Paper short abstract:

Using a corpus of baby names, this paper evaluates claims that Japanese names are becoming difficult to read. A majority of names appeared only once and used kanji in likely difficult ways. The results' sociolinguistic impact will be considered given proposed restrictions to kanji readings in names.

Paper long abstract:

The Japanese government is presently proposing changes to the family register laws that may have the single largest impact on writing in recent years: Adding space in the family register for the phonological forms of names, and, pending recommendations to be announced in 2023, restricting the kanji readings permitted in names for the first time ever. Although partially motivated by concerns that names are increasingly difficult to read (Satō, 2007), the impact of these potential changes has been hard to ascertain given insufficient quantitative research on Japanese names and the lack of reliable public naming data. The question has remained, are recent names really objectively difficult?

This paper approaches the question using data from a corpus of 2,627 matched phonological/graphic forms of children’s names extracted from announcements in municipal newsletters, which have been shown to be a highly reliable resource on Japanese naming (Unser-Schutz, 2018). To determine their difficulty, the names were categorized as phonologically transparent (used established kun/on/nanorikun readings and did not mix reading types, making their phonological forms more predictable) or phonologically opaque (used non-established ateji readings, mixed reading types, or altered established readings, making their phonological forms less predictable).

The majority of names appeared only once, and 57.21% were phonologically opaque, indicating that a majority were both diverse and potentially difficult. However, non-established ateji readings were uncommon compared with mixed kun/on readings or altered established readings, suggesting that while recent names are difficult, most are connected to customary kanji readings. Girls’ names were more likely to be phonologically opaque (63.82% vs. 51.49%), suggesting this may be a gendered trend. However, with the pending legal changes, these trends may be forcefully curtailed: Under the strongest proposal, many such phonologically opaque names would not be permitted. As such, the paper will consider the sociolinguistic impact of the proposed legal changes and the impact they may have on creative kanji uses beyond names.

Satō, M. (2007). Yominikui namae wa naze fueta ka. Tokyo: Yoshikawa-kōbunkan.

Unser-Schutz, G. (2018). Shiryō toshite nihon no nazuke ni kansuru kenkyū ni kōhōshi o mochiiru kanōsei ni tsuite. Risshō Daigaku Shinrigaku Kenkyū Nenpō, 9, 23–35.