Contradictory Images: the Conflicting Influences of Henri Bergson and William James on the aesthetics of T.E. Hulme, and the Consequences for Imagism

CONTRADICTORY IMAGES: THE CONFLICTING INFLUENCES

OF HENRI BERGSON AND WILLIAM JAMES ON T. E. HULME,

AND THE CONSEQUENCES FOR IMAGISM

by matthew gibson

This article considers the evolution of Hulme’s thought from his ‘Lecture on

Modern Poetry’ to ‘Romanticism and Classicism’, including the development of

his belief in poetry as a visual language, why he defines the ‘aesthetic emotion’

in such a limited way, and why he links these ideas to the Classicism of Action

Francaise. What it demonstrates is that William James’s description of the distinc-

tion between Romantic and Classical art in Principles of Psychology played a part in

helping Hulme to reduce Bergson’s concept of ‘aesthetic emotion’ to sensation, and

to see such effects as pertaining to Classical, rather than Romantic art. However,

Hulme failed to realise that Bergson’s ideas on art were in fact incompatible

with William James’s. This incompatibility caused tensions between the theory

and practice of the Imagist movement inspired by Hulme’s ideas. The conflict

was caused partly by James’s denial of the uniqueness of aesthetic emotion, and

partly because, in reducing the aesthetic emotion to sensation, he negates the im-

portance of memory and association to the image’s aesthetic effect, which is by its

very nature temporal in manifestation. The piece concludes by demonstrating how

the poem ‘Hermes of the Ways’, by H.D., does not concord with Pound’s doctrine

of image as an ‘intellectual and emotional complex in an instant of time’, and needs

memory and association to unravel if the image is to work. This critique demon-

strates the divergence between genuine Imagist practice and William James’s

prioritisation of immediate sensation, which Hulme had communicated to Imagist

theory.

The contention of this article is that the ideas of T. E. Hulme, and in particular hisbelief that the ‘dry, hard classical verse’ he preferred involved ‘aesthetic emotion’,

were influenced by William James’s distinction between Romantic and Classical art

in the Principles of Psychology (1890). While no scholars have yet noticed its sig-nificance, or even its existence, James’s distinction not only describes Romanticism

in a way consonant with its description by the austere, Monarchist critics of Action

Francaise, but also understands Classical art as inviting a form of intuitionist,aesthetic response similar to that suggested in the theory of Hulme’s hero

Bergson—but also not entirely congruent with it. This lack of complete con-

gruence with Bergson’s entire theory, and Hulme’s apparent ignorance of thedifference when transposing James’s ideas, was to have contradictory repercus-

sions for the Imagist movement, since Hulme’s writing of this period clearlyaffected the practice of Imagist poets.

The Review of English Studies, New Series! The Author 2010. Published by Oxford University Press 2010; all rights reserveddoi:10.1093/res/hgq099

The Review of English Studies Advance Access published September 30, 2010

by guest on February 7, 2011res.oxfordjournals.org

Dow

nloaded from

Ezra Pound admitted that the Imagist doctrine of his essay ‘A Few Don’ts’, in

which he defined the image as ‘an intellectual and emotional complex in an instant

of time’,1 was influenced by T. E. Hulme.2 Pound had met Hulme as early as 1909

at the Cafe Tour d’Eiffel,3 and, more importantly, attended a lecture by him on

15 July 1912, at a point when Hulme was still under Bergson’s influence,4 and

when Pound himself was becoming more sympathetic to Hulme’s promotion of

vers libre and the prioritisation of the image. The same year a new group, including

Pound, began meeting at the house of Brigit Patmore and working under the ideas

of Hulme’s original coterie.5 When the anthology Des Imagistes was eventually

published in 1914, it included contributions from Hulme, and indeed Pound

published Hulme’s Complete Poetical Works in two pages at the end of his

Ripostes. As a result of all these evident connections between Pound and

Hulme, several critics, including Sanford Schwartz, can also see ‘the imprint of

Bergson’s philosophy’ on Pound’s Imagist poetry and on Eliot’s earlier work.6

Much has been written on Hulme’s use of Bergson in this period and the degree

to which it is consistent with his other views or not. Sanford Schwartz has poign-

antly noticed the inconsistencies in Hulme’s portrayal of Bergson’s theory itself,

writing that Hulme’s obsession with both visual language and spatial form over

flux in the aesthetic effect is in contrast to the more reflective and temporal

elements involved in the Bergsonian experience of aesthetic emotion. Schwartz

attributes Hulme’s modification of Bergson to his reading of Nietzsche.7 Michael

Levenson has noted an irreconcilable contradiction between Hulme’s Bergsonian

1 Ezra Pound, ‘A Few Don’ts by an Imagist’, in Peter Jones (ed.), Imagist Poetry (London,1972), 130–4, at 130.

2 In a letter to Margaret Anderson, 17 November 1917, Pound admitted that, ‘I made theword – on a Hulme basis – and carefully made a name that was not and never had been usedin France’. Qtd in Hugh Kenner, The Pound Era, repr. (London, 1991), 178.

3 J. B. Harmer, Victory in Limbo: Imagism 1908–1917 (London, 1975), 31.

4 Various scholars have seen the article ‘Bergson Lecturing’, first printed in The New Age,2 November 1911, as representing the beginning of Hulme’s break with Bergson. This isdespite the fact, noted by Patricia Rae, that he continued to defend Bergson’s position(although largely in relation to metaphysics rather than art) in The New Age until 22February 1912 (Patricia Rae, The Practical Muse: Pragmatic Poetics in Hulme, Pound andStevens [Lewisburg and London, 1997], 286). A reading of the article itself shows that Raeis right, as the feeling of scepticism which Hulme discusses is a result not of Bergson’s ideas(which Hulme does not attack in any reasoned way), but his being presented before amundane multitude. Hulme concludes by writing that his own scepticism was merely‘childish repugnance’ and that the ancients ‘never allowed these kinds of emotions tocome to the surface. . . We must, then, re-establish the old distinction between thepublic and the esoteric doctrine. The difficulty of getting into this hall should have beencomparable to the difficulty of getting into a harem, not only in appearance, but in fact’(T. E. Hulme, The Collected Writings of T.E. Hulme, ed. Karen Csengeri (Oxford, 1994),154–9, at 158). By 1913, however, Hulme’s interest in Bergson had entirely waned inpreference for Worringer.

5 Harmer, Victory in Limbo, 35–6.

6 Sanford Schwartz, The Matrix of Modernism: Pound, Eliot and Twentieth Century Thought(Princeton, 1985), 31.

7 Schwartz, Matrix of Modernism, 51, 57.

2 of 21 MATTHEW GIBSON by guest on February 7, 2011

res.oxfordjournals.orgD

ownloaded from

theory of intuition in art and his adherence to the anti-intuitionist and

anti-individualist beliefs of Lasserre,8 for whom Bergson was really a dangerous

Romantic, incompatible with the Classicism and Catholicism of Action Francaise.9

Patricia Rae, however, sees no contradiction in either Hulme’s portrayal of

Bergson’s aesthetic theory or in his attribution of Bergsonian intuitionism to the

stringent ethos of Classicism. Unlike Schwartz, she understands Bergson’s aes-

thetic theory as being already entirely sensory, visual and ‘pragmatist’ in its de-

scription of aesthetic reaction, and unlike Levenson sees this theory as leading

necessarily to a James-like, ‘pragmatist’ moral feeling that is entirely congruent

with the ethics and traditionalism of Action Francaise.10

In this article, it will be shown that Hulme was in fact directly influenced by

William James’s theories on art, and that this influence, rather than Nietzsche’s,

was the cause of his coming to describe Bergson’s aesthetic theory as being more

visual and immediate than it actually was when Hulme was trying to reconcile it

with the ethical strictures of Lasserre’s Classicism. It will be shown that the

attempt to merge James’s aesthetic theory with Bergson’s was in reality impossible,

despite Rae’s belief in their compatibility, and was in turn a cause of confusion for

the Imagists. This is because their best poems simply could not conform to

Hulme’s obsession with the visual and the immediate, being dependent upon

temporality and reflection for their emotional effect.

In fact, Hulme’s borrowings from William James’s concepts do no more than

help to subvert his own idea that ‘dry, hard, classical’ poetry can actually promote

aesthetic emotion. This is because James, when himself defining Classical art in the

Principles of Psychology, insisted that aesthetic emotions are not fundamentally

different to other emotions found in life and are also congruent with sensa-

tions—a view utterly refuted by Bergson. Thus, Hulme’s use of James to define

aesthetic emotion is actually a source of conflict with his Bergsonian influence, and

we can also observe this contradiction at work in the Imagist poetry inflected by

Hulme’s ideas.

The development of Hulme’s theory

Hulme’s aesthetic theory developed along an uneven path from 1908 to 1912,

shedding some concepts while retaining certain core notions. His early ‘Lecture

on Modern Poetry’, delivered to Flint, Storer, Connolly and others in 1908 at the

Cafe Tour d’ Eiffel,11 stresses the need for poets to embrace the spirit of the age

8 Michael Levenson, A Genealogy of Modernism: A Study of English Literary Doctrine1908–1922 (Cambridge, 1984), 85.

9 T. E. Hulme, ‘Mr. Balfour, Bergson, and Politics’, Collected Writings, 158–64 (at 165);first published in The New Age, 9 November 1911.

10 Rae, Practical Muse, 47.

11 Harmer, Victory in Limbo, 24–5.

CONFLICTING INFLUENCES OF HENRI BERGSON AND WILLIAM JAMES 3 of 21 by guest on February 7, 2011


ownloaded from

and accept that old models have become exhausted, since there can be no such

thing as absolute beauty or perfection:

We are no longer concerned that stanzas shall be shaped and polished like gems, butrather that some vague mood shall be communicated. In all the arts, we seek for the

maximum of individual and personal expression, rather than for the attainment of anyabsolute beauty.12

To this end—and, as Csengeri notes, in accordance with the ideas of Gustav

Kahn13—Hulme advocates a poetry of irregular metre, rather than those ‘pol-

ished. . .gems’ of traditional stanzas and versification. He furthermore suggests

that this poetry should juxtapose certain images in separate lines to evoke the

state felt by the poet, making a kind of ‘visual chord’ in which a wholly different

image is suggested. The distinction between this type of poetry and that which

preceded it is one of both emphasis and effect: ‘The difference between the two is,

roughly, this: that while one arrests your mind all the time with a picture, the other

allows the mind to run along with the least possible effort to a conclusion.’ For

Hulme, the rhythm is predicated on the image, or visual suggestion, rather than

the image being predicated on the necessity imposed by the rhythm. It is image

that is most important, and by implication what contributes most to suggesting the

‘vague mood’ which modern poetry must translate over.14 He finally illustrates this

form of poetry by drawing an analogy which was to be amplified by Pound some

years later:

This new verse resembles sculpture rather than music; it appeals to the eye rather than tothe ear. It has to mould images, a kind of spiritual clay, into definite shapes. This material,

the [hule] of Aristotle, is image not sound.15

One possible source for Hulme’s prioritising of image over sound is Theodule

Ribot (1839–1916), a French psychologist much impressed by evolutionary theory,

who is renowned for having promoted empirical and pathological psychology over

spiritualist trends. J. B. Harmer argues that Ribot’s description of mysticism in

Essai sur l’imagination creatrice (1900) as an experience which is initially concrete

rather than symbolic, influenced Hulme’s belief that the language of poetry should

be concrete and visual in opposition to the mundane language of prose, which is

strewn with dead metaphors that are thought and not felt.16 Harmer’s suggestion

12 T. E. Hulme, ‘Lecture on Modern Poetry’ (1908), Collected Writings, 49–56, at 53.

13 Karen Csengeri, ‘T.E. Hulme’s Borrowings from the French’, Comparative Literature,34 (Winter, 1982), 16–27, at 20.

14 Hulme, Collected Writings, 55, 53.

15 Ibid., 56.

16 Harmer, Victory in Limbo, 117. Harmer’s identification centres mainly on Ribot’s de-scription of the mystical, which involves the ‘imagination diffluente’ rather than simply‘imagination plastique’. Ribot writes: ‘L’originalite de l’imagination mystique consisteen ceci: elle transforme les images concretes en images symboliques et les emploie commetelles.’ (Theodule Ribot, Essai sur l’imagination creatrice (Paris, 1900), 187).



ownloaded from

that Ribot influenced this early essay is convincing, although perhaps more be-

cause Ribot elsewhere describes the poetic imagination as ‘another form’ of the

‘plastic imagination’, which stimulates visual impressions through language.17

Ribot also details how writers—particularly the Parnassians—illustrate the effects

of sound with images of movement. Nevertheless, when outlining the ‘plastic

imagination’, Ribot talks of ‘Inferiorite de l’element affectif’ and how art is

‘denuee de cette marque interieure qui vient du sentiment’,18 which certainly

contradicts Hulme’s belief here and elsewhere that poetic imagery can suggest

emotion—or ‘vague mood’—through its form.19

It is in fact more than likely that, for Hulme, Ribot’s ideas were merely an

embellishment of Bergson’s aesthetic theory; they may have helped him to stress

the importance of the visual and the sculptural over the aural, but were not a

significant influence. In Essai sur les donnees immediates de la conscience (1889),

Bergson defined aesthetic feeling—‘sentiment esthetique’—as involving an inten-

sity beyond the merely sensory, which was capable of piercing through ‘surface

consciousness’ to the unities of ‘real duration’. Bergson asked:

D’ou vient le charme de la poesie? Le poete est celui chez qui les sentiments se developpent

en images, et les images elles-memes en paroles, dociles au rythme, pour les traduire.

En voyant repasser devant nos yeux ces images, nous eprouverons a notre tour le sentimentqui en etait pour ainsi dire l’equivalent emotionnel.20

Here we see a presentation of ideas similar to those in ‘Lecture on Modern

Poetry’—that the language of poetry constitutes images presented through

words, which are themselves ‘dociles au rythme’, stirring ‘les sentiments’ that

can be translated or carried over: a language which uses words to promote

images that convey emotions with the subordinate service of rhythm. He also

presents the ‘sentiment esthetique’ as being like an ‘etat d’hypnose’,21 similar to

Hulme’s ‘vague mood’. The only major difference is Hulme’s further—and

17 Earlier, in defining the plastic imagination (that of the fine arts), Ribot outlines how theplastic imagination realises itself in spatial form, and understands it as including the literaryand poetic form of imagination which ‘emploie les mots comme evocateurs d’impressionsvives et nettes de vision, de contact, de mouvements’. As an example, he writes of howHugo always wished to ‘voir les mots’ (Ribot, Essai sur l’imagination creatrice, 157).

18 Ibid., 154, 158.

19 Ribot does admit that there are some writers, like Shakespeare, Carlyle, and Michelet,who manage to combine the ‘sensoriel’ and the ‘affectif’ (Ribot, Essai sur l’imaginationcreatrice, 154–5), but does not see them as being linked so much as entwined causes.The ‘affectif’ for Ribot is involved in ‘l’imagination diffluente’, which is spurred by emo-tional intensity, ‘sensoriel’ the result of the intellectual ‘imagination plastique’, which isgoverned by dissociation, analogy, ideas and external stimulus (ibid., 19, 184). He does notattempt to explain literary works which contain both as being spurred by the same root, asdoes Hulme.

20 Henri Bergson, Essai sur les donnees immediates de la conscience (Paris, 1889), 11.

21 Bergson, Essai, 11.



ownloaded from

ominous—emphasis of the visual over the phonic: poetry’s ‘sculptural’ over its

‘musical’ properties, a preference he was to embellish later.22

In the essay posthumously titled ‘Bergson’s Theory of Art’ (1911), Hulme

described the importance of art in his master’s philosophy as lying in its ability

to restore freshness of perception and real consciousness to the perceiver. Hulme

draws upon Bergson’s distinction between consciousness as ‘real duration’ and the

more mechanised ‘surface consciousness’, or consciousness as action. The former

is the intense encounter with the flux of experience in which no two moments of

consciousness are ever the same, and the latter the limitations imposed upon the

mind for the sake of its action within reality. Poets are those who are able to

explore the former, past the ‘crystallised shapes’ of ‘surface’ consciousness, and

who in doing so cast in permanent form certain moments of truth: ‘the function

of the artist is to pierce through here and there, accidentally as it were, the veil

placed between us and reality by the limitations of our perception engendered by

action’.23 Employing a phrase also used in ‘Romanticism and Classicism’, Hulme

stresses that the artist’s material is like ‘architect’s curves’: the attempt to give a

unique turn to wood or steel—a metaphor adapted to show the difficulty of ex-

pressing perception through language. This specific turn of the wood or steel is

what imparts the ‘aesthetic emotion’.24 While this type of emotion may be overlaid

and combined with other emotions, it is nevertheless a unique element of the art

product:

The particular kind of art we are concerned with here, at any rate, can be defined as anattempt to convey over something which ordinary language and ordinary expression lets slipthrough. The emotion conveyed by an art in this case, then, is the exhilaration producedby the direct and unusual communication of this fresh impression.25

Although he does not believe that imagery is the only technique for expressing

such veil-lifting perceptions, he nevertheless centres on the need for concrete

images in poetry and the power of visual stimulus to refresh perception over

and above all other devices, even more completely than in the earlier ‘Lecture

on Modern Poetry’, where he at least detailed his views on rhythm.

22 Patricia Rae also suggests as an influence James’s A Pluralistic Universe (1909), in whichJames praised the power of concrete analogy for expressing ideas without abstraction,exploiting sameness and difference. ‘Hulme conceives the analogy as a kind of ‘‘visualchord’’, offering harmony in its patterning, yet preserving difference through its concrete-ness.’ She states, before continuing her description, that ‘Hulme’s vision of analogy’s the-oretical line and its concrete fringe may be the strongest evidence we have of a direct linkbetween James and his poetic theory’ (Rae, Practical Muse, 73, 74). While I believe that laterworks by Hulme display a very concrete influence from James, it seems to me that hisdescriptions of the end of poetry in ‘Lecture on Modern Poetry’ as being a ‘vague mood’,and of the power of the ‘visual chord’ to provoke a wholly new image, suggest a continuingRomantic and organicist trend (recalling at once the Coleridgean notion of a tertium aliquidand the image of the imagination, the Eolian Harp). This contradicts the notion of‘difference through. . . concreteness’ and the full absorption of pragmatism.

23 T. E. Hulme, ‘Bergson’s Theory of Art’, Collected Writings, 191–204 (at 193, 194).

24 Ibid., 198, 199.

25 Ibid., 201.



ownloaded from

Thus, he continues to emphasise the importance of the visual in poetry for

conveying emotion when outlining Bergson’s theory, even though this now goes

beyond the role accorded it by Bergson. The philosopher asserted that while

poetry was a visual language, its imagery could never convey emotion ‘sans les

mouvements reguliers du rythme par lequel notre ame, bercee et endormie,

s’oublie comme en un reve pour penser et pour voir avec le poete’26—an idea

not excluded by Hulme, but certainly not mentioned by him either, since his

emphasis is now even more firmly on the visual rather than on sound.

Concentrating mainly on passages taken from ‘Romanticism and Classicism’,

but also on one from ‘Bergson’s Theory of Art’, Sanford Schwartz notices how

Hulme’s development of the visual language theory and obsession with immediacy

are counter to the sense of abstracted time and infelt movement in Bergson’s

philosophy, a difference he attributes to Nietszche’s influence. ‘In short’, writes

Schwartz, ‘Bergson evokes the presence of real duration by appealing to temporal

rather than spatial forms, musical sequence rather than visual image’, while

‘Hulme has no interest in restoring the temporal flow of real duration’, and sees

the artist as making a ‘fixed model’.27 However, we can go further than Schwartz

by noticing another more salient deviation in ‘Bergson’s Theory of Art’ from the

French philosopher’s actual ideas. In formulating new ways of expressing ‘aes-

thetic emotion’, Hulme writes of the need to create new metaphors. ‘Metaphors

soon run their course’, he states, even though they were originally ‘used for the

purpose of conveying over a vividly felt actual sensation’, and ‘the reality of the

sensation experienced’. A few paragraphs later he writes: ‘It is only where you get

these fresh metaphors and epithets employed that you get this vivid conviction

which constitutes the purely aesthetic emotion that can be got from imagery.’28

Later he defines ‘the essentially aesthetic emotion as the excitement which is

generated by direct communication’, and finally declares that the ‘aesthetic emo-

tion’ consists in ‘the actual character of that communication, the fact that it hands

you over the sensation as directly as possible’.29 Hulme not only underlines the

visual, spatial and immediate at the expense of the musical and temporal, but also

equates ‘aesthetic emotion’ almost entirely with ‘sensation’ and its immediate effect

(‘exhilaration’/‘vivid conviction’), and indeed does not really distinguish the

‘aesthetic emotion’ from the sensation itself.30


27 Schwartz, Matrix of Modernism, 56, 57.

28 Hulme, Collected Writings, 195.

29 Ibid., 201.

30 Although Hulme declares that what concerns him is ‘only the feeling’, or ‘purely aestheticemotion’ caused by the use of fresh metaphors (ibid., 195), he fails to describe this emotionas qualitatively different to the sensation itself, except in the sense of ‘freshness’, ‘excitement’or ‘sense of exhilaration’ which accompanies it immediately. His stress on the poet’s ‘visualsensation’ and its ‘direct. . .. communication’ (ibid., 201) means that the emotion reallyappears to be inbound with the sensation which causes it – hence, perhaps, his failure toattempt to pursue any form of separate definition in ‘Romanticism and Classicism’, beyondthe ‘zest’ one feels on seeing a girl’s moving skirts (ibid., 70).



ownloaded from

Hulme’s definition of ‘aesthetic emotion’ is completely opposed to that of

Bergson, who described the fundamental difference between ‘des intensites de

nature tres differentes, l’intensite d’un sentiment, par exemple, et celle d’ une

sensation ou d’un effort’. The one—in its more intense forms, such as ‘sentiment

esthetique’—is an organic intensity, which seems ‘self-sufficient’[‘se suffit a lui-

meme’], and which arises from the intuition of real duration, the other the result of

the stimulation of simple surface consciousness.31 As late as 1902, in ‘L’Effort

intellectuel’, Bergson outlined the importance of abstract relations in poetic cre-

ation, and declared that ‘quelque chose de simple et d’ abstrait, je veux dire d’

incorporel’ existed in the soul of the poet,32 effectively embellishing his argument

that neither intellectual nor creative effort involve the purely sensory.

We might attribute Hulme’s linkage of ‘sensation’ with ‘aesthetic emotion’ to

Ribot, a psychologist also cited approvingly by Bergson. As was detailed above,

J. B. Harmer understood the visual and sculptural elements of Hulme’s theory in

‘Lecture on Modern Poetry’ as being directly motivated by Ribot’s concretising

description of mystical experience in L’Imagination creatrice, although it is more

likely that Hulme was impressed by Ribot’s description in the same book of poetic

language being a ‘plastic’ medium, as opposed to a temporal medium like music.33

Nevertheless, Ribot completely opposed ‘sentiment’ to ‘sensation’ in a way counter

to Hulme’s near equation of the two, and despite understanding aesthetic imagin-

ation in the largest sense as having emotion as a potential ‘moteur’, ruled emotion

out of the plastic in its ultimate realisation and banished it from the literary

imagination qua creative act.34 For Ribot, since literary imagination constituted

another version of the plastic and visual, it was driven mainly by intellect and

manifested its ideal conception through sensation, not emotion. Music, by

contrast, was driven entirely by emotion that it then conveyed through sound.35

Thus, completely unlike Bergson, Ribot saw literary language proper as provoking

not emotion but rather the purely sensory.36 While Ribot admitted that Symbolist

poetry attempted to depict emotion through vague images, he noted that it

31 Bergson, Essai, 9–10.

32 Henri Bergson, ‘L’Effort intellectuel’ (1902), in Andre Robinet and Henri Gouhier (eds),Oeuvres, 2nd edn (Paris, 1962), 930–59, at 947.

33 Ribot, Essai sur l’imagination creatrice, 157.

34 Ibid., 29–30, 154, 184. Csengeri notes Ribot’s initial attribution of emotion as a force foraesthetic imagination, but does not really examine his careful division of plastic imaginationand ‘diffluent’ imagination in Essai sur l’imagination creatrice, nor his discussion ofSymbolist Art (Csengeri, ‘T.E. Hulme’s Borrowings from the French’, 22). Inexplicably,no critic, when discussing Hulme, has as yet taken into account Ribot’s careful descriptionof Symbolist art and literature.

35 ‘. . .l’antithese est irreductible. Le langage intellectuel, la parole, est un agencement desmots qui sont les signes d’objets, de qualites, de rapports. . .Le langage emotionnel, lamusique, est un agencement de sons successifs ou simultanes, de melodies et d’harmoniesqui sont les signes d’etats affectifs’ (Ribot, Essai sur l’imagination creatrice, 183–4).

36 As an illustration of the difference between the two thinkers, Bergson specifically stressesthe mobile and elastic nature of Ribot’s ‘conception ideale’ which governs the work of art in‘L’Effort intellectuel’ (Bergson, Oeuvres, 948). This is an emphasis not apparent to Ribot’s



ownloaded from

frequently failed in its objectives,37 remarking that Symbolists were trying to make

words provoke emotion in the same way that sounds or gestures more normally

do.38 Therefore, although Hulme’s aesthetics clearly bear some similarities

with Ribot’s plastic imagination and emphasis on sensation over sound, when

Hulme repeatedly insists on equating Bergsonian ‘aesthetic emotion’ with the

direct effect of ‘sensation’, his error must spring from an entirely different

source than Ribot.Hulme’s essay on Bergson provides a more elaborate explanation of the artistic

procedure recounted in ‘Romanticism and Classicism’ (1911–12), a work which

effectively places Hulme’s own understanding of Bergsonian aesthetic theory

within a historical and political context, and which was also most probably the

lecture delivered on 15 July 1912 at Clifford’s Inn Hall and attended by Pound and

others.39 ‘Man is an extraordinarily fixed and limited animal’ argues Hulme,

‘whose nature is absolutely constant. It is only by tradition and organisation that

anything decent can be got out of him.’ This view tends towards Classicism rather

than Romanticism: ‘Put shortly, these are the two views, then. One, that man is

intrinsically good, spoilt by circumstance; and the other that he is intrinsically

limited, but disciplined by order and tradition into something fairly decent.’ The

one is what Hulme labels Romanticism, the other the Classicism he claims is

contained in the Church’s ‘adoption of the sane classical dogma of original

sin’.40 He also excoriates Romantics for their vagueness and mystification:

‘Verse to them always means a bringing in of some of the emotions that are

grouped round the word infinite’, while Classical verse never goes beyond the

purely human (although is not humanist), and does not attempt to ‘drag . . . in the

infinite’.41

Here, of course, the fundamentally right-wing and Catholic aspects of Hulme’s

aesthetic theory, and his public attack on Romanticism, certainly add to Bergson’s

theory yet more concepts not actually present in the philosopher’s work.

Nevertheless, the main core of the theory presented still appears to be in keeping

with Hulme’s previous understanding of Bergson. The language of poetry is,

Hulme declares, essentially a visual, concrete one: a compromise for what ideally

would be a ‘language of intuition’, or what Hulme calls ‘the exact curve’ of wood.

The attempt to find the ‘exact curve’ is part of ‘this struggle against the ingrained

own definition of the term, mobility and elasticity being confined to the imagination whichrealises it.

37 Ribot, Essai sur l’imagination creatrice, 169–70.

38 ‘Ordinairement, la pensee s’exprime par le mot, le sentiment par les gestes, les cris, lesinterjections, les differences d’intonations; il trouve son expression complete et savant dansla musique’ (ibid., 170).

39 T.E. Hulme, ‘Romanticism and Classicism’ (1912?), Collected Writings, 59–73 (at 59n).

40 Ibid., 61.

41 Ibid., 66, 68.



ownloaded from

habit of the technique’ which the concentrated mind seeks to master.42 Again,

Hulme’s choice of words here is inflected by the Bergsonian distinction between

surface consciousness, which involves crystallisations, or habitual experiences,

and real duration, in which consciousness is freed and can discover the aesthetic

emotion.

The articulated theory once again diverges from Bergson’s doctrine, not simply

in the over-dependence upon the visual and immediate in poetic language at the

expense of the musical and the temporal, but in the continued and more pro-

nounced equation of ‘aesthetic emotion’ with the immediate effect of ‘sensation’.

In describing this language which imparts aesthetic emotion, Hulme calls it, much

as in ‘Bergson’s Theory of Art’, one which ‘would hand over sensations bodily’.43

The example of aesthetic emotion Hulme provides is the ‘zest’ with which one

experiences the lifting of a woman’s skirt by the wind. Having explained that in

this specific example it is the ‘extraordinary interest’ in ‘that peculiar kind of

motion’ which makes the artist search for the ‘exact epithet which hits it off’,

and which will express the ‘properly aesthetic emotion’, he further argues that:

It doesn’t matter an atom that the emotion produced is not of dignified vagueness, but onthe contrary amusing; the point is that exactly the same activity is at work as in the highestverse. That is the avoidance of conventional language in order to get the exact curve of thething.44

Bergson himself wrote that: ‘Dans les procedes de l’art on retrouvera sous une

forme attenuee, raffines et en quelque sorte spiritualises, les procedes par lesquels

on obtient ordinairement l’etat d’hypnose.’45 Despite Rae’s attempt to demystify

Bergson and see his theory as anti-spiritual and entirely compatible with pragma-

tism,46 his views here and elsewhere are continually closer to the quasi-mystical

understanding of poetry put forward by Mallarme rather than the rejection of the

‘vague’ and the ‘dignified’ for the ‘amusing’ specified by Hulme in his own ex-

emplification of the suggestive power of imagery.47 Notwithstanding this differ-

ence, Hulme still declares that the fundamentals of an intuitive and nonintellectual

poetic language for expressing ‘vital complexities’, which language he has

described earlier as incorporated within Classicism, are ‘all worked out in

Bergson’,48 although—perhaps carefully—he does not attribute the other, more

political aspects of his theory to the French philosopher.49

42 Ibid., 69, 70.

43 Ibid., 70.

44 Ibid., 70–1.


46 Rae, Practical Muse, 59–60, 64.

47 Such as ‘l’incorporel’ in ‘L’Effort intellectuel’ (Bergson, Oeuvres, 947).


49 Patricia Rae argues that Bergson refused to see the moment of artistic intuition as apiercing of the veil to a spiritual dimension, crucially distancing himself from Schelling andSchopenhauer. ‘Unlike Schopenhauer, Mallarme or Baudelaire, however, and contrary to



ownloaded from

One can see that despite the continuing preoccupation with poetry being a

visual language, in both these later essays far-reaching changes have been made

since the earlier ‘Lecture on Modern Poetry’. Hulme has dropped the disrespect

for tradition he held in 1908, and has also refined his understanding of the emo-

tional content of art from ‘some vague mood’ to ‘sensation’ and its immediate

effects. As Levenson notes, there is a contradiction between Hulme’s professed

admiration for tradition and obsession with Bergson,50 a potential criticism that

Hulme himself noted in his essay ‘Mr Balfour, Bergson, and Politics’, when re-

counting a conversation with Lasserre. The Classicists of Action Francaise did not

appreciate Bergson’s ahistoricism, and so understood his intuitional theory of art

as being a form of Romanticism—a contention Hulme determined to oppose.51

In his critique of Romanticism, Lasserre objected to imaginative literature since

it records the intimations of infinity experienced by an individual apprehending

nature rather than reflecting on the moral lessons acquired from ‘une communi-

cation abondante et genereuse avec la vie’.52 He particularly objected to Rousseau

and Senancour, mocking the latter’s constant, unrequited desire for ‘l’infini’

(which one? he asks) and ‘ce chaos, uniquement tissues de rayons et des ombres

que, selon l’heure, il projette sur le vague univers’.53 Hulme’s own denigration of

Romanticism as anti-traditional, hubristic and as promoting the ‘vague’ and the

‘infinite’ is clearly gleaned from his reading of and conversations with Lasserre.

That said, there is nothing in Lasserre to suggest the limitation of Bergson’s

‘sentiment esthetique’ to the immediate impression of ‘sensation’, which Hulme

in any case ascribes to Bergson himself in both ‘Bergson’s Theory of Art’ and

‘Romanticism and Classicism’. Nor is there any suggestion in Lasserre that

intuitionism can have any place in Classical art, his rejection of any form of indi-

vidualist impressionism being consistent throughout Le Romantisme francais.

Hulme’s reduction of Bergson’s theory of ‘sentiment esthetique’ and ‘l’equivalent

emotionnel’ to sensation, his continued linking of its intuitional foundation to

the Classicist programme of Action Francaise, and his further pitting of this

the view of many critics who have written of Bergson’s impact on Hulme, Bergson does notrepresent that higher reality as a transcendental spirit. Where the Symbolists have the artistpenetrating a veil of particulars to grasp universal Ideas, he defines quite different bound-aries for the event’ (Rae, Practical Muse, 57). However, Bergson’s differentiation betweenhis own Post-Hegelian views and two Post-Kantian philosophers in La pensee et le mouvant(to which she is referring), does not mean that his interest in empirical experience deniesart’s ability to touch a spiritual realm, merely that he understands the spirit which isrevealed through the ‘moi fondamentale’ as being a form of becoming—‘la realite dans lamobilite qui en est l’essence’—rather than a fixed structure of being (Henri Bergson,La pensee et le mouvant: essais et conferences (Paris, 1934), 33–4). The specific passage towhich Rae refers, in which Bergson attacks the claims of Schopenhauer and Schelling thatintuition can transcend time, was also written in 1922, long after Hulme had died (Bergson,La pensee, 113n).

50 Levenson, Genealogy of Modernism, 85–6.

51 Hulme, Collected Writings, 164–5.

52 Pierre Lasserre, Le Romantisme francais (Paris, 1907), 17.

53 Ibid, 85, 93.



ownloaded from

hybrid against Lasserre’s definition of Romanticism, have never been sufficiently

explained. Before analysing the motives for Hulme’s changes we must try to

examine to what extent his ideas were followed by the Imagists, and with what

consequences.

Hulme and Imagism

At the end of ‘Romanticism and Classicism’, Hulme writes: ‘I prophesy that a

period of dry, hard, classical verse is coming.’54 Although he never concerned

himself with attending meetings of the Imagist group, the fact that he allowed his

Complete Poetical Works to be published in Pound’s Ripostes in 1912 under the

term ‘Imagistes’ shows he was at the very least interested in the alignment.

Furthermore, the two manifesto-like essays on Imagism by Flint and Pound in

the periodical Poetry (March 1913) certainly show Hulmean/Bergsonian ideas.

Flint’s essay delivers three maxims that the poet should treat the thing directly,

use only necessary words and write in accordance with the musical phrase,55 while

Pound more or less parallelly exhorts poets to avoid mixing abstractions with the

concrete, to avoid superfluous words and not ‘chop your stuff into separate

iambs’.56 The first two commands in both essays are obviously in keeping with

Hulme’s insistence that verse should use a concrete, visual language and avoid

words which evoke the vague rather than the concrete. With regard to the third,

Hugh Kenner assumed that the importance of musicality in Imagism was to have

‘[w]ords set free’ so that they may ‘assert themselves as words, and make a nu-

minous claim on our attention, from which visual, tactile and mythic associations

radiate’57—a technique which anticipates Vorticism and diverges from Hulme’s

promotion of ‘new metaphors’. The promotion of free verse at the Poet’s Club—

principally by the mediocre practice of Storer, and by Hulme in his ‘Lecture on

Modern Poetry’—certainly predates any named link with Bergson, and was mainly

employed by Hulme to promulgate the idea of imagery being served by rhythm

rather than the other way round. Nevertheless, a more obvious conclusion than

Kenner’s is that free verse helps promote Bergson’s understanding of flux and

‘real duration’ over surface consciousness, which is integral to an Imagist poem’s

peculiar emotional effect.

One can see this working in ‘In a Station of the Metro’: the three strong stress

syllables at the end of the line ‘Petals on a wet, black bough’ suggest the sudden

slowing of association as the speaker’s mind lingers over the image and impres-

sions,58 thus incorporating the element of changeable time involved in the recalled

images of consciousness. Thus, it also incorporates the sense of what Pound called


55 F. S. Flint, ‘Imagisme’, in Imagist Poetry, 129–30.

56 Pound, in Imagist Poetry, 130–1.

57 Kenner, The Pound Era, 187.

58 Ezra Pound, Collected Shorter Poems, 2nd ed. (London, 1968), 119.



ownloaded from

‘freedom from time limits and space limits’ (Bergson’s surface consciousness) in

his doctrine of the image as ‘that which presents an intellectual and emotional

complex in an instant of time’59—similar in terminology to the Bergsonian notion

of ‘vital complexities’ in aesthetic intuitions. Therefore, Flint’s and Pound’s

insistence that poets should not write in accordance with ‘the metronome’ can

easily be seen as an attempt to portray Bergson’s understanding of consciousness

being a seamless, undulating stream of phenomena, over which either habit or the

necessities of ‘action’ impose ‘crystallisations’, rather than being a series of indi-

viduated associations. It is not entirely clear whether or not Pound’s ‘Doctrine of

the Image’ veers towards a respect for a form of concrete mysticism, as Patricia Rae

maintains, or simply towards the unconscious.60 Nevertheless, by his own admis-

sion, his view that ‘an image is that which presents an intellectual and emotional

complex in an instant of time’ bears the impress of Hulme’s Bergson-inspired

theory that the language of poetry should be a visual one, a substitute for a lan-

guage of direct intuition, and one which conveys unique ‘aesthetic emotions’

which ordinary language simply cannot.61 Where we might charge Pound with

swerving from his theory in his poetic practice is in the idea that the complex is

really an ‘instant’—for the slowing of time in the above poem clearly does not

represent simultaneity and immediacy so much as gauge a change in the speed and

experience of time. It is therefore more in keeping with the musically minded

Bergson than the sculpturally minded Hulme.

Thus the Imagism of Pound is close to, although not congruent with, Hulme’s

Bergson-inspired ‘Classicism’. Consequently it is all the more important to inquire

how Hulme came to define Classicism as he did, since it is neither quite the same

as Lasserre’s definition, nor is it a link drawn by Bergson himself. Furthermore,

we still need to inquire why Hulme should have seen Bergson’s ‘aesthetic emotion’

as an effect of the ‘direct communication’ of sensation through visual language.62

From the desired ‘vague mood’ of his ‘Lecture on Modern Poetry’ in 1908 to the

denial of vagueness in Classical poetry around 1912, some other influence appears

to have made itself known which allowed Hulme—momentarily at least—to

reconcile Bergson’s ideas with the tenets of Action Francaise.

Hulme and James

Hulme’s equation of Classicism with a ‘dry, hard’ verse of concrete language,

which attempts to tear the veil from the world through evoking emotions that

are akin to vivid sensation, but which still free us from ‘the ingrained habit of the

technique’,63 was probably gleaned from Bergson’s great complement, and ally,

59 Pound, in Imagist Poetry, 130.

60 Rae, Practical Muse, 86–8.

61 Kenner, The Pound Era, 178; letter to Margaret Anderson, 17 November 1917.


63 Ibid., 69.



ownloaded from

William James. James’s theory of ‘stream of consciousness’ was explained through

physiology rather than through ‘elan vital’, but was nevertheless an influence on

Hulme’s own ideas and thus upon the Imagists as well. Hulme had reviewed

James’s A Pluralistic Universe positively in 1909, largely because it celebrated

Bergson for giving James ‘a complete system to justify the vague anti-intellectualist

sentiment that James has always felt’.64 However, despite James’ praise for

Bergson, there were notable points of divergence between the two philosophers

which Hulme’s review completely ignored, particularly with regard to emotion.

When describing the cause of emotion in the Principles of Psychology (1890),

James uses his reductivist theories to make one of his rare statements on art.

Having set himself the task of explicating the ‘subtler emotions’—the moral, in-

tellectual and aesthetic (which he explains in purely physiological fashion as being

those ‘whose organic reverberation is less obvious and strong’ than the ‘coarser’ like

grief and anger)65—James continues with his demystification of experience:

we must immediately insist that aesthetic emotion, pure and simple, the pleasure given us bycertain lines and masses, and combinations of colors and sounds, is an absolutely sensationalexperience, an optical or auricular feeling that is primary, and not due to repercussionbackwards of other sensations elsewhere consecutively aroused. To this simple primary andimmediate pleasure in certain pure sensations and harmonious combinations of them, theremay, it is true, be added secondary pleasures; and in the practical enjoyment of works of artby the masses of mankind these secondary pleasures play a great part. The more classic one’staste is, however, the less relatively important are the secondary pleasures felt to be incomparison with those of the primary sensation as it comes in. Classicism and romanticismhave their battles over this point. Complex suggestiveness, the awakening of vistas ofmemory and association, and the stirring of our flesh with picturesque mystery andgloom, make a work of art romantic. The classic taste brands these effects as coarse andtawdry, and prefers the naked beauty of the optical and auditory sensations, unadorned withfrippery or foliage. To the romantic mind, on the contrary, the immediate beauty of thesesensations seems dry and thin.66

The concordance with the views expressed by Hulme in ‘Romanticism and

Classicism’ is striking. First, James’s attribution of the ‘secondary pleasures’,

which include ‘picturesque mystery and gloom’, to Romantic art corresponds

64 T. E. Hulme, ‘The New Philosophy’ (ibid., 85–8, at 86). This review, published in TheNew Age 1 July 1909, is the earliest recorded evidence we have that Hulme had read James’swork. In it Hulme writes: ‘In Chapter V James describes in detail the problem which hehimself originated in his ‘‘Principles of Psychology’’ – the difficulty of explaining on anintellectualist basis the compounding of different states of consciousness’ (ibid., 85).Whether he had already read the Principles, or was encouraged to read them from the‘Bergsonian’ influence he discerned within A Pluralistic Universe, is difficult to establish.In ‘Notes on the Bologna Conference’ (The New Age, 27 April 1911), he quotes Bergsonquoting an anecdote of William James’s that ‘ ‘‘one look at a man is enough to convince youthat there you need trouble no further’’ ’ (ibid., 104–9, at 105). In the later essay, ‘BergsonLecturing’ (The New Age, 2 November 1911), Hulme also refers to the Varieties of ReligiousExperience when describing the sudden crisis of scepticism he felt before seeing Bergsonappear (ibid., 154–9, at 154).

65 William James, Principles of Psychology, 2 vols (New York, 1890), vol. II, 449.

66 James, Principles of Psychology, vol. II, 468–9.



ownloaded from

neatly to the impressions of ‘vagueness’ and ‘the infinite’ against which Hulme,

inspired by Lasserre, objected.67 Secondly, James’s insistence that the ‘classic

taste’ enjoys ‘the naked beauty of the optical or auricular sensations’, which is

purely a ‘sensational experience’ taken ‘as it comes in’, is close to Hulme’s belief

that the language of Classical poetry should be concrete and use visual stimulation

in order to express emotion, as a ‘compromise’ for a ‘language of intuition’ that

‘would hand over sensations bodily’.68 Thirdly, James’s rejection of ‘complex

suggestiveness’ and ‘vistas of memory and association’ in favour of immediate

‘sensational experience’ also explains why Hulme, when continuing to define the

Bergsonian ‘language of intuition’, states that it ‘makes you see a physical thing, to

prevent you gliding through an abstract process’.69 In other words, like James he

confines the pleasure of the image and the emotion it conveys to its immediatevisual effect, banishing the temporal from classical aesthetics, and he also explicitly

aligns the ‘aesthetic emotion’ with ‘sensation’, or ‘a physical thing’, rather than

reflection. Lastly, James’s understanding that the ‘immediate beauty of these sen-

sations seems dry and thin’ to the Romantic mind surely has its echo in Hulme’s

confident prediction of a ‘dry, hard, classical verse’ in ‘Romanticism and

Classicism’. Importantly, James does not specify whether he includes poetry or

not—a point either ignored by Hulme, or an active encouragement for him to

stress again the visual and concrete element of poetic language so as to extend this

definition to the medium of language, so that it too can appear to ‘hand over

sensations bodily’,70 despite its necessary supplementation.

Unlike Bergson and Ribot, James links Classicism directly to concrete, sensory

impressions, and Romanticism to the second order, mental experience of reflection

and vague impressions; he also equates ‘aesthetic emotion’ directly with the ‘pri-

mary’ pleasure of concrete sensation, without Bergson’s mystifying ‘spiritua-

lises’.71 In ‘Romanticism and Classicism’ this helps Hulme both to refine

further what the visual language of poetry can do, including the range of (concrete)

emotions it can provoke, and to define how the specific achievement of such

concretion is precisely part of the Classical agenda. Thus despite obvious differ-

ences with Bergson, James’s description of the ‘aesthetic emotion’ as a form of

immediate, sensory response helped Hulme to modify Bergson’s intuitional theory

to include only the sensational, align it with Classicism and distinguish it from

a Lasserre-like description of Romanticism. As such, it explains the anomaly

noted by Levenson that Hulme should have continued to profess Bergsonism

for so long,72 even when the French philosopher’s tenets increasingly jarred

with all else that Hulme was by then absorbed in.


68 Ibid., 70.

69 Ibid.

70 Ibid., 69, 70


72 Levenson, Genealogy of Modernism, 86.



ownloaded from

Indeed, James’s analysis of aesthetic emotion may also have contributed to the

inconsistencies of Hulme’s essay ‘Bergson’s Theory of Art’, in which the temporal

and reflective elements of Bergson’s theory are already marginalised by the spatial,

visual and immediate, and where ‘aesthetic emotion’ is equated with the ‘freshness’

and ‘exhilaration’ caused by the ‘direct communication’ of ‘sensation’73 (or James’s

‘optical and auricular feeling’)74—something which Bergson himself had consist-

ently refuted. This points to Hulme having read James’s book before needing

to align Bergson’s theories on art with those of Action Francaise, and may ex-

plain why he felt so confident about opposing Lasserre’s claim that Bergson was a

Romantic.75

However, it is not difficult to see where James’s distinction also diverges from

Hulme’s desired aims, helping to cause irreconcilable contradictions in the latter’s

theory which were unnoticed by the English theoretician when appropriating

James’s ideas. The unique nature of the aesthetic emotion which Hulme observes

in good poetry is not for James in any way separate from normal emotions. James has

already spent several pages at this point in the Principles discussing the primary,

‘coarser’ emotions such as grief, joy, fear and anger, and has just explained that they

are in fact illusions: vague names awarded to physiological reactions, themselves the

results of external stimuli. When introducing the so-called ‘ ‘‘subtler’’ emotions. . .the moral, intellectual, and aesthetic’, he denies that they show any fundamental

difference from the emotions he considers to be ‘primary’, and in any case scotches

their reality as purely mental acts. The idea that art is capable of conveying complex

aesthetic emotions—as Hulme believed—is wrong since the aesthetic is actually

‘primary’ ‘pure and simple’. Therefore, like other emotions, aesthetic emotions are

simply straightforward ‘organic reverberation[s]’ to ‘certain lines and masses’: an

‘absolutely sensational experience’.76 The contrast with Bergson’s denial that in-

tense emotion is of the same order as sensation in Essai sur les donnees immediates de la

conscience could not be more complete.77

Furthermore, insofar as he attributes the ‘mystery and gloom’ and ‘repercussion

backwards of other sensations elsewhere’ to ‘tawdry’ Romantic art, James effect-

ively excludes ‘complex suggestiveness awakening vistas of memory and associ-

ation’ from Classical art. This ‘complex suggestiveness’ may include the sense of

infinity or ‘vagueness’ in language condemned by Hulme—the ‘slither’ of Ezra

Pound78—but also surely includes the ‘vital complexities’ that Hulme, referring

to Bergson’s own terminology, describes as inherent to the best kind of poetry,

73 Ibid., 201.

74 James, Principles of Psychology, vol. II, 469.


76 Ibid., II 449, 467–8.





ownloaded from

and which Hulme, curiously, still calls ‘intensive’ (i.e. involving memory and real

duration), rather than ‘extensive’ (spatial).79

The dismissal of ‘complex suggestiveness’ is as much at odds with Modernist

aesthetics as it is with the Kantian sublime. While both James and the Imagists

might agree that Classical art should treat its subject directly and should preclude

‘picturesque mystery and gloom’, James’s denial of recalled sensations also rejects

aspects of the image which are still essential to the supposedly ‘dry, hard’ Imagism.

James’s dismissal of the vague and secondary certainly does cause some contra-

diction in Hulme’s essay between its rejection of all that is not immediate and

concrete, and its still stated belief in the rarity of aesthetic emotion and ‘vital

complexities’ caused by the image. However, in the movement of Hulme’s

views on concrete, visual language to Pound’s ‘Doctrine of the Image’, James’s

exclusion of the non-immediate becomes peculiarly inapposite and helps to pro-

mote aesthetic tensions. For Pound and the Imagists, memories, associations

and all sorts of ‘secondary pleasures’ are inbound with the aesthetic image, even

if they are implausibly raised out of the limits of time and space in an ‘instant’.

Pound’s central definition of ‘the Image’ as ‘intellectual and emotional complex’80

actually has a lot in common with the ‘complex suggestiveness’ denied to

Classicism by James.

Understanding the incompatibility of James’s views on Classical art which

Hulme communicated to Imagism, and Hulme’s further promotion of unusual,

aesthetic emotions through poetry, explains much of the early disillusionment with

the Imagist project registered by H.D., Aldington and others, and Pound’s

own movement towards the more process-centred ideas of Vorticism. H.D. com-

plained late in life to ‘Delia Alton’ (really herself) about the inappropriate way her

79 Hulme, Collected Writings, 70, 71, 72. Hulme’s description of these complexities inpoetry (or ‘multiplicit[ies]’) as ‘intensive’ rather than ‘extensive’, and his brief discussionof them in the essay (ibid., 72), is firmly in keeping with Bergson’s understanding thatintensive states or complexities are grasped by intuition and have a qualitative and organicfeeling, while extensive states are spatial and grasped by the intellect (Bergson, Essai,171–3). However, the continued use of this term contradicts Hulme’s attempt to portrayaesthetic emotion as immediate and inbound with visual effect, since Bergson insiststhat: ‘Isoles les uns des autres, et consideres comme autant d’unites distinctes, les etatspsychologiques paraissent plus ou moins intenses. Envisages ensuite dans leur multiplicite,ils se deroulent dans le temps, ils constituent la duree’ (Bergson, Essai, ibid., 170). In‘Romanticism and Classicism’ Hulme in fact allows no true aesthetic epithet such organicduration and development in creating its effect. This is somewhat ironic, since in his essay‘The Philosophy of Intensive Manifolds’ (23 November to 14 December 1911), Hulmeoutlines both the importance of ‘real time’ over space (defined as a ‘ripening’) for revealingthe ‘interpenetration’ of ‘the more fundamental self’ presented in intensive ‘manifolds’, andalso the importance of memory over matter for its realization (here referring, of course, toBergson’s later work, Matter and Memory [1896]). (Hulme, Collected Writings, 170–90, at181, 176, 184). He notes that such intensive manifolds are also revealed in the intuitions ofliterary creation (ibid., 178).




ownloaded from

poetry had been called ‘crystalline’ (a term implying stasis), but then offered an

alternative:

For what is crystal or any gem but the concentrated essence of the rough matrix, or the

energy, either of over-intense heat or over-intense cold that projects it? The poems as awhole. . .contain[s] that essence or that symbol of concentration and of stubborn energy.

The energy itself and the matrix itself have not yet been assessed.81

As Gary Burnett writes:

Crystalline poetry, as she points out, suggests a poetry of stasis, a poetry metaphorized as

‘crystal or any gem’, as an objectified artifact. Such a false suggestion does recall the dom-

inant version of Imagism; T.E. Hulme’s earliest formulations of the new verse underscoreits sculptural stasis.82

Further declaring that Hulme’s Classicism was too hard for H.D. and as repel-

lent to her as the planes of Futurism and Vorticist style, Burnett nevertheless

perceives in H.D.’s own concept of a matrix an attempt to underline the dynamic

creative process and imaginative re-creation of impressions involved in the concept

of the vortex. Similarly, Burnett shows how Pound’s need to develop logopoeia

alongside phanopoeia—that is, poetry aspiring to music and poetry aspiring to

sculpture83—demonstrates the insufficiency of the Hulmean emphasis on sculp-

ture and concreteness to create dynamic poetry.84 It also shows the extent to which

the devotion to direct treatment of the thing and purely visual language simply

cannot allow the concept of the image as ‘presented complex’ to combine intel-

lectual and emotional ideas as though lifted out of time.85 Contrary to Patricia

Rae’s opinion, the ‘pragmatic turn’ in the theories of Hulme and Pound clashes

with their attempt to create unusual emotional effect, the William James and Henri

Bergson inheritances being ultimately irreconcilable due to different understand-

ings of both what ‘aesthetic emotion’ actually is, and of how it may be delivered

through the poetic artefact.

81 ‘H.D. by Delia Alton’ [‘Notes on Recent Writing’], Iowa Review, 16 (1986), 174–221,at 184.

82 Gary Burnett, H.D. Between Image and Epic: The Mysteries of her Poetics (Ann Arbor/London, 1990), 12.

83 Ezra Pound, ‘The Later Yeats’, in T. S. Eliot (ed.), Literary Essays of Ezra Pound(London and Boston, 1954), 378–82, at 380.

84 Burnett, H.D. Between Image and Epic, 14.

85 Patricia Rae shows the extent to which Pound may have attempted to invest his doctrineof the image with ideas gleaned from the empirical observations of mysticism, and how farhis respect for mystical experience was mirrored in James’s own explorations in the Varietiesof Religious Experience (Rae, Practical Muse, 79–88). That said, the overt discussions ofaesthetic experience, as opposed to the psychical, in James’s work, veer widely from theattempts to replicate the mystical and uncanny which critics discern in Pound’s ‘Doctrine ofthe Image’, with the goals of Pound’s ‘complex’ necessitating further temporal experienceand association.



ownloaded from

James and Imagism

While we may accept that Pound’s views on art cannot ultimately cohere with

Hulme’s James-inspired definition of Classicism, nonetheless we can at least see

the working of James’s principles of psychology in the best Imagist poems, even if

these principles involve aspects from the Principles of Psychology which James

thought best kept out of ‘Classical’ art. While both James and Bergson dismissed

the atomic structure of association, James nevertheless understood association as

an important aspect of thinking. Its principles were primarily ‘mental habit’

(including the old principle of contemporaneity or ‘contiguity’, in which when

we think of one thing we also immediately recall what was next to it on account of

the central nervous system’s laws of habit), ‘recency’ and ‘vividness’.86 Also

included was the falsely conceived ‘emotional congruity’, in which our mood on

seeing an object helps us to recollect certain ideas which we also relate with that

mood, when a different mood would provoke wholly different ideas on our seeing

the same object.87 Such principles of association we can see working in the ‘energy’

of the first stanza of H.D.’s ‘Hermes of the Ways’, a poem which Kenner correctly

understands as using images throughout to express the process of her thought, and

with ‘no detail not germane to such thinking, no detail obligated merely by

pictorial completeness’.88 However, it is worthwhile re-examining the first

stanza to understand how this presentation of process is achieved:

The hard sand breaksAnd the grains of itAre clear as wine.89

The immediate principle of ‘vividness’, in which two similar images recall each

other by image association, can be seen as working here in the grains of sand and

the wine. Or, we can understand both images as being associated with the same

mood.However, the major reason for the imagery is an ‘awakening of distant vistas of

memory and association’,90 making the comparison of grains of sand to wine a

form of transumption, passing over what John Hollander would call ‘unstated

middle terms’, which are nevertheless an integral part of the image’s effect.91

The image, in keeping with H.D.’s Hellenic interests, is influenced by Homer’s

famous description of the sea as ‘wine-dark’, first employed when Achilles gazes

over the Aegean after the cremation of Patroclus. Here, however, the epithet is

reversed to represent clarity: or, ironically, opacity (since wine is not clear), and

86 James, Principles of Psychology, vol. I, 561, 575.

87 Ibid., vol. I, 576–7.


89 H. D., Collected Poems, 1912–1944, ed. Louis L. Martz, 2nd impr. (New York, 1991), 37.


91 John Hollander, The Figure of Echo: A Mode of Allusion in Milton and After (London andBerkeley, 1981), 114.



ownloaded from

thus by reference lack of communication. Thus the image is a complex due to its

synecdochal relation of the grains of sand to the sea and the Homeric description

of the sea as the ‘wine-dark deep’.92 Both could be understood as being associ-

ations of ‘neural habit’ as much as of ‘vividness’, since the first stage of the asso-

ciation relates the grains of sand synecdochally to the sea which is breaking into it,

and the second part is a relation of the sea itself to a traditional literary epithet.

However, in both cases, despite its concrete, visual nature, the image conforms

more to James’s definition of Romanticism than his understanding of Classicism,

since in alluding to and transuming the Homeric epithet, the simile is opening a

‘vista of memory and association’ beyond the immediate context, and in doing so

attains ‘complex suggestiveness’ by indicating the original context in which it was

made (i.e. Achilles grieving Patroclus). Furthermore, it suggests opacity both

visually and metaphorically (literally and symbolically an image of gloom), thus

explaining why Kenner can conclude that the poem ‘is ‘‘about’’ her taut state of

mind, a wried stasis like a sterile homecoming’.93 The preciseness of the image is

indeed not crystalline: its allusive nature and metaphorical associations fill it with

secondary pleasures, which also cannot be appreciated simply in an instant of time,

but must be dwelt over and allowed to resonate.

If we take the image as expressing an unusual ‘complex’ of emotion through

objective imagery, it is one which thwarts both James and Imagist doctrine by

implying gloom and vagueness, and James specifically by relying on reference and

consecutive association. If, however, we read these lines purely visually, then it

portrays no especial emotion at all but is a ‘harmonious combination’ of pure

sensations which makes a sensory and not an emotional impact.94 Therefore,

whatever our attitude to the peculiar emotional effect of the poem, the

James-inspired prioritisation of primary sensation latent in Hulme’s theory and

Pound’s Imagist ideals has to be broken for that effect to be conveyed.

In summation, the Imagist doctrine of objectivity and immediacy does not de-

liver the emotional peculiarity which the writers insist it can, unless the image

complex is allowed to float temporally in the mind of the reader beyond Pound’s

‘instant of time’.95 This incongruity between Pound’s ideals and the poets’

92 The passage occurs just after the Achaeans and Myrmidons have prepared the funeralbarrow for Patroclus. Achilles cuts off his blond locks by the river Spercheius. ‘Then inheaviness spake he, gazing on the wine-dark deep: ‘‘Far otherwise, Spercheius, did myfather Peleus vow to thee that returning thither to my dear native land I should cut my locksfor thee and sacrifice a holy hecatomb, and there offer fifty rams at the founts, where is thyplot and fragrant altar. Thus vowed the old man, but thou hast not fulfilled his desire.Wherefore now since I shall not return to my dear native land, I would give my locks tothe warrior Patroclus to take unto himself.’’ ’ (Homer, Iliad Books XXII and XIII, tr.H. Hailstone (Cambridge and London, 1890), 22–3).






ownloaded from

practice occurs precisely because Imagism’s guru, Hulme, attempted to align his

Bergsonian intuitionism with Action Francaise’s Classicism, and was dependent

upon William James’s description of aesthetic emotion in Classical art for doingthis. In the process, however, he reduced Bergson’s own understanding of aes-

thetic emotion to immediate sensation, and deprived the Imagists of the temporal

and reflective elements which Bergson understood as important to art’s aestheticeffect.

University of Hull



ownloaded from

Contradictory Images: the Conflicting Influences of Henri Bergson and William James on the aesthetics of T.E. Hulme, and the Consequences for Imagism

Documents