Teaching school students how to write has been one of the main tasks of education for hundreds of years.
One view is that writing should be taken down to its 'nuts and bolts'. Students should be told what these nuts and bolts are called. They should do nuts-and-bolts exercises involving spotting the nuts and bolts in specially devised sentences and phrases. There should be nuts-and-bolts tests. There should be writing tasks where students should show that they can include the nuts and bolts that they've learned. Job done, the students will be able to write.
Are there problems with this?
First, let's deal with the nuts and bolts question. What are we talking about here? Usually, this means the so-called 'rules' of language, the names for the words, and longer clusters of words: phrases, clauses, sentences, paragraphs. Intermingled with this, there will be some talk of 'functions' which may mean the functions of words within phrases, clauses and sentences, or it may also mean functions in terms of the jobs that words, phrases, clauses and sentences are thought to do in more human terms eg ask questions, issue commands etc.
This is what education calls 'grammar' but which in actual fact is a very limited descriptive apparatus that is restricted to descriptions of sentences in formal written standard English. If this is how this 'education grammar' was served up, we would at least know how limited its field is. That's to say, it doesn't cover the language of conversation and speech, nor the written languages of song, poetry, plays, films, TV drama, advertising, signage, slogans, mottos.
So when we say school-based 'grammar' teaches us the nuts and bolts of language, this is at the very least an exaggeration or distortion. In addition, it may or may not teach us the nuts and bolts of formal, written sentence language.
Let's deal with this first. Where do we find these formal, written sentences that most people are agreed that students should learn how to write? You or I can make a list: most narration in most fiction; most non-fiction - history, geography, science, philosophy, self-help, gardening, cookery, sport etc; most instructions and information in relation to the things we use - medicines, fridges etc; newspapers, magazines and online versions of these; administration - government, management, the law and justice system, finance and reports from within these worlds...and so on. It's clear from this list that we ask of this written, standard English to do what society regards as the important stuff. In fact, some linguists call this 'prestige'.
The question is, does teaching (and testing) the grammar of formal, written standard English help students to write this way?
One huge body of evidence emerged after 20 years or more: the results of the old O-level exams that the grammar school students who lasted in school till they were 16 and the fairly small cohort of secondary modern school 16 year olds in the post-war years. The three main questions of the English Language paper were 'Precis', 'Grammar' and 'Composition'. The précis was an exercise in reducing an unseen text to about a third of its original length. The grammar question was to answer questions according to what was at that time (not now) thought to be the most important elements of formal, written standard English. The composition question was a piece of formal writing. Students were set an unseen title or theme as with say, 'Trees' or 'An Enjoyable Weekend' or 'Fox-hunting'. In other words, the topics varied across non-fiction themes, to personal and to subjects that you could debate. It was made clear to us that this writing had to be correct - according to the rules of grammar, spelling and punctuation we had learned over the previous nine years or so of schooling. I don't know how marks were awarded or detracted on this matter, but the marking also addressed the matter of whether candidates 'wrote well' according to criteria agreed by examiners and education authorities.
After more than 20 years of this, involving of course tens of thousands students, the examiners could find no correlation between the scores on the grammar question and the scores on the composition question. Of course, this flies in the face of the claim that teaching the nuts and bolts leads directly to enabling students to write formal standard English.
Before I jump to the conclusion that this proves that there is now an open and shut case that teaching the nuts and bolts does NOT ensure that students end up writing formal standard English well, I should complicate matters.
1. The lack of correlation might be that there was something wrong with the teaching methods of teaching grammar and if this was tweaked, a better correlation would result.
(To which I would say: from my personal experience at a grammar school 1957-62, the teaching was thorough, consistent, constantly marked. We were given systems and routines every week for five years (eg box analysis, clause analysis and that because we also all learned at least one foreign language for the whole 5 years we were exposed to the terminology of eg cases, subject and object, conjugation, tense and so on in these lessons too.)
2. The lack of correlation might be because the criteria for marking the composition was not finely enough tuned to loading the marking towards those students who produced grammatically correct compositions. Similarly, that it was loaded towards subjective judgements about 'good writing' or 'writing well'.
(To which I would say: even in the most narrow way of marking grammatically correct writing, the issue of 'good writing' has to be factored in. It is, after all, quite possible to write incomprehensible gobbledegook that is grammatically correct. Let me have some fun doing it on the topic of 'Rain' (one that I remember having to do in English).
Rain is a strange banana. If telephones capture lemurs, the rain which has up till now been solid, may explode. However, rain can also be found on the ocean floor. It was Jacques Cousteau who said, 'Better the shark you know than the onion you don't.'
I'll leave that one with you.
Now let's switch tack and consider the argument that there are better ways of teaching the grammar of formal, written standard English with a view to teaching how to write formal, written standard English.
Here are some them:
1. Demand that students in class only talk in 'complete sentences'. They must not speak using slang or teen talk. They must try to speak with the accent of southern British, educated English (BBC English or Queen's English, so-called). They must not use 'dialect' language eg Cockney, Yorkshire, Geordie etc.
(To which I would say that the problem with this is that it assumes people speak standard English. No one does. Not even Boris Johnson. Speech involves repetition, self-correction, a lot of umming and erring, insertion of many phrases like 'kind of', 'you know', adjustments like 'well', and 'you see'; tailing off into seemingly unfinished parts; using words and phrases that aren't part of complete sentences; using gesture, facial expression and tone of voice to indicate meaning and intention, to structure utterances in such a way as we tend to put the theme/subject/topic at the front and the provisos, conditions or caveats after; we interrupt each other mid-flow and so on. It is unrealistic to expect conversation to follow the 'rules' of written standard English. The main way we hear standard English is in the scripted talk of radio and TV announcers or in audio books or indeed, people reading to us.
The further problem is pyscho-social and political. We all speak a dialect. Boris Johnson's dialect is the dialect of a public school educated person who has stayed in that milieu for the whole of his life. Other dialects don't have the 'prestige' of his, but they are no less part of the identity and ethnicity of people. The question we have to address is whether it's advisable to tell young children and students that they shouldn't use spoken language that is part of their identity and ethnicity. (I'm not talking here about obscenity or offensive language.)
When we ask all students to write formal standard English, it is for every single one of them a different form of language from the one that they speak. There is a valid argument for saying that some versions of spoken English diverge more widely from formal written standard English than others. The question then, is whether it's right and worthwhile to ask of those who diverge most to self-correct. An example would be the London 'we was' construction. Many teenagers in and around London say this. Do we know that if we ask of them to never say 'we was' in school, that such students will write formal standard English better? Do we have the evidence for that? Are there other ways of exposing students to the idea that when we write formal standard English we do not ever write 'we was'?
One suggestion is that we teach codes, registers, dialectology and what is in effect a form of sociolinguistics, particularly in relation to dialect. That's to say, we ask of students to compare speech and writing, to compare their versions of spoken English with the English of formal, written standard English. The idea here is that we give students the 'toolkit' to see that all language involves 'choice'. Ultimately, no matter what we are told in school, how we speak and how we write will always be as a consequence of what we want to do. After all, in various ways students in schools have been told for the last 100 years or more that speaking in certain ways are wrong, and yet regional and class variations in speech go on including forms like 'we was' and we go on producing slang words, phrases and expressions that don't find their way into formal, written standard English. Education might be about equipping students with what it takes to make informed choices about all this.
2. Try to integrate the 'grammar' teaching more with 'writing or a purpose'. I'm sympathetic with this because it integrates the abstractions of grammar with real language use. It veers away from the absurd invented sentences that students are invited to read or write in grammar exercise books, tests and high stakes exams. It also takes us towards seeing that the language we use or create comes in what Michael Halliday called a 'mode' - a word he used to cover the types of text we read and create, whether that's in terms of genre or overall purpose or form. So even with formal written standard English, there are variations between, say, the narration of a piece of adventure fiction, an explanation of a statute, a political manifesto and the instructions on how to take thyroxine pills. The grammar of these different types of formal written standard English is indeed worth looking at.
My argument with this is that grammar may be a necessary condition for this type of writing but by no means a sufficient one. I've already touched on this by referring to Halliday's 'mode'. Because all language, and within that all formal written standard English comes in a form or type or genre (or hybrids and combinations of these), there will always be other nuts and bolts other than what is available to us through grammar. One way to think of this is in terms of bottom-up versus top-down. Or, if we don't want to think of it as an opposition - how do we combine bottom-up with top-down so that we end up with the 'whole'?
If I accept for the moment that 'grammar' gives us bottom-up ways of constructing formal written standard English (I don't but I'll leave that to one side), what gives us the top-down of overall purpose, theme, type, genre of the passage that we're trying to write?
In my experience the best way for top-down is through immersion, imitation, investigation, interpretation and invention.
Immersion is reading loads of standard English and hearing it read to us - as much as possible both in and out of school ie a lot of 'reading for pleasure' and all that that entails.
Imitation is just that: let's write like the instructions of on a box of pills, let's write like a sports journalist in the Daily Mirror, let's write like the opening of a novel etc.
Investigation is asking students to find patterns in formal written standard English. Some of these tools are: prosody, lexical field, patterns of imagery, patterns of 'affect' (language about feeling and/or designed to make us feel something), code-switching, use of rhetorical devices like hyperbole, bathos etc, thematic similarities and contrasts, using terms like instruction, command, imply, suggest, evoke, reveal, digress, headline, foreground, focalise, time-frame, point of view, implied audience, message, elaborate, figurative language (eg metaphor, simile, personification)/literal language, detail, specific, general, abstract, questioning, tone, speculative, argue, emotive, justify, illustrate, instruct, analytic, imaginary, evaluative, inclusive, non-sequitur, analogy, example, tentative, theme, motif, symbolic, literal, allusion, illusion, echoing, pre-figuring, time-frame, red herring, point of view...and so on.
In my experience, learning this material is best done from two ends at the same time - investigating from one side and being told (instructed) from the other.
Investigation can lead to interpretation - what is this piece of writing? What is it setting out to do? Does it succeed? (if yes, how?If not why not?) Another form of interpretation is invention.
Invention can of course be in many different directions and methods. But if, let's say, we are looking at the formal standard English of popular music criticism. We can have a go at inventing a band, inventing its music and style and writing a piece that would be acceptable in a magazine or online. We can then use the top-down and bottom-up tools to see if it fits what seems to be required for such an article.
My argument is that we won't get that bit of writing right unless we do both.
