Literal vs. Liberal: What is a faithful interpretation?

Literal vs. Liberal: What is a faithful interpretation? Bill Moody Free-lance interpreter, New York City Abstract. Fidelity to the source message, in both interpreting and translating, for both spoken language and sign language (SL) interpreters, has been at the core of our conception of the role of the translator/ interpreter. This article presents a selection of research and writings on the theme of the “faithful” interpretation in an effort to bring this research to the attention of the practicing interpreter. It includes brief sections on the history of conference interpreting and community interpreting, the professionalization of interpreting, models of the interpreter’s role, consumer expectations of interpreting services, expectations of working interpreters, the unique situation of SL interpreters in regard to transliteration and to educational interpreting, and the measurements applied to fidelity in interpreting. The author concludes that a faithful interpretation is ideally one that is co-constructed between the speaker and the interpreter. 1. Introduction One of the tenets of our profession is that we should interpret “faithfully” from source language to target language. This has been understood to mean that we should not interject our opinions, nor should we add or subtract anything from the message. But what exactly is a “faithful” interpretation? Of course, we know that interpretation is very rarely literal, i.e., word-for-word or word-for-sign. We know from long experience that we interpret meaning rather than words because we expect the people for whom we are interpreting to understand each other. But how liberal can we be in interpreting the meaning? How much adding, subtracting, or explaining is still faithful to the message? How “sacred” is the source language message, and how much responsibility do we have to adapt it for the audience of the target language? Does our audience expect to hear (or see) the message not only in their own language but also expressed with the underlying schemas, beliefs and preferences to which they are accustomed? The tenet of fidelity (completeness and accuracy) in interpreting goes straight to the heart of our role, and our vision of what it means to be an interpreter. No one pretends to be able to answer these questions definitively, even for a given situation, but the fact that we are still struggling with such questions is a reminder that our work is extremely complex and requires constant self-monitoring to be as accurate and fair as possible. For the past 15 or 20 years, I have heard numerous comments from spoken language interpreters in conferences and courts, expressing shock at how we SL interpreters describe what we do. We may say that we are “cultural mediators”, or that we sometimes have to “expand” the message in order for it to be clear and natural in the target language. We sometimes admit to making the implicit explicit. We even sometimes talk about being “allies” of the deaf person I will not distinguish between culturally Deaf (with a capital D) and audiologically deaf (with a lower case d) people in this paper: when I use the word deaf, I am referring to any deaf, Deaf, or hard of hearing person using a SL interpreter. Distinctions between people who had deaf parents or not, grew up in the deaf community or not, use SL as their primary, secondary or tertiary language, etc. are best left to articles for which that information is essential to the content., which makes some spoken language interpreters’ hair stand on end. Ideas about the role and responsibilities of interpreters are evolving among both spoken and signed language interpreters. This article will explore the history of interpreting and some of the research on the role and function of the interpreter (in both spoken languages and signed languages). Because I am neither an academic nor a researcher, I have no new research to present, but I believe a review of the work done over the years on the interpreting process will clarify our beliefs and practices about faithfully interpreting in a professional manner in our respective communities. 2. Historical background: from helpers to professionals Interpreting between languages has been a human activity since pre-history. For most of history, it was usually a pretty informal affair: someone who knew two languages (more or less) was asked, sometimes even employed, to act as an interpreter (or, as the situation warranted, a guide, messenger, negotiator, or even spy). Such interpreting sometimes occurred on a high level indeed, e.g., meetings of political or business leaders, military campaigns, treaty negotiations, or international conferences. More often, it simply occurred as “natural interpreting” or “lay interpreting” between interlocutors who enlisted bilingual friends or family to relay messages for them. Historical records indicate that interpreters were recognized, sometimes paid or even honored by name, in numerous civilizations: the ancient Egyptians (especially in their dealings with the Nubians) and Romans (the empire even employed salaried interpreters); the Spanish empire, which explicitly regulated the practice of interpreting in its colonies in the 16th and 17th centuries; the French and Austrians, who used ‘language boys’ in dealings with the Ottoman Empire; the Vatican, whose Lateran Councils required interpreters; the English, French, Spanish and Dutch, who recruited interpreter/guides from the indigenous Americans; etc. (Kurz 1985, Bowen 1995, Pöchhacker, 2004:27). But throughout most of history, interpreting was not accorded the status of a profession, in the sense of an occupation requiring specialized training, knowledge, and skills. Professional interpreter associations of licensed or accredited practitioners who consciously adopted codes of ethics or codes of good practice did not appear until the 20th century. 2.1 Conference interpreting The modern profession of interpreting was born during the 1919 peace treaty negotiations following the “Great War” in Europe and during the early meetings of the League of Nations. The need for interpreters increased as the communication needs of international politics and global trade expanded. By the 1920s, there was enough demand that professional spoken language interpreters began to establish free lance businesses. Professional interpreting schools were founded as early as the 1930s to fill the growing need for interpreters in business and diplomacy. As interpreting became professionalized, interpreters at the League of Nations and then at the United Nations became the standard for the public’s conception of interpreters in general. They worked in the consecutive mode, that is, they would wait until the speaker had finished his speech before rendering the interpretation. In the international bodies, the interpreters took notes during the speeches (which could run as much as an hour or more) and then took the speaker’s place at the podium to deliver a consecutive interpretation of the entire speech. Called “conference interpreters,” their work was highly visible and prestigious. After World War II, simultaneous interpreting, made possible by the introduction of headsets and microphones, proved its efficiency at the Nuremberg trials and, despite some resistance from the entrenched consecutive interpreters at the UN, eventually became the modus operandi for interpreting at UN meetings and other formal conferences of politicians and business leaders (Baigorrie-Jalón 2005). Conference interpreters had gained enormous cachet by the middle of the 20th century and were paid commensurate with their professional status. As the market for professional conference interpreters grew for both diplomacy and trade, professional associations were established. The Fédération Internationale des Traducteurs (FIT) and the Association Internationale des Interprètes de Conférence (AIIC) were established in 1953. AIIC promulgated a code of ethics and professional standards in 1957, and still plays a significant role in the training and employment of high level conference interpreters today. What did this early and influential AIIC Code of Professional Conduct have to say about message accuracy and fidelity? Interestingly, it emphasizes professional confidentiality and working conditions, but remains conspicuously silent about the role of the interpreter and the question of accuracy of the interpretation (the current full text of the Code can be seen on the AIIC website, www.aiic.net). However, in a 1996 survey of users of interpreting services at meetings and conferences interpreted by AIIC members, researcher Peter Moser found that fidelity was the most important criterion for a good interpretation. Of the 201 participants interviewed, “faithfulness to the original” was cited as the first duty of the professional conference interpreter. 2.2 Community interpreting When lay people think of interpreters, they usually envision those who work at high level conferences or for politicians on television. However, most of the interpreting in the world today is done by community interpreters, often with little or no special training, who typically interpret for friends, family or other members of a minority language community who are not fluent in the majority language and who need help in dealing with institutions, schools, and government agencies. Community interpreters, sometimes called liaison or dialogue interpreters, are bilinguals who have usually grown up in the minority community and also understand the language and culture of the majority. Recognition and remuneration has come more slowly for community interpreters than for those interpreters who work at international conferences, and in fact some community interpreters are not paid at all for their work. Even today, many community interpreters who work in legal, health, social service, and other community-oriented settings are not considered professionals. The pioneering countries in the movement to recognize, train, and compensate community interpreters in the 20th century were Australia, Sweden, and the United States. In the 1960s and 70s, laws in the US began mandating services for the “vocational rehabilitation” of disabled people (‘handicapped’ was the term used at the time), including improving education and job opportunities for deaf people. Boyce Williams, an ambitious deaf man who served in government for 38 years, worked his way up to the position of head of the rehabilitation section of the US Department of Health, Education, and Welfare, and he constantly advocated for the inclusion of deaf people in the government’s plans for the disabled. Government planning included funds for the establishment of the National Theater of the Deaf in 1967 (NTD toured for decades all over the world, spreading an image of deaf people as creative, attractive, and artistic); the inclusion of fluent American Sign Language (ASL) users as counselors for job training; the establishing of the first interpreter association, the Registry of Interpreters for the Deaf (RID); and the creation of pilot interpreter training programs around the country. Within a decade of its establishment in 1964, the RID had promulgated a Code of Ethics for ASL interpreters (which mandated not only fidelity of interpretation but also confidentiality and conduct befitting a professional, and which had a great influence on other countries’ subsequent interpreter associations’ codes of ethics or codes of good practice). By 1972 the RID established its own certification system to evaluate its members’ professional competence. From the 1970s through the 1990s, national laws increasingly mandated interpreting services for deaf people, creating an explosion of paid SL interpreter positions. By 2006, the RID had almost 6,000 members who had been certified as professional interpreters, though there is still a shortage of qualified interpreters (the demand is so great that many individuals who have never joined the organization and never been certified are also working as interpreters). In the US, while professional signed language interpreters are becoming ever more organized, community interpreters working between spoken languages in immigrant communities continue to struggle. Court interpreting is beginning to be recognized and organized on a state-by-state basis (some states organize interpreting services better than others), and interpreters in hospital settings have organizations in a few states, but most community interpreters in spoken languages in the US still work without the support of an association and often without training. The situation for spoken language community interpreters is better in the pioneering countries of Australia and Sweden, where the development of spoken language and signed language interpreters has evolved hand in hand. In Australia, anti-discrimination laws mandating access of services (social, educational, and medical) to Aboriginal peoples, linguistic minorities, and immigrants began to be passed in the 1970s. Interpreters were hired for telephone interpreter services, schools, and hospitals. In 1977, the National Accreditation Authority for Translators and Interpreters (NAATI) was established to evaluate levels of competence of community interpreters (from para-professional to professional to conference levels) for both spoken and signed languages. There is a basic and significant difference between the American system and the Australian (and Swedish) systems: the US professional association of interpreters itself develops and administers the tests and certification system for the accreditation of sign language interpreters, whereas in Australia and Sweden, an independent agency tests all interpreters, both spoken and signed language interpreters. In Sweden by the mid 1970s, local governments were routinely hiring interpreters for both spoken and signed languages in medical, legal, religious, and educational settings, and interpreter training programs were established in adult education centers. These developments in Australia and Sweden, and the increased market for interpreting services that followed, began to have an impact on the public’s recognition of interpreters who had been serving minority communities. Whether employed by the government or working as free-lancers, community interpreters working in institutions (hospitals, social security offices, courts, for example) have been recognized and often been viewed as advocates or cultural brokers who go beyond the traditional neutral role of the interpreter (Roberts: 1997). As early as 1978, the American interpreter Robert Ingram was advising the spoken language research community that “no description (practical or theoretical) of interpretations which fails to take account of sign language interpretation can be regarded as complete.” With the proliferation of research on SL and interpreting in the deaf and minority language communities of the world in the 1980s and 90s, the mutual influences between researchers on interpreting in both spoken and signed languages have greatly enriched the research on interpreting in general (Wadensjö and Roy are exemplary in this regard). The growing interest in community interpreting in the field of both spoken and signed languages led to the 1995 “First International Conference on Interpreting in Legal, Health, and Social Service Settings” at Geneva Park near Toronto, Canada. The conference title, “Critical Link” evolved into a series of Critical Link conferences: Vancouver (1998), Montreal (2001), and Stockholm (2004), and Sydney (2007). 2.3 The professionalization/academization of SL interpreting In quite a few countries, community SL interpreters are finally beginning to be recognized as professionals both by the general public and by law. Professionalization is quite an achievement and represents a very different paradigm from the historical model of families and friends volunteering to help deaf people communicate with the hearing. However, the transition from community volunteers to professional interpreters has not always been smooth. SL interpreters before the 1970s were bilinguals (a spoken language and a signed language) who were members of the deaf community, either because their families or friends were deaf, or because they worked with deaf people as educators, social workers, or religious workers. There were no tests of professional competence for sign language interpreters, so members of the deaf community would simply ask those hearing people around them who were fluent in SL and who had demonstrated their trustworthiness to interpret for them (Cokely 2005). Since interpreters were almost never paid for these services, community interpreting was not recognized as a job, much less a profession, either by society or by the interpreters themselves. For these interpreters, “it was a way of contributing to the general welfare of deaf people … in keeping with societal norms of the time in which good deeds were a matter of private, and not corporate, concerns. We felt it was our duty to do it, and if we did not do it, the deaf person would suffer and we would feel responsible.” (Fant 1990:10) Cokely and Fant were writing about the situation in the US, though the same pattern among community interpreters in sign language has been remarked upon in countries throughout the world. Volunteer interpreters with no formal training and with close ties to the community rarely requested payment for their services. They were simply repaying a debt to the community (for their parents’ sake if they had deaf parents, or in exchange for their acceptance into the community as friends or workers learning the language) or they were donating services because they knew that deaf people were traditionally under-employed and would not be able to afford payment. Given SL interpreters’ roots in the deaf community and the interactive natural learning of the language from constant contact with deaf friends and family (rather than from an academic course in SL or interpreting), it is not surprising that the early pioneers in our field conceived of interpreting as helping people understand each other rather than simply translating the words or signs. They naturally interpreted linguistic and social meaning. They advocated for the deaf person because they knew intimately the potential catastrophes awaiting powerless minorities who struggled to understand the “system” and whose intentions were often foiled as a result of gross misunderstandings. For the SL interpreter, being “faithful” meant making sure, as far as possible, that deaf and hearing people could understand and get along with each other. To that end, certain assumptions about the mainstream social system (social services, schools, hospitals, courts, etc.) had to be explained to deaf people, and certain norms in the deaf community (expectations of reciprocity, conversational rules regarding turn-taking or eye gaze, use of interpreters, etc.) had to be clarified for hearing people. Cultural differences had to be smoothed over to allow for successful communication. That which was implicit in either language often had to be made explicit in the other language for the interlocutors to understand each other. Moreover, certain things would simply be too time-consuming to explain or too irritating to one or the other and could very properly be left out in order for the interpreted conversation to reach a conclusion satisfactory to both parties. A “faithful” interpretation, then, was the result of being faithful to the community and its values in addition to being helpful to people who were trying to communicate. By the 1970s, sign languages began to be recognized by linguists as complete natural languages. In a few countries, as sign languages were recognized as true languages, interpreting services began to be legally mandated and interpreters began to be paid. Sign languages in those countries became more standardized as deaf people began to take pride in their language and culture. As a result, instead of interpreters being directly “invited” by deaf people to help out in the deaf community, the accreditation of interpreters began to be a product of government licensing or university degrees. Interpreter education programs in universities or training centers in non-profit associations were created to meet the demand for interpreters. Interpreters began to consider themselves “professionals.” As the legal mandates for interpreting services in countries like the US, Sweden, the UK, and Australia spread, the demand for SL interpreting services blossomed, and interpreters began to consider themselves as “professionals.” As experienced interpreters began teaching and developing curricula, research and reflections about the profession and practice of interpreting grew. New codes of ethics or professional codes of conduct began to influence the attitude of interpreters toward their work. Unfortunately, there were not enough candidates already fluent in SL to fill the new interpreting classes offered in interpreter education programs, and the students who wanted to become interpreters but were not fluent in SL had to learn the language while mastering the theory and mechanics of the interpreting process in an impossibly short period of time (some training programs were once-a-week night classes, while university programs included some night programs and some full-time programs, generally from 2 to 4 years in length). Becoming fluent in a new language is hard enough to achieve in a few years, but learning the language and mastering the skills of interpreting in that short time period was a real challenge. Upon graduating from a university with an interpreting degree, students were too often surprised to find that they were ill-prepared to practice the profession. Little wonder, then, that some deaf people complained that their traditional “friend and helper” interpreters were turning into signing robots who could not meet the needs of the community. Having been often ignored or maltreated by “professionals” throughout their history, one can understand how some deaf people were unhappy that many of these new “professional” interpreters were now too detached from the deaf community. The problem of preparing competent interpreters who can meet the burgeoning demand for quality interpreting services has been a hot topic in the interpreting community during the last decade. Improving the teaching of SL in public schools, improving the curricula in university programs, and internships and mentoring for student interpreters who have graduated from interpreter education programs are part of this on-going discussion (cf. Marschark et al 2005 and Witter-Merithew and Johnson 2005). In spite of such problems, the professionalization of community interpreting has yielded some positive results. The public has realized the need for professional interpreters, and deaf people are beginning to demand that their interpreters act professionally. And one of the most important aspects of this professionalization and academization is that recognition of the profession is leading to increased research. And yet we must recognize that in most countries where norms have been established for the provision of interpreting services, accreditation of interpreters, and curricula for training interpreters, there is still not enough meaningful research into exactly what we do, how we ought to be doing it, and how we ought to be training the next generation of interpreters. Hopefully, this journal, in conjunction with the increasing exchanges among interpreters occasioned by the establishment in 2005 of the World Association of Sign Language Interpreters (WASLI), will spur researchers and practitioners to collaborate and disseminate their findings, and encourage the application of those findings to the field. The last 30 years have seen an enormous amount of research on sign language, including some ground-breaking studies about sign language interpreting, but 30 years is a relatively short time for scientific inquiry into a field, especially one so complex. The cross-fertilization of research in both community interpreting and conference interpreting and across both signed and spoken languages, will be enormously healthy for the field. 3. “Faithfulness” as part of the interpreter’s role Rendering an “equivalent” message has traditionally been considered a purely linguistic activity: the interpreter understands the source language message and translates it in real time in a specific situation for a specific audience into the same message expressed in the target language. The information content is identical. Fidelity, then, is usually portrayed as an ideal where the quality of interpreting can be analyzed by the number of “mistakes” or deviations from the source message (omissions, additions, or substitutions). Given the context, speed, register, intent, and emotional effect of the original, the interpretation should be the virtually the same. The source message is sacred. Jean Herbert, the doyen of UN conference interpreters and author of the 1952 interpreters’ Handbook, affirmed the tenet of “fully and faithfully” interpreting a speaker’s original idea (Herbert 1952:4). He did note, however, that a faithful interpretation in consecutive mode need not take as much time as the original – by speaking a bit faster, avoiding repetitions, hesitations, and redundancies, the interpreter could reduce the original speaker’s time in a consecutive interpretation by about 25% and still be interpreting “faithfully.” (p.67) Danica Seleskovitch, one of the co-founders of AIIC and its long-time Executive Secretary, in “L’interprète dans les conferences internationales” (1968) called for the interpreter to work with “fidelité absolue”, translated in the 1978 English version as “total accuracy”. Seleskovitch was a self-taught conference interpreter who began teaching interpreting in the 1950s. She was the founder of the early and influential “Paris school” of conference interpreting centered around her Ecole Supérieure d’Interprètes et de Traducteurs (ESIT) in Paris. She taught that “absolute fidelity” could be achieved only through interpretation of meaning (la théorie de sens), rather than a mere “transcoding” of words (a literal word-for-word translation). Her “theory of meaning” assumed that the interpreter, once she had understood the source message, would parse the meaning into “de-verbalized” chunks, jettisoning the linguistic form of the original message, so that the pure meaning, determined by the source utterance in conjunction with the interpreter’s knowledge of the context of the particular situation and her general world knowledge, would yield a totally accurate interpretation in the target language. “Faithful”, for Seleskovitch, never passes through a literal word-for-word translation. Instead, the transition from source to target language goes through a cognitive process where the meaning is stripped of its linguistic packaging. Her théorie de sens gained momentum in the sign language interpreting community when she was invited as the keynote speaker for the US Registry of Interpreters for the Deaf conference in 1991 (the English translation of her book with Lederer, “A Systematic Approach to Teaching Interpreters,” was published by RID in 1995). Much research on the ideal of faithful interpreting from Seleskovich’s “Paris school” has centered on information processing during simultaneous interpreting for conferences in the real world. Practical considerations for the conference interpreter have included coping with speed of delivery, density or structural complexity of the text to be interpreted, user expectations that interpreted renditions be fluent and intelligible, interpreting between languages with very different grammatical structures, ambient noise or faulty sound equipment, etc. Such constraints for conference interpreters in the simultaneous mode have led to a variety of interpreter coping strategies. One strategy is waiting, or increasing lag time to understand the speaker’s point before committing to an interpretation (German, which tends to place the verb at the end of the sentence, is often mentioned as a source language requiring the interpreter to wait for a full sentence). Even stalling or using “padding expressions” to fill time while waiting (Glémet 1958, Kirchhoff 1976/2002 and others) can be effective, though both waiting and stalling increase the memory retention load on the interpreter. Chunking, another technique, involves taking phrasal elements out of long complex sentences and interpreting them one at a time in a re-structured rendition of the message with shorter sentences (Kirchhoff, 1976/2002, Jones 1998, Setton, 1999, Ishikawa 1999), and is often cited in writings on simultaneous interpretations from Asian languages. One of the most important strategies employed during successful simultaneous interpreting is prediction, or anticipation. Descriptions of interpreters’ production of an element of the interpretation appearing before the corresponding element in the original is often seen in the literature, and working interpreters periodically realize that they have interpreted a phrase before the speaker said it, sometimes surprising themselves and assuming that they are just “in synch” with the speaker. Such cases can be both an anticipation of words based on familiar lexico-grammatical patterns or a more general “sense expectation” which leads the interpreter to complete a thought to which the speaker has been building but has not yet expressed (Llewellyn-Jones 1981, Cokely 1992, Pöchhacker 2004:134). As long as the original message is “synthesized” accurately, all of the words need not be expressed in the interpreted rendition. Ghelly Chernov, a Russian researcher and interpreter at the UN, wrote about compression or omissions in the interpreted version as a result of processing dense content at high speeds (1978, 1994). Sergio Viaggio (1991:51) argued that “saying it all” was not always necessary to “convey all of the sense [meaning]” (also Isham 1994, Dam 1993). Indeed, some writers have argued that, depending on the language pair involved, “what needs to be said or remain unstated depends on the language and culture in question” (Pöchhacker 2004:135). In other words, what is explicit in one language may be rendered implicit in the other in a condensed or reduced form, and conversely, what is implicit in one language community may need to be made explicit in another, resulting in an expanded version. Advocates of “free interpretation” would argue that the terms “reduced” or “expanded” are unnecessarily negative, that such adaptations are in fact necessary for a faithful interpretation. Daniel Gile, who first coined the term interpreting studies for a field of study separate from translation studies Translation as a generic term has been used to include both written translations and oral (or signed) interpretations. Common usage in the field now separates translation (written or, in the case of signed languages, filmed) and interpretation (oral or signed, and interpreted either simultaneously or consecutively for the target audience in attendance). In the process of translating, source texts may be re-read (or re-played) and will be read (or seen) by audiences at a later date. Interpreters, on the other hand, work from a real-time source language message and the target interpretation is produced live under time pressure for immediate use by the target audience. (in 1992 at the Translation Congress in Vienna), added that the interpretation should be faithful to the original both in “message and style” (1992:189). Though he gives priority to the informational content, he affirms that the form of the target language product should nevertheless be natural and native-like. Canadian Brian Harris (1990:118) invokes the paradigm of the “honest spokesperson” as a standard for faithfulness. In his view, the speaker’s representative (the interpreter) should “re-express the original speaker’s ideas and the manner of expressing them as accurately as possible and without significant omissions.” Most of these reflections on faithful transmission of the message are based on the practice of interpreting between languages spoken by many millions of people (e.g., the “official languages” of the UN) in which “equivalent” expressions have been developed as part of the conference interpreter’s vocabulary. Interpretation services between more ‘exotic’ languages of groups whose norms and languages are structured completely differently, and whose cultures may reflect entirely different schemas, or ways of viewing the world, represent an even greater challenge to the interpreter. 3.1 Target or interactional orientation By the 1980s, some of the work of scholars and researchers in the field of translation studies began to shift from the study of the interpreter’s cognitive processing (how does the interpreter translate from source to target language, producing a target message equivalent to the source) to a more functional approach (Vermeer 1989/2000, Kirchhoff 1976/2002, Stenzl 1989, Shlesinger 1989). The function, or purpose, of the text or the message was justifiably more important than the traditional belief that fidelity to the source message was the goal of all translation. In other words, the target message should serve the same purpose for the audience as the original message, considering “interpretation in the context of the entire communication process from speaker through the interpreter to the receiver.” (Stenzl 1989:24). In the research on community interpreting, the target orientation has evolved into a focus on the interaction of both parties in the dialogue. After all, both have goals that they want to accomplish in the exchange. Since the interpreter is the only participant in the triad who understands the language and culture of both primary participants, she is the one best placed to mediate the exchange so that their goals may be realized. As we shall see, the interactional perspective on liaison interpreting became highly influential in determining the model of how the interpreter in the community functions. The research on interpreting in the last half of the 20th century has pointed the way to an expanded view of the question of fidelity. From the conception of a purely linguistic activity in which informational content is conveyed from one language to another, the interpreter’s role has evolved into the more complex task of juggling the goals and intent of participants from two different cultures under the pressures of time and environment. 4. Models of the interpreter’s role So what is the ideal role of the interpreter? Since the 1970s, there have been several models in a continuing evolution of the research on this question. These models assumed certain norms which informed the interpreter’s behavior. High-level conference interpreters have traditionally been seen as neutral conduits, or machines – and although their task may be complicated by the speed of delivery, the density of the text, and/or the pressures of the environment, they generally do not intervene or manage the flow of talk. The ideal of the “neutral” interpreter became widespread with the professionalization of the field of conference interpreting by mid-century In actual practice, even conference interpreters and high-level international tribunal interpreters find themselves filtering out extraneous information, explaining terms, correcting anomalies or errors, and even intruding on the proceedings (c.f. Jansen 1995 on Spanish/Dutch court proceedings, Morris 1989 and Shlesinger 1991 on the Demjanjuk trial in Jerusalem, and Diriker 2001 on Turkish/English conference interpreters)., described metaphorically as “interpreter-as-window”, “interpreter-as-phone-connection,” or interpreter as “machine” (Pöchhacker 2004). All these images imply that the interpreter simply relayed the message in a clinical fashion. This framing of the interpreter’s role has filtered down to community interpreters. In many community situations, especially in courts, interpreters often heard (and still hear) the admonition: “Just interpret what I say!” The expectation is that the interpreter should be a neutral conduit for the source message to pass to the target language, and only the primary participants in the exchange have the power to construct or manipulate that message. This is in stark contrast to the original role of interpreters before the 20th century, who tended to be highly involved participants (helpers, intermediaries, negotiators, and guides) in the interactions for which they interpreted. SL interpreters and friends and family of immigrant communities serving as “natural” interpreters before the 1970s saw their role as anything but neutral. They saw themselves as members of the minority community and considered it their responsibility to make sure that there was no breakdown in the communication flow, that the speakers in a dialogue got their needs met and intentions fulfilled as much as possible. They were personally invested in the outcome of the exchange; they were there to help. Before any code of ethics was written, their personal ethics required that they merit the trust placed in them by the person for whom they were interpreting. Between the passive conduit or machine (for the most part, humanly impossible), and the involved helper, other interpreter roles have been widely discussed in the interpreting community. In the SL interpreting community, the model of communication facilitator was widespread by the 1980s (Caccamise 1980,Witter-Merithew’s “Claiming Our Destiny” 1986 articles in the RID Views, McIntire & Sanderson 1995, Roy 1993/2002), a model which accorded the interpreter an active role in managing the conversation. The concept of cultural mediator or advocate was also proposed for South African spoken language interpreters (Drennan and Swartz 1999) in a movement of advocating for patients’ rights (and the term “cultural mediator” was also widespread in the SL interpreting community). Lastor and Taylor (1994) and Mikkelson (1998) have argued for the interpreter to take an active role in helping to ensure the rights of minority clients confronted with powerful unfamiliar institutions, especially the judicial system. Barsky argued (1996) that refugees entering Canada needed interpreters to serve as intercultural agents in order to empower the refugees. Redressing power imbalances was a major topic in the sign language community in the 1990s (c.f. Baker-Shenk,“The interpreter: machine, advocate, or ally?” delivered at the 1991 RID conference in the US). These newer models incorporated a more active role for interpreters, highlighting the need for “smoothing over cultural differences,” “bridging cultural gaps” and “serving not only as linguistic but also cultural mediators” (Kondo and Tebble 1997). Cultural mediation is a responsibility that requires a thorough understanding of both cultures. For true bilinguals who are sensitive to cultural differences, this role is possible and may even be natural, as it was for many of the children of deaf parents who became professional interpreters. But let us be honest: for the majority of hearing interpreters who learn sign language as adults, taking on the role of cultural mediator may sometimes be more than they can handle. Novice interpreters without proper training and exposure to the community and culture, even those who have graduated from a 4-year interpreting education program, should be particularly cautious in taking on such a responsibility. Another more recent model of interpreting, the open process model, proposes that interpreters include their deaf clients in on the process of interpreting “to reveal the decision making, struggles and vulnerabilities inherent in the translation process, affording all stakeholders in the communication the ability to participate in and influence the accuracy and integrity of the interpretations delivered.” (Molly Wilson, pers. comm.) Taking as its stance the deaf community’s value of participation and collaboration, the model aims to be faithful to the interlocutor’s messages by including them in the process, effectively co-constructing the message with the participants. Rather than the interpreter taking sole responsibility for the interpretation (and, indeed, sometimes covering up doubts and mistakes in order to be seen as competent and “professional”), the interpreting process becomes a shared discourse between interpreters and participants. Models or metaphors for the interpreter’s role can be helpful, but they tend to be theoretical ideals which prescribe a certain role for the interpreter. Experienced community interpreters may, in fact, switch models in the middle of an assignment, depending on the circumstances and the expectations of the consumers. In determining what the interpreter should do, it is preferable to study and describe what the really good interpreters actually do, and begin to base our conceptions of the role and tasks of the interpreters on such descriptions. Some researchers have decided to do exactly that: to describe rather than prescribe, and we will now focus on the recent literature on interpreting as not only a linguistic activity but also a task of facilitating social interaction. 5. Interpreting as a discourse process Historically, research on fidelity in interpreting focused on linguistic fidelity. However, interpreting has always encompassed much more than being faithful to a source message. Interpreters not only provide clarification, they also actively manage the flow of talk and turn-taking, a coordinating role which incorporates far more than linguistic processing and relaying information. SL interpreters who have deaf parents, (called either CODAs [“Children Of Deaf Adults”] or IDPs [“Interpreters with Deaf Parents”]) have acted as advocates operating under the aegis of the trust of the community – they were in fact entrusted to “manage” interpreting situations. By the 1980s, an interpreter in a US school setting was quoted as saying: “For example, in a classroom, a big problem that deaf people often complain about is that they feel that the interpreter interrupts at the wrong time or they don’t interrupt at all, in which case the interpreter tells the deaf person that it’s really their responsibility and they raised their hand [to signal to the teacher that they wanted to intervene] at the wrong time, and their timing is all off, and deaf people get very embarrassed and … stop asking questions. … How could they know when to interject? The intent is to interrupt [take a turn] but not to make a big scene. I like using the analogy of the traffic cop: if there’s no one in charge of the signals, the cars’ll be going back and forth. Accidents are bound to happen.” (Frishberg , 1986:27) All language communities organize how “talk” happens. Each community has standard openings and closings for different situations. We have rules for turn-taking, interrupting, and signaling that we have finished our turn or that we are not finished. We know how to hedge and stall in order to signal that we want the other person to verbalize something. We instinctively know when it is more polite to be indirect or when we need to be more blunt. We know when to look directly at our interlocutor or when we can avoid eye contact, and we know how much physical or emotional distance to put between ourselves and another person. We are conscious of strategies to save face or help another save face. “What interpreters actually know (intuitively or objectively) and do is complex from both the perspective of psycholinguistic processes and also from the perspective of interactive communication systems as a whole. Interpreters are not simply processing information and passively passing it back and forth. Their task requires knowledge of a discourse system that includes grammar, language use, organization, participant relationships, contextual knowledge, and sociocultural knowledge. Interpreters must also have the ability to adapt this knowledge quickly to size up a situation, anticipate problems, and decide on solutions within seconds…” (Roy 2000:103) Language can serve many functions in addition to informing: it is used to get something, to project a self image, to make the interlocutor do something, to discourage the interlocutor from fully participating in the exchange or to dissociate oneself from a partner, to fulfill a desire for social approval, to incite the interlocutor to react, to insult, to invite, to promise, to threaten, to give directions… and the intent of an utterance will be reflected in the language used. Considering the various functions of language, it is incumbent on us as professionals to ask: what is being faithful to the participants’ messages? As we have seen, most of the academic research on interpreting until the 1990s focused on either translating texts or interpreting at high-level conferences, and the bulk of that research was centered on the cognitive processes of interpreters, interpreter errors (deviations from the source message), and the study of translating texts or monologic speech (rather than dialogues or interactive social situations). Moreover, interpretation has been seen as occurring one speech or one utterance at a time: the complications of dialogic conversations (overlapping speech, awkwardness in turn-taking, interpersonal conflicts, and difference in status of the speakers) were rarely taken into consideration. With the rise of the field of socio-linguistics, however, the view of interpreting as a purely linguistic event began to change. In the field of translation studies, as early as 1964, Eugene Nida called for translations to be studied as communicative events (a text, after all, even a classic text, was translated for readers of a particular time and place and meant to communicate to that audience). For Nida, faithfulness to the original could only be measured by “the extent to which people really comprehend the meaning.” (Nida and Taber 1969:173) And in the field of interpreting studies, even Seleskovitch’s dictum of “fidelité absolue” requires the interpreter to “have his listeners understand it as well as it was understood by those who heard it directly from the speaker himself.” (1978:102) In 1976, the American sociologist Bruce Anderson noted that spoken language interpreters agreed that only a small percentage of their work was conference interpreting, and that their typical community assignment primarily consisted of three people: two primary participants (an authority from the majority culture – such as a doctor, social worker, or government employee – and a member of a linguistic minority seeking services) plus an interpreter. Most interpreting events, therefore, consist of a social interaction between participants of different status, each playing a prescribed social role. He also noted that the interpreter, as the sole bilingual in the triad, had at least some power over the flow and even the outcome of the interaction. 5.1 Socio-linguistic research on discourse The focus on interpreting as mediated discourse was influenced by the work of Sacks, Schegloff, and Jefferson (1974), published in the journal Language as “A simplest systematics for the organization of turn-taking in conversation.” Their research was an investigation into the structure of direct (rather than interpreted) conversation between interlocutors, though its impact on the work of the interpreter will become evident. The structure of a dialogue is a sequence of turns, proving the “thoroughly interactional character of conversation.” Conversations require the participants to “display to each other, in a turn’s talk, their understanding of the other turn’s talk.” A turn can be introducing a new topic, contributing or questioning an idea, changing the subject (for any number of reasons), or simply a comment to keep the conversation moving forward. Their work in categorizing types of turns, silences, and overlapping talk laid the foundation for examination of how interpreters deal with such complexities in interpreted conversations. This type of conversational analysis provides the researcher with concrete points of inquiry to compare how interpreters mediate between the differing and sometimes conflicting conversational norms of participants from different language communities. John Gumperz, who coined the term interactional sociolinguistics, showed that participants in a conversation engage in an on-going process of listening to assess the intentions of their interlocutor in order to formulate a response to accomplish their own intentions. What a person means must be determined not only by linguistic output (what is said) but also by knowledge of the expectations, social roles, and world view of the listener (Gumperz 1982). Deborah Tannen, the researcher and author of several popular books on conversational style, including You Just Don’t Understand: Women and Men in Conversation and That’s Not What I Meant!, proposed that speakers with different conventions for signaling meaning have different conversational styles, and these differences can cause significant misunderstandings not only between speakers of different language communities, but also between speakers of the same country with different characteristics: class, gender, race, age, etc. (Tannen 1984). Like Gumperz, Tannen’s method consisted of recording naturally occurring conversations, identifying problems in communication, then analyzing how the participants felt about each moment of dialogue with input not only from the participants but also from outside observers familiar with one or another of the conversational styles. 5.2 Discourse research applied to community interpreting Susan Berk-Seligman’s 1990 study of English/Spanish interpreters in the ‘adversarial’ courtrooms in the US, “The Bilingual Courtroom,” was the first to analyze the actual performance of interpreters, gleaned from hundreds of hours of recorded data. She found that interpreters were in fact active participants in the proceedings: requesting clarification, asking witnesses or lawyers to repeat a statement, clarifying ambiguities, asking for permission to speak to the court to report grammatical problems or difficulty with dialects, etc. In addition, she found that interpreters even managed the testimony by asking witnesses if they understood or by prompting them to answer. Cecilia Wadensjö’s studies from a large data base (audio recordings) of Russian/Swedish interpreted situations in Swedish health care clinics and police stations (1992,1998), grounded in Erving Goffman’s work on the roles of speakers in face-to-face interaction (1981), have been influential in both spoken and signed language interpreting communities. She posits that what the interpreter actually says or does in the course of the interpretation can be described as either translating (relaying the message) or coordinating the interaction (or more precisely, co-coordinating the flow of talk with the participants). Interpreters, in other words, are working with the participants to accomplish their goals in the interaction; helping them project the image of themselves which they think will further their intentions in the situation given the social roles of the participants; relaying the information that will help each assess the other’s intentions; orchestrating their turn-taking (in addition to dealing with utterances directed to the interpreter or even turns initiated by the interpreter); and even deleting utterances which may violate social roles or expectations. She categorizes turns initiated by the interpreter as either “text oriented” (requests for clarification, comments on prior utterances) or “interaction-oriented” (requests to go ahead and talk, or to stop talking; management of turn-taking, etc.). She describes the basic interpreting situation as a “pas de trois”, or “trio dance”. Wadensjö’s work coincides with the increase in attention to and research on community interpreters, and she has been actively involved (with Canadian Brian Harris, Roda Roberts of the University of Ottawa, and Holly Mikkelson of Monterrey, California) with the “Critical Link” movement for the development of such research. 5.3 Discourse research applied to SL interpreting Tannen and Wadensjö’s work in the hearing world has been applied to the world of SL interpreting by Cynthia Roy, among others. Roy’s 1989 dissertation analysis of a short interpreted meeting between a hearing professor (Deborah Tannen) and a deaf student (Clayton Valli, a noted linguist and ASL poet) was updated and published as “Interpreting as a Discourse Process” in 2000. Roy video-taped the meeting, and, like Gumperz and Tannen, analyzed how the participants felt about each moment of dialogue: did they feel the meeting went smoothly (and where did they feel any conversational discomfort); did they feel that their goals were being met; did they feel that the interpretation allowed them to assess the other’s intentions and respond accordingly; was the meeting “successful”, and, in particular, why did they decide to speak, or take a turn, at a particular moment, and how did they feel about the flow of turn-taking? Like Sacks et al, she analyzed: regular turns (those with smooth transitions which felt like the participants were talking directly to each other without an interpreter); turns with an appropriate lag time; turns with what was perceived as a long lag time; preventing or ignoring a turn; and overlapping talk. She also noted when the interpreter initiated a turn in order to manage the flow of conversation, when the interpreter prompted a participant to take a turn, and when the interpreter yielded a turn to the more powerful participant (the professor) in order to further the goals of the encounter for both participants. Although there is a widespread assumption that the participants in an interpreted event are talking directly to each other, and although they sometimes feel that they are, Roy concludes that each participant is in fact exchanging turns with the interpreter in accordance with the norms of their own language. “Both speakers nod their heads, smile and silently laugh … at moments that co-occur with utterances they understand in their own languages… phenomena around turns, such as pauses, lags, overlapping talk, and simultaneous turns, are going to occur naturally and as they are created by all three participants. The on-going recognition of such discourse features are part of an interpreter’s competence and the resolution of discourse confusion, if necessary, belongs primarily to the interpreter.” (p.67, 68, 72) She concludes that turns taken by the interpreter were often a combination of the interpreter’s decisions and the primary speakers’ tacit agreement to accept those decisions. “This means that the interpreter is an active third participant who can influence both the direction and outcome of the event, and that event itself is intercultural and interpersonal rather than simply mechanical and technical.” (p.6) Melanie Metzger’s 1995 study of an interpreted pediatric examination reaffirmed that the interpreter initiated turns (e.g. to get the attention of the deaf participant when the doctor had not finished talking). However, while the interpreter was definitely acting as a participant, she notes that interpreters are “far more constrained in their participation than other participants” (p.248). Her analysis showed that fully 8% of the professional interpreter’s total talk during the examination was not relaying the talk of either of the primary participants: most of this interpreter-initiated talk was either directed at the deaf participant or requests for clarification directed to either the hearing or deaf person. Metzger’s 1999 book, Sign Language Interpreting: Deconstructing the Myth of Neutrality, concluded that it was unrealistic to expect the interpreter to remain completely neutral. Like the participants, interpreters have their own “frames of reference”, which have been shaped by their own life experiences. The act of interpreting is inherently an act of making choices as to what the speaker’s utterance means to himself, to the interpreter, and how it might be understood by the listener. In fact, one can generalize that all utterances, whether they are direct or relayed by an interpreter, must be mentally processed to arrive at meaning. Further, any understanding of meaning will perforce be influenced by the listener’s world knowledge and awareness of the speaker’s perspective (particularly important in cases of irony or humor). The meaning of any interpreted utterance will pass through several additional filters on its way to its target: it is perceived through the interpreter’s senses, understood in the interpreter’s mind (as images, concepts, or feelings), then formulated into a target language utterance (with contextual information and whatever cultural adaptations are required for comprehension). The interpreted utterance in the target language will, in turn, be perceived by the interlocutor through his or her senses, interpreted into images, concepts, or feelings before being formulated into a final mental representation according to the schemas and world knowledge of the interlocutor. (Further filters in the relaying of messages in the real world may also arise from visual distractions or noisy environments, participant or interpreter fatigue, etc.) How does this new perspective of interpreting in both spoken and signed language communities as a discourse process among people in the real world, rather than as a purely linguistic activity, influence our view of faithful interpretation? Being faithful to the source message, a linguistically equivalent interpretation of words or signs, is clearly not sufficient for all interpreting situations. Let’s review some of the literature on what participants in interpreted situations expect of their interpreters, and then review the research literature on fidelity or accuracy, and see if we can come up with some guidance on how we can provide a genuinely faithful interpretation. 6. Expectations of users of interpreting services Our consumers’ expectations vary depending on the situation and on their notion of the process of interpreting. As we have seen, there are those who assume that the interpreter is like a machine, completely neutral while rendering a fully literal rendition of what is said. This conception of an interpreter may be only an illusion, but it is nevertheless still widespread. On the other end of the spectrum is the vision of an interpreter as not only a relayer of messages, but also an advocate for the views of one or the other, or both, primary participants. Somewhere in between these two is the model of interpreter as faithful conveyor of message and “honest spokesperson” who may smooth over cultural differences, correct speaker errors, eliminate repetitions or extraneous information, and help navigate awkward moments. 6.1 Expectations of literal interpretations A strictly literal translation or interpretation favors form over meaning. People are sometimes so attached to the way they express themselves that they want their message interpreted exactly “how I said it”. 6.1.1 COURTS AND TECHNICAL CONFERENCES Judges and magistrates in many countries are notorious for their insistence on “verbatim” interpretations. Laster and Taylor (1994:112) describe the verbatim requirement of some court systems as “a legal fiction necessitated by the inadmissibility of hearsay evidence (i.e. information reported by someone other than the witness) in the common-law courtroom.” Certainly, judges desire to maintain their exclusive right to “interpret” the law in their own courtroom, and lawyers jealously guard their prerogative to “interpret” both the testimony and the law for the benefit of judge and jury. Woe to the interpreter who dares to explain what a judge, lawyer, or witness says in language different from the original form! And yet, surveys of expectations of users of interpreting services indicate that even in the courtroom, opinions vary as to how far the interpreter may legitimately stray from the form of the original message. Fowler (1997) in surveying magistrates in England found that although they conceive of the interpreter as merely a conduit, “they certainly made it clear that when breakdowns occurred it was the responsibility of the interpreter to rectify them.” Arlene Kelly, studying interpreting in Massachusetts courts, found that a majority of the 53 legal professionals in her 2000 survey were opposed to the interpreter including “cultural mediation” in their interpretations. Mira Kadric, in a survey of 200 local judges in Vienna in 2001, found they would accept that the interpreter assume tasks such as “simplifying the judge’s utterances” and “explaining legal language” for their clients, and 85% of them expected the interpreter to explain cultural references to the court. Several other studies have described the various ways in which interpreters function as more than mere conduits (Berk Seligson 1990, Jansen 1995, Shlesinger 1991, Morris 1989). Delegates to a highly technical conference full of jargon-laden speeches for specialists may also expect literal interpretations. Those specialists may require this sort of word-for-word transcoding as they cannot expect the interpreters to understand everything. They realize some interpretations will be more literal renditions from source to target language, from which they will be able to re-construct the meaning from their own specialized knowledge. Conference interpreters are the first to admit that in situations where complete understanding by the interpreter is impossible, and where the speaker and audience share information that is inaccessible to the interpreter, they will opt for more word-for-word transcoding. In these instances, literal translations of at least part of the content are inevitable, and the audience adjusts to the literal interpretation. In such a situation, a substantial décalage, or lag time, is impossible, because the interpreter’s short term memory cannot long retain information that is not processed for meaning. Even in non-technical situations, experienced interpreters faced with a long list of numbers or names will resort to the same strategy: they cut their lag time considerably and transliterate the form. Similarly, deaf participants familiar with technical jargon which is undecipherable to the SL interpreter will expect the interpreter to tend towards transliteration with a severely curtailed décalage. 6.1.2 TRANSLITERATION FOR DEAF ADULTS In the world of SL interpreting, where some of our deaf clients are truly bilingual (having mastered the written form of the spoken language as well as the national SL), some deaf or hard-of-hearing people prefer to see the interpreter render a literal sign-for-word transliteration Transliteration, a term historically meaning changing letters or words into the corresponding characters of another alphabet (e.g., Cyrillic to Roman), is used in the field of SL interpreting to mean literally translating word-for-sign or sign-for-word between a spoken and a signed language. The resulting transliteration, a visual signed version of the spoken language, is referred to as “signed English,” “signed French,” etc., depending on the country. Such strictly literal translations between spoken/written languages are rare, usually reserved for educational purposes, as, for example, when a student in Latin passes through a literal translation phase in order to better understand the structure of the Latin sentence. of the message, even when the content is non-technical and the interpreter understands enough to interpret in a natural sign language. One instance in which our deaf consumers may prefer transliteration is when the deaf consumer has more status or prestige than the hearing participant. In several countries, quite a few deaf people have risen to positions of authority in government agencies, schools, or organizations. Deaf people who are highly skilled in the language and cultural norms of the majority often want to know literally how the hearing people around them are communicating in the majority language. They may wish to make their own judgments about their hearing interlocutor through evaluating their exact word choices or may want to “test” hearing persons to see how much implicit information they can comprehend or how much they know about deaf culture. In such situations, these deaf people prefer dealing with cultural differences themselves rather than deferring to the interpreter to make such decisions. Other instances in which the bilingual deaf clients may ask for or use transliteration is when they simply do not trust the interpreter’s fluency in the natural sign language, and prefer to effectively do the “interpreting” work themselves from a “signed English” (or “signed Swedish”, etc.) version, or when they are in an either highly technical or highly contextual situation where the interpreter simply does not have the background knowledge to adequately interpret the nuances of meaning in the exchange. Though such a literal interpretation may enable the deaf bilingual access to form as well as content, strict transliterating is not well suited to the visual medium. In a natural sign language, the compacting of information through the use of space and the incorporation of myriad details in, for example, the location and movement of a signed verb, is generally not possible in transliteration. Strict transliterating tends to be rapid and both mentally and physically demanding for the interpreter, and at the same time, visually “dense” for the deaf consumer. Worse, a strict sign-for-word transliteration from a SL into a spoken language would result in something that would sound uneducated or pidgin-like to the hearing consumer. In all signed languages, there is no precise boundary between the “natural” sign language and a signed transliteration of the country’s spoken language (“signed Italian,” “signed English,” etc.). As a result, the boundary between a “free interpretation” and a “literal interpretation” is also imprecise (cf. Lucas and Valli 1992, and Davis 1989/2005 for detailed descriptions of contact language in the deaf community and code mixing in interpreting between signed and spoken languages). In Jemina Napier’s study of Australian university lectures interpreted into Auslan, she found that skilled interpreters switched between free and literal interpretation methods as a linguistic strategy in order to ensure access to content while still providing subject-specific terminology through fingerspelling. She argues that university students and adult deaf people express themselves along a continuum of signing styles which includes both Auslan and signed English, and that they expect interpreters to do likewise as the situation warrants (Napier 2001/2005). 6.1.3 TRANSLITERATION FOR DEAF STUDENTS The question of transliterating for deaf students has been particularly contentious in primary and secondary schools, where the children may not yet have clear expectations of the interpreter’s role. While the consensus among scholars and practitioners of interpreting overwhelmingly favors the free interpretation method for interpreting as a general rule, some educators have proposed using strict transliteration in classes where deaf students are integrated into hearing classrooms. The generation of deaf students taught with this philosophy, however, has not shown any significant increase in literacy (Marschark and Spencer 2003). Indeed, there is little research on how valuable any “interpreted education” (some even use the pejorative term “second hand education”) can be for a very young deaf student learning through an intermediary, whether it be transliterated or interpreted. Marc Marschark and his collaborators (2005) point out that sign language interpreters in the classroom may want to carefully consider how much “explaining” or “expanding” (techniques of free interpretation often used with adults) should be used in interpreting for students. They emphasize, for example, that implicit information from the teacher is often a crucial part of the student’s learning process, and that while school interpreters may consider that making such information explicit facilitates comprehension, “providing such inferential links may not be in the students’ best interest [because it may] discourage them from doing so themselves in other settings or perhaps even [prevent them from] being aware that such links are helpful.” (p.68) 6.1.4 TRANSLITERATION : A FAITHFUL INTERPRETATION ? Transliteration, both for deaf people in positions of authority who prefer literal interpretations in some situations and for deaf students in educational settings, constitutes a significant difference between signed- and spoken-language interpreters’ practice. Community interpreters in minority spoken languages may use some code-switching and code-mixing (as do their clients), but they are never expected to provide full-fledged literal interpretations in their everyday practice. Members of spoken language minorities who achieve positions of power in the majority culture usually are assimilated and speak the majority language fluently, functioning without interpreters, whereas deaf people in such positions, although perfectly competent in the majority written language, generally still prefer interpreters or transliterators for face-to-face verbal exchanges. Interpreters in the US were routinely accredited separately for interpreting and transliterating skills in the RID certification process until 2006; most other countries have not tested separately for transliteration skills, considering a sign language to be a continuum between the natural sign language and “signed English” (or “signed Japanese”, etc., depending on the country). The certification in the US, revised in collaboration with the National Association of the Deaf (NAD), now follows more closely the Australian model: a 3-tier generalist test with interpreters’ skills and knowledge rated as “para-professional,” “professional,” or “conference level” interpreters (the American levels are labeled “professional,” “advanced,” and “master,” dispensing with the entry-level para-professional). In the UK, interpreters are qualified as “junior trainee,” “trainee,” or “member of the register.” In all 3 accreditation systems, transliterating now is incorporated into the generalist test as simply a part of what may be expected of interpreters in their day-to-day work. Clearly, the choice of literal or liberal interpretation for sign language interpreters depends on the situation and the expectations of the deaf participant, whether child or adult. If the goals of the deaf person in the interaction include a knowledge of the exact expression used in the spoken language, then a more literal version of the hearing person’s utterances would be the more “faithful” interpretation in the eyes of the deaf participant. What the client expects should have great influence on what the interpreter delivers. In the US, another consideration on how liberally interpreters can interpret, is in the relatively new area of video relay interpreting for telephone calls. There are legal constraints on how “free” a translation can be when a deaf person uses a Video Relay Service (VRS) to call a hearing person through an interpreter. Government regulations for VRS calls in the US are based on an older legally-mandated system of telecommunication between hearing callers (with a standard voice telephone) and deaf callers (with a text telephone) through an intermediary Communication Assistant (CA) who typed the voice message from the hearing caller and then voiced a typed response from the deaf caller. The CA was required by government regulations to type exactly what the hearing caller said, word for word, and say exactly what the deaf caller typed. Funded by a minimal surcharge on all telephone calls in the nation and regulated by a federal agency, the video relay SL interpreter, if the regulations were to be strictly applied, would have to interpret word-for-sign, sign-for-word, even though such literal interpretations are virtually impossible and the consumers would surely not be satisfied. 6.2 Expectations of “free” interpretations Some users of interpreting services are less concerned with exact expressions in their interlocutor’s language and more concerned with the fluency and flow of conversation. Such consumers expect a more “free” or liberal interpretation: the interpreter is expected to mediate the interchange, to make all the cultural adjustments necessary, so that no time is lost in needless confusion or misunderstanding. Clayton Valli, linguist and ASL poet, expressed it this way: “I feel that interpreters should side with Deaf people. If something goes wrong, it’s usually that hearing people don’t understand Deaf Culture. Interpreters need to lean more toward Deaf people and make sure that things are working out. Yes, I feel that interpreters need to help, well, not help but add things that inform Deaf people about the ways of hearing people. When interpreters hear something and they know that the Deaf person is probably going to have a difficult time understanding it, then they should add information that explains things so that Deaf people get it.” (Roy 2000:60) The interpreter, after all, is the only one who has full access to both languages and “thought-worlds”, so it is often more efficient and sensible for the interpreter to assume the task of ensuring that the participants understand each other, rather than letting them flounder as they sort out the source of a misunderstanding. Conversational strategies which are perfectly appropriate in one culture but might be face-threatening or awkward in the interpreted situation can be adapted by the interpreter to the circumstances, permitting the participants to comfortably get on with achieving their goals. When the goals of the participants include cooperation and efficiency, then the interpreter’s fidelity to their intentions will include tasks that are not strictly linguistic renderings. Certainly, community interpreters hired by institutions such as hospitals (Davidson 1998/2002, Galinda Bolden 2000), police (Donk 1994, Scheffer 1997), universities (Roy, 2000), etc. find that they often become advocates for institutional efficiency. In order to save time or to avoid discomfort or irritation on the part of professionals or clients the interpreter may initiate follow-up questions, omit extraneous narratives, or explain unfamiliar vocabulary to keep the interview on task. 6.3 Surveys, research on user expectations Surveys of users of spoken language interpreting services have attempted to determine consumer expectations for settings as diverse as community health centers in South Africa (Drennan and Swartz 1999), health care workers in Austria (Pöchhacker 2000), formal conferences (Kopczynski 1994, Moser 1996), and community service providers in Canada (Mesa 2000). In general, these consumers reported that they expect less cultural mediation and interpreter intrusion than the interpreters themselves report that they actually do, but interviewees’ answers, interestingly, depend on who is being surveyed and in what setting. Kurz found that delegates at a medical conference, engineers, and diplomats (1989/1993/2002) differed in their expectations of conference interpreters depending on their professions. Expectations for media interpreting, studied by Kurz and Pöchhacker (1995) gave priority to criteria such as “pleasant voice”, “native accent”, and fluency (media, after all have audience ratings to monitor). A “pleasant voice” and “native fluency” in the target language for hearing participants is analogous to native-like sign production and aesthetic spatial organization of discourse for users of SL. While it may seem axiomatic that accuracy is more important than fluency, consider that (especially for non-technical meetings), awkward phrasing, unnatural pausing, or inelegant intrusions from the source language can be so off-putting that the audience may be unable to pay attention to the message. As we have seen, users of interpreting services provided by AIIC interpreters in a 1996 survey responded that “faithfulness to the original” was the first duty of the professional conference interpreter (the exact count was 45%), followed by fluency and clarity of delivery (34%). The more experienced conference-goers were even more exigent: 53% of the experienced users demanded fidelity compared to 35% of the less-experienced delegates. Other qualities of a good interpretation mentioned in the interviews included synchronicity (avoiding extended lag times), rhetorical skills, and voice quality, though expectations varied greatly depending on the meeting type (large v. small, general v. technical), age, gender and experience of the responder. In another survey of users of conference interpreting services, Collados Ais (1998/2002) found that even if participants rated fluent delivery as less important in surveys, a “lively delivery” in fact affected their ratings of interpretations more than content errors (which they may not even have perceived at all!). In other words, it doesn’t matter how accurate the message is if it is expressed in an awkward manner or is difficult to understand, and conversely, an interpretation with many inaccuracies delivered fluently and with confidence was often seen as a “good” interpretation. In contrast to the surveys conducted in conference settings, Edwards, Temple, and Alexander (2005) studied the expectations of users of community interpreting services in two cities in the UK. They found that members of minority communities using interpreters for social services noted “personal character” and “trust” as the qualities of a good interpreter, ahead of linguistic competence or fidelity. This predilection often leads to the utilization of family members or friends for interpreting, similar to what occurred within the deaf community prior to the professionalization of SL interpreting services. Deaf people surveyed about SL interpreting mention the “attitude” of the interpreter as often, or more often, than fidelity to the message. “Attitude” usually implies not only respect for deaf people, their language, and their community, but evidence that the interpreter’s cultural competence comes from authentic interaction with that community (Smith & Savidge 2002, Witter-Merithew & Johnson 2005). Anna Witter-Merithew and Leilani Johnson, in focus group interviews with American deaf consumers of interpreting services in the US, found relationships between community interpreters and members of the community to be complex (payment for services often comes from a third party and the deaf consumer has a particular commitment to the on-going provision of services), but it was clear that deaf consumers stressed the importance of linguistic and cultural competence, professionalism, and interpersonal skills. Jemina Napier and Meg Rohan designed a study of Australian deaf users of interpreting services using a diary of 31 deaf persons’ reactions to actual interpreted meetings or events during a 6-week period, with follow up focus groups to elicit additional responses about the consumers’ perceptions of the quality of the interpretations. Not surprisingly, deaf consumers were generally more satisfied with interpreters of their own choosing (which was the case 34% of the time for these respondents, many of whom were deaf professionals) and with accredited professional interpreters (rather than para-professional or liaison interpreters). In positive comments about interpreters, deaf consumers cited professionalism (punctuality, attire, etc.) as their first comment 40.7% of the time, while efficacy of message translation was cited only 12.3% of the time in their first comments. Most comments centered on professionalism, adequacy of the message translation, personal characteristics, and signing skills. It is not always certain that what users indicate on a survey or what consumers tell a researcher actually reflects how they respond in a real interpreted event, but such studies do provide an overall sense of what our some of our consumers say they prefer. 7. Expectations of working interpreters Expectations of consumers can vary widely depending on their goals in the interpreted interaction and on the situation, and the attitudes of interpreters themselves may also vary widely. Conference interpreters, media interpreters, and court interpreters usually consider that they are hired to relay information in a “standard” form of the language. The major world languages, those in which the high-level conference interpreters work, are not only recognized, but have evolved into languages with the rich vocabularies needed for every occupation or profession practiced in the modern world, as well as a full range of registers for formal and informal situations. Even the least educated members of those language communities usually have had some exposure to the “standardized form” of such languages, either oral or written. Adapting to individual consumers or “explaining” is considered patronizing or even dangerous. Interpreters are professionals who facilitate the linguistic transfer of messages, but endeavor not to affect the outcome of the interchange. Community interpreters, on the other hand, may feel torn between a perceived duty to the minority community and the demands of the profession. Community interpreters may be more willing to forego faithful renditions of linguistic utterances in order to enable the participants to achieve their aims as efficiently as possible – they may even come to the assignment with an expectation of fairness and equality. In the SL interpreting community, interpreters who are deaf themselves are increasingly being hired as relay (intermediary) interpreters. These deaf interpreters are finding employment in conferences (“platform interpreting”), working with a hearing “feed” interpreter, to assure maximum audience identification with an interpreter who is not only a native signer but who identifies as “Deaf with a capital D”. Deaf interpreters are also used in dialogic meetings as “relay” interpreters (especially with deaf clients whose language is idiosyncratic or non-standard and difficult to understand for some hearing interpreters), in meetings where the deaf relay interpreter may need to function as an advocate as well as an interpreter, and as relay interpreters for deaf-blind clients. These deaf interpreters, especially in community interpreting situations, may have a different perspective on the role of the interpreter with a deaf client, though the practice of using deaf interpreters is new, and there is, thus far, precious little descriptive research on exactly what the deaf interpreter does in practice (cf. Eileen Forestal in Marschark et al 2005). 8. Translations In the field of written translation, the debate between free translation and literal translation has been raging for centuries. Even without the interpreter’s time constraint which forces an immediate rendition, translators of the written word have engaged in the same debate between literal and free translations as we interpreters. 8.1 Sacred texts The paradigm for the sanctity of the source text is scripture, or religious texts. How can the sacred word of God be translated faithfully? Would altering the structure, or form, of God’s language be “unfaithful” to the meaning? St. Jerome (c.340-420 CE) believed that translation in general should prioritize meaning over words, but he could not bring himself to accept free translation in the case of Holy Scripture, where word order constitutes a “mysterium” and therefore could not be changed (Vermeer 1994). Historically, translating sacred texts has been a dangerous business. Even Jerome had to flee Rome to complete his Latin Vulgate Bible translation in Bethlehem. During the 14th century, John Wycliffe translated the Bible from Latin to English, and his body was dug up and his bones burned for his efforts. William Tyndale’s translation of the Bible appeared a few years later and his reward was a death sentence for heresy. 8.2 Secular texts Secular classic texts have presented translators with difficulty, too. While the classics are with us for all time, translations are usually written for a contemporary audience (which explains why classics have to be re-translated from time to time to speak to a modern audience). It might be useful for a university student to read Euripides or Molière in a literal translation, but it would be deadly for a regular theater audience. The translator who adapts the work according to the playwright’s intentions and beliefs and who achieves the same emotional impact is more faithful to the work than a literalist. As an example, consider Vladimir Nabokov. In 1922, after graduating from Cambridge, he produced a Russian translation of Lewis Carroll’s Alice in Wonderland, in large part on a dare from his father, and for his father’s enjoyment. Of the over 70 translations of Alice, it is considered one of the best. Full of life and fun, puns in English that would have been deadly translated literally into Russian were liberally adapted, and details in the English version that would have puzzled Russians were adapted for the Russian reader (e.g., Carroll’s “French mouse, come over with William the Conqueror” in the 1066 Norman invasion of England was translated in Russian as a mouse left behind during Napoleon’s retreat from Russia in 1812). Nabokov’s written translation from English to Russian calls to mind a story that is often told among UN interpreters: a Russian delegate in the 1950s quoted a line from Pushkin’s verse drama Boris Godunov about the Tsar’s guilt in spilling blood to ascend to the throne, and George Sherry, one of the most cultivated and experienced interpreters in the UN booths spontaneously substituted the famous “Will all great Neptune’s ocean wash this blood clean from my hands…” speech from Shakespeare’s MacBeth! (Frishberg 1986:51) The translation was certainly not a literal rendering of Pushkin, but an amazingly equivalent literary passage using the same metaphor from a verse drama in the English tradition: a true tour de force! By the 1950s, Nabokov, after living in the United States and seeing how “unfaithfully” some of his precious European classics had been translated into English, saw such liberties taken with an author’s text as criminal. His translations into English became much more literal. His former friend and collaborator Edmund Wilson, reviewing his translation of Eugene Onegin in The New York Review of Books in 1965, wrote: “One knows Mr. Nabokov’s virtuosity in juggling with the English language, the prettiness and wit of his verbal inventions … and one suspects that … he seeks to torture both the reader and himself by flattening Pushkin out and denying to his own powers the scope for their full play … passages sound like the products of those computers which are supposed to translate Russian into English … when he tries to translate Onegin ‘literally,’ what he writes is not always really English.” quoted in The New Republic, January 17, 2005, by Wyatt Mason, “Swann’s Ways: Adventures in Literary Translations” The young man who translated Alice for the pleasure of his reader had evolved into an author who translated to bring justice to the writer. This balance for the translator between fidelity to the author and fidelity to the reader is as delicate as the balance for the interpreter between the intent of the speaker and the needs of the listener. Wilson’s allusion to the results of computer translation, or speech-to-speech machine translation using speech recognition techniques and artificial intelligence, experimented with since the 1990s in the US, Japan, and Europe, reminds us of the value of such work for interpreters. Suzanne Jekat (1996) and Birte Schmitz (1998) elucidate the weaknesses of literal computer-generated speech-to-speech translations and argue for an ‘intended interpretation’ incorporating the context and the intentions of the speakers. Computer generated translations may be useful for technical information or simple tourist questions, but machines have a long way to go before they can render interpretations faithful to the speaker’s intent. 9. Measuring fidelity In the 1980s, after three decades of Paris school domination in the field of conference interpreting, practitioner-researchers like Jennifer Mackintosh, Barbara Moser-Mercer, Catherine Stenzl, and Daniel Gile began calling for a more science-based approach to studying interpreting, including more empirical research in the cognitive sciences (psychology and the psycho-, neuro-, and socio- branches of linguistics). By the time of the landmark 1986 Trieste Symposium on conference interpreting, organized by the Scuola Superiore di Lingue Moderne per Interpreti e Tradittori of the University of Trieste, this school of research was dubbed the “Trieste school” and it ushered in a new era of research in conference interpreting, including research on fidelity. Measuring lexical accuracy scientifically is problematic since word-for-word correspondences are rarely exact from language to language. Mackintosh (1983) created a scoring system to measure “semantic equivalence” on the clause or phrase level, with judges giving points for each unit of information reiterated in the target output (the scores in her study averaged from 70% to 90% “accurate”). Other researchers have attempted to measure fidelity based on propositional analysis, the segmentation of discourse into “predicates” (or heads) and “arguments” (adapted for sign language interpreters by Strong and Rudser, 1985; also Lambert, 1989). However, these approaches are strictly linguistic and do not take into account pragmatic concerns like intentionality, interactivity, and the overall success of the exchange in the minds of the participants. Comparison of information from source to target language has resulted in research on identifying interpreting “errors” or “miscues”, typically categorized as omissions, additions, corrections, or substitutions (Barik 1972, 1975/2002, Gerver 1969/2002). Dennis Cokely (1992), studying ASL interpreters, also includes intrusions from the source language which sound or look awkward in the target language. Cokely adds the notion of “reparable” and “irreparable” errors, i.e. those errors that the audience can mentally “repair” when they hear or see the interpretation, and those errors which pass unperceived and are mistakenly assumed to be the message from the speaker. In her study of Swedish/Russian interpreters during police interviews, Wadensjö (1998) viewed certain omissions as deliberate choices on the part of the interpreters. She coins the term reduced renditions to describe interpretations with “less explicitly expressed information than the preceding ‘original’ utterance” and she argues that reduced renditions are sometimes necessary to meet the communicative goals of the interaction. Elsewhere, reduced renditions have been called “condensing strategies” (Sunnari 1995) or “selective reductions” (Hatim and Mason 1990). Jemina Napier (2002) proposes that some omissions are actually coping strategies (c.f. Chernov 1978) employed by the interpreter committed to a successful communication, as, for example, when the omitted information is redundant, irrelevant, or would lead to confusion or misunderstanding. These “strategic intentional omissions” (among her proposed taxonomy of five types of omissions) she argues, like Wadensjö, are not “errors” at all: purposeful omissions contribute to the success of the communication. Thus again, being faithful to the message is far more than linguistic fidelity: it necessarily includes conforming to the participants’ goals for the interaction. 10. “Acting” or taking on the role of the speaker One possible framework for interpreter fidelity is to conceive of the interpreter as assuming the persona of the speaker. Like an actor, the interpreter takes on the role of the speaker and expresses his or her utterances in the first person, instead of adding “he says” or “she says.” A common perception among users of interpreting services is that the use of first person (“reported speech” or direct quoting) is a significant sign of professionalism (Knapp-Potthoff and Knapp 1986, Harris 1990). Reported speech can take 4 forms (in increasing order of involvement): she did this-or-that; she says that she did this-or-that; she says, “I did this-or-that”; and (taking on the role of the doer) “I did this-or-that.” The interpreter can use any of these 4 levels, even alternating between them, depending on the goals of the interaction and the needs of each participant (e.g., using the first person with the participant who is accustomed to communicating through interpreters, and adding “she says” for the participant who has little or no experience with interpreters). Individual interpreters have different styles of indicating that they are “not the authors of the words that they interpret.” (Bot 2005) Hanneke Bot finds in her study that the skilled interpreters who deviated from the ideal first person form usually follow with the direct quote: “she says, ‘I did this-or-that’” rather than resorting to indirect quotation (“she says that she did this-or-that”). This practice serves to emphasize the role of the interpreter as “reporter.” While the simple “reporting” of someone else’s speech may be seen in most Western countries as the original speaker’s words at a remove, Deborah Tannen (1989) indicates that other cultures may react differently to an utterance conveyed by a third party. She writes that “any anger or hurt felt in response to reported criticism is, for Americans, at least, typically directed toward the quoted source rather than the speaker who conveys the criticism. In contrast, according to an Arab proverb, “the one who repeats an insult is the one who is insulting you.” (p.105-6) Thus, the interpreter reporting speech who is seen as merely a “conduit” in American culture may be perceived as an involved “messenger” in other cultures. While generalizations like Tannen’s contain some truth, I would argue that there are instances in all cultures where reported criticism may be interpreted as a personal insult on the part of the reporter, depending on the situation and the intonation of the speaker, and this can have a significant impact on interpreting strategies. I remember an instance when I was interpreting for a meeting between a school headmaster and a deaf student in danger of expulsion. The headmaster was lecturing the student as if he were the boy’s father, and I was interpreting the diatribe in the first person as if I were the headmaster, trying, as I saw it, to project the full emotional import of what was being said. The student (who was on psychotropic medicine at the time, a fact which I did not know) looked directly at me and signed angrily, “Don’t you dare sign that way to me!” I immediately began adding the signs: “he said” in order to mitigate his anger. An interpreter’s close identification with a speaker, regardless of which culture the interpreting takes place in, can become an issue for either primary participant or for the interpreter (cf. Wallmach 2002 on the emotional toll on the interpreter in using the first person during interpreted interviews before the South African Truth and Reconciliation Committee). Interpreters who are trained to use the first person often find themselves needing to add the “s/he says” in order to emphasize that the words are not their own. This practice may be self-protection for the interpreter, but it also can address the need to emphasize that the goals of the interaction and the primary participants are more important than the presence of the intermediary. How far can one go in identifying with a speaker and taking on his/her role? Obviously, the interpreter has to be judicious when taking on the role of a participant. An adult interpreting a for a child, for example, probably would not use complex words or sentence structure, but certainly would rarely imitate a child’s voice in the interpreted version of the child’s utterances. It would sound ridiculous, as the other participants can see that the exchange is being interpreted by an adult. As an interpreter who often interprets for live theater performances in New York, I would hasten to add that the deaf audience in the theater expects polished literary interpretations in performance which at least approximate the interpretation of the actor on stage: the interpreter takes on the role of the character as fully as possible without distracting from the action onstage. In such a case, the interpreter might well consciously imitate the signing of a child, or interpret with a stooped back for an older character, etc. The emotional impact and style of signing is crucial in the interpretation if the goals of directors, actors, and audience are to be fulfilled. In addition to the use of “s/he says,” interpreters may also distance themselves from participants by utilizing an interpreter intonation style, which is an “I-am-just-interpreting-and-this-is-what-they-are-saying” kind of monotone. This “interpreter intonation” distancing may also, by the way, be the result of the interpreter’s being so busy processing content that intent and emotion are lost, resulting not only in a monotone, but also in non-standard pausing, anomalous stress, low-rise non-final pitch movement, awkward phrasing, and source language intrusions (Shlesinger 1994). This interpreting style, whether a result of task overload or purposeful non-involvement, serves as a continual reminder to the participants that the communication is interpreted and that the participants are responsible for the content. In effect, the interpreter is signaling: “You have to deal with this, not me. I am not involved.” The interpreter’s role is seldom rigid. While it may be important to distance oneself from the proceedings in one situation, it may be just as important to be fully involved in another, depending on the goals of the participants. The involvement or detachment of the interpreter may even change from minute to minute as the feelings of the participants evolve. Certainly interpreters whose goal is to remain utterly uninvolved, setting the use of first person as a strict rule and never speaking on their own behalf, risk alienating the consumer who expects a normal response to an ordinary greeting like “Hello, how are you?” directed to the interpreter. The refusal to respond as a normal person in that situation is apt to cause as many problems and confusions as the over-involvement of an intrusive interpreter in other situations. 11. Conclusion “The interpreter and the principal addresser function as a single entity…they exhibit behaviors of coordination and synchronicity.” (Danielle-Claude Bélanger 2004) “Although it may be possible for speakers [in an interpreted event] to feel that they are talking directly to each other, they are not. They are always exchanging speaking turns with the interpreter … in terms of the norms of their own language.” (Cynthia Roy 2000 p. 67, 72) 11.1 Interpreting is a partnership Rather than rendering isolated linguistic utterances, the successful interpreter partners with each participant in turn.  The interpreter is faithful to the goals of each speaker, hence, the interpreter is in effect an ally of both participants alternately, and therefore, in a very real sense, neutral.   A forthcoming case study from Australia of the cooperation between a deaf presenter and his interpreting team (Napier, Carmichael & Wiltshire, in press) will be a welcome addition to the empirical evidence of the importance of the interpreter's partnership with a speaker. The partnership between an interpreter and a participant begins with a time of preparation, which may be brief or extensive depending on the situation.  The interpreter who knows the participants and their backgrounds and relationship, who has been introduced to the jargon, acronyms, and the spelling of proper names which may be discussed, and who is aware of previous encounters or discussions between the participants, will have more contextual information to ensure that the information is conveyed accurately and faithfully. There are situations, especially very formal or structured events, when the participants can not or do not desire to “partner” with the interpreters. Any partnership, even a temporary one, requires mutual commitment.  When either of the partners chooses not to cooperate, then the interpreter can only process linguistic and cultural references to the extent possible without help and with only the clues at hand for guidance. Far from functioning like a conduit or telephone wire, in an ideal interpreting situation the interpreter is a co-constructor of meaning, actively aware that the goals and intentions of the interlocutors supersede the interpreter's own personal beliefs.  In co-constructing messages the interpreter not only refines his or her sensitivity to others' communication, but also increases the chances that the parties will truly understand each other. 11.2 A faithful interpretation depends on balance Radical swings between the extremes of either overly literal or unduly free translations do not lead the participants to confidence in the interpreting process.  Complete freedom (at the extreme, chaos) is as unacceptable as perfect order (at the extreme, totalitarian literalism).  Depending on each situation, the interpreter must consider the balance between literal and liberal, between accuracy and fluency, between encouraging autonomy and helping, between neutrality and advocacy, between managing and letting go.  This web of balances is not fixed.  It is constantly changing, depending on the dynamics and the goals of the primary participants and on the interpreted event itself. The definition of liberal as it applies to translation is “free, not strict or literal.”  Dictionary definitions of liberal also include:  “favoring or permitting freedom of action, especially with regard to freedom of expression or belief”; “freedom from prejudice or bigotry”; “tolerant, not bound by conventional ideas or values”; or “generous.”  These are ideals that can and should be embraced by interpreters who consider themselves partners of the consumers with whom they work, and who are committed to providing a truly faithful interpretation. Acknowledgements Many thanks to Natalie Atlas, a colleague who scrupulously edited the manuscript, to Cynthia Roy for her careful reading of the manuscript, and to Franz Pöchhacker, whose Introducing Interpreting Studies is a primary resource for any interpreter interested in interpreting research. As always, I am most indebted to Lillian Beard, who taught me respect and love for deaf people and their language. REFERENCES: Anderson, R. W. Bruce (1976) ‘Perspectives on the Role of Interpreter’, in Richard Brislin (ed.) Translation: Application and Research. New York: Gardner Press, 208-17. Baigorri-Jalón, Jesús (2005) ‘Interpreters at the United Nations: A History’, Interpreting 7(1): 141-146. Baker-Shenk, Charlotte (1991) ‘The Interpreter: Machine, Advocate, or Ally?’, in Jean Plant-Moeller (ed.) Expanding Horizons: Proceedings of the 1991 RID Convention, Silver Spring, MD: RID Publications, 120-140. Barik, H.C. (1972) ‘Interpreters Talk a Lot, Among Other Things’, Babel 18(1): 3-10. ------ (1975/2002) ‘Simultaneous Interpretation: Qualitative and Linguistic Data’, in Franz Pöchhacker and Miriam Shlesinger (eds) The Interpreting Studies Reader, London & New York: Routledge, 79-91. Barsky, Robert (1996) ‘The Interpreter as Intercultrual Agent in Convention Refugee Hearings’, The Translator 2(1): 45-63. Bélanger, Danielle (2004) ‘Interactional Patterns in Dialogue-Interpreting’, Journal of Interpretation, Silver Spring, MD: RID Publications, 1-18. Berk-Seligson, Susan (1990) The Bilingual Courtroom: Court Interpreters in the Judicial Process, Chicago & London: The University of Chicago Press. Bolden, Galinda (2000) ‘Toward Understanding Practices of Medical Interpreting: Interpreters’ Involvement in History Taking’, Discourse Studies 2(4): 387-419. Bot, Hanneke (2005) ‘Dialogue Interpreting as a Specific Case of Reported Speech’, Interpreting 7(2): 246-258. Bowen, Margareta (1995) ‘Interpreters and the Making of History’, in Jean Delisle and Judith Wordsworth (eds) Translators Through History, Amsterdam & Philadelphia: John Benjamins, 245-73. Caccamise, F. (1980) Introduction to Interpreting, Silver Spring, MD: Registry of Interpreters for the Deaf. Chernov, Ghelly (1978) Teoriya i praktika sinkhronnogo perevoda [Theory and Practice of Simultaneous Interpretation], Moscow: Mezhdunarodnyye otnosheniya. ------ (1994) ‘Message Redundancy and Message Anticipation in Simultaneous Interpretation’, in Sylvie Lambert and Barbara Moser-Mercer (eds) Bridging the Gap: Empirical Research in Simultaneous Interpretation, Amsterdam & Philadelphia: John Benjamins, 139-53. Cokely, Dennis (1992) Interpretation: A Sociolinguistic Model, Burtonsville, MD: Linstok Press. ------ (2005) ‘Shifting Positionality: A Critical Examination of the Turning Point in the Relationship of Interpreters and the Deaf Community’, in M. Marschark, R. Peterson, and E. Winston (eds) Sign Language Interpreting and Interpreter Education, London and New York: Oxford University Press, 3-28. Collados Aís, A. (1998/2002) ‘Quality Assessment in Simultaneous Interpreting: The Importance of Nonverbal Communication’, in Franz Pöchhacker and Miriam Shlesinger (eds) The Interpreting Studies Reader, London & New York: Routledge, 327-36. Dam, Helle V. (1993) ‘Text Condensing in Consecutive Interpreting’, in Yves Gambier and Jorma Tommola (eds) Translation and Knowledge, University of Turku, 297-313. Davidson, Brad (1998) Interpreting Medical Discourse: A Study of Cross-linguistic Communication in the Hospital Clinic, Unpublished Doctoral Dissertation, Stanford University. ------ (1998/2002) ‘A Model for the Construction of Conversational Common Ground in Interpreted Discourse’, Journal of Pragmatics 34(1): 273-300. Diriker, Ebru (2001) De-/Re-contextualising Simultaneous Interpreting: Interpreters in the Ivory Tower?, Unpublished Docgtoral Disseration, Istanbul: Bogaziçi University. Donk, U. (1994) ‘Der Dolmetscher in kriminalpolizeilichen Vernehmugen. Eine ethnographische Strukturrekonstruktion’, in N. Schröer (ed.) Interpretative Sozialforschung, Opladen: Westdeutscher Verlag, 130-50. Drennan, G. and L. Swartz (1999) ‘A Concept Over-burdened: Institutional Roles for Psychiatric Interpreters in Post-Apartheid South Africa’, Interpreting 4(2): 169-98. Forestal, Eileen (2005) ‘The Emerging Professional: Deaf Interpreters and Their Views and Experiences on Training’, in M. Marschark, R. Peterson and E.Winston (eds) Sign Language Interpreting and Interpreter Education, London & New York: Oxford University Press, 235-58. Edwards, R., B. Temple and C. Alexander (2005) ‘Users’ Experiences of Interpreters: the Critical Role of Trust’, Interpreting, 7(1): 77-96. Fant, Lou (1990) Silver Threads: A Personal Look at the First Twenty-five Years of the Registry of Interpreters for the Deaf, Silver Spring, MD: RID Publications. Fowler, Yvonne (1997). "The courtroom interpreter. Paragon and intruder?". The Critical Link: Interpreters in the Community. Proceedings of the First International Conference on Interpreting in Legal, Health and Social Service Settings. Orillia, Ontario, Canada. 1-4/4/95. 191-200. Amsterdam: John Benjamins Publishing Co. ------ (1997). "The courtroom interpreter. Paragon and intruder?". The Critical Link: Interpreters in the Community. Proceedings of the First International Conference on Interpreting in Legal, Health and Social Service Settings. Orillia, Ontario, Canada. 1-4/4/95. 191-200. Amsterdam: John Benjamins Publishing Co. Frishberg, Nancy (1986) Interpreting: An Introduction, Rockville, MD: RID Publications. Gerver, David (1969/2002) ‘The Effects of Source Language Presentation Rate on the Performance of Simultaneous Conference Interpreters’, in Franz Pöchhacker and Miriam Shlesinger (eds) The Interpreting Studies Reader, London & New York: Routledge, 53-66. Gile, Daniel (1992) ‘Basic Theoretical Components for Interpreter and Translator Training’, in Cay Dollerup and Anne Loddegaard (eds) Training Talent and Experience. Papers from the First Language International Conference, Elsinore, Denmark, 1991, Amsterdam & Philadelphia: John Benjamins, 185-94. Glémet, Roger. (1958) ‘Conference Interpreting Aspects of Translation’, in Studies in Communication 2. The Communication Research Centre, University College, London: Secker & Warburg, 105-122. Goffman, Erving (1981) Forms of Talk. Philadelphia: University of Pennsylvania Press. Gumperz, John (1982) Discourse Strategies, Cambridge: Cambridge University Press. Harris, Brian (1990) ‘Norms in Interpretation’, Target 2(1): 115-19. Hatim, Basil and Ian Mason (1997) The Translator as Communicator, London & New York: Routledge. Herbert, Jean (1952) The Interpreter’s Handbook: How to Become a Conference Interpreter, Geneva: Georg. Ingram, Robert (1978) ‘Sign Language Interpretation and General Theories of Language, Interpretation and Communication’, in David Gerver and H. Wallace Sinaiko (eds) Language Interpretation and Communication, New York & London: Plenum Press, 109-18. Isham, William (1994) ‘Memory for Sentence Form after Simultaneous Interpretation: Evidence both For and Against Deverbalization’, in Sylvie Lambert and Barbara Moser-Mercer (eds) Bridging the Gap, Amsterdam: Johns Benjamins, 191-211. Ishikawa, L. (1999) ‘Strategies for Interpreting a Japanese Complex Sentence into English – A Linguistic Study’, Interpreting Research 8(2): 74-95. Jaket, Suzanne and A. Klein (1996) ‘Machine Interpretation: Open Problems and Some Solutions’, Interpreting 1(1): 7-20. Jones, Roderick (1998) Conference Interpreting Explained, Manchester: St Jerome Publishing. Kadric, Mira (2001) Dolmetschen bei Gericht. Erwartungen, Anforderungen, Kompetenzen, Vienna: WUV-Universitätsverlag. Kelly, Arlene (2000) “Cultural Parameters for Interpreters in the Courtroom”, in The Critical Link 2: Interpreters in the Community, Selected papers from the Second International Conference on Interpreting in legal, health and social service settings, Vancouver, BC, Canada, 19–23 May 1998, Roda P. Roberts, Silvana E. Carr, Diana Abraham and Aideen Dufour (eds), 131-48. Amsterdam: John Benjamins Publishing Co. Kirchhoff, H. (1976/2002) ‘Simultaneous Interpreting: Interdependence of Variables in the Interpreting Process, Interpreting Models and Interpreting Strategies’, in Franz Pöchhacker and Miriam Shlesinger (eds) The Interpreting Studies Reader, London & New York: Routledge, 111-19. Knapp-Potthoff, A. and K. Knapp (1986) ‘Interweaving Two Discourses – The Difficult Task of the Non-professional Interpreter’, in Juliane House and Shoshana Blum-Kulka (eds) Interlingual and Intercultural Communication, Tübingen: Gunter Narr, 151-68. Kondo, Masaomi and Helen Tebble (1997) ‘Intercultural Communication, Negotiation, and Interpreting’, in Yves Gambier, Daniel Gile and Christopher Taylor (eds) Conference Interpreting: Current Trends in Research, Proceedings of the International Conference on Interpreting, Turku, 1994, Amsterdam: Johns Benjamins, 149-66. Kopczynski, Andrzej (1994) ‘Quality in Conference Interpreting: Some Pragmatic Problems’, in Mary Snell-Horby, Franz Pöchhacker and Klaus Kaindle (eds) Translation Studies: An Interdiscipline, Amsterdam: John Benjamins, 189-98. Kurz, Ingrid (1985) ‘The Rock Tombs of the Princes Elephantine. Earliest References to Interpretation in Pharaonic Egypt’, Babel 31(4): 213-18. ------ (1989) ‘Conference Interpreting: User Expectations’, in D.L. Hammond (ed.) Coming of Age. Proceedings of the 30th Annual Conference of the American Translator Association, Medford NJ: Learned Information Inc., 143-148. ------ (1993/2002) ‘Conference Interpretation: Expectations of Different User Groups’, in Franz Pöchhacker and Miriam Shlesinger (eds) The Interpreting Studies Reader, London & New York: Routledge, 313-24. ------ and Franz Pöchhacker (1995) ‘Quality in TV Interpreting’, Translation. Nouvelles de la FIT-FIT Newsletter 14(3/4): 350-58. Lambert, Sylvie (1989) ‘Information Processing among Conference Interpreters: A Test of the Depth-of-Processing Hypothesis’, in Laura Gran and John Dodds (eds) The Theoretical and Practical Aspects of Conference Interpretation, Udine: Campanotto Editore, 83-91. Laster, K. and V. Taylor (1994) Interpreters and the Legal System, Leichhardt, NSW: The Federation Press. Llewellyn-Jones, Peter (1981) ‘Simultaneous Interpreting’, in Bencie Woll, Jim G. Kyle and Margaret Deuchar (eds) Perspectives on British Sign Language and Deafness, London: Croom Helm, 89-104. Mackintosh, Jennifer (1983) Relay Interpretation: An Exploratory Study, Unpublished MA Dissertation, London: Birkbeck College, University of London. Marschark, M. and P.E. Spencer (eds) (2003) Oxford Handbook of Deaf Studies, Language, and Education, New York: Oxford University Press. Marschark, M., P. Sapere, C. Convertino and R. Seewagen (2005) ‘Educational Interpreting: Access and Outcomes’, in M. Marschark, R. Peterson and E. Winston (eds) Sign Language Interpreting and Interpreter Education, London & New York: Oxford University Press, 57-83. McIntyre, Marina and Gary Sanderson (1995) ‘Bye-bye! Bi-bi!: Questions of Empowerment and Role’, in A confluence of Diverse Relationships. Proceedings of the 13th National Convention of the Registry of Interpreters for the Deaf, 1993, Silver Spring, MD: RID Publications, 94-118. Mesa, Anne-Marie (2000) ‘The Cultural Interpreter: An Appreciated Professional’, in Roda P. Roberts, Silvana E. Carr, Diana Abraham and Aideen Dufour (eds) The Critical Link 2: Interpreters in the Community: Selected Papers from the Second International Conference on Interpreting in Legal, Health and Social Service Settings, Vancouver, BC, Canada, 19-23 May 1998. Amsterdam & Philadelphia: John Benjamins, 67-79. Metzger, Melanie (1995) The Paradox of Neutrality: A Comparison of Interpreters’ Goals with the Realities of Interactive Discourse, Unpublished Doctoral Dissertation, Washington, DC.: Georgetown University. ------ (1999) Sign Language Interpreting: Deconstructing the Myth of Neutrality, Washington, DC: Gallaudet University Press. Mikkelson, Holly (1998) ‘Towards a Redefinition of the Role of the Court Interpreter’, Interpreting 3(1): 22-46. ------ (1999) ‘The Professionalization of Community Interpreting’, Journal of Interpretation, Silver Spring, MD: RID Publications, 119-133. Morris, Ruth (1989) ‘Court Interpretation: The Trial of Ivan John Demjanjuk’, The Interpreters’ Newsletter 2: 27-37. ------ (1995) ‘The Moral Dilemmas of Court Interpreting’, The Translator 1(1): 25-46. Moser, Peter (1996) ‘Expectations of Users of Conference Interpretation’, Interpreting 1(2): 145-78. Napier, Jemina (2001) Linguistic Coping Strategies of Sign Language Interpreters, Unpublished Doctoral Dissertation, Macquarie University. ------ (2005) ‘Linguistic Features and Strategies of Interpreting: From Research to Education and Practice’, in M. Marschark, R. Peterson and E. Winston (eds) Sign Language Interpreting and Interpreter Education, London & New York: Oxford University Press, 84-111. ------ (2006) ‘“I like it/him/her because…”: An Exploration of Practitioner and Consumer Attitudes towards Signed Language Interpreting in the Community’. Paper presented to Supporting Deaf People Online Conference, 6-10 November 2006. ------ and Meg Rohan (forthcoming) ‘An Invitation to Dance: Deaf Consumer Perceptions of Sign Language Interpreters and Interpreting’, in Melanie Metzger (ed.) Research on interpretation, Vol. 3, Washington, DC: Gallaudet University Press. ------, Carmichael, A., & Wiltshire, A. (in press). ‘Look-Pause-Nod: A linguistic case study of a Deaf professional and interpreters working together’. In P. C. Hauser, K. L. Finch, & A. B. Hauser (eds), Deaf professionals and designated interpreters: A new paradigm. Washington, DC: Gallaudet University Press. Nida, Eugene A. and Charles R.Taber (1969) The Theory and Practice of Translation, Leiden: Brill. Pöchhacker, Franz (2000) ‘The Community Interpreter’s Task: Self-perception and Provider Views’, in Roda P. Roberts, Silvana E. Carr, Diana Abraham and Aideen Dufour (eds) The Critical Link 2: Interpreters in the Community: Selected Papers from the Second International Conference on Interpreting in Legal, Health and Social Service Settings, Vancouver, BC, Canada, 19-23 May 1998, Amsterdam & Philadelphia: John Benjamins, 49-65. ------ (2004) Introducing Interpreting Studies, New York: Routledge. Roberts, Roda P. (1997) ‘Community Interpreting Today and Tomorrow’, in Silvana E. Carr, Roda Roberts, Aideen Dufour and Dini Steyn (eds) The Critical Link: Interpreters in the Community, Amsterdam & Philadelphia, John Benjamins, 7-26. Roy, Cynthia (2000) Interpreting as a Discourse Process, Oxford: Oxford University Press. Sacks, H., E. Schegloff and G. Jefferson (1974) ‘A Simplist Systematics for the Organization of Turn-taking in Conversation’, Language 50(4): 696-735. Scheffer, T. (1997) ‘Dolmetschen als Darstellungsproblem. Eine ethnographische Studie zur Rolle der Dolmetscher in Asylanhörungen’, Zeitschrift für Soziologie 26(3): 159-80. Schmitz, Birte (1998) Pragmatikbasiertes Maschinelles Dolmetschen, Heidelberg: Groos. Seleskovitch, Danica (1968) L’interprète dans les conférences internationales: problèmes de langage et de communication, Paris: Minard Lettres Modernes. ------ (1978) Interpreting for International Conferences, Washington, DC: Pen and Booth. ------ and Marianne Lederer (1995) A Systematic Approach to Teaching Interpretation (trans. J. Harmer), Silver Spring, MD: RID Publications. Setton, Robin (1999) Simultaneous Interpretation: A Cognitive-pragmatic Analysis, Amsterdam & Philadelphia: John Benjamins. Shlesinger, Miriam (1989) ‘Extending the Theory of Translation to Interpretation: Norms as a Case in Point’, Target 1(1): 111-16. ------ (1991/1994) ‘Intonation in the Production and Perception of Simultaneous Interpretation’, in Sylvie Lambert and Barbara Moser-Mercer (eds) Bridging the Gap: Empirical Research in Simultaneous Interpretation, Amsterdam: John Benjamins, 225-36. ------ (1994) ‘Interpreter Latitude vs. Due Process. Simultaneous and Consecutive Interpretation in Multilingual Trials’, in Sonja Tirkkonen-Condit (ed.) Empirical Research in Translation and Intercultural Studies, Tübingen: Gunter Narr, 147-55. Smith, T. and E. Savidge (2002) ‘Beyond Knowledge and Skills: Teaching Attitude’, in L. Swabey (ed.) New Designs in Interpreter Education: Proceedings of the 14th National Convention of Conference of Interpreter Trainers, Minneapolis, Minnesota: CIT Publications, 17-32. Stenzl, C. (1989) ‘From Theory to Practice and from Practice to Theory’, in Laura Gran and John Dodds (eds) The Theoretical and Practical Aspects of Conference Interpretation, Udine: Campanotto Editore, 23-26. Strong, M. and S. Rudser (1985) ‘An Assessment Instrument for Sign Language Interpreters’, Sign Language Studies 49: 344-62. Sunnari, Marianna (1995) ‘Processing Strategies in Simultaneous Interpreting: ‘Saying It All’ vs. Synthesis’, in Jorma Tommola (ed.) Topics in Interpreting Research, University of Turku, 109-19. Tannen, Deborah (1984) Conversational Style: Analyzing Talk among Friends, Norwood, NJ: Ablex. ------ (1989) Talking Voices, Repetition, Dialogue and Imagery in Conversational Discourse, Cambridge: Cambridge University Press. Vermeer, Hans J. (1989/2000) ‘Skopos and Commission in Translational Action’, in Lawrence Venuti (ed.) The Translation Studies Reader, London & New York: Routledge, 221-32. ------ (1994) ‘Translation Today: Old and New Problems’, in Mary Snell-Hornby, Franz Pochhacker and Klaus Kaindl (eds) Translation Studies: An Interdiscipline, Amsterdam & Philadelphia: John Benjamins, 3-16. Viaggio, Sergio (1991) ‘Teaching Beginners to Shut Up and Listen’, The Interpreters’ Newsletter 4: 45-58. Wadensjö, Cecilia (1992) Interpreting as Interaction: On Dialogue-Interpreting in Immigration Hearings and Medical Encounters. Unpublished Doctoral Dissertation, Linköping, Sweden: Linköping University. ------ (1998) Interpreting as Interaction, New York: Longman. Wallmach, Kim (2002) ‘“Seizing the Surge of Language by Its Soft Bare Skull”: Simultaneous Interpreting, the Truth Commission and Country of My Skull’, Current Writing (special issue on Translation and Power) 14(2): 64-82. Wilson, M. (2007) takebackdeafworld.com. Witter-Merithew, Anna and Leilani Johnson (2005) Toward Competent Practice: Conversations with Stakeholders, Alexandria, Virginia: Registry of Interpreters for the Deaf. PAGE 30