from language and discourse to power and vision but, more fundamentally, from around the ways in which Foucault understands the key relationship between saying and seeing in the shaping of power-knowledge (see Deleuze 1988).
According to Deleuze, Foucault’s early work is principally concerned with issues of saying and with the emergence of discourse from speech as something independent of the speaking subject. To try to summarize his position: discourse becomes the “outside” of speech that is other to the speaking subject while being at the same time the space in which subjectivity is itself defined as such (Foucault 1990). In many respects, Hooper-Greenhill’s analysis of the changing épistème in relation to the museum is also an analysis of the changing face of subjectivity constituted through the varying museum discourses of the three épistèmes she describes (see also Hetherington 1999). What this approach of Foucault’s to power and subjectivity leaves out, Deleuze argues, is an understanding of the operation of the nondiscursive, notably with what is made visible and knowable through the operations of power which, he suggests, haunts all of Foucault’s work (Deleuze 1988, 32). This is still the terrain of discourse, but of discourse operating not through texts but through the materialities and figures or technologies of power.
The interest in the nondiscursive, with the nondiscursive environment (what others have subsequently called materiality), is very much the preoccupation of Discipline and Punish. In employing the example of the prison (rather than the school, factory, or other institutionalized apparatus of power), Foucault was simply making use of the most obvious and extreme “diagram” of a disciplining institution, in which the nondiscursive operation of power on subjects can be most clearly seen (on diagram, see Deleuze 1988, and in this context Hetherington 2011). In this respect, Bennett is right to observe that the museum does not operate in the same way as a prison: it isn’t a carceral institution in the same manner. And yet, by employing the idea of the technology of power derived from Foucault’s reading of the diagram of the panopticon with its central focus of discipline, Bennett still uses it to inform our understanding of the museum’s foremost function as disciplining institution. It is not, I would argue, that Bennett is wrong in his reading of the museum but that, in drawing as he does on this aspect of Foucault’s work from his later work, he transforms one element of the operation of the museum – its disciplinary function – into a definition of the museum project as a whole. Museums may have had such a function but, I would argue, they operate in other, less clearly defined, ways too (see also Witcomb 2003). If Hooper-Greenhill, following the early Foucault, overemphasizes the articulation of discourse in the shaping of knowledge, then Bennett, following the later Foucault, can be said to overemphasize the technologies of vision and the operation of disciplinary power through a delimited nondiscursive terrain.
If we want to find an alternative take on the museum, one that retains a recognition of the key observations around discourse and power from both of these eras of Foucault’s work (and the subsequent readings they have influenced) but which offers different readings and possibilities, we need to seek it in the missing and more transitional period in Foucault’s writing – the time where he did, albeit fleetingly, speak of the museum himself (see Hetherington 2011; 2014). Of course, there is no book by Foucault on the museum in the way there is on the asylum, the clinic, or the prison. There are, however, two places where Foucault speaks directly on museums, both from essays written in 1967 when he was just beginning his ill-fated and unrealized study of the painter Édouard Manet and thinking about the relationship of modern art to knowledge constituted within the gallery (see Foucault 2009; Shapiro 2003). What we can add, though, is that there is a book on the archive, from the same period as these museum fragments, that contains suggestive ideas on how Foucault might be usefully applied to the museum in relation to his comments on it from this period – The Archaeology of Knowledge (1974).
Before turning to that, it is important to see first what Foucault did have to say about the museum. The first fragment on the museum is in an essay, written as an afterword, on Flaubert’s novel The Temptation of Saint Anthony (Foucault 1998b). In it Foucault makes reference to two leading nineteenth-century French artists, the novelist Gustave Flaubert and the painter Édouard Manet, and how they have produced novels and paintings that are both directly engaged with the nineteenth-century project of the archive – the library in Flaubert’s case and the museum in Manet’s. A key passage on the function of museums is worth repeating here (for other discussions see Donato 1979; Crimp 1993; Shapiro 2003):
Dejeuner sur l’herbe and Olympia were perhaps the first “museum” paintings, the first paintings in European art that were less a response to the achievement of Giorgione, Raphael and Velasquez than an acknowledgement … of the new and substantial relationship of painting to itself, as a manifestation of the existence of museums and the particular reality and interdependence that paintings acquire in museums … Flaubert is to the library what Edouard Manet is to the museum. They both produced works in a self-conscious relationship to earlier paintings or texts – or rather to the aspect in painting or writing that remains in definitely open. (Foucault 1998b: 107)
This suggests a particular kind of space in which things become visible and around which the discourses of art and literature are formed. As Gary Shapiro (2003, 223ff.) has suggested, Foucault’s thoughts on the museum here were probably influenced by one of the leading commentators on museums within French society after World War II, André Malraux. Malraux was very much a celebrator of the museum both as a universal space in which all cultures from all times could be brought together, and as out of time, for the sake of learning and appreciation (Malraux 1978). He also thought that such an idea could be extended beyond the physical space of a museum through the ability to mechanically reproduce images of art through the use of photographs in art books. In a manner not dissimilar to Walter Benjamin’s essay on the impact of the mechanically reproduced image on the aura of museum art ([1934] 1973), Malraux, writing in 1947, was therefore a champion of the museum project, very much against the grain of over a century of French critique from Quatremère de Quincy to Paul Valéry (see Adorno 1967; for an overview see Maleuvre 1999). Malraux believed that the ability to reproduce artworks in print made them accessible to all. For him, the reproduction of images also universalized the museum principle as a musée imaginaire, or “museum without walls,” allowing people to experience art from across space and time in a single imaginary space that provided opportunity for comparison and new insights into the universal creativity and cultural understanding of Man (sic) across the ages (Malraux 1978).
Foucault would unquestionably have been an opponent of Malraux’s humanism, given the themes of his then major work, The Order of Things ([1966] 1989). He is also far more ambiguous than Malraux in his reception of the museum. However, his vision in this fragment of the modern library as an enclosed but infinite archival space in which objects can be seen diversely in relation to each other, certainly bears at least a passing resemblance to Malraux’s model of the musée imaginaire, at least in formal terms.
Writing some 20 years before Foucault, Maurice Blanchot, one of the key inspirations on his thought at this time (see Foucault 1990), had made a similar, albeit more critical, point about the museum space, in direct response to Malraux’s celebration of the open museum principle. Rather than a space without walls, Blanchot suggested, the museum (and he used the art museum as his model, as does Foucault) is a “monad without windows” (1997, 22). For Blanchot, the museum makes art accessible but it does so in a way that allows art to lose its radical alterity to tell us something different about the world. What art does in the museum, he suggests, is tell us the story of museum art (schools, styles, traditions, developments, etc.). Foucault is less overtly pessimistic than Blanchot on the stultifying effect of the museum on the art that it houses but, for him, it is nonetheless an enclosed archival space of power. The question is how it might operate.
Deleuze suggests that this period of Foucault’s work is still dominated by an interest in the discursive at the expense of the nondiscursive, as in his earlier writings. He singles out The Archaeology of Knowledge to make this point (Deleuze 1988, 32). However, in this
