It's not often I envy science teachers but in one respect I do: there is a general agreement that under the umbrella of "science" lurks at least three, if not four or five, subjects. The acknowledgement that "science" covers more than one discipline has many implications, not least more lesson times at GCSE than for other subjects.
English, on the other hand has always sought to corral a variety of disciplines under one heading. Glance at the curriculum, however, and it is plain that the subject known as English at school, if studied at university, might be called literature or linguistics, creative writing, drama or media studies, to name but a few.
Drama can be taught separately at school and sometimes media studies can be, too. But the most likely place for each to be found is in an English lesson because all come under that broad heading in the national curriculum.
And then there is linguistics. At AS, A2 and degree level, the study of language is happily seen as a discipline in its own right. Lower down the academic food chain, however, grammar - a subset of linguistics - is seen as integral to the subject "English" even though it is ill-defined at this level. That is because, these days, the most likely justification for teaching grammar -found in the letter pages of The TES last week and advocated in the literacy framework - is that it improves pupils' writing.
This particular belief in the benefit of grammar is, however, mistaken. A recent report by Richard Andrews, of York University, has no quarrel with grammar teaching per se but it does show that there is no significant correlation between good writing and a knowledge of grammatical terminology. Nor is there any direct link between bad writing and ignorance of the parts of speech.
Writing and understanding parts of speech demand two very different skills: one is about the use of language, the other its analysis. So the fact that the latter does not automatically lead to the former should not surprise us. What is, perhaps, more surprising is that anyone thinks it should. The York Report is just the most recent in a series of similar studies over the past 50 years that have come to the same conclusion. These have been based, not just on one study, but on reviews of a wide range of research. All have come to a similar verdict. We should now stop debating the issue.
For there is a much better justification for teaching grammar: it's worth studying for its own sake. Children should know something of how the language they speak works, how it has evolved over time and how it continues to change in common usage. They should examine the principles that underpin the way we communicate and reflect on issues such as register and discourse.
The question, then, is how much time in an overcrowded English curriculum should be allotted for such study? In a week that has seen the Qualifications and Curriculum Authority comment on the lack of fiction reading at key stage 2 and 3, it seems clear that the balance is wrong. As long as we persist in the erroneous belief that naming parts of speech improves writing, we will not find time to read fiction to children and each new novel will be just another primer. Let's keep grammar in its place and make room for stories.
Bethan Marshall is a lecturer in English education at King's college, London