One of the signs of age is that we start to feel that the young use language less effectively than we do. Because we don’t understand the subtleties of the new expressions they use, we feel excluded - and tend to notice what has been lost but not what has been gained. We cultivate our pet aversions - to split infinitives, to the replacement of “have” with “of” (as in “she should of been”), to misuses of the apostrophe. It is partly a game, partly snobbery, and it ensures that there is always a large body of people ready to say that standards of literacy are falling.

This isn’t new. Ancient inscriptions have been found on Egyptian tablets of stone complaining about the standards of education and the behaviour of the young. The same gripes crop up in every generation. Comments about the decline of language skills appeared in the Newbolt Report in 1921 and again in the Norwood Report of 1943. Most newspapers have made the theme a favourite.

Nevertheless, it still has the power to menace. Most of us, after all, agree with Secretary of State Gillian Shephard that “if children are to learn to express themselves clearly and effectively and to make full use of our wonderful language, they need to be taught how the English language works”. It is what follows that is objectionable. Her idea that language study was “lost sight of by trendy teaching of the Sixties and Seventies” and that teaching free of that dreaded trendiness will somehow bring back to the diverse population of present-day Britain the language structures of a well-educated elite in the Fifties is highly questionable.

This summer, Mrs Shephard was quoted as saying that a report from the University of Southampton had prompted her to “look again at the key stage 3 tests and the emphasis the curriculum puts on correct English - basic grammar, spelling and punctuation”. Most people assumed that the report - Knowledge about Language: language learning and the national curriculum, published by the university’s Centre of Languages in Education - had been released recently and threw some light on the tests. In fact, it appeared in January 1994, and was not about the tests at all. It was mainly a discussion of the teaching of English and of foreign languages, in the course of which it stated that, although many good English lessons extending the pupils’ knowledge about language had been observed, formal grammar was more often taught through modern languages than through English. Since the report came out, much has changed. There is more emphasis on grammar, punctuation and spelling both in actual lessons and in schemes of assessment. It is also the subject of more primary and secondary in-service training courses .

What is the point of disguising the situation by presenting a two-and-a-half-year-old report as recent evidence? Presumably, it is to create panic - but panic is usually an unproductive state. It will not propel us forward but will drive us into the old strategies of the past which, as we become calmer, we remember didn’t work.

Some reports have been more immediately useful. The University of Cambridge Local Examinations Syndicate, for instance, produced a report comparing aspects of writing in 16-plus English examinations between 1980 and 1994. The researchers concluded that candidates in 1994 wrote shorter sentences and used less elaborate vocabulary than the candidates of 1980. They also found spelling to be less accurate.

The report was interesting for the amount of precise information it produced about specific areas of language and also for its recognition that it is easy to oversimplify and that there are many different aspects of English that have competing values. But this avoidance of over-simplification was predictably absent from the press coverage of the findings.

More recently, in his much publicised but actually rather dull paper on “Standards of English and Maths in Primary Schools for 1995”, John Marks worked out a set of figures to show that standards of literacy and numeracy in this country are patchy and generally too low. He advocates “significant changes in teaching styles, classroom organisation and the use of simple teaching aids such as blackboards and textbooks”.

What does he think teachers have been using for the past 20 years? It is true that teachers vary in their methods. They are always seeking new ways of organising their classrooms for more effective learning. They think hard about the balance of whole-class teaching with group work, deciding which strategy will be most useful for which stages of learning and which will most help the particular group of children sitting in their classroom.

Dr Marks, however, simply reaches the conclusion that some primary schools get better results than others and that this is because “some higher-achieving schools may well have a higher proportion of pupils with above-average ability, while some of the lower-achieving ones may have very few pupils with significantly above average ability”. This is hardly a new insight but, oddly, it is now beginning to be used to suggest that all teachers should teach in the same way.

Everyone wants standards to rise. Teachers devote their working lives to that end. It isn’t just a new idea thought up by Chris Woodhead and the Office for Standards in Education. Standards, however, are complex, and literacy skills particularly are many-faceted. It does not help to pick out one or two aspects that are easily spotted and then make the measurement of them alone stand for a statement about literacy as a whole.

Every year, there is more for children to learn - so teachers’ expectations of their pupils are compelled to keep rising. Of course it is important that we have information about standards. But we must make that information as reliable, as comprehensive and as balanced as possible, and make sure that it looks to the future as well as the past.

Anne Barnes is general secretary of the National Association for the Teaching of English.