanish and English, are at divergent points along the cline of manner salience. In addition, Slobin (1996b) suggests dividing MANNER into tier one (T1) manner and tier two (T2) manner. T1-MANNER is available in both Spanish and English, but T2-MANNER is not readily available in Spanish.

Thus, it is postulated that if L2 transfer of manner salience occurs, a strong piece of evidence would be to observe an increase in the encoding of T2-MANNER in Spanish. In order to test this idea, the methodology and some of the stimuli from Sakurai (2014) were adapted. The experimental group consisted of adult L1-Spanish/L2-English bilinguals (n = 11 females; n = 19 males; M = 34.23 years of age, with SD = 10.32) and the control group consisted of adult L1-English speakers (n = 7 females; n = 13 males; M = 33.55 years of age, with SD = 11.91). There were two linguistic tasks and two non-linguistic tasks. The linguistic tasks involved narrating stories from Mayer’s (1969) picture book Frog, Where Are You? and from a custom-made animation created from episodes of Gazoon (Villemaine & Trouvé, 2007). The two non-linguistic tasks consisted of a MANNER/PATH categorical task and a MANNER/PATH similarity task originally designed by Sakurai (2014) and modified in the current study. In addition, the Bilingual Language Profile, BLP (Birdsong et al., 2012) was administered to the experimental group in order to assess the relationship (if any) between the L2 dominance scores and the performance in the tasks.

The results show that both groups encoded more MANNER in the second linguistic task (the animation) as compared to the first linguistic task (the picture book). There are no statistically significant differences between groups for the proportions of MANNER encoded in the linguistic tasks. However, there is a significant positive correlation (p < 0.01) between the L2 dominance scores and the encoding of T2-MANNER in the animation. This suggests that the effect goes beyond variances in stimulus type: subjects with high L2 dominance scores produced more MANNER expressions characteristic of their L2. The results for the non-linguistic tasks show that the L1-English speakers preferred MANNER to a significantly greater degree than the bilinguals in the first task (p < 0.01). These results conform to the expected lexicalization patterns. There are no significant differences among groups for the second non-linguistic task in regards to MANNER. However, there are significant correlations (p < 0.05) between L2 dominance score and these results. The higher the L2 dominance score, the higher the average MANNER rating and the lower the average PATH rating. Further analyses reveal that the initial between-group difference in the categorical task disappears when the degree of L2 dominance is taken into account. That is, the subjects with L2 dominance scores above the median preferred and rated MANNER in a similar way to the control group in both non-linguistic tasks. Overall, these preliminary findings support the idea of a L2 effect on motion event cognition which could make MANNER more salient in the L1. These results have implications in the fields of cognitive linguistics, linguistic relativity, linguistic typology, second language acquisition, and motion-event experimentation. More data needs to be collected to further validate these results.