Interpreting Impoliteness : Interpreters ’ Voices

Interpreters in the public sector in Norway interpret in a variety of institutional encounters, and the interpreters evaluate the majority of these encounters as polite. However, some encounters are evaluated as impolite, and they pose challenges when it comes to interpreting impoliteness. This issue raises the question of whether interpreters should take a stance on their own evaluation of impoliteness and whether they should interfere in communication. In order to find out more about how interpreters cope with this challenge, in 2014 a survey was sent to all interpreters registered in the Norwegian Register of Interpreters. The survey data were analyzed within the theoretical framework of impoliteness theory using the notion of moral order as an explanatory tool in a close reading of interpreters’ answers. The analysis shows that interpreters reported using a variety of strategies for interpreting impoliteness, including omissions and downtoning. However, the interpreters also gave examples of individual strategies for coping with impoliteness, such as interrupting and postponing interpreting. These strategies border behavioral strategies and conflict with the Norwegian ethical guidelines for interpreting. In light of the ethical guidelines and actual practice, mapping and discussing different strategies used by interpreters might heighten interpreters’ and interpreter-users’ awareness of the role impoliteness can play in institutional interpreter– mediated encounters.


Introduction
Interpreters in Norway's public sector work in a variety of contexts, including courts, hospitals, police stations, schools, social offices, private homes, and jails, and they interpret hundreds of institutional encounters daily (NOU, 2014: 13).The majority of these encounters are polite according to interpreters' perceptions and reports.However, interpreters evaluate some encounters as impolite and challenging for interpreters when it comes to interpreting impoliteness.This recurring issue is also voiced in interpreting courses, which originally motivated this study (see Felberg, 2016).
These challenges raise a basic question: Should interpreters (a) render everything faithfully, thus taking no stance on their evaluation of impoliteness, or (b) should they interfere in the communication between the primary communication participants, thus making an evaluation of the impoliteness?According to Code of Professional Conduct (Retningslinjer for god tolkeskikk, 1997), interpreters should render the meaning faithfully and impartially.Nevertheless, the results of this study indicate that some interpreters choose to interfere in communication.The question that naturally follows is why some interpreters choose to interfere even though they have agreed to follow the ethical guidelines.Are they incompetent and/or unprofessional?Have they misunderstood their function and forgotten that they are neutral participants?Or are they too preoccupied with making an impression and saving their own or others' face?
In reference to this, we argue that an important element in answering this question lies in the interpreter's underlying moral order.The notion of moral order-that which is taken for granted-as described by Kádár and Haugh (2013: 67) and further elaborated by Kádár (2017: 2) applies to the way "we assign and keep things in their place, occasionally even without explicitly recognising the values that we re-enact through maintaining this order."and we use it as an explanatory tool in a close reading of the surveyed interpreters' answers.Kádár further claims that [i]n terms of interaction, this implies that any individual is surrounded by a cluster of perceived moral orders, and uses language according to the moral order that he perceives to be triggered by a given context or interpersonal relationship.(Kádár 2017: 2) Thus, in interpreting settings, the interpreters also face different moral orders that influence interaction: The ethical guidelines based on a particular moral order of not interfering in interpreter-users' encounters might conflict with another moral order, which emphasizes that politeness is good and impoliteness is bad and that the interpreter should act accordingly, that is, interfere.Given all of the above, the main aims of this study are (a) to map interpreters' discursive constructions of what they consider their own strategies for interpreting impoliteness and (b) to analyze interpreters' discursive constructions in connection with the Norwegian ethical guidelines and the notion of moral order.
In order to achieve these goals, we sent a survey to all of the interpreters registered in the Norwegian Register of Interpreters (NRI).The interpreters' answers show that they use a variety of strategies for interpreting impoliteness, including omissions and downtoning.Therefore, our research confirms the idea of an unwritten moral order that seems to be based on the expectation, held by some interpreters and interpreter-users, that the interpreter will fix and/or downplay impoliteness.
The interpreters' reported strategies do not appear to be unusual: Research on interpreting (Jacobsen, 2008) has shown that some interpreters tend to neutralize or downplay impoliteness in order to save their own face or that of the interpreter-users.However, in our study, interpreters give examples of discursively constructed phenomena that can be understood as behavioral coping strategies, rather than interpreting strategies.These phenomena include interrupting and postponing interpreting.
After providing an overview of our theoretical and methodological approach, we present an in-depth analysis of the interpreters' reported solutions and strategies, and discuss the results in light of the ethical guidelines for interpreting.

Theoretical approach and data
Previous research on interpreter-mediated impoliteness has primarily focused on examples drawn from conference and court interpreting (Jacobsen, 2008;Magnifico & Defrancq, 2016).The connection between face issues and interpreting has been addressed in discussions centered on questions such as how face-threatening acts are modified during interpretation (Mason & Stewart, 2001) and what role politeness plays in forming witnesses' impressions (Berk-Seligson, 2002).A recent article by Magnifico and Defrancq (2016) explicitly approached the topic of impoliteness in connection with interpreters' gender.
However, to the best of our knowledge, no studies have explored interpreters' constructions of their own solutions and strategies when the interpreters encounter what they evaluate as impoliteness (while interpreting in the public sector).Therefore, we argue that this study is important for raising interpreters' and interpreter-users' awareness of how impoliteness can influence communication.
Our understanding of impoliteness follows Kádár and Haugh (2013), who claimed that politeness and impoliteness are forms of social action that cannot be found in particular behaviors or linguistic forms themselves but in evaluations of behaviors and linguistic forms.Thus, the central object of this study is to analyze interpreters' constructions of their own evaluations of impoliteness.In addition, evaluations are usually subjective and differ from person to person, relation to relation, and situation to situation.Thus, (im)politeness can be understood as a form of social practice.
According to Kádár and Haugh (2013: 61), there are four dimensions of interpersonal evaluations: persons and relationships, categorization, valency, and normative frame of reference.The first dimension refers to the importance of how one conceptualizes a person and the type of relationships in which a person engages.These relations differ across societies and individuals.In the context of interpreter-mediated institutional encounters, there are at least three persons (i.e., communication participants) with different types of functions, responsibilities, and expectations.By definition, institutional encounters are asymmetrical because public service employees are given power by the state to decide upon goods and services.However, the employees' power is limited by the rules of conduct.How much one can bend the rules-that is, how much personal elements influence professional relationships-is a matter of individual, cultural, and situational context.All of this is likely to influence the evaluations of impoliteness (for a discussion of culture and evaluations of impoliteness, see Spencer-Oatey & Kádár, 2016).
Interpersonal relations in interpreter-mediated encounters involve categorizations about the kind of relationship one should expect from an interpreter, at the personal and professional levels.This implies that all three parties categorize the encounters in their own ways and that their categorizations do not necessarily coincide.Categorizations are valenced; that is, they vary on a scale from good to bad (Kádár & Haugh, 2013: 62-63).Valency can be related to the following example experienced by one of the authors of this article that shows how categorization of one event can be evaluated differently in distinct situations.The author and her friend were talking in Montenegrin in a shop in Norway.A fellow shopper who happened to speak the same language made a comment in Montenegrin: "It is so good to hear our language spoken here in Norway!"In another setting, such as a workplace, speaking a foreign language could have been evaluated as an act of impoliteness or exclusion by Norwegians who do not understand the language being used.The final dimension of interpersonal evaluations, the normative frame of reference, exists at the social and situational levels.The normative frame of reference is "the perception that others from the same social group would evaluate a person or relationship in the same way" (Kádár & Haugh, 2013: 64).Interpreter-mediated encounters, as a specific kind of communication, often require ad hoc negotiation among communication participants about understandings of language and communication, the public sector, and the interpreters' function.The general frames of reference (including the normative frame of reference) of the communication participants are often different, and this is related to several factors: membership in different communities of practice, a shared set of practices (or not shared), and diversity of the communication participants.
First, interpreters in the public sector in Norway represent a community of practice in making (Skaaden, 2016); this is in contrast to conference interpreters, for example, who are referred to as a community of practice in existing research (Magnifico & Defrancq, 2016: 28).In contrast to conference interpreters, interpreters in the public sector seldom work in teams, have various levels of competence, and are not dependent on each other (except in rare cases when they work in booths).Furthermore, they seem to have different sets of practices, and they develop various discourses about their professional activity, as we show.Registered interpreters are obligated to follow ethical guidelines, although these guidelines leave some room for various interpretations, which again can result in various practices.
Second, employees in the public sector represent a range of communities of practice, each with its own sets of practices.Not all are trained to communicate via an interpreter, which leads to the development of various interpretations of what an interpreter's function should be.Regarding behavioral norms, all employees in the public sector are obligated to adhere to ethical guidelines that, among other things, state public service officials should be polite to the public (Ethical Guidelines for Public Service, 2006: 5).
Third, public service users (i.e., speakers of minority languages) come from different countries and different social groups.Users' experiences with interpreters vary considerably, and one can assume that users also have different expectations and understandings of an interpreter's function.
In public encounters with these different parties, moral order plays a significant role.All evaluations are based on people's ideas about what particular situations and relations should be like; in other words, evaluations are premised on assumptions about a certain moral order that frames the situation.Following the ethnomethodologist Harold Garfinkel, Kádár and Haugh (2013: 67) defined moral order as what members of a sociocultural group or relational network "take for granted", or what is referred to as the "seen but unnoticed", expected, background features of everyday scenes.(Kádár and Haugh 2013: 67) Communication participants usually have difficulty identifying and expressing these "seen but unnoticed" expectations.In institutional encounters, each communication participant evaluates (im)politeness based on his or her own expectations of what such an encounter should look like, including how other participants should behave.Because communication participants have different backgrounds, the participants' expectations often do not coincide.Furthermore, it is sometimes difficult to recognize or anticipate the other person's expectations.However, it seems that the social actions on which moral order is based are usually understood by members of the same group, be it a community of practice or some other type of group (Kádár & Haugh, 2013: 67).

Data
The data were obtained through an online survey that contained eighteen questions and was conducted in LimeSurvey.The survey was sent to 1,174 interpreters registered in the NRI in 2014.We received twenty-eight answers in the pilot study and 286 in the main study.These answers represent approximately 27% of the study population.The respondents represent interpreters of more than fifty languages.Their answers were mainly in Norwegian, with some examples in other languages.
Because we use elicited data, in this study we cannot discuss moral orders in interaction or naturally occurring scenarios.Furthermore, our data do not reflect interactional valences.However, we analyze assumptions about moral orders by studying the interpreters' retrospective views represented in the form of the interpreters' metadiscourse.
The interpreters participated in the study on a voluntary basis.The questionnaire and participants were anonymized, and the participants consented to the research.The survey design incorporated yes/no, Likert scale (ranked 1- 6), and open-ended questions in Norwegian.The open-ended questions were designed to elicit interpreters' subjective responses.The survey was divided into three sections: (a) basic information about the interpreter, (b) the interpreter's experience with politeness and impoliteness, and (c) the interpreter's strategies for interpreting impoliteness.The third section was aimed at encouraging respondents' reflections on concrete situations, comments, and examples.Many respondents sent incomplete responses: They provided responses to the questions in which the Likert scale was used and no responses to the questions to which definitions, descriptions, or reflections were required.
This article mainly addresses the third section of the questionnaire.The survey results were analyzed by conducting a content analysis of the interpreters' answers and comments.
We analyzed the answers to the survey question concerned with solutions and strategies that interpreters use while interpreting what they consider impolite: "Describe some of your solutions or strategies you use in interpreting impoliteness."We used two words in this question.The first, solution, can be considered neutral in the sense that it does not imply a well-developed plan made in advance with a particular goal in mind (which the second word, strategy, can imply but does not necessarily do).We did not provide a definition for solution and strategy1 in the questionnaire (for more about strategies, see the following section).
The actual responses given by the interpreters formed the basis for our categorization and labeling of strategies.We tried to find common denominators for the reported examples that, in our understanding, could be classified in the same group.This means that in the accounts the interpreters did not necessarily mention the strategy of using equivalents: What we label as using equivalents was described by interpreters as, for instance, "interpreting directly," "interpreting both the content and tone," and "transferring everything exactly."

Interpreters' reflections about their interpreting strategies
In this section, we first discuss the notion of strategies and then analyze the survey results.

Strategies
In order to obtain a broader picture of what interpreters do when they evaluate an encounter as impolite, we deliberately excluded the definition of the term strategies from the questionnaire so that the notion of strategy could be open to different interpretations.In this study, strategy was not intended to mean a pre-defined or well-developed plan of action made before interpreting.Instead, we conceived of the term was simply as a solution, decision, In translation and interpreting studies, the term strategy is used in different contexts.An example of a strategy employed in the translation process could be a device that solves "the problem of bridging the gap across cultures," that is, a problem that occurs when a source language has a term for a phenomenon that is known in the source culture but unknown in the target culture (Fernández Guerra, 2012).Chesterman (2000, Memes Are Conceptual Tools section, para.5) defined a translation strategy as "any well-established way of solving a translation problem," emphasizing that strategies are potentially conscious and intersubjective.The term has been used in a similar manner by, for instance, Wang (2012), who stated: [s]trategies in interpreting refer to the tactics interpreters employ to deal with problems in cognitive processing and inter-lingual and intercultural communication in interpreting."(Introduction section para.2) We analyzed a total of sixty-nine answers that thematized strategies interpreters reported employing when dealing with what they perceive as impoliteness.2Some of the strategies we identified based on the interpreters' feedback overlap with those identified in the research conducted by Magnifico and Defrancq (2016), whose study on how conference interpreters handle interpreting face-threatening acts (FTAs) identified the following interpreting strategies: omission, downtoning, translation, strengthening, and addition.In addition to the strategies discussed by Magnifico and Defrancq (translation, which we call using equivalents, omission, and downtoning), we identified two other strategies: meta-commenting and interrupting and postponing interpretation.Because these two additional strategies are on the border between interpreting and behavioral strategies (they show how interpreters behave or cope with impoliteness), we label them coping strategies.
The study results suggest that interpreters' responses and reflections about their own interpreting strategies were influenced by their understanding of the concept of strategy, that is, what they do or do not do.This understanding was clearly indicated in some responses.More concretely, some responses revealed that interpreters were unsure how to understand the term strategy (see (a), below); other responses pointed to interpreters' understanding of strategy as "a very specific method" (see (b)), and in some responses, strategy was related to interpreters' meta-representation of an implicit moral order requiring "diminishing of impoliteness" (see (c)): (a) Strategy . . .if one can call it so . ..3(b) None of the other strategies were used apart from interpreting all of the things expressed.
(c) I do not have any strategies to downtone impoliteness.
Interestingly, some respondents reported the absence of any strategy, although at the same time they reflected on concrete solutions in interpreting impoliteness.For example, one interpreter started his comment with the following reflection: "I have not been in need of any strategy, and I do not see any need for it."However, in the continuation of his reflections, this interpreter mentioned the method he usually employed when dealing with what was considered impolite.
Before continuing with the description and analysis of the reported strategies, a note about how interpreters define (im)politeness is in order.The interpreters defined (im)politeness in general and specific terms.Generally, they defined politeness as, for instance, thanking, opening doors, using the formal address form, offering coffee or water, and receiving information about breaks, whereas they perceived impoliteness as the opposite of politeness, including disrespect for others, lack of good manners, and lack of friendliness.Specifically, impoliteness is defined as the act of not respecting the interpreter's profession, that is, the act of not allowing interpreters to do their job before, during, and after an assignment.Thus, general terms include personal, everyday experiences, whereas specific terms have to do with being an interpreter and a professional (see Felberg, 2016, for a discussion of face issues and interpreters' definitions of impoliteness).
In the following section, we reflect on the strategies we identified in our study, starting with the one termed using equivalents.

Using equivalents
Equivalence is not a straightforward notion: It is related to a number of issues, as well as strategies or methods of rendering an utterance in a communicative event.The idea of equivalence is one of Chesterman's (1997) translation supermemes, that is, pervasive ideas that repeatedly appear in the history of the subject.It is still a major translation theory topic (Baker, 2011).Although largely absent from the discourse of interpreting . . .[equivalence] seems to have been a tacit assumption underlying much work on "accuracy" and "errors" in conference interpreting research.(Pöchhacker, 2009: 52) In translation studies that rely heavily on the concept of equivalence as a phenomenon existing at different language levels, "strategies" should not be understood as ideal solutions but as solutions that are actually used and have advantages and disadvantages.
As Hale (2007) explained in her analysis of challenges that court interpreters face when attempting to interpret truly and faithfully, the concept of faithfulness or accuracy of interpretation is inextricably linked to the context of the communicative event.This means that a literal word-for-word translation does not necessarily imply faithful rendition.Nonetheless, many interpreter-users have difficulty understanding and accepting this idea, and thus, interpreters are often told to interpret "word for word."Even translations that, on the surface, seem to be accurate do not necessarily render all of the aspects of the meaning of the original utterance.A relevant example was given by Kádár and Ran (2015), who discussed different ways in which the metalexeme "heckling" is translated and appropriated in Japanese and Chinese: The authors rightly pointed out that the process of translation involves intercultural appropriation.
Regarding equivalence, Hale (2007) emphasized the difference between semantic and pragmatic translations.The former transfers the semantic meaning of utterances, looking at utterances at the sentence level but out of context.Although grammatical and comprehensible, such semantic transfers are not necessarily accurate.In pragmatic translations, the entire message of the utterance beyond the sentence level is considered, as is the intention behind the message.Thus, such translations require a thorough analysis by an interpreter that needs to consider not only the grammatical differences across languages, but also the cultural and pragmatic differences reflected in aspects such as different social conventions, ways of expressing politeness, and forms of address according to participants.(Hale, 2007, Understanding the Meaning of Faithfulness section, para.7) The witness's remark in Spanish included a neutral tone and the imperative mood, and was intended as a request: It was not intended to be impolite or aggressive.However, the interpreter either misunderstood the original intention (a request) or was unable to reproduce it in a pragmatically appropriate manner in English, and used the imperative in addition to putting stress on the pronoun in the imperative phrase (which is a sign of impoliteness in English).Not only does this example illustrate the difficulties with finding equivalents (which is one of the topics of interpreters' reflections on their interpreting solutions in our material), but it also emphasizes the difficulties of another phenomenon worth examining: how a specific interpreting solution can produce impoliteness effects.
In this respect, body language may also constitute part of an original message indicating emphasis, hesitation, irritation, and so on.Several interpreters mentioned body language as an important element in their evaluation of impoliteness.
Using equivalents is the strategy most widely reported by the study respondents.In thirtynine examples, the respondents elaborated on the use of the strategy in question (see, e.g., 1 and 2): (1) Everything should be interpreted.If a person is swearing, I have to say the same thing in Norwegian.If someone uses bad words, I have to translate them in the same manner.
(2) I constantly try to interpret all that is being said, both good and bad . . .I monitor exactly everything that is being said, and I transfer everything exactly.
In some examples, additional reflections and details show which phenomena interpreters perceive as impolite.Some respondents provided examples by quoting parts of the utterances made by interpreter-users (see 3 and 4).As can be seen below, impoliteness is, for instance, related to swearing in (3), "content and tone" in (4), and "words and expressions" in (5).
(3) It is most often swearing [I interpret].For example, when the accused said to the judge in his native language "blet, tu meningitas" this was interpreted directly as "Faen, for en mongo."(Fuck, what an idiot.) (4) An interpreter-user said, "I do not want to hear your nonsense, can't you tell the truth?"My strategy is to interpret both the content and tone.
(5) One interprets [impoliteness] the way it is [expressed] if one can do it by using words and expressions with a similar meaning and goal.
In addition, some respondents reported certain behavioral phenomena other than verbal communication as realizations of impoliteness.These phenomena are participants' impatience or irritation in the communicative event, and their body language (see 6 and 7): (6) I have often experienced that the suspect / the accused expresses impoliteness through impatience or irritation during the hearing.
(7) Impoliteness from persons with authority towards the client is rarely verbalized, but it appears through the way something is said and/or body language that shows impoliteness, or offence towards the subject.
Examples (1-7) show that, in their perceptions and evaluations, interpreters link impoliteness to various language units (such as words and phrases) and semantic subsets of such units (such as swear words), but they also relate impoliteness to how these units are realized in a concrete communicative event by a concrete user (e.g., the user's accent and pitch).Furthermore, the interpreters connect impoliteness with behavioral phenomena, such as impatience, irritation, and body language.Most of these constructions of impoliteness that are related to utterance realization and body language were mentioned in the context of using equivalents as an interpreting strategy (see 8 and 9): (8) I try to imitate the "tone" and "way" something is being said . . .my role is to be a parrot, so I repeat as precisely as possible and with as similar body language as possible-in both directions.
(9) I try to find equivalent expressions and body language, tone, and the voice value.
However, some respondents reported other strategies they employ when dealing with impoliteness, such as euphemization.This strategy is addressed in the "Modifying Interpreter-Users' Impoliteness: Downtoning" section.
The discussion and examples given thus far show that the strategy of using equivalents is in line with the ethical guidelines for interpreting in Norway.The notion of interpreters rendering everything faithfully enacts the moral order of not interfering in the communication of the primary communication participants.This is why interpreters do not take a stance on their own evaluation of impoliteness.

Omissions
A loss of information that occurs while interpreting from the source to target version is usually referred to as omission (Englund Dimitrova, 1995).Wadensjö (1998: 107-108) defined several types of "renditions" in the dialogic interaction that can be considered omissions.They are reduced renditions (information is less explicit than in the original), summarized renditions (the original contains more information), and zero renditions (the original utterance is not translated).Napier (2004) identified five types, of which four (b-e) imply a loss of meaningful information: (a) conscious strategic omissions (interpreters make a choice based on their knowledge and omit redundant information), (b) conscious intentional omissions (interpreters choose to omit something due to, for example, the lack of understanding a lexical item), (c) conscious unintentional omissions (interpreters postpone interpreting an item, and forget it later), (d) conscious receptive omissions (due to poor sound quality), and (e) unconscious omissions (the interpreter does not recall hearing the particular lexical item).
In interpreting studies, omissions are frequently studied under the heading of deviations, errors of translation, and discontinuities (Barik, 1975/2002, Gerver, 1969/2002;cited in Pöchhacker, 2009: 142).However, some researchers emphasize the pragmatic dimensions of omissions: Omissions can be related to an interpreter's wish to produce more coherent speech and/or to reduce redundancy (Viaggio, 2002;Visson, 2005).
Omissions are a dominant topic in the Norwegian ethical guidelines, which stress that nothing should be omitted in interpreting.As already mentioned, the majority of respondents claimed that they follow the guidelines mentioned above.However, the wording and the idea that "nothing" should be omitted bear traces of a monological and linear view of language premised on the position that meaning is constructed by each communication participant alone.This view has been challenged in the literature on interpreting by a dialogical understanding of language and interpreting-interpreting is the product of all of the participants in the communication (Wadensjö, 1998).The study results reveal that only a few respondents reported rendering perceived impoliteness with omissions (see 10): (10) My usual solution is to ignore the impolite words and forget it immediately after I finish interpreting.(Felberg, 2016) Respondents reported they used omission as a deliberate strategy, and in a few cases, they considered it a result of evaluating the interpreter-users' intentions ( 11): (11) I believe that I possibly ignored indirect impoliteness at times if I understood that the speaker's intention was not to be impolite, and that his ignorance could have an unintended effect on the hearer.
The reflection in (11) indicates that the interpreter differentiates between indirect and direct impoliteness and that he or she chooses to ignore indirect impoliteness.Interestingly, the interpreter links these categories not to language means but to speakers' intentions and considers intentional impoliteness to be more problematic.That consideration seems to be the primary reason for choosing omission as a strategy.In making assumptions about moral order, protecting hearers from statements that may exert potentially undesirable effects is the right thing to do.The same respondent reported rendering swearing (which would, in his or her understanding, be direct impoliteness because it is intentional) by using equivalent forms of particular expressions all the time.
A respondent described a type of omission as a technical necessity in one case ( 12): (12) The only cases in which I refrain from interpreting is when a flood of words happens and everyone talks at the same time.It is not necessary to try to interpret anything in these cases [including impoliteness] because no one listens.
Summarizing, according to the interpreters' reports, omission is a strategy they use infrequently.In a few responses in which interpreters reported "zero renditions," this strategy can be related to Napier's (2004) conscious receptive omissions and conscious strategic omissions (see 11).

Modifying interpreter-users' impoliteness: Downtoning
Interpreters reported and reflected on the strategy of downtoning in a few examples.In several examples, the interpreters modify interpreter-users' perceived impoliteness as euphemization or downtoning: (13) I interpret rude remarks "away" . . .I nuance the language use and choose a formulation that is less rude.But there is a limit to this, and it must be in accordance with body language and voice.
As can be seen in ( 13), the second part of the explanation ("but there is a limit to this") indicates the interpreter's representation of a moral order that requires him or her to use euphemization as a regular method or a norm when interpreting impoliteness.Moving on to (14), another respondent related impoliteness to the expression of emotions.More precisely, he or she reported using a euphemization strategy making it possible to render the message but lessening its intensity.The interpreter justified the use of the reported strategy by relating emotions and impoliteness: (14) . . . the interpreter should mediate them [emotions, impoliteness] by his word choice, body language, and word accent, but not necessarily with the same intensity.(15) I raise my voice a bit higher, but not as high as the one talking.
As the interpreters did not provide examples of willingly escalating impoliteness, we concluded that the main moral order underlying the strategy of downtoning is "politeness is a norm."Apart from understanding downtoning through the moral order of "politeness being an expected norm," one can approach these examples from the standpoint of the interpreters' (lack of) awareness of the ethical guidelines.Even though all interpreters registered in the NRI agreed to follow the ethical guidelines, actual use in practice varies.This topic should be explored in face-to-face interviews with interpreters in future research.

Meta-commenting
Meta-commenting refers to interpreters' method of providing comments about what they perceive as impolite language instead of interpreting it.Respondents reported that they used this solution infrequently.Thirteen responses signalized or elaborated on use of this strategy, some of which also indicated the use of other strategies, although the interpreters reported meta-commenting was the central one.According to the interpreters' reflections, the interpreters realize meta-commenting in verbal language and other means of expression, such as gestures.
In some cases, the subject of meta-comments was impolite language, as in ( 16): (16) I say often that "he directs bad words towards you," and then I hear that I do not have to translate that.
Meta-comments on impolite language can, in theory, resemble the strategy of paraphrase, which is used in translating written texts: The disadvantage is the inability to convey expressive, evoked, or associative meaning.That strategy is also cumbersome and awkward because it involves filling a short slot with a (much) longer explanation (see Baker, 2011: 41).The data contain too few reflections on meta-commenting related to language to draw extensive conclusions on similar matters.Finally, meta-commenting conflicts with the ethical guidelines to which interpreters are obligated to adhere.
Meta-commenting on impolite language was also realized as an announcement to a participant in the interaction of what was perceived as impolite; that is, by meta-commenting, the interpreter warns other interaction participants about perceived impoliteness (see 17): (17) The most usual thing is to say in advance, before interpreting, that what was uttered can hurt the person who hears the interpreted utterance.
A topic elaborated in the context of meta-comments is personal identity versus professional identity.In an example in which an interpreter reported on a situation in which impoliteness was introduced and (one of the) interaction participants was warned, the interpreter emphasized his or her professional role in a situation with perceived impoliteness: (18) I started the sentence as follows: "The interpreter apologizes; everything has to be interpreted the way it is said"-and then I interpreted everything that was said.
In (18), meta-commenting is reported as an additional strategy that supplements the reported strategy of using equivalents ("I interpreted everything that was said").
In some examples, meta-commenting is realized by using body language, as (19) illustrates: (19) Because it is explained in the guidelines, I try to use the same [language], but when I interpret very impolite language to a female interpreter-user, I show with my body language/gestures that I have to interpret something distasteful.
Interestingly, the report in ( 19) related (the need for) meta-commenting to the gender of the communicative situation participants.
The relation between professional and personal identity was also a topic of some interpreters' reflections, in which impoliteness phenomena (or the absence thereof) are linked to the interpreter's personality: (20) I reproduce the tone as well I can in translation.Personally, I do not get carried away.I have a courteous and calm tone of voice, so I seldom end up in such situations.
Interpreters' reflections, in which the interpreters reported meta-commenting as an interpreting solution, indicate a conflict between different constitutions of the moral order in the same situation.The moral order that the interpreter appeals to is based on the ethical guidelines-the interpreter is concerned with the principle of fidelity and aims to follow it; and at a meta-level, he or she confirms this aim to other interaction participants (either by issuing a "warning" or reminding them of his or her role).However, this very act (warning, explaining the role, etc.) is an appeal to the moral order based on an assumption of what is good and positive (politeness) and what is bad or negative (impoliteness).

Interrupting and postponing interpreting
A few respondents reported using the strategy of interrupting the interpreting.The interpreters pretend that they did not understand an impolite utterance (see 21): (21) I say that the interpreter did not understand [in order to give them time to reconsider].
By doing this, interpreters give interpreter-users more time to reconsider and reformulate their original utterance, as indicated in ( 22): (22) A child welfare officer was called a whore; the interpreter asks: "What did you say?"The interpreter-user says, "nothing," and the interpreter says: "The interpreter did not understand what was said." In some cases, the strategy of postponing is related to meta-commenting about the interpreter's role: Interpreters explain their role, non-partiality, and the need to interpret everything that is said again, at the moment of a perceived conflict.In other words, they remind the interpreter-users of what they had said when presenting the interpreter role before the interpreting even started, as illustrated by ( 23): (23) I explain the interpreter's role once again by saying "the interpreter is impartial and must interpret everything.That means everything that you say.It is you who are supposed to think about what you say.Not the interpreter."Oddly enough, this helps!As indicated below, some accounts showed a combination of different strategies.In a few examples, interpreters reported using a combination of an adequate translation (close to the original) of the impolite expression (using equivalents) and meta-commenting (see 24 and 25).To sum up, the respondents did not frequently report interrupting and postponing interpreting as a strategy for coping with what the interpreters evaluated as impolite encounters.In the few cases in which interpreters reported using this strategy, they postponed interpreting in order to give other communication participants time to reformulate their utterances.The interpreters occasionally used this strategy with another strategy or related interrupting and postponing to meta-commenting about an interpreter's role.

Discussion
Interpreters' evaluations of (im)politeness, similar to all evaluations, involve an implicit appeal to the moral order (Haugh, 2013) and the practices that give rise to it.Haugh emphasized that group members who identify with different groups do not necessarily constitute the moral order in the same way.Interpreters in the public sector are a professional community of practice in making, and one can expect them to identify with that community and its moral order, which is based on ethical guidelines.However, interpreters' descriptions of solutions that they reportedly use in interpreting impoliteness indicate that interpreters, although theoretically in the same community of practice, constitute the moral order in different ways.The moral order that many respondents implicitly appeal to in their evaluations is related to the ethical guidelines that promote fidelity: Everything that is being said should be interpreted in the way it is being said.In that way, interpreters do not interfere in the communication between the parties.These ethical guidelines provide the cornerstone of the interpreters' moral order to which they as a community of practice members appeal.However, within the same community of practice, the moral order is constituted by some members for some norms and expectations that are beyond the guidelines and beyond the rule of conduct of the community of practice.For that reason, interpreters' evaluations implicitly appeal to what is considered good and a norm in a given broader community, and/or society in general.When this happens, interpreters interfere in the communication between the parties involved.In other words, there are different assumptions about moral orders, which are reflected in the need to interfere or not interfere in communication: Not interfering: interpreters should render everything that is being said the way it is said (this implies no stance on the evaluation of impoliteness).
Interfering: interpreters should tone down impoliteness (this implies a stance on the evaluation of impoliteness).
The idea underlying assumption (b) is that politeness is good or positive and impoliteness is bad or negative.Therefore, interpreters who appeal to that aspect in their evaluations report reducing the intensity of impoliteness or entirely ignoring what they perceive as impolite; they consider doing this advisable and in accordance with the broader moral order.

Conclusion
According to interpreters' retrospective reflections, politeness seems to be a ruling norm in interpreter-mediated institutional encounters in Norway.However, interpreters occasionally encounter the need to interpret impoliteness.They report using different solutions to do so, including using equivalents, meta-commenting, omissions, downtoning, and interrupting and postponing interpreting.Even though meta-commenting and interrupting and postponing interpreting are not categorized as interpreting strategies in the literature on interpreting, some interpreters continue to perceive them as strategies.We labeled these two strategies coping strategies in order to differentiate them from interpreting strategies.However, interpreting and coping strategies overlap and signal the main dilemma that interpreters face when they encounter impoliteness.This study interpreted different solutions as interpreters' appeals to different moral orders.Interpreters appeal to a certain moral order every time they refer to ethical guidelines and the main principles of fidelity and neutrality.This means that only one of the strategies described-that of using equivalents-can be directly connected to that moral order.However, by appealing to another moral order, interpreters excuse themselves for not "repairing and downtoning" impoliteness.This is motivated by the assumption that the interpreter should downplay impolite speech and behavior, which implies that politeness is perceived as a default, ruling norm.The four remaining strategies mentioned above-meta-commenting, omission, downtoning, and interrupting and postponing interpreting-can be associated with this moral order.
Having to balance various strategies based on different moral orders, especially if one is not aware of this balancing act, can lead to personal and relational conflicts.In personal terms, interpreters may feel the tension between their personal and professional identities.In relational terms, interpreter-users and interpreters may experience conflicts when they implicitly appeal to different moral orders and evaluate others' actions based on these.