Lashed into line by grim grammarians

These six reprints appear in the new Oxford Language Classics series. Judged as language textbooks rather than documents of the past, they are of distinctly uneven usefulness. But the first thing both teachers and students should beware is the inflated language of Oxford University Press's cliche-ridden publicity blurb, which misleadingly describes them as "seminal works", providing "all-round essential guidance on the grammar, linguistic structure, and syntax of the English language". They do nothing of the kind. Nor are they "introduced by some of today's language experts", unless the likes of Roy Hattersley, Ned Sherrin and Matthew Parris regard themselves as language experts, which I doubt.

This mixed bag ranges from the undoubtedly important to the frivolous. Into the latter category falls The Devil's Dictionary , a rather laboured joke by a cynical San Francisco journalist. It began as a newspaper column in the late 19th century and earnt Ambrose Bierce a somewhat overblown reputation as a wit. His "shocking" definitions include:

"J is a consonant in English, but some nations use it as a vowel - than which nothing could be more absurd." "Peace, n. In international affairs, a period of cheating between two periods of fighting." Why anyone, now or then, should regard this work - or be persuaded to regard it - as a "language classic" is difficult to see, and the new introduction, mainly devoted to biographical details, fails to solve the mystery.

Into the category of important texts, on the other hand, falls the work of a more distinguished journalist: William Cobbett's Grammar of the English Language , first published in 1820 and written in the form of a series of letters to his "dear little James", a lad of 14. It is worth reading if only to put present-day complaints about the dumbing down of education in some kind of perspective.

I would guess Cobbett would be hard going, if not almost incomprehensible, to most 14-year-olds today, and to a fair number of university students as well. This is not because he is poor at elementary exposition - on the contrary, he is excellent - but because educational fashions have outlawed the traditional metalanguage of English.

Cobbett plunges young James deep into this terminology and sometimes even assumes his 14-year-old is already partly familiar with it. Roy Hattersley, in his witty and light-fingered introduction, seems to me to have missed out altogether on what makes Cobbett a landmark text in the history of English. First, it reflects not, as Hattersley suggests, Cobbett's "didactic" proclivities, but his recognition that "the Soldier, the Sailor, the Apprentice and the Plough Boy" must indeed become masters of English if their social and political aspirations are to be fulfilled. Second, Cobbett's Grammar was written long before Disraeli's famous education act that introduced the three Rs, and it recognises the primary duty of the parent to make sure that all the "little Jameses" could stand on their own two feet when it came to expressing themselves in their native tongue. That duty many parents today have forgotten.

In between Bierce and Cobbett fall what are nowadays called the "verbal hygienists", of which the Fowlers are paradigm examples. Their works are classics; but classics only as illustrations of the pretentious prescriptivism that dominated British thinking about language in the early 20th century. They must have done more to make honest citizens fearful of committing some dreadful linguistic solecism than any other pair of martinets since the fall of the Roman empire.

It goes without saying - or should do - that as guides to the usage of today, the Fowlers are useless. If their works were ever to be recycled nowadays as course material, it could be only on the basis of exercises in determining how language usage has changed. The Fowlers have an understated linguistic arrogance that defies comprehension. As educational examples of how to think about language, these manuals would be better burnt than reprinted in paperback at £7 or £8 a copy.

Robert Burchfield's book is a dull establishment account of the history of English. Years of toil at the desk of drudgery have left the former editor of the Oxford English Dictionary with little ear for the expository registers of the language. His prose is a jarring cacophony of obscure technicalities ("aphesis", "dissimilation", and so on) and a kind of coy philological Christopher Robinspeak in which words can be "brought out and dusted" and "glued together" to form sentences, but sadly have parts that eventually "decay". The brief chapter on syntax reveals the author out of his depth in any forms of linguistic analysis that postdate his own schooldays. At the end there is a stodgy update by Burchfield's current successor at the OED, John Simpson. It takes no account of how present-day English is being changed by immigration, globalisation or the media, and its bibliography is years out of date.

Noblesse Oblige is a collection of essays edited by the arch-snob Nancy Mitford. In many ways this is the most interesting volume in the series, because it sets in aspic a certain "sociology" of language that was current in the postwar period. These were the days of "U" versus "non-U", a distinction to which A. S. C. Ross, whom Mitford blandly patronised, lent whatever academic reputation he then enjoyed. All the participants are in turn patronised in the introduction by Ned Sherrin, to whom the whole linguistic merry-go-round seems to be a name-dropping game.

It would not be impossible to structure a course in modern English on the basis of these six texts. But it would be a course skewed by the omission of anything addressing the fundamental issues of language that were raised in the course of the 20th century.

Roy Harris is emeritus professor of general linguistics, University of Oxford.