A Case for Teachers’ Language Input and Students’ Production


Chuen-Teng Huang(黃春騰)     Claudia Ling-xian Liu劉凌嫻)

National Changhua University of Education

huangtt@cc.ncue.edu.tw          claudialiu@yam.com


This is a case study aiming at the examination of the relationship between the type of the teachers’ oral English input and the students’ oral production. Participants of the present study were students of two English classes, one from a bilingual elementary school and the other from an EFL elementary school. Both groups of the learners had approximately the same instruction hours before. The data collection methods included classroom observation and a story retelling task. The results showed that the learners’ input type and their output are strongly correlated with respect to the formulaic expressions they produced.



Krashen’s (1980) Input Hypothesis states that a learner acquires a language only when s/he understands what s/he hears or reads; and sufficient language input, oral or written, is believed to be beneficial for the learner’s language development (Ellis, 1994; Spolsky, 1989).  Comprehensible input is regarded as a sufficient condition for language acquisition by Krashen.  However, diverse conclusions are derived when some researchers, such as Brown (1973), Huttenlocher (1998), and Peters & Menn (1993), examine spontaneous production corpora by looking closely at the properties of language learners’ input and its relationship to the quality and features of teachers’ output.

An earlier study by Brown (1973) indicates that there is no direct implicational relationship between input and the course of individual children’s linguistic development and production.  According to his viewpoint, what language learners learn in classroom may not straightly affect their language production.  On the other hand, Peters et al. (1993) suggest that the emergence of certain English prepositions in children’s early speech appears to be reasonably related to different inputs they receive from their respective environments.  Similarly, Huttenlocher (1998) argues that differences in language input within the normal range are related to the growth of language learners’ vocabulary and syntactic skills.  Comprehensible input, according to his observation, is indispensable as a resource of target language.  Although views on the role of input may be different, comprehensible input is commonly believed to be necessary for children’s language development.

Additionally, ample evidence from both first and second language acquisition studies (e.g. Peters, 1983; Henry, 1996) indicates that children make use of whole prefabricated utterances (i.e. formulaic speech or FLS) in appropriate social contexts, although they also construct speech from discrete grammatical units.  Related literature has widely observed that L2 learners begin the process of SLA by storing formulaic expressions, i.e., the morphosyntactic speech which the learners memorize as a single form to be produced under proper circumstances.  Such a phenomenon indicates the significance of formulaic speech in the evolutionary process of children’s grammar acquisition.  These unanalyzed chunks of language are believed to “maximize communication with a minimum of language” (Peters, 1983).  The memorized chunks and patterns are able to be retrieved rapidly by language learners and thus allow them to express themselves for communicative needs.   Later on the learner employs the acquired grammatical rules to divide these chunks of language into frames with slots, lexical items and syntactic structures (Henry, 1996). The understanding of the role of formulaic speech is, thus, considered prominent for the understanding of creative, rule-governed processes in SLA.

Krashen (2002b) argues that when teachers are not native speakers of a target language, there is a concern that students may be exposed to imperfect models.  It is because that NNS teachers may not be able to provide the same quality and quantity of input needed for language acquisition and development as a NS teacher. In Taiwan, except for a few Chinese-English bilingual elementary schools, most English teachers in elementary schools are NNS and classrooms serve as the main circumstance wherein students are exposed to the target language.  One of Streven’s (1986 ) six postulates for a model of language learning and teaching, states that “the manner of presentation of language input to a learner affects comprehension and therefore learning” (p. 51).  Thus, the researchers were interested in finding out whether there exists a relationship between teacher’s input and student’s output in the perspective of formulaic speech and creative expressions.



This study mainly attempts to explore the differences between the effect of NNS and NS English teachers’ inputs on the students’ formulaic speech production.  The research questions are stated as follows:


1. Whether NNS and NS English teachers would provide significantly different amount of language input for their students with respect to formulaic speech expressions?

2. Whether the formulaic speech expressions from different kinds of teacher would have an effect on the students’ output?

3. How would the formulaic speech and creative expressions produced by the two groups of students be different?



In the present study, the teachers’ input is the independent variable, and the students’ output is the dependent variable. The students’ individual differences are all treated as constants.  The hypotheses were stated as follows:

1. NNS and NS English teachers may provide different language input with regard to the amount of formulaic speech in the similar instruction sessions.

2. Students receive different type of input with regard to the formulaic speech (in contrast to created speech) may produce different type of output in the same regard.



The current study was conducted mainly through classroom observation, and students’ story-retelling and interviews. The participants were two groups of students selected from two different schools, one an ordinary elementary school, and the other a Chinese-English bilingual elementary school in central Taiwan. Five students from each group were recommended as interviewees by their instructors for story-retelling. The related information about the participants is shown in the following table:


Table 1. Students’ English Learning Experience


Group 1

Group 2


bilingual learners

EFL learners




Home Room Class Size



Grade to Start

1st Grade

3rd Grade

Accumulated Learning Hours

Approx. 180 hours

Approx. 184.8 hours


NS (T1)

NNS (T2)


5 bilinguals

(SA, SB, SC, SD & SE)

5 EFL learners

(SF, SG, SH, SI & SJ)

Legend: SA, SB, SC, etc. = individual students for interview


The instructor for Group 1 is NS of English, while the instructor for Group 2 is NNS of English, each labeled as T1 and T2 respectively. Their information is as follows:


Table 2. Teachers’ Background Information







Experience as a Teacher

6-10 years

1-5 years

Experience of Teaching Eng



Academic Degree


(Early Children Education)


(Language Education)

The story used in the study is The Magic Key written by Roderick Hunt and Susan Sprengeler, illustrated by Alex Brychta, and published by Oxford University Press.



For the instructors’ input, the formal data collection sessions began a week after the preparatory session was conducted, which was meant for both the researchers and the participants to get to feel comfortable with each other. One of the researchers observed the two types of class, each for five and four hours respectively, and the data were later transcribed verbatim.

For collected students’ output, each of the selected student interviewees was told a story individually in mandarin Chinese before they were instructed to retell the same story a week later with the prompts of the ordered drawings based on the storylines.  The story retelling data were also audio-taped and later transcribed verbatim. All the interviewees’ language production data were then categorized, analyzed, and compared with the data collected from their teachers’ input (i.e., those from the classroom observations.)



Identifying Students’ Syntactic Development

In order to examine the students’ oral production in terms of their syntactic development, the students’ story-retelling data were coded following Clyne’s “productive phases” framework (Clyne, Jenkins, Chen, Tsokalidou, & Wallner, 1995; Hagino, 2002). With the researchers’ modification, it includes the following four phases:

Phase 0: Responses only in Chinese (although comprehension may be fairly well developed);

Phase 1: One- or two-word, unanalyzed and formulaic L2 utterances, with the rest in Chinese;

Phase 1a: The entire discourse is in Chinese but some Chinese words are pronounced with L2 pronunciation;

Phase 2: The matrix language of the discourse is Chinese, but individual L2 items, for example, nouns, noun phrases, uninflected adjectives, or infinitives, are transferred;

Phase 3: An attempt is made to speak L2, with frequent code-switching to L1 within sentences, as well as patterns to integrate Chinese words into L2:

Phase 4: The matrix language of the discourse is clearly L2, but occasionally Chinese words are transferred, and sometimes integrated into the L2 phonological and/or grammatical system.

(adapted from Clyne et al., 1995, pp. 31-32 and from Hagino, 2002, pp. 4-5)


Identifying Formulaic and Non-Formulaic Speech

Identifying language learner’s chunks and finding out particular rule-driven or rote-learning utterances may raise some problems if to infer his/her process from production. However, a number of scholars (Myles, Hooper & Mitchell, 1998) have suggested certain range of criteria for identifying chunks which were adopted by the current study. They are:

 1. at least two morphemes in length;

2. phonologically coherent, that is fluently articulated, non-hesitant;

3. unrelated to productive patterns in the learner’s speech;

4. greater complexity in comparison with the learner’s other output;

5. used repeatedly and always in the same form;

6. may be inappropriate (syntactically, semantically or pragmatically) or otherwise idiosyncratic;

7. situationally dependent;



According to the transcription of classroom observations the learners’ task, the following findings of various categories were drawn on.


Language Use in the Two Types of Classrooms

Based on the collected data, the following conditions are observed:

Teacher 1 used English exclusively; on the other hand, Teacher 2 talked more Chinese than English in the classroom; (the ratio is 58:42). When teaching and explaining a new word, T1 gave explanation in English with vivid facial expression and gestures for students to guess; on the other hand, T2 preferred direct translation to avoid any misunderstanding.   In terms of teacher-student interaction, Teacher 1 never spoke Chinese, even though students sometimes talked with her or asked her questions in mandarin Chinese and Taiwanese.  Therefore, students used English to talk to their teacher more often. They would sometimes imitate the teacher’s words and speech. Situations in Group 2 were different.  Nonetheless, the bilingual learners still often used L1 to talk with their classmates, to ask questions, to confirm their teacher’s instructions or commands, and to inform their peers what to do.  On the other hand, Teacher 2 used English, mandarin and Taiwanese in classroom, and mixed Chinese with a relatively high proportion, i.e., 70%.  When the EFL students talked with their teacher, they had a preference to speaking their L1, unless they were asked to speak in L2 or to repeat after their teacher.  As for students’ use of language, both bilingual and EFL learners used L2 mainly in the context of teacher-directed activities.  In the T2 classroom, the researchers did not find clear chances or activities for the students to use the L2. The bilingual students’ use of English would extend to the class recess while this did not happen in the Group 2 setting.


Teachers’ Language Input

As said previously, Teacher 1 is an experienced NS of English, thus she gave extensive and comprehensible target language input exclusively in the classroom. In addition, she did not have an obvious preference of speech repetition as T2 did. Instead, she used a large number of elaborations, questions, and attention-getters to induce the children’s response.  Also, she avoided long and tedious word explanation. She preferred to give examples with rich facial expressions and gestures instead.

In the G2 class, substitution drills were the most frequent activity. T2 often read a short dialogue after introducing and explaining a new word. After this, she would ask the students to repeat after her again and again until they were familiar with the pronunciations of the new words. During each class, her students were required to answer questions derived from the textbook for meaning and grammar checks.

Compared to T1, T2 provided L2 input that was mainly from the textbook and employed the most frequently used classroom expressions, e.g., “turn to page #,” “please sit down,” “open your book,” “it’s ok,” “quiet,” “try,” again,” “good,” etc. She mixed both Mandarin Chinese and English to give instructions and provide explanations. Both the quantity and the variety of her English were limited and closely constrained to the textbook use.

The data from the classroom observations revealed some interesting dissimilarities between T1 and T2’s language use.  For example, the diversity of language, the amount of English use, the use of comprehension check, the teaching style, classroom management tactics used, students’ practice opportunity, error correction approach, etc. were all very different.


Quantity of Students’ L2 Utterances

Table 3 shows the number and the percentage of students’ use of both Chinese and English in the story-retelling task. As indicated in the table, the bilingual learners used English significantly more than their counterparts. In addition, they retold the story in much more full sentences than their Chinese sentences.


Table 3. Students’ Utterances in Chinese/English







Total # of Utterances



English Utterances

















































































Note: % = the percentage of the English used in the total number of utterances.


Syntactic Development

When it comes to the students’ syntactic development of English, it is obvious that bilingual children’s oral production is generally better than EFL students’. According to Clyne’s framework and the story-retelling task used in the current study, the bilingual children’s oral English had reached Phase 3 or higher. Most of the bilingual children in the study were found to be characterized with the use of L2 syntactic structures and the frequent code-switching to L1 lexical items. For example,


SC: John推a window….John run fast. John take a big pen.


On the other hand, except for SF, most of the Group 2 counterparts’ oral production skills have not yet gone beyond Phase Two, in which some single English words were used in the structure of their L1. Most of them still used one or two words in an English utterance, or formulaic and unanalyzed expressions.

For example, in SF’s, SI’s and SJ’s storytelling, they inlaid English words into their Chinese utterances very often. The following is an example of such an utterance:


SI: 結果John和Mary說,這把key是魔法 key…Mary和John see a key說,這是一把魔法的key。


Such examples show that the children at this stage seem to recognize the L2 expression as a single and individual item, and use it by itself or transfer it as a lexical item in their L1 utterances. As an obvious fact, most of the EFL students cannot even produce L2 at the sentence level. This finding implies that they were not provided with enough practice and language input which is necessary for scaffolding their syntactic development. The reason for saying this is because all the sentences structures required for the task had already been taught in their English class and had been repeated many times.

Our data indicated that surprisingly the bilingual children’s progress in L2 oral production was rather fast, faster than the subjects in Clyne’s study (1986). In Clyne’s (1986) study, the children participants moved slowly across the phases over three years. They can merely reach Phase 2 at the beginning of the second year, and, to reach Phase 3, it would require one to two years. However, the bilingual children in this study have reached beyond Phase 3 at the time we collected the data. This may be the combined results of the high effectiveness and efficiency of the teaching/learning in the bilingual program and of the students’ early beginning of their L2 learning.


Type of the L2 Utterances

In terms of the bilingual learners’ L2 production, the bulk of their utterances in the story-retelling task are formulaic expressions, which include fixed expressions of the story discourse, a repetition of the characters’ names in the story, and the words their teachers use in classroom. It is found that they used these formulaic expressions to convey their own meaning and make their story narration comprehensible, even though there are still many grammatical errors in their production. The formulaic utterances frequently used by them are “sth is…,” “look/looking,” “be take,” “sth/sb + adjective,” “see/take + sth,” “is + verb” and so forth.

The students, moreover, often used “is + verb” to indicate present progressive. In SA’s

task, examples like the following can be found:


SA: And Mary is push the door…and push and push and push.


In contrast, most of the EFL learners, except for SF and SH who had attempted to produce L2, could hardly express themselves at the sentences level. Perhaps due to the lack of English vocabulary and insufficient English proficiency, they tended to speak English in the construction of Chinese and inlaid English words to fill up their utterances in order to finish the narration. Seldom formulaic expressions occurred were found in the transcripts.

Some interesting observations were found in the students’ story narration. For example, three of the bilingual students made the L1-L2 transfer. In the story The Magic Key, the word “a pin” occurred. It is called da-tou-zhen in Chinese. Even though the students know the English word “pin,” they did not say “pin.” Instead, they translated the Chinese term literally back to English and said “big head pin.” The example is as follows:


SB: Jane take a pencil and…Pick up a big head pin.


Such original/creative utterances were not produced in a large amount, yet this type of creativity may serve as an indicator that the students were more or less aware of the basic rules of word formation and syntactic structure of both L1 and L2. Moreover, it implies that the storylines were stored conceptually, (obviously in their L1 because the story was told to them in their L1,) and were retrieved through the concepts instead of the language. These learners were able to spontaneously trying out the learned structures, phrases, and words to carry out the story narration.

In class, some English formulaic speech was also employed by the bilingual children to communicate with their classmates, and the unanalyzed expressions of one or two words were found. For example, “be quite,” “page #,” “I don’t know,” and so on. The fact that the students made often use of the formulaic production spontaneously in the same contexts may suggest that such language use plays an important role in their communication with their teacher and classmates, which meets the pragmatic needs in the bilingual environment (Clyne et al,, 1995; Hagino, 2002).

Children’s use of word segments suggests their syntactic skills are not yet fully developed. Even so, formulaic language appears to play an important role for the young learners to participate the classroom activities and communication discourse.


Relationship between Input and Output

In a manner, the relationship between language input and output cannot be easily observed and measured.

First of all, there is much teacher-student interactions within each group. In G1, T1 spoke only English to her students, while the students used mostly Chinese to interact with their teacher. In G2, except in the drills and practice activities, Chinese was used much more frequently than English when the teacher and the students made meaningful communications with one another. Although teaching techniques many have influenced language use and discourse in the classroom, the EFL class seems not to promote L2 acquisition effectively and efficiently. Three of the EFL subjects could barely produce L2 utterances, (except SJ and SF.) Some of them (SG, SI, and SJ) even have difficulty saying a full English sentence. The classroom observation indicated that the words and sentence structures that are necessary to be used for the story narration had already been taught and intensively drilled in the previous lessons, yet it seems that they did not turn into the students’ intake, which can be used in their later language performance.

For another, the teacher’s massive use of L1 in the classroom may have several influences of the students’ acquisition of L2. On the one hand, the quantity and quality of the L2 input are far from sufficient for the learners’ L2 to develop. On the other, the teacher has served as the role model that L2 is not supposed to be used for the communicative function in the classroom.

T2’ repetition drills did not seem to reveal their effect in the learners’ output. This may respond to Lightbown’s argument which says,

“Practice does not make perfect. Even though there are acquisition sequences, acquisition is not simply linear or cumulative, and having practiced a particular form or pattern does not mean that the form or pattern is permanently established. Learners appear to forget forms and structures which they had seemed previously mastered and which they had extensively practiced (1985, pp. 177).”

Obviously the teaching techniques should be further assessed and examined cross-linguistically

On the other hand, totally speaking, the children from G1 did a better job in retelling the story. Most of the pupils’ utterances were target-like, though the students inevitably showed difficulties with some grammatical items, such as, subject-verb agreement, present progressive tense, articles, and individual word usage, etc. It indicates certain connections exist between the learners’ language production and their input.



By using classroom observation and a story-retelling task, the present study attempted to examine the relationship between the effect of the teacher’s input and the students’ output by comparing two different groups of learners.

In answer to the first research question which inquires whether NNS and NS English teachers give different language input, the data collected in the study have suggested that there are significant differences between the two kinds of input. The NS English teacher provided much better input for their students in both quantity and quality; whereas, the NNS English teacher provided little L2 input, which is necessary for the students’ L2 development. When teaching and interacting with the students, the NS teacher in the bilingual school spoke only the target language, i.e., English, no matter what language her students used to communicate with her. On the other hand, the NNS teacher used mostly mandarin Chinese to make meaningful communication. Although she gave students a large number of English language drills and practices, her language use hardly went beyond the scope of the textbook. As a result, from the evidence found in the study, it can be concluded that the language input provided by T1 and T2 for their students is quite different in both quality and quantity.

With respect to the students’ language production, both the quantity (i.e., the amount of the English utterances) and the quality (i.e., the formulaic speech) of Group 1 learners are far better than those of the learners in Group 2. The findings from the study also indicate that the students’ output did show obvious effects from their input although the students’ output may not be so straightforwardly contributed to the teachers’ input alone.

Regarding the third research question, inquiring the bilingual and the EFL language learners’ formulaic speech and creative expressions, the data indicated that most of the EFL learners could not produce L2 at the sentence level. They tended to use L1 syntactic structures with limited L2 vocabulary. Seldom formulaic expressions were found in their narrative production. On the other hand, most of the bilingual students were able to use L2 sentence structure and vocabulary to tell the story.



The current study only examined a few learners in different L2 learning settings.  Much more similar studies have to be conducted to further consolidate the findings in the study. The current study may have simplified the relationship between input and output by downplaying the role of other L2 acquisition processes, such as interactions and output but this is not our intention. The findings did show an obvious relationship exists between the two variables although other variables may also have played a role.


Pedagogical Implications

Some pedagogical implications can be drawn from the findings of the present study.

In the first place, in order to facilitate students’ L2 development, teachers are suggested to improve English instruction in terms of providing sufficient input in both quality and quantity. The teacher’s continuous language input and interaction with her/his students may play a crucial role in developing students’ L2 proficiency. Many researchers of ESL/EFL fields (e.g., Ellis, 1994; Ellis, Tanak & Yamazaki, 1994; Hagino, 2002; Henry, 1996; Huttenlocher, 1998; Krashen, 2002a; Pica, 1983, etc.) have pointed out that if a teacher can continuously call the children by their names, praise them, courage them, ask them questions, and provide them sufficient input, the children can be more spontaneous in trying out the target language and producing L2 utterances.

Secondly, more classroom activities, which can activate students’ involvement and induce meaningful communications by using their limited L2 knowledge and skills should be encouraged for their L2 development. However, this can only be achieved when the L2 teacher is able and willing to use more L2 in their instruction.



Brown, R. (1973). A first language: The early stages. Cambridge, Massachusetts: Harvard University Press.

Clyne, M., Jenkins, C., Chen, I. Y., Tsokalidou, R., & Wallner, T. (1995). Developing second language from primary school: Models and outcomes. Canberra, Australia: National Languages and Literacy Institute of Australia.

Ellis, R. (1994). The study of second language acquisition. Oxford, New York: Oxford University Press.

Ellis, R., Tanaka, Y., & Yamazaki, A. (1994). Classroom interaction, comprehension and the acquisition of L2 word meanings. Language Learning, 44, 449-491.

Hagino, S. (2002, June). Young children’s L2 oral production in Japanese immersion classrooms at a Victorian primary school. ASAA E-Journal of Asian Linguistics and Language Teaching, 2, Retrieved November 19, 2003, from http://www.arts.unsw.edu.au/languages/ asaa_ejournal

Henry, A. (1996). Natural chunks of language: Teaching speech though speech. English for Specific Purposes, 15(5), 295-309.

Huttenlocher, J. (1998). Language input and language growth. Preventive Medicine, 27, 195-199.

Krashen, S. D. (1980). The input hypothesis. In J. Alatis (Ed.), Current Issues in Bilingual Education. Washington, DC: Georgetown University Press.

Krashen, S. D. (2002a). Explorations in language acquisition and use: The Taipei Lectures. Taipei: Crane Publishing Co., Ltd.

Krashen, S. D. (2002b). Foreign language education: The easy way. Taipei: Crane Publishing Co., Ltd.

Lightbown, P. M. (1985). Great expectations: Second-language acquisition research and classroom teaching. Applied Linguistics, 6, 173-189.

Myles, F., Hooper, J., & Mitchell, R. (1998). Rote of rule? Exploring the role of formulaic language in classroom foreign language learning. Language Learning, 48(3), 323-363.

Peters, A., & Menn, L. (1993). False starts and filler syllables: Ways to learn grammatical morphemes. Language, 69, 742-778.

Peters. A. M. (1983). The units of language acquisition. Cambridge: Cambridge University Press.

Pica, T., Young, C., & Doughty, C. (1987). The impact of interaction on comprehension. TESOL Quarterly, 21, 737-758.

Ritchie, W. C., & Bhatia, T. K. (Eds.) (1996). Handbook of second language acquisition. San Diego, California: Academic Press, Inc.

Spolsky, B. (1989). Conditions for second language learning: Introduction to a general theory. Walton Street, Oxford: Oxford University Press.

Streven, P. D. (1986). Learning English better through more effective teaching: Six postulates for a model of language learning/teaching. World Englishes, 7(1), 51-61.

創作者 HuangCT 的頭像


HuangCT 發表在 痞客邦 留言(0) 人氣()