Semiotics and the Social History of Art*
Keith P. F. Moxey 
New Literary History, 1991, 22: 985-999

*This essay by a Commonwealth Center lecturer is an expanded version of a paper presented at
the Twenty-Seventh International Congress of the History of Art held in Strasbourg, 1989.

THE THEORETICAL PARAMETERS of much of the social history of art currently being written
in the Anglo-American world have largely been defined by the work of T. J. Clark. In two
important books, Image of the People: Gustave Courbet and the 1848 Revolution and The
Painting of Modern Life: Paris in the Art of Manet and his Followers, Clark provides both a
theory and a practice for an art history that transcends its traditional preoccupations with style on
the one hand and iconography on the other.' Clark's books not only served to bring out the
political implications of the work of Courbet and Manet within the historical horizons in which
they were produced, but they suggested that the paintings of these artists may have served an
active role in the creation of social and political attitudes. There are, however, aspects of Clark's
theory that betray a continuing adherence to some of the metaphysical values that have
characterized both formalist and iconographic histories of art, values that compromise our capacity
to fully historicize our understanding of the work of art. I should like here to question these values
in order to propose a semiotic theory of representation as the basis for a social history of art.
     It is clear from both his theory and his criticism that Clark subscribes to a notion of
immanent aesthetic value, one which serves to distinguish works of art from the rest of the
cultural artifacts belonging to the culture of which they are a part. He shares this attitude with a
long history of Marxist culture criticism, one which can be traced back to the Frankfurt School as
well as in the Anglo-American tradition constituted by Terry Eagleton and Fredric Jameson. Clark,
however, inherits this most directly from the French Althusserian literary critic Pierre Macherey.2
In the opening chapter of the Courbet book, entitled "On the Social History of Art," Clark sets out
the theoretical framework of his analysis. According to Clark the book is to be an account of the
mediations or links that join the work of art to other social processes. There is one respect,
however, in which the work is said to be independent or autonomous of the social setting in which
it was produced: "A work of art may have an ideology (in other words, those ideas, images and
values which are generally accepted, dominant) as its material, but it works that material; it gives
it a new form and at certain moments that new form is itself a subversion of ideology."3 Art is
thus defined as something quite distinct from an important aspect of the social formation from
which it is derived. Art may be bathed in the values of the class that is responsible for its creation
yet instead of being enmeshed in the transactions that constitute social life, it manages to escape
those circumstances in order to become an active agent which can "work" or manipulate the class
values with which it is associated. The consequences of such a strategy are significant. If the
aesthetic value of the work is an ahistorical constant, something that is always already there rather
than something that is ascribed to it by the culture of which it is a part, then the social history of
art will be insensitive to the ways in which the aesthetic value of the work is actually an
ideological construct. That is, it will have no way of articulating the way in which aesthetic value
was constructed within the historical horizon in which the work was produced nor a way of
accounting for the social function served by the aesthetic value ascribed to the work in the
present.
     In The Painting of Modern Life, Clark's criticism was enriched by his subscription to the
semiotic concept of representation. Works of art, he claimed, are to be regarded as cultural
representations that are similar to all the other processes of cultural signification. However, despite
the fact that on a semiotic model the notion of the work as a representation would define it as a
cultural construct, one that inevitably manifested the social values of the circumstances in which it
was produced, Clark sought to maintain a distinction between representation and ideology. Clark
claimed that ideology and the work's status as a representation were separate and distinct entities.
Ideology, however, worked to naturalize certain values available in artistic representations in the
interests of a hierarchically organized society: "Ideologies naturalize representation, one might say:
they present constructed and disputable meanings as if they were hardly meanings at all, but rather
forms inherent in the world-out-there which the observer is privileged [sic] to intuit directly."4
Clark's distinction is hard to justify in terms of semiotic theory. Only if we consider it possible to
define representations in terms of the signifying systems to which they belong, rather than in
terms of the relations they bear to all the other representations that social life, can we subscribe to
his view. To do so would be to adopt a formalist view of signification. When judged by any
historicized notion of representation, however, we must object that all representations are
"constructed and disputable," not just those that have been "naturalized" by ideology. Clark's
strategy, however, is a revealing one, for as in the art and ideology distinction of the Courbet
book, it enables him to lift the work of art from the embrace of its social and political setting so
as to recuperate it as an aesthetic object endowed with metaphysical value.
     Clark's decision to distinguish between ideology and the work as a representation represents
a conscious rejection of one of the most important developments in the theory of ideology, namely
that put forward by Louis Althusser. In his essay "Ideology and the Ideological State
Apparatuses," Althusser revolutionized the traditional Marxist notion of ideology by suggesting
that far from being a kind of "false consciousness," that is, a system of ideas sponsored and
supported by the dominant classes as a means of concealing the true nature of class relations,
ideology is an integral part of all social relations.5 By suggesting that ideology was to be found at
every level of social interaction, Althusser effectively challenged Marx's view that society was
constituted by a "base," which included the material basis of human existence such as the means
of production and the organization of labor, and a "superstructure," which included the world of
ideas or "ideology" such as religion, philosophy, law, literature, and art.6 By collapsing the
base/superstructure distinction in claiming that ideology was as much a characteristic of the base
as it was of the superstructure, Althusser abandoned one of the central tenets of Marxist culture
criticism, namely, that developments in the base of society, political revolutions and so forth, were
necessarily reflected in the superstructure. He replaced it with an expanded notion of ideology that
went way beyond its traditional identification with the superstructure, so that it now covered all
aspects of life in society. Althusser's model was deeply influenced by semiotic theory, so that in
his view ideology becomes a kind of cultural semiotics in which ideology and sign systems
become synonymous with one another.
     Althusser's identification of Ideology with systems of signification affords us a theoretical
basis for a genuinely revolutionary history of art, one with radical implications for the traditional
underpinnings of the discipline, that is, its Kantian, or immanent, theory of aesthetics. If ideology
is present in all forms of cultural representation, then there is no way of surrounding the work of
art with the cordon sanitaire of aesthetics. There is no way to make the work safe from
the rest of the signifying systems that constitute the culture in which it is located. The work shares
the ideological burden of all forms of cultural representation and is thus equally implicated in the
articulation and perpetuation of notions of race, class, and gender. While eliminating the notion of
immanent aesthetic value, such a view would not deny the existence of aesthetic value as such.
Rather, it would insist that aesthetic value was a social construct. Aesthetic value would be
regarded as a value ascribed to the work by the culture in which it was produced. These values
would not be fixed but would be continually renegotiated through the course of time. Each
generation or alien culture through which the work passed would ascribe it a different value so
that it might either be treasured, for reasons that were radically different from those invested in it
in the first place, or it might lose its value altogether and be reduced to the status of a
non-aesthetic cultural artifact. The history of art is littered with examples that might be used to
illustrate this point: medieval altarpieces which were originally revered because of their role in the
devotional aspects of religious practice, now divorced from their ecclesiastical settings and located
in the clinical context provided by nineteenth and twentieth-century museums for their formalist
narratives of artistic development; or Roman temples cannibalized for their copper or limestone
during the Middle Ages and the Renaissance, so that they were either reduced to ruins or
disappeared altogether.
 If we accept a semiotic definition of representation as the basis for a future history of art, then it
becomes important to examine semiotic theories on which this discipline might be based. It seems
clear that the semiotic tradition that depends ultimately on the linguistic theory of Ferdinand de
Saussure cannot serve as an adequate basis for this enterprise.7 By positing that the linguistic sign
or word draws its meaning from its relation to all other words in the language rather than from its
referents, Saussure not only dramatized the arbitrary nature of the linguistic code but severed that
code from the circumstances in which it is located. According to Saussure, the linguistic sign is
composed of a material soundimage or signifier and an ideal component or concept, the signified.
The meaning of the sign or word, which is associated by his theory with the concept of the
signified, is thus derived from its location within the language rather than from its location in the
process of social communication. The word's meaning is thus isolated and safeguarded from the
significatory interference or "noise" that results from its articulation in a social setting
characterized by differences of race, class, and gender.
 
     There are, however, other semiotic alternatives more suited to a historical project. One of
these is to be found in the history of the Soviet reception of Saussure's linguistics. In contrast to
Saussure, both Valentin Volosinov and Mikhail Bakhtin argued that the study of language as a
signifying system could not be divorced from a consideration of the other aspects of the culture of
which it formed a part. According to this view, language is shaped and colored by the social
location of its utterance so that the lexical sign or word draws its meaning from the full
complexity of its social function rather than from its location within a hermetic system. According
to Bakhtin: "The word, directed toward its object, enters a dialogically agitated and tension-filled
environment of alien words, value judgments and accents, weaves in and out of complex interrela-
tionships, merges with some, recoils from others, intersects with yet a third group: and all this
may crucially shape discourse, may leave a trace in all its semantic layers, may complicate its
expression or influence its entire stylistic profile."8 Since for these Marxist authors ideology was
regarded as an integral part of social life, it was possible for them to conclude that language was
also ideological. In Marxism and the Philosophy of Language, Volosinov writes: "The domain of
ideology coincides with the domain of signs. They equate one another. Wherever a sign is present,
ideology is present, too. Everything ideological possesses semiotic value."9
     The semiotic status of the model used by Bakhtin and Volosinov was always threatened by
the specter of the base/superstructure distinction which would insist that the play of signs in
ideology depends upon a more fundamental reality such as the class struggle. It may be usefully
supplemented by reference to the work of the American philosopher Charles Sanders Peirce.l 
Peirce had a far broader, more universal notion of the sign than did Saussure. Rather than restrict
the definition of the sign to linguistics, Peirce claimed that the notion could be used to refer to all
kinds of coded behavior. His division of the concept of the sign into a triad consisting of the Sign,
its Object, and the Interpretant means that a sign cannot be defined in terms of the system to
which it belongs, say gestures or dance steps, but must be interpreted in the light of the way it has
previously been understood. Peirce's notion of the Interpretant insures that our understanding of
the sign is always a mediated one. The significance of the signs that constitute signifying systems
is not immediately accessible to us for the signs themselves carry the interpretations that have
already been placed on them. Our very perception of the sign is conditioned by the ways in which
our culture has taught us to recognize it. This means that the sign is involved in a process of
limitless semiosis. Signs engender other signs ad infinitum. Interpreters of signs are caught up in
this process so that while being the users and manipulators of signs, they in turn produce signs
that are interpreted and manipulated by others. More importantly perhaps, this view of the endless
proliferation of signs is responsible for severely restricting the autonomy granted the human
subject in the humanist tradition. Rather than being an all-seeing observer whose account of the
play of signs has the finality of an objective and unquestionable truth, the interpreter is reduced to
the status of a sign for those who interpret his or her interpretations in the future.
 Semiotic theory thus prompts us to reevaluate the role of the historian as well as the function of
interpretation. For a semiotic theory of the interpreter as a social subject we can turn to the way in
which Jacques Lacan applied Saussure's linguistics to psychoanalysis." According to Lacan the
human subject is created when it is subjected to the signifying power of language. It is the
realization that one is an "other" to the "Other" that splits the subject between the Real and the
Imaginary. While language, or the Symbolic, grants the subject access to cultural and social life, it
also defines the subject in terms of what it is not. On this view, the subject becomes part of a
signifying system that alienates it from itself. Language serves to mark and maintain a subject's
inability to know the Real. Such a view of human subjectivity implies that the sign systems
manipulated by the subject must always be unstable in their meaning. If the subject is both the
subject as well as the object of language, then the possibility of univocal meaning is necessarily
lost. Language speaks the author's subjectivity to the same extent that the author's subjectivity
manipulates language. Language supports and structures the utterance as much as the utterance
instantiates and articulates the language. What the author wishes to say is never found in the place
in which it is said.
     If we accept that, like lexical signs, visual signs are not only defined by the circumstances
of their social production but that their status as systems of communication means that they bear
the traces of the ways in which they have previously been understood, then it is possible to
conclude that not only are the sign systems studied by the art historian ideological in nature but
that the historian is also an ideological sign engaged in the production of ideological signs. Not
only is the historian's understanding of the past an ideology critique but the critique itself is
ideological. The writing of history is, on this view, identical with the production of ideology. Our
preference for one interpretation over any other would thus depend upon the ideological
formations that have shaped us as well as those with which we identify. Interpretations will be
favored or rejected according to whether or not they offer an account of the past that corresponds
with the interpreter's views on the political needs of the present and the future. The prospect of
invoking a political criterion for the validity of interpretation should not be regarded as favoring
the straightforward imposition of the political values of the author on our understanding of the
past. A valid historical interpretation would be one that made every effort to grapple with the
strangeness or "otherness" of the historical horizon it sought to interpret. The thrust of
interpretation must involve a recognition of the "other." It is only through an appreciation of the
radical alterity of the political circumstances of the past that historical interpretation can serve the
political interests of the present. Those interpretations that flatten the texture of the historical
horizon through the imposition of a political agenda do violence to the complexity of the horizon
to be interpreted as well as to the complexity of the political situation in which they are
formulated. A failure to do justice to the ways in which the politics of the past differ from the
politics of the present serves merely to caricature both the past and the present.
     The adoption of a socially and historically specific notion of the sign implies that the study
of visual representation will approach visual signs as if they were contiguous to and continuous
with the signifying systems that structure all other aspects of the historical horizon that is the
object of study. The work of art will be read as if its surface were part of the social fabric of
which it was once an organic whole. The dominant metaphor for this type of investigation lies in
tracing or delineating. As Roland Barthes put it: "In the multiplicity of writing everything is to be
disentangled, nothing deciphered; the structure can be followed, 'run' (like the thread of a
stocking) at every point and at every level, but there is nothing beneath."'2 The focus on the
function of the work within a broader pattern of social communication means that there is no
attempt to look through the network of signs in order to reify or fetishize the intentions of the
artists involved in their production. Far from seeking to enhance the aesthetic value of the work
by discussing the hidden intentions of that absent personification of creativity, the artist, this type
of analysis will emphasize the opacity of the work, the way in which its historical conventions
resist our gaze. It is only through a recognition of the unqualified alterity of the work that we can
hope to understand the cultural forces that once flowed through it.
       The approach to a social history of art outlined above would involve rethinking some of
the heuristic tools currently available to the discipline. While at first glance the iconographic
method might be regarded as sympathetic to semiotic interpretation, it is after all involved with
the sifting and sorting of the conventions utilized in the representation of different subjects; its
epistemological basis that is, the truth claims with which iconographic knowledge is associated,
cannot be reconciled with a semiotic model.l3 While Panofsky drew a distinction between
iconography, which he regarded as the study of conventional subject matter, and iconology, which
he defined as the study of intrinsic meaning, the one allegedly dealing with the surface of culture
and the other penetrating to the "essential tendencies of the human mind," his actual practice
tended to blur this distinction so that iconography, like iconology, became the means by which the
historian could obtain access to the mind of the artist and period under consideration.l4
Iconography and iconology purported to allow the historian to see through the work into the
mental landscape of the culture that produced it. The knowledge resulting from the application of
the iconographic method thus depended on movement from one level of culture to another. Its
governing metaphor was one of penetration: one laid bare secrets that were hidden from view or
delved beneath mere appearances in order to establish the truth. The doctrine of"symbolic forms,"
which Panofsky derived from Cassirer, enabled him to claim that iconographic and iconological
meaning had the status of a deep mental structure. According to neo-Kantian epistemology,
symbolic forms were the means by which mind and nature were bonded together. While
unmediated access to nature was regarded as impossible, symbolic forms were alleged to fit,
match or correspond with the underlying structure of reality. If iconography is to be understood as
a semiotic study, then it is clear that its idealist epistemology will have to be abandoned. A
semiotic iconography will consist of the study of the pictorial conventions used by a particular
culture in the process of encoding the values that structure its reality. The formulae used in the
articulation of different themes will be studied as signifying systems that shaped the character of
social life. The focus in other words, would be on the social work performed by these structures
of signification, the way in which they intersected with all the other signifying systems that made
up social life, rather than upon their "intrinsic meaning."
     Another strategy available to the social history of art, one developed by Michael Baxandall,
namely, the interpretation of the "period eye," or the "cognitive style" of a particular historical
period, on the basis of an analysis of the various systems of signification that configured its social
life, would also be in need of revision.l5 While the value of Baxandall's method lies in the way in
which it suggests that cultural practices as disparate as gestures, dance steps, and forms of spatial
computation, may be relevant to our understanding of the work of art, the implication (one that
Baxandall himself seeks to qualify) that knowledge of these disparate practices affords us access
to a world view that is manifested in art must be challenged. Rather than using such information
to suggest the ways in which they give expression to the intellectual attitudes and habits of a
period in such a way as to stabilize the meaning of works of art belonging to that context, a
semiotic approach would attempt to define the ways in which works of art actively worked to
generate meaning and thus to define the values of the society in which they were produced. Rather
than study social custom in order to elucidate art, an attitude which would attribute the work a
fixed and passive role in the analysis, a semiotic view would ascribe to art the same function as
social custom and view it as actively engaged in the construction of culture. The goal of the
analysis would not be the work itself so much as the way it performed in a social setting.
     Finally we must examine T. J. Clark's use of art criticism as an interpretive method in his
book on Courbet. Clark draws a distinction between the audience for a work, that is, those who
actually saw it upon its completion, and the public, by which he means something altogether more
elusive. By public, Clark refers to the unconscious values of the period which are betrayed by "the
faults, silences, and caesuras of normal discourse."'6 In these places, where critical writing turns
opaque and difficult, where the author appears to struggle to express or repress things that cannot
find their way into the text, Clark claims that the unconscious values of the period are to be read.
By invoking a psychoanalytical model Clark claims to penetrate the surface of conventional signs
in order to read those that lie in a different realm. What Clark has in mind however has less the
quality of psychoanalytic truths than cultural attitudes that are either expressed with inordinate
passion or absent as a consequence of repression. Like Panofsky's iconographic method, Clark's
analysis claims to be able to see through culture to the mind of the period. The concept of the
public allows him to use his familiarity with the intellectual history of the nineteenth century to
afford us a sensitive assessment of the values that dominate the criticism of Courbet. If we wished
to recuperate this method for a semiotic approach we would have to forego the possibility of
movement from the world of cultural signs to some underlying realm of the mind. Instead of
claiming access to what the public thought, we would evaluate criticism in the light of other
strands of cultural signification. The analysis, in other words, would have to take place on the
surface of culture rather than behind it.
     By way of conclusion I will illustrate the terms of this analysis with reference to the
interpretation of a specific work, namely Hieronymus Bosch's Garden of Earthly Delights which
was produced in The Netherlands about 1500. In choosing this work as one which I can assume is
familiar to most, I do not rely on some notion of immanent aesthetic value but on the way in
which this painting has been ascribed aesthetic value by our culture. One of the ambitions of the
social history of art outlined here would be the attempt to articulate the way in which aesthetic
value intersected with social value. Such studies would document the ways in which the cultural
status accorded to the work has risen and fallen according to the social and political circumstances
in which it has been viewed. Such studies would also be interested in understanding the value we
ascribe to the work today. What role does it play in defining an image of the late Middle Ages as
a dark age, one that was plagued by superstition and the belief in demonic forces, that can
usefully be played against the emerging Renaissance as a period that saw the triumph of reason?
Does it function as a metaphor for the existence of the unconscious and the threat of its irruption
into the conscious world of social life? What meanings does it possess for people of different
social classes, and does it have gendered meanings for members of different sexes? Does the
Garden serve as a Utopian vision of the classless society or as a vision of an egalitarian relation
between the sexes freed from patriarchal domination?
     If we analyze the pictorial components that make up the composition we observe that they
contain mimetic elements--that is, accurate, naturalistic images of considerable illusionism, and yet
it is hard to understand the significance of the ways in which these pictorial elements have been
juxtaposed. The traditional approach to this painting has been to ascribe to it symbolic meaning.
Scholars have attempted to read the various elements in the composition as symbols representing
hidden realities. The classic expression of this attitude is that voiced by Erwin Panofsky: "In spite
of all the ingenious, erudite and in part extremely useful research devoted to the task of 'decoding
Jerome Bosch,' I cannot help feeling that the real secret of his magnificent nightmares and
daydreams has still to be disclosed. We have bored a few holes through the door of the locked
room; but somehow we do not seem to have discovered the key."'7 Most current interpretations
therefore suggest that Bosch's visual forms are symbols that can be explained by reference to
esoteric knowledge which was part of Bosch's culture but which is unknown to us today. As a
consequence, the literature is littered with attempts to decipher Bosch's painting in terms of
astrology, alchemy, rare forms of heresy, and so forth . I f we replace the pictorial symbol with the
pictorial sign, then it is possible to draw attention away from the depths of meaning that are said
to lie beneath the surface of Bosch's visual forms in order to focus on the surface. Rather than
looking through the work, we would interpret the way it resists our gaze. Instead of valuing its
transparence, the way in which interpreters have suggested that it makes accessible ideas that are
apparently hidden, I will emphasize its opacity or the way in which it insists that the interpreter
create meaning before it. In contrast to Panofsky I will quote the approach to interpretation
formulated by Michel Foucault: "The contemporary critic is abandoning the great myth of
interiority: . . . He finds himself totally displaced from the old themes of locked enclosures, of the
treasure in the box that he habitually sought in the depths of the work's container. Placing himself
at the exterior of the text, he constitutes a new exterior for it, writing texts out of texts.''l8
     Since I have been working on a new interpretation of The Garden of Earthly Delights for
some time now, I can at least attempt to summarize the thrust of the argument for you. First I
have tried to discover all those aspects of the work that belonged to discernible codes or that make
use of recognizable conventions. Some of these codes belonged to the internal history of artistic
production in The Netherlands prior to Bosch's time and others are related to the signifying
systems that structured the society within which the work was produced. The triptych format,
together with the fact that the central panel, in which a landscape filled with human figures was
located between panels that can be identified as representations of Paradise and Hell, belongs to
the codes of ecclesiastical art and to the iconography of the Last Judgment. The original location
of the work however, in the palace of the Burgundian aristocrat Henry I I I of Nassau as part of a
collection that included paintings by Italianizing artists representing mythological subjects,
suggests that the traditional meaning associated with the painting's format as well as its subject
had been manipulated and transformed so as to produce new meaning. What, in other words, was
the significance of the reworking of these traditional codes and their transfer from a religious to a
secular context? The clash of codes that prompted this investigation reappeared as a theme at
every point in the analysis. For example, if we turn our attention to the visual fabric of the
central panel, we find that the scene consists of a series of interwoven sets of incongruent sign
systems. The conventions of illusionistic art, so well-established in The Netherlands by the end of
the fifteenth century, would have insisted that all elements in the composition be rendered to a
scale that could be reconciled to their location within a perspectivally organized space. In Bosch's
painting, however, we find birds and fruits whose scale is several times larger than that of human
beings. The very principles of mimesis which led the artist to depict these birds and fruits with
meticulous attention to their appearance are defied when we consider their relation to the
landscape. Bosch appears to have applied different notions of representation in the execution of
different parts of his panel. In certain parts of the composition observation and imitation seem to
have been important, in others such principles seem to have no relevance. Why did this painting
break so elaborately with the conventions of visual representation that were current in its day? Are
there cultural codes belonging to this period that might account for this apparent disjunction?
     One of the places in which the mimetic principles of medieval art were habitually violated is
to be found in the world of manuscript illumination, particularly in the images that populate the
margins. In this context the principle of fantasy seems to have been extensively used in the
construction of images that had little to do with the world of perceptual experience. The function
of fantasy, of the deliberate avoidance of mimesis in such images, can be construed from the type
of subject matter they represent. Most of these images are satirical in nature and are mainly
directed at two classes. That is, they poke fun at the social roles and cultural ideals of the clergy
and the aristocracy. For example, in the Gorleston Psalter, which was produced in East Anglia in
the fourteenth century, a procession of rabbits wearing ecclesiastical garb mimic a funeral
procession as they bear one of their number to its final resting place. The substitution of rabbits
for humans is the means by which a disparaging humor is directed at the solemn rites performed
by the clergy. In a Pontifical illuminated for the bishop of Metz dating from the same time, the
satirist's aim is the aristocracy. In this case an army of rabbits lays siege to a castle defended by
knights. The characterization of the most timid of beasts as the victor in a confrontation with
armed knights pokes fun at the ideals of chivalric valor to which the aristocracy subscribed. The
function of fantasy then, as an alternative to representation based on mimesis, is to provide a
mechanism by which established social hierarchies might be overturned in the imagination without
actually overturning them in reality. It is a means by which the status quo can be challenged,
disparaged, ridiculed, and rejected without effectively changing anything at all. It is the means, in
other words, of turning the world upside down. In light of this traditional use of fantasy we must
ask why this principle was useful to Bosch in the execution of The Garden of Earthly Delights. If
we look at the inversions of scale and the apparently pointless activities in which the figures in the
central panel indulge, it is difficult to discern the social function such fantastical goings-on might
have served. While the iconographic clarity of the Last Judgment theme is clouded by the role of
fantasy, its absent presence as the ecclesiastical code within which the codes of artistic
representation are scrambled serves to stabilize meaning and suggest a level on which the religious
function of the image still persists. If this is the case, then it is possible that it is the dramatization
of the conflict between representational codes that is the most important aspect of the altarpiece's
production of meaning.
     Part of the new humanist culture to which both Bosch and his patron belonged was
concerned with a reevaluation of the status and function of the artist. From being an artisan who
served the interests of the faith in the production of ecclesiastical furniture used in the promotion
of religious devotion, the artist had come to be regarded as a practitioner of one of the liberal arts,
arts which enjoyed great cultural prestige because of the esteem in which they had been held in
antiquity. During the course of the fifteenth century Italian artists and theorists had continually
stressed the notion of invention and the importance of fantasy for artistic creation. The concept of
fantasy became the basis on which artists made exaggerated claims for the importance of their art
and for the exalted social status to which they themselves aspired. Given this intellectual context, I
should like to suggest that the representational incoherence of Bosch's Garden, the conflict
between mimetic and fantastic modes of representation, should be understood as the way in which
humanist culture articulated its claims to a new status for the artist. The world of fantasy, hitherto
used as a vehicle for satire, as a means of turning the world upside down, is the means by which
Bosch could turn the tables on the artistic culture of his day. The clashing codes serve to draw
attention to the role of the artist in the construction of visual culture. It is no surprise that Bosch
should have used fantasy as a metaphor of his exceptional imagination in an aristocratic context
well versed in the pictorial conventions of the illuminated manuscript. By transferring the world of
fantasy from the margins of artistic production to their center, Bosch commodified and fetishized a
mechanism hitherto associated with the inversion of the social hierarchy. In Bosch's hands fantasy
is no longer used to turn the world upside down but rather to inscribe the importance of artistic
autonomy and inventive genius within the ideology of the dominant culture.
 Finally, what is the role of the interpreter, that is, of the author of this piece of cultural
entertainment? The analysis I have presented here depends upon the theoretical foundations
outlined in the first part of this paper. On the one hand, I have tried to read the image and the
culture it belonged to in terms of codes and conventions. On the other, I have made personal
claims as to which codes and conventions I wanted to ascribe importance to. My narrative
depends to a large extent on the way in which I have imposed my own values upon the
interpretation of the material at hand. In stressing the way in which fantasy or the mechanism for
turning the world upside down can be coopted so as to be used not to question the status quo, but
to support and strengthen it, I wanted to draw attention to the ideological significance and social
function of representation. That is, I wished to draw attention to the fact that far from deriving its
significance from some transcendental notion of the value of artistic creativity, The Garden of
Earthly Delights actively participated in the creation of the myth of the artist as an autonomous
creator.
     In conclusion I should like to return to where I began, namely to the criticism of the notion
of immanent aesthetic value as the foundation for a history of art. The elimination of the view that
aesthetic value is an immanent characteristic of works of art allows us to decenter art history.
Instead of concentrating on the work as a repository of social value, the work of interpretation is
dedicated to its role in history. The work is viewed as an agent rather than the consequence of
social change. Just as semiotic theory can be used to decenter the work of interpretation, it also
serves to decenter the role of the historian. The assumption behind the interpretation is no longer
that the interpreter hopes to impose closure, to put an end to the work of interpretation, but rather
that the interpreter is only another sign in the endless process of semiosis. Rather than having to
find what Panofsky called an "Archimedean point," a position outside the cultural practice that is
the object of study, the position of the historian is now incorporated within the act of
interpretation. Such a view not only enables us to open up new avenues of art historical research
but it allows us to foreground the ways in which an understanding of our present circumstances,
our position within the ideological struggles of our own time, is both the basis and the medium for
our interpretation of the past.