I was pleased to read this article by Philipp Angermeyer, whose overall moral—that ‘translation practices’ deserve more attention from sociolinguistics—could be accepted almost without argument. Whether such practices ‘further social justice’ is a different, inherently political matter, and to consider how such a question even arises, and what our response should be, is a welcome addition from Angermeyer to our shared discussions.
I appreciate Angermeyer's juxtaposition of what at first seem quite different phenomena. The first are bureaucratic contexts where translation and interpretation are officially mandated, although, in Angermeyer's analysis, rarely to the equal advantage of all participants. The second are instances of public signage which exemplify what he felicitously dubs ‘punitive multilingualism’: situations where deploying marked languages appears, indexically, to stigmatize the users of some of the languages involved. I reverse his order of presentation, because the latter phenomenon, while it needs amplification, also—as Angermeyer's analysis demonstrates—draws attention to wider issues that evoke but also transcend ‘translation’, or as Angermeyer's rubric has it ‘the parallel availability of multiple languages’.
Angermeyer notes that Hill's (Reference Hill1998) notion of double indexicality in some uses of ‘Mock Spanish’ also exemplifies his notion of ‘punitive multilingualism’, insofar as public signs in some parts of the United States may be written in a Spanish that betrays an apparently wanton indifference to the grammar of the language and, despite appearing to acknowledge them, can simultaneously communicate a disrespect for both Spanish and its speakers. Thus, a sign in a public restroom that says Lava sus manos indexes both the fact that a Spanish-speaking audience is targeted and that the authors of the sign have, perhaps perversely, calqued an English admonition into butchered Spanish. That is, the sign acknowledges, in a left-handed way, that Spanish is likely to be spoken where the sign appears, but simultaneously ‘punishes’ the sign's recipients (by reminding them, and perhaps just them, ungrammatically, to wash their hands).
The matter can, of course, be more complex. Figure 1, for example, is a public sign from the Naples train station that presents a prohibition whose main formulation—its presumed ‘referential content’—is meant to require no ‘translation’ at all. That is, the ‘message’ of the sign—the picture—assumes no specific language. Therefore, the accompanying multilingual textual parts are presumably designed to accomplish something more than, in this case, just issuing a prohibition on smoking.
Figure 1. Sign from Naples train station (photo: John B. Haviland, July 2008).
This item from the linguistic landscape, of course, presents defining features of the most famous of Peircean trichotomies as applied to semiotic ‘signs’. The non-textual part of it utilizes a stylized iconicity to depict the smoking cigarette; it depends on an international symbolic convention (the red circle with a diagonal slash) to express prohibition. And its very location on a wall in the train station indicates that here is the place that one cannot smoke.Footnote 1 Now consider the different kinds of accompanying text. The prohibition is spelled out in words, in the appropriate un-calqued registers of only standard Italian and English. Although speakers of dozens of different languages presumably visit Naples every day, only English has been chosen to address those who can supposedly interpret neither the diagram nor the local Italian, indexing a particular set of presuppositions by the authors about their audience. The remaining small print is only in Italian, but for those who can recognize at least the numbers in the text, there seems again an indexical suggestion that, among other things, laws, police, and fines are involved. These ‘non-natural’ signs are thus necessarily authored, targeted, emitted, and received in differentiated ways by a range of participants.
Consider a further language-free street sign (Figure 2) from Naples, where the Camorra (the Naples Mafia) hold considerable sway. This sign is ostensibly meant to prohibit various sorts of vehicular traffic from the narrow streets of the center of town, but the apparent bullet hole indexes what one assumes is at least one sort of demonstrative addressee response,Footnote 2 as is the simultaneous presence of many bicycles, trucks, and especially the ubiquitous motorcycles on the street itself. The indexicality of these signs, whether multilingual or not, patently invites inferences about the authors, their intended addresses, and the mutual relationships between them.
Figure 2. Street sign from Naples old town (photo: John B. Haviland, July 2008).
Angermeyer's neologism, of course, also opens the possibility of various flip sides—or perhaps an entire cline—to elaborate the ‘punitive’ character of public multilingual communications: one variety might be optimistically called ‘celebratory multilingualism’, for example. In addition to everyday institutionalized multilingualism, there are also other contexts for multilingual ‘signage’ which further elaborate our understanding of sociolinguistic indexicality. Consider, for example, the ‘acknowledgements’ now obligatory in some parts of the world—for example in Australia or California—where First Nation peoples’ proprietary interests in place and land prescribe linguistic (if largely token) nods in the direction of local languages, often taking the form of originary language prefaces to public talks, university lectures, or even syllabi (Moody Reference Moody2022), although all too often followed by an immediate lapse back into a dominant language.
Street signs—even those with very limited referential content—can also modulate the flavor of such putative ‘celebrations’. In northeastern Italy toponyms in local languages are publicly acknowledged on street signs. Routinely the topmost name on a sign is written in standard Italian and the second in a local language, where the choice of the local language indexes quite diverse historical linguistic hegemonies. Jordan (Reference Jordan2022) exhibits an Italian signpost not far from the Austrian border that groups together bilingual signs pointing to three nearby places. The first is a town (Sauris/Zahre), where the written name in the local archaic German dialect follows the Italian place name. In the two other towns, (Lateis/Latais and Ampezzo/Dimpeҁ) the second name is written in Friulano, the regional Rhaeto Romance language. Different audiences may read, in these signs, celebration (or at least official recognition) of local tradition, an intertwined history of linguistic oppression and conquest, or simply bureaucratic lip-service to minoritized languages and their speakers.
The authorship (and authority) of such signs can also be indexically contested. Consider the isolated commune of Resia, in the province of Udine, Italy and not far from the multiple toponymic sign just mentioned. A centuries-long history of isolation has allowed the community to preserve its highly endangered Balto-Slavic Slovene dialect called Resian.Footnote 3 Figure 3 shows the sign at the entrance to one of its villages, citing the name first in Italian and then in Resian. It is evident, however, that the sign has been defaced, bearing (and baring) orthographic disagreement, by at least some Resians, about appropriate ways to write the language.Footnote 4 Thus legitimate authorship of such ‘translation practices’ can be questioned and revised. We can therefore add several flavors of ‘ambivalent multilingualism’—even in such public signage—to round out Angermeyer's catalogue.Footnote 5 Indeed, his coinage suggests an entire arena of detailed ethnography devoted to linguistic landscapesFootnote 6 in a more ample sense, questions of authority, voicing, implied addressivity, and their connections, as Angermeyer himself notes, with ‘the nationalist language policies of liberal democracies’ (see Silverstein Reference Silverstein, Brenneis and Macaulay2018).
Figure 3. Street sign, northeastern Italy (photo: John B. Haviland, July 2022).
Angermeyer's article ends with these public signs but begins with translation in power-laden bureaucratic processes that exacerbate rather than reduce inequalities between speakers. He locates his arguments alongside a wide literature, to contrast Sapir's foundational stance that all languages are expressively equal to the work of authors as diverse as Blommaert, Briggs, Duranti, Eades, Gal, Hill, Rickford, Piller, and Silverstein, who remind us that language is a central nexus of inequality and oppression, especially in an increasingly monoglot world. Based on my own professional experience as an interpreter for speakers of Tzotzil (Mayan), an indigenous language from Mexico, I direct my remaining remarks to one specific variety of ‘translation practices’: legal interpreting. I focus on participation frameworks and interactional configurations, professionalization, and transduction.
Participation frameworks for interpreting
Angermeyer points out how participation frameworks are crucial and consequential for interpreting, but again the matter appears more complex than his essay suggests. Unlike the asynchronous production of parallel texts in translation, interpretation implies a minimally triadic and necessarily interactive configuration of face-to-face (or ear-to-ear) participants, each with their own identities and agendas (Haviland Reference Haviland, Haas and Shuman2019). In Goffman's (Reference Goffman1979) standard model of the ‘speaker’ end of things, as Angermeyer writes, ‘interpreters act primarily as animators and “authors”, at least analytically, for minimally two other “principals”’ (p. 841). However, due to the ideology of ‘referential transparency’, which Angermeyer also mentions, the ‘author role is ideologically vitiated or dismissed—that is, rendered invisible by fiat.’ The triadic nature of the interaction also compounds the framework, since an interpreter establishes one sort of relationship with the source language speaker (something more than an “overhearer” but less than an “addressee”, at least in principle, and despite occasional lapses), and another (sometimes multiple) relationship with the target language speaker. Furthermore, most often both principals periodically switch places. As Angermeyer also notes, these participation roles are complicated by the inevitable asymmetries (in who can speak, when, to whom, and with what content) between the principals’ roles—examining attorney versus witness, for example—as revoiced (that is, transduced) by the interpreter.
The cast of characters in legal interpreting does not end here, of course. A courtroom may contain, both as sanctioned participants and overhearers, many co-present parties to an interpretation: judges, lawyers for multiple participants, witnesses, jurors, prosecutors, court reporters charged with rendering words into an official record, even other interpreters,Footnote 7 along with various publics. Worse still, legal interactions suffer from Gibbons’ (Reference Gibbons2003:174) ‘two audience problem’, by which he meant that, for example, a judge directly delivers instruction to a jury, but also indirectly but consciously to an invisible audience of potential appellate judges who may consequentially evaluate his or her precise words.Footnote 8 In all such situations, the details of an interpreter's language will index—and thus potentially have ‘creative’ effects in redefining—the nature of the interaction as a whole.
As Angermeyer discusses at length, there are established ‘modes of interpretation’—consecutive or simultaneous—that define the turn-structure of an interpreting episode, and the modes themselves enforce unequal opportunities for participation on the different parties, in what they have the opportunity to say, when, and whether or not they or others can respond. How these different modes are selected is supposedly a matter of acoustic access, but both modes depend on the theory that truth propositional information can be indiscriminately chunked into parts,Footnote 9 interpreted, and then re-assembled with no loss (or addition) of content. There are also different standards for when interpreting is deployed at all. In US immigration courts, for example, respondents have legal representation only if they find and pay for it themselves. A statutory requirement that the judge's final determination must be simultaneously interpreted in the respondent's best language can rarely be met without an interpreter physically present, usually employing chuchotage or audio headsets.Footnote 10 Judges can, however, ask a respondent's lawyer to waive such interpretation, just as an interpreter can be instructed to interpret some things but not others for the respondent's benefit, to save time, or when such simultaneous interpretation is impossible.
A court interpreter, after being ‘sworn in’, is supposed to recede into the background as a kind of non-person, no longer a true participant but rather an echoing device, transparent and institutionally effaced. Courts have different rules about when, and how, an interpreter can initiate any sort of non-interpreting interaction (complaining that someone is talking too fast, for example, or requesting clarification). By contrast, interpreters are routinely asked to help suppress ‘interruptions’ by telling recipients just ‘to listen and not respond’.Footnote 11
What Angermeyer describes as ‘maintaining person deixis’ (or in interpreter lingo ‘first person interpreting’) is a shibboleth for interpreter professionalism. (Don't get this wrong when doing your certification exam!) The convention can confound participants who are unfamiliar with it because an interpreter following the professional rules re-animates a participant's words without transposing the original pronouns, as Angermeyer's transcript (1) illustrates.Footnote 12 When a lawyer asks a witness “Did you hit her?” as the interpreter, following the rule, I must translate—without evidential or verba dicendi—the target equivalent, which in Tzotzil would be mi a-maj? (int 2e-hit), where both the perfective aspect and the third person object argument are realized silently.Footnote 13 As in Angermeyer's Polish example, this would be maximally confusing for a Tzotzil addressee, since elements of the original English question are elided (for example, the gender of the object referent, because Tzotzil does not mark gender in pronouns and thus does not distinguish her from him), and others insufficiently specified (for example, the fact that I—the interpreter—am not asking the question, but that someone else is.)Footnote 14 Making sense of the convention thus requires that the person for whom the interpretation is intended follow closely who is talking to whom, and what their respective roles are in the event—something that is taken for granted in most courts, but rarely if ever made explicit in interpreted utterances for the uninitiated. ‘Addressvity’ is always tricky in multiparty talk, of course, and the problem is exacerbated in circumstances like courts where people have different licenses to speak, roles, and rules known only to some and enforced by even fewer.
The limits of professionalization
Angermeyer points out that scholars often believe that problems with interpreting in legal contexts can be solved, in part, by better ‘professional’ interpreter training.Footnote 15 This last example suggests why I am dubious. To me, more important than the limited number of ‘professional’ interpreters available (and in fact their total absence for many languages for which they are urgently needed) is the nature of ‘professional training’ itself. What does it teach? What does it assume, given such desiderata for professional interpreters as ‘accuracy’, ‘impartiality’, and ‘effacement’? The problems of ‘accuracy’ in all translation are legion and unnecessary to rehearse here. Consider just Tzotzil ‘yes’ or ‘no’ or the lack of gendered third person pronouns as examples of radical incommensurability, where the target language fails to index categories obligatory in the source language; or vice versa, as when a target language requires indexing gender for first and second persons (in Slavic or Romance languages, for example)—an area in which English is deficient (pace Sapir) in a parallel way, and where as a result something must be added when grammatically interpreting from English.
Another feature of professional interpreter training is also theoretically problematic: the injunction to ‘maintain the (sociolinguistic) register’. Such a notion immediately raises the issue, mentioned by Angermeyer, of transducing incompatible indexical systems.Footnote 16 My colleague Rebecca Calderón—for many years the head of the interpreting section for a large Federal Court District—often tried to help me professionalize my own interpreting skills. She pointed out the special dilemmas posed by idioms. Once, when a lawyer told a witness, “You have to shit or get off the pot”, she struggled (but failed) on the fly to dredge up from her mind either an equivalent Spanish dicho or an appropriate paraphrase that ‘maintained the register’. A parallel problem derives from another convention—in US courts, at least—that the default tone of the courtroom be hyper-polite, reflected among other things in vocatives: ‘your honor’, ‘Mr. Smith’, ‘Miss Pérez’, ‘Dr. Haviland’, and the like. Effacement is not always possible when physical faces are co-present. The rules of courteous mutual address in Tzotzil are strict and use wholly different valences and linguistic conventions. Since I am an aging male, I cannot politely even start to address Tzotzil speakers without using their first names accompanied by appropriate honorifics, which usually means I have to ask a judge or lawyer to supply the first name, whose use would normally be inappropriate (or heavily limited) in a US court.Footnote 17 Other sorts of interpreter training more directly address transductional issues: trans-modal contexts (involving sign languages, or Angermeyer's written street signs, or situations when other sorts of non-verbal diagrams need to be explained); or contextual transduction involving specialized bureaucratic contexts with different rules and purposes: legal venues in criminal, civil, family law, or immigration courts, for example, or depositions and probation interviews; or insurance mediations, psychological evaluations, emergency room visits, medical discharges, even real-time medical proceduresFootnote 18—all situations where specialists are sometimes needed to interpret. I do not question the advantages of professional training designed for such situations, but it is neither readily available, nor, really, even feasible for most of the languages where it would be most desirable.
Professionalization, for interpreters, seems to me instead primarily a kind of institutionally favored indoctrination. It greases the wheels of bureaucracy and routinizes otherwise difficult decisions by supplying ready-made solutions, mostly with respect to comparative lexicology. Most professional interpreters in courts, for example, work in languages with their own developed legal vocabularies, a fallback jargon that assumes (and in fact tries to legislate) that all speakers remain on the same homogeneous terminological page.
Conclusion and remedies
Angermeyer argues that ‘translation is an important topic for sociolinguistic research that should not be left entirely to scholars in the field of translation studies’ (p. 853). I agree, although his confidence that ‘sociolinguistics’ is well enough placed to propose remedies for the problems he raises may be misplaced.
One of those problems is possibly amenable to some sort of sociolinguistic education: the fact that multilingualism is neither an exception, far less a defect, in human beings, but instead the normal, if not default condition of many. I myself, however, have been taken to task for chiding lawyers for their ‘bad theories of language’, on the grounds that complaints about flawed language ideologies and poor understanding by the law of the situations of multilingual participants in the legal system neglect what might be linguists’ own flawed theories or ‘ideologies of the law’ (Conley & O'Barr Reference Conley and O'Barr2005:155). There are good reasons, for example, for the law to insist on reducing multilingual babble to a fixed and privileged monolingual transcript (for example, to insure the possibilities of appellate review). Lawyers, in doing their best for their clients, also must decide whether or not even to employ interpreters; some Tzotzil speakers in US courts readily elect to use Spanish over Tzotzil, even if their Spanish is minimal, because it dramatically speeds up resolution of their cases. Moreover, the ‘goodwill and cooperation’ of interpreters is far from guaranteed, as is evident in poisoned courtrooms when, say, an indigenous language interpreter, for whatever reason, evinces hostility to other speakers of the language.Footnote 19
Let me end on a topic Angermeyer himself raises laterally to conclude his essay. What would ‘uptake in the wider field of sociolinguistics’ require or entail to address, for example, just one of the problems that Angermeyer has mentioned: ‘when individuals don't understand their rights as told by the police’? The problem arises routinely in the United States when suspected criminals are ‘Mirandized’ on arrest, and the issue has generated a large and influential body of international sociolinguistic work under the rubric of ‘police cautions’. One possible remedy, mentioned at the very end of Angermeyer's essay, is to ‘argue for a legal standard of “demonstrated understanding”’ (p. 854). But achieving such a ‘legal standard’, which was promulgated in Communication of Rights Group (2016), does not seem to be a job simply for sociolinguistics. This is all the more evident from the recent 2022 US Supreme Court decision (Vega vs. Tekoh Reference Vega v. Tekoh2022) which rules that a violation of the Miranda procedure (i.e. not being read one's rights to silence upon arrest) ‘does not necessarily constitute a violation of the Constitution’. Indeed, the direction of the court's reasoning seems to vitiate the need for such cautions at all, deeming them merely ‘prophylactic’. So perhaps most of the remedies required are not in the realm of sociolinguistics at all, but in our political institutions.
Angermeyer concludes by advocating ‘more emphasis on the study of comprehension and understanding, and [to] advocate for understanding between people, rather than translation between languages’ (p. 854). Again, I agree, although Green's (Reference Green2022) cautions about the necessary mutual moral positioning (albeit in a very different communicational context) involved in ‘being rendered intelligible or unintelligible’—part of the main business of interpreters—suggest that the very notion of ‘understanding’ stands in need of critical transdisciplinary, institutional, ethnographic, political, and ethical scrutiny.Footnote 20
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
You have
Access
Open access
I was pleased to read this article by Philipp Angermeyer, whose overall moral—that ‘translation practices’ deserve more attention from sociolinguistics—could be accepted almost without argument. Whether such practices ‘further social justice’ is a different, inherently political matter, and to consider how such a question even arises, and what our response should be, is a welcome addition from Angermeyer to our shared discussions.
I appreciate Angermeyer's juxtaposition of what at first seem quite different phenomena. The first are bureaucratic contexts where translation and interpretation are officially mandated, although, in Angermeyer's analysis, rarely to the equal advantage of all participants. The second are instances of public signage which exemplify what he felicitously dubs ‘punitive multilingualism’: situations where deploying marked languages appears, indexically, to stigmatize the users of some of the languages involved. I reverse his order of presentation, because the latter phenomenon, while it needs amplification, also—as Angermeyer's analysis demonstrates—draws attention to wider issues that evoke but also transcend ‘translation’, or as Angermeyer's rubric has it ‘the parallel availability of multiple languages’.
Angermeyer notes that Hill's (Reference Hill1998) notion of double indexicality in some uses of ‘Mock Spanish’ also exemplifies his notion of ‘punitive multilingualism’, insofar as public signs in some parts of the United States may be written in a Spanish that betrays an apparently wanton indifference to the grammar of the language and, despite appearing to acknowledge them, can simultaneously communicate a disrespect for both Spanish and its speakers. Thus, a sign in a public restroom that says Lava sus manos indexes both the fact that a Spanish-speaking audience is targeted and that the authors of the sign have, perhaps perversely, calqued an English admonition into butchered Spanish. That is, the sign acknowledges, in a left-handed way, that Spanish is likely to be spoken where the sign appears, but simultaneously ‘punishes’ the sign's recipients (by reminding them, and perhaps just them, ungrammatically, to wash their hands).
The matter can, of course, be more complex. Figure 1, for example, is a public sign from the Naples train station that presents a prohibition whose main formulation—its presumed ‘referential content’—is meant to require no ‘translation’ at all. That is, the ‘message’ of the sign—the picture—assumes no specific language. Therefore, the accompanying multilingual textual parts are presumably designed to accomplish something more than, in this case, just issuing a prohibition on smoking.
Figure 1. Sign from Naples train station (photo: John B. Haviland, July 2008).
This item from the linguistic landscape, of course, presents defining features of the most famous of Peircean trichotomies as applied to semiotic ‘signs’. The non-textual part of it utilizes a stylized iconicity to depict the smoking cigarette; it depends on an international symbolic convention (the red circle with a diagonal slash) to express prohibition. And its very location on a wall in the train station indicates that here is the place that one cannot smoke.Footnote 1 Now consider the different kinds of accompanying text. The prohibition is spelled out in words, in the appropriate un-calqued registers of only standard Italian and English. Although speakers of dozens of different languages presumably visit Naples every day, only English has been chosen to address those who can supposedly interpret neither the diagram nor the local Italian, indexing a particular set of presuppositions by the authors about their audience. The remaining small print is only in Italian, but for those who can recognize at least the numbers in the text, there seems again an indexical suggestion that, among other things, laws, police, and fines are involved. These ‘non-natural’ signs are thus necessarily authored, targeted, emitted, and received in differentiated ways by a range of participants.
Consider a further language-free street sign (Figure 2) from Naples, where the Camorra (the Naples Mafia) hold considerable sway. This sign is ostensibly meant to prohibit various sorts of vehicular traffic from the narrow streets of the center of town, but the apparent bullet hole indexes what one assumes is at least one sort of demonstrative addressee response,Footnote 2 as is the simultaneous presence of many bicycles, trucks, and especially the ubiquitous motorcycles on the street itself. The indexicality of these signs, whether multilingual or not, patently invites inferences about the authors, their intended addresses, and the mutual relationships between them.
Figure 2. Street sign from Naples old town (photo: John B. Haviland, July 2008).
Angermeyer's neologism, of course, also opens the possibility of various flip sides—or perhaps an entire cline—to elaborate the ‘punitive’ character of public multilingual communications: one variety might be optimistically called ‘celebratory multilingualism’, for example. In addition to everyday institutionalized multilingualism, there are also other contexts for multilingual ‘signage’ which further elaborate our understanding of sociolinguistic indexicality. Consider, for example, the ‘acknowledgements’ now obligatory in some parts of the world—for example in Australia or California—where First Nation peoples’ proprietary interests in place and land prescribe linguistic (if largely token) nods in the direction of local languages, often taking the form of originary language prefaces to public talks, university lectures, or even syllabi (Moody Reference Moody2022), although all too often followed by an immediate lapse back into a dominant language.
Street signs—even those with very limited referential content—can also modulate the flavor of such putative ‘celebrations’. In northeastern Italy toponyms in local languages are publicly acknowledged on street signs. Routinely the topmost name on a sign is written in standard Italian and the second in a local language, where the choice of the local language indexes quite diverse historical linguistic hegemonies. Jordan (Reference Jordan2022) exhibits an Italian signpost not far from the Austrian border that groups together bilingual signs pointing to three nearby places. The first is a town (Sauris/Zahre), where the written name in the local archaic German dialect follows the Italian place name. In the two other towns, (Lateis/Latais and Ampezzo/Dimpeҁ) the second name is written in Friulano, the regional Rhaeto Romance language. Different audiences may read, in these signs, celebration (or at least official recognition) of local tradition, an intertwined history of linguistic oppression and conquest, or simply bureaucratic lip-service to minoritized languages and their speakers.
The authorship (and authority) of such signs can also be indexically contested. Consider the isolated commune of Resia, in the province of Udine, Italy and not far from the multiple toponymic sign just mentioned. A centuries-long history of isolation has allowed the community to preserve its highly endangered Balto-Slavic Slovene dialect called Resian.Footnote 3 Figure 3 shows the sign at the entrance to one of its villages, citing the name first in Italian and then in Resian. It is evident, however, that the sign has been defaced, bearing (and baring) orthographic disagreement, by at least some Resians, about appropriate ways to write the language.Footnote 4 Thus legitimate authorship of such ‘translation practices’ can be questioned and revised. We can therefore add several flavors of ‘ambivalent multilingualism’—even in such public signage—to round out Angermeyer's catalogue.Footnote 5 Indeed, his coinage suggests an entire arena of detailed ethnography devoted to linguistic landscapesFootnote 6 in a more ample sense, questions of authority, voicing, implied addressivity, and their connections, as Angermeyer himself notes, with ‘the nationalist language policies of liberal democracies’ (see Silverstein Reference Silverstein, Brenneis and Macaulay2018).
Figure 3. Street sign, northeastern Italy (photo: John B. Haviland, July 2022).
Angermeyer's article ends with these public signs but begins with translation in power-laden bureaucratic processes that exacerbate rather than reduce inequalities between speakers. He locates his arguments alongside a wide literature, to contrast Sapir's foundational stance that all languages are expressively equal to the work of authors as diverse as Blommaert, Briggs, Duranti, Eades, Gal, Hill, Rickford, Piller, and Silverstein, who remind us that language is a central nexus of inequality and oppression, especially in an increasingly monoglot world. Based on my own professional experience as an interpreter for speakers of Tzotzil (Mayan), an indigenous language from Mexico, I direct my remaining remarks to one specific variety of ‘translation practices’: legal interpreting. I focus on participation frameworks and interactional configurations, professionalization, and transduction.
Participation frameworks for interpreting
Angermeyer points out how participation frameworks are crucial and consequential for interpreting, but again the matter appears more complex than his essay suggests. Unlike the asynchronous production of parallel texts in translation, interpretation implies a minimally triadic and necessarily interactive configuration of face-to-face (or ear-to-ear) participants, each with their own identities and agendas (Haviland Reference Haviland, Haas and Shuman2019). In Goffman's (Reference Goffman1979) standard model of the ‘speaker’ end of things, as Angermeyer writes, ‘interpreters act primarily as animators and “authors”, at least analytically, for minimally two other “principals”’ (p. 841). However, due to the ideology of ‘referential transparency’, which Angermeyer also mentions, the ‘author role is ideologically vitiated or dismissed—that is, rendered invisible by fiat.’ The triadic nature of the interaction also compounds the framework, since an interpreter establishes one sort of relationship with the source language speaker (something more than an “overhearer” but less than an “addressee”, at least in principle, and despite occasional lapses), and another (sometimes multiple) relationship with the target language speaker. Furthermore, most often both principals periodically switch places. As Angermeyer also notes, these participation roles are complicated by the inevitable asymmetries (in who can speak, when, to whom, and with what content) between the principals’ roles—examining attorney versus witness, for example—as revoiced (that is, transduced) by the interpreter.
The cast of characters in legal interpreting does not end here, of course. A courtroom may contain, both as sanctioned participants and overhearers, many co-present parties to an interpretation: judges, lawyers for multiple participants, witnesses, jurors, prosecutors, court reporters charged with rendering words into an official record, even other interpreters,Footnote 7 along with various publics. Worse still, legal interactions suffer from Gibbons’ (Reference Gibbons2003:174) ‘two audience problem’, by which he meant that, for example, a judge directly delivers instruction to a jury, but also indirectly but consciously to an invisible audience of potential appellate judges who may consequentially evaluate his or her precise words.Footnote 8 In all such situations, the details of an interpreter's language will index—and thus potentially have ‘creative’ effects in redefining—the nature of the interaction as a whole.
As Angermeyer discusses at length, there are established ‘modes of interpretation’—consecutive or simultaneous—that define the turn-structure of an interpreting episode, and the modes themselves enforce unequal opportunities for participation on the different parties, in what they have the opportunity to say, when, and whether or not they or others can respond. How these different modes are selected is supposedly a matter of acoustic access, but both modes depend on the theory that truth propositional information can be indiscriminately chunked into parts,Footnote 9 interpreted, and then re-assembled with no loss (or addition) of content. There are also different standards for when interpreting is deployed at all. In US immigration courts, for example, respondents have legal representation only if they find and pay for it themselves. A statutory requirement that the judge's final determination must be simultaneously interpreted in the respondent's best language can rarely be met without an interpreter physically present, usually employing chuchotage or audio headsets.Footnote 10 Judges can, however, ask a respondent's lawyer to waive such interpretation, just as an interpreter can be instructed to interpret some things but not others for the respondent's benefit, to save time, or when such simultaneous interpretation is impossible.
A court interpreter, after being ‘sworn in’, is supposed to recede into the background as a kind of non-person, no longer a true participant but rather an echoing device, transparent and institutionally effaced. Courts have different rules about when, and how, an interpreter can initiate any sort of non-interpreting interaction (complaining that someone is talking too fast, for example, or requesting clarification). By contrast, interpreters are routinely asked to help suppress ‘interruptions’ by telling recipients just ‘to listen and not respond’.Footnote 11
What Angermeyer describes as ‘maintaining person deixis’ (or in interpreter lingo ‘first person interpreting’) is a shibboleth for interpreter professionalism. (Don't get this wrong when doing your certification exam!) The convention can confound participants who are unfamiliar with it because an interpreter following the professional rules re-animates a participant's words without transposing the original pronouns, as Angermeyer's transcript (1) illustrates.Footnote 12 When a lawyer asks a witness “Did you hit her?” as the interpreter, following the rule, I must translate—without evidential or verba dicendi—the target equivalent, which in Tzotzil would be mi a-maj? (int 2e-hit), where both the perfective aspect and the third person object argument are realized silently.Footnote 13 As in Angermeyer's Polish example, this would be maximally confusing for a Tzotzil addressee, since elements of the original English question are elided (for example, the gender of the object referent, because Tzotzil does not mark gender in pronouns and thus does not distinguish her from him), and others insufficiently specified (for example, the fact that I—the interpreter—am not asking the question, but that someone else is.)Footnote 14 Making sense of the convention thus requires that the person for whom the interpretation is intended follow closely who is talking to whom, and what their respective roles are in the event—something that is taken for granted in most courts, but rarely if ever made explicit in interpreted utterances for the uninitiated. ‘Addressvity’ is always tricky in multiparty talk, of course, and the problem is exacerbated in circumstances like courts where people have different licenses to speak, roles, and rules known only to some and enforced by even fewer.
The limits of professionalization
Angermeyer points out that scholars often believe that problems with interpreting in legal contexts can be solved, in part, by better ‘professional’ interpreter training.Footnote 15 This last example suggests why I am dubious. To me, more important than the limited number of ‘professional’ interpreters available (and in fact their total absence for many languages for which they are urgently needed) is the nature of ‘professional training’ itself. What does it teach? What does it assume, given such desiderata for professional interpreters as ‘accuracy’, ‘impartiality’, and ‘effacement’? The problems of ‘accuracy’ in all translation are legion and unnecessary to rehearse here. Consider just Tzotzil ‘yes’ or ‘no’ or the lack of gendered third person pronouns as examples of radical incommensurability, where the target language fails to index categories obligatory in the source language; or vice versa, as when a target language requires indexing gender for first and second persons (in Slavic or Romance languages, for example)—an area in which English is deficient (pace Sapir) in a parallel way, and where as a result something must be added when grammatically interpreting from English.
Another feature of professional interpreter training is also theoretically problematic: the injunction to ‘maintain the (sociolinguistic) register’. Such a notion immediately raises the issue, mentioned by Angermeyer, of transducing incompatible indexical systems.Footnote 16 My colleague Rebecca Calderón—for many years the head of the interpreting section for a large Federal Court District—often tried to help me professionalize my own interpreting skills. She pointed out the special dilemmas posed by idioms. Once, when a lawyer told a witness, “You have to shit or get off the pot”, she struggled (but failed) on the fly to dredge up from her mind either an equivalent Spanish dicho or an appropriate paraphrase that ‘maintained the register’. A parallel problem derives from another convention—in US courts, at least—that the default tone of the courtroom be hyper-polite, reflected among other things in vocatives: ‘your honor’, ‘Mr. Smith’, ‘Miss Pérez’, ‘Dr. Haviland’, and the like. Effacement is not always possible when physical faces are co-present. The rules of courteous mutual address in Tzotzil are strict and use wholly different valences and linguistic conventions. Since I am an aging male, I cannot politely even start to address Tzotzil speakers without using their first names accompanied by appropriate honorifics, which usually means I have to ask a judge or lawyer to supply the first name, whose use would normally be inappropriate (or heavily limited) in a US court.Footnote 17 Other sorts of interpreter training more directly address transductional issues: trans-modal contexts (involving sign languages, or Angermeyer's written street signs, or situations when other sorts of non-verbal diagrams need to be explained); or contextual transduction involving specialized bureaucratic contexts with different rules and purposes: legal venues in criminal, civil, family law, or immigration courts, for example, or depositions and probation interviews; or insurance mediations, psychological evaluations, emergency room visits, medical discharges, even real-time medical proceduresFootnote 18—all situations where specialists are sometimes needed to interpret. I do not question the advantages of professional training designed for such situations, but it is neither readily available, nor, really, even feasible for most of the languages where it would be most desirable.
Professionalization, for interpreters, seems to me instead primarily a kind of institutionally favored indoctrination. It greases the wheels of bureaucracy and routinizes otherwise difficult decisions by supplying ready-made solutions, mostly with respect to comparative lexicology. Most professional interpreters in courts, for example, work in languages with their own developed legal vocabularies, a fallback jargon that assumes (and in fact tries to legislate) that all speakers remain on the same homogeneous terminological page.
Conclusion and remedies
Angermeyer argues that ‘translation is an important topic for sociolinguistic research that should not be left entirely to scholars in the field of translation studies’ (p. 853). I agree, although his confidence that ‘sociolinguistics’ is well enough placed to propose remedies for the problems he raises may be misplaced.
One of those problems is possibly amenable to some sort of sociolinguistic education: the fact that multilingualism is neither an exception, far less a defect, in human beings, but instead the normal, if not default condition of many. I myself, however, have been taken to task for chiding lawyers for their ‘bad theories of language’, on the grounds that complaints about flawed language ideologies and poor understanding by the law of the situations of multilingual participants in the legal system neglect what might be linguists’ own flawed theories or ‘ideologies of the law’ (Conley & O'Barr Reference Conley and O'Barr2005:155). There are good reasons, for example, for the law to insist on reducing multilingual babble to a fixed and privileged monolingual transcript (for example, to insure the possibilities of appellate review). Lawyers, in doing their best for their clients, also must decide whether or not even to employ interpreters; some Tzotzil speakers in US courts readily elect to use Spanish over Tzotzil, even if their Spanish is minimal, because it dramatically speeds up resolution of their cases. Moreover, the ‘goodwill and cooperation’ of interpreters is far from guaranteed, as is evident in poisoned courtrooms when, say, an indigenous language interpreter, for whatever reason, evinces hostility to other speakers of the language.Footnote 19
Let me end on a topic Angermeyer himself raises laterally to conclude his essay. What would ‘uptake in the wider field of sociolinguistics’ require or entail to address, for example, just one of the problems that Angermeyer has mentioned: ‘when individuals don't understand their rights as told by the police’? The problem arises routinely in the United States when suspected criminals are ‘Mirandized’ on arrest, and the issue has generated a large and influential body of international sociolinguistic work under the rubric of ‘police cautions’. One possible remedy, mentioned at the very end of Angermeyer's essay, is to ‘argue for a legal standard of “demonstrated understanding”’ (p. 854). But achieving such a ‘legal standard’, which was promulgated in Communication of Rights Group (2016), does not seem to be a job simply for sociolinguistics. This is all the more evident from the recent 2022 US Supreme Court decision (Vega vs. Tekoh Reference Vega v. Tekoh2022) which rules that a violation of the Miranda procedure (i.e. not being read one's rights to silence upon arrest) ‘does not necessarily constitute a violation of the Constitution’. Indeed, the direction of the court's reasoning seems to vitiate the need for such cautions at all, deeming them merely ‘prophylactic’. So perhaps most of the remedies required are not in the realm of sociolinguistics at all, but in our political institutions.
Angermeyer concludes by advocating ‘more emphasis on the study of comprehension and understanding, and [to] advocate for understanding between people, rather than translation between languages’ (p. 854). Again, I agree, although Green's (Reference Green2022) cautions about the necessary mutual moral positioning (albeit in a very different communicational context) involved in ‘being rendered intelligible or unintelligible’—part of the main business of interpreters—suggest that the very notion of ‘understanding’ stands in need of critical transdisciplinary, institutional, ethnographic, political, and ethical scrutiny.Footnote 20