Chapter Two: SLA and Related Disciplines PDF

Chapter Two SLA and related disciplines  Second language acquisition (SLA) is an autonomous field with roots in language teaching and has been influenced by linguistics and psychology. It has a special relationship with child language acquisition, as it formed the basis of research in SL...

Chapter Two SLA and related disciplines  Second language acquisition (SLA) is an autonomous field with roots in language teaching and has been influenced by linguistics and psychology. It has a special relationship with child language acquisition, as it formed the basis of research in SLA. Other areas, such as third language acquisition and heritage language acquisition, have a derivative relationship with SLA, while bilingual acquisition blends issues related to second language acquisition and first language acquisition. These areas have well-developed histories and journals devoted to their issues. The relationship between these areas and SLA is different, with some developing out of related but specific concerns, while bilingual research has parallel development with related disciplines. Third language acquisition/multilingualism Second language acquisition, also known as third language acquisition, is a complex process involving multiple languages and individual histories. Factors influencing the acquisition of a third language include age, context, individual characteristics, and language distances. Sociolinguistic, psycholinguistic, and cross-linguistic influences also play a role in this process. Sociolinguistic issues involve learning the language for its purpose, while psycholinguistic perspectives consider lexicon organization. Cross-linguistic influences include proficiency in the target language, user age, and linguistic closeness. Heritage language acquisition Heritage language speakers are individuals raised in a non- English language-speaking home, typically bilingual in both languages. Their acquisition is a form of second language acquisition and bilingualism, with them having knowledge of both their home language and the language of the environment/school. They form a heterogeneous group with a subtly different knowledge base and differ from monolingual speakers of their heritage language. Recent studies have investigated linguistic differences between heritage and non-heritage language learners. Bilingual acquisition Bilingualism is the study of an individual's knowledge and use of two or more languages, often incorporating multilingualism. However, the term 'bilingualism' is problematic as it means 'two'. Bilingualism can refer to both the process of learning and the end result, the product of learning. It is difficult to categorize all types of bilingualism due to various situations, such as growing up with two languages and having the second language as their only language. The terminology used in bilingualism overlaps with second language acquisition research. Definitions of bilingualism (Table 2) Bilingualism can be categorized into several types, including achieved, additive, ambilingual, ascendant, ascribed, asymmetrical, balanced, compound, consecutive, coordinate, covert, diagonal, dominant, dormant, early, equilingual, functional, horizontal, incipient, late, maximal, minimal, natural, passive, primary, productive, receptive, and continuous.  Some individuals may be able to be achieved bilingual due to increased use of their first language, while others may struggle with the second language due to lack of use. Secondary bilinguals may have their second language added to their first language through instruction, while semibilinguals may have insufficient knowledge of either language. Simultaneous bilinguals may have their two languages present from the onset of speech. Subordinate bilinguals may exhibit interference in their language usage, while subtractive bilinguals may acquire their second language at the expense of their first language's aptitudes. Successive bilinguals may have their second language added at a later stage after the first language has begun to develop.  Bilingualism offers numerous benefits, including communicative, cultural/economic, and cognitive advantages. Communicative advantages include better communication with family members, sensitivity to communicative needs, understanding cultural differences, and economic advantages in various fields. Cognitive advantages include divergent thinking, creative thinking, and metalinguistic awareness. Bilingual children have superior abilities in judging grammatical accuracy and counting words, which are essential for selective attention when dealing with competing information.  Early language development, such as babbling, occurs towards the end of the first year of life, and children exposed to two languages from birth show language-specific patterns. Matching the appropriate language to speakers and context is found in children as young as 2. Code-switching, a common phenomenon among bilingual speakers, refers to using more than one language in conversation, often due to lack of concepts in one language, humor, or social context. Language mode refers to the state of activation of the bilingual's languages and language processing mechanisms at a given point in time.  The native language, or base language, is always fully activated, controlling linguistic activities. The guest language can be low to high activation depending on the context. In bilingual language mode, almost equal activation occurs, causing code- switching. Multicompetence, the "normal" propensity for humans to know multiple languages, is a concept proposed by Cook. This multicompetence affects how individuals process their native language, even those with minimal second language knowledge. Cook argues that monolingual orientation of second language acquisition belies the reality of language learning in many contexts. First language acquisition This chapter discusses child language acquisition (SLA) as a crucial field in the development of second language acquisition (SLA). Learning a first language is a complex task, with children having to determine the language they hear and express their needs. The complexity of language learning can be seen in the concept of plurality and the language needed to express it. Children communicate long before they have language, using various means such as crying, smiling, and cooing to communicate their needs. Cooing sounds suggest that infants are aware of sounds and their potential significance, as they use these sounds to play with language-related phenomena like loudness and pitch from approximately four to seven months. Babbling Babbling, a language-like sound that begins around six months of age, is often mistaken for words by parents or caregivers. Early babbling sounds, such as bababa, dadada, and bada, can be interpreted as words. Innation is another early tool children use to express meaning, even before they have grammatical knowledge. For example, a child can say dada with stress on the second syllable, indicating a command. The relationship between babbling and word usage is gradual, with a decrease in babbling and an increase in words during the five-month period. However, there is a point where each child "gets" the concept of words as referring to something, and after this, there is a drop-off in the amount of babbling. Words Words in early child language serve various functions, including referring to objects, indicating grammatical functions, and serving social functions. However, words in an adult's language may not always correspond to words in a child's language. For example, "allgone" is typically produced at the one-word stage in child language, even though it comprises two words in the adult language. Children often overextend the meanings of words they know, such as using "bunny" to refer to various objects or "bear" to refer to a stuffed toy. Additionally, children often underuse words, using words with more restricted meanings than in adult usage. Sounds and pronunciation In early stages, children's pronunciation of words differs from adult speech. They must learn to distinguish sounds early, such as consonants in words like [ta] and [da]. Even when using words similar to adult words, there are pronunciation differences, such as substitutions, deletions of syllables, deletions of sounds, and simplifyation. The explanation for these differences depends on whether they are motor control or perception. For example, a child might not say puddle as puggle, but instead use puddle for puzzle. Children can perceive differences in pronunciation, but not in their own speech. Syntax The transition from babbling to words begins with the one- word stage, where children use content words like nouns and verbs. As children move beyond this stage, speech becomes telegraphic, using only the bare minimum utterances. The mean length of utterance (MLU) is the standard measure used to determine complexity. Further syntactic development includes stages like question formation, which is predictable for all children.  Adults learning a second language also have predictable sequences in question formation, including inversion, which maintains declarative order but not in wh-questions. Second language acquisition is more complex due to factors involving the native language, making the situation more complex. The acquisition of certain structures is predictable for all children, but the situation with second language learners is more complex due to factors involving the native language. Morphology The study of second language acquisition was influenced by Brown's observation of a predictable order of acquisition of certain inflectional morphemes in English. Brown's research showed that children, Adam, Sarah, and Eve, learned English morphemes in roughly the same order, despite not always occurring at the same age. This order may be related to their overall development, measured in Multilingual Units (MLUs). The order does not reflect the frequency of these morphemes in the speech of the children's parents. Another well-known study is Berko's "wug" test, which showed that even preschool children could correctly form plurals, demonstrating understanding of plurality and grammatical form. Overgeneralization of irregular forms can occur, and children may say something like mices without recognizing that the word is already plural.  There are often prerequisites for learning certain forms and interrelationships among forms. For example, children go through the same developmental stages, create systematicity in their language, and develop rules to govern their language knowledge and use. However, these rules do not necessarily correspond to the rules of the adult language. There is overgeneralization of grammatical morphemes, and processing constraints govern acquisition and use. Second language acquisition is not determined by intelligence, and correction does not always work.  Chapter Three SECOND AND FOREIGN LANGUAGE DATA  Data analysis This chapter explores data analysis and interpretation in Second Language Acquisition (SLA) using data from second language learners. It discusses different types of data and elicitation methods. Despite ambiguity in interpretation, there are no "correct" answers, and good argumentation helps lessen ambiguity. The chapter presents several data sets and provides a step-by-step map for readers to understand and challenge the logic and argumentation. Data set I: plurals The study analyzed the use of -ing marking on English verbs in Cairene Arabic, a language spoken by three adult native speakers of English (three adult native speakers of Cairene Arabic, intermediate to advanced speakers of English). The data was collected from compositions and conversations, and the utterances included phrases like "There are also two deserts," "I bought a couple of towel," and "There is many kind of way you make baklawa." The analysis revealed that some sentences were English-like, with an s plural marker on the noun, while others were non-English-like. The study also found interlanguage variation within the same sentence, with months being non-English-like and years being English-like. The IL hypothesis was proposed, marking all plural nouns with /s/ except those preceded by a quantifying phrase or nonnumerical quantifying word. Further data is needed to understand individual utterances and alternative explanations. Data set II: verb + -ing markers -A native Arabic speaker in early English language learning produced sentences with two forms, using Verb + -ing structures. -The acquisition of grammatical forms is variable, with Verb + -ing occurring in intransitive sentences and the simple form in transitive sentences. -This suggests processing limitations or the storage of units as single lexical items, which may not be problematic for this learner. Data set III: prepositions The study examines the use of prepositions in Arabic-English sentences, revealing that learners' behavior is influenced by geographical and temporal semantic areas. The data includes examples of sentences with prepositions from Morocco to Saudi Arabia, and sentences with prepositions in English. The learners' behavior is influenced by geographical versus temporal semantic areas, with some using from for geographical locations. The third set of data presents correct TL forms, although some may appear target-like. The place expressions in Egypt, in Dallas, and in this area negate the simple hypothesis of using from for geographical locations. However, learners may make a distinction between direction and location or produce more TL-like preposition usage when other than "obligatory" prepositions are required in the second language.  The study suggests that learners adopt a strategy of using no preposition except in specific instances, as seen in the first set of sentences Some use from for geographical locations, while others use no preposition except in specific instances. The study suggests that learners use correct TL forms, but may differentiate between direction and location when other prepositions are needed. What data analysis does not reveal The text focuses on data analysis and interpretation in learner-language studies, aiming to reduce ambiguity. It discusses data sources, such as compositions and conversations, and their importance for a thorough analysis. However, it also addresses challenges like pooling data across subjects and identifying individual learners. Further data collection may be needed to address specific questions, such as plural expression in Arabic-English. The authors also discuss the contexts in which interlanguage sentences were produced and the formulation of descriptive interlanguage rules.  Data collection In recent years, there has been a growing focus on data collection and analysis in second language studies. This is due to the difficulty in aligning various elicitation tasks and data analysis methods with the desired constructs. There are numerous ways to elicit second language data, with many having origins in other disciplines. The choice of data collection method depends on the research question being asked. There are several ways to categorize second language data, including context (e.g., classroom data versus naturalistic data), performance types (real speech samples and reactions to target-language data), and "thinking" data (learners' statements about their learning)  Longitudinal studies are case studies that collect data from a single participant over a prolonged period of time. They typically provide detailed information on a learner's speech, setting, and other relevant details. For example, Patty, born in Indonesia, was studied in a longitudinal study. She was taught Chinese as a primary language at home, in the local community, and with friends. After emigrating to Shanghai, China, she emigrated to Hong Kong, where she completed secondary school, acquired Cantonese, and received some English classroom instruction. At 22, she moved to the United States, where she attended junior college and received ESL instruction. At the time of the first recording, Patty was 32 and had lived in the United States for about 10 years. Her language environment was mixed, with Cantonese in her urban Chinatown neighborhood, English at work and school, and both Cantonese and English as lingua franca at home.  Patty, a Chinese-American woman, spent her first six years in an English- speaking environment, speaking English at home, work, and with her family. Her primary Chinese-speaking contact was with a cousin who was also an American, has an English-speaking child, and lives and works in an English-speaking environment. Patty's primary Chinese-speaking contact was with a cousin living in another part of the state, who was also an American.  Longitudinal studies often collect data from spontaneous speech, which is not a suitable experimental paradigm. Researchers can ask specific questions during data collection to generate specific structures, such as determining the frequency of occurrence of certain forms. Analyses of data obtained through longitudinal studies often involve descriptive qualitative comments or narrative expositions, which can be useful in determining developmental trends and interpreting social constraints and input influences on the learner's speech. However, longitudinal studies have drawbacks, such as time involved in collecting data and transcription, and a lack of generalizability. Additionally, spontaneously produced data may not allow for further probing of the learner's knowledge beyond what they produce spontaneously.  Cross-sectional studies, on the other hand, have four identifiable characteristics: number of subjects and time frame of data collection, type of data, descriptive detail, and analysis of data. A cross-sectional study is a research method that gathers data from a large number of subjects at a single point in time, allowing for a slice of development to be pieced together. This format is often based on controlled output, with participants not identified individually or detailed descriptive information provided. The data is presented in tabular form and has an experimental format, with results being more quantitative and less descriptive than longitudinal studies. A pseudolongitudinal study can be created using a cross-sectional design, focusing on language change and proficiency levels. This method can be used to investigate the acquisition of progressive language skills or compare proficiency levels among a large group of subjects.  A cross-sectional approach has advantages, such as generalizability and the ability to provide detailed information about subjects and linguistic environments. However, it also has disadvantages, such as lack of detailed information about subjects and the linguistic environment. Longitudinal data are often associated with descriptive or qualitative data, while cross-sectional and pseudolongitudinal data are often associated with quantitative or statistical measures. Researchers may choose one type of data-collection procedure over another based on the relationship between a research question and methodology  Two studies have examined the range of data in second language acquisition, one qualitative and one quantitative. The first study by Kumpf (1984) found that nonnative speakers express temporality differently in English, with certain verbs marked with tense. The second study by Gass and Ard (1984) examined learners' knowledge of progressive and other verb forms, using data from 139 participants' responses to four tasks. The authors found that controlled observations of spontaneous speech may underestimate a learner's linguistic knowledge, especially when tasks are insensitive or demanding. Eliciting speech samples Grammaticality judgments are metalinguistic performances that help infer grammatical properties of a language. Native speakers' responses help determine which sentences are part of a language, but judgment data can reveal more about a language than production data alone. However, judgment data can also reveal what is not possible in the given language, which raises questions about the validity of judgment data as measures of a learner's grammar at a given point in time.  The need to understand individual learners' knowledge about a second language is not controversial. Two methods to determine this knowledge are truth-value judgments and sentence matching. Truth-value judgments are used to investigate learners' knowledge of reflexives, which refer to the same person in different languages. This method has been the subject of numerous articles. Verbal report data Verbal reports are introspective data that involve individuals describing their thoughts during a task. There are two types: recalls and think-alouds. Recalls involve recollecting the original event, while think-alouds report on the event as it occurs. For example, Spanish students report on grammar problems, revealing ambiguity in word form and verb stems. Verbal reports offer valuable insights into individuals' thoughts and behaviors. Measuring non-linguistic information Questionnaires Questionnaires are used to gather information about a learner's attitudes towards language learning, either generally or a specific language. There are various types of questions, such as asking about the benefits of studying French, understanding French people, meeting diverse people, and becoming a better educated person. Attitudinal ratings are also used in research on motivation and attitudes, with respondents given polar opposites and asked to judge their impressions of a group on an evaluation scale. Interviews are conducted orally and one-on-one, with fixed or less structured questions. Researchers often use a standardized test of language proficiency to measure the attribute under investigation. Diaries, which are journals a learner keeps, are a good source of information about attitudes and motivation, providing an open-ended forum without specific questions.  Measuring general proficiency: standardized language tests Standardized language tests are not commonly used for second language data due to their objective nature and lack of easily analyzed data. Instead, they are used to measure proficiency levels, such as advanced learners with a TOEFL score. However, there is no absolute cutoff point for advanced, intermediate, or beginner proficiency. There are four common ways to assess proficiency: impressionistic judgments, institutional status, specific research-designed tests, and standardized tests. This makes it difficult to compare studies in the field of second language acquisition. Replication The scientific method involves observations that can be repeated and verified by others. Second language acquisition (SLA) research is often empirical and inconsistent due to the lack of participants and the nature of second language knowledge. Many studies have fewer participants, making it difficult to draw generalizable conclusions about acquisition. Learners' indeterminate knowledge can cause inconsistent linguistic behavior. Replication studies are essential for new researchers to gain hands-on experience with actual data, but exact replication is impossible due to the inclusion of different individuals. Issues in data analysis Second language acquisition (SLA) analysis faces challenges due to the lack of an index of development and the complexity of determining oral proficiency. Standardized tests and categorization based on placement scores for specific language programs are common ways to determine a learner's proficiency. Determining oral proficiency is more complex due to different measures depending on whether one is considering monologue or dialogue data. Data analysis in second language research is challenging due to the lack of unique results. Researchers must decide on the relevant data for analysis and choose a point of reference for comparison. Early research focused on comparing learner output with the National Language (NL) and learner output with the Standard Language (TL), but this approach may miss generalizations that learners have constructed for themselves.  The analysis of article use in English can be categorized into four types: Category 1, Category 2, Category 3, and Category 4. In Ge's native language, "topic-comment" is an important concept for a sentence, and the study by T. Pica (1983) demonstrated this discrepancy in determining the acquisition of morphemes. Determining the targeted structure of a learner's produced language and the role of the native language is also difficult. Dialects and language variety can affect the phonology of a second language, and assessing the learner's target language variety and knowledge base is complex. What is acquisition? The concept of second language acquisition is complex and requires careful consideration of various factors. For instance, learners may "backslide" when correct forms appear but disappear, highlighting the need for precise definitions. Acquisition can range from linguistic knowledge to the ability to use that knowledge in speech and real-time language processing. Various definitions of acquisition include the first appearance of a correct form, a certain percentage of accurate forms, the first of three consecutive two-week samples where the morpheme is supplied in over 90% of obligatory contexts, and the presence of at least two examples of structures in two different posttests. However, focusing solely on language forms is limiting, as it does not consider the context in which the forms occur. Researchers use various criteria to determine when acquisition has taken place, but it is crucial to consider the stages a learner goes through in acquiring a particular form. Conclusion This chapter discusses data analysis methods and methodologies, allowing readers to analyze SLA research. It also discusses the role of native language in second language acquisition, examining its historical context in chapter 4.  Chapter Four THE ROLE OF THE NATIVE LANGUAGE: AN HISTORICAL OVERVIEW  Introduction Language transfer, a subfield of second language acquisition research, has a complex history influenced by varying theoretical perspectives on second language acquisition. I t is assumed that learners heavily rely on their native language and culture in second language learning. Lado's work focused on producing pedagogically relevant materials by conducting contrastive analysis of the native and target languages. This involved comparing the two languages to determine similarities and differences. The acceptance of language transfer as the mainstream view of language learning requires understanding the psychological and linguistic thinking at the time. The term "transfer" has been used in the literature to describe two different learning processes: positive and negative transfer. However, the actual determination of whether a learner has transferred is based on the output compared to the target language output. This terminology suggests a confusion between product and process. Behaviorism Early research on language learning, particularly first language learning, was heavily influenced by linguistic and psychological paradigms. Bloomfield's behaviorist view suggests that language is speech rather than writing, and that children without cognitive impairment learn to speak before they learn to write. This theory explains the interrelationship between speech and action through the concept of babbling, which involves mimicking and analogizing. The child learns to make the stimulus-response connection, either by uttering the word doll when the object is seen or by getting the doll when the word is heard. Language-learning terminology and concepts, such as interference and facilitation, are derived from behaviorism.  The notion of transfer is central to behaviorist theory, which asserts that the learning of task A will affect the subsequent learning of task B. Positive transfer refers to correct or incorrect outcomes, while negative transfer refers to incorrect outcomes. Contrastive analysis emerged from this framework, focusing on the North American tradition, which focuses more directly on second language acquisition. Contrastive analysis is a method of comparing languages to identify potential errors and predict areas of difficulty in learning a second language. It involves a structure-by-structure comparison of sound, morphological, syntactic, and cultural systems to identify similarities and differences.  Contrastive analysis Contrastive analysis is a method of comparing languages to identify potential errors and predict areas of difficulty or ease in learning a second language. It involves a structure-by-structure comparison of sound, morphological, syntactic, and cultural systems to identify similarities and differences. The primary source of error in learning a second language is the native language, and the more differences between the two languages, the more errors will occur. The Contrastive Analysis Hypothesis (CAH) framework has two positions: the a priori versus the a posteriori view, the strong versus weak view, and the predictive versus explanatory view. The strong view predicts learning success based on comparisons between two languages, while the weak view analyzes learners' recurring errors based on NL-TL differences. The weak version gained credibility due to the failure of predictive contrastive analysis, which focused on learners' self- reported forms and strategies.  Contrastive analysis, a theory that assumed native language was the driving force of second language learning, faced criticism for its theoretical underpinnings. In the 1960s, the behaviorist theory of language and language learning was challenged, leading to a shift towards structured rules instead of habits. This change had significant implications for second language acquisition, as it suggested that children were not imitators and were not significantly influenced by reinforcement as they learned language. Researchers found that second language learner data reflected errors beyond those in the surrounding speech and, importantly, those in the native language. For example, Czech speakers learning English did not transfer bound morphemes, while Czech learners of Russian did. Another example was found in data from French speakers learning English and English speakers learning French, where object pronouns precede the verb, which is not possible in French.  Difficulty was also criticized for its association with errors, which were seen as signals of difficulty. However, difficulty cannot be unilaterally equated with errors, as differences are based on formal descriptions of linguistic units selected by linguists, teachers, or textbook writers. To equate difference with difficulty attributes a psycholinguistic explanation to a linguistic description, confusing the product (a linguist's description) with the process (a learner's struggle with the second language). The Contrastive Analysis Hypothesis (CAH) by Lado (1957) suggests that second language acquisition is not a simple 1:1 correspondence between the native and target languages. Factors such as innate principles of language, attitude, motivation, aptitude, age, and other languages are more complex than the early version of the CAH. Language learning cannot be seen as just a matter of "linguistic hiccups" from native to target language.  The complexity of comparing languages became apparent early in works like Stockwell, Bowen, and Martin (1965a, 1965b), who established a hierarchy of difficulty and learning. This hierarchy includes different ways in which languages can differ, such as differentiation, new categories, absent categories, coalescing, correspondence, and plurality. The significance of the Contrastive Analysis Hypothesis lies in its importance of empirical validation and the limitations that Lado himself attributed to his work. The list of problems resulting from the comparison of the foreign language with the native language must be considered a list of hypothetical problems until final validation is achieved by checking it against the actual speech of students. This final check will show in some instances that a problem was not adequately analyzed and may be more of a problem than predicted.  Historically, Lado's hypothesis inspired a generation of second language researchers to conduct linguistic field work, checking hypothetical contrastive analysis statements against the actual speech of language learners. This acceptance of the centrality of individual variation in second language acquisition is reflected in the development of error analysis.  Contrastive analysis: summary The Contrastive Analysis Hypothesis (CAH) framework has two positions: the a priori versus the a posteriori view, the strong versus weak view, and the predictive versus explanatory view. The strong view predicts learning success based on comparisons between two languages, while the weak view analyzes learners' recurring errors based on NL-TL differences. The CAH has faced criticism for its theoretical underpinnings, as it assumed native language was the driving force of second language learning. However, it has been criticized for its association with errors, as differences are based on formal descriptions of linguistic units. The significance of the CAH lies in its importance of empirical validation and its acceptance of individual variation in second language acquisition. Error analysis Error analysis is a type of linguistic analysis that focuses on the errors learners make in producing the second language (TL) and the TL form itself. It differs from contrastive analysis, which compares errors with the native language, while error analysis compares them with the TL. Corder's article "The significance of learners' errors" (1967) highlighted the importance of errors as red flags and evidence of a learner's attempt to figure out an underlying rule- governed system. Errors are not solely a product of imperfect learning but rather indicate a learner's attempt to figure out some system, imposing regularity on the language they are exposed to. This focus on errors is the beginning of the field of second language acquisition, which has theoretical implications for fields such as psychology and linguistics.  Errors are not just one-time-only events, but systematic ones that occur repeatedly and are not recognized by the learner. They are only errors with reference to an external norm (in this case, the TL). Error analysis provides a broader range of possible explanations than contrastive analysis, as it identifies two main error types: interlingual and intralingual. Interlingual errors involve cross-linguistic comparisons, while intralingual errors are due to the language being learned, independent of the NL. Error analysis has been criticized for its reliance on errors to the exclusion of nonerrors, which is essential for understanding a learner's linguistic behavior. A 1974 article by J. Schachter collected 50 positions from four groups of English learners: native speakers of Persian, Arabic, Chinese, and Japanese. The research focused on the use of English restrictive relative clauses (RC) by each group. The findings showed that Japanese and Chinese speakers had control over the formation of English restrictive relative clauses, while Persian and Arabic speakers did not. However, Schachter's analysis went beyond errors to look at the total production of relative clauses, including error-free relative clauses.  The Chinese and Japanese produced roughly half as many relative clauses as the Persian and Arabic groups. This discrepancy is significant because Japanese and Chinese form relative clauses by placing the modifier before the noun it modifies, while Persian and Arabic relative clauses are similar to English in that the relative clause is placed after the noun it modifies. Error analysis also faces difficulties in determining what an error is an error of. For example, Chinese learners of English may interpret certain sentences as passive constructions, while Arabic speakers may interpret them as transitive and intransitive variants. This can lead to a discrepancy between the target structure determined by the researcher and what the learner was actually attempting to produce.  Error analysis is an inadequacy in understanding the causes of errors. It assumes that correct usage is equivalent to correct rule formation. However, this assumption is not valid, as the learner may not have correctly sorted out the facts of English. The absence of errors in the previous two stages does not mean correct rule formation; it only suggests a limited sampling bias. Another problematic area of error analysis is relating to the source of errors. Dulay and Burt (1974b) established a category called ambiguous goofs, which can be categorized as either Interference-like Goofs or L1 Developmental Goofs. An example of an interference-like goof is hers pajamas produced by a Spanish-speaking child, while an L1 developmental goof is a typical overgeneralization error of English L1 children. An ambiguous error can be interpreted as either an interference error because it reflects a Spanish structure or as a developmental error because it is also found in English-speaking children learning their first language.  Learner production may be influenced simultaneously by multiple sources. For example, the English article system in Czech, a language without definite or indefinite articles, can be influenced by the lack of a comparable system in the Netherlands.  Error analysis has shown that the acquisition of English negation is influenced by both the native language (TL) and the non-native language (NL). In a study of Spanish speakers, Schumann (1979) found similar patterns of development but also found important differences. Negative utterances are formed by using the word "no" before the verb, followed by the use of "don't" and variants of "don't." Spanish is a language with preverbal negation, with the negative element being "no." When compared to native speakers of languages other than Spanish, slightly different facts emerge. For speakers of languages with preverbal negation, Stage 1 is more persistent than for speakers of languages without preverbal negation. In the case of Spanish speakers, two forces converge: the native language and facts of development. However, in speakers of languages like Japanese, only development plays a role. Error analysis is important in recognizing that learners are more than passive hiccupers of NL forms and functions, but it only provides a partial picture of the second language learner's output.  Chapter 5 RECENT PERSPECTIVES ON THE ROLE OF PREVIOUSLY KNOWN LANGUAGES Theories of learning This chapter discusses various approaches to second language learning, including the behaviorist theory of language and the role of native languages. During the 1950s and 1960s, challenges to the behaviorist theory of language and language learning emerged, with language being seen as a set of structured rules learned through innate principles and exposure to the language being learned. Three examples from child language literature challenge the imitation view of language acquisition, demonstrating that neither imitation nor reinforcement is sufficient to explain a child's linguistic behavior.  In the 1960s, it became common to see children actively involved in creating grammars of their language, rather than passively imitating their surroundings. Children construct grammars, make generalizations, test these generalizations, and alter or reformulate them when necessary. Early utterances by children, such as "no shoe and no book," are not best described as faulty imitation but as representing the child's attempt to systematically express negation. These assumptions have guided work in second language acquisition as well. The chapter also discusses other approaches to the study of child language acquisition, including innatist and interactionist approaches. Research on child second language acquisition is of interest to those interested in SLA. Child second language acquisition Child second language acquisition is a crucial aspect of language learning, focusing on the acquisition of two languages in childhood. Research on adult second language acquisition has influenced child second language studies, showing that strategies of language transfer, simplification, and overgeneralization of target language rules affect second language production in 7- to 8-year-old children. The quality of input to the learner is seen as a central variable in second language outcome. McLaughlin (1978a) claimed that language transfer in child second language acquisition is not possible unless the child is isolated from peers of the target language. R ecent empirical work shows that child second language learners display morphological sensitivity and are influenced by language transfer, which can involve grammatical lexical prototypical links. Child second language morpheme order studies Morpheme order studies by Brown in 1973 were influential in second language acquisition, arguing that child second language acquisition is similar to first language acquisition. However, Chomsky (1959) criticized behaviorism and suggested the learner should be an active participant in the learning process. Critics of contrastive analysis, particularly the creative construction theory, argued that children gradually reconstruct rules for speech guided by universal innate mechanisms. Morpheme order studies were used to verify hypotheses about child second language acquisition, and the Bilingual Syntax Measure (BSM) showed similar development patterns between Spanish and Chinese children, suggesting less significance of native language factors. Adult second language morpheme order studies Dulay and Burt's work on prepubescent children influenced the shift from a behaviorist view of language acquisition to a more mentalist one. Bailey, Madden, and Krashen's influential study on adult second language acquisition found consistency with the Dulay and Burt studies and similar results for the non- Spanish group. This led to the mentalist position embodied in Dulay and Burt's Creative Construction Hypothesis, which posited a "natural order" of English morpheme acquisition. However, morpheme order studies also suggested a more or less invariant order, although far from rigid. Larsen-Freeman found that native speakers of Japanese learning English had lower accuracy scores on English articles than other groups. The morpheme order studies faced challenges, including the possibility that the results obtained may be an artifact of the Bilingual Syntax Measure, which may have biased the results. However, other L2 studies not using the BSM as a data-elicitation measure obtained results similar to those of the BSM.  The Bailey, Madden, and Krashen study specifically incorporated learners from a wide variety of language backgrounds, which may have obliterated differences due to the L1 order. Morphomes with different meanings were categorized together, and the study's findings suggest that these morphemes should not be grouped as a single grammatical structure. The morpheme order studies have been critical in understanding the nature of developmental sequences, but their methodology has been criticized for not considering the total picture of a learner's use of a form. Correct forms may not always signify acquisition of correct rule structures, and the method of counting correct use in obligatory contexts misses instances where learners generalize a form to inappropriate contexts. Another criticism is that there appears to be individual variation in learner data, yet individual data is obscured with grouped data. Evidence for individual variability was provided by Hakuta (1974b) and Larsen- Freeman (1978).  The type of data elicited also appears to be problematic. Two final criticisms are that the studies investigated a limited number of grammatical morphemes and extended the implications to acquisition in general. While it may be the case that there is a predicted order of the acquisition of English morphemes, it is not the case that all of acquisition takes place in a predicted order. Additionally, the major theoretical significance of the studies was to demonstrate that the native language was an insignificant influence and that behaviorism could not be maintained to account for the process of second language acquisition (SLA). In conclusion, while morpheme order studies have been influential in our understanding of developmental sequences, they have failed in their completeness due to attempts to ascribe singular causality. Factors such as perceptual saliency, native language influences, semantic factors, syntactic complexity, and input frequency contribute to acquisition order Revised perspectives on the role of the native language The role of the native language (NL) in forming interlanguages has been a topic of debate since the late 1970s. Some argue that language background does not significantly affect the way ESL learners order English morphemes, while interference from the mother tongue plays a small role in language learning performance. However, some researchers argue that the role of the NL can be considered a creative process, with learners avoiding use based on the facts of the native language. Studies have shown that Finnish-speaking Finns learning English make transfer-induced errors that can be traced to Swedish (their L2), while Swedish-Finnish bilinguals make transfer-induced errors traceable to Swedish (their L1). This suggests that both groups rely more on Swedish than Finnish, depending on the learner's judgment. Avoidance The study by Kleinmann (1977) found that native language (NL) influences the use of structures, such as passives, present progressives, infinitive complements, and direct object pronouns, in Arabic and Spanish/Portuguese speakers. The differential behavior between groups could not be attributed to a lack of knowledge but rather to the choice to use or not to use particular structures to express given concepts. The source of avoidance is in dispute, with some arguing that differences between the L1 and L2 are the major source of avoidance, while others believe that avoidance is based on the complexity of the L2 structures. Dagut and Laufer (1985) found that Hebrew-speaking learners of English preferred the one-word equivalent of phrasal verbs, preferring those that are semantically more transparent. Hulstijn and Marchena (1989) found differences between transparent and nontransparent phrasal verbs, but learners did not accept them when there was close similarity between Dutch and English. Laufer and Eliasson (1993) found that the best predictor of avoidance is the L1-L2 difference, despite L1-L2 similarity and inherent complexity playing a role. Differential learning rates Ard and Homburg (1983, 1992) emphasized the importance of transfer in the psychology of learning, focusing on the facilitation of learning. They compared the responses of Spanish and Arabic learners to the vocabulary section of a standard English test, focusing on response patterns to different items. Spanish learners performed better on the second language acquisition (ACQ) task than Arabic speakers, indicating that learning time and accelerated learning rates are key factors. Differential learning rates can also be attributed to the internal system of the learner's L2 grammar, which exhibits delayed reorganization. Zobl (1982) discussed the concepts of delayed rule reorganization and the number of structures in a given developmental sequence. Henkes (1974) observed three children (French, Arabic, Spanish) in their acquisition of English, particularly the acquisition of the copula (the verb to be). Zobl noted that the Arabic child's lack of use of the copula was not native language-related, but rather due to the absence of the category in the L2 grammar Different paths The rate of acquisition of English definite article and demonstrative in second language acquisition varies across different languages. In a study by Zobl (1982), a Chinese-speaking child and a Spanish-speaking child were compared. The Chinese child showed early evidence of a form that appears to serve the function of a definitizer, while the Spanish-speaking child showed frequent use of both the definitizer and the demonstrative. The differences between the two children suggest that facts of their native languages lead them down two different paths. Wode (1977) argued that there is a predictable order of structures and that certain developmental structures must be used by learners before the second language (NL) can be expected to have an influence on second language production. He discussed the acquisition of English negation by German L1 children, focusing on the first stage of negation, where there is no evidence of NL influence. At a later stage, German-speaking children produce sentences that are clearly influenced by German, such as "That's no right" and "It's no Francisco." This suggests that learners must see some resemblance between the language they are learning and their native language before they recognize that the NL might be "useful" to them. This can be stated as the Transfer to Somewhere Principle, which is discussed in section 5.5.5. Overproduction The development and use of forms in native languages vary significantly. For instance, Chinese and Japanese speakers in English use a topic-comment structure, leading to an overproduction of sentences like "It is very unfortunate" and "there is or there are." This structure is used to express a discourse function, despite the use of other forms. The National Language (NL) function influences this structure, which becomes more like a target-like passive as learners become more sophisticated. Han (2000) found both true passive and topic-comment structures in spontaneous writing of Chinese learners of English, demonstrating that the L1 function has a subtle influence even at later proficiency stages. Predictability/selectivity In the late 1970s, the role of the second language (NL) in language transfer evolved from a dichotomous perspective to a when and under what conditions perspective. Andersen's Transfer to Somewhere Principle states that grammatical forms will occur consistently and significantly in interlanguage as a result of transfer if there already exists the potential for (mis-)generalization from the input to produce the same form or structure. Kleinmann (1977) suggested that when something in the L2 is very different from the L1, there is a "novelty effect," such as the progressive in Arabic, which is absent in Arabic but learned early and well. Bardovi-Harlig (1987) examined differences in the order of acquisition between sentences like 5-27 and 5-28, finding that salience, defined as the availability of input, is the main contributing factor to the unexpected outcome. Kellerman's 1979 proposal in the field of cross-linguistic influences focuses on the learner's perception of the distance between the first and second languages. The degree of language closeness is based on a learner's perception of both the distance between the languages and their perception of the potential transferability of linguistic elements. Second language processing The field of Second Language Acquisition (SLA) has shifted focus to how learners process language, particularly the role of the first language (L1). Studies show that learners resemble native speakers of the first language (L1), while those of the second language (L2) rely on the second language (L2). For example, Spanish and French native speakers interpret the relative clause as the first noun phrase, while English learners interpret it as the second. This suggests that L1 processing plays a crucial role in understanding sentences. Additionally, VanPatten and Keating's experiments show that L2 learners start with a universal processing principle, abandoning their L1 strategies with greater proficiency. Interlanguage transfer Interlanguage transfer refers to the influence of one language over another, affecting the acquisition of a third, fourth, or fifth language. It raises questions about how knowledge of a prior interlanguage is used in the addition of a third, fourth, or fifth language, and how this relation to influence from the first language is related. Current theories of language transfer based on only two languages in the multilingual mind need to understand principles that block native language transfer and encourage or discourage fossilization. Studies have shown the positive and negative effects of interlanguage transfer on the mental structuring and organization of the bilingual lexicon, with some functors coming from the second language in early L3 production.