Identifying and addressing equivocal trouble in understanding within classroom interaction

Maintaining intersubjectivity is crucial for accomplishing coordinated social action. Although conversational repair is a recognised defence of intersubjectivity and routinely used to address ostensible sources of trouble in social interaction, it is less clear how people address more equivocal trouble. This study uses conversation analysis to examine preschool classroom interaction, focusing on practices used to identify and address such trouble. Repair is found to be a recurrent frontline practice for addressing equivocal trouble, occasioning space for further information that might enable identifying a specific trouble source. Where further information is forthcoming, a range of strategies are subsequently employed to address the trouble. Where this is not possible or does not succeed, a secondary option is to progress a broader activity-in-progress. This allows for the possibility of another opportunity to identify and address the trouble. Given that misunderstandings can jeopardise interactants’ ability to mutually accomplish courses of action, these practices defend intersubjectivity against the threat of equivocal trouble.


Introduction
Shared understanding is a mundane, routine and largely unnoticed social accomplishment. This intersubjectivity enables people to transcend their private perspectives and collaborate on mutually recognisable courses of action (Heritage, 1984). Through detailed investigation of conversational repair, research undertaken within conversation analysis has demonstrated a unique capacity to generate an empirical account of the practices people use to maintain intersubjectivity (Heritage, 1984;Schegloff, 1992Schegloff, , 2006. In contrast to this existing research, however, which has largely focused on ostensible threats to intersubjectivity, this article illustrates that identifying such threats is not always straightforward. Nevertheless, using the method of conversation analytic research, we show how a recipient to a potential trouble in understanding responds in systematic ways to identify the nature of the problem and to address it accordingly. A foundational method that people use to establish and maintain intersubjectivity is through contributions to their interactions with one another. By making a contribution, whether through a turn at talk or some other communicative act like gesturing, a participant displays their understanding of what has preceded their contribution (Edwards, 1993;Heritage, 1984;Macbeth, 2011;Moerman and Sacks, 1988;Sacks, 1992;Schegloff and Sacks, 1973). For example, in a routine home visit following the birth of a baby, a community nurse observes the baby sucking or chewing an object and remarks 'he's enjoying that isn't he'. The child's father responds 'yes, he certainly is', displaying an understanding of the nurse's remark as a mere observation. In contrast, the child's mother responds 'he's not hungry cuz he's just had his bottle', displaying an understanding that the nurse may be implying the child requires feeding (Drew and Heritage, 1992). In making sense of one another's conduct, the sequential ordering of contributions to an interaction can be an important resource. This enables participants to interpret one another's contributions on the basis of its congruence with prior contributions, including the immediately prior turn (Schegloff and Sacks, 1973) as well as broader contributions to an extended sequence of action (Schegloff, 1990). In the same way that participants' contributions display their understanding of others' contributions to the interaction, participants' conduct can also display understandings of the social and physical world beyond their current interaction (Davidson et al., 2014;Hester and Francis, 1997).
Many types of educational encounters, especially those that are classroom-based, are hallmarked by a three-turn sequence (Bellack et al., 1966;McHoul, 1978;Mehan, 1979;Sinclair and Coulthard, 1975), comprising an initiating move (e.g. a question), a responsive move (e.g. an answer) and a reactive move (e.g. an assessment). Among other things, these sequences provide for displays of understanding (Edwards, 1993;Hester and Francis, 1997;Macbeth, 2011;Mushin et al., 2013). In particular, the response components of these sequences, which are overwhelmingly produced by students following teachers' questions, display understandings of whatever is being discussed. These displays can convey different levels of understanding (Koole, 2010;Sacks, 1992). They might demonstrate understanding by providing information that establishes the speaker's understanding as independent of the information already available within the interaction. For example, following a teacher's description of how to locate eight o'clock on a sunshine chart, a student demonstrates her independent understanding of how to interpret that chart by claiming 'this is nine o'clock' (Koole, 2010).
In contrast to demonstrations of understanding, participants can also claim understanding, but without providing information that would enable an interlocutor to identify the basis for that understanding. For example, following a teacher's explanation of how to draw a graph, a student responding 'yes yes I get it' claims understanding, but does not actively demonstrate an understanding of how to draw a graph (Koole, 2010). Participants' contributions therefore convey greater or lesser details about their level of understanding. The sequential organisation of participant's contributions to an interaction thus provides an 'architecture of intersubjectivity' (Heritage, 1984), which enables participants to determine -to greater or lesser extents -whether they share an understanding of some aspect of the world they are discussing. This can be especially important for educational encounters involving young children. This study examines how intersubjectivity can be defended in a setting where one party's competence -that of a child -is liable to being questioned (Baker and Freebody, 1989;Mackay, 1974).
Insofar as sequence organisation establishes an 'architecture of intersubjectivity', this organisation also enables identification of gaps, breakdowns and divergences in intersubjectivity. Where this occurs, repair organisation provides for the defence of intersubjectivity (Schegloff, 1992). Although it is recognised that repair is particularly important for intersubjectivity in educational contexts (Macbeth, 2004(Macbeth, , 2011McHoul, 1990), further research is required to understand how practices of repair relate to broader activities and roles pursued by educators and learners (Gardner, 2013). Further exploration of alternative practices to repair, such as those that create spaces for alternative understandings to be proffered (Hester and Francis, 1997), is also needed. Our study attends to this by considering how participants address equivocal trouble in the classroom.

Method
The data reported here were collected as part of a study that investigated how teachers and students engage in Web searching in Australian preschool classrooms that cater for children aged between three and five years. Between May and November 2012, approximately 170 hours of classroom interaction were recorded across nine preschool classrooms in south-east Queensland. Two video cameras were used to record different perspectives. Research assistants operated the cameras and moved around the classroom in response to shifts in activity. An additional perspective of computer use was recorded using screen capture software.
This study utilises the methodology of conversation analysis (Sidnell and Stivers, 2013). This approach provides an alternative to those that principally focus on pre-theorised global orders of discourse, such as asymmetrical power relations between teachers and students. Instead, the primary analytic focus in conversation analysis is how discursive practices are used to accomplish recognisable social actions (Macbeth, 2003;Schegloff, 1997c). The aim of this study is to utilise established findings to progress understanding of practices that are used for addressing equivocal sources of trouble. Although informed by an examination of the entire corpus of data, we achieve this aim through single case analysis. This approach is well suited to utilising existing findings to identify novel interactional practices (Schegloff, 1987a), particularly when focusing on the broader organisation of extended sequences of action (Psathas, 1992). A single case study thus enables an exploration of aspects of interaction that have relevance for organising the use of language within social encounters.
The data were transcribed according to the Jeffersonian system used within conversation analysis (Hepburn and Bolden, 2013), details of which are available as an appendix to this article. One adaptation was to use all capital letters for teacher speaker labels, in order to make clear the difference between teachers and students. Where it is unclear which student contributed to the interaction, this was transcribed with the label 'St' rather than a pseudonym. Close analysis of the video recordings and their transcriptions facilitated development of a multimodal analysis considering verbal and non-verbal conduct, and computer screen activity.

Addressing unequivocal trouble
The following fragment is included as a point of contrast for the subsequent focus on attempts to address equivocal sources of trouble. This instance illustrates how teachers can readily address unequivocal misunderstandings. Extensive conversation analytic research investigating repair and correction in both mundane social encounters (Schegloff, 1992;Schegloff et al., 1977) and classroom interaction (Macbeth, 2004;McHoul, 1990) establishes how these practices can be used readily in response to the emergence of an apparent source of trouble, including misunderstandings. In Fragment 1, an apparent source of trouble is a student's understanding of the communication technology they are about to use. Around 100 seconds prior to the beginning of this fragment, the teacher, Miss Sally, has been explaining to the class that a former teacher, Miss Pam, has asked the class to send her an email. This fragment begins with a teacher's aide, Miss Linda, crouching under the table to plug a device into the computer. The references made by Sally at lines 3 and 4 and one of the students at line 6 are in relation to this device:  At line 6, a student displays an understanding about a device being plugged into the computer that has just been referred to by Sally at lines 3 and 4. The student's claim about the device, 'It's for seeing Miss Pa:m. >On the< computer.', displays an apparent understanding that it is a webcam. 1 Sally's next turn is a third position repair (Schegloff, 1987b(Schegloff, , 1992(Schegloff, , 1997b, correcting the student's understanding by rejecting the claim that they will be able to see Miss Pam (line 8), and explaining they will not be using Skype TM (lines 8, 9, 12, 13 and 15) but rather will send her an email (line 15).
The student's claim at line 6 thus displays an unequivocal misunderstanding that Sally can correct using third position repair. A primary constraint on this particular repair practice is an interlocutor's display that the participants are not utilising the same understanding of some aspect of the world (Schegloff, 1992) -in this case, what the 'things' being plugged into the computer will enable the class to do. In contrast to unequivocal troubles such as these, which have been the predominant focus in existing research on repair (cf. Schegloff, 1992Schegloff, : 1331Schegloff, -1332, our study focuses on practices used where there is equivocality about the nature of a misunderstanding. In particular, although previous research has considered difficulties in identifying the specific basis of an otherwise ostensive misunderstanding (Schegloff, 1992(Schegloff, : 1331(Schegloff, -1334, we focus on equivocality in relation to whether there is even misunderstanding in the first place.

Addressing equivocal trouble
The remainder of this article focuses on a single episode of interaction in which a teacher attempts to address an equivocal source of trouble. The primary reason for using a single case analysis approach (Schegloff, 1987a) is frequency. In the 170 hours of classroom data collected, we only located this one instance where a teacher was addressing an apparently equivocal trouble. This phenomenon may be relatively infrequent because the moment-by-moment progress of interaction provides for the progressive determination of intersubjectivity (Heritage, 1984;Schegloff, 2006). Where there is a possible threat to intersubjectivity, people generally identify an ostensible source of trouble and address this with an appropriate repair practice (Schegloff, 1992(Schegloff, , 2006. Although they may be relatively uncommon, equivocal sources of trouble warrant analysis as they afford opportunities to extend understanding of the methods that people use to promote intersubjectivity. Another reason for a single case study approach is that breakdowns in intersubjectivity can become protracted (Ekberg, 2012;Schegloff, 1992). A detailed focus on how such breakdowns are occasioned and sustained, and whether and how they are resolved, is necessary to determine resources people use in extended attempts to identify and address equivocal trouble. Our analysis identifies a set of practices used to address such trouble. Across several fragments, we document how each practice is used, sometimes recurrently, and eventually overcomes a potential and yet equivocal trouble.
The focal episode involves a teacher called Sheree and three students: Hanna, Naomi and Sally. These participants are searching the World Wide Web for an image of a tick resembling the one that had bitten Hanna the previous evening. In just under nine minutes, they locate an adequate likeness. Before this is achieved, however, Hanna consistently rejects a series of candidate likenesses, which poses a problem for Sheree's facilitation of the search.
As the episode progresses, it becomes apparent there are two possible explanations available for Sheree to appreciate why they have failed to achieve the goal of their search. The first is simply that they have not yet located an adequate likeness. A second is that there is a problem with Hanna's understanding of the images on the screen, which impedes locating an adequate likeness. These alternative possibilities provide an analytic opportunity to explore practices people use to address equivocal sources of trouble in interaction.
In documenting the practices that Sheree employs to address the equivocal trouble, we consider five fragments from this episode (a transcript of the broader encounter from which these fragments have been taken is available as an online appendix). The first fragment follows a collaborative task in which the teacher supported the students to type the word 'ticks' into an image search engine, followed by an initial inspection of the search results. Hanna has already selected an image that Sheree copied into a word processing program. As Sheree returns to view the search results, she asks whether the image they have just copied resembles the tick that had bitten Hanna the previous evening. This occasions space in which Sheree seems to infer that Hanna has misunderstood the scale of the image on the screen: 184 Han: ((Pointing)) I don't th:ink it's ((starts moving finger)) that 185 one. It's ((points to another part of the screen)) that ((stops  186 pointing)) one ma:ybe:,= 187 SHE: =You think it might be that one, 188 Han: ((Nods once)) Yes:.
Our goal in analysing this fragment is to account for Sheree's explanation, at lines 181-183, of how the tick on the screen could seem bigger than the tick that was on Hanna. 2 In particular, we show that this explanation can be attributed to an inference, made by Sheree, that Hanna has misunderstood the scale of the image on the screen. The context of Sheree's explanation is occasioned by her question at line 167, asking whether the tick on the screen (Figure 1) resembles the tick that was on Hanna. The rejection of this likeness (line 168) poses a potential problem for their search. Although it is clear that they have not yet identified an adequate resemblance, it is not clear how the image on the screen is different.
Sheree's next turn is designed in response to an unsubstantiated rejection that has been made by Hanna. Her prosodically marked question (Selting, 1996) 'It doesn't?' (line 169) occasions space in which Hanna might further comment on the likeness of the two ticks. This is successful, insofar as Hanna subsequently explains that the tick on her 'was a <little one.>' (line 173).
Hanna's reference to size can be understood as a relevant account for the difference between the two ticks. If the tick that was on Hanna was 'little', then the tick on the screen must be a different size (i.e. bigger). Superordinate categories like 'size' and subordinate categories like 'little' can enable inferences that extend beyond what a speaker has said (Sacks, 1992: Vol. I, 113-125). In this instance, Sheree has already been informed that the ticks are different and, having occasioned space in which Hanna might account for that difference, is then informed that one was 'little'. Given the ticks are different and one is described as 'little' as an apparent point of difference, an inference about the other tick can be made by selecting a contrastive category within a superordinate category that subsumes both (Bilmes, 2009). In this case, 'little' can be subsumed under the superordinate category 'size', which contains the contrastive category 'big' (Murphy and Jones, 2008).
So if the tick that was on Hanna was little, there is scope for inferring the tick on the screen must be big or, at least, bigger. 3 Indeed, Sheree later clearly indicates, at line 183, that she has understood Hanna's claim to imply that the tick on the screen is big. More immediately, such an inference seems to underpin Sheree's acceptance, at line 174, of this explanation as an adequate account for the difference between the ticks. The impediment to completing their search is thus established as a failure to locate an adequate likeness, and Sheree addresses this by continuing their activity.
Having established the impediment as failure to locate an adequate likeness, there are subsequently grounds for Sheree to infer the impediment might actually be a misunderstanding by Hanna of the scale of the image on the screen. This is occasioned by Hanna's explanation that the tick that was on her 'ha:s little cla::w' (line 179). Although Hanna's explanation has only shifted from the generic littleness of the tick that was on her to the littleness of its 'claw', 4 this results in Sheree responding in a substantively different way. Using the same inferential process described earlier, Hanna's mention of the little claw on the tick that was on her provides a basis for Sheree to infer that the tick on the screen must have a big, or bigger, claw. Such an inference, however, leads to a problem, as the tick on the screen (Figure 1) does not have a particularly big claw.
Where Hanna's previous explanation of generic size made it possible to attribute the impediment for the search to failure in locating an adequate likeness, her subsequent reference to the size of the tick's claw makes it possible to attribute the impediment to a misunderstanding. This shift accounts for the different approach Sheree takes in lines 181-183 to her earlier response at line 174. She now explains how the two ticks could be the same size, although the former might look bigger. Sheree's explanation treats Hanna as having made a scale error 5 by failing to appreciate that the tick on the screen is bigger than it would appear when seen without magnification.
The problem with inferring a misunderstanding, however, is that there is no basis for being certain that such a misunderstanding has occurred. It is also possible that Hanna perfectly understands the scale of the image and the problem is simply that the image does not resemble the tick that was on her. Hanna has a much higher epistemic status (Heritage, 2012) in this regard: she was present when the tick was on her, Sheree was not. The design of Sheree's turn at lines 181-183 suits an equivocal misunderstanding. She does not use a practice like third position repair, which is suited to addressing clear misunderstandings (Schegloff, 1992). Rather, Sheree's explanation does not orient to a clear misunderstanding but does provide information that may correct any misunderstanding resulting from a scale error. If, however, there is no misunderstanding, then Hanna can respond in a way that does not orient to the possible correction.
On this occasion, Hanna responds in a way that is indicative of the latter option. She does this by rejecting the image they have been looking at, and suggesting a possible alternative likeness (Figure 2). Irrespective of whether she has misunderstood the scale of the image on the screen, her turn at lines 184-186 leaves little scope for Sheree to continue addressing such a misunderstanding. At line 187, Sheree therefore aligns with Hanna's suggestion of an alternative possible likeness. Hanna and Sheree's continued discussion of this alternative likeness, however, is circumvented by Sheree's subsequent noticing of another image. In data not shown here, but which is available in the online appendix to this article (see from line 189), Sheree's noticing eventually leads to the discussion of another image, which is the focus of Fragment 2B.
The practices for addressing an equivocal trouble a identified so far can be repeatedly observed throughout the broader episode of interaction. Where an equivocal source of trouble emerges, Sheree attempts to occasion spaces in which Hanna could modify a prior claim (Hester and Francis, 1997), or provide information in support of that claim (Robinson, 2009). The former outcome is not manifest in this interaction, although the latter possibility is often realised. When this provides scope to do so, Sheree infers a source of trouble and addresses it appropriately (e.g. lines 181-183). When this is not possible, or an attempt to identify the trouble is unsuccessful, a secondary option is to progress an activity without addressing the equivocal trouble (e.g. line 174). By progressing the activity, it is possible another opportunity will arise to identify and address the trouble (Schegloff, 1992). The remainder of our article establishes how these practices are used repeatedly in Sheree's attempts to address this impediment to their search. We will argue that these are systematic practices for identifying and addressing equivocal trouble in understanding.
The next fragment begins around 90 seconds after the end of Fragment 2A and involves discussion of another image now displayed on the screen. This fragment affords further opportunities to examine how Sheree addresses the equivocal source of trouble impeding their search:  Similar to Fragment 2A, at lines 250 and 251 of Fragment 2B Sheree again asks about the likeness between the tick on the screen (Figure 3) and the tick that was on Hanna. As in the previous fragment, Hanna responds by disconfirming the correspondence, and again without accounting for the difference. This rejection poses a problem, again, for the image search. Without a basis for appreciating the difference between the two ticks, it is not possible for Sheree to clearly identify the impediment to achieving the goal of their activity and what might be required to address that impediment.
In a context of ambiguity about the impediment to the image search, Sheree again occasions space in which reconciliatory information might be provided. This time, at line 253, she uses a questioning repeat, which initiates repair on Hanna's prior turn (Jefferson, 1972;Keel, 2011;Robinson, 2009Robinson, , 2013Robinson and Kevoe-Feldman, 2010;Schegloff, 1997a;Schegloff et al., 1977). Sheree reproduces the entire trouble-source turn, establishing that her difficulty was not hearing the turn but rather making sense of it (Robinson and Kevoe-Feldman, 2010). Similar to Sheree's earlier question (line 169), this repeat seeks clarification (Robinson, 2013). Although this is followed by the possible beginning of an explanation (line 255), Hanna's eventual response disclaims relevant knowledge about the matter (line 258). In reply, Sheree again uses a questioning repeat to create further space for reconciliatory information (line 259). No response, however, is forthcoming from Hanna (line 260).
Sheree's pursuit of reconciliatory information to this point in Fragment 2B has not resulted in the outcome observed in Fragment 2A. In that earlier instance, Hanna provided an account for the difference between the two ticks -an account based on sizethat enabled Sheree to infer and address an equivocal trouble. This has not been possible in Fragment 2B. The inability to identify divergent understanding means there is no basis to address a breakdown in intersubjectivity (Pollner, 1975;Schegloff, 1992). As established earlier, where it is not possible to infer and address an equivocal trouble, a secondary option is to progress the interaction without addressing that trouble. In doing so, it is possible that another opportunity will arise to identify and address the trouble. This is the course Sheree takes from line 261, which soon occasions another opportunity to identify a source of trouble.
Following Sheree's attempt to progress the interaction by reading information about ticks displayed on the screen (lines 265-271), Hanna again rejects the correspondence between the image on the screen and the tick that was on her (lines 273 and 274). Sheree in turn responds with a questioning repeat (line 275). As she repeats Hanna's prior turn, Sheree scrolls down the webpage, which displays a label 'paralysis tick' for the image they have been inspecting. Sheree's informing of this additional detail (lines 275 and 276) may help overcome the impediment to their search, if Hanna knows the type of tick that was on her. This does not happen, however, with Hanna instead repeating her rejection of a likeness (line 278). Sheree again employs a questioning repeat to occasion space for reconciliatory information (lines 279 and 280). In this instance, however, the negative grammatical form of Sheree's repeat makes disconfirmation the preferred response (cf. Heinemann, 2005;Heritage, 2010;Raymond, 2003). Hanna's production of this preferred response (line 281) means no reconciliatory information is forthcoming.
In spite of multiple attempts to occasion space for reconciliatory information to aid identifying and addressing the equivocal trouble, the absence of such information leaves little scope for Sheree other than to progress their activity. In this case, in data not shown here (but available in the online appendix), Sheree continues to scroll down the same  The first part of this fragment involves the same trajectory repeatedly observed earlier. Sheree refers to a potential correspondence (lines 306 and 307) and Hanna discounts this likeness without accounting for the difference (line 308). Sheree replies with a questioning repeat (line 309), occasioning space for reconciliatory information. No such information, however, is forthcoming (line 310). Once again, there is little scope to infer and address the equivocal trouble, and so again Sheree responds by progressing to a broader activity (from line 313).
To this point, over three and half minutes have passed since the first instance (at line 179) in which there was scope for Sheree to infer an apparent misunderstanding. Now, at lines 325-328, a second opportunity arises, which is again occasioned by Hanna's reference to size. 6 This is another basis for Sheree to return to attributing their impediment to a misunderstanding by Hanna based on a scale error. Where Sheree previously (at lines 181-183) addressed this apparent misunderstanding by explaining the scale of the image on the screen, here (at lines 330 and 331) she takes the approach of attributing confusion (Drew, 2005). 7 Their difficulty in locating an adequate likeness is thus accounted for by explicitly attributing misunderstanding to Hanna.
As it happens, Hanna does not respond to Sheree's attribution. As shown in the following, she instead orients to another image. In their ensuing interaction, however, there is another opportunity to address the equivocal trouble: Hanna and Sheree's discussion of another image follows a similar trajectory to that observed earlier and exposes another basis for inferring a misunderstanding based on a scale error. Sheree initially asks, at lines 339 and 340, whether the tick on Hanna was 'really big and swo:llen' like the image displayed on the screen ( Figure 5). Hanna possibly confirms this with a slight nod (line 341). Following this, however, Sheree comments that the tick on the screen is 'really big' (line 342), to which Hanna responds by disconfirming a likeness (line 345). In the same way that Sheree has previously responded to disconfirmation of a likeness, here she occasions space in which reconciliatory information might be provided (line 346). As Sheree produces this repeat, however, Hanna continues to expand her turn (line 347), providing the very account Sheree is soliciting.
Similar to the explanations provided at line 179 and lines 325-328, Hanna's explanation here provides a basis for Sheree to infer that Hanna has misunderstood the scale of the image on the screen. This time, however, Sheree responds in a way that proves to be successful in addressing the apparent impediment to their search:  Hanna's third mention of the size of the tick that was on her (line 347) occasions another opportunity for Sheree to infer and address the equivocal trouble impeding their search. To reiterate, that trouble is equivocal because there are at least two possible explanations that could account for the impediment. The first is that they simply have not managed to locate an adequate likeness. A second possibility is Hanna has misunderstood the scale of the images they are looking at, and this is impeding their ability to locate an adequate likeness.
In her previous attempts to identify and address this equivocal trouble, Sheree utilised two different approaches. Her first approach involved explaining how a tick on the screen could be the same size as the tick that was on Hanna although they appeared different (lines 181-183). Sheree's second approach involved attributing confusion to Hanna (lines 330 and 331). Her third attempt involves circumventing the basis for a misunderstanding by suggesting they locate images of ticks on a person's skin (lines 352-355). If Hanna has indeed made a scale error, Sheree's suggestion of focusing on images of ticks in scale should circumvent the basis of this error. This approach proves successful. At lines 361 and 362, Sheree nominates a candidate likeness. She then continues to scroll down the screen and, at line 364, nominates another possible likeness (Figure 6), which Hanna confirms (line 366). The practices used by Sheree have culminated in the resolution of an apparent trouble by circumventing the basis of that trouble, rather than finding a more direct way to address it.
Although Sheree seems to have identified a possible misunderstanding in the early stages of their image search, it takes time and repeated attempts to implement an approach that appears to successfully remove the impediment to the search. Her eventual success comes from circumventing the conditions under which Hanna could be potentially misunderstanding what they are looking at, rather than directly addressing that possible misunderstanding. The approach taken by Sheree suits the equivocal nature of the trouble she seeks to address. Although there is evidence supporting Sheree's apparent inference that Hanna has made a scale error, at no point does this misunderstanding become absolutely clear. The strategies employed by Sheree suit this equivocality.

Conclusion
Our analysis progresses understanding about addressing breakdowns in intersubjectivity, predominantly studied in the past with a focus on repair of ostensible trouble, by examining how equivocal trouble can be addressed. Exploring an interaction between a teacher and preschool-aged student in which it was unclear whether they shared the same understanding of the images they were examining, this study has identified practices used by one party in attempts to identify and address a possible misunderstanding that may be impeding the success of their activity. This analysis contributes to existing research that explores practices people use to preserve intersubjectivity (Heritage, 1984;Schegloff, 1992Schegloff, , 2006. It provides further demonstration of the potential for conversation analytic research in this area, by extending existing findings about how repair practices are used to defend intersubjectivity by exploring how equivocal troubles are addressed. In each instance where an equivocal source of trouble emerged, the recipient to that trouble initially occasioned space in which reconciliatory information could be provided (Robinson, 2009). Most often, this was accomplished by repeating a version of the turn exposing the equivocal source of trouble. Although these repeats initiated repair (Jefferson, 1972;Keel, 2011;Robinson, 2009Robinson, , 2013Robinson and Kevoe-Feldman, 2010;Schegloff, 1997a;Schegloff et al., 1977), they did not resolve the equivocal trouble. Rather they created space for information that might facilitate identification of the trouble. Where information was forthcoming, the party who initiated repair used this to infer and address a source of trouble.
Where further information was not forthcoming, or where an attempt to identify and address a source of trouble failed, a secondary option was to continue some broader activity without attempting to identify and address the source of trouble. By progressing the activity, it is possible another opportunity may arise to address the trouble (Schegloff, 1992). The practices used by Sheree suggest a particular type of preference for progressivity (cf. Stivers and Robinson, 2006): where equivocal trouble emerges and cannot be addressed, a recipient of an equivocal trouble can progress a broader activity-in-progress with the prospect that further opportunities may arise to identify and address that trouble.
In taking a conversation analytic approach, this study has identified practices that are demonstrably relevant to participants. Although informed by relevant prior research, the specific avoidance of pre-theorised global orders of discourse enables an analytic enterprise for identifying methodical procedures people use to deal with the local contingencies of social interaction (Macbeth, 2003;Schegloff, 1997c). Our study explores one particular contingency -equivocal misunderstanding -and methodical procedures for ordering a social interaction affected by this contingency: the use of repair to create space for reconciliatory information and, in the absence of any reconciliatory information, progress of some broader activity without attempting to identify and address the source of trouble.
Each instance where the recipient to an equivocal trouble eventually manages to infer and address that trouble was achieved with alternatives to repair: explanation (lines 181-183), accounting (lines 329-331) and finally circumventing the potential basis for a misunderstanding (lines 351-368). These alternatives suit contexts in which it remains unclear whether there indeed has been a misunderstanding that needs to be addressed. Our expectation is that interactants may not adopt such approaches when solicitations of reconciliatory information occasion displays of ostensible rather than equivocal misunderstandings. Where this occurs, a major constraint on the use of third position repair is overcome (Schegloff, 1992). What appears crucial, then, is the strength and quality of any reconciliatory information that becomes available.
The three practices used to address the equivocal trouble also appear to involve not treating the student as being competent in appropriately understanding what she has looked at on the computer screen. Although her general capacity to identify a tick that resembled the one that was on her is recognised, her specific ability to comprehend the scale of the images on the screen is questioned. Students -particularly young students -appear to be routinely subject to such circumstances where their competence is simultaneously assumed and denied (Baker and Freebody, 1989;Mackay, 1974). We therefore expect that the practices we have considered may be more prevalent in settings where one party's competence is liable to being questioned.
Identifying and addressing misunderstandings is an important interactional achievement, insofar as misunderstandings can jeopardise interactants' ability to mutually accomplish a course of action (Ekberg, 2012;Pollner, 1975;Schegloff, 1992). Although the turn-by-turn display of participants' understandings and practices of repair provide a basis for defending intersubjectivity, our study explores instances that prove difficult to address with this 'self-righting mechanism' for interaction (Schegloff, 1992(Schegloff, : 1299. Such instances can be pivotal in educational settings. Here, the core business is to increase and enhance knowledge, and yet this achievement is contingent on participants establishing intersubjective understandings that enable the shared activities of the classroom to progress (Macbeth, 2011). By focusing on particular equivocal trouble in understanding, we have identified practices used in attempts to preserve the intersubjectivity required in education and, presumably, in social life more generally. searching for images. In correspondence with the references used by participants during this particular episode (e.g. at line 167), we refer to 'the tick that was on Hanna'. 3. It may well be the case that the tick on the screen is 'bigger' than the tick that was on Hanna.
In data not shown here (but which is available at line 116 of the online appendix), Sheree commented that the tick displayed in Figure 1 was 'all swollen up'. So it is possible that the tick that was on Hanna was not as swollen as the tick on the screen. As will become apparent in the subsequent interaction, however, Sheree soon appears to supplant this interpretation with an inference that Hanna has misunderstood the scale of the image on the screen. 4. Hanna's mention of the tick's 'claw' is most likely a reference to its mouthparts (Sonenshine and Anderson, 2014), an illustration of which is available in Figure 4. 5. Although previous experimental research has identified scale errors being made by children up to the age of two and a half years old (DeLoache et al., 2004;Ware et al., 2006), we are not aware of any research exploring scale errors in children aged three to five years, the age range of participants in this study. This study is not designed to establish whether and how frequently children in this age group might make scale errors, but does show they are liable to having such an error attributed to them. 6. Sheree has just been reading information displayed on the screen, and Hanna's mention of size is disjunctive with the topic of that information. Her use of past tense at lines 325-328, however, makes it clear that Hanna is referring to the tick that was on her the previous evening, rather than the tick currently displayed on the screen. 7. Although Sheree begins her turn with a repeat of Hanna's prior turn, this is not produced with rising intonation like the other repeats we have considered. The falling intonation at the end of this repeat appears to accomplish confirmation (Schegloff, 1997a), rather than pursuit of the matter being repeated.