4

Participation Frameworks

The Role of the Interpreter

IN ORDER to examine the sociolinguistic question of whether an interpreter can interpret interactive discourse without influencing it, this study has applied frame theory to two cases of interpreted encounters. Interpreters, like the other participants in the two cases under examination, bring their own individual frames and related schema to interpreted encounters. In each case, the mismatches in schema among participants were related to the interpreted encounter frame. The linguistic evidence of these mismatches, including pronominal reference, raises questions regarding the ways in which participants’ utterances impact their relationship to, and perceptions of, one another. The relative participant structures that occur within an interpreted interaction can be examined in terms of Goffman’s concept of footing (1981).

Participation Frameworks: Frames As Footing

Linguists have focused on essentially two aspects of frames. Some scholars, such as Fillmore (1976) and Tannen (1979), allude to frames in terms of activities, as in commercial transactions and the film-viewing frame. Frames have also been referred to in terms of participants, as in Fillmore (1976) and Chafe (1977). These two constructs are consistent with Goffman’s discussion of the framing of events as multilayered and complex (1974); an event, such as a fight, can be framed as a transformed activity (i.e., play) and on the basis of participants (i.e., player vs. onlooker). It is the latter of the two, the participation framework, that provides a foundation for Goffman’s notion of footing.

Goffman defines footing as “the alignment we take up to ourselves and the others present as expressed in the way we manage the production or reception of an utterance” (1981, 128). To clarify this definition, Goffman points out that traditional views of interaction, based on speaker-heaier dyads, are too simple to describe real interactive discourse. For example, although a ratified addressee might be identifiable, there might also be unratified addressees (bystanders) who access a conversation. Similarly, a speaker can be discussed in terms of more than one simple role.

Hearer roles include ratified and unratified status, and the question of whether or not someone is an addressed recipient. Speaker roles refer to the person who is the principal, author, and animator of utterances. Understanding these relationships within interactions can help clarify what linguistic evidence of footing might look like, and when shifts in footing occur. Ratified hearers are those who have an official place within a social encounter (Goffman 1981). Once ratified, an individual can choose whether or not to attend to the discourse. In addition, participants can choose to address certain individuals, leaving other ratified participants as unaddressed recipients. Thus, Goffman makes two relevant hearer distinctions: ratified-unratified addressee, based on access to official status within an encounter, and addressed-unaddressed recipient, which is reserved for ratified participants. The remaining possible hearer status, then, is the unratified participant who has access to the encounter. Goffman refers to such an individual as a bystander. Although Goffman discusses additional complexities for hearer status, by considering a variety of social situations, the divisions discussed here are sufficient for the purpose of this study.

Goffman describes three roles that a speaker can fulfill: animator, author, and principal. These three roles are not necessarily satisfied by the same person at the same time. A speaker generally functions as an animator, the “talking machine” (144) that actually produces an utterance. A speaker can also function as an author of an utterance, the originator of the content and form of the utterance. While a speaker can function as both animator and author, it is possible for a speaker to function only as animator. For instance, when an actress speaks the lines of Shakespeare, she functions as the animator, but not the author of her utterances. Finally, Goffman describes the role of principal as the one who is responsible for or committed to what is being said. Thus, when someone reads a statement as a stand-in for a political figure, it is presumably the politician whose views are being expressed; the politician is, thus, the principal. By distinguishing among these three roles, what Goffman calls the production format, it can be seen that a speaker’s relationship to utterances and addressees can be quite complex.

Understanding the production format within an encounter offers a way to analyze a speaker’s alignment with other interlocutors. Each potential alignment represents a unique way of framing the encounter. For example, by quoting someone else’s words, a speaker can imply a lack of responsibility for the content, denying principal status, even though the speaker is responsible for deciding the comment was worth reporting. Tannen (1986) provides a clear example of this. When one person tells another about a comment or criticism made by a third person, it is generally the third person who receives blame for the comment, rather than the one who has repeated it in the new context. Thus, it is the principal who is held accountable, rather than the animator. Evidence of various alignments is available through linguistic analysis. In fact, Goffman (1981) suggests that changes in footing are often evident through paralinguistic features of discourse. Researchers have applied the notion of footing to interactive discourse to identify a variety of linguistic features as evidence of footing and footing shifts.

Footing in Discourse

Gumperz (1982) discusses a variety of features that provide information regarding the footing between a speaker and addressees. In Gumperz’ example of an African American graduate student who code-switches from Standard English to a black dialect, he points out that the student utters the sentence “Ahma git me a gig” with the singsong rhythm of a stereotyped African American character. Thus, the student is actually mocking his own role and making it clear to insiders that he is totally in control of his own situation (34). By borrowing the dialect and rhythmic prosody of a stock character, he is, in a way, not the principal of the utterance. This is an example of how code switching can represent a shift in footing.

Footing shifts have also been described as changes in participant frames. In examining a pediatric medical interview, Tannen and Wallat (1982, 1983, 1987, 1993) identify these footing shifts through changes in register. The pediatrician uses “motherese” with the child patient, a conversational register with the mother, and a reporting register to the video camera (which is recording the encounter for use by doctors-in-training). Goffman (1981) refers to this work as evidence of the complexities of footing shifts in interactive discourse.

In an examination of footing in news-interview discourse, Clayman (1992) found that interviewers maintained their own uninvolved neutrality by attributing strong opinions to others through the use of constructed dialogue of unnamed parties. Examples of this include indefinite or unspecific noun phrases, as in “It is said …” (170) and “critics on the conservative side have said …” (171). Evidence that interviewers are striving to remain neutral is based on the occurrence of, and self-repairs toward, such constructions.

In an examination of footing within sermons, Smith discusses the use of a variety of forms, including pronouns, rhetorical question-answer sequences, and discourse markers (1993, 160). An example of the influence of pronoun reference occurs when preachers say “I think” or “It’s interesting to me” versus “We can see.” Smith suggests that the use of first-person singular identifies the preacher as author of the utterance, while use of first-person plural implies that the preacher speaks on behalf of the audience as well. The various strategies addressed by Smith appear to demonstrate ways in which preachers represent roles in the preaching task (for example, whether or not they present themselves as an authority or mediator).

The role of pronoun reference in footing shifts has also been found in the examination of footing in discourse of boys engaged in “sportscasting play.” Hoyle (1993) describes shifts between first- and third-person pronouns as evidence that the boys are shifting footing. As they play Ping-Pong, the boys comment on the game as if it were a tennis match on television and, in so doing, refer to themselves in the third-person during “sportscaster talk” segments in the data: “They’re hitting it back and forth!” (117). The occurrence of questions, response cries, asides, and explicit frame-bracketing with terms such as “Time out” all provide evidence of shifting footing between the boys animating their own utterances and animating utterances of an imaginary sportscaster.

The concept of speaking for another is explicitly addressed in Schiffrin (1993), in an analysis of sociolinguistic interviews. Schiffrin discusses shifts in footing as interactional moves. That is, in two examples in which one person speaks for another, one can be seen as helpful and supportive and an extension of help provided in daily activities. Another example can be seen as “putting words in someone’s mouth,” sharing information that might have been private, and seemingly doing so for some benefit of the speaker rather than the spoken for (238–39). Schiffrin demonstrates that just as linguistic markers might identify shifts in footing, footing shifts themselves can assist in the understanding of discourse.

Locker McKee (1992) examines footing in ASL lectures using Goffman’s (1981) discussion of lectures as a base. Locker McKee addresses two types of footing in ASL lectures: quotations and asides. Locker McKee discusses many linguistic and paralinguistic cues that mark shifts in footing, including body leans and stepping to the side as spatial markers of changes in footing, and eye gaze to specify a particular addressee for asides. The use of performatives (for constructed dialogue), the lexical marker QUOTE, code switching, and prosodic changes are indicators of quotations, a shift in which the signer is animating an utterance attributed to another author. Locker McKee also discusses the use of STOP (this gloss referring to the one-or two-handed sign in which the signer’s palm is forward facing the addressee) and INDEX-HOLD as discourse markers used to identify changes in footing.

Footing within English and ASL can be identified on the basis of prosodic, lexical, and other features. While these languages exhibit parallel features in this regard, the linguistic and paralinguistic markers within each language are different. These differences make ASL-English interpreted discourse a unique type of interaction in which to examine footing.

Footing in Interpreted Encounters

Goffman discusses the various roles that a speaker can fulfill in his description of production format. The production format involves unique dimensions when applied to the task of interpreting, a situation in which one individual relays the utterances of others. Few have attempted to analyze footing in interpreted interaction. However, Keith (1984), Edmondson (1986), and Wadensjö (1992) address the notion of production format in interpreted encounters, although they bring different perspectives to the task.

The notion of footing as applied to interpreting by Keith (1984) is very focused on the task of interpreting itself. Keith suggests that an interpreter operates within two distinct footings: translation of utterances and comprehension of utterances. The latter might result in requests for clarification or repetition, whereas the former refers specifically to rendering meaning equivalents. In this way, Keith seems to separate footings on the basis of authorship. That is, when translating, the utterance originates within someone else. When requesting clarification, the interpreter is the original author of the utterance.

Edmondson (1986) attempts to apply Goffman’s conception of the three speakers’ roles—animator, author, and principal—to the process of interpretation. In addition, he discusses Goffman’s identification of hearer roles, addressee, hearer, and overhearer. Edmondson suggests that while interpreters are responsible for the formulation and production of utterances, they are not responsible for utterance meaning. Edmondson concludes that interpreters are not involved in interactions, and that they are neither speakers nor hearers; rather, they depend on a completely unique cognitive process that requires both speaking and hearing be accomplished simultaneously.

Wadensjö draws a different conclusion. In her data-based research, Wadensjö discusses the interpreted encounters as “conditioned by the co-presence of at least … three persons, and one of these (the Dialogue Interpreter) characteristically relays between the others” (1992, 65). The interpreter’s talk is analyzed as two types, relaying and coordinating talk. Through an examination of the relaying done by interpreters, Wadensjö finds that interpreters’ renditions sometimes closely parallel an original participant’s utterance, sometimes contain somewhat more or less information than the original, sometimes summarize prior talk, and sometimes the interpreter’s utterance is not based on a prior utterance. Based on the various types of renditions, Wadensjö concludes that interpreters do not function simply as “translation machines” (72).

In examining an interpreter’s coordinating function, Wadensjö again identifies a taxonomy. For example, an interpreter might ask for clarification, prompt a response or turn from a primary party, explain what one party or another means, or explain that one party does not appear to understand another. In addition, Wadensjö points out that an interpreter influences the coordination of talk simply by relaying utterances; the course of a conversation is influenced, in part, by the content and form of the interpretation.

Footing, as it has been applied to interpreted interaction, can be seen to reflect two different views of the interpreter. Edmondson’s work, which is not data-based, seems to reflect the more traditional notion of interpreter as an uninvolved relayer of messages. Keith and Wadensjö seem to view interpreters in a more interactive light, shifting footings as they attempt to comprehend and relay conversation. The ways in which interpreters negotiate footing shifts is still a relatively unexplored area.

Footing within Interpreter-Generated Utterances

Research regarding footing in English and ASL discourse provides some examples of the types of linguistic evidence of footing that can be found in conversational interaction. Research regarding the potential footings in interpreted encounters has focused, in part, on the various potential footings within an interpreter’s talk. For example, Wadensjö (1992) describes an interpreter’s interpretation as one of two types of utterances that are generated by interpreters. She describes interpretations as renditions, indicating that while a rendition has originated outside of the interpreter, the interpreter still authors the form and content of that rendition (in the choice of form, for instance). Just as an interpreter provides renditions of what participants say, an interpreter can also omit another participant’s talk. Wadensjö refers to this as a lack of rendition. A lack of rendition is distinguishable from the second type of utterance identified by Wadensjö, which is nonrenditions. Where a lack of rendition refers to an omission, a nonrendition is an additional utterance, generated by the interpreter, that has not originated with anyone else.

Although some researchers have begun to explore the negotiation of footing in interpreted encounters, no one has examined footing in ASL-English interpreted interaction. Moreover, in signed-spoken language interpreting there is often an added complexity of potential bimodal utterances on the part of the interpreter.1 The purpose of this portion of the study is to analyze the two ASL-English interpreted encounters in order to identify and categorize evidence of footing. In order to examine ways in which interpreters influence an interaction, the primary focus of this analysis will be on interpreter-generated utterances, paying particular attention to the form and function of the interpreters’ utterances that are nonrenditions of spoken or signed language.

Footing in Interpreted Interaction: Mock Medical Encounter

Examination of interpreter-generated nonrenditions in the mock medical interview reveals that interpreters do contribute to interactional discourse. In these data, the interpreter frequently treats the Deaf interlocutor as a ratified addressee, excluding the hearing interlocutor. Conversely, there is only one utterance for which the interpreter treats the doctor as a ratified addressee while excluding the Deaf interlocutor. Most utterances directed to the hearing interlocutor are both spoken and signed, allowing the Deaf interlocutor to access the utterance as an unaddressed recipient. Thus, the interpreter establishes different footings between herself and each of the other participants.

The total number of utterances produced by the student interpreter is 117. Of these utterances, 102 are clearly motivated by the utterances of other participants and are classifiable as renditions. Thus, the total number of interpreter-generated nonrenditions is 15, representing 13 percent of her utterances. The analysis of this percentage of the interpreter’s discourse, and its influence within the interaction, is the focus here. Of the 15 nonrenditions, 11 are directed to the Deaf patient and are signed only, thus denying the hearing interlocutor access as a ratified unaddressed recipient (see table 4.1). One utterance is directed to the doctor only and is spoken but not signed. Interestingly, this utterance is actually a repetition of an utterance generated by the Deaf interlocutor. Because the Deaf patient signs the comment only once, but the interpreter decides to repeat it, the repetition is considered here to be an interpreter-generated utterance. For 3 of the 15 interpreter-generated utterances, the interpreter attempted to sign and speak simultaneously. For each of these utterances, the ratified addressed recipient was the hearing interlocutor, and the combination of speaking and signing created a footing in which the Deaf interlocutor was a ratified but unaddressed recipient.

These results indicate that the interpreter creates different footings with each interlocutor. She almost never allows the Deaf interlocutor to become an unratified addressee. However, the hearing interlocutor frequently receives unratified status on the basis of the interpreter’s footings. In order to get a better sense of the footings the interpreter creates, and how they are situated within the interaction, it is useful to categorize the interpreter’s footing types and their functions.

Table 4.1 Occurrences of Interpreter-Generated Nonrenditions in Mock Medical Interview

One of the ways in which Goffman distinguishes footing is on the basis of speaker roles. As described earlier, a speaker might employ any or all of the three roles of animator, author, and principal. The interpreter’s utterances, for which the interpreter is primary author, seem to vary in terms of the principal role. Certain interpretergenerated utterances seem to function as a part of the interpretation process. That is, some information is available within the interaction and originates among the interlocutors, but for some reason the interpreter must generate an utterance in order to fulfill the goal of relaying that information. For at least some relayings the interpreter functions as animator and author, but not as principal. For other interpreter-generated utterances, the interpreter appears to be managing some aspect of the interaction. For these utterances the interpreter appears to fulfill all three speaker roles.

Relayings

When an interpreter relays what other people say, generally the original speaker can be thought of as a primary author while the interpreter is a secondary author and animator. Thus, relayings for which the interpreter is primary author are somewhat unique. Examination of the mock medical interview reveals three types of relayings where the interpreter is clearly the primary author. These types include source attribution, explanations, and repetitions (see table 4.2).

Source Attribution

In interactional discourse, people are generally able to identify speakers on the basis of voice recognition and location. When discourse is funneled through a single individual, the interpreter, information regarding the location and identity of the source (the original animator) is not inherently discourse-bound. In other words, the interpreter can relay the content of the discourse without necessarily imparting the source of that content. Moreover, if the interpreter engages in self-generated utterances, there is potential confusion over whether a particular utterance has originated from the interpreter or another source. Thus, it is not surprising to find that some of the interpreter-generated utterances are devoted to source attribution. Of the fifteen utterances, four specifically identify the original animator of the upcoming utterance (see table 4.2). The interpreter appears to fulfill all three speaker roles for this utterance, since only the interpreter has contributed this information to the discourse. If the interpreter provides incorrect information about the source, it is the interpreter who is responsible for the incorrect content. Thus, the interpreter is not only animator and primary author, but also principal for such utterances.

Table 4.2 Occurrences of Relayings in Mock Medical Interview

The most frequent form of source attribution is a single indexical point in the direction of the speaker. In example 4.1 below, the interpreter has just finished introducing herself to the doctor and patient, and the doctor begins the medical interview:

Example 4.1

3

P:

D:    And how are you feeling this morning?

I:

I:       (point right) HOW FEEL ALL RIGHT MORNING?

She said, “How are you feeling? Are you all right this morning?”

The point to the right is directed toward the doctor. Just prior to this example the interpreter has been functioning as author and principal of her own introduction, and the pointing indicates a shift in footing such that the doctor is the primary author and principal of the upcoming utterance. Index pointing in ASL is comparable to the use of pronouns as markers of footing in English (cf. Hoyle 1993; Smith 1993).

It is interesting to note that all four occurrences of source attribution are directed to the Deaf interlocutor. The interpreter never authors any utterance designed to clarify for the hearing interlocutor whether an utterance originated from the patient or the interpreter. In addition, although the interpreter provides this type of information to the patient, she does not do so consistently. That is, for every utterance produced by the interpreter, an inherent question regarding the source of that utterance exists (is it motivated by the interpreter or by another participant?). Thus, it is noteworthy that of 117 utterances for which source could be attributed, the interpreter attributes the source only 4 times. Future research might seek to identify whether professionally certified interpreters provide source attribution, whether they do so consistently, and if not, what circumstances elicit such utterances.

Explanations

Explanations are a second type of relaying provided by the student interpreter. There are two types of explanations in the data: those in which the interpreter explains event-related information, and those in which the interpreter explains why the doctor has spoken to the interpreter as a ratified addressee. An example of the former can be seen in example 4.2, in which the interpreter informs the patient that a third person, a nurse, has just entered the room:

Example 4.2

25

P:

D:    Uh, I do have a list of uh, food that I’d like you to … try to stick to-

N: Excuse me, doctor, can I see =

I: HAVE LIST FOOD RIGHT - NURSE CL:1 (walk up)

I have a list of food right- a nurse just came in

26

P:

D:

N:    = you out in the hall for just one second? I need to ask you something about another patient?=

I:

I:         DOCTOR      CAN      SEE      PRO.1      H-A-L-L-W-A-Y      ONE-      MINUTE …

After the interpreter explains that the nurse has entered, she goes on to relay what the nurse has said to the doctor. This type of explanation occurs once in the data.

The second type of explanation occurs when the doctor directs an utterance to the interpreter as addressee. An example of this occurs when the doctor asks the interpreter how to sign a word:

Example 4.3

22

D:   What is- is there a sign for ulcer?

I:     PRO.3 PRO.1 SIGN FOR U-L-C-E-R PRO.2? ASK-PRO.1 QUESTION

        She asked me, “What’s your sign for ulcer?” She asked me a question.

The interpreter signs the question, including the footing shift on the part of the doctor in which she shifts from treating the interpreter as an unratified addressee or bystander to a ratified addressee. The interpreter explains to the patient that the doctor has asked her a question.

Another example of this type of explanation occurs in example 4.4. In this example, the doctor is responding to a question from the patient. When the interpreter accidentally touches the doctor’s arm while signing “WELL,” the doctor stops in the middle of her explanation, pauses, and asks the interpreter if there is a problem (line 57):

Example 4.4

54

P:     = AWKWARD ME

D:         Well- y’know it just depends on whether you really =

I:      = on coffee. If I don’t have it, I’m a- I’m a wreck.

I: UH =   

55

P:

D:    = wanna heal or not, uh, r- as I said before, right now this isn’t a big problem an ulcer is =

I:

I:      = TRUE DEPEND PRO.1 WANT HEALTH …                  TRUE U-L-C-E-R =

56

P:

D:    = just a little open sore. If you don’t wanna follow these lists, or take the medicines, then =

I:

I:             = NOTHING. SUPPOSE NOT FOLLOW LIST, =

57

P:

D:    = there-… is there a problem? Oh. -then it can become infected and then you do have a problem.

I: ‘scuse me.

I:      = WELL … EXCUSE. PRO.1 TOUCH (to doctor) Doctor-LOOK-AT-interpreter … THEN CAN

WORSE. =

Excuse me. I touched the doctor and she was, like, looking at me.     

The interpreter apologizes to the doctor and then explains to the patient what just happened. Although the apology is both spoken and signed, the explanation is rendered only in ASL. The doctor, unaware of this subordinated communication, has continued with the medical interview. Part of the doctor’s utterance is not rendered in the interpretation as a result of the overlap. The data contain two occurrences of this type of explanation.

The two types of explanations in these data include explanations about the event and explanations regarding why the hearing interlocutor has spoken to the interpreter. Given a larger body of data, it would be interesting to examine whether or not explanations are ever provided to hearing interlocutors, and if so, under what circumstances.

Repetition

The last type of relaying to be discussed consists of repetitions. Obviously, some repetitions in the data originate from participants other than the interpreter. However, there is one example in the data in which the interpreter, as secondary animator, decides to reanimate the utterance apparently due to an overlap in the talk:

Example 4.5

40

P:

D:     = have milk, but have it with a meal, and try to limit how much milk you have,
so that you’re =

I:

I:                     CAN HAVE …         WITH FOOD …         =

                You can have milk with food but try to limit   

41

P: LIMIT? MEAN LIMIT? NOT UNDERSTAND LIMIT.

Limit? What do you mean by that? I don’t understand what you mean.

D:     = not just … uh, y’know, maybe drinking a gallon of milk on an empty stomach.

I: What do you mean, limit? What do you mean limit?   

I:       = TRY LIMIT++

The repetition occurs in the interpreter’s English discourse in line 41. It is interesting to note that the Deaf interlocutor responds at an appropriate moment in the interpreter’s discourse, after she signs “LIMIT” (line 41). His response consists of a request for clarification of the doctor’s indication that he should limit his milk intake. His request is translated by the interpreter into a single English question, “What do you mean, limit?” However, the interpreter necessarily lags behind the doctor’s speech; in line 41 the interpreter is still animating what the doctor said in line 40. When the Deaf interlocutor, who, from his perspective, does not interrupt the conversation, is immediately reanimated by the interpreter, the rendition occurs during the doctor’s utterance, causing an overlap in the English dialogue. The doctor does not yield her turn. When the doctor does complete her turn, the interpreter then reanimates her interpretation of the patient’s question a second time (though he does not pause and then repeat his utterance) apparently so that it can be heard by the doctor. In this manner, the interpreter has taken responsibility for resolving the overlap, by removing it. If she did not, it is conceivable that another of the interlocutors would have taken responsibility for the repetition. For example, the doctor might have asked the patient what he just said, a question that the interpreter could have interpreted. Conversely, if the doctor did not respond to his question, the patient could have authored a repetition himself. Roy (1989a) discusses the role of the interpreter in turn-taking. An examination of the affects of footing shifts on turn-taking in interpreted interaction would be another area for future research.

The most common of the interpreter footings that occur as relaying tasks are source attribution and explanations. In considering the Interpreter’s Paradox, it is interesting to note that at least one of these categories, source attribution, appears to be a required component of interpreted interaction. That is, interlocutors cannot make sense of an interaction if they do not know who is responsible for the utterances. Clearly, some interpreter-generated nonrenditions are essential to the task of interpreting interactive discourse.

Interactional Management

In addition to relaying information about what is said or what is happening within an event, an interpreter’s nonrenditions can also be related to the structure of the interpreted encounter. The footing types that occur within this category can be classified as introductions, responses to questions, and interference (see table 4.3).

Table 4.3 Occurrences of Interactional Management in Mock Medical Interview

Introductions

A professional interpreter will become accustomed to interacting with professionals from various fields, including medical practitioners with a variety of specialties. It is less common, however, for these other professionals to have experience working with interpreters. As a result, participants in a given interaction might not understand who the interpreter is or what particular task he/she will fulfill without some sort of introduction.

In the mock interview, the interpreter introduces herself by signing and speaking simultaneously (example 3.5). This yields a footing in which the interpreter is addressing both interlocutors. The code choice is somewhat awkward, and evidence of this appears in the form of errors or self-repairs in both languages, as can be seen in example 4.6 below:

Example 4.6

1

P:

D: Oh, you’re the interpreter for today.

I:    … and I’m gonna be the sign lang- language interpreter for today.

I:    POSS. I SIGN LANGUAGE INTERPRETER LANGUAGE NOW. PRO.1, PRO.3, PRO.1 OH INTERPRETER =

I’ll be the sign language interpreter- language for today. I- She said, “Oh, I’ll be the =

2

P: (nods)

D: It’s nice to meet you.

I: Thank you, it’s nice to meet you too.   

I: NOW PRO.1, #OH #OK. PRO.1 - NICE MEET PRO.1. THANKS NICE MEET -TO (doctor) SAME.   

interpreter for today, oh, okay.” It’s nice to meet me. Thanks, it’s nice to meet you, too.

In line 1 the interpreter treats both interlocutors as ratified addressees. In line 2, the interpreter responds to the doctor as they exchange greetings, resulting in an unaddressed recipient status for the patient. It is interesting to note that this footing is apparently noticed by the Deaf interlocutor, and in a later reenactment of the interview,2 the introductions are initiated by the patient, not by the interpreter (see example 4.7):

Example 4.7

D:

P:   HELLO PRO.1 INTRODUCE PRO.3 POSS.1 INTERPRETER (name) PRO.3 (nods)

Hello. I’d like to introduce you to the interpreter. Her name is -.

I:   Hi. I’d like to introduce you to the interpreter. Her name is -.

I:

In example 4.7 the Deaf patient handles the introduction of the interpreter. Thus, the interpreter does not generate any nonrenditions, and does not have to make code choices that start the interview on any particular footing. Introductions can be handled in other ways as well. For instance, the interpreter could choose to introduce herself in one language at a time. Each of these options seems to result in a different footing, whether interpreter-initiated or not. Future research regarding the impact of the various footings on different genres of interaction could clarify the impact various types of footing have with regard to the Interpreter’s Paradox.

Responses to Questions

Another type of structural footing shift occurs when questions are directed to the interpreter. These questions can come from either the hearing or the Deaf interlocutor, and examples of both occur in the data, though they are somewhat different in character.

In the mock medical interview, the doctor directs questions to the interpreter during the course of the interview. This was seen in example 3.7 and is repeated here as example 4.8 for convenience:

Example 4.8

22

P

D: What is- is there a sign for ulcer?   

I:

I: PRO.3   PRO.1   SIGN =   

She asked me, “Wbat’s your sign

23

P: (hand flip)

(Whatever.)

D: No, huh?  …  Oh, okay =  

I: Umm …

I:

I: FOR U-L-C-E-R PRO.2? … ASK-TO (interpreter) QUESTION. NO SIGN? ASK TO (interpreter)

for ulcer.” She asked me a question. There isn’t one? She asked me.

In this example, the interpreter does not respond to the doctor’s question. She interprets what the doctor said, and adds her explanation, but all this is done in ASL and is not accessible to the doctor. Thus, the interpreter shifts footing, but not in harmony with the doctor’s shift to interpreter as an addressed recipient. This example differs somewhat from those occurrences in which the patient asks a question of the interpreter.

There are two differences between the previous example and example 4.9 below. First, the patient never asks a question of the interpreter during the course of the interview. The only time the patient treats the interpreter as an addressed recipient is when the doctor is temporarily called out of the room. The second difference between these two examples is that the interpreter responds to the patient with a much different footing. While the interpreter does not comply with the patient’s request, she does provide an answer to his question, filling the empty slot in the patient-initiated adjacency pair (Schegloff and Sacks 1973):

Example 4.9

31

P: UH U-L-C-E-R TRUE MEANING RIGHT PRO.3?

hmm, is an ulcer really what she said it is?

I:

32

P: ASK-ASK? =      

Ask the doctor? =

I:    PRO.1 TRUE EXACT KNOW ULCER? NO. PRO.1 ASK-ASK POSS.3.

I just don’t know much about ulcers, it’s better to ask the doctor.

In example 4.9, the interpreter accepts the footing established by the patient and responds to his question. In these data, there are three occurrences in which the patient asks the interpreter a question and she generates a response, albeit without complying with the request for information or assistance. In addition, there is one example in which the doctor asks the interpreter a question. In this example, the interpreter never responds to the doctor, but tries to elicit a response from the patient by explaining the situation. This footing type is included in the taxonomy on the basis of function (explaining) as are the other types of footing that occur in the data (see table 4.3).

Interference

The last type of footing shift identifiable in the mock medical interview occurs as a result of the physical environment. During the course of the interview, the interpreter accidentally touches the doctor’s arm while signing. This segment of the transcript is repeated as example 4.10 below:

Example 4.10

57

P:

D:    = there-… is there a problem? oh. -then it can become infected and then you do have a problem.

I: ‘Scuse me.

I:    = WELL … EXCUSE. PRO.1 TOUCH (to doctor) doctor-LOOK-AT-interpreter … THEN CAN
WORSE. =

Excuse me. I touched the doctor and she was, like, looking at me.   

This segment of the data, seen in example 4.4, includes an explanation within the utterance. However, prior to the explanation the interpreter shifts footing by code switching and simultaneously signing and speaking a comment directed to the doctor as an addressed recipient, with the patient as unaddressed recipient. The interpreter shifts footing by excusing herself for accidentally touching the doctor. It is interesting to note that after excusing herself, the interpreter does not provide an explanation to the doctor for why the doctor was seemingly tapped on the arm.

The types of interference currently under discussion reveal footing shifts resulting from changes in the physical environment (i.e., being seated next to someone while signing). These types of interference are similar to Hoyle’s discussion of footing shifts resulting from interference (1993, 128), as when a dog gets in the way of a Ping-Pong game and interrupts children engaged in sportscaster talk.

Relayings and Interactional Management in the Mock Medical Interview

The two different categories of footing shifts, relayings and interactional management, are determined on the basis of fifteen interpreter-generated nonrenditions within the data. Of the fifteen utterances, six functions have been identified, three categorized as relayings, and three as related to management of the interaction. The existence of these utterances and the variety of footings that they represent suggest that the student interpreter is one of the interactional participants. However, the function of many of the footings appears to be related to the interpreter’s goal of providing access to the interaction while minimizing participation in it. Nevertheless, it is interesting to note that almost three-quarters of the interpreter’s nonrenditions are directed to the Deaf participant only, and are not accessible to the hearing participant. Having examined the student interpreter’s nonrenditions, the next question is whether or not the professional interpreter exhibits nonrenditions, and if she does, what types of footings they represent.

Footing in Interpreted Interaction: Actual Medical Encounter

Like the student interpreter in the mock medical encounter, the professional interpreter in the actual medical encounter contributes nonrenditions to the interactional discourse. As in the mock interview, the interpreter frequently treats the Deaf interlocutor as a ratified addressee while excluding the hearing interlocutors as a result of code choices. However, unlike the student interpreter, the professional interpreter never attempted to both speak and sign simultaneously while in the presence of either the doctor or the nurse. Thus, the professional interpreter’s nonrenditions, on the basis of language code, treat either the Deaf mother or the hearing medical practitioner (but never both at the same time) as ratified addressees. Although the nature of the footings differ, the professional interpreter, like the student interpreter, establishes different footings between herself and each of the other participants.

The professional interpreter produces a total of 387 utterances. Of these utterances, 358 are renditions of discourse produced by other participants. Thus, there are a total of 29 interpreter-generated nonrenditions in the actual medical interview. The interpreter’s nonrenditions represent 80 percent of her utterances. The focus here will be on the analysis of this percentage of the interpreter’s discourse and its influence on participant alignments within the interaction. As in the case of the mock medical interview, a majority of these utterances are directed to the Deaf interlocutor and are signed only (see table 4.4). Four of the utterances are spoken only. Of the 4, 1 is directed to the doctor, one to the nurse, and 2 are directed to the researcher during a period when neither the doctor or nurse is present. These last 2 utterances occur during a frame break, since the linguistic evidence suggests that the participants view this period as an out-of-frame (that is, the interpreted encounter frame) activity. Finally, 3 of the 29 utterances are both spoken and signed. These occur during the interpreted encounter frame-break, while the interpreter is communicating with the Deaf mother and the hearing researcher. Unlike the “simultaneously” produced utterances generated by the student interpreter, in which the hearing and Deaf participants were all ratified but only the hearing participant was an addressed recipient, in the three occurrences generated by the professional interpreter both participants appear to be both ratified and addressed. Since some of the interpreter’s utterances are both spoken and signed, while others are one or the other, this represents interesting code switches on the part of a native bilingual. Although it is beyond the scope of this study, it would be interesting to determine what code choices native bilinguals who are hearing make when communicating with Deaf and hearing bilinguals, and what the function of such footing shifts might be in noninterpreted discourse.

Table 4.4 Occurrences of Interpreter-Generated Nonrenditions in Actual Medical Interview

These findings are similar to the results of the mock medical interview analysis. For example, the interpreter-generated nonrenditions consist primarily of utterances directed to the Deaf participant and are inaccessible to the doctor or nurse. However, since a footing is a highly situated part of an interaction, it is critical to examine the form and function of the nonrenditions generated by the professional interpreter. As in the case of the student interpreter, the professional interpreter’s utterances under examination fulfill two goals: relaying and interactional management. However, the specific functions within each of these two categories are not identical to the first case.

Relayings

Most of the relayings produced by an interpreter consist of renditions. However, an examination of the interpreters’ nonrenditions reveals that some clearly serve a relaying function. In the actual medical interview, the utterances categorized as relayings consist of two of the functions found in the mock medical encounter: source attribution and repetitions. In addition, the professional interpreter’s nonrenditions include requests for clarification (see table 4.5).

Table 4.5 Occurrences of Relayings in Actual Medical Interview

Source Attribution

Monolingual native English interlocutors are able to identify speakers on the basis of voice recognition and location; therefore, discourse funneled through a single source (the interpreter) does not inherently include such information. However, since that single source also generates nonrenditions, there is the potential for confusion over who is the original author of each utterance produced by the interpreter. Therefore, it is not surprising to find that source attribution is one type of utterance found in both the interpreted mock and actual medical interviews. What is surprising, perhaps, is the relatively limited number of occurrences of source attribution in the actual medical interview. Only three of the interpreter’s nonrenditions attribute source. The attribution of source takes two forms: pointing and body shifting.

In the following example, the interpreter incorporates a body shift as well as a point in the direction of the doctor, who is the original author of the utterance. At this point in the interaction, the doctor and nurse are examining the child, who is crying:

Example 4.11

94

M:

D: Well, let’s have-

N: Oh, darn it.

C:             (crying)(coughs)

I:

I: #WELL

95

M:

D: He’s cutting four teeth and that’s contributing to the mucus, =

N:

C: (cries)

I:

I:     #OH D-A-R-N #IT (shifts to side) #HE HAVE FOUR EYETEETH C-U-T =

96

M:

D:     = and it’s easy for mucus to get in, behind the eardrums and and, so teething infants frequently =

N:

C:

I:

I:      = INDEX (doctor) THAT PROVIDE SOME M-U-C-O-U-S THAT EASY M-U-C-O-U-S BEHIND =

97

M:

D:     = have that.

N:

C:

I:

I:      = E-A-R-D-R-U-M-S #SO TEETH BABY THAT TEETH OFTEN HAVE THAT

In the first line, the interpreter begins by fingerspelling an utterance originally authored by the nurse, “Oh, darn it,” and then goes on to provide a rendition of the doctor’s utterance regarding the child’s eyeteeth. Just prior to beginning the rendition of the doctor’s utterance, the interpreter shifts her body slightly to the right. In this manner, she is able to differentiate between the two English speakers. However, the body shift alone does not distinguish which of the two is speaking, merely that the original author has changed, thus, the interpreter includes a point in the direction of the doctor in the second line. The interpreter is able to make a distinction between the two speakers, including two discourse-relevant pieces of information not otherwise accessible to the Deaf interlocutor: that the primary author has shifted, and identification of the primary author.

The use of the body shift is comparable to what Wadensjö refers to as reported speech in interpreter’s renditions. When the words or actions of more than one individual are being reported or constructed, body shifting is a strategy used for distinguishing multiple characters in ASL narratives.3 In addition, both Winston (1991) and Locker McKee (1992) discuss the use of body shifts in ASL lectures as markers of changes in footing. It is interesting to note the use of a narrative discourse strategy on the part of the interpreter. Since the interpreter is a single individual relaying into one language all the utterances originating in another, she is functioning somewhat like a narrator who is constructing the dialogue of a specific event. The difference between the interpreter’s narrative and “conventional” narrative discourse is that the interpreter is narrating an event while it occurs in the present time and space (cf. Wolfson 1979).

In the previous example, the interpreter used a combination of linguistic devices to attribute the source of the utterance being rendered. However, some examples include no attribution of source. In the following example, the doctor, nurse, and mother all are involved in the examination of the child. The child is crying loudly:

Example 4.12

86

M:

D:    Ask her to just hold his knees.

N: Aw, sweetheart.

C:             (crying) (wails)

I:

I:                     JUST HOLD K-N-E-E-S SWEETHEART

The interpreter provides a rendition of the doctor’s utterance, followed by a rendition of the nurse’s utterance. No marker is used to indicate to the Deaf participant that the same person did not originally author both utterances. Although the doctor and nurse are not in the room at the same time throughout the duration of the interview, there is always potential confusion as to who is the originator of a given utterance. Not only is the hearing researcher present as a potential English speaker, but also the interpreter mixes renditions with nonrenditions during the interview. Thus, every utterance produced by the interpreter can be questioned regarding source. For this reason, the omission of source indication is an interesting finding in the actual interview. Is source information not pertinent to Deaf consumers? Is the determination of source available through other means, such as looking to see whose mouth is moving? The presence or absence of source attribution, and its potential impact on interactive discourse, is an area for further investigation with a larger body of data.

In keeping with the identification of a limited amount of source attribution in ASL for the Deaf interlocutor, a limited amount of source attribution is provided in English for the hearing interlocutors. Just as the potential exists for the Deaf participant to be confused between nonrenditions and renditions, or to whom original authorship of a given rendition is attributable, hearing participants can also become confused about who is the author and/or principal of an utterance. In example 3.15 (repeated here as example 4.13), the doctor becomes confused about who the principal was for a given utterance:

Example 4.13

55

D:    You (points to interpreter) tried or she (points to mother) tried?

In this example, the doctor specifically asks for clarification regarding the source: “You tried or she tried?” The doctor appears to be uncertain as to whether the interpreter is providing a rendition of the mother’s utterance or a nonrendition for which the interpreter herself is the sole author. Since there is evidence that the potential for confusion exists for hearing participants, it is worth pursuing an examination of source attribution in both ASL and English utterances in future research.

Source attribution is a category of interpreter-generated nonrendition that is essential to the relaying of interactive discourse. Interpreters can use a variety of linguistic strategies to attribute source. Moreover, interpreters sometimes do not provide attribution of source at all. The inclusion or exclusion of such information has the potential to influence the comprehensibility of the interactive discourse. Thus, further investigation of various strategies for providing source information, and when such information is and is not provided, could assist in elucidating the kinds of influences interpreters have on interactive discourse.

Repetition

Repetition serves as a second function of interpreter nonrenditions. In the mock medical interview, there is one example of a repetition resulting from an overlap in the dialogue that can be found. The interpreter generates a repetition of her English rendition, although the original author does not. In the actual medical interview, four occurrences of repetitions can be found. Of these, one is the result of an overlap in the discourse.

Just prior to the following example, the mother has asked the doctor if her son’s lungs are clear. The doctor provides an answer, during which the mother attempts to initiate a turn:

Example 4.14

177

M:

D: Yeah he may cough, but his lungs are clear. I mean =

N:

C:

I:

I: NO PRO.3 MAYBE COUGH BUT LUNGS TRUE =

178

M:

D:    = He’ll cough and sound like there’s something there an- it has nothing more to do than there’s stuff =

N:

C:

I:

I:      = CLEAR MAYBE COUGH SOUND SAME BUT =

179

M: BEC- BECAUSE PRO.1 HUSBAND #WAS SAY THAT PRO.3 YEL-

D:     = draining down from his nose and his ears if we clear up the ears we’ll clear up the cough.

N:

C:

I:

I:      = JUST DRIP-FROM-EAR FROM EARS NOSE #IF #IF CLEAR EARS BUT- COUGH =

180

M: OH-I-SEE BECAUSE PRO.1 HUSBAND #WAS SAY =

D:

N:

C:

I: ’Cause my husband was saying that =

I:      = FUTURE CLEAR SAME

The doctor answers the question in the first line, “Yeah, he may cough, but his lungs are clear.” The answer is followed by a more detailed explanation. During this explanation, the mother begins to sign BECAUSE while gazing at the interpreter. The mother then shifts her gaze to the doctor and begins her utterance. Whether or not the doctor is aware of the attempted interruption, he pauses slightly, but does not yield his turn. He adds the conditional “if we clear up the ears we’ll clear up the cough.” The interpreter pauses and appears ready to begin her English interpretation of the mother’s comment when the doctor utters this conditional statement. The interpreter then signs #IF, and as the mother shifts her gaze back to the interpreter, the interpreter repeats the #IF and continues on with her rendition. This brief repetition is a result of the overlapping dialogue. The overlap that occurs in this example is an overlap between the interpreter and the mother. It is conceivable that neither the doctor nor the nurse are aware of the overlap in the ASL discourse. Because interpreters must follow somewhat behind the original utterances in order to understand and then render them, the mother might begin her turn as an overlap with the interpreter’s utterance in hope of making up for the lag in time and effectively gaining the floor. The interpreter’s repetition effectively regains the Deaf participant’s eye gaze in order to allow for rendering of the doctor’s utterance. Although this example of repetition is an interpretergenerated nonrendition (the doctor does not repeat this part of his original) related to the relaying of an utterance, it clearly bears a relationship to the management of the interaction and the talking and yielding of turns. For an in-depth discussion of an interpreter’s strategies for managing turns in an interpreted encounter see Roy (1989a).

The second type of repetition is related to the redoing of a rendition. This can be seen in the following example, in which the doctor is eliciting detailed information about the child’s symptoms:

Example 4.15

52

M:

D:    Okay, how many times did he vomit from the time it started?

N:

C:

I:

I:     #OK HOW- MANY- HOW OFTEN VOMIT. START SICK, VOMIT, HOW OFTEN?

In her original rendition of the doctor’s utterance, the interpreter produces a transliteration or contact variety in a manner consistent with the code choices made by the Deaf participant. However, the code choice causes a problem in the production of a rendition. The interpreter begins her first rendition by signing HOW MANY, which is a direct lexical transliteration of the doctor’s original phrase in English, “How many …” However, as the doctor continues his utterance, adding the word “times” to the phrase “How many times …,” the interpreter initiates a repair. She changes her utterance to HOW OFTEN VOMIT, which is a direct translation of the doctor’s utterance up to that point. Finally, as the doctor completes his utterance with a reference to time, “from the time it started,” the interpreter repeats her rendition.

In ASL, time-related events are generally uttered in the sequence in which they occur. In addition, ASL makes use of a spatial time line in which a signer can move forward to show the passage of time (cf. Friedman 1975; Baker and Cokely 1980; Winston 1991). The interpreter has the option to continue with her contact variety/transliteration, or to redo the entire rendition in order to incorporate the ASL-appropriate spatial and sequential markers of time. She opts for the latter, signing START SICK, VOMIT, HOW OFTEN? while shifting her body from a backward lean to a forward lean. Thus, this repetition is the result of the redoing of a rendition. It has been categorized as a nonrendition because the primary author’s utterance did not include such a repetition.

Although there are a limited number of repetitions in the data, it is interesting to note the two types of repetitions that do occur. The first example of repetition is due to overlap. Roy has described the interpreter’s role with regard to overlaps and other aspects of turn-taking as one in which the interpreter is not a neutral conduit providing access between two members of a dyad, but rather that “the interpreter is an active member of interpreted conversations” (1989a, 263). Similarly, when the interpreter redoes a rendition in order to codeswitch, this repetition demonstrates that the interpreter, and not the original author of the utterance, is making this conversational choice.

Requests for Clarification

The third type of relaying identified in the professional interpreter’s nonrenditions involves requests for clarification. The interpreter makes three such requests: once to the nurse, once to the doctor, and once to the mother. There are two reasons why the interpreter makes such requests—distractions and lack of relevant background knowledge.

There is evidence that the interpreter frames the interview, in part, as a research study. Her schema for this frame causes her to respond differently to the researcher’s utterances than to the utterances of the other participants. That is, when the researcher addresses the interpreter during the medical interview, the interpreter does not respond until the nurse leaves the room. Nevertheless, when the researcher distracts the interpreter from the task of interpreting, she misses something that the nurse is saying and is unable to render that:

Example 4.16

21

M: (points left to interpreter)

N:    = boys- a hundred boys his age. As far as weight and height are concerned.

C:

I:

I:    = COMPARE (..?..) LIST: (?) MEASURE (?) (looks left)

22

M:

D:

N:    He’s off the chart for his weight (laughs” He’s off the chart, he’s - he’s way at- at uh, =

C:

I: (to nurse) Pardon?

I: PRO.3 FINISH (?) TOOK-OFF

In this example, the interpreter gazes to the left, toward the researcher, at the point when the researcher (off-camera) produces an ASL utterance. Rather than interpreting the utterance, or regulating the overlap, the interpreter pauses momentarily and then asks the nurse to repeat, when she says, “Pardon?” This is the only point at which the interpreter appears to be distracted in this manner.

The second request for clarification arises from the interpreter’s need to either obtain additional background information or to compensate for gaps in her knowledge. In the following example, the mother is explaining to the doctor that she will get her son’s records from another local pediatric center where she has taken her son in the past:

Example 4.17

138

M: YES (+ voice) PRO.1 FUTURE GET #IT =

D: It would be nice to have, uh, some of that history because I mean I can’t -

N:

C:

I:     = infection, that he had. Yeah, =

I: FUTURE NICE ABLE HAVE SOME LIST HISTORY

139

M:    = FROM INDEX (rt) + POSS.1 DOCTOR PRO.1 GET CL:G FROM #GT UNIVERSITY =

D:

N:

C:

I:      = I will get it- I’ll get from my other doctor. I have an appointment - =

I:

140

M:     = P-E-D- (to interpreter) #GT UNIVERSITY P-E-D-I-A-T-R-I-C CENTER INDEX (rt)=

D:     All right =

N:

C:

I:     = (head tilt left) (head nodsj from Georgetown uh, the pediatric center there?

I:

In this example, the mother, who grew up locally, uses a local sign for the name of the organization, #GT. However, the interpreter is not from the local area, and so is not familiar with the local sign and its referent. She requests clarification by tilting her head slightly to the left, at which time the mother shifts her gaze to the interpreter and responds to the request. The interpreter is able to complete the sentence after the repetition, although her English rendition of the name of the organization is not quite the same as the English one would hear from a local resident, due to her omission of the word “University.” Because the background knowledge of the signer and the interpreter are not identical, the interpreter requested clarification. This negotiation led to a successful rendition of the mother’s original utterance.

A similar circumstance occurs in the following example in which the doctor is giving the mother a diagnosis of the child’s problem:

Example 4.18

161

M:

D: So … Um …         P.E. stands for physal- physical exam

N:

C:

I:

I:     = TO BREATHE THROUGH NOSE P-E MEAN PHYSICAL SEARCH (body) =

162

M:

D:     = he has bilateral otitis, bilateral, both sides- both ears, otitis immediate- the right =   

N:

C:

I: (to doctor) he has what?

I:     PRO.3 HAVE B-I-L-A-T-E-R-A-L BOTH EARS (two hands) O-T-I-T-I-S =   

163

M:

D:     = is a little worse than the left, and I’m putting him on (?)… that’s a penicillin =   

N:

C:

I:

I:     = RIGHT LITTLE WORSE THAN LEFT     PRO. 1 GIVE-TO (baby) A-U-G-M-E-N-T-A-I-N =

The doctor indicates that the child has “bilateral otitis.” This condition appears to be unfamiliar to the interpreter, who requests clarification from the doctor: “He has what?” It is possible that the interpreter is familiar with the condition, but has simply not heard what the doctor said. However, the doctor appears to treat the request as if it is related to background knowledge, and in his response he defines the terms in an attempt to clarify them, as in “bilateral, both sides, both ears.”

Whether the mother is familiar with this condition does not become clear. If the interpreter had provided an accurate rendition to the mother, three possible outcomes might have resulted. First, the mother might have understood and allowed the interaction to continue. Second, the mother might not have understood, yet might have, for some reason, decided not to comment. Third, the mother might not have understood and requested clarification herself. Although it is impossible to know what might have happened, it seems clear that the interpreter, who makes a legitimate request for clarification, has had an impact on the outcome of this portion of the interaction.

The interpreter requests clarification from each of the participants in the interview. One reason for requesting clarification is due to distractions from what the interpreter is focused on as the task at hand. A second reason for requesting clarification is to assist in understanding an utterance that contains information that might be familiar to the other participant(s), but that is new to the interpreter. Both of these causes for clarification requests are inherent to the task of interpreting. That is, interpreters can never have background knowledge that is identical to that of any one or all of the participants. In addition, while one can attempt to limit distractions, it would be impossible to eliminate the potential for distractions (e.g., phones ringing, rainstorms, failing books or chairs, and so forth). Thus, the particular context of a given interaction will have an impact on all the participants, including the interpreter. Likewise, all the participants, including the interpreter, will have an impact on the situated interaction.

Summary

Both source attribution and repetition occur in the actual medical interview, just as they are identifiable functions of the interpreter’s footings in the mock medical interview. In addition, the professional interpreter’s footings include requests for clarification resulting from distractions of the interpreter or distinctions between the interpreter’s and the other participants’ background knowledge. The latter category, like source attribution, indicates that interpreters cannot be neutral in the sense of simply rendering decontextualized utterances. As Seleskovitch (1978) has said of spoken language conference interpreters, an interpreter must be able to understand in order to interpret.

Interactional Management

In the actual medical interview, there are nineteen interpretergenerated nonrenditions that pertain to interactional management. Although the actual medical encounter does not contain any introductions, as did the mock interview, both responses to questions and interference occur in the data. Summonses represent an additional function that did not occur in the mock case but does occur in the actual medical interview (see table 4.6).

Table 4.6 Occurrences of Interactional Management in Actual Medical Interview

Responses to Questions

Like the student interpreter, the professional interpreter generally does not respond to questions directed at her during the course of the medical interview. In the actual medical encounter, the doctor and nurse address questions to the interpreter, but the Deaf participant does not. In addition, the hearing researcher addresses a question to the interpreter. The latter of these is the only question that the interpreter responds to in the entire interview.

Throughout the encounter, the doctor and nurse direct numerous questions to the interpreter. Many of these are questions for which the mother has the answers. These types of questions take a form in which the mother is talked about rather than talked to. An example of this can be seen in line 10 below:

Example 4.19

10

Nurse: Does she have his shot records?

The interpreter merely relays, rather than redirects, these types of questions. They are then answered by the mother and rendered by the interpreter as a direct response. This is noteworthy because this is what the student interpreter appeared to be striving toward in the mock medical interview. In that case, however, the Deaf person did not always respond. In at least one example in the mock medical interview, this left an unfilled slot in a question-answer adjacency pair.

Because the questions from the doctor and nurse in the actual medical interview are generally answered by the mother, there appear to be no major problems in the flow of the interaction. Nevertheless, it is not clear whether the doctor and nurse understand who is the principal for the interpreter’s utterances. In the earlier example of the doctor requesting clarification regarding the referent of a first-person pronoun, the interpreter does not generate a response for which she is the principal. However, since she is authoring the mother’s discourse as constructed dialogue, using first person to refer to the mother, the doctor obtains the information he needs without further confusion. The example below begins with the mother’s response to the doctor’s question: “You tried or she tried?” The mother is explaining the fact that she herself is the principal of the interpreter’s utterances:

Example 4.20

57

M:     = LET PRO.1 EXPLAIN. WHEN INTERPRETER TRUE INTERPRET, PRO.3 WILL =

D:

N:

C:

I:       = Now let me explain. When the interpreter is interpreting, she will be speaking =

I:

58

M:     = BE TALK A-S I-F PRO.3 NOT I-N ROOM PRO.3. TALK FOR PRO.I, #SO (?off screen) =

D: (chuckles)                   Okay, gotcha.

N:

C:

I:       = as if, she’s not in the room.                   She’s speaking for me. So, if it =

I:

59

M:     = CONFUSE TRUE TALK PRO.1, A-S I-F PRO.3 PRO.1

D: I’m just interested in the pronoun, that’s all.

N:

C:

I: = confuses you, the interpreter’s really speaking as if

I: PRO.1 (REALLY) INTEREST =

60

M: #OK PRO.1

D: You’re not taking care =

N:

C: (cries)

I:

I:       = WORD ”I” (i on chest) PRO.3 (left hand) I (i on chest)

61

M: NO PRO.1 RECENT MEET- (interpreter) #HER FIRST TIME =

D: = of him, you’re just interpreting.

N:

C: (crying continues)

I: I just met =

I:       NOT PRO. 1, PRO. 1 NOT TAKE-CARE -

62

M:     = NOW.

D:                 okay I’ve been in practice thirty-five years, I’ve =

N:

C: (screams then crying stops)

I:       = her for the first time today, so

I: #OK                   PRO. 1 …                   PRO.1 WORK++ =

In line 60, the doctor directs another question to the interpreter: “You’re not taking care of him, you’re just interpreting.” Once again, the interpreter does not respond to the question. As the mother responds that she has only just met the interpreter for the first time, the interpreter’s rendering “I just met her for the first time today, so,” likely answers the doctor’s question for the wrong reason. That is, the referent for the interpreter’s first-person pronoun is the mother, not the interpreter. Moreover, the interpreter is referring to herself in the third person, with “her.” It is likely that the doctor mistakenly assumes that the interpreter is referring to herself in the first person, and that the third-person pronoun refers to the mother. Because the interpreter and the Deaf participant are the same gender, there is no way to know for certain who the intended referents for these pronouns are. Nevertheless, on the basis of the interpreter’s eye gaze and the content of the mother’s recent utterance, it seems likely that the interpreter did not generate the response. Fortunately, the response provided the doctor with accurate information, regardless of whether or not he was entirely clear about who originated it.

The only time that the interpreter actually answers a question is when the researcher asks her if she can move into view of the video camera. The interpreter waits to respond until a break in the interpreted encounter frame, when the nurse leaves the room:

Example 4.21

32

M:

D:

N:     Doctor should be in in just a few minutes (exits)

C:

I: (to researcher behind camera) Wha’d you say?

I:     DOCTOR COME FEW MINUTE

33

M: FINE WHERE?

D:

N:

C:

I: (moves so visible) How’s this? (nods) If we can rid of this chair, ’cause the doctor’s not =

I: (arms moving but signs not visible)

34

M:             TEND SIT INDEX (to another seat) (shakes head)

D:

N:

C:

I:     = gonna sit in it.

I: (to mother) PRO.2 SIT (IN CHAIR)?

35

M:     U-N-L-E-S-S #DR WANT PRO.1 HOLD PRO.3 (to baby) (shrugs)

D:

N:

C:

I: If we can- I’ll just (?)

I: MOVE                          

36

M: PRO.2 STAR! SAME FATHER =

D:

N:

C: (sees video camera)

I:      (moves chair) Better (laughs)

I: BETTER

The interpreter responds by complying with the researcher’s request. The interpreter moves and asks, “How’s this?” When the move is not sufficient to resolve the logistical problem, the interpreter asks the mother if she will use one of the two chairs in the room, and then moves it out of the way in order to situate herself such that she can be seen by both the mother and the camera.

Although numerous questions are directed to the interpreter in the actual medical interview, her only response is directed to the researcher and occurs during an out-of-frame period of the interview. Her lack of responses to the doctor and nurse, unlike the student interpreter’s nonresponses, do not include explanations about the current circumstance. However, whether or not the professional interpreter would include such explanations if the Deaf participant did not intervene with responses and explanations is not clear from this data. This would be an interesting area for future research. Since the Deaf participant does not direct any questions to the professional interpreter, it is also not clear how the interpreter would respond in such a situation. This also would be another area for additional research. It is worth noting, however, that the Deaf participant in the mock interview directed utterances to the interpreter only during frame breaks (when the doctor left the room). The Deaf participant in the actual interview used this time to communicate with the researcher, who is of relatively long acquaintance, rather than to the newly met interpreter. If the researcher had not been present, it is conceivable that the Deaf participant might have interacted with the interpreter on a different footing during these out-of-frame periods. Perhaps a larger corpus of data and different videotaping circumstances would allow for such interaction to be caught on videotape.

Interference

Interference, as described with regard to the mock interview, refers to footing shifts that result from the physical environment. Although no blatant environmental interferences occur, two examples of interference are responsible for the professional interpreter’s nonrenditions. One of these examples is the result of the physical presence of the camera and researcher, while the other results from the use of an underarm thermometer in an encounter that involves ASL-English interpretation.

The presence of the researcher and camera add an additional logistical burden to the interpreter. While ASL-English interpreters must consider numerous logistical issues in terms of both visual and auditory accessibility (cf. Frishberg 1990), the researcher puts an additional demand on these logistical considerations by requesting that the interpreter remain in view of the video camera. The small size of the examination room and the need to see participants at a clear angle (allowing for readability of manual and nonmanual linguistic information) necessitated such a request. All the interpreter’s nonrenditions that address the logistical issues related to the presence of the camera are the result of this interference.

The second example of interference occurs when the nurse attempts to take the child’s temperature using an underarm thermometer. In this example, the mother is watching the interpreter’s renditions, holding her son, attempting to hold his arm down so that the thermometer remains in place, and also conversing with the nurse. Because the mother uses her hands to communicate, the attempt to hold her son and his thermometer at the same time seems analogous to a hearing English speaker attempting to converse with a dentist with the dental instruments in his or her mouth:

Example 4.22

25

M:     ALWAYS B-E-E-N THAT W-A-Y

D:

N:     And he’s just below the ninetieth for his length but he’s way off the chart for his weight.

C:

I:

I:             (?) UNDER             (?) #TH MEASURE

26

M:

D:

N: Okay, did that stop =

C:

I:

I:     WEIGHT TOOK-OFF LOOK-AT (baby’s underarm thermometer)

27

M: I D-O I-N R-E-A-R INDEX

D:

N:     = beeping? No … this is gonna be impossible. (?)

C:

I:

I: INDEX IMPOSSIBLE         WELL #OK

28

M:             EASIER FINISH EASIER #IF PRO.1 =

D:

N: (laughs) We try to get ’em not too excited before … they see the doctor.

C:

I:       I do it in his rear.

I: WE TRY- DON’T-WANT OR EXCITE BEFORE-

The interpreter is rendering the nurse’s report regarding the child’s weight. At the end of this rendition, the interpreter generates a nonrendition, "LOOK-AT (baby’s underarm thermometer),” directing the mother to look at the thermometer, which is slipping out from under the child’s arm. Information about the physical environment can be considered an interference to the existing footing. Nevertheless, in this case the interpreter is providing information about the physical environment that might not otherwise be noticed by the mother, who cannot look in two places at the same time.

Two examples of interference occur in the actual medical interview. In the first example, the presence of the researcher and video camera are responsible for the shift in the interpreter’s footing to a self-generated nonrendition. In the second example, it is the fact that the Deaf participant might not notice relevant aspects of the physical environment while maintaining eye contact with the interpreter. In future research it would be interesting to determine the extent to which interpreters provide certain contextual information to Deaf participants.

Summonses

The third type of interactional management identified in the actual interview is a type of nonrendition that did not occur in the mock medical interview. Summonses, or attention-getting strategies, comprise twelve of the interpreter’s twenty-nine nonrenditions. In the discussion of source attribution it became clear that information regarding the identity and location of the source (or original author) of an utterance is not necessarily accessible to the other participants without additional information being provided by the interpreter. An adjunct to this is whether or not the participants know when an interlocutor begins an utterance.

Hearing English speakers may or may not be aware that signing has originated from the interpreter or the Deaf participant, but once the interpreter begins to speak, they know that an utterance is beginning. Conversely, for a Deaf participant, presence in the room is not enough to provide information that someone is signing an utterance. The issue of utterance accessibility becomes relevant with regard to the channel in which the message is being sent. If it is sent through the visual channel, the receiver’s eyes must be directed at, and within the discriminatory range of, the source of the visible message. If it is sent through the acoustic channel, the ears of the receiver must simply be within the discriminatory range of the acoustic message source. Thus, it is essential that the Deaf interlocutor receive a visual cue to attend to the upcoming discourse (cf. Baker 1977; Haas, Fleetwood, and Ernest 1995).

Baker (1977) analyzes this phenomenon in ASL discourse and identifies three types of “initiation regulators,” or summonses, that serve this function: indexing, touching the addressee, or waving a hand in front of him or her. In addition, Mather (1994) examines the attention-getting strategies of Deaf native signers with Deaf toddlers. One of the strategies that the Deaf adults use is what Mather terms “eye-level gaze” or EL-GAZE. When a Deaf child is engaged in an activity, the Deaf adult moves into the child’s line of vision in order to begin signing to the child. In the interpreted medical encounter, indexing occurs two times, touching occurs one time, hand waving occurs eight times, and EL-GAZE occurs one time. The interpreter produces a total of twelve summonses in the data.

An example of both indexing and EL-GAZE can be seen in the following example. At this point in the interview the doctor is prescribing treatment to help curtail the child’s vomiting. The mother is trying to indicate that she has resolved that part of her son’s illness, and that he is no longer experiencing this particular symptom. The interpreter, obviously unable to manage two linguistic messages simultaneously, is shifting between renderings of the doctor’s and the mother’s utterances so that while neither participant gets it all, both participants get some semblance of the overlapping discourse:

Example 4.23

144

M:

D: ah- when a- when a baby is vomiting, first of all, ah, we immediate- the =

N:

C:

I:

I: BABY VOMIT FIRST, (palms out) BABY VOMIT =

145

M:

D:    = treatment of- of uh vomiting, like with diarrhea is the diet. Uh, and, the diet, basically =

N:

C: (starts crying)

I:

I:      = DO-DO? TO HELP-TO (baby) VOMIT SAME WITH DIARRHEA SAME D-I-E-T FOOD =

146

M:    (head forward, brows raised)

D: = with vomiting the diet is small quantities, of clear liquids, frequently. =

N:

C:

I:

I:      = IMPORTANT WHAT EAT IMPORTANT WITH VOMIT MUST PROVIDE =

147

M: PRO.1-         (TO DOCTOR) PRO.1 GIVE WATER-

D:      = So that you don’t fill his, stomach up, and then it’ll bring it up. =

N:

C:

I:

I:        = SMALL-AMOUNT CLEAR L-I-Q-U-I-D-S HOLD PROVIDE + OFTEN PRO.2 NOT =

148

M:

D: = Uh, in just little, bits, =

N:

C:

I:

I:        = FILL-TO-CHEST DON’T-WANT VOMIT (inflected, from abdomen) AGAIN

149

M: (to doctor) WATER VOMIT T-H-A-T U-P =

D:       = frequently.             And uh, Okay, better to give him, uh =

N:

C:

I: I’ve been giving him water … and he threw that up, too: =

I:        = LITTLE + + BUT OFTEN

150

M:       = (head shake) THAT- PRO.3 NOT VOMIT UP-TO-NOW WEDNESDAY, PRO.1 TRY THAT =

D: =       either       Pediolite or           Gatorade? =

N:

C:

I:

I: INDEX (held up) EITHER THAT #OR PROVIDE P-E-D-I-O-L-I-T-E AND =

151

M:       = NOT WORK. FINISH D-I SAME D-I-D I-N-F-A-N-T-I-L

D:       = uh, she could use either one, Gatorade is over age twelve months Gatorade is- =

N:

C:

I: I’ve already tried those and he threw that up.

I:         = G-A-T-O-R- (shift into mom’s view) G-A-T-O-R-A-D-E FINISH

152

M:       PRO.3 (head shake) B-E-E-N VOMIT UP-TO-NOW LAST-WEEK WEDNESDAY

D:       = in little quantities frequently, little quantities. Uh, and um,

N:

C:

I: He hasn’t been throwing up =

I:         LITTLE + + OFTEN

153

M:       = #SO PRO.1 NOT T-O-O CONCERN ABOUT THAT ANY MORE

D: okay so-        okay        okay

N:

C:

I:         = since last Wednesday, now, so: I’m not too concerned about that anymore.

I: #OK+ +

In line 147 the mother attempts to initiate a turn. The interpreter retains the mother’s content, and renders it in line 149, by completing a part of the doctor’s utterance and talking the floor during the doctor’s moment of hesitation, “And uh,-.” The doctor quickly resumes his turn while the mother continues her explanation. Since the mother is not gazing at the interpreter during her utterance, she cannot know that the doctor has resumed unless she is summoned by the interpreter in order to access the rendition of the doctor’s talk. It is worth noting here that the interpreter could have yielded the turn to the mother, continuing to render the mother’s utterance into English (and possibly overlapping with the doctor’s talk). Roy (1989a) finds that an interpreter interpreting in an academic meeting frequently yields turns to the hearing professor rather than to the Deaf student, and suggests that the relative status of the interlocutor’s social roles might be the rationale behind such a choice. If true, a similar consideration could be at work here. In line 150 the interpreter holds up her hand, index finger extended, to get the mother’s attention and then begins to render the doctor’s portion of the overlapping dialogue. The mother watches the interpreter’s rendition, but when mention is made of “Pediolite” the mother immediately explains that she tried the doctor’s suggestion before and it was ineffective. The mother’s gaze leaves the interpreter after she renders the doctor’s first suggestion (Pediolite), but before she renders the second suggestion (Gatorade). Once again, the mother cannot know that the doctor has made a second suggestion, unless she is looking when the interpreter renders it. In line 151 the interpreter, noticing the lack of eye gaze, shifts into the mother’s line of vision (EL-GAZE) as she resumes her rendition of the second item in the doctor’s list, “G-A-T-O-R- … G-A-T-O-R-A-D-E.” It is interesting to note that this example includes both indexing and EL-GAZE within a short span of discourse. It would be interesting to determine how much the various attention-getting strategies vary in degree, intensity, or politeness. This question is also relevant to the next example, which induces occurrences of both touching and hand waving.

In the following example (seen in parts in examples 3.14 and 3.15) the doctor interrupts the interpreter’s rendition and the mother’s original utterance in order to request clarification of the referent of a first-person pronoun:

Example 4.24

52

M: PRO.3 START VOMIT =

D:     = times did he vomit from the time it started.

N:

C:

I:

I:      = MANY- HOW OFTEN VOMIT. START SICK, VOMIT, HOW OFTEN?

53

M:     = TUESDAY NIGHT, FINISH BY WEDNESDAY AFTERNOON … FINISH

D: Okay.

N:

C:

I:      He started vomiting Tuesday night, and finished Wednesday late afternoon.

I: #OK     

54

M:     VOMIT EVERY TIME PRO.1 TRY GIVE PRO.3 (neutral) SOMETHING VOMIT … #SO =

D:     And the fever’s -

N:

C:

I:      And he threw up, um, and every time he threw up I tried to give him something, so: =

I:

55

M:     = PRO.1 GIVE-UP …                E-V-E-N L-I-Q-U-I-D-S PRO.1

D: You (points to interpreter) tried or she (points to mother) tried?        

N:

C:

I:        = I don’t know: I- (looks at doctor) I tried =

I: (waves)       

56

M:     = PRO.1 + + (glances to interpreter) NO (taps doctor’s arm) PRO.3 INTERPRETER

D:

N:

C:

I:        = to give him liquids too.

I: (taps mother’s knee) PRO.1 TRY OR PRO.2 TRY?

As the interpreter finishes rendering the mother’s utterance, she also waves for the mother’s attention in order to inform her that the doctor has produced an utterance. At this point, the mother has apparently read the doctor’s lips, and in line 55 she indicates that it is she, the mother, who is the child’s caretaker. The interpreter does not see the mother sign her response to the doctor’s not yet rendered question. Both the mother and the interpreter glance back and forth between each other and the doctor, as the mother begins her utterance. The interpreter, intent on rendering the doctor’s question, first waves for the mother’s attention and then taps her on the knee. Interestingly, the mother tries to initiate an explanation two times before the interpreter is able to get her attention long enough to render the doctor’s request. The mother even touches the doctor’s arm and begins a third attempt at explanation by signing to him directly. Unfortunately, the doctor is not visible on the screen at this point in the data. Therefore, there is no way to know at whom he is gazing or what his response is to the mother’s touch. In any case, however, he does not respond verbally. At this point the interpreter touches the mother’s knee, and as the interpreter renders the doctor’s request for clarification, the mother briefly glances at the interpreter, then gazes down and begins her fourth (and successfully rendered) attempt to respond.

The doctor’s request, which is an other-initiation of a self-repair (Schegloff, Jefferson, and Sacks 1977), causes some disruption for the mother and the interpreter. Both the mother and the interpreter recognize the request for a repair. However, the doctor has addressed this request to the interpreter, as the author of the English utterance, including the pronoun in question. The interpreter is not the principal of the utterance, however. The mother and interpreter both exhibit increased eye gaze shifts, and there is a negotiation of turns between them while the doctor waits for a response. After 6.4 seconds, the interpreter and the mother resolve their disruption and the mother begins the repair that is rendered by the interpreter. The fact that the doctor is referring to the English discourse between himself and the interpreter, and that the interpreter and the mother negotiate in ASL the handling of the response, indicates that the interpreted encounter does not represent a single dyad between the mother and the doctor. Moreover, the indication here is that the interpreted encounter does not represent triadic discourse either. This is a critical issue regarding the interpreter’s influence on the interactive discourse and will be addressed in greater depth in chapter 5.

In both of the previous examples there are two types of summonses within close proximity to one another. In future research, it would be interesting to investigate whether or not the four types of attention-getting devices have different interactional functions. It is also noteworthy that while there is such a high occurrence of summonses among the interpreter’s nonrenditions in the actual medical interview, there are no occurrences of such devices in the mock medical interview. One reason for this could be that the mock interview is specifically designed for the purpose of interpreting, with all three participants facing one another and intent on the interpreted encounter frame. An additional explanation could be that there are no children present in the mock interview. In the actual interview, the Deaf interlocutor is responsible for and concerned about her young son, who is crying for much of the encounter. When the child needs her attention, she looks away from the interpreter, thus providing more opportunities for the interpreter to summon her. Nevertheless, this is not the only reason for a summons. The mother attempts to initiate a turn and the interpreter summons her in order to finish a rendition of the doctor’s utterance. Regardless of the causes for a summons, if the interpreter omitted it, the mother would unknowingly miss information. Clearly, these nonrenditions are critical to the management of the ASL-English interpreted interaction.

Like the student interpreter, the professional interpreter’s footing types include those related to the task of managing the interaction. Two categories of these footing types, responses to questions and interference, are found in both the actual and the mock medical interviews. An additional type of footing found in the actual medical interview is the interpreter’s use of summonses. Where summonses provide information that is necessary when the Deaf participant is not gazing at the interpreter, interference provides environmental information for the Deaf participant who is gazing at the interpreter. Thus, these two footings both indicate that certain paralinguistic or contextual information is conveyed to the Deaf participant by the professional interpreter. The absence of responses to questions directed to the interpreter in the actual medical interview is also notable in that the student interpreter attempts the same strategy somewhat unsuccessfully where the professional interpreter succeeds. It seems that a major difference in these results originates in the schema and behavior exhibited by another participant, in this case the Deaf participant. Because the Deaf participant responds to the questions posed by the medical practitioners, and because explanations regarding the interpreter’s footing are provided by the Deaf participant, the interaction is able to proceed relatively smoothly. Through examination of a larger corpus of data, it would be interesting to identify various strategies used by different participants and the effect they have on interpreters’ strategies for interactional management.

Relayings and Interactional Management in the Actual Medical Interview

As in the case of the mock interview, both the existence and the variety of the nonrenditions and the footings they represent indicate that the interpreter does influence interpreted interactive discourse. Nevertheless, the function of many of the footings is related to the goal of providing access to aspects of the interaction that would be unknown to one or another of the participants if the interpreter merely rendered utterances. This supports Roy’s contention that interpreters are not merely conduits providing access to linguistic communication between interactional partners (1989a). Nevertheless, it is interesting to note that, as in the mock case, approximately three-quarters of the interpreter’s nonrenditions are directed to the Deaf participant only. Does this finding indicate that these interpreters are not impartial in the provision of their services? If interpreters demonstrate a partiality for one or another participant (e.g., Deaf over hearing), what kind of influence does this have on the interaction and the participants’ perceptions of interpreted encounters? The next section will address the first of these issues by providing a comparison of the two cases under examination. The latter question will be addressed in chapter 5.

Comparison of Footing Types in the Two Cases of Interpreted Encounters

Examination of the code used in the interpreter-generated nonrenditions indicates that both the student and the professional interpreter directed a majority of nonrenditions to the Deaf participant in each encounter (see table 4.7). In addition, there is a smaller but notable difference in the occurrence of spoken only and simultaneously signed and spoken utterances. Because the choice of one or another language code necessarily ratifies some participants and denies access to others, it might appear on the surface that simultaneously produced utterances allow for ratification of all participants. For an interpreter striving toward neutrality, this would appear to be a potentially effective strategy, and in fact, is a strategy used by the student interpreter on various occasions within the mock interview. Nevertheless, research has indicated that although signed and spoken languages appear to be simultaneously producible as a result of their distinct media of articulation (visual and acoustic), their diverse linguistic structures do not allow for effective simultaneous production (see Johnson, Liddell, and Erting 1989; Johnson and Erting 1989). Thus, it is important to note that the student interpreter produced a higher ratio of simultaneously signed and spoken utterances (20 percent) than the professional interpreter (10.3 percent). Further, the student interpreter’s use of simultaneous code choices tended to result in utterances with flawed and potentially confusing linguistic form. Moreover, the professional interpreter’s use of simultaneously signed and spoken utterances had an apparently different function from those of the student interpreter. The professional interpreter never produced such utterances in the presence of the hearing participants who were not fluent in ASL. Thus, on the basis of these two cases, the simultaneously signed and spoken utterances seem to be problematic for the student interpreter and not employed by the professional as a strategy for ratifying all participants at the same time.

Table 4.7 Comparison of Occurrences of Interpreter-Generated Nonrenditions in Mock and Actual Interviews

In terms of utterances produced only in spoken English, the professional interpreter apparently produces a higher ratio (13.8 percent) than the student interpreter (6.7 percent). However, only two of the four spoken utterances occur with the doctor and/or nurse present in the room. The other two are directed at the hearing researcher during a frame break. Therefore, the professional interpreter produced just two spoken-only utterances of the twenty-nine interpreter-generated nonrenditions found within the context of the medical interview itself (6.9 percent). Based on this comparison, both interpreters produced a comparable percentage of spoken-only utterances. This percentage, less than one-eighth of all the interpreter-generated nonrenditions combined, is quite low. The spoken- only utterances consisted solely of relayings (requests for clarification on the part of the professional interpreter, and a repetition on the part of the student interpreter). This is inconsistent with the function of the signed-only utterances.

The greatest proportion of interpreter-generated nonrenditions are signed only. Both interpreters produced approximately three quarters of their utterances as signed only. Like the spoken utterances, these included relayings (requests for clarification and repetitions). Unlike the utterances addressed to hearing participants, the utterances addressed to the Deaf participants consist of all six types of footings identified in the data, as well as both relaying and interactional management functions. Some of these footing types are applicable only to the Deaf participants. For instance, a summons can serve a unique function in ASL. Hearing participants know that when the interpreter produces a spoken utterance, there is information to which they can attend, whereas Deaf participants who are not gazing at the interpreter will not know unless the interpreter informs them visually.

Other types of footing are not so language- or modality-specific. For example, without clarification from the interpreter, source attribution can be equally confusing to both hearing and Deaf participants. If the interpreter provides more of this type of information to Deaf participants, leaving hearing participants confused, such potentially favorable partiality toward Deaf participants would indicate that the interpreters’ influence on the interaction is not neutral.

By examining the two cases of interpreters’ nonrenditions, it has become clear that the interpreters’ choice of code does have an influence on the participation framework of the interactive discourse. Since the interpreter must make a choice that frequently leaves one or another participant as unratified, and since both interpreters generally decide to sign their nonrenditions, there is a tremendous potential for participants to sense partiality on behalf of the Deaf participants. Moreover, since the hearing participants in both encounters demonstrate interpreted-encounter schema that are at odds with the interpreters’ schema (possibly because any given Deaf person and any given interpreter encounter interpreted events more frequently than does the average hearing person), it would seem likely that the hearing participants could benefit from information related to the management of the interaction. Nevertheless, examination of the interpreters’ code choices indicates that both interpreters prefer two types of interactional alignments: interpreter-Deaf participant and Deaf participant-hearing participant. The interpreters so pointedly attempted to avoid interpreter-hearing participant interactions that they actually avoided responding to interpreter-directed questions from each doctor. The preference for establishing certain participation frameworks is clearly related to the issue of interpreter neutrality.

Comparison of Footing Types: Relayings

In both the mock and actual medical interviews, three types of footings categorized as relayings are identifiable. However, the three types are not identical in each case. The two types of footings that occurred in both cases are source attribution and repetitions. Explanations are produced only by the student interpreter and requests for clarification are produced only by the professional interpreter (see table 4.8). Although both interpreters generated nonrenditions that functioned to attribute the source of rendered utterances, it is interesting to note that almost half of the student’s utterances are devoted to this function while less than a third of the professional’s utterances serve this function. Because the data indicate that confusion regarding source of a message is a real issue, it is not clear why this footing type comprises such a low percentage of the professional interpreter’s utterances. It is possible that the professional interpreter, as a native bilingual with Deaf parents is more aware of strategies used by Deaf adults in determining source (e.g., through visual clues). On the other hand, it is possible that source attribution is a strategy that could improve the professional interpreter’s interpretation. This is an area that would benefit from future research.

Table 4.8 Comparison of Occurrences of Relayings in Mock and Actual Interviews

With regard to explanations, it is notable that these also comprise almost half of the student’s nonrenditions. This is especially interesting since none of the professional interpreter’s nonrenditions consist of explanations. One possible explanation for this is that the professional interpreter is working with a Deaf participant who shares a similar schema regarding the interpreted encounter frame. Thus, explanations regarding the interpreter’s presence or role are actually handled by the Deaf participant, rather than the interpreter, in the actual medical encounter. An interesting area for future research would be to identify whether or not participants (Deaf or hearing) who share a schema of the interpreted encounter frame are more likely to provide such explanations than are interpreters. In addition, it would be interesting, in a larger corpus of data, to analyze the types of explanations that are provided by interpreters and whether or not these types of explanations are similar for Deaf and hearing participants.

The largest percentage of relayings produced by the professional interpreter are repetitions. This is interesting because it is the smallest category of footing types displayed by the student interpreter. Nevertheless, both interpreters produced repetitions for similar purposes (e.g., as a result of overlap), though the professional interpreter also produced repetitions as a way of redoing an interpretation of a given rendition. This strategy might be more challenging for a student interpreter who is not natively fluent in both languages. It would be interesting to conduct a longitudinal study of a group of student interpreters to determine whether this type of repetition becomes more prevalent at some point in their professional development.

The final type of relaying found in the data involves requests for clarification. This type of footing comprised nearly a third of the professional interpreter’s nonrenditions. However, the student interpreter produced no such requests. One possible explanation for this is that the student, who is coping with the processes of interpretation, does not yet have the ability to make such a request, process the information, and render it while remembering and catching up with the utterances missed in the meantime. A second possibility is that the student interpreter intentionally avoids such a footing, in the same manner that she avoids interacting directly with the hearing participant. A third, and more simple explanation is that the student interpreter does not need information clarified. In fact, since the role play is designed to focus on the interpreting task, there is nothing happening in the room aside from the role play itself. Conversely, in the actual medical encounter, the professional interpreter is coping with a small space, the additional presence of a researcher, and a baby crying almost nonstop throughout the encounter. If this latter reason is the case, it is worth investigating to what extent interpreter education programs offer these realistic contextual factors while training students through role plays. For example, do students practice pediatric examinations in which the needs of a real child are incorporated into the technical redoing?

Many similarities and differences can be identified with regard to relayings between the interpreted mock and actual medical encounters. The findings here indicate that future research regarding these similarities and differences would be worthwhile, not only to better understand the ways in which interpreters influence interactions, but also to determine the relationship between how professional interpreters function and how students of interpreting are being taught to function.

Comparison of Footing Types: Interactional Management

Three types of footings are categorized as interactional management in both the mock and actual medical interviews. As in the findings regarding relayings, the three types are not identical in each case. The two types of footings that occurred in both cases are responses to questions and interference. Introductions are produced only by the student interpreter and summonses are produced only by the professional interpreter (see table 4.9). Unlike the student interpreter, the professional interpreter does not provide any introduction of herself or the reason for her presence. It is not clear why the student provides such an introduction and the professional does not. One explanation is that the student has time to do so, since in the role play the doctor does not bear the realities of the hectic schedule of a medical practice. Another possibility is that the professional interpreter knows the interpreted encounter has been arranged, in part, by the researcher. Since the researcher is also a professional interpreter, it is conceivable that the informant has assumed that an explanation has already been provided to the Deaf and hearing participants. Future research regarding the presence of such introductions in interpreted encounters, as well as the impact of their presence and absence, would be useful in understanding the ways in which interpreters influence interactive discourse.

Table 4.9 Comparison of Occurrences of Interactional Management in Mock and Actual Interviews

Both the student and professional interpreter respond to questions. The student interpreter responds only to questions from the Deaf participant, which occur only during the frame break (no doctor in the room). The professional interpreter also does not provide responses to the hearing medical practitioners. However, the Deaf participant does not address questions to the professional interpreter, so it is not clear whether or not she would respond to the Deaf participant in such an event. Nevertheless, it is interesting to note that the one response generated by the professional interpreter, directed toward the researcher, occurs during a frame break in which no medical practitioners are present. It would be useful, in future studies, to determine whether or not professional interpreters respond to Deaf participants and not to hearing participants. It would also be interesting to study the footing between in-frame and out-offrame portions of medical interviews.

A second type of footing generated by both interpreters is the result of interference. Although over one-third of the professional interpreter’s nonrenditions fit this category, the majority of these are the result of logistical issues presented by the research study. Within the medical interview, there is only one occurrence of such a footing on the part of the professional interpreter. This is comparable to the single occurrence found in the mock interview. The types of interference differ somewhat, however. The utterance resulting from interference on the part of the student is the result of an interference generated by the interpreter herself (touching the doctor’s arm while signing). In the case of the actual medical interview, the professional interpreter’s utterance is the result of the physical logistics of the situation. That is, whether or not the nurse is aware of it, the Deaf participant must watch the interpreter while the interpreter renders the nurse’s discourse. Furthermore, the Deaf participant must use her hands while holding her son and the thermometer under his arm. Perhaps, in an attempt to compensate for the logistical dilemma faced by the mother, the interpreter informs the mother when the thermometer begins to fall from under her child’s arm. This type of interference will likely always be a part of ASL-English interpreted encounters. The type of interference generated by the student interpreter is, perhaps, more avoidable.

The final type of footing related to the management of interaction involves summonses. Although the student produced no such utterances, they comprise well over half of the professional interpreter’s utterances. Perhaps the student interpreter has no need to summon the Deaf participant because the classroom has been designed for ease of visual accessibility. The Deaf participant is not there out of real concern for his health, but rather to assist in creating an interpreted encounter. All of the participants are seated and facing one another, and there is no need to summon the Deaf participant to regain his attention. An interesting area for further investigation would be to determine whether or not interpreter education programs provide technical redoings that include circumstances requiring summonses. For interpreting students who are not native users of ASL, practice with such a footing could be very beneficial.

There are many interesting areas for further investigation with regard to the footing types that function as interactional management. Perhaps the most outstanding issue of all with regard to this category of footing types is the fact that the majority of them are addressed only to the Deaf participants. Nevertheless, regardless of the recipients of such utterances, the presence of footing types that influence the sequence of the interaction clearly indicates that interpreters are not simple conduits of language.

Implications

An examination of the production format within an interpreted encounter, and the different types of footing established by the interpreter with each interlocutor, demonstrates that the interpreter participates, in some capacity, in the interpreted interaction. The interpreter generates utterances for a variety of purposes, including both relaying information and management of the interactional structure. Because a given footing represents a participant-based frame of an event, the interpreter’s footing types provide some insight into the interpreter’s negotiation of the Interpreter’s Paradox.

The quantitative analysis of interpreter utterances indicates that a relatively small percentage of interpreter utterances are nonrenditions: 13 percent in the mock medical encounter and 80 percent in the actual medical encounter. Comparatively speaking, it is interesting to note the greater percentage of occurrences of nonrenditions on the part of the student interpreter. This could be the result of the nature of the interviews; if the mock interview is intended to challenge the student interpreter, she might face a greater pressure to generate nonrenditions. It is also conceivable that the professional interpreter generates a smaller percentage of nonrenditions in an effort to minimize her influence on the interaction (an ability developed over years of experience as a practitioner). This represents an interesting area for further investigation. It would be useful to conduct a cross-sectional analysis of the percentage (and footing types) of nonrenditions among a group of professional, certified, native signing interpreters. Similarly, it would be useful to conduct a longitudinal study of student interpreters, to determine if and how the percentage of nonrenditions (and their footing types) change over time.

The findings here indicate the benefit of additional research in the area of interpreters’ footings. In future research, with a larger corpus of data, it would be beneficial to attempt to identify consistent patterns within which various types of footing occur, and within which interactional outcomes appear to be effectively realized. This will assist in determining how interpreters’ choices regarding code and footing type influence interactive discourse. While an interpreter might make conscious choices about the inclusion of certain nonrenditions (such as a summons or an explanation), an interpreter cannot help but make choices about linguistic code when generating utterances. Research regarding the impact of code choices on interactional discourse indicates that code switches can have a profound impact upon the outcome of an encounter (Gumperz 1982). Thus, an interpreter cannot help but make choices that influence the outcome of the interaction, and that, in some way, influence the participants’ perceptions of one another. This being the case, a profound question emerges for professional interpreters. Should interpreters, recognizing that they cannot help but function as a participant within an interpreted encounter, no longer strive to be a neutral, uninvolved participant or should they recognize the paradox of neutrality and strive to minimize their influence on interactive discourse?