Your Turn at Talk

Skip to content Skip to main navigation
Teachers College, Columbia University
Printer-friendly Version
Teachers College, Columbia University Logo

The Language and Social Interaction Working Group (LANSI)


YTAT and Other CA materials

Writing CA

A handout and activity worksheets created for a CA writing workshop for novice CA researchers.

Working with Framegrabs

A handout providing a step-by-step guide to the production of framegrabs to be used in CA research, writing, and presentations. The technical specifics are a bit dated now but it still offers a useful introduction.

Lecture Handouts

A set of materials from the introduction to CA course I taught for Temple University Japan’s MATESOL program.

For more information, please contact  


A brief history of Your Turn at Talk (YTAT)

The beginnings of the YTAT booklet go back to around 1999 when I first started exploring the application of conversation analysis to language teaching. The booklet itself was occasioned by a deep dissatisfaction with all commercially available speaking or conversation textbooks. The problem with virtually all “conversation” textbooks is that they don’t actually teach conversation, which is to say that they don’t show students how to navigate talk-in-interaction in natural and socially appropriate ways. Many so-called speaking textbooks offer mainly thinly disguised grammar practice. Other textbooks equate conversation with discussion, failing to appreciate that discussion is a specialized genre of talk-in-interaction. My feeling was that while decades of conversation analytic research had revealed a new and exciting grammar-of-interaction, these valuable insights were being withheld from language learners...and for that matter from language teachers (see Carroll, 2010). So, week-by-week, I started putting together materials for a false-beginner conversation class I was teaching at a Japanese university. This began with experimenting with ways to teach the timing and turn-taking which form the foundation of almost everything else in conversation (see Carroll, 2011a). By the end of the semester, I had YTAT.


Organization of YTAT

The booklet opens naturally enough with a unit on openings. To the extent possible in printed material, these openings reflect CA research into English conversational openings, both face-to-face and on the telephone. The remaining units are each organized around a single adjacency pair and its possible expansions. In its present state YTAT looks at four sequence types in the following order: 1) invitations 2) requests 3) assessments and 4) news-tellings. Throughout YTAT, preferred responses are presented, and ideally taught, before and separately from dispreferred responses. Each unit begins with a small set of two-turn (invented, although I hope realistic) “mini-logs” that illustrate the base adjacency pair, as well as presenting some of the language typically employed. The reason for using mini-logs instead of the multi-line dialogs typical of many ESL textbooks is that any particular dialog is the outcome, the remaining residue so to speak, of a specific instance of social interaction, and thus vanishingly unlikely ever to reoccur in the same form. On a smaller scale, however, our daily talk is massively repetitious; we say what we have said before. The mini-logs are thus easily learned, borrowed, and adapted.

After the basic adjacency pair has been taught, including its preferred and dispreferred responses, the unit expands on them by looking at prototypical pre-sequences: pre-invitation sequences (go-ahead and block), pre-request sequences (go-ahead and block), pre-announcement sequences (go-ahead and block). YTAT does not work with insert expansions or post-expansions, though those are both of great interest and could be included in further development of YTAT. Throughout YTAT I have used what I call “eye-a-lect” to capture something of what conversational English actually sounds like. This is a compromise between presenting the sorts of nonorthodox (and often confusing) spellings found in CA transcripts and the unnaturally “expanded” speech common to ESL textbooks. 

YTAT begins with invitation sequences because they offer a prototypical illustration of an initiating action with two (and only two) possible responding actions, one preferred (accept) and one dispreferred (reject). Invitation sequences are also perfect for introducing learners to the basics of preference organization, e.g. preferred and dispreferred turn-shapes, including the importance of transition timing (see Carroll, 20011b).

Next, YTAT looks at request sequences. From the point of view of frequency and urgency for learners, it might seem better to begin with request sequences. However, there are reasons not to do so. For one, requests are marked in that they are a dispreferred first pair part, which is to say that people actively work to avoid doing requests, for example, by telling-trouble as a way to elicit a pre-emptive offer of assistance. In addition, when they are done, requests are frequently formatted using features common to the dispreferred turn-shape. (Unfortunately, when I wrote YTAT in 1999 I wasn’t aware of this fact so some aspects of this unit need to be revised.) 

In the next unit we look at a sequence that is almost never explicitly taught in textbooks: Assessment sequences in which a first pair-part assessment (“opinion”) is followed in the next turn by a second pair-part assessment (see Pomerantz, 1984). It is amazing that even advanced intermediate students can be at a complete loss for words when confronted with a simple assessment about a shared experience. Many assessments use very simple language and the responses can be equally simple. However, textbooks rarely present students with anything other than “I agree” and “I think so too” both of which can be heard as same evaluations, which are a weak form of agreement and often preface subsequent disagreement. This unit presents three simple formats for making assessments and then looks at the practices employed in doing next-turn upgrades, same evaluations, and downgrades.

Next up are news-telling sequences. We fill our daily talk with news, large and small. In fact, casual conversation is much more characteristically composed of tellings and respondings rather than of a string of questions and answers. The unit looks at first and second tellings (e.g. “I went to the gym yesterday.” “Really?” “Yeah, I worked out for about two hours.”), including some common news-response tokens, and common formats for pre-announcements and their go-ahead responses. YTAT finishes up with a rather short and insubstantial look at story prefaces.


What I would do differently next time

As stated above, YTAT was written almost 18 years ago! Since then I have completed my Ph.D. in conversation analysis, published a number of papers, taught courses on conversation analysis at the undergraduate and graduate level, presented at conferences, and continued to explore new ideas for teaching conversational interaction in my own ESL classes. So unsurprisingly, if I were to rewrite YTAT today, I would do a few things differently. First, I would remove the full telephone call from the first unit. This was intended as an example of a prototypical purpose-driven call with an unproblematic opening, no-gap speaker transitions, and a two-step closing where everything goes “smooth sailing.” Instead, I would focus more on the operation and variability of the component sequences that comprise typical face-to-face openings, e.g. greeting-greeting, how-are-you sequences, including their various formats (e.g. “How’s it going” “What’s up” “How are things”) and the fact that these are not inquires about health but rather are involved in negotiating a “first topic” for the conversation. Perhaps the largest change, and a real failing of the current version, would be an improved treatment of requests. I would insert a whole additional unit on “Telling your problem” (see Carroll, 2012) as a way of soliciting assistance rather than directly asking for help. Research on various types of “service-desk” encounters shows that overwhelmingly people begin by telling their problem, which is usually enough to get what they need, i.e. “I need an application form.” I teach a course on “practical English” where the entire one-semester course is organized around the one goal of “getting help in English.” This begins with teaching nine simple formats for stating trouble. This is followed with three common ways of offering help: 1) an immediate solution (“Here you go”), 2) a promise of assistance (“I’ll bring one tomorrow”) or 3) a suggestion (“You could…”). Only after this do we look at the use of formal requests. Very little of this language is new to university students, or high school students for that matter, but they’ve never been shown how to task the language for this purpose. 

Requests, then, can be seen as a next (dispreferred) action when the trouble-telling has failed to elicit a pre-emptive offer of help. Since requests sequences are themselves dispreferred, i.e. a request is a dispreferred first pair-part, I would rewrite at least some of the mini-logs to reflect the more hedged dispreferred turn-shape found in many requests, e.g. “Um…I’m sorry to bother you, but, um, I wonder if you could…” 

I would also want to include at least an introduction to how the sequences already dealt with can be, and often are, extended through insert and post expansions. As Schegloff (2007) makes clear insert expansions quite regularly intersect with conversational repair, particularly in terms of adumbrating an incipient dispreferred response. And this brings me to the final area where YTAT could use improvement: Repair. Not simple repair as in “correcting an error” or “checking for understanding” but rather repair as a strategic resource in managing interaction (“Let’s have pizza for breakfast” “What?” / “Pizza?” / “For breakfast?”).

Perhaps I could also work in some of the things my students and I been doing in another semester-long course exploring different question formats in English and the interactional goals they pursue, the idea that questions are almost always vehicles for other social actions, and in fact, may not even be questions at all (See Carroll 2016a, 2016b). This also includes looking at canonical and non-canonical response formats (see Fox and Thompson, 2010, Lee, 2013). I would also make numerous changes to the layout and design, as well as correcting a few typos. However, at this time I have no specific plans to develop YTAT further. 

(YTAT was originally created in PageMaker. I no long have the program nor the original PageMaker file, so I would essentially be starting from scratch.)


Notes on the use of YTAT

As should be the case with the use of any textbook, what is actually in the text is only the bare-bones floor plan upon which the structure of the lesson is built. It was never intended as a “page-turner” textbook, which probably renders it unpublishable in the current ESL publishing environment. Furthermore, it depends fundamentally on the instructor being knowledgeable about conversation analysis, or at least willing to learn. That is obviously a huge drawback since so few of the world’s language teachers have even heard of conversation analysis (“Oh, do you mean turn-taking?”) much less studied it.  (It is my view that every MATESOL program should include a course specifically on conversation analysis in the same way that every program includes a course on pedagogic grammar.) 

I should also say that I’ve moved away from using the booklet itself in class. I still hand out copies to the students, but rarely use them in class.  I’ve come to view the teaching of conversation as analogous to teaching a performing art, such as dance, or to coaching sports. You can’t use a textbook to teach dance or soccer. Also it is my experience that as soon as students have a textbook gripped in their hands, they lose all spontaneity, which is the heart and soul of interaction. My classes have become much more physically animated with students spending a large part of the class out of their seats. My ongoing research into novice L2 embodiment suggests that there is a close tie between physical engagement and conversational participation. As with dance or sports, each class begins with warm-ups. Specifically, we begin every day with 10 x 1 minute one-on-one conversations (rotating to a new partner each time), building towards the end of the semester to 3 x 5 minute conversations in groups of three or four. These warm-up conversation are usually followed with various activities designed to encourage SSJ rule 1b (“self-selection,” see Carroll, 2011a).

I’ll offer one more comment about the use of YTAT. It is in no sense “teaching CA” and there is no need whatsoever to burden learners with CA terminology. While a few basic CA terms do appear in YTAT, I don’t highlight them or insist that students learn them.



Dr. Donald Carroll is a professor at Shikoku Gakuin University in Japan, where he has worked for the past 20 years. Prior to moving to Japan, Don taught in Mexico at the Universidad Autonoma de Guadalajara and at ITESM Guadalajara. Before that Don spent 12 years teaching at universities in Saudi Arabia, Kuwait, and Oman. He holds a BA and MA in Linguistics from California State University Fullerton and earned his Ph.D. from the University of York (UK). During a 2003-2004 sabbatical, Don was a Visiting Scholar at UCLA where he was fortunate to attend Manny Schegloff’s renowned lectures on sequential structures as well as benefiting first-hand from Chuck Goodwin’s groundbreaking work on embodiment in interaction. Don has been a long-time advocate of looking beyond CA as a research methodology to actually applying its observations to language teaching. For several years Don taught a course entitled “Conversation Analysis: Observations and Implications for the Language Classroom” for Temple University Japan’s MATESOL program.


Publications related to conversation analysis and language teaching

Carroll, D. (2016a). Angling for an answer: Preference in polar question design. In D. Tatsuki & D. Fujimoto (Eds.) in Back to Basics: Filling in the Gaps in Pragmatic Teaching Material. JALT Pragmatics SIG, special publication book.

Carroll, D. (2016b). Please don’t answer in a complete sentence! In D. Tatsuki & D. Fujimoto (Eds.) in Back to Basics: Filling in the Gaps in Pragmatic Teaching Material. JALT Pragmatics SIG, special publication book.

Carroll, D. (2012). What’s your problem: Getting help the easy way. In J. Ronald, C. Rinnert, K. Fordyce, and T. Knight (Eds.). Pragtivities: Bringing Pragmatics to Second Language Classrooms. JALT Pragmatics SIG, special publication book.

Carroll, D. (2011a). Taking turns and talking naturally: Teaching conversational turn-taking. In N. Houck & D. Tatsuki (Eds.). Pragmatics: Teaching natural conversation. Alexandria, VA: TESOL

Carroll, D. (2011b). Teaching preference organization: Learning how not to say no. In N. Houck & D. Tatsuki (Eds.). Pragmatics: Teaching natural conversation. Alexandria, VA: TESOL

Carroll, D. (2010). Conversation analysis and language teaching: A call to action. In Greer, T. (Ed.). Observing talk: Conversation analytic studies of second language interaction. Tokyo: JALT Pragmatics SIG.

Carroll, D. (2006). Backwards-oriented self-repair in Novice L2 interaction. Treatises. (Shikoku Gakuin University).

Carroll, D. (2006). Co-constructing competence: Turn construction and repair in novice-to-novice second language interaction. Unpublished Ph.D. dissertation. University of York, UK.

Carroll, D. (2005). Vowel-marking as an interactional resource in Japanese novice ESL conversation. In K. Richards, K. and P. Seedhouse (eds.) Applying conversation analysis. New York: Palgrave-Macmillan.

Carroll, D. (2004). Restarts in novice turn beginnings: Disfluencies or interactional achievements? In Gardner, R. and J. Wagner (eds.) Second Language Conversations. Continuum: London/New York.

Carroll, D. (2000). Precision timing in novice-to-novice L2 conversations. Issues in Applied Linguistics: Special Issue on Nonnative Discourse (UCLA), Vol. 11, No. 1:67-110.



Data Session

Our next data session will take place on Saturday, July 21, from 1-3 pm at Grace Dodge (GDH) 273B.