What does literature tell us about the concept of chance? And how is chance relevant to literary theory and comparison? Relating chance to theory is counter-intuitive, because the very attempt to theorise seems to aim at reducing the coincidental and advancing systematic interpretations of literature that emphasise the latter’s coherence. Randomness in a literary work is typically either associated with the writer’s lack of control over their materials and expressions, or recognised only when it translates into serendipitous moments that improve the work’s quality. For the critic’s theorisation of literature, chance is viewed even more suspiciously, if not downright negatively, for its presence in a theory suggests the critic’s inability to offer rigorous explanations. Even after the death of the author, and in our era when Romantic notions of individual genius are extensively challenged, the view that literature and literary theory are shaped by particular design rather than random forces remains prevalent, as shown by the contemporary proliferation of author interviews and creative writing courses, and positive comments about how well a critic grasps their material.

But writings about stones tell us otherwise. By highlighting stones’ aesthetic qualities and quasi-artistic status, these writings raise questions about the role of chance in the natural processes that form stones, and in the human perception that frame them as aesthetic objects. Although stones are typically considered devoid of artistic intention, unless they are cut and polished into jewels, they have sparked an extensive ‘mineral imagination’ in both modern French and classical Chinese literature and artistic practices. In his Pierres réfléchies, Roger Caillois (2018, p. 152) describes stones as “the most ancient real things, from the most ancient order of what is real, that which presides over the angles and facets of crystals, and which is an order that emerged spontaneously from chaos.” Stones are primordial and they present images of geological, cosmogonic time. Caillois not only emphasises the “spontaneous” emergence of order from chaos, which evokes a chance occurrence that is beyond causality and human control, but also discerns chance in the specific shapes of crystals. This suggests that the concrete forms and textures of stones:—varied, intriguingly patterned, singular in each instance yet evocative of repetitive structures across different stones,—reflect the shape of chance.

Precisely because stones are formed by tectonic movements instead of an artist’s hands, they relate to a common understanding of chance as the absence of human intentionality. Caillois’s “écriture des pierres” (“writing of stones”) returns repeatedly to the question of chance in the formation of stones’ visuality, and how it informs but also goes beyond human aesthetic appreciation. Caillois’s lithic writings provide my focal point in this essay. As a lithophile, Caillois was a fervent stone collector and wrote several texts articulating his pareidolic contemplation of curiously patterned stones, drawing connections between stones, art, and imagination. Caillois also explicitly inserts himself into a long and global tradition of geological poetics by referring to art historical practices involving stones in early modern Europe and premodern China. In particular, he cites Chinese scholars’ rocks, meaningfully termed “pierres de rêve” (“dreamstones”) in French, and empathises with the Song literatus Mi Fu’s legendary mania for them (2008, p. 1076). This reference leads me to consider the aesthetic value of rocks in premodern Chinese poetics and Chinese discourses about literary creativity that draw connections between natural imagery, chance, and spontaneity. These connections are central to the third-century Chinese poet Lu Ji’s 陸機 (c. 261-303) Rhapsody on Literature (文賦 Wenfu), the first Chinese text that outlines a theory of creativity and writing. In Rhapsody, we find stones, pearls, and foliage—natural imagery evoking objects formed without human intervention and by unpredictable events,—presented as analogies to poetic works that explain the emergence of literary creativity.

Although Caillois and Lu are separated by radical differences in language, intellectual context, and time, they are connected via a geological imagination expressed in poetic language and theoretical efforts to understand creativity in a critical framework that blurs human and non-human spheres. Notably, their writings raise specific questions that shed light on the intersections between chance, writing, and literary theory: what can stones tell us about our understanding of chance and creativity? Can creativity and criticism flourish by promoting instead of calculating or eliminating chance? If yes, how? By reading Caillois alongside the poetics of Chinese scholars’ rocks and Lu’s Rhapsody, I discuss how lapidary writings and practices posit stones as poetic and discursive objects that engage with chance reveries, aesthetic sensations, and effects of human-mineral symbiosis.

Caillois’s dreamstones: unpredictability and order

By 1966 when Caillois’s most famous text on stones Pierres was published, major theories of chance were widely known in artistic and literary circles and hotly debated. Firstly, there was the enduring view of chance as the absence of human intentionality, which traces back to Lucretius’s cosmology where chance is the absence of design and causality, since the universe is influenced by clinamen, or unpredictable swerves of atoms, and is therefore fundamentally uncertain.Footnote 1 The Surrealists, with whom Caillois was initially associated, very much aligned with this view. For them, the profound uncertainty of the world was cause for celebration, for it translated into “amour fou” (“mad love”), automatic writing, and poetic encounters of visual shock as exemplified in “cadavre exquis” (“exquisite corpse”) drawings that transgressed conceptual categories.Footnote 2 Indeed, Surrealism was very much a pursuit of chance, as Richards (2020, p. 189) observes, “the incursion of chance into the work of art—as a narrative device, a procedure, a logic of association—was imagined as a countermeasure to means-end rationality.” In early Surrealism as articulated by André Breton and illustrated by Nadja’s protagonist’s random vagaries and encounters in Paris, chance was defined against rationality, mechanistic explanations of cause and effect, and against rules that limit imagination and creativity to a pre-determined world. Chance was therefore almost synonymous with spontaneity, dream, magic, irrationality,—all key words of Surrealism. On the occasion concerning Mexican jumping beans which triggered Caillois’s rupture from Surrealism in 1934, however, Caillois’s understanding of chance was revealed to diverge widely from Breton’s. The beans would inexplicably jump when held on the palm. As Caillois recounts in “Intervention surréaliste” (2008, p. 214): “Breton voulait qu’on rêvât, qu’on s’extasiât sur le prodige. Je préconisais qu’on ouvrit plutôt une des graines pour vérifier si un insecte ou une larve n’y serait pas contenu, ce qui était le cas” (“Breton wanted us to dream and be ecstatic about the wondrous phenomenon. I suggested that we open one of them to see if it contained an insect or a larva, which was the case”).Footnote 3 Breton condemned Caillois as a tedious empiricist who wanted to kill magic by construing rationalising explanations, whereas Caillois argued (2008, p. 224) in his riposte that his approach did not undermine the marvellous in the least because he envisages “un merveilleux qui ne craint pas la connaissance, qui au contraire s’en nourrit” (“a marvellousness that does not fear knowledge but is instead nourished by it”). For Caillois, the random twitches of jumping beans could nevertheless be produced by a principle of movement, as he expounds (2008, p. 223): “L’irrationnel: soit; mais j’y veux d’abord la cohérence, la surdétermination continue, la construction du corail” (“the irrational, so be it; but I firstly want it to have coherence, continuous overdetermination, the structure of coral”). While acknowledging the existence of the unpredictable and incalculable, Caillois pursues a basic cosmic pattern and is, in this sense, fundamentally a rationalist.

Caillois’s demand for rigour even when interpreting apparently illogical and mysterious phenomena makes his position towards chance similar to the Oulipo’s. The Oulipo group, established in 1960 and very much positing a wholesale refusal of the Surrealists’ views and methods, sought to eliminate chance in their literary creations. In Claude Berge’s self-definition of Oulipo (cited in Bens, 1980, p. 136): “nous sommes essentiellement anti-hasard” (“we are essentially anti-chance”). Oulipians reflect the scientific discourses about chance since the nineteenth century, which engaged with the absence of causality and the question of human ignorance. The view that the universe is determinist,—one which questioned chance’s existence,—emerged as a strong competitor to the Lucretian view outlined above. That chance is “the individual’s lack of predictive power over events” rather than “the absence of cause” (James, 2009, p. 36) in Georges Perec’s La vie echoes Henri Poincaré’s view. In Raymond Queneau’s (1938/1973) criticism of Surrealism, he raises the same question that Caillois evokes: how are we sure that what we believe is spontaneity or chance in dreams and automatic writing is not in fact our inability to discern the underlying constraints and real causes of these phenomena? Perhaps we are “slaves” to the unconscious (Queneau, 1938/1973, p. 94), shaped by factors outside our control, and therefore what Surrealists call “chance” is simply human ignorance? Queneau’s scepticism echoes Caillois’s view that an apparently inexplicable and random phenomenon does not necessarily mean chance, and that chance should be investigated with all available methods. And so the intellectual context for Caillois’s lithic writings encompassed two opposing views of chance: that chance seemed to promise a universe that was fundamentally non-deterministic (an idea that writers and artists very much favour); or, that chance might be an illusion created by the failure of human knowledge.

Ultimately Caillois rejects both positions and expounds a theory where rules and chance co-exist. This is articulated in the complementary relation between agon (competitive play) and aléa (games of chance) in Caillois’s Les jeux et les hommes, and reaffirmed in his ekphrastic and theoretical writings about stones, where he paradoxically discerns in stones both unpredictability and order. As Tritsmans argues (1992, p. 113), the very choice of stones as the subject of writing is “une tentative de soustraire l'écriture à l'aléatoire de la communication intersubjective” (“an attempt to exempt writing from the randomness of intersubjective communication”). Nevertheless, in Caillois’s texts, stones emerge as fluid and organic rather than petrified. As the title L’Ecriture des pierres implies, this is writing about stones, writing constituted of stones (mineralised writing), and writing by stones (stones write themselves). This blurring of the organic and inorganic spheres goes together with Caillois’s suggested co-existence of chance and order. To begin with, while Caillois emphasises (2008, p. 501) that visually interesting stones should be appreciated as artworks of nature and not seen as mere accidents that conform to human fancy, he also recognises the uniqueness and surprising discoveries offered by each stone. Among Caillois’s numerous stones with intriguing shapes and picturesque patterns, one striking example is a limestone which he calls “Le Château” (“The Castle”) (1970, pp. 106–107).

In this stone, Caillois (1970, p. 108) discerns the image of a fortress, thick foliage in the background, and curious silhouettes “semblables aux ‘bonhommes’” (“like small human figures”), slotted into notches resembling crenels of a fortress wall. “Comme s’ils étaient aveugles” (“As if blind”), these humanoid figures lack pictorial depth, like “ombres chinoises” (“Chinese shadow puppets”) (ibid.). With arms suspended in mid-air and oriented uniformly towards the right, does this gesture indicate “protection ou […] vénération” (“protection or worship”) (ibid.)? On the far right, an intriguing figure with a white face turns to the left, facing the other figures as if reciprocating their gesture. Everyone who saw this stone, Caillois remarks (ibid.), thought that it was painted “par quelque artiste naïf […], enfant ou néophyte” (by an artist of naïve art […], a child or novice”); and that “il leur semble inconcevable que le hasard seul ait pu produire un dessin qui paraisse à ce point l’œuvre de l’homme” (“it seemed unthinkable to them that chance alone could produce a picture that so strongly resembled the work of human beings”) (ibid., p. 109). Qualifying the stone’s image as a “tableau naturel” (“natural painting”), Caillois affirms (ibid., p. 110) that it is not a creation of human artifice. But he does not think it is pure coincidence that the stone has repetitive anthropomorphic figures. Rather, they reflect a cosmic pattern that “obliges” different forms to “resemble each other” (“il est même presque obligatoire, qu’une forme ressemble à une autre”), so that the likeness between the mineral and the human is due to “nul miracle ni mystère” (“neither a miracle nor a mystery”) but “le jeu des analogies” (“the play of analogies”) between forms (ibid.). Although the stone’s image seems like a representation, it is “un […] concours de signes sans signification” (“an assembly of meaningless signs”) (ibid.) that is interpreted as figuration by human perception.

In contrast, Caillois’s jasper from Oregon (Figure 1) offers an apparently non-figurative image riddled with fluid lines. But Caillois discerns (ibid., p. 84) a range of intensely organic and bodily motifs in this “univers de volutes” (“universe of scrolls”), “d’où émergent des visages écorchés, l’éventail de muscles à vif dans les cavités de l’os” (“from which flayed faces and a variety of muscles laid raw in the bone’s cavities emerge”). The stone’s shades of colour form “la gamme entière des teintes d’hématome (“the entire range of hues of hematoma”), which suggest “une éruption […] de bubons sur un épiderme infecté” (“an eruption of […] boils on infected skin”), “mollusques sans coquille” (“shellfish without their shells”), and disembowelled organs (ibid., pp. 87-88). These descriptions are flights of ekphrastic imagination that could compare with Chinese literati’s poetic appreciation of scholars’ rocks, and expressed in emotionally charged language that vividly translates the jasper,—of “imperturbable” inorganic material and “qui ne sent ni ne sait” (“which neither feels nor knows”),—into “la chair meurtrie” (“murdered flesh”) (ibid., p. 88). Not only does mineral material bleed “une vie démente” (“deranged life”), it also conveys nervous convulsions and painful sensations to its viewer, for the contemplation of a wound is affective, as shown by the disturbingly visceral slitting of a calf’s eye in the opening scene of Un chien andalou.

Fig. 1
figure 1

© Photo by Paul A. Harris.)

Jasper, Caillois (1970, pp. 86-87) (Image from: http://substancejournal.sites.lmu.edu/home/media/jasper-oregon [accessed 8 Jan 2022].

Full size image

Are the jasper’s tortuous creases the results of chance—random flows of matter during the consolidation process that produced unique patterns that represent nothing but are infinitely suggestive to the imaginative viewer? While Caillois (ibid., pp. 87-88) celebrates the stone’s fluidity and unpredictability, as it is like “les nuées” (“clouds”) and “rempli de surprises” (“full of surprises”), he also detects in it “la formule du règne monstrueux” (“the formula of monstrosity”), suggesting it is an instantiation of a larger natural order.

A third example, an “eyed” agate (Caillois, 1970, p. 71) with two quasi perfectly round ring patterns, presents an example that clarifies Caillois’s articulation of chance in his poetic ruminations on stones. Caillois (1970, p. 71) starts by acknowledging that agates’ images are extremely varied and “ambiguës” (“ambiguous”), he stresses that “il faut, pour les arrêter, que l’imagination y mette du sien et qu’elle se tienne au simulacre qu’elle a choisi de déchiffrer” (“to fix these [fluid images], imagination should be put to work and adhere to the simulacrum it has chosen to decipher [from the agate]”). The contrast here is between the indeterminateness of the stone’s image and the delineating force of the viewer’s imagination. This is illustrated by the interpretation of the “eyed” agate. Unlike the previous two stones, this agate displays neither a pictorial scene nor intensely abstract and expressive motifs but a basic geometric shape: the circle. Its twin circles contain concentric rings and are almost equally sized, which make them look like a pair of eyes. How the circles are understood depends upon the viewer. For Caillois, circles are central to his discussion of the unsettling ocelli on butterfly and moth wings in Méduse et cie. (The Mask of the Medusa). Caillois argues that the ocelli are signs worn by butterflies to hypnotize and terrorise their predators. For “tout cercle fixe est naturellement hypnotisant. Le contempler longuement trouble, paralyse, endort” (“every immobile circle is automatically hypnotising. Gazing at it continuously troubles, paralyses, and sedates”) (Caillois 2008, p. 538). This is reflected in the human fear of “le mauvais œil” (“the evil eye”) (ibid., p. 490), for instance the disquieting effect of staring at the enormous eyes of owls, which are often considered sinister for this reason. To master this fear, human beings devise myths such as Perseus decapitating the Medusa to “domesticate” creatures with ocelli, or draw circles on masks to empower the mask wearer in rituals (ibid., p. 542). In brief, ocelli are instruments of intimidation that function like the Gorgon’s head: a “mask” that terrifies (ibid., p. 540). Caillois thus argues that ocelli are a universal motif repeated in different contexts and by different creatures in the world. A necessary connection arises between butterfly ocelli, the owl’s head, the Medusa myth, and the “pierres-images” (“stone-images”) as in the agate above (ibid., p. 502). This is what Caillois terms “mimetism,” a cosmic principle substantiated by mutual resonance and visual imitation across the animal, mineral, and human realms. Seen in this light, the agate’s two ocelli are not coincidental but an example of mineral mimetism. Nevertheless, this interpretation is not a necessary result of the agate’s image, for it depends on the particular viewer.

What theoretical reflections on chance can we extrapolate from Caillois’s descriptions of stones? The first question is whether these intriguing patterns on the stones are due to chance. If chance denotes meaningless coincidence and the absence of any principle in the workings of nature, then Caillois’s answer is emphatically no. As Marguerite Yourcenar says (in Caillois 2008, p. 27) in her prologue to Caillois’s Œuvres: “Les pierres, comme nous, sont situées a l’entrecroisement […] d’un nœud de forces trop imprévisibles pour être mesurables, et que we désignons gauchement du nom de chance, de hasard, ou de fatalité” (“Stones are situated at the intersection […] of a node of forces that are too unpredictable to be measurable, which we awkwardly designate as fortune, chance, or fate”). Caillois refuses (2008, p. 487) to use “chance” in such a convenient and vague way: “Invoquer une coïncidence n’est jamais qu’un pis-aller, sinon un aveu d’impuissance” (“Claiming it is a coincidence is always the last resort, if not a sign of weakness”). Repeatedly, he stresses that stones’ images are not random but the material manifestation of a natural principle such as mimetism or a “formula of monstrosity” (Caillois, 1970, p. 88) inherent in the world. In his conclusion to Pierres, he (2008, p. 1084) mentions “un principe régulateur” (“a regulating principle”) that is primordial, resulting in that “quelque chose sorti du chaos enfin se répète et se reflète. La nature cesse d’être la fruste qu’elle était d’abord” (“something which finally emerged from chaos repeats and reflects itself. Nature ceases to be the messiness that it was at first”). Stones bear the imprint of this necessary transition from formlessness to form. There is nothing accidental about their shapes and patterns.

Caillois does not, however, eliminate chance altogether from his cosmological vision of stones. This is shown, firstly, in his recognition of wide individual variations among stones; and secondly, by his remarks in Pierres réfléchies (2018, p. 153) on accidents that “wound” stones. Here, Caillois offers possibilities of theorising chance other than meaningless randomness. Specifically, he compares (2008, p. 1083) a stone’s “dessin” (“drawing”) to “le profil changeant des flammes” (“the changing contour of flames”) and considers it “l’imprévisible résultat d’un jeu de pressions inexpiables et de températures” (“the unpredictable result of a play of implacable pressures and temperatures”). That a fundamental cosmic principle exists does not mean all forms of materiality are homogeneous. Room must be left for particular lines, nuances of colour, and textures on individual stones, for no two stones are the same. Caillois visualises (2008, pp. 1084-85) the world as a “dragonnier bifide” (“bifurcated dragon tree”) ramifying into numerous branches and paths that, according to his theory of mimetism, echo each other in a “univers de miroirs” (“universe of mirrors”). The singular patterns of each stone are results of these infinite possible paths. Stones’ patterns parallel “radiolarian” seashells; “le peintre qui hésite à placer la touche juste” (“the painter who hesitates to make the appropriate brushstroke”); and “le poète […] qui trouv[e] le mot exact” (“the poet […] who finds the right word”) (ibid.). Capricious exceptions exist within a system of overall coherence, which partly explains why the cosmic principle manifests in this particular form on one specific stone. Here, chance may be understood as the possibilities of nature’s “embranchement” (“ramification”) that has no fixed number, which construct a(n) “labyrinthe infini” (“infinite labyrinth”) that cannot be fully predicted (2008, p. 1084). Nature’s unpredictability is affirmed by Caillois (1970, p. 21) when he uses the term lusus naturae (“sport of nature”) to describe eccentrically-shaped stones, for lusus means “amusement” and “play,” which suggests that chance is a playful force in “l’immense engrenage” (“the huge apparatus”) (Caillois 2008, p. 558) of nature that spontaneously generates infinite possible deviations.

Simultaneously, Caillois differentiates (2018, p. 153) between “tectonic clashes” that form and “scar” stones through geological processes and accidental marks on stones due to “mistreatments” such as being chipped or abrased by “human beings or bad luck.” Caillois acknowledges the accidental but sees it as “short-lived” and insignificant, for “the scars that stir us in stones are those told by conflicts between forces equal in dignity and power: volcanic ire and patience, the subterfuge of pseudomorphoses” (ibid.). Namely, imperfections that result from larger-than-human and necessary forces, elusively characterised by Caillois (ibid., pp. 152-153) as “ascetic fatalities” or “ancient brutality,” are not accidents but add value to a stone by making it “singular” and “incomparable.” Caillois then draws cosmic connections between mineral and human realms (ibid.): “All things that exist, from impervious stone to melancholic imagination, have been or will be accountable to [these necessary forces] at least once”; and that “there are perhaps no unscathed artists and poets.” This suggests that poets are organically related to sediments of the earth and, like stones, are scathed by cosmic movements. Just as wounds make stones more interesting, tribulations enhance the richness and creativity of the poet’s work.

We might pause now and ask whether Caillois’s views on chance really square with each other. How, for instance, do accidental marks on a stone not also increase its singularity? And what distinguishes “bad luck” from violent cosmic “fatalities” that deform and shatter matter, resulting in wounded stones? Why should not accidents and exceptional occurrences be valued equally as necessary and mutually mimetic phenomena, since both are the numerous possible manifestations developed from the underlying cosmic principle? Caillois does not provide clear answers, and he indirectly acknowledges his epistemological shortcoming when describing himself (2008, p. 1078) as striving “à en deviner les secrets” (“to guess [stones’] secrets”). Despite their symbiotic relation to humanity and other forms of life, stones remain enigmas and always resist discourse. Caillois does not have the last word on stones nor on the shape of chance to which they bear witness. Indeed, the question raised by Caillois’s lithic writings is less about what concept of chance stones demonstrate than whether the reveries conjured up by the stone’s viewer are purely arbitrary. Are the myriad images in the monstrous jasper random wanderings of the mind? Is the “eyed” agate’s mimetism not Caillois’s projection of his theory on the stone? What necessary connection exists between Caillois’s reveries and the stones themselves?

This shift from stones per se to the writer’s response to stones and creative process is evoked by Caillois himself. As Memdouh argues (2005, p. 82), Caillois’s mineral universe “aboutit à une écriture du ‘je’” (“culminates in a writing of the Self”), which is suggested by Caillois’s (1975, pp. 142-143) heightened self-awareness during his scrying: “En ce miroirs, c'est mon reflet […] que je tente d'apercevoir” (“In these mirrors [of stones], it is my reflection […] which I try to perceive.”) The contemplation of stones establishes “une sorte de courant” (“a certain current”) between “la fixité de la pierre et l’effervescence mentale” (“the immobility of stone and mental effervescence”) (2008, p. 1078). This suggests that “the pareidoliac’s interpretation of a stone’s pattern depends upon her own personal internalized database of stored images, […] embellished by personal memory, emotion and psychical topography” (Prudence, 2018, p. 71). “Je me sens devenir un peu de la nature des pierres” (“I feel myself becoming something of the nature of stones”), Caillois professes (2018, p. 1078), revealing the fusion between the viewer’s sensibility and the stone itself. Specific variables dependent upon the individual viewer and the circumstances of viewing coincide to produce a mind that ferments ideas that are varying, subjective, and even whimsical. Chance is a meaningful co-incidence – a meeting of different factors – that operates as a creative force generating highly individualised associations.

Chinese rock aesthetics: pivoting creativity in Lu Ji’s Rhapsody

Caillois’s lithic reveries lead me to consider how the writer’s creative process is stimulated by stones, often in surprising ways where chance plays a major role. The relation between stones and creativity can be further explored by examining Chinese scholars’ rocks, which are often cited by Caillois and provide a bridge between the latter’s surreal reveries and Chinese poetics. Referred to as 供石 gongshi (“tribute stones”), 奇石 qishi or 怪石 guaishi (“marvellous stones”), and 靈石 lingshi (“spirit stones”), scholars’ rocks have a special place in premodern Chinese aesthetics because rocks were seen as miniature mountains and embodied cosmic energy (qi 氣). In Chinese cosmogonic myth, the sky is a vault that is patched up by the mother goddess Nüwa using five-coloured rocks. The canonical classical novel Dream of the Red Chamber (紅樓夢), also entitled Story of the Stone (石頭記), has as its protagonist the human incarnation of a spirit stone left over by Nüwa. These depictions show that rocks are considered animate beings that connect the microcosm and macrocosm, an idea that similarly exists in Caillois’s mineral-human resonance. Unsurprisingly, rocks frequently feature in premodern Chinese texts and anecdotes. Marvellous rocks (guaishi) were mentioned as tributary gifts to the mythical emperor Yu, and they decorated imperial gardens as early as the Han (Schafer, 1961, p. 5). Confucius’s aphorism in the Analects: “仁者樂山” (“the benevolent person loves mountains,” 雍也 chapter) merges personal virtue with natural entities and explains the literatus’s love of rocks. Famous litholaters (“stone worshippers”) include Huizong (1082-1135), the Northern Song Emperor who was more aesthete than ruler; and Mi Fu, cited by Caillois as his spiritual predecessor, who allegedly saluted a strange rock as his “Elder Brother,” a scene frequently depicted by Chinese painters who see an “isomorphism between human and stone” (Parkes, 2005, p. 77). By the late Ming (c. 1570s-1644), the literati’s appreciation of rocks became obsessive and was described as 癖 pi, a “mania” for stone-collecting, as evidenced by texts on rocks such as Stone Manual of the Su Garden (素園石譜, 1613) and Ming collectors such as Mi Wanzhong whose lithophilic passion made him feel being “on the verge of becoming a stone” (cited in Little, 1999, p. 24). This feeling is precisely what Caillois also professes to experience, which suggests that affective relations between stones and humans are transhistorical and transcultural.

Like Caillois, Chinese literati also valued the unique visuality of rocks. “Rocks that appear ‘wrinkled, perforated, lean and translucent’ (zhou, lou, shou, tou) are prized the highest” (Tan, 2016, p. 231). These evocative shapes are often described by analogies to birds, plants, clouds, and human figures, indicating a cosmological inter-connectedness which resembles the mutually-mirroring nature reflected by Caillois’s monstrous jasper. Additionally, as Parkes emphasises, rocks were valued for their energy (qi), as Du Wan’s Cloud Forest Catalogue of Rocks (雲林石譜) states: “The purest energy of the heaven-earth world coalesces into rock” (cited in Parkes, 2005, p. 78). Indeed, there is no transgression of the organic and inorganic binary in Chinese rock aesthetics because in classical Chinese cosmology this distinction does not exist. Humans and stones are of the same origin, which is qi, the pervasive energy flows that form everything in the world.Footnote 4 Stones’ shapes are therefore extremely meaningful, for “rugged surfaces, ridges, and crevices grant access to the mechanisms of the cosmos and can, conversely, affect the course of human affairs” (Matteini, 2017, 82). Scholars’ rocks that decorate the literati’s studios are not only objects that trigger reveries and inspire the literatus’s practices such as writing and calligraphy, they also emanate spiritual power that directly influences the literatus’s creative process and self-cultivation. This aligns with the notion of 感應 gan’ying, “sympathetic resonance” between myriad things and existences in Chinese correlative thought and implies that rocks possess a degree of creative agency in the literatus’s artistic activities.

The implication of scholars’ rocks in the Chinese poet’s creativity already envisages creativity in a depersonalised way that expands beyond the individual human agent and encompasses chance elements if chance is understood as the absence of human intentionality. But the scholars’ rock is absolutely not a random element that is dispensable, for its very existence in the literatus’s studio is artificial and curated. Each rock is carefully selected, mounted on tailored stands, placed in relation to the room space and other artefacts. While rocks were above all appreciated for looking unhewn and unpolished (Zeitlin, 1993, 79), the appearance of non-artifice was so sought-after that sometimes rocks were deliberately cut to look more raw. The rock’s physical form is thus not always free of human intervention, plus its display depends on the collector’s choice of stone and perception of what the stone means for them. Unlike Caillois’s stones that are seen as bearers of marks of natural forces that are incalculable or capricious, scholars’ rocks are not natural if “natural” means untouched by human artifice. Nevertheless, room for unpredictability and spontaneity emerges in exactly how rocks interact with their surroundings and affect the poet’s aesthetic sensibility and practices. While the rock acts as a stimulus to creativity and the poet becomes a medium through which creative energy is channelled and takes form in writing, this process is by no means certain as the rock does not necessarily lead to inspiration that produces beautiful writing.

That aesthetic inspiration is stimulated by external objects instead of rising from the poet’s individual genius is an enduring view that is prominently featured in one of the founding texts of Chinese aesthetics: Lu Ji’s Rhapsody on Literature. Rhapsody is the first Chinese text that aims at theoretically and systematically examining literary writing and the process of literary imagination and creation. Lu’s (Rhapsody, section 9) famous dictum that “賦體物而瀏亮” (“the rhapsody [fu] embodies things and gives them clear and bright form”)Footnote 5 indicates the importance of embodying and visualising things in Lu’s own Rhapsody. A stone, for instance, could be one of the “nature images” that are “accompanied by an influx of emotion,” entailing an “emotion-visuality symbiosis” (Cai & Wu, 2019, pp. 4-5).

That writing emerges from meaningful stimuli of external objects and unplanned-for effects is supported by the fact that Rhapsody surprisingly does not offer any prescriptions about what the poet should do to produce good writing. Rather, the poet’s creative process is spontaneous, uncertain, and mysterious, described by analogies involving jade, pearls, and foliage, i.e. phenomena which fall outside the poet’s self-conscious efforts:

石韞玉而山輝, 水懷珠而川媚。彼榛楛之勿翦, 亦蒙榮於集翠。

When rocks embed jade, the mountain glows; when the stream is impregnated with pearls, the river becomes alluring; when the hazel and arrow-thorn bush are spared from the sickle, it will glory in its foliage.

In Chinese, beautifully crafted expressions are described as “pearls and jade” (珠玉). Excellent writing glows like “a rock” that contains jade and is polished and rounded like a string of pearls, which makes the writing “alluring.” But the existence of jade in a rock and pearls in the river is entirely beyond the design and intentionality of the writer, and Lu gives no indication of how a rock becomes embedded with jade. The natural imagery extends to editing one’s own writing, which is likened to pruning a shrub: sometimes when you leave the plain bits or odd parts (“hazel and arrow-thorn”) uncut, they later form a beautiful bunch of “foliage.” An unexpectedly nice section of writing emerges when you do not calculate every word you write nor try to control your writing in a determined form or length. This intriguing view is contrary to the consensus now that good writing results from much deliberation and the writer’s efforts to maintain stylistic uniformity and concision.

Unexpected and uncontrollable moments in the writing process are made explicit by Lu in a crucial passage that describes how the writer tries to grasp fleeting inspirations and mental images, sometimes in vain:

若夫應感之會, 通塞之紀, 來不可遏, 去不可止, 藏若景滅, […] 方天機之駿利, 夫何紛而不理?[…] 紛葳蕤以馺遝, 唯豪素之所擬。 […] 雖茲物之在我, 非餘力之所戮。故時撫空懷而自惋, 吾未識夫開塞之所由。

As for the stimulus and response being in accordance with each other, the principles of smooth flowing or obstructed thoughts, when they come they cannot be restrained, when they leave they cannot be stopped; when they hide they are like vanished shadows, […]. When the heavenly pivot is swift and sharp, how can any confusion not be brought to order? […] The scattered and profuse clusters of [ideas, images, and words] that swiftly gallop around and crowd each other can only be imitated by the brush and paper. […] Though this thing is within me, it is not in my power to grasp it. So I strike my empty chest and sigh again and again, for I do not yet know the reason why [inspiration] arises and stops. [my emphasis]

Lu's observation of the poet's experience of inspiration or his lack thereof affirms unpredictability in the process of literary creation. In Swartz’s view (2021, pp. 98-99), the “ebb and flow of the creative force” described in this passage “hold[s] the writer in suspense,” and the “writer’s inability to control, even predict, the flow of creativity can lead to bitter frustration.” This inability of the poet to assert their agency is due to the fundamentally mysterious origins, trajectories, and duration of the creative process, which is cryptically described as dependent on the “heavenly pivot” (天機 tianji). Tianji encompasses connotations such as “lucky timing,” “a clever device created by superhuman force,” “an opportunity that is gifted by heaven,” “a divine or supreme secret.” Namely, tianji precisely falls outside human calculation, and its causality (if there is one) is beyond human knowledge. Unpredictable thoughts do not necessarily generate productive inspiration as they might simply continue to crowd against each other in a disoriented state. The poet does not know why inspiration comes and goes, nor can control it.

So what can the poet do? Lu does not negate the poet’s skill and knowledge, which are indeed crucial. Firstly, the poet should be the apperceptive recipient and medium of stimuli, to “become receptive to opportunities” (Swartz, 2021, p. 100). The poet reacts to external stimuli such as rocks and other environmental phenomena, stirring up ideas and words that appear like a brainstorm of changing clouds and “shadows.” “Sympathetic resonance” (gan’ying) is crucial here, for it entails “resonance with objects” (感物 ganwu), which correlates “between humans and nature” and explains the “poet’s emotions and aesthetic sense” as stirrings from “objects and circumstances” (Jia, 2016, pp. 462–463). While this notion of resonance runs deep in premodern Chinese aesthetics, it also echoes various expressions of mutual resonance in modern French literature and thought. Notable examples are found in Surrealist images of the “vases communicants” (“communicating vessels”) and “champs magnétiques” (“magnetic fields”), both empirical facts observed through scientific experiments as well as poetic metaphors for the mutual stimuli and reflections between psychic reality and material reality, the human and non-human, as illustrated by Breton’s scrying in a girl’s eyes “a drop of storm-cloud-sky-colored water falling on a body of calmer water” (Caws, 1988, p. 98) in Les Vases communicants (1932). These affective resonances envisaged by Breton, Philippe Soupault, etc. draw analogies between different phenomena and ontological categories and ultimately fuse them. This conjures a cosmic relation between all things that parallels Caillois’s natural mimetism as well as sympathetic resonances between the poet and their physical entourage in Lu Ji’s poetics.

Secondly, the poet tries to “imitate” the images and movements in their mind. Imitation (ni 擬) is a major function of the rhapsody form, which follows excellent models, expressed by Lu’s phrase “操斧伐柯” (“hewing an axe handle with an axe in hand”), and aims at depicting its object “brightly” (mentioned above). The poet harnesses their knowledge of past writers’ remarkable works and imitates the imagery and ideas stimulated by objects. Indeed, the character for “writing” (wen 文) also means “pattern” and describes material surfaces and visual patterns of stones, brocades, clouds, and phoenix feathers. As Hay observes (1987, p. 9), wen in “the language of literary criticism” was “closely related to” the “artistic form” of scholars’ rocks. An imitative poetics as Lu proposes would produce writing that is beautifully “patterned” like a marvellous rock. Creativity is not primarily attributed to the poet’s individual genius, and authorial intention is not about expressing personal feelings but “to engage with […] cultural life” through the writing of literature as a “system of interacting images, devices, and allusions” (Williams, 2015, pp. 28–29).

Is the poet’s agency, according to Lu, therefore limited to being an imitator of illustrious predecessors and a receptor of chance occurrences in the creative process, sparked by felicitous factors such as wondrous rocks? Agency remains significant, for the poet must be able to recognise and seize the opportune moment (ji 機) in the fluid and whimsical process of literary creation, which requires a keen aesthetic sensibility that differentiates between numerous unpredictable instances where images and words bustle around to articulate “smoothly flowing expressions and clear forms” (“辭達而理舉,” Rhapsody) from the morass. The term ji plays an important role in outlining a concept of chance and its interactions with the poet’s agency. Ji is a polysemantic word that typically denotes “opportunity” in a positive way (時機 shiji “timeliness”; 機遇 jiyu, “favourable circumstance/encounter”), but it also encompasses a host of meanings such as luck; “probability” (機率 jilü); the “cause” of something (機由 jiyou); “secret” (機密 jimi); cunning (機巧 jiqiao); a clever machine such as a trap; a “turning point” (轉機 zhuanji) such as a pivot which can tilt in different directions (as in tianji); coincidence, or the “intersection of various circumstances and fortunes” (機緣湊巧 jiyuancouqiao). As Lisa Raphals remarks (2003, pp. 551-552), ji semantically relates to other Chinese terms such as 幸偶 xing’ou and 巧 qiao that denote “good and bad luck that result from accidents,” the “opportune” or “coincidental.” While ji is not a translational equivalent of “chance” or “le hasard,” its polysemy helps us re-conceptualise chance as a combination of factors outside one’s individual control and one’s ability to exploit these factors for their own needs. This is to “act in response to changing circumstances” (隨機應變 suiji yingbian), which is a valued skill demonstrating sharp perception and adaptability. A person who can perceive ji and employ it to their own advantage is typically a diviner, i.e. somebody with secret and superior knowledge, who has access to techniques limited to an exclusive group and can assess and calculate situations appropriately.

The mysterious “heavenly pivot” (tianji) in Rhapsody thus refers to a secret opportunity offered by “heaven” (天 tian), a term literally meaning the “sky” and used widely in premodern Chinese literature to denote non-human power of a cosmic and sacred nature. While tian stands in contrast to 人 ren, “the human” and “artificial,” in that the human cannot forcibly change the course of heaven, the heavenly and the human can complement each other and work together in non-conflictual and propitious ways. The “heavenly pivot” implies that circumstances produced by heaven can be used like a pivot by the writer to tilt in their favour, thus enhancing their creativity and shaping their writing beautifully. While a pivot can turn in many possible directions, Rhapsody suggests that if the writer acts in a timely way, they can make the pivot turn from negative unpredictability to serendipity:

雖逝止之無常, […] 苟達變而相次, 猶開流以納泉。如失機而後會, 恆操末以續顛。’

Although inspiration and moods come and go without constancy, […] if you perceive and adapt to the changes and can differentiate the sequence, it will be like opening a channel to receive spring water. If you miss the opportunity [ji] and understand it belatedly, [it would be as disjointed as if] you are putting the tail on the head.

The writer is like the diviner in that they also have the cultivated sensibility and deep knowledge to perceive ji and employ it advantageously to develop its artistic potential. The writer’s vigilance and agency remain crucial for determining the precise form of writing.

Instead of chance as pure randomness, Rhapsody suggests that chance means opportunity, “a set of circumstances permitting or favourable to a particular action or purpose” (OED). One concrete manifestation of this opportuneness is the contemplation of scholars’ rocks, which triggers wondrous images and meaningful associations. Not all contemplators of rocks see the rich range of possible imagery and ideas that rocks offer, even fewer could craft these wandering reveries into beautiful language and well-structured writings. The poet’s repertoire of literary knowledge and cultivated aesthetic intuition are indispensable, despite the poet’s incomplete control of the creative process. Chance does not annihilate the writer’s intention or skills. Indeed, the writer’s specialised knowledge enables them to recognise that a plain rock is potentially precious and inspiring, as illustrated by the legend of Bian He,Footnote 6 a jade master who insisted that an unremarkable rock contained a magnificent jade while everyone else was unconvinced. Bian’s expertise was affirmed when the rock was opened and revealed an invaluable jade that became the imperial seal. Likewise, the poet’s knowledge allows them to discern the aesthetic and creative potential of the objects and phenomena that they encounter, and craft these into writing. This is why a tall rock appears simply as a rock to most people, whereas the lithophilic Mi Fu sees in it an image of his “elder brother.” Geological imagination requires having an eye for the scholars’ rock’s aesthetic value and visualising it in its particular environment to stir up sensory experience and emotional responses. This skill is essential for the poet, painter, as well as the exhibition curator and literary critic.

Comparative reflections

Read alongside scholars’ rocks and Lu’s Rhapsody, Caillois’s lithic writings can be interpreted in several ways: as the product of a combination of factors including dreamstones and their particular textures, shapes, and colours; Caillois’s own paradoleic scrying and theorisation of stones that neither speak nor represent figuratively; and Caillois’s own intellectual context where the notion of chance as either meaningless coincidence or significant but unknown causality was fiercely debated. What Caillois rejects is not so much the coincidental as the dismissal of coincidence as fruitless for the poet’s imagination and human rationality. Coincidence as a significant encounter, though unpremeditated by the writer, is not always adverse to literary production, as Rhapsody proposes. Nor do authorial intentions solely derive from the writer’s individual agency and capacities. Stephen Owen’s view (1985, p. 20) that the classical Chinese poet is a medium of the manifestation of “the world’s coming-to-be” proposes a view of creativity where the artist’s person is only one instead of the central factor in the process. This is especially appropriate for mineral poetics à la Caillois, which depersonalises the viewer in a mineral-human cosmic connectedness and inserts the human into deep geological time. The poet is both medium and agent, which entails understanding artistic intentionality as partially formulated by chance, sometimes even felicitously reduced by chance.

Reading Caillois with Chinese rock aesthetics and Lu Ji therefore has critical implications for literary theory and comparative criticism. For this encounter between twentieth-century French thought, premodern Chinese poetics, and stones is by no means obvious or inevitable. Caillois has been meaningfully compared to other twentieth-century French writers, dissident Surrealists and geologically-inspired writers such as Victor Segalen and Francis Ponge; whereas Lu is almost exclusively examined within Six Dynasties poetics and Chinese literary criticism. Their critical convergence here shows that both writers recognise there is always something that falls outside human agency, whether it is primordial natural forces or the “heavenly pivot,” thus outlining a non-anthropocentric view of creativity. Likewise, their focus on evocative objects and external stimuli that elicit aesthetic responses also acknowledge existences that fall outside human language and theorisation to an extent. The enigma of stones that remains unfathomable for Caillois and the mysterious flows of inspiration for Lu affirm an untheorizable surplus, especially in the form of opaque stones and non-human phenomena that manifest themselves but resist speech. Perhaps chance is also this ungraspable excess, and is not always serendipitous for literature, aesthetic appreciation, and critical thought, as in Lu’s expression of frustration when the poet is disoriented and misses a good opportunity. But acknowledging this excess is indeed required by the non-anthropocentric depersonalisation arising from Lu’s “resonance with objects” and Caillois’s “mental effervescence” with stones. This acknowledgement chimes with contemporary theories of aesthetics, such as Object-Oriented ontology, which affirms that “objects […] cannot be reduced to literal paraphrase” (Harman 2019, p. 592), thus shifting the critical focus from the attempt to exhaustively articulate an object to “aesthetic experience as a non-literal access to the object,” where the object’s “sensual qualities” are transferred to the “aesthetic beholder” (Harman, 2018, p. 260). Moreover, in recent World Literature criticism that questions determinist interpretations based upon historicist and contextual accounts of literature, there is also more acknowledgement of the importance of chance in the critic’s approach to literary texts. As Hoyt Long writes (2021, p. 4), determinist modes of interpretation miss the “‘differing and often unpredictable ways’ that texts, ideas, and people ‘disconnect and reconnect’ (Felski, 2015, p. 162, original emphasis),” “which are not reducible to the effects of historical and social structures.” We therefore cannot assume that chance always plays a negative or negligible role in an artwork, in literary theory or the critical construction of comparisons.