So HANDSOMELY produced that it should set a new standard for biographies, and so thoroughly documented that it must be one of the most detailed accounts yet given of a modern writer, the O.U.P. biography of Keith Douglas also presents, in its biographical material, evidence of some critical importance. This admission I make through my teeth, for in general I hold firmly to Auden's view that biographies of writers are always superfluous and usually in bad taste. Superfluous it may be to have a glimpse of Douglas at Christ's Hospital, 'tapdancing alongside the billiard table, a copy of New Verse beside him. "Anyone can write like the stuff in here," said Douglas', but I find the picture irresistible. It may even be in bad taste to be told that Douglas lost his father at an early age, and that, through the prolonged financial and housing problems that followed, the toughness he developed merely overlay a chronic emotional insecurity: but that information is valuable in view of the still unsatisfactory critical assessment of his work. Critics have mistaken his masterly verse control for a cerebral detachment. They have pointed to his succession of aristocratic heroes, from 'Famous Men' and 'Haydn: Military Symphony' to 'Aristocrats' themselves, as further evidence of an emotional detachment, without, apparently, noticing that defeat is implicit for all Douglas's hero figures, that they are votive symbols of survival much more than emblems of invulnerability. Douglas's emotional history confirms what I have always felt to be the animus of his poetry, that the detachment is not cerebral but is rather a strategy deployed against the strength of feeling, a means of controlling it and making it positive.

To see this is central, I would have thought, to any appreciation of Douglas: yet it is just this aspect of his work that has given critics most trouble. The hare started by Ian Hamilton, in his essay on the Forties (London Magazine 1964, reprinted in A Poetry Chronicle, Faber 1973), when he dismissed Douglas as 'at his best when he can be remote and speculative', has misled the field ever since: it was still running in the previous issue of Poetry Nation, when Jon Glover spoke of Douglas's response being 'almost withdrawn in its meticulous calm', and its tracks can be seen intermittently through Douglas Graham's biography. In particular, he gives only a muddled account of Douglas's irony, exaggerating its detachment and not perceiving the sometimes startling connections of feeling that it makes possible. The result is that, for all his painstaking discussion, Graham never really defines Douglas's essential qualities: in a biography intended to be definitive this amounts to a serious critical failure.

The uncertainty begins in his handling of some of the earlier poems. In discussing 'Russians' he speaks of Douglas retreating into a protective insensitivity, and 'drawing a grim virtuosity from designedly bad writing, proving with each false tone his refusal to make contact with the scene he described'. This ignores the final line, which Graham fails to quote: 'Well/at least forget what happens when it thaws'; a deliberately delayed connection, that explodes the preceding ironies. Irony, in fact, is too intellectual a term to catch the full quality of Douglas's voice here, which comes closer to the sardonic.

There is a similar mishandling in Graham's discussion of 'John Anderson': he speaks of the poem's 'immunity from reality', first signalled through the clichés 'the bloody dust' and 'desperate final stare', which 'dissociate the poet from what he portrays'. The poem's beautiful final transformation is only possible, he argues, because Anderson's death is essentially unfelt. On the contrary, I would argue that the transformation only has meaning in so far as the death is felt, which is not to say that it is felt uncritically. Rather, the ironies gather around it: death occurs 'on a fine day . . . and warm' ('the bloody dust' has to retain a certain impact for that irony to work) and is seen as a helpless disarticulation (again, 'his desperate final stare' is picked up in 'he only eyed the sun', by which time it is only half-ironic). The poem's final resolution thus depends upon our sympathies being enlisted very precisely, not for the death itself, but for the indignity of it. The satire does not deflect us from the pathos but leads us into it. It is just this quality of discriminate feeling that makes Douglas's work so unusual: when critics misread it as a lack of feeling they obscure a delicate and valuable balance. Ian Hamilton, for instance, speaks of 'reticence stiffening into the tight-lipped insensitivity of the officer's mess' - a criticism that Douglas had already encountered in his lifetime and quite rightly dismissed, as being itself 'astonishingly insensitive'. Douglas Graham nowhere establishes the importance of discriminate feeling to Douglas's achievement, though he does finally bring it into focus in a good discussion of 'Aristocrats', in which he shows how Douglas builds his compassion 'out of a sharply evaluative irony'.

Comparatively slight as they are, 'Russians' and 'John Anderson' indicate that even before he had experienced action, Douglas had sensed that death in battle is not simply death but distortion, and had already evolved his method of dealing with it: a kind of documentary anger, whose expression is made possible by a light sardonic tone, by a sleight-of-wit. The anger in Douglas is something that Douglas Graham never mentions; he merely points out that there is no protest: but anger is a more integral emotion than protest, and it is there in 'Vergissmeinicht', even in 'Simplify Me When I'm Dead', and is part of the driving force of 'Christodoulos', 'Egypt', 'Cairo Jag' and 'Dead Men'. 'Egypt' should dispose once and for all of the charge of cerebral detachment. The verse is so achieved as fairly to be called elegant: yet clearly that elegance does not represent coldness, but an articulation of feeling at its most intense. It is a perfect example of compression and formal control creating, in Douglas's own phrase, 'significant speech'. The cursory treatment that Graham gives it suggests that he isn't sensitive to this quality in Douglas either, and, incredibly, that seems to be the case. Yet, as Ted Hughes pointed out in his introduction to the Selected Poems, Douglas's style represents an important advance, the organisation of everyday diction and colloquial rhythm into a language that was immediate, flexible and eloquent. His work remains a model of how contemporary poetry should be written.

One can see the style beginning to form in 'The Prisoner', to which Graham gives the now familiar reaction, remarking on its 'disquieting detachment'. Once again, a dangerous half-truth. The poem has a considerable emotive charge, but contained within the language. In the final lines intense verbal organisation creates an active force that could almost be called expressionist:

But alas, Cheng, I cannot tell why,
today I touched a mask stretched on the stone

person of death. There was the urge
to break the bright flesh and emerge
of the ambitious cruel bone.

The strong syllables in 'urge' and 'emerge' make one very aware of the polysyllabic structure, particularly of the vowel changes, of 'ambitious cruel', which are then set against 'bone'; while between the hard consonants the single 'I' of 'cruel' acquires an almost threatening resonance. Detachment is merely the first stage towards enactment, towards a poetry in which the energy of personal expression is subsumed within language of an achieved physical shape. Graham himself shows, by reference to the early drafts, how the writing of 'How to Kill' involved the suppression of personal vulnerability, of the image of Douglas's own death, in favour of an objective enactment of vulnerability in general: while in 'Simplify Me When I'm Dead' one can see that the strength of the final stanzas comes from a characteristic combination of an objectifying rhetoric, in the rejection of self, with intensive formal devices; the inversion of word order, shortening line length, alliteration and internal rhyme. The poem's momentum proceeds directly from what Alvarez called Douglas's 'driving use of his talent'.

A clue to Douglas's strategy perhaps lies in his own phrase 'analysis is worshipping'; in other words, in the belief that the imagination's degree of response lies in its degree of control. Thus his conviction that one must enquire accurately even into those experiences that seem inhuman, and destitute of meaning: 'I will sing/of what the others never set eyes on'. This is not to deny feeling: it is to give to feeling a foundation in fact, and at the same time to make of those facts something that can exist in spite of them; to create something to stand against what is 'slaying the mind'.

Since this biography fails to establish a secure base for Douglas's reputation, it becomes all the more necessary, as Vernon Scannell pointed out in his New Statesman review, to refute Ian Hamilton's curious devaluing of Douglas in favour of Alun Lewis. The central thesis of Hamilton's Forties essay, that for poets caught up in a remote global mechanism of war the essential quality was the determination 'to be articulate and intelligible, a sensitive, fully-functioning participant' in a world where 'personal identity could only be preserved by a sustained effort of sheer accuracy', has an emphasis with which one would not want to argue. Indeed, as we become similarly enmeshed in economic mechanisms beyond our control, it is an emphasis that we would do well to remember. None the less, Hamilton seems largely to have mistaken Alun Lewis's intentions for their achievement.

Lewis, like Douglas, eschewed a staff job and sought the experience of action: and just as Douglas felt compassion for the fellahin of Egypt, so Lewis sympathised strongly with the peasants of India and detested the pretensions of the British Raj. Yet oddly, although Lewis came from South Wales, an area of strong working-class traditions, he seemed unable to experience Army life or India with the same level perspective that Douglas preserved. In letters home he exalted and patronised his men, his 'Welsh boys', whereas Douglas simply got on with them. And yet, writing of himself 'I felt sensitive and somehow apart', Lewis could write of his fellow-soldiers: 'the cheapest dance song utters all they feel'. This imbalance in his own attitudes dehumanises what he sees, quite as much as the Army threatens to dehumanise him: if one compares Lewis's 'Infantry' with Douglas's 'Christodoulos' or 'Dead Men', his 'Jason and Medea' - 'the cheap and placid aroma of her smile' - with Douglas's 'Cairo Jag', or his 'Indian Day',

Dawn's cold imperative compels
Bazaars and gutters to disturb
Famine's casual ugly tableaux.
Lazarus is lifted from the kerb [. . .]

with Douglas's 'Egypt', one finds at every point that Douglas is at once more incisive on human corruption and markedly more humane.

Lewis's misjudgements do not occur to anything like the same extent in his short stories. The imbalance seems to spring from his almost Victorian conception of poetry, which involves him in a certain pose, a certain elevation of the tone. This is reflected in his high-flown diction, so that, for all his emphasis on being true to the particular, abstraction sits at the very heart of his method: his Infantry are 'Sharing Life's iron rations', involved in 'an oceanic tide of Wrong', thinking back to a time 'before the soul of things turned bad'. It is no wonder that his best personal lyrics are highly literary in style, or that when he does manage to elevate a realistic treatment successfully, as in 'The Mahratta Ghats', the rhythm and diction seem to belong to the years before the First World War rather than the middle years of the Second.

One begins to question, in fact, just how accurate Hamilton's outline of Lewis's development is. He suggests that the development between Raider's Dawn and Ha! Ha! Among the Trumpets is towards 'the devitalised and prosy', and remarks 'it is an image of Lewis's personal predicament that this should be so'. In fact Lewis's two most prosy poems, and two of his most successful, 'All Day It Has Rained' and 'To Edward Thomas', are in Raider's Dawn. Ha! Ha! begins with poems in the same style, 'The Departure' and 'On Embarkation', but moves away from that towards a more measured use of the high literary style, which accommodates his response to India well in such poems as 'Peasant Song'. A kind of moral maturity is achieved in Ha! Ha! , as in the late stories, but it seems to reflect the impact of India more than a coming to terms with the Army:

What stays of the great religions?
An old priest, an old birth.
What stays of the great battles?
Dust on the earth.

This is just one aspect, however, of that unpredictable mechanism that dispatched poets at random, Roy Fuller to Africa, Alun Lewis to India, Keith Douglas to the Middle East. No experience could conclusively meet Hamilton's paradigm because no experience was typical. In the position of having a valid point to make without a sufficient exemplar, Hamilton seems simply to have adapted Lewis to fit the bill.

It is a pity that in doing so he appeared to value Lewis's theorising on integrity above Douglas's realisation of it through form. Douglas's poetry is almost an imitation of action in the vigour and compactness of its language, yet fully human in its response. It is poetry, not of a studied detachment, but of an achieved intensity. Accepting death as levelly as he accepted all experience, Douglas created a poetry that would be hard and good when he's decayed. Lewis's work is wishful by comparison.