Asking the Obvious: Other-Repeats as Requests for Reconfirmation

Co-participants may reproduce (part of) a prior speaker’s turn to various interactional ends. The focus here is on a specific subclass of sequence-initiating other-repeats, namely second sayings asking the original speaker to reconfirm what they just said. Such reconfirmation-seeking other-repeats have hardly been studied in their own right but typically either been treated in studies on repair initiations or ‘newsmarks’. In approaching these ostensibly different types of reconfirmation-seeking other-repeats collectively, this interactional-linguistic study reveals that the distinction between problem-indicating – and thus, in a broad sense, repair-implicated – and newsmarking other-repeats is gradual rather than binary. Drawing on video recordings of informal German and English face-to-face conversations as data, it further shows that co-occurring multimodal marking may contribute to enhancing action recogni-sability and displaying (dis)affiliation and that the study’s findings may be relevant across languages.


Introduction
Other-repeats -that is, full or partial reproductions of a co-participant's prior turn -are a common phenomenon in everyday social interaction.In previous conversation-analytic (CA) and interactional-linguistic (IL) research, the interactional deployment of such second sayings has been found to vary depending on their position, their prosodic design, and the overall sequential environment (e.g., Betz et al., 2013: 135-136).Based on the literature, we can broadly distinguish between other-repeats which initiate a 'retro-sequence' (Schegloff, 2007: 217-219) from second position, other-repeats which respond to a sequence initiation, and other-repeats which receive and register a response (e.g., Schegloff, 1996: 177-178;Rossi, 2020: 496).While responsive other-repeats as well as repetitional response receipts need not be taken up by the co-participants and may thus entail sequence closure, sequence-initiating other-repeats seek a response in that they reissue (part of) a prior speaker's turn for (re)confirmation, clarification, or completion (e.g., Couper-Kuhlen, 2020;Rossi, 2020).The focus of this paper will be on one of the first types of sequence-initiating other-repeats, namely repeat-format requests for reconfirmation (RfRCs).The following extract from a conversation between three friends at breakfast aptly illustrates the phenomenon of interest.The original saying is marked with a double arrow (»), the repeat with a single arrow (→) and bold font.Note that next to the repeat-format RfRC in line 16, which is focal here (see section 4.1 below for a full analysis of this instance), the extract also features a responsive repeat in line 17.
Extract 1: BADELATSCHEN (LoE_VG_02, 39:00-39:27 min) As part of a humorous telling about a series of unfortunate events at a recent party, Lori (Lor) states that she ended up walking home in her friend's flipflops that evening (line 05).After a side sequence, which is omitted here for the sake of brevity, Bastian (Bas) repeats Lori's assertion in a slightly modified form and with rising final intonation (line 16), thereby prompting her to reconfirm her previous statement with a head nod and a modified repeat in line 17.While RfRCs, like the sequence-initiating other-repeat in line 16 of Extract 1 above, and related phenomena, such as so-called 'newsmarks' (e.g., Jefferson, 1981;Heritage, 1984) and certain other-initiations of repair (OIR s) (e.g., Benjamin and Walker, 2013;Robinson, 2013;Selting, 1996; see section 2), have often been viewed as a subtype of requests for confirmation (RfCs), or polar questions more generally (e.g., Raymond and Stivers, 2016;Steensig and Heinemann, 2013;Stivers and Enfield, 2010), there are clear sequential grounds for treating RfCs and RfRCs as distinct phenomena (see also Gipper et al., this issue).In fact, a more fine-grained differentiation between various yes/no-type initiating actions, including Rf(R)Cs, requests for information, and requests for affirmation, has already been called for elsewhere (e.g., König and Pfeiffer, 2019, this issue).In the light of the aforesaid, the aim of this paper is twofold: First, it makes a case for differentiating RfRCs from RfCs by pointing out similarities and differences between these devices (section 2), and second, it illustrates the different interactional uses of RfRCs in German and English conversation on the basis of one particular format that may be employed to request reconfirmation, namely other-repeats (section 4).In doing so, it exposes striking similarities between the two languages under investigation rather than pointing to possible cross-linguistic differences.The data and methods used for this study will be outlined in section 3. The paper will be rounded off by a discussion of the main findings and implications for future research in section 5.

Requests for (Re)Confirmation
As indicated above, RfCs and RfRCs are commonly lumped together in largescale studies on polar questions (e.g., Stivers and Enfield, 2010) and OIR s (e.g., Dingemanse et al., 2014).In fact, no systematic distinction is being made between 'confirmation' and 'reconfirmation' in the CA/IL literature (but see Heritage, 1984;Seuren et al., 2018;Sørensen, 2021).These two responsive actions certainly share a number of features: They both presuppose a speakertilted epistemic asymmetry (e.g., Barnes, 2012: 246;Küttner, 2016: 54)  typically even performed by means of the same interactional resources, namely response tokens like yeah and uh_huh as well as repeats and head nods (e.g., Betz, 2015: 252;Rossi, 2020: 507; see also Deppermann et al., in prep.).Yet they differ in terms of their sequential placement: Confirmation is generally due after B-event statements, or questions that introduce a proposition about a state of affairs pertaining to the confirming party's experience, personal history, emotions, or expertise (Labov and Fanshel, 1977: 100;also Deppermann, 2018: 120).The confirming party is thus concerned with assuring the validity or accuracy of a statement made by a co-participant, also in the context of candidate repairs (e.g., Dingemanse et al., 2014: 26).Reconfirmation, on the other hand, is brought about by questioning turns that follow up on A-event statements (Labov and Fanshel, 1977) by simply returning them to their original speakers without adding any propositional content.The reconfirming party is thus concerned with assuring the validity or accuracy of their own statement made in prior talk.1 To illustrate this difference, let us contrast the RfRC in Extract 1 above with the RfC starting in line 04 of Extract 2 below, which is taken from a conversation between three flatmates at breakfast.Franz (Fra) is in the process of telling the others how he walked a long distance through Berlin the previous night while trying to come up with an alternative evening program after his original plan of attending a concert had failed.Just before this extract, Franz reported how he and Hajo (Haj) finally agreed on meeting at the event location "Kabine" in Berlin-Friedrichshain with some other friends, which is about four kilometres away from the last location Franz mentioned in his telling, namely Greifwalder Straße in Berlin-Prenzlauer Berg.
Extract 2: ZUR KABINE (LoE_VG_03, 10:19-10:50 min) 1 This includes both the de-facto and the pro-forma assurance of validity, since RfRCs may display genuine as well as mock disbelief (see also Thompson et al., 2015: 76).In this context, it can also be noted that while RfRCs generally invite reconfirmation, they exhibit varying degrees of 'sequential implicativeness' (Schegloff and Sacks, 1973: 296) and may not always be followed by explicit reconfirmation (e.g., Gubina and Betz, 2021;Kaimaki, 2011aKaimaki, , 2011b; see also Gipper et al., this issue; Marmorstein and Szczepek Reed, this issue).As there are no instances of RfRCs without reconfirmation in the current collection (see section 3), suffice it to say that there appear to be different factors at play in such cases, such as the doublebarrelledness of the RfRC, the lack of response mobilising features, or the resistance of the RfRC recipient regarding the pursued course of action.(Car), who first displays a critical stance towards the reported endeavour by shaking her head, waving her hand in front of her face with her palm facing in, and clicking (lines 03-06) and then tries to mark out the final part of Franz' walking tour on an imaginary map (lines 07-13), Hajo initiates repair with an ach-prefaced candidate understanding in lines 04 and 05, which he further expands in line 08, replacing the vague demonstrative da 'there' with a place name to specify his initial place reference (e.g., Enfield and Roque, 2017: 583).
Hajo thus requests Franz to (dis)confirm the proposition that he walked all the way to the venue kabine as well.Since Franz only stated that he kept on walking without explicating up to where, Hajo's proposition comes across as an inference put up for validation by Franz, who has epistemic authority over the matter at hand.In line 11 and following, Franz, after some delay, partially confirms and partially disconfirms Hajo's candidate understanding with a transformative answer (Stivers and Hayashi, 2010), clarifying that he did cover part of the distance by foot, but also used the tram as an additional means of transport.
As we have seen, RfCs and RfRCs relate back to prior talk in different ways and, by implication, seek different responses: While RfCs introduce propositions or interpretations of their own and put them up for confirmation by a more knowledgeable [K+] (Heritage, 2012) co-participant (see Extract 2, lines 04ff.),RfRCs simply take up and return others' propositions for reconfirmation, asking about something that is supposedly obvious from the surface of the conversation (see Extract 1, line 16).Accordingly, the sequential circumstances and implications of RfCs and RfRCs differ from each other quite considerably.Yet it should also be noted that neither RfCs nor RfRCs form a homogeneous class of utterances.In fact, Rf(R)Cs may be vehicles for various actions and therefore make relevant additional response components next to the (re)confirmation as such (e.g., König and Pfeiffer, 2019, this issue).On these grounds, they are best conceived of as 'base-level' or 'basic' actions that may implement different 'additional' actions depending on the specific sequential contexts in which they occur (see Stivers et al., 2023 on the multi-layeredness of actions).
Perhaps, it is for this very reason that the phenomenon of interest, RfRCs, has hardly been studied in its own right but only been treated in the context of studies on more specific interactional practices and devices.These include, on the one hand, studies on newsmarks (e.g., Jefferson, 1981;Heritage, 1984;Local, 1996;Maynard, 1997Maynard, , 2003;;Kaimaki, 2011aKaimaki, , 2011b; see also Marmorstein and Szczepek Reed, this issue), topicalising responses (e.g., Button andCasey, 1984, 1985;Schegloff, 2007: 155-158), responses to informings more generally (e.g., Imo, 2009Imo, , 2011;;Thompson et al., 2015), and displays of surprise and related Downloaded from Brill.com09/23/2023 03:50:17AM via free access affective stances (e.g., Svennevig, 2004;Wilkinson and Kitzinger, 2006; see also Robinson, 2013), which principally describe specific subsets of RfRCs as "assertions of ritualized disbelief" (Heritage, 1984: 339) that constitute aligning and/ or affiliative responses to deliveries of new information.And on the other hand, certain types of RfRCs have been discussed in studies on repair initiations, in a broad sense, where they have been treated as disaligning and/or disaffiliative moves addressing ostensible problems of hearing (e.g., Dingemanse et al., 2014: 23-26;Dingemanse and Enfield, 2015: 100f.), expectation (e.g., Selting, 1996) or acceptability (e.g., Svennevig, 2008;Robinson, 2013;Benjamin and Walker, 2013).A closer look at these various objects of investigation reveals a substantial overlap with regard to turn formats, which, among others, include repeats, minimal tokens (e.g., really), and minimal clauses (e.g., did you) (e.g., Heritage, 1984: 339;Schegloff, 2007: 155;Thompson et al., 2015: 57, 59; see also Gipper et al., this issue), as well as sequential implications.As a matter of fact, it has previously been noted that the boundaries between these ostensibly distinct phenomena are not always clear (Thompson et al., 2015: 60-64, see also Couper-Kuhlen, 2020: 536;Dingemanse and Enfield, 2015: 100f.;Kendrick, 2015: 181-187;Rossi, 2020: 512;Wilkinson and Kitzinger, 2006: 170).There are even cases in which the exact status of a reconfirmation-seeking turn remains obscure, or at least negotiable, altogether (e.g., Thompson et al., 2015: 64, 98f.).Therefore, approaching RfRCs collectively, without focussing on a specific function (e.g., marking prior talk as 'news' , 'surprising' , or 'unacceptable') or sequence type (e.g., informing or OIR sequences) from the outset, may help to arrive at a more comprehensive picture of such turns and their interactional import (see also Gubina and Betz, 2021, whose focus on German echt as one particular newsmark-type response token similarly allows for cross-categorical observations).As will be demonstrated in section 4 below, the findings of this study strongly support Thompson et al.'s (2015: 61) idea of a continuum ranging from turns which bear little to no resemblance to OIR s, in that they merely treat a prior statement as news-or noteworthy, to turns that are akin to OIR s, in that they treat a prior statement as in some way problematic.2This notion of a continuum of uses is developed further in the present study and shown to be of relevance across the two languages under investigation.In this way, the study also complements previous cross-linguistic research on OIR and other-repetition in CA/IL, which typically includes English but not German and hardly presents qualitative findings for individual languages in direct comparison (e.g., Dingemanse and Enfield, 2015;Rossi, 2020).

Data and Method
As part of ongoing PhD research, this paper reports some preliminary results on the interactional uses of repeat-format RfRCs in German and English conversation.The data used for this study comprise video recordings of approximately nine hours of German and eight hours of English face-to-face interactions between friends and family members.The English data involve native speakers of North American English and are taken from the Arizona Tapes, which were recorded under the direction of Elizabeth Couper-Kuhlen, and the Languages of Emotion (LoE) corpus, which was compiled under the direction of Elizabeth Couper-Kuhlen and Margret Selting, both for the project Emotive involvement in conversational storytelling as part of the cluster of excellence Languages of Emotion at the Free University Berlin, as well as my own supplementary recordings.The German data involve native speakers of (German Standard) German and are exclusively taken from the LoE corpus.
In a first step, relevant instances of other-repeats, that is, other-repeats clearly designed and treated as asking the original speaker to reconfirm their original saying, were extracted from the data.Other types of sequence-initiating other-repeats, i.e., other-repeats seeking completion or clarification (see section 1) as well as other-repeats clearly designed and treated as proffering a candidate hearing (see section 2), were not included,3 neither were registering other-repeats, i.e., other-repeats that merely acknowledge the original saying without explicitly inviting a response (see section 1).For the purpose of this study, other-repeats were defined as full or partial reproductions of a prior speaker's turn with what the repeating speaker appears to intend as virtually identical wording (see also Rossi, 2020), meaning that only repeats with no or minor modifications, such as deictic shifts, pronominalisations, and changes in word order, were included.4For the time being, I excluded repeats involving additions of other lexical material, such as prefaces (e.g., really, what) and Contrastive PragmaticS (2023) 1-33 | 10.1163/26660393-bja10088 adverbials, to get a clear understanding of the interactional work done by the repeat as such.This procedure yielded 30 instances of repeat-format RfRCs in total, 16 in the German and 14 in the English dataset.Relevant data extracts were transcribed according to the GAT 2 transcription conventions (Selting et al., 2009;Couper-Kuhlen and Barth-Weingarten, 2011), supplemented by notations of weak cesuras and cesural areas (Barth-Weingarten, 2016) where deemed necessary, and on the focal lines, I added information on the participants' embodied conduct using Mondada's conventions for multimodal transcription (Mondada, 2019).Drawing on the theoretical and methodological framework of IL (e.g., Couper-Kuhlen and Selting, 2018), all cases were then submitted to close sequential and multimodal analysis in order to determine which actions the other-repeats were recognisably accomplishing.

Repeat-Format RfRCs in German and English Conversation
In the following, we will consider the different interactional uses of RfRCs in the form of other-repeats in detail.As suggested above (see section 2), there is a continuum between 'newsmark-like' repeat-format RfRCs, which treat the original saying as news-or noteworthy and will be referred to as 'newsmarking other-repeats' in the following, and, in a broad sense, 'repair-related' repeat-format RfRCs, which treat the original saying as counter to expectation or difficult to accept and will be referred to as 'problem-indicating otherrepeats' in the following.While different sets of RfRCs and related practices have been referred to as OIR s used as "vehicles for other actions" or as "pseudo OIR s" elsewhere (Kendrick, 2015), I am purposely refraining from using the label 'repair initiation' here, since there are different understandings of what constitutes a proper instance of OIR (e.g., Rossi, 2020: 512;Thompson et al., 2015: 61) and conceptualising the instances at hand in repair terms does not seem to add any analytical value.Instead, I will show how the base-level action 'RfRC' , which involves the preliminary receipt and inquiry about some bit of previously unknown information, routinely takes on one of the following two additional action layers: marking the news-or noteworthiness of the received information, with different degrees of affective involvement, or marking its problematicity, with different degrees of problem severity.I will also show instances of repeat-format RfRCs that are underspecified in terms of their placement and design and thus fall somewhere in between the clearly newsmarking and problem-indicating other-repeats.This section will be structured by reference to the continuum of RfRC uses outlined above, in that the first two subsections will be devoted to the two endpoints of the continuum, that is, newsmarking other-repeats (see section 4.1) and problem-indicating other-repeats (see section 4.2), and the third subsection will address intermediary cases (see section 4.3).I will be presenting one German and one English case per section to demonstrate that repeat-format RfRCs may overall be employed and designed comparably in these two languages, which also suggests that the continuum of RfRC uses may be of cross-linguistic relevance.5

4.1
Newsmarking Other-Repeats Newsmarking other-repeats ordinarily occur after news announcements or informings, also in the context of tellings, and engender further, often affiliative engagement with the information just imparted.They are thus not only followed by reconfirmation on the part of the original speaker, but typically entail at least some form of assessment of the matter in question by either of the participants, often in combination with elaboration by the producer of the original saying.6If elaboration is not volunteered by the latter, it may be pursued further through follow-up questions on the part of the repeating speaker.However, if the original saying occurs in the midst of a telling, newsmarking other-repeats usually only lead to brief side sequences (Jefferson, 1972) after which the telling activity is continued.
To illustrate this RfRC use, we will first reconsider the example from the introduction, which is provided here again as Extract 3 in an expanded version.Recall that the conversation is revolving around a recent party -an outdoor party with a rented hot tub, to be precise.Said party was organized by Bastian's (Bas) flatmate Ben and attended by all three participants.Lori (Lor) is in the middle of telling the others how that evening came to a rather unpleasant end for her, which she, however, frames as total witzig 'really funny' from the outset (data not shown).Note that she had put some of her belongings right next to the hot tub, which ended up overflowing.Also note that the party took place at cold outside temperatures.At the beginning of Extract 3, Lori informs the others that some of her things, including her shoes, had become wet at the party (lines 01-04) and that she ended up walking home in her friend's flipflops (line 05) as part of a humorous telling about fortune in misfortune.The projected continuation of her telling (line 06) is suspended by a side sequence, in which Bastian checks an inference about an unknown pair of shoes which was apparently left behind at their shared flat after the party (lines 07-15).The details of this side sequence need not concern us further here.What matters is that afterwards, Bastian resumes Lori's telling by means of a clausal other-repeat (line 16), treating her prior informing as noteworthy and asking her to verify it once more.7 Through its multimodal design with the extra strong focus accent and the combination of head poke and grin (see figures 1-2), Bastian's repeat comes off as signalling interest in a particular story detail and endorsing Lori's humorous stance rather than as addressing a possible problem.And indeed, Lori treats Bastian's repeat as a welcome invitation to reconfirm (line 17) and elaborate on the consequences of wearing inapt footgear (line 20f.), before she goes on with the next story detail as projected in line 06 (not shown).The affiliative character of this reconfirmation sequence is also underlined by the fact that Lori's talk in lines 17ff. is infused with laughter and smile voice and further evidenced by Bastian's affective stance display in line 19 and the ensuing laughter (line 22).Extract 4 is another case in point.It is taken from a conversation between a married couple, Jason (Jay) and Kelsey (Kel), and their friend and host Drew (Dre).Jay and Kelsey have been talking (bad) about different former neighbours of theirs.Kelsey just complained about the family next door, whose dog used to jump over their fence and break things, and who once even dared to ask them to pay for the damage.In the following, Jay complains about the lady next door more generally.
Extract 4: BAD PARENT (LoE_VG_a, 01:09:33-01:09:54 h) other-repeats, however), a different RfRC format, such as a freestanding minimal token, would not work in this context (see also Thompson et al., 2015: 121f.).In line 02, Jay projects an account for why he had a low opinion of that neighbour beyond the dog-related issues.As he only gets to say much later, the mother of three would let her young children roam around the neighbourhood all by themselves (data not shown).Jay sets the scene by providing some general, mostly negative characterisations of the lady next door (lines 01 and 04-06) and goes on to expose her arguably objectionable religious affiliation (line 10).He does so in interrogative form, asking his wife to confirm his claim, which she does with a particle-prefaced repeat in line 11 that is noticeably lengthened on the final consonant.Drew picks up on this religious ascription with his highrising clausal-other repeat in line 12, which is produced with a wide pitch span and a smiley facial expression (see figure 3), treating this information as noteworthy while inquiring into its validity.In opening up a side sequence, the RfRC creates a sequential slot for intermediate affiliation with Jay's negative stance towards the neighbour.However, perhaps due to the potential delicateness of religious matters, none of the participants openly discredits Mormonism in the following.What we get instead is two reconfirmations by Kelsey and Jay in lines 13 and 15, which are slightly marked through lengthening, and a visual stance display on Drew's part in line 17 (see figure 4), implying a negative, or at least sceptical attitude towards the respective religious group and its members.While Jay then sets out to continue his report (lines 16 and 18), Kelsey does some additional work to pre-empt a possible misreference (lines 19 and 23), explicating that they are now talking about the family of five with one dog and no longer about the childless couple with two dogs that they had been talking about earlier.5), who can be seen to be jumping ahead of his story by calling attention to the children before the time.After providing a synoptic assessment, which he reiterates thrice (lines 22-25), Jay then begins to tell everything in order (lines 26ff).In comparison to Extract 3, the affiliative quality of the reconfirmation sequence in this extract is slightly less apparent due to the more subtle stancetaking.
Overall, the sequential contexts in which newsmarking other-repeats tend to occur are fairly restricted, but there is great variability in terms of their lexicosyntactic, prosodic-phonetic, and bodily-visual design.The only consistency is that all newsmarking other-repeats in the current collection are produced with rising final intonation and that the facial expression and voice quality of the repeating speaker are neutral to smiley,8 but neither of these individual design features is restricted to this usage type.Nonetheless, it appears that there are certain clusters of features that are more closely associated with newsmarking than with problem-indicating uses, particularly combinations of eyebrow raises or head pokes forward with smiles in response to positively valenced information.Apart from that, noticeable aspects about the prosodic-phonetic configuration of and the embodied conduct around repeat-format RfRCs seem to function as additional markers of (dis)affiliation across the different usage types.With regard to newsmarking other-repeats, their realisation with prosodic upgrading (e.g., strong accentuation, wide pitch span) and/or additional bodily-visual marking typically adds to their affiliative reception, whereas their realisation with prosodic downgrading (e.g., flat pitch contour, low volume) and the absence of additional bodily-visual marking can make them come off as purely aligning (in contexts where some type of information receipt is expectable) or even blur the line between newsmarking and problem-indicating other-repeats altogether (in contexts where the other-repeat opens up a side sequence to an ongoing activity).

4.2
Problem-Indicating Other-Repeats Compared to the newsmarking other-repeats discussed in the previous section, problem-indicating other-repeats are the more frequent usage type of repeat-format RfRCs in the current collection (see also Gipper et al., this issue) and can, at least in principle, occur after any type of assertion made from a [K+] position.Still, they are mostly found in the vicinity of informings.Since they treat (part of) a prior utterance as in some way problematic -be it contrary to expectation, possibly faulty, or unacceptable -, they are often followed by explanations, accounts, or justifications in addition to the reconfirmation proper.On occasion, the original speaker even self-corrects or backs down from their original assertion, thus we may also get disconfirmation or qualified reconfirmation in response to a problem-indicating other-repeat.9Depending on whether such repeats address problems of expectation or problems of acceptability and whether they are designed to display higher or lower degrees of disaffiliation, they lead to more or less pronounced sequence suspension.
Let us begin with a problem-indicating other-repeat addressing a problem of expectation in a fairly unmarked form.Extract 5 is taken from the very beginning of a conversation between two sisters Sandra (San) and Emma (Emm).The recording researchers Moni (Mon) and Michael (Mic) are still in the room and in the process of announcing what will happen once they finish setting up the recording equipment.
Extract 5: PANCAKES (LoE_VG_04, 00:11-00:16 min) 9 This especially holds true for problem-indicating other-repeats addressing problems of acceptability which turn out to be justified and may preempt more dispreferred courses of action involving explicit disagreement (see also Schegloff, 2007 on pre-disagreements and disagreement-implicated other-initiated repair).In Extract 5, the recording researchers are jointly informing Sandra and Emma that they will be leaving soon and ring the doorbell upon their return in an hour's time (lines 01-02).While Sandra receipts this information by claiming mutual epistemic access with her reduplicated genau in lines 03 and 04 (Betz, 2014, as cited in Oloff, 2017: 210f.),Emma treats the time reference as problematic with her intended repeat in line 05, asking for reconfirmation of this bit of information while indexing a clash of expectations.This is not only suggested by the sequential and epistemic context in which this partial repeat occurs, that is, after an A-event statement imparting information in a matter-of-fact manner, but also made clear by its multimodal design with the minor head poke in combination with slightly lowered eyebrows (see figures 6-7) and the high-rising final intonation.Emma's problem of expectation turns out to be justified, inasmuch as she actually misheard the trouble source.Her purported repeat is therefore not reconfirmed but treated as correctable by the recording researchers (lines 06-07), which Emma acknowledges with a change-of-state token (line 08) and a positive assessment (line 09).
An instance of a more marked and, by implication, more disaffiliative problem-indicating other-repeat, which is followed by another, slightly downgraded repeat of the same trouble source only moments later, can be found in Extract 6.This extract is taken from a dinner-table conversation between five friends at a restaurant.The conversation is revolving around Asian eating habits, including their full utilisation of animals such as cows, pigs, or even cats.In the context of a discussion about Asian eating habits, Lucy presents her boyfriend Timmy with a 'fait accompli' proposal, stating that they will consume horse meat during their upcoming trip to Japan (line 01).After a display of indifference by means of shrugging on Timmy's part (line 02), Lucy adds information on the mode of consumption, which is raw for the country-specific dish she seems to have in mind (line 03).Timmy receives this latter information with a phrasal other-repeat (line 05), which is marked by a striking rise-fall contour, lengthening, and frowning and/or nose wrinkling,10 as well as a slight head poke forward (see figures 8-9).By requesting reconfirmation in a sequential slot where an acceptance (or rejection) would be relevant next, Timmy does not only implicitly take issue with Lucy's proposal but also cast doubt on the information provided, namely that eating raw horse meat is a valid option.In response, Lucy stands her ground by simply reconfirming her prior assertion without further substantiation (line 06), as John (Joh), who is a Japanese expert as well, comes in to further underscore the adequacy of her proposal with a third-party confirmation (line 07).Still, Timmy continues to treat the idea of eating raw horse meat as problematic by producing another other-repeat, now directed at John (line 09).In contrast to the first repeat in line 05, however, this second repeat is less marked in terms of both prosody and bodily-visual conduct: The rising pitch contour is relatively flat, Timmy rubs his mouth abashedly, and even though his eyebrows are still lowered, he is no longer frowning (see figure 10).In this way, the repeat in line 09 comes off as merely indexing divergent expectations rather than genuine disbelief.After John's implicit reconfirmation in line 11, Timmy produces yet another RfRC in the form of the minimal token really (line 14) with an even less pronounced pitch contour.Thus, as the sequence unfolds, Lucy is clearly positioned as more knowledgeable regarding Japanese cuisine, whereas Timmy backs down from his sceptical stance step by step and ends up justifying (line 17) and explaining himself (line 27) rather than further questioning the acceptability of Lucy's proposal.Unlike newsmarking other-repeats, problem-indicating other-repeats, akin to OIR s in the narrower sense, are more flexible in terms of the sequential contexts in which they occur but similarly variable in design.Again, the majority of cases are produced with rising final intonation, often with contours rising from low (see also Couper-Kuhlen, 2020), though there are cases aldrup 10.1163/26660393-bja10088 | Contrastive PragmaticS (2023) 1-33 with falling final intonation as well.In contrast to newsmarking other-repeats, eyebrow raises and head pokes are typically not accompanied by smiles with the problem-indicating other-repeats.What is more, there are a number of features found with the problem-indicating other-repeats that are not attested with the newsmarking other-repeats, at least not in the current collection.For instance, problem-indicating other-repeats are regularly accompanied by lowered eyebrows or frowns, sometimes also squints or nose wrinkles as well as tense or whispery voice, and while head pokes or torso movements towards the original speaker are predominant, repeating speakers may also back away instead.Apart from that, noticeable prosodic-phonetic and bodilyvisual features of turn design can, as pointed out above, be used as additional markers of (dis)affiliation across the different usage types.However, with problem-indicating other-repeats, it is prosodic upgrading and the presence of additional bodily-visual markers that can make them come across as highly disaffiliative, whereas prosodic downgrading and the absence of additional bodily-visual marking may contribute to mitigating their disaffiliative force and, for instance, make an other-repeat that questions the acceptability of what was said come off as a mild challenge.

4.3
Intermediary Cases While most other-repeats in the current collection agglomerate near either pole of the continuum, i.e., the 'newsmarking' pole (see section 4.1) or the 'problem-indicating' pole (see section 4.2), there are a handful of cases that fall in between these two extremes and, by all appearances, do nothing more than tentatively receiving and inquiring about the information just imparted.With these intermediary cases -which, like the other RfRCs, may in principle follow up on any type of 'informing' , be it question-elicited or volunteered -, it is difficult to tell on sequential and compositional grounds whether they treat the original saying as news-or noteworthy or as problematic, though they may slightly tend towards one of the poles.We will first consider an intermediary case in which the repeat-format RfRC may be seen to have a topicalising function and thus rather lean towards the 'newsmarking' pole.Then, we will examine an intermediary case which is more prone to be taken as a problemindicating other-repeat, in that it may be seen to seek assurance in connection with an uncertainty after an implicit other-correction.
Extract 7 is taken from a conversation between a newly-wed couple, Norbert (Nor) and Eleana (Ele).The two seem to be running out of topics here and there, which is why there are extensive lapses in between.The following extract sets in after such a lapse with a duration of around 15 seconds.and 16), readily going along with the teasing frame.Unlike the cases discussed so far, the other-repeat in line 08 neither comes off as signalling genuine interest in Eleana's relatively inconclusive response nor as taking any issue with it.
In its local sequential context, this preliminary information receipt can rather be construed as a fairly non-committal way of further topicalising the motives for Eleana's laughter, which is also suggested by Norbert's additional attempts at pursuing the topic.At a point where the conversation is at risk of coming to a standstill, the ambiguity of Norbert's uptake in line 08 may serve as a resource for keeping the talk going while leaving its exact direction open to negotiation.In this particular case, it sets off a brief teasing sequence.
Let us turn to another intermediary case which rather leans towards the 'problem-indicating' pole, however.Extract 8 is taken from the same interaction as Extract 4 above.The conversation is revolving around Drew's (Dre) teaching at different universities, including Polytechnic University in New York.

Discussion
The aim of this study was to show the relevance of differentiating RfRCs from RfCs (see section 2), and to explore how RfRCs in the form of other-repeats are employed in German and English conversation (see section 4).Across the two languages under investigation, repeat-format RfRCs turned out to fall on a continuum between 'newsmarking' and 'problem-indicating' turns with a rather seamless transition between these ostensibly distinct uses (see figure 12) -an aspect that is easily overlooked if such utterances are not studied in direct juxtaposition.This continuum emanates from the core action of RfRCs, which is shared across all uses, namely the preliminary receipt and inquiry about a piece of previously unknown information.
This study thus exposed that even though 'newsmark-like' and 'repairrelated' RfRCs have often been treated separately in previous research, there is no clear-cut boundary between them.This finding is in line with Thompson et al.'s (2015) research on responses to informings and expands their results by including RfRCs outside informing sequences and by taking a cross-linguistic and strictly multimodal approach.In this way, the present study offers a more systematic account of other-repeats that go 'beyond initiating repair' (e.g., Rossi, 2020: 512-514) than previous studies, among others by presenting empirical evidence for how newsmarking displays of surprise or interest can be differentiated from problem-indicating displays of divergent expectations, both of which have been treated collectively as indexing unexpectedness or (mock) disbelief on occasion (cf.Robinson, 2013).What is more, the study shows that restricting one's focus to either 'newsmarks' or repair initiations from the outset may lead to unwarrantedly simplified descriptions of participants' choices in responding to assertions imparting new information.For instance, the ambiguity of intermediary cases that are neither clearly newsmarking nor Making a binary distinction between 'newsmark-like' and 'repair-related' other-repeats entails the risk of missing out on such nuances that may be relevant for action formation and ascription as well.A collective approach to RfRCs thus seems beneficial, in that it allows us to carefully tease apart the different uses of these formally and functionally closely related turns-at-talk in their local sequential contexts.
Since newsmarking and problem-indicating other-repeats may occur in similar sequential environments, it is all the more important to understand which resources participants can draw on to accurately ascribe an action to a repeat-format RfRC and to respond accordingly.Through detailed multimodal analyses, this study has shown that while there appears to be no clear association between an other-repeat's interactional use and individual turn design features, it is possible to identify certain clusters of features that tend to be found with newsmarking rather than problem-indicating other-repeats and vice versa.Especially in contexts in which the sequential position of the otherrepeat does not suggest a particular reading, aspects of turn design may help to disambiguate between newsmarking and problem-indicating uses (see also Stivers et al., 2023): Other-repeats that are accompanied by eyebrow raises and head or body movements towards the original speaker with a smiling facial expression are inclined to be taken as 'newsmark-like' , whereas other-repeats that are produced with a non-smiling facial expression in combination with raised or lowered eyebrows and head or body movements towards or away from the original speaker, as well as frowns, squints, or nose wrinkles, are likely to be treated as problem-indicating (see also Kendrick, 2015: 178).Likewise, smile voice seems to be clearly associated with 'newsmark-like' uses, whereas tense or whispery voice as well as falling final pitch or pitch contours rising from low may be indicative of problems of expectation or acceptability (see figure 12).Intermediary cases, on the other hand, are often characterised by the absence of additional prosodic-phonetic and bodily-visual marking, leaving some room for interpretation.In such cases, the possibility of initially responding with a plain reconfirmation, allowing for a turn-by-turn negotiation of the further course of action, may come in useful.
Apart from reducing ambiguity in action ascription, noticeable modulations of an other-repeat's prosodic-phonetic design or the repeating speaker's embodied conduct can be shown to take on additional functions, particularly in cases in which the action underway is already disambiguated by the sequential context.In such cases, multimodal marking is commonly used to upgrade the other-repeat's (dis)affiliative implications (see also Thompson et al., 2015: 135-137, 283, 290): With newsmarking other-repeats, co-occurring prosodic-phonetic and bodily-visual marking typically implies a higher degree of affiliation, while with problem-indicating other-repeats, co-occurring prosodic-phonetic and bodily-visual marking typically implies a higher degree of disaffiliation.
Importantly, the above observations about the continuum of RfRC uses and the multimodal design features of repeat-format RfRCs seem to hold for German and English alike.Future research based on larger case numbers and incorporating other, unrelated languages as well as other RfRC formats is needed to shed further light on the possible cross-linguistic relevance of these findings (see also Gipper et al., this issue) and hoped to provide additional insights into the paradigm of choices available to participants when responding to new information in everyday talk-in-interaction.

Figure 12
Figure 12 Continuum of RfRC uses with localisation of Extracts ❸ to ❽ (line 21)and immediately orients back to Jay (see gaze shift in line 21), thereby rendering her further efforts at explication dispensable.In overlap, Jay reinforces his claim to the floor with slightly increased volume and an embodied halt signal directed as Kelsey (see figure Drew only acknowledges Kelsey's first explication about the Downloaded from Brill.com09/23/2023 03:50:17AM via free access aldrup 10.1163/26660393-bja10088 | Contrastive PragmaticS (2023) 1-33 number of children As pointed out above, Extract 7 sets in after some awkward silence.Following prolonged mutual gaze and some self-conscious laughter (lines 01-04), Norbert inquires into the reason for the laughter on Eleana's part (line 05), who claims that there is none (line 07).Norbert picks up on Eleana's smile-infused, fairly non-informative response with a phrasal other-repeat in line 08, which does not exhibit any noticeable prosodic-phonetic or bodily-visual features: It is produced faster than Eleana's original saying with a narrow pitch span and lax articulation, a slightly amused facial expression (see figure11), and eyebrows in home position.Eleana, who continues to smile throughout, responds by simply repeating her original statement (line 09).After a gap (line 10), Norbert goes on to teasingly make his own assumptions about the matter by formulating two B-event statements (lines 11 and 14), which Eleana confirms smilingly (lines 12 due to the overlap in lines 03-04 cannot be ruled out, there is no clear orientation to an actual problem of hearing on part of the co-participants: The other-repeat in line 06 is neither presented as an approximation, e.g., through a hesitant or particularly careful articulation, nor does its recipient see the need to repeat the possible trouble source once again in response.Still, this repeat-format RfRC likely addresses some sort of problem that may not be immediately apparent, thus bearing resemblance to the 'problem-indicating' other-repeats.Its unmarked design suggests that it is at best dealing with a minor problem of certainty, however, and leaves open whether Kelsey takes issue with Drew's corrective response in any other way. Downloaded from Brill.com09/23/2023 03:50:17AM via free access aldrup 10.1163/26660393-bja10088 | Contrastive PragmaticS (2023) 1-33 issues