The English-Learning and Languages Review Æ Homepage
There Ain't No Grammaticality Here
The following piece first saw the light in
1999 in English Today, No 59, as a postscript to a
debate that took place in four articles from 1997 to 1998 in that journal. I
had begun it by attacking the Chomskyan ideas that language was a
"rule-based" system and that grammatical structures were genetically
innate. Chomsky was defended, and I was attacked, by Patrick Honeybone, of the
There are some uses that provoke no controversy as, for example, when you speak of a "grammatical analysis"— of a sentence, for example. In that sense, the term ‘ungrammatical’ is unusable. Conflict of approach arises about how they can be used of examples of language. In English Today, No 55, I had explained briefly why I thought ‘ungrammaticality’ was an impossible concept. In reply, in ET57, Patrick Honeybone quoted the example *"Sid inquired me a question", saying of it "we know the syntax is wrong - it is ungrammatical." Now, what follows is almost exactly as it appeared in English Today.
If the example, "Sid inquired me a question", had really occurred in real life, it might have been said by a foreign learner, in which case the teacher might have commented "That’s wrong—it’s ungrammatical." But that would be no more than a shorthand way of saying that ‘inquire’ is not currently used in standard English in that way to express that sort of idea. There is nothing in the word itself to limit its use, no hidden force, for example, compelling words beginning ‘inq-’ to be used in certain ways but not in others. For all I know, "Sid inquired me a question" might be a normal form in some current variety of English somewhere or might have become a standard, educated form in a hundred years from now.
There is variety and continuous change in all languages, and there are no places or times where something suddenly changes from being ‘grammatical’ to ‘ungrammatical’ or vice-versa. Take the present-day standard British English examples "She married Arthur, didn’t she?", "I omitted to tell the others" and "The wall is being built": the first is ungrammatical for Welsh English—"She married Arthur, isn’t it?", the second for USA English, which does not allow ‘omit’ plus the infinitive, and the third for early eighteenth-century English, which would have said "The wall is building". And goodness knows how all three will compare with how people are speaking in five hundred years time.
The word ‘ungrammatical’ implies ‘wrong’, but in what sense? Clearly, not a moral one, but not a technical or systematic one either, in the way that "2 + 2 = 5" is wrong for the internal logic of mathematics. It only means ‘wrong’ in a social way. It implies that something does not conform to the transient linguistic conventions of some group at some time. There is nothing in language itself to decide the issue. If a sentence is comprehensible, it is bound to have a grammatical structure and cannot therefore be ‘ungrammatical’ in that sense. If it diverges from some standard usage, we would be better using the shorthand phrase ‘grammatically incorrect’ rather than ‘ungrammatical’. If the ‘incorrect’ sentence is an invention of a grammarian and no-one has ever meant it, then, although we could use it for teaching purposes to show what to avoid, we need not consider it as a real part of language.
Imagining sentences that are grammatically bizarre is a bit like imagining how a language is unlikely to develop. Certainly, we can guess that some developments in English are unlikely, such as the reduplication of the article, making "the cake" into "the the cake", but there is nothing in language to stop it; it is just a reasonable guess about human behaviour. Guessing correctly is harder. It is unlikely, for example, that the average Greek in classical times would have predicted accurately the forms that would eventually replace the infinitive, or the Dative Case.
I should not like to be taken here as deprecating standard English. I think that British English is fortunate in having a standard, educated variety as a reference point, though, like every variety, it is slowly changing and has many fuzzy edges. With my students, I am a stickler about it, always correcting them if they write "a good deal of people" for "a good number of people" or "she convinced him to do it" for "she persuaded him to do it". If they challenge me, I tell them that what they have written is not inherently incorrect, since it is comprehensible, but that it diverges from recommended standard usage.
Can we all agree, then, that: (i)‘grammatical’ can be used descriptively (e.g. "a grammatical approach"), but that ‘ungrammatical’ cannot; (ii) that neither of them can be used for defining the consequence of grammatical analysis; (iii) that both may be used evaluatively (though ‘grammatically incorrect’ is preferable to ‘ungrammatical’), but only in a loose, colloquial way to mean conforming to, or not conforming to, the vaguely defined, temporary conventions of some variety of a language?
Revised layout 20 April 2010
The Editor welcomes your comments or contributions to discussion of this article.
The English-Learning and Languages Review and its individual contributors assert their Copyright © on all the material published in it
The English-Learning and Languages Review Æ Homepage