1.Introduction

This paper offers a conversation analytic study of the Japanese turn-final construction tteyuu as a conversational practice of formulation. Tteyuu is a complementizer, composed of the quotative particle tte and the verb yuu “say.” Tteyuu is normally placed between its preceding clausal component and a following head noun, and it functions in clausal noun modification. However, the use of the utterance-final tteyuu, which lacks a following head noun, in both written and spoken data has been reported in recent literature (Kato 2010Kato, Yoko 2010 Hanashi-kotoba ni okeru Inyoo-hyoogen [Quotative Expressions in Spoken Language]. Tokyo: Kurosio Syuppan.; Kim 2014Kim, Joungmin 2014 “Kankokugo no inyoo-shuushokusetsu no shusetsu-ka: Nihongo tono taihi o tsuujite. [Main clause phenomena of quotatitve clauses in Korean: Contrast with Japanese].” In Nihongo fukubun-koobun no kenkyuu [Form and meaning in Japanese complex sentence constructions], ed. by Takashi Masuoka, Motoo Oshima, Osamu Hashimoto, Kaoru Horie, Naoko Maeda, and Takehiko Maruyama, 695–717. Tokyo: Hituzi Syobo.; Koda 2015Koda, Naomi 2015 “Using Reported Thought and Speech to Enhance a Story.” The Japanese Journal of Language in Society 17 (2): 24–39.; Matsumoto 2018Matsumoto, Yoshiko 2018 “The Form and Meaning of the Dangling Mitaina Construction in a Network of Constructions.” In Pragmatics of Japanese: Perspectives on Grammar, Interaction and Culture, ed. by Mutsuko E. Hudson, Yoshiko Matsumoto, and Junko Mori, 75–98. Philadelphia: John Benjamins. ; Ohori 1995Ohori, Toshio 1995 “Remarks on Suspended Clauses: A Contribution to Japanese Phraseology.” In Essays in Semantics and Pragmatics: In Honor of Charles J. Fillmore, ed. by Masayoshi Shibatani, and Sandra A. Thompson, 201–218. Amsterdam: John Benjamins. , 1997Ohori, Toshio 1997 “Framing Effects in Japanese Non-final Clauses: Toward an Optimal Grammar-Pragmatics Interface.” Proceedings of the Twenty-third Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on Pragmatics and Grammatical Structure: 471–480. ; Uemura 2014Uemura, Takashi 2014 “On Functions of Japanese -Toiu in So-called Suspended-sentence: An Investigation on Usage as a Sentence-final Particle.” Papers in linguistic science 20: 31–48.), and it is considered as an “innovative construction” (Matsumoto 2018Matsumoto, Yoshiko 2018 “The Form and Meaning of the Dangling Mitaina Construction in a Network of Constructions.” In Pragmatics of Japanese: Perspectives on Grammar, Interaction and Culture, ed. by Mutsuko E. Hudson, Yoshiko Matsumoto, and Junko Mori, 75–98. Philadelphia: John Benjamins. , 92). This study documents a hitherto unstudied aspect of the turn-final tteyuu as a formulation device.

Formulation is a meta-communicative description with which people explain, characterize, explicate, summarize or furnish the gist of some part of their conversation (Garfinkel and Sacks 1970Garfinkel, Harold, and Harvey Sacks 1970 “On Formal Structures of Practical Actions.” In Theoretical Sociology, ed. by John D. McKinney, and Edward. A. Tiryakian, 337–366. New York: Appleton-Century Crofts.). Heritage and Watson (1979)Heritage, John, and Rod D. Watson 1979 “Formulations as Conversational Objects.” In Everyday Language: Studies in Ethnomethodology, ed. by George Psathas, 123–162. New York: Irvington. identified two general subclasses of formulations: formulations to which a speaker has primary epistemic access and formulations to which a recipient has primary epistemic access. The former is used to explicate, clarify, specify or generalize a speaker’s own prior version of a reference or description, while the latter is used to display understanding of an interactant’s earlier statement by summarizing or developing the gist of it. Following Deppermann (2011)Depperman, Arnulf 2011 “The Study of Formulations as a Key to an Interactional Semantics.” Human Studies 34: 115–128. , this article refers to the former type of formulation as ‘same-speaker formulation’ and the latter as ‘other-speaker formulation.’

This study focuses on other-speaker formulations designed with the turn-final tteyuu, and explores those formulations’ roles for interaction by investigating their compositional and sequential features. In particular, this article shows how tteyuu formulations fall into categories of previously researched other-speaker formulations (Section 4). Then, it examines what conversation participants accomplish by mobilizing this particular type of formulations (Section 5). In so doing, this research reveals the interactional significance of tteyuu formulations in Japanese talk-in-interaction.

2.Background

2.1The utterance-final tteyuu

Tteyuu morphologically consists of the quotative particle tte and the verb yuu “say,” and it is prototypically placed between a modifying clause and its head noun, as indicated in Excerpt (1).

Excerpt 1.

(author’s corpus)

Michi:

are that tte tp na p ayatsuru control tteyuu tteyuu ji? letter Is that the Chinese character of ‘control’?

However, tteyuu has been observed at an utterance-final position without a following head noun. See Excerpt (2) below. (‘Lawson’ is one of the major convenience stores in Japan.)

Excerpt 2.

(author’s corpus)

Mika:	saikin recently konbini convenience.store bakka(h)ri(h) always Recently I always (eat food from) convenience stores,
	cho-   rooson Lawson ni p akite tired kita has.got tteyuu. tteyuu I’m getting tired of Lawson tteyuu.

Ohori (1995)Ohori, Toshio 1995 “Remarks on Suspended Clauses: A Contribution to Japanese Phraseology.” In Essays in Semantics and Pragmatics: In Honor of Charles J. Fillmore, ed. by Masayoshi Shibatani, and Sandra A. Thompson, 201–218. Amsterdam: John Benjamins. appears to be the first study that pointed out the use of the utterance-final toiu in spoken discourse. Toiu and toyuu are phonological variations of tteyuu. Ohori categorized the utterance-final toiu as a kind of suspended clause, i.e., grammatically incomplete patterns that have acquired their own pragmatic functions as independent constructions. Ohori first compared two invented utterances of (3) and (4), and claimed that (3), a suspended version of (4), is possible because the head noun wake ‘story’ has only a general meaning and is not very informative and thus makes it possible for the complement clause to stand by itself. Toiu in Ohori’s (1995)Ohori, Toshio 1995 “Remarks on Suspended Clauses: A Contribution to Japanese Phraseology.” In Essays in Semantics and Pragmatics: In Honor of Charles J. Fillmore, ed. by Masayoshi Shibatani, and Sandra A. Thompson, 201–218. Amsterdam: John Benjamins. examples quoted below is bolded by the author of this article.

Excerpt 3.

([31] in Ohori 1995Ohori, Toshio 1995 “Remarks on Suspended Clauses: A Contribution to Japanese Phraseology.” In Essays in Semantics and Pragmatics: In Honor of Charles J. Fillmore, ed. by Masayoshi Shibatani, and Sandra A. Thompson, 201–218. Amsterdam: John Benjamins. , 214)

Daremo	tasuke	ya	shi	nai	toiu.
anybody	help	prt	do	neg	toiu
(The story is) nobody helped (me).

Excerpt 4.

([32] in Ohori 1995Ohori, Toshio 1995 “Remarks on Suspended Clauses: A Contribution to Japanese Phraseology.” In Essays in Semantics and Pragmatics: In Honor of Charles J. Fillmore, ed. by Masayoshi Shibatani, and Sandra A. Thompson, 201–218. Amsterdam: John Benjamins. , 214)

Daremo	tasuke	ya	shi	nai	toiu	wake	yo.
anybody	help	prt	do	neg	toiu	story	prt
The story is nobody helped (me).

Excerpt 5.

([33] in Ohori 1995Ohori, Toshio 1995 “Remarks on Suspended Clauses: A Contribution to Japanese Phraseology.” In Essays in Semantics and Pragmatics: In Honor of Charles J. Fillmore, ed. by Masayoshi Shibatani, and Sandra A. Thompson, 201–218. Amsterdam: John Benjamins. , 214)

Daremo	tasuke	ya	shi	nakat	ta.
anybody	help	prt	do	neg	past
Nobody helped (me).

Then Ohori compared (3) with (5), and argued that with toiu attached utterance-finally, a speaker may imply detachment or discommitment to a conveyed message. In (3), a speaker reports an event as if she was an observer when it happened. Even though the one who suffered is the speaker herself, the use of toiu renders the utterance as if it were about somebody else. In (5), in contrast, this detachment is not felt, and a speaker simply talks about her experience from her own point of view. Based on this analysis, Ohori claimed that the function of toiu in the suspended construction is anti-evidential because it obscures the credibility of information by impersonalizing it. In other words, “by adding toiu, she is conveying that information as if it were hearsay,” even though what was described in (3) is the speaker’s own experience (Ohori 1997Ohori, Toshio 1997 “Framing Effects in Japanese Non-final Clauses: Toward an Optimal Grammar-Pragmatics Interface.” Proceedings of the Twenty-third Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on Pragmatics and Grammatical Structure: 471–480. , 478). Similarly, Uemura (2014)Uemura, Takashi 2014 “On Functions of Japanese -Toiu in So-called Suspended-sentence: An Investigation on Usage as a Sentence-final Particle.” Papers in linguistic science 20: 31–48. contended that the utterance-final toiu indicates a speaker’s psychological distance from information marked by toiu and that this type of toiu can be grammatically categorized as a sentence-final particle.

Kim (2014)Kim, Joungmin 2014 “Kankokugo no inyoo-shuushokusetsu no shusetsu-ka: Nihongo tono taihi o tsuujite. [Main clause phenomena of quotatitve clauses in Korean: Contrast with Japanese].” In Nihongo fukubun-koobun no kenkyuu [Form and meaning in Japanese complex sentence constructions], ed. by Takashi Masuoka, Motoo Oshima, Osamu Hashimoto, Kaoru Horie, Naoko Maeda, and Takehiko Maruyama, 695–717. Tokyo: Hituzi Syobo. reported that the utterance-final toiu does not necessarily follow a description of the speaker’s experience. Kim discussed that the utterance-final toiu can mark the speaker’s (i) expression of his/her inferences of another’s thought, (ii) unexpectedness and surprise, (iii) presentation of his/her known information as new information, or (iv) description of his/her personal information from another’s point of view. Furthermore, from the perspective of construction grammar, Matsumoto (2018)Matsumoto, Yoshiko 2018 “The Form and Meaning of the Dangling Mitaina Construction in a Network of Constructions.” In Pragmatics of Japanese: Perspectives on Grammar, Interaction and Culture, ed. by Mutsuko E. Hudson, Yoshiko Matsumoto, and Junko Mori, 75–98. Philadelphia: John Benjamins. proposed a broader definition of the utterance-final toyuu (or what Matsumoto called “the dangling toyuu construction”), which encompasses the four usages provided by Kim (2014)Kim, Joungmin 2014 “Kankokugo no inyoo-shuushokusetsu no shusetsu-ka: Nihongo tono taihi o tsuujite. [Main clause phenomena of quotatitve clauses in Korean: Contrast with Japanese].” In Nihongo fukubun-koobun no kenkyuu [Form and meaning in Japanese complex sentence constructions], ed. by Takashi Masuoka, Motoo Oshima, Osamu Hashimoto, Kaoru Horie, Naoko Maeda, and Takehiko Maruyama, 695–717. Tokyo: Hituzi Syobo. mentioned above; the utterance-final toyuu is used “to present to the audience the speaker’s view/stance toward the situation in his or her words as the summary or the essence of the situation” (Matsumoto 2018Matsumoto, Yoshiko 2018 “The Form and Meaning of the Dangling Mitaina Construction in a Network of Constructions.” In Pragmatics of Japanese: Perspectives on Grammar, Interaction and Culture, ed. by Mutsuko E. Hudson, Yoshiko Matsumoto, and Junko Mori, 75–98. Philadelphia: John Benjamins. , 92).

By utilizing a wide variety of data resources including transcripts of conversation data and broadcasted TV programs and interviews, personal blogs, and invented written examples, these studies provide important insights into the functions of the utterance-final tteyuu with regard to a speaker’s perspectives on and attitudes towards the information marked by the tteyuu. The present study aims to contribute to the literature by elucidating interactional aspects of the turn-final tteyuu, specifically as to (a) how the turn-final tteyuu is used in contexts of other-speaker formulations, (b) how tteyuu formulations mark a specific relationship between themselves and their prior speakers’ turns, and (c) how tteyuu formulations shape interactional trajectories of in-progress courses of action.

2.2Formulation

This study provides a conversation analytic description of the turn-final tteyuu as a conversational practice which belongs to the domain of formulation (Garfinkel and Sacks 1970Garfinkel, Harold, and Harvey Sacks 1970 “On Formal Structures of Practical Actions.” In Theoretical Sociology, ed. by John D. McKinney, and Edward. A. Tiryakian, 337–366. New York: Appleton-Century Crofts.; Heritage and Watson 1979Heritage, John, and Rod D. Watson 1979 “Formulations as Conversational Objects.” In Everyday Language: Studies in Ethnomethodology, ed. by George Psathas, 123–162. New York: Irvington.). With formulation, a speaker generalizes, abstracts or specifies the meaning of a first version produced before, either by the speaker him/herself or by another interactant. Formulation of what another interactant said, i.e., other-speaker formulation, “involves summarizing, glossing, and developing the gist of the informant’s earlier statements” (Heritage 1985Heritage, John 1985 “Analyzing News Interviews.” In Handbook of Discourse Analysis. Vol. 3, ed. by Teun. A. van Dijk, 95–117. London: Academic Press., 100). Thus, it serves as a sequence-initiating action of confirmation request, making a response of either confirmation or disconfirmation conditionally relevant from the interactant (Drew and Heritage 1992Drew, Paul, and John Heritage 1992 “Introduction: Analysing Talk at Work.” In Talk at Work, ed. by Paul Drew, and John Heritage, 3–65. Cambridge: Cambridge University Press.; Labov and Fanshel 1977Labov, William, and David Fanshel 1977 Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press.).

Other-speaker formulations are often observed in institutional settings (Heritage 1985Heritage, John 1985 “Analyzing News Interviews.” In Handbook of Discourse Analysis. Vol. 3, ed. by Teun. A. van Dijk, 95–117. London: Academic Press.). In those settings, institutional agents (e.g., therapists, counselors or interviewers) formulate what lay persons say for institutional purposes, such as focusing on therapy-relevant matters in psychotherapy sessions (Antaki et al. 2005Antaki, Charles, Rebecca Barnes, and Ivan Leudar 2005 “Diagnostic Formulations in Psychotherapy.” Discourse Studies 7 (6): 627–647. ), eliciting concerns and feelings during child counseling (Hutchby 2005Hutchby, Ian 2005 “ ‘Active Listening’: Formulations and the Elicitation of Feelings Talk in Child Counselling.” Research on Language and Social Interaction 38 (3): 303–329. ) or committing to newsworthy aspects during interviews (Heritage 1985Heritage, John 1985 “Analyzing News Interviews.” In Handbook of Discourse Analysis. Vol. 3, ed. by Teun. A. van Dijk, 95–117. London: Academic Press.). In institutional talk, formulations are driven by those producers’ underlying agenda, and thus they can be cooperative, uncooperative, or argumentative and contentious (Antaki et al. 2005Antaki, Charles, Rebecca Barnes, and Ivan Leudar 2005 “Diagnostic Formulations in Psychotherapy.” Discourse Studies 7 (6): 627–647. ; Drew 2003Drew, Paul 2003 “Comparative Analysis of Talk-in-Interaction in Different Institutional Settings: A Sketch.” In Studies in Language and Social Interaction: In Honor of Robert Hopper, ed. by Phillip. J. Glenn, Curtis. D. LeBaron, and Jennifer Mandelbaum, 293–308. Mahwah, NJ: Erlbaum.; Hutchby 2005Hutchby, Ian 2005 “ ‘Active Listening’: Formulations and the Elicitation of Feelings Talk in Child Counselling.” Research on Language and Social Interaction 38 (3): 303–329. ).

Previous studies have also documented various practices employed for other-speaker formulations in both institutional and non-institutional interactions. For instance, interactants can explicitly refer to a prior speaker’s words by framing them with a format of “what you’re saying is …” or “you mean …” (Antaki et al. 2005Antaki, Charles, Rebecca Barnes, and Ivan Leudar 2005 “Diagnostic Formulations in Psychotherapy.” Discourse Studies 7 (6): 627–647. ; Steensig and Larsen 2008Steensig, Jakob, and Tine Larsen 2008 “Affiliative and Disaffiliative Uses of You Say X Questions.” Discourse Studies 10 (1): 113–33. ). Interactants can also formulate a candidate continuation of a prior speaker’s turn-constructional unit or TCU (Sacks et al. 1974Sacks, Harvey, Emmanuel A. Schegloff, and Gail Jefferson 1974 “A Simplest Systematics for the Organization of Turn-taking for Conversation.” Language 50: 696–735. ) with ‘appendor questions’ (Lerner 2004Lerner, Gene H. 2004 “On the Place of Linguistic Resources in the Organization of Talk-in-Interaction: Grammar as Action in Prompting a Speaker to Elaborate.” Research on Language and Social Interaction 37 (2): 154–184. ; Sacks 1992Sacks, Harvey 1992 Lectures on Conversation. Cambridge, MA: Blackwell.; Schegloff 1996aSchegloff, Emmanuel A. 1996a “Turn Organization: One Intersection of Grammar and Interaction.” In Interaction and Grammar, ed. by Elinor Ochs, Emmanuel A. Schegloff, and Sandra A. Thompson, 52–133. Cambridge: Cambridge University Press. ). Making an assertion about something a prior speaker alluded to can be another type of formulation (Schegloff 1996bSchegloff, Emmanuel A. 1996b “Confirming Allusions: Toward an Empirical Account of Action.” American Journal of Sociology 104 (1): 161–216. ). Similarly, and-prefaced formulations can articulate what a prior speaker did not say but meant to say or should have said (Bolden 2010Bolden, Galina 2010 “ ‘Articulating the Unsaid’ via And-prefaced Formulations of Others’ Talk.” Discourse Studies 12 (1): 5–32. ). Formulations are also often marked by so at a turn-initial position, and they present a gist or an upshot of a prior speaker’s talk (Antaki et al. 2005Antaki, Charles, Rebecca Barnes, and Ivan Leudar 2005 “Diagnostic Formulations in Psychotherapy.” Discourse Studies 7 (6): 627–647. ; Hutchby 2005Hutchby, Ian 2005 “ ‘Active Listening’: Formulations and the Elicitation of Feelings Talk in Child Counselling.” Research on Language and Social Interaction 38 (3): 303–329. ; Raymond 2004Raymond, Geoffrey 2004 “Prompting Action: The Stand-alone ‘So’ in Ordinary Conversation.” Research on Language and Social Interaction 37 (2): 185–218. ) and may close an in-progress course of action (Schegloff 2007Schegloff, Emmanuel A. 2007 Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge: Cambridge University Press. ).

Prior research has provided great insights into sequential organization and interactional workings of formulations in various types of institutional and non-institutional settings. And yet, as Deppermann (2011Depperman, Arnulf 2011 “The Study of Formulations as a Key to an Interactional Semantics.” Human Studies 34: 115–128. , 125) states, “we know very little about [...] the language-specific resources used for them [turn-constructional formats of formulations] and how these design-features relate to what formulations do and mean in the interaction.” Building on the literature of formulations, this article examines one kind of other-speaker formulation in Japanese designed with the turn-final tteyuu.

3.Method and data

This research adopts Conversation Analysis (CA), which aims to describe perspectives of members of groups, rather than initiating analysis based on some prescribed theory, and to explore the systematicity of social interaction (Heritage 1984Heritage, John 1984 Garfinkel and Ethnomethodology. Cambridge/New York: Polity Press.). This study is based on 53 instances of the turn-final tteyuu used in the context of other-speaker formulations. These instances are culled from the author’s data collection of 16 hours of video-recorded conversations and from 50 hours of audio-recorded conversations from the BTSJ-Japanese Natural Conversation Corpus (Usami 2021Usami, Mayumi ed 2021 BTSJ-Japanese Natural Conversation Corpus with Transcripts and Recordings (March, 2021), NINJAL Institute-based Projects: Multiple Approaches to Analyzing the Communication of Japanese Language Learners.). All data are recordings of non-institutional interactions among adult native speakers of Japanese. Throughout this article, tteyuu is kept untranslated in the transcripts.

The rest of this section clarifies how the author distinguished cases of the turn-final tteyuu from those of the prenominal tteyuu. In the database, the tteyuu-marked utterances are occasionally found in overlap with the next turn by another participant. Then, a question arises as to whether the tteyuu ended up being located turn-finally as a consequence of being left unfinished without a following head noun due to the overlap. The author distinguished the turn-final tteyuu from the prenominal tteyuu based on the timing of the overlap onset and on the intonational design of tteyuu-marked utterances. In fact, in all cases of overlapped tteyuu-marked utterances, overlaps occur as a ‘recognitional overlap’ (Jefferson 1983Jefferson, Gail 1983 “Two Explorations of the Organizations of Overlapping Talk in Conversation: (a) Notes on Some Orderliness of Overlap Onset and (b) On a Failed Hypothesis: ‘Conjunctionals’ as Overlap-vulnerable.” Tilburg Papers in Language and Literature 28: 1–33.). Recognitional overlap happens at the point where the reminder of the ongoing turn is predictable for an overlapping speaker, and thus it is not considered as an interruption by conversation participants. During recognitional overlaps, the turn-final tteyuu shows two intonational characteristics. The tteyuu is designed with a final intonation contour (see Excerpts [6] to [9] and [11]) but not with a continuing intonation contour nor with a cut-off as a result of self-interruption to resolve the overlaps. When the turn-final tteyuu is not accompanied by a final intonation contour, it is produced with and/or followed by its speaker’s laughter (see [10] and [12]). These observations indicate that the turn space around the focal tteyuu is designed by tteyuu-speakers and treated by overlapping participants as a transition relevance space (Ford and Thompson 1996Ford, Cecilia E., and Sandra A. Thompson 1996 “Interactional Units in Conversation: Syntactic, Intonational and Pragmatic Resources for the Management of Turns.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 134–184. Cambridge: Cambridge University Press. ). On the contrary, the prenominal tteyuu is produced with a continuing intonation contour, which projects a following head noun (see lines 2 and 3 of [6], for example). Therefore, the prenominal tteyuu rarely gets overlapped and is designed and treated as a turn-internal component at which an overlap would otherwise be considered as an interruption.

4.Compositional features of tteyuu formulations

The CA literature identifies two types of other-speaker formulations: gist formulations and upshot formulations. The former summarizes what the other person said, while the latter draws out an implication from what the other person said (Heritage and Watson 1979Heritage, John, and Rod D. Watson 1979 “Formulations as Conversational Objects.” In Everyday Language: Studies in Ethnomethodology, ed. by George Psathas, 123–162. New York: Irvington.). In the present database of tteyuu formulations, another type was also identified in addition to gist- and upshot-formulations. This third type of tteyuu formulations not only presents the implied gist of given information but also indicates its speaker’s independent knowledge about the topic being discussed by providing more than what was inferable from what another conversation participant said.11.The author owes this observation to one of the reviewers. This section will introduce these three types of tteyuu formulations, while providing detailed descriptions of their turn compositional features. Excerpt (6) is a case of gist formulation, with which a speaker summarizes what an interlocutor said in the prior turn(s).

Prior to this excerpt, two participants talked about some people who have been in college for 5 years (as opposed to the majority of students who graduate after 4 years). Then in lines 1–4 Ken tells Jiro that in the past there were students who did not graduate even after 6 to 8 years. After Jiro’s news receipt (line 5), Ken in lines 6–12 provides a reason why those students could stay in college for that long period: because they were financially independent, they were free of pressure from their parents. In line 15 Jiro checks his understanding, and Ken confirms it in lines 16–18. Subsequently, Jiro formulates the information provided by Ken with the turn-final tteyuu in lines 19–20.

Jiro says sore dat tara sono ki ni nare ba hachi nen ireru tteyuu “if so, if they wished, they could stay for eight years tteyuu .” “So” in “if so” refers to Jiro’s prior candidate understanding that Ken confirmed (lines 15–16): the condition that expulsion was the only external force for students to leave college. “Eight years” is the maximum length of the period that Ken mentioned at the beginning of his informing in line 1. Formulations are usually selective and they focus on a particular element as a point of the prior talk (Heritage and Watson 1979Heritage, John, and Rod D. Watson 1979 “Formulations as Conversational Objects.” In Everyday Language: Studies in Ethnomethodology, ed. by George Psathas, 123–162. New York: Irvington.). With his gist formulation in lines 19–20, Jiro summarizes Ken’s informing, highlighting the longest possible length of college repeaters in the past as well as the condition that made the college repeating that long. It is also noteworthy that Jiro’s tteyuu-formation was designed based on his already-confirmed candidate understanding in line 15.

As for the compositional features, the following points should be noted. First, the turn-final tteyuu of Jiro’s formulation at lines 19–20 is designed as a declarative assertion. In Japanese, there are two utterance-final question particles ka and no. Ka is used when the predicate of the utterance is in the polite form, and no is used for utterances in the non-polite, plain form (Hayashi 2010Hayashi, Makoto 2010 “An Overview of the Question-Response System in Japanese.” Journal of Pragmatics 42: 2685–702. ). In this excerpt, Jiro uses the polite form, but the particle ka is not placed after tteyuu. Second, the formulation carries a downward intonation contour (marked by the period in line 20), which indexes Jiro’s high degree of epistemic access to the formulated information (Raymond 2009Raymond, Geoffrey 2009 “Grammar and Social Relations: Alternative Forms of Yes/No-type Initiating Actions in Health Visitor Interactions.” In Why Do You Ask?: The Function of Questions in Institutional Discourse, ed. by Alice F. Freed, and Susan Ehrlich, 87–107. New York: Oxford University Press. ). This intonational feature is distinctive from that of Jiro’s earlier candidate understanding which featured a turn-final upward intonation contour (line 15) that indexed his lower degree of epistemic access to the information given thus far. Third, Jiro’s formulation is in a sentential TCU. The formulation, as an assertion about Ken’s informing, works as a request for confirmation, and Ken simply confirms its accuracy in line 21. After 1.5 seconds, their conversation shifts to a topic of current restrictions on college repeaters.

The next example represents a case of upshot formulation, with which a speaker draws out some implication from what an interactant said. Excerpt (7) is a conversation between two college students. They belong to the same student club, to which Kei appears to be new. In the focal segment, Sato, who is senior, tells Kei how seriously the club members play soccer during their summer camps. Right before the segment, Sato said that unlike Sato and Kei, other members have played soccer in high school and thus are much better players. What Sato means by yat(h) te nai to “if you didn’t do it” (line 9) refers to the fact that Kei did not play soccer in high school.

Sato in lines 9–10 emphasizes Kei’s potential difficulty in playing soccer with more experienced members. Although Sato’s multi-unit turn starting from line 1 is not yet syntactically and prosodically complete in line 10, Kei preemptively initiates his turn with a tteyuu formulation: ato bate chau tteyuu “Also you get worn out tteyuu ” in line 11. The turn-initial conjunction ato “also” semantically projects that what will follow is something additional to Sato’s prior talk. With his formulation, Kei presents his inference about potential physical exhaustion (in addition to and/or as a result of the technical hardship of playing with experienced fellows). In response, Sato in line 12 quickly produces confirmation with a soo-type token, and he further validates Kei’s formulation by starting to tell a story.

Kei’s formulation shares compositional features with Jiro’s formulation in (6): a declarative assertion with a downward intonation contour in a sentential TCU. While Jiro’s formulation in (6) exemplified a gist formulation, Kei’s turn in line 11 provides an upshot formulation, which draws out certain elements implied in Sato’s prior talk. Responses to the formulations are also different. Jiro’s gist formulation received a simple confirmation (n::n “uh huh”), but Kei’s upshot formulation receives Sato’s response with a soo-token. According to Kushida (2011)Kushida, Shuya 2011 “Confirming Understanding and Acknowledging Assistance: Managing Trouble Responsibility in Response to Understanding Check in Japanese Talk-in-interaction.” Journal of Pragmatics 43 (11): 2716–2739. , there is a distinction in use between nn-type and soo-type tokens as a response to candidate understanding. While a speaker uses an nn-type token to simply confirm a recipient’s understanding, the speaker uses soo-type token to agree with a recipient’s understanding as an independent formulation and acknowledges the recipient’s assistance in reformulating the speaker’s turn. In the present database, Kushida’s findings generally apply to responses to tteyuu formulations as well. While gist formulations tend to receive nn-type confirmations, upshot formulations receive soo-type responses.

The next case exemplifies the third type of tteyuu formulations, which not only explicates the gist of what the other person said but also provides its speaker’s (i.e., an informing recipient) independent knowledge regarding the given information. Yoshi in Excerpt (8) works for a trading company, and he studied Korean at a community school when he worked at a branch in South Korea. Nao is an Indonesian language instructor at a college, and he has a little experience learning about the Korean language. Prior to the excerpt, they speculated why Korean people are better at speaking English as compared to Japanese people; the Korean language has more phonetic variations, which may overlap with those of English.

In lines 1–4, Yoshi says that the Japanese language rarely has pronunciations which require ‘blowing’ (i.e., aspiration). Yoshi marks the end of his TCU with kedomo (line 4), which leaves recipients to figure out the implications of its contrastive semantics (Ono et al. 2012Ono, Tsuyoshi, Sandra. A. Thompson, and Yumi Sasaki 2012 “Japanese Negotiation through Emerging Final Particles in Everyday Talk.” Discourse Processes 49 (3–4): 243–272. ). Given the content of their prior talk (about phonetic variations of Korean), this description about Japanese is understood as being contrastive to Korean. After Nao’s continuer (Schegloff 1982Schegloff, Emmanuel. A. 1982 “Discourse as an Interactional Achievement: Some Uses of ‘Uh huh’ and Other Things That Come between Sentences.” In Georgetown University Roundtable on Languages and Linguistics: Analyzing Discourse: Text and Talk, ed. by Deborah Tannen, 71–93. Washington D.C.: Georgetown University Press.) in line 6, Yoshi says sooyuu no o yarasare mashita yo “I was made to do things like that.” “Things like that” refers to iki o hukikakeru yoona hatsuon “pronunciations like blowing” mentioned in his prior TCU. Concurrent with Yoshi’s laughter at the end of his turn (line 7), Nao displays his understanding with an interjection a:: “oh yeah” and laughs. Then, in the subsequent turn (lines 9–11), Nao, with tteyuu formulation, explicates what Yoshi said in line 7 by referring to his knowledge which is independent from what Yoshi has told thus far.

First, Nao assesses what Yoshi “was made to do” as omoshiroi “interesting.” With the particle yone, which indexes an equivalent access to the reference (Hayano 2011Hayano, Kaoru 2011 “Claiming Epistemic Primacy in Japanese: Yo-marked Assessments in Japanese.” In The Molarity of Knowledge in Conversation, ed. by Tanya Stivers, Lorenza Mondada, and Jakob Steensig, 58–81. Cambridge: Cambridge University Press. , 2013Hayano, Kaoru 2013 “Territories of Knowledge in Japanese Conversation.” PhD diss. Radboud University Nijmegen., 2018Hayano, Kaoru 2018 “When (Not) to Claim Epistemic Independence: The Use of Ne and Yone in Japanese Conversation.” East Asian Pragmatics 2 (2): 163–193. ), Nao presents his equal epistemicity toward the assessable. Then, Nao says chirigami nanka me no mae de burasagete tteyuu “hanging tissues or things like that in front of eyes tteyuu .” With this, Nao describes a way of practicing Korean aspiration sounds in classroom settings. In other words, Nao reflects his independent epistemic access to the topic in his formulation by providing more than what was inferable from Yoshi’s talk, i.e., how the practice can be done instructionally. With regard to compositional features, similar to gist- and upshot- tteyuu formulations examined earlier, Nao’s formulation here is also featured with a declarative assertion with a downward contour in a sentential TCU. Like upshot formulations (such as Excerpt [7]), this third type of tteyuu formulations typically receives soo-type token responses. In this example, Yoshi responds to Nao’s formulation with successive soo-tokens in lines 12–13.

As seen in the previous three excerpts, tteyuu formulations predominantly receive confirmation responses rather than disconfirmations. In the present database, there is only one case of tteyuu formulation that did not receive a straightforward confirmation. Excerpt (9) is the case in point. In the segment, participants talk about Fumi’s hobby, motorcycling. Fumi does not speed when no cars are around, but he gets tempted to chase cars when they go faster (lines 1–6). Then, Jun, with a tteyuu-marked upshot formulation, proffers his understanding of a reason for Fumi’s temptation to chase cars: charenji shitai tteyuu “you want to challenge (them) tteyuu ” in line 8. Like tteyuu formulations in (6) to (8), the design of Jun’s tteyuu formulation as a declarative assertion (along with his turn-initial a:: “oh I see” in line 7) indexes a high degree of his epistemic access to the formulated information.

In response, Fumi first repeats the term charenji “challenge,” and then provides a more appropriate description of his behavior, saying that chasing cars helps him stay awake (lines 9–11). With the use of the preface tte yuu ka (indicated in bold in line 9), a speaker suggests a better alternative while claiming some degree of legitimacy or warrant with an item that is being replaced (Hayashi et al. 2019Hayashi, Makoto, Yuri Hosoda, and Ikuyo Morimoto 2019 “ Tte yuu ka as a Repair Preface in Japanese.” Research on Language and Social Interaction 52 (2): 104–123. ). Thus, Fumi replaces Jun’s tteyuu formulation with a better suited statement while also acknowledging that Jun’s inference is somewhat legitimate.

This section exemplified three types of tteyuu formulations. A speaker of gist formulation summarizes an interactant’s prior talk by recycling a particular word used by the interactant. A speaker may proffer upshot formulation by drawing out some implication from an interactant’s prior talk. With the third type of tteyuu formulation, a speaker not only explicates the implied gist of given information but also indicates his/her independent knowledge about the topic of the conversation by providing more than what was inferable from what an interactant has told. All types of tteyuu formulations share compositional features of being declarative assertions with a falling intonation, indexing speakers’ high degree of epistemic access to the formulated information. Generally, other-speaker formulations serve as a confirmation request, and thus make a confirmation or disconfirmation response from a speaker of the formulated talk conditionally relevant in the next turn. In the present database, tteyuu formulations are typically responded to with confirmations; nn-type tokens are frequently employed as responses to gist formulations, while soo-type tokens are used with upshot formulations as well as with the third type formulations. A case of upshot formulation was found without a straightforward confirmation, although compositional features of the formulation still indicated its speaker’s high epistemicity. Finally, it should be noted that while this section presented clearer cases, the distinction between the three types of tteyuu formulations often appears to be semantically and pragmatically blurred. Such indistinctiveness between gist- and upshot-formulations in other languages has been also reported in literature as well (Antaki et al. 2005Antaki, Charles, Rebecca Barnes, and Ivan Leudar 2005 “Diagnostic Formulations in Psychotherapy.” Discourse Studies 7 (6): 627–647. ; Deppermann 2011Depperman, Arnulf 2011 “The Study of Formulations as a Key to an Interactional Semantics.” Human Studies 34: 115–128. ).

5.Sequential environments of tteyuu formulations

Tteyuu formulations are commonly observed during extended informings, i.e., informing turns consisting of more than one TCU. The previous section has shown that speakers of tteyuu formulations regularly indicate their high epistemicity toward formulated information by using a declarative form with a turn-final falling intonation. This suggests that tteyuu formulations are launched at a sequential place where informing recipients (i.e., speakers of tteyuu formulations) have been provided with enough information that they can be self-assured of their thus-far understanding of the information being told. This section will show two sequential configurations wherein tteyuu formulations are commonly found: tteyuu formulations for a closure of informing (5.1) and tteyuu formulations as an initiation of further talk (5.2).

5.1 Tteyuu formulations for a closure of informing

Tteyuu formulations are found most frequently at a possible completion of informing (n = 33). A possible completion of informing is regularly indicated syntactically, prosodically, and pragmatically, although these three resources may not always converge (Ford and Thompson 1996Ford, Cecilia E., and Sandra A. Thompson 1996 “Interactional Units in Conversation: Syntactic, Intonational and Pragmatic Resources for the Management of Turns.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 134–184. Cambridge: Cambridge University Press. ; Tanaka 1999Tanaka, Hiroko 1999 Turn-taking in Japanese Conversation: A Study in Grammar and Interaction. Amsterdam: John Benjamins. ). At such a possible completion of informing, recipients are normatively expected to provide a relevant response (Jefferson and Lee 1992Jefferson, Gail, and John R. E. Lee 1992 “The Rejection of Advice: Managing the Problematic Convergence of a ‘Troubles-telling’ and a ‘Service Encounter’.” In Talk at Work: Interaction in Institutional Settings, ed. by Paul Drew, and John Heritage, 521–48. New York: Cambridge University Press.; Maynard 2003Maynard, Douglas. W. 2003 Bad News, Good News: Conversational Order in Everyday Talk and Clinical Settings. Chicago, IL: University of Chicago Press.; Stivers 2008Stivers, Tanya 2008 “Stance, Alignment, and Affiliation during Storytelling: When Nodding is a Token of Affiliation.” Research on Language and Social Interaction 41 (1): 31–57. ), and tteyuu formulations can be deployed as such responses. Previous excerpts (6) and (8) (partially reproduced below) are cases in point. In both cases, informing sequences reach their closures after informing speakers’ (Ken in [6] and Yoshi in [8]) confirmations to their preceding tteyuu formulations.

Excerpt 6.

(BTSJ/355-25-JM053-JM055)

17	Ken:	sooyuu such hanashi story wa tp kiita heard kotoa(h)ru(h) have.done I’ve heard such stories
18		[chokuse(h)tsu  directly ne. p  directly.
19	Jiro:  →	[sore  that dat cp tara if sono ki ni nare ba if.one.wishes  If so, if they wished,
20	→	hachi eight nen years ireru can.stay tteyuu. tteyuu they could stay for eight years tteyuu.
21	Ken:	n::n. Uh huh.
22		(1.5)

Excerpt 8.

(BTSJ/186-13-JMB003_JMO002)

7	Yoshi:	sooyuu like.that no n o o yarasare be.made.to mashita was yo p [hhhhhhhhh   I was made to do things like that.
8	Nao:	[a::                                       Oh hhh yeah
9	→	ano well omoshiroi interesting desu cp yone p ano:: like well interesting, right? Like,
10	→	chirigami tissue ka or nanka something me eye no lk mae front de at
11	→	burasagete hang tteyu [ u. tteyuu hanging tissues or something in front of eyes tteyuu.
12	Yoshi:	[soo                  Right soo right soo right soo right soo right
13		soo right soo. right.
14	Nao:	ee. Yeah.
15		(1.0)

Excerpt (10) below shows a similar sequential configuration. Before line 1, Shige and Ryo talked about how unique Shige’s family name is. Then in lines 1–5 Shige says that, contrary to his name, ‘nothing is interesting’ about where he is from because he was born and raised in Tokyo. In response to Ryo’s follow-up question in line 7, Shige from line 8 starts providing a list of all the places that he has lived in Tokyo, beginning with his birthplace. Then in line 25 Shige mentions a place he moved this year (Azabu), which is recognizable as the end of the list and thus as the completion of Shige’s answer to Ryo’s question.

Excerpt 10.

(BTSJ/190-13-JMB003_JMY001)

1	Shige:	shusshinchi place.one’s.from nikanshite about yuu say to if Speaking of where I’m from,
2	Ryo:	hai. Yes.
3	Shige:	boku I wa tp umare mo sodachi mo born.and.rased tookyoo Tokyo na cp n n de cp I was born and raised in Tokyo, so
4	Ryo:	hai Yes ha[i. yes.
5	Shige:	[ano::        Uhm nanmo anything omoshiroi interesting koto thing ga sp na[i neg        Uhm, nothing is interesting.
6	Ryo:	[A HA HA HA
7		dono a- where dono where atari around desu cp ka? q Where- around where?
8	Shige:	e::tto well umareta was.born no n ga sp bunkyooku Bunkyo-ward de, cp Well I was born in Bunkyo-ward, and
		((lines 9–24: Shige lists three more places he has lived.))
25	Shige:	kotoshi this.year ni p natte became azabu Azabu ni p hikkoshite. move I moved to Azabu this year.
26	Ryo:	hai Yes hai. yes.
27		(2.0)
28	Ryo:   →	˚haa (  )˚ I.see dakara so iroiro various mawatteru move.around I see, so you’ve moved around various places,
29	→	kedomo but tookyoo Tokyo tte(h)yu(h)[ u(h) tteyuu but Tokyo tteyuu.
30	Shige:	[tookyoo                             Tokyo tte(h)yu(h)u(h) tteyuu.
31		(1.5)

Ryo treats this possible completion point of Shige’s informing as a relevant sequential location for his tteyuu formulation (lines 28–29), which follows after his acknowledgement tokens (line 26). An absence of Ryo’s earlier uptake, which is indicated by the noticeable silence in line 27, may be due to a possible continuation of Shige’s list: Shige could have moved from Azabu to another place before the summer when this conversation was recorded. It may also be because of the incomplete syntax of Shige’s turn in line 25, as it ends with hikkosite, the -te form22.The -te form is considered as one of the clause-chaining devices (e.g., Iwasaki 1993Iwasaki, Shoichi 1993 Subjectivity in Grammar and Discourse: Theoretical Considerations and a Case Study of Japanese Spoken Discourse. Amsterdam: John Benjamins. ; Ono 1987Ono, Toshio 1987 “TE, I, and RU Clauses in Japanese Recipes: A Quantitative Study.” Studies in Language 14: 73–92. ). (or gerund or continuative form) of hikkosu “move.” In fact, discontinuation of Shige’s list becomes apparent in line 27, and that leads to Ryo’s tteyuu formulation, which demonstrates Ryo’s understanding of not only the gist of Shige’s informing but also an interactionally relevant action expected in this sequential position.

Another point to note is about Shige’s response to Ryo’s formulation, which manifests two important facets of tteyuu formulation. First, rather than simply confirming the accuracy of Ryo’s formulation (e.g., with nn- or soo-type token), Shige in line 30 responds with a partial repetition of Ryo’s formulation with tteyuu (tookyoo tte(h)yu(h)u(h)). This is considered as the action of ‘confirming allusions’ (Schegloff 1996bSchegloff, Emmanuel A. 1996b “Confirming Allusions: Toward an Empirical Account of Action.” American Journal of Sociology 104 (1): 161–216. ), with which Shige indicates that Ryo’s offered understanding could have been legitimately inferred from Shige’s prior talk. In other words, Shige’s response evidently shows that Ryo’s tteyuu-marked turn serves as formulating the point Shige aimed to convey through his informing. Second, Shige repeats tteyuu as well in his response. When a speaker repeats an utterance, s/he typically dispenses with some turn-constructional components (Schegloff 2004Schegloff, Emmanuel A. 2004 “On Dispensability.” Research on Language and Social Interaction 37 (2): 95–149. ). Like Shige in (10), however, informing speakers often retain tteyuu, when they respond to the prior tteyuu formulations with a (partial) repetition.33.See also Excerpt (12), where a recipient of the first tteyuu formulation produces the second tteyuu formulation that clarifies the first formulated information. This means that tteyuu is an indispensable constitutive component of the action of formulation.

In all three excerpts shown in this section, tteyuu formulations are produced upon a possible completion of informing, where an uptake of the informing turns is due. The analyses of these cases and other similarly positioned tteyuu formulations in the database indicate that producers of these formulations (i.e., informing recipients) align with the informing speakers’ course of action (cf. Stivers 2008Stivers, Tanya 2008 “Stance, Alignment, and Affiliation during Storytelling: When Nodding is a Token of Affiliation.” Research on Language and Social Interaction 41 (1): 31–57. ) by reaffirming and/or inferring the point of the informing with or without demonstrations of their independent knowledge about the informed matter. While the turn-final design is consistent with tteyuu accompanied by a falling intonation, the turn-initials of the formulations vary case by case. In (10), the tteyuu formulation is prefaced with dakara (line 28), which is roughly equivalent to the English so. So-prefaced formulations are frequently characterized as offering an upshot or a gist of the other speaker’s talk (Antaki et al. 2005Antaki, Charles, Rebecca Barnes, and Ivan Leudar 2005 “Diagnostic Formulations in Psychotherapy.” Discourse Studies 7 (6): 627–647. ; Hutchby 2005Hutchby, Ian 2005 “ ‘Active Listening’: Formulations and the Elicitation of Feelings Talk in Child Counselling.” Research on Language and Social Interaction 38 (3): 303–329. ; Raymond 2004Raymond, Geoffrey 2004 “Prompting Action: The Stand-alone ‘So’ in Ordinary Conversation.” Research on Language and Social Interaction 37 (2): 185–218. ).44. Deppermann (2011)Depperman, Arnulf 2011 “The Study of Formulations as a Key to an Interactional Semantics.” Human Studies 34: 115–128. discusses the German also (translated into the English so) as the most commonly used formulation index in the German language. Thus in (10), the dakara preface serves as another formulation marking in concert with tteyuu. Similarly, the tteyuu formulation in (6) is prefaced with sore dat tara “if so” (line 19), which prepares recipients to hear the upcoming formulation as a consequential summary of information provided in the preceding talk. In other cases (Excerpt [8], for example), informing recipients may place tteyuu formulations after assessments, which are frequently employed for topic closure as they allow recipients to show attention to the preceding talk while also delivering a summary of it (Jefferson 1984Jefferson, Gail 1984 “On Stepwise Transition from Talk about a Trouble to Inappropriately Next-positioned Matters.” In Structures of Social Action, ed. by J. Maxwell Atkinson, and John Heritage, 191–221. Cambridge: Cambridge University Press.; Sidnell 2010Sidnell, Jack 2010 Conversation Analysis: An Introduction. Malden, Massachusetts: Wiley-Blackwell.). Tteyuu formulations in this particular sequential position are responded to and most of the time confirmed by the informing speakers’ nn- or soo-type tokens or a (partial) repetition of the formulations.

5.2 Tteyuu formulations as an initiation of further talk

As discussed earlier, tteyuu formulations are frequently observed at a possible completion of informing. These tteyuu formulations receive informing speakers’ responses of a single TCU consisting of a confirmation token or a partial repeat of the formulation, and thus they contribute to leading the informing sequence to its closing. However, tteyuu formulations can also serve to develop ongoing topics into further related talk by tying the formulation to what was mentioned in the prior talk (n = 11). This section introduces three examples of this type of tteyuu formulation. The first case is Excerpt (7), which was shown in Section 4. The excerpt is partially reproduced below.

Excerpt 7.

(BTSJ/364-25-JM075-JM076)

9	Sato:	suGGE extremely taihen hard dakara so yat(h)te do nai no to if it’d be extremely hard, if you didn’t do it,
10		MAji[de seriously  [nori  impulse dakara so because they do it on the spur of the moment.
11	Kei: →	[hhh     [ato    also bate worn.out chau end.up tteyuu. tteyuu Also you get worn out tteyuu.
12	Sato:	soo right bate worn.out chatte end.up ore I ikkai once ni two nen year no lk toki when
13		ha-   haite vomit n(h) n da(h) cp yo f hhhh   Right, I got worn out and once threw up when I was a sophomore.

Before line 10, Sato explained to Kei, a new member of their student club, the technical hardship of playing soccer with their more experienced fellows in the club. Then in line 11, Kei provides his tteyuu formulation, presenting his inference about potential physical exhaustion (in addition to and/or as a result of the technical hardship). In response, Sato in line 12 quickly confirms Kei’s formulation with a soo-token, and he further validates Kei’s formulation by starting to tell a story (which continues after line 13). In other words, Kei’s tteyuu formulation triggers Sato’s further discussion about another aspect of his experiences in the club.

Two graduate students in Excerpt (11) belong to the same (presumably a linguistics-related) department but are meeting for the first time. Prior to the segment, Asa said that she does not conduct research on the Japanese language although she is in the Japanese program within their department. That leads to Mai’s question in line 1.

Excerpt 11.

(BTSJ/195-14-JFB023-JFSa016)

1	Mai:	ja then nani what [nani what kenkyuu research shiteru doing n n desu cp ka? q Then what are you researching about?
2	Asa:	[nanka             like gengo language sesshoku contact o o yattete doing          Like I’m doing language contact
3	Mai:	he↑:[:::: Oh
4	Asa:	[dakara      so shingapooru Singapore o o So Singapore
5	Mai:	a: oh [tagengo:: (.) multilingual [kokka nation mitaina like kanji? feeling Oh it’s like a multilingual nation?
6	Asa:	[(     )         [soo        right desu cp soo right desu cp                    Right right
7		soo right desu cp ne. p right.
8	Mai:	sugoi amazing desu cp yone hhh[h p Isn’t (Singapore) amazing?
9	Asa:	[n:n.                       Yeah.
10	Mai:  →	eh oh sore that gengo language ga sp majiriau intersect to when Oh that, when languages intersect,
11	Asa:	[nn.  Uh huh.
12	Mai:  →	[sono  like hokano other gengo language ni p au meet to when doo how yuu say
13	→	huuni way na[ru become no n ka q tteyuu. tteyuu when they meet other languages, how they’d become tteyuu.
14	Asa:	[>soo           right soo right soo right soo< right           Right right right right

After receiving Asa’s answer (gengo sesshoku “language contact”), Mai in line 3 provides a token he↑:::::, which treats the provided information as newsworthy while also prompting for clarification (Mori 2006Mori, Junko 2006 “The Workings of the Japanese Token Hee in Informing Sequences: An Analysis of Sequential Content, Turn Shape, and Prosody.” Journal of Pragmatics 38: 1175–1205. ). In the next turn, Asa provides her research target (i.e., Singapore). In line 5 Mai, by framing Singapore as tagengo kokka “a multilingual nation,” checks the appropriateness of her association of Singapore with language contact. After Asa’s confirmation (lines 6–7), Mai submits her assessment of Singapore’s multilingualism by saying sugoi “amazing” with the turn-final particle yone, which indexes a speaker’s claim of equal epistemic access to the proposition being discussed.

It is after this pre-shift assessment (Jefferson 1993Jefferson, Gail 1993 “Caveat Speaker: Preliminary Notes on Recipient Topic-shift Implicature.” Research on Language and Social Interaction 26: 1–30. ) that Mai launches a tteyuu formulation (lines 10, 12–13), whereby she brings their topical focus back to “language contact.” Mai employs the eh-prefacing and a demonstrative sore “that” at the turn-beginning as tying strategies. Eh-prefacing serves to frame a forthcoming question as being outside its sequentially most natural position (Hayashi 2009Hayashi, Makoto 2009 “Marking a ‘Noticing of Departure’ in Talk: Eh-prefaced Turns in Japanese Conversation.” Journal of Pragmatics 41 (10): 2100–2129. ). Sore also links between the current turn and previous talk (cf. Sacks 1995Sacks, Harvey 1995 Lectures on Conversation. Vol. 2. Oxford: Basil Blackwell. ; Sidnell 2010Sidnell, Jack 2010 Conversation Analysis: An Introduction. Malden, Massachusetts: Wiley-Blackwell.). The most direct target of Mai’s formulation is language contact (appeared in line 2), as the formulation follows up on the term and seeks confirmation on her elaboration of the term. In other words, the tteyuu formulation in this excerpt works to topicalize what was previously mentioned. The formulation is then followed by Asa’s multi-unit turn consisting of her confirmation and further explanation of language contact in Singapore.

In Excerpt (12), the four participants are colleagues who work at the Cabinet Office. They are routinely assigned to work at different locations, including the Diet Building, to which none of the participants has been assigned. Prior to the excerpt, participants said that they want to go to see the Diet Building at least once. Then, Naka talked about how they can go into some parts of the building which are not accessible by general visitors: they need ano bajji “the badge” (line 3), that is given to the Cabinet Office employees assigned to work at the Diet Building. A tteyuu formulation is produced by Tamu in lines 8–9 after all the participants appeared to reach their understanding of “the badge.”

Excerpt 12.

(author’s corpus)

1	Naka:	sore that igai other.that no lk tokoro place (0.5)   o o mireru can.see ka[ra:: so You can see places other than that,
2	Hiro:	[o: o:                                              Oh oh
3	Naka:	ano the bajji badge o o hhh   [motte(h)  bring ike go ba if if you bring the badge.
4	Tamu:	[a                   oh bajji badge motte bring Oh you bring the badge.
5	Hiro:	ano that kusari chain no lk yatsu thing ne?= p The chained thing, right?
6	Naka:	[=soo   right desu. cp Right.
7	Kato:	[((nods))
8	Tamu:  →	demo but sore that tte tp tsumari in.other.words kokkai the.Diet
9	→	atatta assigned t [ teyu [ u hhh hhhh tteyuu But that means (you’re) assigned to the Diet tteyuu.
10	Naka:	[A h [h h h h
11	Hiro:	[soo(h)                 Right soo(h) right soo(h) right.
12		zenzen at.all ano (.) uhm ureshiku[nai not.happy tteyuu tteyuu uhm (you’re) not happy at all tteyuu.
13	Tamu:	[sonna hhh                         Such
14	Naka:	u(h)re(h)shi(h)ku(h)na(h)i(h)[jookyoo not.happy                situation Unhappy situation.
15	Tamu:	[nanka                               like haitta:: got.in nante like
16		[yuuyuuto  leisurely miteru seeing yoyuu space nasa no [soona  seem [kigasuru  feel.like like, I feel like there won’t be time for us to look around in a leisurely manner.
17	Kato:	[n::n.  Uh huh.
18	Hiro:	[nai                                no ne. p                                No.
19	Naka:	[soo                                       right desu cp ne: p
20		tashikani. certainly Yeah, certainly.

Tamu’s tteyuu formulation in lines 8–9 starts with a conjunction demo “but” and sore tte, a demonstrative “that” with a topic marker, which prepares recipients to hear that Tamu will, with the rest of this turn, most likely state something contrastive to a certain aspect of their prior talk. Then he says tsumari kokkai atatta tteyuu “means (you’re) assigned to the Diet tteyuu .” Tamu’s tteyuu formulation clarifies their unsaid shared understanding that having “the badge” means that they work there. Tamu’s formulation is treated as laughable and confirmed with soo-tokens (lines 10–11). Then in lines 11–12, Hiro, with another tteyuu formulation, overtly states the reason for their laughter: being assigned to the Diet is not something they desire (because of its hard work, as it will be revealed later in this conversation). Hiro’s formulation is then confirmed by Naka’s partial repeat in the subsequent turn in line 14 (cf. Schegloff 1996bSchegloff, Emmanuel A. 1996b “Confirming Allusions: Toward an Empirical Account of Action.” American Journal of Sociology 104 (1): 161–216. ). In this excerpt, by formulating an unsaid shared assumption with tteyuu, Tamu and then Hiro foreground the reality of what they wish for (i.e., that their wish to view special parts of the Diet Building would force them to engage in unwanted work), and that brings about a shift in their topical focus.

As demonstrated in the above three cases and others like these, tteyuu formulations serve to shape the topical trajectory of ongoing talk. They are compositionally characterized with certain turn-initial elements of tying devices including the use of demonstratives, connectives, and/or tokens that indicate a particular relation between what was already said and what will be formulated. Responses to these tteyuu formulations are designed with initial confirmations and further development of their topical talk. Moreover, these formulations deal with semantic units of different sizes and kinds (Bilmes 2011Bilmes, Jack 2011 “Occasioned Semantics: A Systematic Approach to Meaning in Talk.” Human Studies 34: 129–153. ; Deppermann 2011Depperman, Arnulf 2011 “The Study of Formulations as a Key to an Interactional Semantics.” Human Studies 34: 115–128. ; Hauser 2011Hauser, Eric 2011 “Generalization: A Practice of Situated Categorization in Talk.” Human Studies 34 (2): 183–198. ). They can be employed to formulate meanings of a single word or phrase that previously appeared (like Excerpt [11]), a part of a description (like [7]), or a whole informing talk (like [12]). As has been discussed in CA literature, a new topic is not abruptly introduced upon a completion of a previous topic, but a stepwise move is regularly arranged to connect what has been talked about to what is about to be talked about (Jefferson 1984Jefferson, Gail 1984 “On Stepwise Transition from Talk about a Trouble to Inappropriately Next-positioned Matters.” In Structures of Social Action, ed. by J. Maxwell Atkinson, and John Heritage, 191–221. Cambridge: Cambridge University Press.; Sacks 1995Sacks, Harvey 1995 Lectures on Conversation. Vol. 2. Oxford: Basil Blackwell. ). Tteyuu formulations are one way in which such stepwise transitions are accomplished.

6.Concluding discussion

This study investigated the Japanese turn-final construction tteyuu as an interactional practice of formulation. While tteyuu is normally placed in clausal noun modification with a following head noun, its use at a turn-final position has been reported as “innovative construction” (Matsumoto 2018Matsumoto, Yoshiko 2018 “The Form and Meaning of the Dangling Mitaina Construction in a Network of Constructions.” In Pragmatics of Japanese: Perspectives on Grammar, Interaction and Culture, ed. by Mutsuko E. Hudson, Yoshiko Matsumoto, and Junko Mori, 75–98. Philadelphia: John Benjamins. , 92). This study documented a hitherto unstudied aspect of the turn-final tteyuu as an other-speaker formulation device.

In the CA literature, other-speaker formulations have been classified into two general types: gist formulations and upshot formulations (Heritage and Watson 1979Heritage, John, and Rod D. Watson 1979 “Formulations as Conversational Objects.” In Everyday Language: Studies in Ethnomethodology, ed. by George Psathas, 123–162. New York: Irvington.). This study specifically examined tteyuu formulations, and identified one additional type besides the previously recognized two types. This article began with presenting tteyuu formulations representing those three types of formations. Tteyuu-marked gist formulations summarized preceding informing, while tteyuu-marked upshot formulations drew out informing speakers’ implications. The third type of tteyuu formulations indicated its speaker’s independent knowledge about the topical matter by providing more than what was inferable from a preceding informing. All types of tteyuu formulations shared compositional features of being a declarative assertion with a falling intonation contour, indexing informing recipients’ high degree of epistemic access to the formulated information. Tteyuu formulations were overwhelmingly responded to with confirmations.

Other-speaker formulations serve as a confirmation request for understanding because the formulated information originally belongs to other interactants in conversation. By investigating the two most common sequential environments of tteyuu formulations, however, this study revealed that an intrinsic interactional purpose of using the focal formulations is not merely a confirmation request. First, this research examined tteyuu formulations appearing upon a possible completion of informing. In such occasions, informing recipients who produced tteyuu formulations aligned with the informing speakers’ course of action by reaffirming and/or inferring the point of the informing, and those tteyuu formulations led the informing to its closure. Furthermore, this study elucidated that tteyuu formulations serve to shape the topical trajectory of ongoing talk. These formulations worked as a tying device to indicate a particular relation between the preceding informing and to-be-formulated contents, functioning as a pivot via which such stepwise transitions get done. It is also important to note that the turn-final tteyuu was, as clearly shown in Excerpts (10) and (12), an indispensable constitutive component to accomplish these interactional tasks. Further research needs to be conducted to compare tteyuu with other linguistic elements employed for Japanese other-speaker formulations and to investigate if, and if so, why tteyuu is specifically used for the interactional purposes observed in this article.

Throughout this article, it was demonstrated that tteyuu formulations serve as a means to demonstrate intersubjectivity while also advancing sequence progressivity. Intersubjectivity refers to shared understanding between conversation participants as a product of communication (Duranti 2009Duranti, Alessandro 2009 “The Relevance of Husserl’s Theory to Language Socialization.” Journal of Linguistic Anthropology 19 (2): 205–226. ). A general feature of conversation is to favor implicit over explicit methods of accomplishing interactional tasks, and thus it is often the case that the management of given information as intersubjectively secure is done implicitly (Jefferson 1987Jefferson, Gail 1987 “On Exposed and Embedded Correction in Conversation.” In Talk and Social Organization, ed. by Graham Button, and John. R. E. Lee, 86–100. Clevedon, UK: Multilingual Matters. ; Levinson 1987Levinson, Stephen C. 1987 “Pragmatics and the Grammar of Anaphora.” Journal of Linguistics 23: 379–434. ). However, tteyuu formulations are identified as one of the conversational practices with which interactants explicitly demonstrate and negotiate their possible arrival at a shared understanding. At the same time, tteyuu formulations are also tied to the concept of progressivity, which refers to the forward-moving nature of interaction. As shown in this article, tteyuu formulations assist conversation participants to move on to a next interactional agenda in a timely manner. Generally, participants’ intersubjectivity is claimed and embedded in sequences for the sake of progressivity (Heritage 2007Heritage, John 2007 “Intersubjectivity and Progressivity in Person (and Place) Reference.” In Person Reference in Interaction: Linguistic, Cultural and Social Perspectives, ed. by Tanya Stivers, and N. J. Enfield, 255–280. Cambridge: Cambridge University Press. ). Nevertheless, tteyuu formulations tacitly work as a vehicle to advance sequence progressivity, while providing participants with an interactional space to negotiate their understanding of each other.

Discourse-functional studies have provided great insight into how speakers utilize utterance-final tteyuu to present certain information from different points of view (see Section 2 for a review of the literature). These studies seem to agree that the utterance-final tteyuu has become a construction not only grammatically but also pragmatically independent from the tteyuu used for noun-modifying constructions.55.However, see also Matsumoto (2018)Matsumoto, Yoshiko 2018 “The Form and Meaning of the Dangling Mitaina Construction in a Network of Constructions.” In Pragmatics of Japanese: Perspectives on Grammar, Interaction and Culture, ed. by Mutsuko E. Hudson, Yoshiko Matsumoto, and Junko Mori, 75–98. Philadelphia: John Benjamins. , who discusses that the original meaning of tteyuu in noun-modifying constructions as quasi-quotation may still persist in the utterance-final tteyuu. Building on this line of research, the present study used the analytical method of CA and documented another hitherto unstudied interactional use of the turn-final tteyuu as other-speaker formulations. It is hoped that this study also contributed to the research of formulation by unveiling one example of how a language-specific item, with its relatively new grammatical structure, can be deployed as a resource for turn-constructional format of formulation.