© 2026 McGill Journal of Education / Revue des sciences de l’éducation de McGill. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0).

LANGUAGE CONTACT AWARENESS IN A MULTILINGUAL ENVIRONMENT: THE PERCEPTION OF CROSS-LINGUISTIC SIMILARITIES BY PRE-SERVICE ENGLISH EFL TEACHERS

GUILÈNE RÉVAUGER University of Reunion Island

Whether one considers languages as resources, fundamental rights, or, inversely, as impediments, the stakes are high. Some fundamental ideological and methodological disputes reflect a wide diversity of challenges and contexts. Today, the focus is no longer on bilingualism, but on multilingualism and plurilingualism. In the Common European Framework of Reference for Languages (CEFR), “multilingualism” refers to the environments in which many languages co-exist and “plurilingualism” focuses on the individuals. In other contexts, however, “multilingualism” is broadly used for both. Twenty-first century developments in language learning are characterized by a positive reassessment of multilingualism, the consideration of language interactions within “the multilingual mind” (Siemund, 2023, p. 64), and the emergence of ecological approaches. The traditional barriers erected between languages thus appear to be breaking down.

The questions I ask in this article are not whether or why a holistic plurilingual approach should be more coherent than a monolingual approach, or whether languages could either bring on social determinism or, conversely, trigger equal opportunity. Rather, multilingualism is taken as a genuine reality within Reunionese classrooms, and our goal is to tackle difficulties as they arise and to nurture cross-linguistic similarities to facilitate language teaching. There is, however, one important prerequisite to this improved pedagogy: identifying linguistic contacts. Therefore, this study explores prospective English teachers’ perceptions of similarities. It investigates whether cross-linguistic connections are made, and at what linguistic levels. Being the foreign language most widely taught in Reunionese schools, English may have a particular role to play as a bridge language (Forlot, 2009; Robert, 2008) in linguistic, psycholinguistic, and metalinguistic terms. Hereafter, the focus will be on the connections between English and a selection of other languages present in the multilingual Reunionese environment.

LOOKING AT CROSS-LINGUISTIC SIMILARITIES

Perspectives

Studying cross-linguistic similarities from a didactic perspective implies comparing languages and identifying points of convergence and divergence. It is an ambitious enterprise involving different sciences and carried out in the continuity of debates on foreign language learning discussed from the 17th century onward. “Only care is to be taken whilst he is learning these foreign languages, by speaking and reading nothing else with his tutor, that he do not forget to read English,” recommended Locke (1693/1996, p. 120). Monolingualism, bilingualism, and multilingualism were explored first by psychologists (Siemund, 2023). Strongly rejected and devalued at first (often within the historical context of imperialism and acculturation strategies), bilingualism has been subject to reassessment and valourization since McGill University’s Professor Wallace E. Lambert and graduate student Elizabeth Peal’s seminal studies (in six Montreal schools) demonstrated the general intellectual advantage of bilinguals (Peal & Lambert, 1962).

Contrastive linguistics emerged with Weinreich’s (1953) and Lado’s (1957) studies on language contacts. Comparing the micro-systems of two or more languages, it sprang from applied linguistics and was practice oriented. Early contrastive studies focused on the negative transfers which occurred when an L2 (second language) was learnt, and error analyses were conveyed to identify L1-based (first language) deviations. It was admitted, first, that the more different the L1 and L2 were, the harder it would be to learn the L2, and, second, that errors could be predicted. However, the widest gaps did not eventually appear the most difficult to breach. Due to insufficient metalanguage, and since certain elements bore little comparison, the enterprise proved trickier than expected. Contrastive linguistics was rejuvenated in the 1970s when scientists put aside error prediction and made use of developments such as Selinker’s (1972) interlanguage theory.

Contemporary studies on cross-linguistic similarities are now grounded in contrastive and comparative linguistics, language typology, psychotypology, psycholinguistics, sociolinguistics, metalinguistics, language awareness, studies on language acquisition from the perspective of multilingual speakers, and studies on the cerebral activity of polyglots by neurolinguists. Recent studies on second language acquisition and foreign language learning have called for attention to environmental, sociocultural, and socioecological factors (Beatty-Martínez & Titone, 2023).

The absence of tabula rasa

Rather than starting from scratch, language learners rely on what they already know to try and connect new knowledge (Ringbom, 2006). The notions of “transfer,” “influence,” and “interactions” refer to these connections. The term “transfer” was first coined by behaviourist psychologists. As it is associated to a great variety of theories, its suitability is questionable (Odlin, 1989; Ringbom, 2006). The notions of “cross-linguistic influence” and “cross-linguistic interactions” are today favoured as more neutral and more encompassing terms (Siemund, 2023). Although we prefer the concept of cross-linguistic similarities to that of transfer, the two terms are here used interchangeably. Rather, it is the notion of transfer per se which is of interest to us. Its raison d’être may however be compromised if one admits the existence of the individual’s holistic plurilingual repertoire rather than the presence of juxtaposed elements which need to be transferred.

Multilingualism initially conveyed threats of “subtractive multilingualism” (when languages are learnt at the expense of the first language / languages) and “semilingualism” (little proficiency or deficiencies in all languages) rather than the optimistic prospect of “additive multilingualism” (when language learning develops in all the languages). When research began focusing on what languages had in common, the aim was more often than not to identify language universals. This objective had little in common with the practice-oriented approach of contrastive analysis. Early contrastive studies focused on negative transfers from L1, and in particular on the risks stemming from overgeneralization. The concept of “semantic equivalence hypothesis” assumed that L2 conceptual patterns and linguistic codes relied on those of the L1 (Hasselgren, 1994; Ijaz, 1986).

The tide is ebbing today as facilitation is increasingly studied. Yet, a review of the literature still shows the scarcity of studies on positive transfer. The scarcity of research on convergence and facilitation rather than divergence and interference may be explained by methodological difficulties which remain relevant today: the identification of facilitative influence is delicate, transfers are different according to the activities performed (some cross-linguistic similarities may be very useful when listening, but less relevant when speaking), and some positive transfers probably occur even when no similarity is perceived (Kellerman, 1995; Ringbom, 2006).

Research now presumes that learners first search for similarities between languages. Therefore, Chaudenson (2008) recommends making use of similarities between Reunionese Creole and French, in the early classes of Reunionese schools, before later having children identify differences. For Ringbom (2006), the initial focus of learners is on formal cross-linguistic and intra-linguistic similarities (what is already known of the new language); formal similarities are then associated to semantic and functional similarities. Difficulties arise when one considers the perception of similarities as a biased process. Perceiving similarities can be considered as a personal enterprise which may suffer both from a lack of objectivity and accuracy.

Researchers now consider the mutual influence of languages in both L1–L2 and L2–L1 directions. The study of language acquisition has extended its scope to third language and additional language acquisition, as cross-linguistic influence (CLI) also occurs for L3 learning (Ecke, 2015) and subsequent languages. Current CLI models in L3 acquisition studies include the L1 transfer model, L2 status factor model, the typological primacy model, the linguistic proximity model, the scalpel model, and the cumulative enhancement model. Cumulative enhancement illustrates the change of tide within the historical reassessment of multilingualism, as it considers facilitation alone and acknowledges no negative inference in L3 acquisition (Flynn et al., 2004).

THE REUNIONESE LINGUISTIC CONTEXT

In 2018, 706,400 inhabitants lived on Reunion Island, among which 99,230 were born in mainland France, 38,160 in another country in the Indian Ocean, and 12,200 elsewhere. The great majority of Reunionese were thus born on the island (82.5%). The official language and language of instruction is French, yet Reunionese Creole is largely spoken. Reunionese Creole is a French-based creole with Malagasy, Hindi, Portuguese, Gujrati, and Tamil influences (Chaudenson, 1992). Migratory flows originate from France, Mayotte, Madagascar, Mauritius, Comoros, Morocco, Sri Lanka, Tunisia, Algeria, India, China, East Africa, and so forth. Hence, Reunionese Creole and French are in contact with Tamil, Chinese, Hindi, Gujrati, Shimaore, Comorian, Kibushi, Malagasy, and Arabic, amongst others.

In 2020, 90% of Reunion Island natives considered the Creole language to be important for the region’s identity (Insee, 2022). According to that same study, 90% said they easily understood and spoke Creole; 49% expressed that Creole should be taught in schools (compared to 40% in 2010), whereas 35% expressed that it should be restricted to the private sphere (compared to 47% in 2010).

The foreign languages the most taught in secondary schools are English and Spanish. In September 2020, there were 8,088 students enrolled in their final year of general or technological education in private and public schools. Of those, 95% had chosen English as their first foreign language, 3.1% Spanish, 1.3% German, and 0.6% Chinese. However, in total, whether the language was chosen as a first, second, or third option, 100% learned English, 78.7% Spanish, 18.2% German, 3.2% Chinese, 0.3% Arabic and Creole, 0.2% Tamil, 0.1% Italian, 2.5% Latin, and 0.4% Greek (Division des structures et des moyens, 2020). Reunionese Creole is taught in primary and secondary schools, without being compulsory. Its status varies: Creole may be (a) a school subject (e.g., Créole option langue vivante régionale), (b) the language of instruction (e.g., through content and language integrated learning programs / enseignement d’une matière intégré en langue régionale and enhanced dual-language programs / parcours bilingues), (c) a support or developmental language / langue d’appui (e.g., early education in areas where Creole is predominantly spoken).

The recommendations of the CEFR are followed, a task-based approach is required, and languages are conceived as vehicles for culture, which may partially explain the apparent scarcity of linguistic reflection carried out in language classes. Rethinking instructional strategies in multilingual classrooms appears necessary, as a monolingual tradition of foreign language teaching prevails (see Cummins, 2007).

CROSS-LINGUISTIC SIMILARITIES

Languages under scrutiny

Particular attention is paid in this study to the relationships between English and French, Reunionese Creole, Morisyen, Spanish, German, and Shimaore. Similarities with Korean were succinctly looked at, thus helping consider seemingly unrelated languages. Numerous studies have shed light on language learning and teaching considering the relationships between English and French. Research in this field is heavily indebted to the studies on bilingualism and dual language education programs carried out in the linguistic landscapes of Canada since the 1960s (Cummins, 2000, 2007; Duff, 2007). Researchers have also scrutinized the relationships between English and German (Hufeisen & Neuner, 2003), English and Spanish, French and Reunionese Creole (Gaillat, 2014; Lebon-Eyquem & Adelin, 2015; Souprayen-Cavéry, 2020), and English and Morisyen. There are, to a lesser extent, studies on Reunionese Creole and English within an educational context (Rolland, 2013, 2014; Sournin, 2010). However, we have found little on Shimaore and English, which may be due to apparently limited historical and linguistic connections.

According to historical and comparative linguistics, English, French, Spanish, and German all belong to the Indo-European language family and share the common reconstructed Proto-Indo-European protolanguage. The Indo-European family is divided into eight subfamilies, including the Germanic languages to which German and English belong, and Romance / Latin languages, which include French and Spanish. Korean belongs to the Koreanic language family and bears little similarity to English. The phonological system is fundamentally different, and Korean follows a subject–object–verb canonical sentence order.

Shimaore (also known as Maore Comorian and Mahorais in French) is an indigenous language of Mayotte, a French overseas department and region contested by the Union of the Comoros. Shimaore is a dialect of the Comorian language, which is a Bantu language belonging to the Niger-Congo hypothetical language family and sharing the common Proto-Niger-Congo protolanguage. Shimaore is a stable language widely spoken as a first language. Traditionally an oral language, since 2020 its official orthographies are both in Latin and Arabic scripts. Shimaore is an agglutinative language making use of multiple morphemes; it has an elaborate system of noun classification (noun classes or nominal classes, with singular or plural prefixes) and does not rely on the use of articles. Like Spanish, Shimaore phonology follows a five-vowel system, and it includes specific consonants such as the B-hook (/ɓ/), a voiced bilabial implosive also found in the African alphabet (Mori, 2023). In contrast, French has 13 oral vowels and four nasal vowels; English comprises at least 14 monophthongs and diphthongs, and German has a 15-vowel-sound system, although these numbers may vary depending on the dialects.

Morisyen is spoken in Mauritius. Although a French-based creole language, it has been in contact with English. Colonized by the Netherlands then the French, Mauritius was taken by the English in 1810; when slavery was abolished, intensive use was made of indentured labourers from China, Malay, Africa, Madagascar, and then, to a larger extent, India. The name “Morisyen” is preferred to “Mauritian Creole,” as it appears more inclusive in Mauritius, where the term “creole” is perceived as excluding Indo-Mauritians. Some linguists consider that Reunionese Creole influenced Morisyen (Chaudenson, 1992); however, this analysis is not shared by all (Baker, 1972). Morisyen and Reunionese Creole are both classified within the Bourbonnais Creole or Mascarene Creole group, which comprises the French-based creoles of the Indian Ocean. Both Reunionese Creole and Morisyen have a rather straightforward grammar and follow a subject–verb–object canonical word order. Morisyen verbs are invariable, and tenses are indicated by particles or the context. Watbled (2021) considers Morisyen more distant to French than Reunionese Creole, and Chaudenson (1992) reckons that there is little intercomprehension between Reunionese Creole and Morisyen speakers.

The relationships between Reunionese Creole and French have been subject to many studies — because of their proximity, and to tackle educational challenges and prejudiced representations. According to a 2021 report produced by Lofis la lang kréol La Rényon, 29% of the respondents to a survey (n = 503) totally agreed (“total d’accord”) and 12% strongly agreed (“tout à fait d’accord”) on the fact that Creole is an obstacle to learning French (“Le créole est un frein l'apprentissage du français”). The belief however appears on the decline considering that a decade prior in 2009 43% totally agreed and 24% strongly agreed (SAGIS, 2021).

Analyzed by researchers first through the prism of diglossia (although the two languages coexist, Creole is assigned an inferior status to French), Creole and French were then positioned at each end of a continuum of practices (Carayol & Chaudenson, 1979). The continuum of practices paradigm provided for an implication analysis which took into account intermediary states. The notion of “interlect” was then introduced, and it endorsed intertwined interlectal forms and mélanges (mixing; Prudent, 1981; Lebon-Eyquem, 2016). Code-switching, translanguaging, and mélanges are frequent in Reunion; these practices, the proximity between the languages, and language variation make the formal identification and categorization of French and Creole elements challenging (Lebon-Eyquem, 2016).

The nature of cross-linguistic transfers

Cross-linguistic transfers are manifold: They comprise linguistic but also psycholinguistic and metalinguistic transfers (which include cognitive, socio-affective, and metacognitive strategies). In the context of CLI studies, Siemund (2023) distinguishes between grammatical and lexical knowledge, and language proficiencies. At this point, we are convinced of the central role of metalinguistic transfers in language learning (Révauger, 2023). However, these transfers are not the subject of the present article. Ringbom (2006) identifies three types of cross-linguistic relations: a similarity relation (either partial or full), a contrast relation, and a zero relation. Formal similarity is not a prerequisite, as functional and semantic similarities may occur despite the absence of any apparent similarity of forms, as is often the case for unrelated languages.

We will here briefly look at some examples of phonological, lexical, and grammatical similarities. Formal phonological similarities may be found in the pronunciation of sounds; although English has a certain number of specific sounds (for example /θ/, /ð/, /ə/, /i:/), French, Creole, and English have a great number of consonants and vowels in common (/e/, /ɪ/). Similarity may also occur at other levels, such as sentence stress, intonation, emphasis, and questions.

Language comprehension first and foremost relies on words (Hilton, 2022), and researchers have described lexical integration processes (Ecke & Hall, 2014; Laufer, 1990; Ringbom, 2006). The parasitic model of vocabulary acquisition for L2 and L3 learners describes processes based on prediction at the three stages of the lexical integration process (Ecke & Hall, 2014). During the initial identification phase, learners rely on their previous knowledge and, to a large extent, on scripted and phonological similarities. Thereafter, semantic and/or functional similarity or even identity is expected. Words identified thanks to cross-linguistic scripted or phonological similarities are often called “cognates” or “transparent words.” According to its Latin origins (cognatus), the term should normally be reserved for words “of common descent,” that is to say words which have a common heritage but which may differ in meaning. Cognates should be considered as cousins rather than twins. Sometimes their similarity may not be blatant (as for “mater,” “mutter,” “mother,” and “mère”). Researchers have studied the positive impact of cognate status on word recognition, translation, and learning, and have identified what is known as the cognate advantage, and the cognate facilitation effect (Dijkstra et al., 1999; de Groot, 2011; Marecka et al., 2021). Think of the cardinal number “six.” It translates as “sis” in Morisyen, “siss” in Reunionese Creole, “sita” in Shimaore, “sechs” in German, “seis” in Spanish, “sei” in Italian, and “sex” in Swedish and Latin; it is also pronounced “sitta” in Arabic and “sei” in Basque, a seemingly isolate and unrelated language. The origins of the words are traced to the Proto-Indo-European “*swéḱs,” Proto-Semitic “*šidṯ-,” and Proto-Germanic “seks.”

The degree of transparency varies. Lexical cross-linguistic similarity may be total, both morphologically and semantically; a partial morphological similarity may be associated to a total semantic correspondence; and both the morphological and semantic correspondences may be partial (Caddéo & Jamet, 2013). In addition, there are a certain amount of “false cognates,” “pseudo-cognates,” “false friends,” and “interlingual homographs / homophones” responsible for semantic opacity. They may show total or partial morphological similarity but diverge in meaning. Swan and Smith (2001) identify “worse friends” as “unreliable friends,” words meaning almost the same, or “the same in one context and not in another” (p. 69).

The similarities may also be grammatical, and for instance, syntactic. “Word order” is used in syntactic typology to classify languages according to the canonical order of words in a sentence. Although it helps identify convergences and divergences, word order is of course a complex notion. It is the placement of groups of words, words, or morphemes which is described, according to the subject, verb, or object functions they are endowed with. The word order of simple, affirmative, declarative, and non-emphatic sentences is a subject–verb–object order in French, English, German, Spanish, Reunionese Creole, Morisyen, and Shimaore (“The cat eats the mouse”; “Le chat mange la souris”), as is the case for most Indo-European and many African languages. Yet, these are merely canonical patterns. Spanish often omits the subject. Shimaore uses inflections, and the word order is therefore easily subject to change. In French, the use of determiners, adjectives, and nominal or pronominal objects imply different word orders (“Je mange la pomme” but “Je la mange”). Hence, although common canonical patterns can be found, it is often hazardous to assume word order similarity.

Nurturing cross-linguistic similarities within the classroom

From there, it appears necessary to discuss similarities within the classroom. As far as vocabulary is concerned, it seems rational both to recognize the presence, at different degrees, of phonological, morphological, or semantic correspondences, but also to acknowledge their limits. Nation (2013) advises teachers to limit the teaching of vocabulary and encourage its incidental learning, which also raises the question of the implicit or explicit instruction required (Veronique, 2019). Although recent studies have shown that awareness of cognateness does not necessarily enhance the learning of cognates (Otwinowska et al., 2020), researchers seem to agree on the fact that attention to metaphonology (Rolland, 2014) and metalinguistics facilitates many learning and production strategies (Chaudenson, 2008; Ringbom, 2006; Swan & Smith, 2001). The role of teachers should not be underplayed, and at least two fundamental conditions need to be met to ensure such teaching. The first one is ideological: Teachers need to be willing to acknowledge the plurilingual repertoire of their students and to nurture transfers. As rational human beings, they will not do so unless staunch benefits are expected. The second condition is practical: Beforehand, teachers need to be able to identify language contacts and integrate them within lesson plans.

RESEARCH QUESTION & METHODOLOGY

Research questions

This article’s focus is on the identification of similarities, as it appears to be a prerequisite. The central research question asks: To what extent are pre-service teachers able to detect similarity between English and other languages? There are three subsequent questions:

1. At what linguistics level is similarity detected?

2. What language proficiency level is required to identify similarity?

3. What pedagogical implications can be inferred from the data?

Participants

The participants were 19 pre-service English teachers enrolled in the first year of a master’s degree program in secondary education (Master MEEF anglais) at the Institute of Education of Reunion Island. The research began with the analysis of anonymized presentation posts in a forum written during a hybrid course in language didactics, which gathered students specializing in either English, Spanish, German, or Creole. We selected the 19 posts written by the students specializing in English, between August 23rd and October 30th, 2023. Only the posts of the students specializing in English were selected, as the same students, minus one, were to take part in the following step of the research (four contrastive exercises addressing cross-linguistic similarities). Sixteen students then carried out the third step (group work identification of transfers).

Procedure

Step 1: The first step consisted of a preliminary analysis of the language biographies of the participants. The objective of this biographical approach was to gather contextual elements and identify potential points of study.

Before posting on the forum of the course, students were asked to present themselves through language biographies. The instructions were written in French (“Avant de participer au forum, présentez-vous en nous donnant des éléments de biographie langagière!”) and all students answered in French. Therefore, the Step 1 quotations below were translated by me. The posts were anonymized and subject to textual data thematic coding through the atlas.ti (https://atlasti.com/) software. Each post was numbered S1 to S19, according to the date of publication (S1 being the first post published). Deductive coding was used to identify the languages mentioned; it was then enriched with inductive open coding, together with sentiment and word-frequency analyses. Forty-two language codes and 12 thematic codes were generated. The posts were expected to show bias, as the last presentations may have been influenced by earlier ones, the biographies presented by the students implied subjectivity, and the students may have conformed to what they considered the teacher’s or peers’ expectations.

Step 2: The objective of the second step was to gather data on the identification of cross-linguistic contacts. Four contrastive exercises were given in an 80-minute time span. These contrastive exercises were inspired by an activity proposed by Bailly et al. (2009). The instructions were in English. The first exercise asked the students to highlight, in a 270-word excerpt from the beginning of Roald Dahl’s Charlie and the Chocolate Factory (1964), “everything which [reminded them] of the French language (this may include words, grammar, phonology …)” and “explain, at least briefly, why [they had] highlighted these elements.” Inversely, in the second exercise, students had to highlight in a 304-word excerpt from the opening page of Antoine de Saint-Exupéry’s Le Petit Prince (1943), “everything which [reminded them] of English.” The same instructions were given in the third exercise, although this time another language could be chosen among Reunionese Creole, Spanish, German, Malagasy, Comorian, Morisyen, or Japanese. Students were orally encouraged to choose any other language. Two complementary questions were asked (“Do you want to tell me why you’ve chosen this language?” and “Which level of proficiency do you think you have in this language?”). The last exercise asked students to “find on the internet a text in the other language [they had chosen] (article, song lyrics …). This document [had] to be authentic (it [could not] be a translation), and in a rather ordinary language.” Students were requested to “a) copy and paste the document … and indicate its source; b) highlight … all the elements which [reminded them] of English. Explain, at least briefly, why [they had] highlighted the elements.” The 18 students handed back a document which was then coded manually with atlas.ti to fulfil a qualitative analysis. In total, 887 quotations were analyzed. Inductive coding was used in the hope that generating codes from scratch and using the qualitative data itself would offer a more thorough analysis than deductive coding. Some codes were generated to cover the first sample; they were then applied to the next document, subject to modification when they did not match, and enriched by the creation of additional codes. The creation of new codes triggered the recoding of the samples already coded, implying an ongoing process which also attempted to prevent definitional drift. Identical occurrences were only counted once.

The coding frame included 284 codes which eventually permitted:

• language identification;

• the identification of the level of proficiency declared;

• the identification of the nature of the citation (<single word> was used to identify isolated words, separated by white space; <more than one word> was used when more than one word was highlighted, including nouns preceded by pronouns, nouns preceded by elided prepositions [e.g., “d’explications”], nouns followed by numbers [e.g., “numéro 1”], phrases [e.g., “six ans,” “vyé gramoun”], and sentences [e.g., “Their names are Grandpa Joe and Grandma Josephine”]; the code <morpheme, word part, punctuation> was also used);

• the identification of the word(s) cited;

• the identification of the nature of the similarity (a hierarchical frame was used when contacts were explicitly mentioned — top level: <explicit contact explained by student>; middle level: <grammatical structures or word order>, <phonology>, <word resemblance>; third level: <similar plural>, <word formation (construction, affix)>, <etymology>, <word order-adjective> …);

• the identification of similarities which were, in our opinion, questionable.

Step 3: In a third step, the students formed groups according to the languages chosen and, as homework, drew a “list [of] positive and negative transfers between [the language chosen] and English.” They were asked to do so by reflecting upon their own experience of the previous exercise, by predicting transfers learners could possibly make, and distributing these transfers into different categories. Group documents were handed back after a week, tackling the relationships between English and German (four students), English and Spanish (five), English and Korean (one), English and Shimaore (one), and English and Creole (five). The documents were subject to manual inductive coding. Thirty-nine quotations were looked at through 35 codes. The answers of Steps 2 and 3 were then compared.

RESULTS

Language portraits

The coding of the presentation posts revealed that all 19 students mentioned English in their language biographies. Seventeen students mentioned Reunionese Creole, 16 mentioned French, 15 German, 14 Spanish, nine Japanese, six Korean, six Portuguese, six Morisyen, five Arabic, five Latin, five Mandarin, four Russian, three Haitian Creole, three Comorian, three Malagasy, three Italian, three Martinican Creole, two Ancient Greek, two Thai, two Chinese, two Polish, two Breton, and two Picard. There was only one occurrence of Afrikaans, Australian, Bal, Corsican, Gaelic, Hungarian, Irish, Kanak, Kichwa, Kiswahili, Māori, Seychelles Creole, Shibushi, Shimaore, Sinhalese, Swedish, Tagalog, and Urdu. The biographies comprised a total of no less than 42 different languages. The lowest number of languages mentioned by a student was four (S16), the highest 23 (S2). Twelve students mentioned between four and 10 languages, and seven mentioned between 10 and 23.

Points of interest

The data set revealed the importance of the notion of choice (being able to choose or not one’s languages), as mentioned by seven students out of 19. It also shed light on a high number of sociolinguistic reflections on Creole and French (13/19 students):

Although a few Creole words occasionally slipped into our interactions, my parents — perhaps regrettably — preferred the national language, which they conventionally saw as superior to Creole. (S4)

At school, French became more than just a mother tongue, it was the train to take for great discoveries and the imagination. (S10)

Creole is my mother tongue, which I use mainly to communicate with my family and occasionally in the street. French is also my mother tongue, but unlike Creole, it was more widely used at school and in various institutions and businesses. (S3)

The results acknowledged translanguaging (S16: “When I'm with my close family, I tend to speak Creole, but the switch to French happens very quickly), and mélanges:

First of all, at home and in my family circle, I very often speak Creole; but I can resort to French if a member of my family isn't used to Creole … I often mix the two without realising it. (S11)

There was also a recurrent mention of everyday exposure to language (8/19), and of the part played by music, TV series, films, or social networks (15/19). A sentiment analysis showed that learning a language was associated, in varying degrees, to positive or negative experiences (S8: “I fell in love with this language;” S19: “I'd say I'm angry with Chinese”). Four students reported negative experiences, while 13 mentioned positive sentiments. Finally, the notion of usefulness was also recurrent (5/19): “As far as Latin is concerned, it has served me well (and still does) in terms of etymology both in French and English” (S4), and “Spanish has come in handy on trips to Spain and Peru” (S6).

Nature of language contacts

Most contacts were identified at the word level, as, in total, 607 single words were highlighted; 211 citations were composed of more than a word and thus evoked grammatical similarities. Eight morphemes, word parts, or punctuation marks were highlighted.

When they read Charlie and the Chocolate Factory in English, students found more grammatical and syntactic similarities with French Creole, German, and Spanish than the other way round (Exercise 4). In Exercise 1, 48.8% of the citations highlighted were groups of words recalling Creole, and 41.9% of the citations were single words. In comparison, 10.8% of the citations highlighted in Exercise 4 were groups of words recalling English, and 84.6% were single words. A similar trend was identified with German and Spanish. The number of single words identified for French largely exceeded the number of groups of words, without any major difference according to the way of the translation (English recalls French: 24.6% of groups of words, 68.8% of single words; French recalls English: 25.8% of groups of words, 73.5% of single words). A partial explanation for such discrepancy may reside in the documents themselves, or the fact that this exercise was the last one to be done.

The students explicitly named the similarities of 50.2% of the total quotations. Of these explanations, 31% identified grammatical similarities. Specific mentions were made of word order, the position of adjectives, prepositions, comparative forms, similar introduction of subordinate clauses, similar structures (S4: “Elle a bien besoin d’être consolée, similar to Be+en”), similar distinctions (S17 translated “This is Mrs Bucket” as “Unu de Monyé/Mzé Bucket” and explained that “in Mahorese we also have the demonstrative pronoun ‘uwo’ which reminds me of the distinction of proximity between this/that”), and similar negative forms (S11: “As for the construction of negation, it is also the same between both languages [English and Reunionese Creole]: V + negative marker → ‘wasn’t’ or ‘was not’ is translated by ‘lé pa’ whereas in French the negative markers are located before and after the verb ‘ne sont pas.’”).

Phonology was mentioned only in 7.4% of the citations (S7: “These words are very similar in phonetics and etymology [German-English]. Ex: Töchter/daughter; Sohn/son; Nacht/night”; S11: “Phonetically, ‘foutbol’ [Creole] is ‘football’ and the meaning is still the same; S8: “Phonetically, to live = vivre, consonant sound /v/ similar”).

Of the similarities explicitly identified, 67% implied word resemblance, among which 5% referred to formation and/or affixes. Most words showed total morphological and semantic similarity (“six,” “image,” “large,” “question”); others had partial morphological and total semantic similarities (“house” / “Haus” in German; “Family” / “familia” in Spanish; “uncomfortable” / “inconfortable” in French). The identification of the resemblances of “mattresses” / “matelas” (S1); “night” / “nuit” (S6); “name” / “nombre” (S9); or “empleadas” / “employees” (S16) seem to prove that etymological links are identified even when formal likeness may not be apparent at first sight. Few words showed partial morphological and semantic correspondence (“histoires”/ “history”). The top 10 words identified per language revealed similarities across many languages (such as “six”). Morisyen, Shimaore, and Korean are excluded from Table 1, as only single occurrences were obtained — English recalls Morisyen: “so” (1), “to” (1), “you” (1); English recalls Shimaore: “family” (1), “Grandma” (1), “Grandpa” (1), “but” (1), “this” (1); and English recalls Korean: “Mr or Mrs” (1), “they” (1), “bed” (1), “he” (1), “the house” (1).

TABLE 1. Top 10 words identified (excluding proper nouns or single occurrences)

Other similarities also included sociocultural and formal items: punctuation (question marks, inverted commas), a chapter number and title, letter structure (S14: “À Léon Werth, structure similar to French, ‘to whom’ implies the nature of the document, an address or a letter”), title prefixing (Mr/Mrs), self-introductions (S9: “The formality of this phrase is somewhat reminiscent of French politeness in introductions”), and capital letters and proper nouns (S14: “Both ‘Grandpa’ and ‘Grandma’ are quite alike their French counterparts in terms of etymology, the fact that they’re followed by a name may be relevant as well”; S5: “‘Mr Bucket’ … would be ‘Mr. Bucket’ in French too. These nouns have capital letters at the start which is similar to French and makes us understand that these are names of characters”).‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬

Relevance

Although most resemblances appeared coherent, the relevance of 13.5% of the citations explicitly justified was questioned. We doubted relevance mostly when the explanations given did not seem adapted to the transfer identified, when the metalanguage used was irrelevant, or when more than two sentences were highlighted in a block, lacking precision. Without further details, we also doubted the relevance of the identification of conjunctions such as “and,” “because,” “but,” the French preposition “dans” (S5), or “the” and “he is” (S11).

Some students defined syntactic similarities, sometimes offering possible translations to prove the parallelism. S13 thus wrote that “This is Charlie” followed the “same *lexical structure” in French. However, if “C’est Charlie” would be grammatical, it sounds unidiomatic in this context (unsurprisingly the French translation of the book preferred “Voici Charlie”).‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬‬

We question the word resemblance of “people” / “demoune” (S9, Creole) (countability rather than meaning may be the source of the resemblance identified). The transparency of “old people” was also mentioned in French. However, the French use of “people” (“la presse people,” People magazine) refers to stars, people in the news, and gossip press; it therefore has another meaning in another context. The equivalence of “He is pleased to meet you” / “Fait longtemps pas encore vu aou” (rather, “long time no see”), and “better house” / “caze lontan” (S9: rather, “house of yesteryear”) also seem highly questionable.

Thus, resemblance was sometimes admitted despite little formal similarities or important semantic differences.

Language proficiency

The four students who chose German declared A1 (2 students), A2 (1), and B1 (1) levels of proficiency; the six students who chose Creole: B1+ (1), B2 (1), C1 (1), C1+ (1), C2 (2); and the five students who chose Spanish: A1 (1), A1+ (1), A2 (1), B1 (1), B1+ (1). Three students focused on Shimaore (C1-), Morisyen (A1), and Korean (A1). Although the small-size sample calls for precaution and hinders comparison, data analysis did not show important variations according to proficiency. Whatever the languages, similarities were identified in Exercise 3 despite low levels of proficiency. It thus appears that no strong expertise is required to identify cross-linguistic similarities, which seems to corroborate the theses supported by CLI models considering similarity detection. The Sankey diagram (Figure 1) thus shows no divergent trends depending on proficiency.

FIGURE 1. Language proficiency and language contact identification in Exercise 3

Group identification of similarities

Step 3 revealed that students identified slightly more positive transfers (23/39) than negative transfers (16/39). Figure 2 offers an example of the list produced by the students who looked at Reunionese Creole / English influence. Although short, this list, compiled by six students, comprises more potentially positive transfers than negative ones.

FIGURE 2. List provided by the group studying Reunionese Creole/English influence

FIGURE 3. Step 3 transfer identification — group work

Throughout all languages, 12 positive transfers referred to grammar, eight to vocabulary, but only one to phonology; there were five grammatical negative transfers identified, eight vocabulary negative transfers, and two phonological difficulties. The relevance of four items was questioned.

DISCUSSION

The previous results offer avenues for discussion. Our central research question asked to what extent pre-service teachers were able to detect similarity between English and other languages. We also wondered at what linguistic level similarity was detected, what level of proficiency was required to identify similarity, and what pedagogical implications could be inferred.

First, the students showed multilingual awareness (S11: “I'm passionate about discovering and learning new languages, and today I realise that I've already been in contact with many languages thanks to this language biography”; S19 called themselves “a silent polyglot”). The linguistic profiles of pre-service English teachers on Reunion Island are remarkably plurilingual; although this was expected, the number of 42 languages mentioned by 19 students exceeded all predictions. Some pre-service teachers appeared to be true language lovers, while others mentioned languages, such as Kichwa, we would not have guessed. The unapparent, sometimes hidden, knowledge of languages of pre-service teachers, and by extension students, should therefore not be underestimated. Most participants declared they had a relatively high level of proficiency in Reunionese Creole. With hindsight, we question the relevance of differentiating Creole and French similarities. Considering the proximity between the two languages, some English items could merely recall French as well as Creole. Yet, the data collected through the testimonies confirmed that the students appeared cognizant of the presence of mélanges. It also shed light on the impact of language colonial history, of representations, and on different personal viewpoints as far as the regional language is concerned. Multilingual awareness was associated with awareness of the incidence of psychological factors on language learning. Many language portraits mentioned the usefulness of languages, as well as pleasant or negative experiences.

Second, the data showed that many cross-linguistic contacts were perceived; alas, it also confirmed that the relevance of the connections was at times questionable. In fact, these data reflect learners’ psychotypological beliefs and perceptions. The identification of similarities may sometimes be replaced by less grounded assumptions. However, such assumptions may be considered as normal aspects of the learning process; they may be revisited or abolished at a later stage (Ecke & Hall, 2014).

While some focused merely on transparent words, others offered detailed analyses and fine distinctions to appreciate connections. It seems students successfully apprehended positive transfers and identified more positive transfers than negative ones. Likeness stretched across different domains; it would be erroneous to consider that word resemblance only was spotted. In fact, a high number of grammatical connections were identified, with related and unrelated languages. The little mention of phonology probably stems from the fact that written documents were used; results would have probably been entirely different had the students been asked to perform listening tasks.

Yet, some missed even the most transparent words: Eight students missed the connection between the English and French words “six”; five did not identify any French connection for “family” (“family” / “famille”); and six for “large” (“large” / “large”). Some similarities may have been quickly defined, hence the subtle difference between “histoires” (stories) and “history” was not mentioned. Inversely, some similarities were mentioned despite important semantic difference: the “case lontan” (house of yesteryear) / “better house” example mentioned above seems to illustrate a dire need to find resemblance. Two words that sometimes go together in a language were associated, yet the cross-linguistic association thus made bore little similarity. Unsurprisingly, the usual risks of inference strategies may arise when one tries to find associations, and risk takers may draw conclusions despite little proof.

Third, it seems that nurturing connections remains a pedagogically difficult enterprise. If analyzing and grouping are not innate skills, they need to be acquired, maybe through explicit instruction. At this point, the only firm recommendation that can be given in terms of training and pedagogy is that, as is often the case with education, no turnkey solution exists. Whatever the direction, precautions need to be taken. Nurturing cross-linguistic similarities in the classroom is a process which implies many steps, and defects may occur at all levels. Pre-service teachers first need observation skills. In this study however, pre-service teachers appeared to successfully rely on similarity detection to form connections, as psycholinguistic models of vocabulary acquisition assumed. The teacher’s role then extends to students’ detection facilitation. Nurturing similarities may however remain incomplete without adequate metalanguage to describe connections and practical or reflective exercises to ensure students detect, correct, and consolidate their initial observations.

CONCLUSION

As all learners rely on similarities, and for all the reasons evoked above, acknowledging and discussing connections appear as fundamental assets to facilitate language learning. Typological similarity may empower learners, increase their confidence, and motivate language learning. However, acknowledging similarities or wanting to nurture them is not enough. Teachers seem to encounter practical difficulties first to identify them, and probably also to integrate them within the classroom. Making students detect difference and detail in apparently similar structures represents a real challenge for teachers. If plurilingualism is to be encouraged, the objective should then be two-fold: Teachers need to be convinced of its benefits — which to some extent relies on the perception of cross-linguistic similarities — and teachers need to be accompanied to integrate its teaching. This calls for the endorsement of more action-research projects.

We have here suggested that proficiency in all languages was not required to identify cross-linguistic similarities. Acknowledging the presence of a plurilingual repertoire may raise fears, as some are afraid the language teacher could become a jack of all trades, the master of none. Rather, reflective skills and an awareness of similarities and differences are needed. This does not compromise in-depth language teaching. Standing at a linguistic crossroads, the English as a foreign language teacher is in a prime location to look at multilingualism from a positive perspective.

REFERENCES

Bailly, S., Boulton, A., Chateau, A., Duda, R., & Tyne, H. (2009). L’anglais langue d’appui pour l’apprentissage du français langue étrangère. In G. Forlot (Ed.), L’anglais et le plurilinguisme : pour une didactique des contacts et des passerelles linguistiques (pp. 35–57). L’Harmattan.

Baker, P. (1972). Kreol: A description of Mauritian Creole. C. Hurst & Co Publishers.

Beatty-Martínez, A. L., & Titone, D. A. (2023). De-generacy as an organizing principle of bilingual language processing: Evidence from brain and behavior. In K. Morgan-Short & J. G. van Hell (Eds.), The Routledge handbook of second language acquisition and neurolinguistics (pp. 260–273). Routledge.

Caddéo, S., & Jamet, M.-C. (2013). L’intercompréhension : une autre approche pour l’enseignement des langues. Hachette.

Carayol, M., & Chaudenson, R. (1979). Essai d’analyse implicationnelle d’un continuum linguistique : français—créole. Collection IDERIC, 1, 129–172. https://www.persee.fr/doc/oeide_0762-2287_1979_ant_1_1_891

Chaudenson, R. (1992). Des îles, des hommes, des langues : langues créoles-cultures créoles. L’Harmattan.

Chaudenson, R. (Ed.). (2008). Didactique du français en milieux créolophones : outils pédagogiques et formation des maîtres. L’Harmattan.

Cummins, J. (2000). Language, power, and pedagogy: Bilingual children in the crossfire. Multilingual Matters.

Cummins, J. (2007). Rethinking monolingual instructional strategies in multilingual classrooms. Canadian Journal of Applied Linguistics, 10(2), 221–240. https://journals.lib.unb.ca/index.php/CJAL/article/view/19743

de Groot, A. M. B. (2011). Language and cognition in bilinguals and multilinguals: An introduction. Psychology Press. https://doi.org/10.4324/9780203841228

Dijkstra, T., Grainger, J., & van Heuven, W. J. B. (1999). Recognition of cognates and interlingual homographs: The neglected role of phonology. Journal of Memory and Language, 41(4), 496–518. https://doi.org/10.1006/jmla.1999.2654

Division des structures et des moyens. (2020). Tableaux statistiques n° 371 : langues enseignées dans les lycées : constats de rentrée 2020–2021. Région académique La Réunion.

Duff, P. A. (2007). Multilingualism in Canadian schools: Myths, realities and possibilities. Canadian Journal of Applied Linguistics, 10(2), 149–163. https://journals.lib.unb.ca/index.php/CJAL/article/view/19739

Ecke, P. (2015). Parasitic vocabulary acquisition, cross-linguistic influence, and lexical retrieval in multilinguals. Bilingualism: Language and Cognition, 18(Special Issue 2), 145–162. https://doi.org/10.1017/S1366728913000722

Ecke, P., & Hall, C. J. (2014). The Parasitic Model of L2 and L3 vocabulary acquisition: Evidence from naturalistic and experimental studies. Fórum Linguístico, 11(3), 360–372. https://doi.org/10.5007/1984-8412.2014v11n3p360

Flynn, S., Foley, C., & Vinnitskaya, I. (2004). The cumulative-enhancement model for language acquisition: Comparing adults’ and children’s patterns of development in first, second and third language acquisition of relative clauses. International Journal of Multilingualism, 1(1), 3–16. https://doi.org/10.1080/14790710408668175

Forlot, G. (2009). Le rôle de l’anglais dans les apprentissages linguistiques : appropriation et capitalisation d’une culture du contact des langues. In G. Forlot (Ed.), L’anglais et le plurilinguisme : pour une didactique des contacts et des passerelles linguistiques (pp. 7–33). L’Harmattan.

Gaillat, T. (2014). Didactique du français en milieu créolophone : Quels principes ? Quelles implications au niveau de l’enseignement apprentissage du vocabulaire à la Réunion ? Travaux & Documents, 46, 97–115. https://hal.science/hal-00907993v1

Hasselgren, A. (1994). Lexical teddy bears and advanced learners: A study into the ways Norwegian students cope with English vocabulary. International Journal of Applied Linguistics, 4(2), 237–258. https://doi.org/10.1111/j.1473-4192.1994.tb00065.x

Hilton, H. (2022). Enseigner les langues avec l’apport des sciences cognitives. Hachette Éducation.

Hufeisen, B., & Neuner, G. (2003). Le concept de plurilinguisme : apprendre une troisième langue — l'allemand après l'anglais. European Centre for Modern Languages.

Ijaz, I. H. (1986). Linguistic and cognitive determinants of lexical acquisition in a second language. Language Learning, 36(4), 401–451. https://doi.org/10.1111/j.1467-1770.1986.tb01034.x

Insee (2022). Panorama des évolutions de la société réunionnaise de 2010 à 2020 (Insee Analyses Réunion No. 79). https://www.insee.fr/fr/statistiques/6676745#onglet-1

Kellerman, E. (1995). Crosslinguistic influence: Transfer to nowhere? Annual Review of Applied Linguistics, 15, 125–150. https://doi.org/10.1017/S0267190500002658

Lado, R. (1957). Linguistics across cultures: Applied linguistics and language teachers. University of Michigan Press.

Laufer, B. (1990). Why are some words more difficult than others? Some intralexical factors that affect the learning of words. International Review of Applied Linguistics in Language Teaching, 28(4), 293–308. https://doi.org/10.1515/iral.1990.28.4.293

Lebon-Eyquem, M. (2016). Quelle prise en compte par l’école réunionnaise de la diversité des profils linguistiques de ses élèves? Cahiers de Linguistique, 41(2), 139–156. https://univ-reunion.hal.science/hal-01501103v2

Lebon-Eyquem, M., & Adelin, E. (2015). La traduction dans les classes réunionnaises : pratiques efficaces ou illusions pédagogiques? In C. Pic-Gillard & H. Ponnau-Ravololonirina (Eds.), Traduction-trahison : théories et pratiques : la traduction en contextes multilingues (pp. 165–188). Epica.

Locke, J. (1996). Some thoughts concerning education and of the conduct of the understanding (R. W. Grant & N. Tarcov, Eds.). Hackett Publishing. (Original work published 1693)

Marecka, M., Szewczyk, J., Otwinowska, A., Durlik, J., Forys-Nogala, M., Kutylowska, K., & Wodniecka, Z. (2021). False friends or real friends? False cognates show advantage in word form learning. Cognition, 206, Article 104477. https://doi.org/10.1016/j.cognition.2020.104477

Mori, M. (2023). The acoustic characteristics of implosive and plosive bilabials in Shimaore. Journal of the International Phonetic Association, 53(3), 950–976. https://doi.org/10.1017/S0025100322000184

Nation, I. S. P. (2013). Learning vocabulary in another language (2nd ed.). Cambridge University Press.

Odlin, T. (1989). Language transfer: Cross-linguistic influence in language learning. Cambridge University Press.

Otwinowska, A., Forys-Nogala, M., Kobosko, W., & Szewczyk, J. (2020). Learning orthographic cognates and non‐cognates in the classroom: Does awareness of cross‐linguistic similarity matter? Language Learning, 70(3), 685–731. https://doi.org/10.1111/lang.12390

Peal, E., & Lambert, W. E. (1962). The relation of bilingualism to intelligence. Psychological Monographs: General and Applied, 76(27), 1–23. https://doi.org/10.1037/h0093840

SAGIS. (2021). Étude d’usages et attitudes concernant la langue créole de La Réunion. https://lofislalangkreollarenyon.re/sondage/

Prudent, L.-F. (1981). Diglossie et interlecte. Langages, 15(61), 13–38. https://doi.org/10.3406/lgge.1981.1866

Révauger, G. (2023). Compétence plurilingue et langue anglaise : pour un enseignement stratégique des traversées (méta)lexicales et (méta)cognitives. Les Langues Modernes, 2023(3), 71–79.

Ringbom, H. (2006). Cross-linguistic similarity in foreign language learning. Multilingual Matters. https://doi.org/10.21832/9781853599361

Robert, J.-M. (2008). L’anglais comme langue proche du français? Éla, 149(1), 9–20. https://doi.org/10.3917/ela.149.0009

Rolland, Y. (2013). Apprentissage phonologique dans une approche didactique plurilingue. Recherches en didactique des langues et cultures, 10(1). https://doi.org/10.4000/rdlc.1497

Rolland, Y. (2014). L’apprentissage phonologique plurilingue en contexte institutionnel multilingue : réflexions. Travaux & Documents, 46, 149–161.

Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics in Language Teaching, 10(1–4), 209–232. https://doi.org/10.1515/iral.1972.10.1-4.209

Siemund, P. (2023). Multilingual development: English in a global context. Cambridge University Press.

Souprayen-Cavéry, L. (2020). Enseignement-apprentissage du français en milieu créolophone : le cas de La Réunion. Le français aujourd’hui, 208(1), 63–73. https://doi.org/10.3917/lfa.208.0063

Sournin, S. (2010). L’enseignement d’une langue vivante seconde à des élèves créolophones : le cas de l’anglais à La Réunion. Alternative Francophone, 1(3). https://doi.org/10.29173/af9616

Swan, M., & Smith, B. (2001). Learner English: A teacher’s guide to interference and other problems (2nd ed.). Cambridge University Press.

Veronique, G.-D. (2019). L’enseignement-apprentissage implicite et explicite des langues vivantes. https://hal.science/hal-01968803v1

Watbled, J.-P. (2021). La créolistique : arguments pour une approche sociohistorique. Contextes et Didactiques, 17. https://doi.org/10.4000/ced.2668

Weinreich, U. (1953). Languages in contact: Findings and problems. Mouton Publishers.