Conceptions of language and language learning

Conceptions of language and language learning

System 27 (1999) 459±472 www.elsevier.com/locate/system Conceptions of language and language learning Phil. Benson*, Winnie. Lor English Centre, Hon...

132KB Sizes 1 Downloads 115 Views

System 27 (1999) 459±472

www.elsevier.com/locate/system

Conceptions of language and language learning Phil. Benson*, Winnie. Lor English Centre, Hong Kong University, Pokfulam Road, Hong Kong

Abstract This paper questions whether the notion of learner beliefs as conceived in the SLA literature is adequate to capture the complexity of learners' thinking about language learning. It proposes as an alternative, an analytical framework based on three levels: conception, approach and belief. The notion of conceptions of language and language learning is proposed as higher level category conditioning speci®c beliefs. The notion of approaches to learning is proposed as a category describing the level at which beliefs are made manifest in speci®c contexts of learning. These categories are illustrated with reference to the authors' data drawn from interviews with university students in Hong Kong. # 1999 Elsevier Science Ltd. All rights reserved. Keywords: Learner beliefs; Conceptions of learning; Approaches to learning

1. Introduction Research on learner beliefs about language learning is based on the cognitivist assumption that learning attitudes and behaviours are conditioned by a higher order of mental representations concerning the nature of language and language learning. If learners believe that the best way to learn a foreign language is to memorise its component parts, it seems likely that they will hold positive attitudes towards vocabulary and grammar learning and that they will be predisposed to adopt a range of strategies involving analysis, memorisation and practice. If learners believe that the best way to learn a foreign language is to absorb it in natural contexts of use, it is likely that they will hold positive attitudes towards communication with speakers of the language and that they will be predisposed to adopt a range of social and communication strategies. Although it is generally acknowledged that learners can learn equally well by following their own preferences and styles, it is also assumed that certain attitudes and behaviours may be more enabling than others. The practical consequence of these assumptions is that, if language teachers wish to *Corresponding author. 0346-251X/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S0346-251X(99)00045-7

460

P. Benson, W. Lor / System 27 (1999) 459±472

in¯uence learners' attitudes and behaviours, they will need to address the underlying beliefs on which they are based. The typical research strategy in the ®eld of learner beliefs is to talk to learners about language learning in interviews or focus group discussions and analyse what they say (e.g. Wenden, 1986, 1987; Benson and Lor, 1998). Sometimes, learners state their beliefs explicitly. In our interview data, gathered from ®rst-year undergraduate students at the University of Hong Kong, we ®nd the following, ``Whenever we decide to learn a language, we must put in e€ort''. This learner explicitly states his belief that e€ort is important in language learning. Learners also express their beliefs less directly. In the same data we also ®nd, ``The teacher should not always give many things. We should ®nd them ourselves''. In this case, the learner is making a statement about how teachers should teach, but it is evident that she also believes that self-motivation is important in language learning. A second-order research strategy is to compile an inventory of possible statements of belief and ask learners whether they agree with them or not (e.g. Horwitz, 1987, 1988; Cotterall, 1995). This strategy forces subjects to consider beliefs that may not come up in the course of an interview or discussion, but it may also lead them to agree with beliefs that they do not actually hold, or that have no practical consequences for them. Learner beliefs have also proved dicult to identify and classify in any systematic way. In her interview-based study of adult ESL learners in North America, Wenden (1986, 1987) identi®ed and grouped beliefs about the best way to learn foreign languages and established that there was some correlation between groupings of beliefs and preferences for learning strategies. In a study of the beliefs of ®rst-year university undergraduates in Hong Kong, however, Benson and Lor (1998) identi®ed di€erent beliefs and grouped them di€erently. These di€erences may be due to the cultural contexts of the learners or to di€erent approaches to the data by the researchers. Nevertheless, they lead us to ask whether there is any principled way of identifying and grouping learner beliefs about language learning. Research has also tended to see beliefs as being enabling or disabling independently of the contexts in which they are held. Questionnaire research, in particular, tends to give a snapshot of a learner's beliefs without telling us much about their functions or the ways in which they are open to change. This leads us to ask whether the notion of belief as conceived in the SLA literature is adequate to capture the complexity of learners' thinking about language learning, In this paper, we attempt to tackle these questions on the basis of literature from the ®eld of educational psychology concerned with approaches to and conceptions of learning. We are principally interested in exploring whether or not we can identify a higher order of conceptions of language and language learning that can be said to condition beliefs and whether the notion of approaches to learning can help us to understand the functions of beliefs in context. In the ®rst part of the paper we outline what we understand by approaches to and conceptions of learning and in the second part we use our data to illustrate some of the ways in which these conceptions and approaches appear to work in the context of language learning.

P. Benson, W. Lor / System 27 (1999) 459±472

461

2. Conception and approach Research on conceptions of learning originates in the ®eld of educational psychology within a tradition known as `Student Approaches to Learning' (SAL) (Watkins, 1996), SAL theory de®nes itself as an alternative to information processing (IP) theory, which is said to propose a set of ideas about learning that apply regardless of its content and context. In contrast, SAL theory begins from the perspective of the learner and recognises the crucial importance of the content of the learning task and its context. The emphasis on the learner's perspective is re¯ected in the view that ``learning should be seen as a qualitative change in a person's way of seeing, experiencing, understanding, conceptualising something in the real world'' (Marton and Ramsden, 1988, p. 271) rather than as a purely quantitative addition to the learner's existing store of knowledge. In an often-cited experiment, Marton and SaÈljoÈ (1976) asked students to read a passage and then interviewed them to ®nd out what they had learned and how they had approached the learning task. Responses were coded into four categories of content and two categories of processing: surface-level and deep-level. In surfacelevel processing, learners direct attention towards ``learning the text itself (the sign)'' while in deep-level processing attention is directed towards ``the intentional content of the learning material (what is signi®ed)'' (p. 7). Using the same data, but focusing on learning outcomes, Svensson (1977) made a similar distinction between `atomistic' and `holistic' cognitive approaches. The terms `surface approach' and `deep approach' have now become standard in the literature and Biggs (1987) has also added an `achieving approach', in which students make use of surface and deep approaches as needed in order to achieve higher grades. Biggs (1987, 1992) has developed two questionnaire instruments to assess approaches to learning: the Learning Process Questionnaire (LPQ) for secondary level and the Student Process Questionnaire (SPQ) for tertiary level. Biggs (1993, p. 6) observes that the term `approaches to learning' has come to have two di€erent meanings: (1) ``the processes adopted prior to, and which directly determine, the outcome of learning'', studied by Marton and SaÈljoÈ (1976), and (2) ``predispositions to adopt particular processes'', studied through instruments such as the LPQ and SPQ, which ask students how they usually go about learning. Biggs (1993, p. 10) also observes that: A predisposition to this or that learning approach is the individual student's way of achieving balance in the system as perceived by the student. Given an individual's goals, self-perceptions as to ability, the mode of teaching and assessment, the outcome, and the student's attributions for that outcome, so the student will after exposure to a particular teaching/learning environment ®nd a certain approach to be viable and personally comfortable in day-to-day coping with that environment, and thus be predisposed to use deep or surface strategies for particular tasks in that context. This observation has important consequences for our understanding of the status of learner beliefs. Many of the items in the LPQ and SPQ re¯ect what we might call

462

P. Benson, W. Lor / System 27 (1999) 459±472

beliefs about learning. However, Biggs's comment suggests that such beliefs are always contextualised in relation to some learning task or situation. The beliefs articulated by students are not necessarily held to under all circumstances. Rather they can be understood as cognitive resources on which students draw to make sense of and cope with speci®c content and contexts of learning. In recent research, Marton and his associates have shifted their focus to the description of conceptions of learning (Marton et al., 1993) using a methodology they call phenomenography (Marton, 1981; SaÈljoÈ, 1988). Phenomenography aims to investigate the ways in which people construe or conceive of phenomena and is based on the assumption that subjective interpretations of reality are more important in analysing actions than any underlying objective reality. According to Marton (1981, p. 180), phenomenography does not necessarily require that we make phenomenological assumptions about the nature of reality, merely that we focus attention on the `second-order perspective' of the ways in which people experience reality rather than on the `®rst-order perspective' of the reality itself. SaÈljoÈ (1988, p. 42) places phenomenography within the tradition of qualitative research: The concrete praxis of phenomenography implies that the variation in forms of talking about phenomena is reduced to a limited set of categories (usually between three and ®ve) that depicts signi®cant di€erences in ways of construing this phenomenon. The assumption is that conceptions of reality can be expressed in a large number of linguistic forms without necessarily changing the basic way in which the phenomenon is construed. In order to investigate conceptions of phenomena, subjects are interviewed to probe the ways in which they understand them. Transcripts are read repeatedly until categories begin to emerge following a methodology similar to that used in grounded theory (Glaser and Strauss, 1967). The transcripts are coded and theory is developed on the basis of interrelationships between categories identi®ed. Typically, the categories and their interrelationships are the outcome of phenomenographic research. Phenomenography also makes two key assumptions: 1. Conceptions do not reside within individuals. Although people tend to use particular conceptions of reality in certain settings or in relation to certain problems, they cannot be assumed to hold these conceptions under all circumstances. 2. Conceptions are relational rather than inherent qualities in the mind of the thinker or in objects themselves (SaÈljoÈ, 1988). Individual interviewees may, therefore, express fragments of more than one conception, and the conceptions they express will not necessarily be consistent over time. The objective of phenomenographic research is, however, not so much to understand individuals in terms of the conceptions they hold, as to understand the nature of the conceptions themselves. Much of the research in the ®eld of conceptions of learning has been concerned with conceptions of scienti®c theories, but Marton and his associates have recently

P. Benson, W. Lor / System 27 (1999) 459±472

463

turned their attention to conceptions of the phenomenon of learning itself (Marton et al., 1993, 1996, 1997; Purdie et al., 1996). Summarising conclusions from earlier studies, Marton et al. (1997) refer to six distinct conceptions of learning as: A. an increase in knowledge; B. memorising; C. acquiring facts, procedures, etc., which can be retained and utilised in practice; D. the abstraction of meaning (understanding); E. an interpretative process aimed at the understanding of reality (seeing something in a di€erent way); and F. a change in the person. These six conceptions are assumed to be ordered hierarchically in relation to approaches to learning. Conceptions A, B and C are described as quantitative and are associated with surface approaches to learning. Conceptions D, E and F are described as qualitative and are associated with deep approaches to learning. Watkins (1996, p. 6) makes a strong claim for this relationship, stating that: . . .there is clear evidence that conception, approach, and outcome are linked by a chain of functional relationships. It seems that students who are only capable of conceiving a quantitative conception of learning only achieve a surface approach to learning, and that awareness of a qualitative conception of learning is a necessary, but not sucient condition for the adoption of a deep-level approach. In Marton et al. (1997), the model of six conceptions of learning has been revised and now takes the form of a matrix based on the distinction between the temporal dimension of learning and a dimension of depth characterised by four ways of experiencing learning (Fig. 1). The distinctions between ways of experiencing learning represented in this matrix are based on distinctions related to the subject, act and object of learning. In relation to the subject, the important distinction is whether the teacher or student is the agent of learning. Ways of experiencing learning involving understanding are student based by de®nition, whereas ways of experiencing learning involving memorisation are more likely to be teacher based. In relation to the act of learning, the important distinction is between the intention to commit to memory and the intention to understand. The di€erence between memorising (meaning) and understanding (meaning) is, therefore, a di€erence of emphasis (understanding in order to remember or understanding for its own sake). The important distinctions in relation to the object of learning are between the words of the text (the sign) and the meaning of the text (the signi®ed) and between the meaning of the text and meaning of the phenomenon with which it is concerned. Each of the four ways of experiencing learning can also be characterised in terms of `acquiring', `knowing' and `making use of ' phases.

464

P. Benson, W. Lor / System 27 (1999) 459±472

Fig. 1. The outcome space of learning (from Marton, Watkins & Tang, 1997).

Marton et al. (1997) note that this two-dimensional model is very di€erent from the model of six conceptions of learning proposed in earlier work. They also note that the model, based on interviews with Hong Kong secondary-school students, does not include conceptions of learning as seeing things di€erently (E) and personal change (F). Despite these di€erences, the important distinction between quantitative (committing signs to memory) and qualitative (understanding the meaning of signs) conceptions of learning remains, together with their association with surface and deep approaches to learning. On the basis of the research reviewed so far, we can draw three broad conclusions regarding the nature and function of learner beliefs. First, it is helpful to distinguish between two levels of representation in learners' thinking about their learning: conception, and belief. Conceptions of learning are concerned with what the learner thinks the objects and processes of learning are, whereas beliefs as conceived in the SLA literature are concerned with what the learner holds to be true about these objects and processes, given a certain conception of what they are. Second, conceptions of learning characterise learners' thinking at a higher level of abstraction than beliefs. Whereas beliefs can be inferred more or less directly from data, conceptions of learning call for a further level of analysis. Conceptions of learning are also organised hierarchically and can be characterised in terms of a basic distinction between quantitative and qualitative. Lastly, conceptions and beliefs are understood as relational and responsive to context. In other words, beliefs are made manifest in approaches to learning, which can also be analysed in terms of a quantitative/qualitative distinction. Approaches to learning are functional in a given context of learning and are constrained by the range of conceptions available to the learner at a given moment in time.

P. Benson, W. Lor / System 27 (1999) 459±472

465

In the context of foreign language learning, conceptions of learning can be understood as conceptions of what a foreign language is and of what the process of learning a foreign language consists of. In the remainder of this paper, we will try to show on the basis of our data that beliefs about language learning inferred from learner talk can best be categorised in terms of basic distinctions within these conceptions and that the enabling or disabling character of beliefs can best be understood in terms of approaches to learning. 3. Learner beliefs and conceptions of language learning The data that we discuss in this section were collected from a series of interviews conducted with 16 ®rst-year Arts Faculty undergraduates at the University of Hong Kong during the academic year 1996±97 to evaluate students' responses to a programme designed to encourage independent learning. The interviews were conducted in the students' ®rst language, Cantonese, and extracts have been translated into English. Although the data sample is small, we believe that the range of beliefs expressed by these 16 students is broadly representative of the wider group from which they are drawn. In our data, we identi®ed three broad domains of belief: beliefs about language learning, beliefs about self and beliefs about the learning situation. Beliefs about language learning are expressed in prescriptive statements about what learners should do in order to learn a foreign language well or about the conditions that must be present for good learning to take place. In contrast, beliefs about self and the learning situation are expressed as statements that are speci®c to the current internal and external context of learning. Although we may assume that beliefs about language learning are also constrained by the context of learning, they are generally expressed as general truths about the process of learning a foreign language. Within the domain of beliefs about language learning, 14 discrete beliefs came up repeatedly in the data. Although some beliefs were expressed more frequently than others, we were unable to identify any distinctive pattern of beliefs characterising the group as a whole. Rather, the 14 beliefs can be considered as a set available to the group as a whole, on which each individual has the potential to draw. These 14 beliefs are listed in Fig. 2, where they are categorised under three major headings: work, method and motivation. In Fig. 2, beliefs about language learning are represented by statements using the impersonal second-person pronoun and a modal auxiliary of necessity. In the data itself, statements of belief were generally signalled by the equivalent Cantonese grammatical form, although ®rst-person plural pronouns and passive constructions were also often used. The three broad categories of belief (work, method and motivation) can also be expressed as explicit statements: ``you need to work hard'', ``you need to have a good method of learning'' and ``you need to have a source of motivation''. However, we also observed that a belief in the importance of work tended to invade the other two categories. A good method of learning was often the one

466

P. Benson, W. Lor / System 27 (1999) 459±472

Fig. 2. Learner beliefs about language learning in the authors' data.

that helped the learner to organise work while a good source of motivation was often the one that encouraged the learner to work hard. This emphasis on work is not unexpected and ®ts in with research which shows that Hong Kong learners tend to associate success in learning with e€ort and application (for a review, see Salili, 1996). Having identi®ed and classi®ed discrete beliefs characteristic of this group of learners, however, understanding how these beliefs might be ordered in relation to each other and related to more speci®c beliefs about self and situational context remained a problem. We also observed that the variety of beliefs identi®ed in the data obscured a rather crucial distinction apparent in the two extracts below: When we decide to learn, we should devote time to constructing our foundations like having clear concepts of grammar, word patterns, etc. During the learning process, we have to collect, absorb and assimilate all the knowledge that we get too. Studying language, is based on your attention in your daily living. It's a kind of sense. If you've got that kind of sense, no matter who speaks a sentence, or if you notice any word you don't know in the street, you will take the initiative in learning it. These two extracts represent two very di€erent beliefs about language learning. The ®rst extract illustrates the belief that ``you have to build a good foundation'' and the second illustrates the belief that ``you have to pay attention to language in use''. The two extracts also appear to represent two very di€erent ways of construing language as an object of learning. The ®rst extract conveys a clear conception of

P. Benson, W. Lor / System 27 (1999) 459±472

467

language as a collection of things (grammatical concepts, word patterns, etc.) for the learner to ``collect, absorb and assimilate''. The second conveys an equally clear conception of language as an environment to which the learner must be responsive in order to learn. Di€erent beliefs about language learning methods thus appear to be constrained by di€erent conceptions of the object and process of learning. In the ®rst extract, language as the object of learning is conceptualised quantitatively as a collection of things to be accumulated, and in the second extract it is conceptualised qualitatively as an environment to which the learner responds. In the ®rst extract, language learning is conceptualised quantitatively as a process of collection, absorption and assimilation, and in the second it is conceptualised qualitatively as a process of making sense of an unfamiliar environment. The belief that work is the most important factor in language learning does not necessarily discriminate between quantitative and qualitative conceptions of learning, although it seems to be more strongly associated with quantitative conceptions in our data. Looking at the category of beliefs associated with method, three itemsÐ ``you need to work with a teacher'', ``you need to build a good foundation'' and ``you need to pay attention to all aspects of the language''Ðappear to be consistent with a conception of language as a collection of things. The other three items in this categoryÐ``you have to pay attention to your needs'', ``you have to pay attention to language use'' and ``you have to expose yourself to the language''Ðand are more consistent, with a conception of language learning as making sense of an unfamiliar environment. Four of the ®ve beliefs about motivation in Fig. 2 can be associated with qualitative conceptions of learningÐ``you need to be in an environment that forces you to use the language'', ``you need to be self-motivated'', ``you need to gain a sense of self-satisfaction from learning'' and ``you need to follow your own interests''. The basic distinction between quantitative and qualitative conceptions of language and language learning, therefore, gives us a framework for classifying the beliefs listed in Fig. 2. We also need to bear in mind, however, that beliefs expressed as general truths about language learning also occur within speci®c contexts of learning. In this light, it is interesting to observe that the quantitative/qualitative distinction also appears to be fundamental to speci®c beliefs about self and the learning situation in our data. Looking at beliefs about self in our data, we ®nd, for example, characteristically quantitative and qualitative ways of talking about one's own pro®ciency in the foreign language: Erm, for general communication purpose, [my English is] okay. But I think there's a lot to learn in English. It's not only for simple communication. So many things that you may not know. That's why I suggest learning more new things. Knowing how to speak, so many other people know how to speak. But there's a lot to learn. Even if my English is an A, still mine is Hong Kong English. I still cannot be like those people [overseas-born Chinese]. So I become very sensitive. That's the kind of English I want.

468

P. Benson, W. Lor / System 27 (1999) 459±472

In the ®rst extract, the learner's approach to self-assessing pro®ciency is conditioned by the idea of language as a collection of things to be learned. In the second extract, however, the learner is less concerned with the things that she knows than with the quality of her English in comparison to that of a group she has taken as her target. Looking at beliefs about the learning situation, we ®nd that the learners in our study orient towards a shared experience of language learning within the Hong Kong educational system. All of the subjects have passed through English-medium secondary education, in which a great deal of emphasis is placed on English language skills, and have received more than 2000 hours of English language lessons before entering the university. All have reached a sucient standard in English language examinations to enter university, a standard which less than 60% of candidates achieve. Although they have been relatively successful in learning English, the learners in our study commonly construed the context of their learning as one in which little further learning could be expected: I just think that it's at a stop now. I cannot absorb anything. In the past I absorbed everything. The things that I can gain at this stage are not many. Therefore, my interest in learning English isn't very much. But it's like a curve. At ®rst, you might like it very very much but then later, if too much you may not like it. Bear with me if I put it this way. So many years, more than ten years of learning English, if some more, I can't stand it. Such interpretations of the context are understandable when it is kept in mind that the English examination that determines whether students are able to enter university or not is likely to be the last major linguistic hurdle they face in their educational career. In this situation, many university students appear to see further study of English as a `burden', particularly when the other diculties of organising their study and social lives are taken into account. However, we also observe that the metaphor of English as a burden tends to be supported by a quantitative assumption of a declining rate of return on e€ort: the more you know, the less you are able to learn for a given amount of e€ort. Although this assumption is counterintuitive from the standpoint of qualitative conceptions of learning (where the emphasis is on understanding and personal change), it is rational as long as the learner holds to a conception of language as consisting of a de®nite number of items to be learned. Interestingly, the few students who did not regard further study of English as a burden were also those who expressed qualitative conceptions of language and language learning. The following extract links the two explicitly: If I can follow my interest to learn, I don't think it's a burden. It's something we learn naturally, unconsciously.

P. Benson, W. Lor / System 27 (1999) 459±472

469

The preference for `natural' learning expressed in this extract was shared by several of the interviewees, for example: Because I don't like people pushing me, I don't like following a schedule. I think erm, learning English, Chinese, languages, there's no speci®ed way to learn. It's natural that you need to hand in homework if you study. You need to read books if you need to do homework, digesting it and writing it. Or you need to present your work. In fact in this process, we really have learnt English. It isn't really reading books, listening to the radio, reading newspaper. Learning English is not like a formula. Other students prefer to be `pushed': Erm, in fact, if, or maybe people need some external force to push them so that they can do something. For me, it's true of me. For learning, I think, there should be some external things. Like the Independent Learning Project, you are asked to hand in a report, or to talk about the progress. Anyway, you need to hand in something to some people. Perhaps, for me, I can learn something. If you ask me to take initiative, self-learning, I don't think I can do it. Although the issue here is essentially one of a preference for the mode of direction, it is again interesting to observe that the two extracts involve very di€erent ways of conceptualising language and language learning. To the student in the ®rst extract, it does not make sense to be pushed or follow a schedule because she conceptualises the object of her learning in qualitative terms. For the student in the second extract, being pushed does make sense because the object of learning is conceptualised quantitatively in terms of things to be `learned' and `handed in'. The literature on approaches to learning and conceptions of learning strongly implies that qualitative approaches and conceptions are in some sense `higher' than quantitative approaches and conceptions. However, we can also understand conceptions and approaches in terms of their functionality in a given context of learning. In other words, a `lower' conception or approach may be more functional than a `higher' one in a context that supports it. In the Hong Kong context, quantitative conceptions and approaches may be functional within an examination-oriented secondary education system in which language is presented to learners as a collection of items to be learned. Qualitative conceptions and approaches are more likely to be functional where the context is one in which learners are readily able to immerse themselves in target language use. The learners interviewed in our study predominantly expressed quantitative conceptions of language and language learning while only a few expressed qualitative conceptions. A number shifted from one to the other depending on the area of learning they were talking about. One student, for example, expressed qualitative conceptions of learning when talking about learning English from movies, but reverted to quantitative conceptions when talking about vocabulary learning. The speci®c problem that many ®rst-year undergraduates in Hong Kong face in relation

470

P. Benson, W. Lor / System 27 (1999) 459±472

to English, however, is the need to make a transition from a context in which English is learned as a subject primarily for the purposes of passing examinations to one in which it is used as a means of communication within educational and professional settings. We are, therefore, able to observe that, in this speci®c context, quantitative conceptions of language learning tend to be dysfunctional. Indeed, those learners interviewed who seemed strongly committed to quantitative conceptions of language learning seemed particularly prone to a sense of demoralisation arising from the feeling of a declining return on e€ort and the feeling that English has become a burden to them. One response to this sense of demoralisation, however, appears to be for learners to shift away from e€ort-based, quantitative conceptions of language learning towards exposure-based, qualitative conceptions. Qualitative conceptions often took the form of statements such as ``English is a language which has no speci®ed curriculum'', ``studying language is a kind of sense'' and ``there is no formula to improve''. The belief expressed by some students that English should not be learned `deliberately' may well be a strategic response to a conception of the context of learning as one in which quantitative approaches to language learning are no longer functional. This observation leads us to the conclusion that the value of research on learner beliefs may lie not so much in an understanding of the enabling or disabling attributes of beliefs as in an understanding of the ways in which learners put their beliefs to use. Two examples from our data will illustrate this. The ®rst is taken from a transcription of a classroom discussion in English among four students, of whom J and C are currently holding the ¯oor. J: C: J: C: J:

Um, but watching, watching tapes is just like watching, but is just like, quite entertainment. Do you think we can learn di€erent things from entertainment? Yes, yes, but it's quite strange to me and er Do you like to see Western ®lms? Well of course. But I'm afraid I'll learn nothing from it because I, I enjoy watching too much.

In this extract, J and C are discussing the value of watching movies for language learning and it is interesting to observe how the two positions taken on the issue are rooted in quantitative and qualitative conceptions of learning. J appears to have diculty reconciling the idea of enjoyment and learning because he expects to learn `something' in a learning activity, while C's questions suggest that she is able to view entertainment as a form of learning without worrying too much about what exactly is learned. The point of interest here is, however, that although J is not convinced of the value of watching ®lms, he is open to a dialogue which may lead him to shift towards a more qualitative conception of learning, In the second extract, the learner counterposes methods of learning based on quantitative and qualitative conceptions in a kind of interior monologue: For example, when I learn English, now I won't deliberately open a grammar book and memorise all the things. On the contrary, when I am free, I may read

P. Benson, W. Lor / System 27 (1999) 459±472

471

the South China Morning Post to `see' the English, such as `bid a piece of land' means put in a tender for land. But you won't deliberately look up `bid' in the dictionary and then memorise all its meanings. It's a lively method. You see the things around you. Whenever you see something you don't know, you will be sensitive to remember it and you will ask others. In this extract, the learner shows evidence of her own transition from quantitative to qualitative conceptions by describing not only what she does (a method based on sensitivity to the linguistic environment), but also what she does not do (open the grammar book and memorise). Dialogue (with others and with oneself) is thus one of the ways in which learners appear to be able to manage shifts from one conception of language and language to another. This dialogue occurs, however, at the level of approach, where conceptions and beliefs are contextualised within concrete experiences and situations of learning. 4. Conclusion Our aim in this paper has been to suggest that we can make better sense of the complex area of learner beliefs about language learning if we consider three levels of analysis: conception, belief and approach. Conception constitutes a higher and more abstract order of representation that constrains beliefs. The notion of conceptions of learning is of value in helping us to classify beliefs inferred directly from data hierarchically. Approach constitutes the level at which conceptions and beliefs are made manifest. The notion of approaches to learning is of value in helping us to understand the functionality of conceptions and beliefs and the ways in which they may be open to change. From a practical point of view, teachers need to know not only what their learners believe about language learning, but also whether their beliefs are functional or dysfunctional and how dysfunctional beliefs can be modi®ed. Our conclusions in this regard are tentative, but we would suggest that in order to modify beliefs, the learner must also modify the underlying conceptions on which they are based and pay attention to the context in which they function.

References Benson, P., Lor, W., 1998. Making Sense of Autonomous Language Learning: Conceptions of Learning and Readiness for Autonomy (English Centre Monograph No. 2). University of Hong Kong. Biggs, J.B., 1987. Student Approaches to Learning and Studying. Australian Council for Educational Research, Hawthorn, Victoria. Biggs, J.B., 1992. Why and How do Hong Kong Students Learn? Using the Learning and Study Process Questionnaires (Education Paper No. 14). Faculty of Education, University of Hong Kong. Biggs, J.B., 1993. What do inventories of students' learning processes really measure? A theoretical review and clari®cation. British Journal of Educational Psychology 63, 3±19. Cotterall, S., 1995. Readiness for autonomy: investigating learner beliefs. System 23 (2), 195±206.

472

P. Benson, W. Lor / System 27 (1999) 459±472

Glaser, B.G., Strauss, A.L., 1967. The Discovery of Grounded Theory: Strategies for Qualitative Research. Aldine, Chicago. Horwitz, E.K., 1987. Surveying student beliefs about language learning. In: Wenden, A., Rubin, J. (Eds.), Learner Strategies in Language Learning. Prentice Hall, London, pp. 119±129. Horwitz, E.K., 1988. The beliefs about language learning of beginning university foreign language students. The Modern Language Journal 72 (3), 283±294. Marton, F., 1981. PhenomenographyÐdescribing conceptions of the world around us. Instructional Science 10, 177±200. Marton, F., Ramsden, P., 1988. What does it take to improve learning? In: Ramsden, P. (Ed.), Improving Learning: New Perspectives. Kogan Page, London Marton, F., SaÈljoÈ, R., 1976. On qualitative di€erences in learning: IÐOutcome and process. British Journal of Educational Psychology 46, 4±11. Marton, F., Dall'Alba, G., Beaty, E., 1993. Conceptions of learning. International Journal of Educational Research 19, 277±300. Marton, F., Dall'Alba, G., Tse, L.K., 1996. Memorizing and understanding: the keys to the paradox? In: Watkins, D.A., Biggs, J.B. (Eds.), The Chinese Learner: Cultural, Psychological and Contextual In¯uences. Comparative Education Research Centre, Hong Kong/Australian Council for Educational Research, Melbourne, Australia, pp. 69±84. Marton, F., Watkins, D., Tang, C., 1997. Discontinuites and continuities in the experience of learning: an interview study of high-school students in Hong Kong. Learning and Instruction 7 (1), 21±48. Purdie, N., Hattie, J., Douglas, G., 1996. Student conceptions of learning and their use of self-regulated learning strategies: a cross-cultural comparison. Journal of Educational Psychology 88 (1), 87±100. Salili, F., 1996. Accepting personal responsibility for learning. In: Watkins, D.A., Biggs, J.B. (Eds.), The Chinese Learner: Cultural, Psychological and Contextual In¯uences. Comparative Education Research Centre, Hong Kong/Australian Council for Educational Research. Melbourne, Australia, pp. 85±106. SaÈljoÈ, R., 1988. Learning in educational settings: methods of inquiry. In: Ramsden, P. (Ed.), Improving Learning: New Perspectives. Kogan Page, London, pp. 32±48. Svensson, L., 1977. On qualitative di€erences in learning: IIIÐStudy skill and learning. British Journal of Educational Psychology 47, 233±243. Watkins, D., 1996. Learning theories and approaches to research: a cross-cultural perspective. In: Watkins, D.A., Biggs, J.B. (Eds.), The Chinese Learner: Cultural, Psychological and Contextual In¯uences. Comparative Education Research Centre. Hong Kong/Australian Council for Educational Research, Melbourne, Australia, pp. 3±24. Wenden, A., 1986. What do second language learners know about their second language learning: a second look at retrospective accounts. Applied Linguistics 7 (2), 186±201. Wenden, A., 1987. How to be a successful language learner: insights and prescriptions from L2 learners. In: Wenden, A., Rubin, J. (Eds.), Learner Strategies in Language Learning. Prentice Hall, London, pp. 103±114.