Re-codified standards from the perspective of language experts: credentials, practice and attitudes amongst translators and interpreters of the Bosnian, Croatian and Serbian languages

Jim Hlavac

Re-codified Standards from the Perspectiveof Language Experts: Credentials, Practiceand Attitudes amongst Translators andInterpreters of the Bosnian, Croatian andSerbian Languages

Abstract: This article examines aspects of linguistic behaviour, attitudes andprofessional practices amongst a group of 47 “expert users” who are translatorsor interpreters for one, two or three of the following languages: Bosnian, Croatianand Serbian. The official terms for these languages in the respective successorstates of Socialist Federal Republic of Yugoslavia and elsewhere reflect not onlyhistoric but also popular designations. Incumbent on translators and interpretersare the professional regulatory norms that require practitioners to follow thesedesignations. An overview of the re-codification of the three languages is pro-vided, followed by a discussion of models that account for how speakers and text-producers negotiate verbal and written interactions with speakers or text ofdifferent language varieties. Data are elicited on the following: informants’reported behaviour in professional and non-professional situations; unanticipateddifferences in the language for which an assignment was accepted and its actualform; attitudes on assignments with unofficial or unclear language designations;others’ assumptions of informants’ native speaker competency and ethnicity; andattitudes towards the distinctness of the three languages. Accommodation toclients’ language varieties is reported by half of all informants, and those withmultiple accreditations report converging to others’ language varieties more sothan those with accreditation in one language only. Metalinguistic talk, with orwithout accommodation, is also a common practice in the initial stages of inter-preted interactions or the negotiation of translation assignments. The reportedbehaviour and practices of translators and interpreters are likely to be indicativeof “lay speakers’” and marketplace requirements, and therefore reflective of actuallanguage use amongst users of these three languages when interacting withone another.

Jim Hlavac, Department of Translation and Interpreting Studies, Monash University, Caulfield,VIC 3145, Australia, E-mail: [email protected]

Multilingua 2015; 34(1): 61–91

Keywords: translation norms, close-related languages, translators and inter-preters, Bosnian, Croatian, Serbian

DOI 10.1515/multi-2014-0201

1 Introduction

This article examines the credentials (or number of accreditations), personal andprofessional practices and attitudes of translators and interpreters of theBosnian, Croatian and Serbian languages. These three languages underwentre-codification in the 1990s at the same time that speakers of these languagesbecame residents of the newly independent states of Bosnia-Herzegovina,Croatia and Serbia after the break-up of the Socialist Federal Republic ofYugoslavia (hereafter: SFRY). One characteristic of the political system of theSFRY was a top-down policy of linguistic unitarism that prescribed the use of theterm “Serbo-Croatian” as the official designation of the first language of most ofthe inhabitants of the SFRY. Notwithstanding periods of transition (e.g. 1945–1954) and liberalisation (late 1960s to 1971), linguistic unitarism was very muchthe dominant language policy pursued by the political leadership of SFRY, andthe term Serbo-Croatian was one that had currency not due to popular senti-ment, but to its imposition as the official designation. The violent break-up ofSFRY in 1991 and armed conflict in Croatia (1991–1995) and Bosnia-Herzegovina(1992–1995) have had commensurate effects on many speakers of the threelanguages investigated in this article. Perhaps expectedly, during and afterthese conflicts a number of normative publications appeared that describedand codified the three languages as separate and distinct entities (e.g. Kačić1995; Halilović 1998; Radovanović 2000).

The attribute “Bosnian” had been widely used to refer to the native lan-guage of inhabitants of Bosnia-Herzegovina (including non-Muslims) since thetime of the Ottoman occupation, as well as being the term used by mostHabsburg and local administrators after Austro-Hungarian annexation in 1878.As a linguistic designation, Bosnian was consigned to the status of a regional ordialectal variety of Serbian in inter-war Yugoslavia and to the status of aregional variety of Croatian during the time of the Independent State ofCroatia (1941–1945). Further, the status of Bosnian as a recognisably distinctcode appeared, in formal terms, buried by the double double-barrelledterm “Serbocroatian-Croatoserbian” (see Table 1) in the Socialist Republic of

62 Jim Hlavac

Table1:

Previous

andcurren

tde

sign

ations

ofofficial

lang

uage

sin

four

successo

rstates

ofSFRY

Bos

nia-Herzego

vina

Croa

tia

Serbia

Mon

tene

gro

Officialpre-1991

design

ation

“Serbo

croa

tian

orCroatos

erbian

lang

uage

withIje

kavian

pron

unciation”

(Article

4of

the

Con

stitutionof

SRBos

nia-

Herzego

vina

1974

[Bug

arski&

Haw

kesw

orth

2004])

“Croatianliterarylang

uage

,the

stan

dard

form

ofthena

tion

allang

uage

ofCroatsan

dSerbs

inCroatia,also

know

nas

Croatianor

Serbian

”(Article

38of

theCon

stitutionof

SR

Croatia

1974

[Bug

arski&

Haw

kesw

orth

2004])

“Serbo

croa

tian

lang

uage

”(Article

10of

the

Con

stitutionof

the

Repu

blic

ofSerbia1990

[Bug

arski&

Haw

kesw

orth

2004])

“Serbo

croa

tian

lang

uage

with

Ijeka

vian

pron

unciation”

(Article

182of

the

Con

stitutionof

SR

Mon

tene

gro1974

[Bug

arski&

Haw

kesw

orth

2004])

Current

offic

ial

lang

uage

/s“B

osnian

”“C

roatian”

“Serbian

”“C

roatian”

“Serbian

”“M

ontene

grin”withthe

adde

ndum

:“S

erbian

,Bos

nian

,Alban

ianan

dCroatianalso

enjoyofficial

status

”.

Re-codified Standards and Language Experts 63

Bosnia-Herzegovina within SFRY. But the emergence of a Sarajevo-inspiredstandard in the 1970s (see Šipka et al. 1973), and the overwhelming percentageof Bosnian Muslims – 80% (Jurić-Kappel 2003: 95) – who, in spite of thedecades-long imposition of Serbo-Croatian and no official support for linguisticseparatism, declared their language to be Bosnian, indicates that Bosnian hadbeen dormant rather than dead. The distinctness of Croatian had already been awidely advocated position, both inside Croatia amongst linguists (Babić 1990)and amongst Croat émigrés (see Franolić 1980a). In regard to Serbian, there waslittle opposition amongst Serbs towards the term Serbo-Croatian, whose formand attributes reflected their own language tradition to a degree greater than forother groups. A tacit claim amongst some Serbian linguists (e.g. Ivić 1971) hasbeen that the Štokavian dialectal basis for all three languages is a uniquelySerbian one: “the Štokavian dialect [is] eminently the dialect of organicallySerbian origin” (Radovanović 1996: 182). In Serbia and in the Bosnian-Serbentity, reversion to the popular term “Serbian” as the official language designa-tion has been accompanied not only by some changes in corpus and languageplanning but also by a concerted campaign to portray Serbian linguistic heritageas superior to that of others and therefore vulnerable to appropriation. Examplesinclude, “Croats adopted the Serbian standard language to bolster their claimsto Bosnia-Herzegovina” (Brborić 1999: 341–342), and the claim by Kovačević(1999: 100) that the “Montenegrin language is supported by Croats and Muslimswith the aim of weakening the Serbian language” (all translations mine). There-codification of languages in the successor states of Bosnia-Herzegovina,Croatia and Serbia has received attention in various branches of linguistics:linguistic planning (Samardžija 1999), sociolinguistics (Tolimir-Hölzl 2009), dia-lectology (Kovačević 1998), as well as in translation studies (Dragovic-Drouet2007). In the main, these studies recognise and acknowledge the separateness ofthe three standard languages and, often with a comparative approach, trace theseparate paths that each language has been taking. Only a small number stilladvocate linguistic unitarism (e.g. Okuka 1998) and favour the term Serbo-Croatian, which has not been in official use in any of the successor statessince 1992 and which, outside these states, has almost fallen into disuse.

This article is a contribution to the small number of studies on intra- andinter-group language differences for these languages. It examines languageattitudes and reported practices from a group of informants who are “linguisticexperts”, congruent therefore to linguists and other philologists, but whoseoccupational practices also include regular and intense contact with “naïvespeakers”. (I define naïve speakers here as those speakers whose relationshipto language and responses to questions about language are primarily affectiverather than analytical.) At the same time, translating and interpreting (hereafter:

64 Jim Hlavac

T&I) practitioners do not usually have the research and time resources availableto linguists to engage in in-depth studies of language users. T&I practitioners arealso guided by and generally obliged to follow codified norms different to thoseof linguistics researchers who may enjoy different degrees of (academic) free-dom. Within the translation and interpreting sector, official designations andstate language policies have a regulating effect such that the official names ofthe languages in Bosnia-Herzegovina, Croatia and Serbia are used by T&I bodiesand practitioners in these countries, as well as accrediting or professionalorganisations in other countries where T&I services for these languages areprovided (e.g. ATA 2012; NAATI 2012; Universitas n.d.). (“Accrediting” and“accreditation” are Australia-based terms1 referring to formal certification ofability to work professionally. The term accreditation is used here as a hypernymto refer to formal recognition of interlingual transfer skills whether throughtesting or through a university degree or other formal qualification.)

This article is organised in the following way. Section 2 provides a descrip-tion of the features of T&I practitioners and why they constitute a group ofinformants that can provide insights into language use and attitudes as bothinsiders (i.e. language experts) and outsiders (i.e. practitioners who use lan-guage in a literally instrumental sense). Section 3 provides a brief overview ofthe previous and current designations of the languages of the Bosniaks, Croatsand Serbs, some aspects of now defunct Serbo-Croatian and a presentation ofthe historical standardisation and recent re-codification of the Bosnian, Croatianand Serbian languages.

In Section 4, the topic of different varieties that are indexical of locality oforigin and ethnicity is introduced with theoretical models that describe thefeatures which pertain in situations in which speakers of different varietiesmay or may not accommodate to each other’s variety. This section discussesaccommodation, markedness and audience design as models that account forminimisation and accentuation of differences in language varieties based on theascertainable features of other interlocutors or communication recipients.

Section 5 outlines the professional identity, protocols and interactionalnorms that regulate the T&I sector in a general sense, through international,

1 Accreditation is the term used in Australia to refer to recognition of a test candidate’sstandard of performance that entitles him or her to seek professional employment:

The National Accreditation Authority for Translators and Interpreters Ltd (NAATI) is thebody responsible for setting and monitoring the standards for the translating and interpret-ing profession in Australia. It does this through its system of accreditation. NAATI accred-itation is the only credential officially accepted by employers for the profession oftranslation and interpreting in Australia. (NAATI 2011)


national or local guidelines, and that pertain to T&I practitioners in their indivi-dual interactions with clients, colleagues and other service providers. This isfollowed by Section 6, the methodology, which describes how data were elicited,the components that make up the data sample for this article, general detailsabout the informants and how these are shown in the discussion of data.

Section 7 presents data from informants about their reported experiencesand behaviour, responses to how they would behave in hypothetical situations,responses in respect of their own attitudes and responses to others’ attitudes.The content of these data relates to informants’ self-reported speech innon-occupational (e.g. social) situations; their professional protocols whenconfronted with a language different from the one for which they acceptedan assignment; attitudes towards accepting assignments in languages withunofficial or unclear designations; others’ assumptions of their native speakercompetency and ethnicity; their own and others’ attitudes towards the distinct-ness of Bosnian, Croatian and Serbian as separate languages; and informants’attitudes towards the future development of the three languages.

The conclusion collates and contextualises the responses from the discus-sion section and summarises them as findings relevant to perceptions andattitudes towards contemporary Bosnian, Croatian and Serbian. The amenabilityof T&Is as providers of data and their dual role as linguistic experts and serviceproviders is also discussed.

2 Naïve speakers, expert language users andtranslators and interpreters

No occupational group is closer to that of the linguist (or philologist) than T&Is.In lay and even specialist descriptions, T&Is are often termed “linguists” (e.g.the UK professional organisation of T&Is is designated the Institute of Linguists).This article adopts a conventional definition of T&Is as those who engage in theinterlingual transfer of written texts and spoken or signed speech. As such, T&Isare not naïve speakers or language users. Historically, translation or interpretinghave been viewed in Western countries as activities that do not constitute asingle, discrete profession, but that may be commonly performed by philolo-gists, writers, businesspeople, diplomats or merchants, to name but a few; T&Ishave been philologists as well as non-philologists, but their capacity to performT&I endows them with not only interlingual but also metalinguistic skills. Theyare therefore not naïve speakers, who, in recent years, have become the primarysource for linguistic fieldwork samples.

66 Jim Hlavac

This article reclaims T&Is as a sample group of providers of linguistic data,whose proficiency and occupational practices qualify them as expert users. Aswell as being expert users, they are at the same time contemporary languageusers who on a regular basis deal with speech and text from one, two or threelanguage groups. Their occupational practices reflect not only official conven-tions for linguistic designations but also clients’ vernaculars and texts, as wellas marketplace demands.

3 Language policy in the SFRY: official andpopular designations of Bosnian, Croatianand Serbian

In the immediate post-World War II years, SFRY had four national languages:Serbian, Croatian, Slovenian and Macedonian. By the 1950s, the first two hadbeen officially merged into one, known as Serbo-Croatian; Slovenian andMacedonian remained separate languages as their features were consideredtoo distinct and different from those of Serbian and Croatian to warrant linguis-tic merging. Convergence of the linguistic standards of the Croats and the Serbsbecame official policy, culminating in the Novi Sad Agreement of 1954 (Okuka1998). The Novi Sad Agreement sought to officially sanction linguistic unity. Itstated that the Croats, Montenegrins and Serbs speak one language (BosnianMuslims remained unmentioned), with two literary centres (Belgrade andZagreb), two variants for pronunciation (eastern Ekavian and westernIjekavian), with two alphabets of equal standing (Greenberg 2004). Oppositionto and abandonment of this agreement by Croats in 1967 entrenched the positionof Serbian as the dominant component of Serbo-Croatian (Brozović 2008).Constitutional reform in 1971 led to the emergence of Sarajevo and, to a lesserextent, Titograd (now Podgorica) as centres of Bosnian-Herzegovinian andMontenegrin linguistic practice respectively (Radanović-Kocić 1986; Čirgić2011). These two smaller republics, while still part of SFRY, also publishedtheir own, adapted versions of grammars and orthographies of theSerbo-Croatian language.

From 1953 to 1991, SFRY listed three official national languages: Serbo-Croatian, Slovenian and Macedonian, while at the same time recognising thestatus of the languages of other nationalities such as Albanian, Hungarian andItalian. But the dominant language of the state remained Serbo-Croatian.Analogous to the name of the state, the term “Yugoslav” was used by some


locals at home or abroad when in contact with foreigners, who could not beexpected to know or remember the various languages spoken in SFRY.

In SFRY, the single largest national group were the Serbs (36% of thepopulation, Tabeau and Bijak 2005) and the Serbian capital, Belgrade, was alsothe national capital of SFRY. The concentration of political and military power2 inSerbia had linguistic consequences. When the number of texts for domestic andinternational readers that appeared in the language called Serbo-Croatian isquantified, it is striking how disproportionately large is the number that werewritten in Serbian (Ekavian) standard compared to the number of Serbian(Ekavian) speakers that made up but one part of SFRY (e.g. the language usedby Javarek 1974 and Babić 1978 in textbooks intended for foreign learners). Serbo-Croatian embodied and upheld Serbian linguistic traditions in a way that it didnot for other ethnic groups, most notably Croats but also Bosniaks, and mostrecently Montenegrins. The estrangement of non-Serb national groups from Serbo-Croatian is not only – as many would see it – a political act, but a consequence ofthat linguistic variety primarily representing the largest ethnic group rather thanother, less numerous ethnic groups. Overall, in most republics and in most non-official communication, Serbo-Croatian remained an unused and unloved term(see Kapović 2011). There are no popular or official movements to reinstate thisdesignation in any of the successor states of SFRY.

A detailed description of designations of the languages of the Bosniaks,Croats, Montenegrins3 and Serbs goes beyond the bounds of this article. Theinterested reader is referred to the many texts that deal with the union (basedoriginally on romanticism and pan-Slavicism and later on socio-political unitar-ism and Yugoslav federalism) and dissolution (based on preferences for ethni-city and language designation to be co-terminous) of the languages of theBosniaks, Croats, Montenegrins and Serbs (e.g. Bugarski 2002; Okuka 1998;

2 The concentration of political and military power in Serbia had linguistic consequences. Forexample, the political leader of SFRY from 1945 to 1980, Marshall Tito, a Croat from nearZagreb, spoke a laboured variety of Serbian after coming to power and taking residencein Belgrade. In the Yugoslav People’s Army, Serbs were disproportionally over-represented(65–70%) in the upper and mid-level ranks (Goldstein 2003) and vojnička ekavica (militaryEkavian pronunciation) was a euphemism to refer to the Serbian vernacular that was mostwidely used in the Yugoslav armed forces.3 In Montenegro, nearly one half of the population (approximately 350,000) declare their firstlanguage to be Montenegrin, and around a third Serbian (approximately 250,000), while theremainder of the population speaks minority languages such as Albanian, Bosnian, Croatianand Romany (Nikčević 2009). The development of a codified standard of Montenegrin is ofrecent vintage with the first orthography released in 2009 and the first grammar published in2010. Due to the small number of speakers of Montenegrin and an absence of informants bornor raised in Montenegro (see Table 3), Montenegrin is not further examined in this article.

68 Jim Hlavac

Neweklowsky 2003; Bugarski and Hawkesworth 2004; Greenberg 2004;Badurina et al. 2009). Table 1 provides the pre- and post-1991 designations forthe official languages of four successor states of SFRY.

4 Inter-group communication in SFRY andtheoretical models of variety selection inrelation to speakers of Bosnian, Croatian andSerbian

Notwithstanding the diversity of vernaculars that were subsumed under the termSerbo-Croatian and the variables that accounted for different varieties, vernacularsthat revealed a speaker’s ethnicity were commonly recognised as such. Diversity oflanguage varietieswas usually very apparent inmixed-ethnic verbal interactions andwritten exchanges. In these inter-ethnic interactions, a gamut of optionswere open tospeakers when communicating with others: no change to speech (zero accommoda-tion); avoidance of forms specific to one’s own variety; adoption of some forms fromanother’s variety without abandonment of one’s own forms; widespread adoption ofanother’s forms and conscious avoidance of one’s own forms (accommodation);comprehensive adoption of another’s forms (full convergence); and imitation andsocio-ethnic co-identification (passing). The term “accommodation” is used hereaccording to Giles et al.’s (1991) employment of the term to refer to the reduction oflinguistic dissimilarities in Communication Accommodation Theory (formerlyknown as Speech Accommodation Theory). The term “passing” is employed hereusing Piller’s (2002) notion of it to refer to non-locals who successfully imitate andconvincingly co-identify with locals to the extent that the locals view them as one oftheir own. Perhaps themost noteworthy recent example of passing in ex-SFRY is thatof Radovan Karadžić, who shed his native Montenegrin dialect after moving toBosnia-Herzegovina at the age of 15, only to then discard his unmistakeably urban,Sarajevo accent for Ekavian Serbian from the age of 51 onwards, which he spokefaultlessly while living as a fugitive in and around the Serbian capital, Belgrade.4

4 Radovan Karadžić’s abandonment of his Bosnian-Serb vernacular and acquisition of thevariety of Serbian spoken around Belgrade (Ekavian Serbian) was reported by both localnews sources (e.g. Sve Vesti 2008) and international ones (e.g. Kulish & Bowley 2008). Theformer source reports that he spoke Ekavian Serbian when appearing for the first time at theInternational Criminal Tribunal for the Former Yugoslavia in The Hague in 2008, while the lattersource quotes an unsuspecting acquaintance in Belgrade saying that “he did not speak with aBosnian accent”.


This begs the question: what were the features of the speech of thosespeakers of different language varieties in SFRY, and, in particular, of speakersof Bosnian, Croatian and Serbian when they verbally interacted with each other?This is a difficult question to answer because the state policy of unitarist Serbo-Croatian and socialist ideology of fraternity and unity prohibited systematicinvestigation of inter-ethnic interactions consisting of different language vari-eties. Investigation of ethnic groups’ linguistic differences was anathema to theofficial policy of ethnic equality and linguistic sameness. Information that canpartly address this question is available from other fields of linguistic research.As in many other countries, until the latter part of the twentieth centurylinguistic fieldwork in SFRY concentrated largely on dialectology and the record-ing of rural dwellers’ vernaculars (e.g. Finka & Moguš 1981; Kalogjera 1989;Dragičević 1986). A limited amount of sociolinguistic fieldwork was undertakenin SFRY (often not by citizens of SFRY, e.g. Magner 1966; Friedman 1993) thatfocussed on innovations in urban dialects due to rapid urbanisation and internalmigration of village dwellers to towns and cities. Again, this sociolinguisticresearch focussed on speech differences based on features such as place oforigin of speakers (or their parents) rather than more frequently employedsociolinguistic variables such as gender, age, class or ethnicity.

On the ground, the nominal ideology of fraternity and unity rarely resultedin linguistic egalitarianism. Within Serbo-Croatian, speakers of non-Serb vari-eties accommodated to Serbian, more so in the case of Bosnian and Montenegrinspeakers (Halilović 1998), and usually less so in the case of Croatian speakers(Franolić 1980b). Sometimes accommodation to Serbian was met with accom-modation by Serbian speakers towards the other variety, but rarely was itequally reciprocal. Asymmetrical accommodation to Serbian, as the numericallyand politically dominant variety, by speakers of languages other than Serbian isin line with findings from LePage and Tabouret-Keller (1985) and Sebba (1993),who report that speakers of prestigious or standard varieties who attempt toaccommodate in the direction of less prestigious or non-standard varieties aresuspected of mockery, while accommodation in the direction of the more power-wielding variety does not carry this implication. The changes that speakers ofthe less dominant Bosnian, Croatian or Montenegrin languages vis-à-vis speak-ers of Serbian can also be seen in terms of Myers-Scotton’s (2006: 154) morespeaker-centred model in which changes enacted by subordinate groups reflecthow they wish to present themselves to others, which include “negotiations ofself-identity and desired relationships with others”. Political conformism tothe communist Yugoslav supra-ethnic but Serb-dominant apparatus very oftenmotivated linguistic conformism amongst non-Serb aspiring functionaries (seeAuburger 1999).

70 Jim Hlavac

In the post-1991 setting, with three re-codified and separate standards,power dynamics have been re-aligned so that Serbian no longer occupies thestatus of dominant code that it did in SFRY, and each language standardoccupies the function of dominant code within each ethnic group. Further, inCroatia and Serbia where the majority ethnic group is co-terminous with thename of the language, Croatian and Serbian respectively assume the positionof dominant code for all citizens in these countries, regardless of ethnicity (e.g.Serbs in Croatia and Croats in Serbia). T&Is working in one, two or threelanguages, are required, as is normal practice for all T&Is in any situationanywhere, to employ a variety that is either the standard form of the languagethat they are working from and/or into or a form that is characteristic of thatspeech community. T&Is cannot, as they often may have done before 1991,employ Serbian for a Bosniak audience, or Croatian for a Montenegrin one, or ahybrid variety that bears features of convergence to another group’s vernacu-lar. Instead, the separation of the languages and the ethical standards andmarket forces that require practitioners to use the standard for which anassignment is requested (or market that is to be serviced) means that T&Isnow do not “cross-service” other groups through the use of Serbo-Croatian as ahypernym. Instead, T&Is are usually native speakers in one language (usuallytheir ethnic and/or chronologically first learnt language), expert users inanother, usually formally learnt language (e.g. English, German) and wherethey work in another, re-codified language from ex-SFRY, they do so as expertusers who rely on their native proficiency in a cognate language together withovert acquisition of forms, norms and rules specific to the language cognate totheir own native one. This situation is comparable to T&Is with three workinglanguages.

For interpreters, a further consequence of the separation of the three lan-guages is the protocols and norms that they apply when interlocutors use alanguage different from the one in which they accepted an interpreting assign-ment, and where a negotiation of the language form (and/or its designation)occurs. Consequences of this negotiation between the interpreter and otherinterlocutors may be anything from zero convergence to full adoption of theothers’ language, as well as the interpreter and/or the interlocutors declining tofurther engage in the interpreted interaction (Hlavac 2010a). Here, professionalnorms may contrast with desired linguistic choices, and interpreters may need tochoose whether to follow guidelines from codes of ethics (e.g. AUSIT 2012) thatprohibit interpreting in a language in which an interpreter does not haveaccreditation or whether to accede to clients’ requests to perform interpretingservices in languages cognate to a practitioner’s stated working languages(Hlavac 2010b).


Translators are similarly guided by recommendations that they work only inlanguages for which they have accreditation (or certification, e.g. ATA 2010). Atthe same time, as practitioners in languages for which demand is smaller know,the attraction of broadening their market base and/or requests for translations inother languages can motivate translators to seek and accept work in languagescognate to their own. As translators provide written rather than verbal(or signed) interlingual transfer, models based on spoken, face-to-face andusually dialogic interactions such as Giles et al.’s (1991) “CommunicationAccommodation Theory” and Myers-Scotton’s (1993) Markedness model havelimited applicability. The asynchronous and non-proximate nature of text trans-mission, from commissioner to recipient via translator, is more amenable tocommunication models that include unknown or remote addressees. For thisreason, Bell’s (1984) Audience Design model can be applied to the situation oftranslators, where the audience is the target audience or recipient specified bythe commissioner in the brief to the translator. Further, where the commissionerhas proficiency in the language into which the translator is translating, thecommissioner (as the commercial client whom the translator is serving) canassume the role of addressee, auditor and/or “over-reader” (analogous toBell’s “overhearer”) of the translated text. The Audience Design model includesnot only a responsive but also initiative dimension to the communicative choicesthat writers and translators (as well as speakers) may draw on: translators areoften supplied with little specific information about the profiles of consumers oftheir translations and the form, style, register, format and even content of thetranslation (or “target text”) are features that often only the translators them-selves are left to decide on. This is analogous to the remote, uni-directional andpublic nature of newsreaders’ speech examined by Bell (1984) as a basis forhis model.

5 Translation norms and occupationalmacro-pragmatics

In Translation Studies, the term “norms” applies to the regularities of behaviourthat T&Is exhibit in their approach to a text (written or spoken) and in theirpractice. “Competence norms” (Toury 1980) refer to those options that areavailable to T&Is in a given context; “performance norms” refer to the subsetof options that T&Is select in real life (Toury 1980: 63). As the term suggests,norms relate to the professional role that T&Is adopt to ensure that practitionersare able to work competently and accurately and that they act in a proper way

72 Jim Hlavac

towards all parties and uphold the ethical standards of the profession. As inother countries with a developed T&I infrastructure (i.e. training, testing andmarket sector), in Australia there is a professional code of ethics (AUSIT 2012)which recommends that practitioners should accept assignments only in lan-guages which they are competent to perform in. At all levels of government andamongst major T&I agencies in Australia there is a policy of assigning onlypractitioners that have accreditation in the required language. The workplaceand ethical duties that practitioners have to themselves, their clients and theprofession and the way that these guide their behaviour in interactions withothers are subsumed under a term congruent to Chesterman’s (1993) definitionof professional norms, that I label “occupational macro-pragmatics”. This is aterm analogous to “communities of practice” (Lave & Wenger 1991), but differentin that T&I practitioners often perform their work in isolation from other peers.Occupational macro-pragmatics refers therefore to the practices and strategiesthat practitioners have developed primarily on the basis of their own, personaloccupational experiences. It is not a prescriptive set of guidelines in the way thata code of ethics or statement of duties is. In a macro sense, the T&I sector doesnot have a community of practice, nor even communities of practice. Instead,the total number and wide variety of occupationally based practices amongstT&I practitioners that ensure professional and ethical standards in the perfor-mance of their duties calls for a more abstract hypernym. Thus, the termoccupational macro-pragmatics encompasses both regular patterns and irregularincidences of behaviour, where the latter are likely to occur in unusual or highlyspecific contexts.

6 Methodology

Invitations to participate in this research were sent firstly to Bosnian, Croatianand Serbian T&Is who advertise their services in the Australian Institute forInterpreters and Translators (AUSIT) and National Accreditation Authority ofTranslators and Interpreters (NAATI) online directories.5 Two agencies and twohealth services with a large number of Bosnian, Croatian and Serbian speakingclients were contacted in Australia. Further, T&I colleagues known to the authorlocated in Austria, Croatia and Serbia were contacted. The response rate from

5 Approval to contact potential informants and collect data was granted by the StandingCommittee on Ethics in Research Involving Humans (SCERH), Monash University. ProjectNumber 2007002093.


contacted potential informants – 47 out of approximately 120 ≈ 40% – is reason-ably high. However, the sample cannot be considered representative of T&Is ofthese three languages in Australia or elsewhere. Thirty-eight of the informantsare Australia-based, and the remaining nine are located in Europe – only four ofthese are still resident in Bosnia-Herzegovina, Croatia or Serbia. Due to theuneven distribution of informants from the émigré, European and homelandsettings, the only variable that is employed to compare groups of informants isthe category “no. of accreditations”. As outlined in Section 1, this article seeks tocompare the practices of T&Is across three languages with single or multipleaccreditations, and this variable alone is applied in the presentation of data.Other possible variables such as exact place of origin from an ethnically andlinguistically mixed versus homogenous area were not applied due to a lack ofdata from informants about their exact place of origin. Table 2 presents thenumbers of informants according to their focus of employment and number ofaccreditations. Informants who completed questionnaires as both translatorsand interpreters are included as single informants only and only the responsesfrom their major focus of employment (in all cases interpreting) are presented inthe data sample. In Section 7, data presented in the tables do not distinguish theresponses of translators from interpreters as there is little substantial variation inthe responses between the two groups. In the evaluation of responses presentedin Section 7, comparative data about the two groups are discussed only wherethere are substantial differences in responses between the two groups.

The author is an Australia-based practising Croatian-English interpreter and bi-directional Croatian < > English translator with accreditation (at the NAATI profes-sional level) in this language only. However, his own contacts include practi-tioners of all language backgrounds and the higher number of practitioners withCroatian as one of their languages is due to the larger number of Croatian speak-ers in Australia compared to the number of Bosnian and Serbian speakers.6

Questionnaires were collected from informants from July to September 2010.

Table 2: Number of informants according to area of work and number of accreditations

One accreditation Two accreditations Three accreditations Total

Translators 16 3 5 24Interpreters 14 5 4 23Total 30 8 9 47

6 Data from 2011 Census in Australia show that there are 16,269 Bosnian speakers, 61,547Croatian speakers and 55,114 Serbian speakers in Australia (ABS 2012).

74 Jim Hlavac

In the questionnaire, the terms “native speaker” and “near-native speaker”were used as approximate but commonly accepted yardsticks of “complete”proficiency and “near-complete” proficiency in a language (see Love &Umberto 2010). Informants were free to interpret these terms to apply to theirchronologically first learnt language/s, the language that they feel is their mostdominant, and/or the language/s of the ethnic group/s that they belong to. Theterm near-native speaker refers to languages that informants report having ahigh level of proficiency in, sufficiently high for informants to employ theselanguages in their work.

All interpreters work bi-directionally, that is, from and into their A language(s).All Australia-based translators have NAATI bi-directional (from and intoEnglish) accreditation. One translator in Germany has NAATI uni-directional(into English) accreditation only. The other translators based in Croatia andSerbia are holders of philological degrees with a T&I component. The Europe-based informants were left to select which languages they work in (from andinto) which always reflected the country in which they completed their philolo-gical degree and in which they currently reside or previously resided. Thus,Europe-based translators who selected one language in which they work havebeen allocated to the “single-accreditation” group of the sample. Those whoselected two languages have been allocated to the “double accreditation” groupand so on (see Tables 3 and 4). The Croatia- and Serbia-based practitionersgenerally translate into their A language. However, as translators of languagesfor which demand is smaller well know, the Croatia- and Serbia-based transla-tors often consider and accept assignments into their B language, and all ofthem identify themselves as bi-directional translators. Only one informant, the

Table 3: T&I practitioners’ number of accreditations and country of birth

Country of birth One accreditation Two accreditations Three accreditations Total

Bosnia-Herzegovina 5 3 1 9Croatia 8 3 6 17Serbia 8 1 1 10Australia 4 1 1 6Macedonia 2 0 0 2USA 2 0 0 2Poland 1 0 0 1

Total 30 8 9 47

Median no. of yearsworking as T&I

18.6 years 24.5 years 15.2 years 18.9 years


Germany-based one, stated an inability to answer questions about translationinto their B language.

Table 3 shows that 30 (64%) of the T&Is have one accreditation, 8 (17%) havetwo accreditations and 9 (19%) have three accreditations. Thirty-eight (81%) areborn in successor states of SRFY: Bosnia-Herzegovina 9 (19%); Croatia 17 (36%);Serbia 10 (22%); and Macedonia 2 (4%). Data in Table 2 and other qualitative datacontained in informants’ responses indicate that the vast majority (at least 41 ofthe 47 informants) are L1 or “A language” speakers of Bosnian, Croatian and/orSerbian. Table 3 contains information about informants’ self-described languageproficiencies. (Multiple responses were permitted.)

Table 4 shows that overall, the average number of accreditations per infor-mant is just under three. Informants generally possess accreditation in threelanguages: their native language, a cognate language and the language of theirhost society (usually English). Amongst the 47 informants, 9 claim nativeproficiency in two languages (2 � Bosnian þ Serbian, 2 � Bosnian þ English,1 � Croatian þ Serbian, 1 � Croatian þ German, 1 � Croatian þ English, 1 �Serbian þ English, 1 � Serbian þ French) and 4 informants claim nativeproficiency in three languages (2 � Bosnian, Croatian þ Serbian, 1 � Bosnian,Croatian þ English, 1 � Croatian, English, German).

7 Data and discussion

This article contains a large amount of mostly quantitative data. There are alsosolicited and unsolicited comments from informants which add to the picture

Table 4: Combined totals of self-rated proficiencies and accreditations according to language

Native speaker Near-native speaker Accreditation

Bosnian 8 15 18Croatian 22 15 30Serbian 16 17 25English 11 34 45German 2 6 6Macedonian 2 0 2Hungarian 1 0 1French 1 0 1Polish 1 0 0

Total 64 87 128(Total no. of informants: 47)

76 Jim Hlavac

provided by the quantitative data. Interpreter or translator status, informantnumber and accredited languages are presented after quotes from informants.Data presented in this section are only briefly contextualised with the theoreticaldiscussion presented in Sections 4 and 5. Section 8 contains a fuller interpreta-tion of the data and commentary on the practices and strategies of the overallsample of T&I informants. The figures shown in Tables 5–12 are percentages thatrelate to the number of responses provided by each group of informants, that is,the different responses from all 30 informants with one accreditation only arebroken up into percentages that reflect the distribution of responses; responsesfrom the 8 informants with two accreditations are similarly presented in percen-tage form; responses from the 9 informants with three accreditations are alsopresented in this way. In the last column of each table are the overall averageresponses from all 47 informants. For one question, informants could selectmultiple answers and for this reason, some columns in Table 5 have totalsthat are greater than 100%.

The content of the data presented below focusses on informants’ self-reported speech in non-occupational (e.g. social) situations, their professionalprotocols when confronted with a language different from the one for which theyaccepted an assignment, attitudes towards accepting assignments in languageswith unofficial or unclear designations, others’ assumptions of their nativespeaker competency and ethnicity, their own and others’ attitudes towards thedistinctness of Bosnian, Croatian and Serbian as separate languages, and

Table 5: When you are not interpreting or translating, but communicating with someone whouses a language different from your own, do you avoid using forms that are specific to yourlanguage? Do you change your speech/text or expect the other person to change their speech/text in any way? (Multiple responses allowed.)

Oneaccredit.

Twoaccredit.

Threeaccredit.

Overallaverage total

percent

Yes. I adapt my speech/text to make it similarto that of the person that I’m speaking to.

40 50 44 40

No. I don’t expect the other person to adapttheir speech/text.

37 38 44 38

Yes. I avoid words or forms that are specific tomy language only.

33 25 44 34

No. I don’t adapt my speech/text. 30 38 22 30Yes. I expect the other person to also adapt

their speech/text to make it closer to mine.9 13 11 11


informants’ attitudes towards the future development of the three languages.Table 5 contains informants’ selected responses to their linguistic behaviour, notwhen interpreting, but when interacting with others whose speech is differentand identifiable as a variety spoken by another linguistic group. Multipleresponses are permitted.

Table 5 shows a rough 2:1 divide between the informants. Nearly two-thirdsof informants report changes in their own language, with relatively few expect-ing the other interlocutor to reciprocate. This indicates that self-rated conver-gence is a characteristic that a majority of informants engage in, with a lowerlevel of expectation that the same may be expected of the other interlocutor,although this may still be forthcoming. There is little difference in the selectionof responses from each of the groups according to number of accreditations.There is only a slight tendency for informants with one accreditation to reportadaptation of their own speech less often and to report no adaptation of speechmore often in comparison to the other two groups.

It is not uncommon for T&Is for the three languages to accept assignmentswhich may later reveal themselves to be assignments for a language other thanthe one accepted. Responses from Table 6 show that in the first place, almosthalf of the T&Is check with the client the language they wish to use or checkwith the agency that they know which language the text is written in. Informantswith one accreditation are more likely to do this than informants with multipleaccreditations. At the same time, a third of the informants with multiple accred-itations report that they would not do anything and proceed to interpret (in thelanguage for which the assignment was booked) or translate (from or into thelanguage for which the assignment was accepted) without further referral. For

Table 6: You have accepted an interpreting/translation assignment. When you commenceinterpreting for the client/receive the text and look at the language, you realise that thelanguage is different from the language for which you had accepted the job. What do you do?

Oneaccredit.

Twoaccredit.

Threeaccredit.

Overallaverage total

Check with the client which language theywant to use/that they know whichlanguage the text is in.

53 25 44 47

Do nothing and interpret/translate asnormal.

10 38 33 19

Check with the client which language theywant me to use

0 0 11 2

Other/no answer 37 38 11 32

78 Jim Hlavac

informants with multiple accreditations, code-switching to another languageneed not be overtly requested or justified. However, overall nearly half ofall informants respond that metalinguistic comment rather than convergenceto or adoption of another language variety is required to establish thevariety that the T&I should use. Comparison of the responses between transla-tors and interpreters not shown in Table 6 reveals that metalinguistic commentis more common amongst translators (57%) than interpreters (36%) as transla-tors cannot rely on locally negotiated verbal convergence as a strategy toestablish a commissioner’s target audience. For translators, a brief that containsnot only the target language but also the function of the text for a particularaudience is required before translation can be attempted. A number of infor-mants, particularly those with one accreditation, give specific accounts of suchexperiences.

Amongst those who check with the client on the language designation, thereis a range of comments about what this may entail:

(1) I explain that I don’t have accreditation for the language that they’re speaking butif they accept that I speak my language then we can continue. (Interpreter 7,Croatian)

This comment offers a speaker-centred account for code negotiation and theoffer that both parties proceed with the need to accommodate to the other. Forthis informant, the unmarked choice for the interaction remains Croatian only.Another interpreter claims to accommodate to the language of the client, butdownplays the accommodation as a strategy that does not threaten the practi-tioner’s face as a professional interpreter:

(2) I interpret in the language that the client is using. Sometimes I am booked as aCroatian interpreter for Bosnians. I then interpret for them in Bosnian, but in my caseit doesn’t make any difference to me. (Interpreter 18, Croatian+Serbian)

Another informant identifies agreed-upon mutual comprehensibility withoutfurther negotiation as the norm that they usually follow.

(3) If we understand each other I interpret as normal. (Interpreter 16, Bosnian+Serbian)

The following interpreter suggests that uni- or bi-directional convergence as wellas non-accommodation are possible outcomes, depending on the reception ofthe client to these offers:

(4) I negotiate with the client on what the best way is for us to be able to understand eachother well. (Interpreter 22, Bosnian+Croatian+Serbian)


Two further interpreters and one translator with multiple accreditations alsoclaim accommodation as the strategy that they pursue:

(5) I adapt to the language of the client. (Interpreter 15, Croatian+Serbian)

(6) I have lived in Serbia and later in Bosnia, so I don’t find it difficult to change mylanguage to my client’s language. I have often stayed in Croatia as well and was exposedto Croatian a lot too through the media. (Interpreter 23, Bosnian+Croatian+Serbian)

(7) If working for an agency I inform the agency that the language in actual fact is notCroatian, if that is the case, and then leave it up to them whether they want toproceed. Especially if the document is older, the official language was then Serbo-Croatian no matter which republic, and personally I have no problem with translatingthat, except if the alphabet is Cyrillic which I find harder and usually decline.(Translator 21, Bosnian+Croatian)

Comments (4)–(7) from informants suggest that accommodation, a strategyreported by half of the informants when not interpreting or translating, occursalso in their professional practice. Informants here only report their own con-vergence to others’ speech or text. Other informants are mindful of the proce-dural recommendations that are given to interpreters, and call on a third party,usually a T&I agency, to be part of the negotiation of language code:

(8) If there is any language difficulty, I report it to the agent. (Interpreter 16, Bosnian+Serbian)

(9) I contact the agency that arranged the assignment. (Interpreter 14, Serbian)

One translator mentions that only exceptional circumstances justify acceding toa client’s request:

(10) If it is an emergency and it is a simple text, then yes, as I am proficient in Serbianand the Cyrillic alphabet. (Translator 12, Croatian)

The responses from Table 6 show that informants expect the anticipated lan-guage of an assignment to be the unmarked choice. Any other code is markedand the marked code precipitates negotiation, usually led by the T&I practitionerwho also usually accommodates to the code of the client (or agency). Single-accreditation informants are more likely to invoke others (agencies, other thirdparties) in their negotiation of the language variety. Convergence and metalin-guistic speech/text appear as the most common strategies that informantsemploy in other-language situations.

Tables 7–9 present informants’ responses to assignments which containformer and now largely disused, unofficial or non-standard designations for

80 Jim Hlavac

the language in which the assignment is sought. Interpreter informants weresupplied with the following question: “An agency says that a client wants aninterpreter for language x. Would you accept this request?” Translators wereprovided with the following question formats: “A client wants a translatorfor work from language x into English. Would you accept this request?” and“A client wants a translator for work from English into language x. Would youaccept this request?” Translator informants were provided with two formats as allbut one translator work bi-directionally, notwithstanding the general preferenceamongst translators to translate into their A language. (All but one translatorresponded to the questions contained in Tables 7–9; see Section 6.) There weredifferent responses from almost all groups of informants according to a translationrequest from compared to into the other language. The statistics below contain anaveraged response from the translator informants for translation from and into the

Table 9: An agency says that a client requires a T&I for Yugoslav. Would you accept thisrequest?

One accredit. Two accredit. Three accredit. Overall average total

Yes 33 18 22 29Possibly 23 38 50 31No 44 44 28 40

Table 7: An agency says that a client requires a T&I for Serbo-Croatian. Would you accept thisrequest?


Yes 40 50 78 49Possibly 17 25 22 19No 43 25 0 32

Table 8: An agency says that a client requires a T&I for “Bosnian-Croatian-Serbian”. Would youaccept this request?


Yes 40 38 56 43Possibly 23 12 22 23No 37 50 11 32No answer 0 0 11 2


other language together with the interpreters’ responses. Informants were pro-vided with three options: “yes”, “possibly” and “no”.

Tables 7–9 reveal that in overall terms, relative majorities of the informantswould accept assignments under the designations Serbo-Croatian and Bosnian-Croatian-Serbian but reject assignments that used the designation Yugoslav.There are differences between the groups of informants. For Serbo-Croatian,roughly equal numbers of informants with one accreditation would accept anddecline a request while those with three accreditations are overwhelminglylikely to accept such a request. Those with two accreditations occupy a mid-point between the other two groups. For Bosnian-Croatian-Serbian, informantswith one accreditation are less likely than those with three to accept such arequest, but those with two accreditations are less likely than the other twogroups to accept such a request. Similarly, responses to an assignment inYugoslav are most often negative from those with two accreditations, whilethe other two groups have mixed or negative responses.

In general terms, informants with three accreditations are found to have a lessrestrictive attitude to assignments with non-standard designations and they alsorecord the highest percentage of possibly responses, indicating accommodation orpreparedness to negotiate aspects of the assignment where they may be reluctantto provide immediate acceptance. Those with one or two accreditations are morerestrictive, with degrees of variation ranging from moderate with Serbo-Croatianand Bosnian-Croatian-Serbian to low acceptance for Yugoslav.

There are differences between interpreters and translators in regard to theresponses. Interpreters, regardless of number of accreditations, were 10% lesslikely to give yes responses compared to translators in relation to Serbo-Croatian, but were almost 20% more likely to accept assignments under thelabel of Bosnian-Croatian-Serbian than translators. Translators appear to bemore reluctant than interpreters to translate from, but especially into a codesuch as Bosnian-Croatian-Serbian, which is an uncodified hypernym. The reasonfor these differences is the historical legacy of Serbo-Croatian as an officialdesignation that encompassed different but codified language varieties. This isnot the case for Bosnian-Croatian-Serbian, which is used only as a hypernym bysome bodies (e.g. International Criminal Tribunal for the former Yugoslavia) andeven then its use is accompanied by a disclaimer (Draženović-Carrieri 2002: 49).The following comment from one translator expands on this:

(11) The written languages of Bosnia, Serbia and Croatia are more different than thespoken word. Translating contracts has to be taken very seriously, as one cannot“jump” from one language to the other, as may be possible in case of interpreting “inbetween” these languages. (Translator 12, Croatian)

82 Jim Hlavac

There are a number of reasons why T&I practitioners show non-accommoda-tion to unofficial designations. These may include: ethical concerns; doubtsabout proficiency; negative responses from agencies, commissioners and cli-ents; fear of refusal from other parties on the basis of proficiency level orethnicity; and beliefs that the differences between the three languages aresuch that the same practitioner cannot service all three adequately. Thefollowing tables address the last three of these. Table 9 contains informants’responses in regard to refusals or assignment withdrawals due to informants’language proficiency.

Table 10 shows that 42 (89%) of the 47 informants do not report refusalsfrom clients in relation to perceptions of their proficiency. For the most part, thisis due to the circumstance that many of the informants are, much of the time,interpreting or translating into and from their native language.

Two informants who reply that they have not encountered refusals imply thatsome clients may register that they are not native speakers of one of theirlanguages and that uni- or bi-directional accommodation is the strategy thatthey (and others) employ:

(12) No. We show flexibility and mutual respect. (Interpreter 17, Bosnian+Croatian)

(13) No. Most of them did not mind. I make sure first that it’s okay by them. (Interpreter18, Croatian+Serbian)

One of the informants who reports refusal from clients states this in the contextof over a decade of working in the profession:

(14) In the last 14 years of phone interpreting two people refused to work with me for thatreason. (Interpreter 10, Serbian)

One informant mentions that attributes other than proficiency can bequestioned:

Table 10: While interpreting or when negotiating (or undertaking) a translation assignment, hasa client or other party ever refused to work or cancelled an assignment with you because theybelieve that you are not a native speaker of their language/the language required?


Yes 3 0 11 4No 90 100 78 89No answer 7 0 11 7


(15) No. They just sometimes questioned my ethnicity/religion. (Interpreter 23, Bosnian+Croatian+Serbian)

Another informant speaks at length about the overall context of another’slanguage, suggesting it may be a reason for clients to be initially uncertain.This informant explains that refusal could occur where a client has sufferedtraumatic experiences at the hands of members of another language group:

(16) I believe that the professional quality of an interpreter overrides the client’s unwill-ingness to accept “the other language” interpreter. The client may at first questionthe language of the interpreter but accepts him or her during the professional sessionif the job is done well. (Interpreter 16, Bosnian+Serbian)

Examples (15) and (16) indicate that non-accommodation from clients (and alsopractitioners) need not lead to the termination of the assignment; instead,speakers negotiate a lingua receptiva situation, that is, speakers agree thateach speaks his/her own language (see Rehbein et al. 2012).

As stated in Section 3, in SFRY a speaker’s “native code” was strongly,almost axiomatically, linked with that speaker’s ethnicity. The wars in the 1990shave led to a homogenisation of each ethnic group, with commensurate effectson linguistic behaviour: speakers of each ethnic group converge now morefrequently towards the specific characteristics of their group’s standard anddiverge more frequently from “shared-ethnic” regional or local varieties (seeDragosavljević 2000). The following question seeks to elicit whether informantshave experienced refusals from clients who believe that the informant belongs toa different ethnic group than themselves.

Table 11 shows that almost 80% of informants do not report refusals from clientsdue to differences in ethnicity. Amongst the responses that do not recordrefusals, there are informants who state plainly that differences in ethnicityare not problematised by clients. At the same time, there are others who reportthat clients’ questions about ethnicity need not lead to a refusal:

Table 11: While interpreting or negotiating a translation assignment, has a client or other partyever refused to work with you because they believe that you are of a different ethnicity to theirown?


Yes 3 12 11 6No 87 76 56 79No answer 10 12 33 15

84 Jim Hlavac

(17) No. I never hide my ethnicity. Clients never refuse me. (Interpreter 17, Bosnian+Croatian)

(18) No. As they are happy with the quality of work, there is no need to question thetranslator’s origins. (Translator 20, Bosnian+Croatian)

(19) Yes. I’ve been asked many times, but out of curiosity, not with hostility. (Translator13, Bosnian)

(20) Usually in the health care area of interpreting for clients from Serbia, Bosnia orCroatia, clients do not pose a great problem with accepting/refusing interpreters notof “their” “origin”. They are usually very accommodating. Although, how they reactto an interpreter not of their “origin” is very individual. (Interpreter 19, Croatian+Serbian, Original punctuation)

(21) Once a client objected that I wasn’t a real Bosnian but still accepted my service.(Interpreter 21, Bosnian+Croatian+Serbian)

Three further informants report on instances of refusal:

(22) Yes. Bosnian in court because I am not a Muslim. (Interpreter 16, Bosnian+Serbian)

(23) Occasionally a Croat would refuse my services because I’m not Croatian, eventhough I’ve lived in Croatia. (Interpreter 15, Croatian+Serbian)

(24) There are situations in which interpreters are not accepted by clients because theyare of a different background and so doubts arise as to their ability to impartially dotheir job, or simply because they have a general feeling of mistrust due to wartimeevents or for historical reasons. (Translator 3, Croatian)

While refusals are unusual, Table 11 also shows that informants with threeaccreditations are the group from whom a third do not provide a response tothis question. Responses (19)–(21) suggest that ethnicity is registered by othersand that it is a feature that can be verbalised, at which point, linguistic accom-modation is likely to cease as interlocutors negotiate the inter-ethnic interactionnot in the form of a linguistic variety but in the content of their speech.

As “expert language users”, T&Is are keenly aware of the similarities anddifferences between the languages and are familiar with the re-codifications ofthe three languages. Table 12 elicits their responses about future directions.

Table 12: Do you think that in the future, the differences between Bosnian, Croatian andSerbian will continue to increase, decrease or stay as they are now?


Increase 50 50 67 53Stay as they are now 43 50 33 43Decrease 7 0 0 4


Table 12 shows that over half believe that the differences between thelanguages will continue to increase while 43% believe that current differenceswill not increase.

8 Conclusions

This article has examined the reported practices and attitudes of a group of T&Ipractitioners and how they engage with speakers of and/or texts in the Bosnian,Croatian and Serbian languages as separate, distinct languages. Further, thearticle has examined whether there are differences in the way T&I practitionersengage with these languages on the basis of possessing accreditation in one, twoor all three languages.

T&I practitioners are service providers operating within a marketplace whichincludes external regulation of two kinds: standards (or elements of a code ofethics) of the professional association and/or credentialing authority that theT&I received accreditation from; and clients’ and other commissioners’ (usuallyagencies’) requirements. The first regulating authority invariably follows official(legislated) language designations and in all homeland countries as well as inmost countries of Western Europe, North America and Australia, these autho-rities classify Bosnian, Croatian and Serbian as separate, distinct languages. Inregard to the second regulating mechanism, clients and agencies, the data showthat these also employ the distinct designations for the languages that were oncesubsumed under Serbo-Croatian. Questions relating to the acceptance of assign-ments according to non-official designations show that only half of informantswould possibly accept an assignment under the title of Serbo-Croatian and fewerstill would accept one labelled Bosnian-Croatian-Serbian or Yugoslav. These arethe reported practices of the T&I practitioners in regard to their professionalprocedures.

In communicative non work-related contexts, informants generally accom-modate, that is, adapt their speech or text to reduce linguistic dissimilaritiesbetween their own speech or writing and that of their interlocutors or readers.Accommodation takes the form of adoption of forms similar to those used byothers and/or avoidance of forms specific to one’s L1. The same type of accom-modation is less often expected from the other interlocutor. There are no sub-stantial differences between informants with one, two or three accreditations.

Nearly half of informants overtly question their clients or others whenconfronted with speech or text that is in a language different from the one inwhich an assignment was accepted. Less than 20% of informants do not further

86 Jim Hlavac

question unexpected changes in the language of an assignment – the ones whodo this tend to be informants with multiple accreditations for whom a change inlanguage does not present a prohibition to continue the assignment. Thissuggests in the first place that metalinguistic comment or text is the mostcommon means of negotiating unexpected language varieties. Unsolicited com-ments from informants indicate that accommodation also occurs as a commonstrategy. Accommodation is more common amongst those with multiple accred-itations, who are less likely to question or check the code being used thaninformants with single accreditation.

Almost 90% of informants report that they have not encountered refusalsfrom clients or others on the basis that clients or others believe that theinformants are not native speakers of the language in which they have acceptedan assignment. Almost 80% of informants report that in situations where othersperceive an informant to be of a different ethnicity this does not lead to othersrefusing the informant’s services. “Other-group-ethnicity”, more often than“non-native proficiency”, is something that informants (interpreters more thantranslators) report as a possible feature of some assignments. Ethnicity can beremarked on by others, and sometimes problematised, as shown in some infor-mants’ recollections. Unilateral accommodation by the T&I, more so than bi- ormulti-lateral accommodation from other interlocutors, is evident in many infor-mants’ responses. These as well as professional attributes contribute to a com-missioner’s or client’s negotiation of acceptance or refusal of a T&I on the basisof different ethnicity. Although perceived non-native speaker competence is veryrarely reported as a reason for others to refuse informants’ services amongst allthree groups, there are differing patterns amongst the three groups. Informantswith one accreditation are least likely to report that others have refused them onthe basis of ethnicity. Those with two accreditations report a higher incidence ofrefusal, while a third of those with three accreditations do not supply an answerto this question.

Lastly, just over half the informants believe that current differences betweenBosnian, Croatian and Serbian will increase in the future, while almost all othersbelieve that differences will stay at the same level as they are now. Only twoinformants believe that they will decrease. Overall, T&Is with one accreditationgenerally engage in metalinguistic talk or text as a strategy to negotiate inter-actions with speakers of other, congruent languages and have a more restrictiveattitude to accepting assignments that carry former or non-formal designations.Single accreditation also limits the number of situations in which a T&I is likelyto encounter speakers of other languages and ethnicities, which accounts for alower number of reported refusals from others for their services on the basis ofdifferent language or ethnicity.


Informants with two or more accreditations employ accommodation as aprimary strategy, with or without metalinguistic talk or text, usually followed bycode-switching to the desired language of the client or commissioner.Informants with multiple accreditations are more open to requests that carrythe designation Serbo-Croatian than singly accredited informants. Multiple-accreditation holders are also unlikely to meet refusal from potential clients orcommissioners on the basis of their linguistic proficiency in one of their lan-guages. Likewise, their ethnicity may be rarely problematised by others, but asubstantial number of these informants avoid providing a response to thisquestion. Surprisingly, an overall majority of the multiple-accreditation infor-mants believe that the differences between the languages will increase.

Finally, the close but distinct relationship that the three languages occupyin state legislation, marketplace practices and in the informants’ linguisticrepertoires bears evidence of the following features. First, T&Is, and to a lesserextent, their clients or fellow interlocutors, engage in metalinguistic talk or texton the language to be used by the T&I. This feature is common to other, reportedinstances of lingua receptiva in which speakers (unexpectedly) encounter thespeech of other language varieties. Second, T&Is practise linguistic accommoda-tion while reciprocal accommodation practised by their clients is something thatis less expected and perhaps less extensive and/or frequent. This occurrence is afeature that is not shared with other lingua receptiva situations in which this isgenerally reported to be a reciprocal and dynamically negotiated process.

References

ABS (Australian Bureau of Statistics). 2012. Reflecting the nation: Stories from the 2011 census.Retrieved 2014-06-24, from http://www.abs.gov.au/ausstats/[email protected]/Lookup/2071.0main+features902012-2013

ATA (American Translators Association). 2010. Code of ethics and professional practice.Retrieved 2014-06-24, from https://www.atanet.org/governance/code_of_ethics_commentary.pdf

ATA (American Translators Association). 2012. Online directories. Retrieved 2014-06-01, fromhttp://www.atanet.org/onlinedirectories/individuals.php#translators

Auburger, Leopold. 1999. Die Kroatische Sprache und der Serbokroatismus. Ulm: Gerhard HessVerlag.

AUSIT (Australian Institute of Interpreters & Translators). 2012. AUSIT code of ethics and code ofconduct. Retrieved 2014-06-19, from http://ausit.org/AUSIT/Documents/Code_Of_Ethics_Full.pdf

Babić, Slavna. 1978. Serbo-Croat for foreigners. Belgrade: Kolarčev Narodni Univerzitet.Babić, Slavna. 1990. Hrvatski Jezik u političkom vrtlogu [The Croatian language in a political

maelstrom]. Zagreb: dr Ante Pelivan.

88 Jim Hlavac

Badurina, Lada, Ivo Pranjković & Josip Silić (eds.). 2009. Jezični varijeteti i nacionalni identiteti[Linguistic variants and national identities]. Zagreb: Disput.

Bell, Alan. 1984. Language style as audience design. Language in Society 13. 145–204.Brborić, Branislav. 1999. Das Serbische. In Uwe Hinrichs (ed.), Handbuch der Südosteuropa-

Linguistik, 339–382. Wiesbaden: Harrassowitz Verlag.Brozović, Dalibor. 2008. Jesu li Beči i Novosadski “dogovori” samo beznačajne epizode i činovi

unitarističkog nasilja – ili jedine osnovne točke u hrvatskoj novoštokavskoj standardiza-ciji? [Are the Vienna and Novi Sad “agreements” just meaningless episodes and acts ofunitarist force – or the only fundamental events in the standardisation of the Croatianlanguage based on Neoštokavian?]. In Anita Peti-Stantić (ed.), Identitet Jezika JezikomIzrečen [The identity of language expressed through language], 33–42. Zagreb: SrednjaEuropa.

Bugarski, Ranko. 2002.]Nova Lica Jezika [The new faces of language]. Belgrade: Slovograf.Bugarski, Ranko & Celia Hawkesworth (eds.). 2004. Language in the former Yugoslav lands.

Bloomington, IN: Slavica.Chesterman, Andrew. 1993. From “is” to “ought”: Translation laws, norms and strategies.

Target 5(1). 1–20.Čirgić, Adnan. 2011. Crnogorski jezik u prošlosti i sadašnjosti [The Montenegrin language in the

past and today]. Podgorica: Isik/Matica crnogorska.Dragičević, Milan. 1986. Govor ličkih jekavaca [The vernacular of speakers of Jekavian in the

region of Lika]. Srpski Dijalektološki Zbornik 32. 178–189.Dragosavljević, Andjelija. 2000. Language policies and academic responses: The Ekavian

debate in Republika Srpska. Australian Slavonic and East European Studies 14(1–2). 1–27.Dragovic-Drouet, Mila. 2007. The practice of translation and interpreting during the conflicts in

the former Yugoslavia (1991–1999). In M. Salama-Carr (ed.), Translating and interpretingconflict, 29–40. Amsterdam: Rodopi.

Draženović-Carrieri, Maja. 2002. BCS – A practical approach. In Radovan Lučić (ed.), Lexicalnorm and national language. Lexicography and language policy in South Slavic languagesafter 1989, 49–52. Munich: Otto Sagner.

Finka, Božidar & Milan Moguš. 1981. Karta čakavskog narječja [A map of the Čakavian dialect].Hrvatski dijalektološki zbornik 5. 49–58.

Franolić, Branko. 1980a. A short history of literary Croatian. Paris: Nouvelles Éditions Latines.Franolić, Branko. 1980b. Language policy and language planning in Yugoslavia with special

reference to Croatian and Macedonian. Lingua 51. 55–72.Friedman, Victor. 1993. Dialect variation and questions of standardization in

Macedonia: Macedonian, Albanian and Romani. Zbornik za filologiju i lingvistiku36(2). 7–35.

Giles, Howard, Nikolas Coupland & Justine Coupland. 1991. Accommodation theory:Communication, context and consequence. In Howard Giles, Nikolas Coupland & JustineCoupland (eds.), Contexts of accommodation: Developments in applied sociolinguistics,1–68. Cambridge: Cambridge University Press.

Goldstein, Ivo. 2003. Hrvatska povijest [Croatian history]. Zagreb: Inter Liber.Greenberg, Robert. 2004. Language and identity in the Balkans. Oxford: Oxford University

Press.Halilović, Senahid. 1998. Bosanski jezik [The Bosnian language]. Sarajevo: Baština.Hlavac, Jim. 2010a. Shifts in the language of interpretation with bi- or multi-lingual clients:

Circumstances and implications for interpreters. Interpreting 10(2). 186–213.


Hlavac, Jim. 2010b. Ethical implications in situations where the language of interpretationshifts: The AUSIT code of ethics re-visited. Translation and Interpreting 2(2). 29–43.

Ivić, Pavle. 1971. Srpski narod i njegov jezik [The Serbian people and their language].Belgrade: SKZ.

Javarek, Vera. 1974. Serbo-Croatian reader. London: St Paul’s House.Jurić-Kappel, Jagoda. 2003. Bosanski ili bošnjački? [Bosnian or Bosniak?]. In Gerhard

Neweklowsky (ed.), Bosanski – Hrvatski – Srpski. Aktuelna pitanja jezika Bošnjaka,Hrvata, Srba i Crnogoraca [Bosnian – Croatian – Serbian. Current questions on thelanguages of the Bosniaks, Croats, Serbs and Montenegrins]. Wiener SlawistischerAlmanach 57. 95–102.

Kačić, Miro. 1995. Hrvatski i srpski. Zablude i krivotvorine [Croatian and Serbian. Delusions anddistortions]. Zagreb: Institute for Linguistics, Faculty of Arts, University of Zagreb.

Kalogjera, Damir. 1989. Serbo-Croatian dialects and school in Yugoslavia. Sociolinguistica 3.152–158.

Kapović, Mate. 2011. Čiji je jezik? [Whose language is it?]. Zagreb: Algoritam.Kovačević, Marko. 1998. Hrvatski jezik između norme i stila [The Croatian language between

norm and style]. Zagreb: Nakladni zavod Globus.Kovačević, Miloš. 1999. U odbranu jezika srpskoga – i dalje [In defence of the Serbian language

– and further to that]. Belgrade: Trebnik.Kulish, Nicholas & Graham Bowley. 2008. The double life of an infamous Serbian fugitive. New

York Times, 23 July. Retrieved 2013-12-21, from http://www.nytimes.com/2008/07/23/world/europe/23karadzic.html?pagewanted=all&_r=0

Lave, Jean & Etienne Wenger. 1991. Situated learning: Legitimate peripheral participation.Cambridge: Cambridge University Press.

LePage, R. B. & Andrèe Tabouret-Keller. 1985. Acts of identity: Creole-based approaches toethnicity and language. Cambridge: Cambridge University Press.

Love, Nigel & Ansaldo Umberto. 2010. The native speaker and the mother tongue. LanguageSciences 32. 589–593.

Magner, Thomas. 1966. A Zagreb Kajkavian dialect. Philadelphia, PA: Pennsylvania StateUniversity.

Myers-Scotton, Carol. 1993. Social motivations for codeswitching. Oxford: Clarendon Press.Myers-Scotton, Carol. 2006. Multiple voices. An introduction to bilingualism. Malden, MA:

Blackwell.NAATI (National Accreditation Authority for Translators and Interpreters). 2011. Methods

of NAATI accreditation. Retrieved 2012-06-17, from http://www.naati.com.au/accreditation.html

NAATI (National Accreditation Authority for Translators and Interpreters). 2012. Accreditation bytesting. Retrieved 2012-12-01, from www.naati.com.au/PDF/Booklets/Accreditation_by_Testing_booklet.pdf

Neweklowsky, Gerhard (ed.). 2003. Bosanski – Hrvatski – Srpski. Aktuelna pitanja jezikaBošnjaka, Hrvata, Srba i Crnogoraca [Bosnian – Croatian – Serbian. Current questions onthe languages of the Bosniaks, Croats, Serbs and Montenegrins]. Wiener SlawistischerAlmanach 57.

Nikčević, Vojislav. 2009. Crnogorski interdijalektni/naddijalektni (koine) standardni jezik [TheMontenegrin interdialectal/supradialectal (koine) standard language]. In Lada Badurina,Ivo Pranjković & Josip Silić (eds.), Jezični varijeteti i nacionalni identiteti [Linguisticvariants and national identities], 147–168. Zagreb: Disput.

90 Jim Hlavac

Okuka, Miloš. 1998. Eine Sprache. Viele Erben. Sprachpolitik als Nationalisierungsinstrument inEx-Jugoslawien. Klagenfurt: Wieser Verlag.

Piller, Ingrid. 2002. Passing for a native speaker: Identity and success in second languagelearning. Journal of Sociolinguistics 6(2). 179–208.

Radanović-Kocić, Vesna. 1986. Synonym split in the dialect of Bosnia and Hercegovina: a studyof a change in progress. Studies in the Linguistic Sciences 16(1). 123–131.

Radovanović, Milorad. 1996. Srpski jezik [The Serbian language]. Opole: Uniwersytet Opolski –Instytut Filologii Polskiej.

Radovanović, Milorad. 2000. From Serbo-Croatian to Serbian. Multilingua 19(1–2). 21–35.Rehbein, Jochen, Jan ten Thije & Anna Verschik. 2012. Lingua receptiva (LaRa) – Remarks on the

quintessence of receptive multilingualism. International Journal of Bilingualism 16(3).265–286.

Samardžija, Marko. 1999. Norme i normiranje hrvatskoga standardnoga jezika [Norms and thesetting of norms for the Croatian standard language]. Zagreb: Matica hrvatska.

Sebba, Mark. 1993. London Jamaican: Language systems in interaction. London: Longman.Šipka, Milan, Mevlida Karadža-Garić & Josip Baotić (eds.). 1973. Mostarsko savjetovanje o

književnom jeziku: referati, diskusija, zaključci [The Mostar roundtable on literary lan-guage: Papers, discussion, resolutions]. Sarajevo: Institut za jezik i književnost u Sarajevu.

Sve Vesti. 2008. Karadžić pričao ekavski u sudnici [Karadžić spoke Ekavian in the courtroom].Retrieved 2013-12-21, from http://www.svevesti.com/a96554-karad%C5%BEi%C4%87-pri%C4%8Dao-ekavski-u-sudnici

Tabeau, Ewa & Jakub Bijak. 2005. War-related deaths in the 1992–1995 armed conflicts inBosnia and Herzegovina: A critique of previous estimates and recent results. EuropeanJournal of Population 21(2–3). 187–215.

Tolimir-Hölzl, Nataša. 2009. Bosnien und Herzegowina. Sprachliche Divergenz auf demPrüfstand. Munich: Otto Sagner.

Toury, Gideon. 1980. In search of a theory of translation. Tel Aviv: Porter Institute.Universitas. (n.d.). Verzeichnis-Suche. Retrieved 2012-10-19, from http://www.universitas.org/

de/service/uebersetzerinnen-und-dolmetscherinnen/einfache-suche/


Copyright of Multilingua is the property of De Gruyter and its content may not be copied oremailed to multiple sites or posted to a listserv without the copyright holder's express writtenpermission. However, users may print, download, or email articles for individual use.