institutions and ways of seeing 1 discourse and visual culture: a reprise The previous chapter began with a brief introduction to the work of Michel Foucault, and suggested that there are two methodologies that have developed from his work. Although these two are related and overlap -most particularly because both share a concern with power/knowledge as it is articulated through discourse - these two methodologies have tended to produce rather different sorts of research. The first type of discourse analysis, discussed in Chapter 6, works with visual images and written or spoken texts. Although it is certainly concerned with the social positions of difference and authority that are articulated through images and texts, it tends to focus on the production and rhetorical organization of visual and textual materials. In contrast, the second form of discourse analysis, which this chapter will explore, often works with similar sorts of materials, but is much more concerned with their production by, and their reiteration of, particular institutions and their practices, and their production of particular human subjects. This difference can be clarified by looking at how two exponents of these two kinds of discourse analysis use the term 'archive'. In her discussion of the first type of discourse analysis, Tonkiss (1998: 252) describes the material which that sort of analysis works with as an 'archive'. While Tonkiss puts the term in inverted commas, clearly aware that it carries a certain conceptual baggage, she nevertheless uses it to refer to her collection of data, and then moves on to consider what the data shows about certain discursive formations. However, a different kind of discourse analyst, like Alan Sekula (1986, 1989), would spend some time examining the archive itself as an institution, and unpacking the consequences of its particular practices of classification for the meanings of the DISCOURSE ANALYSIS II 1B5 things placed within it. Referring to archives of photographs in particular, he argues: Archives are not neutral; they embody the power inherent in accumulation, collection and hoarding as well as that power inherent in the command of the lexicon and rules of a language . . . any photographic archive, no matter how small, appeals indirectly to these institutions for its authority. (Sekula, 1986: 155) No doubt Tonkiss would agree with this comment. However, Sekula is at pains to explore the effects of 'archivalization' on texts and images in a way that Tonkiss is not. Sekula and writers like him make that analytical move because they place their understandings of discourses firmly in relation to the account of institutions given by Foucault. Archives are one sort of institution, in the Foucauldian sense, and this second sort of analysis would not treat them as transparent windows on to source materials in the way that Tonkiss seems to (see also Rose, 2000). As we have seen, several of Foucault's books examine specific institutions and their disciplines: prisons, hospitals, asylums. For writers concerned with visual matters, perhaps the key text is Discipline and Punish (Foucault, 1977). Subtitled The Birth of the Prison, this is an account of changing penal organization in post-medieval Europe, in which alterations to the organization of visuality (and spatiality) are central. The book begins by quoting a contemporary account of a prolonged torture and execution carried out as a public spectacle in 1757. Foucault then quotes from a prison rulebook written 80 years later which is, as he says, a timetable. Foucault's questions are, how (rather than why) did this change in penal style, from spectacular punishment to institutional routine, take place? And with what effects? Through detailed readings of contemporary texts, Discipline and Punish traces this shift. By the mid-nineteenth century: The punishment-body relation is not the same as it was in the torture during public executions. The body now serves as an instrument or intermediary: if one intervenes upon it to imprison it, or to make it work, it is in order to deprive the individual of a liberty that is regarded both as a right and as a property. The body, according to this penality, is caught up in a system of constraints and privations, obligations and prohibitions. Physical pain, the pain of the body itself, is no longer the constituent element of the penalty. From being an art of unbearable sensation punishment has become an economy of suspended rights ... As a result of this new restraint, a whole army of technicians took over from the executioner, the immediate anatomist of pain: warders, doctors, chaplains, psychiatrists, psychologists, educationalists. (Foucault, 1977: 11) The prison was born. As well as a new institution and a new understanding of punishment, in Discipline and Punish Foucault describes the emergence 166 VISUAL METHODOLOGIES of a new set of professions who defined who needed punishment and who could exercise that punishment, and of a new subjectivity produced for those so punished: what he called the 'docile body'. This was the body subjected to these new penal disciplines, the body which had to conform to its 'constraints and privations, obligations and prohibitions'. A key point of Foucault's argument is that in this new regime of punishment, these docile bodies in a sense disciplined themselves. Foucault argues that this was achieved through a certain visuality (for general discussions of the role of visuality in the work of Foucault, see Jay, 1993 and Rajchman, 1988). Once defined by the new 'expert' knowledges as in some way deviant, these bodies were placed in an institution that was 'a machine for altering minds' (Foucault, 1977: 125). Foucault (1977: 195-228) expands this point, and demonstrates the importance of a visuality to it, by discussing a plan for an institution designed by Jeremy Bentham in Panopticon 1791. Bentham called this building a panopticon, and suggested it could be used as the plan for all sorts of disciplining institutions: prisons, but also hospitals, workhouses, schools, madhouses. The panopticon was a tall tower, surrounded by an annular building. The latter consisted of cells, one for each inmate, with windows so arranged that the occupant was always visible from the tower. The tower was the location of the supervisor, but because of the arrangement of its windows, blinds, doors and corridors the inmates in their cells could never be certain that they were under observation from the tower at any particular moment. Never certain of invisibility, each inmate therefore had to behave 'properly' all the time: thus they disciplined themselves and were produced as docile bodies. 'Hence the major effect of the Panopticon: to induce in the inmate a state of conscious and permanent visibility that assures the automatic functioning of power' (Foucault, 1977: 210). This sort of visuality, in which one subject is seen without ever seeing, and the other sees without Surveillance ever being seen, Foucault called surveillance, and he argued that, since it was an efficient means of producing social order, it became a dominant form of visuality throughout modern capitalist societies. Through its operation, says Foucault (1977: 200), in an echo of Lacan, 'visibility is a trap'. Foucault suggests that institutions work in two ways: through their apparatus and through their technologies. This is a distinction this chapter will use. However, Foucault was rather inconsistent in his use of these terms, and the distinction made here between them is clearer than that institutional found in his work. An institutional apparatus is the forms of power/ apparatus knowledge which constitute the institutions: for example, architecture, regulations, scientific treatises, philosophical statements, laws, morals, and so on, and the discourse articulated through all these (Hall, 1997b: 47). Hence Foucault described Bentham's panopticon as an apparatus: at once institutional an architectural design and a moral and philosophical treatise. The institu- technoiogies tional technologies (sometimes difficult to differentiate from the apparatus) are the practical techniques used to practise that power/knowledge. DISCOURSE ANALYSIS II 167 Technologies are 'diffuse, rarely formulated in continuous, systematic discourse ... often made up of bits and pieces ... a disparate set of tools and methods' (Foucault, 1977: 26). An example might be the design of the windows and blinds in the panopticon. It has been argued by some historians of photography that it must be understood as a technology in this Foucauldian sense. John Tagg, for example, writes: Photography as such has no identity. Its status as technology varies with the power relations that invest it. Its nature as a practice depends on the institutions and agents which define it and set it to work . . .. Its history has no unity. It is a flickering across a field of institutional spaces. It is this field we must study, not photography as such. (Tagg, 1988: 63) For Tagg, photography is diffuse; it is given coherence only by its use in certain institutional apparatuses. He elaborates this claim by studying photographs as they were used in the nineteenth century by police forces, prisons, orphanages, asylums, local government's medical officers of health, and newspaper journalists and publicists. Its uses in these institutions, Tagg argues, give photography its status as a unified something rather than a diffuse no one thing, and that coherent something is, according to Tagg, the belief that photographs picture the real. (Hence he is very critical of Barthes's 1982 assertion, discussed in section 3.2 of Chapter 4, that the punctum of a photograph is a trace of an uncoded referent.) The apparatus of these various institutions - the police, prisons, orphanages, asylums, local government, the emergent mass media -asserted the truth of their claims to be able to detect, punish, or cure the criminal, the ill, the orphaned, the mad, the degenerate (in part by relying on the scientific status of the discourses of physiognomy and phrenology, discussed in the previous chapter). Producing a certain regime of truth, these institutions used photography as a crucial technology through which these distinctions were made visible. The related opposite of this, as Sekula (1989) notes, was the detection, celebration and honouring of the moral, the familial and the proper in bourgeois photographic portraiture. Thus the institutional uses of photography make us think photographs are truthful pictures, not photographic techniques themselves. For Tagg, then (and see also Lalvani, 1996; Sekula, 1989), Foucault's emphasis on institutions and power/knowledge is crucial for understanding the belief that photography pictures the real. This emphasis on institutional apparatus and technologies gives a different inflection to this second kind of discourse analysis. It shifts attention away from the details of individual images - although both Tagg (1988) and Sekula (1989) describe the general characteristics of particular types of photographs - and towards the processes of their production and use. That is, this type of discourse analysis concentrates most on the sites of production and audiencing, in their social modality. In their discussion of 168 VISUAL METHODOLOGIES nineteenth-century police photography, for example, both Sekula and Tagg pay a good deal of attention to the processes used to classify, file, retrieve and use photographs of those who had been pictured as 'criminal'. Both also argue that photography was only one part of what Sekula (1989: 351) calls 'a bureaucratic-clerical-statistical system of "intelligence"', and he suggests that the filing cabinet was actually a more important piece of institutional technology than the camera. They discuss other technologies -such as phrenology and fingerprinting - that were used alongside photography, and explore other aspects of institutional apparatuses in their studies too. This means that the sources used in their accounts are as eclectic as those of the discourse analysts discussed in Chapter 6. However, certainly in the case of Tagg and Sekula, their work is held together by an insistence on the power relations articulated through these practices and institutions. Visual images and visualities are for them articulations of institutional power. This is one aspect of their work that has been criticized. For although both take care to distinguish their Foucauldian understanding of power from those who see power simply as repressive, nonetheless there is very little sense in either of their work of the possibility of visualities other than those of dominant institutions. Lindsay Smith (1998), for example, takes them to task for not looking at a wide enough range of nineteenth-century photographic practices, and in particular for neglecting the kinds of domestic photography practised by a number of women in the mid-nineteenth century. These women photographers can be seen as producing images that do not replicate the surveillant gaze of the police mug-shot or the family studio portrait: they thwart that classifying gaze by strategies such as blurred focus, collage and over-exposure. Moreover, like their discourse analyst cousins whose work was discussed in the previous chapter, there is very little reflexivity in this second type of discourse-analytic work. Ironically, considering their critique of truth claims, Tagg and Sekula both make very strong claims themselves about the veracity of their accounts. Tagg (1988: 1-2) in particular is quite scathing about Barthes, implying that Barthes's insistence on the uncoded quality of certain photographs was merely an emotional response to his search for a photograph that would remind him of his mother after she had died. 'I need not point out,' says Tagg (1988: 2) 'that the existence of a photograph is no guarantee of a corresponding pre-photographic existent.' Tagg here counterposes the self-evident ('I need not point out'), which he later expands at great length with the use of much theory, to the emotional need driving Barthes's work. As I read it, Tagg is making an opposition between his masculinized rationality and what he sees as the effeminate emotionality of the grieving Barthes. Hardly a self-reflexive strategy, I think. This chapter though will not focus on the work of Tagg or Sekula and their interest in photographic archives. Rather, it will turn to work that considers two other kinds of institutions which deal with visual objects -the art gallery and the museum - and which have also been subject to DISCOURSE ANALYSIS II 169 Foucauldian critique by writers such as Tony Bennett (1995) and Eilean Hooper-Greenhill (1992). (Other important discussions include the essays collected by Barker, 1999; Greenberg et al., 1996; Sherman and Rogoff, 1994; Vergo, 1989.) These accounts explore how visual images and objects are produced in particular ways by institutional apparatuses and technologies (as 'art', for example) and how various subjectivities are also produced, such as the 'curator' and 'the visitor'. However, these are institutions, which while of course not free from the workings of power, are not as obviously coercive as those examined by Tagg and Sekula. Their disciplines are more subtle, and thus they provide a more fruitful ground for exploring the extent to which this second type of discourse analysis can address questions of conflicting discourses and contested ways of seeing. The particular case study will be the American Museum of Natural History in New York (hereafter referred to as the AMNH), as seen by Donna Haraway (1989: 26-58), Ann Reynolds (1995) and Mieke Bal (1996: 13-56) (although Bal's account also incorporates a semiological approach). Their accounts will allow another opportunity to consider the possibility of a reflexive discourse-analytic practice. The status of the art gallery and museum as institution provides a way of examining the methodology of this second kind of discourse analysis. So, this chapter will: • examine ways of describing the apparatus of the art gallery and the museum; • examine ways of describing the technologies of the art gallery and the museum; • examine how this second kind of discourse analysis argues that these institutions produce and discipline their visitors; • assess the strengths and weaknesses of this type of discourse analysis of institutions. 2 the sources for discourse analysis II The kinds of sources used for this type of discourse analysis are as diverse as those deployed by the discourse analysis discussed in Chapter 6. A key Foucauldian account of the emergence of the art gallery and the museum as particular kinds of institutions is Tony Bennett's The Birth of the Museum (1995), and he is typical in his use of a wide range of sources. He undertakes a careful reading of the many written texts that discussed museums and galleries in the second half of the nineteenth century. These were produced by reformers, philanthropists, civil servants and curators who were all arguing, though often in different ways, for the establishment of galleries and museums that were open to the public. Studies of current discussions about museums and their practices supplement this sort of historical written source with other types of documents available now, 170 VISUAL METHODOLOGIES such as the annual reports of galleries and museums and their mission statements. Interviews with the directors, curators and designers of museums and galleries can also be used in contemporary studies. Both historical and contemporary studies often use photographs or other visual images of buildings, rooms and displays too, sometimes simply as illustrations to their written accounts, and both also pay attention to the architecture of the institution: its design, decorations, inscriptions, layout and so on. Studies of contemporary museums and galleries also often rely on visits to the institution and observation of the way people visit and work in them. In relation to studies of the AMNH, both Haraway (1989) and Reynolds (1995) are historical accounts of particular halls of that museum, and so they use written texts such as the autobiographies of curators, the minutes of museum committee meetings, scientific texts and the museum's annual reports; Haraway (1989) supplements this with an account of what the hall she is interested in looks like to the visitor now: or, at least, what it looks like to Haraway. Both illustrate their arguments using photographs of museum displays and other images. Bal's (1996) account is a reading of a few halls of the museum based entirely on their layout and the displays on show to the visitor in late 1991. (Her study is also interesting in the way it uses illustrations to make her points, as well as written text.) focus Visit a gallery or a museum. When we visit a museum or a gallery, it is somehow clear that certain things are 'the objects to be looked ať: the paintings, the objects, the items in the shop. This time, spend time looking at other things: the architecture of the building, for example, its floor plan, its warders, its other visitors. 3 the apparatus of the gallery and the museum As Stephen Bann comments, the history of museums can be interpreted: grosso modo in terms of two conceptually distinct phases. The first, roughly speaking up to the end of the eighteenth century, qualifies as a 'prehistory' in the sense that the collection and display of objects appears to answer no clear principles of ordering by genre, school, and period. The second, which represents an almost irresistible movement towards conformity over the course of the last two centuries, is a history in which the museum has developed and perfected its own principles of ordering by giving spatial distribution to the concepts of school and period, in particular. (Bann, 1998: 231; see also Hooper-Greenhill, 1992) DISCOURSE ANALYSIS II 171 Bennett's (1995) discussion of museums and galleries focuses on the second of these phases, and draws much theoretical inspiration from Foucault's Discipline and Punish: The Birth of the Prison (1977). Bennett points out that both prisons and modern museums were born in broadly the same historical period, and he argues that they deployed a similar disciplining surveillance. In making this claim, Bennett interprets his sources using the kinds of methods discussed in the previous chapter. Thus he too looks for key themes, for truth claims, for complexity and for absences (see Chapter 6, section 4.1). He pays attention to the diversity of ways in which public museums and galleries were justified by nineteenth-century commentators, noting, for example, that they were defended as an antidote to working-class men's drunkenness, as an alternative to working-class disaffection and riot, and as a means to civilize manners and morals. But his overall emphasis is very much on the way this discursive formation produced the museum as a disciplining machine: The museum, in providing a new setting for works of culture, also functioned as a technological environment which allowed cultural artefacts to be refashioned in ways that would facilitate their deployment for new purposes as part of governmental programmes aimed at reshaping general norms of social behaviour. (Bennett, 1995: 6) His concern, then, is with the power that saturated the museum and gallery, and he explores that power in terms of those institutions' apparatuses. In particular, he focuses on particular discourses of culture and science that shaped their design and practice, and also produced certain subject positions. Hooper-Greenhill (1992: 176) too is interested in the way 'new technologies and new subject positions were constituted through the adminstration of [a museum's] newly acquired material'. Bennett argues that there was a specific discourse of 'culture' which saturated the births of the museum and gallery. Using the sources mentioned in section 2, he argues that the power of museums and galleries had the same aim: both use 'culture' as a tool of social management. He notes that the definition of 'culture' used in the two sorts of institutions is somewhat different and that does produce some differences between them, especially in the sorts of objects they display. In the museum, 'culture' tends to refer to that later nineteenth-century understanding of culture as 'a whole way of life', and museums often collect objects that are meant to exemplify the way of life of particular social groups. In the nineteenth century, this meant that museums collected and displayed the artefacts of colonized peoples, but these peoples were also seen as less cultured and more natural than those of the West. (Annie Coombes, 1994 discusses nineteenth-century displays of African artefacts in European and North American museums in her book Reinventing Africa and Cathine Lutz and Jane Collins (1993) in their study of National Geographic magazine demonstrate the persistence of this notion.) Bal's (1996) account of her 172 VISUAL METHODOLOGIES 1991 visit to the AMNH emphasizes its continued articulation of imperialist, white discourse, noting that halls showing the way of life of certain third world peoples are entered directly after halls displaying stuffed mammals and birds, thus implying that certain cultural groups are closer to nature than others. Galleries, on the other hand, work with an older definition of 'culture' as that which can ennoble the human spirit, and the objects they display are those defined as Art (see the focus in Chapter 2 section 3.5 for more on this notion of Art). Such objects - usually paintings and sculpture from Western traditions - are then also constituted as 'Art', and as noble and uplifting, by being on display. Bennett also discusses, more briefly, a specific discourse of science that was part of the museum's apparatus of power. In museums, he notes, objects are always classified according to what are claimed to be 'scientific' or 'objective' principles, whether they be drawn from notions of historical progress, scientific rationality or anthropological analysis. Bal (1996) remarks that differentiations made by the complex discourse of culture are expressed in the gallery and museum that flank either side of Central Park in New York; on the one side, the AMNH, on the other, the Metropolitan Museum of Art: By this very division of the city map, the universal concept of 'humanity' is filled with specific meaning. The division of 'culture' and 'nature' between the East Side and the West Side of Manhattan relegates the large majority of the world's population to the status of static being, assigning to a small portion only the higher status of art producers in history. Where 'nature', in the [AMNH] dioramas, is a backdrop, transfixed in stasis, 'art', presented in the Met as an ineluctable evolution, is endowed with a story. (Bal, 1996: 15-16) Bennett (1995) also pays much attention to the way the architecture of museums and galleries articulated these various discourses of culture, art and science. As well as the distinction between two sorts of building - the museum and the gallery - there are the imposing facades and entrance halls of many nineteenth-century galleries and museums, for example, which were designed to be as inspiring and uplifting as the understanding of culture and science articulated within. Haraway looks at the facade of the Theodore Roosevelt Memorial - the main building of the AMNH -and considers the effects of its design: The facade of the memorial ... is classical, with four Ionic columns 54 feet high topped by statues of the great explorers Boone, Audubon, Lewis and Clark. The coin-like, bas-relief seals of the United States and of the Liberty Bell are stamped on the front panels. Inscribed across the top are the words TRUTH, KNOWLEDGE, VISION and the dedication to Roosevelt as 'a great leader of the youth of America, in energy and DISCOURSE ANALYSIS II 173 fortitude in the faith of our fathers, in defense of the rights of the people, in the love and conservation of nature and of the best in life and in man'. Youth, paternal solicitude, virile defense of democracy, and intense emotional connection to nature are the unmistakable themes. (Haraway, 1989: 27) The internal layout also echoes the discourses of science and culture. In the case of galleries, for example, paintings are hung in groups in separate rooms according to periods and (often national) schools, and this works to naturalize these periods, schools and nations, and also to produce a narrative of development from medieval painting to the present day (Bal's art production in history; see also Bann, 1998). As well as these architectural articulations, Bennett (1995) is especially concerned to examine the social subjectivities produced through these discursive apparatuses. The strong emphasis he places on how discourse produces social positions, and the consequences for how museums were designed and policed, distinguishes his study from many of those that rely on the type of discourse analysis examined in Chapter 6. He identifies three subject positions produced by the museum and gallery. First, there were the patrons of these new institutions. Thus he is clear that the emergent 'experts' on museum and gallery policy and patronage were white middle-class men, their social position produced through their claims to 'expert-ness' as well as through the larger discourses of capitalism, patriarchy and racism. Similarly, Haraway (1989: 54-8), in her discussion of the AMNH as 'institution' in the early twentieth century, carefully explores the intersecting discourses of eugenics, exhibition and conservation that were mobilized to justify the founding of the museum, and also notes those three discursive themes were all 'prescriptions against decadence, the dread disease of imperialist, capitalist, white culture' (Haraway, 1989: 55). The museum's funders were precisely representatives of 'imperialist, capitalist, white culture', and thus she too is clear on the coincidence between the discourses of the museum and the wider power relations of society. Richard Bolton (1989) offers a more recent example of the effects of exhibition patronage in his discussion of the sponsorship of an exhibition of Richard Avedon photographs at the Institute of Contemporary Art in Boston by a local department store. Second, there were the scientists and curators: the technical experts, if you like, who operationalize those discourses of culture and science in their classifying and displaying practices (section 4.5 will return to these latter practices; Bennett pays them little attention). Third, there are the visitors. The visitor with whom the nineteenth-century patrons of museums and galleries were most concerned was produced as the morally weak, probably drunk, working-class man. The contemplation of art and the appreciation of museums' knowledge was constructed discursively by these patrons as involving particular ways of visiting museums and galleries, and Bennett (1995) argues that these ways involved orderly appreciation rather 174 VISUAL METHODOLOGIES than unruly entertainment. In ways he less than convincingly demonstrates, he argues that both sorts of institutions disciplined their visitors into what were seen as civilized ways of behaving. Bennett again pays some attention to the visual and spatial aspects of museums and galleries when making this argument, examining architectural plans and noting the way that surveillance of other visitors was often built into the designs of these institutions; he also reproduces some contemporary photographs of museums and exhibitions taken from positions which he claims again articulate the surveillant quality of these spaces. He thus suggests that museums and galleries worked to regulate social behaviour by producing docile bodies. Reynolds (1995) discusses a hall of the AMNH in the 1950s, and notes how it too assumed, addressed and produced a very specific audience, again one in apparent need of education: city dwellers. Bennett (1995) also makes a distinction between the construction of the gallery visitor and the museum visitor, though. Galleries, he argues, rely on a notion of Art that always remains implicit: In art galleries [Art] theory, understood as a particular set of explanatory and evaluative categories and principles of classification, mediates the relations between the visitor and the art on display in such a way that, for some but not for others, seeing the art exhibited serves as a means of seeing through those artefacts to see an invisible order of significance that they have been arranged to represent. (Bennett, 1995: 165) Following the work of Bourdieu and Darbel (1991), who found that the visitors to art galleries were overwhelmingly bourgeois, he argues that this particular sort of Art theory is understood only by middle-class gallery-goers because only they have been allowed access to the sort of education that considers Art. This is a problematic claim and Bennett himself worries that it is too crude in the class categories it uses; nevertheless, Bennett concludes that art galleries remain obscure places to some social groups, and that this is a contradiction at the heart of their institutional apparatus. In contrast, museums often do make their classification systems as explicit; Henrietta Lidchi (1997), for example, in her account of an exhibition that opened at the Museum of Mankind in London in 1993 which sought to portray the way of life of the Wahgi people on Papua New Guinea, shows the way the exhibition admitted to its own practices of collection and reconstruction. This admission produced a visitor capable of critique, a possibility Bennett suggests is not available in galleries. However, the question of how visitors actually do look in museums and galleries is one that none of these writers address; indeed, Bennett (1995: 11) notes explicitly that he is less interested in the visitors to museums and galleries than in their institutional apparatuses. No reason is given for this absence, and it is an absence that occurs in all the studies of the AMNH. Section 4 of this chapter will return to it. DISCOURSE ANALYSIS II 175 This section's discussion of the discourses that were part of the institutional apparatus of the museum and gallery has been partial. Bennett (1995) ranges more widely in his book; for example, he explores the role of national government in funding public museums and galleries, and notes that this makes the visitors to museums and galleries citizens instead of, or perhaps as well as, docile bodies, and was therefore a potentially democratizing move. Similarly, writers on the AMNH draw on a range of institutions, practices and sites in order to describe the multiplicity of meanings residing in that institution. Haraway (1989), for example, suggests that in order to understand the dioramas in the Akeley African Hall, it is necessary to understand not only the practices of diorama and taxidermy, but of early twentieth-century safaris too, the role played in them by photography, and the wider discourses of nature, culture, patriarchal masculinity, eugenics, conservation and so on that were articulated through them. However, the broad aims of these discussions of the institutional apparatus are I hope clear. In their explorations of institutional apparatuses, these discourse analysts of institutional power/ knowledge focus not only on discourses about museums and galleries, but also on how those discourses are materialized in the forms of architecture and subject positions. Their concern is always with the intersection of power/knowledge and with the production of differentiated subject positions. 4 the technologies of the gallery and museum Section 1 of this chapter defined institutional technologies as the practical techniques used to articulate particular forms of power/knowledge: 'the techniques of effecting meanings' (Haraway, 1989: 35). Foucault described them as diffuse and disparate sets of bits and pieces, and this section will enumerate some of these bits and pieces as they work in museums and galleries. The question posed by this second type of discourse analysis is, again, what the effects of certain technologies are in terms of what they produce; and Bann (1998) insists that this question demands carefully detailed and historically sensitive empirical answers. All the studies of museum and gallery technologies discussed here focus on the public display areas of the institution in question. 4.1 technologies of display Section 3 has already touched on some aspects of how images and objects are displayed in museums and galleries, but on the large scale: how buildings are differentiated into museums or galleries, how whole rooms 176 VISUAL METHODOLOGIES are labelled and how this then classifies objects and paintings in particular ways. This section will focus instead on more small-scale techniques of display. These are usually accessed by researchers through visits to museums or galleries, or through historical documentation. In museums, several technologies of display are available (Lidchi, 1997: 172): 1 display cases, mounted either on walls or on tables. 2 open display, with no protective cover. 3 reconstructions, which are supposedly life-like scenes. The dioramas discussed by Haraway (1989) in the AMNH are a particular sort of reconstruction. 4 simulacra: objects made by the museum in order to fill a gap in their collection. Each of these different display techniques can have rather different effects, and their precise effects very often depend on their intersection with other technologies, especially written text. For example, Lidchi (1997: 173) suggests that reconstructions in museums usually consist of everyday objects put together with some kind of reference to their everyday use. Reconstructions thus depend on the presence of 'real' artefacts in an 'accurate' combination, and this makes their display seem truthful; although, as Lidchi also points out, this effect also depends on the visitor's prior faith in the accuracy of the anthropological knowledge used to make the display. Glass display cases, on the other hand, produce a truth not in relation to the apparent representational accuracy of what is on display, but in relation to the classification system of the museum. When placed in a case, an object is dislocated from the everyday context that reconstructions attempt to evoke, and is instead placed in the classificatory schema of the museum. Again though, given the truth regime of the museum as an institution, the effect on the visitor is of a truth: an analytic one this time rather than a representational one. All the discussions of the AMNH pay a good deal of attention to the social meanings produced through the 'truthful' display of exhibits in their cases or dioramas. These discussions often focus on the effects of the spatial organization of displays: how different objects are placed in relation to one another. Haraway (1989: 30), for example, says that in the dioramas showing stuffed large African mammals against painted backdrops of their natural habitat, 'most groups are made up of only a few animals, usually a large and vigilant male, a female or two, and one baby . . . The groups are peaceful, composed, illuminated . . . Each group forms a community structured by a natural division of function . . . these habitat groups . . . tell of communities and families, peacefully and hierarchically ordered. Sexual specialization of function - the organic bodily and social sexual division of labour - is unobtrusively ubiquitous, unquestionable, right.' Thus patriarchy is naturalized, she says. Similarly, Bal (1996: 40-2) looks at a glass display case in the AMNH's Hall of African Peoples which, DISCOURSE ANALYSIS II 177 according to its caption, contains objects that show the hybridization of Christianity with indigenous African religions. However, Bal notes that the display is dominated by a large carving in the centre of the case of a Madonna and child: thus 'my overall impression of this exhibit is its emphasis on Christianity' (Bal, 1996: 42). Reynolds's (1995) discussion of the Felix Warburg Man and Nature Hall in the AMNH, which opened in 1951, is an especially detailed exploration of the way of seeing invited by a particular group of displays. The displays in this hall refuse the apparent reality of the dioramas that Haraway (1989) discusses. Instead, Reynolds shows how they offer a visually and spatially fragmented, and clearly illusionistic, series of views of a landscape that draw the visitor closer in for a detailed look at each of the component parts. The effect, 'through foregrounding the very devices of illusionism', says Reynolds (1995: 99), is to transform 'the visitors' eyes into magnifying glasses, microscopes, or scalpels, which could reveal the invisible workings of a previously familiar but superficially understood natural world'. Hence the spatial organization of these displays still produces a reality effect, but it is a rather different one from those that Haraway (1989) and Bal (1996) explore. In the case of the gallery, consider how the images are framed and hung. Paintings are now very often hung in a single row around the walls of a room, inviting you to follow them round, looking at each one in turn. That is, they are hung as individual images. This is a twentieth-century practice (Celant, 1996; Waterfield, 1991); in the nineteenth century, it was very common instead for the walls of galleries to be packed almost from floor to ceiling with paintings. This change is associated with increasingly detailed modes of classification and changing notions of Art. The discourse of Art as something to be contemplated for universal truths, which section 3 of this chapter described (see also section 3.5 of Chapter 2), became widespread in the twentieth century, and it changed hanging practices. If paintings are hung side by side, it is possible to contemplate each of them individually as pieces of Art. This also has an effect on the viewer: to encourage that contemplative way of viewing (Duncan, 1995). The combination of this kind of hanging with the layout of galleries often heightens this effect. As Jean-Francois Lyotard says of the spectator at an exhibition: - the visitor is an eye. The way he looks, not only at the works exhibited but also at the place where the exhibition takes place, is supposedly governed by the principles of 'legitimate construction' established in the quattrocento: the geometry of the domination over perceptual space. (Lyotard, 1996: 167) Thus it could be argued that both the image and the viewer are individualized through this technology of hanging, and that viewers are produced as contemplative eyes and paintings as objects to be contemplated. 178 VISUAL METHDDOLGGIES focus ■'■■ What technologies of display are used in the gallery or museum you visited? Is the list of possibilities provided in this section adequate to their description? Or are there technologies of display that you want to consider? 4.2 textual and visual technologies of interpretation These sorts of display effects always work in conjunction with other technologies, especially written and visual ones. There are a number of textual technologies to consider, and they can be interpreted using the tools of the first kind of discourse analysis, described in Chapter 6. 1 labels and captions. These are a key way in which objects and images are produced in particular ways. For example, in a gallery, a painting will always have a caption with the name of the artist; it will almost always have the date of the painting and its title, and very often the materials it was made with. These apparently innocuous pieces of information nonetheless work to prioritize certain sorts of information about paintings over others. In particular, it makes the artist the most important aspect of the painting, in accordance with the notions of Art and genius examined in section 3.5 of Chapter 2; whereas Chapter 1 was at pains to suggest that there are many other aspects of an image which are much more important than who made it. In a museum, labels have similar effects; they make some aspects of the objects on display more important than others. Bal (1991: 32) notes that labels and captions at the AMNH almost always deploy a rhetoric of realism - 'realism, the description of a world so lifelike that omissions are unnoticed, elisions sustained, and repressions invisible' - which makes it difficult for visitors to question the kinds of knowledge they offer. 2 panels. Both galleries and museums often have large display panels of text in their display rooms. These often provide some sort of wider context for the objects or images on display. In the case of the exhibition discussed by Lidchi (1997), the panels were where the exhibition's practices of representation were made explicit. Panels are often more explicitly interpretive than labels and captions. 3 catalogues. Most larger exhibitions, and many galleries and museums, produce catalogues for sale. These too are part of their technologies of interpretation. Like labels, captions and display panels, though, they convey very particular kinds of knowledge. DISCOURSE ANALYSIS II 173 focus Look at the labels and captions in the museum or gallery you're visiting. What might be the effect of taking all the labels and captions away? Take two or three images or objects and invent some new labels for them. What kind of effects are you aiming for in your new text? Bal (1996) also suggests some strategies for undermining the realism of museum labels and captions. Visual technologies can also shape the effects of a museum or gallery. Museums often use photographs as part of display panels or catalogues to show what the use of an object 'really' was, or to assert the authenticity of an object on display by showing a picture of it, or one like it, in its original context of use. Galleries use photographs in display panels much less often, but their catalogues often have them, again usually as apparently documentary images. All of these visual and textual technologies can be examined using the method of discourse analysis described in Chapter 6. Read them for their key themes, their claims to truth, their complexities and their silences. 4.3 technologies of layout Section 3 has already touched on aspects of the overall layout of museum and gallery space. Here some of its smaller-scale spatial and visual effects will be explored. First, there is the layout of an individual room. As Kevin Hetherington (1997: 215) says, 'as classifying machines, museums have to deal with heterogeneity through the distribution of effects in space'. Hence the importance of the spatial organization of displays and buildings, but also of rooms. Haraway's (1989) discussion of the Akeley African Hall in the AMNH describes the effect of its spatial organization by means of an analogy: The Hall is darkened, lit only from the display cases which line the sides of the spacious room. In the center of the Hall is a group of elephants so lifelike that a moment's fantasy suffices for awakening a premonition of their movement, perhaps an angry charge at one's personal intrusion. The elephants stand like a high altar in the nave of a great cathedral. The impression is strengthened by one's growing consciousness of the dioramas that line both sides of the main Hall and the spacious gallery above. Lit from within, the dioramas contain detailed and lifelike groups of large African mammals - game for the wealthy New York hunters who financed this experience . . . each diorama presents itself as a side altar, a stage, an unspoiled garden in nature, a hearth for home and 180 VISUAL METHGDOLGGIES family . . . Above all, inviting the visitor to share its revelation, each tells the truth. Each offers a vision. Each is a window into knowledge. (Haraway, 1989: 29) Here, Haraway considers the relation established between elements in the room, and writes to convey the effect of their combination. She emphasizes the coherence of this hall, both in its spatial organization and in its effects. Hetherington (1997), on the other hand, reminds us that museum and gallery spaces can also be incoherent. Particular objects can disrupt the symmetry or the clarity of the museum or gallery layout, for example. Rooms can also be decorated in particular ways, with particular effects. In galleries of modern art, and also in galleries showing photography as art, the walls are often painted white and any seating is modern and minimal. This practice of display became common after World War II, and Duncan (1993) argues that it was encouraged by the insistence of the Museum of Modern Art in New York that that was how its big touring exhibition of post-war abstract expressionist American art should be shown. Duncan places this exhibition in the context of US attempts to assert its cultural dominance in the Cold War. The effects of this mode of display are suggested by Brian O'Doherty (1996: 321-2): 'the new god, extensive, homogeneous space, flowed easily into every part of the gallery. All impediments except "art" were removed ... the empty gallery [is] now full of that elastic space we call Mind.' O'Doherty is suggesting that the minimality of the white gallery space again produces the Art work as something to be contemplated separately from any other distractions; and again, it produces the visitor to such galleries as simply an eye unencumbered by considerations other than looking (see also Grunenberg, 1999). focus By no means all galleries have white walls, and few museums do. In the museum or gallery you visited, what other elements of decoration were important? "What about coloured wall coverings, lighting, carpet, screens, other objects? What effects did they produce? If you visited a gallery that had white walls in some of its rooms and not in others, what was the difference between the white and non-white rooms, in terms of their objects on display and the effects created? 4.4 tactile technologies One of the most important disciplines of museum and gallery spaces for visitors is the almost universal rule that you cannot touch the exhibits. This is enforced in a number of ways: objects are placed in glass cases, ropes are placed in front of paintings, warders watch visitors. Again, the Foucauldian question must be, what kind of subjectivities does this DISCOURSE ANALYSIS II 181 produce? Obviously, it produces a visitor who looks rather than touches (again). focus So far, this section has listed a number of 'bits and pieces' that are used in museums and galleries. It has focused on their possible effects in terms of the productivity of their power/knowledge; that is, on how they produce certain knowledges about paintings and objects, and certain subjectivities of visiting and curating. Does the gallery or museum you have visited use any other technologies to produce particular interpretations of its contents or visitors? 4.5 spaces behind the displays The rooms in which objects are displayed are of course only some of the spaces through which a museum's or a gallery's power/knowledge works. There are also the stores and the archives, the laboratories and the libraries, the offices and service areas. As Hooper-Greenhill (1992: 7) notes, these spaces are not open to the public (although researchers can often gain access) because they are the spaces in which the museums and galleries produce their knowledges. They are the spaces in which the museum professionals such as curators, restorers, designers and managers work; the spaces in which the classification schemes that structure the public display areas are put into practice: A division [is] drawn . . . between knowing subjects, between the producers and consumers of knowledge, between expert and layman ... In the public museum, the producing subject 'works' in the hidden spaces of the museum, while the consuming subject 'works' in the public spaces. Relations within the institution are skewed to privilege the hidden, productive 'work' of the museum, the production of knowledge through the compilation of catalogues, inventories and installations. (Hooper-Greenhill, 1992: 190) Yet very little attention is paid by Foucauldian studies of museums and galleries to these spaces and their particular technologies. Indeed Bal (1996: 16) argues that the curators and other museum staff who work in these spaces are 'only a tiny connection in a long chain of subjects' and are therefore not worth studying in any detail. Bann (1998) however demurs, and I too find this rather an odd omission. While writers like Bal (1996) and Hetherington (1997) are happy to explore the discursive contradictions of museums' and galleries' display spaces, they seem uninterested in the possibly more subversive contradictions at work in the behind-the-scenes practices that operationalize those institutions' regimes of truth. If, 182 VISUAL METHODOLOGIES as Bann (1998: 239) argues, there are 'internal contradictions built into the development of the modern museum', they too require investigation and might perhaps be best seen in these hidden spaces. focus Few of these accounts of museums and galleries deal in any detail with what are now surely two more key spaces which visitors to these institutions encounter: the shop and the café. Visit the shop and café of your museum or gallery. "What sorts of discourses are at work here? What sorts of practices? Are they connected to those of the display spaces? If so, how? If not, how not? Could you use the methods used by the discourse analysts in this chapter to examine the productivities of these spaces? 5 the visitor Sections 3 and 4 have both noted that, according to these Foucauldian accounts of museums and galleries, as well as producing the images and objects in their possession in particular ways, these institutions also produce a certain sort of visitor. This visitor is perhaps above all constituted as an 'eye': someone who sees and, through seeing, understands. Museums do this explicitly, precisely offering their objects to their visitors as a kind of educational spectacle. According to Bennett (1995), things are slightly more complicated in the case of galleries, where the knowledge that produces the 'good eye' is kept invisible in order to maintain the gallery as a space where the middle class can distinguish itself from other social groups by displaying apparently innate 'taste'. There are though more prosaic ways in which visitors to galleries and museums are disciplined. Section 4.4 noted some of these in relation to the prohibition on touching objects and images. There are many other rules about what visitors can and cannot do in galleries and museums, and these are enforced by warders. Picnicking and playing music, for example, are forbidden: the effect of this prohibition is to reiterate the 'higher', contemplative or pedagogic, aims of the institution. Other forms of discipline include the spatial routeing of visitors. Often galleries and museums invite visitors to follow a particular route, either through the layout of rooms or through the provision of floor plans marked with suggested walks (this is common for very large galleries which expect visitors with little time: routes are suggested which ensure that sort of visitor will see (what are constructed as) the highlights of the collection). Some galleries also give you a clue as to which paintings are especially deserving of this kind of viewing by providing seating in front of them. As section 3 of this chapter noted, Bal (1996) pays a lot of attention to the effects of this sort of spatial routeing of visitors at the AMNH. DISCOURSE ANALYSIS II 183 Bennett (1995) argues that there are other, less overt forms of disciplining behaviour in museums and galleries, though. From his historical work, he argues that the contemplation of art and the appreciation of museums' knowledge were expected to involve particular ways of visiting these places, and that these ways were policed not only by rules and warders but also by other visitors. That is, he reworks Foucault's discussion of the way surveillance makes the operation of power 'automatic' by suggesting that the regulation of social behaviour in these museums is conducted as much by the visitors' knowledge that they are being watched by other visitors, as it is by more obvious forms of discipline. This emphasis on the productivity of the museum or gallery as institution in relation to its visitors raises a key question though. Just how effective are these disciplining technologies? Chapter 6 noted that Foucault insisted that wherever there was power, there were counter-struggles. But a common criticism of Foucauldian methods is that they concentrate too much on the disciplining effects of institutions and not enough on the way these disciplines may fail or be disrupted. This is a criticism which can be made of nearly all the accounts of museums and galleries cited in this chapter. The previous section remarked on their frequent uninterest in exploring the working practices behind the scenes in museums and galleries, for example. It seems to be assumed that in those spaces, classifying systems and rhetorics of realism are successfully coherent, even by those writers who question its success in the more public spaces of these institutions. Similarly, few of these studies consider the possibility that visitors may be bringing knowledges and practices to the museum or gallery that are very different from those institutions' knowledges and practices. Bennett (1995: 11) is quite clear that this is not an issue his book is concerned to address: My concern in this book is largely with museums, fairs and exhibitions as envisaged in the plans and projections of their advocates, designers, directors and managers. The degree to which such plans and projections were successful in organizing and framing the experience of the visitor or, to the contrary, the degree to which such planned effects are evaded, sidestepped or simply not noticed raises different questions which, important though they are, I have not addressed here. (Bennett, 1995: 11) Hooper-Greenhill's (1994) book on Museums and their Visitors focuses on recent attempts by museums and galleries to attract more visitors by increasing the relevance of their displays to potential visitors' lives (and suggests in passing that this involves the decentring of curatorial power), but says little about how visitors respond to their efforts. This neglect parallels the critique made by Smith (1998) of the Foucauldian histories of photography offered by Tagg (1988) and Sekula (1986, 1989). There too, the diversity of engagements with particular fields of power/knowledge is underestimated. 184 VISUAL METHODOLOGIES There are a few exceptions to this neglect of visitors as subjects constituted through discourses other than those of the museum or gallery. There are a number of case studies that have focused on exhibitions which have been especially controversial (see, for example, Lidchi, 1997). Several recent exhibitions displaying the artefacts of native peoples, for example, have been heavily criticized for their continued naturalization or exoti-cization of those peoples, and Elsbeth Court (1999) discusses both this accusation and some artistic and curatorial responses to it in a case study of displays of art by Africans. However, much less attention has been paid to less organized forms of resistance to the museum and gallery's disciplines. One exception to this general neglect is the study by Gordon Fyfe and Max Ross (1996); they interviewed a range of people who visited museums in Stoke-on-Trent, England, in order to explore the particularities of their ways of seeing. Their study invites more general questions about the visitors to museums and galleries. Do they critique the particularity of the sort of knowledge about Art offered by a gallery, for example? If so, how? Through their own experience? Through boredom? Through more formalized kinds of understanding, wondering why almost all the artists produced by galleries as great were men, or white? Do visitors touch objects on display surreptitiously? Do they find routes around museums they shouldn't, or sneak a sandwich while a warder looks the other way? And what are the effects of these possible strategies on the visuality and spatiality of the museum and gallery, and on their paintings and objects? These sorts of questions are not made impossible by this second type of discourse analysis, but they have been pursued only very rarely. Hence none of these studies offer any methodological clues as to how such questions might be answered. focus This section has noted the consequence of the emphasis in this second kind of discourse analysis on the institution rather than the visitors. What did your visit to a gallery or museum suggest about the power of the institution over its visitors? Did all the visitors you saw behave 'properly'? If not, how not? Were there certain groups allowed to behave differently - children, for example? How were any deviations policed, if at all? 6 discourse analysis II: its own productivity This second type of discourse analysis follows Foucault in understanding visual images as embedded in the practices of institutions and their exercise of power. It thus pays less attention to visual images and objects themselves than to the institutional apparatus and technologies that DISCOURSE ANALYSIS II 185 surround them and which, according to this approach, produce them as particular kinds of images and objects. This approach is thus centrally concerned with the social production and effects of visual images, and to that extent conforms to one of the criteria set out in Chapter 1 of this book for a critical visual methodology. It offers a methodology that allows detailed consideration of how the effects of dominant power relations work through the details of an institution's practice. However, this type of discourse analysis pays little attention to the specific ways of seeing invited by an image itself. Nor, as sections 4.5 and 5 have noted, has it paid much attention to the way that 'power is exercised from innumerable points, in the interplay of nonegalitarian and mobile relations' (Foucault, 1979: 94). Foucault's own arguments do not rule out this latter as a topic of research, but it has not so far been developed by these Foucauldian analysts. Finally, there is the question of reflexivity. The kind of discourse analysis discussed in this chapter does not spend time on reflexive contemplation. This is no doubt for the same reasons as section 5 of the previous chapter outlined: many of the assumptions underlying the conventional forms of reflexivity in the social sciences are not tenable within a Foucauldian framework. However, unlike the 'certain modesty in our analytic claims' nonetheless advocated by Tonkiss (1998: 260) in her discussion of the first type of discourse analysis, discussed in section 5 of the previous chapter, this second type of discourse analysis tends, if anything, to the immodest. The introduction to this chapter noted as an example of this analytical self-confidence the stinging critique of Barthes made by Tagg (1988). But all the writers on museums and galleries cited in this chapter appear equally confident that the claims they make about the effects of these institutions are correct. Haraway's (1989) essay, for example, makes some highly coloured assertions about the effects of the AMNH's Akeley Hall that give me pause. Here's a taster of her style: Scene after scene draws the visitor into itself through the eyes of the animals in the tableaux. Each diorama has at least one animal that catches the viewer's gaze and holds it in communion. The animal is vigilant . . . but ready also to hold forever the gaze of meeting, the moment of truth, the original encounter. The moment seems fragile, the animals about to disappear, the communion about to break; the Hall threatens to dissolve into the chaos of the Age of Man. But it does not. The gaze holds, and the wary animal heals those who will look. (Haraway, 1989: 30) While Haraway here may be attempting, in the Foucauldian manner advocated by Kendall and Wickham (1999: 101-9), to give co-authorship of her encounter with the Akeley Hall to its inanimate objects, she might also be read as offering an account of the effects of the hall that is somewhat ungrounded in the details of its apparatus or technologies. 186 VISUAL METHODOLOGIES Moreover, I suspect that this sort of writing makes the AMNH a lot more exciting - and powerful - than it is to the vast majority of its visitors. Hence, this second form of discourse analysis focuses very clearly on the power relations at work in institutions of visual display. However, this focus produces some absences in its methodology too: an uninterest in images themselves, a lack of concern for conflicts and disruptions within institutional practices, a neglect of the practices of viewing brought by visitors to those institutions, and a lack of any form of reflexivity. 7 summary • discourse analysis II focuses on the articulation of discourses through institutional apparatuses and institutional technologies. • in doing this, it utilizes similar methods to discourse analysis I. . discourse analysis II pays much attention to the powerful discourses that saturate institutions and apparatuses to produce their subjects. . discourse analysis II is less interested in the site of the image itself, and in practice seems uninterested in the complexities and contradictions of discourse. • discourse analysis II is not concerned with reflexive strategies. further reading Henrietta Lidchi (1997) provides a detailed study of a particular museum exhibition that is carefully grounded in the details of the exhibition's apparatus and technologies, and also makes some connections with other methods of looking at museum and gallery spaces, while Mary Anne Staniszewski (1998) discusses the effects of different display practices at the Museum of Modern Art in New York from 1921 to 1970. 1 introduction This chapter ends the book by both rehearsing its central themes and by introducing some new considerations. Each chapter has explored a particular method for interpreting a particular kind of visual imagery, and the first section of this chapter will compare the methods a little more systematically than previous chapters have done. For each of these methods has its strengths and weaknesses not only in relation to the criteria for a critical visual methodology laid out in Chapter 1, but also in terms of what it is most effective in exploring empirically. These empirical focuses do not concern the kinds of visual images on which each method can be deployed. Although most chapters have concentrated on only one sort of visual image, every method discussed here can be applied to images other than the sort discussed in that method's chapter. Rather, the specificity of the empirical orientations of these methods concerns the sites and modalities of visual meaning-making, and this specificity leads to the two other considerations of this chapter: the possibility of mixing methods, in order to broaden the empirical scope of a study; and of using other methods in order to access issues otherwise neglected by the methods so far discussed. Of these neglected issues, perhaps the most obvious is the site of audi-encing. Although several of the methods discussed in previous chapters make certain assumptions and claims about the effects of images on audiences, none of them are directed exclusively at the site of audiencing and none of them aim to explore the ways in which audiences make sense of images. This final chapter will therefore also examine ways of exploring this site. Thus this chapter will: • briefly rehearse the arguments of Chapter 1 concerning the sites and modalities of the meanings of visual images, and place the methods so far discussed in relation to them.