On moths and butterflies, or how to orient oneself through images. Georges-Didi Huberman’s art criticism in context Vlad Ionescu ‘Yes. When all else fails, philosophise.’ J.M. Coetzee, Disgrace (In the loving memory of Marc Verminck, 1954-2014) 1. Art historians question two fundamental objects: firstly, history as narrative, and secondly, the image as a form of representation. While most art historians are implicit philosophers of history and the visual, very few are so explicit about their philosophical background as Georges Didi-Huberman. After all, his intellectual acumen is well known. The question that this essay addresses concerns a concrete episode in Georges Didi-Huberman’s reflection on images, namely the frequent elaborations on butterflies and moths as objects of visuality.1 What do these images of fragile insects add to our understanding of images and to art historiography? My hypothesis is that they question the status of the image in the modern art historical discourse. The seemingly chaotic movements of butterflies and moths denote the fact that images are apparitions whose potentiality is unravelled when seen and thought of in broader relations to other images. To think the potential of images means to ‘weave’ them in wider constellations. Moreover, these constellations of images modify the temporality implicit in any history of art. Instead of thinking temporality as a diachronic narrative (the chronological story of artefacts), an alternative would be to conceive it as the anachronistic montage of heterogeneous images. With Didi- Huberman, the mode of writing history is that of the atlas of images and its significance consists in the ability to distil remerging visual forces. The visual has in Georges-Didi Huberman an anthropological and hermeneutic value as it pictures how humanity represents itself throughout history. In order to see this, we need to integrate his conception of art history into the broader context to which it directly or indirectly relates. * An earlier version of this essay has been published in the Romanian journal Images, Imagini, Images 5/2016. 1 We refer here to the two volumes entitled Essais sur l’apparition: Phasmes (1998), which consists of essays written between 1983 and 1997, and Phalènes. Essais sur l’apparition (2013) that collects works published from 1998 to 2007. Journal of Art Historiography Number 16 June 2017 Vlad Ionescu On moths and butterflies, or how to orient oneself through images 2. Let us start with three episodes that elucidate much of Didi-Huberman’s writing about art. The first episode concerns the Eskimos at Bering Strait and the Kwakiutl Indians of British Columbia. The Eskimos used to throw the bladders of seals and whales back into the sea because they believed that the animals’ souls were attached to these organs. By returning them to the sea, the souls will reincarnate and the bodies could be caught again. Once a year, a ritual would take place to celebrate the bladders of the animals, so that they could come back as whales and seals that could be killed again (Frazer 526-527). The Kwakiutl Indians also throw the salmon offal and bones back into the sea because these remains supposedly contain the soul that returns into the realm of the salmon. If they were burned, then the soul would not reincarnate. Even more, the Ottawa and the Huron Indians think that burning the bones will allow the souls of the dead fish to warn the living fish against forthcoming danger. However, the Hurons have preachers who talk to the fish and convince them to let themselves be caught. These mediators between man and animal are important members of the community because sustenance depends on their persuasive eloquence. They urge the fish to be courageous because they provide for friends ‘who honour them and do not burn their bones’ (Frazer 527). By attaching consciousness to inert matter and animals, these preachers contribute to the food sustainability of the community. Yet, isn’t this transferring of consciousness quite similar to the work of the art historian who sees in images more than they actually show? Of course, no art historian takes this transference literally – as a Kwakiutl Indian would do – yet both point to an over-determination of appearances. Does not this principle of interpretation reflect the ‘act of lending a soul’ (der Akt der Seelenleihung) that Fr. Th. Vischer described in his epochal essay Das Symbol (1887)? Symbolisation, understood as the act of animating the inanimate, is essential to humanity, even when impersonal nature is no longer viewed as a god.2 There are many names for this process in the nineteenth century: Karl Köstlin called it Formsymbolik and Volker Einfühlung. Yet animating the inanimate does not mean merely transferring consciousness to inert matter. It designates a process that emerges in between moments and aspects of things as they show themselves to us, partially and fleetingly. Nothing appears to consciousness completely and eternally. Symbolisation occurs when the mind capitalizes on the surplus of presence that it desires but it does not have. This notion therefore denotes the most common human experience, the confrontation with something that is either absent or only partially present, something that, in any case, is not yet consumed or fixated. We speak to the dead because there is always more to the living than meets the eye. In his fascinating 2 Follow Vischer’s cautious formulation: ‘This act of lending a soul remains the naturally necessary trait of humanity long after the myth is left behind; yet now with what we call reservation (nur jezt mit dem, was wir Vorbehalt nennen); thus also the I foisted under the impersonal nature is not a deity (so wird den auch das der unpersönlichen Natur untergeschobene Ich nich zu einer Gottheit), neither are poems composed from that place, nor do myths appear – something emerges that is similar to this, yet that does not belong to this context, but it is rather a matter of clear symbolism without illusions’ (täuschungslos hellen Symbolik; 435, my translation). See also Götz 53 ff. 2 Vlad Ionescu On moths and butterflies, or how to orient oneself through images analysis of Tony Smith’s minimalist blocks, Georges Didi-Huberman advances the idea that there might be something inside these cubes, an equivocal surplus that resists the objectifying gaze and points to the ‘obscure interior’ of these sculptures that look like tombs (Ce que nous voyons 79). Minimalist art is comparable to the bladders of whales and with it art history becomes a theatre of the sensible that looks for its meaning on the stage of time. 3. The second episode concerns art historians who talk to butterflies. These colourful insects were present in Aby Warburg’s life at a tragic moment when, as a patient in Binswanger’s clinic, he ‘practiced a cult with the moths and butterflies that [flew] into his room at night. He speaks to them for hours. He calls them his little soul animals (Seelentierchen) and tells them about his suffering’3 (Michaud 171). The hallucinations of a tormented psyche are transferred from the doctor’s testimonies into art historical records. The symptom, a central notion in Didi-Huberman’s art history, is deeply inscribed in the Warburgian psyche. These insects form the background of a positive transference: they are animated and they move erratically all over the room. Here we can already detect distinctive features of the image both in Warburg and in Didi-Huberman: the perceived movement throughout history of a pulsating and dynamic tension that re-emerges. Art history does not merely organise artworks according to styles and epochs like a pharmacist who arranges medicines in different drawers; actually, it is the associative and vibrant connection of figures and affects that is prototypical to human visuality. The writing of history is not meant to attach a univocal iconological meaning to a specific image but rather to weave through the works and animate them in endlessly new constellations. Writing art history is a constant search for connivances and affiliations between motifs, the systematic plundering of the past in order to generate alliances between images. Finally, the third episode concerns another art historian, Aloïs Riegl, who was also talking to his doctor a few years before succumbing to cancer at the age of forty- seven. Unlike Warburg’s, the tone of his conversation is lucid and rational. Riegl opens his essay Kunstgeschichte und Universalgeschichte (1898) with a comment of his family practitioner who was interested in natural history but disliked art history: ‘He was not fond of my job’, he writes (Gesammelte Aufsätze 3). The physician adds that art history is a futile attempt to ‘describe the indescribable;’ it is a copy of a work created in a moment of intoxication and it should be experienced in the same way. However, true to the attitude of a real scientist who does not criticise that which he has not seen, the doctor has taken a course in Dutch painting. His diagnostic was that those forty hours of biographical overviews and anecdotes about artists had been futile, yet he valued the comparisons between Rembrandt’s portraits and the representations from the Roman Iron Age, the fact that ‘the most distant phenomena’ had been brought together. Although the physician was but a dilettante in this field, 3 Besides Michaud’s book (originally published in 2000), two other seminal studies on Warburg have been issued in France during the last ten to fifteen years: Georges Didi- Huberman’s L’image survivante (2002) and Marie Anne Lescourret’s Aby Warburg ou la tentation du regard (2014). 3 Vlad Ionescu On moths and butterflies, or how to orient oneself through images Riegl argues that his attitude is ‘symptomatic’ for the most modern developments in art history (Gesammelte Aufsätze 4). After a period of specialised study and historical overviews of artefacts, art history realises that determining the immediate causes and effects of artworks, the socio-cultural context in which the artists worked, does not explain them. Instead, argues Riegl, art history should correlate ‘artistic phenomena’ (Kunsterscheinungen) according to their ‘consolidating generalising characteristics’ (vereinigenden verallgemeinernden Merkmake, Gesammelte Aufsätze 6). After all, ‘there is no ascending development in the human fine arts’ (es gibt keinen aufsteigenden Entwicklungsgang in der bildenden Kunst der Menschkeit), only reoccurring laws that constitute the pertinent object of art history (Gesammelte Aufsätze 9). Riegl is obviously associating his own rigorous art historical project to the realm of science. Yet is he also announcing – already in 1898 – the type of atlas of symptoms that Warburg will later initiate and that Didi-Huberman will return to? When art history becomes a science of art, it appears to the physician’s gaze as symptomatology. With Warburg and Riegl, art history makes for deficiencies associated with scientific thought that even Aristotle had recognised, namely that science is never just a collection of facts, regardless of how extensive or exhaustive it might be. There is no science of singular objects or juxtaposed moments. In order to become a science, the thinking about art has to reconfigure the temporality presupposed in its narrative, that is to say the diachronic arrangement of artefacts. For the historicist approaches that precede Riegl and Warburg, the time of art is a straight line and art history is a taxonomy of these man-made objects. Alternatively, the time of this thing called art is the synchronic determination of motifs that sporadically re-emerge. The method of this new art history is the construction of analogies and contrasts between different images in order to extract these fundamental figures. While Warburg correlates images to rituals and poetry, Riegl argues for a history of the image as a visual phenomenon and – in the famous last chapter of the Late Roman Art Industry (1900) – he interprets the Early Christian art as the visual analogue of St. Augustine’s theology. In both cases, the revolutionary methodological moment consists of connecting the artwork to other forms of thought that actively mediate the representation of the world. In both cases, the history of this thing called art is a dialogue of humanity about how the world appears to the consciousness. Whereas Riegl argued for subsuming artistic phenomena under the general laws of form, Warburg detected in their specific features a dynamic polarization between past and present. Georges Didi-Huberman’s art history follows this dynamics; the French thinker makes manifest its latent visual energies and regenerates its fluttering course. 4. These three episodes crystallise two essential points that re-emerge in Georges Didi- Huberman’s practice of art history: first, that images carry a porosity of meaning that becomes manifest when they are considered within larger constellations of images; secondly, that the modern history of art rethinks the temporality of its objects. In this context, the image of the butterflies, moths but also the hardly detectable phasmida (or stick insects) seems out place. Yet they appear in the writings of Georges Didi- Huberman as ways of questioning the image and its treatment in art history. The phasmida, for instance, illustrate an essential aspect of images, namely the 4 Vlad Ionescu On moths and butterflies, or how to orient oneself through images phenomenon of apparition as opposed to mere appearance. Stick insects have no head or tail so one cannot associate a soul to them. They camouflage themselves so that, in order to perceive them, one must focus on everything around them and still remain sensitive to any differentiating outline emerging from the background4 (Phasmes 17). This intuition also informs Riegl’s entire model of the image defined as the relation between ‘outline and colour on the plane and in space’ (Die Spätrömische Kunstindustrie 6, 392, 229). The history of visual arts is the history of the open space that emerges when a figure emanates from the plane and signals a minimal density. For instance, even though the Egyptian relief represses all sense of space and reduces shapes to the surface, the hands overlapping with a body already signal a minimal contrast between background and foreground. The history of art becomes with Riegl the opening up of the space between a rigid outline and the background from which it emerges. Furthermore, the drawn outline is a human invention that does not exist in the model the image is based on. It proves that the image is grounded on dissemblance and not on the emulation of a natural model, for instance a plant (Stilfragen. Grundlegung 2). The acanthus ornament does not copy but transforms the existing plant and the arabesque is the product of a stylistic development in time. Dissemblance from the pre-existing natural models and images justifies the metamorphosis of artistic forms. Analogically, Georges Didi-Huberman argues that the vivarium might seem inert until a slight movement terrifies the child who knocks on the glass pane. On the one hand, the body of the phasmida extends into the scenery where it lives and feeds. Even though we might intuitively think that the insect imitates the environment in which it lives, it actually subverts imitation because there are no discontinuities between branch and insect, between copy and model. On the other hand, the insect scares us because of an unexpected and momentary dissimilarity. We could say that appearance becomes an apparition due to a sudden fleeting movement.5 The phasmida denies the possibility of becoming an image because it does not emerge from the background, all in order to protect itself. Humans emulated their example by using the camouflage, literally or figuratively. Think of Humbert Humbert in Vladimir Nabokov’s Lolita (1955) who chooses the town of Beardsley for his nymph guided ‘not only by the fact of there being a comparatively sedate school for girls located there, but also by the presence of the women’s college. In my desire to get myself casé, to attach myself somehow to some patterned surface which my stripes would blend with, I thought of a man I knew in 4 Boehm too conceives the image as the ‘iconic difference’ emerging as a fundamental contrast between figure and background (in Alloa 33-34). Only the camouflage, he argues, suppresses this distinction because the point is to dissolve the figure into the background. 5 Didi-Huberman evokes Roger Caillois’s reference to the excessive mimetism of some insects that absorb the environment like the eggs of moths that emulate the image and even organic structure of seeds (Phalènes 62). The idea of incorporating the other also appears in Disparates sur la voracité (Patchwork on voracity), an essay from 1991, where various cases of people eating what they wish to become are discussed. Voracity is explained as an act of emulating and incorporating in order to become the other (Phasmes 169-170). Yet the emulation of the other can also be interpreted as an act of representation: when Christ shares his own body and blood, this emulation anticipates his own representation. 5 Vlad Ionescu On moths and butterflies, or how to orient oneself through images the department of French at Beardsley College’ (Nabokov 174-175). In the imagination of the writer who is also a passionate lepidopterist, the predator hides as a stick insect in a community of preys so that he cannot be detected. The understanding of image as appearance and apparition is also a central distinction in Lyotard’s aesthetics. While the appearance refers to a degree of likeness between the image and its model, the apparition is a sudden emergence of the sensible that disturbs all appearances. While the pragmatic gaze reads the objects that it expects to unfold in time, ‘the apparition arises as the ruin of appearances’ (Que Peindre? 416), like Daniel Buren’s columns in the Palais-Royal that remind us that their apparition is temporary but defies the chronology of appearances. In his captivating La parabole des trois regards (The Parable of the Three Ways of Seeing, 1987), Didi-Huberman tells the story of a ‘surveyor of the visible’ (arpenteur du visible) who meets Aphrodite in an ‘anadyomenic sleep.’ She grants him three wishes: he chooses first a form of absolute vision that allows him to see everything at once, even during sleep. This proves inappropriate for humans because, instead of perceiving the world in perspective, he sees every single detail equally clear but never the thing as a whole. This plainness is so tiring that his second wish is to sleep. Instead of an infinite proliferation of images, now he sees no logical relations between them. Like in a manifest dream, Aphrodite’s hips melt into the foam that turns into a shoreline (Phasmes 117). The third way of seeing is to look beyond the visible into the visual, a distinction that recalls the Augustinian immanence of light where each section of space folds into a coruscation of virtual figures.6 In the visual realm that the surveyor inhabits there is no differentiating distance so that he dwells in a living whiteness that constantly changes. The price he pays for looking at Aphrodite Anadyomene is to become a spatter of foam and blood, to dissipate into formlessness (Phasmes 119). It ensues that the visual is a virtual excess of sensations that appears to us humans in adumbrations. There is no reference to Lyotard’s similar understanding of the visual and the visible in The Exposure (Que Peindre? 344-78). Lyotard uses these terms in the opposite sense to Didi-Huberman’s, yet he argues that all visual implies a surplus that we will or would be able to see (347). The invisible and the unrepresentable in Lyotard, however, do not designate the incapacity to represent but the fact that the visual testifies to an inexhaustible surplus of sense that makes dialectical images possible. This conception of the visual as an abrupt discontinuity with the perceived environment (as it is the case with the phasmida) and as latency of sense resurfaces in twentieth-century French aesthetics. 5. Butterflies are distressing apparitions on all levels of human experience. From a phenomenological perspective, they signal a perplexing movement that disturbs the viewer’s attention. When looking at flying butterflies, we perceive highly irregular sensings that make it difficult to distinguish the continuity of their forms and movement. From a phenomenological perspective, consciousness perceives flying butterflies as unforeseeable lines of flight. From a semantic perspective, the idiom 6 Saint Augustine’s theory of vision is discussed in the essay Les paradoxes de l’être à voir (The Paradoxes of the Being-to-Be-Seen, 1988) published in Didi-Huberman 1998, 121-36. 6 Vlad Ionescu On moths and butterflies, or how to orient oneself through images ‘butterflies in the stomach’ denotes the euphoric effect that throws the body off- balance. It points to a slight imbalance due to a sensitive indeterminacy (a look, a touch) that unsettles the body’s centre and causes one to remain briefly speechless: a ‘guizzo’ (as the Italians say). The young also speak of ‘butterfly kisses’, the fluttering of eyelashes on the skin (preferably all over the body) that triggers an obfuscating pleasure. Now, butterflies and moths also constitute a figural conception of the image as such. Their irregular fluttering around points to an inherent indeterminacy specific to images whose latency of sense remains to be discovered. No authentic image is fixed and framed, dated and situated in history once and for all. On the contrary, the erratic image has a dynamic relation to other images, and thus it establishes a connection between past and present. Georges Didi-Huberman reverses the Romantic conception of art history founded on the metaphor of life (life of the spirit or life of an organism) that progresses in time following a straight line from past to present. Just as Warburg and Riegl, he sees the life of images as an equivalent to their ability to engender new meanings when reassembled in new constellations. As Didi- Huberman explains in L’armoire à mémoire (The Memory Cabinet, 1995), when St. Paul writes that certain events took place ‘in figura’ rather than ‘in imago’, ‘figura’ denotes not an object but a ‘temporal relation’ between a past in close connection to the future, in much the same way as the New Testament was ‘prefigured’ in the Old Testament7 (Phasmes 143). Georges Didi-Huberman compares the image to a flying butterfly whose movements are grasped together in a perpetual motion of opening and closing, appearing and disappearing, systole and diastole, inhaling and exhaling (Phalènes 9, 10, 13). The cinema is another figural conception of the image since – following André Labarthe – a motion picture presents us with a perpetual rhythm of apparitions and disappearances. In his 1999 L’être qui papillone (The Fluttering Being), Didi-Huberman detects this joint movement in Alain Fleischer’s video L’homme du Pincio (1991-93). A man walks around the Pincio Gardens and moves intermittently all around Rome without a specific destination. This aimless walk is a fluttering movement, like the polymorphous, seductive yet purposeless flight of a butterfly8 (Phalènes 157). 6. Both the stick insects (phasmatodea) and the moths (phalaena) point to light but also to phantasm. They represent a metamorphosis from the formlessness to the formed, from the chrysalis to the final imago (Phalènes 12). Butterflies follow a different generative process than people: they emerge from the ground as chrysalides and grow into adult imagines, symmetrical and harmonious creatures (Phalènes 16). Even though these aspects are perfectly visible when the insect is spread out and pinned 7 This essay is an analysis of Fra Angelico’s Armadio Degli Argenti (1451-53) that was originally conceived as a decorative piece for a silver chest and depicts the life of Christ. 8 Similarly, in The Acinema (1973), Lyotard described this polymorphous process as a release of energy without any exchange value. He illustrated it with the image of a child lighting a match without a specific goal in sight. 7 Vlad Ionescu On moths and butterflies, or how to orient oneself through images down in the specimen box, the butterflies’ hovering around is essential in perceiving their dynamic beauty where form and formation are intertwined. We could evoke the butterfly hunt in Benjamin’s Berlin Childhood around 1900 and the disorientation of little Walter, who feels that the roles have been switched: while he becomes the butterfly, the latter ‘took on the colour of human volition […] its capture was the price I had to pay to regain my human existence’ (351). Like the science of butterflies, the science of images captures this dynamic and unstable state. Both images and butterflies imply this constant migration and transformation.9 This is an essential feature of the analogy between the two and Didi- Huberman – deliberately following Warburg – is quite clear about the metamorphosing power of images throughout history: The image wanders around, it comes and goes from here and there, it spreads itself without obligation and plainly. In short, it flutters around (elle papillonne), as we say. This does not mean at all that it is inexact, improbable or inconsistent but that all knowledge of images must in general be established as a knowledge of exploratory movements – of migrations, said Aby Warburg – of each image in particular. (Phalènes 17) This passage is significant because it illuminates the model of art history that both Warburg and Didi-Huberman are pursuing, a model where the image is in an unstable state and where historical time knows no terminus ad quem. Instead of fixating its ‘immanent sense’ (Panofsky) and subsuming it to a chronological taxonomy of styles, this kind of art history understands the image as the apparition of a virtual sense whose realisation depends on its correlation with other works. For Warburg, just as for Riegl, this approach was a reaction against the ‘universal history’ model and the specialised accounts that it produced, often imbued with subjective aesthetic taste. D’Agincourt’s Histoire de l’art par les monuments (1810- 1823) is such a systematic overview of thirteenth- and fifteenth-century artists, Giotto and Pierro della Francesca, which disregards other historical styles. D’Agincourt’s narrative aimed at providing an exhaustive arrangement of only those artefacts that conformed to the historian’s own aesthetic taste. The epistemology of art history integrated aesthetic normativity into a one-dimensional narrative: history moves forward and the task of historiography is to provide a full account of this movement while at the same time dismissing artistic debris. Friedrich Von Rumohr and Franz Kugler strived for situating artworks in their historical context and their works are monuments of this universal art history. However, whereas Kugler idealises the Germans and ignores the Renaissance, Karl Schnaase continues these systematic overviews in a history of art that attained monumental dimensions. An army of art historians co-edited a project that in its second version (published between 1866 and 1879) counted eight volumes. Romantic in tone, such projects are driven by the desire to incorporate separate artefacts in a unified diachronic narrative. They are imbued with aesthetic and nationalistic biases. Nineteenth-century art history examines the 9 On the description of butterfly as stable apparitions see James Elkin’s wonderful chapter on moths’ wings from How to Use Your Eyes? (2000), pp. 182-89. 8 Vlad Ionescu On moths and butterflies, or how to orient oneself through images past by projecting on it its current desires, Romantic notions of originality, and ideal forms or national styles. Yet as this universalism fails, art history starts to speculate and conceive visual meaning as constellations of disparate motifs. With Didi- Huberman, art history imitates Walter Benjamin’s butterfly hunt: the historian- collector has to follow the movement of the butterfly-image as the only viable strategy to capture something new… In his 1996 Don de la page, don du visage (Gift of the Page, Gift of the Face), Didi- Huberman relates the figural potential of Victor Hugo’s stained manuscripts to the Rorschach cards. The written text melts into ink stains and the page opens up towards the tension between text and form, thus multiplying the possibilities of reading and seeing it (Phasmes 153). Rorschach cards maintain the inscribed figure in a state of active potentiality: the inkblots are ‘figuring’ (figurantes) rather than ‘figured’ or fixed in the way that sentences consist of words that follow a strict set of syntactic rules. An irregular figural force disturbs the regularity of the discourse and transforms the text into a figure-image, as Lyotard argues in Discours, Figure (1971). After all, these figuring images concern the non-iconic potential of images that reveal a ‘purely visual and virtual field’ (Phasmes 166). Is this not the same sensible indeterminacy detected by Lyotard in the dream-work, as shown in his analysis of Mallarmé’s Un coup de dés jamais n’abolira le Hasard?10 7. Georges Didi-Huberman’s essays remind of Adorno’s analysis of a form of writing that resembles a ‘carpet’ of thoughts woven together (160). But, while they are indeed constructed as a dense fabric of textual and visual references, Didi- Huberman’s essays nevertheless lack the experimental character that Adorno placed at the centre of his investigation of a type of writing exposed to error in much the same way as learning a language by using words picked up in different contexts. The experimental character is more typical of the montages that Didi-Huberman thoroughly addressed11. These montages of images are the objects of a dynamic imagination that perceives connivances between things. This dynamic imagination, associative and hypothetical, emerges when Didi-Huberman traces relations between seemingly heterogeneous events. For instance, he finds a latent relation between the fact that Goethe wrote on butterflies and caterpillars roughly at the same time that he described the Laocoön as representing the ‘transitory moment.’ Freely building upon the same motif, Didi-Huberman refers next to Adriaen Coorte’s Three Medlars with a Butterfly (c. 1705), a still life that includes a colourful unmoving butterfly that seems to look at the viewer (Phalènes 26). By capturing this improbable moment of 10 See Discours, Figure 60-72. In his analysis of Mallarmé’s ‘object-book’, Lyotard tackles the tension between an ‘object of signification’ that can be read and understood (the text) and an ‘object of significance’ that consists of plastic signifiers (blanks, typeset). The two categories mirror each other but not in the same way that a signifier refers to a signified. Rather, this relationship bears a certain resemblance to the transformation of the discursive order by the sensible order, just like in the dream-work. 11 I refer here to the lecture on Jean-Luc Godard’s Histoire(s) du Cinema that Didi-Huberman gave at the conference What Images Do in Copenhagen (19-21 March 2014). Godard’s film is the subject of the fifth volume of the series L’oeil de l’histoire entitled Passés cités par JLG (2015). 9 Vlad Ionescu On moths and butterflies, or how to orient oneself through images motionlessness, the image crystallises imitation and imagination. Whereas the real insect is the model of the painting and can be dissected as any other organism, its motionless presentation is interpreted as an instant when the psyche is abandoned to the unconscious (28). To put it simply, the insect is not presented as an object among other objects but as the correlate of a dream because the very moment that the painting presents is never perceivable as such. Based on the (Benjaminian) notion that the past, just like the image itself, is an unstable apparition, this interpretative model feeds on free association and perpetual transformation of the past. Here, the function of history is not to take possession of the past but rather to display its figural alliance with the present. Like the butterfly, the past follows unpredictable directions, enchanted as it is by its ineluctable acrobatics. The ambivalent re-emergence of motifs is the prerequisite of this interpretative model that cultivates their excess and latency of sense. The image of the butterfly is illustrative for this: the insect’s wings show symmetrical patterns when raised, and yet, during flight, they create complex shapes and chaotic lines as the ‘eyes’ painted on them seem to multiply. Didi-Huberman evokes here the Oudemans phenomenon described in Adolf Portmann’s Animal Forms and Patterns (1948): when spread, the butterfly wings display the motif on their outer margins, but when at rest, the insect becomes invisible on its branch (Phalènes 68). We are dealing here with a canonical polarity in art historiography (that divides researchers even today). On the one hand, there is the positivist investigation of primary sources in their historical context. Here, art historians interpret artworks by stabilising them, just like a dead butterfly is pinned down on a collector’s board. On the other hand, there is the speculative iconological model that discovers ‘adventurous coherences’ (in Caillois’ terms) and visual connections. This is no longer a history of artefacts fixated in their cultural environment but the analysis of visual forms that pulsate like organisms. They ‘flutter around’ amongst images and across historical time (Phalènes 77). 8. Georges Didi-Huberman’s art history reincarnates a truly modern ideal because it sees the past as a malleable given related to the present. For him, the time of artworks is the present perfect continuous, an ongoing movement from past to present. It is only when a motif from a different time re-emerges into the present that its past is not complete and that its sense is not fixed, dead like a pinned down butterfly. This line of thought is usually traced back to Warburg but it actually originated in the work of Franz Wickhoff and Aloïs Riegl, central figures of the Viennese School hardly mentioned in Georges Didi-Huberman’s work. In his Roman Art (1900 [1895]), Franz Wickhoff introduced the notion of ‘illusionism’ by alternating examples of second- and third-century Roman art with modern Western artworks and comparing the plein air Impressionism of his time to Japanese art. Writing roughly at the same time as Warburg, Wickhoff defines illusionism as the fleeting ‘appearance [of an image] at a given moment’ (18), thus echoing Warburg’s ‘transitory movement in hair and garment’ discussed in his 1893 essay on Botticelli (19). This variety of illusionism emerges in the Roman period but has different historical manifestations. Japanese painting also allows us to understand both the 10