2/17/2016 TRANSFORMATIONS Journal of Media & Culture

ISSN 1444­3775

2016 Issue No. 27 — Thing Theory, Material Culture, and Object-Oriented

Ontology
Thinking Things: Images of Thought and Thoughtful Images
By Chari Larsson

Soviet director and theorist Sergei Eisenstein famously asserted the possibilities of montage could
theoretically extend well beyond film to encompass all art forms (36). French philosopher and art
historian Georges Didi-Huberman has taken Eisenstein’s claim seriously, enigmatically proposing
“[m]ontage is the art of producing this form that thinks” (Images 138). Didi-Huberman’s
comments were in relation to a series of World War II archival photographs, but have evolved
over the past decade to form an important cornerstone in his ongoing investigation into images.
What, then, does it mean for an inanimate thing, an image, to think? My hypothesis here is that
Didi-Huberman is entering into direct dialogue with Gilles Deleuze’s critique of the philosophical
assumptions of what thinking is. This article will proceed in two sections. The first will examine
Didi-Huberman’s claims in relation to recent shifts across the humanities that have sought to
restore agency to images. The second section will examine Didi-Huberman’s response, arguing
that he turns to some of the great experimental projects of the early twentieth century, such as
Walter Benjamin’s literary montage The Arcades Project (1927–1940) and Aby Warburg’s
photographic collection Mnemosyne Atlas (1925–1929). The governing principles of montage, such
as juxtaposition, the assemblage of heterogeneous parts, and emphasis on the spectator’s
participation are well known. Perhaps, however, there is another way to speak of montage, when
it is employed as a mode of knowledge. Didi-Huberman identifies the epistemological capacity of
montage as a mechanism to free the image from the dogma of art-historical discourse. What is at
stake here is a model of representation that is no longer imitative, but capable of generating its
own theoretical and intellectual undertaking. By emphasising montage’s capacity to create new
meaning and generate new lines of thought, the image becomes a theoretical object, a thing that
“thinks.”

The notion that images are endowed with a life of their own has enjoyed increasing currency
amongst art historians and visual theorists in recent years. Loosely assembled under the rubric of
object-orientated ontology, theories as diverse as Bill Brown’s “thing theory,” W.J.T. Mitchell’s
“pictorial turn,” and Gottfried Boehm’s “iconic turn” have sought to restore agency to the image.
Keith Moxey has extended this line of thought, arguing images are endowed with presence.
Fatigued with the “linguistic turn,” and the idea that images are a language that can be “read,”
Moxey observes that images are “more appropriately encountered than interpreted” (132). In
France, Didi-Huberman is one theorist who has maintained an ongoing commitment to describing
the power of images. Emerging from the general critique of representation that was underway
during the 1980s and 1990s, Didi-Huberman has sought to articulate alternative frameworks for
understanding images beyond the idealist legacies bequeathed by traditional art history.

One of the lines of thought unifying this cross-disciplinary impulse is a reconsideration of the
subject-object relationship. The subject is bracketed and the object becomes the focus of enquiry.
As Bill Brown observes: “The story of objects asserting themselves as things, then, is the story of a
changed relation to the human subject and thus the story of how the thing really names less an
http://www.transformationsjournal.org/issues/27/02.shtml 1/10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

object than a particular subject-object relation” (4). The reconfigurement of the subject-object
relationship enjoys a long history in French philosophical, literary and psychoanalytical
traditions. ecall Jacques Lacan’s anecdote concerning the glimmering sardine can he saw floating
on the water as a young man. In his theorisation of the gaze in eminar I, Lacan reversed
dominant theories of vision, arguing that “[t]hings look at me” (109). The subject, no longer the
originating point of vision, is ensnared in the gaze of the object. Alternatively, consider the tactile
power assigned to objects in Sartre’s 1937 novel, a sea. The main protagonist, Antoine
oquentin, suffered from acute bouts of nausea as a response to the sudden awareness of being
touched by objects. oquentin describes his horror:

Objects ought not to to ch, since they are not alive. ou use them, you put them back
in place, you live among them: they are useful, nothing more. But they touch me, it’s
unbearable. I am afraid of entering in contact with them, just as if they were living
animals. (22)

The simple act of holding a pebble on the beach is acknowledged by oquentin as “sweet disgust.
How unpleasant it was ” (22).

If Lacan’s sardine can has the capacity to “look at me,” and objects are responsible for generating
nausea, we are not far away from Didi-Huberman’s proposition that images can think. As
Hanneke Grootenboer and Matthew Ancell have recently demonstrated, the notion of thinking
images occupies a privileged position in French thought, stretching from theorists, artists and art
historians historically and theoretically diverse as icholas Poussin, oland Barthes, Hubert
Damisch and Jacques Derrida. Didi-Huberman’s claim that montage is a thinking form
simultaneously contributes to this lineage, as well as breathing new life into old debates between
members of the Frankfurt School in the 1930s concerning the efficacy of representation. Georg
Luk cs famously condemned montage, relegating it to little more than a “one-dimensional
technique,” rapidly acceding to monotony. Luk cs writes: “The details may be dazzlingly
colourful in their diversity, but the whole will never be more than an unrelieved grey on grey.
After all, a puddle can never be more than dirty water, even though it may contain rainbow tints”
(43). How, then, can montage be anything more than a nostalgic revival of the goals and
ambitions of the historical avant-garde? To push the question even further, how is it possible to
imagine a contemporary relevancy for montage practices?

Against Luk cs’ condemnation, montage as a tool for revising art’s historiography has drawn
increasing attention by French scholars. Philippe Alain-Michaud has traced the theme of
movement in his reconsideration of German art historian Aby Warburg’s project. For Michaud,
Warburg’s work is best understood in cinematic terms, culminating in his great unfinished
montage experiment, Mnemosyne Atlas. Michaud argues that Warburg’s interest in movement
motifs in the enaissance was influenced by the growth and influence of the nascent film
technology in the 1920s and 1930s. eaching back further in art’s history, Giovanni areri draws
from Sergei Eisenstein’s “montage of attractions” to re-examine the work of seventeenth-century
sculptor and architect Gian Lorenzo Bernini. Traditional Bernini scholarship has emphasised the
holistic unity of sculpture, painting and architecture of Bernini’s oman chapels. areri departs
from this, instead working in an explicitly Einsteinian register, underscoring the juxtaposition
between Bernini’s forms, with the spectator left to reconstruct the heterogeneous components.

A somewhat different treatment of montage has been advanced by Didi-Huberman, who has
elected not to restrict his discussion to Soviet avant-garde cinema. Instead, Didi-Huberman
retrieves the general principles of the procedure itself, emphasising the shock or juxtaposition
between individual parts to generate new lines of thought. Out of the collision, something new is
created. At no point does Didi-Huberman get drawn into a detailed discussion of montage,
preferring instead to distil its salient features as general operating principles. As a result he places
himself at risk of oversimplifying the concept, ignoring the form’s individual nuances,
complexities and mutations as it was taken up in other geographies. The strategy, however,
http://www.transformationsjournal.org/issues/27/02.shtml 2/10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

affords Didi-Huberman an enormous degree of freedom. He looks and finds montage practices
where we least expect: the collision created in the layout of text and print in Georges Bataille’s
journal oc ments (1929–1930) (La essem lance In orme) Bertolt Brecht’s press images in
riegs i el (1955) ( and les images) and Andr Malraux’s Le M s e imaginaire (1952–1954)
(L al m).

Against a purely aesthetic understanding of montage, Didi-Huberman is instead drawing close to

Deleuze’s famous conclusion at the end of The Time Image: “ inema’s concepts are not given in
cinema. And yet they are cinema’s concepts, not theories about cinema” (269). inema does not
illustrate philosophy, but produces its own philosophical project. inema is understood by
Deleuze here less as a spectacle, but as a form of thought. inema thinks in images, but these
images are not imitative or derivative. Instead, they are generative, initiating their own thinking
forms.

The origins of Deleuze’s claims may be located in his 1968 i erence and e etition. For Deleuze,
the task assigned to philosophy was to overcome the image preceding thought, or thought based
on resemblance, recognition and identity. Before we even begin thinking, our thought is already
predetermined by a series of pre-existing postulates, or assumptions. This is Deleuze’s “image of
thought” and formulated in the following terms: “We live with a particular image of thought, that
is to say, before we begin to think, we have a vague idea of what it means to think, its means and
its ends” ( esert Islands 139). Deleuze argues that lying at the origin of all thought is a pre-existing
image serving as the fundamental ground for thought. He adds his voice here to a long strain in
continental philosophy: the moment we begin to think, we have already made an assumption of
what we call thinking. The artesian method of doubt, for instance, was designed to ensure there
were no presuppositions left. Descartes’s cogito, “I think,” is preceded by assumptions of “I,”
“think” and “am” or self, subjectivity and being. These assumptions are predicated on the notion
that the human subject is universal (everybody knows what it means to think), stable and centred.
Deleuze’s point is that whilst philosophy has taken care to eliminate preceding suppositions, it
has never been entirely successful in escaping them completely. The “I” of the humanist subject
can no longer be identified as the foundational source for all knowledge.

Deleuze’s attack on the subject as the origin of knowledge points to a decentering of subjectivity
that corresponds with his contemporaries. In a similar vein, Michel Foucault recognised that the
artesian “I think” no longer equates to the “I am.” Foucault argued that the “modern cogito”
would not necessarily be human:

In this form, the cogito will not therefore be the sudden and illuminating discovery
that all thought is thought, but the constantly renewed interrogation as to how
thought can reside elsewhere than here, and yet so very close to itself how it can e in
the forms of non-thinking. (324)

Foucault’s “death of man” thesis points to a shift or reorientation from the subject to the object.
This has important consequences for later scholars like Didi-Huberman, who are working
through the wake of Deleuze and Foucault’s respective legacies. Didi-Huberman recognised that
traditional art history, practiced in the humanist tradition, was no longer viable. Art history could
no longer lay claim to a rational, all-knowing mode of subjectivity and dominance over the image
could no longer be comfortably maintained.

eturning now to i erence and e etition, it is an image that resides at the heart of philosophy’s
residual subjectivity, an “[i]mage in general which constitutes the subjective presupposition of
philosophy as a whole” (131). The image is understood by Deleuze in representational terms. To
overcome the image of thought, therefore, the image must be destroyed. This is Deleuze’s
iconoclasm and the task he assigns to philosophy: to displace the idea of what thinking looks like.
Deleuze continues:

http://www.transformationsjournal.org/issues/27/02.shtml /10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

[c]onceptual philosophical thought has as its implicit presupposition a pre-

philosophical and natural Image of thought, borrowed from the pure element of
common sense. According to this image, thought has an affinity with the true it
formally possesses the true and materially wants the true. It is in terms of this image
that everybody knows and is presumed to know what it means to think. (131)

The image of thought relies on a series of pre-existing “postulates” that exist prior to thinking.
Therefore philosophy must generate “[a] new image of thought – or rather, a liberation of thought
from those images which imprison it” (Deleuze xvi–xvii). How, then, to free thought from what
Deleuze designates as the “dogmatic” and “orthodox” image that occurs prior to thinking (131)?
Deleuze gives us an important clue, drawing an analogy with modernist painting’s passage
towards abstraction: “The theory of thought is like painting: it needs that revolution which took
art from representation to abstraction. This is the aim of a theory of thought without image” (276).
In painting’s withdrawal from the demands of mimesis, it was gradually freed from the strictures
of imitation. For Deleuze, this shift encapsulated philosophy’s task of thinking prior to the image,
language and representation.

Deleuze’s arguments hold enormous implications for Didi-Huberman. If art history is organised
via a series of pre-existing conventions, these conventions predetermine outcomes. In an
interview with obert Maggiori, Didi-Huberman described this circularity:

rosso modo, I am interested in the image in that it moves the foundations of

representation, that is to say our idea of representation... What often fascinates me is
the way an image is capable of inventing... configurations that, literally, defy thought.
This is why I less have the impression of projecting a philosophical gaze’ onto
images, than returning the power to the image – if it is strong – to upset, thus literally
recommencing ith tho ght itself, on all levels.

There are two crucial points to recognise in this passage. The first is Didi-Huberman’s desire to
displace pre-existing ideas of what representation can possibly be. From his earliest writings in
the 1980s, Didi-Huberman has sought to pursue a concept of representation departing from
classic understandings of mimesis. For art history, this has been the idealism underwriting the
discipline since asari and codified by Erwin Panofsky via Kant (Con ronting Images). Idealism is a
notoriously overdetermined term in the history of philosophy. In its broadest sense, idealism
privileges the notion that ideas are primary, while non-ideas, such as physical and material
things, are secondary. We can easily recognise this line of thought descending from Plato who
derided images as secondary and derivative as opposed to higher Forms. The second point in the
conversation with Maggiori reminds us that Didi-Huberman is not simply trying to import a pre-
existing “philosophical gaze” and apply it to images. Instead, he is attempting to recommence
with thought itself, what it means to think free from disciplinary conventions that predetermine
outcomes. To reframe this in Deleuzian terms: disciplinary conventions blind the art historian to
an actual image. It is therefore necessary to free the image from the discourse preceding it.

This is not a matter of simply opposing words and images, a long tradition that has been explored
in various ways by art historians and theorists. W.J.T. Mitchell famously mapped the tension
between words and images, arguing “The history of culture is in part the story of a protracted
struggle for dominance between pictorial and linguistic signs, each claiming for itself certain
proprietary rights on a “nature” to which only it has access” (43). Mitchell did not attempt to
neutralise or resolve the relationship, instead elected to historicise the conflict, charting the
historical struggle for dominance between word and image. Alternatively, Svetlana Alpers
approached the issue in Foucauldian terms of knowledge and power: “For as scholars, art
historians all too often see themselves as being in pursuit of knowledge without recognizing how
they themselves are the makers of that knowledge” (6). Alpers recognised the unequal
distribution of power exerted by the language-based discipline over images. It would be a
mistake, however, to position Didi-Huberman’s work in terms of this intellectual lineage. or
http://www.transformationsjournal.org/issues/27/02.shtml /10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

may it be measured as a deconstructionist gesture aimed at recovering the image from its
subordination to text. Instead, Didi-Huberman tries to approach the image free from the
conventions that determine the discourse of art history. For Didi-Huberman, montage’s potential
lies in reimagining the discipline, beyond its linguistic and logocentric biases. As he puts it:
“Beyond the model of the graphic or cinematographic process as such, we ought to conceive of
montage as doing for the field of images what signifying difference did for the field of language in
the post-Saussurian conception” (Images 121). To be clear, what is at stake here for Didi-
Huberman is an epistemological shift on par with structuralism’s impact on the life sciences in the
1950s and 1960s.

How, then, does a deeply visual tradition like art history rethink its dependence on mimetic
representation? If idealism implicitly determines art history’s methodological approach, how may
the image be perceived prior to its theorisation? How to overcome the image of thought?
Deleuze’s answer is the encounter. He writes “[s]omething in the world forces us to think. This
something is an object not of recognition but a fundamental enco nter” ( i erence 139). What is
this encounter Deleuze speaks of? A genuine encounter is a shock or jolt, forcing a rupture and
breaking with convention. As opposed to the representation which reinforces the same,
conventional habits of thought, the affective force of the encounter forces us to think otherwise,
and develop a new image of thought. The conjunction between the shock of the encounter and
montage is made clear in Deleuze’s Cinema books. Deleuze writes “Montage is in thought the
intellectual process’ itself, or that which, under the shock, thinks the shock” (Time Image 153).
Here, Deleuze was able to crystallise this new non-representational image of thought in terms of
cinema. Deleuze’s anti-representationalism is given full expression in the image’s capacity to
generate new lines of thought through the shock of montage. Deleuze describes two moments
here. The first is the shock of montage’s juxtaposition that jolts the spectator: “The complete
circuit thus includes the sensory shock which raises us from the images to conscious thought, then
the thinking in figures which takes us back to the images and gives us an affective shock again”
(156). The second moment moves in the opposite direction, back to the image itself, “from the
concept to the affect, or which returns from thought to the image” (154). Deleuze relocates the
origin of thought from the spectator to image. Images no longer passively “represent” or
“imitate” philosophical thought, but are capable of generating their own theoretical project. We
might say cinema consists of images, but images are not simply passive reflections. Images are
what cinema thinks with.

Deleuze’s critique of the image of thought holds important implications for what philosophy can
be. Philosophy now demands a new approach that is non-dogmatic and non-systematic. For Didi-
Huberman, montage is an exemplary form of intervention. Firstly, it is anti-mimetic, therefore
aligned with his critique of idealist formulations of representation. Secondly, montage is
discontinuous. It can jolt, shock, reconfigure and reassemble. Montage cuts through existing
knowledge structures, refusing to allow anything to settle. This extends to mental images, with
Didi-Huberman identifying a montage impulse in harles Baudelaire’s description of the
imagination. Writing with regard to Edgar Poe, Baudelaire observed:

Imagination is not fantasy, nor is it sensibility, difficult though it would be to

conceive of an imaginative man who was not sensitive. Imagination is a virtually
divine faculty that apprehends immediately, by means lying outside philosophical
methods, the intimate and secret relations of things, the correspondences and
analogies. (199)

The behaviour of Baudelaire’s “intimate and secret relations”, “correspondences” and “analogies”
is akin to montage. Montage produces, not simply reproduces, an existing idea. Montage allows
for the recommencement of thought.

It is the montage practices of Walter Benjamin and Aby Warburg where Didi-Huberman detects a
style of thought deliberately eschewing recognition through representation. During the 1930s,
http://www.transformationsjournal.org/issues/27/02.shtml /10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

Benjamin took up the theme of montage in respect to his discussions on Marcel Proust and his
friend Bertolt Brecht. [1] His commitment to montage is given its most complete expression in his
monumental Passagen er The Arcades Project onstructed in the late nineteenth-century out of
iron and glass, the Parisian arcades were a visible representation of the industrial revolution and
considered the pinnacle of luxury and leisure culture. Despite this, the arcades had long been in
physical decline. By the time of Benjamin’s permanent relocation and exile to Paris in 1933, the
arcades had become fusty relics, symbols of the aspirations of a tired modernity. onsisting of
more than 900 pages, Arcades Project is a literary montage, a collage of incomplete aphorisms,
historical quotations and Benjamin’s own commentary. Benjamin’s sources are many and varied
as he assembles a dizzying array of diverse fragments ranging from nineteenth-century tourist
guides, extracts of Baudelaire’s poetry and architect Haussmann’s memoires. As a literary
montage, the reader is provided with an inexhaustible series of combinations and opportunities
for potential assembly. As a result, the act of reading the text is a restless, fragmentary experience
of competing voices and sources of material.

One of Benjamin’s most important operative concepts for Didi-Huberman is the “dialectical
image.” Didi-Huberman pulls the dialectical image in the direction of Deleuze, animating a nexus
between the two that privileges the shock of an encounter. The dialectical image is not understood
by Didi-Huberman in representational terms, as a traditional “image”, or “picture”. Instead it is
imagined as a “dazzling shock” ( e ant le tem s 241). Didi-Huberman is building on Benjamin’s
own comments that present the dialectical image as participating in a certain “caesura in the
movement of thought” (475). Benjamin repeatedly emphasises the disruption performed by the
dialectical image: “Where thinking comes to a standstill in a constellation saturated with tensions
– there the dialectical image appears” (475). eturning to Deleuze’s challenge issued in i erence
and e etition, Benjamin’s dialectical image does not depend on memory or recognition, but is the
result of a “fundamental encounter” (139). This encounter is formulated as a jolt or disturbance.

Benjamin demands an active participant – at no point is the reader permitted to settle into a
passive or complacent state. This is, of course, entirely Benjamin’s point. The montage was
designed to jolt the reader from their anaesthetisation. By rejecting the form of traditional
chronological narrative, Walter Benjamin demonstrates the power of montage to reconfigure and
reassemble material anew. This strategy suggests a clear shift from spectator to text. As Benjamin
describes it: “Method of this project: literary montage. I needn’t say anything. Merely show” (460).
o longer imitative, Benjamin’s quotations collide, producing new associations and lines of
thought for the reader.

ontemporaneous with Benjamin’s project, the Mnemosyne Atlas was a famous collection of
images that Aby Warburg was working on in the final years of his life (Fig. 1). In Warburg’s
experiments with montage, Didi-Huberman detects an anti-idealist sensibility eschewing
recognition through imitation.The Mnemosyne Atlas was conceived by Warburg in 1925 and
remained unfinished at his death in 1929. It consisted of about one thousand images pinned over
seventy nine panels covered in black fabric. Dismantled with the Warburg library’s shift from
Hamburg to London, it exists today only as a series of photographs. By mapping out his ideas in
arrangements of images, Mnemosyne Atlas allowed Warburg the flexibility to pursue his ideas
visually with a freedom not afforded by the written text. Mnemosyne gives visual form to
Warburg’s notion of Pathos ormel (pathos formulas) and the idea that certain motifs and gestures
survive through the history of images. By drawing together constellations of images, Warburg
was able to trace the recurrence and return of gestures, drawing new connections and relations
between them.

http://www.transformationsjournal.org/issues/27/02.shtml 6/10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

ig. 1 Aby Warburg, ilderatlas Mnemosyne, 1925-1929. Panel 79. ( opyright: eproduced with
permission from The Warburg Institute, niversity of London.)

Didi-Huberman is quick to differentiate the atlas from its cousins, the encyclopaedia and the
archive. The atlas does not behave systematically like the encyclopaedia or exhaustively like the
archive. The atlas “chooses.” From the 25,000 photographic reproductions available in Warburg’s
personal image library, the Mnemosyne Atlas consisted only of 1,000 carefully selected images. An
atlas refuses to be “read” in the sense one would read a novel or an essay from first page to last.
Its point of entry is random, and the pathway through not linear or predetermined. ejecting a
strict chronological structure, the image configurations could be continually changed around,
altered, photographed and dismembered, only for another series of images to be repositioned,
explored, documented and dismantled.

Warburg’s Mnemosyne serves as a springboard for how new ideas are created through montage
and juxtaposition. The form of montage in Mnemosyne is more than the result of an artistic
process, but “a visual form of knowledge” (Atlas 14). Didi-Huberman resists the temptation to
construct an image by image reconstruction, instead accentuating the non-axiomatic openness of
http://www.transformationsjournal.org/issues/27/02.shtml 7/10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

the images and their ability to be constantly rearranged. In its capacity to conjure unforseen
relations and give form to previously unnoticed affinities, Didi-Huberman argues montage is a
“procedure capable of putting into movement new thought spaces’” (182). The montage effect
does not illustrate pre-existing ideas and concepts, but generates new relationships between the
constellations of images. Akin to Benjamin’s Arcades Project, Mnemosyne ensures there are no
predictable or stable outcomes. Instead, it asks the spectator to proceed heuristically, eschewing
pre-determined rules.

The montage process is not imitative, but productive and generative. o longer dependent on the
idealism of mimetic forms, montage brings Didi-Huberman’s work directly into dialogue with
Deleuze’s claim that cinema generates its own ideas. Mnemosyne no longer “represents” or
“imitates” pre-existing ideas, but allows new relations and unforseen connections to flourish. The
images do not simply reproduce, but rearrange and reorganise in ways that render them new. By
generating new relations between images, Mnemosyne posits a shift in the origin of knowledge
and disrupts the traditional subject, object relations. The spectator is no longer the privileged
source of meaning. Instead, it is through the juxtapositions and collisions of the montage process
that fresh lines of thought are generated by the images themselves.

We can recognise here Didi-Huberman’s commitment to Deleuze’s call to overcoming the image
of thought:

Warburg expresses in his atlas a fundamental com le ity – of an anthropological order

– that was neither to be synthesised (through a unifying concept) nor described
exhaustibly (in an integral archive), nor classed from A to (in a dictionary).
Warburg’s atlas does not perform according to an pre-axiomatic structure. It is not
predictable in its outcome. (19)

The atlas is dynamic. It is through the montage and collision of images that previously
unanticipated dialogues and relationships are produced. In this way, the atlas intersects with
Deleuze’s claim that cinema generates its own concepts. This results in a shift or displacement in
the source or origin of knowledge. o longer located in the subject, the atlas itself is a visual form
of knowledge. The atlas “thinks,” performing its own philosophical and epistemological project.

Firmly eschewing the notion of the image as a passive reflection, or illustration of pre-existing
philosophical concepts, Gilles Deleuze privileges the cinematic image as a preferred mode for
overcoming the image preceding thought. o longer understood in mimetic or imitative terms,
the image is capable of generating its own thought processes. This essay has traced the trajectory
of Didi-Huberman’s invocation of montage as a general disruption to mimetic understandings of
representation, extending to the discipline of art history itself. By way of two case studies, The
Arcades Project and the Mnemosyne Atlas Didi-Huberman responds to Deleuze’s challenge,
identifying the epistemological capability of montage as a mechanism to free thought from the
image that precedes it. Benjamin and Warburg’s montage practices produce cinematic modes of
thinking. Performing their own philosophical and epistemological projects, these great
experimental montage projects eschew the idealism that has implicitly informed the discipline of
art history. The process signals a radical displacement from the spectator to the image as the
source of knowledge.

r Chari arsson has recently completed her PhD examining the work of French
philosopher and art historian Georges Didi-Huberman. Her research focuses on theories of
images and representation, specifically in the areas of art historiography.

http://www.transformationsjournal.org/issues/27/02.shtml /10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

ndnotes

1. See, in particular, Benjamin’s discussion of Proust’s m moire in olontaire (Image o Pro st)
Benjamin underscores the disruptive nature of the involuntary experience, akin to the shock
of montage, writing: “The true reader of Proust is constantly jarred by small shocks” (208).
On the montage impulse in Brechtian theatre, see Benjamin’s 1934 essay “The Author as
Producer” where Benjamin aligns epic theatre with what he argues are the most advanced
modernist techniques, such as John Heartfield’s photomontages and Soviet film and radio.

orks Cited

Alpers, Svetlana. “Is Art History?” aedal s 106.3 (1977): 1–13. Print.

Ancell, Matthew. “Credo rgo m: Faith, Blindness, and Pictorial Logic in Derrida s Memoirs
of the Blind.” ord Art o rnal 37.2 (2014): 193–210. Print.

Baudelaire, harles. “Further otes on Edgar Poe.” elected ritings on Art and Artists
ambridge: ambridge P, 1981: 188–208. Print.

Benjamin, Walter. “The Author as Producer.” nderstanding recht. Trans. Anna Bostock.
London: erso, 1998. 85–101. Print.

---. The Arcades Project. Trans. olf Tiedemann. ambridge, Mass.: Belknap, 1999. Print.

---. “The Image of Proust” Trans. Harry ohn. Ill minations. Ed. Hannah Arendt. ew ork:
Schocken, 2007. 201–15. Print.

Boehm, Gottfried and W. J. T. Mitchell. “Pictorial ersus Iconic Turn: Two Letters.” C lt re
Theory and Criti e 50.2 (2009): 103–21. Print.

Brown, Bill. “Thing Theory.” Critical In iry 28.1 (2001): 1–22. Print.

areri, Giovanni. ernini lights o Lo e the Art o e otion. Trans. Linda Lappin. hicago: of
hicago P, 1995. Print.

Deleuze, Gilles. Cinema The Time Image. Trans. Hugh Galeta and obert Tomlinson. ew
ork: ontinuum, 1989. Print.

---. i erence and e etition. Trans. Paul Patton. ew ork: olumbia P, 1994. Print.

---. esert Islands and ther Te ts – . Trans. Michael Taormina. ambridge, Mass.: MIT
Press, 2004. Print.

Didi-Huberman, Georges. La ressem lance in orme o le gai sa oir is el selon eorges ataille.
Paris: Macula, 1995. Print.

---. “Des Gammes Anachroniques (Entretien Avec obert Maggiori).” Li ration 23 ovember
2000. Web.

---. e ant le tem s istoire de l art et anachronisme des images. Paris: Les ditions de Minuit,
2000. Print.

---. Con ronting Images estioning the nds o a Certain istory o Art. Trans. John Goodman.
niversity Park: Pennsylvania State P, 2005. Print.

http://www.transformationsjournal.org/issues/27/02.shtml /10
2/17/2016 TRANSFORMATIONS Journal of Media & Culture

---. Images in ite o All o r Photogra hs rom A sch it . Trans. Shane B. Lillis. hicago: of
hicago P, 2008. Print.

---. and les images rennent osition. Paris: Les ditions de Minuit, 2009. Print.

---. Atlas o to Carry the orld on ne s ac Madrid: Museo acional entro de Arte eina
Sofia, 2011. Print.

---. L al m de l art l art l o e d “M s e Imaginaire.” Paris: Hazan: Mus e du Louvre, 2013.

Print.

Eisenstein, Sergei. “Word and Image.” Trans. Jay Leyda. The ilm ense. leveland: Meridian,
1957. 3–65. Print.

Foucault, Michel. The rder o Things An Archaeology o the man ciences. intage Books ed.
ew ork: andom House, 1971. Print.

Grootenboer, Hanneke. “The Pensive Image: On Thought in Jan an Huysum’s Still Life
Paintings.” ord Art o rnal 34.1 (2011): 13–30. Print.

Lacan, Jacques. The o r ndamental Conce ts o Psychoanalysis. Trans. Alan Sheridan. ew

ork: W.W. orton, 1998. Print.

Luk cs, Georg. “ ealism in the Balance.” Aesthetics and Politics. London: LB, 1977. 28–59.
Print.

Michaud, Philippe-Alain. A y ar rg and the Image in Motion. Trans. Sophie Hawkes. ew

ork: one, 2004. Print.

Mitchell, W.J.T. Iconology Image Te t Ideology. hicago: of hicago P, 1986. Print.

---. Pict re Theory ssays on er al and is al e resentation. hicago: of hicago P, 1994.

Print.

Moxey, Keith. “ isual Studies and the Iconic Turn.” o rnal o is al C lt re 7.2 (2008): 131–46.
Print.

Sartre, Jean-Paul. a sea. Trans. obert Baldick. London: Penguin Modern lassics, 2000.
Print.

http://www.transformationsjournal.org/issues/27/02.shtml 10/10