Acquisition of the English Article System by Speakers of Polish in ESL and EFL - Assignment Example

ACQUISITION OF THE ENGLISH ARTICLE SYSTEM BY SPEAKERS OF POLISH IN ESL AND EFL INTRODUCTION - An attempt has been made by the researcher to understand the order of acquisition of the articles in English, ( a, the and zero ) by adult speakers of Polish language in two different environments, EFL ( English as a foreign language ) and ESL ( English as a second language ). Polish has no articles or article-like morphemes. The researcher argues on the basis of previous studies that English article system is considered to be one of the most difficult structural elements for ESL learners, on account of several reasons, which are as follows: 1. Articles occur quite frequently in discourse and learners find it difficult to apply rules continuously. 2. Articles being function words, (not content words) are not stressed in speech, hence difficult to discern 3. There is no one-form-one-function correspondence with articles. On the contrary, articles are burdened with multiple functions. In addition to these difficulties are those inherent in the process of SL/SF learning. BACKGROUND – Article acquisition is generally studied in the contexts in which they appear, that is, in the Noun Phrase. The researcher has used a modified version of Heubner’s Model ( 1983 ). Four basic NP contexts can be determined on the basis of whether a noun is specific referent [+/- SR] and whether it is known to the hearer [+/- HK]. They are as follows: Type 1 – [-SR+HK] – ( generics ) – a, the, zero Type 2 – [+SR +HK] – ( referential definites ) the Type 3 – [+SR-HK] – ( first mention nouns by speaker ) – a, zero Type 4 – [-SR-HK] – ( non-referentials ) – a, zero Type 5 – Idiomatic expressions and conventional uses. Most researchers have assumed the theory of UG (Chomsky) which governs the acquisition of L1 and L2. For Polish [-Art] speakers existence of articles may be strange but definiteness and indefiniteness in language is a universal property. Distinction between specific and non-specific referents and shared and unshared background knowledge is accomplished in Polish through word order, verbal aspects and demonstratives. The researcher proceeded to compare specific stages of acquisition across EFL and ESL to understand the natural order of acquisition. He selected the English article system because it is a linguistic form that is susceptible to environmental differences. Some researchers believe that English articles are un-teachable while others believe that they can be taught in classroom situations. Some are of the opinion that the process of language acquisition does not change with the environment and that there is no need to distinguish between ESL and EFL. METHOD: SAMPLE – total 25 participants were selected. 10 adult Polish learners of ESL – 3 low ability + 3 intermediate ability + 4 high ability level learners. Their average length of stay in America was one year; they used English daily and were employed in English speaking environments. They attended English classes at least 3 times a week. 10 adult Polish learners of EFL Their average length of stay was five years and they had not been out of Poland for more than a month and not used English outside the classroom. They got instruction two hours per week. 5 native English speakers served as the control group. TOOL – It consisted of 42 sentences with a total of 75 deleted obligatory uses of a/an, the and zero articles across 5 semantic types, with 15 instances of each type. No blanks were given otherwise participants would fill up every gap, including zero article. ANALYSIS – Percentage scores were calculated for each type. They were also plotted on line graphs separately for ESL and EFL. The data were combined according to proficiency levels and percentage scores for each type were calculated. They were also plotted on the line graph. This helped to follow the developmental sequence in the two different environments. The mean of overuse of each article (a, the, zero) was calculated for each proficiency level. CONCLUSION – It was observed that both ESL and EFL learners follow the same route of acquisition of English articles, irrespective of the environments. Sequence of L2 article acquisition is not different from L1 article acquisition. However, the findings are to be taken with caution because of the limitations of this research like small sample size, ignorance of individual differences in background etc., consideration of only written data and a strictly quantitative analysis. The researcher concluded that there exists a natural sequence in L2 acquisition which does not change from environment to environment. Hence, distinction between ESL and EFL is irrelevant. Majority EFL learners performed better than their ESL counterparts. (To be seen on the background that some researchers believe that English articles are un-teachable and can only be learnt through exposure). Distinction between ESL and EFL is useful to compare proficiency gains across two different settings and to understand the extent to which L1 affects inter -language outcome, if variable are kept constant. SOURCES OF LINGUISTIC KNOWLEDGE IN THE SECOND LANGUAGE ACQUISITION OF ENGLISH ARTICLES INTRODUCTION – There are 3 sources of acquisition of English articles by L2 learners – L1 transfer, L2 input and Universal Grammar (UG). Earlier researchers have proved that speakers of [-ART] L1s omit English articles in obligatory contexts to a greater extent than [+ART] L1 speakers. This study tries to find out the effect of L1 transfer on English semantics. It also tries to find out whether L2 learners have access to UG in absence of L1 transfer. The distinction between the and a is not only that of definiteness and indefiniteness but also that of specificity. In order to understand definiteness it is necessary to evaluate the discourse situation. English marks specifics only in the spoken or colloquial register. The problem is that definites as well as indefinites can be either specific or non-specific. Specificity, like definiteness is also a cross linguistic semantic universal on which the choice of articles depends. In some languages the definiteness distinctions are marked morphologically, as in English but in some other languages, the specificity distinctions are marked instead. Hence two types of errors are generally committed – overuse of specific indefinites or overuse of non-specific definites. The researcher observed that in absence of L1 transfer, the learners access both definiteness and specificity which are semantic universals. But since they cannot identify which of the two is relevant for choice of a or the, they fluctuate between the two. Assuming the effect of L1 transfer in L2 acquisition, the researcher hypothesized two possibilities – either fluctuation overrides transfer or transfer overrides fluctuation. METHOD SAMPLE – It consisted of 23 adult speakers of Russian and 24 adult speakers of Spanish and a control group of 6 adult native English speakers. TOOLS - The tools used were two written tests – an elicitation test of English articles and a cloze test of L2 proficiency. The cloze test consisted of 4 paragraphs with a blank after every 10 words. There were 30 blanks in each passage and total 120 blanks. The acceptable word criterion and not the exact word criterion was considered while correcting the test. The elicitation test consisted of 60 short dialogues, with a blank in the target sentence to be filled with any appropriate word or a dash. There were three categories of 6 items each for eliminating the and a separately. ANALYSIS – the participants were divided into 3 proficiency groups – beginner, intermediate and advanced. 4 L1 Russian and 4 L1 Spanish participants were eliminated because they gave less than two-thirds of acceptable responses. The percentages for all 4 types ( a, the, no article and other word ) were calculated for all 3 groups. The native English speakers performed as expected. They supplied the with all definites and a with all indefinites. The L1 Russian group committed two main error types – overuse of the with specific indefinites and overuse of a with non-specific definites. The L1 Spanish speakers were more accurate on non-specific definites and specific indefinites but this error was due to high article omission. Otherwise they were quite accurate on both types. In order to examine the role that proficiency plays in the degree of fluctuation, only the percentages of the L1 Russian speakers were taken into account. (Because L1 Spanish speakers do not show fluctuations). It is observed that as proficiency increases, use of the goes up with definites and use of a goes up with indefinites. Fluctuations gradually decrease with increase in proficiency; however, the differences in performance are not statistically very significant. This may be due to small sample size. Separate ANOVA tests were conducted on the Russian and Spanish groups in order to test the effects of definiteness and specificity. They are seen to interact significantly when use of the is measured and marginally when use of a is measured. This is because specificity effects are stronger with indefinites than with definites. It can be concluded that Spanish speakers are more accurate and increase on use of articles at all proficiency levels because accuracy is directly proportional to proficiency. Therefore they become more attuned to the discourse contexts. Results of individuals reflected fluctuation at individual level, but the errors found were minor in each case and present only in case of beginners. Recent Research on various aspects of language acquisition has suggested that some aspects of language acquisition are achieved through statistical learning. Results of child and adult learners of a mini language in a laboratory were compared. The and a could mean several things for the learners but the data provide evidence that the learners think only in terms of definiteness and specificity. The difference in children and adults is due to domain-general rather than language specific learning mechanisms. Children imposed a system on input, unlike adults. In case of adults, output closely resembled input. The can be associated either with assertion of existence or with modification. Results showed overuse of the largely confined to specific indefinites. This suggests that specificity and not assertion of existence influences choice of articles. Overuse of a was largely confined to non-specific definites hence it can be ruled out that learners use the wherever they see some kind of modification. Learners finally settle on precise patterns which they retrieve directly from UG. Innate linguistic knowledge does not guide the learners in choosing the actual specifications for L2. Language input plays an important role here. The input triggers lead the learners to decide that English articles are distinguished on the basis of definiteness and not specificity. It is concluded that both input and UG access are important in acquisition of the English articles. UG tells which article specifications are possible and input tells which article specifications are actually used in the language. Learners fluctuate between the and a until input tells them that choice of English articles is based on definiteness and not specificity. THE ACQUISITION F ARTICLES IN CHILD SECOND LANGUAGE ENGLISH: FLUCTUATIONS, TRANSFER OR BOTH? INTRODUCTION –The aim of this study is to determine the role of L1 transfer, test the Fluctuation Hypothesis ( FH ) and compare the process of acquisition of English articles in Children and adults. This study is heavily based upon previous research findings, which are, in fact, a starting point and a necessary background for this study. The relevant findings are briefly summarised as follows – Article distinctions based on definiteness and specificity. ( Ionin et al, 2004 ). [+definite] – the speaker and the hearer pre-suppose the existence of a unique individual in the set denoted by NP. [+specific] – The speaker intends to refer to a unique individual in the set denoted by the NP and considers this individual to possess some noteworthy property. English marks definiteness in its article system and not specificity. According to Ionin et al ( 2004 ), article systems of languages with two articles encode either definiteness or specificity but not both. The FH – Learners with [+ART] L1 transfer knowledge of articles in English L2 acquisition. But for those of [-Art] L1s there is no question of transfer. Hence, they fluctuate between a and the until they converge on a target system as a result of sufficient L2 input. If it is hypothesized that fluctuation overrides transfer, then all L2 learners should fluctuate between article choices. But Ionin et al found evidence to support the hypothesis that transfer overrides fluctuation, that is, L2 learners who’s L1 has articles, transfer article semantics from L1 to L2. The same results were obtained by Hawkins et al ( 2006 ) Ionin et al ( 2004 ) did not carry out a full investigation on child learners but they expected child L2 learners to converge on the definiteness pattern with greater success than adult L2 learners from the same L1, because children are better at making grammatical generalizations from various input. Previous studies do not focus on infelicitous use of null articles. The FH is also formulated as an account of article misuse, and not article omission. Incorporation of misuse/omission of null article is possible in the approach of Hawkins et al (2006). It is a feature assembly based approach. Schwartz and Sprouse (1996) proposed Full transfer/full access (FT/FA) based account. According to this, the L2 learner’s knowledge of L1 grammar as well as UG accessed directly play a role in L2 acquisition. Hence, both approaches are mutually compatible. Lardiere (2004), Robertson (2000) and White (2003) found the directional difference in accuracy. Their studies have widely reported that L2 English learners are more accurate in choosing the definite article in definite contexts than they are in choosing the indefinite article in indefinite contexts. This might be because English need not take into account the number/count/mass in case of the definite article the. Hawkins et al (2006) also concluded that a is inserted in singular contexts. Hence, learners have to identify the feature (singular) for accurate use of a. If this featural complexity is responsible for errors, then L2 learners could be expected to display this pattern regardless of L1 background. RESEARCH QUESTIONS – To test whether [+Art] and [-Art] L1 groups fluctuate between article parameters. To test whether [-Art] group is more likely to omit articles than [+Art] groups. To test whether there is a difference between accuracy in article use in definite and indefinite contexts, and whether this changes as a function of L1 background. To test whether child L2 learners acquire English articles more rapidly and successfully than their adult counterparts. METHOD SAMPLE – 17 children learning English as a second language were studied for two years. Some used [-Art] L1 (Chinese, Korean, Japanese) and some used [+Art] L1 (Spanish, Romanian, Arabic). The performance of learners was studied after every 6 months. The children were divided into 2 groups depending on whether their L1 was [+Art] or [-Art]. TOOL – picture books designed as part of ENNI project were used. Each book contained a set of 3 series of line drawings. Children narrated the story by looking at the pictures which were not visible to the interviewers who were also instructed to use only neutral cues. The study was limited to singular common nouns. Use of articles was recorded for both definite and indefinite contexts. Null articles were always considered incorrect since proper nouns were not included. ANALYSIS – The mean percentage of accurate use of articles in obligatory contexts was prepared separately for a and the. An ANOVA test was also conducted to understand the accuracy level better depending on L1 of participants. An analysis of individual scores showed that most children followed the group trend. Error type relative distribution was also studied. The percentage scores confirm that learners grew steadily more accurate over time with their use of articles. They were more accurate with the than with a. [-Art] group showed lower accuracy with definite article in the first test round. A post-hoc independent sample t-test was conducted and it showed that overall, L1 did not much influence acquisition of articles. Only children with [-Art] L1 lagged behind [+Art] group at first round in use of article the. Hence it can be said the article type was a more important factor than L1. Most children in both groups were more accurate with the than with a, throughout the 2 years. Misuse, rather than omission was clearly the major error type for both the groups. CONCLUSION - Regarding fluctuation it can be said that both L1 groups demonstrated fluctuation pattern in article choice, and little L1 influence. Hence, in case of children, fluctuation overrides transfer because unlike adults they are more efficient in accessing UG to establish a new, language specific grammar. The was used more often in definite contexts than in indefinite contexts at all stages of acquisition by learners in both groups. This could be due to greater semantic complexity of a and has nothing to do with L1 transfer. When results were compared with those of adult L2 learners it was found that children converge faster. Some similarities and some differences were observed in adults and children. There was limited role of L1 influence in children. Children showed opposite pattern regarding fluctuation and transfer. THE ENGLISH ARTICLE SYSTEM: ACQUISITION, FUNCTION AND PEDAGOGY INTRODUCTION – This is not a research study but a descriptive paper that deals with the acquisition, function and frequency of articles in English. On the basis of this explanation, it provides a framework for the presentation of articles in the classroom at 3 levels based on proficiency – beginner, intermediate and advanced. It has been repeatedly said that the English article system is difficult to acquire and since article errors rarely lead to miscomprehension, there is no point in wasting valuable classroom time in teaching articles. While language acquisition is a natural and unconscious process, language learning is a conscious process and can be influenced only by classroom instruction. Shwartz (1993) has termed it as learned linguistic knowledge (LLK). Moreover, a second language is always a second one and inter-language grammar is different from native competence. LLK is thus necessary to supply competence. The explanation – production model was followed traditionally in classrooms but Van Patten and Cadierno (1993) followed explanation with non-production activities. They recommend instruction focussed on input processing. Whatever the method, it is now agreed that formal instruction does have a positive effect. It must be admitted that in highly sophisticated and academic writing, article errors can affect the quality and create a bad impression. It is the responsibility of teachers to help their students to understand the English article system. Acquisition – Acquisition of articles in L2 English is different as it is governed by the presence or absence of articles in the L1. Articles are learnt earlier by [+Art] speakers than by [-Art] speakers. Overuse of zero article is observed in the initial stages in case of [-Art] L1 speakers. But zero use can actually be omission of articles since they do not exist in the L1. Hence it can be called acquisition by default. Overuse of zero article decreases with increasing proficiency. Performance of [+Art] L1 speakers shows a U shaped curve on graph in the use of zero article. When [-Art] learners realize that zero article is not always appropriate, they overuse the with all nouns. [+Art] users overuse the from the start. They are ahead of [-Art] learners because it takes the latter some time to realize that something like an article system exists. But this difference is observed only in the initial stages. According to Chaudron and Parker (1988) a is learnt only after the is mastered. Pica (1983, 1985) found that acquisition of a was unique because it did not seem to be influenced by instruction. It is different because it is a marker of countability. L2 English learners do understand that context is important in the choice of indefinite article but they fail to understand how English speakers use this context. More research is necessary on this psycholinguistic mechanism. In English, countability, number and definiteness, all 3 elements must be considered in arriving at the correct choice of articles. Frequency and Function – Articles are most frequently occurring free morphemes in English. According to COBUILD the is the most frequent word in a corpus of 20 million words and a is in the fifth position. Among the 3 English articles, zero is most frequent, followed by the and then a. Palmer ( 1939 ) suggested that zero article has 2 forms – one used with non-count plural nouns ( milk, eggs ) and the other used with singular count and proper nouns ( lunch, London ). The first is referred to as zero article and the second is called null article. This paper discusses the function and characteristics of each article in detail. We shall summarise it in brief. The zero article is the most indefinite. Its function is to remove the boundaries that make nouns discrete. It occurs with indefinite count nouns and plural count nouns. If zero article is placed before a singular count noun, the noun is rendered as a mass noun. The null article is the most definite of articles. It generally names a one-member set and occurs with singular count nouns and with singular proper nouns. The indefinite a signifies a boundary that makes an entity concrete and countable. It occurs with singular count nouns. Definite the indicates an identity intended by the speaker and/or shared by the hearer. The is not limited by countability or number. It is primarily used with non-generic nouns. Generic the indicates a class of entities and must be used with singular count nouns. PEDAGOGICAL FRAMEWORK – Peter Master has suggested a framework for formal instruction in the classroom. This is in accordance with the acquisition, frequency and function of articles described above. Beginning level – At this level, it is not advisable to emphasise rules of article usage. Words should be introduced in the classroom with the help of actual objects or pictures and should always be used with the articles, as and how they occur with the words. Beginning with concrete countable nouns, the teacher should proceed to mass nouns. Focus on the should be avoided at this stage. Intermediate level – Systematic teaching of articles using cognitive methods is more effective than marking errors in composition. Sufficient practice until a single distinction is mastered is necessary. Berry (1991) has provided several interesting article tasks to encourage comprehension and application rather than a simple selection of correct articles. Van Patten and Cadierno’s model (1993) has designed an exercise based on their notion of input processing which is also useful at this stage. Advanced level – Rules are not useful – they are often forgotten at this stage. Focus should be on lexical items instead of syntactic rules. Rules are internalised at the advanced level. Students should be encouraged to maintain records so that they will avoid them on their own. But errors should not be criticised, as Lewis argues, (1993) “ --- error is essential part of learning process.” Finally, it is suggested that English articles are difficult to learn and teach and should be introduced over a long period of time. Articles can and should be taught in the classroom. The end Read More