When Phonation Matters: The Use and Function of yeah and Creaky Voice Colorado Research in Linguistics. June 2004. Volume 17, Issue 1. Boulder: University of Colorado. © 2004 by Tamara Grivičić and Chad Nilep. When Phonation Matters: The Use and Function of yeah and Creaky Voice Tamara Grivičić and Chad Nilep University of Colorado at Boulder This paper illuminates the conversational functions of the combination of creaky voice quality and the response token yeah. Jefferson (1984) described yeah as an acknowledgement token that also projects “a preparedness to shift from recipiency to speakership” (p. 200). This speaker incipiency is not consistent, though. While yeah is sometimes used to indicate a shift from recipient to speaker, it is sometimes used simply as an acknowledgement token. This difference in function of apparently similar items may be related to token shape. This paper examines several telephone interactions and finds the use of yeah with creaky voice to indicate passive recipiency and either a dispreference to continue the current topic, or a disalignment with the primary speaker. This analysis contributes to the study of phonetics in interactional linguistics. In addition, it supports the notion that token-shape distinctions can account for functional differences within token types. It suggests that phonation or other behavior below the word level may be significant in verbal interaction. 1. Introduction Interactional Linguistics is the emerging endeavor to study traditional linguistic interests, including syntax and phonology, using the tools of Conversation Analysis (Selting and Couper-Kuhlen 2001). At the level of sound patterns, much work has clustered around analyses of prosody and intonation (e.g. Couper-Kuhlen and Selting 1996), and their functions in conversation. To date, there has been less attention paid to narrower, phonetic analysis of speech sounds1. This paper presents an analysis of the conversational functions of the combination of creaky voice quality and the response token yeah. While response tokens are extensively studied within Conversation Analysis, voice quality has not been widely considered from an interactional point of view. Nonetheless, this paper will suggest that qualities such as creaky voice are available to speakers as resources, and that voice quality may interact with the word yeah to perform a set of conversational functions. Within the field of phonetics, creaky voice, or laryngealization, has been described as one of a number of phonation types, or voice qualities. Ladefoged (1982) catalogs a variety of phonation types, including modal voicing – the normal vibration of the vocal folds which, according to Ladefoged, occurs in all spoken languages – as well as aspiration, murmur, "glottal catch," pharyngealization and laryngealization2. However, in his brief discussion, Ladefoged mentions only cases in which these glottal distinctions are phonemic; he makes no mention of the occurrence of such phonation types in languages whose speakers do not systematically utilize or orient to them as distinctive. However, implicit in Ladefoged's opening remarks is the suggestion that such varieties might occur 1. However, see, for example, Local and Kelly 1986, Fox Tree and Clark 1997, Bybee and Schiebman 1999, and Jurafsky et al. 2001. 2. This compares to five features for glottal stricture described in Ladefoged (2001): [voiceless]; [breathy voice]; [modal voice]; [creaky voice]; and [closed], the setting for glottal stops. 1 Grivic?ic? and Nilep: When Phonation Matters: The Use and Function of yeah and Creaky V Published by CU Scholar, 2004 Colorado Research in Linguistics, Volume 17 (2004) 2 in any normal speaker's repertoire. Indeed, Ladefoged (2001) notes that creaky voice "occurs at the ends of falling intonations for some speakers of English," (125) even though English has no laryngealized phonemes. There is, though, no discussion of the functions of creaky voice in languages where it is not distinctive. Other scholars have suggested that creaky voice can have communicative function in English. Pittam (1987) suggests that, for Australian speakers, creaky voice indexes low solidarity and is associated with male speakers. Blount and Padgug (1976) describe creaky voice as characteristic of English care-giver speech. Duncan and Fiske (1977) suggest that, when coupled with low pitch, creaky voice can signal the end of a conversational turn. From an interactional point of view, two works bear particular mention. Ogden (2001) suggests that among Finnish speakers, creaky voice often co-occurs with syntactic completion, pragmatic completion, and sentence final intonation at the end of a turn- constructional unit (TCU; Sacks, Schegloff and Jefferson 1974). Such a combination of potential turn-end markers indicates a Complex Transition Relevance Place (CTRP; Kärkkäinen, Sorjonen and Helasvuo, to appear), where a current speaker typically gives way to a new speaker. This use of creaky voice contrasts with glottal stops, which are generally not treated as transition relevant, even when followed by a long pause. Furthermore, when creak co-occurs with one or more of these elements, but speaker transition is not affected, TRP is retracted by, for example, rushing through the next TCU. To date there are no widely reported findings for English orientation to creaky voice which would compare to Ogden's findings in Finnish. However, Jasperson's (1998) work on repair in English suggests that glottal stop may function in a comparable way in each language. In several types of focus-repairs described by Jasperson, a speaker may produce a significant pause after a glottalized cut-off. According to Jasperson (personal communication), Closure cut-off (which can, under the right conditions, be realized by glottal stop) is routinely used to initiate same-turn repair of the TCU-so- far, and to that extent projects more talk to come (the repair), the continuation of the TCU. Silences that may follow closure cut-off, before the resumption of phonation, get interpreted as belonging to the speaker, because she has not brought the turn to possible completion. Thus, in English as in Finnish, glottal closure is not treated as transition relevant. It remains to be seen whether English speakers treat other glottal strictures, such as creaky voice, as marking transition places. Unlike creaky voice, the lexical item yeah has inspired a great deal of writing by linguists. In fact, the shear volume of information precludes a thorough review here. However, two studies that bear on the issues discussed here should be mentioned. Jefferson (1984) offers a preliminary analysis of the interactional work which speakers can accomplish through the deployment of acknowledgement tokens mm hm, uh huh, and yeah. According to Jefferson, mm hm and uh huh mark "passive recipiency" (202). Yeah, on the other hand, is said to mark "imminent speakership" (202); that is, a recipient who produces yeah as an acknowledgement token also projects an assumption of primary speakership. However, Jefferson points out that not all tokens of yeah prefigure a change in speakership. What accounts for this variability? The answer is not 2 Colorado Research in Linguistics, Vol. 17 [2004] https://scholar.colorado.edu/cril/vol17/iss1/10 DOI: https://doi.org/10.25810/36ed-z920 When Phonation Matters: The Use and Function of yeah and Creaky Voice 3 entirely clear, but Jefferson suggests that "token-shape" may account for differences in function of apparently similar tokens. This suggestion leaves open the possibility that phonation or other behavior below the word level may be important to the form and function of acknowledgement tokens. The work of Drummond and Hopper (1993a) is in some ways a continuation and expansion of work begun by Jefferson (1984, 1993), particularly Jefferson's suggestion that there is a continuum from the passive recipiency of mm hm, to go-ahead markers such as oh really? to the speaker incipiency marked by yeah. Drummond and Hopper find oh and okay frequently at the end of tellings, often projecting a change in speaker and/or topic3. The tokens mm hm, uh huh, and yeah all occur earlier in the telling, and prefigure a continuation of the telling. When all three tokens occur during the course of an extended telling, mm hm tends to be realized earliest in the sequence, and yeah latest. Further, yeah may signal a shift in speakership, with the participant who utters the token taking over as primary speaker. As Jefferson (1984) found, though, yeah can also prefigure a continuation of the telling. 2. Data and analyses The data for this study consist of approximately twenty-five minutes of telephone conversations. This may be considered ‘found’ data; it was not recorded by the investigators for the purpose of analysis. Instead, phone calls were recorded by men who were ‘teasing’ telemarketers, attempting to keep them on the line for as long as possible with no intent of buying the service advertised. The peculiar nature of these conversations may make it impossible to generalize about much of the behavior recorded. However, since speakers seem not to have any metalinguistic knowledge of their ability to manipulate phonation type (despite the facility of manipulation found by Jasperson 1998 and Ogden 2001), we assume that the particular phonetic behavior described here is not affected by the nature of the conversation4. The investigators worked with audiotapes of the conversations; the tapes were transcribed and coded for the occurrence of various discourse markers. ‘Discourse marker’ was defined to include items such as oh, ok, really, mm, mhm, uh huh, and yeah5. Within the transcripts, yeah was the most frequent lexical discourse marker, accounting for 87 of the 260 markers coded. Also coded was the occurrence of creaky voice, determined impressionistically (see Local 1996). A word was coded for creaky voice when creak was hearable over at least one syllable of the word. Distributional analyses (see below) showed that yeah with modal voicing tended to be followed by additional speech much more often than yeah delivered with creaky voice. 3. This is comparable to the go-ahead responses that Schegloff (1995) describes in pre-expansion sequences and minimal post expansions. 4. Reviewers have also pointed out potential ethical dilemmas related to the use of recorded telephone conversations. This is certainly an issue that researchers should be sensitive to. However, all names and individual identifiers have been suppressed from the data. Furthermore, since both the company employing the telemarketers and the customers themselves reserved the right to record the interactions, we have decided to use the data. Both federal and state law allow for such recording when, as in this case, at least one party grants consent. 5. For a fuller description of discourse markers, see Jucker and Ziv (1998). 3 Grivic?ic? and Nilep: When Phonation Matters: The Use and Function of yeah and Creaky V Published by CU Scholar, 2004 Colorado Research in Linguistics, Volume 17 (2004) 4 Conversational Analysis was then carried out on portions of the data that featured the word yeah, both with and without creaky voice. (1) Example 1. Creaky and non-creaky yeah. Track_02:45-50 44 M: mean if you only make five calls uh you know five times thirty five that's like ((water stops))(0.80) you kno::w (0.6) m'sorry (0.72) little slow on m'math . it's like= 45 R: =oh, it's like a dollar fifty somethin'\ [(>you know uh<) 46 M: [yeah, it's like a dollar fifty five, 47 R: %ye:[ah,% 48 M: [th]at's for five calls at dollar fifty fi[ve, ] 49 R: [%;ye::ah;%] 50 M: that# that's a lot less then paying like four ninety five a month for you just for making like . you know five ca:lls you know, In example 1, neither instance of creaky yeah is accompanied by other speech. Both function as minimal responses and indicate passive recipiency (Jefferson 1984). However, we would expect the function of the non-creaky yeah to be different. Although it still functions as a minimal response acknowledging the past turn, it signals high speaker incipiency. That is, it secures that the speaker who has just uttered yeah will remain or will become the primary speaker. Indeed, this is the case following the non- creaky yeah in example 1. In the interaction so far, M has been the primary speaker, describing a telephone rate plan to R. At line 45, R co-constructs M’s turn by supplying the answer to M’s calculation. So, M’s yeah at line 46 demonstrates his intent to remain the primary speaker. On the other hand, creaky yeah in line 47 has the complimentary function indicating passive recipiency. R does not express interest in taking the floor, but rather he merely gives an indication that he accepts or is following M’s ongoing explanation. Similarly, after the creaky yeah at line 49, M continues his sales pitch with no interruption from R. (2) Example 2. No uptake. Track_02:96-101 96 M: =yeah\ so right now you are probably payin' about three ninety five if ya have qwest for your long distance\ 97 (0.62) 98 M: of course i'm# i'm not ((dishes start up again)) exactly certain cause . i mean unless you know which program you have 99 (0.85) 100 R: ri:ght, (0.27) *%yeah right%*= 101 M: =but like i said right now i can getchyou seven cents no monthly fees or minimums\ Example 2 illustrates another case of creaky yeah used to indicate passive recipiency. Of particular interest in this example is the sequence organization between the two speakers. Although M is providing a space for R to reply -- this space is evident in the silences between M’s turns -- there is no uptake by R. When R finally produces a reply 4 Colorado Research in Linguistics, Vol. 17 [2004] https://scholar.colorado.edu/cril/vol17/iss1/10 DOI: https://doi.org/10.25810/36ed-z920 When Phonation Matters: The Use and Function of yeah and Creaky Voice 5 and M does not resume speaking, R produces creaky yeah signaling his intent to remain a passive recipient. From these examples it appears that creaky yeah functions to mark passive recipiency, not projecting further speech from the speaker who produces it. Furthermore, creaky yeah may be seen as an attempt to close out a sequence or discontinue the current topic. As seen in Example 1, however, this attempt by the primary recipient to close a topic may or may not be respected by the primary speaker. That is, the attempt to close a sequence is not always successful, since it depends on concurrence of the interlocutors. It is often the case, as in Example 2, that creaky yeah displays dis-alignment or dispreference for continuing the sequence. Thus, creaky yeah can be seen to function both as a marker of passive recipiency and as a tool to accomplish sequence closings. The following example shows a possibly deviant case. In Example 3, which precedes and includes a portion of Example 1, M produces creaky yeah and follows it with a substantial turn. (3) Example 3. Deviant case analysis. Track_02:31-44 31 R: [and then . on top ] of the seven cents there [and so was like] forty [two cents] 32 M: [yeah, yeah ] [there . is a . ] there is a thirty-five cents surcharge yeah, 33 R: we::ll there you go, that's what I'm [gettin at, 34 M: [ye:ah,] 35 M: but that's only when you use it i mean you say you don't make many calls i mean you make an an average amount of calls though right? 36 R: i don't know wha[t an average amount of calls is, 37 [((dishes banging continuously))] 38 (0.38) 39 M: you said about five to ten right? 40 R: ((long inhalation)) sss i don't kno:w,= 41 M: =that's about average for most people,= 42 R: =is it? 43 (0.49) 44 M: %ye:ah,% i mean it's# it's not a lot ((water running))of calls really? (0.49) mean if you only make five calls uh you know five times thirty five that's like ((water stops))(0.80) you kno::w (0.6) m'sorry (0.72) little slow on m'math . it's like= In this example, M is the primary speaker. Again, he is making a sales pitch, to which R displays recipiency. Unlike previous examples, where the primary recipient utters creaky yeah, here M uses the token, at line 44. Lines 42-44 constitute an insert sequence within the on-going interaction. At line 42, R produces a checking question, which M answers at line 44 with creaky yeah before resuming the larger activity in which he has been involved. Thus, the creaky yeah at 44, followed by sentence final intonation, accomplishes much the same function as that displayed in Examples 1 and 2. In terms of speaker incipiency, creaky yeah appears to be similar to mm hm or uh huh, as described by Jefferson (1984, 1993) and Drummond and Hopper (1993a, 1993b). That is, a participant who utters either mm hm, uh huh, or creaky yeah, continues to be a 5 Grivic?ic? and Nilep: When Phonation Matters: The Use and Function of yeah and Creaky V Published by CU Scholar, 2004 Colorado Research in Linguistics, Volume 17 (2004) 6 recipient, and allows her conversational partner to continue as the primary speaker. There appears to be a slight difference between these tokens in terms of alignment, however. Example 4 shows that the use of uh huh indicates alignment with the primary speaker. (4) Example 4. Alignment with uh huh Track_04:127-135 127 R: so thirty five cents no matter what, 128 M: right thirty five cents no matter what\ [and then ( )] 129 R: [and then seven cents] on top of that, 130 M: right [but you know ] 131 R: [so if i ] if i just call one minute\ 132 M: uh-hmm? 133 R: it's seven= 134 M: =forty two,= 135 R: =seven cents plus thirty five, Here, R is checking his understanding of the sales-pitch-so-far. At line 128, and again and at 131, M has produced speech in overlap with R. In both instances, M drops out, and allows R to speak. M’s uh hmm? at line 132 indicates passive recipiency, allowing R to continue as primary speaker. Further, uh hmm indicates alignment with the turn that R is in the midst of producing. This alignment is further evidenced by the co-construction of the number and its significance in lines 133-135. Contrast the alignment shown by uh hmm with the dis-alignment and topic transition that creaky yeah marks in Example 5.6 (5) Example 5. Disalignment with creaky yeah. Track_03:358-368 358 M: [yaa::h, . heh, ] =yaa::h\ that's a good college, 359 R: you know i TELL you ma:n\ every weekend\ those# those da:mn kids are up there BURNin' the DA:MN hill down\ 360 R: you [know, kickin' in the windo:ws] 361 M: [he he he ] 362 R: and drinking bee:r and throwing up and all over the damn street\ and [burnin'] sofas/ 363 M: [hhhhh ] 364 (0.598) 365 R: those {expletive deleted} crazy over there those BA:stards, 366 M: %yaa::h%, 367 (0.694) 368 M: so like what you think about that other pla:n, (unintelligible) the surcha:rge with thirty five cent connec[tion fee:? ] In response to M’s observation in line 358, ‘Yeah, that’s a good college,’ R produces a telling that may be seen as disagreeing. R describes some negative aspects of college 6. A potentially offensive expression has been removed from Example 5. 6 Colorado Research in Linguistics, Vol. 17 [2004] https://scholar.colorado.edu/cril/vol17/iss1/10 DOI: https://doi.org/10.25810/36ed-z920 When Phonation Matters: The Use and Function of yeah and Creaky Voice 7 life, and ends with an unflattering characterization of the students, which includes an offensive racial epithet. This remark is followed by M’s creaky yeah at line 366. The use of creaky yeah signals M’s desire for R to end the current topic, but allows for R to remain the primary speaker. Following the creaky yeah there is a significant silence, during which R fails to resume speakership. After a pause, M self-selects and begins a new topic. 3. Results The data from this pilot study revealed a high proportion of creaky voice tokens occurring on the word yeah (n=5 of 8 creaky voice tokens). As mentioned on the previous section, we conducted a distributional analysis on the audio data comparing occurrences of creaky and non-creaky yeah. Tables 1 and 2 show the results of this analysis. The above tables show that instances of non-creaky yeah are likely to be followed by additional speech. This finding is analogous to Drummond and Hopper’s (1993b) observation that 46% of yeah tokens are followed by speech. The instances of creaky yeah, on the contrary, tend not to be followed by speech. In fact, they appear to show preference to occur alone. speech final non-final total count 37 56 93 % 40% 60% Table 1. Number of speech-final occurrences of yeah versus yeah followed by speech. speech final non-final total count 4 1 5 % 80% 20% Table 2. Number of speech-final occurrences of creaky yeah versus creaky yeah followed by speech. 7 Grivic?ic? and Nilep: When Phonation Matters: The Use and Function of yeah and Creaky V Published by CU Scholar, 2004 Colorado Research in Linguistics, Volume 17 (2004) 8 Although our research is preliminary and based on a very small set of data, it nevertheless shows a clear distinction between creaky and non-creaky yeah and functional interaction between voice quality and lexeme. This distinction is not unlike the one presented in Jefferson’s (1984) analysis, which showed a distinction between yeah and Mm hm. Jefferson suggested, “This systematic distinction, raised in a single-instance analysis which generated a collection, can now serve as a resource to be turned to further single-instance analysis, where some otherwise obscure interaction-bits can be brought to focus” (206). Our analysis shows that voice quality may interact with words to perform a set of conversational functions. Discourse uses of the combination of creaky voice quality and response token implicate the following semantic functions: passive recipiency, a dispreference to continue the current topic, or a disalignment with the primary speaker. These findings support the notion that token-shape distinctions can account for functional differences within token types and that qualities such as creaky voice are available to speakers as resources. Our research contributes to the study of phonetics in interactional linguistics as it pays attention to the narrower, phonetic analysis of speech sounds. In addition, it suggests that phonation or other behavior below the word level may be significant in verbal interaction. 4. Conclusion This study set out to illuminate the functions of a response token, yeah, and a phonation type, creaky voice. We have demonstrated that creaky yeah is not identical in function to yeah with modal voicing or other types of glottal stricture. While Jefferson (1984) suggests that yeah generally signals high speaker incipiency, yeah in conjunction with creaky voice signals passive recipiency. This observation may support and explain Jefferson’s suggestion that token-shape distinctions can account for functional differences within token-types. Considerable work exists to describe the functions of yeah. A recurrent suggestion (Jefferson 1984, 1993; Drummond and Hopper 1993b, 1993c; Gardner 1998, 2001) is that yeah marks speaker incipiency. To date, however, analysts have described this incipiency as variable. While yeah can both respond to a previous utterance and project continued speech, it is not always followed by further talk. The present analysis suggests one possible reason for this: creaky yeah features speaker incipiency so low that it has the complementary function of indicating recipiency. Very little interactional research has been done on creaky voice. Ogden’s (2001) research on glottal phonation and turn transition in Finnish is one example of such work. Ogden’s suggestion that creaky voice occurs at the ends of turns, and signals transition relevance, is compatible with our findings. In Finnish, creaky voice signals the end of a speaker’s turn. This invites an interlocutor to take the floor, and projects recipiency on the part of the speaker who produces creaky phonation. We have suggested that, in English, creaky yeah similarly signals recipiency and requests a change in topic. Our further suggestion that creaky yeah indicates dis-alignment or dispreference may be limited to English or to this token alone. It remains to be seen what, if any other functions creaky voice has in English talk in interaction. 8 Colorado Research in Linguistics, Vol. 17 [2004] https://scholar.colorado.edu/cril/vol17/iss1/10 DOI: https://doi.org/10.25810/36ed-z920 When Phonation Matters: The Use and Function of yeah and Creaky Voice 9 Appendix A Transcription Conventions ? terminal rise , terminal fall / non-terminal rise \ non-terminal fall . short pause (< 0.2) (0.6) pause in seconds # cut off or interruption (( )) transcriber’s notes ( ) transcriber’s best guess (unintel) unintelligible speech * * low volume ; ; low pitch % % creaky voice > < relatively fast speech BAstards relatively high volume pho:ne long segment ! alveolar click hh exhaled breath NB: These transcription conventions were designed for easy reproduction in ASCII character sets. Thus, they can be used with most transcription software and most email or other file-sharing software. Note also that, unlike earlier systems, each mark of punctuation has only one function. 9 Grivic?ic? and Nilep: When Phonation Matters: The Use and Function of yeah and Creaky V Published by CU Scholar, 2004 Colorado Research in Linguistics, Volume 17 (2004) 10 References BLOUNT, BEN AND ELISE PADGUG. 1976. ‘Mother and Father Speech: Distribution of Parental Speech Features in English and Spanish.’ Papers and Reports on Child Language Development 12: 47-59. BYBEE, JOAN AND JOANNE SCHEIBMAN. 1999. ‘The effect of usage on degrees of constituency: the reduction of don't in English.’ Linguistics 37 (3): 575-596. COUPER-KUHLEN, ELIZABETH AND MARGRET SELTING. 1996. Prosody and conversation. Cambridge: Cambridge University Press. DRUMMOND, KENT AND ROBERT HOPPER. 1993a. ‘Acknowledgment Tokens in Series.’ Communication Reports 6 (1): 47-53. DRUMMOND, KENT AND ROBERT HOPPER. 1993b. ‘Back Channels Revisited: Acknowledgment Tokens and Speakership Incipiency.’ Research on Language and Social Interaction 26 (2): 157-177. DRUMMOND, KENT AND ROBERT HOPPER. 1993c. ‘Some Uses of Yeah.’ Research on Language and Social Interaction 26 (2): 203-212. DUNCAN, STARKEY AND DONALD W. FISKE. 1977. Face to Face Interaction: Research Methods and Theory. New York: Wiley. FOX TREE, JEAN E. AND HERBERT H. CLARK. 1997. ‘Pronouncing “the” as “thee” to signal problems in speaking.’ Cognition 62: 141-167. GARDNER, ROD. 2001. When Listeners Talk: Response Tokens and Listener Stance. Amsterdam: John Benjamins. GARDNER, ROD. 1998. ‘Between Speaking and Listening: The Vocalizations of Understanding.’ Applied-Linguistics 19 (2): 204-224. JASPERSON, ROBERT. 1998. ‘Repair after Cut-off: Explorations in the Grammar of Focused Repair of the Turn-constructional Unit-so-far.’ Ph.D. Dissertation, University of Colorado at Boulder. JEFFERSON, GAIL. 1984. ‘Notes on a Systematic Deployment of the Acknowledgement Tokens “Yeah” and “Mm hm.”’ Papers in Linguistics 17 (1-4): 197-216. JEFFERSON, GAIL. 1993. ‘Caveat speaker: Preliminary Notes on Recipient Topic Shift Implicature.’ Research on Language and Social Interaction 26: 1-30. JUCKER, ANDREAS AND YAEL ZIV Discourse Markers: Descriptions and Theory, 171-201. Amsterdam: John Benjamins. JURAFSKY, DANIEL, ALAN BELL, MICHELLE GREGORY, AND WILLIAM D. RAYMOND. 2001. ‘Probabilistic relations between words: evidence from reduction in lexical production.’ In Joan L. Bybee and Paul J. Hopper, eds., Frequency and the emergence of linguistic structure. Amsterdam: Benjamins. KÄRKKÄINEN, ELISE, MARJA-LEENA SORJONEN AND MARJA-LIISA HELASVUO. To appear. ‘Discourse Structure.’ In T. Shopen, ed., Language Typology and Syntactic Description. LADEFOGED, PETER. 1982. ‘The Linguistic Use of Different Phonation Types.’ University of California Working Papers in Phonetics 54: 28-39. LADEFOGED, PETER. 2001. A Course In Phonetics. Fort Worth: Harcourt College Publishers. 10 Colorado Research in Linguistics, Vol. 17 [2004] https://scholar.colorado.edu/cril/vol17/iss1/10 DOI: https://doi.org/10.25810/36ed-z920 When Phonation Matters: The Use and Function of yeah and Creaky Voice 11 LOCAL, JOHN. 1996. ‘Conversational Phonetics: Some Aspects of News Receipts in Everyday Talk.’ In Elizabeth Couper-Kuhlen and Margret Selting, eds., Prosody in Conversation: Interactional Studies. Cambridge: Cambridge University Press. LOCAL, JOHN AND JOHN KELLY. 1986. ‘Projection and 'silences': notes on phonetic and conversational structure.’ Human Studies 9: 185-204. OGDEN, RICHARD. 2001. ‘Turn transition, creak and glottal stop in Finnish talk-in- interation.’ Journal of the International Phonetic Association 31(1): 139-152. PITTAM, JEFFERY. 1987. ‘Listeners' Evaluations of Voice Quality in Australian English Speakers.’ Language and Speech 30 (2): 99-113. SACKS, HARVEY, EMANUEL SCHEGLOFF AND GAIL JEFFERSON. 1974. ‘A Simplest Systematics For the Organization of Turn-taking in Conversation.’ Language 50: 696-735. SCHEGLOFF, E. A. 1995. ‘Sequence Organization.’ Manuscript. SELTING, MARGRET AND ELIZABETH COUPER-KUHLEN. 2001. Studies in Interactional Linguistics. Amsterdam: John Benjamins. 11 Grivic?ic? and Nilep: When Phonation Matters: The Use and Function of yeah and Creaky V Published by CU Scholar, 2004 Colorado Research in Linguistics 6-2004 When Phonation Matters: The Use and Function of yeah and Creaky Voice Tamara Grivičić Chad Nilep Recommended Citation Microsoft Word - paper_GRIVICIC_NILEP.doc