About CA
Contact CA

Search Journal

Editorial Board

Permission to Reprint


Site Map


Rhizomatic Mnemosyne: Warburg, Serres, and the Atlas of Hermes

  María del Carmen Molina Barea

Author’s note: I would like to express my gratitude to Professor Charles Burnett (The Warburg Institute, London)

This essay aims to examine Aby Warburg’s Mnemosyne Atlas according to two conceptual perspectives that seem deeply interwoven, Deleuze and Guattari’s notion of rhizome and Michel Serres’s metaphor on Hermes. Both theoretical approaches cast light on the epistemological implications of the Mnemosyne Atlas and explore its intriguing composition from an innovative point of view. Specifically, this paper excavates the disrupted nature of the Warburgian Atlas, paying particular attention to the schizophrenic proliferation of unexpected connections. In this scenario, it will be necessary to elucidate the terminological opposition between ‘atlas’ and ‘archive,’ as studied by Boris Groys, Foucault, and Derrida, without leaving aside Didi-Huberman’s pioneering research on Warburg.

Key Words
archive; cartography; Gilles Deleuze; Félix Guattari; Hermes; Mnemosyne Atlas; rhizome; schizophrenia; Michel Serres; Aby Warburg


1. Warburg’s rhizomatic anti-method

Comment organiser l’interdisciplinarité?[2]

When Aby Warburg was twenty years old, he traded his birthright as the firstborn son in exchange for his brother’s promise to buy him books for the compiling of a library. The result of such an exchange was the world famous KWB Warburg Library of Hamburg. Its founder articulated the collection following an apparently random order that did not fit a uniform pattern. The books were displayed on the shelves with no regard to any homogeneous model; Warburg himself constantly changed the location of the books. In doing so, Warburg intended to invite visitors to make inspiring connections between diverse topics and generate new ideas when going through the corridors full of books.

In fact, the library did not work according to any standard cataloguing system. To a certain extent, it could be said that the library took on a life of its own.[3] “In brief, Warburg orders the Library in such a way that it ‘wants not only to speak, but also to listen attentively’….”[4] By the same token, from 1924 to his death in 1929, Warburg devoted his efforts to a specific project that went in tandem with the library, the Mnemosyne Atlas. Composed of sixty-three mobile panels, the Bilderatlas, as it is also called, put hundreds of photographs related to several research themes side by side. The purpose of such an apparatus was to ultimately build up a transversal history of the survival of psychological expression in visual culture.[5] What is interesting about the Atlas is that Warburg frequently changed the position of these pictures, removing and detaching them according to the development of his own scientific work.[6] “He repeatedly rearranged these images, just as he repeatedly rearranged the books in his library and even the order of words and phrases in his written texts.”[7]

As can be seen, the Mnemosyne Atlas is not properly a book or atlas in the traditional sense. It is rather a deconstructive space, a milieu for contrast and dialogue, and a battleground of images and mutable concepts that proceeds according to connections and disjunctions. As is well known, Warburg called such mechanism the "law of the good neighbor." For his part, Georges Didi-Huberman, who has carried out exhaustive research into the Atlas of Aby Warburg, alludes to a dialectical montage aimed at dealing with discontinuities and partial knowledge. In Didi-Huberman’s view, Warburg shows a destructive behavior that paradoxically makes room for the appearance of creative relations. Thus, thanks to this anti-method, Warburg promotes the arousal of unpredictable events within the epistemological realm. In this context, it is highly significant that Warburg’s disorganized procedure bears a strong resemblance to Gilles Deleuze and Félix Guattari’s rhizomatic proposal.[8] As it is put by those authors, a rhizome is not a root and neither a tree, both of which grow vertically. On the contrary, the rhizome grows horizontally, connecting and disconnecting diverse points.

In this respect, Deleuze and Guattari distinguish some rhizomatic principles: the principle of connection and heterogeneity, the principle of multiplicity, and the principle of asignifying rupture. As they say, the rhizome works by means of productive sequences and connective discontinuities (‘and… and… and…’). In the words of Simon O’Sullivan:

A rhizome is a system, or anti-system, without centre or indeed any central organising motif. It is flat system in which the individual nodal points can, and are, connected to one another in a non-hierarchical manner. A rhizome then fosters transversal connections and communications between heterogeneous locations and events.[9]

Thus, it is possible to state certain similarities between the Bilderatlas and the rhizome, inasmuch as they share the same fragmentary connectivity. Put bluntly, the law of the good neighbor works mostly the same as the rhizome. To quote Deleuze and Guattari, “any point of a rhizome can be connected to anything other, and must be. This is very different from the tree or root, which plots a point, fixes an order.”[10] This is also the case of Warburg’s methodology, which disrupts hierarchical structures in favor of nondefined assemblages. The Atlas of Aby Warburg is produced in an unorganized manner, so the result is not an organism but, above all, a vague entity, a diffuse body of knowledge. As Peter Krieger has remarked, this is not inconvenient for epistemological purposes, in the sense that the law of the good neighbor opens a wide range of promising paths.[11] So Warburg breaks with fixed structures in his Atlas, as he already did in his library, being aware that every pre-established order entails a predetermined criterion. Rather, he goes about producing knowledge through connectivity and transversality, the principles of the rhizome, or, as Deleuze and Guattari would say, through resonance, proximity and neighborhood.[12]

2. Nomadic maps

L’atlas ne dessine plus les mêmes cartes.”[13]

If, according to the above, the Mnemosyne Atlas proceeds rhizomatically, then such a device necessarily moves away from traditional epistemology, since it is not a vertical or hierarchical one-sided system, like the tree, but multiple, horizontal, and transversal, like the rhizome. This procedure is against fixed cognitive models, usually grasped through the metaphor of the genealogical tree, which is effectively vertical and hierarchical.[14] This is the dictatorship of the arborescent structure. As has been said before, the epistemological mode of Aby Warburg has nothing to do with the tree; the rhizome is an anti-genealogy.[15] In other words, the rhizome is free from the subjection to the tree that continuously repeats the same trajectory, what Deleuze and Guattari call tracing or decalcomania. Rhizomatic knowledge, on the contrary, is not fixed. The rhizome does not produce decalcomanias. It produces maps, instead. Not for nothing, as Deleuze and Guattari remark, among the principles of the rhizome we also find the principle of cartography. So, rather than drawing tracings, that is, reproducing the same model to infinity, the rhizome composes maps. That said, these maps are not fixed images but concern fluctuations, oscillations, and interconnections.[16] They vary constantly, insofar as the map is rhizomatic by definition. According to Deleuze and Guattari:

Unlike the graphic arts, drawing, or photography, unlike tracings, the rhizome pertains to a map that must be produced, constructed, a map that is always detachable, connectable, reversible, modifiable, and has multiple entryways and exits and its own lines of flight.[17]

Here it is possible to maintain that the Mnemosyne Atlas functions like a rhizomatic map. In the same way that the Warburgian Atlas connects diverse images, maps as conceived by Deleuze and Guattari outline changing routes between different points, intensive zones, or, put more simply, plateaus. “We call a ‘plateau’ any multiplicity connected to other multiplicities by superficial underground stems in such a way as to form or extend a rhizome.”[18] That is because the rhizome is an open and de-centered system of multilateral connections. In fact, the role of the plateaus is to be in between. Undoubtedly, the Bilderatlas exhibits this productive model, thus becoming the perfect framework where plateaus take place, that is, the vibrant space where the rhizomatic connections occur, just like in the  KWB Warburg Library, which has a life of its own. Hence it is no coincidence that Didi-Huberman has accurately noticed the close relationship between Mnemosyne’s project and the notion of rhizome taken from Deleuze and Guattari.[19]

Finally, the plateaus are what Deleuze and Guattari also call smooth space, in opposition to the arborescent striated space. The smooth space is the habitat of the nomad, who becomes responsible for connecting and disconnecting the polyvocal flows traversing through the plateaus. So the nomad is in constant movement; he never takes root. “The life of the nomad is the intermezzo.”[20] His labor is to plug in a multiplicity of coexisting possibilities, an endless variety of plateaus. Consequently, it is thanks to him that the rhizome successfully works. In this context, Aby Warburg seems to undertake the same role. He is also a nomad, in between the panels of Mnemosyne, forging continuously changing networks between its images. It is important not to lose sight of the striking similarities between Warburg and the nomad. Both of them take a similar stance: Warburg puts together and also divides different plateaus through the cohabitation of images. By means of this method, Warburg makes possible a nomadic circuit in which he himself takes part. In fact, “during his work sessions, Warburg was constantly in motion, handling books, comparing photographs, and writing and classifying reports.”[21] Thus, it can be stated that the nomad completely disturbs the given order, the institutional settings. Warburg is clear about this: “It is a matter of perpetual ‘migrations’ (Wanderungen), as he liked to say.”[22]

3. Hermes and the Black Box

Warburg décomposait, déconstruisait subrepticement tous les modèles épistémiques en usage….”[23]

Going further still, the nomad is clearly illustrated by the reflections of Michel Serres upon Hermes, the nomadic god par excellence. As is well known, Hermes’s fundamental task consists of delivering messages between and among gods and mortals, thus connecting and disconnecting networks of fluxes. So, he is the god of communication, transport, commerce, travelers, and sailors, the god whose statue was placed at the crossroads in ancient times. His life, then, is also the intermezzo. Not in vain, as Serres points out, Hermes is precisely the god of migration. Therefore, Hermes works “as an échangeur, a point and instrument of transmission, of communication, a facilitator of circulation.”[24] Indeed, Hermes’s mission is to facilitate connections. For this reason, he always stands at the intersections, "crossroads," to quote the Serresian term. This concept refers to a point of junction where things come together, a sort of maze of connections or simply a multiplicity of crossings. As a result, it could be argued that, like the nomad, Hermes connects plateaus and produces rhizomatic maps, quite similar to the Atlas of Aby Warburg. Also worth mentioning is the fact that if Serres speaks about Hermes, Warburg mentions the Nympha, the divinity who moves forward, Gradiva, who lives in motion.[25] By extension, by emulating Hermes’s nomadism we will be able to achieve the “law of the good neighbor," particularly bearing in mind that, according to Didi-Huberman, this Warburgian procedure is meant to stay at the crossroads (‘croisée des chemins’):

Cette pensée tranche, disloque, surprend, mais elle ne prend aucun parti définitif dans la mesure même de sa nature expérimentale et provisoire, dans la mesure où, née d’une pure transformation topique, elle se sait recombinable, elle-même modifiable, toujours en mouvement et en chemin, “toujours à la croisée des chemins."[26]

This is the sense of topology in Serres: a new cartography traced by Hermes, who thus becomes the author of infinite maps. It is therefore a new form of mapping knowledge closely linked to the rhizome. Small wonder then that Serres spells out the specificity of this phenomenon when talking about the atlas. In his book, entitled precisely Atlas, Serres outlines a new cognitive methodology and explains the significant changes we have experienced in relation to the contemporary production of knowledge. He holds that epistemology has undergone a profound transformation: Now we live in the virtual; as a consequence, we have abandoned traditional forms of fixed knowledge. In his words, there has been a metamorphosis from the hard to the soft. The hard is energy and materiality, and it is associated with words such as matter, finite, and local. The soft is intelligible, indefinite, and global, and it is frequently described as information and meaning, concepts and signs. Simply put, knowledge that was once plainly delineated is now imprecise and diffuse. What once was local is now global. Such a process takes place within a conceptual dispositif that Serres calls Black Box, a term coined to refer to an obscure device that works in between the hard and the soft, transforming the former into the latter through unknowable fluctuations and interchanges.[27]

Note that this fundamental transformation also happens within the changing frames of the Bilderatlas, which somewhat becomes a Black Box in itself. The Atlas of Aby Warburg dilutes and intermingles knowledge in search for the twinning with an abstract and mutable mosaic of kaleidoscopic voices. The Mnemosyne Atlas is actually a polyvocal Black Box or, better said, a unlimited set of Black Boxes, in whose interior the hard is transformed into the soft. Hence, it might be said that the Warburgian Atlas is composed of many Black Boxes, one box inside another, to infinity, as if they were an endless chain of Russian dolls. Each image, each panel, constitutes a plateau ready to be connected, or, in other words, a hard element ready to be transformed into the soft. Furthermore, this makes sense, since Warburg’s main interest was to unveil through the analysis of images the subtle links lying in the heart of the psychohistory of the socius.[28] So, his objective was as though to study the hard in order to achieve the soft. This explains Warburg’s effort of looking for the universal (global) in the particular (local), because, as he used to say, “The Dear Lord nestles in detail." Not surprisingly, Serres shares a similar insight: “Behind the thickness of things, the one called God is almost infinitely hidden.”[29]

In this regard, we could finally say that Warburg is not only a nomadic Hermes but also an angel. Following Serres, the angel is the person who enables the transformative connection between the hard and the soft. So, like the nomad, the angel works on the permanent conjunction and disjunction of the rhizomatic maps: “This person thus fluctuates between the collective and the individual.”[30] Not for nothing, angels are also divine messengers. As Serres says, they carry messages all throughout the Black Box, that is, a sort of Jacob’s Ladder, an endless row of angels going up and down a ladder that connects earth and heaven, turning the hard into the soft. This is precisely the way in which the Bilderatlas maps knowledge. Ultimately, reading Serres’s work, one tends to merge this idea of constant mobility with the immobility of the mythological titan Atlas, condemned to carry the world globe, that is, the traditional atlas, the atlas made a tree. In contrast with this, we found restless angels and, leading them, their predecessor, the winged god Hermes. It is thanks to him that the cartographic rhizome is possible, inasmuch as he works in the intersections; remember that Hermes inhabits crossroads. Because of this, if Warburg’s philosophy was the “law of the good neighbor,” Hermes develops the philosophy of prepositions:

“Where are you?” “What place are you talking about?” I don’t know, since Hermes is continually moving on. Rather, ask him, “What roadmap are you in the process of drawing up, what networks are you weaving together?” No single word, neither substantive nor verb, no domain or specialty alone characterizes, at least for the moment, the nature of my work. I only describe relationships. For the moment, let’s be content with saying it’s “a general theory of relations.” Or “a philosophy of prepositions."[31]

4. The schizophrenic trace

“All mankind is eternally and at all times schizophrenic.”[32]

It clearly follows that the law of the good neighbor is Hermes’s main task. “He produces, alone, a relation among an incongruous mixture of subjects and practices and an incongruous set of objects… .”[33] This melange produces, in effect, a continuous connection between unexpected elements. As far as Warburg is concerned, such a procedure is carried to the extreme, as can be seen in the Bilderatlas, which creates an endless circuit pushing the envelope of connectivity and embracing cognitive production in a paroxysm of infinite possibilities. Let us say that the Atlas of Aby Warburg takes the form of a hyperbolic bunch of rhizomes. At first glance, it looks like a piecemeal labyrinth, responsible for the unleashing of incessant enchainments. This suggests that the Mnemosyne Atlas not only fulfills the law of the good neighbor but also the eel-soup style (Aalsuppenstil). This is another expression used by Warburg to refer to his extremely associative-rhizomatic way of thinking. In this sense, the eel-soup style indicates the kaleidoscopic nature of the Warburgian system that is intrinsically related to the obsessive image of the Laocoön and the Hopi snake dance, widely studied by Warburg.[34] It therefore appears that the nomadic principles of the rhizome are actualized in the Mnemosyne Atlas whereby a patchwork of meandering “snakes” takes place. So it is impossible to find a rational order within such an intricate mass of confusing directions. In these circumstances, Serres could not have overlooked the fact that Hermes’s stick is decorated with a double coiled serpent: “Look at the caduceus of Hermes. Two snakes cross repetitively on it.”[35]

Namely, the eel-soup style of the Warburgian Atlas is reflected in the dramatic saturation of images displayed in illogical order throughout the numerous panels of photographs. The resulting horror vacui has a connotation of excessive disarticulation, altered thinking, mental impairment, and "intrinsic madness,” according to Didi-Huberman. It also inspires certain intellectual anguish; Warburg’s mind seems to be as strangled by the images of the Atlas as Laocoon’s body by the snakes. Indeed, the Mnemosyne Atlas is composed of what Didi-Huberman has named a “manic enchainment of thoughts.” Not in vain, the constitutive system of the Bilderatlas shows a slightly pathological nature. In this respect, too, it is related to the rhizomatic dynamics that Deleuze and Guattari also call schizophrenic or schizoid.[36] Basically, the rhizome is schizophrenic because its core features meet those of the illness, that is to say, a multidimensional phenomenon of disruption, disorganization, and fragmentation that makes unexpected connections between disparate elements. From a clinical perspective, schizophrenics constantly merge hallucinatory ideas driven by dreamlike experiences, delusions of persecution, and paranoid fears. In this process schizophrenics feel like their ideas are melting and disappearing as they constantly migrate and metamorphose. What is more, schizophrenics are unable to grasp and fix their own thoughts because of an increasing flight of ideas; their thoughts become, as it were, liquid and filter through a sieve.

It is worth recalling that for five years Warburg himself suffered from these schizophrenic symptoms. Due to a severe mental crisis, Aby Warburg was committed first to several asylums in Hamburg and Jena and finally hospitalized in the psychiatric clinic Bellevue, in Kreuzlingen, under the care of Doctor Ludwig Binswanger. Initially, Warburg was treated for acute schizophrenia, although the final diagnosis changed into a manic depression disorder with possibilities for improvement, which was indeed the case. Fortunately, the patient overcame the illness, thus putting an end to long years of terrible mental darkness.[37] However, it is certain that Warburg had always showed a remarkably schizoid constitution, and he had suffered from hypochondria and obsessive neurosis from early childhood. Interestingly enough, he even considered himself a "schizo." In his notes for the conference The Serpent Ritual that Warburg gave at the time when he was recovering from his paranoia, he wrote: “They are the confessions of an (incurable) schizoid, deposited in the archives of mental healers.”[38] In fact, Warburg himself was fully aware of the palpable consequences of his schizophrenic tendency: “My illness consists in losing my capacity to link things according to their simple causal relations, which is reflected in the spiritual domain as well as the real.”[39]

In this sense, Warburg declared having experienced racing thoughts and flights of ideas, that is, completely disconnected thoughts, whose result can be appreciated in the profusion of multiple images of his Atlas, a phenomenon that has been defined by Didi-Huberman as a "migration of images.”[40] So we meet migration and nomadism once again. Precisely, Deleuze and Guattari stress this ability of migration when talking about the fluidity of associations within the schizophrenic thinking: “It might be said that the schizophrenic passes from one code to the other, that he deliberately scrambles all the codes, by quickly shifting from one to another, according to the questions asked him, never giving the same explanation from one day to the next, never invoking the same genealogy, never recording the same event in the same way.”[41] Consequently, in this pathological phenomenon we can easily find the seeds of the polymorphous eel-soup style. Thus it could be sustained that the Bilderatlas, with its convoluted and short-circuited network of images, is partly a consequence of Warburg’s propensity to schizophrenia.[42] In the opinion of Didi-Huberman, it is impossible to separate Warburg from his illness, often considered an embarrassing biographical fact. So it is of greatest importance to bear this in mind in order to undertake an in-depth analysis of his work inasmuch as it plays a key role in the production of Warburg’s main project, the Mnemosyne Atlas. Nonetheless, Didi-Huberman cautiously remarks that we should not fall into the trap of thinking that Warburg’s work is simply the result of a hidden sickness or an inner decay but of an acute intelligence.

5. Atlas or archive?

There would of course be no atlas possible without the archive that precedes it… .[43]

Thus, there is no doubt about the schizophrenic implications of the Warburgian endeavor. Moreover, the Mnemosyne Atlas is undertaken at the precise moment when Warburg was just about emerging from his psychosis. The saturation and juxtaposition of images derives, then, from a kind of pathological compulsion of compilation, as the consequence of a sort of disorganized schizophrenic thinking, or, more accurately, a particular way of rhizomatic organization. By means of the excessive gathering of images, Warburg exhibited signs of an obsessive connectivity that almost bordered on madness. Such profusion of elements is closely akin to collecting purposes and echoes the principles of the archive.[44] It is in this sense that we understand Didi-Huberman’s definition of atlas: “An atlas is neither a dictionary nor a scientific manual nor a systematic catalogue. It is a collection of singular things, often extremely heterogeneous, whose affinity produces a infinite (never closed) and strange knowledge… .”[45] Hence, it could be argued that the Warburgian Atlas becomes a sort of dysfunctional collection, like the cabinets de curiosités, wonder chambers (Wunderkammern) and studiolos, and other spaces for the amalgamation of unusual elements. As Suely Rolnik has noted, this archival compulsion continues nowadays, inextricably bound up with the idea of atlas, in the work of many artists.[46]

However, there is a fundamental difference in nature between the atlas and the archive. The latter codifies the rhizomatic knowledge into a fixed corpus whereas the former puts infinite fluxes into circulation. Briefly said, the archive catalogues, lists, and indexes every single element in its interior according to a previously established discourse. The atlas, instead, promotes a heterogeneous polyphony. Let us say that the archive is therefore a genealogical system, an arborescent tree, whereas the atlas constitutes a rhizomatic map.[47] In other words, the atlas is a crossroad open to the connection of plateaus. In contrast, the archive entails epistemological coercion and domination. Boris Groys has thoroughly studied the nature of this controlling system.[48] As he explains, the archive selects and guards systematically valuable cultural things that thus become separated from the rest of mundane objects, concepts, and ideas, and in so doing it not merely conserves certain elements but also institutionalizes them. It does not simply take them from the reality but more importantly produces reality thanks to them. Thus, by means of the deliberate and intentional organization of the collected items, the archive generates a biased view and spreads a univocal discourse. As put by Michel Foucault: “The archive is first the law of what can be said, the system that governs the appearance of statements as unique events.”[49]

In this same vein, Jacques Derrida refers to the archive as origin and command (Arkhé), and highlights the archival system as the site for consignation. In his words: “Consignation aims to coordinate a single corpus, in a system or a synchrony in which all the elements articulate the unity of an ideal configuration.”[50] Put it in a different way, the archive is also what Giorgio Agamben calls an oikomanía: “a set of practices, bodies of knowledge, measures, and institutions that aim to manage, govern, control, and orient - in a way that purports to be useful - the behaviors, gestures, and thoughts of human beings.”[51] For its part, the atlas adopts a completely different method that is open and disfunctional by definition and does not conform to any kind of mandatory structure. It does not codify inasmuch as it works according to constant connections and juxtapositions, so it never produces a corpus of normative knowledge. It never fixes, it never takes root. Warburg said it clearly: the Bilderatlas is not simply a treasure chamber where to stack and classify different things.[52] The atlas is dynamic in itself. That is why the Mnemosyne Atlas is ungraspable; it rhizomatically escapes from codification. It works in motion, in a way, since it is guided by the dancing nymph or the nomadic Hermes. In this sense Warburg realized that the images should not be fixed; consequently he never imposed a specific structure on them. On the contrary, he let the images hold fascinating conversations.[53]

6. Conclusion: The new atlas

Le nouvel atlas dessine cette mappemonde.”[54]

In sum, the Atlas composed by Aby Warburg blurs the hierarchical archive and delineates a new one based on nomadic and abstract principles: the soft, as Serres would say.[55] Therefore, the new atlas looks like a map of limitless boundaries, a "world map" or "knot of intersections," in Serres’s (2008) view, or, differently put, a fluid topography traced by Hermes. Finally, such a map strongly resembles the schizoanalytic cartographies that Deleuze and Guattari define as a “schizo stroll.” Basically, these are intensive maps resulting from the connection and rupture of different plateaus, like the smooth space of the nomad, the territory of the flâneur, or the Situationist psychogeographies. It might be of interest to add that Warburg himself used to draw these kinds of migratory maps, as can be appreciated in the Schemes of Personal Geographies that he outlined in his diaries between 1895 and 1928. This is not strange since it is well known that Aby Warburg had a strong preference for the term bewegtes Leben (“life in motion" or “animated life"). So, in opposition to the arborescent archive, Warburg seems to repeat the Serresian saying: “Comment capter, sur les pages de cet atlas, trop solides, ces jolies cartes agiles?[56]

Warburg’s erratic dynamics answer this question. As stated earlier, the rhizomatic tendency developed by Warburg gives form to a specific epistemological foundation that works according to the law of the good neighbor or the eel-soup style, designated by Serres as a philosophy of prepositions. Not casually, Warburg has also described the result of his methodology as the iconology of the intervals: “An iconology founded on ‘conaturality, the natural coalescence of the word and the image’… .”[57] In this context, as far as Michel Serres is concerned, he supports the idea of the philosophy of chaos that attacks traditional methods consisting of reducing multiplicities to hierarchical structures. All considered, it is according to these criteria that one can figure out the nature of the new atlas, the new way of producing knowledge, that can be found in the Warburgian Atlas. “It deliberately ignores any definitive axioms. For it has to do with a theory of knowledge devoted to the risk of the sensible and of an aesthetic devoted to the risk of disparity.”[58] For this reason, Giorgio Agamben has referred to Warburg’s iconology as "the nameless science,” inasmuch as it entails a unprecedented strategy for approaching cognitive production.[59] In this sense, the Bilderatlas calls into question the generative procedure of knowledge and makes a fundamental contribution to an epistemological mutation. Thus, the Mnemosyne Atlas builds a new machinery for the production of what will be called knowledge, and we find Hermes at the control of this new machine.


María del Carmen Molina Barea

María del Carmen Molina Barea is Assistant Professor in the Department of Social Sciences and Humanities at the University of Córdoba, Spain. She holds an International Ph.D. in aesthetics and art theory, an MA in contemporary art theory from Goldsmiths College University of London, and a BA in art history. Her research focuses on avant-garde aesthetics, postmodern ontology, gender theory, film philosophy, and visual culture studies.

Published on January 1, 2018.



[1] I would like to thank the anonymous reviewers for Contemporary Aesthetics, who provided encouraging and helpful comments that improved the essay at key points.

[2] “How to organize interdisciplinarity?,” Georges Didi-Huberman, L’Image survivante: Histoire de l’art et temps des fantômes selon Aby Warburg (Paris: Les Éditions de Minuit, 2002), p. 42, my translation.

[3] See José Francisco Yvars, Imágenes cifradas. La biblioteca magnética de Aby Warburg (Barcelona: Elba, 2010).

[4] Christopher D. Johnson, Memory, Metaphor, and Aby Warburg’s Atlas of Images (New York: Signale, Cornell University Library Ithaca, 2012), p. 67.

[5] See Aby Warburg, Der Bilderatlas Mnemosyne (Berlin: Akademie Verlag, 2012).

[6] “It is unlikely that the atlas of images was thought of -and must be thought of - in strict relation to the collection of books organized, as we know, along principles that were as disconcerting for a standard librarian as Menmosyne is for a standard iconographer.” Georges Didi-Huberman, Atlas. How to Carry the World on One’s Back? (Madrid: Museo Nacional Centro de Arte Reina Sofía, 2011), p. 166.

[7] Philippe-Alain Michaud, Aby Warburg and the Image in Motion (New York: Zone Books, 2004), p. 277.

[8] See Gilles Deleuze and Félix Guattari, A Thousand Plateaus. Capitalism and Schizophrenia, 2 (Minneapolis: University of Minnesota Press, 1987).

[9] Simon O’Sullivan, Art encounters Deleuze and Guattari. Thought beyond representation (Hampshire, New York: Palgrave Macmillan, 2006), p. 12.

[10] Deleuze and Guattari, (1987), p. 7.

[11] Peter Krieger, “El ritual de la serpiente. Reflexiones sobre la actualidad de Aby Warburg, en torno a la traducción al español de su libro Schlangenritual. Ein Reisebericht," Anales del Instituto de Investigaciones Estéticas, 88 (2006), 239-250; ref. on 241.

[12] See Gilles Deleuze and Félix Guattari, What is Philosophy? (London: Verso, 1994).

[13] “The atlas does not design the same maps anymore.” Michel Serres, Atlas (Paris: Éditions Julliard, 1994), p. 207, my translation.

[14] “Deleuze and Guattari compared the dominant Western model of thinking to the tree. This image refers not only to the literal shape of a tree (the seed is the cause, the tree the effect), but also -for instance- to the genealogical lineage attributed to ancestry in the family tree. …Thus the image of the tree expresses how the dominant model of Western thinking creates a single version of the truth… .” David Martin-Jones and Damian Sutton, Contemporary Thinkers Reframed: Deleuze Reframed (London: Tauris, 2008), pp. 3-4.

[15] The foundations of this anti-genealogy can be found in Michel Foucault’s text, “Nietzsche, Genealogy, History” (1971).

[16] “The map is open and connectable in all of its dimensions; it is detachable, reversible, susceptible to constant modification. It can be torn, reversed, adapted to any kind of mounting, reworked by an individual, group, or social formation. …A map has multiple entryways, as opposed to the tracing, which always comes back ‘to the same.'” Deleuze and Guattari 1987, pp. 12-13.

[17] Ibid., p. 21.

[18] Ibid., p. 2.

[19] Didi-Huberman continuously refers to the rhizome in his book, L’Image survivante, in which he explicitly connects the Mnemosyne Atlas with rhizomatic processes. See, for example: “Dialectique du temps qui n’a besoin ni du bien ni du mal, ni des debuts ni des fins pour exprimer sa impureté: faite de rhizomes, répétitions, symptômes.” (“Dialectics of time which do not require neither the good nor the bad, neither the beggining nor the end, to express their own impurity: make rhizomes, repetitions, symptoms.”) 2002, p. 112, (my translation). He expands this idea in Atlas. How to Carry the World on One’s Back?: “The atlas is guided only by changing and provisional principles, the ones that can make new relations appear inexhaustibly - far more numerous than the things themselves - between things and words that nothing seemed to have brought together before.” 2011, p. 16.

[20] Deleuze and Guattari (1987), p. 380.

[21] Michaud (2004), p. 235.

[22] Didi-Huberman (2011), p. 20.

[23] “Warburg broke down, deconstructed surreptitiously all epistemological modes in use… .” Didi-Huberman (2002), p. 24, (my translation).

[24] Abbas Niran (ed.), Mapping Michel Serres (Michigan: The University of Michigan Press, 2008), pp. 86-87.

[25] See Aby Warburg, The Renewal of Pagan Antiquity: Contributions to the Cultural History of the European Renaissance (Los Angeles: Getty Research Institute for the History of Art and the Humanities, 1999).

[26] “This way of thinking slices, dislocates, surprises, but it does not take a firm stand because of its experimental and provisional nature, because it emerged from a pure topical transformation so it considers itself recombinant, modifiable, always in movement, always on the way, always ‘at a crossroads.” Didi-Huberman (2011), p. 122, (my translation).

[27] Michel Serres describes the Black Box as follows: “Take a black box. To its left, or before it, there is the world. To its right, or after it, travelling along certain circuits, there is what we call information. The energy of things goes in: disturbances of the air, shocks and vibrations, heat, alcohol or ether salts, photons… . Information comes out, and even meaning. We do not always know where this box is located, nor how it alters what flows through it, nor which Sirens, Muses or Bacchantes are at work inside; it remains closed to us. …Before the box, the hard; after it, the soft.” Michel Serres, The Five Senses. A Philosophy of Mingled Bodies (London and New York: Continuum, 2008), p. 129.

[28] “The originality of Warburg’s approach lay precisely in the attempt to get through art at the mental image behind it, to question not only paintings but also literature, festivals, anything that might reflect the ideas these people had in their minds.” Ernst H. Gombrich, “Warburg Centenary Lecture," in Art History as Cultural History. Warburgs’ projects, ed. Richard Woodfield (Amsterdam: G+B Arts, 2001), pp. 33-54; ref. on p. 40.

[29] Michel Serres, The Parasite (Baltimore and London: The Johns Hopkins University Press, 1982), p. 46.

[30] Michel Serres, Angels. A Modern Myth (Paris, New York: Flammarion, 1995), p. 295.

[31] Michel Serres and Bruno Latour, Conversations on Science, Culture, and Time (Michigan: The University of Michigan Press, 1995), p. 127.

[32] Aby Warburg quoted in Ernst H. Gombrich, Aby Warburg. An intellectual biography, with a memoir on the history of the Library by F. Saxl (London: The Warburg Institute, 1970), p. 223.

[33] Serres, (1982), p. 42.

[34] See David Freedberg, Las máscaras de Aby Warburg (Spain: Sans Soleil, 2013).

[35] Serres, (1982), p. 42.

[36] The schizo in Deleuze and Guattari refers to the virtual undercurrent of desiring production in each of us that is actualized in neurotic “machines,” which is also the key of the central theory of the "body without organs." This is directly opposed to the idea of desire as lack as it is formulated by the Oedipus Complex. That is why Deleuze and Guattari propose schizo-analysis as a means of counteracting Freudian psychoanalysis. For further clarification, see Gilles Deleuze and Félix Guattari, Anti-Oedipus. Capitalism and Schizophrenia. (London: Continuum, 2004).

[37] Warburg’s clinical case has been carefully studied in Davide Stimilli (ed.), La curación infinita. Historia clínica de Aby Warburg (Buenos Aires: Adriana Hidalgo Editora, 2007).

[38] Warburg quoted in Gombrich (1970), p. 227.

[39] Warburg quoted in Didi-Huberman (2011), p. 177.

[40]Mnemosyne saved him from his madness, from the 'fleeting ideas' so well analyzed by his psychiatrist Ludwig Binswanger. But at the same time, his ideas continued to 'stream out’ uselessly, like dialectical images, from the shock of the assembling of particularities.” Ibid., p. 20.

[41] Gilles Deleuze and Félix Guattari, Anti-Oedipus. Capitalism and Schizophrenia. (London: Continuum, 2004), p. 15.

[42] “The fact that the configuration of images can always be changed around in the Mnemosyne Atlas is a sign in itself of the heuristic fecundity and the intrinsic madness of such a project.” Didi-Huberman (2011), p. 20.

[43] Ibid., p. 187.

[44] Warburg himself, neurotic from his early youth, was keen on collecting things. As Gombrich recalls, throughout his life Warburg obsessively kept copies of his paperwork and letters, he also collected stamps and even made a gigantic archive where he accumulated articles, journals, and daily press.

[45] Didi-Huberman (2011), p. 284.

[46] Some of these artists have been studied by Didi-Huberman, for example, Marcel Broothaers, Gerhard Richter, Christian Boltanski, Sol LeWitt, Sophie Calle, Hanne Darboven, Susan Hiller, On Kawara, Hanna Höch, André Malraux, and Hans Peter Feldmann. To these we could add Ydessa Hendeles, Walid Raad, Taryn Simon, Simon Evans, Andreas Seltzer & Dieter Hacker, Dora Maurer, Eva Koťátková, Fiona Tan, Hans Haacke, Juan del Junco, Richard Hawkins, Robbert Flick, Olafur Eliasson, Roni Horn, Arman, Mark Dion & Robert Williams, Damien Hirst, and Taryn Simon. See Suely Rolnik, “Furor de archivo," Revista electrónica Estudios Visuales, 7 (2010), 115-129.

For an insider account of this issue, see Anna Maria Guasch, Arte y archivo 1920-2010. Genealogías, tipologías y discontinuidades (Madrid: Akal, 2011) and Eivind Røssaak (ed.), The Archive in Motion: New Conceptions of the Archive in Contemporary Thought and New Media Practices (Oslo: Novus Press, 2010).

[47] See Michel Foucault, The archaeology of knowledge and The discourse of language (New York: Pantheon Books, 1972).

[48] See Boris Groys, Art Power (Cambridge, Massachusetts, London: The MIT Press, 2008), Boris Groys, On the New (London and New York: Verso, 2014), and Boris Groys, Under suspicion: a phenomenology of the media (New York: Columbia University Press, 2012).

[49] Michel Foucault, (1972), p. 128. By archive, Foucault does not mean here a set of documents which must be stored, kept  and preserved given their cultural importance, but more accurately the system of possibility of discourses.The archive fixes the boundaries of what might be said, the form according to which it can be said, the way in which it can be appropriated, etc. It is the archive that owes such a power.

[50] Jacques Derrida, Archive Fever: A Freudian Impression (Chicago and London: University of Chicago Press, 1996), p. 3.

[51] Giorgio Agamben, What is an apparatus? and other essays (Stanford: Stanford University Press, 2009), p. 12.

[52] See Salvatore Settis, Fritz Saxl, and Eric M. Warburg, Warburg Continuatus. Descripción de una biblioteca (Madrid: La Central, Museo Nacional Centro de Arte Reina Sofía, 2010).

[53] This could be seen as the "anarchival discourse" or "un-archival order" that has been theorized by Wolfgang Ernst, Digital Memory and the Archive (Minneapolis: The University of Minnesota Press, 2013).

[54] “The new atlas designs this world map,” Serres (1994), p. 128, (My translation).

[55] In this regard, Warburg could even be considered the precedent of the “cultural software," in the terms of Jack M. Balkin (Cultural Software: A Theory of Ideology. New Haven, Conn, London: Yale University Press, 1998), or the antecedent of the "digital archive,” as has been suggested by Franco Speroni (“El archivo post-textual: Aby Warburg y su Atlas de la memoria," Revista de Occidente, 381 (2013), 53-65).

[56] “How to capture, in the pages of this atlas - which are too fixed - these wonderfully agile maps?” Serres (1994), p. 275, (my translation).

[57] Didi-Huberman (2011), pp. 17-18.

[58] Ibid., p. 15.

[59] See Giorgio Agamben, “Aby Warburg and the Nameless Science” in Potentialities: collected essays in philosophy (Stanford: The Stanford University Press, 1999), 89-103.