Language transfer

Introduction

The title of this research article has caught my attention because it talks about language transfer which is an area I am always interested in knowing more about it. Also, being an ESL teacher, it is definitely beneficial for me to study this article as it specifically investigates the cross-language and writing system transfer between Chinese and English.

This paper consists of two main parts. Part one is a description of the article itself. The background and details of the research will be summarized and presented here. Part two is a critique of the article. Various parts of the research will be analyzed and, hopefully, modifications and suggestions can be made to better the research design and generate ideas for further investigation into this topic.

Part One: The article itself

Contextualization

According to the introduction made by Wang, Perfetti, & Liu (2005), various studies conducted in the 1990s have tried to investigate whether there is a relationship between learning to read and the orthography and phonology of a language.

As cited in Wang et al. (2005), certain studies (Adams, 1990; Perfetti, 1992; Treiman, 1993) have said that learning to read is actually learning to match the phonological form with the orthographic form of the language. This has also been claimed as language-specific, i.e. subject to the nature and writing system of the language (Chen & Tzeng, 1992; Feldman, 1987; Frith, Wimmer, & Landerl, 1998; Goswami, Gombert, & Barrera, 1998; Leong & Tamaoka, 1998; Perfetti, 1999; as cited in Wang et al., 2005).

However, recent studies (Cisero & Royer, 1995; Durgunoglu, Nagy, & Hancin-Bhatt, 1993; Durgunoglu, 2002; as cited in Wang et al., 2005) have found that cross-language transfer of phonological skills does take place when one is learning to read two alphabetic languages, like Spanish and English.

These findings have aroused the researchers’ (Wang et al., 2005) interest in investigating whether there is cross-language and writing system transfer when one is learning to read Chinese and English, i.e. acquiring biliteracy.

Background in the Literature

To prepare for their in-depth research on the topic, Wang et al. (2005) have conducted a literature review focusing on the following three aspects:

(i)The orthography and phonology of Chinese

A number of studies (Perfetti, 1999; Perfetti, Zhang, & Berent, 1992; Defrancis, 1989; Mattingly, 1992; Perfetti, Liu, & Tan, 2005; as cited in Wang et al., 2005) have been reviewed by Wang et al. (2005) in order to produce the following general overview of the Chinese orthography.

It has been found that Chinese is a nonalphabetic writing system, in which the basic unit is the character. Each character consists of basic strokes, which can be combined to form one or more component radicals. A character which is composed of two or more component radicals is called a compound character. The structure of the component radicals in these compound characters are usually left-right or top-bottom.

Wang et al. (2005) have also reviewed certain studies (Hanely, Tzeng, & Huang, 1999; Taylor & Taylor, 1995; as cited in Wang et al., 2005) so as to offer a clear introduction of the Chinese phonology. It is now being summarized below.

The basic speech unit of Mandarin Chinese is the syllable which is made up of the onset and the rime. Very often, the onset is a single consonant while the rime is composed of vowels. As a result, the number of syllables in Mandarin Chinese is pretty small (i.e. around 400) whereas the number of homophones is quite large. Fortunately, the use of the four tones in Mandarin Chinese (i.e. 1. high level; 2. high-rising; 3. falling-rising; 4. high-falling) have successfully reduced the large number of homophones. Yet, one should note that the tones are not shown in written Chinese, i.e. they are not part of the written characters. Thus, they will not help learners distinguish the characters.

(ii)Different cognitive demands involved in learning to read the Chinese and English writing systems

According to Wang et al. (2005), there are three main contrasts between the two target language and writing systems.

**This table is a simplified version of Table 1 (Wang et al., 2005, p. 69)

As suggested by Wang et al. (2005), these contrastive differences are bound to cause differences in learning to read these two languages.

First, studies (Byrne & Fielding-Barnsley, 1995; Hulme et al., 2002; Lundburg, Frost, & Peterson, 1988; Muter, Hulme, Snowling, & Taylor, 1998; Huang & Hanley, 1994; as cited in Wang et al., 2005) have discovered that children’s ability to read English can be predicted by their skills in processing the phonemes (i.e. the small phonological units) and phonological awareness.

On the contrary, studies about Chinese reading (Perfetti et al., 2005; Taft, Zhu, & Peng, 1999; as cited in Wang et al., 2005) have found that the phonological and meaning information can only be activated after the orthographic representation has been recognized. Then, Wang et al. (2005) have concluded that graphemic information and visual skills are essential for learning how to read Chinese.

Despite these previous studies, Wang et al. (2005) have mentioned that recent studies (Ho & Bryant, 1997; Hu & Catts, 1998; McBride-Chang & Ho, 2000; Shu, Anderson, & Wu, 2000; as cited in Wang et al., 2005) have indicated that phonological information and early phonological skills such as rhyme processing do play a part in Chinese reading acquisition.

(iii)Cross-language transfer in bilingual and biliteracy acquisition

Wang et al. (2005) have obtained two key findings about cross-language transfer through reviewing numerous studies (Durgunoglu et al., 1993; Cisero & Royer, 1995; Ganschow & Sparks, 1995; Comeau, Cormier, Grandmaison, & Lacroix, 1999; D’Angiulli, Siegel, & Serra, 2001; Geva & Siegel, 2000; as cited in Wang et al., 2005).

First, Wang et al. (2005) have found that a clear phonological relationship exists between two alphabetic languages such as Spanish-English, French-English, English-French, English-Italian, etc. This means the phonological skills of one language are highly related to those of the other language.

Second, Wang et al. (2005) have discovered that phonological skills of one language can help with the word reading skills of the other language.

Yet, Wang et al. (2005) have failed to figure out whether there is such cross-language transfer existing in Chinese and English as there is little research done in this area and the two language systems are contrastive in nature.

Besides that, certain studies (Wydell & Butterworth, 1999; Liow & Poon, 1998; Liow, 1999; as cited in Wang et al., 2005) have tried to argue that there is dissociation or even a negative transfer from the nonalphabetic first language (L1, Chinese) to the alphabetic second language (L2, English).

Nevertheless, Wang et al. (2005) have finally justified their research topic by saying that these previous studies have failed to test the cross-language and writing system transfer as they did not investigate the phonological and orthographic processing skills of the learner’s L1 and L2.

Research Hypotheses

Wang et al. (2005) have hypothesized that ‘bilingual reading acquisition is a joint function of shared phonological skills and writing system specific skills’ (p. 72). They have particularly predicted that ‘sensitivity in English and in Chinese to onset and rime, common linguistic units in both languages, will be correlated’ and ‘Pinyin reading skills will correlate with English word reading, since the two systems share the alphabetic principle’ (p. 72). However, Wang et al. (2005) have estimated that orthographic skills ‘are writing system and script specific skills’ (p. 72).

Details of the Research

(i)Research design

The design of this research is mainly quantitative in nature. A number of experimental tasks were completed by the participants in order to obtain data in objective ways.

(ii)Setting & Participants

As reported by Wang et al. (2005), the forty-six participants came from the Washington, DC area and they were Chinese immigrant children consisting of 24 boys and 22 girls with the mean age of 8 years and 2 months (SD = 9.1 months). All of them learned Chinese as their first language and had developed normal English proficiency. They attended English classes in public schools and weekend Chinese schools in which the Pinyin systems and a simplified version of Chinese characters were taught and used. It should be noted that forty-two of them spoke both Mandarin Chinese and English at home while the other four only spoke Mandarin Chinese at home.

(iii)Methodology

Referring to the description given by Wang et al. (2005), the participants were tested in two 30-minute sessions in which they had to complete a set of Chinese or English experimental tasks in each session in a language laboratory. The list below has summarized the various experimental tasks adopted by Wang et al. (2005).

According to Wang et al. (2005), there was also a nonverbal skill test which was used to test the participants’ nonverbal ability.

(iv)Analytic procedures

As mentioned by Wang et al. (2005) in the presentation of the research results, several measures were adopted to analyze the quantitative data collected from those experimental tasks.

First, means and standard deviations were used to show the participants’ performance in the Chinese and English language and reading tasks.

Second, the bivariate Pearson correlations were used to measure the correlations Chinese and English tasks.

Third, the stepwise regression analyses were adopted to find out the best predictors for Chinese and English reading.

Finally, the hierarchical regression analyses were used to explore (a) whether Chinese phonological and orthographic processing would affect English word reading; and (b) whether English phonological and orthographic processing would affect Chinese character reading.

(v)Results

With the help of the bivariate Pearson correlations, Wang et al. (2005) found that there was a significant correlation between the Chinese onset matching skill and English onset and rime matching skill. Moreover, another significant correlation was found between Chinese orthographic choice and Chinese character naming. Furthermore, Pinyin was found highly correlated with the pseudoword reading in English whereas the English phoneme deletion task was also found highly correlated with English real word and pseudoword naming. Last but not least, Chinese tone was found correlated with character reading as well.

Through the stepwise regression analyses, Wang et al. (2005) found Chinese orthographic processing to be the best predictor for Chinese character reading while the phoneme deletion skill was the best predictor for English real word and pseudoword reading.

Consequently, by using the hierarchical regression analyses, Wang et al. (2005) found that only Chinese tone processing skill, but not orthographic skill, could affect English pseudoword reading. They could not find any cross-language influence of the English tasks on Chinese character reading.

(vi)Discussion

Wang et al. (2005) have concluded that the findings of this research do support their hypothesis.

Specifically, Wang et al. (2005) have discussed the three significant findings (i.e. the significant correlations between Chinese onset, English onset, and rime matching skills; the significant influence of Chinese tone processing skill on English pseudoword reading; and the significant correlation between Pinyin and English pseudoword reading) and derived the implication that phonological awareness and skills are not language-specific but shared between Chinese and English reading when bilingual reading acquisition is taking place.

Besides that, based on Wang et al. (2005), the finding of orthographic skills being unable to affect English reading skills has also supported the hypothesis proposed by Wang et al. (2005). This has been explained by Wang et al. (2005) who has cited a study (Liu & Perfetti, 2003; as cited in Wang et al., 2005) to show that the orthographic skills of Chinese and English are language-specific as Chinese reading requires the use of both the left and right occipital brain areas while English reading mainly involves the use of the left occipital brain area.

Part Two: The Critique

Assessment of the Research’s Internal Validity

Based on my analysis, this research done by Wang et al. (2005) is considered to be a good research which has demonstrated a high level of internal validity. The following is a detailed critique of the various parts of the research.

(i)High correspondence between the research hypotheses and the findings

As reflected from the previous session, the findings generated from this research are highly related to the research hypotheses established beforehand. Actually, they have successfully confirmed the hypotheses, proving that there is a certain degree of cross-language transfer in Chinese-English biliteracy acquisition.

(ii)In-depth literature review

Wang et al. (2005) have done a detailed and in-depth literature review on Chinese phonology and orthography, and previous studies related to the cross-language transfer in bilingual and biliteracy acquisition. This certainly helps lay down a good foundation for the later data analysis and discussion.

However, it may be better if Wang et al. (2005) can provide the audience with a more comprehensive literature review by describing the English phonology and orthography in greater detail. In this way, the audience can have a clearer picture of the phonology and orthography of these two language systems. This can then enhance the audience’s understanding of the research findings.

(iii)Well-constructed and clear research hypotheses

The research hypotheses of this study are well-constructed with clear predictions made about the relationship among the variables mentioned. This can certainly help increase the internal validity as this has given clear directions for the development of research design, measurement tools and analytic procedures.

(iv)A well-selected group of participants

The participants are well-selected with little variation in their background and learning profile. It is very wise for Wang et al. (2005) to study bilingual children as this can minimize the dominance of L1 over L2, which might affect the findings of the research.

Moreover, the sample size is big enough for making generalizations. And, the number of male participants is similar to that of the female participants. This can help reduce the gender influence on the findings.

(v)Carefully-designed measurement tools

Obviously, the large number of Chinese and English experimental tasks and methods for data analysis are carefully designed and chosen. As shown from the results, the Chinese and English tasks are highly correlated, indicating a high level of internal validity.

Nevertheless, it is thought that the tasks are cognitively or mentally too demanding for the young participants. This is because they are required to process a number of tasks and a large amount of information and instructions within a short period of time.

(vi)Variable overlooked — gender difference

According to Skaalvik & Rankin (1994), girls have been found to be more motivated to study language than boys. One may then wonder if this gender difference would affect the level of cross-language transfer in bilingual reading acquisition. Since Wang et al. (2005) have recruited a similar amount of boys and girls as participants, they may also take a look at the influence of this variable.

Assessment of the Research’s External Validity

Although this research enjoys a high level of internal validity, its external validity is relatively low and limited. The reasons are as follows:

(i)Findings are limited to learners with a similar background or learning profile

As mentioned in the above, the background (e.g. age and family background) and learning profile of the selected participants are more or less the same due to the careful selection done by Wang et al. (2005) to avoid the research findings being affected by the individual differences among the participants. As a consequence, the research findings derived from this specific group of participants can only be applied to learners with a similar background and learning profile.

(ii)Findings may not be able to make generalizations in the actual learning context

In this research, Wang et al. (2005) have conducted all the Chinese and English experimental tasks in a language laboratory in a bid to minimize the influence of the numerous factors which may affect the participants’ cross-language transfer in the actual learning context. For instance, in the real learning context, learners’ cross-language transfer in the process of language acquisition can be affected by their learning environment, teachers’ teaching methodologies, peer influence, learning materials, etc. Hence, the research findings derived from tasks conducted in the language laboratory which keeps the other factors constant may not be able to contribute to making generalizations in the actual learning context.

(iii)Few implications can be derived for language learning and teaching

Few implications can be generated from this research for language learning and teaching as the findings can only be applied to a specific type of learners (i.e. bilingual Chinese-English learners with a particular background and learning profile). Besides that, the experimental tasks conducted are all at word level and this definitely does not resemble the actual reading acquisition in language learning, which is usually at sentence level or discourse level.

Suggested Modifications for the Research

With reference to my analysis shown in the above, certain modifications can be made in the different parts of the research mentioned below.

(i)Literature Review

It is suggested that Wang et al. (2005) may also give the audience an introduction on the English phonology and orthography instead of simply describing English as an alphabetic, nontonal language with phonemes as the smallest basic speech units in the language. This is because this introduction can help to make this research more complete as the audience are then allowed to make a clearer comparison between the two target language systems and better understand the findings obtained from this research.

(ii)Measurement tools

As mentioned previously, the Chinese and English experimental tasks can be cognitively or mentally too demanding for the young participants as they need to process a lot of information and instructions within a short time. Therefore, it is suggested that Wang et al. (2005) may either simplify the content and instructions of the tasks or lengthen the period of time given to participants for completing the tasks. In this way, the influence of frustration and fatigue on participants’ performance in the tasks can be avoided.

(iii)Discussion

For the discussion part, it is recommended that Wang et al. (2005) can also analyze the data collected and see if there is any influence of gender difference on the cross-language transfer between Chinese and English reading acquisition. By doing so, significant implications may be derived and this will in turn provide useful implications for language learning and teaching in the real learning context.

Suggestions for an Extension Study

After having studied this research and its findings, it is believed that further research can be done in the following areas in order to derive more implications for language learning.

(i)Chinese and English reading at sentence level or discourse level

Referring to what mentioned beforehand, this research done by Wang et al. (2005) focuses on Chinese and English reading acquisition at word level only as all the experimental tasks are designed at word level. However, it is a well-known fact that learning to read may proceed to sentence level or discourse level. Thus, it is definitely worth investigating whether cross-language transfer still occurs when Chinese and English reading acquisition is done at sentence level or discourse level.

(ii)Bilingual children whose first language is English

Since bilingual reading acquisition is the main focus of this research, it is believed that similar research can also be done to study bilingual Chinese-English children whose first language is English and see if similar findings about the cross-language transfer can be generated. In this way, the findings generated may help confirm the findings produced by Wang et al. (2005) and enhance the external validity of the study done by Wang et al. (2005).

(iii)Children with dyslexia or other language learning problems

Likewise, will there be any changes to the present findings of the research conducted by Wang et al. (2005) if the bilingual participants suffer from dyslexia or other language learning problems? It is thought to be a good idea to further research on this group of participants. Hopefully, useful implications can be derived to help enhance the learning effectiveness for this group of learners.

(iv)Reading fluency and comprehension

As mentioned in the above, research on bilingual reading acquisition should not be limited to word-level as reading occurs more often at sentence level or discourse level. In addition, reading acquisition should not be limited to phonological and orthographic skills of individual words only. Actually, reading fluency and comprehension are two important aspects in reading acquisition. So, it is worth researching on these two aspects and figure out if there is any cross-language transfer in these areas of Chinese and English language learning.

Conclusion

Critiquing a research article is indeed a challenging but thought-provoking task to me. Through working on this assignment, I have learned about how a quantitative research can be conducted in the area of language acquisition, and gradually developed a better set of critical analysis skills when writing the critique. I would surely bear in mind the research and analytical skills that I have learned from the study of Wang et al. (2005) and put them into use when I am given a chance to conduct a research on my own.