The use of the non-lexical sound öö in Hungarian same-turn self-repair

Zsuzsanna Németh
University of Szeged | MTA-SZTE-DE Research Group for Theoretical Linguistics and Informatics

This paper explores the use of the non-lexical schwa-like sound öö in three Hungarian self-initiated same-turn self-repair phenomena, namely, searching, replacing, and searching converted into aborting. Since self-repair inherently prevents the turn from progressing towards possible completion, by deploying the non-lexical öö in the course of the three repair operations, speakers can vocalize and promise continuation without producing lexical elements. Öö thus serves as a delaying technique and an indication that the speaker is aware of the obligation to make the unit-under-way recognizably complete, and is ready to satisfy this obligation. When the turn-in-progress is halted and verbally marked with öö in the course of the three repair operations under investigation, the speaker displays to the other participants that the turn is problematic in some respects, but she is ready to attend to the problem in order to produce a recognizably complete action.

Publication history

Date received: 7 May 2023

Date accepted: 1 December 2023

Published online: 18 April 2024

Table of contents

1.Introduction: Theoretical background and aims of the paper

Although the Hungarian non-lexical schwa-like sound öö 11.Since there is no letter for schwa [ə] in Hungarian, I will use the letter ö to indicate the sound in question for two reasons. On the one hand, this is the letter which indicates the closest sound to schwa in Hungarian (IPA [ø]) (cf. Horváth 2014 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar, 25). On the other hand, the literature generally uses ö to indicate the phenomenon I examine. The double letter öö refers to the default use of the sound in the Hungarian dataset, which is a lengthened schwa. has been investigated extensively in the psycholinguistic and phonetic literature (e.g. Deme and Markó 2013Deme, Andrea, and Alexandra Markó 2013 “Lengthenings and Filled Pauses in Hungarian Adults’ and Children’s Speech.” In Proceedings of DISS 2013. The 6th Workshop of Disfluency in Spontaneous Speech, ed. by Robert Eklund, 21–24. Stockholm: KTH Royal Institute of Technology.Deme, Andrea, and Alexandra Markó 2013 “Lengthenings and Filled Pauses in Hungarian Adults’ and Children’s Speech.” In Proceedings of DISS 2013. The 6th Workshop of Disfluency in Spontaneous Speech, ed. by Robert Eklund, 21–24. Stockholm: KTH Royal Institute of Technology.; Gósy 2006Gósy, Mária 2006 “A semleges magánhangzó nyelvi funkciói [The linguistic functions of the neutral vowel].” In Beszédkutatás 2006 [Speech research 2006], ed. by Mária Gósy, 8–22. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google ScholarGósy, Mária 2006 “A semleges magánhangzó nyelvi funkciói [The linguistic functions of the neutral vowel].” In Beszédkutatás 2006 [Speech research 2006], ed. by Mária Gósy, 8–22. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google Scholar; Gósy, Bóna, Beke and Horváth 2013Gósy, Mária, Judit Bóna, András Beke, and Viktória Horváth 2013 “A kitöltött szünetek fonetikai sajátosságai az életkor függvényében [The phonetic characteristics of filled pauses with regard to age].” In Beszédkutatás 2013 [Speech research 2013], ed. by Mária Gósy, 121–143. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google ScholarGósy, Mária, Judit Bóna, András Beke, and Viktória Horváth 2013 “A kitöltött szünetek fonetikai sajátosságai az életkor függvényében [The phonetic characteristics of filled pauses with regard to age].” In Beszédkutatás 2013 [Speech research 2013], ed. by Mária Gósy, 121–143. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google Scholar; Horváth 2014 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar), in the framework of conversation analysis, its use has only been examined thoroughly from the perspective of turn-taking. Németh (2022) 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 194: 23–35. Google Scholar logo with link to Google Scholar 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 1941: 23–35. Google Scholar logo with link to Google Scholar has observed that in overlapping speech it can be a device to keep the floor, and by using öö at the transition-relevance place (TRP), participants may express their orientations to the turn-taking rules (cf. Sacks, Schegloff and Jefferson 1974Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson 1974 “A Simplest Systematics for the Organization of Turn-Taking for Conversation.” Language 50: 696–735. Google Scholar logo with link to Google ScholarSacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson 1974 “A Simplest Systematics for the Organization of Turn-Taking for Conversation.” Language 501: 696–735. Google Scholar logo with link to Google Scholar). If a speaker cannot (immediately) satisfy the normative obligation imposed on her by a first pair part, by using öö she can project a problem in providing an answer (orientation to rule 1a). A participant who self-selects, by using öö, may prevent the others from self-selecting and the previous speaker from adding a new unit to her turn (orientation to rule 1b). Furthermore, öö may help the current speaker to avoid a speakership change (orientation to rule 1c). However, öö is not only used at the transition-relevance place or in overlapping speech, but psycholinguistic and phonetic studies report the high frequency of öö “in the vicinity of speech errors” (Horváth 2014 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar, 115). Its appearance may refer to speech planning or self-monitoring problems (Clark and Fox Tree 2002Clark, Herbert H., and Jean E. Fox Tree 2002 “Using Uh and Um in Spontaneous Speaking.” Cognition 84: 73–111. Google Scholar logo with link to Google ScholarClark, Herbert H., and Jean E. Fox Tree 2002 “Using Uh and Um in Spontaneous Speaking.” Cognition 841: 73–111. Google Scholar logo with link to Google Scholar; Gósy et al. 2013Gósy, Mária, Judit Bóna, András Beke, and Viktória Horváth 2013 “A kitöltött szünetek fonetikai sajátosságai az életkor függvényében [The phonetic characteristics of filled pauses with regard to age].” In Beszédkutatás 2013 [Speech research 2013], ed. by Mária Gósy, 121–143. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google ScholarGósy, Mária, Judit Bóna, András Beke, and Viktória Horváth 2013 “A kitöltött szünetek fonetikai sajátosságai az életkor függvényében [The phonetic characteristics of filled pauses with regard to age].” In Beszédkutatás 2013 [Speech research 2013], ed. by Mária Gósy, 121–143. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google Scholar). In their psycholinguistic study, Gósy, Gyarmathy, and Beke (2017)Gósy, Mária, Dorottya Gyarmathy, and András Beke 2017 “Phonetic Analysis of Filled Pauses Based on a Hungarian-English Learner Corpus.” International Journal of Learner Corpus Research 3: 149–174. Google Scholar logo with link to Google ScholarGósy, Mária, Dorottya Gyarmathy, and András Beke 2017 “Phonetic Analysis of Filled Pauses Based on a Hungarian-English Learner Corpus.” International Journal of Learner Corpus Research 31: 149–174. Google Scholar logo with link to Google Scholar show that öö is significantly longer if the speaker needs more time for speech planning. Horváth (2010)Horváth, Viktória 2010 “Filled Pauses in Hungarian: Their Phonetic Form and Function.” Acta Linguistica Hungarica 57: 288–306. Google Scholar logo with link to Google ScholarHorváth, Viktória 2010 “Filled Pauses in Hungarian: Their Phonetic Form and Function.” Acta Linguistica Hungarica 571: 288–306. Google Scholar logo with link to Google Scholar has found that öö can appear both between two words and within a word, and it may combine with silent pauses in various ways. These findings make it reasonable to explore the occurrence of öö in conversational repair phenomena in the framework of conversation analysis as well.

In the domain of investigating human social interaction, the practices when speakers interrupt the ongoing course of action in order to attend to possible trouble in speaking, hearing or understanding talk-in-interaction have been regarded as one of the fundamental structures of conversation for a long time (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google ScholarKitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar; Schegloff, Jefferson and Sacks 1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 531: 361–382. Google Scholar logo with link to Google Scholar). The phenomenon of repair has been studied across a range of languages (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google ScholarKitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar). Schegloff et al. (1977)Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 531: 361–382. Google Scholar logo with link to Google Scholar showed that self-repair is preferable to other-repair. The most common type of repair is same-turn self-repair. This means that the repair is initiated and accomplished by the speaker who produces the trouble-source in the turn containing the repairable (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google ScholarKitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar). Schegloff (2013) 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar describes the same-turn repair operations that speakers carry through in order to deal with a trouble-source in an ongoing turn-at-talk (Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 43): recycling, replacing, deleting, searching, parenthesizing, sequence-jumping, reformatting, reordering, inserting, and aborting. The only frequency examination of these repair operations in Hungarian has been conducted by Tar (2022)Tar, Cintia 2022 “Az önjavítási jelenségek magyar nyelvű spontán, baráti társalgásokban [Self-repair phenomena in Hungarian spontaneous conversations among friends].” Beszédtudomány – Speech Science 3: 75–104.Google Scholar logo with link to Google ScholarTar, Cintia 2022 “Az önjavítási jelenségek magyar nyelvű spontán, baráti társalgásokban [Self-repair phenomena in Hungarian spontaneous conversations among friends].” Beszédtudomány – Speech Science 31: 75–104.Google Scholar logo with link to Google Scholar. While the most frequent operation in her dataset was recycling repair (34.7%), some operations hardly occurred in her dataset: parenthesizing (2.8%), sequence-jumping (1.6%), deleting (0.7%), reordering (0.1%), and reformatting (0.0%) (Tar 2022Tar, Cintia 2022 “Az önjavítási jelenségek magyar nyelvű spontán, baráti társalgásokban [Self-repair phenomena in Hungarian spontaneous conversations among friends].” Beszédtudomány – Speech Science 3: 75–104.Google Scholar logo with link to Google ScholarTar, Cintia 2022 “Az önjavítási jelenségek magyar nyelvű spontán, baráti társalgásokban [Self-repair phenomena in Hungarian spontaneous conversations among friends].” Beszédtudomány – Speech Science 31: 75–104.Google Scholar logo with link to Google Scholar, 96). Fox and her colleagues (Fox et al. 2009Fox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez 2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google ScholarFox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez 2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar; Fox, Maschler and Uhmann 2010Fox, Barbara, Yael Maschler, and Susanne Uhmann 2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 42: 2487–2505. Google Scholar logo with link to Google ScholarFox, Barbara, Yael Maschler, and Susanne Uhmann 2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 421: 2487–2505. Google Scholar logo with link to Google Scholar) have carried out a cross-linguistic study on recycling and replacing in Bikol, English, Finnish, German, Hebrew, Indonesian, Japanese, Mandarin, and Sochiapam Chinantec. Németh (2012)Németh, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 44: 2022–2034. Google Scholar logo with link to Google ScholarNémeth, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 441: 2022–2034. Google Scholar logo with link to Google Scholar has explored Hungarian recycling and replacing. The main finding of these examinations is that there are underlying universal patterns in the practices of repair across the languages, however, these patterns can be manifested in various ways depending on the linguistic characteristics of languages. For example, Fox, Maschler, and Uhmann (2010Fox, Barbara, Yael Maschler, and Susanne Uhmann 2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 42: 2487–2505. Google Scholar logo with link to Google ScholarFox, Barbara, Yael Maschler, and Susanne Uhmann 2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 421: 2487–2505. Google Scholar logo with link to Google Scholar, 2504) predict that languages which have function words preceding content words (mainly verb-medial and verb-initial languages) show a preference for function word recycling rather than recycling back to content words. Németh (2012)Németh, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 44: 2022–2034. Google Scholar logo with link to Google ScholarNémeth, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 441: 2022–2034. Google Scholar logo with link to Google Scholar has found that, in Hungarian, function words tend to be the most frequent lexical category that serves as the destination of recycling (76%) (cf. Lerch 2007Lerch, Ágnes 2007 “Az ismétlés mint az önjavítás eszköze a magyarban [Repetition as a means of self-repair in Hungarian].” In Nyelvelmélet – nyelvhasználat [Language theory and language use], ed. by Tamás Gecső, and Csilla Sárdi, 123–130. Budapest: Tinta Könyvkiadó.Google Scholar logo with link to Google ScholarLerch, Ágnes 2007 “Az ismétlés mint az önjavítás eszköze a magyarban [Repetition as a means of self-repair in Hungarian].” In Nyelvelmélet – nyelvhasználat [Language theory and language use], ed. by Tamás Gecső, and Csilla Sárdi, 123–130. Budapest: Tinta Könyvkiadó.Google Scholar logo with link to Google Scholar), and Hungarian speakers tend to replace content words (66%). Given that demonstrative determiners and articles project an upcoming noun phrase in the language, and relative pronouns and conjunctions are at the beginning of clauses (Lerch 2007Lerch, Ágnes 2007 “Az ismétlés mint az önjavítás eszköze a magyarban [Repetition as a means of self-repair in Hungarian].” In Nyelvelmélet – nyelvhasználat [Language theory and language use], ed. by Tamás Gecső, and Csilla Sárdi, 123–130. Budapest: Tinta Könyvkiadó.Google Scholar logo with link to Google ScholarLerch, Ágnes 2007 “Az ismétlés mint az önjavítás eszköze a magyarban [Repetition as a means of self-repair in Hungarian].” In Nyelvelmélet – nyelvhasználat [Language theory and language use], ed. by Tamás Gecső, and Csilla Sárdi, 123–130. Budapest: Tinta Könyvkiadó.Google Scholar logo with link to Google Scholar, 127), this finding is in accordance with the prediction of Fox et al. (2010Fox, Barbara, Yael Maschler, and Susanne Uhmann 2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 42: 2487–2505. Google Scholar logo with link to Google ScholarFox, Barbara, Yael Maschler, and Susanne Uhmann 2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 421: 2487–2505. Google Scholar logo with link to Google Scholar, 2504).

Schegloff (2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 59) regards recycling as a repair operation if it is used in the case of overlapping talk in a “surviving turn”. According to Fox et al. (2009Fox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez 2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google ScholarFox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez 2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar, 75), recycling may also be a repair operation if it serves as a delaying device for treating e.g. selection problems. Németh (2012Németh, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 44: 2022–2034. Google Scholar logo with link to Google ScholarNémeth, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 441: 2022–2034. Google Scholar logo with link to Google Scholar, 2028) argues that while article-recycling indicates a word search process in Hungarian, recycling back to relative pronouns or conjunctions indicates a clause search process in the language. After analyzing a dataset of almost five hours of audio and video recorded everyday Hungarian conversations, Németh (2022) 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 194: 23–35. Google Scholar logo with link to Google Scholar 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 1941: 23–35. Google Scholar logo with link to Google Scholar has found (1) twenty-four occurrences of öö at the transition-relevance place with the function of managing the turn-taking system (see above), (2) sixteen occurrences of öö in overlapping talk (either in combination with recycling repair or without recycling) (see above), (3) 134 occurrences of öö within a TCU well before the transition-relevance place with the function of phrase or clause searching (either in combination with recycling repair or without recycling), (4) thirty-five occurrences of öö in word searching, (5) twenty-eight occurrences of öö in other repair phenomena (of which seventeen are in replacing), and (6) twelve occurrences of öö in the phenomenon when searching is converted into aborting. It seems that the function of öö is well describable in searching, replacing, and in the phenomenon when the searching is converted into aborting. The aim of the present paper, therefore, is a qualitative examination of the role of öö in these three operations in the above dataset. Before introducing the dataset and the methodology of the study, a brief overview of searching, replacing, and aborting is provided next.

Schegloff et al. (1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 531: 361–382. Google Scholar logo with link to Google Scholar, 363) describe the phenomenon when an item “is not available to a speaker when ‘due’”. Since this classic study, which opened the way for studying conversational repair as the natural part of everyday social interaction, word search has been extensively studied in the framework of conversation analysis (e.g. C. Goodwin 1980Goodwin, Charles 1980 “Restarts, Pauses, and the Achievement of a State of Mutual Gaze at Turn-beginning.” Sociological Inquiry 50(3–4): 272–302. Google Scholar logo with link to Google ScholarGoodwin, Charles 1980 “Restarts, Pauses, and the Achievement of a State of Mutual Gaze at Turn-beginning.” Sociological Inquiry 50(3–4): 272–302. Google Scholar logo with link to Google Scholar, 1987 1987 “Forgetfulness as an Interactive Resource.” Social Psychology Quarterly 50(2): 115–130. Google Scholar logo with link to Google Scholar 1987 “Forgetfulness as an Interactive Resource.” Social Psychology Quarterly 50(2): 115–130. Google Scholar logo with link to Google Scholar; Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar; Helasvuo, Laakso and Sorjonen 2004Helasvuo, Marja-Liisa, Minna Laakso, and Marja-Leena Sorjonen 2004 “Searching for Words: Syntactic and Sequential Construction of Word Search in Conversations of Finnish Speakers with Aphasia.” Research on Language and Social Interaction 37: 1–37. Google Scholar logo with link to Google ScholarHelasvuo, Marja-Liisa, Minna Laakso, and Marja-Leena Sorjonen 2004 “Searching for Words: Syntactic and Sequential Construction of Word Search in Conversations of Finnish Speakers with Aphasia.” Research on Language and Social Interaction 371: 1–37. Google Scholar logo with link to Google Scholar; Laakso 1997Laakso, Minna 1997 Self-Initiated Repair by Fluent Aphasic Speakers in Conversation. Helsinki: Finnish Literature Society.Google Scholar logo with link to Google ScholarLaakso, Minna 1997 Self-Initiated Repair by Fluent Aphasic Speakers in Conversation. Helsinki: Finnish Literature Society.Google Scholar logo with link to Google Scholar, 2003 2003 “Collaborative Construction of Repair in Aphasic Conversation: An Interactive View on the Extended Speaking Turns of Persons with Wernicke’s Aphasia.” In Conversation and Brain Damage, ed. by Charles Goodwin, 163–188. Oxford: Oxford University Press. Google Scholar logo with link to Google Scholar 2003 “Collaborative Construction of Repair in Aphasic Conversation: An Interactive View on the Extended Speaking Turns of Persons with Wernicke’s Aphasia.” In Conversation and Brain Damage, ed. by Charles Goodwin, 163–188. Oxford: Oxford University Press. Google Scholar logo with link to Google Scholar; Lerner 1996Lerner, Gene H. 1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google ScholarLerner, Gene H. 1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar; Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar). It is typical to initiate word search in the mid-course of a turn-constructional unit (TCU) (Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar), and the target of searching can be a name or a place, but it can also happen that the source of the problem is unclear and the target is not “precise” (Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 50). In Example (1), Béla22. Béla is a Hungarian male name. tries to find the precise name of a parking maneuver when he is describing his driving lessons.33.The transcriptions follow the conventions of conversation analysis (Jefferson 2004 2004 “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 13–31. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar 2004 “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 13–31. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar). The glosses include only the grammatical information that is marked by a non-empty morpheme in Hungarian. Where possible, I give the corresponding English word in an inflected form. Furthermore, the words in the extracts are not segmented into components by hyphens for two reasons.

(1)

bea002f002, 02:47

1 B:     az44.In order for the clause to be grammatical, this word would need a -t at the end to mark the accusative
					case. However, the native Hungarian speaker did not produce it. Taking into consideration that constructing the turn is an online
					activity, and that az nem volt könnyű ‘that wasn’t easy’ would be grammatical without megtanulni
					‘to learn’, it is possible that the speaker was first only to produce the structure az nem volt könnyű ‘that
					wasn’t easy’, then altered the turn-constructional unit by adding megtanulni ‘to learn’ without the correction of
					the first word (the understanding of the action is not endangered).     nem     volt          könnyű          megtanulni.=
         that    not     was.3sg       easy            pvb.learn.inf
         “it wasn’t easy to learn that. =
2        = főleg        az     ilyen     (.)     beparkolásos     manőver
         especially     the    such              parking          maneuver
         = especially this kind of (.) parking maneuver
3       ilyen   (.)   mittöminek     hívják          ezt          .h
        such          whatever.dat    call.3pl.def     this.acc
        this kind of (.) I don’t know what they call it .h”

In line 2, Béla uses the discourse marker ilyen ‘this kind of’, then a micropause, and the expression beparkolásos manőver ‘parking maneuver’. In the next line he repeats the whole noun phrase, but in the slot of ‘parking maneuver’ this time he puts an expression which contains the inflected form of the placeholder mittömi ‘whatever’,55.This placeholder is usually pronounced with an -o, but Béla is producing here an -ö. ‘I don’t know what’ (cf. Amiridze, Davis and Maclagan 2010Amiridze, Nino, Boyd H. Davis, and Margaret Maclagan (eds) 2010 Fillers, Pauses and Placeholders. Amsterdam/Philadelphia: John Benjamins Publishing Company. Google Scholar logo with link to Google ScholarAmiridze, Nino, Boyd H. Davis, and Margaret Maclagan (eds) 2010 Fillers, Pauses and Placeholders. Amsterdam/Philadelphia: John Benjamins Publishing Company. Google Scholar logo with link to Google Scholar). This seems to be an attempt to find a more appropriate or more precise name for the parking maneuver he refers to. Research into the use of ilyen ‘such’ as a discourse marker in Hungarian has shown that it can be used when the speaker has selectional difficulties, i.e. in the course of word searching (Vukov Raffai 2016Vukov Raffai, Éva 2016 “A diskurzusjelölő-választások életkori sajátosságai az így, ilyen, hát, mondjuk, ugye esetében [Age characteristics in discourse marker selections in the case of így, ilyen, hát, mondjuk, ugye ].” Magyar Nyelvőr 140: 483–497.Google Scholar logo with link to Google ScholarVukov Raffai, Éva 2016 “A diskurzusjelölő-választások életkori sajátosságai az így, ilyen, hát, mondjuk, ugye esetében [Age characteristics in discourse marker selections in the case of így, ilyen, hát, mondjuk, ugye].” Magyar Nyelvőr 1401: 483–497.Google Scholar logo with link to Google Scholar). Its use, together with the pause and the placeholder, supports the analysis of this phenomenon as searching.

Goodwin and Goodwin (1986)Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar emphasize that searching for a word is a visible activity and not merely a cognitive process occurring inside the speaker’s head. Consequently, it is not an individual activity on the part of the speaker but the other participants of the conversation can recognize and take part in it (Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar, 52). This recognition can happen not only when the speaker obviously indicates that they are in the activity of searching for a word, e.g. appealing for assistance by shifting their gaze to the recipient (cf. Kurhila 2006Kurhila, Salla 2006 Second Language Interaction. Amsterdam: John Benjamins. Google Scholar logo with link to Google ScholarKurhila, Salla 2006 Second Language Interaction. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar, 96), but uh(m)-s, sound stretches, pauses, cut-offs, word repetitions, and the movements of the eyes, eyelids, and lips may also refer to searching, and may make the recipient offer a candidate solution (Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar; Lerner 1996Lerner, Gene H. 1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google ScholarLerner, Gene H. 1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar, 2013 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar). That is to say, the activity of word search is designed for the conditional entry of the recipient (Lerner 1996Lerner, Gene H. 1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google ScholarLerner, Gene H. 1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar, 261). According to Goodwin and Goodwin (1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar, 57), in various cultures it is recognizable that the speaker withdraws gaze when beginning the process of searching and produces a ‘thinking face’. Lerner (2013) 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar explores cases when speakers utilize turn-constructional delaying practices for interactional purposes. For example, the practices used in searching for a word can be employed to perform another action as well, namely, displaying the speaker’s stance toward a formulation as a delicate. By using practices like uh(m) before voicing a term, the speaker may display reticence about voicing it (Lerner 2013 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 105).

Replacing is the repair operation whereby speakers substitute for a partially or wholly articulated element of the ongoing TCU another, different element with the sense that ‘this is the same utterance’ (Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 43), i.e. without aborting the TCU-in-progress. Replacing has been investigated in various respects in the previous conversation analytic literature, such as the site of repair initiation, the syntactic class of the target word, or the interactional function of the repair operation. Németh (2017) 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar reveals a variety of interactional motivations of replacing in Hungarian. In accordance with previous results in English, she points out that replacing can downgrade the force of the action (cf. Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google ScholarKitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar), and reduce speakers’ responsibility for what they are saying (cf. Jefferson 1974Jefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google ScholarJefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google Scholar). Furthermore, the use of replacing in Hungarian can display the speaker’s unease about what they are about to say when delivering criticism, avoiding offensive language, or delivering self-praise (Németh 2017 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar, 307). Let us consider a Hungarian example of replacing.

Address is possible in various forms in Hungarian (Domonkosi 2002Domonkosi, Ágnes 2002 Megszólítások és beszédpartnerre utaló elemek nyelvhasználatunkban [Forms of address and elements referring to the partner in conversation]. Debrecen: A Debreceni Egyetem Magyar Nyelvtudományi Intézetének Kiadványai [Proceedings of the Department of Hungarian Linguistics of the University of Debrecen] 79.Google Scholar logo with link to Google ScholarDomonkosi, Ágnes 2002 Megszólítások és beszédpartnerre utaló elemek nyelvhasználatunkban [Forms of address and elements referring to the partner in conversation]. Debrecen: A Debreceni Egyetem Magyar Nyelvtudományi Intézetének Kiadványai [Proceedings of the Department of Hungarian Linguistics of the University of Debrecen] 791.Google Scholar logo with link to Google Scholar). The two formal equivalents of ‘you’ are maga (plural: maguk) and ön (plural: önök). Ön is used in polite and/or distant formal situations, and its use indicates that the speaker and the recipient are not on familiar terms with each other. Maga is less polite and/or distant than ön, it can be used in relatively informal situations as well. However, it is much more formal than te ‘[informal] you’. Both maga and ön co-occur with a third person verb. In Example (2), two young women, Márta and Enikő are talking to Tibor, who is an older man. Enikő asks Tibor about Christmas.

(2)

bea004f003, 00:10

1     szeretnénk          önt          megkérdezni     hogy     a     karácsonyt
      like.cond.1pl.def     you.acc       pvb.ask.inf      that     the   Christmas.acc
      “we would like to ask you
2     azt     (.)     hogyan          töltötte         ön?     meg     hogyan
      that.acc         how             spent.3sg.def     you     and     how
      how (.) you have spent Christmas? and how
3     szokott       ma-     önöknél       zajlani?   (.)     az      egész
      used                  you.pl.ade     happen.inf          the     whole
          it is usually celebrated in your family? (.) the whole
4     (.)     ünnep?
              holiday
      (.) Christmas holiday?”

Tibor is addressed three times (in lines 1, 2, and 3). On the first two occasions, Enikő selects ön. It is plausible to assume that in line 3 the first syllable of maguknál ‘in your family’ (ma-) is replaced with önöknél ‘in your family’, which indicates that the use of maga and ön cannot alternate in the same interaction when addressing the same recipient.

Speakers may interrupt the ongoing turn-constructional unit and abandon what they have said altogether, then start the same action in a different form with another TCU. This is the repair operation of aborting (cf. Laakso and Sorjonen 2010Laakso, Minna, and Marja-Leena Sorjonen 2010 “Cut-off or Particle – Devices for Initiating Self-Repair in Conversation.” Journal of Pragmatics 42: 1151–1172. Google Scholar logo with link to Google ScholarLaakso, Minna, and Marja-Leena Sorjonen 2010 “Cut-off or Particle – Devices for Initiating Self-Repair in Conversation.” Journal of Pragmatics 421: 1151–1172. Google Scholar logo with link to Google Scholar; Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar). According to Schegloff (2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 52), there are two possible orientations towards a TCU that is left uncompleted: speakers may abandon what they have said altogether, or they may only abandon the way the turn has been done so far, in favor of another way of doing the same undertaking. Aborting is a similar repair operation to replacing, except that while replacing preserves the syntactic shape of the utterance (e.g. the type of the clause), aborting leaves the syntactic construction altogether uncompleted (Laakso and Sorjonen 2010, 1157). For this reason, it seems that the interactional functions that can be observable in the case of replacing can be fulfilled by using the repair operation of aborting as well. Németh (2017) 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar brings examples from everyday conversations in Hungarian where the speaker changes a stronger formulation to a weaker one in order to downgrade the force of the action by changing the whole TCU-so-far. Abandoning the way the TCU has been done so far can reduce the speaker’s responsibility when invoking other people’s opinions or can help avoiding offensive language in Hungarian (Németh 2017 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar 2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar, 307). In Example (3), Ágnes refers to a Roma showman’s non-Roma manner of speaking.

(3)

SZTEPSZI2, 16:10

1 Á:  de      hogy     ilyen     egész     norm-     tehát     hogy     ö     (.)    nem:    hm:
      but     that     such      quite     norm-     that.is   that     uh           not     u:mm
      “but like quite norm- that is, uh (.) he didn’t u:mm
2     nem:    hasz-    nem     volt     akcentusa:
      not     use-     not     was.3sg   accent.poss.3sg
      didn’t use- didn’t have an accent”

In line 1 Ágnes interrupts the element norm- which follows egész ‘quite’, therefore it must be a partially articulated adjective or adverb. The only other Hungarian lexemes which start this way and are not the derivations of norma ‘norm’ are Normandia ‘Normandy’ and normann ‘Norman’. Since neither of them fit into the conversation, the adjective that is cut off is likely to be normális ‘normal’. Referring to a Roma man’s non-Roma manner of speaking as ‘normal’ is quite offensive: it implies that the Romas’ manner of speaking is not normal. Perhaps that is why Ágnes cuts off the word and starts another TCU that underlies the same action: tehát hogy ö (.) nem: hm: nem: hasz- nem volt akcentusa: ‘that is, uh (.) he didn’t u:mm didn’t use- didn’t have an accent’. The reformulation marker tehát ‘that is’ after the cut-off in line 1 supports the analysis of the phenomenon as aborting.

In this paper I will analyze the use of öö in the course of the three repair operations described above. The organization of the paper is as follows. In Section 2, I introduce the Hungarian dataset of the study and the methodology of my analysis. In Section 3, I analyze the appearance of öö in word search (3.1), the [conjunction + öö + clause] technique which, as I will argue, can be regarded as searching for a clause (3.2), searching converted into aborting (3.3), and when speakers search for a more appropriate candidate after producing a less appropriate word or segment (3.4). In Section 4, I summarize the results of my examination and conclude the study. Before turning to the description of the phenomena listed above, I introduce the Hungarian dataset and the methodology of the research.

2.The Hungarian dataset and the methodology of the study

Although the Hungarian Gigaword Corpus (Oravecz, Váradi and Sass 2014Oravecz, Cs., Tamás Váradi, and Bálint Sass 2014 “The Hungarian Gigaword Corpus.” In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014), ed. by Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Slelios Piperidis, 1719–1723. Reykjavik: European Languages Resources Association (ELRA).Google Scholar logo with link to Google ScholarOravecz, Cs., Tamás Váradi, and Bálint Sass 2014 “The Hungarian Gigaword Corpus.” In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014), ed. by Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Slelios Piperidis, 1719–1723. Reykjavik: European Languages Resources Association (ELRA).Google Scholar logo with link to Google Scholar) is the largest corpus in Hungarian and contains spontaneous speech as well (5.4%), these are radio reports, which are edited, and self-repair phenomena may be cut out of them. The findings on öö presented in this study are thus based on a Hungarian dataset that has two parts. The first part consists of eight conversations, approximately fifteen minutes long each, from the multifunctional spontaneous Hungarian speech database (BEA; Gósy 2012 2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar 2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar). This database contains speech data from nearly 500 participants, representative of the central region of Hungary in sex, age, dialect, and educational background. The eight conversations used in the present study from the BEA database were randomly selected. The second part consists of nine approximately twenty minute long conversations from the SZTEPSZI dataset recorded at the University of Szeged.66.The extracts are indicated as SZTEPSZI1, SZTEPSZI2, etc. and bea001, bea002, etc., respectively. The total length of the dataset is four hours, fifty-eight minutes, and forty-two seconds (4:58:42). Since my dataset does not include the full range of variability of the Hungarian population, and for this reason it does not meet the criteria of representativeness (Biber 1993Biber, Douglas 1993 “Representativeness in Corpus Design.” Literary and Linguistic Computing 8: 243–257. Google Scholar logo with link to Google ScholarBiber, Douglas 1993 “Representativeness in Corpus Design.” Literary and Linguistic Computing 81: 243–257. Google Scholar logo with link to Google Scholar: 243), it cannot be called a corpus. However, it is suitable for the aim of the present paper, which is not to make frequency statements (though the number of occurrences is given), but to describe the function of öö-s occurring in the examples of the present dataset. The arguments used in the analyses are based on the position of öö in the turn-constructional unit, the general regularities of conversation (e.g. the preference for progressivity; Schegloff 1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar), and the action that the turn implements.

There were three participants in each conversation (altogether thirty-eight speakers across seventeen interactions). The age of the participants in the BEA database ranges from twenty to eight-five (Gósy 2012 2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar 2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar, 336). The subjects of the SZTEPSZI dataset were university students, most of whom were friends with each other. While the SZTEPSZI dataset consists of video recordings, the conversations from the BEA database are audio recorded. The fact that only nine conversations are video recorded in the dataset does not influence the reliability of my results, because the participants’ gaze behavior only has a supportive function in my argumentation. In the examples of searching, for instance, the analysis is built on the other participants’ cooperation, the characteristics of the subsequent word, etc. The observation that the speaker usually avoids eye contact during the production of öö in these examples is just a further argument for their status as searching processes.

In the SZTEPSZI dataset the participants were asked to talk freely about anything. However, in order to help them start talking, participants were given a sheet of paper with three questions which were potentially interesting to them. The use of these questions was nonetheless optional, the participants were encouraged to talk about anything they were interested in. The questions were as follows: (i) Before his/her most important exam, your friend asks you to help them cheat in an elaborate way. If the cheating comes to light, both of you will be dismissed from the university. Would you help your friend? (ii) If light drugs were legalized in Hungary, would people drink less alcohol? (iii) Is there a need for further talent shows on Hungarian television? Similarly, the conversations from the BEA speech database were initiated by one of the participants asking a question (e.g. How have you spent Christmas? What is your opinion on getting a driving license in Hungary? etc.). This shaped the course of the interaction to a certain extent,77.For this reason, Németh (2021) 2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar 2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar regards these conversations as semi-spontaneous. but the participants could equally contribute to the conversations in each case.

We can say about both datasets that although the participants were talking in an acoustically controlled room in order to reach the best possible sound quality, and the recordings were made for research purposes, the subjects were not controlled at all: they could talk freely, and before the conversations they did not know what the initial topics would be, therefore they could not prepare for the discussion in advance. For this reason, the conversations can be regarded as spontaneous in both datasets (cf. Gósy 2012 2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar 2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar, 343).

As to the methodology of my research, with a case-by-case analysis I explored all the occurrences of öö. Then, in a qualitative analysis, using conversation analysis, I examined the position of each occurrence of öö in the turn and sequence. When öö occurred as the first element of a turn, I regarded its position as turn-initial, when it was the last element of a turn, it was regarded as turn-final, and all the other occurrences of öö are called turn-medial öö-s in the paper. Then, I examined whether the function of öö can be identified in the particular cases. Since turn-initial and turn-final öö-s may manage the turn-taking system rather than having a repairing function, in the present paper I focus on the occurrences in which öö appears in the course of same-turn self-repair, in turn-medial position, before a transition-relevance place. After finding the öö-s used by speakers in the course of same-turn self-repair, I attempted to find an answer to the following question: What do speakers do by using öö in the particular places in the discourse in the individual cases? Since the analyst does not always possess all the necessary information to answer this question, there were occurrences that showed a resistance to analysis (cf. Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 60), i.e. the function of the öö-s in these examples could not be identified. However, I have identified three repair phenomena, in the course of which speakers use öö with a describable function. In the next section, I aim to describe these phenomena and the role of öö in them.

3.The role of öö in the repair operations of searching, searching converted into aborting, and replacing in Hungarian conversations

3.1Searching for a word

As Goodwin and Goodwin (1986)Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar observed, the activity of searching is not an individual phenomenon, but finding the sought-for item can be the result of the participants’ cooperation, even when the speaker does not obviously appeal for assistance by shifting gaze to the recipient. In Example (4), Márton, Beáta, and Johanna are talking about Hungarian musicians who have become famous outside Hungary. Beáta remarks that Hungarian musicians tend to become world famous only in the field of classical music.

(4)

SZTEPSZI8, 05:40

ex4.svg

When saying hát olyanok vannak akik világhírűek mondjuk ‘well, there are people who are world famous, let’s say’, Beáta is looking at Márton.88.For the sake of simplicity, in the transcription of gaze behavior we can always see only two participants’ gaze directions out of the three. These two participants are the speaker and the recipient with whom the speaker is in interaction regarding her/his gaze direction in the extract. The lines are divided into units. A new unit begins if at least one of the participants’ gaze direction changes. Then, she lowers her gaze and says ööö, while Márton continues to look at Beáta. Right after Beáta’s ööö, while she is still looking down, Márton offers violin and piano as two musical instruments Hungarian musicians are famous for (hegedű meg zongorán ‘for violin and piano’; line 2). With this, Márton treats Beáta’s behavior as searching for the name of musical instruments which Hungarian musicians who became world famous play. In other words, by offering candidates, he displays an understanding of the prior turn as one that involved searching. I argue that the clues which indicate for Márton that Beáta is in the act of searching are the expression mondjuk ‘let’s say’, ‘for example’, which projects that what follows is an example (line 1), the producing of ööö (line 1), and Beáta’s gaze behavior. That is to say, as has already been mentioned, speakers tend to withdraw gaze when beginning the process of searching (Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar, 57), and Beáta lowers her gaze right at the beginning of ööö. As soon as Beáta recognizes that Márton is saying the word hegedű ‘violin’ in line 2, she produces hegedűművészek ‘violin artists’ (line 3). By completing her turn with one of the two items that Márton offered, Beáta not only acknowledges that her earlier conduct was really a word searching but also that she has been looking for one of the very items Márton offered. This shows well that each and every turn in a conversation displays the participants’ way of understanding of the prior turn (Heritage 1984Heritage, John 1984 “Conversation Analysis.” In Garfinkel and Ethnomethodology, ed. by John Heritage, 233–292. Cambridge: Polity Press.Google Scholar logo with link to Google ScholarHeritage, John 1984 “Conversation Analysis.” In Garfinkel and Ethnomethodology, ed. by John Heritage, 233–292. Cambridge: Polity Press.Google Scholar logo with link to Google Scholar, 259). Öö in (4), therefore, is a delaying practice for Beáta to find the next appropriate element. An important argument for this analysis is that Márton analyzes Beáta’s activity as searching.

Given that öö is a non-lexical phenomenon, it is an interesting question why speakers use it as a delaying technique before the transition-relevance place: here the speaker is not in danger of losing the floor, and a pause seems to be more economical. However, in conversation there is a preference for progressivity (Schegloff 1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar), the action should progress towards possible completion, and if a speaker starts a turn, she has not only the right but also the obligation to produce at least one action, one TCU (Schegloff 2007 2007 Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar 2007 Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar). If her turn-so-far is not such a recognizably complete action yet, she has to account for the suspension. Producing öö can be such an account: it indicates that the speaker is aware of the normative expectation that is imposed on her to produce at least one recognizably complete action in the conversation and will satisfy this obligation.

In the next four examples öö appears immediately before longer, and/or rare, and/or foreign-originated words. According to Iványi (2003Iványi, Zsuzsanna 2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 127: 76–91.Google Scholar logo with link to Google ScholarIványi, Zsuzsanna 2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 1271: 76–91.Google Scholar logo with link to Google Scholar, 87), the target words of the ‘tip of the tongue’ phenomenon99.The tip of the tongue (TOT) phenomenon is a type of word search process when speakers cannot verbalize a word that is otherwise familiar to them (Brown and McNeill 1966Brown, Roger, and David McNeill 1966 “The “Tip of the Tongue” Phenomenon.” Journal of Verbal Learning and Verbal Behavior 5: 325–337. Google Scholar logo with link to Google ScholarBrown, Roger, and David McNeill 1966 “The “Tip of the Tongue” Phenomenon.” Journal of Verbal Learning and Verbal Behavior 51: 325–337. Google Scholar logo with link to Google Scholar; Iványi 2003Iványi, Zsuzsanna 2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 127: 76–91.Google Scholar logo with link to Google ScholarIványi, Zsuzsanna 2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 1271: 76–91.Google Scholar logo with link to Google Scholar). in German and Hungarian conversations consist, most of the time, of three or more syllables, they are rarely used or are of foreign origin. In Example (5), the speaker produces öö before a long compound word which contains an internal pause, too, and which is not part of everyday language use (illóolaj(.)mániás ‘essential.oil(.)sucker’).

(5)

bea003n001, 11:19

1 A:  én     nagyon     ilyen     mécses     meg     ilyen     ilyen    kis
      I      very       such      lantern    and     such      such     small
      “I am a sucker for such lampions and such such small
2     öö     illóolaj(.)mániás          vagyok     ((laughing))
             essential.oil(.)sucker     am
      öö essential oils”((laughing))

It is worth noting that the speaker recycles back to the pronoun ilyen ilyen ‘such such’ before producing the adjective phrase. Since the use of ilyen in itself can indicate the speaker’s selectional difficulties (Vukov Raffai 2016Vukov Raffai, Éva 2016 “A diskurzusjelölő-választások életkori sajátosságai az így, ilyen, hát, mondjuk, ugye esetében [Age characteristics in discourse marker selections in the case of így, ilyen, hát, mondjuk, ugye ].” Magyar Nyelvőr 140: 483–497.Google Scholar logo with link to Google ScholarVukov Raffai, Éva 2016 “A diskurzusjelölő-választások életkori sajátosságai az így, ilyen, hát, mondjuk, ugye esetében [Age characteristics in discourse marker selections in the case of így, ilyen, hát, mondjuk, ugye].” Magyar Nyelvőr 1401: 483–497.Google Scholar logo with link to Google Scholar, see above), its repetition as a delaying strategy is an even stronger indicator of word search in the example (cf. Fox et al. 2009Fox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez 2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google ScholarFox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez 2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar; Németh 2012Németh, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 44: 2022–2034. Google Scholar logo with link to Google ScholarNémeth, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 441: 2022–2034. Google Scholar logo with link to Google Scholar). By producing öö, Anna orients herself to the preference for progressivity (cf. Schegloff 1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar), and she can vocalize without producing lexical elements. In this way, she can indicate that in spite of her selectional difficulties, she is to produce the noun projected by the adjective kis ‘small’. In Example (6), öö also projects further talk.

(6)

bea004f003, 07 :00

1 A:   a (.)   mindenféle     öö        marista          iskolatestvérekkel     együtt
       the     every.kind.of            marist           school.sisters.com      together
       “together with various öö marist school sisters
2      pakoltuk     ott     a     teherautókra     a     téglát
       loaded.1pl   there   the    trucks.sub       the    brick.acc
       we loaded bricks on trucks”

Saying öö after the quantifier mindenféle ‘various’ (line 1), the speaker indicates to the other participants that she will continue her turn with the next relevant lexical element, which turns out to be a rare compound word belonging to the Christian jargon (marista iskolatestvérek ‘marist school sisters’). In the last example of this kind (Example [7]), ö precedes a word of Japanese origin referring to Japan’s ancient martial art (szumóztunk ‘we did sumo wrestling’).

(7)

SZTEPSZI1, 02:08

ex7.svg

When producing eleinte pingpongoztunk, .h utána ‘in the beginning we played table tennis, .h then’, Kati is looking upwards, then during the production of Lillával ‘with Lilla’ in line 2 she looks at Ági. Exactly for the production of ö she looks down, then right after ö she looks back at Ági. This means that both before and after ö she is looking at Ági, but during the production of ö she is looking down.

In the above three examples we can observe that öö (in Example [7], ö) occurs before rare words or words of foreign origin. This makes it plausible that the speaker has a problem producing that next term, in other words, they seem to have engaged in a search for that precise term. This is in accordance with Lerner’s (2013 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 102) finding, that turn-compositional delaying techniques can serve for delaying a precise term in the repair operation of searching. By using öö, speakers interrupt the turn-in-progress, i.e. they violate the preference for progressivity on the micro level (cf. Schegloff 2013 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 42) in order to deal with a problem concerning the next relevant item due. In Example (7) the speaker’s gaze aversion also supports the analysis of the example as word searching (cf. Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar, 57). As soon as she is ready to continue the turn, and the interaction can progress towards possible completion again, she directs her gaze back to her recipient. In these occurrences, öö seems to be a delaying technique with the deployment of which speakers can legitimize that the progressivity of the unit-under-way is halted when they are in the activity of searching for the next relevant word or phrase.

Lerner (2013 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar 2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar, 104) points out that speakers may hesitate not only before a precise term but also prior to a delicate term, for instance, when they negatively evaluate somebody’s character or action. This practice can be observed in the Hungarian dataset as well, when öö is used immediately before a potentially offensive word (Example [8]).

(8)

SZTEPSZI2, 07:10 1010. Tök, shortened from tökéletesen ‘perfectly’ and homonymous with tök ‘pumpkin’, in evaluative APs serves as an upgrading modifier in colloquial Hungarian.

ex8.svg

Mariann produces the word mondjuk ‘let’s say’ before mozgássérült ‘disabled’, which can serve as a mitigating device before a potentially delicate expression. The unit-under-way is further marked with öö, which, together with the speakers’ gaze behavior, displays the turn as delicate. When producing tök jó ‘very good’, the speaker, Mariann directs her gaze towards Évi. Évi also looks at Mariann (the third participant is looking downwards). When saying the next words (hangja van és épp ‘voice and he is just’), Mariann is looking to the upper left. Then, she lowers her gaze and is looking down when saying mondjuk öö ‘let us say öö’. Since Mariann is not looking at her, Évi lowers her gaze and is looking down for the period of mondjuk öö moz ‘let us say öö dis’. After producing öö Mariann says mozgássérült ‘disabled’ looking at Évi again, who, after one syllable of the word also looks back at her. This behavior may indicate Mariann’s selectional difficulties in finding an interactionally appropriate word (cf. Jefferson 1974Jefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google ScholarJefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google Scholar).

In view of the above examples, I argue that the function of öö in searching can be complex. On the one hand, it can help speakers delay the sought-for item, and, on the other hand, by producing öö, they can project further talk, and, thus, commit themselves to making the unit they have begun recognizably complete. Furthermore, the use of öö in searching can elicit the recipient’s assistance without the speaker’s explicit solicitation. In these examples, öö serves both as a practice that interrupts the progressivity of the action in order to attend to a problem in producing talk, and as a practice whereby the speaker deals with that problem.

There are 117 occurrences of öö in the Hungarian dataset where öö is next to a conjunction that introduces a clause. These cases can be described as the [conjunction + öö + clause] technique. The function of öö here, as I will argue in the next section, is to delay the next clause due and make it obvious to the other participants that the speaker is ready to fulfill the requirement to make the TCU-in-progress recognizably complete.

3.2The [conjunction + öö + clause] technique: Searching for a clause

Hogy ‘that’ is a typical conjunction in Hungarian. It can introduce subordinate clauses of various types that can be arguments and adjuncts as well (Kenesei 1994Kenesei, István 1994 “Subordinate Clauses.” In The Syntactic Structure of Hungarian, ed. by Ferenc Kiefer, and Katalin É. Kiss, 275–354. San Diego: Academic Press. Google Scholar logo with link to Google ScholarKenesei, István 1994 “Subordinate Clauses.” In The Syntactic Structure of Hungarian, ed. by Ferenc Kiefer, and Katalin É. Kiss, 275–354. San Diego: Academic Press. Google Scholar logo with link to Google Scholar). When öö occurs in turn-medial position, well before the transition-relevance place in the dataset, it can follow hogy ‘that’.

In the next example Zsuzsi tries to find a legitimate excuse for cheating in an exam. She introduces a subjective subordinate clause by using hogy öö ‘that öö’.

(9)

SZTEPSZI5, 15:22

ex9.svg

At the beginning of the turn-constructional unit Zsuzsi is looking down. Then exactly after producing the conjunction hogy ‘that’ she turns her head to the left and looks away from both of the other participants (while Martin sits across from her, Daniel is on her right). This looking away lasts exactly for the period while she is producing öö. After finishing öö, she immediately looks toward Martin. This gaze behavior is similar to what we could observe in the case of searching (cf. Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar). Since öö is frequently accompanied by pauses and repetitions, i.e. turn-constructional delaying techniques in this position (Németh 2020 2020 “A nemlexikális öö hang interakciós szerepének elemzése magyar nyelvű társalgásokban [An analysis of the interactional role of the non-lexical Sound öö in Hungarian conversations].” Jelentés és Nyelvhasználat 7: 23–50. Google Scholar logo with link to Google Scholar 2020 “A nemlexikális öö hang interakciós szerepének elemzése magyar nyelvű társalgásokban [An analysis of the interactional role of the non-lexical Sound öö in Hungarian conversations].” Jelentés és Nyelvhasználat 71: 23–50. Google Scholar logo with link to Google Scholar), it is plausible to assume that it has a similar function here as in searching. However, it is not probable that these öö-s take part in word searching. That is to say, it is not probable that it is almost always a term following a conjunction and being the first word of a clause that the speaker has trouble with and has to search for. In addition, clause-initial conjunctions are frequently followed by another function word (e.g. an article) in Hungarian (Németh 2012Németh, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 44: 2022–2034. Google Scholar logo with link to Google ScholarNémeth, Zsuzsanna 2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 441: 2022–2034. Google Scholar logo with link to Google Scholar), which again reduces the probability that the speaker has trouble with the next word after the conjunction. Furthermore, concerning their intonation, this kind of öö’s belong to the conjunction that they follow, rather than to the word that follows them, as if the conjunction and öö were one single word (hogy-öö ‘that-öö’). In other words, speakers seem to attach öö as a clitic onto the previous conjunction in these examples (cf. ‘and-uh’ in Clark and Fox Tree 2002Clark, Herbert H., and Jean E. Fox Tree 2002 “Using Uh and Um in Spontaneous Speaking.” Cognition 84: 73–111. Google Scholar logo with link to Google ScholarClark, Herbert H., and Jean E. Fox Tree 2002 “Using Uh and Um in Spontaneous Speaking.” Cognition 841: 73–111. Google Scholar logo with link to Google Scholar). This use of öö seems to delay the syntactic unit that the conjunction introduces, i.e. the clause. This suggests that Hungarian speakers in this dataset, when constructing their turns, treat clauses as single units. In this sense we can regard hogy-öö ‘that-öö’ as a ‘clause searching’ technique.

As has been mentioned above, öö can be accompanied by other phenomena in the dataset. Let us consider an example for the [conjunction + öö + pause + clause] technique.

(10)

SZTEPSZI5, 01:06

ex10.svg

As has been mentioned, because of the preference for progressivity (Schegloff 1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar), a pause in the TCU-in-progress is dispreferred. However, if the speaker produces öö+silence, öö can legitimize the subsequent pause, because the speaker using it promises continuation (Németh 2022 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 194: 23–35. Google Scholar logo with link to Google Scholar 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 1941: 23–35. Google Scholar logo with link to Google Scholar). In Example (10), the speaker’s gaze withdrawal supports the clause searching function of hogy-öö (0.5) ‘that-öö (0.5)’. When Zsuzsi starts her turn-constructional unit in line 1, she directs her gaze towards Daniel, who sits on her right and is looking down. Daniel looks at Zsuzsi when he recognizes the word véleményem ‘my opinion’ which projects the woman’s action, namely, that she is about to express her opinion (cf. Jefferson 1984 1984 “Notes on Some Orderlinesses of Overlap Onset.” In Discourse Analysis and Natural Rhetoric, ed. by Valentina D’Urso, and Paolo Leonardi, 11–38. Padua: Cleup Editore.Google Scholar logo with link to Google Scholar 1984 “Notes on Some Orderlinesses of Overlap Onset.” In Discourse Analysis and Natural Rhetoric, ed. by Valentina D’Urso, and Paolo Leonardi, 11–38. Padua: Cleup Editore.Google Scholar logo with link to Google Scholar). Then, when Zsuzsi starts saying kapcsán ‘concerning’ (line 2), she turns her head down and looks in front of herself. She maintains this gaze direction exactly till the end of the pause, then looks back at Daniel. This means that when she produces hogy-öö (0.5) ‘that-öö (0.5)’ she avoids eye contact with the other participants.

Just like pauses, conjunction repetition can also accompany öö in Hungarian (Németh 2020 2020 “A nemlexikális öö hang interakciós szerepének elemzése magyar nyelvű társalgásokban [An analysis of the interactional role of the non-lexical Sound öö in Hungarian conversations].” Jelentés és Nyelvhasználat 7: 23–50. Google Scholar logo with link to Google Scholar 2020 “A nemlexikális öö hang interakciós szerepének elemzése magyar nyelvű társalgásokban [An analysis of the interactional role of the non-lexical Sound öö in Hungarian conversations].” Jelentés és Nyelvhasználat 71: 23–50. Google Scholar logo with link to Google Scholar). In the next example, three young friends are discussing whether alcohol consumption would decrease in Hungary if light drugs were legalized.1111.Since in the extract the speaker does not look at either of the other participants, the transcription shows only the speaker’s gaze behavior.

(11)

SZTEPSZI8, 09:48

ex11.svg

Johanna presents an argument for her earlier opinion which the other two participants agreed with: alcohol consumption would not decrease in Hungary if light drugs were legalized. Her argument is that although they are not legal, some people use drugs even now. When she says mer most hogy- ‘because now that’, she looks straight ahead (the other two participants sit on either side of her). As soon as she starts saying hogy-ööö ‘that-ööö’, she lifts her eyes to the ceiling, and maintains this gaze direction exactly until the end of ööö. Finishing ööö she looks down. This means that during the production of the TCU she does not look at either of her interlocutors. Her gaze lifting suggests the activity of searching (cf. Goodwin and Goodwin 1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar, 57). To sum up, the three phenomena discussed above (conjunction repetition, öö, and pause) seem to strengthen each other’s delaying function when occurring together at the beginning of clauses.

3.3Searching converted into aborting

Schegloff (1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar, 273) describes a phenomenon when the repair operation of searching is converted into an aborting repair. Instead of continuing the search for the next item due, the speaker reconstructs the whole turn-so-far and carries out the same action in a different form. In Example (12), in line 5 the speaker, instead of finding a fitting next element, reconstructs the turn-so-far.

(12)

Schegloff (1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar, 273)

1  Merle:  So how’s Michelle.
2          (1.0)
3  Robin:  They brought her ho:me.
4          (0.7)
5  Robin:  She hadda wait up the:re fo:r- u-she:’s been there since eight uh’ clock 
           this morning’n at six thirty she called me… (0.5) Said “Please com’n get me…

Schegloff (1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar, 273) suggests that the stretches in the words the:re and fo:r may preindicate a search for the next item due. However, the repair is not pursued, and the search is converted into an aborting. This ‘repair conversion’ can be observed in the Hungarian dataset as well. In this section I explore the occurrences of öö in it.

In Example (13), Anna is talking about her father, who is a surgeon at a military hospital. Replying to Feri’s question she tries to explain how her father can be a professional soldier if he has never served as a conscripted soldier.

(13)

bea008f006, 10:19

1   A:  ő     mondjuk        katona  nem     volt,
        s/he  say.imp.1pl     soldier not     was
        “he, by the way, has never been a soldier
2       tehát     sorkatona     sose     volt,
        that.is   conscript     never    was
        that is, he has never been a conscript”
3   F:  Lehet     úgy          valaki       hivatásos     katona
        can.be    in.that.way  somebody     professional  soldier
        “Is it possible for somebody to be a professional soldier
4       hogy     nem     volt    sorkatona?
        that     not     was     conscript
        without being a conscript before?”
5   A:  Hogyne      lehetne       o-   úgy          úgy
        how.not     can.be.cond         in.that.way  in.that.way
        Of course it is possible o- the way, the way (to become a soldier) is
6       lehet     hogy     öö     akkor     (.)     amiko:r     öö     (1.0)
        can.be    that            then              when
        that öö whe:n öö (1.0)
7       na     mindegy      tehát     ő       orvos     ugye
        well   whatever     so        s/he    doctor    you.know
        whatever, that is to say, he is a doctor you know”

In line 5, Anna first starts a subordinate structure (úgy lehet hogy öö akkor (.) amiko:r ‘the way [to become a soldier] is that öö whe:n’). At the end of line 6, with öö and a one second pause she interrupts the syntactic structure she has begun. Before this interruption, in line 5 she already uses a cut-off and replaces the segment o-, and we can also see several turn-constructional delaying techniques both in line 5 and 6 (repetition, micropause, elongation). These seem to be in service of finding fitting elements to complete the structure started at the beginning of line 5. The delaying techniques accompanying the two öö-s in line 6 suggest that these öö-s also have a delaying function in a searching process. With the reformulation markers na mindegy ‘whatever’ and tehát ‘that is to say’ in line 7, Anna launches a different TCU addressing the same undertaking, in other words, she converts the searching into an aborting. She abandons the subordinate structure at the conjunction amikor ‘when’, and starts her explanation from another direction: her father is a doctor and went to work at a military hospital, in this way he became a professional soldier. At the end of line 6, we can see the [conjunction + öö + pause] technique again, but now the pause that the previous öö legitimizes is longer (1 second), which suggests that the speaker may have difficulties in continuing the TCU that she has started.

3.4The use of öö in the repair operation of replacing

In this section, I will explore the appearance of öö in replacing repairs where speakers search for a more appropriate candidate after producing a less appropriate word or segment. In these cases öö occurs just after a cut-off.

Cutting off the TCU-in-progress is a common technique to halt progressivity and indicate a potential repair (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google ScholarKitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar, 239). Let us see examples for the [cut-off + öö] combination in Hungarian. In Example (14), Anna is describing what her family usually eats at Christmas.

(14)

bea003n001, 7:00

1   A:   mi     is     eszünk   mondjuk       év
         we     also   eat.1pl  say.imp.1pl     year
         “anyway, we also eat (lentils) during the year
2        közben     is     de     szi-     öö     ö     január    elsején
         during     also   but    New-                  January   first.poss.3sg.sup
         but on New- öö ö on the first of January
3        mindig     lencsét       eszünk.
         always     lentil.acc     eat.1pl
         we always eat lentils.”

In line 2, Anna interrupts her ongoing TCU by a cut-off (szi-). Then after a longer öö and a shorter one, she produces január elsején ‘on the first of January’. In Hungarian szilveszter means ‘New Year’s Eve’ (from St. Sylvester’s Day, on December 31). Since the appropriate expression, i.e. the repair outcome (‘on the first of January’) is the day after New Year’s Eve, we can assume that the word which Anna interrupts is szilveszter ‘New Year’s Eve’. The cut-off already makes the elements szi- and január elsején as ‘an error and its correction’ recognizable for the recipient (cf. Jefferson 1974Jefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google ScholarJefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google Scholar, 187),1212.The cut-off is a component of the minimal form of the Error Correction Format (Jefferson 1974Jefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google ScholarJefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google Scholar, 187). therefore, I argue that the cut-off has a repair-initiating, trouble-alerting function in this example, and the use of öö is a delaying technique for Anna to find the appropriate time adverb instead of szilveszter ‘New Year’s Eve’. At the same time, öö can be an indication to the other participants that a repair phenomenon is in progress. According to the preference hierarchy model of repair operations (Németh 2021 2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar 2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar), replacing violates the preference for progressivity in more respects than searching: while in searching the action is merely suspended, in the case of replacing the TCU is ‘on the wrong track’: the speaker has produced inapposite segments. The use of öö in replacing repair, before the speaker can produce the repairing segment, may thus be even more warranted than in the course of searching.

In the last example we can observe a similar case. Three university students, Enikő, Karcsi, and Anett talk about their high school experiences. In Example (15), they are describing their high school graduation photo collages.

(15)

SZTEPSZI4, 10:07

ex15.svg

In line 1, Enikő interrupts the word görög ‘Greek’ (gör- ‘Gre-’), produces öö, then replaces the half-pronounced word with római ‘Roman’. In line 3 she makes the replaced segment and its inappropriateness explicit (nem görög ‘not Greek’). This suggests that she assumes the partially articulated, inapposite word to be recognized by the other participants. While the cut-off in the word görög ‘Greek’ seems to have a repair-initiating, trouble indicating function, the öö in line 1 promises continuation with the next relevant and appropriate form: the proper nationality name for the general Enikő mentions. Enikő is directing her gaze downwards when producing the replacing repair. Using öö after a cut-off, therefore, can indicate that, although the speaker finds the partially articulated element problematic, she is ready to address the problem and continue the turn with the next appropriate lexical element(s).

4.Conclusion

In this paper I have examined the use of the non-lexical schwa-like sound öö in the course of three self-initiated same-turn self-repair strategies in Hungarian conversations, namely, in searching, searching converted into aborting, and replacing. The use of öö before the transition-relevance place projects further lexical elements. During the practice of repair, the co-interactants of the conversation need to interrupt the ongoing action in order to attend to possible problems in speaking, hearing, or understanding the talk (Kitzinger 2013Kitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google ScholarKitzinger, Celia 2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar; Schegloff et al. 1977Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A., Gail Jefferson, and Harvey Sacks 1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 531: 361–382. Google Scholar logo with link to Google Scholar). Since in conversation there is a preference for progressivity (Schegloff 1979Schegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google ScholarSchegloff, Emanuel A. 1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar), a repair operation is inherently dispreferred. In Hungarian, when speakers are involved in the process of repair, by deploying the non-lexical öö they can vocalize and promise continuation without producing lexical elements. Öö thus serves as a delaying technique and an indication that the speaker is aware of the obligation to make the unit-under-way recognizably complete and is ready to satisfy this obligation. When the TCU-in-progress is halted and verbally marked with öö in word search processes, the öö can elicit the recipient’s assistance without the speaker’s explicit solicitation. Following conjunctions, öö can delay the clause that the conjunction introduces, and it can combine with conjunction recycling and/or silence. If the speaker produces öö+silence, öö can legitimize the subsequent pause. In the case of searching converted into aborting, öö displays the speaker’s orientation to the continuation of the TCU as problematic, and the aborting repair expresses that the TCU-so-far is sub-optimally formed. Finally, in replacing repair, producing a cut-off exhibits the partially articulated word as problematic, and öö is an indication that the speaker is ready to address the problem and continue the turn.

To sum up, when the turn-in-progress is suspended and marked with öö in the course of the three repair operations under investigation, the speaker displays to the other participants that the turn is problematic in some respects, but she is ready to attend to the problem and make the action-in-progress recognizably complete.

In the conversations where visual analysis was possible, during the production of öö, speakers usually did not look at anybody in particular, and sometimes it was exactly for the period of öö that they avoided eye contact. Since Goodwin and Goodwin (1986Goodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google ScholarGoodwin, Marjorie H., and Charles Goodwin 1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 621: 51–75.Google Scholar logo with link to Google Scholar, 57) observed that speakers tend to withdraw gaze when beginning the process of searching, this gaze behavior supports the analysis of the phenomena described above as including a searching process, and the analysis of öö in these phenomena as an element taking part in the searching processes.

The present paper has aimed to contribute to our knowledge about the non-lexical schwa-like sound öö used in Hungarian spontaneous speech. Beyond its discourse-organizing function (Németh 2022 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 194: 23–35. Google Scholar logo with link to Google Scholar 2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 1941: 23–35. Google Scholar logo with link to Google Scholar), it seems to have a role in self-initiated same-turn self-repair as well. However, in order to see whether the functions explored in the present study can be generalized to Hungarian language use, a qualitative and quantitative examination of a representative sample of the Hungarian population from multiple registers would be necessary.

Funding

This paper was supported by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences. This paper was supported by the Subprogramme for Linguistic Identification of Fake News and Pseudo-scientific Views, part of the Science for the Hungarian Language National Programme of the Hungarian Academy of Sciences (MTA).

Acknowledgements

My special thanks are due to the anonymous reviewers for their insightful comments, Anna Fenyvesi for improving my English, and Enikő Németh T. for her constant encouragement. I also express my gratitude to Ágnes Lerch who was my first mentor and with whom I have made the recordings of the SZTEPSZI corpus. I would like to thank the Institute of Psychology, University of Szeged, for making it possible to record nine conversations, and Mária Gósy for putting eight conversations from the BEA database at my disposal.

Notes

1.Since there is no letter for schwa [ə] in Hungarian, I will use the letter ö to indicate the sound in question for two reasons. On the one hand, this is the letter which indicates the closest sound to schwa in Hungarian (IPA [ø]) (cf. Horváth 2014 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar 2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar, 25). On the other hand, the literature generally uses ö to indicate the phenomenon I examine. The double letter öö refers to the default use of the sound in the Hungarian dataset, which is a lengthened schwa.
2. Béla is a Hungarian male name.
3.The transcriptions follow the conventions of conversation analysis (Jefferson 2004 2004 “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 13–31. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar 2004 “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 13–31. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar). The glosses include only the grammatical information that is marked by a non-empty morpheme in Hungarian. Where possible, I give the corresponding English word in an inflected form. Furthermore, the words in the extracts are not segmented into components by hyphens for two reasons.
4.In order for the clause to be grammatical, this word would need a -t at the end to mark the accusative case. However, the native Hungarian speaker did not produce it. Taking into consideration that constructing the turn is an online activity, and that az nem volt könnyű ‘that wasn’t easy’ would be grammatical without megtanulni ‘to learn’, it is possible that the speaker was first only to produce the structure az nem volt könnyű ‘that wasn’t easy’, then altered the turn-constructional unit by adding megtanulni ‘to learn’ without the correction of the first word (the understanding of the action is not endangered).
5.This placeholder is usually pronounced with an -o, but Béla is producing here an -ö.
6.The extracts are indicated as SZTEPSZI1, SZTEPSZI2, etc. and bea001, bea002, etc., respectively.
7.For this reason, Németh (2021) 2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar 2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar regards these conversations as semi-spontaneous.
8.For the sake of simplicity, in the transcription of gaze behavior we can always see only two participants’ gaze directions out of the three. These two participants are the speaker and the recipient with whom the speaker is in interaction regarding her/his gaze direction in the extract. The lines are divided into units. A new unit begins if at least one of the participants’ gaze direction changes.
9.The tip of the tongue (TOT) phenomenon is a type of word search process when speakers cannot verbalize a word that is otherwise familiar to them (Brown and McNeill 1966Brown, Roger, and David McNeill 1966 “The “Tip of the Tongue” Phenomenon.” Journal of Verbal Learning and Verbal Behavior 5: 325–337. Google Scholar logo with link to Google ScholarBrown, Roger, and David McNeill 1966 “The “Tip of the Tongue” Phenomenon.” Journal of Verbal Learning and Verbal Behavior 51: 325–337. Google Scholar logo with link to Google Scholar; Iványi 2003Iványi, Zsuzsanna 2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 127: 76–91.Google Scholar logo with link to Google ScholarIványi, Zsuzsanna 2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 1271: 76–91.Google Scholar logo with link to Google Scholar).
10. Tök, shortened from tökéletesen ‘perfectly’ and homonymous with tök ‘pumpkin’, in evaluative APs serves as an upgrading modifier in colloquial Hungarian.
11.Since in the extract the speaker does not look at either of the other participants, the transcription shows only the speaker’s gaze behavior.
12.The cut-off is a component of the minimal form of the Error Correction Format (Jefferson 1974Jefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google ScholarJefferson, Gail 1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google Scholar, 187).

References

Amiridze, Nino, Boyd H. Davis, and Margaret Maclagan
(eds) 2010 Fillers, Pauses and Placeholders. Amsterdam/Philadelphia: John Benjamins Publishing Company. Google Scholar logo with link to Google Scholar
①
Biber, Douglas
1993 “Representativeness in Corpus Design.” Literary and Linguistic Computing 8: 243–257. Google Scholar logo with link to Google Scholar
①
Brown, Roger, and David McNeill
1966 “The “Tip of the Tongue” Phenomenon.” Journal of Verbal Learning and Verbal Behavior 5: 325–337. Google Scholar logo with link to Google Scholar
①
Clark, Herbert H., and Jean E. Fox Tree
2002 “Using Uh and Um in Spontaneous Speaking.” Cognition 84: 73–111. Google Scholar logo with link to Google Scholar
① ②
Deme, Andrea, and Alexandra Markó
2013 “Lengthenings and Filled Pauses in Hungarian Adults’ and Children’s Speech.” In Proceedings of DISS 2013. The 6th Workshop of Disfluency in Spontaneous Speech, ed. by Robert Eklund, 21–24. Stockholm: KTH Royal Institute of Technology.
①
Domonkosi, Ágnes
2002 Megszólítások és beszédpartnerre utaló elemek nyelvhasználatunkban [Forms of address and elements referring to the partner in conversation]. Debrecen: A Debreceni Egyetem Magyar Nyelvtudományi Intézetének Kiadványai [Proceedings of the Department of Hungarian Linguistics of the University of Debrecen] 79.Google Scholar logo with link to Google Scholar
①
Fox, Barbara, Fay Wouk, Makoto Hayashi, Steven Fincke, Liang Tao, Marja-Leena Sorjonen, Minna Laakso, and Wilfrido Flores Hernandez
2009 “A Cross-linguistic Investigation of the Site of Initiation in Same-turn Self-repair.” In Conversation Analysis: Comparative Perspectives, ed. by Jack Sidnell, 60–103. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar
① ② ③
Fox, Barbara, Yael Maschler, and Susanne Uhmann
2010 “A Cross-linguistic Study of Self-repair: Evidence from English, German, and Hebrew.” Journal of Pragmatics 42: 2487–2505. Google Scholar logo with link to Google Scholar
① ② ③
Goodwin, Charles
1980 “Restarts, Pauses, and the Achievement of a State of Mutual Gaze at Turn-beginning.” Sociological Inquiry 50(3–4): 272–302. Google Scholar logo with link to Google Scholar
①
1987 “Forgetfulness as an Interactive Resource.” Social Psychology Quarterly 50(2): 115–130. Google Scholar logo with link to Google Scholar
①
Goodwin, Marjorie H., and Charles Goodwin
1986 “Gesture and Coparticipation in the Activity of Searching for a Word.” Semiotica 62: 51–75.Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤ ⑥ ⑦ ⑧ ⑨ ⑩ ⑪ ⑫
Gósy, Mária
2006 “A semleges magánhangzó nyelvi funkciói [The linguistic functions of the neutral vowel].” In Beszédkutatás 2006 [Speech research 2006], ed. by Mária Gósy, 8–22. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google Scholar
①
2012 “Multifunkcionális beszélt nyelvi adatbázis – BEA [A multifunctional spontaneous speech database: BEA].” In Általános nyelvészeti tanulmányok XXIV. Nyelvtechnológiai kutatások [Studies in general linguistics 24: Language technology research], ed. by Gábor Prószéky, and Tamás Váradi, 329–349. Budapest: Akadémiai Kiadó.Google Scholar logo with link to Google Scholar
① ② ③
Gósy, Mária, Judit Bóna, András Beke, and Viktória Horváth
2013 “A kitöltött szünetek fonetikai sajátosságai az életkor függvényében [The phonetic characteristics of filled pauses with regard to age].” In Beszédkutatás 2013 [Speech research 2013], ed. by Mária Gósy, 121–143. Budapest: MTA Nyelvtudományi Intézet.Google Scholar logo with link to Google Scholar
① ②
Gósy, Mária, Dorottya Gyarmathy, and András Beke
2017 “Phonetic Analysis of Filled Pauses Based on a Hungarian-English Learner Corpus.” International Journal of Learner Corpus Research 3: 149–174. Google Scholar logo with link to Google Scholar
①
Helasvuo, Marja-Liisa, Minna Laakso, and Marja-Leena Sorjonen
2004 “Searching for Words: Syntactic and Sequential Construction of Word Search in Conversations of Finnish Speakers with Aphasia.” Research on Language and Social Interaction 37: 1–37. Google Scholar logo with link to Google Scholar
①
Heritage, John
1984 “Conversation Analysis.” In Garfinkel and Ethnomethodology, ed. by John Heritage, 233–292. Cambridge: Polity Press.Google Scholar logo with link to Google Scholar
①
Horváth, Viktória
2010 “Filled Pauses in Hungarian: Their Phonetic Form and Function.” Acta Linguistica Hungarica 57: 288–306. Google Scholar logo with link to Google Scholar
①
2014 Hezitációs jelenségek a magyar beszédben [Hesitation phenomena in Hungarian spontaneous speech]. Budapest: ELTE Eötvös Kiadó.Google Scholar logo with link to Google Scholar
① ② ③
Iványi, Zsuzsanna
2003 “A “nyelvemen van”-jelenség német és magyar nyelvű beszélgetésekben: Nyelvészeti vizsgálatok a konverzációelemzés módszereivel [The “tip of the tongue” phenomenon in German and Hungarian conversations: Linguistic examinations with the methods of conversation analysis].” Magyar Nyelvőr 127: 76–91.Google Scholar logo with link to Google Scholar
① ②
Jefferson, Gail
1974 “Error Correction as an Interactional Resource.” Language in Society 3(2): 181–199. Google Scholar logo with link to Google Scholar
① ② ③ ④
1984 “Notes on Some Orderlinesses of Overlap Onset.” In Discourse Analysis and Natural Rhetoric, ed. by Valentina D’Urso, and Paolo Leonardi, 11–38. Padua: Cleup Editore.Google Scholar logo with link to Google Scholar
①
2004 “Glossary of Transcript Symbols with an Introduction.” In Conversation Analysis: Studies from the First Generation, ed. by Gene H. Lerner, 13–31. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar
①
Kenesei, István
1994 “Subordinate Clauses.” In The Syntactic Structure of Hungarian, ed. by Ferenc Kiefer, and Katalin É. Kiss, 275–354. San Diego: Academic Press. Google Scholar logo with link to Google Scholar
①
Kitzinger, Celia
2013 “Repair.” In The Handbook of Conversation Analysis, ed. by Jack Sidnell, and Tanya Stivers, 229–256. Oxford: Wiley-Blackwell.Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤ ⑥
Kurhila, Salla
2006 Second Language Interaction. Amsterdam: John Benjamins. Google Scholar logo with link to Google Scholar
①
Laakso, Minna
1997 Self-Initiated Repair by Fluent Aphasic Speakers in Conversation. Helsinki: Finnish Literature Society.Google Scholar logo with link to Google Scholar
①
2003 “Collaborative Construction of Repair in Aphasic Conversation: An Interactive View on the Extended Speaking Turns of Persons with Wernicke’s Aphasia.” In Conversation and Brain Damage, ed. by Charles Goodwin, 163–188. Oxford: Oxford University Press. Google Scholar logo with link to Google Scholar
①
Laakso, Minna, and Marja-Leena Sorjonen
2010 “Cut-off or Particle – Devices for Initiating Self-Repair in Conversation.” Journal of Pragmatics 42: 1151–1172. Google Scholar logo with link to Google Scholar
①
Lerch, Ágnes
2007 “Az ismétlés mint az önjavítás eszköze a magyarban [Repetition as a means of self-repair in Hungarian].” In Nyelvelmélet – nyelvhasználat [Language theory and language use], ed. by Tamás Gecső, and Csilla Sárdi, 123–130. Budapest: Tinta Könyvkiadó.Google Scholar logo with link to Google Scholar
① ②
Lerner, Gene H.
1996 “On the ‘Semi-permeable’ Character of Grammatical Units in Conversation: Conditional Entry into the Turn-space of Another Speaker.” In Interaction and Grammar, ed. by Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, 238–276. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar
① ② ③
2013 “On the Place of Hesitating in Delicate Formulations: A Turn-constructional Infrastructure for Collaborative Indiscretion.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 95–134. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤
Németh, Zsuzsanna
2012 “Recycling and Replacement Repairs as Self-initiated Same-turn Self-repair Strategies in Hungarian.” Journal of Pragmatics 44: 2022–2034. Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤
2017 “The Interactional Functions of Four Repair Operations in Hungarian.” In Pragmatics at Its Interfaces – Mouton Series in Pragmatics 17, ed. by Stavros Assimakopoulos, 279–310. Berlin: Mouton de Gruyter. Google Scholar logo with link to Google Scholar
① ② ③ ④
2020 “A nemlexikális öö hang interakciós szerepének elemzése magyar nyelvű társalgásokban [An analysis of the interactional role of the non-lexical Sound öö in Hungarian conversations].” Jelentés és Nyelvhasználat 7: 23–50. Google Scholar logo with link to Google Scholar
① ②
2021 A Preference Hierarchy Model of Same-Turn Repair Operations in Talk-in-Interaction. Sheffield, UK: Equinox Publishing.Google Scholar logo with link to Google Scholar
① ②
2022 “The Conversation-Organising Role of the Non-lexical Sound öö in Hungarian.” Journal of Pragmatics 194: 23–35. Google Scholar logo with link to Google Scholar
① ② ③ ④
Oravecz, Cs., Tamás Váradi, and Bálint Sass
2014 “The Hungarian Gigaword Corpus.” In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014), ed. by Nicoletta Calzolari, Khalid Choukri, Thierry Declerck, Hrafn Loftsson, Bente Maegaard, Joseph Mariani, Asuncion Moreno, Jan Odijk, and Slelios Piperidis, 1719–1723. Reykjavik: European Languages Resources Association (ELRA).Google Scholar logo with link to Google Scholar
①
Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson
1974 “A Simplest Systematics for the Organization of Turn-Taking for Conversation.” Language 50: 696–735. Google Scholar logo with link to Google Scholar
①
Schegloff, Emanuel A.
1979 “The Relevance of Repair to Syntax-for-Conversation.” In Syntax and Semantics XII, ed. by Talmy Givón, 261–286. New York: Academic Press. Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤ ⑥ ⑦ ⑧
2007 Sequence Organization in Interaction: A Primer in Conversation Analysis. Cambridge: Cambridge University Press. Google Scholar logo with link to Google Scholar
①
2013 “Ten Operations in Self-Initiated, Same-Turn Repair.” In Conversational Repair and Human Understanding, ed. by Makoto Hayashi, Geoffrey Raymond, and Jack Sidnell, 41–70. Cambridge: Cambridge University Press.Google Scholar logo with link to Google Scholar
① ② ③ ④ ⑤ ⑥ ⑦ ⑧ ⑨ ⑩
Schegloff, Emanuel A., Gail Jefferson, and Harvey Sacks
1977 “The Preference for Self-Correction in the Organization of Repair in Conversation.” Language 53: 361–382. Google Scholar logo with link to Google Scholar
① ② ③ ④
Tar, Cintia
2022 “Az önjavítási jelenségek magyar nyelvű spontán, baráti társalgásokban [Self-repair phenomena in Hungarian spontaneous conversations among friends].” Beszédtudomány – Speech Science 3: 75–104.Google Scholar logo with link to Google Scholar
① ②
Vukov Raffai, Éva
2016 “A diskurzusjelölő-választások életkori sajátosságai az így, ilyen, hát, mondjuk, ugye esetében [Age characteristics in discourse marker selections in the case of így, ilyen, hát, mondjuk, ugye ].” Magyar Nyelvőr 140: 483–497.Google Scholar logo with link to Google Scholar
① ②

Address for correspondence

Zsuzsanna Németh

Department of General Linguistics

University of Szeged

Egyetem u. 2

Szeged H-6722

Hungary

[email protected]

Biographical notes

Zsuzsanna Németh is interested in self-repair and implicit pragmatic phenomena in naturally occurring conversations and Hungarian fake news headlines. She is an assistant professor at the University of Szeged, Department of General Linguistics, and a member of the MTA-SZTE-DE Research Group for Theoretical Linguistics and Informatics. In her research she combines conversation analysis and pragmatics.

ORCiD logo with linkhttps://orcid.org/0000-0002-8509-7862
 
Mobile Menu Logo with link to supplementary files background Layer 1 prag Twitter_Logo_Blue