Poster #114 - Korean Mothers Place Nouns in the Utterance-final Position Notwithstanding the SOV Word Order

• In Event: 2-110 - Poster Session 07
  In Poster Session: PS 07 Section - Language, Communication

Integrative Statement

Mothers across cultures adopt similar acoustic characteristics in addressing their children but differ in their strategies in teaching words (Fernald&Mazzie, 1991). For example, differences in English as “noun-dominant” relative to more “verb-dominant” languages like Korean (Gopnik&Choi, 1990) could yield variations in maternal speech style. This paper investigates the effects of Korean as a head-final on the mothers’ word teaching strategies based on corpus analyses and word-teaching tasks. We show that Korean mothers often present target words in the utterance-initial position, but they also repeat target words at the end of an utterance, similar to the English tag-question construction.

We first conducted a corpus analysis to investigated how Korean mothers present nouns to their children in comparison to American mothers. The Korean corpus was constructed by recording 35 mothers spontaneously interacting with their children of three age group: 0;9 (12 dyads), 1;1 (11 dyads), and 2;3 (12 dyads). Matching data of American mother-child dyads were selected from the Brent, Brown, Post, and Providence corpus in the CHILDES database (MacWhinney, 2000.). We selected 10 target nouns for analysis based on frequency in each mother-child data. We then coded the position (isolated, utterance-initial, utterance-final) and the presence of repetition (tagged, within-utterance, nearby-utterance). A total of 2812 Korean and 3774 English utterances were coded.

Multiple regression analyses show that age and language significantly predicts the use of repetition (R2=.35, F(3, 66)=11.9, p<0.001). Korean had a higher rate of repetition than English (β=21.9, p<0.01), and it decreased with age (β=-8.2, p<0.05). Both the ratio of tag-repetition and within-utterance repetition was explained by language, with Korean having a significantly higher rate of them than English (p<0.001). An interaction between age and language (β=-3.7, p<0.001) indicates that the ratio of tag-repetition is higher in speech to Korean younger children (Figure 1). The placement of focused words in the utterance-initial position is explained by language (R2=.76, F(3, 66)=70.59, p<0.001). The placement of focus at the right-edge, however, was explained both by language and age (R2=.83, F(3, 66)=106.2, p<0.001). An interaction between language and age (β=4.9, p<0.5) indicates that the ratio of focused words occurring at the right-edge is lower in speech to Korean older children (Figure 2).
In order to corroborate and supplement the findings in our corpus analyses, we devised an experiment where Korean mothers are specifically asked to teach words to their 9-month-old children. We provide a custom-made picture book to the mothers, who teach the labels of novel objects to their children and also to another adult. We will compare the ratio of repeating nouns in child- and adult-directed speech, and also conduct acoustic analyses for the target words in the two different speech registers.

Our findings suggest that Korean mothers mothers might have tacit knowledge of the potential benefit associated with placing the target word in utterance-final position such as edge-prominence and the recency effect. This result echoes previous finding in Turkish mothers who demonstrated a similar mutation of syntactic structure to take advantage of the utterance-final position (Aslin, 1993).

Authors

• Eon-Suk Ko, Chosun University
  Presenting Author

• Jihyo Kim, Chosun University
  Non-Presenting Author

• Jinyoung Jo, Seoul National University
  Non-Presenting Author