schwa character (an upside-down e) for the similar but var ious vowel sounds that together are the most common in English. The system used by Fowler (and, again, COD up to the seventh edition) is simple, yet far more accurate. It


IN THIS GHASTLY world we seek solace in order, beauty, elegance, truth and justice. There is little to be found in the works of man, but one small, enduring refuge has been available since 1926. The first edition of H W Fowler’s A Dictionary of Modern English Usage can be seen in retrospect as a defiantly post-Great War work. Fowler set out to restore order and pride in the British spirit, then as now in darkness lost. Galsworthy too fretted that the time was ‘tight in money and loose in morals’, as he wrote The Forsyte Saga, which chronicles the rise of incertitude. Fowler composed a book of conduct the certainties of which are as absolute as Scripture. Fowler will always tell you to do the best thing – and he will also tell you to do the decent thing when it would be unkind to do the most correct thing, such as not speaking better than your neighbours. As the Bible fell into desuetude, it was replaced, in the hearts of those newly secular yet habitually reverent, by Fowler. Not that it is necessary to choose between these companionable complements.

Fowler has survived time and his revisers – all three editions (1926, 1965, 1996) are still in print. Best is Oxford’s new printing of the first edition, with an excellent introductory essay and notes on the entries by David Crystal. The most annoying, though still indispensable, edition, is the third – the only one to be called ‘Fowler’s’, and the only one to contain very little of him. R W Burchfield, one of the great lexicographers of his day, made several catastrophic mistakes in his revision. One was to adopt the IPA pronunciation scheme. Of all phonetic respelling systems, the IPA is the dodgiest. But British linguists and lexicographers have adopted it even in elementary dictionaries, thereby guaranteeing that the ignorant will never learn how to speak correctly, since the symbols are incomprehensible and unlearnable. Fowler’s scheme, used also in the Concise Oxford Dictionary (COD) up to the seventh edition, is much simpler; it is more comprehensive, accurate, readable, and intuitive.

Two intractable problems must suffice to show why the IPA should be rejected. A special symbol, small capital I, at reduced height (I), represents the short ‘i’ of pin; and that same letter, preceded by a single-storied ‘a’ represents a long ‘i’, as in pine. A special legibility problem asserts itself when I is followed by n. The IPA representation of pin (pIn) looks very much like pm, and pine comes out as ‘paIn’ or pam or pain. These scanning errors plague even readers with perfect vision. The right hand serif of the small cap I meets the left-hand serif of n; together they ineluctably form m. The spread of ink common in book printing makes the problem worse. This gross defect alone makes it impossible to condone the IPA.

Another problem is that the IPA uses a single glyph, the consists of the following possibilities, with italic to represent the weak vowel: about, item, edible, gallop, circus. The COD seventh edition points out that these variants of schwa will only be differentiated by precise speakers. But shouldn’t a dictionary, especially an Oxford dictionary, make such distinctions? How much better to have these vowels particularised, rather than represented by schwas – a situation that worsens in some American dictionaries where the schwa is defined as the ‘uh’ sound of ‘duh’: uh-bout, ed-uh-ble, gall-up, circ-uhs. Now let us consider the great book’s military functions, which were utilised by Churchill. He objected to the language in which the plans for the Normandy landings were written, and instructed the Director of Military Intelligence to read Fowler. Marriage is war, and Fowler is the deadliest weapon my husband holds against me. Oh, the indignity. I am an Oxonian classicist; he is a barely schooled American, whom I rescued snorting and stamping from the prairies. Yet he will tell me, Fowler in hand, that I ought to pronounce ‘myth’ as ‘mythe’. When he was reading my novel, O Caledonia, I observed that Penguin replaced all my English ‘ises’ with vulgar ‘izes’. This was doubtless for the American market.

‘I’m afraid you have got that wrong. Penguin follows Oxford usage. You ought to read Fowler on “ise” and “ize”. I shall read it for you.’ That article is short, but I fell into a trance, for Fowler’s arguments were bulletproof. I woke to hear my spouse demand: ‘Did you not learn “ize” at Oxford? Or don’t they teach it at Somerville?’ Why did I ever let a copy of this baneful book into the house? There is no recourse; its correctness is absolute. Fowler prevails even over David Crystal with his computers and all the scientific apparatus of modern linguistics. Crystal is anxious, in his introduction, to show up Fowler’s flaws – surely one can be found, seventy-five years after publication? See him pounce:

At literally, we read that we ought ‘to be at pains to repudiate’ the emphatic use of this word, but then we encounter it at negotiate, where Fowler condemns a usage which, he says, ‘stamps a reader as literarily [sic] a barbarian’. But ‘literarily’ is not Fowler’s misspelling of literally; it is his correctly spelt use of a word which means ‘in a literary manner or respect’ (OED). Crystal could not have made a more embarrassing mistake. Upstairs, in Heaven, Fowler generously affects not to notice. That would not be the thing. ❑