Language Preference when Learning a Third Language

Abstract

Research demonstrates that learners typically rely on the first language (L1) when learning another language, regardless of whether it is the second (L2), third (L3), and so forth (Hermas, 2010). However, other research found effects related to language preference when considering order of language learned (Rothman, 2010). This study sought to determine whether language type influenced learning a third language when the third language related syntactically to the first or second language acquired. All participants were near bilinguals in Spanish and English and had varying proficiency levels in their L3. Participants rated sentences in a third language based on grammaticality. The results shed light onto the influence of language type above the order of language learned.

Headnote

Abstract

Research demonstrates that learners typically rely on the first language (L1) when learning another language, regardless of whether it is the second (L2), third (L3), and so forth (Hermas, 2010). However, other research found effects related to language preference when considering order of language learned (Rothman, 2010). This study sought to determine whether language type influenced learning a third language when the third language related syntactically to the first or second language acquired. All participants were near bilinguals in Spanish and English and had varying proficiency levels in their L3. Participants rated sentences in a third language based on grammaticality. The results shed light onto the influence of language type above the order of language learned.

Keywords: cross-linguistic influence, third language acquisition, morphosyntactic transfer, processing strategies

Introduction

In the past 20 years, the area of third language acquisition (TLA) has gained popularity due to the growing need to distinguish between bilinguals and multilinguals (Leung, 2007). Results from previous research indicate that bilinguals demonstrate greater success than monolinguals when learning a third language (Cenoz, 2003; Klein, 1995; Sanz, 2000). However, Odlin (2003) claims that it is very difficult to narrow down the most influential factor in TLA and that there is not sufficient evidence to decide which factors are more or less influential. Some research claims that TLA may use or activate any and all known languages (Flynn et al., 2004; Lemöhfer et al., 2004; Rothman, 2010). However, others claim there is a preference for L1 transfer in the initial stages of TLA (Hermas, 2010). Still others show a greater influence of the L2 in TLA (Falk & Bardel, 2010).

The present study sought to clarify the influence of the L1 and the L2 on the L3, when the learners were late bilinguals learning languages that relate syntactically to either the L1 or the L2.

Previous Research

Language similarities are evident in languages that share a proto-language. Languages such as Spanish, Italian, and French (and others) are typically grouped together as Romance languages as they derive from Latin, whereas English and German (and others) are considered Germanic languages as they derive from protogermanic (Rothman, 2010). For these reasons, we chose to examine Third Language Acquisition with different language ancestry (Romance and Germanic) and tested participants in Romance and Germanic languages (French and German.) By doing so, we hoped to gain clarity on how the L1 and L2 function in TLA.

Influence of the L1

Hermas (2010) investigated the contribution of syntax and its affect on the L3 initial state. The study employed acceptability judgment tests and preference tests with Arabic-French bilinguals learning English as an L3. Hermas found that the L1 strictly influenced the L3.

Bardell et al. (2013) investigated the role of metalinguistic knowledge in the first language when learning a third language. All participants were L1 Swedish, were L2 English, knew at least one Romance language, and were learning Dutch. All participants had varying levels of metalinguistic knowledge in the L1. The results demonstrated that the level of explicit knowledge in the L1 is a deciding factor in the beginning learning stages of L3.

Sanz et al. (2014) also investigated the roles of the first and second languages in learning beginning Latin. Participants were English speakers, from classrooms learning either Spanish or Japanese (i.e., late bilinguals). The results demonstrated that the significant improvement was due to influence from the first language (English). Sanz et al. suggest that in order to have higher influence from the second language, a greater proficiency level may be necessary.

Influence of the L2

Fouser (2001) investigated general language transfer in TLA. The participants' L1 was English, L2 was Japanese, and L3 was Korean. Japanese and Korean share syntactic similarities, but neither has syntactic commonalities with English. The results demonstrated that only the L2 (Japanese) had an influence on the acquisition of the L3, Korean.

Bardell and Falk (2007) also found that L2 transfer takes preference, at least at the initial stages of TLA. Bardell et al. (2013) investigated syntactic transfer of L1 and L2 to L3. The study tested 44 French-English or English-French participants (order of acquisition) on the acquisition of object pronoun placement in German. Results demonstrated that the L2 was the privileged source of transfer. These studies indicate that there is a significant influence of the L2 on TLA.

Llama et al. (2010) investigated L2 influence in phonological L3 acquisition. Two groups, L1 English-L2 French and L1 French-L2 English, read a list of words in Spanish with different phonological features and were tested on their ability to determine the difference. The results found L2 as the determining factor in both groups.

Riestenberg et al. (2015) investigated the role of explicit knowledge in order to explain the advantage of previous language experience in learning a third language as well as the potential role of second language ability. The participants were native speakers of English, who were learning Arabic as a second language and Latin as a third. Thirty-five native Englishspeaking learners of Arabic at three different levels completed computer-based training and testing tasks dealing with thematic role assignment in Latin. The results demonstrated that learners in the early stages of L3 development start with a strategy based on their first language, but as proficiency increases, second language strategy played an increased influence.

Influence from any Previous Language

Flynn et al. (2004) investigated the source of L3 transfer in formal syntactic features and functional categories. They examined the production of restrictive relative clauses in participants with L1 Kazakh, L2 Russian, and L3 English. The results demonstrated that any previously acquired language may aid in the acquisition of a new language.

Lemöhfer et al. (2004) investigated the effects of cognates in trilingual processing. The experiment consisted of 28 trilinguals with a background in Dutch, English, and German. Participants were tested on Dutch-German-English cognates and Dutch-German cognates, and were asked to determine whether each word was a real German word. The results demonstrated that all known languages can be accessed simultaneously during lexical processing.

Rothman (2010) investigated two groups of learners. Group 1 participants were L1 Italian, L2 English, and L3 Spanish at beginner to intermediate levels. Group 2 participants were L1 English, L2 Spanish, and L3 Portuguese. Group 2 had the same levels of knowledge in their L2 and L3. The participants learned adjective placement and meaning change in Romance language and completed tests to measure acquisition. The data revealed that L3 transfer can occur from L1 or L2 in the initial stages, and the determining factor during acquisition is language family closeness.

To summarize, previous research reveals that learning a second language involves an influence from either the L1 or the L2, or both the L1 and L2, when considering phonological, syntactic, and lexical acquisition as well as proficiency level and age of acquisition in the second language. The current study sought to determine leaner preference when determining grammaticality in a third language that is related syntactically to either their first or second language.

Research Question

The research question that guided this study is the following:

When determining the grammaticality of expressions in the third language (L3), do learners rely on the grammar of the first language (L1) or second language (L2)?

Hypothesis: If the L3 is related to the L1, learners will rely on the L1 when determining grammaticality. If the L3 is related to the L2, learners will rely on the L2 when determining grammaticality.

Methods

Participants

Participant 1 was a native English speaker. She had 11 years of Spanish instruction as well as experience living and studying in a Spanish-speaking country and described her reading, writing, listening, and speaking as advanced. She also studied 2 years of university level French and lived in a French-speaking country. She described her writing and reading in French as beginner, and her listening comprehension and speaking as intermediate. She had studied 2 months of beginning university Japanese at the time of testing.

Participant 2 was a native English speaker. She had 8 years of Spanish secondary and university instruction, and she described her reading, writing, listening, and speaking as advanced. She also had a beginner level of Polish and Swahili, and was enrolled in a first semester beginning university French course.

Participant 3 was a native speaker of Spanish. She has studied English for 18 years and lived in the United States of America for 4 years at the time of the test. She also had studied Italian for 15 years, which she rates as intermediate in the four areas.

Materials

The treatment covered the Romance (French) and Germanic (German) language families. The researchers chose French, a Romance language, for its syntactic similarity to Spanish. However, there are some very common differences in syntax between French and Spanish. For example, in Spanish, it is acceptable and even preferred to omit the subject pronoun in a sentence.

Hablo español.

speak-1st person singular-present Spanish-noun.

In French, subject pronoun omission is ungrammatical, and it is normally necessary to include the subject pronoun.

Je parle français.

I-nom. pro. 1st pers. singular speak-1st person singular present French-noun.

The same is true in German.

Ich spreche Deutsch.

I-nom. pro 1st pers. singular speak-1st person singular present German-noun.

Therefore, in the experiment below, learners were asked about the grammaticality of French and German expressions that were missing subject pronouns.

*parle français.

speak-1st person singular-present French-noun.

*spreche Deutsch.

speak-1st person singular-present German-noun.

If learners accept the ungrammatical sentences with subject pronoun deletion, this indicates learners are relying on Spanish for grammatical cues. If learners reject the ungrammatical sentences, this demonstrates learners are relying on English.

Another grammatical feature that differs in some Romance and Germanic languages is adjective placement. In general, the adjective comes after the verb in Romance languages. This is true for Spanish and French, with some exceptions. The researchers avoided these exceptions in the experimental materials.

French:

la robe verte

the-det. dress-noun green-adj.

Spanish:

el vestido verde

the-det. dress-noun green-adj.

In Germanic languages, in general, the adjective comes before the verb.

German:

das grüne Kleid

the-det. green-adj. dress-noun

Therefore, if the participants choose the adjective placement in front of the verb, they rely on cues from English. If they choose the adjective after the verb, they are relying on cues from Spanish.

Procedure

Participants completed a language history questionnaire (Appendix A). Participants informed researchers what language(s) they spoke at home or with their family (as they do not live at home during their university study); what other languages they spoke or studied; at what age they began speaking or studying the language; and for how many years they had spoken or studied the language. They answered questions about the countries they lived in and the languages they spoke while there. Finally, they self-rated their level in each language (beginner, intermediate, or advanced) in the areas of writing, reading, listening (comprehension), and speaking.

Next, participants read a list of vocabulary in French (Appendix B). They had two minutes to learn the vocabulary.

Next, participants completed a grammaticality judgment task in French (Appendix C). The expressions were either grammatical or ungrammatical based on subject pronoun deletion. All words on the grammaticality judgment task were from the vocabulary that participants had studied. Participants rated the expression as grammatically correct, questionable, or incorrect, and attempted to translate the expression to English. They saw no information about language

type or family. The type of pronoun deletion on the test was canonical in Spanish, but not acceptable in French.

Next, participants completed a grammaticality judgment task in French (Appendix D). The expressions were either grammatical or ungrammatical based on adjective placement. All words on the grammaticality judgment task were from the vocabulary that participants had studied. Participants circled the correct expression in French and again attempted translate the expressions to English. The adjective placement was canonical in Spanish and French.

Next, participants read a list of vocabulary in German (Appendix E). They had two minutes to review the vocabulary.

Next, participants completed a grammaticality judgment task in German (Appendix F). The expressions were either grammatical or not based on pronoun deletion. All words on the grammaticality judgment task were from the vocabulary that participants had studied. Participants rated the expression as grammatically correct, questionable, or incorrect, and attempted to translate the expression to English. Pronoun deletion items were unacceptable in both German and English.

Next, participants completed a grammaticality judgment task in German (Appendix G). The expressions were either grammatical or not based on adjective placement. All words on the grammaticality judgment task were from the vocabulary that participants had studied. Participants circled the correct expression in German, and attempted to translate the expression to English.

Results

In Table 1, we present the results from the pronoun deletion acceptability test. All three participants accepted 50% of the incorrect pronoun deletion in French. In German, Participant 1 accepted 10% of the incorrect pronoun deletion, Participant 2 accepted 20% of the incorrect pronoun deletion, and Participant 3 accepted 60% of the incorrect pronoun deletion.

In Table 2, all participants correctly rejected incorrect adjective placement in French.

In Table 3, Participant 1 accepted incorrect adjective placement in German for 40% of the items. Participant 2 and 3 accepted 0% of the incorrect adjective placement.

Discussion

Participants in the grammaticality judgment tasks demonstrated a preference for Spanish pronoun deletion and Spanish adjective placement when judging French sentences. When judging German sentences, participants demonstrated a preference for English subject pronoun cues and English adjective placement.

There were some issues that the researchers feel could have influenced the outcome of the study. The first was the lack of French and German language data. For example, though participants were given a vocabulary list, they were not given examples of French or German intact sentences, so the only cue they had to determine the grammaticality of an expression was their first or second language. However, they performed differently in French and German, which indicates the influence of the language family, the main focus of the study, regardless of their experience with another language beyond the languages they considered advanced or native.

Another factor to consider is the participants' language history. All participants have an L3 in a Romance language; however, two have an L3 of French and L1 of Italian. Further testing with more participants with different fluency levels in their L3 could paint a clearer picture of the effects of L1 and L2 on L3, and the effects at the different stages of L3 acquisition.

Conclusion

The goal of this study was to determine learner preference of L3 grammar. Researchers employed grammaticality judgment tests in order to determine whether the first or second language was more influential when learning a third. We hypothesized that based on the similarity to the language being learned, the learner would rely on either the first or second language. Results demonstrate that language type influenced learners' third language acquisition.

Sidebar

About the Authors

Lauren Wileński ([email protected]) is an alumna of the University of Tampa, with a Spanish major and a Political Science minor, and she begins a Master's degree in Spanish in the fall.

Andrew DeMil, Ph.D. ([email protected]), is a professor at the University of Tampa. Dr. DeMil specializes in second language acquisition, with a focus on textbook analysis, processing instruction, and teaching future second language teachers.

Discussion Questions

1. As a language learner, what influences from other languages have you noticed?

2. What role could cognates, or words shared by two or more languages play in language learning?

3. Is more easier to learn a language similar to one you already know, or one that is completely different?

To Cite this Article

Wilenski, L., & DeMil, A. (2020, Fall). Language preference when learning a third language. Journal of Multidisciplinary Research, 12(2), 153-168.

References

References

Bardel, C., & Falk, Y. (2007). The role of the second language in third language acquisition: The case of Germanic syntax. Second Language Research, 23(4), 459-484. https://doi.org/10.1177/0267658307080557

Bardel, C., Falk, Y., & Lindqvist, C. (2013). The role of L1 explicit metalinguistic knowledge in L3 oral production at the initial state. Bilingualism: Language and Cognition, 18(2), 227235. https://doi.org/10.1017/s1366728913000552

Cenoz, J. (2003). The additive effect of bilingualism on third language acquisition: A review. The International Journal of Bilingualism, 7, 71-87. https://doi.org/ 10.1177.13670069030070010501

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

Fouser, R. J. (2001). Too close for comfort? Sociolinguistic transfer from Japanese into Korean as an L3: Cross-linguistic influence in Third Language Acquisition. Korean Education Journal, 11(2), 149-169. https://doi.org/10.21832/9781853595509-010

Hermas, A. (2010). Language acquisition as computational resetting: verb movement in L3 initial state. International Journal of Multilingualism, 7(4), 343-362. https://doi.org/10.1080/14790718.2010.487941

Klein, E. C. (1995). Second versus third language acquisition: Is there a difference? Language Learning, 45, 419-465. https://doi.org/10.1111/j.1467-1770.1995.tb0048.x

Lemöhfer, K., Dijkstra, T., & Michel, M. (2004). Three languages, one ECHO: Cognate effects in trilingual word recognition. Language and Cognitive Processes, 19(5), 585-611. https://doi.org/10.1080/01690960444000007

Leung, Y.-K. I. (2007). Third language acquisition: why it is interesting to generative linguists. Second Language Research, 23(1), 95-114. https://doi.org/10.1177/0267658307071604

Llama, R., Cardoso, W., & Collins, L. (2010). The influence of language distance and language status on the acquisition of L3 phonology. International Journal of Multilingualism, 7(1), 39-57. https://doi.org/10.1080/14790710902972255

Odlin, T. (2003). Cross-linguistic influence. in C.J. Doughty & M.H. Long (Eds.), The handbook of second language acquisition (pp. 436-486). Oxford, England: Blackwell. ttps://doi.org/10.1002/9780470756492.ch15

Riestenberg, K. J., Jan, H. S., Maimone, L., & Sanz, C. (2015). Cross-linguistic effects of Arabic experience on L3 learning and explicit knowledge. Arizona Working Papers in SLA & Teaching, 1-25.

Rothman, J. (2010). L3 syntactic transfer selectivity and typological determinacy: The typological primacy model. Second Language Research, 27(1), 107-127. https://doi.org/10.1177/0267658310386439

Sanz, C. (2000). Bilingual education enhances third language acquisition: Evidence from Catalonia. Applied Psycholinguistics, 21, 23-44. https://doi.org/10.1017/ s0142716400001028

Sanz, C., Hae, I. P., & Lado, B. (2014). A functional approach to cross-linguistic influence in ab initio L3 acquisition. Bilingualism: Language and Cognition, 18(2),116.https://doi.org/10.1017/S1366728914000285

(ProQuest: Appendix omitted.)