RSS feed

“People have probably offered to buy me a dictionary 20 times since I’ve been here”: Risk management within a community of journalists in francophone Canada

Lucile Davier, University of Geneva

ABSTRACT

Gathering and writing news in a bilingual context increases the complexity of a practice already characterised by multitasking. Does this situation create particular risks? How do journalists deal with hazards? This article discusses the strategies of risk management that reporters develop as a community of practice and investigates what these strategies reveal about reporters’ conception of language. To discover these strategies, I carried out fieldwork in a newsroom situated in Canada’s National Capital Region: Ici Radio-Canada Ottawa–Gatineau, which is the francophone public service broadcaster that publishes multimodal content in French on various platforms (radio, television, a website and social media). I conducted semi-structured interviews, sessions of non-participant observation and gathered documents in the field. Participants are especially concerned by the risk of linguistic interference (Anglicisms) because they align with Ici-Radio Canada’s model of linguistic prestige and, therefore, fear complaints from their audience. They mainly share these risks with their direct French-speaking colleagues on an ongoing basis and with the speech community of educated French speakers in a context where French is seen as a minority language and English is seen as a threat.

KEYWORDS

Communities of practice, collaborative translation, news translation, audiovisual translation, risk management, ethnographic approach, non-professional translation.

By way of introduction, I would like to outline briefly the context of the fieldwork I conducted at Ici Radio-Canada, the francophone network of the Canadian Broadcasting Corporation. The newsroom to which I gained access is located in Ottawa, Ontario, and it produces news for the French-language community in the National Capital Region of Ottawa–Gatineau. This region straddles two provinces: Ontario, where English is the sole official language, and Quebec, where French has been the official language since the introduction of Bill 101 in 1977 (Hudon and Behiels 2015).

A brief sociolinguistic description is necessary to understand bilingualism in this region. On the one hand, French is a minority language within the Ontario part of the metropolitan area of Ottawa–Gatineau: English is the mother tongue of 60.5% of the population, whereas French is the mother tongue of 15.3% of the inhabitants (Statistics Canada 2016). Here, one should note that Statistics Canada defines mother tongue as “the first language learnt in childhood and still understood at the time of the census” (Mougeon and Beniak 1991: 21). On the other hand, English is a minority language within the Quebec part of Ottawa–Gatineau: English is the mother tongue of 12.9% of the population and French, 75% (ibid.). Notably, francophones living in Ontario are so proficient in English that nearly all of them are bilingual (Sioufi et al. 2016).

The Canadian Broadcasting Corporation was founded in 1936 (Vipond 2011), and it falls under regulations established by the Broadcasting Act. This Act specifies that the Corporation is expected to “contribute to shared national consciousness and identity” (Broadcasting Act 1991: vi), and to “be in English and in French […], including the particular needs and circumstances of English and French linguistic minorities” (Broadcasting Act 1991: Part I, art. 3, l. m). In other words, the Broadcasting Act clearly mandates that the Canadian Broadcasting Corporation should serve minority language communities. It is essential to understand that “the English-language CBC and the French-language Société Radio-Canada operate as virtually separate entities” (Vipond 2011: 141) and that each target different audiences. As part of their routine, journalists working for these entities have to translate when they cover an event occurring in the other language community.

In this process-oriented study of news translation, I investigate whether journalists perceive the fact that they are working with English and French in their practice of newsgathering and newswriting as a source of risk. If so, what types of risk do they identify? How do they deal with these risks as a community of practice? What does risk management mean to them?

1. Risk within a community of practice: Conceptual framework
1.1. The process-oriented approach in news translation

Different introductions to news translation—or journalistic translation research—concur in their observation that this subfield emerged at the beginning of the 21st century, even though the bulk of publications have appeared from 2010 onwards (Valdeón 2015; Schäffner 2018; Valdeón 2020). The aim of this section is not to review the different approaches to news translation, which previous articles or chapters have done extensively, but to focus on an area that Holland (2013) labels the “process-oriented approach”. This approach is concerned “with questions of how translations are produced, by whom and in what contexts” (Holland 2013: 336).

The multi-site ethnographic project conducted by Bielsa and Bassnett (2009) is usually identified as the first of its kind. After the publication of Bielsa and Bassnett’s seminal book, other scholars looked behind the scenes of news organisations by going into the field (among others, Davier 2014, 2017; Tesseur 2017; van Rooyen 2018; Matsushita 2019; van Rooyen 2019; van Rooyen and van Doorslaer 2021; Xia 2019).

In what they call the “dominant model” (observed at Reuters and AFP), Bielsa and Bassnett found that journalists “do not normally have any specific training as translators” (2009: 81). On the contrary, translation is sometimes even a task left for newcomers to acquaint themselves with journalistic style (Davier 2014; van Rooyen 2018). Journalists tend to share a “view of the source text as raw material”, which enables them to make important changes in content and structure to adapt to different journalistic norms and audiences (Bielsa and Bassnett 2009: 85), as noted also by Davier (2014) and van Rooyen (2018). Both Bielsa and Bassnett (2009) and Davier (2014) have also observed that journalists tend to maintain a narrow, literal understanding of translation (Bielsa and Bassnett 2009; Davier 2017). Nevertheless, Bielsa and Bassnett do not elaborate on the consequences of this conception of translation, while I show that it entails the risk of communicative inaccuracies (Davier 2017).

1.2. Translation as risk for journalists

To the best of my knowledge, Matsushita (2015) is the only news translation scholar who systematically applies a framework of risk management. Drawing on Pym (2015), she distinguishes credibility risk and communicative risk (Matsushita 2015). While credibility risk concerns the fear of losing readers’ trust—a central commercial concern in news organisations—and that of the source, communicative risk refers to the dangers of distorting the source’s message or of creating misunderstandings among the readers. As Matsushita was able to confirm through interviews, Japanese journalists seem to give priority to the comprehension of their readers (communication) to keep their trust (credibility) unless the speakers are prominent domestic figures, since the risk that those figures will check the accuracy of their quote is high (Matsushita 2015). For instance, reporters will maintain the exact wording of a quote from the Japanese Prime Minister (intralingual translation) even if that utterance may pose problems of understanding to their audience. On the contrary, they will insist on the readability of a quote from overseas sources because of the low risk such sources will read the Japanese version of their speech or interview.

To expand on Matsushita’s risk model, I revisited two previous studies with her concepts. In the first study (Davier 2014), I analysed the literal conception of translation shared by news writers as a source of risks entailing possible distortions and cultural approximations (ibid.). I understood these risks, discussed in semi-structured interviews, as communicative risks in Matsushita’s terms. In another publication (Davier 2017), I referred to a journalist working for the international news desk of a national news agency. This journalist mentioned the low-probability risk of losing sources’ trust (credibility risk) if their quotations are misrepresented through translation (communicative risk) in a foreign news organisation. In a domestic context however, one news editor on duty at the national news desk said that she took communicative risks more seriously. These specific examples accord with Matsushita’s findings (2015) but they are only supported by the comments of two interviewees working for news agencies in Switzerland.

It is necessary to add important contextual information. While in the field at these two news agencies, I noticed the embarrassment of some journalists: they feared I would judge their proficiency level in their foreign language, and others, particularly junior employees, said they dared not ask language-related questions out of their fear of being perceived as incompetent (Davier 2017). In both news agencies, several journalists acknowledged limitations to their foreign language skills (ibid.). Similarly, in the Belgian context, some francophone journalists tended to self-assess their proficiency in Dutch (one of the three official languages of Belgium) as limited (Bouko et al. 2019).

In this study, I aim to situate my results in relation to the findings of these previous studies conducted in Europe and Asia, and to discuss what risk means, not to individual journalists but rather to a community of practice.

1.3. The concept of a community of practice

To shift the focus from individual accounts to collective conceptions of risk and translation, I apply the concept of a “community of practice” in the meaning developed by Lave and Wenger (1991) and Wenger (1998) to the newsroom I investigated for this study1.

Not all collections of practitioners qualify as communities of practice. Three conditions must be met: “mutual engagement”, a “joint enterprise” and a “shared repertoire” (Wenger 1998: 73–4). Mutual engagement refers to dense relations between workers who “organize their lives with their immediate colleagues and customers to get their jobs done” (ibid.: 6). In other words, Ici Radio-Canada reporters attending the same daily editorial meetings would meet the criterion of mutual engagement; Ici Radio-Canada and CBC journalists would not, because they do not interact on a daily basis even if they share the same work space. Members must also be engaged in a joint enterprise, which one can interpret as writing news for the same media organisation (e.g. Ici Radio-Canada but not CBC). A shared repertoire corresponds to “routines, words, tools, ways of doing things, stories, gestures, symbols, genres” (Wenger 1998: 83), for instance shared software or language resources.

Wenger has developed another concept for people who interact with each other on a regular basis outside of their primary community of practice: they form part of a “constellation of practices” (Wenger 1998: 126) and “create forms of continuity that take on a global character” (ibid.: 129). I thus consider that Ici Radio-Canada reporters form a constellation of practices, for instance, with their anglophone counterparts from CBC.

2. An ethnographic study in a multi-platform media setting

As shown in the previous section, this study is anchored in the process-oriented approach of journalistic translation research, and it builds on the concept of a community of practice—areas of research in which scholars tend to rely on workplace studies (Wenger 1998) or ethnographies (see, among others, Flynn 2005; Neather 2012; Duflou 2016). In this tradition, I have adopted an ethnographic approach for my research focussing on bilingual practices of newsgathering and newswriting within a regional newsroom of the French-language public service broadcaster in Canada. In this type of approach, writing in the first person is common practice to acknowledge one’s positionality as a researcher (Bensa 2008; Saldaña 2011).

For this article, I conducted 12 semi-structured interviews and 7 sessions of non-participant observation, in which I observed 15 individual participants from January to June 2017 without taking part in their activities. I also collected publicly available and internal documents while in the field. This study is part of a larger project that includes another francophone media outlet and a small number of anglophone journalists based in the same area.

All the recordings of the interviews were transcribed verbatim, including false starts, stutters and redundancies. Nevertheless, one should note that “verbatim” is a social construction—or a form of “entextualisation” (Urban 1996)—since participants usually speak in a flow of language that researchers arbitrarily break down in sentences (Poland 2001).

I obtained ethical clearance from the University of Ottawa Research Ethics Board in October 2016. To guarantee the anonymity of my interviewees in this paper, I refer to all of them using feminine pronouns, and I have deleted all personal identifiers from the passages I quote. Nevertheless, for the sake of transparency, the original quotes in French can be read following their English translations. All translations are mine.

After data collection, I proceeded with qualitative data analysis. Following the principles of grounded theory (Corbin and Strauss 2008), I used the first round of coding to determine initial categories, which I then refined iteratively after several cycles of fieldwork and additional coding. To achieve this, I used the data analysis software QDA Miner (Provalis Research 2012), which is similar to Atlas.ti or NVivo.

3. A community sharing risks: Results

This section is organised around the two concepts defined in Section 1.2: risks of credibility towards the audience (3.1) and communicative risks (3.2). In the context of Ici Radio-Canada, risks of credibility cover risks of language mistakes in the target language (French), including interference from English. Here, communicative risks mainly include risks of distortions for the audience. In addition, I analyse two forms of risk management: risk transfer and risk mitigation. Risk transfer can occur when translators (or, here, journalists) ask someone for help (Pym 2015). Risk mitigation is a complex strategy that implies the acceptance of “one kind of risk to protect against the negative consequences” of the initial risk “without actually removing the initial risk” (Pym and Matsushita 2018: 2).

3.1. Risk of credibility towards the audience

The first sub-section (3.1.1) describes the risks the reporters are facing, while the next two (3.1.2 and 3.1.3) analyse different forms of risk management implemented by Ici Radio-Canada journalists working closely together in the Ottawa–Gatineau newsroom.

3.1.1. Time pressure and risk of interference

Out of my twelve interviewees, six discussed the risks of working under time pressure without elicitation on my part. Time constraints are seen as having a direct impact on language quality: “[…] it’s easy to make mistakes, in the–, when you’re in a hurry” (“[…] c’est facile de faire des erreurs là dans la, quand on est pressé”), acknowledges the director of French services. One digital editor observed the same phenomenon from the opposite perspective: “uh, the level of precision that is required with written language, well, that takes time” (“Puis, euh, le niveau d'exigence que te demande l'écrit, ça, ça prend du temps, là”). The statement indicates that high-quality linguistic output is not easily compatible with speedy production. The time constraint is interpreted by two of my informants as a consequence of digital news:

[…] we don’t have a choice, with the–, given the reality of the internet, we have to be the first ones to have a title, a picture, a lead, and send a news alert. […] Because people will click on the first alert they receive on their phone with that piece of news. ([…] on n’a pas le choix avec le, la réalité du Web, étant ce qu’elle est d’être les premiers à avoir un titre, une photo, une amorce et d’envoyer une alerte. Parce que les gens vont cliquer sur la première alerte qu’ils reçoivent sur leur téléphone avec cette nouvelle-là.)

One journalist also adds that the number of digital stories has “exploded exponentially” (“explosé de façon exponentielle”), which has left designated revisers overwhelmed.

The pressure to release news as fast as possible is all the riskier since Ici Radio-Canada is expected to meet high language standards. Four journalists mentioned these standards without being asked explicitly, and five repeated several times during the interview how much the “quality of French” (“la qualité du français”) matters. For one reporter, Radio-Canada embodies this linguistic quality to the extent that she used a “very high standard of French” (“un très bon niveau de français”) and “Radio-Canada French” (“un niveau de français… radio-canadien”) synonymously. She specified that this type of French, nevertheless, has to be understood by a broad audience. To her, Radio-Canada carries out an “educational mission” (“une mission éducative”) regarding the spread of linguistic correctness.

The director of French Services justified her “alignment” (Wenger 1998: 174) with Radio-Canada’s mission by referring to the expectations of the audience, who are “still very sensitive to the quality of the French” (“[…] le public est quand même très attentif à la qualité du français”). Journalists imagined their audience as “very educated” (“[…] des gens très éduqués”), “particularly demanding” (“particulièrement exigeants”) or “with a higher level of education in French” (“avec un niveau d’éducation plus élevé parmi le français”). They do not only imagine their audience: they concretely hear back from them, as I explain in Section 3.1.2. This vocal part of their audience can be said to form a “speech community” (Morgan 2004): a construct at the intersection “between language, speech, and social structure” (Hymes 1964: 385). This speech community defines itself by publicly asserting a standard form of French free of borrowings from English.

Another journalist provides a contextual justification for this mission: “[…] especially in a minority context, so outside of [the French-speaking province of] Quebec, we kind of have the mandate to promote French […]” (“[…] surtout en situation minoritaire, donc hors Québec, on a un peu dans notre mandat de promouvoir le fait français […]”). Promoting French has to be interpreted as promoting a form of linguistic correctness (‘quality’) or a standard version associated with quality journalism. When I explicitly asked whether she perceived this as “a form of linguistic activism” (“une forme d’activisme linguistique”), she agreed but then specified that this form of linguistic correctness was “seen internally as, as being part, being part of our mandate” (“Je crois que c’est vu à l’interne comme étant, comme faisant partie, comme faisant partie de notre mandat”) rather than an individual form of activism.

The director of French services interprets the practice of translation as a source of problems that affect the correctness of French (“[We see a]ll kinds of errors, grammatical, uh, syntactic mistakes, uh, Anglicisms. We – we have many of them, unfortunately.” [“Toutes sortes de fautes, grammaticales, euh syntaxe, euh anglicismes. On en, on en a beaucoup malheureusement.”]), contrary to what was observed in the Swiss (Davier 2017) and the Belgian media contexts (Bouko et al. 2019). This ideology of translation, or bilingualism, as a threat to the purity of language is expressed by seven other informants, once through a medical metaphor that interprets English as a virus and French as a victim: “there’s still a danger in working in –, in both languages, it’s that, sometimes we contaminate our other language” (“[…] il y a quand même un danger à travailler dans le, dans les deux langues, c’est que des fois, on contamine notre autre langue”).

Notably, all my participants name lexical and syntactic interferences (both called “Anglicisms” by Ici Radio-Canada journalists) as the highest risk to producing news in a bilingual environment. This form of language purism is typical of a context where English is perceived as a threat to the minority language – French. Concurrent with the ideology of Tewa speech described by Kroskrity (1998), this doctrine of purism focuses more heavily on the exclusion of English vocabulary than on grammatical and syntactic problems.

3.1.2. Non-stop live vetting: Risk mitigation

How is this risk managed in the digital era, where speed matters intensely? Whereas stories recorded for TV and radio are systematically vetted, digital news items are run as soon as they are ready and are vetted afterwards, depending on their popularity. This situation reflects an instance of risk mitigation, because the risk of losing a scoop to the competition is considered higher than the risk of carrying a story with typos and mistakes. A digital editor generalises this practice to all media: “it’s pretty well – I think – established in the industry now that we publish a news item first, and someone vets it later” (“[…] c’est pas mal, je pense, établi dans l’industrie maintenant qu’on publie et quelqu’un relit plus tard”). Digital editors take the risk of making mistakes and manage them as soon as possible. According to another digital journalist, a posteriori vetting fulfils the need to transfer the risk of mistakes from one individual journalist to the newsroom.

Vetters in the Ottawa–Gatineau newsroom first revise the most-read digital stories at one given moment. Screens with live information hang in all corners of the newsroom. These screens are a major working tool for vetters: “If there’s a major local story and we see that there are 200 people reading the text, let’s just say that I’m gonna edit it faster than a report that only has one person reading it at the time” (“S’il y a un fait divers majeur et qu’on voit que le texte, il y a 200 personnes dedans, disons que, moi, je vais aller le relire plus vite qu’un texte qui a une personne dedans en ce moment”). I was able to confirm this statement by my observations. One journalist called this a form of “continuous” correction (“[…] on corrige en continu nos textes”).

The French-language network of the Corporation also hires revisers based in Montreal to proof the published stories and correct mistakes. However, my informants experienced this centralised solution as insufficient: first, because “we don’t have an army of revisers and they can’t proofread all the texts, all that we produce in a day” (“Parce que, tu te doutes bien qu’on n’a pas une armée de réviseurs et qu’ils ne peuvent pas lire tous les textes, tout ce qu’on produit en une journée”), and second, because they are based outside the community of practice and “won’t necessarily fully understand the regional context” (“[…] ce sont pas des gens qui vont nécessairement connaître tout le contexte régional”) crucial to spot factual errors.

The risk of mistakes is also partly transferred to the demanding audience of Radio-Canada: readers can simply click on a button at the bottom of each news item to report a typo, and they regularly do so, according to a digital editor. The director of French services agreed that this process could be considered a form of crowdsourced vetting. This type of vetting works because Ici Radio-Canada’s readers are prone to complain when they discover language mistakes. As a digital editor stated: “[…] people who traditionally read us or who, who consume Radio-Canada news, […] are particularly demanding. People have probably offered to buy me a dictionary 15 or 20 times since I’ve been here” (“[…] les gens qui nous lisent traditionnellement ou qui, qui consomment Radio-Canada […] sont particulièrement exigeants. On a dû me proposer de m’offrir un dictionnaire au moins une quinzaine ou une vingtaine de fois depuis que je suis ici”). Audiences also report mistakes using the generic email of the newsroom, Twitter or Facebook.

3.1.3. Shared expertise as a form of risk transfer

In a context of extreme time pressure, relying on colleagues seems to be the most popular technique of risk transfer. All my participants said that they turn to their peers for help, contrary to what I noted among journalists working in Switzerland (Davier 2017). One informant saw this method as especially useful when she had a language question while in a hurry to run a story. For instance, when she was in the field, she would ask someone in the newsroom to look up a word for her on the language intranet site. A colleague of hers also said that she “always like[s] having someone else go over [her] texts” (“[…] j’aime toujours ça que quelqu’un repasse à travers mes textes”) to ensure they are free of language interference. Journalists help each other even when they are working on different stories. For example, I saw a reporter who, while reading the Corporation website, noticed a typo in a story she was not vetting and called the author to inform her.

According to four different informants, a group of “exemplary writers” is designated to reside in the Ottawa–Gatineau newsroom: an “informal group” of employees who have “particular skills in French” and who the “members of the team can turn to if they have particular questions” (“[…] un groupe informel euh d’employés qui sont, qui ont des aptitudes particulières euh en français et qui euh, vers qui les membres de l’équipe peuvent se tourner s’ils ont des questions précises”), according to the director of French services. One of these employees regularly sends language notes related to current affairs to the whole team. Their “language expertise” (Rampton 1990: 98–99) or “communicative competence” (Morgan 2004) is constructed by the hierarchy of each newsroom, with implicit evaluation criteria.

In addition to this pool of expertise constructed in local communities of practice, language expertise is crystallised at a higher level in the French services of the Corporation. Ici Radio-Canada employs linguists who have developed an intranet website with language resources that are adapted to the conventions of Radio-Canada (Villedieu and GTQL 2003). In the field, I learned that, in 2017, this website contained approximately 10,000 entries concerning equivalents for loanwords, correct spelling, typographical conventions, feminised forms, etc. Eight of my participants—who had all published a digital story—reported using it on a regular basis. One digital reporter even viewed it as Ici Radio-Canada’s “bible”, which can be seen as a form of “reification” typical of every community of practice (Wenger 1998: 59).

I have shown in Section 3.1 how central language quality is to Ici Radio-Canada, especially in a regional context where French is partly a minority language. Journalists consider mistakes as threats to the correctness and purity of Ici Radio-Canada French. This risk is transferred first from the individual journalist to the newsroom with informal help among colleagues and with formal help from vetters (and revisers), and second to parts of the audience who form a speech community that hunts language mistakes. One might wonder whether this continuous pursuit of language quality is being prioritised at the expense of issues of meaning.

3.2. Communicative risk

The first sub-section (3.2.1) examined the risks the journalists described in their interviews, while the next two (3.2.2 and 3.2.3) investigate how these risks are managed inside the community of practice and within the constellation of practices.

3.2.1. Workload and specialised contexts: Risk of mistranslations

It may be illuminating to show how the time constraints presented in Section 3.1.1 materialise in the process of production.

First, digital journalists carry four to six 1,000-word stories per day, according to a digital editor. In comparison, a reporter often prepares two or three stories of this length in one day. In addition, digital journalists do not only spend their days drafting texts: they also choose and trim images, edit videos, vet their colleagues’ reports, check information on the phone, attend editorial meetings, etc. Deadlines are not extended when they must translate a full story from their counterparts at CBC.

Second, even if my interviewees do not view translation as a difficulty, they are aware of its risks. Four of them acknowledge the general threat posed by translation: “Of course, it can lead to shifts of meaning, when it comes to that, there’s no doubt about it!” (“Mais c’est sûr que ça peut mener à des glissements sémantiques, ça, il y a pas de doute là-dessus, là”), stated an investigative journalist, for instance. Another investigative reporter talks about the “accountability” (responsabilité) she thinks she and her colleagues bear towards a broad public who believe in the accuracy of the reported quotes. One participant sees the stakes of translation in the light of the Corporation’s Journalistic Standards and Practices (Ici Radio-Canada 2018): “For me, there’s always, like, this weight, no matter what work I’m doing, I always think: right, is there a mistake, have I respected the Standards and Practices […]?” (“[…] moi, j’ai toujours ça, comme, qui pèse, peu importe ce que je fais dans mon travail, je pense toujours : OK, est-ce qu’il y a une faute, est-ce que j’ai pas respecté mes normes et pratiques ?”).

Third, the distribution of risk is situational, as Pym (2015) reminds us: my informants describe contexts implying specialised language as riskier than general topics. Three radio and TV reporters expressed the risk of covering an event in a specialised domain, particularly when no French-speaking press officer is present with whom they can “navigate this jargon” and “have some semblance of vocabulary” to enable later translation (“[…] un expert bilingue et/ou francophone […], avec qui on peut naviguer dans ce jargon-là, […] et avoir un, un semblant de vocabulaire pour pouvoir traduire ce qu’il nous dit”).

3.2.2. Shared journalistic expertise: Risk transfer

In the fight against misunderstandings and mistranslations, turning to a colleague or expert for help seems to occur even more systematically than the use of language resources does.

Help takes place within the community of practice: the seasoned journalists assist their junior colleagues more often. This form of teaching (and learning) between “old-timers” and “newcomers” seems to be taken for granted, even though Wenger (1998: 100) suggests that these efforts should be recognised. I was able to witness different forms of collaboration. One example is a reporter asking a colleague in digital news to assess her translation: “[…] I could’ve just trusted my own judgment that I thought was right at the time, but, uh, I, I preferred to double-check with a colleague. Two heads… uh, are better than, than one” (“[…] J’aurais pu juste me fier à mon jugement à moi que je pensais qui était juste à ce moment-là, mais, euh, je, j’aimais mieux valider avec un collègue. Deux têtes… euh, valent mieux que, que, qu’une seule”). Risk is shared within the community of practice.

Cooperation is also central in situations where a journalist needs to cover a more specialised topic for the first time. In those cases, knowledge is transferred within the community of practice. The following quotation shows how one senior member accompanies a less-experienced journalist so that she can move from a form of peripherality to fuller participation in the community (Wenger 1998: 165–166):

[…] I have a colleague […], she’s more specialised in this, and she’s French-speaking, so we’ve had some exchanges, and when I heard that I was going to cover [X], I asked her how it worked, uh, I spent a morning with her [X] so that she can explain a few things to me.
(“[…] j’ai une, une collègue […], elle est beaucoup plus spécialisée là-dedans, puis elle est francophone, donc on a eu certains échanges, puis, quand j’ai su que je m'en allais au [X], je lui ai demandé un peu comment ça fonctionnait, euh, j’ai passé une avant-midi avec elle au [X] pour qu’elle puisse m’expliquer un peu.”)

Her statement shows that learning is gradual and fully integrated in the newsroom routine to allow the inclusion of newcomers (Wenger 1998).

The same principle also works across communities of practice, for instance at the intersection between the French-language and the English-language networks of the Canadian Broadcasting Corporation. Two journalists who were raised in monolingual backgrounds and who acknowledged their difficulties with English when they began as journalists several years earlier tended to discuss their understanding of English statements with their anglophone colleagues.

Three reporters who had to cover trials or ministry briefings explained that they did not hesitate to contact experts they met in the field to petition them for clarifications. A reporter said that she would regularly speak to an attorney or prosecutor at the end of a trial. These conversations usually happen in English. Thus, risk is shared in the constellation of practices.

3.2.3. Proficiency in English as a favourable environment

I find it revealing that most journalists did not understand my question when I asked them about the challenges they face while working with two languages on a daily basis.

As I conducted more interviews, it became obvious that I was observing a situation different from the one I had witnessed in Switzerland (Davier 2014, 2017) and Bouko et al. had witnessed in Belgium (2019) in regards to mastery of a second language. The six informants who had grown up in monolingual French environments (and who had assessed their level of English as insufficient when they started in this job) had spent at least three years in a minority context (i.e. outside of the French-speaking province of Quebec). Three of my participants had studied journalism at an English-speaking university, one had grown up in a bilingual family, and two had started working for Ici Radio-Canada in a minority context more than 20 years ago. Furthermore, some of them spoke English in their personal life.

Perceived mastery of a second language does not necessarily amount to translation skills; one’s training as a translator does not mean that one identifies as bilingual (Toury 1995; Antonini 2011). However, the reporters I interviewed link their level of English with their confidence in translating, as this statement illustrates: “I’m not bilingual, but my English is better than before, so now, now I don’t really have trouble translating anymore” (“Je suis pas bilingue, mais mon anglais est meilleur qu’avant, puis, maintenant, maintenant, j’ai plus vraiment de la difficulté à traduire”). This idea was confirmed by the director of the French services, who agreed that francophone journalists were proficient in English and that English classes would, therefore, be superfluous in the Ottawa–Gatineau newsroom. This finding is in line with Conway’s historical observations that “francophone journalists bore most of the burden of translating” (2011: 35) because nearly all of them had professional proficiency in English whereas fewer of their anglophone counterparts mastered French (ibid.: 53) and the minority position of French in Canada. In fact, “minority-language cultures are of course translation cultures par excellence as they are heavily dependent on translation”, writes Cronin (2009: 170).

In addition, proficiency in English does not seem to be an explicit part of the recruitment policy in the Ottawa–Gatineau newsroom: according to the managing editors and the staff, English skills are only tested informally through a few questions asked in English during the job interview. As opposed to regional newsrooms in Western provinces such as Manitoba or Saskatchewan (where the French minority communities are much smaller than in Ontario), where there are more new-comers, a couple of journalists acknowledged that the Ottawa–Gatineau Ici Radio-Canada newsroom is traditionally staffed with seasoned journalists. Language expertise appears to be distributed at the national level within the Corporation.

My informants share a repertoire unheard of in the world of journalism until now (Davier 2014, 2017; Matsushita 2015). Contrary to some of Matsushita’s findings, my participants are reluctant to retain some form of “unnatural phrasing” (2015: 133) in the target language. Eight of them explicitly reject literal translation. The following quote is representative of their general conception: “I may get further away from the original if I translate word for word, that’s why, so…, so I try to translate the most precise meaning” (“[…] je peux m’éloigner de la traduction plus en traduisant mot à mot, c’est ça qui est, donc j’essaie de traduire le sens le plus juste, quand je traduis”). Two broadcasting reporters explained that word-for-word translation is nearly impossible in the case of television news because the voice-over is necessarily shorter than the actual quotation. In addition, my informants shared concerns that are not alien to trained translators: producing a translation that does not read as a translation, taking context into account, finding equivalent idiomatic phrases or adapting cultural references. Their conception of translation differs from a standard one among existing studies of news translation, and it seems closer to the one of professional translators.

The risk of mistranslations does exist but is kept low because risks are shared among members of the community and proficiency in the second language is particularly high.


4. Managing risks as a bilingual community of practice: Concluding remarks

The journalists working at the Ici Radio-Canada Ottawa–Gatineau newsroom expressed two types of risk linked with bilingual contexts and translation in their journalistic practice: the risk of interferences (or, in Matsushita’s (2015) terms, credibility risk towards the audience) and, to a much lesser extent, the risk of mistranslations (or communicative risk for the audience). In line with Matsushita’s (2015) findings, the fear of losing audience trust is more relevant than hypothetical concerns about mistranslations (about which the journalists receive scant complaints). Unexplored in previous research on news translation is the focus on language mistakes in the mother tongue (reflected in the high expectations of the audience) and the collective management of this form of credibility risk.

The focus on linguistic interference—or Anglicisms in this particular case—can be interpreted as an ideology of purism with explicit and shared “proscription against the use of foreign words” (Kroskrity 1998: 107). Interferences are envisioned as a risk for two reasons. First, participants describe the way that Ici Radio-Canada stands for a model of “linguistic prestige”, which can likewise occur in other contexts with formal education or elite media (Kroskrity 1998: 108; Spitulnik 1998). My participants share risks within the community of practice and align with the goal of the institution to serve as a model of high-standard French in Canada. While some comply literally with Ici Radio-Canada’s editorial policy, others identify with it through a broader perspective of preserving French, which they view as an endangered language in this minority context.

Second, journalists strive for linguistic correctness in response to regular feedback from their audience, which allocates part of the risk to an elite readership keen to indicate language mistakes, especially mistakes that result from English-into-French transfers. Journalists co-construct Ici Radio-Canada as a model of linguistic prestige with a demanding readership component. In other words (see Figure 1 below), journalists share the risks within the community of practice (with their direct French-speaking colleagues), between communities of practice (with their anglophone CBC colleagues), as well as in the broader constellation of practices (as they intersect with the community of educated French speakers in a context where French is a minority language).

Figure 1. Schematic representation of the management of bilingual risk in the community of practice and across communities in the constellation of practices

The “language expertise” (Rampton 1990) is distributed within the newsroom to a few journalists chosen through hierarchy. Ici Radio-Canada’s audience also continuously assess the language expertise of all journalists producing content for the French-speaking network.

In the case of digital news items (which are increasingly important), Ici Radio-Canada has nevertheless decided to adopt the hazard of a posteriori vetting. This decision functions as a form of risk mitigation, since the risk of losing a scoop or alert to the competition is assessed as higher than the risk of mistakes. Nonetheless, live information about internet users enables digital editors to prioritise vetting. Finally, the high level of English proficiency among francophone journalists working for the Corporation creates a rather favourable environment (Canfora and Ottmann 2015) that seems to reduce mistranslation risks. This finding contradicts Toury’s view (1995) that bilingualism is not a “precondition for the development of translation competence” (Antonini 2011: 102). It may be unique to this context, where French is a partly a minority language.

Because mastery of French appears to be less common among the anglophone reporters sharing the same office space, a comparative study would have yielded revealing results. Despite my attempts, I was not granted access to the English-language section of the newsroom, which is managed by a separate hierarchy.

The conceptual framework of communities of practice places a welcome emphasis on collaboration inside an organisation instead of individual skills and decisions. It also blurs the borders between news producers and news consumers by showing how readers’ feedback is integrated in the strategy of risk management. It would be interesting to undertake further research into these aspects from the newswriting world to investigate whether they are equally relevant in communities of trained translators.

Acknowledgments

I am deeply indebted to all the extraordinary journalists who shared with me their passion for their work without looking at their watch. I am very, very grateful to Peter Flynn for showing me that “nothing is ever only about one thing”. Many thanks also to Aurélien Riondel, for his insightful comments, and Danielle Thien, who kindly took the trouble of revising my translations of the interviews. Of course, any shortcomings still present remain my own responsibility.

References
  • Antonini, Rachele (2011). “Natural translator and interpreter.” Yves Gambier and Luc Van Doorslaer (eds) (2011). Handbook of Translation Studies. Vol. X. Amsterdam/Philadelphia: John Benjamins, 102-104.
  • Bensa, Alban (2008). “Conclusion : Remarques sur les politiques de l'intersubjectivité.” Alban Bensa and Didier Fassin (eds) (2008). Les politiques de l'enquête. Paris: La Découverte, 323-328.
  • Bielsa, Esperança and Susan Bassnett (2009). Translation in Global News. London/ New York: Routledge.
  • Bouko, Catherine, Standaert, Olivier and Astrid Vandendaele (2019). “Being a journalist in a multilingual country: Representations of Dutch among Belgian French-speaking journalists.” Multilingua: Journal of Cross-Cultural and Interlanguage Communication 38(3), 231-251. DOI: 10.1515/multi-2018-0036.
  • Broadcasting Act, Statutes of Canada (1991, c. 11). https://laws-lois.justice.gc.ca/eng/acts/B-9.01/ (consulted 01.03.2021).
  • Canfora, Carmen and Angelika Ottmann (2015). “Risikomanagement für Übersetzungen.” trans-kom 8(2), 314-346.
  • Chartrand McKenzie, Francine (1967). Les Journalistes anglo- et francocanadiens : leurs opinions et leurs comportements vis-à-vis de la coexistence des deux cultures au pays. Report. Ottawa: Royal Commission on Bilingualism and Biculturalism.
  • Conway, Kyle (2011). Everyone Says No: Public Service Broadcasting and the Failure of Translation. Montreal: McGill-Queen's University Press.
  • Corbin, Juliet and Anselm Strauss (2008). Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Thousand Oaks: Sage.
  • Cronin, Michael (2009). “Minority.” Mona Baker and Gabriela Saldanha (eds) (2009). Routledge Encyclopedia of Translation Studies. London: Routledge, 170-172.
  • Cumming, Carman, Cardinal, Mario and Peter Johansen (1981). Canadian News Services. Ottawa: Royal Commission on Newspapers.
  • Davier, Lucile (2014). “The paradoxical invisibility of translation in the highly multilingual context of news agencies.” Global Media and Communication 10(1), 53-72. DOI: 10.1177/1742766513513196.
  • Davier, Lucile (2017). Les enjeux de la traduction dans les agences de presse. Lille: Presses universitaires du Septentrion.
  • Duflou, Veerle (2016). Be(com)ing a Conference Interpreter: An Ethnography of EU Interpreters as a Professional Community. Amsterdam/Philadelphia: John Benjamins.
  • Flynn, Peter (2005). A Linguistic Ethnography of Literary Translation: Irish Poems and Dutch-Speaking Translators. PhD thesis. Universiteit Gent.
  • Holland, Robert (2013). “News translation.” Carmen Millán and Francesca Bartrina (eds) (2013). The Routledge Handbook of Translation Studies. London: Routledge, 332-346.
  • Hudon, R. and Michael D. Behiels (2015). Bill 101 (Charte de la langue française). https://www.thecanadianencyclopedia.ca/en/article/bill-101 (consulted 01.03.2021).
  • Hymes, Dell (1964). “Social structure and speech community: Introduction.” Dell Hymes (ed.) (1964). Language in Culture and Society: A Reader in Linguistics and Anthropology. New York: Harper and Row, 385-390.
  • Ici Radio-Canada (2018). Normes et pratiques journalistiques. http://www.cbc.radio-canada.ca/fr/rendre-des-comptes-aux-canadiens/lois-et-politiques/programmation/journalistique/ (consulted 01.03.2021).
  • Kroskrity, Paul (1998). “Arizona Tewa Kiva speech as a manifestation of a dominant language ideology.” Bambi Schieffelin, Kathryn Woolard and Paul Kroskrity (eds) (1998). Language Ideologies: Practice and Theory. Oxford: Oxford University Press, 103-122.
  • Lave, Jean and Etienne Wenger (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge, MA: Cambridge University Press.
  • Matsushita, Kayo (2015). Risk Management in the Decision-Making Process of English-Japanese News Translation. PhD thesis. Rikkyo University.
  • Matsushita, Kayo (2019). “Globalization of the emerging media newsroom: Implications for translation and international news flow in the case of BuzzFeed Japan.” Lucile Davier and Kyle Conway (eds) (2019). Journalism and Translation in the Era of Convergence. Amsterdam/Philadelphia: John Benjamins, 135-153. DOI: 10.1075/btl.146.06mat.
  • Morgan, Marcyliena (2004). “Speech community.” Alessandro Duranti (ed.) (2004). Companion to Linguistic Anthropology. Oxford: Blackwell, 3-22.
  • Mougeon, Raymond and Édouard Beniak (1991). Linguistic Consequences of Language Contact and Restriction: The Case of French in Ontario, Canada. Oxford: Clarendon.
  • Neather, Robert (2012). “‘Non-expert’ translators in a professional community.” The Translator 18(2), 245-268. DOI: 10.1080/13556509.2012.10799510.
  • Poland, Blake D. (2001). “Transcription quality.” Jaber F. Gubrium and James A. Holstein (eds) (2001). Handbook of Interview Research: Context and Method. Thousand Oaks: Sage, 629-649.
  • Provalis Research (2012). QDA Miner. https://provalisresearch.com/products/qualitative-data-analysis-software/ (consulted 01.03.2021).
  • Pym, Anthony (2015). “Translating as risk management.” Journal of Pragmatics 85, 67-80. DOI: 10.1016/j.pragma.2015.06.010.
  • Pym, Anthony and Kayo Matsushita (2018). “Risk mitigation in translator decisions.” Across Languages and Cultures 19(1), 1-18. DOI: 10.1556/084.2018.19.1.1.
  • Rampton, Ben M. B. H. (1990). “Displacing the native speaker: Expertise, affiliation, and inheritance.” ELT Journal 44(2), 97-101. DOI: 10.1093/elt/44.2.97.
  • Saldaña, Johnny (2011). Fundamentals of Qualitative Research. Oxford: Oxford University Press.
  • Schäffner, Christina (2018). “Language, interpreting, and translation in news media.” Kirsten Malmkjaer (ed.) (2018). The Routledge Handbook of Translation Studies and Linguistics. London: Routledge, 327-341.
  • Sioufi, Rana, Bourhis, Richard Y. and Réal Allard (2016). “Vitality and ethnolinguistic attitudes of Acadians, Franco-Ontarians and Francophone Quebecers: Two or three solitudes in Canada's bilingual belt?” Journal of Multilingual and Multicultural Development 37(4), 385-401. DOI: 10.1080/01434632.2015.1072205.
  • Spitulnik, Debra (1998). “Mediating unity and diversity: The production of language ideologies in Zambiam broadcasting.” Bambi Schieffelin, Kathryn Woolard and Paul Kroskrity (eds) (1998). Language Ideologies: Practice and Theory. Oxford: Oxford University Press, 163-188.
  • Statistics Canada (2016). Language Highlight Tables, 2016 Census: Census Metropolitan Areas and Census Agglomerations. https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/hlt-fst/lang/Table.cfm?Lang=E&T=12&Geo=00&view=2&age=1&RPP=100 (consulted 01.03.2021).
  • Tesseur, Wine (2017). “The translation challenges of INGOs: Professional and non-professional translation at Amnesty International.” Translation Spaces 6(2), 209-229. DOI: 10.1075/ts.6.2.02tes.
  • Toury, Gideon (1995). Descriptive Translation Studies and Beyond. Amsterdam: John Benjamins.
  • Urban, Greg (1996). “Entextualization, replication, and power.” Michael Silverstein and Greg Urban (eds) (1996). Natural Histories of Discourse. Chicago: The University of Chicago Press, 21-44.
  • Valdeón, Roberto A. (2015). “Fifteen years of journalistic translation research and more.” Perspectives: Studies in Translation Theory and Practice 23(4), 634-662. DOI: 10.1080/0907676X.2015.1057187.
  • Valdeón, Roberto A. (2020). “Journalistic translation research goes global: Theoretical and methodological considerations five years on.” Perspectives: Studies in Translation Theory and Practice 28(3), 325-338. DOI: 10.1080/0907676X.2020.1723273.
  • van Rooyen, Marlie (2018). “Investigating translation flows: Community radio news in South Africa.” Across Languages and Cultures 19(2), 259-278. DOI: 10.1556/084.2018.19.2.7.
  • van Rooyen, Marlie (2019). “Tracing convergence in the translation of community radio news.” Lucile Davier and Kyle Conway (eds) (2019). Journalism and Translation in the Era of Convergence. Amsterdam/Philadelphia: John Benjamins, 155-176. DOI: 10.1075/btl.146.07van.
  • van Rooyen, Marlie and Luc van Doorslaer (2021). “News translation as collaboration in multilingual community radio stations in South Africa.” Language and Intercultural Communication, 21(3), 411-425. DOI: 10.1080/14708477.2021.1883046.
  • Villedieu, Yanick and Groupe de travail sur la qualité de la langue (GTQL) (2003). Un français de qualité : une priorité pour Radio-Canada. Report. Montreal: Radio-Canada.
  • Vipond, Mary (2011). The Mass Media in Canada: Who Decides What to Read, Watch, and Hear? 4th ed. Toronto: Lorimer.
  • Wenger, Etienne (1998). Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press.
  • Xia, Liang (2019). A Discourse Analysis of News Translation in China London: Routledge.

Biography

davier_portraitLucile Davier is a Research and Teaching Fellow at the Faculty of Translation and Interpreting of the University of Geneva (Switzerland). She was a postdoctoral researcher at the University of Ottawa (Canada) in 2016–2017 and a visiting scholar at the University of Leuven (Belgium) in 2012–2013. Her research interests include news translation, the ethnography of translation, news agencies and multi-platform media. (Picture by Alexandre Bourquin)

E-mail:Lucile.Davier@unige.ch

Notes

Note 1: For a review of the literature on the use of the concept of communities of practice in Translation Studies, please consult the article introducing this special issue.