It is my intention in this brief paper to present, perhaps in something of a provocative way, some of the difficulties which I perceive with the current state of foreign/second language teaching and therefore language-teacher education in particular the predominance of the linguistic paradigm.
 
 

The linguistic paradigm

It does not seem unreasonable to assert that for a very long time, language-learning and teaching has been dominated by what one might call the "linguistic" paradigm. This paradigm takes as its starting point the notion that language is what comes out of the mouths of people, that this is usually or often heard by other people and that it is then acted upon in various ways. In its most standard form, language according to this paradigm, can be reduced to grammar, vocabulary and phonology.

This approach originates in the desire by people interested in the thing called "language" to develop the scientific study of language. This has given rise to the discipline of Linguistics which purports to perform proper objective and scientific analyses of language. It does so either by focusing on the aforementioned grammar, lexis and phonology or by modifying itself in various ways through the invention of new forms of linguistics such as sociolinguistics, neurolinguistics, psycholinguistics or second language acquisition - all of which use similar intellectual processes to analyse language. Linguistics also combines itself with other areas such as semantics and/or pragmatics in order to take account of the various communicative phenomena that it cannot explain on its own. Ultimately, though, the focus of all these forms of linguistics will be on items (such as words) or acceptable strings of items.

The method of inquiry adopted by this approach to language analysis is supposed to be fundamentally descriptive as opposed to normative as researchers want to know how language really is and not how some people feel that language ought to be. This supposedly objective emphasis on description is identified as a good thing as the process is supposed to reveal to us what makes up the constituent pieces of language and how these pieces relate to one another. The procedure for performing this study is essentially statistical. It relies on the collection of large amounts of string data from native-speaker informants. A statistical analysis of acceptable (and sometimes less-acceptable) sequences is performed and statistical models of how particular languages work are generated.

What is interesting in this process is that the language productions of individuals, with all their idiosyncrasies, are transformed into an object, i.e. language is reified. In the process of reification, individual differences are effaced in favour of the creation of an object which exists only as an abstraction generated by the application of the necessarily arbitrary, and therefore necessarily limited, sets of principles governing the language which have been developed but not discovered by linguists. Reification is effectively an impoverishment of the object of study. Why? Because the act of defining something not only includes certain features in the definition but automatically excludes a huge number of others from it. Linguists then go about studying this reified object which actually exists in nobody's head and in nobody's speech productions. Instead, this object, the specific language of study, can be thought of as a kind of pastiche of all the possible apparently acceptable productions.

Now that language has been made "real" albeit through a process of manufacture rather than discovery, it can be used for purposes other than the study of language for language's sake. An area of obvious relevance is that of language teaching and language-teacher education. The basic reasoning which applies here is that if you are going to teach something then you should know about its structure. Such knowledge should be helpful to you as a teacher and this same knowledge should also be helpful to learners i.e. the basic assumption is that you should know what something looks like in order to be able to teach it. Furthermore, because you now know, or at least believe, that you know what language looks like, you are now able to talk about it. And as we all know, talking about things is an age-old form of sound and commonsense teaching.

Over time, Linguistics has found its place in standard academic structures as the legitimate representative of language studies.
 
 

Problem 1 - the object of linguistic study is not language

The outcome of the process described above is not in fact the discovery of language per se. Instead a representation of language has been produced.

Once the representation of an object has been produced, there is a serious risk that the representation rather than the purported object of study actually becomes the purported object of study in the minds of those who have created the representation. In our case, an arbitrary linguistic description, rather than remaining an arbitrary description, is somehow miraculously transformed into the language itself thus potentially obscuring a myriad of phenomena not accounted for by the description. Thus the manufactured description of the language rather than the language becomes the object of study. If you like, the description of the language has been transformed into the language itself.

This process, which is often invisible as it occurs over long periods of time, seems to satisfy the ambitions of linguists who, consciously or unconsciously, believe that to study their particular description of the language makes good sense. Further, the academy (through the university system), because it values solidity as opposed to vagueness, because it values analysis and because it values the thing which it calls "the pursuit of truth", often rewards this kind of apparently solid, systematic and revelatory study which, it is felt, will lead to the discovery of the "truth" of language - its ultimate structure. In so doing, the academy legitimises the process and entrenches the linguistic paradigm both explicitly and implicitly within the academic system. So now, we not only have a learning and teaching dimension to our problem we also have a significant political dimension involving the relative power of various groups, their careers, the ability to make money and influence others and so on. Of course, in a sense, it cannot be otherwise and it is not even new, but being aware of the phenomenon may be helpful.
 
 

Problem 2 - A learner's perspective

While the sort of study described above may well make sense from the perspective of an academic researcher and is in fact useful in many contexts, the same cannot be said from the perspective of learners faced with the task of learning to communicate proficiently in a foreign language. Interestingly, unlike linguists who are interested in ascertaining the structure of an object which does not exist anywhere other than in statistical formulations, language-learners are faced with what is arguably a much more formidable task.

Language learners have to do all that is necessary to function effectively in a different language and culture. And much is hidden in the phrase "all that is necessary". The learners' focus is not only on correct form but also and primarily on the management of meanings, whether productive or receptive, in order to achieve their personal goals. This effectively requires them to function through the manipulation of symbolic systems which are often unrelated to the "language" of linguists. They have to learn the real language, not its representation. Furthermore, some of the systems which they are required to use, and the variables which they are required to manage, are either undefined or beyond description.

Finally, they also need to achieve this management of meanings at such a speed that they can interact effectively with other people. Let me try to explain a little.

Meaning-making mechanisms (the 3Ms)

Focusing on the management of meanings raises important issues, not the least of which is the question of where meaning is found. There is little space to discuss this in such a short paper but let me assert that meaning is not found in words or dictionaries or texts. It is not found anywhere at all but is generated by meaning-making mechanisms operating in the heads of each and every human being as they go about their daily lives. What makes me say this? Two basic principles which are:

1. All is perception, and
2. Our perceptions of the world are not direct.

In other words, our perceptions are the product of our meaning-making mechanisms. They do not just happen naturally, we do not JUST see or hear or feel, and corrective intervention, to be effective, must necessarily act on our meaning-making mechanisms.

In this perspective, understanding each other, understanding communicative signals, is the product of how we make sense of what people say: how we manufacture meaning from what people say.

How, then, do we understand each other? Space here is very short but the following comments may help to set the scene.

There is an illusion that we understand each other because through experience and observation of the attending practices, we are able to shortcut the whole process of questioning of our "understanding" on the basis of probability. Many of the things that we understand are understood because we have had a long history of dealing with the kinds of signals that come in to us (some more than others) and of assessing and re-assessing over time the probable meanings that we attribute to these signals on the basis of linguistic and non-linguistic clues. While in the beginning we struggle with our understandings of particular things, after some time spent dealing with a specific phenomenon or a related phenomenon we now feel that we "know" and that we no longer need to "understand". However. the fact that we do not understand each other but only think that we do can be illustrated through:

1. The not uncommon need to seek clarification. This requires the re-formulation of something in ways that we have a higher probability of relating to in terms of our internal meaning-making mechanisms. This increase in redundancy is a way of obtaining more "handles" to grab onto in terms of our internal logical and representational systems. That, by the way, is why we have lectures etc where we spend huge amounts of time repeating the same things in different ways.
2. The fact that we say "I agree", then go away and do something which is significantly different from that expected by the other interlocutors.
3. The fact that we cannot perform the linguistic equivalent of the Vulcan Mind Meld (with apologies to Mr Spock in Star Trek) even though we know all the words and all the relevant grammar.

In a foreign/second language context, learners need to be able to generate both productive and receptive meanings from form but they must also be able to generate both productive and receptive meanings from the myriad of other communicative variables which operate at any one time. Some of these variables are either badly studied or badly defined or else they are so strongly context-dependent that they are beyond prediction. The notion of context in this case, includes such things as each individual's personal interpretative mechanisms, their personal history in an absolute sense and their personal histories in relation to any others who may be involved in the communicative event. These context-bound variable are often not constant or stable and keep changing according to circumstances and according to one's developing understandings of the communicative situation in which one is involved. Thus many of the skills required are not merely consciously known or learnt descriptions of linguistic systems but also involve the ability to make subtle and strategic decisions as to language reception and production very rapidly in situations often never encountered before, and unlikely to be encountered again. Further, there are just too many variables, both known and unknown, to cope with quickly. Another way of putting it is that when we become involved in communication, or any other activity for that matter, we are constantly engaged in the construction and re-construction of meaning rather than in the finding of meanings or the remembering of meanings. Thus meaning-making is essentially a dynamic process rather than a static process which relates to the simultaneous management of both known and unknown, perhaps even unknowable, variables.

While this sounds like a daunting task, it is precisely all of these variables and unknowns, the real language if you like, that we all cope with on a daily basis and that proficient language speakers are able to control in order to achieve their communicative objectives. Our best learners achieve this kind of performance and teachers often help them to achieve this. However, I would argue that these students are successful not because they have learnt abstracted grammatical systems or because they or their teachers have had a good dose of linguistic analysis, but because they have somehow integrated certain forms of linguistic, cultural and strategic behaviours into their personal meaning-making mechanisms. These behaviour systems will include, amongst others such non-linguistic systems as gesture, posture, proxemics, gaze, to name but a few of the variables which one can arbitrarily define amongst the myriad of all possible communicative variables operating at any one time.

This dynamic management process is diametrically opposed to models of learning based on synchronic linguistic description which purport to describe a linguistic system, i.e. a static rather than a dynamic entity. The trick for us as language teachers is to find ways of growing the numbers of mysteriously successful learners.

The insufficiency of the linguistic paradigm in achieving communicative outcomes in the terms described above and the obvious necessity to focus on personally-generated meaning makes a strong case for a radical re-thinking of our language-teaching and learning activities and processes. In a very real sense, we now need to go back to first principles.

Returning to first principles will have an impact on (a) our language-teaching and learning practices and (b) our language-teacher education programs.
 
 

An alternative paradigm (based on the 3Ms)

So what might a new-look language-learning system look like? Essentially, it will need to give learners an opportunity to make sensefor themselves and in the contexts of their internal logical and representational systems, of the new communicative world with which they are trying to engage.

How would language-learners make sense of this new world? At least in part by looking, in terms of their personal requirements, motivations and needs, at where and how language in a broad sense is used and what outcomes such usage has in practice, in the real world. They need to be able to confront, contrast and contest (the 3Cs) their personal understandings, beliefs and personal logics against what they are observing. They need to be able to function in a system which answers their own questions and difficulties in ways which are relevant to them rather than having arbitrary or statistically-derived sequences and items imposed on them. And they need to receive appropriate support and assistance when they require it and in a form which is useful to them. In other words, the meanings of form-function relationships are identified not from dictionary definitions but from how those form-function (and other) relationships relate to the real world with all of its complexities and from a multiplicity of sources of information.

Another way of approaching the notion of support is to think of it as a form of learner-generated explanation or self-explanation which enables learners to infer for themselves a picture of what functioning in a particular society feels and looks like. In terms of what was said above, the trick to learning is to make the meaningless meaningful. For this to happen, it is important to provide learners with access to a multiplicity of examples of communicative acts in authentic settings as opposed to the artificiality of textbook settings. To achieve this, teachers and other facilitators of learning, rather than inventing such things as example dialogues, now have the potential to use technology to retrieve large numbers of authentic dialogues which illustrate the point that the learner wishes to investigate or that they wish to make. In retrieving authentic material, they will retrieve not only the current point of focus but also a whole set of other communicative factors which may also be of benefit to the language-learner. Dealing with authentic rather than pedagogic materials is an important way for learners to get at least some access, maybe even unconsciously, to the kinds of undefined and perhaps undefinable (as they are dynamically constructed) communicative events on which their personal logical systems can function and which will empower them to function effectively in a variety of interpersonal contexts in another language.

Thus, in a real sense, effective language-learning requires the learning experience to be complexified rather than arbitrarily simplified. Any simplification which is required will be driven by the needs of the learner within an autonomous learning environment capable of providing personally useful support. Thus the environment will hold and make available within a supportive context not only grammatical or lexical structure but also a whole set of highly complex, often unpredictable sets of communicative relationships.

It is with a view to enabling language-learners to develop personal understandings of these relationships that within the University of Canberra's School of Languages and International Education's Master and PhD and other research programs in language-learning we have set about the research and development of environments which attempt to meet the language-learning needs outlined above and which incorporate a supporting technological infrastructure.

At the core of the learning environments are sets of conditions designed to facilitate the achievement of personal goals and objectives. Such conditions can be created through the generation of personally-negotiated macro-tasks or through the creation of other conceptual spaces for complex communicative activity. One such macro-task is macrosimulation. Macrosimulation, which is described elsewhere in detail (Lian and Mestre, 1985), is an ongoing long-term simulation which provides opportunities for people to develop both common and separate histories and to engage in communicative activities in which they have genuine stakes. These activities will then give rise to questions to be answered and needs to be met as students confront, contrast and contest their understandings about how communication occurs in the foreign language with what they are able to observe and what they are helped to observe.

Three basic principles govern learning activity in the system and its development. They are: Awareness, Autonomy and Achievement (the 3As) (Lian 1987, 1993). These are reported more fully elsewhere. Suffice it to say for our purposes that:

1. Awareness implies that learners need to be able to develop a critical feel for the sorts of things that matter in their interactions with texts from the target language/culture which they are studying.
2. Autonomy implies that learners need to be able to be self-driven and to have the ability to define the kinds of activities which will be of benefit to them.
3. Achievement implies that learners need to be able to gauge what they have been able to do at any particular point during their learning in terms of their current state of communicative competence as well as in terms of their ultimate goals or objectives.

Toward a technological infrastructure for language-learning

Information Technology (IT) is an integral part of the system under development and provides the ability to access information in random fashion and very quickly. Properly organised, it can provide tailored support in an individually-relevant format, and it can offer a dynamic environment capable of being augmented or otherwise modified according to the needs of the learners. Thus the IT system does not provide just a fixed set of resources and procedures but is a growing and dynamic system able to respond flexibly to changing demands by learners. These changing demands are an integral part of its ongoing development.

At the core of the IT infrastructure is an audiovisual database of authentic materials such as films, game shows, advertisements, business letters or newspapers which enable learners to retrieve information at will and in a variety of personally-relevant ways. This database can in turn be linked to support materials such as listening comprehension development systems, awareness-raising systems, standard lessons and, in its most recent form, learner-generated lesson materials for delivery both locally and across the Internet. These systems are currently at the proof-of-concept stage.

It is true that, at this stage, as with traditional systems, much of the material has been gathered by language teachers or curriculum specialists. This will not always be the case as it may be possible to gain access to such resources as the national audiovisual archives of a country as well as much more up-to-date materials such as the latest interviews of people on TV channels such as CNN. Furthermore, and very importantly, the actual pathways that can be followed e.g. subject matter and form of support, are learner-driven.

The next part of this paper describes briefly two of the systems which play a central role in our technological infrastructure. These descriptions are followed by a schematic representation of the total IT infrastructure supporting our system as it is envisaged at present.
 
 

MMBase - an audiovisual database

Here a number of authentic video or audio-based interactions have been recorded onto disk. They are retrieved at will according to categories such as: language, dialect, subject, activity, function, register, participants, power relationships, attitudinal tones, setting - time, setting-place. These categories are supported with a transcript function and a comments function.

All categories are searchable and each can be related to the others according to the wishes of learners.

For instance, learners who may need to prepare a videotaped simulation of "A day on French television" might wish to examine a French television advertisement about dental plaque-removers. They can search according to the criteria of French language, the activity of advertising and a function containing the word "plaque". With a click of the mouse they are able to see the advertisement and read the transcript and comments relating to the particular interaction. If they wish to, they can compare the advertisements they find with similar advertisements in other languages.

Other interesting studies which database systems such as this might facilitate include the possibility of cross-linguistic and cross-cultural comparisons of a broader kind e.g. How does English advertising work as opposed to Chinese advertising? Alternatively, it can also be a valuable tool for answering questions such as: "How are women represented in the advertising discourses of Chinese, French, German and Japanese cultures?". Information can be quickly retrieved and contrasted. Imagination and need will ultimately determine the range of possible uses of systems such as these. Further details may be found in Lian, A.B. 1996.

MMBrowse - a multimedia browser

This is a piece of software designed to develop listening comprehension skills through a self-study approach based on the exploration of authentic audio and/or video text.

An audio or video file is recorded. Learners are provided with a written transcript of the passage where each letter in every word has been replaced with an asterisk (*). The learner's task is to discover the words underneath the asterisks, to gain an understanding of the passage and to come to grips with the features of the text. In order to achieve this the system provides a series of inbuilt facilities which enable the learners to:

1. listen to the recording in its entirety,
2. listen to a selection of words from any arbitrary point in the recording to any other arbitrary point, hence the notion of browsing,
3. listen to, guess and verify individual words,
4. obtain information about individual words,
5. guess and verify "chunks" of language in the text (the underlined portions on the screen),
6. obtain information about "chunks" of language in the text,
7. practise saying selected portions of the text (including comparison of their voice with the original recording),
8. develop awareness of rhythm and intonation structures of authentic text through forward build-up and backward build-up exercises and through listening to electronically-filtered language.

A computer-enhanced language-learning support system (CELLSS)

Below is a diagram indicating the components of an IT system designed to assist with language-learning. The diagram shows the relationship between the various components.

Of particular interest in this system is its ability for the system to generate learning materials according to parameters selected by learners.

There are several advantages in this. The following will illustrate:

1. Rather than doing and re-doing lessons whose content has been pre-determined, learners would be able to generate (i.e. create), on demand, lessons appropriate to their needs.

 

 
 
 

For instance, learners wishing to discover/study specific intonation patterns might request a lesson on "short neutral declarative statements" (selecting other parameters as appropriate to specify the patterns further). The lesson system (designated "Intonation Training" in the above diagram) would then retrieve the required information from the audiovisual database and either make a selection of patterns or list the patterns for learners to choose from on the basis of their personal learning objectives. The source of the intonation patterns would be a range of authentic materials such as televised political speeches, televised game shows, cultural debates, radio news etc. They could also be pedagogic materials designed by teachers and others to support learning.

The system proposed above would then literally construct a lesson consisting of authentic instances of the selected intonation patterns. Thus the models provided would not be contrived or artificial and, in many cases, the learners would also be able to see a video recording of native speakers actually producing the intonations in their natural context. Learners could then be sensitised to the selected intonation patterns and would practise them with all of the support currently provided in Intonation Patterns of French (Lian, 1980 - 1998) (i.e. filtering to enhance perception, intense repetition, forward and backward build-up drills to sensitise to pitch contours and pitch differences, discrimination exercises between filtered and unfiltered patterns etc.).

Further, because the program is designed to be fully integrated with other support systems, learners would be able to seek additional information regarding the various functions of the intonations by interrogating the database system directly e.g. who produced the utterance, in what situation, for what purpose? This would allow learners to relate for themselves, intonative information against situational, discursive and other pertinent information.
 
 

2. The system design ensures that all of its components (i.e. lesson materials and other programs) are linked automatically to one another.

An additional and significant advantage of such a system is that it can be accessed by learners both locally and remotely via the World Wide Web through the use of Web database management systems such as Cold Fusion used in tandem with streaming audio and video (e.g RealAudio and Realvideo). While transmission speeds across the Internet are still relatively slow, the situation will certainly improve in the medium term. It is difficult to predict with any precision the likely impact of such a development as it will open up possibilities which have not yet been thought of. In practical terms and in the foreseeable future, it will be possible to have data sources anywhere in the world which can be manipulated locally for a variety of purposes. An important practical outcome of this arrangement is that it will be possible to share information across very large distances, to share the preparation load in ways which will make it manageable, ensure that information is always up to date (Lian 1988) and, as a result, to increase substantially the amount of rich information available to learners everywhere.

It will also be possible to provide classroom practitioners with simple tools for entering the necessary information thus increasing the number of people able to contribute to the system as a whole.

The explanations provided so far in relation to the above diagram have referred only to the line headed with the labels "Exercise Generator" and "Retrieval". There remain two other labels: "Materials Generator" and "Systems to be conceptualised: Imagination needed!!".

The Materials Generator label refers to the notion of programming computers in such a way as to generate if not authentic materials, at least "authentic-LIKE" materials. The concept here is to write a computer program with enough information to enable it to piece together example interactions between hypothetical human beings in hypothetical life-like situations. The program could produce examples of written language as well as examples of spoken language. Some proof-of-concept systems have already been written (Lian and Joy, 1983, Lian, 1994, 1995).

Finally, the label entitled "Systems to be conceptualised: Imagination needed!!" is probably the most important of all. Technology-based systems lend themselves to all kinds of uses and purposes impossible to imagine before their existence. No one can predict what motivated people will think of once they are given the potential to do interesting things (from Lian and Lian, 1997).
 
 

Implementation

How is all of this put together in practice? In a multiplicity of ways which, ideally, are negotiated between learners and teachers. Here are three possibilities:

1. Learning could be essentially self-managing. Learners negotiate objectives and outcomes with teachers, a broad program of activities is worked out. Assistance with developing autonomy is provided and learners work on their own or on a one to one basis with a teacher, or in groups with no teacher or in groups which include a teacher.
2. The system is integrated into to standard classroom organisation, with groups always made up of the same people, meeting on a regular basis but engaging in a range of different activities based on need and motivation. Autonomy development support is provided.
3. The system is integrated into the standard classroom organisation but the composition of each group varies. In this context, people come together and reorganise themselves into different groups, classes and workshops according to motivation and learning needs.

To summarise, and for those who like formulations of this kind, we can think of the process described above as developing the 3Ms by enabling the 3Cs thanks to the 3As.
 
 

Summary of differences between the two approaches discussed

1. The Linguistics-based approach seeks to provide an object for people to study. It seeks ultimate truth and asserts that such truth is attainable. It focuses strongly on strings of words and grammar to the detriment of other forms of symbolic signification. Individual meaning-making takes a back seat and the teacher/expert is an essential information-giver.
2. The meaning-making approach does not postulate the existence of any object outside the heads of human beings. It is based on the notion that students incorporate, through interaction with communicative phenomena, their conscious and unconscious understandings of these phenomena into their own personal systems of logic rather than having logical systems imposed upon them. They develop an understanding of how to manipulate communicative activities to achieve their personal stakes. The teacher must relinquish control.

Implications for language-teacher education programs

If as I believe, the Linguistics-based programs, imbued, as they are, with the Linguistics way of thinking, are flawed, then it becomes really important for academia to reappraise its value systems in terms of the ways in which it identifies, judges and values its disciplines. For instance, it needs to ask itself seriously questions such as "What is knowledge?" and "What does it mean to pursue truth?".

More specifically, the major implication for Anglo-American language-teacher education programs is that they need to change their focus away from language forms, and strings of words organised according to grammar and move toward an understanding of the importance of meaning-making mechanisms as the base problem to deal with. Linguistics, whatever its virtues, cannot be the portal to language-teaching and learning (Perkins, R. B. 2000, personal communication). In order to do so, programs will have to educate language-teachers to find ways of developing sets of conditions or environments which will enable language-learners to develop, at a personal level, appropriate meaning-making mechanisms.

In practical terms, this means that we need to rehabilitate methodology as a fundamental and important focus of language teacher education programs, perhaps the primary focus. However, such methodology should not be construed as providing future teachers with sets of recipes to be implemented in the classroom. Indeed, methodology needs to be grounded solidly not so much on linguistics and related areas but on critical theory, socio-cultural theory and philosophy of science as a way of providing quality intellectual experiences for future teachers which will enable them to develop the necessary critical apparatus to empower them to remain at the leading edge of their profession through the considerable challenges which we are certain to encounter in the first half-century of the new millennium.

Should Linguistics in its various forms figure at all in the smorgasbord of courses being offered? Yes it should but only to the extent that it is useful and informative and not to the extent that it takes over and diverts attention away from the primary focus of the course: to empower future teachers to develop conditions appropriate for their students' meaning-making mechanisms to develop.

Ultimately, though, we need to appreciate the fragility of the intellectual models that we produce and which inform our actions.
 
 

Conclusion

This paper has sought to problematise the question of the intellectual paradigms used in language-teaching and language-teacher education. It has sought to do so by discussing, in particular, what seems to be the predominant paradigm informing the language-teaching profession: the linguistic paradigm. It has tried to identify some of the weaknesses of that paradigm and to offer a substitute which gives a proper place to meaning-making as the major objective and to the proper recognition of the learner as an individual with specific learning goals and objectives. Finally, it has proposed one possible implementation of the meaning-making paradigm and suggested a set of guidelines for language-teacher education programs.

Some post-conference thoughts/questions

1. What is truth?
2. What is knowledge?
3. What is an object of study?
4. What are the consequences of making assumptions about the nature of objects and the solidity of these objects?
5. What price do you pay for wanting to believe that meaning is solid?
6. What does it mean to be objective.
7. We need to move to a higher level of abstraction.
8. The argument presented in this paper is not about the necessity to categorise in order to make sense of things but about the level of such categorisation.

Bourdieu, P., 1995, The logic of practice. (Translated by R. Nice), Stanford University Press, Stanford, CA.

Guberina, P., 1972, Restricted bands of frequencies in auditory rehabilitation of deaf. Institute of Phonetics Faculty of Arts, Zagreb.

Lian, A. B., 1996, ‘The management and distribution of language-learning resources in the digital era’, paper presented to the National Conference of the Australian Federation of Modern Language Teachers' Association, Perth, October 1994, published in Scarino, A. (ed.): Equity in Languages Other Than English, Perth, pp. 177-182.

Lian, A. B. and Lian, A-P., 1997 >The Secret of the Shao-Lin Monk: Contribution to an intellectual framework for language-learning, revised and expanded version of a paper read originally to the Language Departments at Pennsylvania State University, State College, PA, USA, February 1997, in On-CALL, vol. 11, no. 2, May 1997 pp. 2-18.

Lian, A-P., 1980 - 1998, Intonation Patterns of French. Teacher's Book. River Seine Publications, Melbourne. Digitised version begun in 1998.

Lian, A-P., 1985, 'An Experimental Computer-Assisted Listening Comprehension System', in Revue de Phonétique Appliquée, 73-74-75, 1985, pp. 167-184.

Lian, A-P., 1987, 'Awareness, Autonomy and Achievement'. Revue de Phonétique Appliquée. vol. 82-84, p. 167-184.

Lian, A-P., 1994, ‘Dialogue generators Mark II', paper read to the ALAA National Congress, University of Melbourne.

Lian, A-P., 1995, ‘Dialogue generators: the next generation', paper read to the International CALICO symposium, Middlebury College, USA.

Lian, A-P. and Joy, B. K., 1981, 'Verbo-tonalism, Research and Language-learning', in SGAV Newsletter, July Sydney, pp 7-12.

Lian, A-P. and Joy, B. K., 1982, 'Prosody: the Crossroads of Language Systems', in SGAV Review, vol. 1, no. 1, Sydney, pp. 5-11.

Lian, A-P. and Joy, B. K., 1983, 'The Butcher, The Baker, The Candlestick Maker: Some Uses of Dialogue Generators in Computer-Assisted Foreign Language Learning', in Australian Review of Applied Linguistics, vol. 6, no. 2, pp. 60-71.

Lian, A-P. and Mestre, M-C., 1985, 'Goal-directed Communicative Interaction and Macrosimulation', in Revue de Phonétique Appliquée, 73-74-75, 1985, pp. 185-210.

Renard, R. 1978, Introduction à la méthode verbo-tonale,Didier, Paris.