Navigation – Plan du site

AccueilNuméros4RecherchesTwo Political Itineraries of Euro...


Two Political Itineraries of European Bergsonism: Georges Sorel and Thomas Ernest Hulme

Tommaso Giordani


À travers l’analyse des trajectoires de Georges Sorel (1847-1922) et de Thomas Ernest Hulme (1883-1917), l’article se propose de creuser la question de la politisation de la philosophie d’Henri Bergson dans l’Europe du début du XXe siècle. La première partie examine les parcours qui menaient les deux penseurs à découvrir la pensée Bergsonienne et à en devenir avocats et défenseurs. En dépit de quelques différences – Sorel acceptera seulement les arguments psychologiques des Données, Hulme sera plus intéressé à la métaphysique Bergsonienne – on trouve que les deux étaient attirés essentiellement par les éléments anti-déterministes qu’ils trouvaient dans les œuvres du philosophe français. Mais quand il s’agit de mettre cette opposition au déterminisme au centre d’une théorie politique, leurs parcours divergèrent nettement. Si, d’un côté, Sorel n’aura aucun problème à incorporer des idées Bergsoniennes dans sa révision du marxisme, Hulme, d’autre côté, sera incapable de fonder sa politique sur une métaphysique du mouvant. Cette différence significative est expliquée en termes générationnels. Si la cosmologie ouverte de Bergson fut réputée comme excessivement relativiste par Hulme, c’est en vertu du plus grand cout que l’engagement politique comportait pour sa génération, d’où sa conviction que la politique pouvait seulement être enraciné dans des principes absolus, immutables, et éternels.

Haut de page

Entrées d’index

Mots-clés :

Sorel, Hulme, Bergson, bergsonisme
Haut de page

Texte intégral

Research leading to these results has received funding from the European Research Council under Grant Agreement No. 757873 (Project BETWEEN THE TIMES). The author would like to thank Henry Mead for help with certain subtleties in Hulme’s thought.

Introduction: Bergson, politics, and Bergsonism

  • 1 All translation from French sources are the author’s.

1The question of Bergson and politics occupies a marginal position in Bergsonian scholarship, as engagements with his work tend to portray a figure whose political dimension is tangential at best. For Gilles Deleuze, Bergson was the inventor of “one of the most fully developed methods in philosophy”, and it is easy to find portraits of Bergson as a psychologist, as a philosopher of biology, or as a thinker of time (Deleuze 1966, 1; Kolakowski 1985; Posteraro 2022). It is much harder to find works that frame him as a political writer. Though there is some truth in this apolitical view of Bergson, it is important to appreciate how much this view is wound up with the reception of his writings. The French and European fame of the philosopher was established roughly in the decade before the Great War. Looking at the works written in that period – from the Données to the Introduction à la métaphysique, from Le Rire to L’Évolution Créatrice – little that is immediately political jumps to the eye. The work which deals most directly with ethical and political questions – Deux Sources de la Morale et de la Religion – was published in 1932, in a phase in which, at least in France, the vitality and popularity of Bergsonian philosophy had turned into “the cult of an ossified glory” (Azouvi 2007, 293).1

2And yet, the question of Bergson and politics can be posed, and has been posed, in three different ways. The most straightforward one is to treat Bergson as a political philosopher. This theoretically oriented approach involves focusing on Bergson’s works and evaluating the extent to which they furnish answers to the canonical problems of political philosophy:

Interrogating the stakes and problems of a Bergsonian politics means searching Bergson’s oeuvre […] for theorisations of common life […] it is starting from this work that it becomes possible to confront Bergsonian philosophy and the tradition of political philosophy, that is, to locate the former within the latter (Caeymaex, Janvier, and François 2012, 172).

3This approach has been attempted in both Francophone and Anglophone scholarship by philosophically-oriented researchers and tends to have at its core exegeses of Les Deux Sources de la morale et de la religion (Kisukidi 2012; Zanfi 2012; Lefebvre 2012).

4A second approach consists in seeing Bergson as an historically situated political actor. Here, Bergson is here presented less as a name linking together a body of philosophical writings and more as a man with passions, interests, preferences, axes to grind, and political goals to achieve. The method involves situating this Bergson in a political field and deducing, from his agency within it, a political positioning. The most accomplished work in this direction is certainly Philippe Soulez’s 1989 Bergson politique. Even in an approach of this kind, however, Bergson’s philosophical output remains important. Soulez’s book is based on a distinction between the author – the philosopher of the moi profond and of élan vital – and the historically situated political man – who wrote war propaganda, lobbied the US for intervention, and dedicated energies to the construction of international civil society. But this distinction between philosopher and political man is not posited in order to concentrate on the latter at the expense of the former. Instead, given that “it is as a philosopher that Bergson became an organic intellectual” (Soulez 1989, 35), its point is to explore the dialectic between the two. This way of posing the question illuminates, through historicization, Bergson’s own political commitments. It also sheds light on the dynamics of politicization of thinkers who, like Bergson, dominated the public sphere, as well as offering an interesting case study of the forms of political agency available to them.

5And yet, there is one issue that this approach can touch only tangentially: that of Bergsonism, or, put otherwise, of what Bergson and his philosophy were to their readers. To deal with this question a different approach, focussing more on reception than on intellectual production, is needed. It is this approach that will be adopted in the essay. Studies on Bergsonism have covered substantial ground, from cultural histories and histories of 20th century French philosophy through the category of “post-Bergsonism” (Azouvi 2007; Bianco 2015) to works exploring more canonically political ramifications (Antliff 1993; 2007; Diagne 2011). This essay builds on these latter works but diverges from them in that the space of reference is neither French nor global, but European.

6In this essay, through two case studies – those of Georges Sorel (1847-1922) and Thomas Ernest Hulme (1883-1917) – I intend to examine not so much either Bergson’s philosophy or the thought of these two protagonists. Instead, I want to focus on that ambiguous middle ground which we can label Bergsonism, paying particular attention to its political potential. The Bergsonian trajectories of Sorel and Hulme will, thus, be read in the light of the following questions: how did they arrive at Bergson? In the light of which problems, and against which positions, did they read him? In which ways did they politicize Bergson’s philosophy? Answers to these questions will allow us to go beyond Sorel and Hulme and contribute to a wider understanding of Bergsonism in fin de siècle Europe, clarifying what conceptual and, more importantly, political operations it allowed one to perform.

Roads to Bergson

7For both Hulme and Sorel, the engagement with Bergson was of some importance. Sorel commented on Bergson from his early writing years until his death in 1922, and most conspicuously appealed to the philosopher’s authority for a justification of his arguments in the Réflexions sur la violence (Sorel 1910, 34-35). Hulme discovered Bergson while teaching at a language school in Brussels, after he had been expelled from Cambridge (Ferguson 2002, 70), and first read the Données back in England in 1907 (Martin 1967, 137). He would go on to become a determined advocate of the French philosopher, helping his ideas in crossing the Channel, translating the Introduction à la métaphysique and contributing a bibliography of mostly European studies on Bergson in the English translation of the Données (Bergson 1912; 1910, xiii-xxii). The first question we shall ask our protagonists concerns the reasons that allowed Bergson’s work, or at least certain of his arguments, to make a deep impact on the evolution of their thinking.

8At a very general level, this can be explained by looking at the timing of the respective first meetings with the French philosopher. Both Sorel and Hulme encountered Bergson’s texts at the beginning of their writing careers. Sorel began publishing in the late 1880s and first discussed Bergson’s Essai in 1894 (Sorel 1935), whereas Hulme read Bergson in 1907, two years before his first publication. In both cases, we have thinkers in formative phases, grappling for the first time with some of the central problems of their careers, trying to give them a certain conceptual architecture.

9These central concerns, moreover, largely overlapped. Both Sorel and Hulme expressed reservations about scientific determinism, and both cultivated an interest in a space of experience beyond the causalities of nature. As we shall see, they pursued these concerns in different fields and in different ways. Sorel, at the time, was interested in a general epistemology covering both nature and society. Possibly because the older man, he was the more epistemologically cautious of the two, holding on, until 1896 (Sorel 2019), to a strong conception of science, capable of producing causal laws. Hulme, too started from epistemology, then turned to language and its poetic – and eventually political – use. Much more of a relativist than Sorel, he jotted down in his early diaries that “in spite of pretensions to absolute truth, the results of philosophy are always tested by the effects” (Hulme 1994, 7).

10For Sorel the movement towards Bergson took place in the middle of a sustained epistemological reflection on the possibility of social science (Giordani 2018a). The problem with which the freshly-retired engineer was struggling in the early 1890s was the attempt to reconcile a strong conception of science as capable of producing causal laws with a commitment to the freedom – and indeterminacy – of human agency. The tension between the two – a deterministic cosmology being required for the possibility of science, but at the cost of the freedom of human agency – was evident to Sorel himself: “s’il y a liberté, il n’y a pas de connaissance possible” (Sorel 1935, 177).

11In the spring of 1894, Sorel came up with a solution to this problem in L’Ancienne et la nouvelle métaphysique (henceforth ANM), a work that was the most systematic statement of his epistemological thinking up to that point. It is here that Sorel first mentioned Bergson, whose Données he described as a “vigorous tree rising in the desolated steppes of contemporary philosophy” (Sorel 1935, 167-68). With Bergson, and, as we shall see, Hulme, Sorel shared a pragmatic view of language, conceptuality, and, to a point, rationality. Embedded in this pragmatic view was, however, a vision of human agency as undetermined, and hence always potentially creative. It was to Bergson’s characterization of deep interiority that Sorel turned to vindicate this, as well as to, more generally, reconcile scientific rationality with human creativity. In this reconciliation, determinism remained true, but was circumvented in two ways.

12The first one was that the Bergsonian moi profond constituted a, so to speak, blind spot of science. Since our whole conceptual apparatus evolves in practical confrontation with the world, Sorel argued, it cannot, by definition, understand interiority, because interiority is absence of this confrontation. In these moments of disengagement we access the “fundamental self”, a being “agitated by emotion and freed from all that comes from the outside”(Sorel 1935, 178). Here, though determinism is not broken, it is impossible to see it at work, and we gain a degree of freedom: “[Bergson] is correct in seeing in affective life what is most individual”(Sorel 1935, 178). The impossibility of tracing the causal processes that determine our inner life means that a space for creative agency exists, and it is in this space that Sorel will more fully vindicate human freedom.

13The second way in which determinism was circumvented was through a conception of freedom as environment-modifying agency which, despite being based on Bergsonian ideas of selfhood, moved in a different direction. For Sorel, freedom could not be conceptualized as exception to the laws of nature. Instead, it was the power to arrange the causalities of nature in a new order, creating in this way an artificial environment:

The existence of the artificial milieu is the fundamental requirement for our freedom […] We are free in that we can construct devices that have no model in the cosmic milieu; we change nothing in the laws of nature, but we are able to create sequences whose ordering is our own (Sorel 1935, 164).

14It is worth underlining that, despite the importance of Bergsonian interiority in producing a space in which we are “able to create sequences whose ordering is our own”, Sorel’s notion of freedom moved towards exteriority. As can be seen from the passage, freedom requires both creativity and agency. Creativity, granted, is only possible in interiority. But agency can only be found in the attempt to overcome the practical constraints and resistances offered by the “cosmic milieu”. This confrontation with exteriority, both natural and social, is an essential aspect of Sorelian freedom. Thus, whereas for the Bergson of the Essai “the social is the domain of space, of utility and habit, the reign of regularity, of repetition, of rigid and fixed roles” (Zanfi 2012, 226), for Sorel it is the space where creative, environment-modifying, agency can occur.

15In contrast to Sorel, Hulme discovered Bergson before he began publishing, in a period spent between Cambridge, London, Brussels, and Canada. It is from the then-unpublished writings of these travelling years – two notebooks entitled “Cinders” and “Notes on Style and Language” – that we have to reconstruct the concerns that brought him to Bergson. If we look at the earlier “Cinders”, we see a preoccupation with issues of epistemology, science, and rationality. Unlike Sorel’s careful attempt to balance scientific necessity with human creativity, Hulme’s perspective is more openly relativistic, affirming that “there is no such thing as an absolute truth to be discovered” he affirms (Hulme 1994, 8). His anti-foundationalism – “there are no ultimate principles, upon which the whole of knowledge can be built once and forever” – questions the “pretensions” of science to truth and mocks the “Hegelians who triumphantly explain the world as a mixture of ‘good’ and ‘beauty’ and ‘truth’” (Hulme 1994, 16 and 8).

16If we probe this radical relativism, we find a view in which human rationality is likened to a game played on a chessboard drawn in the ashes of a dying fire: below the ash, lay the cinders, the ultimate and yet unorganized reality. Between our conceptual and linguistic instruments and that which they try to grasp there is, in other words, a chasm. In a highly personal idiom, Hulme portrays language as a “disease” and as intrinsically unable to penetrate into the essence of “the real things”:

There is a kind of gossamer web, woven between the real things, and by this means the animals communicate. For purposes of communication, they invent a symbolic language. Afterwards this language, used to excess, becomes a disease, and we get the curious phenomena of men explaining themselves by means of the gossamer web that connects them (Hulme 1994, 8).

17The reflection on language was, for Hulme, an important, if ambiguous, space of engagement with Bergson. On the one hand, like Bergson, he saw language and rationality as emerging from the requirements of biological existence, and thus as pragmatic, and intrinsically ineffective, instruments, ultimately unable to grasp the real. On the other hand, however, he spent a great deal of time reflecting about language and did not dismiss it as easily as it may at first appear. Already in “Notes on Style and Language”, he suggested a theory of poetic language as, in some sense, being able to go beyond abstraction and “nearest to flux and real basic condition of life” (Hulme 1994, 27). Poetic uses of language would thus break out of the shackles of conceptual thought by turning “words into visions, in realities we can see” (Hulme 1994, 24). This view of poetry as a, so to speak, anti-linguistic use of language not only can be likened to analogous treatments of language in Bergson and Sorel (Fujita 2022), but also, as we shall see, will be the basis on which he will attempt his own application of Bergson’s method of intuition.

18The points of contact between these early philosophical sketches and Bergson’s ideas – especially the pragmatic view of language and the vision of the real as an unstable flux – help us in making sense of Hulme’s first enthusiasm for Bergson. As he would admit, “If the reading of 'Les Données Immédiates' was an influence and an excitement it could only have been so because it bore some relation to something which already existed in my own mind” (Hulme 1911d, 587). In his first published pieces, we see a more mature re-statement of the same ideas, combined with an attempt to push forward the programme of a “new philosophy” of openness and flux, represented by William James’ pragmatism, by Jules de Gaultier’s bovarysme, and above all by Bergson, whose critique of rationalism “is the only radical attack, the only attack which concedes nothing” (Hulme 1909a, 198; 1909c).

19If we compare the roads to Bergson of Hulme and Sorel, it is clear that the similarity of the conceptual puzzle with which all three were dealing is of some importance. This problem – a classical of 19th century, post-Kantian philosophy, and, for some, the fundamental problem of modern philosophy tout court (Habermas 2000, 28) – consisted in asking the question of the relation between human cognitive instruments and the world. As soon as rationality is subjectivized, seen as a human capacity and not as the structuring principle of the real, the question of its relationship to its objects becomes relevant. Bergson, Sorel, and Hulme, like many others in the period, dealt with different versions of this large conceptual problem. All three sought a vindication of human agency, and shared an opposition to the idea that the fundamental structures of the real could unproblematically be garnered from observation – a position which would require, amongst other things, a deterministic cosmology which all three, to different degrees, opposed.

20Both, thus, took anti-deterministic arguments from Bergson, but accepted their validity at different levels. In 1894, Sorel could only read the Données, and therefore accepted essentially a psychological argument. As Bergson moved on to metaphysics and cosmology, Sorel would not follow him, rejecting the cosmological vision of the Evolution Créatrice and suggesting other, earthlier, pursuits to Bergson (Sorel 1907b). Things were different for Hulme, whose early writings made bold metaphysical claims about reality. Not only did he take Bergson’s metaphysics more seriously, but also reflected on the application of the method of intuition far more than Sorel did. His view of poetic – but, as we shall see later, also political – language as an intuitive plunge into the flux of the real, as “a compromise for a language of intuition”, preventing one from “gliding through an abstract process” testify to this (Hulme 1909b, 315).

Political Bergsonisms

21The connection between, on the one hand, discussions on epistemology, science, determinism and, on the other, politics may at first appear enigmatic. How did these anti-deterministic Bergsonian ideas sustain political conclusions in the thought of Hulme and Sorel? How did positions on language, philosophy, and psychology become political? What was the role of an ontology of becoming – whether social or metaphysical is irrelevant – in justifying and sustaining political arguments? In this process of politicization, the two differ substantially. Whereas Sorel drew political conclusions from the psychological argument about free human agency, Hulme, after initially going in a similar direction, eventually ended up convinced that politics could not be grounded in a metaphysics of movement and transformation.

22If we look at Sorel, the path from epistemology to politics is relatively straightforward. In the ANM, he had established a dual epistemology. On the one hand, the causality of nature, while on the other the openness of the artificial, man-made, environment, the two connected by experimentation, where laws of nature could be discovered by examining natural necessities through man-made concepts and tools. As he became increasingly interested in the social sciences, however, he abandoned his view of science as capable of expressing causal laws, and, through the interactions with Antonio Labriola and Benedetto Croce, as well as through his reading of Giambattista Vico, came up with a conception of social scientific explanation of a very different kind, unconcerned with universal laws and focussed on explaining how given historical configurations could be connected to the agency of historical groups (Giordani 2018b). Contrasting what he saw as Marx’s method with that of Durkheim, for example, he wrote that whereas the latter explained social action by appealing to “imaginary entities, to the needs of the collective soul and other sociological absurdities”, Marx’s theory of class struggle allowed explanation in terms of “real humans collected in groups acting upon social life” (Sorel 1895, 24).

23As should be clear, this insistence on agency implied a commitment to its fundamental indeterminacy. This insight – first articulated in 1894 in Bergsonian language – was to be fundamental in Sorel’s revision of Marxism and in the elaboration of his syndicalism, two highly intertwined projects to which he dedicated the last half-decade of the 19th century. In part responding to republican critiques of Marxism as a shallowly materialistic theory, in part pursuing his own intuitions about the fundamental openness of social world, Sorel elaborated a Marxism shorn of economic determinism, incapable of historical teleology, and abandoning its status as a supra-historical scheme of social development in order to embrace a much more embedded vision of itself as the self-consciousness of a specific historical formation, the proletariat. Without going into details, it should be underlined how all of these three elements can be derived from the more fundamental commitment to the freedom of human agency. In the face of this freedom, not only every notion of historical necessity had to be abandoned, but it also resulted impossible to assert that economic and technological factors were the ultimate determinant of human agency. No such ultimate determinant could, in fact, exist. All the social scientist could do was elucidate, on a case by case basis, how different factors constrained and shaped human agency, bearing in mind that, ultimately, this agency always exceeded the conditions in which it was exerted, and could not fully be explained by them. It retained the capacity for discontinuity, for a creative surplus.

24This non-determinist Marxism had political consequences. If the historical process was fundamentally open rather than leading necessarily to socialism, and if, moreover, the modalities through which collective political subjects emerge could not be reduced to economic transformations, it followed that the road to socialism could be travelled only by making sure that a collective subject capable of enacting the socialist transition did emerge. If collective agency was the core of the historical process, then socialism could only arrive through it. Marxism, in short, became a theory of class, and socialist strategy became a matter of class development. This insistence on class, on the autonomy of proletarian institutions, on the development of their political, administrative, and juridical capacities is the defining aspect of Sorelian syndicalism. It is an element that one can find in his earliest political writings from 1898, in the Réflexions, and in his last writings: take, for example, his volume of collected works, fittingly entitled Matériaux d’une théorie du prolétariat (see Sorel 1898; 1919).

25As anticipated above, Hulme’s journey from epistemology to politics is more complex and contradictory than Sorel’s, even if initially he threaded the same path, seeking to establish politics on a Bergsonian view of the self. Illustrative of this early attempt is “A note of the art of political conversion”, a text from February 1911 in which a Bergsonian notion of selfhood was deployed to illustrate how to effectively produce political propaganda. In it, Hulme developed a theory of the political use of language as based on “a recognition of the basic fact of the absolute impotence of a mere idea to produce any change in belief”. To be effective, political language must bypass the purely linguistic-rational sphere and reach “the motor and emotional aspects of mind” (Hulme 1994, 207). In substantiating this position, Hulme predictably mentioned a shift, produced “during the last fifty years in philosophy”, through which “instinct and emotion have asserted their rightful place” vis à vis rationality, to the point that “the intellect is regarded merely as a subtle and useful servant of the will […] Bergson, Le Roy, Croce, Eucken, Simmel, are all anti-intellectualists” (Hulme 1994, 208).

  • 2 The immediate context for this text was the Parliament Act of 1911, a reform that curtailed the pow (...)

26And yet, only a few months later, in August of the same year, we see Hulme going in a different direction. Tackling the issue of parliamentary reform in an article entitled “Progress and Democracy”, he underlined the risks of what he called a “pure democracy”, justifying his position on what at a first glance appears as a critique of progress, a notion against the Hegelian version of which he had, as we have seen, already protested (Hulme 1994, 221).2 Upon further inspection, however, it becomes clear that Hulme is not merely contesting the idea of gradual, or even necessary, social improvement. Much more radically, he is beginning to suggest that change and transformation are invalid categories of political thought:

Here it seems to me that you get the fundamental difference between the Conservative standpoint and the democratic. The Conservative does not believe in progress […] He believes that man is constant, and that the number and types of the possible forms of society are also constant […] The Radical, I take it, is the exact opposite of this (Hulme 1994, 222).

27It really should be underlined how this is much more than an argument against progress: it is an argument against history and time, presenting politics as the art of dealing with two immutable entities, man and society. The contrast between a view of politics and humanity as essentially timeless, and a Bergsonian metaphysics of flux and openness should be apparent to the reader, and it is a contradiction through which Hulme will make his way in the few remaining years of his short life. What prompted this change? In April 1911, Hulme travelled to Bologna to cover for the New Age the international philosophy congress held in the Italian city. On the way, he stopped in Paris, where he met, amongst others, Pierre Lasserre, a prominent Action Française intellectual who, in 1910, had engaged in a long and vitriolic anti-Bergsonian polemic in the pages of the party newspaper (Lasserre 1910).

  • 3 For example, he was sceptical of Action Française’s anti-Semitic attacks on Bergson, writing that “ (...)

28Though Hulme’s agreement with the politics of the Action Française was partial, it was the group’s cultural politics that he endorsed.3 More specifically, he fully took on board – and further developed – a distinction developed by Lasserre in his doctoral thesis between a “romantic” and a “classical” spirit (Lasserre 1907). As Hulme’s own account of the discussion with Lasserre reveals, it was this distinction that first brought him to elaborate doubts regarding Bergson. The Neo-Royalist position was a belief “in the existence of laws which express what we know of the necessary and permanent characteristics of any social and political order”, laws which could be deduced from “the experiences of history” (Hulme 1911c, 40). It was on the basis of these laws that, for Lasserre, an argument against democracy could be developed. However, Bergson changed the rules of the debate between democrats and authoritarians, because, in Lasserre’s view, Bergson allowed one to deny the existence of these laws. “Bergson has now proved that Time is real… that the present moment is a unique moment and can be paralleled by nothing in the past… history would not be history if it were not change itself and perpetual novelty” (Hulme 1911c, 40).

29The position in which Hulme found himself in mid-1911, thus, was between two contradictory allegiances:

M. Lasserre then endeavoured to prove to me that Bergsonism was nothing but the last disguise of romanticism. If I thought this was true, I should be compelled to change my views considerably. I can find a compromise for myself, however, which I can roughly indicate by saying that time is real for the individual, but not for the race (Hulme 1911c, 40).

30What should be garnered from this passage is that, more than a full rejection of Bergson, it consists in a disqualification of its political potential. Bergsonian metaphysics are insufficient for “the race”, i.e., for the collective realm of politics and ethics. Indeed, as politics became more important to Hulme, so did the distinction between the dynamism of the romantic spirit and the stasis of the classical one. Increasingly identifying himself with the latter, his enthusiasm for a Bergsonian metaphysics of flux waned.

31Sometimes, this took the tones of a ruthless – if a bit theatrical – self-criticism, such as when he described the “romantic” temperament as a “disordered state of mental health” in which the subject is addicted to the emotion of novelty, the idea “that something wonderful and extraordinary can and is about to happen to man” being a “necessity” for their well-being (Hulme 1994, 237). It is sufficient to look at the words he had used to describe his earlier enthusiasm for Bergson – a thinker which had given him “the exhilaration that comes with the sudden change from a cramped and contracted to a free and expanded state” (Hulme 1911b, 587) – to understand that he was talking about himself.

32Hulme, however, never fully dismissed Bergson, and worked his way towards a pluralist synthesis, as illustrated by this 1915 sketch:

Let us assume that reality is divided into three regions […] (1) The inorganic world, of mathematical and physical science, (2) the organic world, dealt with by biology, psychology, and history, and (3) the world of ethical and religious values […] The outer and inner regions [1 and 3] […] have both an absolute character, and knowledge about them can legitimately be called absolute knowledge. The intermediate region of life is […] essentially relative; it is dealt with by loose sciences like biology, psychology, and history (Hulme 1915b, 138).

33Despite the attempt to create a space in which Bergsonian insights could remain valid, the distance travelled from a mere decade earlier is noticeable. Most significantly the politics of creative agency sketched in early 1911 were now an impossibility: political action had to be based in the world of “ethical and religious values”, a sphere characterized by immutability rather than by movement. Whereas for Sorel, the syndicalist option was ultimately grounded on the openness of the social world, Hulme required for his politics, “an ethic which, from the narrow rationalist standpoint is irrational, being not relative, but absolute” (Hulme 1916, x).


34Let us recap and compare the trajectories of Sorel and Hulme, both in their roads to Bergson and in the way in which they succeeded, or failed, to derive political conclusions from his arguments. In both cases, it was an interest in questions of science, epistemology, and determinism that led to Bergson. For both, it was a pre-existent uneasiness vis à vis determinism which explains their receptiveness to Bergson’s anti-deterministic arguments. In Bergson, both found a more authoritative and more articulated version of hunches, suspicions, and intuitions which they had formulated previously. The two, however, accepted Bergson’s arguments at different levels. Sorel only accepted a psychological argument, whereas Hulme took on board Bergson’s metaphysics, and reflected much more thoroughly than Sorel on the method of intuition, which he attempted to apply to theories of poetic and political uses of language.

35When it came to politicising Bergson, however, their paths, having initially run in parallel, diverged sharply. In the case of Sorel, the open view of human subjectivity developed by Bergson in the Données eventually became the foundation of a non-determinist Marxism which stressed the centrality of class development. In the case of Hulme, the Bergsonian path was abandoned as soon as politics became of some importance. From 1911 until his premature death in 1917 he worked at the construction of an alternative metaphysics, one capable of vindicating the “absolute values” which he felt were needed for ethical life and political commitment.

36A first general point emerging from our case studies is the capacity of Bergsonism to become the philosophical language of reference for a large number of anti-determinist ideas, or at least the one best suited to polemical argumentation. The anti-determinist visions that Sorel and Hulme connected to Bergson could have been found in many other sources of the period. These alternative sources, in fact, were very much read and discussed extensively by both Hulme and Sorel. An obvious example is American pragmatism, especially in its Jamesian variant (Giordani 2014; Rae 1989). Another example comes from Sorel’s elaboration of a non-determinist epistemology of the social sciences, in which Italian historicism and Karl Marx were much more important than Bergson.

  • 4 I borrow the expression philosophème from Azouvi (2007).

37And yet, both our protagonists turned to Bergson, especially when engaged in polemical interaction. In part, this is a reflection of the quality and intrinsic attractiveness of Bergson’s writings. This is especially true of the earliest engagements, occurring in periods in which Bergson’s fame was either not established or had yet to reach Britain. In 1894, we saw Sorel praise the work of a little-known philosopher as a “vigorous tree” towering over desolated steppes, and in 1909 Hulme would write that Bergson “gives a complete system to justify the vague anti-intellectualist sentiment that James has always felt” (Hulme 1909a, 198). But in a later period, this has to do with fame and with the topicality of Bergsonian philosophèmes, with their capacity to occupy a number of different polemical spaces, both philosophical and political, thereby making a text that displayed them interesting to a large number of different sets of readers.4 As I have argued elsewhere (Giordani 2018a), it is in these terms that the partial instrumentality of Sorel’s Bergsonian references in the Réflexions sur la violence can be accounted for.

38A second important point concerns the difference between Sorel and Hulme in politicizing Bergson: as we have seen, whereas the former did it relatively straightforwardly, the latter came to the conclusion that politics required absolutes not available in a Bergsonian cosmology of flux. How to explain this crucial difference? A possible explanation would be in terms of the kind of politics that the two advocated, and would argue that whereas Sorel’s revolutionary politics to some extent required the openness that Bergsonism provided, Hulme’s more conservative ones could not be sustained on these premises. Despite its plausibility, this view must be rejected for two reasons.

  • 5 Moreover, even in our case studies, the relationship between theoretical preferences and political (...)

39To start with, Hulme’s politics are very ambiguous, and it is unclear that, despite his hostility to progressivism and democracy (a hostility shared by Sorel) and his insistence on discipline and authority, he can be unproblematically labelled as a conservative. More substantially, this explanation suggests that certain forms of politics are implicit in certain philosophical positions. But in intellectual history the relationship between the theoretical and the political is virtually always characterized by a degree of mutual autonomy. Just as it mistaken to see certain forms of theorization as mere ideological covers for given political options, it is incorrect to see political choices made by historical actors as fully determined by theoretical preferences. It is very easy to prove both that no specific political content was implied by Bergson’s philosophy and that there is no necessary connection between philosophies of openness and politics of change, whether revolutionary or reformist. One could point to the fact that sections of French Catholicism and even French nationalism greeted Bergson positively (Antliff 1993, 106-34; Azouvi 2007, 141-47), or to the dominance, in the European Marxism of the time, of deterministic visions of history and economics.5

  • 6 For an argument explaining the diversity in Bergson’s receptions in terms of the multiplicity of st (...)

40It is true that Sorel and Hulme responded to different Bergsons, the former only really accepting the psychological arguments of the Données while the latter engaging much more seriously with the metaphysical and cosmological claims of the Introduction à la métaphysique and Évolution créatrice.6 But this does not satisfactorily account for the divergence on the question of politics. Even if Sorel had accepted Bergson’s later cosmological turn, the lesson he would have drawn would have been the same and, conversely, had Hulme remained on the psychological terrain favoured by Sorel, this would have still clashed with the fixed view of humankind that he started advocating after 1911.

41I have argued elsewhere (Giordani and Mead 2022) that the explanation of this difference in the process of politicization has to be looked for, instead, in the different weight that politics implied for the generations of Sorel and Hulme. If a politics ultimately based on a social ontology of creativity was sufficient for Sorel and his belle époque generation, it is because politics did not carry the enormous consequences that it did for those born after 1880. It is true that, in its demand for some form of rationalism, Hulme’s distancing from Bergson could remind one of Julien Benda’s pre-war critique of Bergsonism. But it is essential to see that Bergsonian relativism, for Hulme, became insufficient as a principle of active, political life, setting him decisively aside from Benda, whose ultimate objection to Bergsonian philosophy was that “if it is very true that ‘to know is not to live’, it is symmetrically true that living is not knowing” (Benda 1912, 101).

42In its search for a rationalism capable of sustaining a political life of considerable intensity, Hulme is closer to the anti-Bergsonism of the younger post-war generation, represented most conspicuously by Georges Politzer (1903-1942) and Paul Nizan (1905-1940). Despite important differences, Hulme shares with Nizan and Politzer the same fundamental objection against Bergsonism, namely that, for all its talk of life and concreteness, it is unable to concretely sustain human life in its most intense, political, manifestations (Politzer 1965; Nizan 2001, 245-48). The politics of Hulme’s generation, in short, had grander and more terrifying stakes than anything Sorel, for all his ardour, had ever conceived. Thus, whereas at the outbreak of the Great War Sorel would confess to Croce that “events crush me” and that he saw himself as a “man of the past” with “nothing more to say”, Hulme would react very differently, finding in the war the confirmation that politics require absolutes: “Individuals in a condition of danger, when these pseudo-absolutes melt away into a flux, require once more a real absolute, to enable them to live” (Sorel 1929, 51; Hulme 1915a, 112).

43In conclusion, it is this difference that tells us most about the political potential of Bergsonism in the period. On the one hand, it confirms what other scholars have argued, namely that the arguments and tones of Bergsonian philosophy had, for the new generations that had meaningfully experienced the transition represented by the Great War, become suspect, when not indigestible (Azouvi 2007, 237-323; Bianco 2015, 109-30). This indigestibility resulted from the perceived inability of Bergsonian openness to sustain serious, important, political commitments. In this, the results of my research go against visions of the intellectual history of the pre-war period in terms of a “crisis of reason” leading to the war and to subsequent catastrophes. (Burrow 2000; Sternhell 1996, 36-130). It was precisely when politics became central that Bergsonism and its critique of analytical rationality showed its limits to a generation. In this respect, Hulme and Sorel represent convenient signposts, not only for tracing the shift in philosophical sensibilities, but also for understanding the type of metaphysics that the protagonists of the “age of extremes” felt were required for politics on that scale.

Haut de page


Antliff, Mark. 1993. Inventing Bergson. Cultural Politics and the Parisian Avant-Garde. Princeton, NJ: Princeton University Press.

Antliff, Mark. 2007. Avant-Garde Fascism. The Mobilization of Myth, Art, and Culture in France 1909-1939. Durham NC: Duke University Press.

Azouvi, François. 2007. La Gloire de Bergson. Essai sur le magistère philosophique. Paris: Gallimard.

Benda, Julien. 1912. Le Bergsonisme ou une philosophie de la mobilité. Paris: Mercure de France.

Bergson, Henri. 1910. Time and Free Will. Translated by Frank Lubecki Pogson. London: Swan Sonnenschein.

Bergson, Henri. 1912. Introduction to Metaphysics. Translated by Thomas Ernest Hulme. New York: G.P. Putnam’s Sons.

Bianco, Giuseppe. 2015. Après Bergson. Portrait de Groupe Avec Philosophe. Paris: PUF.

Burrow, John Wyon. 2000. The Crisis of Reason: European Thought, 1848-1914. New Haven, CT: Yale University Press.

Caeymaex, Florence, Antoine Janvier, and Arnaud François. 2012. ‘‘Présentation - Bergson et La Politique.’’ Annales Bergsoniennes 5: 171-74.

Deleuze, Gilles. 1966. Le Bergsonisme. Paris: PUF.

Diagne, Souleymane Bachir. 2011. Bergson Postcolonial. L’élan vital dans la pensée de Léopold Sédar Senghor et de Mohammed Iqbal. Paris: Éditions CNRS.

Ferguson, Robert. 2002. The Short Sharp Life of T. E. Hulme. London: Allen Lane.

Fujita, Hisashi. 2022. “Diremption and Intersection: The Violence of Language in Bergson and Sorel.” Parrhesia 36: 180-200.

Giordani, Tommaso. 2014. ‘‘Introduction au pragmatisme de Georges Sorel.’’ Mil Neuf Cent. Revue d’Histoire Intellectuelle 32 (1): 93-110.

Giordani, Tommaso. 2018a. “On Sorel and Bergson: The Impact of Bergsonian Ideas on Sorel’s Reflections on Violence.” Lo Sguardo. Rivista Di Filosofia 26 (1): 163-81.

Giordani, Tommaso. 2018b. “Redefining Historical Materialism in the Peripheries of Marxism: Georges Sorel and Antonio Labriola between France, Italy, and Germany.” In Decentering European Intellectual Space, edited by Marja Jalava, Stefan Nygård, and Johan Strang, 88-113. Leiden: Brill.

Giordani, Tommaso, and Henry Mead. 2022. “‘The Regeneration of Society’: Thomas Ernest Hulme and the Early British Reception of Georges Sorel.” Modern Intellectual History: 1-30.

Habermas, Jürgen. 2000. “Richard Rorty’s Pragmatic Turn.” In Richard Rorty and His Critics, edited by Robert Brandom, 31-55. Oxford: Blackwell.

Hulme, Thomas Ernest. 1909a. “The New Philosophy.” The New Age 5 (10): 198-99.

Hulme, Thomas Ernest. 1909b. “Searchers after Reality II: Haldane.” The New Age 5 (17): 315-16.

Hulme, Thomas Ernest. 1909c. “Searchers after Reality III: Jules de Gaultier.” The New Age 6 (5): 107-8.

Hulme, Thomas Ernest. 1911a. “Bergsonism in Paris.” The New Age 9 (8): 189-90.

Hulme, Thomas Ernest. 1911b. “Notes on Bergson (1).” The New Age 9 (25): 587-88.

Hulme, Thomas Ernest. 1911c. “Mr. Balfour, Bergson, and Politics.” The New Age 10 (2): 38-40.

Hulme, Thomas Ernest. 1911d. “Notes on Bergson (4).” The New Age 10 (5): 110-12.

Hulme, Thomas Ernest. 1915a. “A Notebook (1).” The New Age 18 (5): 112-13.

Hulme, Thomas Ernest. 1915b. “A Notebook (2).” The New Age 18 (6): 137-38.

Hulme, Thomas Ernest. 1916. “Translator’s Preface.” In Reflections on Violence, by Georges Sorel, v-xi. London: Allen & Unwin.

Hulme, Thomas Ernest. 1994. The Collected Writings of T. E. Hulme. Edited by Karen Csengeri. Oxford: Clarendon Press.

Kisukidi, Yala. 2012. ‘‘Création, universalisme et démocratie: la philosophie politique de Bergson dans Les Deux sources de la morale et de la religion.’’ Annales Bergsoniennes 5: 245-65.

Kolakowski, Leszek. 1985. Bergson. Oxford: OUP.

Lasserre, Pierre. 1907. Le Romantisme français. Paris: Mercure de France.

Lasserre, Pierre. 1910. ‘‘La Philosophie de M. Bergson.’’ L’Action Française, 9 August 1910.

Lefebvre, Alexandre. 2012. “Bergson and Human Rights.” In Bergson, Politics, and Religion, 193-214. Durham NC: Duke University Press.

Martin, Wallace. 1967. The New Age under Orage. Chapters in English Cultural History. Manchester: Manchester University Press.

Mead, Henry. 2015. T. E. Hulme and the Ideological Politics of Early Modernism. London: Bloomsbury.

Nizan, Paul. 2001. Intellectuel Communiste. 1926-1940. Edited by Jean-Jacques Brochier. Paris: La Découverte.

Politzer, Georges. 1965. La Fin d’une parade philosophique, Le Bergsonisme. Utrecht: Jean Jacques Pauvert.

Posteraro, Tano. 2022. Bergson’s Philosophy of Biology. Virtuality, Tendency and Time. Edinburgh: Edinburgh University Press.

Rae, Patricia. 1989. “The Aesthetics of Comfort: Hulme’s Pragmatism and Contemporary Criticism.” ESC: English Studies in Canada 15 (1): 48-65.

Schwartz, Sanford. 1992. “Bergson and the Politics of Vitalism.” In The Crisis in Modernism: Bergson and the Vitalist Controversy, edited by Frederick Burwick and Paul Douglass, 277-305. Cambridge: Cambridge University Press.

Sorel, Georges. 1935 [1894]. L’ancienne et La Nouvelle Métaphysique. Edited by Édouard Berth. Paris: Marcel Rivière.

Sorel, Georges. 1895. ‘‘Les Théories de M. Durkheim.’’ Le Devenir Social 1: 1-26 and 148-80.

Sorel, Georges. 1898. ‘‘L’avenir Socialiste Des Syndicats.’’ L’humanité Nouvelle 2: 294-307; 432-45.

Sorel, Georges. 1907a. Insegnamenti Sociali Dell’economia Contemporanea. Degenerazione Capitalista e Degenerazione Socialista. Palermo: Remo Sandron.

Sorel, Georges. 1907b. ‘‘L’Évolution Créatrice.’’ Le Mouvement Socialiste XXII-XXIII (October).

Sorel, Georges. 1910. Réflexions Sur La Violence. 2nd ed. Paris: Marcel Rivière.

Sorel, Georges. 1919. Matériaux d’une Théorie Du Prolétariat. Paris: Marcel Rivière.

Sorel, Georges. 1929. “Lettere Di Georges Sorel a Benedetto Croce.” La Critica 27: 47-52.

Sorel, Georges. 2019. Study on Vico. Edited and translated by Eric Brandom and Tommaso Giordani. Leiden: Brill.X

Soulez, Philippe. 1989. Bergson Politique. Paris: PUF.

Sternhell, Zeev. 1996. Neither Right nor Left: Fascist Ideology in France. Translated by David Maisel. Princeton, NJ: Princeton University Press.

Zanfi, Caterina. 2012. ‘‘Le Sujet en société chez Bergson : du moi superficiel à la société ouverte.’’ Annales Bergsoniennes 5: 223-43.

Haut de page


1 All translation from French sources are the author’s.

2 The immediate context for this text was the Parliament Act of 1911, a reform that curtailed the power of the unelected House of Lords over the national budget – a reform which both the Tories and Hulme opposed. When, in 1911, Hulme began writing political texts, he did not do it in the New Age, the socialist publication for which he had written for the past two years about philosophy. Instead, he chose both a different venue – the conservative The Commentator – and another name – Thomas Gratton.

3 For example, he was sceptical of Action Française’s anti-Semitic attacks on Bergson, writing that “all this racial gossip about philosophers is a bit tedious” (Hulme 1911a, 190). See also Mead (2015, 118).

4 I borrow the expression philosophème from Azouvi (2007).

5 Moreover, even in our case studies, the relationship between theoretical preferences and political content is less tight that I may have made it appear, especially in the case of Sorel. Though always grounded in an anti-determinist view of history, Sorel’s Marxism was compatible with a number of contrasting political positions. In 1898, the insistence on class development was an anti-insurrectionist position, which insisted on the importance of social reforms for the possibility of an autonomous class development. In the new century, the same theoretical position and the same imperative of proletarian autonomy was used to argue against reformism, since “bourgeois magnanimity is a certificate of inferiority for the working class” (Sorel 1907a, 38).

6 For an argument explaining the diversity in Bergson’s receptions in terms of the multiplicity of stances present in his work, see Schwartz (1992).

Haut de page

Pour citer cet article

Référence électronique

Tommaso Giordani, « Two Political Itineraries of European Bergsonism: Georges Sorel and Thomas Ernest Hulme »Bergsoniana [En ligne], 4 | 2024, mis en ligne le 01 janvier 2024, consulté le 13 avril 2024. URL : ; DOI :

Haut de page


Tommaso Giordani

Tommaso Giordani is an intellectual historian with a background in philosophy. He specializes in European intellectual history, with a focus on the beginning of the 20th century. He has worked and published on pragmatism in France, on transnational exchanges between French and Italian Marxists, and on Bergson, Sorel, and T.E. Hulme. He is currently working on a monograph on Georges Sorel, a thinker to whom he dedicated his doctoral work.

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search