Early Acquisition of Plural Morphology in a Classifier Language: Data from Korean 2-4 Year Olds

ABSTRACT

Korean is a classifier language in which bare nouns are not obligatorily number-marked. Children learning other classifier languages like Japanese and Mandarin are late in learning the plural morpheme. In this paper, we present two datasets that suggest that Korean plural marker -tul is acquired much earlier, in contrast to what has been previously claimed. In a comprehension study, we find that Korean children begin acquiring this morpheme by age 3, showing adult-like performance by age 4. We suggest that the higher frequency of plural marking on both types and tokens of nouns and the consistent marking of plural in the domain of definite nouns may facilitate Korean plural acquisition.

Acknowledgement

The authors would like to thank Chungmin Lee, Sudha Arunachalam, and the anonymous reviewers for their helpful feedback and suggestions. We would also like to thank Khyunghee Kindergarten, Seoul National University Center for Child Educare Service and Research, and Hoegi Preschool for their help in recruiting participants, and all of our adult and child participants in Seoul, Korea for their active participation in the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ -tul has been shown to indicate distributivity in addition to plurality (C.-E. Kim, 2012, a.o.), especially when it appears on non-nominal elements such as verbs and adjectives. This use is called the Extrinsic Plural Marker (EPM, in contrast to Intrinsic Plural Marker, IPM; C.-E. Kim, 2012) or a ‘plural copy’ (C.-H. Kim, 2005; J.-J. Song, 1997; S.-C. Song, 1975; Sohn, 1999). The plural copy can be added on multiple elements within a sentence, as shown in (a).

(a)namca-tul-iphwungsenhana-lul-tulsasseyo-tul.

man-pl(ipm) -nomballoonone-acc-pl(epm)bought-pl(epm)

‘Each of the men bought a balloon.’ (modified from C.-E. Kim, 2012; (2b))

In this paper, we focus on the nominal –tul (IPM) which contributes a plural meaning, and refer interested readers to C.-H. Kim (2005), C.-E. Kim (2012), Sohn (1999), Song (1975, and Song (1997) for detailed discussion of the other meanings.

² Because there was no variability in the 4-year-olds in the Picture Matching task, the model did not converge. Thus, for calculating the at-chance performance of 4-year-olds in the Picture Matching task, we ran a binomial test treating each participant as a trial.

³ The size of the symbol represents the number of participants in each point: the smallest circle represents one participant, while the largest circle represents 29 participants (the intermediate sizes represent 2, 3, 4, 7, and 8, respectively).

⁴ We also ran a separate analysis comparing the frequency of plural-marked nouns reported in Sarnecka et al. for Japanese with age-matched Korean data, which also turn out to be significant (Fisher’s Exact test, p<.0001).

⁵ A reviewer pointed out that in previous studies such as Suh, 2008, native speakers who were given a fill-in-the-blank task produced bare nouns in contexts requiring specificity, which conflicts with the data presented here. However, in Suh’s study, the contexts did not require an anaphoric reading, enabling an indefinite reading. When we replicated the sentences and provided contextual description ensuring that the noun refers anaphorically to a plural entity, five out of five native speakers consulted separately commented that the plural marking is required in both (i) and (ii).

(i)Context: Yesterday, Hoseok met three students. He met them again.

Hoseok-un ku_________-ul tto mannassta.Hoseok-top dem accagainmet‘Hoseok met ____ again.’

(ii)Context: Seokjin and Taehyung went to a pond with many frogs. Seokjin wants Taehyung to catch five frogs. What should Seokjin say to Taehyung?ce_________-ul/lul capacwe.demacc catch.imp ‘Please catch _______.’