Skip to content
BY 4.0 license Open Access Published by De Gruyter Open Access December 7, 2021

OH + OKAY in informing sequences: On fuzzy boundaries in a particle combination

Elizabeth Couper-Kuhlen
From the journal Open Linguistics


This study explores fuzzy boundaries, or weak cesuras, in the particle combination OH + OKAY as used in informing sequences in ordinary English talk-in-interaction. The focus is on the third position in such sequences, where OH + OKAY responds to information that has been solicited by the speaker in a prior turn. Based on a collection of approximately 45 instances of OH + OKAY in recent American English telephone and face-to-face interactions, this study adopts the methodology of Conversation Analysis and Interactional Linguistics to examine the prosody and phonetics of OH + OKAY as a feature of turn design. It asks: (i) when does the speaker use a combination of OH and OKAY in the third position rather than a simple OH or OKAY? and (ii) to what extent does the prosodic–phonetic delivery of OH + OKAY contribute to an interpretation of what the turn is doing? The study finds that (i) OH + OKAY is used when responding to a solicited informing that not only supplies information but also accomplishes another action with consequences for the recipient and (ii) if OH + OKAY is delivered with a strong cesura between its parts, its actions are distributed in separate turn-constructional units, creating a multi-unit turn that proposes sequence closure. If the cesura in OH + OKAY is weak, the component parts are fused into a compound particle that can preface more talk by the same speaker. Weak boundaries between OH and OKAY can be exploited by participants for the purposes of turn construction and epistemic positioning.

1 Introduction

In his seminal volume on sequence organization in conversation, Schegloff (2007) describes the particles OH and OKAY as devices for minimal sequence expansion. Because they do not per se project more talk, he calls them “sequence-closing thirds”, OH being found primarily in informative sequences and OKAY in directive sequences (2007, 120). Yet, in more recent English talk-in-interaction, OKAY is also increasingly encountered as a sequence-closing third in informative sequences – and not infrequently in combination with OH. It is combinations of OH + OKAY in the third position of solicited informing sequences that are the subject of this study.

As we will see, in such combinations OH and OKAY can be only loosely “juxtaposed” to one another in the third position, with each particle forming a turn-constructional unit of its own and together building a multi-unit turn. Or, they can be more tightly “fused” into a single unit, which can stand on its own as a turn-constructional unit or in cases of maximal fusion, even serve as a unitary item prefacing a turn-constructional unit. The distinction between “juxtaposed” and “fused” is primarily a prosodic–phonetic one; yet, it is by no means binary. Instead, there are varying degrees of integration between the two particles, creating cesuras of differing strength (Barth-Weingarten 2016, Barth-Weingarten and Ogden 2021, this issue), some of which participants can exploit for turn-constructional and social interactional purposes.

The present study grew out of a large-scale exploration of the particle OKAY as deployed in social interaction across a range of typologically diverse languages (Betz et al. 2021). For the English part of the project, over 500 tokens of freestanding OKAY in everyday American English conversation were collected and analyzed in terms of sequential position, prosodic design, and action (Couper-Kuhlen 2021a, 2021b). It was in this collection that combinations of OH + OKAY were first observed. The forty-five instances attested in informing sequences form the basis for the present investigation, which was carried out using the methodology of Interactional Linguistics (Couper-Kuhlen and Selting 2018). The aim of this type of research is to determine how language use is shaped by social interaction and, vice versa, how social interaction is shaped by language use. The approach is radically empirical, data-driven, and micro-analytic; it views language as a set of resources for conducting social interaction and analyzes these resources in ways that are sensitive to their composition and position in turns and sequences (Schegloff 1996, 2007).

Before exploring combinations of third-position OH + OKAY and the varying cesuras between them, we begin by reviewing the work done in the third position by simple OH and simple OKAY in informing sequences. This is necessary to appreciate what a combination of the two particles accomplishes.

2 Simple OH and simple OKAY in the third position of informing sequences

Informing sequences are built around the delivery of a piece of information that the informer treats as something that the recipient does not yet know. This information can be either volunteered or solicited by the ultimate recipient. Here, we consider only cases in which the recipient has requested information, typically by asking a question. Upon delivery of the information, the recipient can respond, inter alia, with simple particles such as OH and OKAY.

2.1 Simple OH

Heritage (1984) has argued that the particle OH marks a “change-of-state” in the knowledge or orientation of a speaker. In response to an informing or telling, it indicates that the speaker now knows something they did not know before. Used in the third position, freestanding OH proposes that the sequence initiated by the solicitation of information has been successful and can now be brought to a close. Here is a representative example to illustrate (the particle is in bold):[1]

(1) “Back from the gym” (Call Home 4092, 59.49)
((It is the Fourth of July: Andrea has announced to her sister Hannah that she has the day off and has just returned from the gym.))

In line 12, Hannah uses a rising-falling and somewhat lengthened OH to respond to Andrea’s news that she received pay for her day off on the day before their phone call. And in line 16, Hannah produces a freestanding, creaky OH in response to the information that Andrea is not being paid for her day off on the day of the phone call. Both these OH tokens are news receipts in the sense of Heritage (1984): in both cases, they indicate that Hannah did not know the information before being told by Andrea and that it has led to a change-of-state in her knowledge. Moreover, both OHs are closure-relevant, as evidenced by what happens next: Hannah initiates a new sequence after the first one (line 13), while a lengthy silence emerges after the second one (line 17) before Andrea resumes her report about the gym (not shown here). They differ, however, in their sequential position: while the first OH receipts volunteered news (line 11), the second OH receipts an informing that has been solicited by a prior query, is it paid today? (line 13). It is the second sequential environment, that of solicited informings, which is the object of this study.

As prior research has shown, the prosody of freestanding OH can be manipulated in ways that suggest an affective coloring. For instance, it can be delivered with extra-high pitch and a “pointed” rising-falling contour implying that the news was exceptional (Reber 2012). Or, it can be delivered with low pitch, soft volume, and a narrow falling contour, which in the context of the rejection of a proposal, will be interpretable as a display of disappointment (Couper-Kuhlen 2009). Yet, OH is not the only option for responding to a piece of solicited news. In the following section, we show that a freestanding OKAY can also occur in this sequential position, although it may do slightly different work.

2.2 Simple OKAY

The use of OKAY in response to solicited informings is a relatively recent development in American English: it is vanishingly rare in this sequential position in ordinary conversation before the 1990s (Couper-Kuhlen 2021a, 2021b). However, since then it has become a widely used device for acknowledging newly received information in both solicited and unsolicited informings.

Three kinds of situations can be distinguished in which OKAY is an appropriate way to respond to a solicited informing.[2] In the first of these, OKAY marks that the information just imparted has consequences for some future project of the speaker’s.[3] This situation occurs in the following extract:

(2) “Spending the night” (Call Friend 6899, 1665.397)
((Mom is about to travel to New Jersey to see Sally’s grandparents. Chris is Sally’s partner.))

In line 08, Sally responds with a freestanding OKAY to the information she has just solicited from Mom by asking are you spending the night? (0.3) in New Jersey then? (lines 02–04). On close scrutiny, it will be seen that Sally’s query is embedded in a larger sequence launched with well I I thought I would (line 01) but then broken off in favor of first determining how long Mom will be staying in New Jersey. Once it has been established that Mom will be there overnight (lines 05, 07), Sally pursues her project with w’then (line 10), but again breaks off when Mom begins to detail her travel plans (line 11, etc.). Ultimately, it emerges that Sally wants to call her relatives while Mom is visiting them in New Jersey (lines 49–51) but will be unable to do so in the evening because she and her partner are going out (line 28). The information Mom provides about spending the night in New Jersey thus allows Sally to form a plan to call there the following morning (see line 52).

In such situations, an OKAY response confirms the consequentiality of the information just imparted for the speaker. Consequential OKAY contrasts with a simple change-of-state OH: while OH blandly acknowledges receipt of a piece of information the speaker did not know before, it does not imply, as does consequential OKAY, that this information is relevant in any way to what the speaker might do in the future (Couper-Kuhlen 2021b).

On other occasions, third-position OKAY following a solicited informing indicates that the solicited information has brought about a revised understanding of a state of affairs, often one implicated in prior talk.[4] The following exchange illustrates the use of a “revised-understanding” OKAY in the third position:

(3) “Bridesmaids” (Call Friend 6938, 1399.810)
((Anabel has asked Renate when their friend Amanda is getting married and Renate has informed her that the wedding is not planned for anytime soon. She (Anabel) indicates that she is relieved there will not be another wedding to attend that summer.))

Renate’s pronouncement in line 02 we’re all bridesmaids indicates that she believes Anabel will also be a bridesmaid in Amanda’s wedding. When Anabel denies this (line 05), Renate expresses disbelief with the query you’re not? (line 07). Anabel’s renewed (but epistemically weaker) disclaimer I doubt it (line 09) now leads Renate to revise her understanding of the situation. This is testified to by her OKAY in line 11. Her subsequent concession with anyway (line 13) indicates that she now provisionally accepts that Anabel will not be a bridesmaid in Amanda’s wedding.

Renate could also have responded with OH in line 11 of extract (4). However, the effect would have been somewhat different: she would be heard as treating the information as something she did not know before but not necessarily as acknowledging that it has forced a revision of her beliefs and assumptions. For this reason, OKAY could be said to contrast with OH in this sequential environment as well (Couper-Kuhlen 2021b).

Finally, third-position OKAY is found in situations where the informing is neither consequential for a future project of the speaker nor corrective of any of their assumptions. Instead, this non-consequential/non-corrective OKAY merely marks that the information has contributed to the recipient’s understanding of the situation. This use of OKAY is exemplified in the following extract:

(4) “Weight Watchers” (Farmhouse, 19.56)
((Mom and Laura have been talking about Weight Watchers and the recommendations their adviser gives them there. Their guest Donna now wants to know more.))

In line 10, Donna responds to Mom’s informing about where she goes for Weight Watchers with OKAY; yet, this information is not consequential for Donna: she herself does not have a weight problem nor is there any suggestion in the talk that she might want this information in order to recommend Weight Watchers to someone else. Furthermore, Mom’s informing does not lead to a revised understanding on the part of Donna: there is no indication that Donna has beliefs to the contrary. Yet, Mom’s informing does contribute to Donna’s understanding of the situation. Here, Donna could easily have responded with OH instead of OKAY: her OKAY is simply receipting the information that she has requested, much as OH would do (Couper-Kuhlen 2021b).

In sum, there are three distinct situations in which a simple OKAY is an appropriate way to respond to a solicited informing: one in which the information is treated as consequential for the OKAY speaker’s future project, one in which the information is treated as corrective of the OKAY speaker’s belief or assumption, and one in which the information is simply treated as having contributed to the speaker’s understanding of the situation. While in the first two cases OKAY contrasts with OH, in the last case it is essentially equivalent to OH in the work it does.[5]

Yet, on occasion, recipients do not choose one simple particle over the other in the third position of informing sequences; instead, they combine the two.[6] We turn now to these combinations.

3 OH + OKAY in the third position of informing sequences

When recipients of a solicited informing respond with a combination of OH and OKAY, it is invariably in the order OH followed by OKAY. That is, they first address the informativeness of the information delivered and then its implications.[7] Schegloff (2007) would argue that such combinations are needed to respond to prior turns that are running on two tracks. And indeed, in such combinations, the individual contributions of each particle can often still be identified, with OH marking that the informing has brought about a change of state and OKAY marking its implications for the speaker’s projected course of action, assumptions, or understanding of the situation. Yet, having seen that also simple OH and simple OKAY can be used as particle responses to solicited informings, we must ask: When and why does a speaker use both OH and OKAY in the third position rather than simple OH or OKAY?

Moreover, the combination of OH and OKAY can take on a variety of different prosodic–phonetic shapes. At one extreme, each particle forms a prosodic unit of its own and together they build a multi-unit turn. At the other extreme, the two particles are fused together so closely that they imperceptibly merge into one. There are also intermediate degrees of separateness or fusion between the two particles. This raises the following questions: when is it appropriate to deliver OH + OKAY in two separate units and when is it appropriate to deliver them as one? And what does the lack of a strong cesura between the two particles allow participants to accomplish?

It is questions such as these that we propose to explore in the following. We begin by examining cases in which OH and OKAY form separate units in a multi-unit turn.

3.1 OH + OKAY with a strong prosodic–phonetic cesura

A strong cesura in the production of OH + OKAY means that each particle will be processed as doing its own work, although together they build a response delivered in a multi-unit turn. Here is a case in point:

(5) “New job” (Call Home 4247, 491.51)
((Alan and Gina are both English language teachers for foreign students; Gina is currently located in Russia.))

In line 02, Alan inquires whether he has told Gina about his new job. Gina’s response is a transformative one (Stivers and Hayashi 2010) introduced by well (Heritage 2015): she affirms that he told her about starting a new job but did not tell her what it was (lines 04–06). While the affirmative part of this response reassures Alan that he was not totally remiss, the negative part comes off as an incipient complaint (Schegloff 2007). Alan replies immediately, first with a high-peaked OH (Reber 2012) and then, following a micro-pause, with OKAY (lines 07–08). After an in-breath and a short delay, he now begins to detail how he came to change jobs (lines 10–17).

As can be seen in Figure 1,[8] OH + OKAY is delivered in this instance with a strong prosodic–phonetic cesura: each particle has a pitch accent of its own, the phonation cuts off after OH before resuming on the first syllable of OKAY, and there is a micro-pause between the two particles. Furthermore, there are three separate peaks of intensity, one for each of the three syllables. At the same time, the two prosodic units are bound together by an overarching line of pitch and loudness declination, which creates a measure of cohesion between them.

Figure 1 
                  Pitch trace and intensity curve for lines 07–08 in Ex. (5) “New job”.

Figure 1

Pitch trace and intensity curve for lines 07–08 in Ex. (5) “New job”.

The separate delivery of OH + OKAY in Ex. (5) is indicative of the composite work it does. With a “high and pointed” OH (Reber 2012), Alan receipts the last, negative part of Gina’s informing turn as a piece of news that is surprising, the implication being that he did not intentionally keep Gina in the dark. With the following lower-pitched OKAY, Alan acknowledges and accepts the consequences of Gina’s informing, namely that she does not know what his new job is: he now proceeds to fill her in on the details. Had Alan responded only with OH, this would have testified to the fact that he now knows something he did not know before. However, there would be no acknowledgment of consequences ensuing from this change-of-state. On the other hand, had Alan responded only with OKAY, this would acknowledge that Gina is now due an explanation, but it would avoid any acknowledgment that he had not given her the details before and with it, any recognition of her incipient complaint. Thus, both parts of this combination are functional: the fact that they are two prosodically separate entities underlines that each particle accomplishes a different job.

OH + OKAY can be delivered as separate entities also in cases where a solicited informing forces a recipient to revise their understanding of a situation:

(6) “Rummikub” (Call Friend 6899, 663.725)
((Sally knows that her mother is looking for a board game called Rummikub like her Aunt Erica’s; in this episode, she announces that she saw one but did not buy it because it was not nice enough.))

In line 15, Mom asks Sally whether the Rummikub game she saw was in a case. However, she words her question in a way that suggests she expected that it would be: it’s not in a case? Sally’s response is significantly delayed and epistemically weak: I don’t think so (line 17), followed by an account for her hesitation it’s hard to tell from the box (line 18). This informing provides a weakly negative answer to a question whose design indicated an expectation to the contrary, in addition to an explanation for the uncertainty of the answer. Mom’s OH (line 19) receipts Sally’s informing as a disappointing piece of news (Couper-Kuhlen 2009): if the game does not have a case, then it is not worth buying. Her OKAY (line 20) marks an acknowledgment of the consequences deriving from Sally’s informing: Mom’s understanding of the situation has now been revised and she will have to keep looking.

Here too, there is a strong cesura between the two particles, as can be seen from Figure 2. OH is stretched, with a low falling glide. The phonation is then discontinued for a brief interval before setting in again for the softer OKAY. There are three peaks of intensity. Yet despite the intervening pause, the two units are linked through the overlay of breathy voice: they thus belong together and form a multi-unit turn.

Figure 2 
                  Pitch trace and intensity curve for lines 19-20 of Ex. (6) “Rummikub”.

Figure 2

Pitch trace and intensity curve for lines 19-20 of Ex. (6) “Rummikub”.

As in (5) “New job”, the separate units OH and OKAY in (6) “Rummikub” execute separate actions that deal with the informing: one addresses its informativeness, the other its implications.

3.2 OH + OKAY with a weak prosodic–phonetic cesura

The particles OH and OKAY can also occur in combination with a weak cesura. Nevertheless, even if the particles are not clearly separated from one another, they still do similar work. Let us examine a representative case:

(7) “Out of touch” (Call Home 4247, 419.117)
((Gina has been describing the pre-departure orientation course she gives for Russian students who are about to study in the States. But, she claims, she is not really up on American attitudes to topics like racial discrimination, since she has been away so long.))

Alan’s query in line 08 so you really are out of touch arncha is at once a request for confirmation from Gina and a tease (Drew 1987). Rather than joining in Alan’s post-completion laughter (line 10), Gina “corrects” Alan’s understanding, claiming that she does not think of herself as completely out of touch (line 13), although she does overlay this with a few aspiration particles. She goes on to add that even when she was in the States she was not particularly in touch (lines 16–17). It is this informing that Alan now receipts with OH + OKAY (line 18). His OH registers the prior turn as new information which has brought about a change-of-state, while his OKAY marks that he has now revised his understanding based on that new information. In fact, he immediately goes on to specify what this new understanding is: he now knows that Gina is “that type” (line 19), an oblique reference to the kind of person who is not fully “with it”. This too, is done as a tease, as his I gotcha and subsequent laughter indicate (line 20).

Alan’s production of OH + OKAY is run-on, as can be seen from Figure 3: the only break in phonation is that associated with the closure phase of the velar plosive /k/. The vowel of OH glides imperceptibly into the first vowel of OKAY via an epenthetic /w/, visible as a slight hump between the two intensity peaks associated with OH and the first syllable of OKAY. There are two pitch accents: a fall from high on OH and a rising-falling contour on OKAY, followed by a final slight rise in termination. But despite there being two pitch accents, this particle combination is produced in one turn-constructional unit, with a rhythmic group created by three strong beats /DA DA DA/. The following unit you’re ↑THAT type is immediately latched on to the unit formed by OH + OKAY.

Figure 3 
                  Pitch trace and intensity curve for line 18 in Ex. (7) “Out of touch”.

Figure 3

Pitch trace and intensity curve for line 18 in Ex. (7) “Out of touch”.

Although Alan’s OH and OKAY in Ex. (7) are run-on, both words are still audibly present. And, they each perform a distinct job: OH treats Gina’s prior claim as a new piece of information, while OKAY acknowledges that it has implications. Had Alan responded only with OH, he would not have marked that the informing has consequences for him; had he responded only with OKAY, he would not have acknowledged that Gina had told him something new.

On occasion the cesura between OH and OKAY is even weaker; when this happens, the particle combination OH + OKAY can function as a turn-initial preface to what follows. Yet to the extent that the two components can still be distinguished, they continue to do distinct work. This is what we find in the next extract:

(8) “Moving” (Call Home 4248, 171.25)
((In the earlier talk, it has been established that Anita will be moving at some point in the future. She wants Brenda to come and visit her before then, but Brenda demurs because her husband still has to defend his thesis.))

In this negotiation over whether Brenda might visit, Anita virtually implores her friend to come with a threefold are you coming or not followed by jocular whining (lines 01–03, 05, 07), while Brenda displays reluctance to commit, claiming that her husband still has to defend his thesis (line 14) and also has a job (line 19). Now Brenda launches an alternative plan with if not (line 21), hinting that she might come at a later date.[9] For this alternative plan to work, however, the visit would need to be before Anita moves. Thus, her turn continuation: when are you guys moving (line 21).

Anita’s response is delayed and framed in the negative: we’re not moving for a year (line 23). It is also done as a full clause rather than as a phrase, e.g., “not for a year”, suggesting that Anita may find Brenda’s question somewhat tedious (Thompson et al. 2015, 28ff). Yet, the negative format is designed to allay any fears Brenda might have that the move could be soon. This may account for Brenda’s response: she receipts the informing with a loud and high-pitched OH followed immediately by OKAY and a positive assessment of Anita’s news as good (line 25).

As can be seen from Figure 4, not only OH and OKAY are run together but also with them the assessment good. All three words are positioned on one line of pitch and loudness declination, with OH being the highest and loudest of the three. The only break in phonation is for the closure phase of /k/ in OKAY, whose articulation, however, is overall weaker than that of OH. Rhythmically, the three words are integrated into a single group /DA da da DA/. There are two pitch accents: a fall from high on OH and a lengthened good. The prominence on good is, however, subordinate to that on OH in terms of pitch and intensity.

Figure 4 
                  Pitch trace and intensity curve for line 25 in Ex. (8) “Moving”.

Figure 4

Pitch trace and intensity curve for line 25 in Ex. (8) “Moving”.

Altogether then, the particle combination OH + OKAY in Extract (8) – because of the very weak cesura between its parts – functions as an indivisible whole. Moreover, the weak cesura between it and the following good leads to OH + OKAY being heard as a preface to the assessment that follows.[10] Yet despite the three words being run together, each makes its own contribution to the work of the turn, with OH treating Anita’s informing as news, OKAY marking its implications for Brenda’s understanding of the situation, and good assessing both positively. If one of these components were lacking, the work being done would change accordingly.

In sum, what we have seen is that the particle combination OH + OKAY can be delivered with varying degrees of prosodic–phonetic integration, with pitch, loudness, rhythm, pausing, and voice quality contributing to cesuras of different strength. The examples we have shown here have been at different points on a continuum between separation and fusion: while the OH + OKAY in Ex. (6), with a 0.2 s break in phonation between the two parts, might be said to be at one extreme, that in Ex. (8), where the two parts are run together with no perceptible break at all, might be closer to the other extreme. Cases like that in Ex. (7), where the two parts are separated by an epenthetic /w/, are located in between, if closer to the “fused” end. Assuming for the moment a binary categorization of the cases in our collection as “separate” or “fused” based on whether there is a break in phonation between OH and OKAY or not, the “fused” instances are by far in the majority, making up slightly more than two-thirds of the collection. The minority of “separate” instances are distributed roughly equally across the three different types of OKAY.[11]

4 On the development of OH + OKAY

What does this tell us about where the particle combination OH + OKAY came from and where it is going? It might be speculated that the use of OH + OKAY developed out of a need to deal with multi-unit informings, i.e., turns with two (or more) turn-constructional units. This was, for instance, the case in (5) “New job”, where Gina concedes that Alan has told her that he was changing jobs but adds that he has not told her what the new job is. For this case, it could be argued that “separate” OH + OKAY allows Alan to deal in turn with each part of Gina’s informing. The informing in (6) “Rummikub” was also multi-unit: Sally first states that she does not think the game she saw had a case and then provides an account for why she is not completely sure. Here too, the “separate” components of Mom’s OH + OKAY response might be said to deal with the two parts of Sally’s informing.

Yet also “fused” OH + OKAYs are found when the informing is multi-unit. This is what we saw in (7) “Out of touch”, where Gina first claims that she is not totally out of touch and then asserts that she was not really in touch before leaving the country. Alan’s responsive OH + OKAY is delivered in one go. So, it does not seem to be the case that multi-unit informings call for “separate” OH + OKAYs to the exclusion of “fused” variants.

But do single-unit informings perhaps call for “fused” OH + OKAY to the exclusion of “separate” alternatives? In (8) “Moving”, for instance, we find a “fused” OH + OKAY being used to receipt an informing turn that consists of just one turn-constructional unit, delivering the information that Anita will not be moving for a year.

However, there are also cases where a “separate” OH + OKAY is used to respond to a single-unit informing. For instance, in the following extract:

(9) “Faculty meeting” (Call Home 4112, 351.38)
((Clara is a counselor for foreign students studying in the States and is planning to visit Denise, who teaches a study-abroad class in Japan. They have agreed on the month and are now trying to settle on the date of Clara’s arrival.))

When Clara proposes to arrive in the late afternoon or evening of the eighth (line 04), Denise initially agrees but then adds that she has a department faculty meeting on the eighth (lines 06, 08, 10). This report hints at a possible obstacle to Clara’s arriving that same day (see also Drew 1984). Clara’s next turn aims to establish how long the faculty meeting will last (line 14), presumably on the assumption that if it does not last long, she could still come that same day. Denise hesitates before answering and prefaces her turn with well, suggesting it will not be straightforward (Schegloff and Lerner 2009). Her answer is transformative (Stivers and Hayashi 2010) in that it does not accept the terms of the question to provide the information Clara is looking for but instead offers information that would allow Clara to infer the upshot. The negative formulation they don’t start til five ten (line 16) implies that although an earlier start might be expected, in actual fact the faculty meeting starts late, the implication being that it will also go late. In the next turn Clara first receipts this informing with OH, marking a change-of-state (line 17), and after the shortest of breaks marks an acknowledgment of what this means for her travel plans with OKAY (line 18). Following a short intervention from Denise, she then proceeds to spell out what the consequences of the informing are: so it would be better to come in on like (0.4) the morning of the ninth (line 20).

Figure 5 shows how Clara’s OH + OKAY is delivered prosodically. There is a small break in pitch and intensity between OH and OKAY. Each particle has its own pitch accent, in both cases a fall from high. There is a slight uptick in pitch and intensity at the beginning of OKAY, meaning that the declination is reset on that word. There are three peaks of intensity, but the beats are not integrated into a unified rhythmic group. Instead, there are two rhythmic units: /DA/ followed by /DA DA/. Yet, we do not hear these units as separated by a transition-relevance place. Although there is a slight break between them, it does not create enough space for the other speaker to legitimately come in (Jefferson 1984). Instead, the second unit is heard as “latched” onto the first forming a multi-unit turn.

Figure 5 
               Pitch trace and intensity curve for lines 17–18 of Ex. (9) “Faculty meeting”.

Figure 5

Pitch trace and intensity curve for lines 17–18 of Ex. (9) “Faculty meeting”.

Denise’s informing in line 16 consists of a single turn-constructional unit; yet, it is responded to with a “separate” OH + OKAY. As we have seen elsewhere, OH deals with the informativeness of the prior turn and OKAY with its implications for Clara’s project. Had Clara responded only with OH, there would be no indication that she perceived or accepted the consequences of what the information means for her arrival. Had she responded only with OKAY, she would have accepted these consequences but would not have acknowledged that Denise’s information was something she did not know before.

In sum, it does not appear to be the structural complexity of the informing turn – whether it consists of several turn-constructional units or only one – that explains when OH + OKAY is fused and when it is not. Both delivery formats are found with multi-unit and single-unit informings. Instead, the explanation would seem to be another:

Certain informings, whether multi-unit or not, can implement “composite” actions (Rossi 2018), and the OH + OKAY particle combination is an appropriate way to acknowledge this. In the examples presented here, for instance, we have encountered turns that not only inform but also do incipient complaining ((5) “New job”); turns that inform and at the same time correct a belief or preconception ((6) “Rummikub”, (7) “Out of touch”); turns that inform and simultaneously disabuse the other of a false expectation ((9) “Faculty meeting”); and turns that inform as well as unexpectedly relieve the other of a worry or concern ((8) “Moving”). Other types of “composite” informings are imaginable. What it boils down to is whether the informing is doing something in addition to informing, something with implications for the recipient. It is in these cases that we can expect to find OH + OKAY used as a response rather than a simple OH or a simple OKAY.

Whether OH + OKAY is delivered as separate or fused is likely to have to do with the way language develops. We know that the use of third-position OKAY in informing sequences is a relatively recent phenomenon (Couper-Kuhlen 2021b). When a piece of information is not only consequential but also “new”, possibly making a display of affect relevant,[12] we can hypothesize that, initially, the response will have been done with OH and OKAY in their own units. But as we know, with increased frequency and routinization, cesuras between components weaken and originally separate units begin to fuse (Bybee 2001, 2006, Haiman 1994). This can lead to a process of grammaticalization, whereby two or more independent units of language structure “grow together” into a single construction with integrated components (Narrog and Heine 2011, 13).[13] Currently, we can assume that this fusion process is well underway in the case of OH + OKAY: Particle combinations with weak cesuras are significantly more frequent than those with strong cesuras in the present collection. Should the process continue, it may ultimately end with the birth of a new particle OOKAY, distinguished from OKAY through a lengthened first syllable.[14] Only the future will tell.

5 What do weak cesuras with OH + OKAY allow speakers to achieve?

At the moment, when responding to composite-action solicited informings, speakers appear to have a “free” choice: they can produce OH + OKAY separately, with a strong cesura separating the parts, or they can deliver the two particles as a fused entity with only weak cesuras. There does not appear to be a functional distinction between the two modes of delivery in terms of what is being done.[15] But as with other types of so-called “free” variation in language, we know that seemingly interchangeable alternatives can be exploited in conversation for sequential and interactional purposes (see Couper-Kuhlen and Selting 2018, 546ff). This is true as well for the production variants of OH + OKAY.

For instance, strong vs weak cesuras in OH + OKAY can be used strategically to manage turn construction. With a strong cesura, each component stands on its own and OKAY carries its full weight as a sequence-closing third. This allows for the possibility of speaker transition afterward, as in (5) “New job”, where a brief lapse can be seen in line 09, or as in (6) “Rummikub” and (9) “Faculty meeting”, where the transition to next speaker actually ensues. On the other hand, with a weak cesura OH + OKAY forms a unitary item, a compound particle, one that can be deployed, e.g., to construct a multi-unit turn. In (7) “Out of touch”, for instance, an integrated OH + OKAY is followed immediately by a new unit formulating an inference that derives from the news and its implications. Moreover, with a maximally weak cesura, OH + OKAY can serve as a compound-particle preface to a turn-constructional unit (Heritage and Sorjonen, eds. 2018). This is what happens in (8) “Moving”, where OH + OKAY essentially functions as a preamble to the following assessment good. In both cases – with OH + OKAY building a turn-constructional unit that is the first part of a multi-unit turn or with OH + OKAY serving as a turn-initial (compound) particle – the weight of OKAY as a sequence-closing third is significantly reduced.

But producing OH + OKAY with a weak cesura can also serve tactical, social interactional purposes. For instance, it can be useful in epistemic positioning, i.e., in negotiating one’s own knowledgeability status relative to that of the interlocutor. This is what we find happening in the following extract, where Brenda, who is pregnant with her first child, has asked her friend Anita how she handled feeding when Anita’s five-year-old daughter was an infant. Anita explains that she nursed the baby, whereupon Brenda inquires whether she pumped her milk when she was at work and how she did it. As we join the conversation, Anita is describing the pumping machine she used.

(10) “Plastic container” (Call Home 4248, 466.26)
((The reference in line 01 is to the pumped milk.))

Although Anita is more experienced in nursing than Brenda, Brenda’s questions are nevertheless indicative of a “partially knowing” stance (Heritage 2012). This is evident in lines 01–02, where she provides a candidate answer for her own question about pumped breast milk: what does it go INto = like a little plastic BAG thingy? And, it is evident in the declarative format of her next question, a request for confirmation: and you can REuse the plastic conTAINer? (line 12). Yet, Anita’s responses do not directly provide the confirmation Brenda is seeking. For instance, Anita’s turn in lines 14–17 does not say whether the plastic container can be reused or not. Instead, it pursues a report of how she brought little bags to work to use with the plastic container.

Following a brief pause (line 18) indicating that some uptake on Anita’s report is now due, Brenda attempts a response, one that emerges as highly ambiguous. On one hearing, she produces a version of OH + OKAY with only the weakest of cesuras. There is no break in phonation between the two words. There are two peaks of intensity, one corresponding to OH_O- and one corresponding to -KAY, but only one pitch accent, a falling contour starting on the first syllable OH (see Figure 6).

Figure 6 
               Pitch trace and intensity curve for line 19 in Ex (10) “Plastic container”.

Figure 6

Pitch trace and intensity curve for line 19 in Ex (10) “Plastic container”.

But there is an alternative hearing compatible with these features, namely that Brenda’s response is a simple O:KAY with a lengthened first syllable. This would make it more like her OKAY in line 11. The rhythm of line 19 is indeterminate between /DA da da/ and /DA: da/. Figure 7 shows a Praat picture of the two response tokens side by side. The duration of the first syllable of the particle (combination) in line 19 is 0.1751 s, while that of the first syllable of the particle in line 11 is 0.1577 s: that is, the former is only roughly 11% longer than the latter.

Figure 7 
               Pitch traces and intensity curves for line 19 (left) and line 11 (right) in Ex (10) “Plastic container”.

Figure 7

Pitch traces and intensity curves for line 19 (left) and line 11 (right) in Ex (10) “Plastic container”.

The ambiguity of this turn allows Brenda to achieve two things at once. If the response is heard as a (fused) OH + OKAY, then she goes on record as acknowledging that she has been told something she did not know before, namely that you use little bags to put the pumped milk in and that this information has implications for her understanding of the situation. But if it is heard only as simply O:KAY, then Brenda goes on record as acknowledging only that the informing has contributed to her understanding of the situation. She avoids acknowledging that (Anita’s) bringing little bags to work to put the pumped milk into is something she did not know about before. This allows her to preserve an epistemic stance of being (partially) in the know about pumping. By later insisting on her understanding that the plastic bag is ultimately thrown away (lines 36–40), Brenda shows a concern to be seen as someone who is not wholly unknowledgeable [K-] about pumping breast milk. In line 19, it is thus in her interest to use a weak cesura that obscures the OH and with it the implication of epistemic inferiority (Stivers et al. 2011).

6 Conclusion

Recapping the argument, we have seen that OH + OKAY is one way of responding in third position to an informing that has been solicited. In contrast to simple OH or simple OKAY in this position, OH + OKAY does two things at once: it acknowledges that the information provided is “news” and has led to a change-of-state. (This change-of-state can be overlaid with affective marking.) At the same time, OH + OKAY acknowledges and accepts that the information provided has implications for one’s own project, beliefs, and/or understanding of the situation. OH + OKAY is called for when the prior turn is doing more than just informing (if it were only informing, a response with simple OH or simple OKAY would suffice). In addition to informing, the turn can be, e.g., incipiently complaining, disabusing the other of a false expectation, correcting a misconception, or conveying something unexpected – to mention some of the cases seen here.

OH + OKAY can be produced in two separate prosodic–phonetic units which together build a multi-unit turn, or its two parts can be “run on”, with a cesura between them of varying degrees of strength. We have seen examples at each end of the continuum and some in between. In maximally “separate” cases, there is a break in phonation at the join between the two words, which may or may not be accompanied by a pause. There are two separate pitch accents, one on each word, and three peaks of intensity corresponding to each of the three syllables. There are typically two rhythmic groups: /DA/ and /DA DA/. However, the separate prosodic units are typically held together by a common line of pitch and loudness declination or by an overlying voice-quality effect – which indicates that they are part of a larger entity, namely a multi-unit turn. In “fused” cases, on the other hand, there is no break in phonation between the two words: they may be linked by an epenthetic [w] or the offset of the vowel of OH may merge directly into the onset of the first vowel of OKAY. The pitch accents may be reduced from two to one, the peaks of intensity from three to two. There is usually only one rhythmic group, which can be /DA DA DA/ or /DA da da/ – to mention the variants seen here. In the most extreme case, the vowel of OH and the first syllable of OKAY merge into a single long [o].

We have argued that all of the production variants of OH + OKAY accomplish the same job of receipting an informing as “news” and acknowledging its implications for the recipient. However, “separate” variants of OH + OKAY have turn-constructional implications that “fused” variants may lack. For instance, OKAY has more weight as a potential sequence-closing third when it is produced as a separate entity. With “fused” parts, OH + OKAY can become a unitary item with positional flexibility: for instance, it can be used to build a turn-constructional unit in a multi-unit turn or it can serve as a turn-constructional preface. Finally, “fused” variants of OH + OKAY can have social interactional advantages, in that a maximally weak cesura obscures the boundary between the two words. Since the vowel at the end of OH is the same as the vowel at the beginning of OKAY, this can lead to the perception of a simple O:KAY (with a lengthened first syllable) rather than of a particle combination OH + OKAY. OH is thereby deleted, and with it, an acknowledgment of epistemic inferiority relative to the other. Weak cesuras between OH and OKAY thus have turn-constructional and social interactional implications that speakers can put to their advantage when responding to solicited information.


Earlier versions of this paper were presented at the workshop “Divergent units and fuzzy boundaries” (University of Potsdam, September 2018) and a similarly named IIEMCA panel (University of Mannheim, July 2019), both organized by Dagmar Barth-Weingarten. I am grateful to the participants at these events, to two anonymous reviewers, and especially to the editors of this special issue for encouraging me to pursue the topic of OH + OKAY and for helping me to hone my argument. All remaining errors and infelicities are my own.

  1. Conflict of interest: No conflicts of interest were involved in this work.


Barth-Weingarten, Dagmar. 2016. Intonation units revisited: Cesuras in talk-in-interaction. Amsterdam: John Benjamins.Search in Google Scholar

Barth-Weingarten, Dagmar and Richard Ogden. 2021. “‘Chunking’ spoken language: Introducing weak cesuras.” Open Linguistics 7(1), 531–48.Search in Google Scholar

Betz, Emma, Arnulf Deppermann, Lorenza Mondada, and Marja-Leena Sorjonen, eds. 2021. OKAY across languages: Toward a comparative approach to its use in talk-in-interaction. Amsterdam: John Benjamins.Search in Google Scholar

Bybee, Joan. 2001. Phonology and language use. Cambridge: Cambridge University Press.Search in Google Scholar

Bybee, Joan. 2006. “From usage to grammar: The mind’s response to repetition.” Language 82(4), 711–33.Search in Google Scholar

Couper-Kuhlen, Elizabeth. 2009. “A sequential approach to affect: The case of ‘disappointment’.” In: Talk in interaction. comparative dimensions, eds. Markku Haakana, Minna Laakso and Jan Lindström, p. 94–123. Helsinki: Finnish Literature Society.Search in Google Scholar

Couper-Kuhlen, Elizabeth. 2011. “Grammaticalization and conversation.” In: The Oxford handbook of grammaticalization, eds. Bernd Heine and Heiko Narrog, p. 424–37. Oxford and New York: Oxford University Press.Search in Google Scholar

Couper-Kuhlen, Elizabeth. 2021a. “The prosody and phonetics of OKAY in American English.” In: OKAY across Languages: Toward a comparative approach to its use in talk-in-interaction, eds. Emma Betz, Arnulf Deppermann, Lorenza Mondada and Marja-Leena Sorjonen, p. 131–73. Amsterdam: Benjamins.Search in Google Scholar

Couper-Kuhlen, Elizabeth. 2021b. “Language over time: Some old and new uses of OKAY in American English.” Interactional Linguistics 1, 33–63.Search in Google Scholar

Couper-Kuhlen, Elizabeth and Dagmar Barth-Weingarten. 2011. “A system for transcribing talk-in-interaction: GAT 2. English translation and adaptation of Selting, Margret et al: Gesprächsanalytisches Transkriptionssystem 2.” Gesprächsforschung Online 12, 1–51.Search in Google Scholar

Couper-Kuhlen, Elizabeth and Margret Selting. 2018. Interactional linguistics: Studying language in social interaction. Cambridge: Cambridge University Press.Search in Google Scholar

Drew, Paul. 1984. “Speakers’ reportings in invitation sequences.” In: Structures of social action. studies in conversation analysis, eds J. Maxwell Atkinson and John Heritage, p. 129–51. Cambridge: University Press.Search in Google Scholar

Drew, Paul. 1987. “Po-faced receipts of teases.” Linguistics 25, 219–53.Search in Google Scholar

Haiman, John. 1994. “Ritualization and the development of language.” In: Perspectives on grammaticalization, ed. William Pagliuca, p. 3–28. Amsterdam: Benjamins.Search in Google Scholar

Heritage, John. 1984. “A change-of-state token and aspects of its sequential placement.” In: Structures of social action. Studies in conversation analysis, eds. J. Maxwell Atkinson and John Heritage, p. 299–345. Cambridge: Cambridge University Press.Search in Google Scholar

Heritage, John. 2012. “Epistemics in action: Action formation and territories of knowledge.” Research on Language and Social Interaction 45(1), 1–29.Search in Google Scholar

Heritage, John. 2015. “Well-prefaced turns in English conversation: A conversation analytic perspective.” Journal of Pragmatics 88, 88–104.Search in Google Scholar

Heritage, John and Marja-Leena Sorjonen, eds. 2018. Between turn and sequence: Turn-initial particles across languages. Amsterdam: Benjamins.Search in Google Scholar

Jefferson, Gail. 1984. “Notes on some orderliness of overlap onset.” In: Discourse analysis and natural rhetorics, eds. Valentina D’Urso and Paolo Leonardi, p. 11–38. Padova: Cleup Editore.Search in Google Scholar

Kaimaki, Marianna. 2011. “Sequentially determined function of pitch contours: The case of English news receipts.” York Papers in Linguistics Series 2, 11.Search in Google Scholar

Narrog, Heiko and Bernd Heine. 2011. “Introduction.” In: The Oxford handbook of grammaticalization, eds. Bernd Heine and Heiko Narrog, p. 1–15. Oxford and New York: Oxford University Press.Search in Google Scholar

Raymond, Geoffrey. 2013. “At the intersection of turn and sequence organization: On the relevance of ‘slots’ in type-conforming responses to polar interrogatives.” In: Units of talk – units of action, eds. Beatrice Szczepek Reed and Geoffrey Raymond, p. 169–206. Amsterdam: Benjamins.Search in Google Scholar

Reber, Elisabeth. 2012. Affectivity in interaction. Sound objects in English. Amsterdam: Benjamins.Search in Google Scholar

Rossi, Giovanni. 2018. “Composite social actions: The case of factual declaratives in everyday interaction.” Research on Language and Social Interaction 51(4), 379–97.Search in Google Scholar

Schegloff, Emanuel A. 1996. “Turn organization: one intersection of grammar and interaction.” In: Interaction and grammar, eds. Elinor Ochs, Emanuel A. Schegloff, and Sandra A. Thompson, p. 52–133. Cambridge: Cambridge University Press.Search in Google Scholar

Schegloff, Emanuel A. 2007. Sequence organization in interaction: A primer in conversation analysis, Vol. 1. Cambridge: Cambridge University Press.Search in Google Scholar

Schegloff, Emanuel A. and Gene H. Lerner. 2009. “Beginning to respond: Well-prefaced responses to wh-questions.” Research on Language and Social Interaction 42(2), 91–115.Search in Google Scholar

Selting, Margret, et al. 2009. “Gesprächsanalytisches Transkriptionssystem 2 (GAT 2).” Gesprächsforschung - Online-Zeitschrift zur verbalen Interaktion 10, 353–402.Search in Google Scholar

Stivers, Tanya and Makoto Hayashi. 2010. “Transformative answers: One way to resist a question’s constraints.” Language in Society 39, 1–25.Search in Google Scholar

Stivers, Tanya, Lorenza Mondada and Jakob Steensig, eds. 2011. The morality of knowledge in conversation. Cambridge: Cambridge University Press.Search in Google Scholar

Szczepek Reed, Beatrice. 2015. “Managing the boundary between ‘yes’ and ‘but’: Two ways of disaffiliating with German ja aber and jaber.” Research on Language and Social Interaction 48(1), 32–57.Search in Google Scholar

Thompson, Sandra A., Barbara A. Fox, and Elizabeth Couper-Kuhlen. 2015. Grammar in everyday talk: Building responsive actions. Cambridge: Cambridge University Press.Search in Google Scholar

Received: 2020-08-23
Revised: 2021-01-07
Accepted: 2021-04-09
Published Online: 2021-12-07

© 2021 Elizabeth Couper-Kuhlen, published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.