Eugenia Diegoli (Bologna): Questioning the question mark: preliminary considerations from a web corpus of Japanese

Join us for the RC21 Project Symposium, where invited speakers and project team members, Poster Presenters will present their work on methodology and applications of concordance analysis!

Eugenia Diegoli (Bologna): » Questioning the question mark: preliminary considerations from a web corpus of Japanese«

Abstract:

Introduction:

Participants in online written contexts use punctuation signs in innovative and channel-specific ways to reach their interactional goals (Androutsopoulos, 2023). This talk employs corpus-assisted methods to investigate the patterns and functions of the question mark (QM) in the Japanese web corpus Yahoo! Chiebukuro data (LY Corporation, 2024).

Data and Methods:

Yahoo! Chiebukuro is a Q&A website with subfora on specific subjects. I analyze two samples (ca. 3m tokens) from the subfora ikikata to renai, ningen kankei no nayami ‘ways of living and romantic relationships, relational worries’ (the Relationship corpus) and, by way of contrast, nyūsu, seiji, kokusai jōsei ‘news, politics, international affairs’ (the News corpus). These were processed using AntConc (Anthony, 2023) and analyzed by combining corpus tools and close reading of (segments of) texts, including concordances, as typical in Corpus-assisted Discourse Studies (CaDS).

Why the Question Mark?

The QM occurs significantly more frequently in both corpora when compared against the same reference corpus, and hence is indicative of style. Moreover, interrogative constructions in Japanese do not require a QM (the interrogative particle ka, to which the QM tends to be attached, has the same function). Therefore, it conveys additional meaning beyond merely marking a question.

Results:

The analysis of 5-grams corroborated the assumption that in both corpora, the QM almost invariably follows the interrogative particle ka, making it redundant as an interrogative marker. It also revealed that the QM tends to collocate with expressions of **modality**, specifically the verb omoimasu (‘think’) and the modal form of the copula deshō, which decreases the producer’s degree of commitment to the proposition while assuming agreement.

The close reading of concordances revealed that the patterns observed in Table 1 have implications in terms of politeness, as they compensate for the imposition of asking a question while conveying uncertainty and doubt, leaving room for the receiver(s) to disagree or not comply. It also showed that they tend to be immediately preceded by highly evaluative elements indicating necessity/ability/desirability, as in:

(1) Kō iu shinbokukai wa, hontō ni hitsuyō na no deshō ka? Kyōsei nan desu ka? Iya desu ne.
Riyū tsukete kotowaru koto wa dekinai?
‘Are these social gatherings really necessary-DESHŌ? Are they mandatory? They’re awful.
Can’t you refuse?’

We can then expand the patterns in Table 1 and annotate them as:

• <evaluation> QT THINK ka ?
• <evaluation> <ADJ> N deshō ka ?

These findings suggest that interactants may not be willing to take full responsibility for the proposition, especially when such a proposition is overtly evaluative. Modal markers allow them to express various degrees of commitment, while the QM, associated with informality and directness, reduces social distance and conveys positive politeness. The evaluative and politeness elements would have been missed without reading concordances.

Conclusion:

This exploratory analysis hints at the significance of punctuation (and the patterns associated with it) as a tool for social interaction, argumentation, and possibly persuasion. Methodologically, it illustrates a replicable process for functional analysis.

References:

• Androutsopoulos, J. (2023). Punctuating the other: Graphic cues, voice, and positioning in digital discourse. Language & Communication, 88, 141–152.
  https://doi.org/10.1016/j.langcom.2022.11.004
• Anthony, L. (2023). AntConc (Version 4.2.4) [Computer software]. Waseda University.
  https://www.laurenceanthony.net/software
• LY Corporation. (2023). Yahoo! Chiebukuro data set（3rd. Ed.） [Dataset].
  Informatics Research Data Repository.
  https://doi.org/10.32130/idr.1.3