The Varieties of Metaphysical Poetry 
by T.S. Eliot, edited by Ronald Schuchard.
Faber, 343 pp., £25, November 1993, 0 571 14230 3
Show More
Show More

Eliot’s Clark Lectures ‘On the Metaphysical Poetry of the 17th Century with Special Reference to Donne, Crashaw and Cowley’ were commissioned in 1925 and delivered at Trinity College, Cambridge in 1926. Since then they have been famous for not being available. Eliot intended to make them into a book called The School of Donne, which would be far longer, partly because – on the face of it unexpectedly, given his title – he wanted to write a lot more about Dante. On Dante, as he remarked in a preface, the whole of his argument depended. But this book was itself to be merely part of a larger project, a trilogy of which the other volumes would deal with the Elizabethan drama (on which he had already written a good deal) and the Sons of Ben. The whole would be known as The Disintegration of the Intellect, a title suggesting an almost Spenglerian ambition, and a scope beyond the usual range of literary criticism as he himself claimed to understand it.

Eliot never did enlarge the Donne book, though in 1929 he wrote a long essay on Dante, by now much more important to him, the model of the great poet. The other members of the trilogy lost their original titles and became The Outline of Royalism and The Principles of Modern Heresy, but they, too, remained unwritten, at any rate in the promised form. By 1931 he had come to think that the occasion for a book on Donne had passed or been seized by others; but lacking time to write brand new lectures, he gave a reduced and somewhat altered version of the old ones as the three Turnbull Lectures at Johns Hopkins University in Baltimore in 1933.

At that time his main reason for being in America was to give the Charles Eliot Norton Lectures at Harvard, duly published as The Use of Poetry and the Use of Criticism. In a year that would have seemed laborious even to a writer not suffering from marital disaster and general ill-health, Eliot added to his Harvard commitment the University of Virginia lectures called After Strange Gods: A Primer of Modern Heresy, a topic that interested him more than his Harvard theme. These three lectures became a book, published at once but never subsequently reprinted. It is hardly surprising that he chose to meet his obligation at Baltimore by potting the old Clark series. The Turnbull Lectures, entitled The Varieties of Metaphysical Poetry, are here printed for the first time. This was worth doing because they contain some lively new material about himself, especially in the last lecture.

To be a Turnbull lecturer you were supposed to have achieved distinction as a poet, and it is interesting to overhear Eliot instructing his audience on a topic he often meditated, the difference between great poets and non-great poets:

There is no reason why one should not try to write great poetry, except that great poetry is not written in that way: I mean that if one cares enough about poetry, ‘greatness’ is not the aim or the criterion. The aim is not to emulate Shakespeare or Homer or Dante or anybody else; for if and so far as one is a poet these criteria and ambitions are nonsense. Poetry is in this respect like science: the aim of the true poet is not to be a ‘great poet’, but to make a contribution to poetry: merely to say the true thing at one’s time, to say the thing to be said in the circumstances, in the right way ... Greatness is not a state that poets really seek; greatness is a matter, so far as we are concerned, of chance, of what happens afterwards when we are dead; and that depends upon a great many things outside of ourselves.

He concludes that ‘the important poets will be those who have taught the people speech,’ alluding for the first time to the line of Mallarmé to be remembered years later in ‘Little Gidding’, ‘to purify the dialect of the tribe’. This act of purification – no less than ‘a perpetual return to the real’ – is to be achieved by devoted technical effort on the part of poets; and Eliot does not forget that when he insists that literary criticism has to do with the study of technique, and is only secondarily involved with sociology or psychology or anything else. This principle he tries but, perhaps inevitably, fails fully to honour in his treatment of Donne in these Clark Lectures; there are a few, but too few, instances of close literary criticism, nose to text, brisk, dogmatic, arguable and fun. Nevertheless, he insists that the lectures are works of literary criticism, not meant to be anything else, and certainly not professing scholarship; but of course he cannot draw the line firmly, especially since he is persuaded that the decline of poetry is a symptom or consequence of a more general intellectual and cultural collapse that was increasingly preoccupying him.

In 1961, when all intention of publishing The School of Donne had vanished, he resigned the title to A. Alvarez, who used it for a book published in that year. The lectures, celebrated but largely unread, remained in two typescript copies, a top copy in King’s and a carbon in the Houghton Library at Harvard. Now they appear in a scrupulously edited text. The editor conjectures that they will ‘have as much impact on our revaluation of [Eliot’s] critical mind as did the facsimile edition of The Waste Land (1971) on our comprehension of his poetic mind’.

The Clark Lectures are generally thought to be, so far as critical celebrity in this country is concerned, the big ones. The series have in recent years got shorter, but in 1925 there were normally eight lectures, enough for a book – for example, in the following year E.M. Forster gave the lectures that made up Aspects of the Novel. Ronald Schuchard, the present editor, is naturally curious as to why a 37-year-old American, a banker, and, as a critic, practised in the journalistic essay rather than the full-scale academic book, should have been chosen for this high office. The short answer is that he had made a great impression with The Sacred Wood and The Waste Land, and had achieved a small but choice audience as editor of the Criterion. The newly established English Faculty at Cambridge, and especially I.A. Richards, had taken him up, and Middleton Murry, his predecessor in the lectureship and still a powerful name, had nominated Eliot as his successor to its sponsors, the fellows of Trinity. They had already offered it to A.E. Housman, himself a fellow of the college. When he turned it down they called on the youthful banker.

The occasion was formidable enough to make anybody nervous. The Trinity fellowship at that time contained, among other great men, J.G. Frazer and Francis Cornford, both of whose work had been an inspiration to Eliot, as well as G.E. Moore and Housman, who, we are told, sat in the front row at all the lectures. Also present were Richards and F.R. Leavis, then a keen admirer; some known enemies, especially F.L. Lucas; and many undergraduates, especially women. The undergraduate William Empson, who did not go to lectures on principle, attended the morning-after discussions of them.

They seem not to have been a spectacular success. Some judged them too recondite, and others were unable to hear much of discourses delivered in a low tone and in a cold hall with bad acoustics. The audience, as lecture-series audiences will, dwindled week by week. Eliot’s own confidence in the lectures was somewhat shaken by the criticisms of his friend Herbert Read and of the Italian polymath Mario Praz, then teaching at Liverpool, whose work in the same field Eliot greatly admired. In 1937 Eliot pronounced them ‘pretentious and immature’. Should one now agree with Schuchard that they contribute to our understanding, if not of Donne, then of Eliot?

Eliot had already published some famous brief remarks about Donne and some of his contemporaries: ‘The Metaphysical Poets’, a review of Herbert Grierson’s anthology Metaphysical Lyrics and Poems of the 17th Century, is at least as well-known as anything else in his critical work, and for many years it set the agenda for discussion of the subject. By 1921 the revival of Donne was well established, and Eliot’s essay was in some ways less revolutionary than it later seemed; but nobody had so persuasively suggested that the sort of praise accorded the best of Donne – those poems or moments of poems when thinking and feeling could be said to coexist – might be systematically worked into a theory of cultural decline from the 13th century to our own. The inability of later poets to achieve Donne’s fusion of thought and feeling was believed to be due to a vast spiritual catastrophe which he, unlike any before him but like thousands after him, called a ‘dissociation of sensibility’. Never having recovered from it, we are condemned to think and feel by turns, for ever unable to synthesise disparate experience in language. This fall or dissociation occurred in the 17th century. In the earlier phase of Eliot’s thinking on the subject – we can now see that he was already, in the Clark Lectures, preparing a change of mind – Donne was a prime example of undissociated sensibility, that ‘direct sensuous apprehension of thought, or a recreation of thought into feeling’ which is not found in later poets. ‘A thought to Donne was an experience, it modified his sensibility.’ That is what poets can or could do, and ordinary people, whose experience is ‘chaotic, irregular, fragmentary’, can’t.

However, the great poet of unified sensibility was not Donne but Dante, who didn’t look very like Donne, but was certainly metaphysical; so that it was necessary to explain both their resemblance and their differences. Also to be accounted for was that moment in the 19th century when, Eliot believed, the true metaphysical quality emerged in France, announced by Baudelaire and coming to flower in Eliot’s admired Laforgue. So the historical moments of metaphysical poetry are three: the age of Dante and some contemporaries, the age of Donne and some contemporaries, the age of Laforgue and some contemporaries (Rimbaud, Corbière). Or rather four, though the lecturer was too modest to say so right out. On the whole, however, the damage done by the 17th-century dissociation, despite isolated efforts to mend it, was as good as irreparable. In terms of poetry it can be fixed at the time when the world of Donne gave way to that of Milton and Dryden.

This idea evidently implied or required some larger theory of the ‘disintegration of intellect’, but the job of working it out was left for the most part to the care of compliant academic disciples. Meanwhile there were those who found the whole set of notions, at first rather cursorily and dogmatically proposed, and later elaborated by unthrilling epigoni, to be vague and ill-formulated. The meanings given to such terms as ‘feeling’, ‘emotion’, ‘sensibility’, ‘intellect’, ‘fusion’ and ‘dissociation’ were not made clear. Given the chance to write a book explaining them, Eliot wrote these Clark Lectures and planned, in the unwritten trilogy, to provide the full-blown historical theory of poetry, culture and religion he thought necessary.

He was not happy with the first steps, confessing to Grierson that the lectures were ‘full of hasty generalisations and unsubstantiated statements’. And he progressively lost confidence, not in the fact of cultural decline, but in the way he had dealt with it; in the Milton lecture of 1947 he said that although he still thought there really was something in the idea of a ‘dissociation of sensibility’ he now regarded its causes as ‘too complex and too profound to justify our accounting for the change in terms of literary criticism’.

However, that is roughly what he set out to do in these lectures. And they do provide a clearer idea of his efforts to give body to his thinking about poetry. The statement at the beginning of the first lecture, that ‘it is valuable to understand the poetry of the 17th century, in order that we may understand that of our own time and understand ourselves’ was one for which, in 1926, there was a fit audience, even though few. But how to go about understanding it in just this way – that is, with a view to understanding ourselves (scil. the mess we’d got into)? That called for long, bold views, and in particular for an acceptance of the central importance of Dante.

Accepting that pre-eminence, he was free to do some comparison and analysis, and offer some limiting judgments on Donne. Dante, he allows, is not a poet one would normally think of under the more usual trade definition of ‘metaphysical’, derived largely from Dryden and Johnson, who had in mind not Dante but Donne, Cowley, Cleveland etc. We have to take account of poetry that is metaphysical in the less specialised sense: philosophical poetry possessing the means to make us feel thought and think feeling – Lucretius, Dante, Goethe. We need not insist on the presence in such poetry of whole systems of thought, but an interaction of thought and feeling we must have. We are looking for poetry in which ‘what is ordinarily apprehensible as an intellectual statement is translated in sensible form’.

This ‘translation’ is what Eliot found in Dante. We know from the essay of 1929 to what degree of subtlety he developed his taste for that poet. He did so in a manner best described by himself: first one surrenders oneself to a line or a stanza, then, possessed by it, seeks a way of saying something relevant about this impassioned moment, showing how it might be related to all else that one thinks about the poet, and about oneself. Eliot did this so well that for many English readers the most familiar, even the most valued, passages of Dante, are the ones he chose to comment on. They are, in a sense that would not perhaps have pleased him, touchstones. But what are the resemblances between these Dantesque ‘bewildering minutes’ and the poems of the conceited Donne? To see them we have to remember and explain that although Dante is ‘metaphysical’ in a much larger sense, the limited sense of ‘metaphysical’ (‘fantastic, elaborated’) is not irrelevant to him. Having done that we may speak of those three phases of metaphysical poetry, and the inexplicit fourth. Eliot was evidently well aware that Dante had somehow to be given a cardinal position in his scheme, and of some difficulty in doing so. Such a position it might not have occurred to others to claim for him. In so far as he manages to do so, his success is due less to the hairsplitting about the meaning of ‘philosophical’ and ‘metaphysical’ as applied to poetry than to the splendour and certainty of his passion for Dante.

When he turns directly to Donne the results are, not unexpectedly, less satisfactory. He is keen to dismiss the arguments of M.P. Ramsay in her important book Les Doctrines mediévales chez Donne (1917), for he does not want a Donne too closely allied to scholastic philosophy, a medieval Donne with a system of doctrine rather like Dante’s. Admittedly Donne did a lot of reading in that area, but his real interest was in the big (and disastrous) theological controversies of his own time; in any case it was not in his nature to have the sort of relation to a body of belief that is to be found in Dante. This is broadly true, and it prompts Eliot to describe Donne as modern, and also as Jesuitical, a modern thing to be at the time.

To be modern in this way is to belong to an inferior civilisation. The Church splits into quarrelling factions, polemic replaces serious theology; these are symptoms of dissociation, like the impending philosophical disaster of Descartes. One consequence of his living at the dawn of dissociation is that Donne, who read so much, never sounds wholly serious. He arrests ideas rather than pursuing them into their larger intellectual contexts, so the result is often a scrappy assemblage. Bits of thought are detained for witty enquiry, grilled as to what emotion they can produce. In this he is, of course, quite unlike Dante.

Accordingly, Dante and his times here get more devoted attention than the 17th century. Praz gently ticked Eliot off for depending too much on secondary sources, especially Rémy de Gourmont, but as an earnest of earnestness, the lecturer, to the recorded discomfort of some of his audience, quotes a great chunk of Richard of St Victor in Latin (apologising for his Italianate Latin accent). He does this because he wishes to contrast the intellectual purity of medieval mystical practices with the ‘spiritual haschisch’ of Ignatius Loyola, whose 16th-century Exercises were more familiar to Donne. Having done so, he can give ‘The Extasie’ an examination minute enough to make you think that in spite of applauding some things in it he doesn’t like it much. The critique is detailed, lively and contestable, but it turns out that what is really wrong with the poem is that it accepts what is called the crude ‘modern’ separation of body and soul, a recent vulgarity impossible to Aquinas. I daresay it might be possible to prove this wrong.

Eventually, in the fourth lecture, Eliot finally tackles the question of the English ‘metaphysical conceit’, and finds that Donne’s lack the ‘rational necessity’ of Dante’s similes and metaphors. Here there is room for argument. For instance, in explaining ‘rational necessity’ Eliot turns his attention for a moment to Shakespeare. He quotes Dante’s passage about Brunetto Latini (famous with us because of Eliot’s admiration):

Poi si rivolse, e parve di coloro
  che coronno a Verona il drappo verde
  per la campagna; e parve di costoro
quegli che vince e non colui che perde.

Then he turned back, and seemed like one of those who run for the green cloth at Verona through the open field; and of them he seemed to be like him who wins, and not like him who loses.

That is, Brunetto is damned, a loser, though he surprisingly looks more like a winner. Eliot, in the later essay on Dante, talks of being hit by these lines, but, as in the case of Francesca, also in hell and lamenting the good times, seems to attribute their force to their salutary emphasis on the severity of the sinner’s punishment. In the lecture his immediate purpose is to compare the Brunetto passage with what Octavius Caesar says about Cleopatra’s body at the end of Shakespeare’s play:

  But she looks like sleep,
As she would catch another Antony
In her strong toil of grace.

Shakespeare’s lines have not the ‘rational necessity’ of Dante’s, but they are admitted to have a different, even a superior, yet presumably not rational, necessity, offering ‘an image absolutely woven into the fabric of the thought’. It is true that Dante is in syntax and vocabulary expansive (parve di coloro – parve di costoro, quegli – colui, the apparently inessential per la campagna); whereas Shakespeare is extremely intensive. But both depend on paradox. On the one hand, there is the loser/winner simile, spelled out with fleeting details of the race and the prize; on the other, the ‘strong toil of grace’, a compressed figure, a sort of oxymoron, with scriptural antecedents. There is an apparent impropriety of sense in both the green cloth and the toil (nets or traps); both are surprising, both hit you. As Octavius remarks, Cleopatra should have some postmortem symptoms but hasn’t, so he is surprised; but our surprise comes rather from the sudden emergence of the great paradox from a less powerful similitude, ‘she looks like sleep’. Thus once more we are hit, and strangeness has much to do with violent success in both passages. The comparison remains interesting, but the passages hardly have that generic differentiation here posited. It may be right to remark that the Shakespeare lines would be ‘impossible for Dante’, and the Dante passage could as well be said to be impossible for Shakespeare. It doesn’t sound much like Donne, either. But it would be hard to agree that the difference is about ‘rational necessity’.

The close readings of Donne are ostensibly based on the dictum, correct I think, that ‘you must construe [Donne] analytically and enjoy synthetically’; but in fact the analyses are a shade disappointing. You feel that Eliot can always surrender to a particular line – ‘A Bracelet of bright hair about the bone’ – but external considerations then induce him to include it in a hostile analysis of its context, regarded as over-conceited, so that the promise of a true union of thought and feeling is not long kept. Such failures are explained by the contention that ‘Donne really doesn’t believe anything; instead he isolates a thought and makes it an object of sense.’ ‘A conceit is the extreme limit of the simile and metaphor which is pursued for its own sake, and not to make clearer an idea or more definite an emotion.’ Here we are moving rapidly towards Eliot’s later and lower estimate of Donne. Rather surprisingly, he argues for the long poems as the best, though one would have thought the ‘Anniversaries’ offered hundreds of examples of thoughts briefly entertained for whatever emotional surprises they can be induced to provide.

Crashaw gets a lecture, and is much admired, though found, not surprisingly, to be ‘on the side of feeling rather than thought’, unlike Dante, for whom feeling and thought are ‘reverse sides of the same thing’. There follows a lecture centred on the ‘mediocre’ but not unimportant Cowley; and the last is devoted to that brief rebirth of metaphysical poetry in 19th-century France. Otherwise all is disintegration, the disintegration of verse following on a greater one, the disintegration of intellect, of which we were to hear more later, though not in the form promised here.

The lectures have occasionally a rather severe jauntiness, as in certain rude remarks about the likes of Middleton Murry and Frank Harris, and some sly digs at Richards in the audience. And Eliot has prefixed a fascinating double epigraph, which couples an exalted passage from the Vita Nuova – ‘Ladies, the end and aim of my love was but the salutation of that lady of whom I conceive that you are speaking’ – with a fragment of some lost ‘popular song’:

I want someone to treat me rough.
Give me a cabman.

The assiduous editor has found no source for this and perhaps it needs none. It makes a calculated bathos of a kind Eliot always enjoyed. This is a rare failure on the part of Mr Schuchard, who offers a great deal of very well-judged annotation. And he dutifully corrects the volleys of misquotation in the lectures. Eliot’s inaccuracies in this respect have been commented on before and found interesting. The most striking in the present volume is in a passage said to be from Chapman’s The Revenge of Bussy D’ Ambois:

  fly, where men feel
The cunning axletree, and those that suffer
Beneath the chariot of the snowy Bear.
And tell them all that D’ Ambois now is hasting
To the eternal dwellers.

Chapman wrote ‘The burning axletree’, and he wrote it in Bussy D’Ambois, not in its sequel. When, seven or so years earlier, Eliot remembered this passage in ‘Gerontion’, he wrote ‘Beyond the circuit of the shuddering Bear’. Referring, in his Harvard lectures of 1933, to Chapman’s original in Seneca, he further remarked that what the imagery meant to Seneca, Chapman or himself, was ‘too obscure to understand’. Certainly, as in other instances, elements of the dying speech of Bussy mingled in his head with other apparently irrelevant associations; but how the axletree became ‘cunning’ seems too obscure to understand.

It is probably true to say that if Eliot had published this book in the Twenties the course of later critical writing about Donne and the Metaphysicals would have been different. As to his own ‘critical mind’, it is illuminated by the book, as Schuchard claims, but also by his rejection of it. The scope of his ‘dissociation of sensibility’ is much enlarged. And it becomes even more clear that Dante, used here us a model, was the end and aim of his love. Donne faded, Dante didn’t; possibly it was, for him, the measure of that difference, repeatedly pondered, between a great and a lesser poet. As to where he stands himself, we can only agree with him that this is a matter for decision elsewhere and at another time.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 16 No. 3 · 10 February 1994

Although Eliot’s criticism is prone to the odd error – and one or two are very odd indeed – it should be borne in mind that, occasionally, these slips were not entirely his own fault. The ‘striking’ error Professor Kermode discusses in his exact, exacting review of The Varieties of Metaphysical Poetry (LRB, 27 January) – the rendering of ‘burning axletree’ into ‘cunning axletree’ which, he pronounces, ‘seems too obscure to understand’ – is simply a case of Eliot falling victim to an unreliable text. In the 1926 Clark Lectures (and in his 1927 essay on ‘Seneca in Elizabethan Translation’) he was quoting from W.L. Phelps’s 1875 Mermaid edition of Chapman’s work, which prints the lines from Bussy as: ‘fly where men feel/The cunning axletree: and those that suffer/Beneath the chariot of the snowy Bear’.

Chapman’s lines appeared again in the Charles Eliot Norton Lectures of 1933. When he introduced his Harvard audience to the passage, and admitted that it lay behind some lines of ‘Gerontion’, Eliot explained: ‘I am glad of the opportunity to use it again, as on the previous occasion I had an inaccurate text.’ The corrected version now read: ‘fly where men feel/The burning axletree, and those that suffer/Beneath the chariot of the snowy Bear’. (The Use Of Poetry and the Use of Criticism). Eliot was probably not too upset; after all, the dubious Mermaid text had provided him with the crucial adjective for one of ‘Gerontion’’s most memorable phrases: ‘History has many cunning passages.’

David Pascoe
Oriel College, Oxford

Vol. 16 No. 7 · 7 April 1994

Frank Kermode’s review of Eliot’s Clark Lectures (LRB, 27 January) gives some of the history of their provenance. It is surprising that he does not note that in fact a substantial section of these lectures was published at the time they were given as ‘Deux attitudes mystiques: Dante et Donne’ in Chroniques in Paris in 1927. This is the section of the lectures that Kermode discusses in some detail, and in which Eliot favours the writings of Richard de Saint-Victor and gives an extensive and unfavourable critique of Donne’s ‘The Extasie’. In 1930 and 1931, Eliot seems to have returned to his Clark Lectures, not merely to produce the Turnbull Lectures, but also to produce six broadcast talks on the Metaphysical poets, published in the Listener in 1930, and to write the better known ‘Donne in Our Time’ that appeared in 1931 in A Garland for John Donne. ‘Deux attitudes mystiques’ was certainly of great interest concerning Eliot’s changing attitude to the Metaphysical poets; but the pieces in the Listener offered few surprises, though Eliot by then favoured the devotional poets.

The non-publication of the Clark Lectures may have been associated with the spiritual crisis Eliot was undergoing at the time: this was the period of his conversion. His changing valuation of Donne in favour of Dante is in keeping with this; and he may have felt insecure about his attitude to his subject. He makes a relevant distinction in another piece published in French at this time, ‘Note sur Mallarmé et Poe’ in La Nouvelle Revue Française of November 1926: ‘Le poète philosophique est celui qui vit d’un système, soit complet et conscient comme ceux de Dante et de Lucrèce … Le poète “métaphysique"… par exemple … Donne, Poe et Mallarmé… ont la passion de la spéculation métaphysique, mais il est évident gu’ils ne croient pas aux théories auxquelles ils s’intéressent.’ The undissociated sensibility of the 17th century, with its capacity to hold in balance a controlled variety of attitudes, seems to have been replaced as the ideal by the unified sensibility of Dante, unironic in its attitude to belief.

In a slightly later piece from 1932, ‘A Note on Two Odes of Cowley’ in Seventeenth-Century Studies Presented to Sir Herbert Grierson, we encounter a clear statement of what lay behind Eliot’s concerns: ‘The transference of attention from theology to science, or to sciences – the beginning of the slow disease which was to separate and confuse thought and feeling …’ Like Yeats, Eliot looked nostalgically back for a time in which the dominant world explanation was a humanly oriented (and religious) one (as it was with Dante), unthreatened by the humanly indifferent explanations of science, which seemed increasingly to both poets to pre-empt the domain of the imagination.

A.T. Tolley
Carleton University, Ottawa

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences