Download files
Abstract
Corpus studies have shown that native speakers’ spontaneous speech contains disfluencies, such as pauses, filled pauses, and various forms of retracing (Candéa 2000, Blanche-Benveniste 2010). Such behaviour is accounted for by the cognitive load required in working memory for the activation of words and the processing of sentences in real time (Levelt 1989). Unsurprisingly, similar phenomena characterize the spontaneous speech of L2 learners, who, depending on their level of proficiency, produce them at significantly higher rates than native speakers (Towell et al. 1996, Temple 2000, Hilton 2014).
The following utterance provides an example of disfluent French L2 speech:
(1) File: UWI Corpus: 0831_Ex.cha (l.177-183):
*STU: et [/-] &hum le gagnant &hum (.) yeah@s &euh peut [//] va &euh (.) faire un &hum (.) cd avec leurs chansons, quelque chose comme ça, &*INV:oui, &euh oui .
%mor: conj|et det|le&MASC&SING n|gagnant L2|yeah v:mdl|aller&PRES&3SV v:mdllex|faire-INF det|un&MASC&SING n|cd&MASC prep|avec det:poss|leurs&PL n|chanson-PL&FEM det:gen|quelque n|chose prep|comme pro:dem|ça adv:yn|oui .
"and there is a (filler) the winner (filler) yeah (filler) can is-going-to (filler) make a (filler) cd with their songs, or something like that, (listener backchannel), (filler) yes."
Research in this domain has usually been linked to the establishment of quantifiable, objective, and reliable temporal measures correlated with the perception of fluency (Lennon 1990, Freed 2000). Tavakoli & Skehan (2005) distinguish three types of fluency: speed fluency, concerned with the rate of speech, breakdown fluency, concerned with silent and filled pauses, and repair fluency, concerned with repairs. Among the latter, Olynyk et al. (1990) establish a distinction between progressive repairs (repeats, but also fillers) that are not necessarily detrimental to fluency as they allow native speakers and highly proficient L2 learners to avoid silent pauses within utterances, and regressive repairs (self-repairs and false starts) that affect fluency.
Per current models of L2 speech production (Kormos 2006), the abundance of disfluencies in non-native speech is accounted for by a lack of automatization of the processes of lexical retrieval and syntactic production. It therefore involves the conscious arrangement of individual lexemes into syntactic units, a laborious process that rapidly overloads the working memory (Hilton 2014). Additionally, as suggested by Temple (2000), self-repairs of L2 learners differ not only quantitatively, but also qualitatively from those of native speakers: the former, again with variations depending on their proficiency, make many corrections of syntax and morphology, while the latter focus on lexis or register.
In the spirit of the grammatical study of self-repairs by Fornel & Marandin (1996), the present study investigates the morphosyntactic properties of retracing phenomena (repeats, self-repairs and false starts) produced by L2 learners of French. It is based on the careful quantitative 87 analysis of data from the UWI learner corpus (Péters 2014), and offers a qualitative analysis based on current linguistic theories, such as Distributed Morphology (Halle & Marantz 1993). Our hypothesis is that these phenomena, usually considered prime examples of performance hiccups, manifest in fact syntactic regularities, which can reveal aspects of the underlying grammatical competence of learners.
The UWI learner corpus is a longitudinal corpus of oral productions by 10 Jamaican learners of French, engaging in 67 one-on-one conversations with the investigator over a nineteenmonth period. This corpus of 54,824 words has been transcribed, annotated, and encoded using the CHILDES conventions (MacWhinney 2000). Each orthographic line is coupled with a morphological decomposition line, pruned of fillers and retracing, as shown in (1) above, and has been linked with the corresponding audio-file in electronic format. The CLAN system provides inbuilt tools for the automatic analysis of the corpus. The participants, all diglossic natives of Jamaican Creole and Jamaican English, were learning French at the University of the West Indies, and had never been immersed in a French speaking country at the time of the interviews.
The corpus was primarily compiled to assess the grammatical characteristics of their Interlanguage, and especially the influence of their L1s on the acquisition of an L2. However, issues around the transcription of disfluencies immediately came to the fore as the transcription conventions in CHILDES provide for researchers to encode retracing in such a way that the original segment is not parsed on the morphological tier.
Our exclusive concern is on the analysis of unexpected asymmetries characterising retracing of morphosyntactic features by learners at different levels of proficiency: more precisely, articles and verbal clitics, two important topics of SLA research on the acquisition of French (Prévost & Paradis 2004).
In confirmation of previous research showing the prominence of the masculine gender in L2 interlanguage grammar of definite articles (Dewaele & Véronique 2001, Granfeldt 2004, Péters & Stewart 2009), the data shows that, in cases of self-repair with a change of gender of a definite determiner, most occurrences go from the masculine le to the feminine la. So, self-repairs of the type: le [//] la cité (the city) are observed, but very rarely of the type: la [//] le cours (the course). In this case, if lexical activation proves difficult, repeats: le [/] le mot (the word) or la [/] la présence (the presence), or repairs involving the whole noun phrase: la droit [//] le droit (law studies) are observed. Another feature, more characteristic of lower proficiency students, that clearly distinguishes them from native speakers, is the tendency to omit the third person subject pronoun when repeating lexical verbs (ils enseignent [/] enseignent). Interestingly, they never omit that pronoun with auxiliary verbs, except in cases of negative or interrogative structures, and never omit the first-person pronoun. These facts inform us on the featural composition given to clitic pronouns and on the default value of masculine definite determiners in the Interlanguage competence of these speakers.
Our goal will be to propose a syntactic representation of these retracing strategies. Such quantitative and qualitative research can have benefits not only for a better understanding of the conscious (Monitor) and unconscious grammatical processes, but also to inform the pedagogy of French as a second or foreign language. Further research will compare these results with other L2 learners of French from other linguistic background and with native speakers.
REFERENCES
Blanche-Benveniste C. (2010). Approches de la langue parlée en français. Paris: Ophrys. 88
Candéa M. (2000). Contribution à l’étude des pauses silencieuses et des phénomènes dits “d’hésitation” en français oral spontané. Thèse de doctorat nouveau régime, Université de Paris III.
Dewaele, J.-M. & Véronique, D. (2001). Gender assignment and gender agreement in advanced French Interlanguage: A cross-sectional study. Bilingualism: Language and Cognition 4, 275–297.
Fornel M. de & Marandin J.-M. (1996). L’analyse grammaticale des auto-réparations. Le gré des langues, 10, 8- 68. Freed B. (2000). Is fluency, like beauty, in the eyes and ears of the beholder? In H. Riggenbach (ed.), Perspectives on fluency (pp.243–265). Ann Arbor, MI: University of Michigan Press.
Granfeldt J. (2004). Domaines syntaxiques et acquisition du français langue étrangère. L'exemple du DP. Acquisition et Interaction en Langue Etrangère, 20, 47-84.
Halle M. & Marantz A. (1993). Distributed Morphology and the Pieces of Inflection. In: K. Hale & S.J. Keyser (eds.), The View from Building 20 (pp.111–176). Cambridge, MA: MIT Press.
Hilton H. (2014). Oral fluency and spoken proficiency: considerations for research and testing. In: P. Leclerq, A. Edmands & H. Hilton (eds.), Measuring L2 proficiency. Perspectives from SLA (pp.2751). Bristol: Multilingual Matters.
Kormos J. (2006). Speech production and second language acquisition. Mahwah, NJ: Lawrence Erlbaum.
Lennon P. (1990). Investigating fluency in EFL. Language Learning, 40 (3), 387-417.
Levelt W. (1989). Speaking: from intention to articulation. Cambridge, MA: MIT Press. MacWhinney B. (2000). The CHILDES Project: Tools for Analyzing Talk. Mahwah, NJ: Lawrence Erlbaum Associates.
Olynyk M., D'Anglejan A. & Sankoff D. (1990). A quantitative and qualitative analysis of speech markers in the native and second language speech of bilinguals. In: R. Scarcella, R. Anderson & S. Krashen (eds.), Developing communicative competence in a second language (pp.139-155). Rowley, MA: Newbury House.
Péters H. (2014). Étude sur la complémentation infinitive des verbes de déplacement sur la base d'un corpus de productions orales d'apprenants jamaïcains de FLE et suggestions pédagogiques. In: L. de Serres, F. Ghillebaert, P-A. Mather & A. Bosch (eds.), Aspects culturels, linguistiques et didactiques de l'enseignement-apprentissage du français à un public non-francophone (pp.62– 84). Québec: Association Internationale des Études Québécoises.
Péters H. & Stewart M. (2009). The syntax of determination in a corpus of oral productions of Jamaican learners of French: word order and agreement patterns in nominal phrases. The International Journal of the Humanities, 7, 51-62
Prévost P. & Paradis J. (eds.). (2004). The acquisition of French in different contexts. Focus on functional categories. John Benjamins.
Tavakoli P. & Skehan P. (2005). Strategic planning, task structure and performance testing. In: R. Ellis (ed.), Planning and task performance in a second language (pp.239–277). Amsterdam: John Benjamins.
Temple L. (2000). Second language learner speech production. Studia Linguistica, 54 (2), 288-297. Towell R., Hawkins R. & Bazergui N. (1996). The development of fluency in advanced learners of French. Applied Linguistics, 17(1), 84–119.
