THE COUNCIL OF EUROPE, THE COMMON EUROPEAN FRAMEWORK OF REFERENCE FOR LANGUAGES AND ENGLISH PROFILE

The Russian Federationis one of the forty-seven member states that make up the Council of Europe. The Council is probably best known for its work in the fields of human rights and legal affairs, but it has also played a key role in language education with many important initiatives, most notably its central role the introduction and promotion of the ‘communicative approach’ to language teaching.

Australian tests now report scores in terms of the CEFR levels. This helps employers or schools in other countries to understand how they compare to local qualifications.

QUESTIONS AND RESERVATIONS
The spread of the CEFR has naturally given rise to criticisms and concerns. Some have worried that the CEFR imposes teaching methodologies that may not suit every context. In my view, this concern is not justified. There is no consensus across Europe on the methods that should be used and the CEFR is not a recipe book that sets out a particular technique or insists on a fixed approach. What it does offer is sets of questions that help users to think about, describe and explain why they choose to learn, teach or test a language in the way that they do. Flexibility and pluralism are seen to be essential. If the illustrative descriptions of what learners are able to do are not suitable for a particular group, the CEFR makes it clear that users are free to develop alternative descriptions that work better for them and suggests ways of doing exactly that.
More worryingly, it has become clear that people working in different places may be interpreting the CEFR levels in very different ways. One would naturally expect that language courses and tests for different groups of learners in different countries would cover different areas of language (a course for doctors should involve different elements to a course for tour guides or a course for priests). On the other hand, if results are going to be meaningful for test users, the B2 level on one test of general language abilities should represent a very similar challenge to a B2 level on any other. If there is no shared understanding of the levels many of the potential benefits of the CEFR will be lost.
Because the CEFR is designed to apply across languages: to learners of Arabic, Bulgarian and Chinese as well as English, it can say nothing about specific words or structures that learners might be able to use at each level. Reports repeatedly show that users of the framework feel the need for more detailed explanations. They would like to know what exactly is meant by 'simple phrases' or 'a wide range of subjects' and they want more concrete examples of what learners at the different levels can do: recordings of learners speaking, samples of their writing or of the texts that they can understand.
Recognizing the need to provide more guidance, the Council of Europe has asked for 'Reference Level Descriptions' which will show in much greater detail how the CEFR applies to specific languages. For English, a good deal of work has already been done. Threshold (updated in 1990) is effectively a specification of the B1 level and other books cover CEFR A1 (Breakthrough), A2 (Waystage) and B2 and above (Vantage). All of these are available as e-books that can be downloaded from the English Profile website: www.englishprofile.org.

ENGLISH PROFILE
Launched in 2005, the English Profile is a collaborative programme of research which aims to take this work further. The core partners include the University of Bedfordshire, the University of Cambridge, the British Council and English UK, but the Profile also involves, through the English Profile Network, researchers from all over the world. A major aim is to identify key features of language use -the English actually produced or understood by learners -that are characteristic of each CEFR level:.
One strand of English Profile research is concerned with input: what learners read or listen to at each level. We are building up a picture of what learners are expected to understand and what conditions apply: how complex are the texts that learners are able to read? What range of vocabulary can they cope with? how fast is the speech they listen to? It appears, for example, that a C1 level of reading may be needed when students collect and evaluate information from a number of advanced texts -as university students need to do when learning about an academic subject.
English Profile researchers are also interested in the input learners receive from teachers and textbooks. What grammar points are being taught to learners at different levels? How many words are learners expected to know or use and in what contexts? What are learners being taught to do with the language they learn: introduce themselves? describe people? exchange opinions? This has indicated, for example, that adapting language to different audiences (being more or less formal, using idioms, slang, and technical language) tends to be a particular focus for teaching at the (highest) C levels.
A second strand is concerned with learner output: the speech and writing that learners produce. Cambridge English Language Assessment and Cambridge University Press have for many years been building up a corpus of learner language (the Cambridge Learner Corpus). This is a computerised collection of the language produced by learners taking Cambridge exams at different levels. The corpus now extends to over 45 million words. Errors of grammar or word choice are coded according to a scheme that has been refined over the years. With help from the University's Computer Laboratory, this process has been partially automated and it is now possible to search the corpus to identify at each level which error types are found most often and which grammatical forms are used accurately at each CEFR level.
Interesting and sometimes surprising findings have emerged and are being incorporated into the English Grammar Profile, which will become a valuable resource for teachers. For example, higher level learners make fewer verb tense errors, but B2 and C1 level learners tend to make more verb agreement errors (three birds is singing) than do B1 learners. It seems that learners at B1 level can accurately use questions (can/will Olga read? what does Olga read?) and negations (Olga can/will/does not drive), while only at C1 level are learners able to use relative clause types like 'the professor whose book I read'.
The English Vocabulary Profile is also based on insights into how language is used by learners at different levels. The different senses of frequent words in English that are taught in language classrooms around the world and which are typically found in learner writing at each level are listed, together with brief definitions and examples of how they are used, both by first language speakers and by learners of English. To take the word 'date' for example, at the A1 level it generally refers to a particular day as in 'the date of the class is 7 June' while phrases like 'up to date' and 'out of date' are found at B1 level.
Reflecting the Council of Europe's concern for practical language use, the English Grammar Profile and English Vocabulary Profile are linked through the developing English Functions Profile, which will show how learners at each CEFR level can use their language resources to communicate their intended meanings. What language can an A1 level learner use when he or she wants to apologise. How can a B2 level learner use his or her resources to make an apology that is more effective or appropriate to the context?
Of course, learners' first languages play a role in how they learn English, as does the kind of teaching they experience. The English Profile research pays particular attention to these. Within the corpus it is possible to search for texts produced by speakers of different languages. It will come as no surprise to Russian teachers and learners of English that, regardless of level, speakers of languages that do not have articles make more missing article errors than speakers of languages with articles.

COLLECTING MORE EVIDENCE
A limitation on what it has been possible to achieve through the English Profile to date ('to date', by the way, is at B2 level) has been the amount of data available. The Cambridge Learner Corpus is predominantly based on Writing tasks from Cambridge exams and Cambridge reading and listening tasks have been analysed as input. For the future, it is vital that English Profile should draw on the widest possible range of teaching and learning contexts so that it can be relevant to all. The English Profile partners are now making a concerted effort to gather material from a wide variety of institutions for analysis. Data collection involves partners from all over the world and a particular an effort is being made to find more spoken language data. As more evidence becomes available, the information the descriptions of learner language can be further refined.
To keep up with developments in the English Profile, please visit the English Profile website: www.englishprofile.org.