1Evaluation has long been a research subject in Languages for Specific Purposes. In the European context, recently, Laurent Rouveyrol (2012) showed that CLES was a means to socially engage the learner. Carlos Meléndez Quero (2012) analysed Spanish certificates from the point of view of reliability, validity, authenticity, interactivity, impact and feasibility. Claire Tardieu (2009), after Albert Bandura (2007), emphasized the importance of developing a positive imaginary in the language classroom to overcome the "depression" which prevails in the French society, hence the importance of raising teachers’ awareness of learners' cognitive processes.
- 1 The Bayrou decree of 1997 reads: "An assessment of teaching and training is organised for each degr (...)
2Yet, even though teachers may fully acknowledge their positive impact on learning by establishing a climate of trust, universities eventually always make binary decisions at the end of the academic year. Individual teachers, and institutions, pass or fail students. Examination juries mitigate the impact of less-than-average marks to some limited extent, and only for those students whose marks lie above a level defined at each examination session. Binary decisions take place at all times in the educational process. Education ministries do or do not authorise universities to create degree courses. Students evaluating their courses, a compulsory practice in France since the Bayrou decree of 19971, reflect on whether learning has been successful or not. Language certificates are granted or refused. Students’ applications for mobility schemes are selected or rejected.
3On the other hand, the Common European Framework of Reference for Languages (2001), subsequently abbreviated CEFRL, which is in use in the educational system in higher education in various European States, has a dual scope, political and scientific. As a political instrument, it aims at "providing a common basis for the elaboration of language syllabuses, curriculum guidelines, examinations, textbooks, etc. across Europe" (Council of Europe 2001: 1). As a scientific instrument, it considers language learning as a continuum between two extremes of the competence spectrum which the CEFRL does not consider, namely a total absence of competence and native speaker competence, probably because such extremes do not reflect the language learning process. Six levels are defined, from A1, the lowest, to C2, the highest. The six levels can be further broken down into an indefinite number of levels to track minute progress, so that there seems to be no limit to the flexibility of the branching scheme of the CEFRL. At the heart of the CEFRL also lies the concept of partial competence, which suggests that, for a given competency (writing, reading, speaking, listening and interacting), learners may have reached a certain level of proficiency in some domains (for instance, for certain specific purposes), and a different level of proficiency in others. For example, a doctoral student may be at C1 level in writing a research paper in one given field, but find it difficult to take messages describing enquiries or problems, which is at B1 level. Such a concept of partial competence encourages a positive appreciation of learners’ achievements.
4The fact that language policy agents make binary decisions in Languages for Specific Purposes (LSP) creates tensions considering the inclusive philosophy of the CEFRL. What does failing mean if the CEFRL encourages us to see only positive outcomes? What are the relationships between success, failure and proficiency? Does proficiency automatically lead to success? Language competence is always situated in space and time, varies accordingly and, thus, is never complete. Even though the CEFRL is promoted by the State, language policy agents, notably supranational organisations, ministries, universities, teachers and learners seem to have different notions of what success and failure in LSP are. What do success and failure mean for them, and what does that entail? How do stakeholders position themselves in relation to the binary opposition between success and failure and to the CEFRL competence continuum? As failure and success are determining factors of social cohesion in Europe, the CERFL, with its positive, inclusive approach, actually represents a change of paradigm in language education, hence Jean-Claude Beacco's stance that "[t]he age of the 'political innocence' of language teaching draws to a close with the CEFRL" (2013: 11). Has such radical logic of the CEFRL been fully taken into account in LSP research and policy? What are the points of tension between language policy agents’ positioning regarding success and failure in LSP?
5A recent subject of investigation in second language acquisition, agency has been defined as "the realized capacity of people to act upon their world and not only to know about or give personal or intersubjective significance to it" (Inden 1990, quoted in Holland et al. 1998). Much attention has been devoted to how learners’ identities are co-constructed through social interaction (see, for instance, Deters et al. 2014, and Duff 2012). However, it is mainly the agency of learners which has been examined, leaving aside that of other stakeholders, however numerous and diverse they may be. Therefore, the present paper aims at filling a research gap in the perception of success and failure in LSP teaching and learning among some other language policy agents.
6For reasons of space and to limit the number of agents studied, we will focus on the French national context and on three main stakeholders, namely supranational organisations, the State and students. The overarching theoretical framework is based on Bernard Spolsky’s language management theory. Our corpus is composed of the many discursive productions of LSP agents, among which legislation, frameworks, and statements and recommendations. The analysis will show that language policy agents have dissenting views over success and failure. We will examine the political implications for success and failure in LSP of two types of English which can be encountered in the academic context, which are English as a Lingua Franca (ELF) and English as a Medium of Instruction (EMI). Regarding the rise of the ELF phenomenon, Alessia Cogo & Jennifer Jenkins (2010) argue that speakers of English may now choose not to abide by the type of native speakers’ norms commonly taught in EFL classes. Beyza Bjorkman (2011) considers that the advent of ELF and the findings of ELF research need to be taken into consideration in English for Academic Purposes (EAP) instruction. We will discuss the crucial position of another language policy agent, namely teachers, in relation to the notions of success and failure in terms of scientific and social positioning. We will conclude on the potential for LSP research to examine language policy issues at the micropolitical level, which is the level of language management agents.
7Spolsky’s management theory (2004, 2009) draws on the distinction between macro and micropolitical aspects of language policy.
- 2 There is another theoretical model of language management, developed by the Prague school (for inst (...)
8Founded on the analysis of the language policies of monolingual States, of States with two or three official languages, and of States with twelve languages or more, Spolsky’s management theory2 describes language policies in terms of agency, listing the implied stakeholders and their relationships. Spolsky’s lists eleven actors, or agents. Those most relevant to our context are the family, companies, institutions, governments, activist groups and supranational organisations.
9Studying the relationships between agents allows to better understand the dynamics of identified language policies which, at universities, are potentially numerous and sometimes conflicting, and which can be described at macro level.
10As noted above, Spolsky also draws on the distinction between macro and micropolitical aspects of language policies, the macro aspect being that of the implemented policies and the micro aspect, that of agents. University language policies are potentially inspired and carried out by several agents. Plurilingualism, for instance, is particularly promoted by the Council of Europe. Social inclusion interests both the Council of Europe (Council of Europe 2013) and the European Union, which is currently co-funding INCLUDE, a consortium operating in the field of language policy and practices in favour of the inclusion of groups at risk of exclusion, particularly migrants (INCLUDE network 2013).
11Macro-type policies morph to adjust to changing political conditions. To name but a few policies, one may cite CLIL, evaluation, intercomprehension, mobility, ICT and translation. Organisations with clearly identified political scope, such as supranational organisations, States, regions and local authorities, are not the only promoters of macro-level policies. Through papers and communications, researchers also define, refine, analyse, help create and comment on such policies and therefore. can also be considered as macro-level language management agents, when working, for instance, on such topics as Institution-Wide Language Programmes (IWLP which, in France, bears the name of the "LANSAD" sector), LSP as distinct from ESP, language certifications, and other subjects. Experts’ works also deeply influence macro policies. This is especially true for the CEFRL, and true also for other language policy themes as they are mediated by LSP teachers in various contexts and through scientific and professional events.
12To define the micro-level, we will rely on Richard Baldauf's definition, according to whom the micropolitical aspects of language policy deal with "processes, relations and dynamic activities" of individuals and groups, while macro-type policies are concerned with "fixed objects and structures" (Baldauf 2006: 153). Such distinction between micro and macro aspects of language policies, in line with Spolsky’s distinction between language policy and language management, allows studying university language policy in France from two angles, namely agents and types of policies. As language policy statements are localised in space, and time, and individuals may simultaneously belong to several categories of agents (LSP teachers being possibly, experts, researchers and parents for instance), and pursue different interests, agents' diverse views on success and failure are worth considering.
13Several definitions of ESP, thus of LSP, exist in the French context, which led the Société des Anglicistes de l'Enseignement Supérieur to establish some clarification in 2011 (Commission formation de la SAES 2011). Alain Cazade (2000) described English for Specific Purposes (ESP) as the fourth field of anglophone studies, besides literature, linguistics and civilisation. ESP has recently evolved under the pressure of several learned societies members of the SAES. In its framing document, the SAES distinguishes IWLP, ESP and didactics. IWLP refers to language teaching aimed at students in a single field of study or in groups of mixed disciplines, for instance in language centres. ESP deals with one domain, such as law or economics. Didactics, which is concerned with IWLP, ESP and also literature, civilisation and translation, lies both within and outside Cazade's fourth field (SAES 2011). It is worth noting that Michel Van der Yeught (2012) distinguishes between ESP and "anglais de spécialité". "Anglais de spécialité", he claims, has an intrinsic existence, whereas ESP is defined in relation to ever-changing professional purposes. Thus, the present study jointly concerns the didactics, IWLP, and "anglais de spécialité" branches of ESP which, itself, is part of LSP.
14Regarding CLIL, first, more than 40 definitions exist in Europe, as noted in a 2008 report of the LANQUA European project. For the sake of this paper, we will use Anna Räsänen's definition of CLIL, which encompasses Pre-CLIL, adjunct CLIL and CLIL. Pre-CLIL refers to:
[C]ourses/programmes provided systematically by subject specialists to mixed, multicultural and multilingual groups (>25 % exposure) [with] language learning expected due to exposure, but outcomes not specified; implicit aims and criteria [and] collaboration possible, but rare. (LANQUA 2008: 6)
15Adjunct CLIL is when
[L]anguage support [is] coordinated with/integrated in subject studies and takes place simultaneously, [with] joint planning between teachers and specified outcomes and criteria for both content and language. (ibidem)
16CLIL is when there is
[A] fully dual approach and full integration of language across subject teaching by subject specialist or via team teaching, with specified outcomes and criteria for both content and language. (ibid.)
17EMI may take place in Pre-CLIL, adjunct-CLIL or CLIL contexts, depending on how much language support is available and how much language and subject are integrated in teaching. As James Coleman (2006) notes, the fact that ELF increasingly diverges from standard varieties in English-speaking countries creates a particular issue in EMI, with native speaker English becoming one standard among the many to which academics are exposed while travelling in their private lives or exchanging for professional purposes, in their institutions and abroad.
18According to Danielle Colardyn and Jens Bjornavold (2004), formal learning is intentional and takes place in a structured context such as a university or a company. Non-formal learning is intentional, happens outside structured contexts and is planned by the learner. Informal learning results from daily activities, is not structured and in most cases, is non-intentional.
19We will present an analysis of the discourses of some French language management agents in a larger international context and will thus focus on international organisations, the State and trade unions. In the analysis, we examine each agent’s views on success and failure and how such views vary across the diversity of agents.
20The views of the Council of Europe on success and failure are expressed in two main instruments, the CEFRL and the guides for relating examinations to the CEFRL.
- 3 The Council of Europe makes no distinction between them on this point. See Council of Europe 2011:1 (...)
21The CEFRL recognises that competences may be innate or acquired, and that learners have different cognitive abilities. Fluency and the ability to make one’s meaning clear are evidence of functional success as a learner/user3. Success at language learning can thus be distinguished from proficiency. In spite of all the positive wording, failure exists in the CEFRL. It concerns those learners who do not engage in communication activities, with no communication strategy, inhibited, with low cognitive abilities, unsuccessful at completing tasks, uninvolved.
22Yet, the focus not just on language learning, but also on partial competence and plurilingual and pluricultural competence, make more difficult the distinction between success and failure. Partial competence refers to the ability to perform limited tasks in an LSP. Pluricultural competence may be non-linguistic. It may refer to the knowledge of the characteristics of a given culture as long as such competence or knowledge are related, directly or indirectly, to the L2 or L3 in question. Some learners may have a limited mastery of a language, or a limited ability to engage in activities in a particular domain, or in a specific task, while having some plurilingual or pluricultural competence. This challenges the construct of LSP examinations and language certifications and any binary distinction between success and failure. The potential far-reaching consequences of CEFRL on the whole educational process, and not just on curricula development, are exemplified in the subtitle: "Learning, Teaching, Assessment".
23Between 2001 and 2009, the Council of Europe developed manuals and studies to relate examinations to the CEFRL (for instance, Council of Europe 2009a, 2009b). By publishing such guides, the Council of Europe accepts its name to be linked to that of the institution who has produced the examinations and tests. Chapter 4 of one of these manuals reads:
This chapter deals with the content analysis of an examination or test in order to relate it to the CEFRL from the point of view of coverage. This might be done by discussion, or by individual analysis followed by discussion. The end product is a claim by the institution concerned of a degree of linking to the CEFRL based on Specification, profiling their examination in relation to CEFRL categories and levels. (Council of Europe, Language Policy Division 2009a: 26)
24The Council of Europe does not impose any control over the claim that the tests or examinations relate to the CEFRL. Any organisation can claim its examinations and tests to be CEFRL-aligned.
25The preface to Relating Language Examinations to the Common European Framework of Reference for Languages: Learning, Teaching, Assessment (CEFRL). A Manual raises several issues:
it indicates that the manual accepts different constructs, examination goals, and conceptions of proficiency;
it does not certify that examinations established following the recommendations are linked to the CEFRL;
it recommends turning to national and international testing agencies for quality standards, and to the scientific literature.
26Failure at a language examination or a language test established along the lines of the CEFRL is thus perfectly possible. In theory also, the CEFRL may be linked to language tests whose ideological underpinnings are not that of the Council of Europe. As Richard Young (2006) noticed, tests may be concerned with validity and reliability, but often fail to deal with equity. In other words, tests and examinations reflect ideologies of what good language and good command of a culture are, however situated in time and space such conceptions might be. Research has recently addressed the issue of developing inclusive tests which would conform to the goals of the Council of Europe (see, for instance, Shohamy 2012 and Jenkins 2007). By accepting all types of constructs, the guides for relating examinations to the CEFRL also address tests whose ideologies contradict its aims, with the idea of overcoming such ideological differences.
27With its inclusive logic, the Council of Europe thus tolerates that some institutions may use its authority to claim that their tests comply with European standards. The paradox is that some test-takers with partial linguistic and pluricultural competences, thus some proficiency, may fail tests that claim to be CEFRL-aligned and thus, the name of the very institution which promotes social cohesion can be associated with some learners’ exclusion.
28The decree in force since August 1st, 2011 specifies that the training should:
make[s] sure that students acquire a diverse mix of knowledge and skills, among which (…) language skills, notably the ability to read, write and express oneself in at least one foreign language.
- 4 The competences to be acquired at Bachelor’s level in a law degree comprise "Understanding a conver (...)
29There is no indication of any progress to be expected, or of any target level at the end of the Bachelor's degree. The 2011 decree is completed by a brochure entitled Référentiel de compétences en licence (Conseil National de l'Enseignement et de la Recherche 2012) which, for Bachelor’s degrees in Law and Economics, contains for the first time a reference to LSP4. Here again, there is no mention of any target level for higher and lower achievers, of progress and failure, let alone of European levels. A learner may well have completed a Bachelor’s degree feeling that language provisions at university have not been sufficient to maintain a level acquired at the end of secondary school, and thus, may consider that language learning has been unsuccessful at university, but successful at secondary school. On the other hand, the learner with an above-standard level after high school, a language input at university insufficient to maintain such initial level, but who has benefited from a mobility scheme, may consider formal language learning at university unsuccessful, but informal and non-formal language learning, in the framework of a mobility scheme, highly successful.
30Success can thus be distinguished from proficiency, which is the ability to interact with other speakers, both native and non-native, in the real world rather than in the world of language tests (Lantolf & Frawley 1988). The French ministry of Education and Research distinguishes objective learners’ academic success, language proficiency in classroom settings, and the subjective perception of success at language learning. This is set out particularly in article 9 of the decree, which makes it compulsory for institutions to facilitate transfer between fields of study and institutions and to accommodate students’ diverse learning histories. By diversifying the delivery of education and training through the development of international training and mobility schemes, and through "internationalisation-at-home" (Crowther et al. 2000, Wächter 2003), the State has contributed to shift the locus of control of success in LSP from the language classroom to the real world.
- 5 "The Master's degree is awarded only on validation of the capacity to master at least one foreign (...)
31Between April 25 2002 and January 22 2014, a decree provided that the Master’s degree would be awarded "upon validation of the capacity to master at least one foreign language. Teaching programmes typically contain[ed] courses which allow[ed] students to acquire such skills" (Ministère de l'Éducation Nationale 2002)5. As noted in Anne-Marie Barrault-Méthy (2013), the decree placed LSP teachers in an untenable position: they were expected to judge students’ level based not on language proficiency, but on the cognitive ability to be proficient at languages.
32The concerned article in the decree has since been abrogated, and the decree in force as of September 2014 contains no mention of any expected language proficiency at Master’s degree level, or of expected language provisions. Master’s degrees can be delivered irrespective of any language competence and in the absence of any language teaching provision. Conversely, LSP proficiency at Bachelor’s level does not preclude success, or failure, at Master’s degree level if the Master’s degree does not contain any language component.
33However, one may consider that the selection to enter a 2nd year of Master’s degree favours those students who have already been selected for mobility schemes, and thus who have a successful informal and non-formal LSP learning experience.
34The same shift of emphasis from classroom proficiency to out of the language classroom successful interaction can be observed at Doctorate level. Unsurprisingly, the decree of August 7 2006 bears no mention of organised language teaching as part of doctoral programmes. Here, non-formal and informal language learning are implied through the international scope of doctoral schools. The term "international" appears seven times in the decree, including five times in relation to joint supervision agreements, through which doctoral students may follow teaching programmes in two universities and receive two doctorates. Joint supervision remains possible outside any formal agreement between two universities, on the basis of an informal agreement between two supervisors and upon agreement of the institution in which the doctoral student is registered.
35Thus, informal and non-formal language learning is encouraged in order to expand the international recognition of the doctorate. Success, at doctoral level, is likely to consist in being able to engage in internationally recognised research.
- 6 "The language of teaching, examinations and competitive examinations as well as of dissertations in (...)
36Following the Toubon law of August 4 1994, the Code de l'éducation contains article L121-3, introducing an exception to the rule according to which French is the teaching language6. The exception concerned classes delivered by visiting professors and the "necessities" of regional or foreign languages and cultures. The Ministry of Higher Education’s white paper of the law of July 22, 2013 allowed derogating to the law of August 4, 1994 by allowing the delivery of more classes through the means of languages other than French. The Ministry changed this however and the law of July 22, 2013 provides in article 3 that the impact of CLIL programmes in France and French CLIL programmes abroad will be evaluated by the Parliament. We can interpret such backtracking in two ways. The Ministry may be concerned over how exclusive such programmes might be for students, particularly from migrant backgrounds. For certain degree courses also, French would have been recognised less likely to favour employability than English.
37There are two main student unions in France, Union Nationale des Étudiants de France (UNEF) and Fédération des Associations Générales Étudiantes (FAGE).
- 7 "(…) In most university degree courses, there are also fields of study which are indispensable to a (...)
- 8 "UNEF is attached to democratisation and access to foreign languages, but does not support the rep (...)
38UNEF considers that LSP as a subject is necessary for all students7 as it may help them get a job8. Success in LSP thus contributes to employment. On English Medium Instruction though, the Union’s position seems to differ: in some contexts, notably for research, mastering the English-speaking literature is indispensable and may justify that certain courses be delivered in English:
- 9 "For certain activities, notably research, the mastery of the English-speaking literature is indis (...)
[c]onversely, the integral replacement of French in some degree courses by English would mean giving up making French a language which counts, and would send the wrong signal to all young people from French-speaking countries for whom French is a gateway to the world. UNEF wishes the debate on article 2 to calm down in order to envisage significant improvements such as the obligation to deliver a considerable part of the teaching in French. (UNEF 2014b)9
39As the mastery of English is necessary in research for understanding the literature and participating in international debates, EMI can then be "justified", but in limited cases, and should preserve a place for French, particularly for migrants from French-speaking countries. It is unclear whether the suggestion also applies for programmes delivered in other languages such as German and Spanish. FAGE supports language teaching for all students in all degree courses but currently seems to have no explicit position on CLIL programmes. It does not distinguish between languages. It advocates the validation of competencies acquired in informal and non-formal contexts, which obviously involves language competencies (FAGE 2015).
40FAGE and UNEF are both members of the European Student Union (ESU), which currently counts 47 member organisations. In order to better equip students with the skills required internationally, ESU recommends tuition-free language courses and certifications for students and migrants prior to their entering university. ESU stresses that languages should be "on offer" to both domestic and international students, that universities should put in place policies of internationalisation-at-home (ESU 2013), but says nothing about whether language classes should be mandatory for certain degree courses, or if a minimum language level should be expected on graduation. ESU also distinguishes small language areas, in which EMI might be beneficial, and larger ones (ESU 2009).
41Faced with so many positions regarding success and failure in LSP and so many definitions of LSP, some of which he or she may not be aware of, the LSP teacher keeps making choices. These are first methodological. Examination constructs heavily depend on the definition of LSP and, as we have seen, several such definitions coexist, grounded in national research and academic traditions. Teachers may or may not align course content against the CEFRL. They may take into account pluricultural and partial competence and ignore certain linguistic limitations. They may create tests that are discriminatory towards low-level learners, composed of questions all requiring a certain level. The choices teachers make are thus also highly political. Teachers may favour the success of low-level learners, or migrant learners, and allow them to reach a pass mark, or choose to conform to institutional pressure and, particularly at Bachelor’s level where students are numerous, create tests with a gatekeeping function. They may create a marking scheme thanks to which most students will reach an average mark, and so leave it to subject teachers to pass or fail students. In any case, teachers choose a definition of LSP based on their initial training, their research fields, possibly also on their personality, and on their ability, and willingness, to resist peer pressure.
42Designing a learning and a testing construct can thus be considered a deeply personal socio-historically situated experience in which teachers negotiate "identity, agency and marginalisation" (Rudolph 2013: 132) with a number of other agents, namely supranational organisations, governments, local institutions, colleagues, and networks. In France, the Ministry of Education and the Council of Europe provide frameworks of "reference", but overall have little grasp on what takes place in the LSP classroom. This, then, raises two questions: has the role of affective variables (Bialystok & Frölich 1978) in LSP not been underplayed, if even recognised? And if we take into account the rise of informal and non-formal learning and of their place in university language policies, how far do LSP teachers contribute to overall success in LSP?
43The study across the discourses of university language policy agents of the seemingly unequivocal notions of success and failure, and their seemingly binary opposition, reveals ambiguity and variation among language management agents and within the discourses of such agents. For the Council of Europe, complete failure seems to be outside the scope of language learning, which implies learner activity, however minimal. Failure at a language examination still implies some degree of success at language learning, even though not institutionally validated. The CEFRL emphasises teachers’ social responsibility of viewing positively any learning, however minimal. The Council of Europe also considers positively the linking to the CEFRL of tests and examinations, even though these may not serve an inclusive purpose, and irrespective of their ideological undertones. In other words, the Council of Europe, for the sake of inclusion of multiple approaches to assessment, and in order not to exclude any approach, encourages tests and examinations which may exclude some learners. In the same way, the French State provides for some language learning at university, but without setting any explicit measurement for success or failure. Implicitly, though, particularly at Master’s and PhD level, success is defined by the capacity to participate in international programmes and engage in international networking, which implies non-formal (autonomous) and informal (through contact) language learning. For student organisations, success seems to be defined in two distinctive ways, as the fact of succeeding at examinations and of finding employment. Non-formal and informal language learning are not considered.
44Being mostly unaware of the various definitions of success, and thus of failure, and because of their immense social responsibility, language teachers’ positioning towards success and failure may be quite uncomfortable. The notions of non-formal, and even more, of informal language learning, question their agency at university and, due to the increasingly international character of higher education institutions, the very interest of maintaining language classes instead of developing informal language learning by having CLIL courses, while also developing non-formal language learning through MOOCs.
45The university is also a language policy agent as regards success and failure by offering role models. The successful academic, active in the international scientific debate, is by definition able to convey research results through the means of other languages, and then, logically, to deliver teaching also in foreign languages. Even at UK universities, such role models may be non-native speakers. Thus, the rise of ELF in university settings also questions assessment and test constructs, and makes language policy agents’ positioning towards success and failure all the more unclear. This, in turn, may contribute to feeding the debate on the social responsibility of universities.
46As we have endeavoured to show, examining what success and failure means at micro level, the level of agents, shows how ambiguous and situated in time and space such notions are. We can conclude, then, that the vision of LSP learning as successful, or as a failure, is highly political, and reveals the tensions between policy agents’ positions on success and failure, with socio-political implications on LSP learning as an ecosystem, and on individual agents. Examining the policy of success and failure at micro level, the level of agents, also seems to offer promising methodological perspectives to account for agents' identity-building decisions. It would also be interesting to observe whether similar gaps exist among agents’ understanding of other notions. The fact that agents may have such conceptual differences may, in practice, eventually inform language planners at institutional level.