Visual Methods in Social Research. Marcus Banks
sockets. Eck notes that Hindu images are imbued with life by opening the eyes with a golden needle, or by the final stroke of a paintbrush; the deity’s first glance is so powerful that it can kill a man and so the image is first shown a pleasing thing, such as sweets, fruit or flowers, or even its own reflection in a mirror (1985: 7). In some Jain and Hindu temples, especially those on busy city streets, screens are placed just inside the threshold to prevent inadvertent darshan on the part of those who are temporarily or permanently impure who may be passing by – menstruating women, for example, or those classed as untouchable. Also in India, and elsewhere in the world, fragments of mirror glass are incorporated into embroidered textiles to divert or reflect back the gaze of the evil eye. Furthermore in India, women, especially childless women, may refrain from looking too long at another woman’s child for fear of witchcraft accusations. Wherever they work, anthropologists and other social researchers need to spend as much time considering how and at what people look, as listening to how and what they say.
While vision may be a privileged sense in some Euro-American contexts, these societies are also strongly in the thrall of language – both oral and written. In many cases the use of vision and appreciation of the visual is compartmentalized or constrained, as appropriate for some contexts but not others. This containment is largely effected by language, by placing the visual and visible aspects of culture within a language-based discourse that has primacy. Such containment of social and cultural activity – of breaking up the business of living and hence the organization of society into named and categorized chunks – is a feature of many Euro-American societies, notwithstanding what others have said about the distinct character of ‘Western’ society (e.g. Jay 1993). While the appreciation of fine art is a social skill and a class-bounding diacritic (as Bourdieu et al. have pointed out [1991]), going to the cinema or taking family snapshots are straightforward cultural practices predicated upon a visual sense. These and many other activities must be enmeshed in language to become meaningful or valuable. Moreover, while cultural activities that centre on vision and the visual are valued in some contexts, they are clearly not in others. Education provides a good example. Preliterate children, and even pre-linguistic infants, are encouraged to engage with picture books, not in order to develop their visual sense but in order to familiarize them with books of words they must learn to value and rely on in later life. As Alice noted before she disappeared down the rabbit hole, the absence of pictures denotes the intended adult readership of a book. While some higher education disciplines such as art history clearly must engage with the visual manifestations of culture, others that are expressly concerned with the organization and flow of social life, such as sociology, place a far greater reliance on language both to investigate and then report on human social relations.
It is almost as though the removal of visual culture from its original context of production and its enwrapping in a discourse of ‘art’ has caused a suspicion of images in other contexts, and a consequent need to constrain and limit the work that they do. This is apparent, for example, in the contrast between the flagrant disregard for language that some artists and fine art photographers employ by captioning their works ‘Untitled’, and the apparently exegetical or descriptive captioning employed by academics (and others, such as newspaper editors) for images that are inserted into primarily written texts (see Section 2.2).
1.4 Reading narratives
The study of images alone, as objects whose meaning is intrinsic to them, is a mistaken method if you are interested in the ways in which people assign meaning to pictures. (Ruby 1995a: 5)
The idea of ‘reading’ a photograph or other visual image merely extends the range of a term normally applied to the written word and is used commonly by commentators on a whole variety of visual forms, from paintings to television soap operas. There are, however, some important differences which are not always made explicit, or which are perhaps not even recognized by some who use the term. First, although we use the term fairly casually throughout this book, we do not wish to suggest that there is a ‘language’ of images or image components that follows some kind of quasi-grammatical rules, either universally or in more socially specific contexts. Within any particular sociocultural environment, we may learn to associate certain visual images with certain meanings, but these are normally highly context dependent and often transient. In popular Indian cinema of the 1980s a sharp camera zoom in onto the face of a character (together with a musical climax) was commonly read as an indication of intense emotion on the part of the character, perhaps associated with the revelation of a hidden fact. On British television by contrast, the same camera movement would be read today as a melodramatic cliché, perhaps prompting associations with amateurish 1960s or 1970s soap opera. Sequences of images, however, or individual pictorial elements, have no inherent para-syntactic or structural association, other than that which an interpretative community – the audience – is educated to expect by convention.
Secondly, ‘reading’ to some extent implies that the ‘message’ being read lies within the visual image, that it is speaking to us and that all we need to do is listen. On the contrary, it is human beings who speak to one another, literally and metaphorically through their social relations. But, as anthropologists and others are well aware, human beings frequently displace those conversations onto inanimate objects, giving them the semblance of life or agency (Kopytoff 1986). When we read a photograph, a film or an art-work, we are tuning in to conversations between people, including but not limited to the creator of the visual image and his or her audience. Those other participants include gallery curators, television producers, aid agencies, and a whole variety of other persons who present images to a viewing/reading public.
In The Photography Handbook Terence Wright describes three approaches to reading photographs: looking through, looking at and looking behind. These approaches he associates with realist, formalist and expressive strategies of authorial intention (Wright 1999: 38 ff.). The labels in themselves do not matter here; what Wright is saying about photographs, which would hold true for any visual representation, is that a reader can consider both their content and their context. For some photographs, or in the eyes of some readers, the content is primarily a matter of information, as though one were looking through a window at some object beyond: this is my partner, this is the house where I stayed on fieldwork. In the eyes of others the way that content is presented is deemed important – the arrangement of elements, the angle of light, and so on: this tiny baby in the crook of that heavily muscled arm, lit to produce deep shadows ‘says’ something about strength and fragility, experience and innocence. With other images, or in the eyes of still other readers, it is the context within which the image was produced that assumes prominence: this image of a naked Aboriginal woman, standing in profile against a measuring rule, was taken in accordance with a now-discredited nineteenth century theory of human biological variation.
The properties of the images, and the interpretation of readers, are not fixed. The nineteenth century anthropometric photograph (reproduced in Spencer 1992: 101) was intended to be read for its informational content, but would now be read as an insight into the social, intellectual and perhaps even sexual background and interests of its unknown photographer and those like him. In what follows we focus in particular on the first and third of Wright’s approaches – looking through and looking behind – but we employ a slightly different terminology, one that stresses the element of readership or audience, and one that is concerned with the social rather than the individual construction of meaning.4 The content of an image we refer to as its internal narrative – the story, if you will, that the image communicates (see Barthes 1981: 40). This is not necessarily the same as the narrative the image-maker wished to communicate, indeed it can often be markedly different. This is linked to, but analytically separable from, what we call the external narrative. By this we mean the social context that produced the image, and the social relations within which the image is embedded at any moment of viewing.
Although these terms may be used in opposition, in practice they are of course intertwined, and elements of external narrative – information about the nature of the world beyond the photograph – are always involved in readings of the internal narrative. Shown a photograph of a woman in a white dress and veil, a man standing beside her in a morning coat, we might reasonably assume that it is a wedding