“Misunderstanding” and (mis)interpretation as strategic tools in intercultural interactions between preschool children

4.1 Legitimate misunderstanding and repair: A brief review

Researchers interested in communication have long studied misunderstanding as a path to understanding both how smooth communication works as well as how misunderstandings occur and are then managed by interlocutors (Humphreys-Jones 1986). Some of these accounts have aimed to create taxonomies of types of misunderstandings, based on various aspects or features, such as the level of language at which they occur (phonology, semantics, etc) (e. g. Banzanella and Damiano 1999), or how aware participants are, given the underlying reasons for the misunderstanding (e. g. Coupland et al. 1991). At the same time, many of these analyses of misunderstanding – particularly those from the conversation analytic tradition (e. g. Schegloff 1987) – also provide detailed accounts of how misunderstandings unfold, are noticed, pointed out, and addressed.

By all accounts, misunderstanding/repair sequences begin with someone noticing the trouble. (Although of course, misunderstandings may occur without anyone noticing, these do not involve repair or clarification sequences and thus are less relevant for our purposes here.) In what Schegloff et al. (1977) called “other-repair” and Hirst et al. (1994) called an “other-misunderstanding,” a hearer notices trouble – that she has not heard clearly, perhaps, or that what she has heard does not make sense within the conversation – and asks for clarification right away (“Pardon?” or “Did you say bees?”). The speaker may then clarify, constituting what Schegloff (1992) called a “third position repair”: the first position was the “trouble spot” (Schegloff et al. 1977), the second position was the request for clarification, and the third is the repair (“No, no, beads”). Third position repair can be also initiated by the original speaker, as a “self-misunderstanding” (Hirst et al. 1994). This occurs when the original speaker (position 1) notices that his interlocutor’s reaction (position 2) is incongruent with what the speaker meant and infers from this logical breakdown that interlocutor misheard or made the wrong interpretation (Banzanella and Damiano 1999). In both cases, after the misunderstanding-repair sequence is complete, the speakers resume the conversation.

Repairs may also occur in the fourth position, if speaker 2 notices an incongruence between her second position interpretation and speaker 1’s subsequent third position turn, or in subsequent positions, if misunderstandings take several turns to notice, or if it takes several turns to navigate to the repair sequence. The third position repair – trouble spot (1), clarification request (2), repair (3) – is the most common, however (Banzanella and Damiano 1999), and is the basis for the discussion in this paper.

4.2 Preschoolers doing legitimate misunderstanding and repair

Work that has approached children and misunderstanding from a developmental perspective has found that preschoolers are quite capable of doing this kind of clarification and repair work. Children as young as 2;6 (2 years, 6 months) are able to notice misunderstandings in conversation, both on their part and by others, and are able to respond to them, through self-repair as well as through requests for clarification (Garvey 1984; Golnikoff 1993; Forrester and Cherington 2009). Although there has been some debate as to what very young children’s motivations are for repairing misunderstandings – Shatz and O’Reilly (1990), for instance, argued that children were more likely to initiate repairs when making requests than assertions – there is not debate over whether they can engage in them at all. And while much of the research on young children and repair examines children speaking with adults, other work also shows children engaging in repair with one another. In one study of spontaneously occurring conversation between 2;8- to 3;6-year-old peers, Aviezer (2003) found that “young children are sensitive to conversational breakdown and recognize different kinds of communicative failure as well as being able to initiate effective clarification procedures as an invitation for others’ repair” (134). Although older children engaged in clarification sequences more frequently, all of the children in his study recognized misunderstandings and engaged in repair, particularly in clarification sequences. From a developmental perspective, therefore, the students who produced the conversations in the present paper (ages 3;6 to 4;10) would be able to engage in negotiation of misunderstandings and repair. As the following example illustrates, that is indeed the case.

4.3 Preschoolers doing legitimate misunderstanding and repair: An example

In this example, the interlocutors are a 4-year-old Nepali speaker, Kritika, and her English-speaking teacher, Lucia. The interaction began as Lucia overheard three students speaking in Nepali about a drawing. Hoping to encourage their use of English, she asked the students what they were saying. Kritika answered, but Lucia did not understand her response, so Lucia and Kritika engaged in a clarification and repair sequence. The multimodal transcript highlights Kritika’s use of multiple modes to help Lucia arrive at her intended meaning. The shape and direction of the over-image text show the intonation of Kritika’s speech. The captions show the timestamp and describe the action taking place.

Example 1

This example shows the kind of misunderstanding/clarification/repair sequence outlined above, in which one interlocutor notices that she may have misheard, she checks with the speaker, the speaker attempts to clarify, and they go back and forth until understanding is reached. Interactions like this one show students and teachers knowing how to “do” – jointly produce – misunderstanding/repair routines. These kinds of misunderstandings/repairs exemplify the cooperative spirit assumed in much work on miscommunication in general and on intercultural miscommunication in particular: Misunderstandings happen, but with enough goodwill and effort, communication can proceed.

4.4 Repairing misunderstanding through third party interpretation or “speaking for”

A different way of resolving misunderstanding in this classroom involved the use of a third, mediating interlocutor. This practice originated with the teachers: Because of the makeup of the classroom, at the start of the year, the teachers had significant trouble understanding many of the students, and thus relied on peers who already spoke some English to help by interpreting, or clarifying what a student said. This did not necessarily mean translating from Nepali to English. Often it meant simply repeating what a student had said, but in clearer English. Like the third position repair sequences outline above, these interactions had a routine format. A student would speak to the teacher (position 1), the teacher would realize she did not understand and would call another student to help (position 2), and the interpreter would clarify (position 3). The position 2 requests for clarification took a routine format: “Dinesh, what’s Monal saying?”; “Dinesh, help me out,”; “Dinesh, I can’t make out what Maiya is saying. Can you help me?” (Field Notes, October 10, 2012). The student, in these cases Dinesh, would typically reply in position 3 in a format for reporting speech – “She’s saying ‘More pineapple!’” – and the interaction would continue. (I label these examples collectively Example 2).

The role of interpreter was a powerful one, particularly when it served as gatekeeper (for instance, of access to the teachers), and sometimes students stepped into the interpreter role of their own accord, as Kritika did in the following interaction. This conversation took place just after morning circle time (the time when the whole class met to talk about the day ahead), as Hande, a Turkish speaker, was attempting to tell the teacher where she intended to play for the next hour of the day.

Although Hande did not acknowledge it herself, Kritika correctly interpreted where Hande wanted to play. This was not lost on Lucia, who smiled and nodded knowingly at Kritika in line 12, after observing Hande’s choice. Whether the interpreter was invited by the teacher or took up the role on her own, her responsibility was to convey the words of a student who was unable (or who the teacher saw as unable) to engage in the clarification sequence herself.

4.5 The participation framework of Speaking For

What unites the cases of two-party clarification/repair and three-party clarification/repair is that both involve the act of Speaking For another person. In the “sun” clarification sequence (Example 1), for instance, Lucia repeated what she believed she heard; in the Dinesh examples (Example 2) and the “sand” example (Example 3), the interpreter recast what he believed he heard in a way that clarified it for the original, intended hearer. In both cases, this voicing of another speaker’s words somewhat complicates the typical notion of speaker, a person whose words represent his own ideas and intentions, and hearer, the intended audience for those ideas.

Goffman (1981) deconstructed the roles of speaker and hearer in ways that are useful here. He argued that “speaker” and “hearer” do not address all of the ways of being a speaker and hearer, particularly in multi-party interactions. Rather than “hearer”, he proposed to distinguish between official audience members (“addressee(s)”, “unaddressed intended hearers”, “intended hearers who are not listening”, etc.) and unofficial hearers (such as “overhearers”, “eavesdroppers”, “bystanders”). He also dissected the role of speaker into three “production formats”: animator (the mouthpiece for the words), author (the creator of the words being spoken), and principal (the one whose interests or beliefs are conveyed by the words being spoken). Goffman explained that all three production formats can be united in one person (as when Kritika declared, “Sun”), or they can be distributed among different people (as when Kritika said “sand,” animating Hande’s words). These ways of being speaker and hearer, when configured in various arrangements, form different participation frameworks. The participant framework of Speaking For, for instance, was a common one in this classroom, and was invoked specifically as a way to repair misunderstanding.

To revisit the examples of interpretation (Examples 2 and 3) then through Goffman’s production formats, Maiya (looking for more pineapple) and Hande (looking to play with the sand) were the authors, animators, and principals behind their original utterances. When they ran into trouble, however, their words were re-animated by another speaker for clarification. Dinesh said “She’s saying ‘More pineapple!” and Kritika said, simply, “Sand.” Dinesh and Kritika were not the authors of the words, nor were they invested, as principals, in the message the words conveyed; they only served as their animators. In terms of recipient formats, Dinesh and Kritika, although they certainly heard the original utterances, were not original addressees. Dinesh was brought in as an unaddressed (but intended) hearer, while Kritika was a bystander who stepped in of her own accord. In both cases, they functioned as a stand-in understander (for an addressed non-understander) and then a helpful mouthpiece.

Even in the case of a two-party interaction, we can understand the clarification sequence as a change in participant framework, from standard conversation – in which speakers function as animators, authors, and principals of their own utterances and hearers function as official addressees – to a Speaking For framework, in which one person becomes an animator for the other in order to clarify what was said. In Example 1, for instance, when Teacher Lucia asked, “Song?” and then “Sauna?,” in each case, she was animating (what she thought were) Kritika’s words.

4.6 Speaking For as double-voiced discourse

As helpful as Goffman’s categories are for understanding each participant’s role in these misunderstanding/repair sequences, this dismantling of speaker and hearer also risks oversimplification. Although Speaking For indeed involved animating another person’s words, in another sense, it also meant issuing a response to those words and, in that response, a social positioning. ^[1] In the case of the “sun/sauna” misunderstanding, for instance, when Lucia said “sauna,” she was not only re-animating what she thought Kritika said, she was also responding that she had not understood, but that she wanted to understand. Additionally, in making this response, she positioned Kritika as a speaker who was difficult to understand, but also worth understanding, despite it taking a bit of work. Speaking For can therefore be seen as what Bakhtin (1981) called double-voiced discourse, or language that “serves two speakers at the same time and expresses simultaneously two different intentions” (324). In the original focus of Bakhtin’s theory of double-voiced discourse – the novel – the two voices are those of the speaking character and of the author, expressing both the character’s intention and the author’s. In the examples here, the two voices belong to the author of the original words and to their re-animator. Thus, when Kritika told the teacher, “Sand,” on Hande’s behalf, she not only animated Hande’s words and shared Hande’s intention with the teacher, she simultaneously conveyed her attentiveness and helpfulness and positioned herself as a capable interpreter, able to understand and to convey a message more clearly than the original speaker. She also, by the very act of Speaking For (and whether she meant to or not), positioned Hande as a less capable communicator and someone in need of assistance.

4.7 Who could speak for whom?

Together, the concepts of participant framework, with accompanying production formats, and of double-voiced discourse provide a way to understand why not all students could take on all of the roles in the Speaking For frame. While anyone technically could repeat another’s words, animating those words required others to step into the accompanying roles of spoken-for-author and of-addressee-accepting-the-interpretation, and to acknowledge the animator as a legitimate interpreter. In the Dinesh examples (Example 2), it was the teacher who invited Dinesh into the role of interpreter, accepting the addressee role for herself. Additionally, students like Maiya and Monal accepted the help, and with it the positioning that came with each double-voiced act of Speaking For: The first voice was that of the Monal, for instance, looking to convey his request, the second was that of Dinesh, positioning himself as helpful, useful, and, importantly, as a more competent English speaker than Monal. Thus, the success of the entire operation depended on whether the addressee (here, teacher) saw the interpreter (here, Dinesh) as a more competent speaker than the spoken-for student (Monal or Maiya). Thus, the power to speak for another student was a relational power. It was not something that a student always possessed, nor did it depend on an objective measure of English proficiency; instead, a student’s ability to successfully claim the role of interpreter depended on his competence, first, relative to the spoken-for student and, second, in the eyes of an addressee. As such, not all students were able to successfully take up the interpreter role, as illustrated by the following transcript.

In refusing to engage in the participant frame of Speaking For that Pooja invoked, the teacher rejected Pooja’s double-voiced speech on both levels: as animation of Anita’s words (prompting Anita to speak for herself) and as a claim to Pooja’s relative competence and Anita’s relative incompetence as English speakers. Despite Anita’s willingness to be spoken for, it was the addressee – in this case, the teacher, one of the listeners with the most power to judge competence – whose ratification of the framework really mattered. In this classroom then, one kind of symbolic power was the power to invoke and draw others into a desired participant frame.

5.1 Strategic misunderstanding, plus power and positioning

What Hinnenkamp’s analysis (or any other grounded only in linguistic data) cannot answer, however, is who can successfully use these strategies, when, and to what broader effects. Answering these questions requires looking beyond a single transcript to the larger social field in which the interaction occurred, including social relations, which are always also power relations. Deborah Cameron’s (1998) work – also about misunderstandings that are not really misunderstandings – introduces power into the equation. Cameron’s work on communication across genders began from a similar premise to Sarangi’s: that misunderstandings in cross-gender communication cannot automatically be assumed to be the result of gender difference, and while there are certainly examples of true misunderstandings across genders, many of the “misunderstandings” between men and women are not explainable by some inherent difference in how men and women speak, but by conflicts in interest.

One of Cameron’s examples of misunderstanding-that-is-really-conflict came from a magazine article containing advice for women about how to communicate at work. It gave the example of a female boss saying to her male subordinate, “Would you like to finish that report today?”, and later returning to find it unfinished. When the boss admonished the employee, he became defensive, saying that if he had to finish today, she should have said so. The advice column then explained that by being more direct (i. e. more like a man), and thus making it clear that the report must be finished that day, the boss could have avoided the misunderstanding. Cameron argued, however, that in the context of a workplace, where one’s personal preferences are generally less relevant than carrying out one’s duties, the man should have had no trouble inferring that his boss was not asking about his taste in work tasks, but requesting that he finish the report. Cameron provided two possible alternative interpretations: Either the man pretended to misunderstand (“misunderstood”), using the form of her indirect request as a cover and an alibi; or the man genuinely misunderstood her request as a suggestion, the result of a general refusal to see a woman as someone in a position to give him orders. Importantly, either way, his response evidences a dispute over power relations between employee and boss, male and female. In a different conversation, in which there was no conflict over power and authority, the man would likely have no trouble interpreting the question, “Would you like to finish that report today?”, as the strong-suggestion-verging-on-order that it was meant to be. As Cameron put it:

The same person can behave very differently depending on who he or she is talking to, from what position, and for what purpose; the utterance which is ‘misunderstood’ by an individual in one situation may be treated as perfectly transparent by the same individual in another.

(451)

This is because all communication creates a frame in which speakers and hearers inhabit particular social positions (e. g. “I am the kind of person who can make requests like this and you are the kind of person who will carry them out”). And while the act of speaking sets up the frame and positions, the act of understanding is a decision whether or not to accept them. For Cameron, one of the ways to resist an undesirable frame or position is to “misunderstand:”

[L]aying claim to a particular intention or interpretation can function as a strategic move in a game of power and resistance. What is commonly called ‘misunderstanding’ may often be better analyzed as a kind of conflict; at bottom, conflict about the social positioning that is always implicitly presupposed when one person addresses another.

(452)

From this perspective, a misunderstanding might be a refusal on the part of a hearer to participate in the conversational frame being set up by the speaker, or to be positioned in the way that the speaker’s utterance, if understood as it was intended, would position him.

In a sense then, strategic non-understanding, or refusing to comprehend in any way at all, could in the most basic way accomplish a rejection of a speaker’s positioning of a hearer (a woman does not turn when a man calls “baby” across the street; a preschool student looks blankly at another who says, “You took my toy.”) Strategic misunderstanding is slightly more complex in that it not only refutes the speaker’s proposed positionings, but offers alternate ones. In the “misunderstanding” between the boss and employee, for instance, the employee’s interpretation of the request (as a question) offers a different interpretation not only of the words, but also of the relationship between the speakers, as if to say, “You are not someone who gives me orders. I am not someone who has to take your requests too seriously.” Thus, while Hinnenkamp interpreted H’s false misunderstanding as a way to get the floor and get a laugh at B’s expense, we might also understand it as a conflict over positioning. Perhaps H did not see himself as the kind of stumbling, backtracking, clarification-seeking non-native speaker that he perceived B to be. Proposing a false misunderstanding, then, not only got a laugh, but also positioned him as facile with German and thus unlike B.

Armed now with Hinnekamp’s and Cameron’s theories, as well as a theory of how legitimate misunderstanding and repair worked in this classroom, I now turn now to one of the troubling interactions that looked, at first glance, like a true misunderstanding and interpretation sequences but that, on closer inspection, was not.

5.2 Preschoolers doing “misunderstanding” and (mis)interpretation

The interaction involved Hande (age 4 ½), the sole Turkish speaker in the class and one of the students that the teachers worried about most. Throughout the year, peers saw Hande as a dispreferred playmate and teachers saw her as somewhat of a loner (Bernstein 2014). Unlike the Nepali speakers, who could showcase their interactional and social competence in their L1, Hande could only interact in English. Although in the end this may have helped her grow more in terms of vocabulary and utterance complexity than the other ELLs (Bernstein 2014), she struggled socially, and this led the teachers and classmates to see her as a less-than-competent communicator. The other participants in the interaction were Tommy (age 3½) a white, male, English speaker and Lucia, the teacher. The interaction occurred in April of the school year, at snack time, where Hande was sitting next to Tommy. Midway through snack time, over the course of several minutes, Tommy took all of the orange slices from the shared serving bowl, squeezed them one-by-one into his milk, and then soaked his napkin in it. As he wrung his napkin onto his plate and began lapping it up like a cat, Hande reached for an orange slice and found that the bowl was empty. Looking at the pile of the discarded orange peels and wet, balled-up napkins in front of Tommy, who was still licking milk and juice from his plate, Hande, appalled, called to the teachers to tattle on him. Researchers who have studied tattling have shown how children use it to resolve conflict and restore the moral order (Danby and Theobold 2012; Kyratzis 2004) of the classroom, particularly when students do not have the resources, linguistic or otherwise, to resolve it themselves (Cekaite 2012). This transcript begins with just such a tattle and with Teacher Lucia approaching the table.

At first reading, I flagged this conversation as a legitimate misunderstanding, because neither the teacher nor Tommy seemed to understand what Hande was trying to convey – that Tommy had made a huge mess and taken all of the oranges. Yet, Hande’s tattle was quite complex and complete, if non-standard, and the evidence of the mess – soaked napkins, discarded orange peels, milky orange juice running down Tommy’s chin – was visible. Thus it ought to have been quite clear to Tommy and Lucia what Hande was saying. Cameron’s perspective can help to explain what happened and how Tommy accomplished it.

A successful tattle involves drawing the teacher in as an objective party who neutrally hears out the conflict, through questions like “What happened?”, and then acts as judge and restorer of moral order through socializing questions, such as “How come X?” or “Why did you X”? (Cekaite 2012; Sterponi 2003). In this case, Hande’s tattle in lines 1 and 2, which opened the interaction and continued into lines 4 and 5, did not cast Tommy in a very favorable light. The conflict of interests here is clear: If Hande’s message were to be understood as a tattle, Tommy would likely be punished; if Hande’s message was not understood as a tattle, Tommy might avoid punishment. In terms of positioning, Hande’s tattle would position her as a reliable source of information and a competent student – someone who knew how to appropriately behave at snack time – in contrast to Tommy, a misbehaver, and thus a less competent student. In line 6, the teacher began to take up the tattling framework, neutrally thanking Hande for the information and taking in the scene.

Tommy, aware of the consequences of Lucia’s understanding, made a strategic move. In line 7, he presented a counter-tattle based on a strategic misunderstanding of what Hande said: That by using the wrong pronoun (she, rather than he), Hande had in effect, called Tommy a girl. Suddenly, Hande was in the position of being accused, rather than accusing. Additionally, Tommy, in “failing” to understand Hande’s utterance, positioned her as linguistically incompetent (or worse, too stupid to know the difference between boys and girls), undermining her authority as speaker and thus delegitimizing her tattle. While the teacher (line 8) did not ratify Tommy’s new frame, she did not contradict it either, nor did she participate further in the tattling frame that Hande proposed. Tommy, still not satisfied that the teachers understood his “misunderstanding” of Hande’s utterance, moved in line 9 into the reported speech formula of (a false) speaking for. In a strategic (mis)interpretation, Tommy spoke for Hande using words that Hande never actually said, attempting to pass them off as an animation, rather than an authoring, thus roping Hande into being the principal behind them. In a parasitic double-voicing, he voiced his interest (highlighting how he was wronged, but also showing himself to be a reliable source, unlike Hande) and purported to voice hers as well (calling him a girl), when in reality his utterance presented only his view.

In line 10, there was a shift, as Lucia was finally drawn into the frame that Tommy set up. Although Lucia tried to defuse the situation, assuring Tommy that Hande did not really believe he was a girl and did not mean to call him one, in doing so, Lucia also positioned Hande as someone who did not know what she was saying, or a confused English learner. Lucia thus ratified a frame in which Hande’s tattle might be unreliable as well. Tommy persisted in his claim, and although, in line 13, Hande resisted his positioning of her as a confused or inaccurate speaker, to do so she was forced to abandon her own conversational frame and to enter Tommy’s, in which she and her credibility, rather than Tommy and his mess, were on trial. In the end, Lucia tried again to diffuse the situation – this time through distraction with bread and jelly – leaving Tommy unpunished, the orange bowl empty, and the power differential between Tommy and Hande intact. By accepting Tommy’s strategic and decidedly parasitic “misunderstanding” and (mis)interpretation as true misunderstanding and interpretation, Lucia underlined that, as an English speaker, Tommy had the power to redefine the situation and to be taken seriously in a way that Hande, a “mixed up” English learner, did not.