Introduction
John Trim was born in 1924. His father was a docker and his mother was the daughter of a printer; both were active in the local Workers’ Educational Association. John described the atmosphere of his home as ‘intellectual, internationalist and socialist’. He won a scholarship from his primary school to Leyton High School, where he learned French and German. For the first term – which John missed because he had pneumonia – his French teacher taught the language entirely in phonetic transcription in order to lay the foundations of accurate pronunciation. In his second year John had to choose between Latin and German. He chose German because he was not yet thinking in terms of university, and here too he encountered a teaching approach that was strongly oral. From these beginnings John went on to become a leading phonetician, a founding father of applied linguistics, director of the Centre for Information on Language Teaching and Research (and ex officio editor of Language Teaching and Research Abstracts), adviser to the Council of Europe's modern languages projects for three decades, and a powerful advocate of communicative approaches to language teaching and learning.
In the summer of 2011 we recorded an extended conversation with John Trim in which we ranged widely over his family background and education, his academic career at University College London and Cambridge, the years he spent as director of CILT, his extensive experience of language education policy development and implementation, and his three decades of work for the Council of Europe. What follows is the second of two extracts from the recorded conversation. The first extract, which draws on all the topics we discussed apart from the Council of Europe work, was published in Language Teaching 46.3.
Applied linguistics, the Council of Europe and BAAL
The Council of Europe was clearly an important part of your life. Could we talk about how you became involved with its work in the 1960s?
In 1969 I organised in Cambridge the Second International Congress of Applied Linguistics, which was the first to arouse world interest. The way that came about was that after the signature of the Convention on Cultural Cooperation, the Council of Europe had become interested in the teaching of modern languages from the point of view of improving the quality of interpersonal communication to support greater interaction and cooperation across Europe. This was spearheaded by the French, who were very impressed by the rise of English and the threat it posed to French as a world language. Particularly they were impressed by the Center for Applied Linguistics in Washington DC, so they proposed that an equivalent centre should be established in Europe. But they couldn't get support from the other member states because of the financial implications, so instead they put money and effort into a major ten-year project in languages and put in charge of it a very brilliant young Swede, Sven Nord, within the secretariat. The project was due to run from 1963 to 1973. Its main priorities were the promotion of audio-visual methodology, and applied linguistics at university level.
The Council of Europe had the idea of launching a series of international congresses of applied linguistics. The first one was held in Nancy in 1964 or 1965. Peter Strevens was very active in the whole affair at that time, and he undertook to organise the second congress in 1968 in Essex. But he couldn't get a government subsidy to do it and so withdrew. In his attempt to get government backing he had, together with Michael Halliday, organised the setting up of the British Association for Applied Linguistics (BAAL). With Halliday and Angus McIntosh, Strevens had published The linguistic sciences and language teaching (1964), which was very influential at that time. Possibly in response to a Council of Europe recommendation but also for other reasons, a lot of money was put into a number of centres for applied linguistics in the early 1960s, keeping them out of London for the most part (they didn't want to have too many foreign students in London). There was a particularly prominent centre in Edinburgh, where there was a whole cluster of linguistic interests. Angus McIntosh as professor of English was important in that, as were Halliday as professor of linguistics and Strevens, who was concerned with phonetics.
I became first treasurer of BAAL, largely because I had, from the floor, opposed the limitation of applied linguistics in its constitution to language teaching and machine translation (the interests of the Council of Europe). My approach was that applied linguistics should be concerned with the application of linguistic knowledge and skills to all social problems involving language, so I proposed a change in the constitution, which was accepted and as a result I got into the BAAL network.
When Strevens said that he was no longer prepared to organise the congress of applied linguistics because of the financial problem, I thought that in Cambridge any congress ought to be able to pay its way without money from outside, so I took it on. I reckoned that with a £10 registration fee I could break even with 300 participants. I had no opposition – it was a year when there were no other congresses of linguistics or phonetic sciences – and I was quite sure that the Americans who wanted to come to Europe that year would come to Cambridge. If the big Americans would come, then everybody would come. I got to 300 participants three weeks before the congress, and in the event 600 came, 100 of them not having said in advance that they were coming. They all paid their registration fee and I was able to start off BAAL with a little nest egg.
Organising the congress brought me into the Council of Europe orbit. They put (I think) about 8000 francs into it, in return for which I had to submit a report. They were pleased with the report, so I was invited to various of the symposia that were being organised as part of the major project.
Creating a methodological consensus
How was it possible to create a consensus in the Council of Europe's work among people from very different academic traditions and with very different theoretical backgrounds? What perspectives were brought to the work by people who were involved from the beginning – yourself and Jan van Ek, and on the francophone side Daniel Coste, Henri Holec, Eddy Roulet and René Richterich? How was the team built? Or did it just emerge?
Well, a good deal of that was done before I arrived on the scene. I’ve spoken of the major project and the promotion of applied linguistics within that project, and all the people you've mentioned and a number of others, particularly from Scandinavia, were already active. I think a good deal of the general approach was shared by most people working within the Council of Europe project at that time. I slotted into it in a way. I didn't myself hold a very strong theoretical or ideological position on language teaching methods. I thought that working within the framework of the Council of Europe, one would accept that its interest in language learning and teaching was to improve the quality of interpersonal communication and cooperation among the populations of Europe. European cooperation was not seen as something that happens only at a high level, leaving the mass of the people unconcerned, but as a matter of getting large populations with very different backgrounds and very different characteristics to interact and cooperate. Improved interpersonal relations and practical cooperation seemed to me to be what we were aiming at.
Communication as a goal
Communication to me was a central objective. I didn't think it was necessarily the central objective of all language learning, but it was the central objective of the language learning that was relevant to the overarching aims of the organisation I was working for. It was an open question how that was to be achieved. I didn't think necessarily that you had to reach the ability to communicate well by indefinitely practising communication. I think you would have to do some, but it was not necessarily the method. There were people who believed that, quite a lot of them, but my own view was that methods must vary with the nature of the learners, teachers and resources involved. Anybody working in the field who was sincere and had successful experience had something to contribute. My job was seeing how it all slotted in rather than establishing a line that I would then expect everyone who came along to adhere to. I was also inspired by the bon mot of Tucholsky: ‘Alles ist richtig, auch das Gegenteil’ (‘Everything is right, including the contrary). I hadn't much time for erudition, but I had a great deal of time for practical experience and insight of all kinds. If there were fresh people coming in, usually because their governments nominated them to participate in symposia, my job was not to say ‘Do they fit in or not?’ but ‘How do they fit in?’. I felt that there must be coherence, but that coherence was not a party line but a matter of seeing how the world fits together, with all sorts of people in it. But perhaps the people who would have been impossible to fit in didn't come to us.
Anglo-French differences
So is the story of – let us say – differences of approach between English and French speakers within the Council of Europe overstated?
I enjoyed that, I found it great, because the two approaches are so different. Philosophy is compulsory in the sixth form in France and unheard of in the English system. The two tendencies are that the French are not prepared to go into action until they have intellectual clarity – they’re prepared to spend quite a long time getting to the point of intellectual clarity, but then they work very fast and very effectively; whereas the English don't work in that way at all – they like to jump in and get experience, and when you've got enough experience you arrive at some intellectual clarity. That means that when they start working together it's very difficult because they are both convinced that the other is on the wrong track and is producing nothing. The French feel that the English – and not just the English, also the Scandinavians quite often, with the Germans somewhere in between: the Anglo-Saxon tendency – are floundering around without knowing what they’re doing. And the English feel that the French are just sitting back and philosophising and doing nothing of any value. If you can keep them together for three or four years, somehow the two approaches cohere. I can appreciate both approaches. But if you take someone like Daniel Coste, he is betwixt and between. He's got strong Anglo-Saxon influences and he combines very well the theoretical and the practical, I find. Some others in the French tradition I found much more difficult, but I’ve enjoyed working with Daniel and I got on very well with Louis Porcher.
Difficulties with AFLA (French Association for Applied Linguistics)
I remember when Pit Corder was president of AILA (International Association of Applied Linguistics), we had great problems with AFLA (French Association for Applied Linguistics), which was in the hands of Professor Culioli. He came to a meeting of the board in Neuchâtel, organised by Eddy Roulet and René Richterich, and we were concerned to know why AFLA, which was so central to the early development of AILA, seemed to have withdrawn after 1969. He said that he was perfectly willing for AFLA to return within the AILA fold provided we agreed on the definition of what applied linguistics was. There is theoretical linguistics and applied linguistics: theoretical linguistics is concerned with the development of models of language, whether or not there are languages that conform to these models. Once you start looking at actual languages, that is applied linguistics. We were invited to accept that definition of applied linguistics and then AFLA would return to the fold. It was rather similar to the definition of theoretical and applied mathematics. We said that this was an interesting concept but it didn't correspond to our present commitments and interests; perhaps we should discuss it after lunch. And after lunch he didn't reappear, and for a long time AFLA was not active within AILA. When AILA recognised other competing groups in France, AFLA took offence and took legal action to be readmitted. But by that time it had all changed. If you're working at an international level you can't be a purist or have a strong theoretical requirement. You have to accept that if people are to cooperate they have to accept each other's differences of tradition, differences of working method quite often, and then see how these different things can in their different ways contribute to an overall project.
Adult education, short-term language learning objectives and The Threshold Level
We started off under the aegis of the Committee for Out-of-School Education and Cultural Development, which was the last of the Council of Europe's major committees to take an interest in languages, and the interest they took was in the application of the ideas on unit/credit organisation developed by Bernard Schwartz in Nancy. That committee had quite a strong ideology – it was strongly learner-centred, not teacher-centred or subject-centred – and was committed to the idea of permanent education, the idea that all kinds of learning, including language learning, were lifelong. Because they were interested in adult education they were interested particularly in establishing relatively short-term goals and recognising relatively short-term achievements – that is the only way you can organise adult education. Because of the conditions of adult life, you couldn't say that you're going to follow a course for seven years and that you don't expect to reach communicative proficiency until seven years’ time. In the meantime, in the first two or three years, you can be working in an insulated classroom environment and you're not expected to communicate with anybody; you're just getting the materials together. You can get away with that in a school environment provided everybody will go along with it, but in an adult environment you can’t. Whatever people want, they want it quickly, so you have got to be able to organise short-term goals and make them very explicit and work towards them – people are happy if they achieve those. This may mean a very restricted form of communication, but it will satisfy people to start with. If you say, ‘No, you're going to pick up the vocabulary and learn the grammatical rules, and only much later will you benefit from this, and how you benefit is up to you’, this would not be acceptable.
Our working group was brought in to consider whether this approach to adult education, associated with Bernard Schwartz in Nancy, also Yves Chalon and Henri Holec, would work in the language field. David Wilkins and I were nominated by the British government to attend the conference convened to discuss the issue at Rüschlikon in Switzerland in 1971. We talked about it in advance and were not satisfied with the model of situational teaching which was proposed by the Eurocentres, so we wrote a short paper. We thought that it was necessary to find situational constants, not individual events. Individual events are not situations, which are constructs that underlie a class of events. You have got to look for those things that can characterise those particular situational structures.
For that symposium I produced an appendix with a classification of language functions which I got from here and there. That was taken up by the working party that was set up. I got the members together in Selwyn College in a freezing January with no heating and filled the room with all the source materials that I could think of. We worked in little groups on these and produced long, unordered lists of categories in a range of different dimensions of communicative competence. Then Jan van Ek was given study leave by the Dutch government and his university to put this material into order. Even the unordered lists were very much welcomed, particularly by the Volkshochschulverein (German Association of Adult Education Colleges) and the International Certificate Commission that worked with them, and that persuaded us and the Dutch government that the material should be put in order and published by the Council of Europe quickly before it got too widely disseminated and out of hand. We published the initial set of papers in 1973; The Threshold Level was produced in 1974 and published in 1975.
The ‘scientific’ status of the Council of Europe work
You say you produced a classification of language functions ‘from here and there’ and people then looked at it. How scientific did it end up being? Or could it not be scientific?
I don't think my mathematics is strong enough for me to be a scientist. But the scientific status of the work has never worried me particularly. We produce something and if you find it useful, use it, and if you don't find it useful, don’t. I grew up in an atmosphere, both at University College in my undergraduate days in German, particularly on the literary side, and also to a large extent in the Phonetics Department, which favoured direct engagement with the phenomena, rather than with secondary literature and academic justification . . . There was a left-wing group which I belonged to, based in Hull – The Hull Linguistics Society – Geoff Ellis, Michael Halliday and a number of other people – and there was no strong idea of intellectual property, at least so far as I was aware. Ideas circulated: you fed ideas in and you took ideas out and you tried to develop knowledge of the field in this way. But the question of who was the first person to set down these things and . . . And you read what everyone else has published and then you publish your own, and you don't tell anyone what you're thinking until you're ready to publish it, and then it belongs to you . . . this whole way of thinking was quite alien to me and, I think, to my generation. You couldn't survive in this way in the modern academic world. I can't really say where all the ideas I have had come from. I don't much mind who takes up those ideas and what they do with them. Once I’ve put them in they lead their own lives. When I started in the 1950s – in the 1940s, even – there were very few journals, just two or three, I think, and you weren't expected to publish until you’d got something important to say. The current academic ethos – it's research that matters, not teaching, and your research is measured by the number of papers and books that you publish – is not the way that I was brought up. I find writing very difficult, because I sit down to write and my mind goes off in all directions and I find that I haven't written anything. There are certain necessary disciplines that I find difficult.
But what was surely important in the process that you've been describing was the decision to systematise all these ideas and raw materials, and then the act of putting some kind of shape on them. The number of available shapes was probably not very many and these unordered lists could be separated out into a rather small number of dominant strands which you could then weave together in maybe three or four different patterns, but not more than that.
I suppose that's probably right, but I don't know that I would necessarily have thought in that way at the time.
The Threshold Level and assessment
The Threshold Level has nothing to say about assessment. Why is that?
Well, we had problems with assessment. The Committee for Out-of-School Education and Herbert Jocher, who was the official responsible for it, had a strong ideology, and formal assessment was not part of it. They didn't like the idea of formal examinations and certification by examination. They were very much more interested in self-assessment by learners and hostile towards its institutionalisation. The form they wanted the unit/credit scheme to take would have a teaching/learning/assessment model of a strongly learner-centred kind. If you look back on the publications of that time, you’ll find the work of Mats Oskarsson on self-assessmentFootnote 1 but you won't find anything much on language testing and language assessment. I did get them to commission a specimen Threshold Level test by Piet de Groot and Andrew Harrison. They did a lot of work on it and it was very good, but I couldn't get the committee to publish it. It's in the archive because it got to the point of being a mimeographed paper, but I just couldn't get them to publish it. They would never say that they wouldn’t, but it always got put off and so never reached the light of day. If you take the collection of papers in Across the Threshold, I had to write the chapter on assessment because the work on that side had not been developed.Footnote 2 There was nevertheless a strong interest from the start of my connection with the Council of Europe from the International Certificate Commission, and Tony Fitzpatrick had an authorial role in Waystage. Later, Cambridge Assessment became very interested, so by the time it came to the Common European Framework the assessment interest was very strong.
The Threshold Level and Un Niveau Seuil
You said earlier that The Threshold Level took a particular shape which was one possible shape among probably a rather limited number. Un Niveau Seuil was published one year after Threshold, in 1976, and Kontaktschwelle in 1980. Were these attempts to open up the model and steer it in slightly different ways?
One of the tensions between the Anglo-Saxon and the French groups was over the status of The Threshold Level. The French were very worried on ideological grounds that although The Threshold Level was stated to be just one possible model, it would be taken in its detail as being a European certificate of proficiency, and that it would become canonical. They felt that was wrong. So while Un Niveau Seuil does contain a fairly close translation of The Threshold Level, it is put in a rather subordinate position. What they wanted was to produce a survey of the resources of French at a fairly elementary level, which could not be used directly by anybody other than as source material, so that people could draw on it in accordance with the needs and characteristics of their learners. But it was too rich to be used as an objective by anybody. The Threshold Level was an objective, but a specimen objective, and we expected that people would use those parts of it they liked and amend it according to the needs of their learners, and so on. We didn't expect anybody to take it exactly as it was, and nobody did, partly because, being in English, the English EFL industry is very strong, very diverse, and subject to copyright laws. When organisations produce courses, those courses must be unique, so they can't take The Threshold Level as it stands, they must amend it in detail. So in regard to English we didn't have to worry so much about it being a unique model. But the French were absolutely determined that there should be no possibility of that, so they organised Un Niveau Seuil in that way. It's much thicker than The Threshold Level and not user-friendly. The simpler model won out among the users because it was straightforwardly usable – you have to be a pretty highly qualified linguist in order to be able to use the French model at all. As a source document Un Niveau Seuil is invaluable, however, and should be far better known.
When producing something like The Threshold Level, do you think there's a tension between making it so all-encompassing that people can clearly do what they like and restricting it to something useful and then people take it and say, ‘That's it. That's what we have to use.’
Well, if they want to they can, but they’re not obliged to. The Council of Europe does not have sticks and carrots to make people do anything. It can only produce models that people can use as they wish or not. If they abuse them, they abuse them. It can't set itself up as an educational authority; it hasn't the structures or the resources to do that. It's always under pressure to establish mechanisms for recognition and so on. I have always resisted such pressures.
The Threshold Level specified objectives for learners of English, but because it was free of the kind of ideology that shaped Un Niveau Seuil, it turned out to be easily transferable, even translatable, into other languages.
Well, it finished up in some 25 or 26 languages across Europe and it survived the transfer into other branches of the Indo-European languages and into non-Indo-European languages. Because it was not concerned with formal linguistic structure and speakers of languages of all kinds essentially want to do the same things, a taxonomy of what was involved in language use was universally useful. What was taken over very readily by the language teaching industry was the functions and the specific notions. General notions were on the whole passed over.
Because they’re more difficult to understand?
They’re not difficult to understand but more difficult to apply, I think.
From adult education to language learning at school
So, if we understand what you are saying, when you became involved with the Council of Europe's modern languages work, the first thing was the interest in the unit/credit system. That then generated a need for definition, categorisation, taxonomy and so on, which yielded The Threshold Level (and as variants, Un Niveau Seuil and Kontaktschwelle). There was a shift through that period in terms of responsibility for the work from the Committee for Out-of-School Education to the Committee for School Education, and that was accompanied by a shift in interest from self-assessment to tests. Presumably that shift also explains why in the 1980s the project moved into school visits and teacher training seminars?
Yes, from 1977 the much more powerful Committee for School Education became interested in our work. They commissioned a variant of The Threshold Level for schools from van Ek, which was published by Longman.
We published a report on the unit/credit scheme for adult education and presented the scheme to an intergovernmental symposium held in Ludwigshafen in 1977; then it was quietly put to bed. The whole direction of work changed to the school sector. Antonietta de Vigili moved out of the adult education sector and became concerned with the administration of the project. She attracted a lot of attention because she was intellectually very strongly committed to the ideas of the project, whereas some people, particularly Cellan Williams, a school inspector representing the UK, felt that the task of the secretariat was not to become engaged with projects. The secretariat should be an administrative apparatus, and the projects should be in the hands of people nominated by governments.
Workshops for teacher trainers
In the 1980s and a fair amount of the 1990s the main activity was the series of workshops for teacher trainers and the like – there were a very large number of them. We identified teacher trainers as the key personnel for educational change. The teaching profession itself is fairly conservative. If you're used to your textbooks and examinations, life is not easy but as straightforward as you can make it, and you don't necessarily welcome people who come along and want to change everything. You particularly don't if you get the idea in your mind – and I felt it very much with the Americans – that these new ideas are going to change every five years. If you attach yourself to some new ideas now, people are going to be laughing at you in ten years' time because by then it's old stuff. Academic neophilia is a barrier to educational change: if ideas change too often, the ship sails on, the dogs bark and the caravan passes. If you really want to carry out educational change you've got to maintain pressure long enough for new ideas to become generalised through the profession. You've got to get the different agencies involved, some of them independent, all to work in the same direction. And that has to go on long enough to enter into the presuppositions of classroom practice. We felt we had got to the stage where the professional agencies involved had been fairly well won over, so it was important to directly affect the ideas of practising teachers. If a body as central as the Council of Europe then starts organising workshops for teachers, they are drops in the ocean, so we felt we had to get at teacher trainers. I was rather conscious of the very limited extent to which we had affected teacher training in this country. The first set of workshops, and then the new-style paired workshops (with some kind of research and development work in between), took up a great deal of the 1980s and 1990s. It was really quite effective. We did enquire at the end what channels the influence had gone through and they were multifarious – all sorts of direct and indirect channels. The fact that the workshops were peripatetic, that they had to be organised by different ministries in turn, meant that you had got people, quite often of increasing influence, within the national ministries who had a direct knowledge of what it was all about. We had well-placed friends in a lot of member governments. More than 200 people acted as workshop animateurs. It was at that stage that the idea of communicative methods became much stronger.
The Common European Framework of Reference for Languages (CEFR)
Could we now focus on the Framework and invite you to share with us your reflections on its impact?
I think the impact of the CEFR has been very strong but not exactly what we had expected or intended. But it may be very much in line with what was intended by the authorities who commissioned it in the first place. I know that it has been attacked in some quarters as arising out of a petty local squabble about qualifications. That is to say, the motivation of the Swiss government in setting up its own research project and inviting the other member governments to the symposium in Rüschlikon in 1991 was the need for some external objective calibration instrument to enable the separate educational qualifications – language qualifications in this case – issued by the independent cantonal authorities to be appropriately interpreted. But Switzerland might be seen as a microcosm of the European or global situation in which we have increasing educational and vocational mobility, with decisions by university admission authorities or employers being based upon the evaluation of qualifications obtained under different authorities in different places. We needed some calibration that enabled them to be roughly compared with each other. The authoring group was conscious of that – and Brian North, working in the Swiss context, was particularly concerned with the scalings. But Daniel Coste was concerned with curriculum design and I was concerned to give a more complete taxonomy of what I saw to be involved in the act of communication, which would be a basis for professionals of all kinds and perhaps learners themselves if they were sufficiently equipped to do so, to consider which aspects of communication are important to them and which should be taken into account in drawing up an appropriate profile for themselves or other people. That idea was certainly taken up in the German Profile (which is plural).Footnote 3
The impact of the CEFR on language testing
The take-up has been very much stronger, it seems, in respect of the calibration aspect than in respect of the basis for reflection on existing and possible practice. The reason for that is really that it has come at a time when the practical problems that are raised by educational and vocational mobility and the portability of qualifications have become an urgent matter for authorities of various kinds. That social requirement is not going to go away. There have been ideological objections to it, quite strong in some quarters, amounting to the argument that you shouldn't be doing this at all. But the demand comes from the field. It has not been a requirement or an objective of the Council of Europe itself, which hasn't the structures or resources to act as an educational authority in the way that some people have wanted. So I think that the impact the Framework has had, which has produced a response from ACTFL (American Council on the Teaching of Foreign Languages), for example, where it impinged on the American market because of the expansion of the idea of ascribing a level on a six-level basis to qualifications and the proficiency of individuals – this has been seen as meeting a very real need.
Operational adequacy of the CEFR
Then the question is whether the Framework as it stands, as well as meeting the various criteria that Brian North set out, is also operationally adequate to perform this task. And there the problem is that the statements that can be made at a non-language-specific, general, global level have to be of a functional-notional character, because once you get down to the details of linguistic structure, generalisation becomes impossible. So these specifications in terms of what degree of structural complexity is appropriate to different levels and what order of grammatical and lexical progression can be established for a particular language will depend on the target language and also, within that, to a very considerable extent on the source language, the mother tongue of the learners. That kind of specification has to be made language by language and learner group by learner group. It can't be done once and for all – one size fits all – across Europe, let alone the rest of the world.
To a certain extent the other aspect, that is, the basis for reflection, communication and dialogue among providers of different kinds with regard to questions of objectives and to some extent methodology, has perhaps become a little better understood as time has gone by, and certainly it has had some effect on initial teacher training in some countries by giving teachers a more global view of what is involved in language use by a competent adult speaker.
Criticisms of the CEFR
The Framework has attracted two lines of criticism. The first is that it is an attempt by an international authority to impose its will and a particular set of assumptions and objectives on the world at large, which is politically objectionable. It may not have been an attempt by the Council of Europe to do so, but it has been interpreted as such by a lot of users, and there are questions of prestige and other corrupting factors involved, which mean that it acquires a status which the administrative structure associated with it is not strong enough to bear. The other criticism is that from the point of view of the various people who have to operate with it, particularly in the testing field, it is very much under-specified. But the answer to that seems to be that the full specification has to be undertaken by users in the concrete situations that face them. And therefore the degree of international calibration that is possible is limited, necessarily and rightly so, but it's still better than nothing.
The six-level system
The essential six-level system, which is very closely related to the traditional distinctions between beginner, intermediate and advanced, and upper and lower sections of these, is fairly coarse and fairly robust, particularly if you think of it as banding rather than a level point – because obviously you have a continuum which is broken into bands; it can't be reduced to a succession of discrete points. If people are talking about (say) A2, this will cover a certain range, not a point, and things that are different from each other but within the same range are, from the point of view of most of the users of the information, sufficient for their needs. But of course, once you get to the actual process of testing and certification, enormous attention attaches to the details of the transition points between pass/qualification and failure and a degree of specification is required that the Framework is not intended to provide, and that gives rise to problems.
Follow-up to the CEFR
The Council of Europe's subsequent work took two major forms. One was the reference level descriptions undertaken in detail for particular languages, English Profile Footnote 4 being one of these and Profile Deutsch another, and the French, Spanish and Italians have things in train. The other was the composition of the manual for relating examinations to the Framework levels.Footnote 5 I was also involved in the final formulation of the Recommendation of the Committee of Ministers – 2008 (7),Footnote 6 addressed to member governments on the use of the Framework, designed to encourage them, in the qualifications that they offer and the tests that they apply to immigrants or other groups, to make reference to the Council of Europe levels. But if they do so, then they incur certain responsibilities for making sure that the procedures by which these levels are ascribed are transparent and professionally respectable. I would rather see those people who have been critical of the abuse of the Framework paying less attention to criticising the Framework and more to ensuring that the criteria for its use by authorities of one kind or another are monitored and its transparency and respectability assured.
It's now ten years since the Framework was published and it had evolved through the previous ten years. Two drafts appeared in rather quick succession.
One of the problems was that the drafts became widely known and were taken to be the Framework. When the eventual publication was different, people were rather taken aback. The main difference – and this I would take responsibility for and would still defend – was that the various scalings for particular competences and particular activities were originally all collected in an appendix and in that way were indicated to be somehow not part of the Framework itself. In the final publication I felt that it would be easier for users if the scales relating to a particular competence or activity were next door to the taxonomy of competences and activities in the body of the Framework itself. So they were transferred from an appendix to Chapters 4 and 5. It's true that to a large extent a number of those scalings were speculative and provisional and not subject to the same kind of Rasch procedures that were applied to Tables 1 and 2. They are in some cases incomplete, with no statements at the A and C levels. With regard to the descriptors as a whole, particularly at the A and C levels, they are not entirely internally consistent. For example, at the A1 level, which always attracts the statements of incompetence, even though our aim is to rephrase them in a positive manner, there are some statements which make it look as if competence at A1 is very largely of a phrase-book type – fixed formulae learnt by rote – and others that imply that it involves the manipulation of a small number of simple structures. Whether the level is creative or simply involves memorising is not fully resolved.
In Chapter 2, where the levels are described in an introductory way, A1 is described as the lowest level of generative language use, and there is a statement to the effect that formulae like ‘hello’ and ‘goodbye’, ‘please’ and ‘thank you’ are somehow prior to generative language use.
Yes, that's a very difficult statement, isn't it? If we don't recognise a level below A1, then they’re part of A1. I don't know how the apparatus actually places them – if they are below A1, what are they? Speaking from memory, I think I would list those things in my Breakthrough definitionFootnote 7 and I would regard them as being part of the A1 level because we have nothing that is regarded as being examinable as a public interest below that level. The order in which you acquire things in getting to that first certifiable level is up for grabs. It's a matter of how the individual teaching institution organises its programme. Whether there are things that are listed as being below A1 and necessarily come before simple structures, I’m not at all sure, because the human mind is creative, and as soon as you have one particular example of a structure, you play with it. When I. A. Richards's travelling circus came to Cambridge I was used as a guinea pig and taught to say ‘Yo soy un hombre; usted es una mujer’(‘I am a man; you are a woman’). As a consequence I would expect to understand a woman who said, ‘Yo soy una mujer; usted es un hombre’. I don't really have to be taught it; it comes out of my general experience of life and of language.
The approach to A1 was that this was the lowest level at which the certification of achievement by viable numbers in formal education could be of public interest, and C2 the highest level at which such public certification was appropriate. The scheme is open-ended at the top. If the Goethe Institut regard their Grosses Sprachdiplom as being above C2, then it will be possible for them to invent a level D, but at the other end, because the alphabet starts at A, it is much more difficult. In Chapter 2 the ‘plus levels’ are introduced (A2+, B1+, B2+) and it's difficult to know what nomenclature to use for divisions that you might want to make for teaching purposes. Before you get to the lowest certifiable level, you've still got to organise your teaching programme. You may want to establish mini-levels below that as on the branching diagram in Chapter 2, but what you actually call them is not so easy. On the American model you would have 0.1, 0.2 etc., leading up to 1, because they use the decimal system.
Now that the Framework has been out there for ten years, perhaps it is time to think in terms of a second edition. Do you have any particular views on focuses for revision?
I think there are untidinesses that have been revealed over the course of time, which is not surprising. Changing things to a decimal system would be a new undertaking, I think; everything would have to be reconsidered from the start. I think the basic 3 × 2 6-level system is enough for the purposes for which it was intended, but I’m sure that it can be improved in detail and it would be useful to have a working party got together. And since the Framework has proved to be of wider than European interest, it's quite possible that if modifications are to be considered, organisations beyond Europe which have used it should be consulted. In a future edition one might consider dropping the word European, which has been interpreted in a way which was not intended. When the Framework is used outside Europe, people's sensibilities are aroused – ‘We're not Europe, why are we adopting a European standard?’ We used the word European because that was the framework within which we were operating and we had no remit to adopt a more general title. But if global interests were consulted and taken into account, then it might not be necessary to continue to impose that restriction in the title. On the other hand, it's now known by that title and to change it might make it look as if it's more different than it actually will be.
The CEFR and young language learners
It has often been noted that the Framework was developed for adolescent and adult learners and that at the lower levels younger learners sometimes do things somewhat differently. Have you got any thoughts about this? For example, there is a lot in early language learning which is about listening to stories and other extended texts in a way that doesn't ‘compute’.
If you are thinking of a core vocabulary for A1, does it include dinosaurs?
Or stick insects. What was that famous phrase? ‘My guinea pig died with his legs in the air’.
Somebody with C2 in German is capable of participating fully in a Hauptseminar at a German university. The requirements of all the C-level specifications I’ve seen are cognitively and rhetorically demanding in a way which would be quite inappropriate to a younger learner. A normal English child in a normal English school at the age of ten would not be C2 in English. You may say that from the point of view of the language the child knows it. When you get towards the higher levels the limitations on attainment by native speakers and foreign learners can merge, whereas if you were establishing an A1–C2 scale for native speakers, at the bottom they would be very different in character. A child beginning school has a knowledge of the spoken language as a native speaker that a foreign language beginner doesn't have at all, which is why there is always this polarity – what George Perren called ‘the space between’ – when you try to bring foreign language teachers and English teachers into the same universe. Though I must say that one of the happiest experiences in my professional life was the languages committee set up by the Royal Society of Arts. We did have the people teaching English to secretaries or to people who were educationally under-achievers in English in the same committee as foreign languages and English as a second or foreign language. By working together we found that we were in the same universe and that in terms of functional–notional objectives we had a lot in common.
The CEFR and the language of schooling
You spoke about a functional convergence between native and non-native speaker requirements at the higher levels. But as you know, the current project of the Language Policy Division has rather excluded the CEFR from its consideration. At an early stage they thought they might develop something like the Framework focused on languages of schooling or perhaps extend the existing Framework to embrace L1 functionalities, but that was very quickly set aside as an impossibly ambitious task.
I wasn't consulted on that, but I would have considered that the part of the Framework with which I was particularly concerned, namely Chapters 4 and 5, is as relevant in that context as in any other. The description of what a mature adult language user is expected to be able to do and what competences are required in order to do it applies as much to mother tongue as to foreign language development. On the other hand, the scalings would have to be radically different, particularly at the lower end. L1 descriptors might be very similar in form but different in content, because in the case of L1 development a great deal of what the Framework specifies at lower levels for foreign language learners will already have been acquired and can be presupposed for people starting an educational process in their mother tongue, or in the language of school instruction once school starts. The basic linguistic development of a child in spoken language is generally thought to be complete by the age of four. The problems of literacy are rather different, but they can still be formulated in functional terms. The way the Framework's scaling is formulated does presuppose a congruity between cognitive and linguistic development, so that you find statements that make it look as if when people begin to learn a language they are much more able to cope with the concrete than the abstract. That is at least open to doubt, I think. If you get an immigrant coming into the school system in their teens, for example, they will be confronted by the abstract right away, and in the school context they will not be confronted with quite a lot of the concrete aspects of everyday transactional language that tends to predominate in the early stages of the learning process as it's envisaged in some of the functional statements of the Framework.
David Little is Fellow and Associate Professor Emeritus at Trinity College Dublin. In 1978 he established Trinity College's Centre for Language and Communication Studies, which followed the example set by John Trim in Cambridge by combining the functions of a language centre and a department of linguistics. Since the early 1980s David Little has been involved in the modern languages projects of the Council of Europe, most recently those concerned with the European Language Portfolio and the linguistic integration of adult migrants.
Lid King has been a language teacher, examiner and teacher trainer. As Director of CILT (1992–2003) he led a range of initiatives, including the establishment of the Comenius network and the initiation of support for Primary Languages. Between 2003 and 2011 he was National Director for Languages in England with responsibility for the implementation of the National Languages Strategy, working with Lord Dearing on the Languages Review. Since 1995 he has also been involved in the languages work of the Council of Europe and the European Commission and has coordinated a number of European projects.
