Theories of Language Learning, Appunti di Lingue

Scarica Theories of Language Learning e più Appunti in PDF di Lingue solo su Docsity! 14.02.2023 Interlanguage (1972)  A type of language produced by L2 learners in the learning process. By studying this process and its production, we can find out more about mental representations of it and mental processes. SLA starts with the notion of interlanguage, before that, the out-dated version  Contrastive Analysis and Error Analysis. CONTRASTIVE ANALYSIS HYPOTHESIS (CAH, 1957) It has been developed within the behaviourist approach. CAH assumes that mistakes made by learners are caused by the differences between L1 and L2, and by the transfer of L1 habits to L2  learners transfer linguistic habits which they gave in L1 to the L2. Consequently, these errors are predictable and avoidable. The different languages can then be analysed in a contrastive way, in order to highlight: Symmetries, similar behaviours, which cause positive transfer. Asymmetries, differences that are a source of negative interference, and they come along with language learning problems. But, many aspects of learners’ language couldn’t be explained by CAH so this led to take a different approach to analyse learners’ errors  ERROR ANALYSIS ERROR ANALYSIS (70’s) When learners produce correct sentences, they maybe simply repeating something they have already heard. BUT, when they produce sentences that differ from target language, we may assume these sentences reflect the learners’ current competence (transitional competence, understanding of the rules and patterns of the language). 1 Selinker called this system an “interlanguage” because the system is neither the L1 nor the L2, but something that the learner is building from environmental data, relying on internal mechanisms.  independently from L1 and L2. INTERLANGUAGE CARACTERISTICS Interlanguage has empirically been found to be: 1) Systematic It is governed by rules which constitute the learner’s internal grammar. Selinker argued that learners’ internal linguistic system is worthy of study in its own right: a language system that had to be taken on its own terms and not as some corrupted version of the L2. Q: How is the IL system organized? Focus on the form-function relationship in the IL. Development of questions in L2 English (Stage 3) Fronting  do- fronting, wh- fronting without inversion. Do you have a shoes in your picture? Where the children are playing? Do you can play tennis? 2) Dynamic Continually evolving as learners receive more input and revise their hypotheses about L2. However, the path through language acquisition is not necessarily smooth and even. 2 The process of self-reorganization of grammar knowledge and representation  As the internal grammars evolve with new forms, the system must adjust by restructuring. Each time a new form is acquired, a new organization may be imposed on the previous knowledge. First phase: present/infinitive for all verbs. Second phase: present for all except for past perfective  past participle 4. U-Shaped Behavior When this happens, learners seem to “unlearn” something as they incorporate new data. Stage 1: “went” (lexicon)  Stage 2: “wented/goed”  Stage 3: “went” // going back to accurate form. In stage 1, accuracy is purely coincidental, because it lacks the full representation of target-like functions and meanings. RESTRUCTING & U-SHAPED BEHAVIOUR Shift from an exemplar-based system to a rule-based system. In stage 1, irregular past tense verbs exist as lexical items: they’re not formed by any rules. When the regular past tense ending enters the system, it spreads as a rule throughout the concept of “pastness”, and thus even irregulars become subject to a rule. – (Barry McLaughlin) 5 INTERLANGUAGE, IMPLICATIONS FOR TEACHING Before starting to teach, we should observe the interlanguage level of the student. We cannot skip the developmental stages (perfective tense before imperfective tense)  natural order of acquisition. Teachability Hypothesis Experimental studies on L2 German, on word order (Pienemann 1984, 1989). He taught inversion (Stage 4) to ten 7- to 9-year-old Italian children in Germany and three adult learners of German in Australia. Those who benefited from the instruction were “ready” learners who exhibited evidence of stage 3 (verb separation) at the time of the teaching. Those who were “unready” learners didn’t gain anything because they had begun the study at stage 2 (adverb preposing). Interlanguage and errors: a phase that we need to go through to acquire the correct form. Making errors is sometimes a good sign. And making fewer errors is not always a good thing for language learning  can impact on development of learning. Accuracy-oriented students might avoid more complex forms and might not be able to learn them!! Advantages of L2 instruction  Instructed learners progress at faster rate along the sequences of acquisition.  They tend to become more accurate than uninstructed learners.  Developing more elaborate language repertoires.  Developing balanced linguistic abilities (oral, written, productive, receptive)  They are less likely to fossilize. 6 NATURAL ORDER HYPOTHESIS (Krashen) The reason why we develop the language always in the same way is because of acquisition process.  Internal strategies (morphology, phonology, syntax etc.) to process the language and these strategies are not influenced by external factors. (such as frequency, salience etc.) Regardless of native language, all L2 learners follow the same predictable order. The order of acquisition remains the same regardless of explicit instruction. Evidence  the studies of learners whose primary learning environment was outside the classroom >>> Spontaneous acquisition in informal setting. Morpheme Studies – Morpheme Accuracy Order Learners from different L1s develop theirs accuracy in using these morphemes following a predictable and universal order of acquisition. In analyzing each learner’s speech, researchers identify the obligatory context (the places in a sentence where morpheme is necessary to make the sentence grammatically correct) for each morpheme. An accuracy is calculated for each morpheme and these can be ranked from highest to lowest, giving an accuracy order for the morphemes. Irregular past is more frequent than the regular one so that’s why it appears before. 7 Possessive Determiners Stage 1: Pre-emergence: No use of “his” and “her”. “your” use for all persons, genders, and numbers. Stage 2: Emergence: Emergence of “his” or “her”, with a strong preference to use one of them. Stage 3: Post-emergence: Use of them in different shapes and different contexts. THE TYPOLOGICAL AND FUNCTIONAL PERSPECTIVE THE ROLE OF UNIVERSALS AND MARKEDNESS Typological Universals: attempt to discover whether there are types of languages.  similarity could be found due to genetic evolution or being influenced. We can make generalization across genetically unrelated and geographically unrelated languages. 10 The Notion of Markedness No language has a trial number unless it has a dual. No language has a dual unless it has a plural  Trial is more marked than dual – Dual is more marked than plural – Plural is more marked than singular. In a specific language, less marked elements/features are more frequent than marked ones. Less marked elements are more frequent cross-linguistically as well. A category is taken as marked if every language that has marked category also has the unmarked one but NOT VICE VERSA. We assume that universals could affect acquisition order, more marked forms would be the last to be acquired and would imply the acquisition of less marked forms. Q: If that’s the case, to what extent is the variability of learner languages limited? Can we find evidence of interlanguages that violate these generalizations? One answer to this question has been formulated as the Structural Conformity Hypothesis  All universal that are true for primary languages are also true for interlanguages. Relative Clauses - Accessibility Hierarchy The universal itself is known as the Accessibility Hierarchy (AH) by Keenan and Comrie (1977) 11 The prediction is: If a language has a relative clause of type X, then it will also have any relative clause type higher on the hierarchy, or to the left of type  implicational universal Thus, if we know that a language has object of preposition relatives (That’s the woman about whom I told you), we know that it also has subject, direct object, and indirect object relatives. Crosslinguistlically, these six possible types of relative clauses are in a markedness relationship which is hierarchical and implicational  Each lower (more marked) types is seen to be possible in a given language only if all other preceding (less marked) types are also possible. RELATIVE CLAUSES IN SLA The study of relative clauses was aimed at verifying whether Accessibility Hierarchy reflects the ease of relativization. Gass presented data from English learners with a wide range of native languages. Her study was based on data from free compositions, sentence combining and grammaticality judgments. It was argued that the production of relative clauses by L2 learners could be predicted on the basis of Accessibility Hierarchy. 12 Pragmatic phase: focus exclusively on vocabulary and pragmatics, no morphological marking. Phonological phase: the phonological regularities drive some (only apparently morphological) word endings in the interlanguages. Lexical phase: the value of number is expressed lexically (numerals, quantifiers), no inflection and no morphological marking for number and gender. (proto) morphological phase: inflexion stops being used randomly or according to purely phonological criteria, the overextension of (apparently) female forms decreases Morphological phase: the categories of the number and gender are expressed on the components of the SN, according to the sequences described. Tense and Aspect: The Aspect Hypothesis It claims that first and second language learners will initially be influenced by the inherent semantic aspect of verbs and predicates in the acquisition of tense and aspect markers. This approach is semantic in nature and focuses on the influence of lexical aspect in the SLA of tense-aspect morphology. Q: How do learners recognize what morphological markers (e.g. past tense, progressive) go with what verbs? Andersen presented a study of two native speakers of English, one child and one adolescent, learning L2 Spanish. the perfective markers (preterit) emerged with accomplishment (punctual verbs) and achievement verbs  TELIC VERBS se partió (s/he left), (accomplishment, punctual verb) enseæó (s/he taught), (achievement, durative verb) the imperfect markers emerged with verbs states and activities  ATELIC VERBS tenía (s/he had), (state) Bardovi-Harlig characterizes aspectual classes as follows: States persist over time without change (e.g., seem, know, need, want, and be, as in be tall, big, green). Activities have inherent duration in that they involve a span of time, like sleep and snow. They have no specific endpoint as in I studied all week and, thus, are atelic (e.g., rain, play, walk, and talk). Achievements capture the beginning or the end of an action as in the race began or the game ended and can be thought of as reduced to a point. Examples of achievement verbs include arrive, leave, notice, recognize, and fall asleep. Accomplishments (e.g., build a house or paint a painting) are durative like activities and have an endpoint like achievements. Based on Andersen’s study, the sequence of developmental stages: Perfective aspect Telic predicates > atelic predicates > states Imperctive aspect States > atelic predicates > telic predicates 15 The developmental of the perfective marking spreads from achievement verbs to accomplishment verbs (telic to activities and finally states (atelic). The situation is different for the imperfective marking, which appears later than the perfect, and spreads in the reverse order  from states to activities to accomplishments, and then to achievements. Thus, Andersen argued that, when tense—aspect morphology emerged in the IL, it was constrained by lexical aspect in terms of the types of verbs described above. Perfective morphology emerges with punctual verbs and verbs indicating achievements and accomplishments (telic verbs). The morphology then gradually extends to verbs expressing activities and states (atelic verbs). Imperfective morphology emerges with durative and/or stative verbs (i.e., activities and states), then gradually spreads to achievement/accomplishment and punctual verbs. Progressive morphology is strongly associated with durative and dynamic verbs (i.e. activities). To sum up: The Aspect Hypothesis claims that learners initially use the semantics of a verb to know which tense and aspect markers to use with that verb. Achievement and accomplishment verbs (i.e., telic verbs) are associated with perfective morphology. Activities and States (i.e., atelic) are associated with imperfective morphology . Activities are associated with progressive morphology. 14.03.2023 Cognitive Interactionist Approach and Emergentism It is adapted to language but still in cognitive studies. The idea of this approach, when we adapt an emergent perspective, we see the point of view that the way we conceive language process. There is no language specific devise or universal grammar. Domain-general abilities (categorization, use of symbols of interference, memory systems, attentional system, social cognition, pattern finding skills and so on), we rely on.  So data from the environment plays a crucial role. Domain-general abilities are those that are used in general cognition (also outside of language). Domain-specific: ones that are specific for language. Frequency: For Emergentism, all learning and representation in the mind is sensitive to frequency. In innatism, frequency does not play a role. We acquire frequent elements/structures before less frequent structures. 16 Link to this idea, Threshold of activation: the highest the frequency, the lowest the activation threshold. The items that are more frequently used, they need less stimulation to be reaccesed or reactivated.  the more we practice the language, the more we are quick to activate the structures. Power Law of Practice: The threshold of activation will be lower in the early phases of acquisition but play a less evident role during the late phases. Statistical Learning: An implicit statistical learning process. Learner is like a human computer that processes and tallies linguistic info in the input.  from this, grammar emerges over time. Under this scenario, everything you need is contained in the input data. Type: the word Token: is the occurrences of the word that takes place in different contexts. Priming: pre-activating word: raggionare – raggionamento. Creating a word with suffixes. For example -ino is a suffix : tavolino but in giardino, it’s not. Salience: Something which is easily perceived. How much it is important to its context. (-mento is more salient) Communicative weight of a form, how communicatively useful it is to understand and use it correctly. Frequent and salient words are going to be acquired before than less frequent ones. Consistency/Contingency: Reliability. How many functions does have “-s” ?  Cats  plural  hurts  3rd person  it’s  to be  Nick’s  possessive  its  pronoun However, it is not easy to acquire “-s”. So it has very low contingency. If a form has a lot of functions, it is not reliable. Before understanding the structure of morphology, you have to get the other forms ending with - ino. Acquisition is the result of frequency, salience, and contingency. This interaction affects the way learners perceive, process, memorize, re-activate (retrieve) linguistic structures and forms. These principles are not specific to language. Q: do we process the language when it is for L2? What is the effect of L1 on L2? 17 vi. Use of questions to verify if they understand vii. Metalinguistic info viii. Intonation emphasis ix. Use of L1 21.03.2023 Modified or Simplified Input Naturally simplified (spontaneous) vs pedagogically simplified (pre-planned) Teacher talk provides input which has been pedagogically simplified. Graded input: pedagogically simplified input but still intended to engage the learner in the search for meaning Dependent exemplification: aims at focusing the learner’s attention on specific properties of the language >>> intended to teach the language rather than to engage the learner in the search for meaning => is this still authentic comprehensible input?  Teachers tend to talk with louder and more distinct voice. Take longer pauses in speaking  Teacher talk presents a number of linguistic adjustments, at the formal level but also at the level the interaction  Use strategies like repetitions, recasts, prompting answers, feedbacks  Ungrammatical speech modification do not (should not) occur in the teacher talk variety 20 Foreigner talk is used by native speakers to L2 learners to facilitate comprehension. Often contains ungrammatical language, native speakers attempt to “mimic” what they think non- natives produce.  “You like soupe?” “No like soupe?” It may be beneficial for acquisition outside of the class but it has to be observed that in class, it can also be the cause of a slowdown and, in some cases, the arrest of the linguistic development  fossilization, stabilization, pidginization. Modified Input: Simplification & Elaboration: Adding details and saying something in different ways should be a part of the modified input. We can’t reduce too much. BACK TO THE NATURAL APPROACH: FROM EXERCISES (OR DRILLS) TO ACTIVITIES We refer to activities as opposed to audiolingual drills or cognitive learning exercise. For acquisition to take place, the topic used in each activity must be intrinsically interesting or meaningful so that the students’ attention is focused on the content of the utterances instead of the form. Affective-humanistic activities Dialogues, interviews, preference ranking, personal charts and tables, activities which use the imagination Problem-solving activities Carrying out a practical task, making a story using pictures, sorting out a problem using charts, graphs, ads, maps Games Guessing games, action games, contests, problem-solving games, etc. Content activities Presentations, reports, ‘show and tell’ activities, storytelling, music and drama activities, subject- based activities, etc. The purpose of activities is to supply comprehensible input, not to teach a specific structure, “since conscious concentration on structure and form may prevent focusing on the message and may thus impede acquisition” (Krashen & Terrell) The effectiveness of any activity can be measured by the interest it evokes: “discussing topics that are of interest to the students is not just a frill: it is essential if language acquisition is to take place”  Each activity focuses on a particular topic and/or situation and students will normally be made aware of the content of the activity.  The activity may also often have a specific form or structure that can be used repeatedly  this will provide learners with a familiar cognitive ‘frame’ to rely on. PROBLEMATIC ASPECTS OF COMPREHENSION INPUT HYPOTHESIS: COMPREHENSION IS NOT ACQUISITION  The Construct 1+1: the hypothesis is not specific as to how to define levels of knowledge so that it is not clear how it will be possible to gear the input provided one degree beyond that actual learner’s level of competence.  Comprehensible and Simplified Input: Comprehensible Input Hypothesis “fails to consider cases where the input does not help at all [role of extra-linguistic factors], and 21 underestimates the problem of the acquisition of form”  Not enough attention on psycholinguistic factors.  Is Input Really Sufficient? The strong claim that comprehensible input is both necessary and sufficient for L2 learning proved to be untenable in light of findings gleaned by Schmidt and by many others, who documented minimal grammatical development despite ample meaningful opportunities to use the language, even with young L2 learners. For example, children attending French immersion and regular English- speaking schools.  Input is undoubtedly necessary, but it cannot be sufficient. Comprehension-based Programs Teaching methods which provide a great deal of comprehensible input and aim for a low-anxiety environment.  Canadian French Immersion programs: the exposure to French is very high. Results with respect to the content: The achievements of students are similar, sometimes better, to those of non-immersion students. Results with respect to L2 language: Students attending immersion programs develop higher levels of proficiency (although not in all the abilities) in French than those non-immersion students studying French as a regular school subject. Limitations of immersion programs in Canada  Learners failed to achieve high levels of performance in some aspects of French grammar after several years of the study in the immersion programs  accuracy vs. fluency  Students just use their incomplete TL knowledge because they are rarely pushed to be more precise or more accurate  as a result of the communicative, focus-on-meaning (only) approach  Because students share the same interlanguage, they have no difficulty understanding each other. Therefore, there is little need for them to use appropriate TL form to negotiate for meaning. CONTENT-BASED LANGUAGE TEACHING Content-based language teaching (CBLT) is an instructional approach in which non-linguistic curricular content (such as geography or science) is taught to students through the medium of a language that they are learning as an additional language. CBLT comes in many different shapes and sizes, and in fact is called by different names and acronyms including Content-based instruction (CBI), English as Medium of Instruction (EMI, specific for English), Content and language integrated learning (CLIL) (teaching the subjects like science or geography and also the language that you are using to teach other subjects at the same time). 22 on forms. Activities such as fill in the blank, in which the learner does not have to pay attention to what the sentence means in order to complete the sentence, are focus on forms exercises. Technique of input enhancement: Input flood, Task-essential language, Input (textual) enhancement, Negotiation, Output enhancement, Dictogloss, Grammar consciousness-raising tasks, Input processing, Garden path. INPUT ENHANCEMENT (Michael S. Smith) A technique that aims at directing learners’ attention to formal properties of language, while also maintaining a focus on meaning. It entails any effort to make formal features of L2 more salient to learners and comes in two varieties: positive and negative. Positive input enhancement involves manipulating input in certain ways to make formal features more obvious to learners Such manipulations include louder voice, acoustic stress on something while the teacher is talking, bolding/highlighting particular features in written input. Negative input enhancement is basically feedback: the teacher draws a learner’s attention to an incorrect production to signal that the learner has violated the target norms. TEXTUAL ENHANCEMENT Boldfacing, underlying, capitalizing, color coding… It may have positive effects but also limited. Textual Enhancement generally helps learners to notice the target forms in input (and in the comprehension of the content). It is controversial whether it is also effective for enhancing learners’ intake new grammatical forms. While some researchers have only manipulated the input by Textual Enhancement (TE) as an independent variable, the other studies investigated effects of TE in the combination with other intervention. Shook investigated the effects of TE on Spanish present perfect tense and relative pronouns (quen/quein). The subjects were 125 university students. The results revealed that the two experimental groups who received the passages with TE performed significantly better than the control group in all tests. However, there was no significant difference between TE group and TE plus focus on form group, revealing that focus on form information played no role in L2 acquisition. DICTOGLOSS Basically, students work in small groups in order to summarize or reconstruct a text in the L2. 1. the teacher prepares a text that contains examples of the grammatical form to be studied. 2. The teacher reads the text to the students at normal speed while they take notes. 3. Students then work in small groups to prepare a summary using the correct grammatical structures. 4. Finally, each group presents their work to the rest of the class. 25 A language teaching technique which is aimed at teaching grammatical structures >> dictogloss activities encourage learners to focus on the form of the L2, while also being based in communication (focus on meaning). Used in task-based language teaching. Dictogloss activities have several advantages: They integrate the four language skills of listening, reading, speaking, and writing. They also give students opportunities to talk about both content and the language itself. Dictogloss activities are a useful way of presenting new factual information to students and encourage them to listen for key points >> focus on meaning. They give support to less confident students, as they are encouraged to participate in their groups as part of the structure of the activity. 28.03.2023 INPUT PROCESSING AND PROCESSING INSTRUCTION (Chapter 3, Nava) Learners have to make the appropriate form-meaning connection in order to process the sentence by assigning each linguistic element its specific function/meaning. This is something that L2 learners might fail or might be challenged. While we are exposed to L2, we might not have enough resources to have input processing (costly) like L1 individuals have. The most widely known model for input processing has been developed by Bill VanPatten. Research in input processing attempts to understand how learners parse, compute (and filter) input as they attempt to make sense out of it. Linguistic processing requires considerable computation power, especially for the computation of grammatical structures a detailed syntactic representation is necessary and processing mechanisms need to be automatized. More precisely, input processing refers to how learners do (or do not) link meaning to a piece of linguistic data — be it a word, a phrase, or a piece of a word (e.g., an inflection, an affix). Input Processing (IP) According to this approach, the human mind is considered as an information processor (as a language processor) that relies on a set of algorithms or grammatical rules in handling linguistic input and output. Limited Processing Capacity Comprehension for learners is initially very effortful in terms of cognitive processing and working memory. Learners are limited capacity processors and cannot process and store the same amount of information as native speakers can during moment- by-moment. 26 Only certain parts of input that are immediately relevant to the message content will be attended. Learners make use of certain universals of IP but may also make use of the L1 input processor. L2 Processing Principles  The Primacy of Meaning Principle a) The Primacy of Content Words Principle  Learners process content words in the input before anything else. For example, “The cat is sleeping.”  which is said in a context of a cat is actually sleeping so learners will try to isolate the lexical forms that encode the meaning of cat and sleep  key elements to the meaning of an utterance. b) The Lexical Preference Principle  relying on lexical items for meaning before grammatical forms when both encode the same sentence. // redundancy occurs as we saw in blocking/overshadowing = fossilized. E.g. I called my mother yesterday. c) The Preference for Non-redundancy Principle  Learners are more likely to process non-redundant meaningful grammatical markers before they process redundant meaningful markers. For example, -ing has more info “in progress” – Thus, if learners are confronted with something like -ing on verb forms, they will be forced to make this form–meaning connection sooner than, third-person -s because the latter is redundant and the former is not. - The First-Noun Principle  The First-Noun Principle  Learners tend to process the first noun or pronoun they encounter in a sentence as the subject. Learners will rely on word order (not case) as a cue to who did what to whom.  The Lexical Semantics Principle  Learners will rely on lexical semantics, where possible, instead of the First-Noun Principle (or an L1 parsing procedure) to interpret sentences. The cake was eaten by Chris.  the only possible agent is Chris. Cake cannot be agent. The cow was kicked by the horse.  forced to rely on grammatical competence. To sum… L2 learners use processing strategies for making form-meaning connections to the task of comprehension. Input processing refers to how learners do (or do not) link meaning to a piece of linguistic form. 27 Learners’ natural processing strategies must be identified at first in order to develop SI activities in which learners’ reliance on the strategy is altered and their attention is appropriately guided to process a particular form to get the meaning. SI activities have to work on particular processing process. The processing problem may vary depending on the specific L2 features that learners might find difficult to process (Word order, Pronouns, Relative clauses, Morphological inflections) The identification of a possible processing problem is thus crucial. Once the processing problem has been identified, teachers will be in a better position to develop suitable SI activity. OUTPUT HYPOTHESIS Starting from the end of the ‘80s, output has been also conceived as one on the factors responsible for L2 acquisition. The Output Hypothesis (also known as the Comprehensible Output Hypothesis or Pushed Output Hypothesis) is associated with Merrill Swain. As a result of immersion education in Canada. Immersion program is a school program that uses both L2 and L1 (in different proportions) during the course of years, from kindergarten to sixth degree. She observed about seventy secondary school pupils who had been in a French immersion program for seven years in Canada. Large-scale assessment of the linguistic outcomes of French immersion schools in Ontario, an English-speaking province of Canada  Comparison between students in immersion schools and French peers on the same tasks Optimal development in discourse competence, as well as optimal comprehension abilities and school content learning, but not in grammatical competence or in sociolinguistic competence for aspects that demand grammatical means. Children, despite the massive amount of input in the target language to which they are exposed daily, show many linguistic deficits in the production in L2 (especially in terms of verbal morphology) - Fluency vs Accuracy and Complexity On the basis of these results, Merrill Swain expressed concern that too much emphasis was placed on the role of input in SLA (following the Comprehension Input Hypothesis). She claimed that what was missing was output, and in direct contrast to Krashen’s concept of comprehensible input, she argued that learners need to be pushed to make comprehensible output during acquisition. Moreover, learners production was not elaborate as expected. Learners were not ‘pushed’ to produce output. OUTPUT AND NOTICING Output draws attention to the linguistic form. When the learner needs to encode meaning into forms or to modify his/her productions, because they are problematic or communicatively unsuccessful, he/she engages in an analysis or a reflection on linguistic forms. Output causes more noticing: “the ‘noticing/triggering’ function or what might be referred to as its consciousness-raising role. As a trigger of noticing target language forms, output is able to guarantee learner’s interlanguage development. Moreover, it is through production that learners are able to receive Negative Feedback (NF) during conversational interactions. 30 More specifically, during production learners notice:  the difference between their productions and those of others comparing what they said to what other say, the way they express  (noticing the gap)  the lack of elements or forms necessary to express the message they intend to convey  (noticing the hole) In the long term, drawing attention to the form promotes the acquisition process: Through the activity of producing the target language, L2 learners may become aware of what they need to acquire in the target language. An opportunity to produce language and gain feedback, which, by focusing learners’ attention on their speech  it may lead them to notice a mismatch between their production and the feedback. OUTPUT FOR HYPOTHESES TESTING Starting from the input, the learner elaborates hypotheses on the relationship between forms and functions. Those hypothesis can be tested through output. Output is an opportunity for learners to see what is acceptable and what is not in the target language. Such an activity of hypotheses testing is not necessarily planned and carried out consciously. The output encourages metalinguistic reflections, especially in interaction. “the metalinguistic function or what might be referred to as its ‘reflective role’”. Metalinguistic reflections: reflective self-hear, self-analyse, self-test, self- review. Negotiations can (even accidentally and consciously) contain or promote explicit metalinguistic reflections. This function can also be called negotiation of form. Learners may receive feedback by way of reflection on the target language form through conversational interactions. FROM SEMANTICS TO SYNTAX Comprehensible output is a necessary mechanism of acquisition. Its role is, at minimum, to provide opportunities for contextualized, meaningful use, to test out hypotheses about the target language, and to move the learner from purely semantic analysis of the language to a syntactic analysis of it. By semantic analysis, Swain meant reliance on words and real-world information to get meaning. By syntactic analysis she meant actual detection of the formal properties of language that help to convey meaning.  learners will pay more attention to how something is supposed to be said and not just to the message. OUTPUT AND AUTOMATIZATION Another function of output is to promote automaticity, which refers to the routinization of language use.  production plays an integral role in automaticity: continued use of language moves learners to more fluent automatic production. Producing language is a useful exercise, which gradually improves the pronunciation of new sounds, new words, phrases, utterances, 31 producing them and arranging them in succession in an increasingly fluent and accurate way, as well as 'automatic’  without conscious effort. Two main lines of Output and the developing of new knowledge (the notion of languaging), Output and the fostering automatization of existing language knowledge (Automatiticity & Fluency). LANGUAGING Recent development of the Output Hypothesis, influenced by new insights on language and language learning afforded by Sociocultural Approach to SLA, lead to the elaboration of the notion of languaging (Swain). The term ‘languaging’ (as a verb) entails a process, is a dynamic activity. Producing language does not simply mean to ‘put thoughts into words’, ready-made thoughts. By producing language (by languaging) we develop new understanding, new knowledge, we are involved in the process of learning.   We naturally ‘language’ to learn, to better understand new concepts, to appropriate knowledge. Through language we better understand language  ‘languaging about language’. Swain (2006) advocates that “languaging about language is one of the ways we learn a second language to an advanced level” and “it is part of what constitutes learning”. By talking about language we learn both through and about the language  developing both implicit and explicit knowledge. Reinterpretation of the ‘metalinguistic’ function of output. OUTPUT, AUTOMATICITY AND FLUENCY Fluency is a result of automaticity. One of the functions of output is to promote automaticity, which refers to the routinization of language use. Continued use of language moves learners to more automatic production, which leads to fluency. Producing language is a useful exercise, which gradually improves the pronunciation of new sounds, new words, phrases, utterances. Producing them and arranging them in succession fosters both a fluent and an accurate use of the language. Pushed output improves the articulatory gesture (the way we control the movements of articulatory organs), but frequency of use makes linguistic elements (especially words) more easily retrievable, more available, more ‘activated’ in the leaners’ mind (mental lexicon and grammar). Non automatic processes are deliberate, controlled, they require a significant amount of time and working memory capacity. Automatic and routinized processes involve less time and attentive capacity.They are hard to control, to alter and to suppress. Because the human mind is a limited processing system, automatization is important to free cognitive resources to process new language. Q: How to define an automatic process? 32 FLUENCY VULNERABILITY POINTS Learners may face challenges in terms of fluency. Kormos claims that whereas lexical, syntactic, morphological and phonological encoding is almost automatic in the L1 speech production, these mechanisms are only partially automatic even in case of advanced L2 learners. Pauses can be associated with online planning (reformulations, replacements, etc.). Pauses or a slower race of oral production can be the result of learners relying on L2 explicit knowledge, especially when deciding on function words (articles, prepositions, etc) or syntactic structures. FLUENCY WITHIN COMMUNICATVE LANGUAGE TEACHING (CLT) The dichotomous relationship between fluency and accuracy is the main concern in Communicative Language Teaching. Language displays for evaluation tended to lead to a concern for accuracy, monitoring, reference rules, possibly explicit knowledge, problem solving and evidence of skill-getting. In contrast, language use requires fluency, expression rules a reliance on implicit knowledge and automatic performance. Fluency activities are often the language activities that we find in the back of textbooks or listed only in teachers’ manuals or resource books.They may be seen as a less important or as optional activities that are worth doing but only if there is enough time. 35 Q: Which kind of practice can improve fluency development? Types of practice that requires learner ‘to use target utterances repetitively while conveying genuine meaning’. Practice should simulate the same conditions that apply in real-life communication, where ‘processing and formulation of utterance must be done in real time’ and where the primary focus is on message rather than linguistic accuracy. Practice activities should not require too challenging language  learners’ attention will be shifted away from the forms that are being practised (‘form defocus’), so that it can be engaged in conveying the message in a fluent way. Fluency activities should facilitate the transfer of automatized skills to contexts of real language use. TASKS DESIGN VARIABLES FOR L2 FLUENCY Q: How can teacher implement the design of fluency task? 36 STRATEGIC PLANNING Strategic planning has a deep effect on fluency resulting in faster speaking rate and fewer disfluencies. Giving learners some time to prepare to perform the task, considering what they are going to say and how they will need to say it. Planning should not be viewed as something ‘detached’ or extremely time-consuming but should be incorporated into the whole sequence of task performance. unguided planning learners can use their own devices or resources when planning a task guided planning leaners are given specific advices about what and how to plan participatory structure the planning may be taken by the learners working individually, in groups or with the teacher REPETITION Repetition is another effective technique to enhance fluency skills. It can be an inherent feature of the task, or can be part of the practice sequence. The repetition of the same task can help learners perform the task in a more fluent and complex way: while during the first performance learners tend to focus on message content and basic language needed to convey it, during the repetition task they can rely on the previous work of conceptualization and encoding and switch their attention to the selection and monitoring of more appropriate linguistic aspects. Gain in oral fluency during task repetition also seem to be regardless of proficiency level or task type. FLUENCY ACTIVITIES 37 Q: How can tasks be implemented? By manipulating task design and implementation options it is possible to engage psycholinguistic processes that underlie the emergency of any of the three aspects of production: accuracy, complexity, fluency. When learners engage in production, they tend to allocate their attention to either what they are going to say or how they are going to say it. Attention is a limited capacity. 40 When learners focus on a message, they do not have ‘enough’ capacity left to pay as much attention to the form of the message (and vice versa)  language production tends to favour one of the three factors  there is a ‘trade-off’ between fluency, accuracy and complexity. Learners how are ‘adventurous’ in their language use and open to taking risks may be more apt to experiment with complex language with respect to learners who are less willing to leave safe territory and are more prone to stich to what they know is correct. When learners produce the L2, they generally focus on a particular aspect of the task, in a mutually exclusive way: focus on accuracy makes it less likely that interlanguage change will occur and may result in slow speech (low complexity, low fluency). focus on complexity boosts the chances that new forms will be incorporated into interlanguage systems (low accuracy, low fluency). focus on fluency often results in faster speech and lower priority being attached to getting language right, or to the use of new forms. Performance on a particular task can help some of the areas of language development, not all of them. FOCUSED VS UNFOCUSED TASKS Tasks are usually taken to be unfocused; “no effort is made in the design and in the execution of the task to give prominence to any particular linguistic feature”. A task is focused when one or a set of linguistic features are made prominent, “although not in a way that causes the learner to pay more attention to form than to meaning”. Q: What do we mean by making linguistic features prominent? The researchers posit three degrees of involvement of a lexicogrammatical feature in the implementation of a task: In task-naturalness, a grammatical construction may arise naturally during the performance of a particular task, but the task can often be performed perfectly well, even quite easily, without it. In the case of task utility, it is possible to complete a task without the structure, but with the structure the task becomes easier. The most extreme demand a task can place on a structure is essentialness: the task cannot be successfully performed unless the structure is used. TASK ESSENTIALNESS AND OUTPUT HYPOTHESIS By performing a task, learners may be led to use output to reflect on their interlanguage and extend their linguistic resources (Samuda 2001). Things in the pocket task: try to profile a ‘mystery person’ on the basis of the things she/he has in her/his pockets. Students are provided with the objects and a chart aimed at guiding their hypothesis. 41 HOW CERTAIN ARE YOU? less than 50% certain (It’s possible) 90% certain (It’s probable) 100% certain (It’s certain) age sex profession In this task, modal verbs are extremely useful to express degrees of possibility/probability (epistemic modality). However, students can convey the same content by means of lexical means (maybe ..., it is possible ...), by ‘miming’ expressions, etc. The task has created the ‘semantic space’ for the epistemic modality: a need to mean, to express specific semantic meanings which the learners did not manage to fulfil, or fulfilled only partially. Learners were led to notice a hole in their interlanguage: the process of focusing the task is carried out, during task implementation. As the students report their findings, the teacher can reformulate the guesses expressed by using modal auxiliaries: she must be a ..., she could be ... TASK IMPLEMENTATION: PRE- AND POST-TASK PHASE Pre-task activities should target both the cognitive and the linguistic challenges posed by a task, and lessen the potential challenges in task implementation. Q: Why using pre-task activities? To introduce new language  New language may be introduced at this stage to kickstart interlanguage change. To increase the chances that restructuring will occur in the underlying language system  Reorganization of the interlanguage system through the integration of partially acquired structures may take place. 42 (1) by modifying input. Input modification occurs when the interlocutor perceives that the learner does not understand what is being said, and restates something by simplifying, exemplifying, or otherwise altering the original statement. (2) by providing feedback. Feedback occurs when the interlocutor uses particular devices to inform the learner about something he or she has said. These include such things as “Do you mean...?”, “I’m sorry. I don’t understand”, and “OK. I get it.” POSITIVE AND NEGATIVE FEEDBACK The term ‘feedback’ refers to some kind of NS response to what the learner has said; the feedback that learners receive during interaction can either be positive or negative. NF is an interlocutor’s interactional move that indicates explicitly or implicitly any nontarget-like feature in the learner’s speech. NF obtained during negotiation work or elsewhere may be facilitative of L2 development, at least for vocabulary, morphology, and language-specific syntax, and essential for learning certain specifiable L1-L2 contrasts”. IMPLICIT VS EXPILICIT NEGATIVE FEEDBACK Explicit NF is an overt error correction per se. The error is corrected and or some kind of metalinguistic explanation of the error is provided. 45 Implicit NF usually takes the form of recasts or negotiation moves (i.e. confirmation checks, CRs and repetitions) that indicate to the learner that there is a problem with his/her output. There is no direct signalling that the a correction is being undertaken. Explicit NF is relatively infrequent in naturalistic Interaction while implicit NF is more frequently provided. *He has dog 1)  reformulating it: ‘A dog’ > recast/reformulation 2)  alerting the learner to the error and providing the correct form: ‘No, you should say “a dog”’ > explicit correction 3)  asking for clarification: ‘Sorry?’ > clarification request 4)  making a metalinguistic comment: ‘You need an indefinite article’ > metalinguistic feedback 5)  eliciting the correct form: ‘He has ...?’ > elicitation 6) repeating the wrong sentence: ‘He has dog?’ > repetition RECAST Recasts are defined as target-like reformulations of ungrammatical utterances that maintain the central meaning of the original utterance. Utterances that repeat a learner’s incorrect utterance, making only the changes necessary to produce a correct utterance, without changing the meaning. Recast is as an implicit and input-providing kind of NF, i.e. , interactional move through which learners are provided with more linguistically target-like reformulations of what they have just said. The learner might repeat the recast, there can be some awareness but what part of recast has been perceived could be blurry so they may have trouble in understanding and addressing the problem. Recasts are an indirect form of correction, it is not clear to what extent they are relevant to acquisition. E.g. Learner: Why the man came to see his teacher? Instructor: Why did the man come to see his teacher? Well, because today is the woman’ s birthday… Possible problems with Recast:  Learners may not perceive them as negative evidence but as further input showing how to say the same thing in a different way.  Learners may look upon recasts as an acknowledgement that their message has been understood.  Learners may simply repeat a reformulation (recast) without true understanding, just to show compliance.  In order to make recasts efficient, they need to be followed by learner response. 46 ELICITATION The (correct) form is directly elicited by asking questions (e.g., "How do we say that in ...?"), by pausing to allow the student to complete the teacher's utterance (e.g., "It's a....") or by asking students to reformulate the utterance (e.g., "Say that again."). Explicit, ouput-prompting. Learners seem more likely to remember their own corrections than the corrections provided by the teacher. When “learners are pushed to self-correct, they may become aware of the gap in their knowledge and their attention may be directed to subsequent input”. It is aimed at eliciting the correct form from the learner through: 1)  The completion of the teacher’s own utterance 2)  The use of questions to elicit correct forms 3)  (The reformulation of the students’ utterance) E.g. LEARNER : what happen for the boat? NS : what? LEARNER : what’s wrong with the boat? CLARIFICATION REQUESTS A CR refers to an interactional move in which a speaker solicits aid in understanding a partner’s previous utterance by means of questions or statements of non-comprehension. It is an implicit and output-prompting kind of NF. The fact that the NNS is induced to make the actual correction, as opposed to hearing about the correct form, is in itself a facilitation for acquisition. E.g. Learner: Does they’re sister? Teacher: Sorry? Learner: What’s the relationship between them? NNS: He pass his house. NS: Sorry? NNS: He passed, he passed, ah, his sign. REPETITION (OR CONFIRMATION CHECK) The teacher (or the NS) repeats the ill-formed part of the learner’s utterance, usually with a change in intonation. Implicit and output prompting feedback. 47 Grammatical feature and linguistic forms targeted: metalinguistic explanations were found to be more effective than recasts in improving learners’ use of some specific forms (e.g., comparative –er,). Salience of form plays a role. Although morphosyntactic errors seem to be the most targeted, learners end up noticing and/or repairing lexical and phonologically errors more successfully. Recast greater improvements have been observed in vocabulary knowledge and pronunciation. Instructional contexts also plays a role (studies based on classroom practice observation and quasi-experimental studies carried out in labs, immersion classes and more traditional ones, etc.). MISMATCHES BETWEEN TEACHER’S INTENTIONS AND LEARNER’S INTERPRETATIONS? Teachers should take research evidence into consideration and become aware of the importance of giving feedback. The question is how and when to provide CF, which implies taking into consideration:  Different cultural backgrounds (relationship with teacher's authority, with idea of knowledge, etc.)  Learners’ personalities (extraversion, etc.) and individual characteristics (learning styles, cognitive styles, attentional control, phonological memory, etc.)  Proficiency levels (literacy levels, accuracy, fluency, etc.)  Age (developmental readiness and prior knowledge, analytical capacities, etc.)  Tasks and linguistic forms Teachers may vary their choise of feedback strategy depending on their assessment of the student’s ability to attend to form being corrected. INTERACTION WITHIN THE SOCIOCULTURAL APPROACH (Lantolf 2006, Landolf & Thorne 2007, McCafferty 2008) This is an approach developed within the cognitive psychology, in which social interaction has a deeper and more substantial role in any learning process. By using L2 we do not only ‘practice’ language but, more importantly in Vygotskian view, we co-construct our knowledge about the language by interacting with other speakers (especially with more experienced ones). Language acquisition and learning does not happen only in the learner’s mind, but is ‘situated’ and co-constructed in the social interaction => situated cognition (in the context) and distributed cognition (co-constructed by different people). Language emerges from social interaction and means, makes sense only in the interaction, in the context. The individual and the environment form an inseparable dialectical unity that cannot be understood if the unity is broken. The mental activity cannot be separated from the external activity. Following the sociocultural approach, the internal, mental, psychological processes cannot be disentangled from the external context in which they take place and by which they are originated. Language has a social origin and is internalized progressively, to become a tool to mediate 50 thought: the internal representation of the linguistic system is shaped by interaction. Language development (as the development of cognition) results from the social and interpersonal activity and is therefore inherently social. Q: What is the role of interaction and feedback in sociocultural approach? In this perspective the negotiation is one of the interactional competences which are at the origin of acquisition and of the co-construction of L2. Feedbacks are something more than simple opportunities to get ‘positive’ or ‘negative’ evidence about the L2  the feedback helps the learner co- construct her/his discourse, to co-reach his communicative aim, to co-complete the task  Something that (s)he would not be able to do alone  notion of Zone of Proximal development. ZONE OF PROXIMAL DEVELOPMENT (ZPD) The space between what a learner can do without assistance and what a learner can do with adult guidance or in collaboration with more capable peers. a conceptual tool that educators can use to understand aspects of students’ emerging capacities that are in the early stages of maturation.  The emergence of self-regulation. SCAFFOLDING ZPD has been associated with the construct of “Scaffolding” which derives from cognitive psychology and L1 research. It refers to the dialogic process by which one expert speaker (the teacher or another learner) assists another speaker (a learner) to perform a task that (s)he is unable to carry out alone. Co-construction of discourse. A scaffolded performance is built through dialogue that promotes the learner’s internalization of knowledge co-constructed in shared activity.Scaffolding and collaborative dialogue during expert-learner and learner-learner interaction. Shift from ‘native’ vs ‘non native’ dichotomy to ‘novice’ vs. ‘expert opposition. SCAFFOLDING AND LEARNING Scaffolding is essential in supporting the learning process. Learning involves moving away from other-regulation (other-repair), toward more reliance on self-regulation (and self-repair). When a non-expert subject (the child, the learner) participates in a new activity and interacts with a more expert subject (the parent, the teacher): many instruction-giving events, modelling, feedback and also taking over for parts of the activity where the less- expert participant is estimated to need help  the activity is other-regulated. An activity is self-regulated when the participant is capable of carrying out an activity independently. The scaffolding practice in the language class results in: Improvement in the use of the relevant linguistic features, 51 The learner’s attempt to move away from the tutor’s help (‘other- regulation) to reliance on personal resources (‘self-reliance). The teacher has an active role in progressively adapting his scaffolded help in response to the learner’s needs and skills. The teacher’s feedback is adapted to the learner’s problems arising during the task and is thus shaped by the dynamics of interaction itself. COLLABORATIVE WORK (DONATO) Collaborative work involving learners provides opportunities for scaffolding help. In the classroom setting, learners are able to construct a scaffold for each other by managing aspects of the problem, identifying critical features of differences between what has been produced and the ideal situation, thus reducing frustration and risk. SCAFFOLDING AND GROUP ACTIVITIES Group work activities should not simply provide opportunities for exchanging linguistic information but should be seen as forms of collective and shared language acquisition. The role of situated interaction as a source of learning. Scaffolding and collaborative dialogue arising expert-learner interactions enable learners to produce and learn linguistic features that they would not be able to use independently. A collaborative interaction involves:  A meaningful interaction activity with a defined outcome, such as in information/opinion gap activities and group decision activities,  The contributions of each individual for the goal of the activity to be achieved,  The building of coherence in terms of social relations and knowledge among the members of the group. 52 This position rejects both the possibility of explicit knowledge transform directly into implicit knowledge and the possibility of implicit knowledge to become explicit. STRONG INTERFACE POSITION: SKILL THEORY The strong interface position claims that explicit knowledge can be converted into implicit knowledge through practice >>> Proceduralization of declarative knowledge Learners can first learn a rule as a declarative fact and then transform it into an implicit representation. Successful L2 users coming out of self-study language programs are thought to be the living proof that learning under explicit conditions can convert into acquisition. In order for this to happen, targeted practice is required: it is the role of practice to gradually bridge the gap between explicit knowledge and use. SKILL LEARNING THEORY Anderson et al. 2004 Theory originated in cognitive psychology. The acquisition of any skill (driving, playing a musical instrument, practicing sport, riding a bike, knitting, etc.) takes place in three stages: (a) declarative stage > (b) procedural stage > (c) automatic stage  A gradual move from the effortful, conscious, and overtly controlled declarative processes of the novice to the smooth, unconscious, and covertly controlled procedural processes of the expert.  Progression of the initial performance that is slow and prone to errors towards fluent, competent and error-free L2 uses. (a) declarative (conscious) stage  In the declarative stage the knowledge is acquired through the observation of others engaged in skilled behavior or ‘transmitted in verbal form from one who knows to one who does not know’.  First trials present a massive load to the working memory in terms of remembering the sequence of things to do and what to pay attention to. The resulting performance is an error-prone, arduous process that (generally and thankfully) does not usually last longer than a few trials. (b) proceduralization, procedural stage  During the procedural phase of the skill leaning (i.e., when we start driving in real traffic with an instructor), we have a practice period.  There is a shift from relying on declarative facts to procedural knowledge as the learner develops efficient procedures of performing the skill, often by skipping or compounding steps that were presented in the first phase. 55  During this phase explicit knowledge becomes at least partly implicit, which translates into the ‘smooth, rapid, procedural execution. (c) automatic stage  The execution of the skill becomes fully autonomous and automatized. Fine- tuning of proceduralized knowledge takes place eventually leading to fluent, errorless performance of the skill.  This automatic stage involves the continuous improvement in the performance of a skill that is already established. The level of automaticity and fluency increases while cognitive involvement decreases.  Learners often lose their ability to describe verbally how they do the task. Reaching this stage, however, requires a large amount of practice.  DeKeyser warns us that even highly automatized behaviors are not 100 per cent error- free. WEAK-INTERFACE POSITION There is no one single weak-interface position. There are different positions that fall somewhere in the continuum between no-interface and strong-interface. Rod Ellis’s weak interface model claims that there can be a communication between knowledge stores only in for some specific features (developmental features vs variational features). Nick Ellis also takes a weak-interface position, in which both implicit and explicit knowledge can work together cooperatively, with implicit knowledge being the most important for learning. MEMORY SYSTEMS: ANATOMICAL STRUCTURES The delay with which the relationship between language and memory has been analysed was due to the fact that no clear interaction seemed to emerge between the two systems: patients with memory problems (amnesia) do not manifest any problem with language (aphasia) and vice versa. When research findings indicated that ‘memory’ (as language) is not a monolithic unity, but is articulated into subsystems which interact, scholars could get to a new understanding of the complex relationships between language and memory. PATIENT H.M. The first convincing experimental evidence that memory is not made up of a single system localized in just one specific area in the brain was provided from Milner (1962). She demonstrated that a severely amnesic patient, H.M. (Henry Molaison), could learn a hand–eye coordination skill (mirror drawing) in the absence of any memory of having practiced the task. She claimed for a dissociation between: declarative memory (‘knowing what’) and non- declarative or procedural memory (‘knowing how’). 56 NON-DECLARATIVE MEMORY Probably the most important form of memory for living beings. It is present in many species  this memory system is considered as the oldest from an evolutionarily point of view. In humans, it is already active in newborn babies. Non-declarative, implicit memory is the first to appear in children and the last to disappear in the elderly people. Experimental studies have shown that children up to 8 -10 months can rely only on this type of memory, while the explicit, declarative memory is still absent or poorly developed. Independent of IQ or other standard measures of cognitive ability. Non-declarative, implicit memory develops independently of consciousness. The acquisition of implicit knowledge happens without paying attention or focusing on the task => without conscious awareness. Children can acquire a large amount of implicit knowledge before developing consciousness (it seems that awareness starts to develop very slowly after the age of 10 months). In children up to the age of 3, the acquisition of implicit knowledge prevails over explicit knowledge. Non-Declarative Memory Subsystems Procedural memory, for cognitive and motor procedures (such as the ability to walk, to articulate sounds, to play an instrument, to drive a car, ...) => skill learning Memory for the acquisition of conditioned reflexes (classical conditioning) Systems underlying the priming (phenomenon that affects the visual and auditory perception) MEMORY CONSOLIDATION  Memory consolidation is a process that transforms novel memories from a relatively fragile state to a more robust and stable condition. For a long time it was believed that the consolidation of procedural memories took place solely as a function of time (and repetition), but more recent studies suggest that for certain forms of learning, the consolidation process is enhanced during sleep. Sleep has been consistently shown to aid in the development of procedural knowledge in humans, especially when sleep follows the initial phase of memory acquisition. The acquisition of non-declarative, implicit knowledge is strengthened and consolidated during sleep. Specifically, consolidation 57 According to Ullman, L2 grammar tends to depend more on declarative memory and less on procedural memory than L1. There is an age effect because learning in procedural memory seems to be established early and then declines, while declarative memory shows the opposite pattern. L1 learners (and early L2 learners) procedural memory for learning grammar. (especially after a reasonable amount of language exposure) L2 learners  declarative memory. When we’re adults we rely on declarative memory because it’s more powerful. Memory systems work differently for different ages. CO-OPERATION OF EXPLICIT AND IMPLICIT LEARNING We cannot develop sufficient implicit knowledge on a L2 without the effective functioning of our explicit learning mechanisms. Explicit practice creates implicit learning opportunities! Explicit info is supposed to fine-tuning  ??? Factors affecting the explicit-implicit co-operation: it varies according to the characteristics of the learners, the learning situations, the L2 area focused on. The positive impact of the explicit learning component of:  learner’s characteristics (age: children learn by receiving, adults need to have an explicit rule in order to learn, metalinguistic sophistication, prop education experience, motivation, cognitive style, language aptitude, especially working memory capacity)  L2-L1 similarities and differences,  Characteristic of the target L2 structure/area (complexity, regularity, form-or-meaning based nature)  Characteristics of the available/accessible natural L2 input (overall amount of frequency and salience)  The length of time available for the learning process. CROSSLINGUISTIC INFLUENCE THE ISSUE If we talk about L2, it is because we already have a L1: we have experienced a previous acquisition process, we have previous language knowledge and we have used the L1 for a considerable (although variable) amount of time. Previous language knowledge is an important source of influence on L2 acquisition  universally true for all L2 learners. 60 If knowledge and capabilities for competent language use are already available to L2 learners through the mother tongue (and/or other languages they may know), how do they affect the development of the new language? KEY TERMS Interference  This term has been replaced in contemporary SLA, to avoid the unwanted implication that knowledge of L1 hinders L2 development Transfer  L1 influences can have both a positive as well as negative consequences for L2 learning. Crosslinguistic influence  Knowledge of the L1 impacts on the L2 learning subtly and selectively, might result in strikingly different consequences (both negative and positive) for different L1 backgrounds: at different stages of development or proficiency - for different areas of L2 CONTRASTIVE ANALYSIS HYPOTHESIS (CAH) 1950-60: before SLA domain was established. The scholars most associated with the contrastive analysis hypothesis are Robert Lado, Charles Fries, Robert Stockwell. The contrastive analysis hypothesis (CAH) assumes that the mistakes made by learners are caused by the differences between L1 and L2, and by the transfer of L1 habits/behaviours to L2. Consequently, these errors are predictable and therefore avoidable. The different languages can then be analysed in a contrastive way, in order to highlight: symmetries, which are assumed to originate positive transfer asymmetries, differences that are a source of negative interference, and therefore of language learning problems and errors. More specifically, comparing L1and L2 and determining structural similarities or differences was aimed to: predict which structures will be easy or difficult to learn, predict when negative transfer will occur and what errors will be produced by particular L1 background learners, identify what needs to be taught in L2 learning, in order to prevent the errors to occur  this will lead to the production of L1-based pedagogical materials. CAH is able to predict difficulties and errors  strong view, also known as predictive or a priori view  not particularly successful. CAH is able to understand and explain errors  weak view, also known as explanatory or a posteriori view  will lead to Error Analyses. CAH was compatible with a behaviourist view of language acquisition  Because language development was viewed as the formation of habits, it was assumed that a person learning a second language would start off with the habits formed in the first language  L2 learning would therefore consist in replacing the responses consolidated in the L1 with those of L2. L1 habits would interfere with the ones needed for the L2  According to the CAH, errors were basically assumed to be the result of transfer from learners’ first language. 61 Until the late 1960s, second language learners’ speech was seen simply as the incorrect version of the target language. Errors were a sign of failure that should be corrected immediately, to prevent the formation of ‘bad habits’. During the 1970s, however, researchers found that many of the errors learners make are not predictable on the basis of their L1  many L2 ‘errors’ were not the result of transfer. These errors either looked like those that children learning the L1 would make (although we don’t really call what children do ‘errors’), e.g., overregularization, U-shaped behaviours  this will lead to the Error Analysis Approach ERROR ANALYSIS (EA) Error Analysis Approach, which developed during the 1970s, involved detailed description of the errors L2 learners made. For Pit Corder, errors are informative clues on the way learners process the L2  when learners produce correct sentences, they may simply be repeating something they have already heard BUT, when they produce sentences that differ from the target language, we may assume that these sentences reflect the learners’ current competence (transitional competence, understanding of the rules and patterns of that language). Much of the early work in error analysis focused on determining whether SLA was the result of L1 transfer or rather a creative construction (i.e., something similar to the processes L1 learners used to acquire language). What early error analysis showed was that not all errors could be attributed to L1 influence and that L2 learners were active creators of linguistic systems  these results were used to argue against the CAH and will lead to the notion of Interlanguage In addition to Corder, other names associated with error analysis during the 1970s are Martina Burt, Heidi Dulay, Elaine Tarone, Jack Richardsand Jackelyn Schachter. Influenced by L1 research – which had repudiated any kind of strict behaviourist account of child language acquisition – Corder suggested that like children, perhaps L2 learners came equipped with something internal, something that guided and constrained their acquisition of the formal properties of language. He called this something ‘the internal syllabus’, noting that it did not necessarily match the syllabus that instruction attempted to impose upon learners. EA paves the way towards the notion of Interlanguage. CRITICISM OF CAH: UNIVERSAL PATHWAYS Universal interlanguage patterns are followed in L2 development, regardless the L1 Acquisition of Negation in English L2 Similar steps are observed in the IL of learners with different L1s. At early stages pre-verbal negation is used (‘I no/not want that’). Attested in languages with preverbal negation; e.g. Italian: Io non voglio quello. 62 CRITICISM OF CAH: SIMILARITY EFFECTS AND INTERLINGUAL IDENTIFICATION Transfer depends on how ‘similarity’ or closeness is perceived by L2 learners Duskova presents data from Czech speakers learning 1) English and 2) Russian. Transfer of bound morphemes in L2 Russian No transfer of bound morphemes in L2 English Hulstiin and Marchena on phrasal verbs in L2 English (Dutch L1) English phrasal verbs were not accepted by Dutch learners when there was close similarity between Dutch and English => ‘disbelief’ that another language can have a structure so similar to the ‘unusual’ one (see later) In order to understand L1 transfer, one needs to go beyond L1-L2. Learners have to make an interlingual identification for transfer to occur. a better understanding of the forces that shape crosslinguistic influence rests not with external comparisons, but with learners’ psychological perceptions of those L1-L2 similarities or differences. For transfer to occur learners have to make an interlingual identification  ‘the judgement that something in the target language is similar’. The judgement might be, but need not be, made as a conscious, strategic choice CRITICISM OF CAH: TRANSFER AS A COMMUNICATIVE STRATEGY A study by Singleton in 1987 on Philip, L1 English professional writer, who had picked up French during three short visits to France. In three half-hour interviews in French, 154 transfer errors were identified. About third of them were accompanied by a hesitation, an apology, interrogative intonation, laughter or some other overt indication that Philip might have been conscious of the less than perfect transfer solution. When reviewing the interviews and commenting on specific French choice retrospectively, Philips said: I knew that it probably wasn’t right, but it was the nearest I could get to something that might be right  transfer as a (more or less conscious) communicative strategy What is transferred is determined by specific constraints, not just by the two languages: Transferability Kellerman (1979) Aspects which are perceived as ‘universal’ are more likely to be transferred than ‘language specific’ aspects. Language-neutral items: considered common across all languages E.g. simple syntactic structures, such as The sky is blue Language-specific items: idioms, collocations, phonology, morphosyntax (e.g. inflectional morphology, some specific syntactic constructions, etc.) E.g. It’s raining cats and dogs  idiom (more language-specific) E.g. That book made a difference  collocation (less language-specific) 65 Transferability interacts with proficiency: Both in English and Dutch, The transitive (he broke a vase) and intransitive (the vase broke) structures are allowed in both languages. Transitive constructions are perceived as language-neutral. Intransitive constructions are perceived as language-specific. Learners who have L1 Dutch and develop higher competence, tend to accept more easily the English transitive than intransitive structures.  intransitive structures are perceived as too Dutch-specific to be transferred. Perception of transferability is crucial for transfer to occur  If a feature sounds too L1-specific, it does not get transferred  The judgement of whether something in the L1 is similar enough to be transferred to the L2 is influenced by other factors beside external similarities between L1 and L2.  Transferability interact with proficiency in shaping what may or may not get transferred Markedness The Markedness Differential Hypothesis (Eckman 2004) A structure X is typologically marked relative to another structure Y (and Y is typologically unmarked relative to X), if every language that has X also has Y, but every language that has Y does not necessarily have X. Successful in explaining directionality effects in L1-L2 transfer Ex. Voiced / Voiceless stops (/b/-/p/) Each language that has a voiced stop, also has a voiceless Voiceless (unmarked) → Voiced (marked)  Voiceless stops are acquired earlier Voiceless (unmarked) → Voiced (marked) Voiced stops are more marked on the basis of the fact that:  All languages have voiceless stops, only some have voiceless and voiced ones  No language exists that has only stops without having voiceless ones  Children whose L1 has both voiceless and voiced stops, will acquire voiceless first  The phenomenon of devoicing: voiced stops are pronounced as voiceless in certain positions, so that unmarked elements are used instead of marked ones Markedness Differential Hypothesis English/German both have voiceless and voiced stops, but consonants in word-final position behave differently: in English, cap vs cab both are allowed. But in German tag /k/ voiced are not allowed  voiced are marked. German L1 – English L2  Difficulties: “tab” > /p/ - German native speaker will have difficulties to grasp the voiced phonetic at the final position English L1 – German L2  No difficulties. 66 According to Eckman’s Markedness Differential Hypothesis (Eckman 2004) 1. Marked forms are more difficult to learn  More interlanguage phenomena 2. A form that is marked in the L2, but not present in the L1, is more difficult to learn 3. A form that is not marked in the L2 will not lead to difficulties Target words IL words plant pilanti fred fired translate tiransilete Vowel epenthesis to transform a marked CCV syllable into a less marked CV syllable. (CCV is more marked than CV)  like in Turkish, it doesn’t have consonant cluster, and learners who has L1 Turkish and learn L2 English tend to insert a vowel (vowel epenthesis) between two consonants to obtain a less marked structure. However, Learners whose L1 has the more marked structure, won’t experience any difficulty in producing the less marked one. When transfer is not easy to detect… Crosslinguistic influence may lead to: Errors of commission (overuse/underuse) Errors of omission Transfer effects may go unnoticed: avoidance  example: production of relative clauses in English L2 Chinese/ Japanese L1: huge differences between L1 and L2  few errors but also few occurrences, avoidance of the difficult structure. Persian/Arabic L1: more similarities between L1 and L2  many errors (but many relative clauses) AVOIDANCE Q: Do learners avoid a structure because it is different from the one present in the L1 or for its inherent complexity? Dagut & Laufer 1985. Hebrew-speaking learners of English acquiring phrasal verbs (come in, take away, lay aside, shut off, let down, mix up, ect.). Hebrew does not have phrasal verbs They observe that these learners prefer the one-word equivalent of the phrasal verbs (enter, remove, save, stop, disappoint, confuse, ect.) and, more importantly, that within the category of phrasal verbs they prefer the more semantically transparent ones (come in, take away). The authors conclude that the complexity of the target language structure had a greater impact on the issue of avoidance that did differences between L1 and L2. 67