Context-Dependent Script Choice in Emails: The Case of Sumimasen

ABSTRACT

This study examines how pragmatic factors influence Japanese script choice by focusing on the representation of ‘sumimasen’ – a phrase used to express apology and gratitude. In standard Japanese, this term is written in hiragana, but use of katakana for the term has been observed. Through a quantitative survey targeting 200 undergraduate Japanese-L1 students and investigating their impressions of different representations of sumimasen in Japanese emails, this study examines how context influences reader impressions of sumimasen representations. Here, ‘context’ specifically refers to three pragmatic factors of communication: situation, writer-reader relationship, and purpose. Through examining how alterations to the situation and relationship factors influence survey responses, the study reveals that reader impressions of katakana sumimasen differ depending on the respondent’s preferred politeness strategy. In focusing on this understudied question of how L1 readers respond to orthographic variation in context, the study expands on existing accounts of how katakana creates meaning by recognizing that interpersonal factors can influence the intent and reception of script variation. Additionally, the study finds that the social meaning of a given script variant can be dependent on the word it represents, with terms like sumimasen becoming active and accepted vehicles for meaning negotiation.

Disclosure statement

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

Notes

¹ These documents can be viewed on the website of the Agency for Cultural Affairs: www.bunka.go.jp/kokugo_nihongo/sisaku/joho/joho/kijun/naikaku/index.html.

² Masuji (2019) contains a comprehensive review of the motivations covered in previous Japanese-language literature.

³ Masuji (2018: 2–3) goes into more detail and provides ten different categories. I have provided a simplified version here.

⁴ Sasahara (2013) reports some regional variation in kanji use and readings, partly deriving from how kanji spread geographically from China to Japan. However, to the best of my knowledge, there is no study that has identified trends in regional variation in script choice with regards to katakana.

⁵ The aforementioned introductory Japanese-language textbook, Genki, for example, lists sumimasen, konnnichiwa, and ohayō in hiragana (Banno, Ohno, Sakane & Shinagawa, 2004: 6–7). The dictionary contained in the kisha handbook [The journalists’ handbook], published by Kyōdō News and used by journalists in Japan, also lists ohayō (2016: 175), gomen (2016: 253) and konnichiwa (2016: 255) in hiragana.

⁶ This research was approved by the Ethics Committee of the Faculty of Human Informatics, Aichi Shukutoku University (05/07/2018, approval no. 2018-007). Informed consent was obtained from all participants before the survey form was completed.

⁷ These were chosen as the mostly likely descriptors for the katakana sumimasen by a group of six Japanese L1 speakers. The two negative descriptors, ‘insincere’ and ‘dislike’, were included to allow the author to check whether or not respondents were providing true responses (and that they were not suffering from respondent fatigue).

⁸ Studies such as Slugoski and Turnbull (1988) and Brown and Gilman (1989) have demonstrated some weaknesses in Brown and Levinson’s FTA calculation formula, especially in relation to the D factor. In short, they report that affect (how much the interlocutors like each other) is a significant aspect that should be factored into the formula. However, as the present study deals with hypothetical addressees, the impact of the level of affection between the speaker and addressee is also considered to be relatively fixed. Admittedly, there is room for distortion in the results if the respondent has a strong dislike (perhaps from prior experience) for people who do not complete tasks on time, for example.

⁹ In other words, respondents felt that the hiragana sumimasen was more thought-out and polite than the katakana sumimasen regardless of changes in the P variable, or whether they were the sender or the recipient of the email. This is probably because the hiragana version is considered the ‘standard’ representation. In Japanese, ‘thought-out (teinei)’ and ‘polite (reigi tadashii)’ are very close in meaning, so similar results would be expected in these two questions. The fact that the results were similar between these two questions throughout the study shows that respondents were not impacted by ‘response fatigue’.

¹⁰ Although there are no academic papers reporting this finding to date, this is likely to be an informal understanding of the katakana sumimasen shared by many individuals. Yahoo! Chiebukuro (the Japanese version of Yahoo! Answers), for example, contains many posts that discuss the katakana sumimasen negatively as ‘impolite’, ‘casual’ and ‘inappropriate’. However, there are also some users who either feel neutrally about the use of katakana or simply assume that the writer is a non-native speaker. This mix of responses supports our conclusion that there is individual variation in the perception of the katakana sumimasen.

Additional information

Funding

This work was supported by the Aichi Shukutoku University [18TT31].