IMAGES Journal for Visual Studies
Published by
cvs_center for visual studies_zagreb
ISSN 1848-9478

Visual studies: Critical and historical assessments


Ilaria Fornacciari: The Complexity and Stark of Pictorial Knowledge: About Foucault Reading Panofsky


In his short and intense critical review (published in 1967) about the two works of the German art historian Erwin Panofsky, Michel Foucault looked at the operation of the father of iconology not only as an innovation in the context of art history, but more generally as a signal of turn in the way we approach the field of relations between texts and images. In this prospective, Foucault's analysis can be read as an astonishing prevision. He pointed out two large displacements following from Panofsky's works: 1) The abolition of the privilege of discourse, and 2) the problematization of the function of representation. I will focus on the epistemological character and outcomes of these two radical operations through comparative analysis between some theoretical notions used by the art historian in his works and some constitutive elements of Foucault's archaeological analysis.

Keywords: Foucault, Panofsky, archaeology, iconology, vision, discourse, representation

This article focuses on a short critical review written by Michel Foucault on two works by the art historian Erwin Panofsky (Foucault, 2001). First of all, I would like to clarify the reasons for which we might be interested in pointing out some aspects of this text. The importance of the role of vision and images in the thinking of Foucault is today widely recognized. From the role of painting in the first chapter of History of madness (Foucault, eng. ed. 2006), to the  notion of regard, translated as “gaze” in The birth of the clinic (Foucault, eng. ed. 2003) the book that starts with the well-known sentence “This is a book about space, about language, and about death; it is about the act of seeing, the gaze” (Foucault, 2003: p.xi); up to the well known chapter on Velásquez’s painting Las Meninas that opens The order of things (Foucault, eng. ed. 1971) with its enigma of representation - the attention of the philosopher on the visual is also fundamental in his genealogical analysis, from the panoptic model in Discipline and punish (Foucault, eng. ed. 1995) to the political role of the notion of “visibility” in the courses held at Collège de France.

All this explains the need for reassessment of what has been untill now described as an attitude or a style, on a more ample theoretical and practical plane1. It demands a reflection that would take into account these elements not as additional, but as an essential part of the so called “boite à outils” (the tool-box for a diagnosis of the present as was conceived by Foucault), recalling the advice of the philosopher Gilles Deleuze, who already admonished us in 1986 in this respect:

 “If we forget the theory of visibility we distort Foucault's conception of history but equally we distort his thought and his conception of thought in general” (Deleuze, 1988: p.43)

In this context, the article of 1967 acquires a particular value. I propose, therefore, to analyse the implications of what we can call “a change of attitude” regarding the question of the relation between seeing and saying, as it was glimpsed through the conceptual instruments of Archaeology (Foucault’s method of analysis up to the ’70) in confrontation with some notions of Panofsky’s studies. This brief investigation will also enable us, I hope, to reveal the astonishing predictive capacities of Foucault's reflections of  '67,  as well as to detect the elements that characterise emerging and problematic connections between the visual and the cultural in his work. Published in Le nouvel observateur, the article I will focus on was written in tribute to the French edition of two essays by Panofsky. The most important agent for the introduction of Panofsky's work in France was the sociologist Pierre Bourdieu, who translated the essay Gothic architecture and Scholasticism (Panofsky, 1951) and directed the collection that contains the volume.2 Furthermore, in his postscript of  “Architecture gothique et pensée scolastique” (Bourdieu, 1967), the sociologist focused on the concept of Habitus, which will become one of the most important notion of his work. The context of Panofsky's reception in France is particularly interesting in order to situate Foucault's reading, or rather, to mark its extreme discrepancy with the most common and resistant reactions. 

It is true that the attention given to Panofsky's studies, after the discussions between experts3 about his works on the Gothic, came more from the side of philosophy, sociology and history of science (that esteem and appreciate its theoretical constructs), rather than from art history.4 The philosopher Merleau-Ponty carefully discussed Panofsky’s work Perspective as a symbolic form in his course of 1954-55 at Collège de France (Merleau-Ponty, 2003), and he will also quote the art historian in his later work L’oeil et l’esprit (Merleau-Ponty, 1964: p.32); in 1956 the science historian Alexandre Koyré wrote a critical comment in Critique about Panofsky’s Galileo as critic of arts (Koyré, 1956) and some years later Lévi-Strauss in an interview called Panofsky “a real structuralist” (Lévi-Strauss, 1965). In this sense we can also mention the success of the concept of Habitus for the historians linked with the revue Annales.5

So, while an important interest in Panofsky's writings came from a certain tradition of history and history of science at the precise moment at which they were calling into question their own epistemological tools, as demonstrated in Bourdieu's postscript, the situation is different for history of art. The development of Panofsky's reception in this field seems to be dominated by what we can call a strong prejudice or a long misunderstanding. If Panofsky's works on gothic were suddenly stigmatised as markedly rationalist (Francastel, 1952), after 1967 his works were understood, by French criticism, more through the opposition between the visual and the discursive elements of culture leading eventually to a kind of criticism that accused Panofsky's method as being overtly “logocentric”6. This interpretational grid, which is a side effect of the tendency to equate Panofsky's method with human sciences (the context in which his works were published), far from being regarded as outdated, seems today still difficult to call into question7. In opposition to all this, and with an ironical effect, it is clear that the interest of the scholars who were engaged in the promotion of Panofsky’s works in France, was not to peacefully assimilate his thinking.

What they underlined was precisely how his works problematized or even surpassed the prevalent epistemological background of human sciences at the time. This is apparent in the title “L’évolution d’un scheme structural”- The Evolution of a structural schema” which Bernard Teyssèdre gave to the second part of his Avant-propos to L’œuvre d’art et ses signification (The French translation of Panofsky’s Meaning in the visual art) in 1969. In fact, two years before, not only did Bourdieu use Panofsky's reflections to rethink some of his methodological instruments, but Foucault’s reading was precisely geared to indicate a new dimension of the relations between visual and discursive disclosed by Panofsky's works.   

In his article, entitled “Les mots et les images” (in tribute to René Magritte, with whom he had a correspondence the year before [Foucault, 1973]) Foucault underlines two aspects of Panofsky’s methods and presents those as representative examples of a change how historians are perceived, a change of position of their habitual way of seeing, reading and knowing. I would like to analyse Foucault's response and look closely at some notions of Panofsky’s Gothic architecture and scholasticism (Panofsky, 1951) and Studies in Iconology (Panofsky, 1972) while pointing out these two major operations.

The suppression of the privilege of discourse

“Panofsky removes the privilege of discourse. Not in order to claim the autonomy of the plastic universe, but in order to describe the complexity of their interrelations: entwinement, isomorphism, transformation, translation, in short, this entire “feston” of visible and  speakable that characterises a culture in a moment of its history” (Foucault, 2001: p.649 [my translation])

The first, and probably the most important aspect of Panofsky's essays that strikes Foucault is that they bring hardship to the discourse's leading position. Between the plasticity and the discursive, the relation is not a quiet domination of the discursive over the plasticity of the figurative; what we are faced with is more a succession of challenges thrown from one at the other and Panofsky’s essays present themselves as a kind of cartographic view of those intricate movements, a model of historical intelligibility of artistic production that takes into account both the homologies and the conflicting aspects of these two dimensions. Foucault wrote clearly in this respect:

“Discourse, therefore, is not the interpretative background of every phenomenon of a culture. To make a form appear is not a devious modality (more subtle or naïve, as one would like) of saying something (…). Discourse and figure each have their own way of being; but they maintain complex and entangled relations. It’s their reciprocal functioning that has to be described.” (Foucault, 2001: p. 649 [my translation])

An example of how the historical interactions of the two dimensions are presented is the process called by Panofsky “Pseudomorphosis8 (Panofsky, 1972) in which the aspect of the historical interaction between figurative components of different traditions is dominant but connected with textual elements in various and unpredictable ways. Another and very disparate example is the principle of manifestations, through which the homology between scholasticism and Gothic architecture is structured in his essay of ’51: in this case, in a certain age and in a precise place, by the action of an institution, it is the same principle of spatialization that dominates both the plasticity and the discursive practices of what he calls “visual logic” (Panofsky, 1951: p.58).

These examples, besides showing the complexity of the mentioned phenomenon urge us to observe that concepts and methodological tools used by Panofsky are resulting from their specific operational field. And, as Didi-Huberman admits: “It happens actually that the concretely obtained results are more rich, opened, than the abstractly professed intentions” (Didi-Huberman, 2008: p. 74 [my translation]).9

As we have already anticipated, the reflection on the complexity of the interactions of discursive events and visible dimensions that mark culture in a precise moment of its history, would become the most intriguing positions of Foucault in his Archaeology of knowledge - the book in which, as Deleuze pointed out, Foucault could not designate the elements of visibility’s systems otherwise than merely negatively, as “non-discursive”. However, as we will soon see, in this work, which clarifies Foucault's conceptual instruments, methodology, and aims of the analysis, a particular place for painting is preserved.

The problem of representation

The second point underlines the necessity of taking into account the representational function as “a complex function that traverses, with different values, the entire formal substance of the painting”. This point marks the impossibility of making the relation of painting's formal aspect with an object of representation as stable and referential. The traditional apparatus of representation, that indicates some kind of connection between what is represented and a means of representation (illustration, depiction, painting as window on the world), that obliges us to question indefinitely the status of “oeuvre”, is now open to a different dimension, that is - the representative function as a productive one - as performative. “These themes donnent lieu (produce - give space), in the strict sense of the term, to a sensitivity, to a system of values, but under the rule of a cultural symptomatology” (Foucault, 2001: p.649 [my translation])

Fig. 1: Muybridge E. J. (1884-1904), Walking, taking off  hat, Date: ca. 1884-1886, Animal locomotion: an electro-photographic investigation of consecutive phases of animal movements. 1872-1885. USC Digital Library, 2010, public domain.

In his famous introduction to Studies in Iconology, Panofsky (through the example of the acquaintance greeting by lifting his hat) distinguishes three levels of meanings: The primary meaning, that is, in turn, separated in factual and expressive; the secondary or conventional subject, that is properly the first intelligible content; and the intrinsic meaning. This last level of meaning, according to Panofsky “may be defined as a unifying principle which underlies and explains both the visible event and its intelligible significance, and which determines even the form in which the visible event takes shape.” (Panofsky, 1972: p.28) Assigned to the analysis of art works, those levels are identified with forms, motifs, themes and concepts, seen as manifestations of principles, in other words, as elements which are susceptible to designate different components of a certain cultural cosmos. The inquiry into those kinds of elements is presented as the aim of the method and has to be regulated by the history of styles and cultural symptoms.

This process, divided into levels for reasons of methodological clarity, is actually an organic activity, here dissected in order to show the components, normally taken for granted, as relevant and permeated by cultural elements through different modalities. Through the recognition of those different levels and through his notion of the artwork as “document”, the action of Panofsky makes the traditional scheme of the representation’s referentiality immediately unsuitable as a stable formal structure, but analysable as an historical function. The unity of the painting as œuvre, guaranteed by this opposition, is now undone.

Fig. 2: Panofsky’s final Schematization of an Interpretive Strategy, 2939, Surce: Panofsky, E. (1972) Studies in Iconology: humanistic themes in the art of the Renaissance. Oxford: Icon Edition.

Panofsky shows us that without escaping the fictional unity of the artwork, without following the  texture of relations, through those elements that have regulatory functions at different levels (styles, themes, symptoms) we would not be able to formulate even the first content, that is, the identification of the objects that are figured by lines and colours.10 The issue of the function of  representation will, of course, bring to mind the position of this concept in Foucault’s The order of things (Foucault, eng. ed. 1971) in which book Velésquez’s painting, incarnating the problematic status of representation, plays the double role of both condition for the historical investigation and document of the classic age. But following this line of thought, we would soon be led - through the hypothetical comparability of Panofsky’s concept of common invariability and Foucault's notion of field of possibility - to the troublesome question of the relation between his concept of episteme and Panofsky's idea of “cultural cosmos” or even Weltanschauung11.

These concerns, even though seemingly directed to an overtly theoretical-comparative task, are also beside the point considering the position taken by Foucault in his Archaeology of knowledge, especially having in mind his strong refusal to offer any kind of cultural totality. What is more interesting, regarding the role of representation, important is the relation between the function of œuvre in his well-known lecture of  '69 “What is an author?” (Foucault, 2001b), in which he developed and specified some arguments contained in his Archaeology of knowledge (Foucault, eng. ed. 2002) regarding the use of the notion of tableau. While in the lecture the notion of œuvre is presented as the duplication of the author (as the construct that is a substitution of author, granting him a reassuring unity), in the opposite way, the notion of tableau in the historiographical reflections of Archaeology is precisely indicated as an open “series of series”. In a polemic and somehow enigmatic note in his introduction12 Foucault wrote:

“Is it necessary to signal to the last flâneurs that a tableau (and in all the senses of the term), is formally a “series of series”? Anyway, it is not at all a small fixed image that is placed in front of a lantern, to the biggest deception of the children, who of course prefer the vivacity of cinema.”  (Foucault, 1969 : p.19 [my translation])

The Tableau is a figure of intelligibility for different historical dimensions and temporal orders. Far from constructing a unified principle, the tableau is meant to show the space of dispersion. It functions as a model of articulation of different historical series that, instead of constructing or fixing its coherence, interrogate its limits and also its mobile and unequal correlations.

The problem of Form

The two aspects that we have followed - the suppression of discourse privilege and the problematization of representation - are integrated in the Foucault’s article of 1967, as the problem of form. Panofsky's essays present the problem of  “form” in the complexity of reciprocal operation between themes and plastic motifs, in which the first, characterised by discursive elements, gain their materiality in the latter. However this reciprocal operation remains always frail and reversible. The problem of form, that is connected with the historiographical concerns of Foucault (as it is demonstrated by his text on Cassirer of 1966 [Foucault, 2001c]13) and which he recognizes to have risen in XIX century art history, presents itself as the necessity to overcome the discourse on the possibilities of language, in order to bring out the complex texture of the games of the visible and the speakable. 

To figure out this operation we must leave behind any static point of view and try to think in terms of strategic relations and reciprocal effects. It is important to underline, in this respect, the use of two expressions recurring in the text: on one hand the word “manifestation” (that is used in reference to the scholastic principle of manifestatio); and on the other hand the expression “donner lieu” mentioned above. These two terms emphasise the importance of exteriority, and they underline the fact that we are not dealing with regularity as laws, combinations, linear relations, but with functions and processes of operation, productive games incorporated in plasticity.

This problem of form is absorbing Foucault in the moment in which he is preparing his analysis of discourses as practices. The work of Panofsky is in this sense a signal, presenting the relation between form and representation as a complex game of mise en oeuvre in which different elements are involved: conventions and formal rules; skilful techniques but also systems of sensitivity and values. Therefore we can indicate Foucault’s high appreciation of those kind of historical studies which open the possibility to analyse or, rather, try to outline the different components of the knowledge that passes through the substance of the plastic work.

In fact, Foucault will also dedicate a page in the last part of the Archaeology of knowledge, to an hypothetical archaeology of painting that:

 “would try to discover whether space, distance, depth, colour, light, proportions, volumes, and contours were not, at the period in question, considered, named, enunciated, and conceptualized in a discursive practice; and whether the knowledge that this discursive practice gives rise to was not embodied perhaps in theories and speculations, in forms of teaching and codes of practice, but also in processes, techniques, and even in the very gesture of the painter” (Foucault, 2002: p. 214)

The comparison of this quote with the early article of Panofsky “On the problem of description and interpretation of plastic arts” (Panofsky, 1992)14, that was, in all probability, unknown to Foucault (in 1967 he affirms to know Panofsky’s works from their French translation), will have a unexpected effect. In his article Panofsky says that the analysis of an artwork “only reached its real aim when all of the effectiveness-moments (so, not only the representational and iconographic ones, but also the purely formal factors of light and shadow distribution, of the surface structure, and even the stroke of the brush, of the chisel or of the graver) are detected and reported as “Document” of a Weltanschauungssinn”.15 (Panofsky, 1992: p. 218 [my translation])

Taking care not to forget the vastness that separates the words of the two thinkers and without any intention to cover the differences of the attitude and the ambitions, we are forced to observe an astonishing proximity. From this proximity, from the possibility of a theorisation of the historical analysis of plastic productions that takes into account their connections with discursive practices trough different dimensions, we are dealing with an approach that will focus not only on the question of the production of meaning but on the effects, that will soon become political effects, of the game between images and language. The possibility of this relocation is illustrated by a juxtaposition of these two quotations, in the middle of which we can place the attention on the formal factors focused on their inscription in the productive activity up to the most minute gesture. 



1 I refer to the hesitation of the French panorama of foucauldians studies in taking into account the importance of the visual aspects in the research of Foucault in an organic way, nevertheless the recent publication of the scénario about Picasso and the notes on Manet, in the volume of Herne (edited by P. Artières, J.F. Bert, F. Gros, J. Revel, 2011), demonstrates an important opening in this direction. In the Anglo-Saxon’s Visual Studies, on the other hand, many scholars where and are engaged in showing the importance of Foucault's reflections for this field of research, among those we can mention: Martin Jay, that seems to maintain a certain ambiguity in his position in order to defend the vitality of the debate (Jay,1994; Jay, 2007); Gary Shapiro (Shapiro, 2003); John Rajchman (Rajchman,1988) and Jonathan Crary (Crary, 2001) that uses the instruments elaborated by Foucault together with others conceptual influences (especially Benjamin) in order to delineate and conduct his historical-critical analysis. I would like also to indicate the work of T.J. Tanke that aspires to construct an exhaustive analysis on Foucault's relation with visual art as a contradictory but organic philosophy of art (Tanke, 2009). The Italian debate on those topics distinguishes himself with a certain vivacity with authors as Miriam Iacomini (Iacomini, 2008); Diego Melegari; Stefano Catucci and Michele Cometa (Cometa & Vaccaro ed.: 2007). In this respect see also the forum “literature and Art in Foucault” on the web site of the revue Materiali Foucaultiani: URL:

2 I would like to mention also the engagement of the philosopher Bernard Teyssèdre in introducing Panofsky’s thought to the French public. He translated, presented and edited Essais d’Iconologie, published in 1967; L’ouvre d’art et ses signification in 1969, and he wrote an article in 1964 entitled “Iconologie: Reflexions sur un concept d’Erwin Panofsky” (Teyssèdre:1964).

3 Among the most important scholars we can indicate André Chastel and Luis Grodrecki that publish some articles on Diogene and Critique, (Grodecki: 1951; 1952).

4 Cf. “Histoire de l’art et sciences humaines” and “La maîtrise d’Erwin Panofsky”, articles written by Chastel in 1968, in which he denounces the scarceness of French translations of Panofsky’s works available until 1967, but is also polemical about the “presentation” of the publications one year before, (Chastel, 1980 [collected in]: pp.99-110). See also the interview with André Chastel, (Chastel: 1991), for a general retrospective point of view of the situation of the field at the time.

5 See the contribution of Roland Recht at the conference “Relire Panofsky” in 2001: (Recht: 2008).

6 See the Avant-propos of  «L’œuvre d’art et ses significations » by Bernard Teyssèdre (Teyssèdre, 1969).

7 Cf. Relire Panofsky: cycle de conférences organisé au Musée du Louvre du 19 novembre au 17 décembre 2001.

8 The process of “Pseudomorphosys” is introduced by Panofsky in Studies in Iconology, he illustrates it in two essays: “Father time” and “Blind cupid” (Panofsky, 1972: pp. 69-95). The notion of Pseudomorphosys is defined as follows: “Certain Renaissance figures became invested with a meaning which, for all their classicizing appearance, had not been present in their classical prototypes, though it had frequently been foreshadowed in classical literature. Owing to its mediaeval antecedents, Renaissance art was often able to translate into images what classical art had deemed inexpressible”. (Panofsky, 1972: p. 70)

9 “Il arrive en effet que les résultats concrètement obtenus soient plus riches, plus ouverts que les intentions abstraitement professes” (Didi-Huberman, 2008: p.74).

10 For an epistemological analysis of Panofsky’s method in comparison with the conceptual instruments of Cassirer and Warburg, and for the use of the notion of “cultural cosmos” see: “Art, histoire et signification. Un essai d’épistémologie d’histoire de l’art autour de l’iconologie d’Erwin Panofsky” (Rieber: 2012).

11 For more information see the debate of Foucault with Raymond Aron: The 17 March 1967. Aron invited Foucault to his seminar at Sorbonne and tried to identify the concept of épistème to the notion of Weltanschauung, causing the refusal and the disappointment of Foucault, their discussion anyway was overdetermined by questions of academic politics. See the Chronology of Dits et écrits (Foucault: 2001, p. 39) and  the chapter “La mer au large” in the Foucault’s biography by Dider Eribon (Eribon: 1991).

12This note is absent in the English version, the polemical correction is directed against  Sartre's critics followed the publication and the great success of The order of things, cf. “Les Mots et les Choses de Michel Foucault. Regards critiques 1966-1968” (Artières [edited by]: 2009).

13 Cf. also: Muriel van Vliet: “L’histoire de l’art  un paradigme pour penser la logique des sciences de la culture? Autour des épistémologies d’Ernst Cassirer et de Michel Foucault” (Vliet, van: 2012)

14 “Zum Problem der Beschreibung und Inhaltsdeutung von Werken der bildenden Kunst” (Panofsky:1992).

15 “Sie hat erst dann ihr eigentliches Ziel erreicht, wenn sie die Gesamtheit der Wirkungsmomente (also nicht nur das Gegenständliche und Ikonographische, sondern auch die rein formalen Faktoren der Licht- und Schattenverteilung, der Flächengliederung, ja selbst der Pinsel-, Meißel- oder Stichelführung) als „Dokumente“ eines einheitlichen Weltanschauungssinns erfaßt und aufgewiesen hat.” (Panofsky: 1992).

An English translation of this essay, with the title “On the Problem of Describing and Interpreting Works of the Visual Arts” is available since 2012 (Panofsky, 2012). Nevertheless, I chose to use my translation because the argument of the quotation seems to be reversed in the version by Jas Elsner and Katharina Lorenz: “It is the ultimate task of interpretation to reach into this level of intrinsic meaning. For it has only reached its actual objective if it covers the aggregate of the different aspects of agency (not only formal factors such as the distribution of light and shadow, or the structuring of surface planes, or even the use of easel, chisel, or engraving tool) and accounts for all these as documents of a unified meaning related to a particular worldview.” (Panofsky: 2012)