Mother tongue interference in the acquisition of grammatical gender by Spanish learners of English

eliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaelia eliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaelia eliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaelia eliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaelia eliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaeliaelia ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 79 Mother tongue interference in the acquisition of grammatical Mother tongue interference in the acquisition of grammatical gender by Spanish learners of Englishgender by Spanish learners of English La interferencia de la lengua materana en la adquisición del La interferencia de la lengua materana en la adquisición del género gramatical por aprendices españolas de inglés.género gramatical por aprendices españolas de inglés. María Luisa Carrió-Pastor María Luisa Carrió-Pastor Universitat Politècnica de València [email protected] AbstractAbstract This study focuses on the acquisition of grammatical gender agreement at different levels of language proficiency in English, paying attention to syntactic constraints and those arising from the influence of the mother tongue. Thus, the objectives are, first, to categorise mismatches in English grammatical gender acquisition when learnt by Spaniards at A2, B1, B2 and C1 levels of proficiency established by the Common European Framework of Reference for Languages (Council of Europe, 2001; 2018). Second, the study aims to determine whether these mismatches are produced by syntactic failure or by mother tongue transfer. Additionally, it explores whether communicative competence may be affected by grammatical gender acquisition mismatches. In this paper, the failures in grammatical gender agreement were identified and classified into syntactic and mother tongue transfer. The methodology followed was based on a corpus-driven approach. First, ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 80 Mother tongue interference in the acquisition of grammatical gender... a corpus of essays written by Spanish learners of English with A2, B1, B2 and C1 language proficiency levels was compiled. The grammatical gender agreement mismatches were then identified to ascertain the students’ ability to assign gender in English. Finally, the frequencies of failure of grammatical gender agreement at the different levels of language proficiency were classified into syntactic and language transfer categories. The results showed not only the grammatical gender agreement mechanisms when acquiring English at the different levels of language proficiency but also which kind of components in communicative competence may be affected by gender agreement processes. Keywords: grammatical gender agreement, language proficiency, Spanish learners, English. ResumenResumen Este estudio se centra en la adquisición de la concordancia de género gramatical en distintos niveles de competencia lingüística en inglés, analizando las restricciones sintácticas y las derivadas de la influencia de la lengua materna. Así, los objetivos son, en primer lugar, categorizar los desajustes en la adquisición del género gramatical en inglés aprendido por españoles en los niveles A2, B1, B2 y C1 establecidos por el Marco Común Europeo de Referencia para las Lenguas (Consejo de Europa, 2001; 2018). En segundo lugar, el estudio desea comprobar si los desajustes se producen por fallos sintácticos o por transferencia de la lengua materna. Finalmente, explora si la competencia comunicativa puede verse afectada por los desajustes en la adquisición del género gramatical. En este trabajo se identificaron los errores en la concordancia gramatical de género y se clasificaron en sintácticos y de transferencia de la lengua materna. La metodología seguida se basó en un enfoque basado en corpus. En primer lugar, se recopiló un corpus de ensayos escritos por estudiantes españoles de inglés con niveles de competencia lingüística A2, B1, B2 y C1. A continuación, se identificaron los fallos ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 81 María Luisa Carrió-Pastor de concordancia gramatical de género para determinar la capacidad de los estudiantes para asignar el género en inglés. Por último, las frecuencias de fallo en la concordancia gramatical de género en los distintos niveles de competencia lingüística se clasificaron en categorías sintácticas y de transferencia lingüística. Los resultados mostraron no solo los mecanismos de concordancia gramatical de género en la adquisición del inglés en los distintos niveles de competencia lingüística, sino también qué tipo de componentes de la competencia comunicativa pueden verse afectados por los procesos de concordancia de género. Palabras clave: concordancia gramatical de género, competencia lingüística, estudiantes españoles, inglés. 1. Introduction1. Introduction Most studies on language acquisition focus on the analysis of lexical gender of languages such as French and Spanish, which clearly differentiate whether the gender of a noun is motivated by biological gender, so that, in Spanish for example, mujer is feminine and hombre is masculine (Corbett, 1991). Some other studies also pay attention to the acquisition of grammatical gender, that is, when there are arbitrary words, without any semantic motivation, in which there is no reason for nevera and mesa , for example , to be feminine rather than masculine (Franceschina, 2001; White et al., 2004). Most languages therefore exhibit lexical and grammatical gender but, specifically, English has a semantic gender system determined by the meaning of the noun in context and does not differentiate grammatical gender in inanimate nouns such as fridge or table . In this vein, it may seem easy for native Spanish speakers to learn English as a foreign language to acquire gender, as there are no gender differences when definite and indefinite articles or determiners are used. However, this is not the case for pronouns or ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 82 Mother tongue interference in the acquisition of grammatical gender... possessive adjectives. Hence, in some cases, Spanish learners of English transfer the gender used in their mother tongue to English inanimate nouns. The Spanish gender system does not have a direct equivalent to the English pronoun it and the possessive adjective its to refer to inanimate nouns and animals. Spanish uses neutral pronouns to refer to inanimate nouns (e.g. tráeme eso [la nevera]), whose meaning is determined in context, but in general inanimate nouns are referred to as feminine or masculine (e.g . tráemela [la nevera ]). Thus, when learning a foreign language that does not follow the same gender agreement ‘rules’ as Spanish, some errors derived from grammatical gender mismatches may arise. Grammatical gender poses a serious problem to novice learners as well as to proficient speakers of a foreign language. This fact may affect learners’ communicative competence in components such as linguistic competence, strategic competence, sociocultural competence, actional competence and discourse competence (Celce- Murcia et al., 1995) and language teachers should be aware of these language failures. Researchers should pay more attention to these aspects, as the cases of grammatical gender acquisition of Spanish and French by learners with different gender systems (Audring, 2008), by bilinguals (Alarcón, 2011) or the comparison of native and non-native speakers (Liceras et al., 2000; Franceschina, 2001) have already been extensively studied. But some grammatical gender problems may be encountered by learners of other foreign languages that also need to be studied and discussed. Thus, in this paper, the first objective is to categorise mismatches in English grammatical gender acquisition when learnt by Spaniards at certain levels of language acquisition established by the Common European Framework of Reference for Languages (2001, 2018), i.e. A2, B1, B2 and C1. The second is to check whether mismatches are produced by mother tongue transfer. Additionally, it explores whether the components of communicative competence, that is, linguistic competence, strategic competence, sociocultural competence, actional competence, and discourse competence, may be ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 83 María Luisa Carrió-Pastor affected by gender failures produced by Spanish learners of English. Accordingly, the following research questions are posed in this study: 1. Does the mother tongue influence the way speakers of a language with grammatical gender agreement acquire a language with a poorer grammatical gender system? 2. Are there any differences in the use of grammatical gender at different levels of foreign language proficiency? 3. Which components of communicative competence should be considered by English teachers to eliminate mismatches in grammatical gender agreement? To answer these research questions and meet the objectives set, this paper has been organised in the following way. First, the introduction gives a general outline of the analysis and describes the most important aims of the paper in the objectives and research questions; second, the theoretical background is addressed in the following section about gender acquisition. The material and method section is then described in the third section. In the fourth section, the results are shown and discussed and, finally, conclusions are drawn. 2. Gender acquisition: Grammatical gender agreement2. Gender acquisition: Grammatical gender agreement Studies on foreign language acquisition have traditionally paid attention to the different problems encountered by learners and have proposed solutions to improve foreign language learning (Corder, 1967; Edge, 1989; Ellis, 1997; James, 1998). But before suggesting solutions, researchers should be aware of the problems faced by foreign language learners. Many of the problems learners come across are related to gender assignment when the target language attributes feminine and masculine gender to nouns (Corbett, 1991). In this sense, gender assignment rules in a language ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 84 Mother tongue interference in the acquisition of grammatical gender... are proposed to decrease the burden of learning and memorising the gender of nouns and to help assign a gender to new words (Audring, 2008). But the most problematic case of gender acquisition is gender agreement. Determiners, pronouns and adjectives must therefore be marked for the same gender as the noun they are associated with. This is part of the functional nature of determiners and adjectives and their relationship with nouns in order to organise discourse in a logical and effective way. Although the focus of this study is related to the acquisition of grammatical gender agreement in English, first it is also necessary to understand the mechanisms of grammatical gender agreement in Spanish to identify the way Spanish learners of English apply these rules when acquiring a foreign language. Spanish learners of English sometimes transfer the characteristics of their mother tongue onto the target language. So, in this section attention is paid to the studies that have focused on the problems of gender agreement when learning Spanish, which are taken into consideration to discuss the findings of this research. While biological and semantic criteria are prevalent in lexical gender agreement, that is, feminine gender is assigned to women and masculine to men, the problems arise in grammatical gender agreement when learners try to retain and assimilate the gender of inanimate nouns in languages, such as Spanish, that do not follow any logical process or semantic criteria. That is, nouns are assigned to gender as a part of their form. In Spanish, nouns have gender while adjectives, pronouns and determiners show gender agreement with the noun. As Grüter et al. (2012) explained, referring to the assignment of gender to Spanish nouns: “In many cases, morpho-phonological properties of the noun can provide a cue to its gender, with nouns ending in –o overwhelmingly belonging to the masculine, and those ending in –a to the feminine class” (p. 192). Besides, they mention that this rule is not fully reliable, that is, the assignment of gender to inanimate ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 85 María Luisa Carrió-Pastor nouns in Spanish seems to be complex and difficult to understand by those learning Spanish due to the existence of many exceptions. In this vein, they indicated that “The only consistent and fully reliable cue to a noun gender class is distributional, consisting of its cooccurrence relations with transparently gender-marked modifiers” (p. 192). In contrast, Harris (1991) considered that, in inanimate nouns, –o and –a endings are word markers rather than gender markers in Spanish, as these endings may be found in other words, such as adverbs, in which gender is not present. Canonical –o masculine endings and –a feminine endings form the ‘inner core’ or most prototypical cases, while nouns ending in –e or a consonant conform the non-canonical or ‘outer core’ and masculine nouns ending in –a and feminine in –o are exceptions, also called ‘residue’ (Montrul et al., 2008). After these reflections, it thus seems that the semantic classification of grammatical gender in inanimate nouns is idiosyncratic and complex in Spanish, and its complexity is disambiguated by the syntax (Ambadiang, 1999). Alternatively, in English, gender is based on a pronominal gender system, that is, it has three genders, feminine, masculine and neuter, marked by the personal pronouns, ‘he’ and ‘she’ for male and female human antecedents, and ‘it’ for all inanimate nouns and animals (Frenck-Mestreet al., 2009). The main difference between English and Spanish is that the former has a semantic gender system, gender being interpreted depending on context and showing a strong tendency to use plural forms to neutralise it. For example, the European Parliament (2018) recommends the use of “officials” instead of “the official” (more examples can be found at https://www. europarl.europa.eu/cmsdata/151780/GNL_Guidelines_EN.pdf). Moreover, English depends on word order rather than agreement to organise discourse and to show syntactic functions. One characteristic of gender agreement, present in Spanish but not in English, is the availability of noun drop or nominal ellipsis (Liceras et al., 2000). As Montrul et al. (2008) explained, “Spanish nouns can be dropped or omitted in discourse because the referent is recovered from the rich ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 86 Mother tongue interference in the acquisition of grammatical gender... gender and number agreement on the determiner” (p. 510-11), as, for example in the expression la violeta , where the implicit meaning in one specific context would be the violet jacket . Thus, the fact that Spanish has a well-identified gender system for inanimate nouns, which is arbitrary and frequently distinct from natural gender while English has no grammatical gender, could create difficulties for Spanish learners of English. They may find it easier to use the grammatical gender reference of their mother tongue, that is, puerta is feminine and it should be the same in another language. Hence, acquiring English gender entails learning that there is no classification of inanimate nouns into feminine and masculine and that they are ‘neutralised’, thereby making it unnecessary for them to agree with pronouns, adjectives and determiners and to recognise agreement in nominal ellipsis. This fact seems to simplify foreign language acquisition processes, but it is an aspect that is questioned and studied in this paper. The difficulties of foreign language learners regarding gender could have different sources (Grüter et al., 2012), such as learners not knowing a word and not knowing its gender (lexical knowledge) or their failure to fully understand gender agreement (syntactic knowledge) or not being able to reproduce lexical and syntactic knowledge in a very specific context. Additionally, some researchers have pointed out that grammatical gender agreement is not difficult to acquire when learners are instructed correctly (McCarthy, 2008; Alarcon, 2011) and that the mistakes that may arise could be due to specific problems mainly in spoken production, when speakers have to process grammatical gender in real time and thus errors or mismatches are more frequent. In this vein, Sabourin et al. (2006) paid attention to transfer effects in learning a second language and indicated that “some L2 constructions are learnable while others seem not to be. This apparently depends partly on the structure of the L1 since it seems that transfer from the L1 can help in L2 acquisition in some cases. An example of an L2 phenomenon that is particularly difficult to learn ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 87 María Luisa Carrió-Pastor is grammatical gender” (p. 2). Sagarra and Herschensohn (2010) also agreed with the idea that grammatical gender is difficult to acquire due to language transfer. Most studies on gender agreement have paid attention to the way speakers of languages that do not have grammatical gender acquire it. But in this paper, I believe that grammatical gender is also difficult to acquire by learners whose mother tongue does have grammatical gender. Thus, the focus of this research is on the influence of Spanish transfer when acquiring gender in English, given that English has natural gender (determiners, pronouns and adjectives do not show grammatical gender agreement with the noun, the gender is shown through its meaning) while Spanish does have such agreement. Sabourin et al. (2006) proposed three degrees of transfer in the target language, that is, no transfer, partial transfer and full transfer. In the present study, only full transfer was considered given the corpus and the fact that English only has gender agreements in pronouns and possessive adjectives and pronouns, which are the items studied here. Franceschina (2001) observed that learners whose mother tongue has gender agreement performed better than those whose mother tongue does not, but the dissimilarities among languages make it difficult to be certain that the concept of gender can be transferred from one language to another. The question at the heart of most studies on L2 gender acquisition is therefore whether abstract properties of grammatical gender need to be present in the mother tongue for late learners to fully acquire grammatical gender and gender concord in a foreign language (Frenck-Mestreet al., 2009). This paper deals with empirical problems regarding gender agreement on pronouns and possessive adjective and pronouns, with the help of writing performed by Spanish learners of English. Even though English gender is not complicated to acquire, I would like to make a point in some cases in which English pronouns, possessive adjectives and possessive pronouns do not match their antecedents in the gender system when used by Spanish learners. Additionally, this paper explores language proficiency associated with ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 94 Mother tongue interference in the acquisition of grammatical gender... mismatches caused by nominal ellipsis were also detected manually. After the manual identification of pronouns, possessive adjectives and possessive pronouns, a mechanical search of these items was also carried out using the search option of the text processor Word to be sure that all the occurrences of failure in grammatical gender agreement were located. The frequencies of mismatches of grammatical gender agreement were counted and normalised to 10,000 words given that the sub-corpora had different numbers of tokens. Once all the mismatches had been found, examples were discussed at each proficiency level to identify the cause of grammatical gender agreement with inanimate objects. The causes identified were related to the transfer of their mother tongue, i.e. Spanish. Finally, the components of communicative competence were considered to study whether mismatches in grammatical gender agreement may affect communication in English. This was discussed bearing in mind the different levels of English proficiency studied in order to identify mismatches that might be caused by a poor knowledge of English. To finish the study, conclusions were drawn. 4. Results and discussion4. Results and discussion The results obtained from the analysis of the four sub-corpora revealed, as expected, that grammatical gender agreement is not a common failure in Spanish learners of English. Table 2 shows the occurrences and normalised frequencies (N) per level of language proficiency: ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 95 María Luisa Carrió-Pastor Table 2. Occurrences of mismatches in grammatical gender agreement Levels of language proficiency Occurrences of syntactic failure Occurrences of mother tongue transfer (N) A2 proficiency level 0 27 (22.79) B1 proficiency level 1 13 (9.07) B2 proficiency level 0 3 (1.91) C1 proficiency level 0 0 (0.00) Total 0 43 (7.32) Only one occurrence caused by syntactic failure was found at the levels analysed, and it could be a mistake (it can be seen in example [2], B1-A). It may therefore be said that the acquisition of the syntactic arrangement of pronouns and possessive adjectives and pronouns by Spanish learners of English is effective and successful even at low levels of proficiency. The occurrences found in the analysis of the corpus also revealed that failure in grammatical gender agreement was more frequent at low proficiency levels than at higher levels, as few or no frequencies were found at higher levels such as B2 and C1. After studying the occurrences in detail, it also seems that the mismatches in grammatical gender agreement only take place when Spanish learners of English transfer the gender of an inanimate noun from their mother tongue. This may indicate that learners with low levels of language proficiency whose mother tongue has grammatical gender agreement tend to translate from their mother tongue and use the gender assigned in it. The fact that these failures disappeared at higher levels indicates that the correct pronouns, possessive adjectives and pronouns are acquired very quickly by Spanish learners. This could be due to the low complexity of grammatical gender in English, so once students increase their knowledge of English, they are able to use grammatical gender agreement correctly in that language. Ayoun (2007) pointed out that “The question is whether adult L2 learners can acquire new functional categories which do not exist in the L1 at all, or functional categories existing in the L1 but with different values in the L2” (p. 143). ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 96 Mother tongue interference in the acquisition of grammatical gender... The results of this analysis have shown that new functional categories existing in the mother tongue but different in the target language can be acquired by learners, given that grammatical gender mismatches appear at low proficiency levels, but they do not appear at higher levels. This shows that, after instruction, Spanish learners acquired grammatical gender agreement in English. Additionally, it was observed (in B1 learners) overuse of the definitive article in English in inanimate nouns, as in “Cables are driven along the * pipelines by the flow of the * water”. This is not related to gender agreement, but it may support the idea of the influence of the mother tongue in low levels of English proficiency. The data obtained here were expected, taking into consideration previous studies that identified gender violations at different levels of language proficiency (Sagarra and Herschensohn, 2010; White et al., 2004) and the fact that these authors also observed that beginners do not show any sensitivity to gender agreement violations. The results obtained here are thus in the same line, evidencing that mismatches decrease at the same pace as grammatical knowledge increases. As mentioned by Ayoun (2007), a strong tendency to use the plural forms of pronouns (e.g. them ) was also observed in the corpus. This use of the plural means that learners do not have to decide whether the inanimate noun is masculine or feminine or if the neuter pronoun (e.g. it ) should be used. Additionally, there was a tendency to repeat nouns instead of using pronouns, which is a strategy that allows learners to avoid mismatches in grammatical gender agreement; for example, in “The use of social media is something usual in young people, but social media can become a danger for the youngest, who try to use social media without the knowledge they need” (B2-23). Furthermore, language transfer was also seen to occur in the agreement of inanimate nouns and pronouns, but learners did not transfer nominal ellipsis. I did find number disagreement at low levels of language proficiency, such as in the sentence “Now the vehicles are more comfortables ” (B1-40), which is also due to mother ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 97 María Luisa Carrió-Pastor tongue transfer, as in Spanish adjectives can be plural ( cómodos ) or singular ( cómodo ). This fact coincides with Sabourin, Stowe and de Haan (2006: 26) Stowe and de Haan (2006, who stated that “making use of a noun’s gender in an agreement context is much harder to do”. When learners rely on their mother tongue to acquire syntacticlevel knowledge in a target language, this may make the learning process harder. Grammatical gender agreement should be learnt as a new process, as there are no similarities between English and Spanish. The transfer of grammatical gender agreement in this case was negative in the learning process. Instances of the occurrences found in the three sub-corpora can be seen in example [1], which shows examples from the A2 proficiency level, example [2], which displays the B1 proficiency level, and example [3], which corresponds to the B2 proficiency level. Examples of the C1 proficiency level are not shown as no grammatical gender agreement mismatches were found. [1] A2[1] A2 A) “Internet is a useful form of work. When she appeared it was a revolution”. (A2-03) In these sentences, the learner refers back to ‘Internet’ with a feminine personal subject pronoun instead of using the English pronoun for inanimate nouns (that is, it ), this may be caused because Spanish speakers use the pronoun ella (she) to refer to the internet (a feminine noun in Spanish: la internet, la red ). The learner therefore transfers the grammatical rule employed in Spanish and uses the feminine personal pronoun to refer to the internet. B) “I can’t take breakfast because if I take him I lose the bus that I take everydays”. (A2-32) In this example, the speaker uses the personal object pronoun him to refer to breakfast , even when in Spanish the particle used is ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 98 Mother tongue interference in the acquisition of grammatical gender... neuter ( si lo tomara ). The learner transfers the grammatical gender agreement from Spanish, but this strategy is not based on translation, as in the previous example. In this case, the learner uses an object pronoun but with the incorrect form, assigning the gender of the inanimate noun in Spanish. C) “The teenagers even don’t have enough capacity to prevent from he [the internet], also it will get them into trouble”. (A2-40) Here, the writer refers to the internet but fails in grammatical gender agreement, using a subject pronoun after the preposition and, additionally, this is a pronoun for the masculine gender. In this case, as pointed out by White et al. (2004), the writer prefers the use of the masculine pronoun even when in Spanish a feminine pronoun or no pronoun is used for the internet. Thus, in this example we can observe failure in gender agreement and gender assignment. [2] B1[2] B1 A) “probably children would spend too much time with the computer, in fact he does it now. But it would be a good hobby because by him that time could be spended for learn other cultures” (B1-08) Here, the subject noun, children , is referred to as him in the second sentence. In this case, the learner uses the masculine object pronoun at the beginning of the clause, therefore giving rise to a mismatch of grammatical gender syntactic position and agreement, by transferring the Spanish expression ‘ para él’ . The learner does not consider children as plural, but as masculine singular; additionally, in the first sentence the learner uses the singular masculine subject pronoun to refer to children : ‘in fact he does’. This case again reinforces the idea that learners prefer the masculine gender, a fact also pointed out by Liceras et al. (2008). In our analysis this happens when learners are not certain about functional aspects. In this ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 99 María Luisa Carrió-Pastor sample, the learner does not acquire new functional categories which do not exist in the mother tongue, but instead adapts the functions of the mother tongue to the target language. B) “Although I have some bad moments in my life as a teenager, I wouldn’t change her parts because I also get good situations.” (B1-28) In this case, the failure is caused by the use of her (feminine singular possessive pronoun) instead of using the possessive adjective for inanimate nouns and animals its . Here, the learner has not acquired the possessive adjectives correctly and then he/she refers to the conceptual image of the noun life in his/her mother tongue and uses the feminine adjective form of the Spanish noun la vida . C) “It has been a major concern for people that they have to teach how to use social media safely and legally. Some people hold that she is very important, especially the teenagers” (B1-40) Here, the learner refers to social media as feminine and singular with the subject pronoun she . This may be caused by the fact that the noun media ends in –a , which is associated to feminine nouns in Spanish (Harris 1991). The use of a singular pronoun is rather odd as in Spanish it is a plural noun ‘ redes sociales ’ or ‘ medios sociales ’, the first being feminine and the second masculine. In English, social media is usually plural when we refer to different media, but it might be used in singular when we are referring to it as a single phenomenon. Perhaps the learner may have acquired the idea that it is singular when reading news on the internet and then decided on the feminine form from the ending. [3] B2[3] B2 A) “It isn’t only a way to spend your time but also a page that allows you to know how our society is. It’s well worth visiting her in order to understand why famous people sometimes feel” (B2-05) ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 100 Mother tongue interference in the acquisition of grammatical gender... Very few occurrences were found in the sub-corpora of learners with B2 proficiency level. The object pronoun is used in its feminine form ‘ her ’ to refer to ‘ page ’ (that is, in Spanish la página ), a mistake that should not be produced by learners with a B2 level. I think this may be a mistake on the part of the learner, given the few occurrences found at this level. I, therefore, consider it an exception, as lexical and grammatical English gender and gender agreement should have already been acquired by the B2 level. B) “In conclusions, if the new transport is better for us we will have to find the correct way for using him ” (B2-27) This is the same case: I think it is a mistake, as the learner employs the object pronoun him to refer to the noun transport , which is singular and masculine in Spanish ( el transporte ). It seems that Spanish learners acquire grammatical gender agreement in the early stages of foreign language acquisition, and this is not a problem for advanced speakers of English. Hence, the data presented here reveal that English learners should be aware of the differences between the Spanish and English grammatical gender systems, which will enable them to avoid mismatches. Regarding the different components of communicative competence, here I studied the written production or performance of English learners with different levels of language proficiency. At each stage of proficiency, I analysed how failure in grammatical gender agreement may affect their communicative competence in order to be able to provide some guidelines to help language teachers and learners avoid mismatches. The discourse component involves the selection, sequencing and arrangement of words, structures, sentences and utterances to obtain a unified written text. Grammatical gender agreement is related to the component of pronouns; in deixis, it is part of the coherence of discourse and links the situational context. Discourse ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 101 María Luisa Carrió-Pastor may therefore be affected by mismatches in grammatical gender agreement as the argumentative flow might be disrupted by mother tongue transfer. Here, special guidelines about the importance of pronouns should be given to learners, stressing the idea that they should not use translation strategies. The linguistic component comprises the basic elements of communication, that is, the sentence patterns and types, the morphological inflections, the lexical resources and the phonological and orthographic systems needed to communicate. Grammatical gender elements are related to lexicon, which includes determiners, pronouns and adjectives, and they are thus related to linguistic competence. In this component, both language learners and teachers should pay special attention to lexicon and to the linguistic elements that play a vital role in grammatical gender agreement, as well as establishing the relationships of these linguistic elements and consequently avoiding failure. The actional component conveys and matches actional determination with linguistic form, based on the knowledge of verbal schemata. It is very close to interlanguage pragmatics and rhetorical competence. This component is not affected by mismatches in grammatical gender agreement, as actional competence is related to pragmatic aspects of language. The sociocultural component refers to the speaker’s knowledge of how to express messages appropriately in a given social and cultural context of communication. Grammatical gender agreement is more related to syntax than to pragmatics, so the sociocultural component is not affected by its failures. Finally, the strategic component could be defined as knowledge of communication strategies and how to use them. As explained before, one of its sub-components is achievement and compensatory strategies, and one of the aspects taken into account is literal translation from the mother tongue and other compensatory strategies carried out by learners to communicate in a foreign ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 102 Mother tongue interference in the acquisition of grammatical gender... language even when their language proficiency is low. The mismatches of grammatical gender agreement found in the corpus may be one strategic component used by Spanish learners to be able to communicate in English. As can be seen, in communicative competence the components related to discourse, linguistics and strategies may be affected by lack of grammatical gender agreement. The actional component and the sociocultural component are more related to functional aspects of language communication and, on the contrary, grammatical gender is more associated to generative and structural characteristics of language. English learners and teachers should thus take into account the fact that grammatical gender agreement may affect communicative competence in several ways and this proposal may serve as a checklist that language teachers can refer to. 5. Conclusions5. Conclusions This study has shown that mismatches in grammatical gender agreement disappear when learners acquire foreign language proficiency. Learners with lower proficiency levels rely on mother tongue transfer on some occasions and transfer mother tongue gender agreement when referring to inanimate nouns. Some of those beginning to learn English display insensitivity to gender concord but this mismatch disappears when knowledge of the language increases. This may be due to the fact that English relies on word order to indicate the syntactic functions of words, and the syntactic form and function of pronouns have already been acquired at higher levels. The objectives of this paper were, first, to identify mismatches in English grammatical gender acquisition when learnt by Spaniards at different levels of language acquisition. This objective has been ELIA 24, 2024, pp. 79-107 DOI: http://dx.doi.org/10.12795/elia.2024.i24.3 103 María Luisa Carrió-Pastor shown in Table 2 and it can be seen that few occurrences of grammatical gender agreement mismatches were found, as expected, given that Spanish learners have a mother tongue with a wellestablished grammatical gender system and English has a poor one. The second objective was to check whether mismatches were produced by mother tongue transfer and, in this vein, the first research question also asked about the influence of the mother tongue on the way speakers with grammatical gender agreement acquire a language with a poorer grammatical gender system. The mismatches of grammatical gender agreement were only related to mother tongue transfer – indeed, only one occurrence was found that showed syntactic failure. From our analysis it may therefore be said that at low levels of language proficiency, learners rely on their mother tongue when they do not know how to use certain functions. Most Spanish learners acquire grammatical gender correctly in English, as it is a poor system that is not difficult to manage. Yet, some cases were identified that transfer grammatical gender agreement to the target language. Regarding the second research question about dissimilarities in the use of grammatical gender at different levels of language proficiency, it was observed that the vast majority of mismatches were found at lower levels of language proficiency and very few were found at higher levels. Given the few differences in grammatical gender in English, learners acquire this very early, as has been shown in the results in Table 2. Regarding the last objective and research question, the components of communicative competence that may be affected by mismatches in grammatical gender acquisition by Spanish learners of English were identified as those related to discourse, linguistics and strategies. All these components include communicative aspects in which pronouns, adjectives and determiners are involved. Additionally, there might sometimes be communicative strategies in which learners transfer certain linguistic aspects from the mother tongue.