precisely to grasp the significance of a visual technology of power in the library/museum in relation to the discourses articulated there in cultural form. The model for the library/museum is not the prison but the archive, that visual space in which all discourse is housed and made available.
Despite what Deleuze has to say about this period in his work (and it is true that there is no understanding of the visual technology that surrounds the nondiscursive operation of power, as there is in Discipline and Punish), there is in the Archaeology a key section where Foucault does try to understand precisely the relationship between saying and seeing. It is to be found in the section where he introduces us to the idea of a surface of emergence (1974, 41ff.; see also Elden 2001 on this theme in Foucault’s work). In the section where he discusses the formation of discursive objects Foucault suggests that there are three practices that need to be identified in the making of any understandable discourse around a discursive object: surface of emergence, authorities of delimitation, and grids of specification (Foucault 1974, 41–42). In the first, surface of emergence, Foucault is interested in the space in which otherwise diverse things first become visible and knowable as a common set of discursive objects. They may have been seen before but they haven’t been known together before this emergence became apparent. The second, authorities of delimitation, relates to the professional individuals who assume the authority to speak about the objects that have emerged and who shape the discourse around those objects. The third, grids of specification, are the forms of discursive knowledge that these professionals use to order and classify the discursive object over which they have claimed authority.
Foucault’s own example to illustrate this understanding of a visual apparatus surrounding the formation of a discursive object is that of madness in relation to the clinic and psychiatry in the nineteenth century: the surface of emergence, in this case, refers to the visible manifestation of different conditions of insanity in the clinic; the authorities of delimitation are those in the medical profession who are able to speak and write about what they observe; and the grids of specification are the ways in which they order, classify, and understand different kinds of madness as claims of truth about them. This is very different from the museum in terms of its discursive object, but the process of ordering and classifying in order to make truth claims about phenomena could be said to reflect the key functions of the modern museum quite clearly: it makes visible a set of things that, when brought together, can have apparent familiarity (pace Malraux); this discourse has a series of professional curators, conservations, and catalogers who together shape it around what has come to be seen; and it has forms of knowledge relating to such things as style, school, period, epoch, genus, and so on through which we come to understand the world and its past through its varied forms of material culture (organic or inorganic). The formation of such discourses takes place, Foucault suggests, in self-enclosed interior spaces (1974, 76). The model here is not the prison or the clinic but the archive.
What Foucault is principally interested in here is the process of making knowable what is already visible. If we read the Saint Anthony fragment in light of the position developed in The Archaeology of Knowledge, we get a glimpse of how the museum might be said to function as a surface of emergence, within a specific delimited field (and the museum has many, from the universal to the highly specific, natural as well as cultural).
It is around this theme of emergence that we might finally begin to understand what Foucault meant in the other fragment where he speaks of a museum as a heterotopia (1986; 1998a; see also Bennett 1995; Hetherington 1997a; 2011; Lord 2006). Originally given as a talk and never developed systematically (see Johnson 2006), his examples of heterotopia (other places/places of otherness/emplacements of the other – it is never fully made clear which, and there are elements of each interpretation in this unfinished essay) are diverse, ranging from brothels to old people’s homes to asylums. Among them he includes the museum as an example of a type of heterotopia that he calls a heterochronia – a space that brings into view the otherness of time all in one place in contradistinction to the flux of time in modernity (1998a, 182). What he says about museums here is also a general statement about the archive: museums – and he also includes libraries – endlessly accumulate times in one space through the material objects they contain and the knowledge associated with them (Foucault 1998a, 182). In so doing, they constitute themselves outside of time, “protected from its erosion” (182). In this way, the museum becomes both a product of modernity and also immune from its conditions of change and flux, and this allows for the uncertainties of the modern world to be measured against them.
Bennett’s brief use of heterotopia in relation to the museum at the opening of his Birth of the Museum (1995, 1) draws out Foucault’s suggestion that museums, which accumulate all times, stand in contrast to spaces of festival, like fairs, in which time is fleeting. He goes on to see this approach to time within the museum’s exhibitionary complex and approach to spectacle as part of its disciplinary power, which contrasts with the temporality and form of exhibition in the undisciplined fair. But there is another reading. What Foucault also says at the outset of this essay is that the second principle of thermodynamics, entropy, is the engine of nineteenth-century time-focused thought, out of which institutions like the museum emerge (1998a, 175). As Donato (1979) has pointed out, the fear of stasis and decadence haunted the nineteenth-century imagination, and ideas about progress and improvement, which were very much articulated by the great nineteenth-century museums, were used to stave off the fear that modern society was in a state of decay. Foucault sees museums and their approach to time as a mirror in which society can see itself. He doesn’t say whether what it sees in the museum is good or bad, though he recognizes that all heterotopia, including the museum, have a utopian function (see also Marin 1984).
In terms of Deleuze’s articulation of the question of saying and seeing, we can use these fragments of Foucault’s on the museum, alongside his observations on the surface of emergence to suggest that he poses the problem of the modern museum thus: the premise of the museum is that it is able to say new things in the space of the already seen. In that sense, it is set up as a surface of emergence for new discursive objects, new discourses and new disciplines, and potentialities of knowledge (see, e.g., Whitehead 2009). The promise of the museum is the promise of the emergence of new discourse amid the display of the already read and of the endless development and progress in knowledge. The other side of the museum is that the mirror reflects back: it is also a space in which new things come to be seen in a space of the already said (see Hetherington 1997b). It thus unsettles existing knowledge and brings it into doubt, thereby threatening to make visible the prospect of stasis, undermine serious collections and make them appear nothing but bric-a-brac (see Donato 1979; Saisselin 1985; Crimp 1993). The relationship is one of irresolvable tension between establishing an impossible primacy between the discursive and the nondiscursive, and in the interplay between the discursive and the figural (see Lyotard 1984). That is perhaps what the self-referential nature of Flaubert’s and Manet’s art is all about – an ironic statement of opportunity and disappointment in the project of the museum to realize encyclopedic understanding that will also endlessly offer up new ways of seeing things within its monadic yet endless capacity. In this dual sense, a surface of emergence is both the beginning of the establishment of relations of power through a process and also the space of the outside of power that is their unravelling (see Hetherington 1997b; 2014).
Conclusion: Seeing in the space of the already said
We have, then, three models of the museum in Foucault’s work: the abstract space of changing discourse (Hooper-Greenhill); the disciplinary space of the “open prison” (Bennett); and the space of the infinite and yet monadic archive, which is ambiguously open to both possibility and doubt in its truth claims. Foucault himself was never explicit on any of these positions; he invited all three and perhaps others besides. If the first of these positions, following Deleuze’s criticisms, can be said to place too much emphasis on the formation of discourse in the making of truth at the expense of an understanding of the role of the nondiscursive (materialities) in relations of power, then the second, associated with Discipline and Punish, can be said to rectify that; but it does it in a way that limits the possible
