1One of the main aims of many introductory courses in sociolinguistics is to put across the idea that a comprehensive definition of any named language cannot be limited to a description of its phonetic, morphological or syntactic features. Rather, the different attitudes, discourses and representations that surround any language must also be taken into account. Embracing these indexical elements, along with the aforementioned linguistic factors, brings us towards observing what Michael Silverstein terms the “total linguistic fact”, that is, an “unstable mutual interaction of meaningful sign forms contextualized to situations of interested human use, mediated by the fact of cultural ideology” (Silverstein, ‘Language and the Culture of Gender’ 220). From this perspective, three crucial components come to the fore when it comes to defining what any particular language is at any given moment: form, use and ideology.
2Despite the importance of each member of this trio (form, use and ideology), it could be claimed that form and use alone tend to take centre stage when it comes to the formal, explicit teaching, learning and testing of languages. However, one might argue that this does not mean that the ideological mediation mentioned by Silverstein simply disappears from teaching/learning/evaluation processes, but rather, it just finds itself sidelined to the realm of the implicit. By way of example, while teachers and students may explicitly focus on language forms and uses in the classroom, any discussion regarding the reasons as to why certain forms and uses are acceptable, whereas others are not, is likely to be characterised by traces of (ideological) language attitudes and representations. The initial motivation for this paper was borne out of personal observations made in a wide variety of contexts during such discussions with colleagues and/or students regarding a language at the centre of my research and teaching: English. In these exchanges, I noticed an almost unanimous convergence towards a set of attitudes and representations that paint a very particular, homogenous picture of what English “is” and/or what English “should be” in terms of how it should be pronounced, written, structured, etc.
- 1 An explanation of the Agrégation and details as to why it was chosen as a case study follow below.
3The aim of this article is to identify and explore some of the ideological dynamics that fuel this uniformised view of English and the attitudes and representations that undergird it. To do this, I focus on a key reference point in the teaching/learning of English in France – the Agrégation competitive exam1 – and show how different discourses contribute to legitimising certain linguistic forms and uses (and delegitimising others), thereby defining what English “is” in this context. Over the following pages, I first briefly introduce the main theoretical tool of “language ideologies” before homing in on the Agrégation and this study’s data, methodology and analytical framework. Through the data analysis, I explore how different language ideologies combine in the context of the Agrégation, leading to the elaboration of a restrictive set of reference models for the English language for this exam. In the discussion, I suggest that, given the influence of the Agrégation in France, the phenomena observed here do not just contribute to the ideological construction of English in this particular context, they potentially contribute to indexically elaborating what goes by the name of “English” in French education, and therefore in France more generally.
- 2 Detailed discussion of the concept of “language ideologies” falls outside the scope of this article (...)
- 3 The use of “beliefs” or “feelings” in these definitions does not (necessarily) imply that language (...)
4The main conceptual tool that forms the backbone of this study is that of “language ideologies”. Language ideologies are a long-standing heuristic in linguistic anthropology and sociolinguistics and can be defined in simple terms as “beliefs, or feelings, about languages as used in their social worlds” (Kroskrity 498), often “articulated by users as a rationalization or justification of perceived language and use” (Silverstein, ‘Language Structure and Linguistic Ideology’ 193)23. Language ideologies can thus be conflated with the representations and attitudes mentioned in the introduction to this article. They can be explicitly thematised, in “epilinguistic discourse” (Canut, ‘Pour une analyse des productions épilinguistiques’) (e.g. a speaker commenting on the language use of another speaker) and can also manifest themselves implicitly in speakers’ language practices (e.g. a speaker selecting one linguistic form over another).
- 4 Indeed, this is one of the main observations made by Silverstein (1985) with regards to the “total (...)
5Given the defining elements mentioned above, it is easy to see how language ideologies might play a role in how a language comes to be defined. From one perspective, language ideologies impact upon the language choices made by speakers, something which necessarily comes to exert an influence on the form and use of a language more generally. Simply put, if speakers employ or avoid certain forms in line with language ideologies, they collectively contribute to including or excluding those forms from the perimeter of the language, influencing the ways in which the language evolves4. From another perspective, as shown by Canut (‘À la frontière des langues’), epilinguistic discourses play a fundamental role in processes through which languages come to be named, identified, categorised and, ultimately, brought into existence. Such ideologically-charged discourses are key to defining what sits inside or outside the borders of a given language, contributing to its very elaboration as a (socio)linguistic object. These different processes contribute then to the ideological constitution of a language, discursively defining what a given language is in a particular context, at a particular time.
- 5 To be clear, Bucholtz and Hall’s article is mainly focused on the indexical relationships between l (...)
6The particular context of interest here is, broadly speaking, that of language education. When exploring language ideologies, education is often highlighted as a key context as it clearly constitutes a domain in which “indexical associations can […] be imposed from the top down by cultural authorities” (Bucholtz and Hall 596)5. One situation in which this often takes place is in the choice, elaboration and transmission of the variety (or varieties) of reference that will be learnt, taught and tested (see Cushing; Lippi-Green; Milroy and Milroy, among others). The process of picking out a variety (or varieties) of reference is intrinsically motivated by language ideologies and, similarly, language ideologies are mobilised to justify and rationalise the chosen model(s). As such, these ideologies are a central force in the very elaboration of the (socio)linguistic object being learnt/taught (all while propagating other beliefs and feelings about (the) language).
- 6 In the interests of space, I have decided to employ relevant references at strategic points of the (...)
7Studies on language ideologies in education and/or the ideological construction of languages and/or the sociolinguistic repercussions of these dynamics are nothing new6. However, to my knowledge, research on these topics is rare in the context of the teaching/learning of additional languages in France, and especially English (see Miras et al.; Dupouy and Wilson, for further insight). This article constitutes then a humble attempt at taking an exploratory step in this direction, with the English Agrégation competitive exam being chosen as a particularly pertinent focus for such an endeavour.
8The Agrégation is a concours (competitive exam) that grants those who pass it (those who are admis or admises) access to teaching positions in the French education system, at secondary or, in some cases, tertiary level. Different Agrégations exist for different subject areas. The English Agrégation is aimed at future English teachers and candidates can choose to specialise in either civilisation (history/culture), littérature or linguistique.
9It should be noted that the Agrégation is not the only concours that provides access to teaching positions in English in France. Indeed, only a minority of English teachers are agrégés or agrégées. Beyond this, it should be noted that, in comparison to the total number of students in France who have English as a major component of their studies, only a tiny proportion actually sit the Agrégation and an even smaller percentage successfully pass the exams to become agrégé or agrégée. By way of example, in 2022, 856 candidates sat the Agrégation externe7 and 151 of those candidates were successfully admitted (17.64%)8, whereas thousands upon thousands of students sign up to begin a degree in English – or a degree in which English forms a major part – every year.
- 9 Interestingly, there is very little scientific research on this issue (though Bourdieu & de Saint M (...)
- 10 Original French (my translation): l’agrégation d’histoire tend à apparaître de plus en plus comme u (...)
10Despite (or perhaps due to) the fact that there are relatively few agrégés and agrégées, the Agrégation benefits from substantial social prestige in France. This prestige is both material – teachers who are agrégés or agrégées get paid more, and teach less, than those who are not – and symbolic, whereby the Agrégation is often considered as a marker of academic excellence and quasi guarantee of expertise in a given subject area9. Indeed, as noted by Colon in relation to the discipline of history, the Agrégation “tends to appear increasingly as a title, rather than as a competitive exam for recruitment” (Colon 37)10. It is then for this prestigious status in French society that the Agrégation has been chosen for the present study, a point on which I elaborate towards the end of this article.
11Within the English Agrégation, the evaluation of candidates is based on various individual exams that all fall within the overarching competitive exam. Candidates first sit written exams, consisting in a dissertation (in French), a text commentary (in English), a linguistic analysis involving both grammar and phonetics/phonology (in English) and literary translations. Those who pass these written exams (those deemed admissibles) then sit a series of oral exams, which involve a commentary/explanation of a text/document in English (followed by an interview in English), a leçon (lesson, or lecture) in English (followed by an interview in French), an oral presentation and accompanying interview in English and an exam in which candidates listen to a recording in English before translating into French with added comments and/or context (followed by an interview in French). In each of the exams that take place in English, a grade is awarded for the quality of a candidate’s spoken English.
12In assessing spoken language quality, the Agrégation (as with most exams) makes use of models of reference when it comes to language form and use. These models – and the necessary sets of prescriptions and proscriptions that accompany them – are transmitted, to a greater or lesser extent, through a variety of channels: the concours syllabus (programme), classes aimed at preparing candidates, pedagogical resources, and a document that forms an integral part of French competitive exams: the rapports du jury (examiner reports).
13The rapports du jury are produced each year by the Agrégation jury and have two main aims: 1) to explain what was expected of candidates in the different exams that make up the concours, 2) to highlight what was done particularly well (or not) by candidates that particular year. As one might imagine, these documents are often perused by future candidates as they function as a “good practice guide”, giving valuable insight into how the different exams take place and what candidates are expected to do or not to do.
14The dos and don’ts contained within the rapports range from practical advice (using certain types of ink, for example) to the identification of common analytical pitfalls when commenting upon texts. They also extend to the form and use of the English language, in terms of both the metalinguistic knowledge that candidates should be able to demonstrate and, crucially, the ways in which candidates are expected to speak and write in English in the context of the concours (as will be explored in depth below). Clearly then, these documents can be seen as key players in the elaboration and transmission of reference models of the English language within the Agrégation. They thus are presented as an ideal basis for a case study aiming to explore the ideological construction of English in this context.
15For this first exploratory study, a small corpus was put together by collating the rapports du jury from both the Agrégation externe and the Agrégation interne. These documents are made publicly available by the French government and are indexed on the website of the Société des Anglicistes de l’Enseignement du Supérieur (or SAES)11. Given that the SAES indexes the rapports from 2010 until 2022 (at the time of writing), this time period was adopted for this first exploratory study, defining a corpus made up of 26 rapports du jury: 13 ‘external’ and 13 ‘internal’.
16In keeping with the exploratory spirit, I have chosen to focus in on parts of the rapports that relate to oral (as opposed to written) English. The reason for this is the importance given to spoken English, not just as a key component of the concours, but as a symbol of the mastery of the language more generally, and as a pointer as to whether a candidate might be considered as a good model for students and pupils, as shown in this extract from the rapport relating to the quality of spoken English during the Agrégation externe of 2015:
- 12 For all the data extracts used in this article, all emphasis and all translations into English are (...)
Extract 1. Rapport du jury – Agrégation ext. 2015 (p.131)12
Le candidat doit être conscient du fait que la note de langue est très importante. Tout d’abord, parce qu’une bonne note en langue apporte des points précieux pour la réussite au concours. Ensuite, parce qu’elle est le signe que le futur agrégé maîtrise la langue qu’il va par la suite enseigner. […] : ce candidat est-il un bon modèle pour des élèves ?
Candidates should be aware that the language mark is very important. Firstly, because a good mark in language contributes valuable points when it comes to succeeding in the competitive exam. Secondly, because it is a sign that the future agrégé has mastered the language that they will be teaching. […]: Is the candidate a good model for students?
17Given this focus on spoken English, two sections of each rapport are of particular interest. The first key section is that which deals with the phonology/phonetics part of the linguistic analysis that forms part of the written exams (see above), where candidates must demonstrate their knowledge of, and ability to describe, the phonetic and/or phonological system(s) of English. The second key section is one in which the jury gives feedback on candidates’ spoken English. The jury provides pointers as to what is considered as good spoken English (or the opposite) in the context of the concours, ranging from detailed phonetic observations to advice regarding the gestural, prosodic or discursive elements of public speaking in exams like this one (numerous examples follow below).
18The main objective of the analysis that follows is to explore the language ideologies present in, and disseminated by, these parts of the rapports. This is done by assembling analytical tools from discourse analysis, sociolinguistics and linguistic anthropology – and especially their critical (Blommaert; Heller, Éléments d’une sociolinguistique critique) and/or political (Canut et al.) incarnations – into a critical apparatus that allows for a discursive approach to the study of language ideologies. In what is a resolutely qualitative approach, I undertook a thematic analysis through close reading of the rapports, identifying and labelling passages that included “epilinguistic discourse” (Canut) or other discursive tropes that might be considered as traces or markers of language ideologies. Each labelled passage was then interpreted through the analytical lens mentioned above, allowing each instance of epilinguistic discourse to be related to a particular language ideology (or ideologies). These interpretations were then set against findings from previous research in the field, bringing to light the ways in which they contribute to defining what English “is” in this context.
19It should be mentioned at this stage that neither this article, nor its author, are positioned primarily within the field of language didactics. Rather, this research deals with the Agrégation, and the teaching/learning of English in France more generally, as a sociolinguistic field. While the conclusions drawn may have didactic and pedagogical implications, this will be the object of future research.
20In what follows, I highlight four principal language ideologies at play in the rapports du jury, addressing each one in turn before arguing that, together, they contribute to painting the picture of an “ideal” speaker of English and, consequently, an “ideal” model of what English should sound like. I then show how, beyond simply providing “ideals”, these dynamics contribute to defining the very essence of “English” as a language in the context of the Agrégation. I close the analysis by exploring how this particular case might bear influence on French higher education, and French society, more generally.
- 13 From a sociolinguistic (or, indeed, any other) perspective, this term is problematic, a point under (...)
21The first language ideology addressed here is what Adrian Holliday (2006) has termed “native-speakerism”. Simply defined, native-speakerism refers to a “pervasive ideology within ELT [English Language Teaching], characterized by the belief that ‘native-speaker’ teachers represent a ‘Western culture’ from which spring the ideals both of the English language and of English language teaching methodology” (Holliday, ‘Native-Speakerism’ 385). In other words, ideologies of native-speakerism position so-called “native” speakers13 as the most, or perhaps only, legitimate speakers – and thus experts – of a given language.
22In the context of the rapports du jury, native-speakerism is present and, in some cases, very explicit, as the following extract shows:
Extract 2. Rapport du jury – Agrégation ext. 2015 (p.131)
Il est […] du devoir du futur enseignant de faire tout son possible pour s’exprimer dans une langue qui soit la plus proche possible de ce qui est produit par des natifs.
It is the future teacher’s duty to do everything in their power to speak a [form of the] language that is as close as possible to that which is produced by natives
23Here, speech as produced by native speakers is clearly placed on a pedestal. While such clear native-speakerism is relatively rare, more implicit legitimising of the native speaker is widespread. An example of this phenomenon can be found in the following extract, taken from a section of the report advising candidates on how to improve their spoken English performance:
Extract 3. Rapport du jury – Agrégation ext. 2019 (p.214)
Les candidats dont la langue orale a besoin d’être améliorée sont fortement encouragés à accroître autant que possible leur exposition à l’anglais, que ce soit en effectuant des séjours en pays anglophones, en recherchant les occasions de pratiquer l’anglais avec des locuteurs natifs, ou en écoutant de l’anglais authentique par le biais d’émissions radiophoniques, de séries ou de films.
Candidates whose spoken language needs to be improved are strongly encouraged to increase their exposure to English as much as possible, whether it be through stays in English-speaking countries, looking for opportunities to practice English with native speakers, or listening to authentic English on the radio or in series or films.
- 14 The mention of “English-speaking countries” gives some idea as to where the authentic, native speak (...)
24In this extract, a direct link is drawn between improving one’s spoken language competency in English and practicing with native speakers of that same language (a theme which appears in almost all of the rapports). This implicitly positions native speakers as both ideal models of, and expert authorities on, spoken English, a clear manifestation of native-speakerism. Furthermore, this native-speakerist discourse is reinforced here by the juxtaposition of “native speaker” and “authentic English”, indexically framing native speakers of English as authentic and non-natives as inauthentic (see Lowe and Pinner)14. As has been argued elsewhere, positioning native speakers as “authentic” legitimises this group of speakers, necessarily delegitimising all other groups of speakers (Creese et al.; Wilson).
25This (de)legitimisation extends beyond speakers in that it is also applied to the language itself. As can be seen in extract 4 below, native speakers are sometimes invoked in order to justify why a certain form is considered correct (the primary stress on certain polysyllabic words in this case):
Extract 4. Rapport du jury – Agrégation ext. 2020 (p.181, emphasis in the original)
Attention néanmoins aux exceptions et à leurs dérivés (Catholic, dont les anglophones natifs tendent à contracter la seconde syllabe, mais Roman Catholicism).
Be careful nevertheless with exceptions and their derived forms (Catholic, where native English speakers tend to contract the second syllable, but Roman Catholicism).
26If summarised simplistically, the extract above argues that these particular stress patterns are to be considered correct for the simple reason that they are the patterns produced by native speakers. This is a form of native-speakerism in that the linguistic productions of native speakers are held up as the most – or sole – legitimate model of English form and use simply because they are productions of native speakers.
- 15 Indeed, even more fundamentally, actually defining who does or does not constitute a “native” speak (...)
27Though probably unnecessary in a journal such as this one, it seems worth pointing out that this legitimisation of (only) native speaker models is ideological in nature. It is not a statement of fact to say that native speakers are the sole bearers of “authentic” English; authenticity can be defined or measured in myriad different ways (see Creese et al.; Wilson for a more detailed discussion). Similarly, it is not a statement of fact to say that native speakers necessarily make the best experts or teachers of English; the reality is much more complex and multifaceted (see Braine, for instance). In sum, native speakers and their language practices are not intrinsically, inherently superior as models of language form and use when compared to non-natives15. Thus, any framing of native speakers and their productions as superior are ideologically – not linguistically – based, and study and critique of such ideologies are a long-standing pillar of work in sociolinguistics and related disciplines (e.g. Holliday, ‘Native-Speakerism’; Jenkins; Leung et al.; Pennycook, The Cultural Politics of English as an International Language; Phillipson; Muni Toke; Derivry-Plard).
28Despite extensive critique of native-speakerism, its workings still seem to be very much at play within the Agrégation, as evidenced by the illustrations above. In this context, English as produced by native speakers is repeatedly framed as the model of reference in terms of language form and use. Simply put, (real) English is English produced by native speakers.
- 16 See the following section for a detailed discussion of questions pertaining to variation in English
29As alluded to previously, defining who exactly is to be considered a “native” speaker is a sociolinguistically delicate question. However, the mention of “English-speaking countries” in extract 3 above offers one potential characteristic that might be taken into account: nationality. Indeed, references to the geographic origin of English and its varieties (and, by inference, its speakers) are present throughout the rapports du jury. For instance, in the extract below, the acceptability of different varieties of spoken English16 is illustrated with a list of examples, all of which take the form of adjectives relating to nation states or countries:
Extract 5. Rapport du jury – Agrégation int. 2019 (p.162)
Pour rappel, le jury de l’agrégation ne privilégie aucun modèle phonétique particulier (américain, britannique, irlandais, écossais, australien, etc.)
As a reminder, the Agrégation jury does not favour any particular phonetic model (American, British, Irish, Scottish, Australian, etc.)
30Representations like this one draw upon, and contribute to, ideologies that indexically position languages (or varieties) as being intrinsically linked to nations or similar geopolitical entities (see Heller, Paths to Post-Nationalism; Joseph, among others). This is a form of “banal nationalism” (Billig), the process through which everyday, mundane symbols or practices contribute to the construction of a national identity. In this case, English is indexically and ideologically positioned as being linked to – perhaps even an innate property of – the handful of nations or countries mentioned in the extract. This same list of territories returns almost systematically throughout the rapports, with the possible addition of Canada, New Zealand and/or (more rarely) South Africa. The extract below, praising candidates for their knowledge of the English language and its diversity, provides another example:
Extract 6. Rapport du jury – Agrégation int. 2015 (p.82)
Comme chaque année, le jury a pu apprécier l’excellente préparation des meilleurs candidats, qui ont su […] améliorer leurs connaissances des variétés de l’anglais actuel (Grande Bretagne et USA bien entendu, mais également Irlande, Canada, Australie, Nouvelle Zélande, Afrique du Sud, etc.).
As is the case every year, the jury was able to appreciate the excellent preparation of the best candidates, who were able to improve their knowledge of varieties of present-day English (Great Britain and the USA of course, but also Ireland, Canada, Australia, New Zealand, South Africa, etc.).
31Similar lists can be found when it comes to advising candidates which media sources they should consult in order to improve their listening and speaking skills in English:
Extract 7. Rapport du jury – Agrégation int. 2021 (p.201)
Par ailleurs, une écoute attentive et régulière des supports audio […] émanant de pays représentatifs de la sphère anglo-saxonne (Grande-Bretagne, Irlande, États-Unis, Australie, Afrique du Sud, Nouvelle Zélande, etc.) […] faciliteront grandement l’identification et la reproduction des phonèmes rencontrés.
In addition, listening attentively and regularly to audio media from countries that are representative of the English-speaking sphere (Great Britain, Ireland, United States, Australia, South Africa, New Zealand, etc.) will greatly help in identifying and reproducing the phonemes encountered.
- 17 Countries traditionally perceived as English-speaking that (usually) provide the norms of reference (...)
- 18 See Schneider (19) on this point, who notes that “[i]n representations of the “Three Circles” model (...)
32Beyond the simple fact of English being indexically linked to certain geopolitical territories, the territories themselves are also of interest. As the above extracts illustrate, this discursive framing is not only systematic, it is also restrictive and selective. Although each of the above lists finishes with “etc.”, only a handful of countries are explicitly named. This equates to repeated promotion of varieties linked to those particular countries (and non-promotion of varieties conceived as being from elsewhere). The countries in question are all “Inner Circle” (Kachru)17 territories firmly situated in the Global North (with the exception of South Africa18). The above extracts are particularly interesting, as this restrictive list is framed as an enumeration of “present-day English” (extract 6) and as “representative of the English-speaking sphere” (extract 7). In other words, the Anglophone world is set out here (and thus in the Agrégation more generally) as being represented solely by these Inner Circle, Global North territories, with all other localities sent to the periphery through the use of “etc.”.
- 19 Of course, it goes without saying that people who live in the above-mentioned territories and/or pe (...)
33Of course, language does not exist in a disembodied manner and any mention of place in relation to language necessarily evokes the speakers of the varieties that are said to be found in those places. Once again setting South Africa aside, it would seem fair to suggest that the countries in the above lists are linked in the popular imaginary, to a greater or lesser extent, with a dominant, homogenous racial and ethnic profile. More specifically, this is where one is likely to find what has been termed the “white native speaker” (Ramjattan)19. These ideological dynamics can be read through what Rosa and Flores term a “raciolinguistic perspective”. Using this perspective, these authors have shown how language and processes of racialisation have become so inextricably intertwined through history as to become “co-naturalised”, with certain ways of speaking becoming indexically linked to certain racial(ised) identities (and vice versa) (see J. Rosa; J. D. Rosa; Rosa and Flores, among many other publications). Such dynamics are necessarily intertwined with the widely-documented, multi-faceted role of colonialism in the spread of English around the world (Mufwene; Pennycook, English and the Discourses of Colonialism). In the extracts above, an indexical link is formed between what are presented as representative, legitimate forms of the English language and white-majority countries and, as a consequence, white speakers.
34It should be noted here that some rapports du jury take a different approach in this regard, employing different formulations when listing acceptable varieties when it comes to the evaluation of candidates’ spoken English:
Extract 8. Rapport du jury – Agrégation ext. 2020 (p.201)
Rappelons que toutes les variantes sont acceptées, à partir du moment où elles sont cohérentes (on pense aussi bien aux modèles du nord ou de l’ouest des Îles britanniques, qu’à ceux d’Amérique du Nord ou d’Océanie, ainsi qu’aux variétés cohérentes d’Afrique ou d’Asie)
We remind the reader that all variants are accepted, so long as they are coherent (this includes models from the north or west of the British Isles, just as one might think of those from North America or Oceania, as well as coherent varieties from Africa or Asia)
35In this case, varieties of English that potentially come from outside the Inner Circle and the Global North are acknowledged. However, this acknowledgement remains vague and implicit. No varieties or specific territories are explicitly named and the further the extract gets from the Global North, the vaguer the references become (moving from specific geographic regions to full-blown continents). Whether varieties from these vast geographic areas are veritably promoted through such vague references is a matter up for debate.
36All in all, despite the slight outlier that is extract 8, the extracts explored through this section show that national and raciolinguistic ideologies in the rapports du jury frame legitimate English as being the linguistic productions of white native speakers from territories in the Inner Circle and the Global North.
37Aside from elements pertaining to national and raciolinguistic ideologies, the final extract above (extract 8) contains two mentions of another discursive trope that plays a key role in defining legitimate English in the context of the Agrégation: coherence. In extract 8, language varieties are declared as acceptable if, and only if, they are judged to be “coherent” (or consistent). For those unfamiliar with the Agrégation (or similar concours in France), what sits behind the idea of “coherence” is perhaps not immediately obvious. The following extract, which refers to candidates’ spoken English in the oral parts of the Agrégation, provides some more context:
Extract 9. Rapport du jury – Agrégation ext. 2015 (p.149)
Tous les accents régionaux sont acceptés, sans distinction. […] Il convient toutefois de veiller à la cohérence de la variété acquise : un accent rhotique doit l’être quasiment systématiquement, de même qu'un /t/ intervocalique réalisé comme une battue alvéolaire voisée, comme il l'est dans bien des accents américains, doit rester identique d’un bout à l’autre de l’exposé.
All region accents are accepted, without distinction. However, attention should be paid to the coherence of the acquired variety: a rhotic accent must be rhotic systematically, just as an intervocalic /t/ realised as a voiced alveolar flap, as it is in a number of American accents, must remain identical from one end of the presentation to the other.
38Here, “coherence” is explicitly linked to the practice of being consistent and systematic in using phonetic features considered as salient in terms of particular varieties of English. In this respect, candidates are advised to remain within the perceived confines of one (and only one) variety of English, using salient features from the chosen dialect (and not mixing in features perceived as coming from different varieties or, presumably, other languages). For instance, based on the example above, candidates who “flap” an intervocalic /t/, typically seen as a salient feature of North American varieties, are implicitly proscribed from pronouncing this phoneme as a voiceless alveolar plosive or as a glottal stop, both of which are typically perceived as salient markers of (various forms of) British English.
39This phenomenon is perhaps best exemplified with the classic example of “MidAtlantic English”, as illustrated in the extract below:
Extract 10. Rapport du jury – Agrégation ext. 2018 (p.229)
Les fluctuations entre anglais britannique et américain (MidAtlantic English), de plus en plus fréquentes, sont à éviter dans la mesure du possible
Fluctuations between British and American English (MidAtlantic English), which are increasingly frequent, are to be avoided as much as possible
40Simply put, if a candidate adopts what is perceived as American English, features seen as characteristic of British English are to be proscribed, and vice versa. In other words, being “coherent” means banishing variants that are perceived as coming from outside of a candidate’s chosen variety of English.
41The above dynamics are an example of “linguistic purism” (Thomas; Walsh), defined as a “manifestation of a desire […] to preserve a language from, or rid it of, putative foreign elements held to be undesirable (including those originating in dialects, sociolects and styles of the same language)” (Thomas 2). In the extracts above, banishing “foreign” elements equates to ensuring that the chosen variety of English contains features perceived as being only from that variety, though this presumably also extends to eradicating elements that might be considered as “foreign” in any variety of English (such as phonetic features of, say, French, or any other language perceived as being not English).
42Appeals for such “coherence” are extremely common throughout the corpus of rapports du jury:
Extract 11. Rapport du jury – Agrégation int. 2021 (p.205)
Si des variations locales, cohérentes dans le système phonologique du candidat sont acceptées […], il faut éviter la francisation de ces phonèmes
Though local variations, [which are] coherent within the phonological system of the candidate, are accepted, the Gallicisation of these phonemes must be avoided
Extract 12. Rapport du jury – Agrégation ext. 2016 (p.129)
toutes les variétés géographiques d’accents sont acceptées, à condition que le modèle proposé soit cohérent
All geographical accent varieties are accepted, on the condition that the proposed model is coherent
43Requiring this “coherence” can be seen as an attempt to outlaw phenomena relating to language and/or variety contact: the ideal model of English is positioned as a variety that is not in any way “tainted” by traces of contact with other languages or varieties.
44This promotion of “pure” language varieties is ideological in nature and does not necessarily correspond to social reality in that traces of language contact in the language practices of English speakers around the globe are common (as they are with speakers of any language). As Hansen Edwards (3) suggests, “[d]ue to increased globalization and migration, many people may speak a mix of different Englishes and have a mixed English accent, with features from different varieties”. Indeed, even speakers that would be classified without hesitation as “native” speakers of English also display inconsistency, as shown through the prevalence of (conscious or unconscious) phenomena such as style shifting, code switching, linguistic accommodation, etc (see Ervin-Tripp; Rickford and Mcnair-Knox; Schleef, among others).
45Despite this, it is the “pure” forms of English, productions perceived as being free of any linguistic influence from outside the chosen variety, that are legitimised in the rapports du jury. To highlight the ideological nature of this positioning, one could just as easily argue, for example, that varieties of English that bear the hallmarks of language contact and/or multilingualism should be seen as superior or desirable because they index a speaker with a diverse linguistic repertoire.
46Of course, as mentioned towards the beginning of this article, one of the concerns on the mind of the Agrégation jury when evaluating candidates’ spoken English is the extent to which each candidate’s spoken English constitutes an acceptable model for pupils and students. The implication here is that said model should be, or must be, “coherent”, although this is a matter up for debate. Is teaching/learning more effective if a teacher avoids MidAtlantic English at all costs? Or less effective if a teacher’s otherwise Australian accent is peppered sporadically with features from Scottish English? These are real questions, in that research on ideologies of “coherence” are (to my knowledge) in their nascent stages, whether it be in terms of the actual language practices of English speakers or the suitability of models used in teaching/learning (but see Glain and Wilson on this latter point). Such ideologies clearly constitute a promising avenue for future research.
- 20 Multilingualism does not necessarily imply the absence of traces of language contact, but the ideal (...)
47As it stands, the purist ideologies that undergird the mentions of “coherence” in the rapports du jury contribute to the elaboration of a model of English for the Agrégation that would correspond to the linguistic productions of a monolingual speaker with no traces of language contact20. This ideal(ised) speaker is necessarily a quasi-fantasy, given the high level of language contact in much of the English-speaking world (whichever way that term is defined).
48The purist dynamics explored above are closely related to another common language ideology, that of “standard language ideology” (Lippi-Green). Following Lippi-Green (1997), standard language ideologies ascribe a superior symbolic value to the codified, homogenous varieties that serve as explicit models of reference for a given language, when compared to (all) other varieties of that language. As with the dynamics explored above, there are no linguistic grounds for suggesting that the standard form of a language is inherently superior; it is simply one variety among many that has been (ideologically) elevated to this position. This elevation is often justified through reference to the supposed superiority of the standard – in terms of, say, clarity, complexity or aesthetics – but much sociolinguistic research has deconstructed this ideological framing, highlighting how all linguistic varieties might be considered equal in linguistic terms (see Crowley; Lippi-Green; Milroy and Milroy, to name just a tiny selection).
49Within the rapports du jury, standard language ideologies are clearly at play, notably through the exclusive promotion (or rather imposition) of just two models of reference when it comes to the “phonology” section of the concours:
Extract 13. Rapport du jury – Agrégation ext. 2020 (p.37)
les variétés servant de référence dans la partie phonologie sont le Southern British English (SBE) et le General American (GA)
the varieties that serve as references in the phonology section are Southern British English (SBE) and General American (GA)
- 21 Defining either of these standard forms from a linguistic perspective with any precision is not eas (...)
50These two – and only these two – varieties of reference for the phonology section of the appear systematically throughout the entire corpus of rapports (though the glossonym used to refer to Southern British English does vary somewhat). This constitutes then a clear promotion of two of the most prestigious standard forms of the English language21.
51Such restrictive promotion of two standard(ised) forms of English would seem to sit at odds with the lived reality of English around the world. On one hand, English is a well-established pluricentric language with multiple varieties of reference around the world (Leitner) and, on the other hand, it is one of the planet’s most widely used lingua francas. Added to the forms of linguistic variation found in any language, English presents itself as a language that is almost defined by its variation and variability (see Glain).
52As seen throughout this article (see extracts 5, 8 or 9 above, for example), these different phenomena are clearly taken into account in the rapports du jury when it comes to advising candidates on the variety used when speaking English. Linguistic variation is not only acknowledged, but rather explicitly framed as something acceptable from the jury’s point of view. Indeed, other parts of the report take this further, positioning variation as something overtly desirable, as shown below:
Extract 14. Rapport du jury – Agrégation ext. 2018 (p.229)
Accents régionaux. Les nombreuses variétés d’anglais participent également de cette langue que nous aimons tous, et toutes sont non seulement acceptées mais aussi appréciées par les membres du jury
Regional accents. The numerous varieties of English also make up part of this language that we all love, and all of them are not only accepted but also appreciated by the members of the jury
Extract 15. Rapport du jury – Agrégation int. 2019 (p.155)
Les candidats à l’agrégation doivent absolument être familiarisés avec la diversité dialectale du monde anglophone
Agrégation candidates must absolutely be familiar with the dialectal diversity of the English-speaking world
53These two extracts shed light on an apparent positive framing of English language variation and “dialectal diversity” within the Agrégation. This positive framing is omnipresent in the corpus studied here, meaning that it has been present since at least 2010. However, Michari shows that this is a relatively recent development, echoing increased attention to English variation in French circles more generally, as evidenced by developments such as the PAC (Phonologie de l’Anglais Contemporain) project22.
54Despite the positive framing of variation in the rapports du jury, there are mitigating factors. In the different extracts explored in this article, all references to varieties of English in which these varieties are explicitly named are to standard(ised) forms of (pluricentric) English, rather than what might be seen as “non-standard” dialects. As alluded to previously, explicitly naming varieties (and not others) is equivalent to promoting them (and therefore not promoting others). Thus, even if the oral expression section of the rapports du jury acknowledges and positively frames varieties beyond SBE and GA, standard language ideologies remain at play.
55Overall, standard language ideologies in the rapports du jury contribute to framing codified, standard(ised) forms of the English language as the legitimate reference models for English in this context. Southern British English and General American are afforded even more legitimacy, thanks to their status as reference models in a particular part of the concours, the phonology exam.
56The sections above have outlined (some of) the language ideologies at play in the rapports du jury of the Agrégation. As mentioned in the introduction to this article, language ideologies play a role in defining how certain forms and uses come to be considered as part of a given language (or not) and are thus instrumental in the very constitution of said language. The ideologies explored here are no different in that they all contribute to the identification and elaboration of a small group of legitimate models of reference for the English language, thus shaping what English is in the context of the Agrégation.
57The different ideological linguistic dynamics explored in this article all contribute to promoting and giving value to certain forms, uses and representations of the English language (and not to others). Taken together, these ideologies paint a picture of the ideal speaker of English: a monolingual white native speaker, hailing from an Inner Circle territory in the Global North (and perhaps especially the UK or the USA), with a perfect mastery of one of the standard varieties of English and with little to no perceptible influence from other linguistic sources. Through the language ideologies at play in the rapports, such a speaker is framed as the best (perhaps only) example of an authentic, and thus acceptable or desirable, speaker of English. It follows then that the (imaginary) language practices and linguistic traits of this (imaginary) speaker constitute an ideal model of what English should sound like. This model becomes the essence of what might be defined as “English” in this context.
58As the word “imaginary” in the above paragraph suggests, such a model does not correspond to the social reality of English in most of the contexts in which the language is used. As alluded to previously, even the vast majority of so-called native speakers are unlikely to speak a version of English that ticks all of the boxes of this ideal model (as many speak a non-standard variety, for example). Despite corresponding to productions that are unlikely to be uttered by most real-life English speakers (however this term is defined), the ideal(ised) models promoted in the rapports du jury are held up as the (only) legitimate form of English in the context of the Agrégation.
59As mentioned previously, the Agrégation is an elite, competitive exam and, as such, it may be understandable that an idealised, normative model is used in order to assess candidates’ ability in the language being evaluated. The question could of course be asked as to whether the specific models put forward in the Agrégation (and analysed in this article) are the best, or only, models for fulfilling this purpose (see Glain and Wilson for a detailed discussion).
60Going beyond this, as mentioned at the beginning of the article, most students of English at university level will not go on to become Agrégation candidates. As such, the group of people who need to learn and acquire the ideal models of English form and use for the purposes of sitting the concours is very small. However, as mentioned towards the beginning of this article, the Agrégation benefits from huge prestige, and thus influence, in French education (and society) more generally. It is therefore not unreasonable to suggest that this influence might carry the Agrégation models, prescriptions and proscriptions beyond the concours itself.
61Aside from the fact that the agrégés and agrégées themselves will go on to be important linguistic models (and vectors of language ideologies) for their own students, anyone familiar with English learning/teaching at the tertiary level in France is likely to recognise the influence of the Agrégation in various places. This influence runs through course syllabi (that often mirror the disciplines present in the concours) to pedagogical resources, via teaching and evaluation practices (for instance, the way in which translation is evaluated often mimics the way in which this is done in the Agrégation). In the same way that analysis methodologies and grading techniques of the Agrégation trickle down (in more or less diluted form) into almost all English teaching in French higher education, so do the norms and models of language form and use. As such, most students and teachers of English in France – at any level and in any diploma featuring English teaching/learning – are likely to have come into contact with the ideal models of English identified here and the language ideologies that underpin them. It is not difficult to see then how the ideological construction of English elaborated by, and within, the Agrégation might come to influence the perception of what English is on a wider scale.
62This article is not in a position to offer empirical verification of the above claims. However, the first steps of a new research project that includes interviews, document collection and sociolinguistic fieldwork, as well as this author’s personal experiences in a number of different French university settings, all point to a situation in which the idealised models of English described throughout this article – and their accompanying language ideologies – permeate well beyond the Agrégation. They imbibe the whole system of teaching, learning and evaluating English in France, shaping how English is conceived of throughout French education and, consequently, throughout French life.
63Such dynamics necessarily engender sociolinguistic consequences. Some potential consequences can be traced to the fact that the prescriptions linked to the idealised reference model of English elaborated ideologically within the Agrégation proscribe many traits exhibited by many English users around the globe, whether so-called native speakers or users of English in its most prevalent social reality, as an international lingua franca (see Seidlhofer, among many others, on this topic). This is paradoxical in that it is this very social reality with which English learners (no matter what diploma they are enrolled in) are likely to be faced when using English. Seen in this light, models such as those propagated through, or by, the Agrégation prepare individuals for interaction with a tiny minority of the English-speaking world as most English speakers (however they are defined) do not speak in ways that correspond to this idealised model.
64If such a model does indeed flow through English teaching/learning in France more generally, there are myriad potential, far-reaching repercussions. To be clear, the language ideologies explored in this article legitimise only a tiny fraction of the language practices that might be considered to make up the English-speaking world, writ large. All other ways of speaking English are necessarily devalued and delegitimised. This positions as “authentically” English-speaking only those who conform to this idealised “legitimate language” (Bourdieu). In other words, conforming to the model becomes the basis for evaluation of an individual’s capacity to be “English-speaking”. The potential consequences in an academic context are clear to see: students with the ability to conform to the model obtain good grades (and everything that that entails), those that do not, do not.
65These dynamics are more or less a classic representation of what Lippi-Green (1997) calls “linguistic subordination”, where one variety (or a select group of varieties) is legitimised (at the cost of all others) and a link can be drawn between conforming with the model (and thus abandoning non-conforming practices) and success, academic or otherwise. Once again, one might argue that such discrimination is the very raison d'être of concours such as the Agrégation, it bestows elite status upon those candidates who are successful. However, if the same idealised model is indeed mobilised in other contexts, this becomes problematic in that only a tiny minority of people learning English in their lives are going to go on to become agrégés or agrégées, yet they are all subject to the same process of subordination.
66Indeed, it is difficult not to link such conclusions with what sometimes appears as a profound sense of linguistic insecurity (Labov) in France with regards to English (see Wharton and Wolstenholme, for instance), or foreign languages more generally (see Frame, for example). The ideological dynamics explored in this article render legitimate only a tiny part of what might be considered to be the English language, framing as authentically “English-speaking” only those that master the resulting restrictive, idealised model. If these models do indeed permeate through the education system, this is likely to lead to consequences, in the form of poor grades and/or negative, sometimes public, comments from teachers (see Wharton and Wolstenholme). This can translate to fear of judgement, ridicule or mockery in later life; to a feeling that one doesn’t really speak English. Within France, this latter sentiment can be linked to the widespread idea, propagated by the media, that French people speak poor English, or do not speak English at all. It is difficult to dislocate these different phenomena from ideological dynamics like those explored in this article.
67At this stage, this is conjecture, and full-blown sociolinguistic fieldwork is required to explore to what extent the idealised model identified in this article runs through the French education system (and possibly beyond), as well as to establish its real-life repercussions and effects. Further research in this direction would seem to be of primordial importance, given how sociolinguistic research has shown that, throughout history, linguistic insecurity can have profound social and psychological effects. These range from self-censorship – a non-negligible issue given how important it might be considered to have a voice in what is increasingly positioned as the world’s global tongue – to influencing the professional, academic and personal avenues individuals choose, or choose not, to explore.
68In this article, I hope to have shown how different language ideologies manifest themselves in the rapports du jury de l’Agrégation d’anglais and how they contribute to the construction of the (socio)linguistic object named “English” in this context. Native-speakerism, national and raciolinguistic ideologies, ideologies of coherence and linguistic purism, and standard language ideologies have been shown to play a role in shaping a restricted set of ideal(ised) models of English. These models frame legitimate English as that which is materialised in the linguistic productions of an (imaginary) ideal speaker: a monolingual, white native speaker from an Inner Circle, Global North territory, speaking a standard variety of English with no perceptible traces of influence from other languages or varieties. These linguistic productions are thus positioned as the very definition of authentic, legitimate English in the context of the Agrégation.
69These ideological dynamics resonate beyond the Agrégation as its status as a top-down source of authority on academic matters seems to lead to its reference models – and the language ideologies that underpin them – bleeding into the teaching/learning of English in France more generally. This might be considered problematic as the idealised model of the Agrégation finds itself at odds with the language practices of most English speakers around the world today, excluding as it does a large array of language forms and uses that might all fairly be considered as falling within the perimeter of “English” defined more broadly. As explored above, such a discrepancy between the idealised models of English in teaching/learning and the actual language practices of English speakers in the real world can potentially be linked to widespread phenomena of language subordination and linguistic insecurity.
- 23 To take an example from overseas, language diversity and sociolinguistic variation are now included (...)
70Indeed, as illustrated through the different references to previous work made throughout this article, similar ideological dynamics to those described here have been uncovered and critiqued in a number of different contexts. This is less so the case for France, and I hope that this article will constitute a humble attempt to open dialogue on these issues. Further research on similar topics in the French context would thus seem particularly fitting for two reasons. First, despite the existence of similar language ideologies elsewhere, the ideological dynamics outlined in this study are very much French in nature in that they are inherently linked to a very French context: the Agrégation. To what extent do similar ideologies manifest themselves in the teaching/learning of other additional languages in France? Or in the teaching/learning of French? How do they contribute to the ideological constitution of these languages, or to conceptions of language more generally? Second, further research on language ideologies in France opens up possibilities for the introduction of more elements pertaining to the ideological, indexical and symbolic elements of language in the teaching/learning of English or, indeed, any other language. Getting closer to a situation in which all the elements of the “total linguistic fact” (Silverstein, ‘Language and the Culture of Gender’) are part and parcel of language education – be it at the Agrégation level or elsewhere23 – opens up the possibility of identifying and challenging language ideologies and their effects, promoting awareness and potentially leading to change in the way we talk about – and conceive of – English, or language in general.