1. Introduction
1.1. Dutch in Flanders: Historical Background and Status Quaestionis
The standardization of Dutch in Flanders has been thoroughly discussed (for example, Willemyns & Daniëls Reference Willemyns and Wim2003, Vandenbussche Reference Vandenbussche, Gilles, Scharloth and Evelyn2010, Grondelaers & Van Hout Reference Grondelaers and Van Hout2011a, Jaspers & Van Hoof Reference Jaspers2013).Footnote 1 The northern, Dutch-speaking part of Belgium is traditionally considered to be a region with a delayed standardization process, compared to the Netherlands (see Van Hoof & Jaspers Reference Van Hoof and Jaspers2012). Dutch became a standard language in the Netherlands much earlier than in Flanders: Standard Dutch was established in the Netherlands from the 16th century onwards, with the bulk of the standardization taking place in the 17th and 18th century. While Standard Dutch developed in the Netherlands, Flanders was ruled by the Spanish, the Austrians, and the French, which held back the development of Standard Dutch in Flanders. In 1830, when Belgium was founded, French became the dominant and most prestigious language. After decades of language struggles, Dutch was eventually recognized as an official language alongside French in 1898, with the Gelijkheidswet or ‘Law of equality’ (Vandenbussche et al. Reference Vandenbussche, Roland, De Groof, Eline, Nils and Davies2005, Jaspers & Van Hoof Reference Jaspers2013). Flanders adopted the standard variety of the Netherlands: Northern Dutch became the linguistic ideal. However, most Flemings were unfamiliar with northern Dutch, and it took until the middle of the 20th century before large-scale initiatives were put into place to actually familiarize Flemings with it.
Today linguists and language advisors no longer consider the language situation in Flanders to be exclusively derived from the language situation in the Netherlands (Geeraerts Reference Geeraerts, De Caluwe, Geeraerts, Kroon, Mamadouh, Soetaert, Top and Vallen2002). The language variety used on the Flemish public-service broadcasting station VRT (Vlaamse Radio- en Televisieomroeporganisatie, ‘Flemish Radio and Television Broadcasting Organisation’)—often referred to as VRT-Dutch—has taken over the position of Netherlandic Dutch as the standard (van der Sijs & Willemyns Reference Van der Sijs and Willemyns2009). Now that Belgian Dutch is following its own course, Dutch can be considered a pluricentric language (Geerts Reference Geerts1992, Deprez Reference Deprez1997, Hendrickx Reference Hendrickx1998, Martin Reference Martin, Els, Hendrickx, Martin, Smessaert, Van Belle and Joop2010, De Caluwe Reference De Caluwe and Muhr2012a, Reference De Caluwe, Matthias, Vogl and Molinerb), in Clyne's (Reference Clyne1992b:1) sense that it is a language “with several interacting centers, each providing a national variety with at least some of its own (codified) norms.” One national variety is spoken in the Netherlands, while the other is spoken in Flanders.
Do Flemings in 2016 actually master Standard Dutch, and is the variety spoken and written? That is not an easy question to answer. A distinction needs to be made between passive knowledge of Standard Dutch and active use of the variety. Flemings have passive knowledge of Standard Dutch and are capable of understanding the variety when they read or hear it (Impe Reference Impe2010). Concerning the active use of Standard Dutch, a difference can be observed between spoken and written language. Most Flemings can write Standard Dutch, but the standard variety is rarely spoken. Grondelaers & Van Hout (Reference Grondelaers and Van Hout2011a:218) even refer to it as “an almost unattainable ideal achieved only by a small minority of Dutch-speaking Belgians in a limited number of contexts” (see also Geeraerts Reference Geeraerts2001, Plevoets Reference Plevoets2008, De Caluwe Reference De Caluwe2009).
In reality, Flemings predominantly use nonstandard language. They mostly speak tussentaal (literally ‘interlanguage’ or ‘in-between-language’), an umbrella term for the extensive array of intermediate language varieties in between (Belgian) Standard Dutch and the dialects (Absillis et al. Reference Absillis, Jaspers and Van Hoof2012b). Tussentaal is increasingly used: it is not only dominantly present in daily (private) life (see, among others, Vandekerckhove Reference Vandekerckhove, De Caluwe and Van2004, Plevoets Reference Plevoets2008), but also in public domains: Tussentaal features are used by politicians (Van Laere Reference Van Houtte2003, Auman Reference Auman2009) and teachers (Delarue Reference Delarue2013, Reference Delarue2014), in TV and radio commercials (Van Gijsel et al. Reference Van der Horst2008) and in TV fiction (Van Hoof Reference Van Bezooijen2013). Alongside tussentaal, dialects are also spoken. However, while Flanders used to be characterized by a striking dialect diversity, nowadays dialects seem to be doomed to disappear eventually. Research has shown that Flemish dialects are suffering from functional loss, that is, dialects are decreasingly used as a means of communication, and from structural loss or dialect levelling, that is, the dialects lose (a part of) their typical characteristics (see Taeldeman Reference Taeldeman, Herman and Adams1991, Reference Taeldeman2005; Vandekerckhove Reference Vandekerckhove2009; De Caluwe & Van Renterghem Reference De Caluwe and Van Renterghem2011).
1.2. Language Debates in Present-Day Flanders
Flemish teachers have an important task when it comes to (standard) language use and norm reproduction: As “the first-line dispensers of standard usage” (Grondelaers & Van Hout Reference Grondelaers and Van Hout2012:48), they are expected to adhere to the standard language variety as closely as possible, setting a linguistic example for their pupils (Delarue Reference Delarue2013). With epithets such as “guardians of the standard language” (Van de Velde & Houtermans Reference Van de Velde, Houtermans, Erica and Weltens1999) and “the last gate-keepers of the standard” (Van Istendael Reference Van der Horst2008:31), teachers are under a considerable pressure to master, use, and support the standard variety. This pressure is prominent in Flemish language-in-education policies (Vandenbroucke Reference Vandenbroucke2007, Smet Reference Smet2011), where the use of the standard variety inside as well as outside the classroom is strongly advocated; it is considered to be a conditio sine qua non for a successful school career, participation in Flemish society, and socio-economic promotion. In other words, Standard Dutch is essential for eliminating social inequality (Delarue & De Caluwe Reference Delarue and De Caluwe2015). However, this egalitarian aspiration of Flemish education (see Hirtt et al. Reference Hirtt, Ides and De Zutter2007) actually is rather meritocratic in nature, in that it reproduces social statuses and provides access to elite linguistic (that is, Standard Dutch) forms.Footnote 2
The strong (but often symbolic, see Blommaert Reference Blommaert2011) role of language in Flanders also causes heated discussions in the broader societal context, regularly stirring up controversy. This is especially true when the language used in public institutions—such as the media or education—comes under scrutiny, the general public lashes out at media figures or teachers who do not properly adhere to the standard variety. A case in point is a dispute that erupted in August 2012, following the publication of a book, in which tussentaal was not rejected, but was instead discussed in a neutral, nondenouncing way (Absillis et al. Reference Absillis, Jaspers and Van Hoof.2012a). During the days following the book's publication, this neutrality was reinterpreted as a positive take on nonstandard language use. The debate focused on its ubiquity (“We spreken allemaal wel eens tussentaal” [We all speak tussentaal sometimes], De Standaard, August 30, Reference De Caluwe and Muhr2012, p. 7) or added value (“Dialect verkleint de kloof met de gewone mens” [Dialect bridges the chasm with the common man], De Morgen, August 31, Reference De Caluwe and Muhr2012, p. 10).
The chapters on education were scrutinized in particular, as they stated that teachers did not need to adhere to Standard Dutch at all times, leaving room for the use of nonstandard features. This apparently controversial statement was discussed in the Flemish newspaper De Morgen under the headline “Tussentaal in klas is heel efficient” [Tussentaal is very efficient in the classroom] (August 29, 2012, p. 4). It caused a significant upheaval and many angry letters from agitated readers, who criticized teachers for speaking tussentaal, rather than “proper” Standard Dutch. The discussion dominated the Flemish newspapers for days and even weeks afterwards, proving the ideological sensitivity of the standardness issue in Flemish (institutional) contexts, especially in the media (Van Hoof Reference Van Bezooijen2013) and in education (Blommaert & Van Avermaet Reference Blommaert and Van Avermaet2008, Delarue Reference Delarue2011).
A second example dates from November 2014, when the newspaper De Standaard published the results of the large-scale language study Hoe Vlaams is uw Nederlands? ‘How Flemish is your Dutch?’ of over 3,000 Flemish language professionals, such as actors, lawyers, journalists, teachers, and linguists (De Schryver Reference De Schryver2015). In the study, participants were asked to assess the standardness of sentences containing a few typical Flemish words or constructions (as opposed to words that are used and accepted as Standard Dutch in both the Netherlands and Flanders), by answering the question, “Do you think the following sentence is acceptable in the standard, for example, in the newspaper or the news?” The results showed that 58% of these language professionals did not object to the presence of such Flemish words or constructions in genres typically reserved for Standard Dutch.
These results led—yet again—to a steady stream of newspaper headlines and opinion pieces: “More ‘Flemish’ Dutch no longer a taboo” (standaard.be, November 3, 2014), “Mutualiteit, vuilbak and autostrade? Should be possible!”Footnote 3 (deredactie.be, November 3, 2014), “Language test: Standard Dutch is considerably Flemish” (deredactie.be, November 7, 2014), and “Stop cooing about Flemish!” (De Standaard, November 8, Reference Delarue2014).Footnote 4 When the results were broken down by profession, teachers appeared to favor more “Flemish” lexical and grammatical elements in the standard than members of other language professions. Although the approval rate of teachers (61,4%) did not differ that much from the mean score of all the informants (58%), teachers were almost immediately scorned on Twitter and other social media platforms. Not only was their status as “guardians” and “gate-keepers” of the standard questioned, some even started doubting the professionalism of the teachers. In several (televised) debates and opinion pieces, teachers were called “lax,” “not language professionals,” “language amateurs,” “sloppy,” and “ignorant.”Footnote 5
These examples clearly show that Flemish teachers are under severe pressure to be standard language authorities: Both language-in-education policy and Flemish society expect teachers to strictly adhere to Standard Dutch, and even the slightest (perceived) deviation from that norm is enough to trigger feelings of disapproval and condemnation. In this paper, we want to shed light on how Flemish teachers deal with this pressure to adhere to Standard Dutch, and how it impacts their identities as teachers. How do teachers envision the role of Standard Dutch in the classroom? Do they feel the use of the standard variety is important while teaching, and/or is there any room for nonstandard varieties or features? What indexes do Flemish teachers attach to the standard variety?
Finding an answer to these questions is interesting and relevant for at least two reasons. First, theories of language and identity are very relevant for research in educational settings, as stressed by Norton (Reference Norton, Hornberger and Sandra2010:364):
Teachers' conceptions of “language” (…) are broad in scope. The teachers conceive of language not only as a linguistic system, but as a social practice in which experiences are organised and identities negotiated.
How teachers perceive language and the usefulness of different language varieties in classroom settings can therefore have major consequences: “There is recognition that if learners are not invested in the language practices of the classroom, learning outcomes are limited, and educational inequities perpetuated” (ibid.).
Second, earlier studies (Delarue Reference Delarue2011, Reference Delarue2012; De Caluwe Reference De Caluwe, Matthias, Vogl and Moliner2012b; Delarue & De Caluwe Reference Delarue and De Caluwe2015; Jaspers Reference Jaspers2015) have shown a large gap between language policy and linguistic practice in Flemish educational settings. In spite of the strictly monolingual language-in-education policy in Flanders, teachers seem to be taking liberties with those strict norms: Nonstandard features are used quite frequently in Flemish classrooms (Delarue Reference Delarue2011, Reference Delarue2013). The diverse ways in which Flemish teachers discursively bridge this gap between policy (being “standard language guardians”) and practice (using nonstandard variants when teaching) are very revealing of how teachers shape their identities. In this paper, we use semistructured interview data from 82 Flemish primary and secondary school teachers (conducted by the first author of this paper). We analyze their discourses on the value and usefulness of Standard Dutch and other language varieties in classroom settings.
In order to do so, we first need to clarify a few key concepts, which have been extensively used within sociolinguistics, linguistic anthropology, discourse analysis, and social psychology, but have proven to be quite “slippery”. Section 2 therefore discusses widely used, yet diversely conceived, concepts such as (standard) language ideology, identity (or rather identities, in plural, see Bucholtz & Hall Reference Bucholtz, Hall and Alessandro2004:381–382), agency, and indexicality. All these concepts are of importance here because they work together in the creation of teacher identities. In section 3, we outline our methodology and data analysis. In section 4, we use this framework to analyze and discuss a number of illustrative interview extracts. Our goal is to gain more insight into (i) how Flemish teachers perceive (the importance of) Standard Dutch and other, nonstandard varieties of Dutch, and (ii) how these perceptions shape teacher identities of authenticity, authority, and professionalism. Section 5 offers some conclusions and a discussion of the analysis.
2. On Standard Language Ideology and Identity
2.1. Standard Language, Standardization, and Language Ideology
The way we speak and write about language is a product of the culture we are part of, and that culture is imbued with language ideology (Bourdieu Reference Bourdieu and Thompson1991, Agha Reference Agha2003), a complex concept associated with different meanings. Following Gal Reference Gal2006:163, we define language ideologies (in the plural) as follows:
[t]hose cultural presuppositions and metalinguistic notions that name, frame and evaluate linguistic practices, linking them to the political, moral and aesthetic positions of the speakers, and to the institutions that support those positions and practices.
In other words, specific language features entail specific presuppositions about the users of those language features (concerning the level of education, social class, etc.), hierarchizing those language features and varieties.
In a Flemish context, language ideologies are very important. As stressed in Blommaert & Verschueren Reference Blommaert, Verschueren, Schieffelin, Woolard and Kroskrity1998, language was an important nationalist motive in the battle for Dutch language rights in Flanders, and, as outlined above, discussions on language policy and language use in public domains (like the media and education) continue to stir up debate. The specific language ideology considered influential in Flanders, is the standard language ideology (SLI), an ideology which we, following Swann et al. Reference Swann, Ana and Rajend2004:296, define as follows:
[a] metalinguistically articulated and culturally dominant belief that there is only one correct way of speaking (i.e. the standard language). The SLI leads to a general intolerance towards linguistic variation, and non-standard varieties in particular are regarded as ‘undesirable’ and ‘deviant’.
Many western languages and cultures know such a linguistic climate (Milroy Reference Milroy2001), generally instigated by Humanism and Renaissance in the 16th century, and the Enlightenment and Romanticism from the 17th century onwards (Bauman & Briggs Reference Bauman and Briggs2003).
In Flanders, the standard language ideology has been very influential in the standardization of Dutch (see section 1.1). According to Jaspers & Van Hoof Reference Jaspers2013, Flanders even knew a period of hyperstandardization from 1950 until 1980, “involving a fiercely propagandistic, large-scale, extensively broadcasted, scientifically supported and enduring ideologisation of language use in all corners of Flemish society” (2013:332). During that period of extreme linguistic purification, an almost complete assimilation to the northern standard norm was pursued (except for pronunciation, where divergence from Northern Dutch was allowed). The Flemish media contributed actively to this massive propaganda by giving linguists the opportunity to address their audience and spread their views. Almost every newspaper in Flanders had a daily column to help Flemings gain proficiency in the northern standard language, and radio and television channels broadcasted language-related programs. With Standard Dutch being part of the mission of the Dutch-speaking public broadcasting channel VRT, linguists kept close control over the language used by presenters of radio and television programs, and programs dedicated specifically to Standard Dutch were broadcasted in prime time (Van Hoof Reference Van Bezooijen2013). In schools, Standard Dutch was heavily promoted as well, by the means of so-called ABN kernen ‘ABN clubs’, youth clubs, where the main “good deed” was to fanatically spread the use of Standard Dutch.Footnote 6 As Willemyns (Reference Willemyns, Christian and Salverda2013) indicates, these youngsters, after becoming parents, started to socialize their children into Standard Dutch and paved the way for the massive wave of dialect loss that was soon noticed. The Flemish media and schools were thus the two main public institutions where Standard Dutch was enforced and reproduced.
The standard language ideology has led to specific perceptions of and attitudes toward Standard Dutch and toward speakers of Standard Dutch in Flanders. Standard Dutch is evaluated as correct, superior, and civilized (see Lybaert Reference Lybaert2014a), and speakers typically score highly for status- and prestige-related characteristics, with “power”, “superiority”, and “influence” as key concepts (for example, Deprez Reference Deprez1981, Vandekerckhove Reference Vandekerckhove2000, Van Bezooijen Reference Van Bezooijen2004, Impe & Speelman Reference Impe and Speelman2007, Ghyselen Reference Ghyselen2010, Grondelaers & Van Hout Reference Grondelaers and Van Hout2010). Standard Dutch is perceived as the most appropriate variety for (very) formal and prestigious situations (Lybaert Reference Lybaert2014a), as a symbol of intelligence and schooling, and as a variety that must be taught and requires an amount of effort (Geerts et al. Reference Geerts, Nootens, Van den Broeck, Guido and Hagen1980, Impe & Speelman Reference Impe and Speelman2007, Ghyselen Reference Ghyselen2010, Lybaert Reference Lybaert2014a, Reference Lybaert2015).
Recent societal changes such as informalization, democratization (Fairclough Reference Fairclough1992), globalization, immigration, and feelings of antiauthority—changes, which are typical of what Giddens (Reference Giddens1991) refers to as the present-day era of “Late Modernity”—influence the position of the standard language and standard language ideology in Europe (compare Deumert & Vandenbussche Reference Deumert and Vandenbussche2003), including in Flanders. The research group Standard Language Ideology in Contemporary Europe (SLICE; Kristiansen & Coupland 2011, Kristiansen & Grondelaers 2013), which focuses on variation in spoken language use in a European context, distinguishes between two possible scenarios for changes in the standard language. The first is demotization, whereby “the ‘standard ideology’ as such stays intact, while the valorisation of ways of speaking changes” (Coupland & Kristiansen Reference Coupland, Kristiansen, Kristiansen and Coupland2011:28). The second is destandardization, whereby “the established standard language loses its position as the one and only ‘best language’” (Coupland & Kristiansen Reference Coupland, Kristiansen, Kristiansen and Coupland2011:28).
In regard to Flanders, the dominance of tussentaal and the limited use of standard Dutch seem to be indicative of changing language standards (De Caluwe Reference De Caluwe, Johan De and Devos2006, Grondelaers & Van Hout Reference Grondelaers and Roeland2011b). However, the (future) status of the standard language ideal and the position of tussentaal are subject to much debate. Grondelaers et al. (Reference Grondelaers and Van Hout2011) consider the destandardization scenario to be the most probable (see also Grondelaers & Van Hout Reference Grondelaers and Van Hout2011a). They say that the present Flemish language situation is best described as a “standard language vacuum” because the uniform, Northern Dutch-inspired standard language is hardly ever spoken in reality, and therefore cannot be considered the norm. For Grondelaers & Van Hout (Reference Grondelaers and Van Hout2011a:219), the “highest” nonvirtual level of Standard Belgian Dutch can be equated with the speech of Belgian teachers, a form of standard Dutch with a regional accent. However, speaker evaluation experiments have shown that none of the regional accents were evaluated as the most prestigious by all the informants, and tussentaal was not evaluated in positive terms either. This has led Grondelaers & Van Hout (Reference Grondelaers and Van Hout2011a:236) to conclude that “[t]here is […] no vital standard variety of Belgian Dutch either from the production or from the perception point of view” (compare Willemyns Reference Willemyns, Christian and Salverda2007, Van der Horst Reference Van der Horst2008, Grondelaers et al. Reference Grondelaers and Van Hout2011).
According to Van Hoof Reference Van Bezooijen2013, however, the destandardization scenario is unlikely. Her research on language use in TV fiction shows a movement toward less dialect and Standard Dutch, and more tussentaal. This evolution should be considered as a sign of a “mixed ideological field” (compare the terminology of Coupland & Kristiansen Reference Kristiansen and Nikolas2011): an area of tension between the established standard language ideology and an alternative behavioral ideology, in which nonstandard language forms are gaining prestige in some contexts (Van Hoof Reference Van Bezooijen2013:104–106). According to Van Hoof Reference Van Bezooijen2013, this area of tension exhibits signs of late or high standardization, by analogy with more large-scale sociological analyses of modernity (Giddens Reference Giddens1991, Bauman 2000). The standard language will encounter more and more pressure, and Van Hoof (Reference Van Bezooijen2013) observes a reconfiguration of indexical values: Some indexical associations are gaining strength, while others are weakening. Standardization and vernacularization tendencies have always been operative, but since the nineties they seem to be more complex than before.
Research by Lybaert (Reference Lybaert2014a, Reference Lybaert2015) on the reported perceptions and attitudes of Flemings toward supraregional language variation in Flanders shows comparable results. While Standard Dutch is still considered the superior variety, a large number of the informants do not think this variety should be spoken in many situations. Instead, nonstandard language variation is considered appropriate for more informal situations; in formal and public situations, speakers' attempts to use less dialect and more standard features are valued highly. Lybaert (Reference Lybaert2014a:157) has called this an “ideology of situational diglossia”: In more formal and/or public situations, speakers need to “make an effort” by using (intended) Standard Dutch, whereas they can “talk like they are used to” in informal and/or private situations.
Finally, Grondelaers & Speelman (Reference Grondelaers and Speelman2013) also formulated a hypothesis on the evolution of language ideologies in Flanders. By analogy with the language situation in Denmark (see Kristiansen Reference Kristiansen2001, Grondelaers & Kristiansen Reference Kristiansen2013), Grondelaers & Speelman (Reference Grondelaers and Speelman2013:184) think the dominance of tussentaal in Flanders is supported by progressive ideologies with dynamism as a key concept: “Tussentaal speakers (know they) are perceived as trendy and assertive by their fellow speakers.” In a matched-guise experiment, two audio fragments with some lexical tussentaal features were considered to be dynamic by the informants. On the basis of the research results, Grondelaers & Speelman (Reference Grondelaers and Speelman2013:184–185) conclude that two ideological systems can be distinguished in Flanders:
We propose that the core of both is the conservative standard language ideology, and that this ideology exists in a public and a private format. Whereas the public ideology is for the most part common knowledge—albeit at different levels of specificity—the private version is more variable because it is entrenched in, and informed by personal value systems which pertain to, among others, matters of identity (‘to what extent do I regard myself as Belgian or Flemish, as Dutch-speaking or Flemish-speaking?’), conformity (‘what is the distance between what I know I should do and what I want to do?’), and comfort (‘what is the distance between what I know I should do and what I am comfortable with?). The answer to these questions determines how close private ideologies are to the public version.
Grondelaers & Speelman (Reference Grondelaers and Speelman2013) thus make a distinction between a public and a private version of the traditional standard language ideology, taking into account personal value systems.
2.2. Identity, Agency, and Indexicality
When referring to these personal value systems, Grondelaers & Speelman (Reference Grondelaers and Speelman2013) indicate that (private) standard language ideologies are strongly linked to identity. Their conception of identity focuses on group membership, that is, whether Dutch-speaking Belgians regard themselves as primarily Belgian or Flemish, in line with the literal meaning of identity: “sameness”. However, sameness alone is not sufficient to talk about identity construction, as identity also needs difference: “[I]dentities can only function as points of identification and attachment because of their capacity to exclude, to leave out, to render ‘outside’, abjected” (Hall Reference Hall, Paul Du, Evans and Redman2000:17–18). In that view, identities are constructed by the interplay of sameness (what we adopt from others, leading to group membership and conformity) and difference (which drives us toward autonomy and making our own choices). Sameness and difference are not mutually exclusive but are present simultaneously in every individual. However, Bucholtz & Hall (Reference Bucholtz and Hall2005:598) want to “call into question the widespread but oversimplified view of identity relations as revolving around a single axis: sameness and difference.” Identity is a far more complex and dynamic notion, they argue.
In the traditional scholarly view, identity is considered to be “housed primarily within an individual mind” (Bucholtz & Hall Reference Bucholtz and Hall2005:587), tying in with the structuralist approach to identity and subjectivity, where the subject is understood as “a real thing, with a fixed structure operating in knowable and predictable patterns” (Mansfield Reference Mansfield2000:9). Hall strongly criticizes this notion of the subject as “a self-sustaining entity” (2000:15); he adopts a poststructuralist view, in which identity is “a construction, a process never completed—always ‘in process'” (2000:16).
The complexity of identity is mirrored in the diversity of definitions of the concept. In this paper, we take on Bucholtz & Hall's (Reference Bucholtz and Hall2005:586) “deliberately broad and open-ended” definition: “Identity is the social positioning of self and other.” This definition may seem deceptively simple, but it is powerful in that it strongly implies a discursive take on identity. Instead of being a stable structure located primarily in the individual psyche or in fixed social categories, identity is considered to be a relational and sociocultural phenomenon that emerges and circulates in local discourse contexts of interaction (ibid.). In order to fully grasp the complexity of identity as a concept, we need to discuss a few aspects of it in more detail.
First, identities emerge from social interaction. Identities are not individually produced or assigned in an a priori fashion, but are interactionally emergent (Bucholtz & Hall Reference Bucholtz and Hall2005:587). In the poststructur-alist sense, identity is a discursive construct: Sameness and difference are phenomenological notions that arise from social interaction. Foucault already points out in his early “archeological” work that “not a theory of the knowing subject, but rather a theory of discursive practice” (1973:172) is required. This knowing subject no longer plays a central, unmediated role, but is produced “as an effect” through and within discourse (Hall Reference Hall, Paul Du, Evans and Redman2000:23). In other words, Foucault insists on a reversal of the subject-statement relationship: The subject has to conform to the conditions dictated by the statement before (s)he can become a speaker of it. From a Foucauldian perspective, discourse prevails over human agency. An implication of this subject-statement reversal might be that the acquisition of social identities is a process of immersion into discursive practice, and of being subjected to discursive practice. For example, becoming a teacher is a process, in which novices gradually adopt and subject themselves to the multiple modes of speaking and writing available in professional contexts.
That means that identities are not attributes of individuals or groups, but are inherent in situations: “As the product of situated social action, identities may shift and recombine to meet new circumstances” (Bucholtz & Hall Reference Bucholtz, Hall and Alessandro2004:376). At the same time, this constitution of identities through social action does not exclude the possibility that resources for identity at work in a given interaction derive from resources developed in earlier interactions (compare Bucholtz & Hall Reference Bucholtz and Hall2005:588): They may also draw on “structure” (for example, ideology, the linguistic system, or the relation between the two).Footnote 7
Second, identities are fractured, discontinuous, and partial. As a consequence of their discursive and situational nature, identities are dispersed. Bucholtz & Hall (Reference Bucholtz, Hall and Alessandro2004:374) therefore challenge the classic view of “social identities … as clearly delineated from one another, internally homogeneous, and linked to discursive linguistic practices.” Instead, identities are fragmented and discontinuous, leading to an equally fragmented social subject that is no longer “whole”, “true,” or “unique.” This view on identities as fractured (Zembylas Reference Zembylas2005:938) automatically follows from the interactional and contextual take on identity as a concept (compare Visweswaran Reference Visweswaran1994): Because identities are inherently relational, they will always be partial, produced through contextually situated and ideologically informed configurations of the self and the other. This ties in with what Bucholtz & Hall (Reference Bucholtz and Hall2005:606) refer to as the “partialness principle”:
Any given construction of identity may be in part deliberate and intentional, in part habitual and hence often less than fully conscious, in part an outcome of interactional negotiation and contestation, in part an outcome of others' perceptions and representations, and in part an effect of larger ideological processes and material structures that may become relevant to interaction. It is therefore constantly shifting both as interaction unfolds and across discourse contexts.
Third, identities are multiple. As identities are discursive, situational, and thus partial, they are also multiple. An individual can enact more than one single identity in a given (discursive) context, depending on the relationships, interactions, and identifications one chooses to foreground: language, ethnicity, gender, social class, etc. (Porto Reference Porto, Rivers and Stephanie2013:104). Furthermore, Bucholtz & Hall (Reference Bucholtz and Hall2005) stress that identities do not emerge at a single analytic level (for example, vowel quality, turn shape, code choice, or ideological structure) but operate at multiple levels simultaneously. These “varied dimensions of identity” (Bucholtz & Hall Reference Bucholtz and Hall2005:607) present a challenge for identity researchers, who can only shed light on some of these dimensions, but every partial account can be a contribution to a broader understanding of identity.
Fourth, identities are dynamic and unstable. Identification is a never completed construction that is always “in process” (Hall Reference Hall, Paul Du, Evans and Redman2000:16). As “the self is continuously constituted, never completed, never fully coherent, never completely centered securely in experience” (Zembylas Reference Zembylas2005:938), identities can also never be fixed, stable, or permanent (Alexander et al. Reference Alexander, van Wyk and Moreeng2014:406). Identities do not signal a stable core of the self that always remains the same, and is identical to itself across time: They are “constantly in the process of change and transformation” (Hall Reference Hall, Paul Du, Evans and Redman2000:17). This dynamic, unstable nature of identities follows logically from viewing identity as a “quintessentially social phenomenon” (Bucholtz & Hall Reference Bucholtz, Hall and Alessandro2004:377): People carry with them the identities they created for themselves, and these are modified and restructured by later experiences. Identities change with every social interaction, with every discursive event. Therefore, Bucholtz & Hall argue that identity needs to be approached as a relational and dynamic sociocultural phenomenon, “rather than as a stable structure located primarily in the individual psyche or in fixed social categories” (2005:586).
From this poststructuralist belief that identities are multiple, social, dynamic, and discursive, it is clear that language plays an essential role as “a carrier and shaper of individual and group identities” (Guiora Reference Guiora2005:185). Kroskrity (Reference Kroskrity2000) even defines identity solely from a linguistic perspective, as “the linguistic construction of membership in one or more social groups or categories” (2000:111). However, this definition does not imply that nonlinguistic criteria are not significant; indeed, linguistic resources are often key in defining group membership: “[A]mong the many symbolic resources available for the cultural production of identity, language is the most flexible and pervasive” (Bucholtz & Hall Reference Bucholtz, Hall and Alessandro2004:369).
These linguistic resources can come from particular languages or linguistic forms (for example, the standard or the local dialect). However, identities can also be constructed by means of communicative practices that are indexed, through members' normative use, to their group. indexicality, “the way in which linguistic forms are used to construct identity positions” (Bucholtz & Hall Reference Bucholtz and Hall2005:594), plays a key role in identity formation: Linguistic forms are semiotically linked to social meanings (Silverstein Reference Silverstein, Elizabeth and Richard1985, Ochs Reference Ochs, Alessandro and Goodwin1992). For example, standard language represents the need for a shared language and the common identity it is supposed to embody; it is called for by nationalists and advocates for nation-states in order to create a common, national identity (Kroskrity Reference Kroskrity2000).
Ideological structures, therefore, play an important role in indexicality and in identity formation, for associations between language and identity are deeply rooted in cultural beliefs and values about which (groups of) speakers can or should produce particular sorts of language (Bucholtz & Hall Reference Bucholtz and Hall2005). As was already discussed in the Introduction, the ideological sensitivity of standard language has proven to be quite vigorous in Flanders, and the social meanings attached to (perceived) nonstandard features are heavily marked. This is especially true when such features are used in (institutional) contexts where the use of the (unmarked) standard language norm is expected, such as the media and—the context we focus on in this paper—education.
2.3. Teacher Identities
In recent years, questions related to teacher identity have received significant attention within educational research (Weber & Mitchell Reference Weber and Mitchell1995, Danielewicz Reference Danielewicz2001, Miller Marsh Reference Miller Marsh2002). The poststructuralist theorizations of identity described above have contributed to the deconstruction of many assumptions about teacher identity. As a result, teacher identities have come to be regarded “as being more contingent and fragile than previously thought and thus open for re-construction” (Zembylas Reference Zembylas2005:936).
In order to gain more insight into these teacher identities, researchers need to focus on the key role of discourse (Alexander et al. Reference Alexander, van Wyk and Moreeng2014:406): Through their participation in discourses on a macro and meso level (that is, government versus school), teachers construct, modify, and deconstruct their identities. To do so, they choose among various discourses available to them, or act to resist those discourses. In this light, Zembylas (Reference Zembylas2005) introduces the Foucauldian notion of resistance, “a form of risk taking that is an important part of how teachers come to understand their professional identity” (Reio Reference Reio2005:987). According to Foucault (Reference Foucault1990:95), wherever there is domination or power, there is also resistance, and agency is derived from this interplay between power and resistance (Zembylas Reference Zembylas2005:938). From a Foucauldian perspective, teachers come to understand their identities through acts of resistance and consent (Zembylas Reference Zembylas2005:946). Resistance has several goals: It functions as a defense against vulnerability, and at the same time as an assertion of power in the face of impositions (compare Boler Reference Boler1999). In order to study teacher identities, it is therefore key to examine teachers' dominant and/or resistant discourses.
In this paper, we take this idea one step further: Instead of studying teachers' language use in the classroom (discourse), we analyze sociolinguistic interviews about their (standard) language perceptions (metadiscourse) in order to shed more light on the multiple (linguistic) identities they construct. The interview setting allows us to discursively contrast teachers' reported perceptions with other (dominant) beliefs or discourses, such as those expressed in the Flemish governmental language-in-education policy documents (Vandenbroucke Reference Vandenbroucke2007, Smet Reference Smet2011). These documents present Standard Dutch as the only language variety appropriate for school contexts, both inside and outside of the classroom.
In the interviews, teachers have the chance to reflect on these policy demands: They can resist (in the Foucauldian sense) or consent to these demands, taking into account both their own language use in the classroom (and perceptions thereof) and the high expectations (and strong sensitivities) in Flemish society concerning standard language use (see section 1.2). Both agency and structure thus play a crucial role in shaping (teacher) identities, showing that it is necessary to “undo the false dichotomy between structure and agency that has long plagued social theory” (Bucholtz & Hall Reference Bucholtz and Hall2005:607). Structure and agency are intertwined: Large-scale social structures only come into being through discursive interaction, and every single everyday interaction is impinged upon by ideological and material constructs (see Ahearn Reference Ahearn2001).
3. A Flemish Case Study: Data Collection, Methodology, Analysis
In order to gain more insight into how teacher identities can be constructed, this section presents a case study of Flemish teachers' metalinguistic behavior. Using the framework outlined in the previous section, a number of illustrative interview extracts are analyzed and discussed. Our goal is to gain more insight into how Flemish teachers perceive (the importance of) Standard Dutch and other, nonstandard varieties of Dutch, and to shed more light on how these language perceptions shape various teacher identities. After a quick look at the research methodology and data analysis in this section, we discuss three typical teacher identities that emerge from the interview data: identities of authenticity, authority, and professionalism (section 4).
For the analyses in this paper, interview data from the Corpus of Flemish Teachers' Language (CFTL) are used (see Delarue Reference Delarue2014). The CFTL corpus was compiled between October 2012 and February 2014 by the first author of this paper, and it contains speech data of 82 Flemish teachers, teaching in 21 primary and secondary schools in 10 different cities: the five Flemish province capitals (Bruges, Ghent, Antwerp, Leuven, and Hasselt), as well as five smaller regional cities (Ypres, Eeklo, Turnhout, Vilvoorde, and Beringen), which attract people from the surrounding communities for school, work, shopping, or leisure activities (see figure 1).
Map of Flanders: The ten cities included in the Corpus of Flemish Teachers' Language.
The schools participating in this study were all Catholic schools, to rule out potential differences between schools from different educational networks.Footnote 8 The teachers themselves were recruited on the basis of their willingness to co-operate; there were no specific requirements as to gender, age, region of birth, or current residence. In spite of this quite random selection, informants were distributed fairly evenly over the different demographic categories. 59% of the teachers in the corpus are female (n=48), and 41% are male (n=34).Footnote 9 The spread of the informants over different age groups is quite even as well, with 14 teachers in their 20s, 29 in their 30s, 22 in their 40s and 17 teachers older than 50.
One informant characteristic was specifically controlled for during selection: In order to investigate whether the teachers' language use was influenced by the age of the pupils in their classrooms, the teachers were divided into three groups. The first group, totaling roughly a quarter of all the teachers involved in the corpus (n=18), taught 6th grade of primary school. The other two groups were secondary school teachers, teaching 3rd grade (n=31) and 6th grade (n=33) of general secondary education (ASO or Algemeen Secundair Onderwijs). In the secondary schools included in the research, teachers of Dutch were selected, as well as teachers of other school subjects (for example, Mathematics, History, Geography, Physics), as long as the subjects were taught in Dutch.
The corpus contains two types of speech data for each of the 82 teacher-informants: lesson recordings and sociolinguistic interviews. During these (semistructured) interviews, four key topics were addressed: (i) an assessment of (the language use of) the pupils; (ii) the education and teacher training of the teacher, and the attention that was given to Standard Dutch; (iii) the knowledge of and views on language-in-education policy, both on a governmental and on a school level; (iv) the personal linguistic background and the language-related expectations for him/herself and other teachers. For the purposes of this paper, the first and the last two parts of the interviews presented us with particularly interesting perspectives.
After transcribing all of the (audio-recorded) interviews, the transcripts were coded and analyzed with the aid of NVivo software, following the principles of thematic analysis (Boyatzis Reference Boyatzis1998, Braun & Clarke Reference Braun and Clarke2006).Footnote 10 We first read and reread the transcripts, in order to become familiar with the data. Then, codes (for example, labels such as language degeneration, correctness, intelligibility) were ascribed to meaningful text units. After coding several interviews in this inductive way, the code list was compared critically with the content of the interview transcripts, to ensure the validity and reliability of the coding procedure. Codes were grouped together, and we searched for themes among them. In this paper, we discuss three of these themes, which were prevalent in the interviews, and which can be identified as discursively constructed teacher identities. To illustrate these identities of authenticity, authority, and professionalism, representative and relevant quotes were selected.
4. Three Teacher Identities: Authenticity, Authority, Professionalism
Before we turn to the discussion of three key identities found in interviews with Flemish teachers, it is important to stress that the different types of teacher identity presented here are neither exclusive nor singular. Individual teachers, as social actors, can experience the multiplicity and interactivity of these different identities (compare Kroskrity Reference Kroskrity2000:112). Although the three themes presented here are prevalent in the interview corpus, we explicitly do not want to generalize from this case study to all (Flemish) teachers. Instead, we want to identify those aspects that are in need of further exploration.
4.1. Authenticity
In the introduction to this paper, we already discussed the gap that appears to exist in Flanders between language-in-education policy (“Standard Dutch only”) and linguistic practice (the frequent use of nonstandard features in most Flemish classrooms). During the interviews, most of the teachers stress the importance of Standard Dutch in classroom contexts, but at the same time, they usually assess their own language use as an attempt to speak Standard Dutch, containing a certain number of nonstandard (that is, tussentaal/dialect) features. The teacher-informants try to justify that discrepancy by using a number of strategies (see Delarue & Van Lancker Reference Delarue and Ghyselen2016 for a more detailed analysis of these strategies). One of them is the claim that Standard Dutch limits fluency and spontaneity in the classroom. Teachers who use this strategy claim that speaking Standard Dutch does not allow them to “be themselves,” and forces them to “play a part” when teaching. Instead of emphasizing their authoritative and hierarchically superior position vis-à-vis pupils, they focus on their own personality, which needs to be reflected in their teacher identity. In extract 1, primary school teacher Frederic denounces the way some of his colleagues patronize their pupils by speaking strict Standard Dutch—for example, they use the standard form of the 2nd person singular pronoun je/jij ‘you’ instead of the more colloquial ge/gij ‘you’.Footnote 11
Frederic, M, 32, primary school teacher, GhentFootnote 12
Frederic feels that by focusing on adhering to the standard as closely as possible, teachers cannot create authentic classroom contexts and instead start “playing a part,” which makes it hard(er) for them to convey their message. It is far from coincidental that he uses the opposition between standard je/jij ‘you’ and nonstandard ge/gij ‘you’ to illustrate this: It is “one of the clearest exponents of the (nonstandard) Flemish Dutch” (Vandekerckhove Reference Vandekerckhove, De Caluwe and Van2004:981, our translation).
In her study on how laymen perceive tussentaal features, Lybaert (Reference Lybaert2014b:199) comes to the same conclusion, pointing out that the ge/gij-system is a relatively salient linguistic feature. It is intriguing to see that in this extract, Frederic rejects the use of the standard forms je/jij, but at the same time uses je/jij in his discourse: “dan ga je taal als je ook . met de ‘je’ en . dan . dan neem je precies een rol in . en ben je niet meer jezelf.” A possible explanation for this apparent contradiction could be that Frederic tries to speak more standard/formal in this interview setting than he would do in the classroom, talking to pupils. In the interview, he speaks hesitantly, with frequent pauses, which seems to support this explanation. At the end of extract 1, however, he explicitly denies any form of linguistic discomfort (“when I'm myself, I talk like I am talking to you now”), which could imply that he is either downplaying his discomfort, or that his discomfort only concerns the content of the interview, and not the language use.
In the second part of extract 1, Frederic hastens to retract his statement that speaking Standard Dutch means patronizing pupils and leads to nonauthentic classroom contexts: Authenticity can also be reached by standard-speaking teachers, but only if speaking Standard Dutch does not require any effort (compare Lybaert Reference Lybaert2014a). Otherwise, teachers have to think constantly about how they are saying something, instead of focusing on the message they want to convey, while the latter seems to be taking priority for most teachers in our corpus (see Delarue & Van Lancker accepted).
In extract 2, chemistry teacher Amber refers to the same feature as Frederic did in the previous extract: the 2nd person singular pronoun je ‘you’. To Amber, using the je/jij-form sounds “very unnatural,” and it does not allow her to teach “in an authentic way” and to stay true to herself.
Amber, F, 45, chemistry teacher, Hasselt
Note how linguistic and pedagogical elements are closely linked in teachers' interview discourses. For example, teachers such as Frederic and Amber choose to use ge ‘you’ instead of je ‘you’ because they want to come across as teachers who stand close to their pupils and have a strong connection with them, not (or not solely) because it is a feature that indexes authenticity. In light of Coupland's (Reference Coupland2003, Reference Coupland2007) model of authenticity, presupposing an automatic link between nonstandard (that is, regionally or socially colored) variants or varieties and authenticity can be seen as problematic. In his model, Coupland (Reference Coupland2003:420) distinguishes between establishment authenticities (which seem to tie in with the SLI discussed earlier) and vernacular authenticities. Footnote 13 He points out that the dominant sociolinguistic agenda is to defend the latter, while opposing the former. Important here is that Coupland's concept of establishment authenticities shows that not only vernacular, but also standard features can be seen as authentic.
The establishment and vernacular sets of authenticities entail a different view on authenticity as a concept, with a more authoritarian perspective in the former (that is, what is proper), and a more egalitarian perspective in the latter (that is, what is rightfully ours). The teachers in our corpus, however, only refer to authenticity from this egalitarian perspective, as is also clear from the two extracts above: For Amber (extract 2), being authentic means “to be who I am,” and Frederic (extract 1) stresses that once you change your language for the child (that is, switch to the standard), the authenticity goes out the window.
The question of why teachers fail to see Standard Dutch as an authentic language variety can probably be answered by looking at Flanders' (historically) problematic relationship with the exoglossic standard (see section 1.1): Instead of starting an autonomous Flemish standardization process, an exogenous norm was adopted, and Flemings were expected to learn a foreign language. To this day, Standard Dutch is generally considered as a variety only spoken in certain formal situations. Following Geeraerts Reference Geeraerts2001, this phenomenon is commonly referred to as the “best-suit” mentality of Flemish speakers: They are convinced you actually need one, but only wear it for special occasions and feel uncomfortable wearing it.
The teachers in our study struggle to identify with Standard Dutch as well. In extracts 3 and 4, two Dutch language teachers emphasize that Standard Dutch is not their mother tongue, and that, as a consequence, they do not expect their pupils to adhere to it either. For Marie (extract 3), it is downright “utopian” to expect pupils to speak Standard Dutch in all classroom contexts—even in Dutch language classes. To illustrate this, she explicitly refers to the 2nd personal singular pronoun je (similarly to the teachers in extracts 1 and 2), and to the deletion of the final t in short function words (colloquial nie ‘not’ instead of standard niet, da ‘that’ instead of standard dat).
Marie, F, 25, Dutch language teacher, Vilvoorde
Simon, M, 37, Dutch language teacher, Turnhout
Simon, another Dutch language teacher, agrees on the “foreign” nature of Standard Dutch. In the second part of extract 4, he explicitly compares speaking Standard Dutch in Flanders to speaking French with someone from Wallonia, the French-speaking part of Belgium. He states that it is impossible for him to express himself as well as he would be able to in his everyday vernacular. Note that while Marie boldly states that Standard Dutch is not her mother tongue, Simon does leave a little “wiggle room” in his statements on Standard Dutch as a foreign language: Regarding the pupils, he states that the standard “is sort of a foreign language” (but at the same time he stresses the is), and further on, the standard is “a foreign language for so many people .”
For teachers like Simon, who desire a “strong classroom dynamic,” Standard Dutch just feels too distant and detached. In extract 4, Simon identifies two criteria in creating such a “strong” classroom context: interaction (pupils need to “speak a lot,” he wants to “express himself (…) spontaneously”) and dynamism (“classroom dynamic”). Standard Dutch scores low on both criteria: Due to its (perceived) foreign character, it is ill-equipped to foster interaction. Grondelaers & Speelman's (Reference Grondelaers and Speelman2013) study mentioned in section 2.1 showed that tussentaal features score much higher on dynamism than standard features.
4.2. Authority
Teachers can also take on a more authoritative identity, through the use of linguistic markers of expertise, such as formal language or specialized jargon (Bucholtz & Hall Reference Bucholtz, Hall and Alessandro2004:386), or by adhering to the (formal) standard language as strictly as possible. In that case, Standard Dutch no longer remains the unmarked variety it is deemed to be in Flemish language-in-education policy documents (see section 1.2; Delarue & De Caluwe Reference Delarue and De Caluwe2015), but instead becomes marked as an index of authority.
However, it is important to note that, just as authenticity cannot be linked exclusively to nonstandard features, authority is not reserved for standard features only: The authority of teachers is an omnipresent feature of classroom contexts that structures classroom life and order. Moreover, teachers can choose to linguistically articulate (voice) that authority by turning to the standard variety (see Milroy & Milroy Reference Milroy and Milroy1985). In the Flemish context, with an exoglossic standard variety, authority becomes closely linked to distance: The standard can be used by teachers to discern themselves linguistically from their pupils (Delarue Reference Delarue2013:212).
In extract 5a, History teacher Sylvie states that a certain distance is needed when teaching, more specifically in situations of instruction (as opposed to “lighter” classroom situations, for example, group work): then the teacher needs to be able to “teach his class,” and pupils need to “pay attention.” In those more formal situations, Sylvie appreciates a larger distance between teacher and pupils. The use of Standard Dutch helps to index that distance, and for Sylvie it “comes in handy” to underline her authoritative identity as a teacher.
Sylvie, F, 26, History teacher, Ghent
In other school contexts, however, Sylvie worries that this distance can lead to a weakening teacher/pupil relationship. In extract 5b, she links Standard Dutch to detachment: Using the standard makes you sound “posh” and “solemn,” while in other contexts—Sylvie explicitly refers to field trips—it is important for pupils and teachers to bond with each other. Using the standard in those situations is far from beneficial: The distance indexed by Standard Dutch now is perceived as a burden, instead of an advantage (as was the case in extract 5a).
Sylvie, F, 26, History teacher, Ghent
Both extracts show that Sylvie's view on Standard Dutch is not without conflict: She wants to use the standard to express authority, thus creating a distance with her pupils, but simultaneously fears this distance, as it would not be beneficial to the classroom atmosphere.
4.3. Professionalism
A third teacher identity we want to discuss in this paper is an identity of professionalism. Teachers are perceived as education professionals (Reynolds Reference Reynolds1995), and part of that professionalism lies in selecting the appropriate language varieties and variants.Footnote 14 In that respect, professionalism proves to be an interesting concept: The teachers in our corpus not only refer to the identity of professionalism to explain why they feel the need to adhere to the standard (see extract 6 below), but also use it as a legitimation to move away from this need to speak Standard Dutch at all times (see extract 8).
Thomas, M, 34, (Catholic) religion teacher, Antwerp
In extract 6, Thomas, a religion teacher, talks about how he often runs into his own pupils after class, on the tram or in the city. When asked about his language use in those after-school contexts, he stresses that he speaks the same to his pupils as while he is teaching, stating that it is important for a teacher “to stay as professional as always” and “to be a role model” for pupils. At the end of extract 6, Thomas states that teachers need to be aware of their exemplary role (that is, speak Standard Dutch), because making an exception to that rule could be punished severely. The story about a teacher, who talked to pupils using nonstandard language would “spread like wildfire,” and he would risk losing his authority—more specifically, the authority to address pupils when their language use is inappropriate. This indicates that for Thomas, professionalism and authority are not two mutually exclusive identities, but that they are coconstitutive: Being professional means not losing your (linguistic) authority.
The connection between being professional and speaking Standard Dutch also emerges in the interviews when teachers refer to the (future) professional life of their pupils. A bit earlier in the interview, Thomas states the following:
[W]hen they start in their professional life, they need to be able to express themselves … and it's important to be able to switch over [use another language variety, that is, Standard Dutch, SD & CL] and some sort of professionalism needs to surface.Footnote 15
Related to this argument of professionalism is the need for teachers to meet linguistic expectations. During the interviews, nearly all teachers talk about how well they are aware that Standard Dutch is expected from them (see Delarue 2014, Delarue & Van Lancker Reference Delarue and Ghyselen2016), and these expectations serve as a strong incentive to adhere to the standard norm as closely as possible. In extract 7, Dutch language teacher Nathan feels that pupils expect their teachers to use Standard Dutch, pointing out that teachers serve as an example on different levels: Expertise, behavior, and language use. In their comparison of Flemish teachers' and pupils' perceptions vis-à-vis Standard Dutch, Delarue & Van Lancker (Reference Delarue and Ghyselen2016) show that most pupils indeed expect their teachers to aspire to Standard Dutch.
Nathan, M, 23, Dutch language teacher, Beringen
In a previous study (on the same interview data used here), Delarue & Ghyselen (Reference Delarue and Ghyselen2016) compared the standard language use of Flemish teachers with that of other highly educated professionals in Flanders, such as lawyers, doctors, and dentists, and concluded that teachers used significantly fewer nonstandard features in their speech. A plausible explanation for that discrepancy is that the nonteachers in the study do not feel equally pressured to adhere to this Standard Dutch norm.
However, there are also teachers in our corpus who boldly state they are not concerned by these linguistic expectations. Interestingly enough, they also use the notion of professionalism to explain why they do not feel the need to (always) use Standard Dutch in the classroom. In extract 8, for example, Gregory (a primary school teacher) indicates that being professional means that as a teacher, you can choose the language variety you address your pupils in, as well as the norm you impose on your pupils. For Gregory, “everyone has his own norm,” and these individual norms are legitimized by referring to the professionalism of teachers—actively dismissing the government's dictum that teachers always have to speak Standard Dutch. Notably, Gregory uses a substantial number of nonstandard features in this dismissal of Standard Dutch.
Gregory, M, 30, primary school teacher, Antwerp
At the end of extract 8, Gregory defines more specifically what he means by “being a professional.” For him this means to know how to “adjust your language use to the children,” in order to convey your message adequately. In other words, professionalism entails being flexible and having a hands-on mentality, and choosing the appropriate language variety—even if that choice conflicts with linguistic requirements.
5. Discussion and Conclusion
In this paper, we have tried to shed more light on three teacher identities that emerge from our corpus of interviews with Flemish primary and secondary school teachers. Frames of authenticity, authority, and professionalism are prominent themes in the interviews, and we hope to have shown how Flemish teachers discursively handle these concepts when talking about their (standard) language perceptions. In the interviews, the apparent gap between language policy and linguistic practice in Flemish educational settings (compare Delarue & De Caluwe Reference Delarue and De Caluwe2015) serves as a backdrop. On the one hand, teachers stress the importance of Standard Dutch in classroom settings. On the other hand, they are concerned that overusing Standard Dutch might damage their relationship with their pupils.
Teachers explain the default choice of Standard Dutch by referring to their teacher professionalism: They need to meet the linguistic expectations stipulated by the Flemish governmental language-in-education policy, as they serve as a linguistic example for pupils. Standard Dutch is also intentionally used (or strived for) when teachers take on an authoritative teacher identity, taking into account that the exoglossic standard variety also indexes distance between teachers and pupils. In some contexts, that distance is welcomed (for instance, in situations of instruction), but at the same time, most teachers fear that it could also hinder spontaneity and interaction. For that reason, most teachers in our interview corpus are reluctant to adhere to the standard continuously, fearing that the use of Standard Dutch would damage their relationship with their pupils. Instead, their language use in the classroom typically contains a certain number of nonstandard (that is, dialect or tussentaal) features.
To eliminate the uneasy feeling of dissonance (confirming the importance of Standard Dutch, while using vernacular in the classroom), teachers make use of an extensive array of strategies (see Delarue & Van Lancker accepted). Many of these strategies pertain to authenticity: Standard Dutch does not allow teachers to “be themselves,” and makes them—by their own account—sound “patronizing,” “posh,” and “solemn.” Moreover, many of the teachers in our corpus seem to perceive Standard Dutch as a foreign language, which makes it difficult or even impossible for them to identify with Standard Dutch.
The need to diverge from Standard Dutch can also be underpinned by feelings of—yet again—professionalism: Part of being an education professional lies in the selection of appropriate language use. For some, Standard Dutch is not (always) the most appropriate variety. Teachers should then have the freedom to choose their own norm, actively dismissing the government's insistence on Standard Dutch.
The fragments analyzed in this paper, which are illustrative of the whole of the interview corpus, appear to confirm what Lybaert (Reference Lybaert2014a:157; section 2.1) has referred to as the “ideology of situational diglossia.” There appears to be a situational dichotomy in classroom settings, where teachers (try to) use Standard Dutch in the more formal settings of giving instructions or transferring knowledge (see Delarue Reference Delarue2013), but fall back on language use containing more tussentaal or dialect features in more informal situations. However, reverting to our discussion of the concept of identity in section 2.2, it is important to note yet again that teacher identities are multiple, ambiguous, and individual. Teachers can take on several identities at once, shift between identities, or discursively (de)construct (partial accounts of) identities. Issues of teacher authority are closely linked to authenticity, legitimacy, and professionalism (compare Creese et al. Reference Creese, Blackledge and Takhi2014).
In recent sociolinguistic research, the concept of authenticity has received a particularly extensive coverage. It has been pointed out repeatedly (see, among others, Heller Reference Heller1996, Blommaert Reference Blommaert2012, Blommaert & Varis Reference Blommaert, Verschueren, Schieffelin, Woolard and Kroskrity2013) that “authenticity is a dynamic process that can no longer be viewed as fixed, or as a romantic notion of heritage that can be straightforwardly transmitted” (Creese et al. Reference Creese, Blackledge and Takhi2014:948). Instead, authenticity is negotiated locally in the classroom. For Blommaert & Varis (Reference Blommaert, Verschueren, Schieffelin, Woolard and Kroskrity2013), authenticity boils down to “enoughness:” Being authentic means you are enough of X, and not too much of Y. The same seems to be true for teachers. They feel the need to balance their language use according to the specific classroom situation. It needs to be standard enough to adhere to the demands of governmental language-in-education policy; at the same time, it has to be vernacular enough to control the distance between teacher and pupils. As these processes involve conflict and contestation, they are highly dynamic. Configurations of features and criteria of enoughness can be adjusted, reinvented, or amended (Blommaert & Varis Reference Blommaert, Verschueren, Schieffelin, Woolard and Kroskrity2013:147). As Heller (Reference Heller1996) points out, teachers need to find out when and how to deploy certain linguistic resources if they want to become legitimate teachers.
From a monolingual point of view, however, shifting between different language varieties is often interpreted as a symptom or a consequence of low language proficiency (compare Creese et al. Reference Creese, Blackledge and Takhi2014). Perhaps that negative connotation can help explain why the teachers in our corpus keep stressing the importance of Standard Dutch in the classroom, while being fully conscious of the fact they do not always use it themselves. In Delarue & Van Lancker (Reference Delarue and Ghyselen2016), we refer to these approaches as strategies, implying that teachers make carefully considered linguistic choices. Mohanty et al. (Reference Mohanty, Minati and Rashim2010:228) describe teachers as follows:
[Teachers are] not uncritical bystanders passively acquiescent of the state practice; in their own ways, they resist and contest the state policy. … It is quite clear that the agency of the teachers in the classroom makes them the final arbiter of the language education policy and its implementation.
The teachers not only explain and justify in the interviews their divergence from the governmental dictum to use Standard Dutch at all times; they also have the power and the agency to change language-in-education policy in the classroom. Reverting to the identities of authenticity and/or professionalism, as some of the teachers in the extracts cited above did, is one of the many possible strategies. The relation between identity and agency will remain a difficult question, however: How free are teachers to actually manipulate a flexible system of identities, and how strong are the influences of political-economic ideologies and other forms of social structure? By paying close attention to the ideologies at work in the classroom, one can gain understanding of the nuanced negotiations through which teachers may assume (or may be ascribed) alignments that include (versions of) certain identity positions (Creese et al. Reference Creese, Blackledge and Takhi2014:947).
In all of this, the role of language policy—both overt and covert (Shohamy Reference Shohamy2006)—cannot be underestimated. Adding to the already well-studied connection between ideology and language policy in education (for example, Tollefson Reference Tollefson1986, Ricento Reference Ricento, Thomas and Burnaby1998, Shohamy Reference Shohamy2006), and the social inequalities created and sustained by language policy (for example, Tollefson Reference Tollefson1986, Martín Rojo Reference Martín Rojo2010), Hornberger & Johnson (Reference Hornberger and Johnson2007) have introduced the “ethnography of language policy.” It can be described as a method for examining the agents, contexts, and processes across multiple layers of what Ricento & Hornberger (Reference Ricento and Hornberger1996) refer to as the “language policy onion.” In the last few years, a substantial body of ethnographic work on language policy has been published, from the role of the local in language policy (Canagarajah Reference Canagarajah2005) to the negotiation of language policy in schools (Menken & García Reference Menken and Ofelia2010). For those interested in gaining more insight in how the language perceptions of Flemish teachers relate to issues of policy and practice on the one hand, and agency and structure on the other, this ethnographical approach to language policy offers promising perspectives.
