THE INDIVIDUAL GRAMMAR APPROACH, OR THE POSSIBILITY OF

AN IDIOLECT-DRIVEN SYLLABUS IN LANGUAGE LEARNING

Camilo Andrés Bonilla Carvajal

University of Delaware, Department of Linguistics & Cognitive Science

Abstract

This paper addresses the concept of facilitating spoken word production by using the most recurrent lexical phrases from the learners’ idiolect. A quasi-experimental design was used in the application of this method over a 24-week period comparing the effects on a group of independent language learners (N = 10) in terms of lexical syntactic recall in semi-directed conversations and interlinear translations. The mixed approach analysis of the results reveals that the method has had until now a marked positive effect on the oral and written production of languages due to the explicit exposure to learners’ own idiolect in the foreign language, together with cognitive learning strategies. Guidelines for the application of an individual grammar approach are outlined, given the factors that influence stand-alone learning.

Keywords: idiolect re-encoding, translation, independent learning, metacognition

Resumen

Este artículo desarrolla el concepto de facilitar la producción léxica oral por medio de traducir las recurrencias léxico sintácticas más frecuentes del propio idiolecto de cada aprendiente. Para ello se utilizó un diseño cuasi experimental durante veinticuatro semanas a fin de comparar los efectos del método sobre un grupo de aprendientes autónomos de lenguas (N = 10). Las unidades de datos fueron la recuperación léxico sintáctica en conversaciones semidirigidas y la elaboración de traducciones interlineadas. Un análisis de tipo mixto revela que el método ha tenido hasta el momento un efecto positivo en la producción oral y escrita debido a la exposición explícita de los aprendientes a su propio idiolecto en lengua extranjera junto con estrategias metacognitivas. Se ofrecen pautas para la implementación de un enfoque tipo gramática individual tomando en cuenta los factores que intervienen en el aprendizaje autónomo.

Palabras clave: recodificación idiolectal, traducción, aprendizaje independiente, metacognición

Fecha de recepción del artículo: 6 de diciembre de 2013

Fecha de recepción de la versión revisada: 31 de julio de 2014

Fecha de aceptación: 28 de octubre de 2014

La correspondencia relacionada con este artículo debe dirigirse a:

Camilo Andrés Bonilla Carvajal

125 E Main St.

Newark, Delaware

USA, 19716

andresca@udel.edu

Introduction

 

The role of idiolect in foreign language (fl) learning has received little attention to date, at least from mainstream theorists or researchers in fl learning. As B. F. Skinner contended, linguistics has utterly ignored the nature of individual speakers (2002: 14); nevertheless, learners’ idiolect might yield an appealing insight in second- and third-language acquisition studies if regarded as a breakthrough for shaping the sought-after inner curriculum (Corder, 1967; Pienemann, 1985), i.e., learners’ inner order of acquisition.

This study aims at describing and testing whether recording, transcribing and translating one’s L1 verbal repertoire is a feasible way to activate lexical access in fl. This study involved the participation of 10 test subjects, who undertook the same procedure (recording, transcribing and translating their own speech) as a metacognitive strategy for spoken-word production in different fls (two learners each for English, Spanish, and Latin, and one for German, Dutch, Afrikaans, and French). Therefore, I shall begin with a brief overview of the main concepts developed in this paper, a description of test-subject characteristics, data collection instruments and procedures, data analysis, results and closing remarks.

 

 

Theoretical framework

 

In this project the notion of translation in fl learning is explored from a new perspective, because translation techniques applied to learning remain nowadays an almost unexplored field, partly because the benefits of translation in language teaching have been totally ignored and debased (Witte, Harden & De Oliviera Harden, 2009: 2). This is reinforced by authors such as Krashen (1981), Chaudron (1988) and Ellis (1988), amongst others, who, in conjunction with biased attitudes towards L1 in the classroom, have stated — with no empirical evidence — that translation leads to interference or error fossilization in fl learning (Bonilla Carvajal, 2014).

 

 

Translation

 

Translation is not, as usually regarded, a barely linguistic undertaking; on the contrary, it is a communicative action yielding cognitive dimensions to fl learning (De Mejía, 1987). Furthermore, translation is an inevitable part of the process itself. As shown by Hentschel (2009: 18), empirical neurologically-based evidence supports the notion that, once a speaker enters puberty, foreign lexical units are accessed from a cortical area differing from those involved in L1, as detected in fMRI scans where different brain regions are activated according to the language in use. It is therefore necessary not to treat foreign lexical units as a single entity, but instead as two interconnected realities or available linguistic forms resulting from the constant use of the two language systems (Ellis, 2006: 165). This is corroborated in the study by Thierry and Wu (2007) on unconscious automatic translation detected as a necessary part of learning. Translation is the mechanism whereby learners can bridge the gap between fl and their mother tongue. This notion is also supported and further explored by Sadeghi and Ketabi (2010), as well as by Hummel (2010). We intend to provide independent language learners with a metacognitive tool for building active vocabulary and context-sensitive discourse strategies (cf. Hummel, 2010; Sadeghi & Ketabi, 2010; Witte et al., 2009: 4). The aim of language learning is to build bilingual and intercultural competence, rather than monolingual encoders (Hentschel, 2009: 28); in this respect, renewing studies on translation in fl learning is relevant, all the more in view of the current discredit of classroom-based practices, which only creates a methodological inhibition for learners and teachers alike (Ibero, 1997: 111): “se han confundido los medios específicos del aprendizaje de la traducción con los del aprendizaje de una lengua extranjera. Y si aún se acepta el que se conciba la traducción pedagógica como recurso posible del aprendizaje de una lengua extranjera, la noción misma de traducción aparece muy deformada en esta práctica…”.1

This traditional proscription (deformation) supposes a harsh barrier to researchers for finding a disciplinary scope to develop translation as an object of research (Sánchez Iglesias, 2009: 23). There are various examples of punishment for L1 usage, or L1 — fl contrasting in classrooms (cf. Prodromou, 2000; Deller & Rinvolucri, 2002). Nonetheless, “translation is not a dangerous technique in itself, provided its nature is understood, and its use is carefully controlled: and translation is, in itself a valuable skill to be imparted to students” (Catford, 1980: viii).

 

 

Inner syllabus

 

The concept of inner syllabus derives from the notion of verbal repertoire, that is, the individual nature of learning. “One should not forget that the process of language learning is continuous and individual. No two users of a language, whether native speakers or foreign learners have exactly the same competences or develop them in the same way. Any attempt to establish ‘levels’ of proficiency is to some extent arbitrary, as it is [in] any area of knowledge or skill” (Council of Europe, 2001: 17, emphasis added). The inner syllabus is the set of words, phrases, idioms, syntactic patterns and semantic domains that are to be taught. These items should not be taken out of a pre-made syllabus unifying different levels, cognitive styles and learning preferences, but out of the salient verbal behaviour of each individual learner (active vocabulary); in other words, their idiolect or inner verbal repertoire: their individual grammar.

Thus, based in a study by Brown, Johnson (2008: 117-120) analyses the possibility that learners might have a previously established acquisition order in fl, just as they have it in L1, akin to a programme inside of each learner’s head, an inner programme:

 

Piense por un momento en las posibles implicaciones de este concepto para la enseñanza de idiomas. A la mayoría de los aprendices en la mayor parte de los salones de clase se les impone un programa de estudios desde fuera. El libro de texto o el maestro (y con frecuencia, en última instancia, la Secretaría de Educación) les dice qué morfemas y estructuras gramaticales van a aprender y en qué orden. En otras palabras, se impone a los aprendices un ‘programa de estudios externo’. Pero si los aprendices realmente cuentan con un orden propio de aprendizaje, ¿tiene algún caso imponerles otro orden? ¿Qué caso tiene darles un programa de estudios externo si ya tienen uno interno? Esta emotiva idea ha rondado muchos debates en torno a la enseñanza de idiomas en años recientes y tiende a evocar intensas pasiones (Johnson, 2008: 120).2

 

The learner–centred (inner) syllabus will focus on this natural choice made by speakers, finding and providing the most suitable translation to re–encode and internalise the idiolect (L1) in the target language (fl). The learners’ task will be to use their own idiolect (by means of spontaneous speech samples, as in recordings) to identify, translate (especially the high frequent phrases and vocabulary), and then render them with a proper equivalence suitable to the intended register in the target language.

In this design learners decide and control the order, pace and selection of what they are willing (or feel confident) to learn, working always inside the scope of their finite (at the moment) idiolects. The individual grammar (inner syllabus) approach thus empowers learners’ responsibility and decision through a learner-centred design based on idiolect, translation and learning strategies.

Considering the aforementioned facts, as they are supported by experimental researchers and theoretical authors alike, the verbal repertoire or idiolect, might be used as an inner curriculum to build communicative competence in a foreign language. Accordingly, the aim in this small-scale exploration study is to answer the following questions:

• What is the impact of translation and an idiolect-driven syllabus to elicit spoken word production in independent language learners?
• How does learners’ spoken lexical richness augment in spontaneous speech during the implementation of idiolect re-encoding protocols?
• How does idiolect translation improve free lexical-syntactic retrieving?

 

Methodology

 

Participants

 

The target population comprised adult non-impaired independent language learners (N = 10), these participants were recruited as a voluntary sample. Learners followed — with different degrees of continuity and commitment — a minimum amount of accumulated hours using the idiolect re-encoding Protocol (see below) to achieve the results.

 

Hypothesis

 

As a summary of conceptual tenets to guide the individual grammar approach, I summarise the following:

 

1. Structural randomness represents an accurate course of action to approach a foreign language, because it is the only one that more closely resembles the nonlinear nature of language structure (see Lewis, 2000: 184). Constant exposure to nonlinear input (e.g. authentic materials) follows the randomised path of interleaved and spaced practice that in actuality enhances oral production and retrieval (Bahrick, Bahrick, Bahrick & Bahrick, 1993; Taylor & Rohrer, 2010).

2. The individual grammar (internal syllabus) consists of continuous translations into a foreign language of all unities of thought comprising our verbal repertoire (idiolect). Interference and transference (semantic, structural and phonemic) are usually associated with negative communication habits in a foreign language (see Porter & Duncan, 1953: 61). The reason for an interference to occur however lies in the actual verbal repertoire of one’s mother tongue, whose structures and semantic patterns are arbitrarily re-uttered using L2 terms (Anderson, 2003). The idiolect, as a result, could potentially be a new horizon to curricular development yet to explore — if observed as the pivotal basis whereby an inner syllabus could be applied. The instrument to attain such re-encoding would be translation.

3. Context of utterance is a pivotal cue for the brain to retrieve passive vocabulary from the internal lexicon (lexical access). This validates applying translation as a tool for triggering associative memory and morphosemantic retrieval from passive vocabulary (Geiger, 2008; Meyer, Schvaneveldt & Ruddy, 1972; Taylor & Rohrer, 2010), due to the activation of lexical accessibility (Aitchison, 2003) by means of contrasting L1 deep structures with L2 surface structures.

4. A speaker is said to be fully competent in a foreign language as soon as the active vocabulary from his/her idiolect is completely reconstructed in that language, and when such reconstructed idiolect has been internalized so as to recall and produce it either orally or in writing in a different context to that where it was initially reconstructed.

 

 

Protocols of idiolect re-encoding

 

These are the activities learners were to complete at home.

 

1st Protocol: verbal distillation

1. Learners ask themselves something (concerning a personal matter, something they have enough knowledge of to speak comfortably free); they can also improvise soliloquies.

2. They try to answer each question (speaking out loud) bilingually, in code-switching. For instance: “I think ich denke that she is a good person dass sie eine gute Person ist and besides… und außerdem…” Another example with Latin: “Tomorrow early, Cras mane I’m going to the University and… Universitatem ī́bō atque… after that… postquam hoc…” Understandably, they will be in need of vocabulary or constructions. When this happens, they switch to the L1 to complete the meaning and switch back to the foreign language to finish the answer. The foreign language is used as much as possible, and every word/expression in native tongue they restore to when feeling devoid of vocabulary is written down as is. Once they have answered the question, those words or expressions are looked up in a bilingual dictionary, a translation corpus, or a phrasebook. Learners write the equivalents in foreign language and re-answer the same question, this time completing what they ignore with the found expressions.

3. It will be necessary to restate and re-answer the same question a number of times before it can be responded comfortably and without interruptions. As calibrations (every time a full answer is given) go by however, learners will see how this becomes easier and more active vocabulary is added to their current speech.

 

 

2nd Protocol: idiolect rewriting

 

Learners record themselves speaking, or using their own writings to know their idiolect (mental lexicon), then they translate each phrase into the foreign language and arrange everything as an interlinear text for further reference and to carry it for reading or listening on a day-to-day basis. They also record the foreign language version of their idiolect to assess pronunciation improvement, e.g.:3

 

 

 

Data collection: instrument and procedures

 

Since our intention was to provide a fuller and richer understanding of the phenomenon (research questions), obtaining and comparing sources (triangulation) suited better this mixed approach of research. The first three instruments (questionnaire of linguistic profile, field notes and learners’ journals) were used to find and name key ideas or recurrent concepts, then organising them into categories to find relationships (patterns) aiming to present data as a reliable foundation to answer the research questions.

Three quantitative instruments of data collection (i.e. diagram of quantitative linear fluency, learners’ interlinear translations, and analysis of lexical richness) were computed and compared weekly to see whether a lexical-semantic growth occurred in learners after the implementation. For the diagram of quantitative linear fluency, learners’ semi-spontaneous speech in foreign language was recorded during guided conversations about themes they have good knowledge about (e.g. What did you do yesterday? What’s your favourite kind of music? What are your reasons for learning this language?), this was tape recorded during 50 seconds and then transcribed, as shown in (2). Two comparative axis show the number of seconds and the number of words the learner gets to produce, such ratio is charted at the beginning (two weeks) during and before the end of the implementation to see results (see Figure 1).4

 

(2)	Interviewer:	What’s your favourite kind of music?
	Learner:	Well / the : / my favorite kind of music is : pop music / and : Nelly Furtado / I like / yeah / I like the other bands / I : I don’t know / but / but / it is something good to hear / that other people like that / and : and : me too / I : I always like that music / and different artist / / yes : I think it is pop music / that is the : the : the best thing and : I was : I was a child / and a young person and I like it / I liked it already / that kind of music / I remember / yes : no / that’s it / / /

 

 

Figure 1. Quantitative model of linear fluency (foreign language)

 

 

50 seconds X 94 words

Ratio = 1.88 words/second

 

Learners were recorded and results compared always inside the same semantic field. In other words, their fluency was measured while they answered the same question — requiring them to use the same scope of vocabulary — in order to follow lexical growth always within a unique pattern.

Learners’ interlinear translations. At a first phase learners were asked to capture their spontaneous speech, either orally or in a written form by means of tape-recording themselves in casual face-to-face conversations, telephone conversations, voice chat or monologues. Personal diaries, chatroom interactions, text-messaging, social networking messages (e.g. Facebook’s wall commentaries, Twitter or written commentaries on You Tube videos, etc.) since they all belong to their lexical repertoire — in which their unities of thought are seen in action in natural contexts of utterance. These data comprised the content upon which an inner curriculum was built: Learners transcribed and printed their oral recorded interactions together with the written interactions (both double spaced). They were then required to translate beneath each word their idiolect into the foreign language. A quantitative analysis of errors (grammatical, stylistic, vocabulary or orthography) was carried out every fortnight. Besides all three learners were coached in translation techniques, tone, register, domain, deverbalization and finding semantic equivalences to translate their idiolect into a foreign language.

Analysis of lexical richness — type-token ratio model. This method for data analysis was employed with full-length recorded conversations in the foreign language. This has an already well stated and confirmed validity in applied linguistics (Read, 2000), as well as in word frequency distribution studies (Baayen, 2001). This is also validated by different studies on bilinguals’ lexical improvement and vocabulary acquisition. “The best known quantitative measure for samples of written texts or speech is the type-token ratio (ttr)” (Daller, Van Hout & Treffers-Daller, 2003: 199).

Nevertheless, the constraints of such methodology have been highlighted, the most noticeable one being that its reliability depends on text length. As texts under analysis increase, type words tend to decrease, leading to lose accuracy when computing lexical richness (Schmitt, 2010: 213). This study used mean segmental type-token ratio (msttr), in order to avoid the length of text effect on type-token frequency (Richards & Malvern, 2002). This is controlled by framing all elicited verbal responses within a semantic field (i.e. a question), knowing that if a learner is asked: what do you see in the picture? (while showing him/her a sunny landscape), s/he would hardly use vocabulary from a different semantic scope to answer, e.g.: political unrest, bailout, literacy, deception, snow, etc., since they are not pertinent to the question; instead the learner will convey a message inside the desired semantic field: trees, mountains, river, sun, birds, sky, light, warm, etc. The analysis will only consider time spans from 50 seconds up to 1:10 minutes per answer. This will limit the number of words used during the protocols.

Next on, specific samples from each instrument, the categories, patterns and results of the study are presented to offer a detailed view of the conclusions of the implementation and the hypothesis.

 

 

Data analysis

 

Categories extracted from learners’ journals and field notes

 

These are some categories extracted from learners’ journals and field notes. These data are drawn from constant patterns present all along the intervention.

Category 1. Improving by learning how to learn.

Before completing the protocols of idiolect re-encoding, learners were coached on translation techniques and learning strategies. Test subjects reported in their journals this previous training had a marked positive effect in their success when completing the protocols and overall performance.

Category 2. Learning by constantly using the language.

Although error avoidance was present from the very beginning, learners soon understood they had to regularly (daily, if possible) use the language (not merely study it), even if that meant error emergence. They realised errors are in fact an opportunity to learn.

Subcategory. Lack of time and absence of time management skills as critical drawbacks

Learners claimed lack of time as the chief hindrance for not completing the full battery of protocols. When they were reminded about techniques to manage time, they averred not being capable of controlling their schedules due to external commitments. This is a major concern for independent language learners willing to engage in autonomous, systematic language learning regimes.

 

 

Quantitative linear fluency

 

We recorded interviews with learners, choosing semantic fields close to their everyday life, work, and social relationships (i.e. questions about personal information), as part of the verbal distillation exercise (protocol N° 2), these questions were recorded after three days, after seven days, and at the end of the implementation.

The words learners used to convey the answer in each recording of 50.0 seconds were transcribed, and later computed as an indicator of performance and lexical retrieving. The longer the pauses, false starts and silence, the harder the accessibility to the mental lexicon. Figure 2 shows the progress acquired before, during and after the usage of the protocols.

 

Figure 2. Words used at each intervention stage (50.0 seconds/recording)

 

 

 

 

The semantic fields (topic questions) were the same for all participants, albeit different questions were asked at each stage. The numbers in columns represent the number of words uttered (following a logical, coherent structure) in each answer required. Pseudo words and half uttered words were not counted.

However clear, there are some variables that should be also considered: not all languages use the same amount of words to express the same meaning, for example:

 

(3)	Latin:	Heri mea matre sororeque computatro loquebar = 6 words
	Spanish:	Ayer hablé con mi mamá y mi hermana del computador = 10 words
	English:	Yesterday I talked to my mom and sister about the computer = 11 words
	German:	Gestern habe ich mit meiner Mutter und Schwester über den Rechner gesprochen = 12 words

 

Every language has different number of words to convey the same message, in this example Latin takes exactly the half of words of German to express an identical meaning. And that also applies to semantic fields (not every question demands the same amount of words). This ought not cloud the understanding of progress, that is, all languages in the above table should not be cross compared one another, but every one with itself, knowing that if in the first stage the learner used 15 words, and in the last 89, that is a progress in terms of acquired quantitative fluency within his/her target language.

 

 

Interlinear translations

 

As mentioned before, based on translations of learners’ transcribed interactions, a calculation of their orthographic, morphosyntactic error-free clauses and error-free verb forms was made to see progression or stalemate throughout the intervention. Texts under analysis were no longer than 2000 characters each:

 

Table 1. Number of errors in idiolect translation over 24 weeks

	Beginning	Middle	Ending
	2 weeks	12 weeks	24 weeks
English	19	61	94
English	12	48	80
Spanish	25	66	97
Spanish	30	67	101
Latin	15	51	89
Latin	13	54	90
German	22	70	115
Dutch	11	42	67
French	25	80	96
Afrikaans	16	34	62
	What did you do yesterday?	Tell me something interesting about yourself	What’s your favourite kind of music?

 

 

Translated texts were analysed every fortnight, however the axes in every line represent a control reading by the consultant-researcher. In the case of languages like German, French, Afrikaans, and Dutch, native speakers were asked to check the errors, as for Latin, advanced learners. These proofreaders were contacted through the web pages: Correct my text (<www.correctmytext.com>) and Conversation exchange (<www.conversationexchange.com>).

It can be seen how all languages presented a constant decreasing pattern of errors. This is due to the number of repeated words in texts. Learners’ task was enormously facilitated due to the rather reduced scope of vocabulary to translate (their own conversational idiolect, the inner syllabus). Lexical coverage of spoken discourse takes only around 2000 to 5000 word families (see for details and discussion Adolphs & Schmitt, 2003; Hirsh & Nation, 1992) and this was easily done during the 24 weeks, owing to in a 1.5 spaced text, with font size Arial 12, around 355 + words fit. If learners translated approximately one page a day, after 14 days (two weeks = 14 pages) they had covered ca. 4970 words. This far exceeds the vocabulary needed to start a conversation equal to a B1 level (according to the Common European Framework). Learners soon found, repeated, translated and internalised high-frequency phrases such as: I think, I am, this is not, Let’s do this, in my opinion, in other words, excuse me?, etc., because they all are inside contexts of utterance belonging to their general daily situational context. Learners reported the second or third time they encountered these expressions in their idiolect, it did not take the effort of looking them up in translation corpora or dictionaries — they simply knew them already — thus working within the scope of conversations, after the seventh week learners’ errors started to dramatically decrease, also meaning more acquired passive vocabulary.

As with the diagram of quantitative linear fluency, Spanish as a foreign language learners had the fastest degree of augmenting error-free clauses and error-free verb forms, due to constant exposure (aural immersion 24/7) and eve­ryday need of using oral and written Spanish. On the other hand, Latin learners had little progress with respect to error reduction, however the progress in the language was perceived due to the proximity of their L1 (Spanish) to Latin.5 The Afrikaans learner had also good percentage of decrease in errors, but this was also affected by his previous knowledge of English (that has mutual cognates with Afrikaans).

Statistically, all learners presented a perceivable enhancement of mediation competence in translation, orthography, and morphosyntactic accuracy. As a general measure they had a mean of 10.0 errors less every control revision. And a total of 60-70 errors (every 2000 words) less after the intervention. These results suggest that:

1. Working within the limited scope of conversations (oral idiolect) allow learners to a rapid progression in terms of writing skills, building vocabulary and phraseology.
2. It takes around two weeks of translation on a day-to-day basis — regardless of the linguistic family of the language, Germanic or Romance — to see an improvement of vocabulary and syntactic knowledge by means of translating one’s idiolect.
3. Learners presented a rather reduced degree of errors since the earlier weeks, in spite of not being advanced learners. This was possible with the usage of offline and online aids (translation software, translation corpora and memories of translation, as well as techniques like synonymy and paraphrase), which indicates that a previous work on metalinguistic knowledge and translation planning leads to better results from the onset in language learning.
4. Therefore a syllabus based on learners’ idiolect seems to be a feasible course of action to elicit written production, metalinguistic awareness, and vocabulary enhancement. On top of that, vocabulary closer to learners’ immediate daily situational context is more significant and consequently easier to recall. As they are exposed everyday to use and hear expressions, idioms and words, those semantic equivalences are less difficult to be retrieved and used when translating.

 

Analysis of lexical richness

 

For the purpose of this study, answering the research question with a positive or negative statement is simply not enough to give a proper account of activation of spoken word production. Nor is it merely counting the number of words in the speech line within 50 seconds. Notwithstanding spoken fluency is not a clear-cut notion to be assessed in second language acquisition (sla) studies (Riggenbach, 2000). There is a solid quantitative evidence of progress with the captures of spontaneous speech, and the number of words learners could use, however good, we have yet to see how the actual substance of such speech is aiming towards progress.

Test subjects were recorded when speaking during the control sessions with the consultant-researcher. In semi-directed conversations (interviews) they were elicited to activate monologic production. I followed the fluency measure model of lexical richness proposed by Yuan and Ellis (2003) and Richards and Malvern (2002).

 

 

A qualitative measure of fluency

 

Fluency was measured with the lexical richness type-token ratio model (for an in-depth discussion and analysis of this model see Daller et al., 2003). In this model the total number of words of a text are recorded and transcribed as tokens, i.e., total running words, and types: the total number of words not repeated. Types are then divided by tokens and the result multiplied by 100.

Semantic fields were controlled by recording and analysing only one question with one answer that covered the topic. Questions differed over the control sessions but always were within the range of knowledge of participants. The name of each language represents each learner. The decimals are percentages of lexical richness in oral answers.

 

Table 2. Type-token ratio to assess lexical richness in monologic production

	After (weeks)
	4	10	15	19	24
English	49.4	50.1	51.9	78.8	79.0
English	44.7	55.3	56.6	61.0	65.8
Spanish	82.5	80.2	83.0	84.1	90.3
Spanish	81.1	82.4	87.1	89.9	92.7
Latin	15.2	14.0	20.5	23.3	30.3
Latin	13.5	15.2	17.3	23.1	26.7
German	36.7	45.5	53.0	57.7	65.5
Dutch	33.4	36.3	42.0	44.3	55.5
French	45.7	54.3	56.8	66.2	71.1
Afrikaans	19.2	26.9	27.8	38.9	53.3

 

The above table shows a considerable lexical growth throughout the weeks that learners followed the protocols. These indexes of progression should not be cross compared between languages, but seen in isolation, for even the slower lexical richness growth always goes up. Independent variables in the case of Spanish learners (living in Santafé de Bogotá, a Spanish speaking city) and the easier access to internet, plus a surrounding environment requiring them to constantly express themselves orally, explain the quicker enhancement in terms of lexical development. But even contrasted with Afrikaans and Latin learners (languages with a higher level of grammatical difference with Spanish and with scarce resources to ensure constant exposure), learners also presented an important percentage of growth:

 

 

Latin learner 1	=	15.2 %	I	30.3%
Latin learner 2	=	13.5%	I	26.7%
Afrikaans learner	=	19.2%	I	53.3%

 

 

This improvement, and spoken word production activation by translating their idiolect and following the protocols leads to conclude that granted the resources and the force of necessity to translate and use equivalences from the mental lexicon to use a language, oral skills can be rapidly activated as a consequence of using the correspondences in fl of the unities of thought comprising the verbal repertoire in L1.

 

 

Final comments

 

This investigation was led by the question: What is the impact of translation and an idiolect driven syllabus to elicit spoken word production in independent language learners?

The proposed idiolect-driven syllabus consisted of a set of activities (a regime of study) at home and during the sessions with the consultant-researcher, intended for learners to systematically capture their oral (transcribed) and written everyday interactions in L1 to be translated. These activities, idiolect capture, and translation, are framed in the regime of study which is the central component (the techniques) of the stand-alone language learning model proposed as the individual grammar approach.

Learners found the task of collecting and transcribing their idiolects simple and straightforward. Translation (as mediation competence) raised linguistic and cultural awareness about meanings intended to be understood by a target population (foreign speakers). On the other hand, centring the vocabulary on learners’ idiolect, means centring the focus of attention on expressions of the highest frequency and meaningful vocabulary — considering the definition of new knowledge linked and processed with previous knowledge in meaningful verbal learning (see Ausubel, 1963). Spoken word production improved in terms of word/minute ratio and lexical richness. Learners’ translations also support this view, on account of the progressive reduction of morphosyntactic errors and vocabulary choice.

The word count average increased (improving quantitative fluency) due to the spaced-repetition of speech formulae (or lexical items) from learners very idiolect in L1. They soon learnt them together with their context of utterance
(in L1) strengthening the cognitive-semantic relation that allows to retrieve the expressions subsequently as a desirable verbal behaviour. Languages rely on ready-made constructions (Nattinger & DeCarrico, 2005) suitable to contexts of utterance. Working with expressions bound to contexts of utterance from learners’ daily situational context induced high probable word combinations (lexical items) to be relatively easy to incorporate into learners’ mental lexicon for later use.

The initial hypothesis of idiolect as a finite mechanism whose re-encoding with translation was a feasible means to activate spoken word production in independent language learners is partially confirmed. The experimental group only completed a portion of the protocols and yet, they had an appreciable rank of oral performance and accuracy in translations. Therefore a first generalization can be drawn that under semi-controlled conditions, if learners receive proper metacognitive and linguistic tools, they can discover and use the building blocks of oral skills. Nonetheless granted the higher level of responsibility needed in the individual grammar approach, it is yet to be answered whether learners’ performance is not due to internal factors (locus of inner control, self-esteem, resources or force of necessity) and not entirely because the inherent distribution of time and activities of the protocols.

Thus, to know precisely whether all sorts of learners would have the same (or better) levels of commitment to complete the protocols on their own, further examination is needed. Additionally, present results are only preliminary since more research is a must considering larger populations, having all the different variables isolated, so that their connections and hierarchies of influence to oral activation in independent settings can be established.

This study was conducted with a well-defined learning environment (10 independent language learners) and in a specific span of time (24 weeks), although arguably a longer testing period with different population spheres would authenticate the above results. Taking these data into account, I therefore share the opinion of Segalowitz (2003: 402) that “future research will have to determine which dimensions of psychological similarity (e.g., whether the learners’ intentions, feelings, etc., are important, or whether only linguistic contexts are important) are relevant…”, this is indeed an appealing field to sla studies because a better insight of learners’ limitations to access autonomy and self-determination can be gained.

As it is obvious, independent language learners must make many environmental adjustments to be surrounded by contingencies of reinforcement and reasons to use the language. Such adjustments are to be made once the force of necessity to utilise the language is already present on them, having said this, I do not advocate a unique or stiff perspective on stand-alone learning — not even from a sheer linguistic standpoint.

 

 

Limitations of this study

 

The chosen languages to be learnt were somehow close to learners’ L1; another study would have to confirm whether results vary in learning a language not from the Indo-European family, like Chinese, Hungarian, Finnish or Basque, and conversely if learners from different linguistic backgrounds experience difficulty apart from the evident morphosyntactic distance of languages. Plus it would be advisable a longitudinal study to test retention and lexical-syntactic retrieving in longer spans of time to confirm the capacity of remembering vocabulary when working with the mental lexicon.

 

Bibliography

 

Adolphs, S. & N. Schmitt (2003). Lexical coverage of spoken discourse. Applied Linguistics, 24 (4): 425-438.

Aitchison, J. (2003). Words in the mind: an introduction to the mental lexicon. Malden: Wiley-Blackwell.

Anderson, M. C. (2003). Rethinking interference theory: executive control and the mechanisms of forgetting. Journal of Memory and Language, 49: 415-445.

Ausubel, D. P. (1963). The psychology of meaningful verbal learning. New York: Grune & Stratton.

Baayen, H. R. (2001). Word frequency distributions. Dordrecht: Kluwer Academic Publishers.

Bahrick, H. P., L. E. Bahrick, A. S. Bahrick & P. E. Bahrick (1993). Maintenance of foreign language vocabulary and the spacing effect. Psychological Science, 4: 316-321. [Versión electrónica. Consulta: 20 de octubre de 2014 <http://www.pashler.com/Articles/Cepeda%20et%202008_psychsci.pdf>]

Bonilla Carvajal, C. A. (2014). “Grammar-translation method”, a linguistic historic error of perspective: origins, dynamics and inconsistencies. Praxis y Saber: Revista de Investigación y Pedagogía, 4 (8): 200-213.

Catford, J. C. (1980). A linguistic theory of translation: an essay in applied linguistics. Oxford: Oxford University Press.

Chaudron, C. (1988). Second language classrooms: research on teaching and learning. Cambridge: Cambridge University Press.

Corder, S. P. (1967). The significance of learners errors. iral, 4: 161-170.

Council of Europe (2001). Common European framework of reference for languages learning, teaching, assessment. Cambridge: Cambridge University Press.

Daller, H., R. van Hout & T. J. Treffers-Daller (2003). Lexical richness in the spontaneous speech of bilinguals. Applied Linguistics, 24 (2): 197-222.

Deller, S. & M. Rinvolucri (2002). Using the mother tongue: making the most of the learner’s language. London: Delta Publishing.

De Mejía, A. M. (1987). Reflexiones sobre la utilidad de la traducción en la enseñanza del inglés como lengua extranjera. Lenguaje, 16: 25-30.

Ellis, N. C. (2006). Selective attention and transfer phenomena in L2 acquisition: contingency, cue competition, salience, interference, overshadowing, blocking, and perceptual learning. Applied Linguistics, 2 (27): 164-94.

Ellis, R. (1988). Classroom second language development. New York: Prentice Hall.

Geiger, J. A. (2008). Neural encoding of contextual cues during fear extinction and reinstatement. (Doctoral dissertation, The University of Wisconsin-Milwaukee, 2008).

Hentschel, E. (2009). Translation as an inevitable part of foreign language acquisition. In A. Witte, T. Harden & A. Ramos de Oliveira Harden (eds.). Translation in second language learning and teaching (pp. 15-30). Bern: Peter Lang.

Hirsh, D. & P. Nation (1992). What vocabulary size is needed to read unsimplified texts for pleasure? Reading in a foreign language, 8 (2): 689-96.

Hummel, K. M. (2010). Translation and short-term L2 vocabulary retention: hindrance or help? Language teaching research, 14 (1): 61-74.

Ibero, J. (1997). Por entre las rendijas de la traducción. En L. F. Fernández & E. O. Arjonilla (eds.). Lecciones de teoría y práctica de la traducción (pp. 111-119). Málaga: Universidad de Málaga.

Johnson, K. (2008). Aprender y enseñar lenguas extranjeras: una introducción. México: Fondo de Cultura Económica.

Krashen, S. (1981). Second language acquisition and second language learning. Oxford: Pergamon.

Lewis, M. (2000). Learning in the lexical approach. In M. Lewis (ed.). Teaching collocation: further developments in the lexical approach (pp. 155-84). Hove: Language Teaching.

Meyer, D. E., R. W. Schvaneveldt & M. G. Ruddy (1972). Activation of lexical memory. Psychonomic Society Meetings, St. Louis, 1: 20-25.

Nattinger, J. R. & J. S. DeCarrico (2005). Lexical phrases and language teaching. Oxford: Oxford University Press.

Pienemann, M. (1985). Learnability and syllabus construction. In K. Hyltenstam & M. Pienemann (eds.). Modelling and assessing second language acquisition (pp. 23-99). Clevedon: Multilingual Matters.

Porter, L. W., & C. P. Duncan (1953). Negative transfer in verbal learning. Journal of Experimental Psychology, 46 (1): 61-64.

Prodromou, L. (2000). From mother tongue to other tongue. Teaching English. [Retrieved: february 19, 2008, from <https://www.teachingenglish.org.uk/article/mother-tongue-other-tongue>]

Read, J. (2000). Assessing vocabulary. Cambridge: Cambridge University Press.

Resnick, M. C. (1991). Introducción a la historia de la lengua española. Washington: Georgetown University Press.

Richards, B. & D. Malvern (2002). Investigating accommodation in language proficiency interviews using a new measure of lexical diversity. Language testing, 19 (1): 85-104.

Richards, J. C. & R. Schmidt (2010). Longman dictionary of language teaching & applied linguistics. London: Pearson Education.

Riggenbach, H. (2000). Perspectives on fluency. Ann Arbor: University of Michigan Press.

Sadeghi, S. & S. Ketabi (2010). Translation: towards a critical-functional approach. Babel, 44 (3): 4-13.

Sánchez Iglesias, J. J. (2009). La traducción en la enseñanza de lenguas extranjeras: una aproximación polémica. Biblioteca 2009, núm. 10. Red Electrónica de Didáctica del Español como Lengua Extranjera [Retrieved: july 3, 2009 from <http://www.mecd.gob.es/redele/Biblioteca-Virtual/2009/memoriaMaster/1-Semestre/SANCHEZ.html>]

Schmitt, N. (2010). Researching vocabulary: a vocabulary research manual. New York: Palgrave Macmillan.

Segalowitz, N. (2003). Automaticity and second languages. In C. J. Doughty & M. H. Long (eds.). The handbook of second language acquisition (pp. 382-408). Carlton: Blackwell.

Skinner, B. F. (2002). Verbal behavior. Acton: Copley Custom Textbooks.

Taylor, K. & D. Rohrer (2010). The effects of interleaved practice. Applied Cognitive Psychology, 24: 837-848.

Thierry, G, & Y. J. Wu (2007). Brain potentials reveal unconscious translation during foreign-language comprehension. PNAS: Proceedings of the National Academy of Sciences of the United States of America, 104 (30): 12530-12535. doi: 10.1073/pnas.0609927104.

Witte, A., T. Harden & A. R. de Oliviera Harden (eds.) (2009). Translation in second language learning and teaching. Bern: Peter Lang.

Yuan, F. & R. Ellis (2003). The effect of pre-task planning and on-line planning on fluency, complexity and accuracy in L2 monologic oral production. Applied linguistics, 24 (1): 1-27.