POP 8 (1+2) pp.

171–182 Intellect Limited 2017

Philosophy of Photography
Volume 8 Numbers 1 & 2
© 2017 Intellect Ltd Article. English language. doi: 10.1386/pop.8.1-2.171_1

Johannes von Müller

The Warburg Institute, London

Metadata: New perspectives on

Aby Warburg’s ‘critical iconology’

Keywords Abstract
iconology The article reassesses the renown ‘critical iconology’ practiced by Aby Warburg and confronts it with recent
digital art history theories regarding metadata and the curatorship of information. A close examination of Robert Kitaj’s
digital humanities portrait Warburg as a Maenad (1961–62) affirms Warburg’s ‘combinatory experiments’ (Christopher D.
cultural history Johnson) to be a main characteristic of his practice. Warburg seems to have operated by applying what can
Aby Warburg now be described as metadata to the processes of transmission of culture he was accessing in form of migrat-
post internet art ing images. Further, the insights he was able to gain, namely his famous notion of the ‘Pathosformel’, may
have to be considered as a specific form of metadata too. This is of particular interest acknowledging that in
the digital world the migration of images takes mainly place in the realm of metadata. Hence ideas put
forward by Warburg are able contribute to a deeper understanding of such phenomena.

171
Johannes von Müller

‘Warburg as Maenad’
In 1958, most likely inspired by lectures given by the art historian Edgar Wind in Oxford (see Engel
2014: 104; Chaney 2013: 11–13), R. B. Kitaj painted a portrait of Aby Warburg, whose assistant Wind
had been in Hamburg. More of a study than a finished painting, it was soon to be followed by a
second, more ambitious portrait, presenting the cultural theorist in a quite unusual way, depicting
the acclaimed founding father of the iconological method (Schmidt 1993) as a maenad. Its flamboy-
ant colours – the orange framing of the red backdrop, the pink flesh, the red moustache and, not to
forget, the purple head – match the distorted position of the figure: standing with most of the weight
on one leg while bending the other, the back arched in an almost unnatural manner, the left arm
held up, the right hanging down. These details do indeed allow the depicted figure to be identified
as a maenad. Its classical origin as well as its ecstatic expression link it without question to the work
of Warburg (Chaney 2013: 13–14). This might slightly alleviate the surprise of finding a masculine
head with a moustache and other facial features of Warburg on a clearly female body.
The figure of the maenad quotes a classical form. By doing so it, takes on a key point in Warburg’s
thinking, the survival of the pagan antiquity in European art, its types moving through space and
time (see Didi-Huberman 2016). Being further an expression of a state of ecstasy, the dancing
maenad – a theme famously studied by Warburg (Figure 1) – represents one particular aspect of the
dynamics that, according to Warburg, act as driving forces of the aforementioned movement. An
aspect that is reinforced, moreover, by the androgyny accomplished by placing the thinker’s head on
the raving woman’s body, quite literally embodying the bipolarity of Warburg’s thinking. The latter is
even verbalized and simultaneously transferred to Warburg’s personal condition by the slip of paper
glued on to the screen to the maenad’s right. In the artist’s handwriting, a paragraph taken from
Gertrud Bing’s essay on Fritz Saxl is quoted. It reports on Warburg’s breakdown following the First
World War and the study on Luther and Melachthon that preceded this event:

Warburg had foreseen the outcome of the war from the beginning, and throughout its course,
watches with growing anxiety every bad omen of political, moral and intellectual decline. In
the autumn of 1918, when the world around him fell to pieces, he broke down – Just before
and during the war he had been occupied with the study of a historical period also filled
with forebodings of catastrophe: he had made a study of Luther’s and Melachthon’s attitude
towards astrology and portents through the imagery found in prognostications, calendars and
the reformers’ letters and scathing criticism.
(quoted by Deppner 1990: 236–37)

Warburg as Maenad refers to Warburg’s mindset in figurative and even textual elements. The painting
can be read as a pictorial commentary. Martin Roman Deppner already described it as such in 1990.

172 Philosophy of Photography

 Metadata

Figure 1: Aby Warburg, the Mnemosyne Atlas, Panel 6, 1929, Courtesy of the Warburg Institute, London.

[Link] 173
Johannes von Müller

Deppner is certain that the assembled image and text sources, all fairly easy to identify, suggest an
understanding of the painting as a loose-leaf system, first accumulated by applying Warburg as a
keyword and then transferred into image form; an operation that was possible due to some knowl-
edge of Warburg’s professional and personal life, even though, Deppner concludes, the painting
lacks a deeper understanding of Warburg’s methodology (1990: 238). Nevertheless Kitaj, by indexing
and commenting on Warburg’s scholarly enterprise, operates in the structural sphere of what may be
referred to as metadata. In doing so, the artist seems to have intuitively drawn on a vital element of
Warburg’s oeuvre – evoking it in its various details but even more importantly re-enacting it as a
whole by the chosen modus operandi.
This article poses the questions whether Warburg’s cultural–analytical approach to art historical
sources can be described in terms of data and metadata and what sort of implications this might
have; questions that touch on the steadily evolving field of digital art history (cf. Manovich 2015).
Nonetheless, they point into different directions than preceding enquiries of the material in ques-
tion; e.g. Stefka Hristova’s analysis of Warburg’s Mnemosyne Atlas focusing on its significance regard-
ing the contemporary praxis of media visualization (2016). The aim is to describe and analyse
processes of cultural transmission taking place in various forms of the movement of images; not
having to rely on traditional concepts of art history may open up new possibilities of understanding
these processes. The article proceeds in three sections: it will give a definition of metadata, then
examine Warburg’s work against this backdrop, focussing on the function of the photographic image,
and, eventually, turn to aspects of metadata in a contemporary context, assessing metadata’s share in
the migration of images in the digital era on the basis of two examples of works of art.

Metadata
Metadata, according to the recently proposed definition by Richard Gartner,

is a human construct and not found in nature. The shape of metadata is designed by human
beings for a particular purpose or to solve a particular problem, and the form it takes is indel-
ibly stamped with its origins. There is nothing objective about metadata: it always makes a
statement about the world, and this statement is subjective in what it includes, what it omits,
where it draws its boundaries and in the terms it uses to describe it.
(2016: 4)

The example Gartner gives is a globe: ‘a simple metadata representation of our own earth’. He continues:

[w]e could think of our planet as an immense body of data; its physical features are too
complex and too large a scale for us to grasp them from our observations of the small part

174   Philosophy of Photography

 Metadata

we experience every day. To make sense of the world we can (make) use of an object such as
a globe which abstracts its important features to make them comprehensible to us. Although
some features found on a globe mirror those found in nature […], many of the details on its
surface add human-created metadata to these representations of physical reality. […] All of
these features are added for a purpose: the lines of longitude and latitude provide a useful
grid for locating points on the Earth’s surface, and the political boundaries reflect a human-
made division of the planet […]. The designer of this artefact has been selective about what
to include and what to omit: equally valid representations may omit political boundaries and
concentrate on the physical or environmental features of the planet.
(Gartner 2016: 4–5)

By applying a carefully chosen selection of human-made parameters or categories, a globe gathers

information about the ‘immense body of data’ that is our world, thus allowing us to experience
something that otherwise would remain incomprehensible. These parameters or categories, this
information in itself is, again, data. Hence, metadata is commonly described as data about data. This
data about data, just like in the example of the globe, is what actually enables human agents to
engage with data. Or to quote Gartner once more:

[m]etadata exists for a reason and that reason lies fundamentally in the limitations of the
human brain. Amazing though it is in many ways, the brain is restricted in the amount of
information it can store and retrieve accurately. […] Once we move beyond the confines of a
single brain, the need for metadata rapidly becomes more pressing.

Moving ‘beyond the confines of a single brain’, Gartner speaks of communities and succeeding
generations who rely on systems for storing and passing on knowledge. For this, metadata is indis-
pensable. But it does not only allow ‘a single fragment of information to be stored away and found
again. It enables these to be linked together to form knowledge and for this knowledge to be consol-
idated into what we understand as culture’ (Gartner 2016). Gartner goes even further, for

culture is always evolving but this evolution relies on the preservation and transmission of its
earlier manifestations […]. To enable this to happen, a culture has to be curated […]. Curation
involves finding the constituents of culture, choosing which one are important, making
connections between them, describing and adding context to them so that they can be under-
stood and sharing them […]. Going through these steps ensures above all that culture can be
[…] transmitted between generations.
(2016: 13–14)

[Link] 175
Johannes von Müller

Metadata allows for the transmission of culture through space and time. The production of meta-
data has therefore a history much longer than that of the term itself. The latter was coined in the
late 1960s within information and computer sciences, originally used in the context of program-
ming languages (Bagley 1968: 26). Metadata is commonly divided into three different types: (1)
descriptive, (2) administrative and (3) structural metadata. (1) Descriptive metadata enables the
finding of data, e.g. an author’s name or a book’s title which help locate it in a library. (2)
Administrative metadata sustains the actual system in which data can be found; in the aforemen-
tioned case of a library this is provided by its classification. (3) Structural metadata acts in bring-
ing ‘together simple components into something larger that has meaning’ – like the paging of a
book (Gartner 2016: 8–9).

Metadata and Warburg

Before moving on to the question of metadata in Warburg’s work and returning briefly to Kitaj’s
painting Warburg as Maenad, it is now possible to identify the various types of metadata in the
artwork itself. To speak of structural metadata in relation to the painting is making use of that
term as a mere metaphor for the artistic act. Administrative metadata could actually refer to the
systems the artist most likely accessed to assemble the information that was needed for the
artwork. Descriptive metadata, however, is the type of metadata most present in the painting.
Not only because the latter can be read as a commentary as suggested by Deppner (1990) but
also because it relies essentially on descriptive metadata, since the artist applied Warburg as a
keyword: the text fragment by Bing is a statement about Warburg, the figure of the maenad
stands for Warburg’s work and the facial features are borrowed from a portrait filed under
Warburg’s name.
Girded by a rectangular frame, the face, as a second image within the painting immediately
evokes its feature’s prototype: the famous picture showing Warburg next to a member of the Hopi
tribe, taken during his journey to America in 1896 (Figure 2). Within the painting the figure of
Warburg therefore achieves the most articulated presence in a form that reveals itself as a fragment,
a quotation that points back to its origin: a photographic image. This is most noteworthy, since the
photographic image is not only the source of Warburg’s facial features – the very site of the survival
of his personal image – but functioned as his central tool. In his famous lecture on astrological
imagery in Palazzo Schifanoia which he gave in Rome in 1912, he introduced for the first time his
new method of, as he called it, ‘critical iconology’, a moment now commonly regarded as the birth
hour of iconology as an art historical practice (Didi-Huberman 2004: 8). Warburg spoke of this prac-
tice informed by the technological facilities that it relied on: it enabled him to ‘illuminate one single
obscurity’, to ‘cast light on great and universal evolutionary processes’ or to catch the item in

176   Philosophy of Photography

 Metadata

Figure 2: Warburg during his journey to America, 1895–96, Courtesy of the Warburg Institute, London.

[Link]   177
Johannes von Müller

question in ‘a cinematographic spotlight’ (Warburg 1999a: 585–86). Being his central tool, the photo-
graphic image also represents the most obvious case of use of metadata in Warburg’s work.
Warburg did not deal with frescos, paintings, sculptures and other material products of the
many branches of the visual arts. They were not the matter of his primary interest. He dealt with the
pictorial data carried by these works of art and other kinds of media. He may have seen frescos,
paintings, etc., but he perceived images. He did not interact with works of art, but with images.
Only in the form of their photographic reproductions did the former actually become accessible to
Warburg’s examination. The photographic image therefore formed one of the conditions of
Warburg’s enterprise. The expression of an assumedly universal state of the human psyche being
torn between emotional and intellectual engagement with the world was to him not specifically
artistic but generally visual, brought to light by an obligatory process of externalization. Hence,
already the product of a movement in itself, the pictorial data then kept moving on, travelling on
‘pathways’, as Warburg called them, through space and time, through: cultural history. This latter
motion could be traced and then studied by Warburg only thanks to yet another movement, the
transfer of the pictorial data from one medium (e.g. sarcophagi, tapestries, prints) onto another: the
photographic picture. It was only through the photographic picture that Warburg was able to
assemble the data in such a way as to enable him to gain insights regarding the complex processes
outline above.
Warburg did so by selecting images – including some, omitting others – a process which took
place on the basis of criteria defined by Warburg himself. A criterion could be a gesture that, once
identified and chosen as a key element, allowed him to assemble a group of images that had this
element in common. To recall Gartner’s description of metadata:

[t]he shape of metadata is designed by human beings for a particular purpose or to solve a
particular problem […]. There is nothing objective about metadata: it always makes a state-
ment about the world, and this statement is subjective in what it includes, what it omits,
where it draws its boundaries and in the terms it uses to describe it.
(2016: 4)

Warburg’s operation of including or omitting images was therefore based on metadata and under-
taken as a binary algorithm performed by Warburg in front of the data to be examined, making
either a positive or negative decision. It is important though to highlight that the formulation of the
criteria did not anticipate the analysis. The creation of metadata that described data and thus allowed
it to be located did not yet reveal the dynamics Warburg eventually discovered. They only became
evident once selected images were assembled in subsets in which they could interact with each
other, so that the space between them emerged and could therefore be studied.

178   Philosophy of Photography

 Metadata

Gartner, in his example of the globe that was cited above, spoke of ‘an immense body of data […] too
complex and too large a scale for us to grasp’ (2016: 4–5). The same could be said about the mass of images
that mirrored the pathways on which Warburg observed both conscious and subconscious legacies passing on
through history. ‘To make sense’ of this Warburg did not make use of a globe ‘which abstracts (the earth’s) impor-
tant features to make them comprehensible to us’. He created an atlas; an atlas designed to make these dynamics
accessible for examination (Warburg 2001). According to Gartner metadata constitutes culture because it allows for
the identification of its constituents, to make ‘connections between them’ and to transmit them between genera-
tions| (2016: 13–14). Against this backdrop a concept like Warburg’s famous ‘Pathosformel’ (see Warnke 1980) may
be understood as a tool to observe such connections. It formulates a cultural historical continuum of image
production by defining certain criteria, which serve to identify and interlink particular images. These criteria, the
‘Pathosformel’, can be consequently comprehended as metadata. As a methodological instrument of interpretation
the metadata created by Warburg reveals the images’ trajectories, making them visible and functioning in this
sense as an indicator. The question remains whether the studied phenomena themselves, the various motifs, forms
and other pictorial elements that are constantly reappearing throughout cultural history also have to be considered
as metadata – they evidently constitute culture by facilitating connections between its constituents, connections
described, if not invented, by Warburg.

Metadata and the migration of images in the digital era

Before coming to a brief conclusion, this article will attempt to cast an momentary glance at the role and the
significance of metadata in the works of two contemporary artists: Richard Prince and Louisa Minkin. Prince’s New
Portraits, particularly present across the media in 2015 because of a series of law suits provoked by it, are pictures
the artist took from other people’s instagram accounts. Prince scaled them up, printed and assembled them, plac-
ing a statement in the social media network’s commentary function to appear in the screenshots as signature. Not
only does Prince make use of descriptive metadata to appropriate the pictures, he is altering both their administra-
tive and structural metadata: transferring them from a digital onto a physical medium and from primarily non-
artistic accounts into art galleries and from there into private collections (sometimes even into the hands of the
original authors). This process or movement is synonymous with the transformation of the snapshots taken by
various persons into a single and coherent portrait series.
Louisa Minkin’ Durer Converted is, by contrast, the product of an entirely different operation. Minkin works with
programmes designed to render animations of three-dimensional objects on the basis of extensive recordings of
these objects in two-dimensional pictures. In this particular case she supplied these programmes with various repro-
ductions of Albrecht Dürer’s Melencolia I. She truly hacks the programmes by uploading information that must,
inevitably, be misinterpreted by the programmes’ structural metadata. Thus, Minkin reveals differences in data invis-
ible to the human eye and therefore impossible to experience for human agents and simultaneously examines
potential agencies of metadata.

[Link]   179
Johannes von Müller

These are only two exemplarily cases of contemporary movements of images that, in the digital
era, seem to take place rather on the level of metadata than of the actual data. A phenomenon by far
not limited to artistic imagery. However, the latter may be able to function as an explicit articula-
tion of more universal but implicit trends. Such is the claim made by the 2016 exhibition Electronic
Superhighway. The catalogue accompanying the exhibition that took place at the Whitechapel
Gallery in London states:

… the project examines artistic processes and formations that have emerged as a direct
result of changing technological conditions. Paying particular focus to how aesthetic forms
have been developed through the accelerating context of mass and distributed production
processes, it questions how constantly emerging technologies might lead us ‘to a sense of a
possible future’.
(Kholeif 2016: 25)

Assuming that within today’s digital era the major movements of images occur in the realm of meta-
data – in the exemplary case of Prince creating new meanings and identities, in that of Minkin even
new forms and realities – it seems essential to reassess Warburg’s approaches and to investigate
them from a changed perspective. Not so much because he extensively studied the migration of
images, but because of the way he revealed the dynamics driving them: by creating metadata and
thus maybe even enabling the identification of the driving forces themselves as metadata. Warburg’s
cultural historical pathways of images run in parallel and cross the ‘electronic superhighways’ – and
his ‘critical iconology’ may reveal itself to be a digital one.

References
Bagley, Philip R. (1968), Extension of Programming Language Concepts, Pennsylvania: University City
Science Centre.
Chaney, E. (2013), ‘R. B. Kitaj (1932–2007): Warburgian artist’, Emaj, 7.1, pp. 1–34, [Link]
[Link]/2012/11/[Link]. Accessed 1 May 2017.
Deppner, M. R. (1990), ‘Bilder als Kommentare: R. B. Kitaj und Aby Warburg’, in H. Bredekamp, M.
Diers and C. Schoell-Glass (eds), Aby Warburg: Akten des internationalen Symposions, Hamburg
1990, Weinheim: VCH Verlagsgesellschaft, pp. 235–60.
Didi-Huberman, G. (2004), ‘Knowledge: Movement (the man who spoke to butterflies)’, in P.-A.
Michaud (ed.), Aby Warburg and the Image in Motion (trans. S. Hawkes), New York: Zone Books,
pp. 7–21.

180   Philosophy of Photography

 Metadata

—— (2016), The Surviving Image: Phantoms of Time and Time of Phantoms: Aby Warburg’s History of Art,
Pennsylvania: Penn State University Press.
Engel, F. (2014), ‘Though this be madness: Edgar Wind and the Warburg tradition’, in S.
Marienberg and J. Trabant (ed.), Bildakt at the Warburg Institute, Berlin and Boston: De Gruyter,
pp. 87–115.
Gartner, R. (2016), From Information to Knowledge: A Beginner’s Guide to Metadata, Heidelberg: Springer.
Hristova, S. (2016), ‘Images as data: Cultural analytics and Aby Warburg’s Mnemosyne’, Digital Art
History, 2, pp. 116–32.
Kholeif, O. (2016), ‘Electronic superhighway: Towards a possible future for art and the Internet’, in
O. Kholeif (ed.), Electronic Superhighway: From Experiments in Art and Technology to Art After the
Internet, London: Whitechapel Gallery, pp. 24–33.
Manovich, L. (2015), ‘Data science and digital art history’, Digital Art History, 1, pp. 14–35.
Schmidt, P. (1993), Aby M. Warburg und die Ikonologie, Wiesbaden: Harrassowitz Verlag.
Warburg, A. (1999a), ‘Italian art and international astrology in the Palazzo Schifanoia, Ferrara’, in:
Ibd.: The Renewal of Pagan Antiquity (ed. K. Forster and trans. D. Britt), Los Angeles: Text &
Documents, pp. 732–757.
—— (2001), Der Mnemosyne Atlas (ed. M. Warnke), Berlin: Akademie Verlag.
Warnke, M. (1980), ‘Vier Stichworte: Ikonologie – Pathosformel – Polarität und Ausgleich –
Schlagbilder und Bilderfahrzeuge’, in W. Hofmann, G. Syamken and M. Warnke (eds),
Die Menschenrechte des Auges: Über Aby Warburg, Frankfurt: Europäische Verlagsanstalt,
pp. 113–86.

Suggested citation
von Müller, J. (2017), ‘Metadata: New perspectives on Aby Warburg’s “critical iconology”’, Philosophy
of Photography, 8:1+2, pp. 171–82, doi: 10.1386/pop.8.1-2.171_1

Contributor details
Johannes von Müller studied Art History and History in Berlin and Rome, received his Ph.D. in Art
History from the Universität Basel and is currently working as coordinator of the international
research project Bilderfahrzeuge. Aby Warburg’s Legacy and the Future of Iconology at the Warburg
Institute, London.

[Link] 181
Johannes von Müller

Contact: Bilderfahrzeuge, The Warburg Institute, Woburn Square, London, WC1H 0AB, UK.
E-mail: vonmueller@[Link]

Johannes von Müller has asserted his right under the Copyright, Designs and Patents Act, 1988, to
be identified as the author of this work in the format that was submitted to Intellect Ltd.

182   Philosophy of Photography