*Keywords:* Translanguaging, French as a foreign language, classroom interaction, factors of influence

‘anne-marie-bezzina’

---

Volume 11, No.1, pp. 75-95 Faculty of Education©, UoM, 2017

Translanguaging Practices in the Teaching of French as a

Foreign Language in Malta

Anne-Marie Bezzina

Faculty of Education, University of Malta

anne-marie.bezzina@um.edu.mt

Abstract: This study reviews beliefs related to translanguaging activities in the French as a Foreign Language (FFL) classroom and suggests cultural reasons why some condemn the concomitant use of previously learnt languages with the target language in FFL learning contexts. A corpus analysis of two Maltese FFL teachers’ recorded lessons attempts a structural categorisation of translanguaging instances according to the classification of classroom translanguaging in Causa (1998). It sheds light on the functions that translanguaging is made to fulfil in the Maltese FFL classroom, in comparison to those proposed mainly in Maarfia (2008). The results are compared to Maltese FFL teachers’ statements about their views on translanguaging in a recently administered questionnaire (Bezzina, 2016 ). Basic statistics reveal that the target language is often far from being the dominant language in the FFL classroom, and that learner talk is hardly encouraged in any language. Observations are interpreted in the light of possible practical factors of influence on translanguaging in the FFL classroom, as proposed in Molander (2004).

Keywords: Translanguaging, French as a foreign language, classroom interaction, factors of influence

Introduction

The aim of this study is to analyse translanguaging practices, or the movement of speech from one language to another, in the French as a Foreign Language (FFL) classroom in Malta. The notion of translanguaging, as opposed to “double or multiple monolingualisms(s)” refers to bilinguals’ ability to strategically employ their linguistic repertoire as a whole, according to their perceived language needs in any communicative situation. The foreign language (FL) classroom is necessarily a place where language contact occurs, and translanguaging takes place in virtually all contexts where two or more languages meet. The frequency, distribution and functions of translanguaging are influenced by diverse socio-individual factors, as well as by factors pertaining to the pedagogical setting in which the teacher and the

learners are interacting (Molander, 2004). The socio-individual factors include the demographic composition of the group made up of the teacher and his/her learners, the learners’ degree of knowledge and practice of the L2, and their motivation to learn, whereas the more institutional factors concern the methodological approach, teaching strategies and tools used. Other factors of influence are the way the teacher shapes interaction in the FL classroom, contact time allotted to FL lessons and syllabus demands (Ibid.).

Moreover, in the Maltese secondary school setting where French is most often learned as a chosen FL from a list of possible languages, one must equally take into account the effects of bilingualism, as most learners speak Maltese and English, which they will also have formally learnt throughout their primary school cycle, prior to commencing their studies of French. Thus this study is based on the analysis of a corpus of recorded FFL lessons in two Maltese secondary schools, and attempts to put observations regarding the frequency, structure and functions of translanguaging in relation to possible factors of influence on the participants’ spoken behaviour patterns.

The study takes into account the perspective that classroom talk is interactional in nature, and that teacher talk influences learner talk and vice versa, but at the same time the corpus analysis focuses here almost entirely on speech produced by the teacher. This is due to one main reason. In the researched context, teachers appear to verbally lead such a dominant role in the recorded lessons (see below), and in this sense they fashion their lessons in a rather traditional way. Consequently it seems that learner initiative is reduced to a minimum. It is the teachers who determine the flow of speech and set the pace, and therefore it is more worthwhile to analyse teachers’ verbal behaviour, from the point of view of translanguaging, because they appear to be setting the example in this regard and to be themselves eliciting instances of translanguaging from the learners’ part.

An initial brief review of literature on the benefits of translanguaging specifically in the French as a Foreign Language classroom is followed by a review of some researchers’ views against the use of the L1 in this context, and reasons are proposed as to why such strong opinions may be held in the FFL context. This is followed by a corpus analysis, with interpretations as to the functions translanguaging is made to fulfil, and categorisations regarding the structure of the translanguaging instances. A limited statistical exercise sheds light upon teacher and learner verbal presence in class and on the two entities’ extent of translanguaging practices. Finally, the attested verbal behaviour is linked to possible factors of influence which give rise to it, and an evaluation is made of the overall role that translanguaging plays in the recorded teacher talk.

Since three languages need to be mentioned in this study, when a distinction needs to be made between Maltese and English use, the L1 will be taken to refer to Maltese as native language, and the L2 will refer to English as a second language, previously learnt throughout the learners’ schooling and often medium of instruction for an array of school subjects. In other instances, when the L1 alone is mentioned, it will be taken to generally include both Maltese and English as a mixed medium of instruction. French, the target language (TL), will be referred to as the L3. For reasons explained in Bezzina (2016), the term translanguaging is the preferred term to refer to the movement from one language to another in the participants’ speech; however, code-switching (CS) will here also be used, especially in the literature review, to respect original authors’ choice of terminology.

Translanguaging in the French as a Foreign Language Classroom

In this paper the focus of the literature review is narrowed specifically to the French didactic context. We will first review a number of works in which translanguaging is viewed positively. Thus Coste (1997), writing generally about what he terms “didactic switching” (my translation), draws up a number of arguments in favour of the practice. He observes that codeswitching is constantly present in FL classrooms, whether admitted or not, and that in many cases, teachers cannot avoid a feeling of guilt in resorting to the L1. This happens despite the fact that discourse and spoken interaction analyses of classroom verbal exchange have highlighted the contribution of the L1 in managing activities and work, and in avoiding misunderstanding while defining tasks. For Coste, CS derives its value not from being a mixture of languages, but from its role in the construction of a multilingual competence set within the frame of a repertoire of distinct varieties, with the speaker developing an ability to manage this repertoire as a whole. This needs to be placed within the perspective of a sensible promotion of multilingualism practices which go well beyond simple juxtaposition. Coste thus advocates “doing” rather than defining CS, applying it by integrating it fully in methodological models.

Mati (2013) finds that Algerian teachers of French are little aware of the language distribution issue and have received no training on it. However they wish there could be a development of the use of the L1 in their lessons.

Several other researchers empirically observe functions being fulfilled by translanguaging in the teaching and learning of FFL. Maarfia (2008) conducts analyses of recorded classroom interactions and interviews with teachers at the second year of the primary cycle in Algerian schools, which lead her to generally positive conclusions regarding CS in the French classroom, in that it facilitates learning by allowing students to overcome obstacles, it enriches learning by allowing them to memorize a greater number of elements, and it

improves learning by allowing them to resort to a wider range of cognitive operations. Another research conducted among very young learners is that of Ehrhart (2002) which assesses the efficiency of different types of teaching of French to sixto eightyear old primary school children in the German Sarre region, which borders France. This early exposure to French takes places on a two-hourly weekly basis and is delivered by native French teachers. It is noted that while absolute

Fig. 1 Teacher profiles vis-à-vis translanguaging in the FFL classroom (Translated from Ehrhart (2002:6)

beginners are still in their “silent period”, CS is found in teachers’ speech. As the learning progresses, both parties resort to CS. Learners do so due to a communicative urge, when they wish to contribute input, but words in the TL fail them. When teachers persist in using CS in spite of their learners’ having reached a certain competence in the TL, their motivations for doing so may be to better explain content or instructions, to structure the lesson’s progress, and to construct proximity with the learners on a relational and affective level. Ehrhart constructs a typology of teacher profiles according to their practices vis-à-vis CS, as in Figure 1.

Thus, for Ehrhart, there are three main types of CS patterns in teachers’ speech: hardliners refuse any use of the learners’ L1, moderate teachers tolerate the L1 whilst resorting to it themselves or refusing to integrate it in conversation, and active users exploit the benefits of the practice. In the case of the latter group, teachers’ resorting to the L1 may stem from the L1 having just been used in a preceding turn by the learners, but not always. Ehrhart

believes it is unjustified to speak of language choice in everyday didactic contexts, because language use is not often conscious; rather, CS is a strategy used according to students’ real or perceived needs.

At a much higher level of education, Yiboe (2010) films FFL sessions in sixth forms in Ghana and observes that CS serves to support utterance construction and comprehension, and enhances metalinguistic and metacognitive activity. Molander (2004) remarks that in FFL immersion classes in Quebec, not only does CS facilitate communication and the teaching / learning of the FL, but it also enriches them. Thus, among other forms of translanguaging patterns, learners may express themselves through what she calls “compensatory CS” (my translation), which is comparable to Ehrhart’s designated learner motivations for using CS. In fact, for Molander too, compensatory CS is produced in the L1 in order to make up for lexical or structural lacunae in learners’ L2 communication. Molander observes that translanguaging manifests itself in teacher talk in the form of translation, reformulation, precisions, explanations and clarification. These are used metalinguistically, to refer to objects of which learners do not know the TL equivalent, to comment on differences and similarities between the L1 and the TL, to explain the right formation of TL words and sentences, to check learners’ understanding and to explain content. They are also used communicatively, in maintaining interpersonal relationships (building complicity with students, joking) and for classroom organization (clarifying future activities, regulating learners’ behavior, reprimanding, and for attracting and maintaining attention).

Soku (20 14 ) describes intra-sentential, inter-sentential and alternate (or speech-turn boundary) CS in beginner FFL classes at High School level in Ghana. An analysis of his corpus of recorded lessons shows that teachers’ CS is pedagogically motivated, and revolves around one main aim, namely that of facilitating comprehension by their public of beginners, who, as can be seen from the transcript excerpts, and given their educational level, are expected to discuss general social issues like unemployment and immigration. As in Ehrhart (2002) and Molander (2004), CS is interpreted as taking place because of learners lacking elements in their lexical repertoire, although the general impression obtained from the transcript excerpts shows an effort on Ghanaian students’ part to use French when they can. Learners also code-switch to encourage their classmates to talk, or when confused. Soku takes stock of the positive contributions of CS to interaction and access to content, and recommends revisiting the question of language choice for the teaching and learning of FFL. The Ghana 2004 syllabus for the teaching of French stipulates the use of French as medium of instruction for the subject in secondary and high schools, and school policy often insists that teachers immerse their students in a situation of exclusive French-language communication. Soku observed that this policy’s effect makes learners lose

interest. On the basis of his corpus-based observations, he advocates toleration and even encouragement of L1 use in the FFL learning context, especially in beginner classes and in certain phases of the lesson, such as the introductory phase, during grammatical explanations and to capture learners’ interest and attention.

Views in favour of limited L1 use in the FFL Classroom

Literature likewise presents the opposite scenario when it comes to judging the utility of translanguaging in the FFL classroom. Notwithstanding the positive situations observed by the above authors among others, one also finds a series of research works pertaining to the pedagogical tradition, perhaps particularly fervent in the teaching of French, that recommends the near-exclusive use of French during FFL lessons, and possibly beyond this context too.

Factors of a sociolinguistic and even political nature may partly account for the background ideology why many believe that French should be the privileged medium of instruction and expression in the FFL classroom, perhaps even more so when English is the L1 or a previously mastered language. Successive French governments have invested and still invest enormously to promote French internationally and to defend it on the international front as well as in its inward form from the danger of invasion from English, considered as a competing language (Judge, 1993). Likewise, beliefs are widespread among the French that their language is a pure, logical and beautiful language (Gadet, 2007), a language of culture which conveys civilization. Such ideas used to underline the traditional teaching of FFL in past colonial times when diffusion of the language was a political priority (Raimond, Anne-Claire, personal communication during lectures of the course Identifier les théories de l’apprentissage et les courants méthodologiques , Université d’été BELC 2012, Les métiers du français dans le monde). This kind of indoctrination has left repercussions on the mentality of a number of French citizens (Gadet, 2007).

What is certain is that one finds in the literature several contributions which hold CS as a hindrance to developing competence in French as a TL. Highly respected experts in the didactics of FFL recommend a limited use of the L1 in the classroom. Galisson (1980), for instance, recommends that the L1 be integrated to make the learner reflect on analogies and differences between the L1 and TL communicative systems, and on the relationship between knowledge of the L1 and acquisition of the TL, particularly in the domain of grammar. It may also ‘occasionally’ (my translation) intervene to allow access to the meaning of the foreign item.

More recently, Turnbull (2001, 2006) has insisted on the importance of maximizing TL use whilst proposing practical strategies which should lead to a “judicious and selective use of the L1” (my translation; Turnbull, 2016, p.612) as well as strategies to help learners function as much as possible in French. Turnbull (2001) reported from an 8-week period of classroom observations that TL use in four Grade 9 core French classrooms in Canada ranged from 9% to 89%, and that the students of the two teachers who spoke French most frequently outperformed students in the other two classes on many measures of general French proficiency and on achievement tests based on the curriculum implemented by all four teachers during the time of observation.

Analyses of learners’ spoken interaction and interviews with teachers of preintermediate level French in Sweden, lead Stoltz (2011) to likewise conclude that a prudent use of the L1 may favour classroom learning of a FL, but it is important to try to use the TL. He claims having observed a tendency to overuse the L1 and, unlike Soku (2014; see above), who recommends a change of TL-only policies in Ghana, Stoltz believes that his study gives sense to the 2011 Swedish syllabi’s recommendations that French must be spoken as much as possible in class, and must be the medium of instruction.

These latter views concur in following an additive bilingualism approach according to which it would be safer and more productive to keep languages separated and to curb the degree of L1 intrusion in the French classroom. In view of the two contrasting positions considered above, it becomes interesting and indeed essential to question what approach teachers of French in Malta adopt as regards L1 use in their classroom.

Functional and structural classification of translanguaging

Maarfia (2008), elaborating on the terminology describing FL classroom switches in Moore (1996), presents a rather simple, bipartite classification of the functions of CS. Analyses of recorded classroom interaction at the second year of learning at primary school level and interviews with teachers in Algerian schools lead her to follow Moore in subdividing CS instances into two main categories: Springboard and Relay Switching.

Springboard switching favours learning and is marked by hesitation, pauses and metalinguistic comments which attract attention to the switch. Relay Switching, which favours communication, is more fluid and tends to ease the flow of discourse and the construction of meaning. Within these two main categories, CS serves specific functions, for both the teacher and the learner, as summarized in Table I.

SPRINGBOARD SWITCHING,

SERVING LEARNING

RELAY SWITCHING, SERVING

COMMUNICATION

For the TEACHER For the TEACHER

• Giving explanation - Reestablishing order in the classroom

• Reminding or explaining rules

• Commenting on learners’ behavior

• Giving tasks - Asking practical questions to individual learners

• Making the message pass - Evaluating learners’ responses

• Guiding the learning

• Clarifying

• Correcting misunderstandings

• Resolving ambiguity

• Avoiding misunderstandings For the LEARNER For the LEARNER

• Checking - Asking for information

• Explaining - Giving information

• Compensating for an unknown item

• Reporting a classmate

TABLE I: Springboardand Relay-type CS, according to their description in Maarfia (2008), based on Moore (1996)

Causa (1998) maintains that in teacher talk, CS occurrences pertain to a homogeneous whole, because all forms she traces in one teacher’s speech can also be found in the discourse produced by the other teachers in her corpus. This homogeneous whole is also structured, as the forms present constitute a coherent set of types which reveal the metalinguistic and lexical density employed by the teacher. As far as content is concerned, they shed light on the metacommunicative dimension in relation to the communicative dimension. In her 1998 corpus, Causa distinguishes four CS forms produced by the teacher, which will be used in the categorisation of translanguaging occurrences in the present corpus analysis (see below). For this reason, the description of these four types in Causa (1998) is here borrowed and reproduced in detail. Examples are not included at this stage though, since in Causa’s corpus they concern French as L1 and Italian as the TL. The four different CS structures will be illustrated concomitantly with the present article’s corpus analysis (see below).

The first of these CS structures concerns what Causa calls acts of bilingual correspondence (my translation; in the original French: “la mise en correspondance bilingue”). This first category openly refers to the two

languages in contact through explicit markers such as in Maltese / in English / in French, you / we say , which in our case would be produced in any one of the three languages. This type therefore involves a high degree of metalinguistic and lexical density.

The second category is that of equivalence (in Causa’s terminology: “les équivalences”). In this case the act of correspondence is effected through a metalinguistically structured comment of the X is Y type. The use of such verbs which render explicit the relationship between the known word in the L1 and the unknown term in the TL, makes this category also heavily loaded from the metalinguistic and lexical points of view.

Thirdly, repetition activities (for Causa: “les activités de reprise”) do not present any markers or verbs with a metalinguistic value; the forms belonging to the two languages in contact are simply juxtaposed and momentarily interrupt the flow of discourse, so they are situated halfway along the continuum of CS forms. They could be ‘word for word’ translations, for instance, in which case they are really and truly repetitions in the TL of what was proposed in the L1, or vice versa. Otherwise, the movement into the other language could contain modifications in form and to a lesser extent in meaning, in which case they can be called reformulations.

Language crossing in the fourth category, simply called bilingual speech (“le parler bilingue”), occurs very smoothly, since language movements no longer imply in this case the reiteration of what has already been said in the other language. These CS instances do not hinder the flow of discourse, as was the case for the three previous categories. However, some explicit markers can at times appear to announce a switch, in which case Causa calls the instance a prepared switch. In other cases, the switch may also be totally unmarked and unprepared.

Data source and methods of analysis

The main data used for the present analysis is a linguistic corpus based on the transcription of FFL lessons delivered in two Maltese schools. The classroom interaction excerpts are derived from the Gauci (2016) corpus, recorded and transcribed in part fulfilment of a dissertation carried out by a FFL teacher trainee. The corpus is comprised of sixteen recordings, made in two teachers’ classes, in a boys’ and a girls’ Maltese church secondary schools. Classroom interaction was recorded as each teacher was holding 40-minute lessons with A1 beginner (Form 1) and A2 (Form 3) classes, after all necessary ethicsrelated permissions were obtained.

Other data used are the results of the questionnaire administered to teachers of FFL, presented in Bezzina (2016). These are discussed in conjunction with

the corpus analysis results, because the latter examples lend themselves to a test aimed to corroborate or refute the stated uses of translanguaging in Maltese FFL teaching as they emerge from the questionnaire (Bezzina, 2016) administered to Maltese FFL teachers. The results of the questionnaire-based enquiry are thus here being verified, as regards the teachers’ claims about what they deem to be the most and the less important functions of translanguaging. This exercise is not statistical in nature but relies on a careful inspection of the functions which translanguaging is made to fulfil in the two recorded participants’ actual teacher talk.

However, a limited exercise of statistical calculations is performed on a sample (around one-fourth) of the Gauci corpus, in an attempt to acquire a clear picture of language distribution realities in the two contexts where the recordings were carried out. Calculations of teacherand learnertalk also give an indication of what may be causing difficulties in learners’ verbal output. Rates of use of the L3 vis-à-vis the L1 and L2 are calculated by a simple addition of the number of words produced in each language in each lesson, separately for the teacher and the group of students. Percentages corresponding to these numbers are calculated to give a clearer picture of the relationships between the uses of each language.

Evidence of the functions fulfilled by translanguaging in the Maltese FFL classroom

The excerpts taken from the corpus of lesson recordings are mainly analysed here in order to illustrate which function(s) is / are being fulfilled by translanguaging in Maltese classes of French. The corpus analysis is to a certain extent based on subjective interpretation, but this is in each case backed by Maarfia’s classification of CS functions into the main categories of springboard and relay switching. The structure of the switch is in its turn classified according to Causa’s typology (see above).

Many times it seems that the sole reason for opting for the use of the L1 corresponds to the teachers’ claim in the Bezzina (2016) questionnaire, that the most important function fulfilled by translanguaging stems from the ideal to create a more relaxed environment in the class whilst putting the learners at ease. In several parts of the sample, one can appreciate the contribution of translanguaging in this respect, as in example one where the explicit translation into the L1 allows the learners to verify their comprehension of the teacher’s initial statement in the TL, and therefore helps them not to get lost at that point:

This exercise in clarification for ease of comprehension from the learners’ side is an example of what Maarfia classifies as springboard switching, and is structurally carried out through what Causa terms the equivalence type of

switching ( X meaning Y ): [… maintenant on va conjuguer les deux ensemble / iġifieri se ngħaqqduhom ma’ xulxin {… now we are going to conjugate them both together / which means we are going to join them together }].

(1) T: … on a conjugué le verbe partir et puis on a conjugué le verbe aller / n’est-ce pas? / maintenant on va conjuguer les deux ensemble / iġifieri se ngħaqqduhom ma’ xulxin BOYS FORM 1

we conjugated the verbe partir and then we conjugated the verb aller / right? / now we are going to conjugate them both together / which means we are going to join them together

In some cases, even simple matters which would easily be understood by learners in the TL are often expressed in the L1, so it is questionable whether these FL teachers are not perhaps taking the practice of translanguaging to an extreme. In the sample retained for basic statistical scrutiny in this study (ca. one-fourth of the Gauci corpus), the overall percentage of teacher-talk in French in any one single lesson can be as low as 30.6% (for the teacher in the boys’ school)! In example 2, the simple explanation and elicitation are of the type that Form 3 learners would understand in French. The fact that Maltese and English are resorted to in such a context is quite reminiscent of a more traditional approach where teaching about the FL dominates teaching the FL:

(2) T: eħe issa inti għandek żewġ “options” / ġieli jkun hemm mistoqsija magħhom din / jgħidlek quel est le moment de la journée? / xi tfisser din il-“question”? / quel est le moment de la journée? BOYS FORM 3

yes now you have two “options” / sometimes these are accompanied by a “question” / he tells you which part of the day is it? / what does this “question” mean? / which part of the day is it?

This example of translanguaging is equally of the springboard category, and according to Maarfia’s classification it appears that the teacher is guiding the learners. It can be said that the example reflects the reverse side of the medal when it comes to Causa’s typology of structures, since here we see that in her utterance the teacher asks the learners to produce the equivalence, rather than providing it herself.

Although it features in the penultimate position in the teachers’ ranking of the useful functions of translanguaging in the Bezzina (2016) questionnaire, in the corpus, expressing positive evaluation in the L1 appears to be one method teachers frequently use to boost learners’ self-confidence and motivation whilst creating closeness with them (see example 3), through a recurrent use of words like “eżatt” [precisely], “tajjeb” [good], “bravu” [well done]. Example 3 is another case where the teacher and student build equivalence together, in a further illustration of springboard switching involving

explanation [un avion / “plane / ok?” {a plane / “plane / ok?”}], although the positive evaluation in itself (bravu) is of the relay type.

(3) T: … avion x’inhi għidna? S: “I leave by aeroplane” T: bravu / un avion / “plane / ok?” BOYS FORM 1

T : … plane what did we say that it is? S: “I leave by aeroplane” T: well done / a plane / “plane / ok?”

Nevertheless, as in example 4, positive evaluation is also sometimes carried out in French:

(4) S: il est quatorze heures T: bravo très bien BOYS FORM 3

S: it is fourteen hundred T: well done very good

The teachers’ stated second most important function of translanguaging in the Bezzina (2016) questionnaire does figure very often in the corpus, with the teacher resorting to the L1 to ensure comprehension, for instance by translating a phrase (as in example 5 [il est fou / “he is crazy”]). In other instances she verifies comprehension of vocabulary (by eliciting a translation, as in example 6) or of reading texts (as in example 7):

(5) T: … flok qalilna il adore / les jeux vidéo / qalilna il est fou / “he is crazy” / orrajt? GIRLS FORM 3

instead of telling us he really likes / video games / he told us he is crazy / “he is crazy” / ok?

In example (5), the function (giving explanation) and the pause reflect a springboard switch, which, as regards structure, is also of the repetition type. Another springboard switch is that of the teacher guiding the learning in example (6), where once again the teacher elicits a reply which, along with her question, will construct the ‘X means Y’ equivalence type of structure: [xi tfisser inconnu? {what does unknown mean?}].

(6) T: … il est le personnage inconnu / xi tfisser inconnu? GIRLS FORM 3

he is the unknown character / what does unknown mean?

(7) T: … aqrawhom it-tnejn ġo qalbkom u araw jekk għandkomx xi “difficulties” GIRLS FORM 1

read both of them silently and check whether you have any “difficulties”

This last example is one of the rare cases of relay switching in this choice of corpus excerpts, with the teacher asking a practical question. Causa’s type

termed “bilingual speech” is here seen to correspond to Maarfia’s relay category; the structurally uneventful insertion of the L2 term fits the description of this category as the more fluid kind of switch, which eases the flow of discourse.

It is evident in the corpus sample that explanations of grammar points and complex concepts are often conducted by the teacher while heavily relying on the L1. It is interesting to note that a pattern seems to be followed while negotiating metalinguistic content: more general comments are made in Maltese, whilst technical or more pedagogy-related terms and word-for-word translations are often rendered in English. This to and fro movement between three languages does not produce any confusing results; on the contrary, learners constantly respond by showing understanding of the concepts being discussed. The triple translanguaging mix, which at first sight may appear to complexify the teacher’s explanation, in reality exerts a simplifying effect. This can be illustrated through example 8, where the teacher is reminding students about the recently presented contracted articles, with the added complexity of providing examples through the notion of holiday destinations. The simplification function accomplished by translanguaging is equally clear in example 9, when the teacher is explaining the need for coherence in opting for either the twelve-hour or the twenty-four hour clock:

(8) T: … u l-bieraħ / hier / on a discuté les prépositions bil-à hux vera? / u għidna li dawn irridu nitgħallmuhom bħala “expressions as they are” / iġifieri jekk irrid ngħid “by the sea” / mal-“beach” / à la plage / à la montagne / à la campagne / tħalluhom bħala “expressions” kif inhuma // u à le / à le teżisti ħdejn xulxin? S: le T: bravo / u x’ngħidu minflokha? S: au T: u à les / x’għidna kif tiġi? S: aux T: tajjeb BOYS FORM 1

T: … and yesterday / yesterday / we discussed prepositions with the to (at) right? / and we said that we need to learn these as “expressions as they are” / which means if I want to say “by the sea” / with the “beach” / to (at) the beach / to the mountains / to the countryside / you leave them as “expressions” as they are // and (à le) to the / (à le) to the do they exist next to each other? S: no T: well done / and what do we say instead of it? S: (au) to the T: and (à les) to the (plural) / how did we say it comes? S: (aux) to the (plural) T: good

The first two instances of switching in example (8) are of the springboard type, with the teacher engaged in explanation, in what is structurally a repetition type of switch: [l-bieraħ / hier {yesterday / yesterday}]; [“by

the sea” / mal-“beach” / à la plage … {“by the sea” / with the “beach” / to (at) the beach}]. The positive evaluation (bravo) corresponds to the relay type, whilst springboard switches involving the notion of saying [x’ngħidu minflokha? … x’għidna kif tiġi?… {what do we say instead of it?… how did we say it comes?}] structurally belong to the bilingual correspondence type as defined by Causa.

(9) T: iġifieri inti se tgħid jew it“twenty-four hour clock” bħal ma qalilna XXX [triple letters substitute learners’ names] per eżempju il est douze heures “it’s twelve o’clock” le? / issa “twenty-four hour clock” jekk ngħidlek “seventeen hundred” jien x’ħin ikun? S: forsi l-ħamsa? T: il-ħamsa / ta’ x’ħin? / ta’ filgħodu? S: ta’ filgħaxija T: mela jekk jien ngħidlek “seventeen hundred” / il est dix-sept heures m’hemmx għalfejn ngħid de l’aprèsmidi għax hija ovvja li après-midi / hux vera? / jekk ħa tgħid il est dix heures S: trid tgħidha T: hemmhekk eħe trid iddaħħalha BOYS FORM 3

T: which means you are either going to say the “twenty-four hour clock” as XXX told us for instance it is twelve hundred hours “it’s twelve o’clock” no? / now “twenty-four hour clock” if I tell you “seventeen hundred” what would the time be? S: maybe five o’clock? T: five o’clock / of what time? / in the morning? S: in the evening T: so if I tell you “seventeen hundred” / it is seventeen hundred hours there is no need to say in the afternoon because it’s obvious that it is the afternoon / isn’t it so? / if you say it is ten o’clock S: you have to specify T: in such a case yes you have to add that

In example 9, the first two instances of translanguaging [il est douze heures “it’s twelve o’clock”]; [“seventeen hundred” / il est dix-sept heures] are structurally of the repetition type, and functionally belong to the springboard category, their purpose being clarification in order to make the message pass. Bilingual discourse is on three occasions the privileged structure later on in the example [m’hemmx għalfejn ngħid de l’après-midi; hija ovvja li aprèsmidi {there is no need to say in the afternoon because it’s obvious that it is the afternoon}]; [jekk ħa tgħid il est dix heures {if you say it is ten o’clock}], although this fluid movement from one language to another still needs to be functionally classified among springboard switching, where the teacher’s aim is to explain and clarify expressing time in the TL.

In the Maltese FFL teachers’ questionnaire results, the use of translanguaging fared much higher for classroom management from the point of view of dealing with disciplinary issues, than it did for task management, which for instance involves giving instructions. In actual fact, the corpus sample shows

that in the two recorded teachers’ talk, translanguaging was very common in both instances.

The teacher in the boys’ school has to handle her learners’ behaviour; this does not apply in the girls’ school where the sample revealed no discipline problems. In the case of the former teacher, the L1 does reveal itself as a useful tool for managing the class at times when there is disruption, for instance to pronounce classroom rules and to give directives for better behaviour. The translanguaging instances in the case of example 10 are structurally of the more fluid bilingual discourse type, and functionally belong to the relay category, where the purpose is to reestablish order. It could be that the teacher feels she can create more closeness with the learners when she needs to put them back on track if she uses the L1:

(10) T: … XXX “sit down” / ejja ejja x’int tagħmel? / qed taħli l-ħin XXX / u YYY / ejja /// donc / le verbe partir … … x’qed tagħmel ZZZ? S: le miss għax qed infittex fulskep T: EJJA / suppost diġà kienet fuq ilmejda / filkas oħroġ oħra / ma naħlux ħin / ejja ZZZ / hemm bżonn tkunu iktar organizzati / xbajt ngħidilkom / donc le verbe aller… BOYS FORM 1

T: … XXX “sit down” / come on come on what are you doing? you’re wasting time XXX / and YYY / come on /// so / the verb partir… … what are you doing ZZZ? S: no miss because I am looking for a foolscap T: COME ON / it should already have been on your table / if need be take out another one / no time wasting / come on ZZZ / you all need to be more organized / I’m fed up of telling you / so the verb aller…

However, the corpus also shows that the TL is sometimes at least partially resorted to when dealing with disciplinary issues (example 11). In this case also, the bilingual speech structure type reflects non explicit switching which functionally can be categorised among relay switches. The teacher’s aim is in fact to reestablish order while at the same time commenting on learners’ behaviour.

(11) T: tu ne peux pas boire / c’est pas possible / tinsewx ir-regoli / irregoli ma ħriġtomx jien imma rridu nimxu magħhom / issa x’ħin iddoqq il-“bell” ixrob ftit ok? BOYS FORM 1

T: you cannot drink / it’s not possible / don’t forget the rules / it’s not I who came up with the rules but we need to follow them / then when the “bell” rings drink a little ok?

The corpus does however, at least in the case of the two teachers whose lessons were recorded, contrast with Maltese FFL teachers’ claims that compared to switching for class management purposes, translanguaging is not very common while managing tasks related to content. In fact, in the questionnaire results (Bezzina 2016 ), translanguaging fulfilling the function of dealing with disciplinary issues ranked at the relatively high fifth position, while translanguaging for discussing tasks ranked at the relatively low eleventh place. It is clear that both teachers tend to mix the L1 and the TL in giving instructions too. It was noted that especially in the case of the girls’ teacher, languages are switched intrasententially: thus, in example 12, the springboard switching involving task-giving is structurally of the bilingual (or even trilingual) discourse type.

(12) T: tajjeb / issa d-dar se tagħmlu page soixante exercice trois / méthode orrajt? tridu tagħżlu wieħed / iġifieri “one way” / mela mhux l-ewwel se nużaw it-“twelvehour clock” imbagħad it-“twentyfour hour clock” / wieħed li trid inti // issa *page cinquante-huit / tournez la page… * GIRLS FORM 1

T: good / now at home you are going to do page sixty exercise three / from the textbook ok? you have to choose one / that is “one way” / so you don’t first use the “twelve-hour clock” and then the “twenty-four hour clock” / whichever one you want // now page fifty-eight / turn the page

Intersentential switching occurs more frequently in the boys’ teacher’s speech, and gives longer stretches in the TL (example 13). Task assignment is once again accomplished here through the smooth bilingual discourse structure.

(13) T: l-aħħar darba konna qed nagħmlu tal-verb aller u partir hux vera? / jekk għandkom l-istess karta oħorġuha / jekk le oħorġu oħra / regardez le tableau s’il vous plaît // good / alors / l’autre fois on a discuté le verbe aller et le verbe partir n’est-ce pas? BOYS FORM 1

T: last time we were doing about the verb aller and partir right? / if you have the same paper take it out / if not take out another one / look at the board please // good / so / last time we discussed the verb aller and the verb partir isn’t it so?

In example 13 and even more extensively in example 14, the corpus excerpts include instructions almost entirely delivered in the TL, with the L1 appearing briefly to allow general comprehension. In example 14, the teacher clarifies what is going to happen next in class and makes the message pass by resorting to a springboard switch of the equivalence structure type [… et puis on va faire des phrases / ħa nagħmlu ftit sentenzi “as we go along” {…and then we are going to build sentences / we will build a few sentences “as we go along”}].

(14) T: alors on va discuter les transports / puis je vais vous donner les notes S: ħa niktbu n-“notes”? T: non pas pour ce moment / regardez seulement le tableau et concentrez-vous / donc on va discuter les transports qu’on utilise normalement et puis on va faire des phrases / ħa nagħmlu ftit sentenzi “as we go along” / ça va? BOYS FORM 1

T: so we are going to discuss means of transport / then I’m going to give you the notes S: are we going to write down the “notes”? T: not for now / just look at the board and concentrate / so we are going to discuss the means of transport which we normally use and then we are going to build sentences / we will build a few sentences “as we go along” / is that ok?

Statistical indications regarding language distribution in FFL classes

Generally, the corpus suggests that the L1 is the dominant language in these two Maltese FFL teachers’ classes. A percentage calculation exercise carried out on a sample of the Gauci corpus in fact revealed that on average, the girls’ teacher used the L1, Maltese, 45% of the time, and the L2, English, a meagre 5.9% of the time. The L3, French, was used on average nearly half of the time (49%). The boys’ teacher used the L1, Maltese, much more extensively, 54% of the time, the L2, English, 12% of the time, and the TL, French, only 33% of the time! The teachers’ verbal behaviour as regards TL use was observed to be rather constant from one lesson to another. In the case of the girls’ teacher, there is only an 8% difference in the use of French from one lesson to another, while in the case of the boys’ teacher, this difference even goes down to 5%.

The two rather different tendencies in teacher talk vis-à-vis translanguaging do not seem to have an evident effect on the amount of talk realized by the students in the TL and the two other languages. Thus the girls use the L3 rather less than their teacher, at 33% of the time, and Maltese and English more than her, at respectively 50% and 17% of the time. However, marked fluctuations were noted in the girls’ levels of use of the L3 between one lesson and another (for example, 50% vs. 17% rates of use of French). The boys used the TL slightly more than their teacher, that is 38% of the time, though in their case too, there were large gaps between their verbal behaviour in individual lessons. Use of French by the boys was noted to reach 59% in one lesson and to be as low as 16.5% in another lesson. The boys used English slightly more than their teacher, for an average of 17% of the time, and Maltese significantly less than her, at 37% of the time.

These results suggest that although the potential positive contributions of the L1 and L2 are doubtlessly being put to fruition by these teachers, there can be

an improvement in the rate of use of the TL. This applies particularly in the case of the boys’ teacher. The low rates of TL use by the learners themselves are quite worrying, and adjustments would need to be made in order to promote confident L3 use among them. This need emerges as all the more pressing when it is considered that learner talk (by the whole group of learners) amounts to a meagre average of 12% in the girls’ case, and to the comparable rate of 15% in the boys’ case. Absolute teacher talk dominance is known to be unhealthy as it stifles interaction from the learners’ part, and appears to be seriously hindering learners’ intervention in any language, apart from the much-needed spoken practice in the TL.

Without generalizing, it is possibly legitimate to think that these observations reflect a much wider practice. In fact, when, as part of their University course, student teachers of French try to use the TL as dominant language in their teaching practice assignments in Maltese schools, they often meet with resistance and lack of understanding from the learners. This suggests that learners in schools may tend to be used to French lessons being delivered mostly in a combination of the L1 and the L2. There also needs to be more formal preparation of student teachers on how to carry out translanguaging in FFL, and consensus among teacher practice examiners as to how to assess its management.

Conclusion

The corpus generally corroborates, by evidence of its frequent use in the classroom, what Maltese FFL teachers expressed about translanguaging practices and their usefulness, in the questionnaire administered to them (Bezzina 2016). Their opinions expressed in the questionnaire, namely that translanguaging is particularly useful on the affective and class management levels, besides ensuring faster and easier comprehension and vocabulary learning, are reflected in the uses that translanguaging is put to in the recorded lessons. An attempt can be made to trace a relationship between some of the factors of influence related to translanguaging, as enumerated in Molander (2004; see above), and the extremely marked frequency of translanguaging instances observed in the lesson recordings carried out in the Maltese educational context. Among the socio-individual factors, the homogenous demographic composition of the teacher and learner group, all Maltese bilinguals, allows the use of Maltese L1 and English L2 besides French as the L3. It is statistically shown here (see above) that the degree of spoken practice of the TL, and indeed of any verbal expression by learners, is unfortunately kept low by the teachers’ design of teacher and learner talk in their classes. Elicitation is mostly carried out in the form of closed questions. As regards learners’ motivation to learn, this will differ in each case, however, interviewed in Gauci (2016), the two participating teachers do claim that with time, learners gradually lose some of the excitement they initially feel about

the fact of learning a FL, and that the more they grow, the more wary they become of committing mistakes and consequently of being sneered at by their classmates.

As regards the more institutional factors of influence, the methodologically rather traditional approach adopted, in spite of the communicative approach textbooks and syllabus design, also contributes to the observed state of affairs. The focus is on grammar and vocabulary, and subsequent phrases from reading passages are translated almost word for word. The oral component for its own sake, as in role play or discussions, is rarely reported to be observed in Maltese classes. Thus in the sixteen lessons observed by Gauci (2016), there were few instances of assignment of pair or group work, where learners had more opportunity to talk, and more comfortably, in smaller groups. This restricted practice may be due to teachers’ concerns about syllabus demands which have to be covered in the limited time allocated to TL lessons. However, the syllabus does include oral practice skills and therefore teachers are prioritizing grammar and writing skills to the detriment of spoken interaction. They probably effect these choices for quite pragmatic reasons, since the current exam marking scheme of school end-ofyear examinations and of the national examinations learners sit for at the end of their secondary school reserves the vast majority of marks for the written component and a very small percentage for the oral component. This therefore suggests that the overall setup of summative evaluation has to change, in favour of a more balanced weighting of spoken, listening, reading and writing skills (this has started through the Subject Proficiency Assessment project which is however for the moment only being offered to learners who show evident difficulty in FL learning). The substitution of ageing textbooks and their accompanying material needs to be envisaged, to make place for more appealing, taskand project-based language learning methods which would give space to discussions. Most importantly, Continuing Professional Development needs to focus on developing teaching strategies so that teachers cease to prioritize grammar and writing and shape their classroom interaction patterns in such a way as to encourage extended learner talk.

As things stand, the low percentages at which French features in the recorded FFL lessons in Malta, if these results are applicable to a wider context, are cause for concern. However, the recordings also reveal that teachers are positively aware of the good uses that translanguaging can be made to fulfil and are capable of drawing upon them.

The results seem to confirm that functionally, switching of the springboard type is much more frequent than the relay type. This is in keeping with Maarfia’s listing of many more functions carried out by springboard switching, to enhance the learning of content. Structurally, a mix of acts of

bilingual correspondence, equivalence, repetition, and bilingual speech types is observed. As Causa (1998) reports, translanguaging is not restricted to beginner level. However, Causa notices a transformation in the types of structures of translanguaging instances at beginner and more advanced levels. It is not the scope of the present study to focus extensively on the structure of the switches, but this could be investigated in further studies carried out on corpora of transcribed FFL lessons in Malta.

According to Ehrhart’s typology of teachers, the two participant practitioners can be classified among the active user type, as they resort freely to the L1, both through self-initiation of switches and when they are stimulated to use the L1 due to previous initiation by the learners. As Molander (2004) states for her observed situation, not only do these two teachers exploit instances of translanguaging for explanation of TL forms, but they also accept and acknowledge the content of their learners’ switches, and encourage learners to produce them, through elicitation, whilst producing them very often themselves. It would be interesting to compare these observations with transcribed classroom interaction involving other FFL teachers in Malta, possibly from the State and Independent school sectors, in an attempt to get a clearer picture of how widespread this free use of translanguaging and L1dominance are in Maltese FFL classes.

Transcription Conventions

/, //, /// : pause – brief, medium, long

XXX, YYY, ZZZ : students’ names

… : words, utterances or parts of utterances produced in French

“…” : words, utterances or parts of utterances produced in English

? : instances of interrogative intonation

References

Bezzina, A.M. (2016). Translanguaging in the teaching of French as a foreign language in Malta: Teachers’ perspectives and attitudes. Malta Review of Educational Research 10 (2), 277296. Causa, M. (1998). Maintien, transformation et disparition de l’alternance codique dans le discours de l’enseignant, du niveau débutant au niveau avancé. Etudes de linguistique appliquée, 108 , 457-465. Coste, D. (1997). Alternances didactiques. Etudes de linguistique appliquée, 108 , 393-400. Ehrhart, S. (2002). L’alternance codique dans le cours de langue : le rôle de l’enseignant dans l’interaction avec l’élève – Synthèse à partir d’énoncés recueillis dans les écoles primaires de la Sarre. In A. M. Lorenzo Suarez, F. Ramallo, X. P. Rodriguez-Yanez (éds.). Proceedings from the Second International Symposium on Bilingualism, University of Vigo, October 23-26. Gadet, F. (2007). La variation sociale en français. Paris: Ophrys.

Galisson, R. (1980). D’hier à aujourd’hui la didactique générale des langues étrangères: Du structuralisme au fonctionnalisme. Paris: Clé International. Gauci, J. (2016). L’alternance codique en classe de FLE dans des collèges maltais : représentations et constatations. Bachelor of Education (Hons.) dissertation, unpublished, University of Malta. Judge, A. (1993). French: a planned language? In C. Sanders (ed.), French today: Language in its Social Context (7-26). Cambridge: CUP. Maarfia, N. (2008). L’alternance codique en classe de français en deuxième année primaire: Entre fonction communicative et fonction didactique. Synergies Algérie , 2 , 93-107. Mati, N. (2013). Entre deux langues. L’alternance des codes comme aide à l’appropriation de la langue de l’autre. Sciences Humaines Combinées [online], 12. Proceedings from The Université de Bourgogne Interdoctoral Conference

2013. http://revuesshs.u-bourgogne.fr/lisit491/document.php?id=1096 Molander, L. (2004). L’alternance codique en classe d’immersion : délimitation, interprétation et fonction interactionnelle. Sociolinguistica 18 (1), 86-105. Moore, D. (1996). Bouées transcodiques en situation immersive ou comment interagir avec deux langues quand on apprend une langue étrangère à l’école. AILE 7 , 95-121. Soku, D. (2014). Alternance codique en classe de FLE : Raisons d’ordre pédagogique chez les enseignants et facteurs de motivation chez les apprenants. Developing Country Studies, 4 (20), 1-10. Stolz, J. (2011). L’alternance codique dans l’enseignement du FLE. Etude quantitative et qualitative de la production orale d’interlocuteurs suédophones en classe de lycée. Göteborg : Linnaeus University Press. Turnbull, M. (2001). There is a role for the L1 in second and foreign language teaching, but … The Canadian Modern Language Review , 57 (4), 531-540. Turnbull, M. (2006). Employons le français en français de base! The Canadian Modern Language Review , 62 (4), 611-628. Yiboe, K. T. (2010). Enseignement/apprentissage du français au Ghana : écarts entre la culture d’enseignement et la culture d’apprentissage (Doctoral dissertation). Université de Strasbourg, Strasbourg. http://scd-theses.u-strasbg.fr/2419/