Visual Sociology The visual culture G, The visual in sociology 66 Three facets of visual data 68 Varieties of visual data 69 ■Wow can we use visual data? 70 Analyzing the visual 72 Content analysis 73 Semiotics 7i Conversation analysis 75 Technical consideration in doing photographic research 76 Vtewfinder, single lens reflex and digital cameras 77 Camera controls and other considerations 7g A few notes on videography 7g Chapter summary ___ 7„ Suggested readings 80 Exercise __ A group of tourists take pictures of the Eiffel Tower, family members gather around a Christmas tree to 'capture the moment,' a photo journalist in a war zone frantically points his camera at the horrors of war. What do these examples have in common? They all involve documenting moments in the flow of social life. Still photos, videos, cinematic images and television programs are ways through which we are regularly exposed to the visual in our daily lives. Furthermore, lower costs and more accessible technology have made it possible for many of us to create images of our own surroundings. We are no longer just consumers but also producers of visual media. In a matter of a several decades, the visual has become much more widely used and available than the text.To the chagrin of academics, most people spend more hours watching'the tube' than reading printed media. For example, when a friend recommends a good%ook, we might respond with,'111 wait till the movie comes out.' What is the impact of this trend on the qualitative study of social life? This chapter explores this question from several angles. I begin by considering the VISUAL SOCIOLOGY S3 place of the visual in the-broader cultural contexts of our time (i.e.,'What does it mean to be visually oriented?'). I then offer an overview of the history of visual work in the discipline of sociology. The next topic of interest is how 'pictorial and filmic materials' (a phrase borrowed from Ball and Smith 2001: 302) are used by social scientists. After this discussion, the chapter offers specific how-to, technical advice on the analysis and production of visual data. The visual culture This section briefly discusses the place of the visual in Western societies. Let us begin by thinking about the different pictorial and filmic images we come across in an average day. If you are like me, you start your day with the latest news on television (local and international news, the weather report, and so on). Or perhaps you read the newspaper and there on the front page, you find a photo that grabs your attention. On the way to work or school, there are advertisements everywhere, rypically portraying young attractive people consuming products that supposedly we all want and cannot do without. Of course, for most of us, getting home at the end of the day is synonymous with sitting in front of the television set and watching our favorite programs. It may not occur to you at first glance that what you watch throughout the day forms an essential part of your knowledge about your social world. Many years ago, my college roommate insisted that we did not have a TV in our apartment because he thought it would interfere with our studies. Reluctantly, I went along. Toward the end of this six-month no-TV regimen, it was glaringly clear that I was out of touch with the cultural tempo of my society. Not watching TV meant that I was culturally illiterate. For example, during this period, there was a famous commercial that had worked its way into the everyday vernacular. It showed a frail woman in her seventies at a fast food restaurant. Annoyed with the small size of her hamburger meat, she yells at the workers,'WHERE'S THE BEEF?'The question had gradually become a hip way of asking about the substance of an argument. Ignorant of all this, during a class discussion about social inequality in the U.S., I was dumbfounded when a classmate interrupted me with the question, 'Where's the beef?' 'At a grocery store, I presume.' I responded matter-of-facdy. Everyone laughed. Of course, I had no idea what any of this meant. The point is that a general awareness of the visual aspects of one's culture has become an essential criterion for full membership in modern societies. This tendency is what some researchers have labeled 'visual culture' (Evans and Hall 1999; Rose 2001).The phrase 'visuaŕculture*brings up two questions: 1. what is 'visual?' and 2. what is 'culture?' Regarding the first question, the word 'visual' for our purposes refers to pictures, video, film, TV programs and ads. The answer to the second question is more difficult since the concept of culture has been defined in many different ways by sociologists. For example, some define culture as a shared way of life and others may regard it as the 64 QUALITATIVE RESEARCH IM SOCIOLOGY everyday practices that help us interact with others and make sense of our 'social world. We will not setde this question here. Suffice it to say that the latter interpretation (i.e. culture as a set of practices and ways of understanding) is more likely to make its way into the discussions in this chapter. With these rudimentary concepts in mind, we might ask: What are the implications of a'visual culture?'An excellent discussion of this topic is offered in Rose's Visual Methodologies (2001), where she argues that the notion of a visual culture implies that: 1 Visual imagery is a powerful medium worthy of investigation. 2 The visual image is constructive of reality rather than simply being descriptive «of it. 3 Y Visual images offer particular ways of seeing social issues. 4 " The visual is embedded in a wider cultural context. 5 The influence of the image is at least in part dependent on its audience: different people view and understand images differendy Let us examine each of these premises. According to Rose, that we belong to and participate in a visual culture means the status of the visual has reached, if not surpassed, that of the written word. Many of your textbooks may use photos as a way of illustrating the text or 'jazzing up' an otherwise boring discussion. However, from the perspective of a visual culture, the pictures are just as important as the words that surround them. To put it another way, the illustrations tell their own stories. For example, consider the magazine cover photos you see at a newsstand. How often do you construct a story just from looking at the picture? A picture of a female model on a fashion magazine, let's say. You don't have to do much reading to know you are looking at a cultural symbol of beauty whose attributes are to be desired. Notice that the cover doesn't tell you about her intelligence or her professional accomplishments, but that she is exceptionally beautiful. Presumably, the model's attractiveness is inherently good, for her as a woman, for the men 'who lust after her, and for other women who should want to be like her. That is the story. Indeed, such images are designed with the explicit intent of making words seem redundant. This may seem like an obtuse way of making the simple argument that 'a picture is worth a thousand words,' but the point is somewhat more profound than that. The notion of the visual culture means that the picture is a thousand words. It doesn't have to compete with the words. In Rose's words, 'visual images can be powerful and seductive in their own right' (2001: 10). Rose's second assumption regarding visual culture is that images, still or moving, are constructive of reality, they create what we see as much as they are 'realistic'1 representations of it. Picture a war zone, for example. On one side a group of women mourning the loss of their children after a recent aerial bombing. The women are hysterical. They have to be held back by the crowd as they lunge forward, sttaining to throw their bodies atop their dead loved ones. Farther ahead, a mob of menacing-looking men is shouting slogans. They VISUAL SOCIOLOGY 65 fire their machine guns into the air with one hand and raise a clenched fist with the other. You are the photojournalist reporting this event for an international newspaper. Which sight makes a better picture, that of grieving women or the rabid men? Perhaps you decide to shoot both. But what if your editor decided there was only room for one of the two on the front page?Your decision could mean the difference between 'constructing' your subjects as victims or potential aggressors. In this way, your visual work both creates and describes the reality of what you observe. The third aspect of visual culture, according to Rose, is that images provide us with ways of seeing a particular issue. Returning to the example of a cover girl, die picture doesn't point to potential weight problems or the fact that the model might be malnutritioned. What we see, given the lighting, the clothing, and the makeup used in the setting, is physical attractiveness. Also note that in such photos the women are not involved in any activities (they are not doing anything).They simply stand there as objects to be admired, things of physical beauty. Another example of how the visual dictates a way of seeing can be found in sports car commercials that typically target young men. What you see in the TV advertisement is a young man who, after driving a certain car, becomes more exciting, more attractive to the opposite sex, more virile. In one ad run on American TV, two women are shown fighting over a man's attention at a dinner party after they notice he drives an expensive car. In this commercial the driver is never shown. His physical appearance is irrelevant, it is his car that defines his presence for the audience and the women in the commercial. Thus the visual, in this case a shiny sports car, does not just provide what we see, but it also gives us a context for seeing. Regarding the car, it is not just a means of transportation, but it becomes visually defined as an object that magically transforms a dull person into a sexy adventurer. Rose's fourth point is reflected in the phrase 'visual culrure' itself. The ways of seeing the visual described above are culturally bound. When we see a picture, we view it using our knowledge about social relations and meanings in our culture. The picture of a model on a magazine cover receives its meaning from the structure of male-female relations in Western societies. That a woman's body can be viewed as an object tells us a good deal about her place in society in relation to men. Look at your own family photos. For example, what is the place of your father in relation to the rest of the family? Is he usually in the center of the photographs, surrounded by the other members of the family? If that is the case, what cultural value can be deduced from this observation? Even the way we pose for pictures is very telling about our cultural conventions. I am sometimes irritated by the one^ťwo-three-say-cheese routine. It seems disingenuous to fake a smile by saying a word that forces your mouth into position. But from a cultural standpoint, smiling at the camera is a required friendly gesture. It tells the would-be viewer that you didn't mind being photographed and that you are pleased to share your likeness with them. On the other hand, a stern look could mean you are taking the whole thing too seriously. In the case of my eighty-year-old father and his generation, photography m ifumiiMiTt KtstflNĽH IN SOCIOLOGY was a serious business almost exclusively used for official purposes. Therefore, to this day, he insists on having his finest suit on for a photo shoot, when he - gets in front of the camera, he stands fully erect, shoulders and chin up, without a smile. My father's behavior reflects his cultural interpret tion of the occasion of having a picture taken. In the above example, we should also consider how material culture or technology plays a part in the way we pose for pictures. Early photography was slow. It took several minutes for things to be recorded on the film during which time the subject had to stay motionless in front of the camera or else the picture would come out blurred (Chaplin 1994: 204). So people like my father were trained to be stiff when they had their photos taken. Looking at the old photos of your grandparents, you may have noticed that they also appear rigid or even uncomfortable in font of the camera. What I am suggesting here is that the production and meaning of visual materials are never without a cultural context. ■Rose's last point concerning a visual culture is that it matters a great deal which audience is viewing the image. Men might look at a photo of a female fashion model with lust, whereas for women the focal point could be the model's choice of shoes or other attire. In the example of a photo from a war zone, the : person sympathetic to their cause would view the shouting men with their machine guns as resistance fighters. By contrast, to the eyes of another viewer they might seem like dangerous zealots. The visual and its creators are not sole arbiters of how an image is perceived. Instead, the visual product is interpreted in different ways depending on the audience and its cultural sensibilities. My students often recommend, or even demand, that I see a particular movie because they feel that it is a good illustration of an issue we discussed in class. Occasionally, I follow their recommendations, but many rimes I sadly report back that I saw the movie and just didn't get it. In this case, the widening rntergener-anonal gap between my students and me means that we see the same visual piece very differently. I see gratuitous violence and they see a poignant social commentary on the human condition. According to Rose, then, we cannot dismiss the capacity of the audience to view the visual m novel and unexpected ways. In summary, Rose alerts us to the significance of a growing visual culture with its own sensibilities. She explains how the visual can be much more than a method of illustrating the written text. With this general appreciation for the visual and its accompanying analytical insights, let us now probe the history of the visual in the discipline of sociology. The visual In sociology While the visual has always had a place in sociology, its use and analysis have fluctuated over the history of the discipline. More than a hundred yean ago the American Journal of Sociology, the flagship journal of the discipline, published a number of articles that used photos as data (Stasz 1979) According to Chaplin (1994: 201), the first manuscript of this type was E Blackmar's The VISUAL SOCIOLOGY 67 Smoky Pilgrims', published- in 1897. The study depicted poverty in rural Kansas using posed photographs. Yet, this early interest in the visual waned as the written word accompanied with numerical analysis became the dominant mode of sociological analysis. In a sense, statistical figures, charts, and tables became the visual centerpieces of professional sociological publications. It is worth noting that this trend was not followed in the related discipline of anthropology. In fact, among the social sciences, anthropology is a leader of the use of pictorial and filmic materials. One of the more notable visual anthropological studies is Bateson and Mead's BaHnese Character: A Photographic Study (1942).This study was exceptional in its use of photos as an integral part of the story. It juxtaposed text and the visual in a complementary way, so that one would enhance the meaning of the other. In the words of the authors, We arc attempting a new method of stating the intangible relationship among different types of culturally standardised behavior by placing side by side mutually relevant photographs.,.. By the use of photographs, the wholeness of each piece of behavior can be preserved. {Bateson and Mead 1942: xii, as quoted in Harper 1994: 404) Thus by showing a series of photos of a native ritual, for example, on one page and related text on the opposite page, Bateson and Mead encouraged their readers to see and read the story simultaneously. Following their footsteps, a number of sociologists in recent decades have revived the interest in the visual. For example, in a 1974 article Becker (1974) called for bridging the gap between photography as an art form and photography as a mode of understanding and analyzing social reality. He also promoted greater appreciation for the role of social theory in producing and analyzing photographic images (cited in Harper 1994: 406). Becker subsequently published Exploring Society Photographically (1981), an edited book that follows a visual presentation style similar to that of Bateson and Mead, Another notable sociologist that encouraged the use of visual analysis was Goffman, whose landmark sociological study. Gender Advertisements (1979), looked at how gender roles and expectations are reflected in magazine ads. Using over 500 photos, Goffman underlined the taken-for-granted nature of gender relations in Western societies. For example, he showed how magazine ads in the late 1970s, depicted men in active roles (doing things like helping patients or playing in sports), whereas the women were shown as mere spectators, passively watching the men's activities. Any discussion of important works in visual sociology should also include Denzin's contributions. In books like Images of Postmodern Society (1991) and Cinematic Society: The Voyeur's Gase (1995) he rejects the notion that cinematic representations are mere entertainment with no social value. Instead, he argues that we understand and express ourselves and our social settings through Hollywood films. According to Denzin, cinematic representations both describe social re all ties and m in da t c a way of seeing or accepting these realities. Consider, for example, his analysis of the movie When Harry Met Sally: 68 QUALITATIVE RESEARCH IN SOCIOLOGY The movie,., is a'Field Guide to Single Yuppies',.,, As such it takes a stand on and defines the following problematic terms; being single versus being married; sexuality and i women's orgasms; love, sexuality, and friendship; life after divorce, or after breaking up with a lover. These terms are presented as obstacles.... The solution^ are gender specific. Women must not be single, must learn how to fake orgasms, so that males think they have sexual power.... Men, on the other hand, must have a woman who lets them think they can make them sexually happy They need male friends to talk to, because women don't understand male sexuality. In this battle between the sexes, sex must be overcome, before love and friendship can be achieved. (Denzin 1995:117) In this analysis, the movie more than entertains; it mandates a way of thinking about male-female relationships. It becomes a how-to guide on heterosexual relations, constructing and describing the reality of how men and women shc^ild relate to one another. For Denzin, the cinematic representations berime taken-for-granted truths. Indeed, I remember many of my friends citing When Harry Met Sally as empirical evidence for the inherently conflict-ual nature of heterosexual relationships. They would say,'Didn't you see the movie? That's how we are.' Denzin s sociological analysis shows how cinematic images both construct and validate what we know about society. As a whole, the visual is making a gradual comeback in sociology. It now has its own journals (e.g., Visual Sociology and International Journal of Visual Sociology) and annual conferences. However, as Ball and Smith (2001: 305) note, where anthropology has expanded its interest in the visual into moving images and ethnographic films, the field of sociology has more or less remained tied to still photography. The following section describes ways in which images could be conceptualized in social research. Three facets of visual data One way of assessing the research value of visual materials in qualitative sociology is to think about the different aspects of a picture or a movie. According to Rose, any visual piece can be judged along at least three dimensions: 1. the production; 2. the image; and 3. the audience (2001: 16). Regarding production, we may begin by noting that visual images do not appear on the airways, in newspaper or magazines, or in theaters spontaneously. They have creators and a process of creation. As Rose puts it: All visual representations are made in one way or another, and rhe circumstances of their production may contribute towards the effect they have. Some writers argue this case very strongly. Some, for example, would argue that the teaiuoiogies used in the making of an image determine its form, meaning and effect. Clearly visual technologies do matter to how an image looks and therefore to what it mightJ8o and what might be done to it. (2001: 17) Let us evaluate the production differences between a still photo and a videotape using the example of the terrorist bombing of the Twin Towers in VISUAL SOCIOLOGY 69 New York on September MjtgOOl. Still pictures of the tragedy tend to provide snapshots of the moment of horror and peoples reaction as they stare at or rún away from the site. There is intensity in all the photos, and they are solemn and lurid. The moving images and videos, on the other hand, have a different quality. Perhaps, because we have seen many scenes like them in Hollywood motion pictures, they seem more incredible, like a violent scene from a movie that appears terrifying and unbelievable. We should also consider that videographers and photographers have differing skills and operational needs. The photographer might look for a single image that tells 'the whole story,' whereas the videographer works on constructing a narrative that resulrs from piecing together multiple action shots and sounds. Furthermore, the two bring into the field different material resources and organizational priorities. One might be taking pictures for a newspaper with a very small and politically partial readership, and the other could be videotaping for an undifferentiated mass market. Therefore, as Rose would suggest, the technique of production is a crucial part of what we see and is worthy of investigation in its own right. Another elementary consideration in choosing and analyzing visual data is the image itself. The size, composition, color, texture and other characteristics of an image all figure into how it communicates its message. For example, a close-up of someone's face conveys a very different message than, let's say, a wide angle shot that shows the whole body. In fact, close-up images may be used deliberately to imply intensity or intimacy. Another example would be the use of bright red lighting in the background of a photo or a video. For many of us, that would suggest a sense of impending danger.Thus, in addition to the conditions and necessities of production, wc can empirically study the properties of the image itself. The third area of interest in researching visual materials, according to Rose, is the audience. Putting it simply, one cannot accurately evaluate the meaning of an image without knowledge about the audience. What a seventy-year-old sees in a music video is very different from the perceptions of an eighteen-year-old.The two have different levels and kinds of visual literacy and, therefore, would arrive at different conclusions about the meaning of the work. It is possible, then, to conduct a qualitative visual research purely based on variations in audience perceptions. No doubt, creators of commercial advertisements spend endless hours carefully assessing the effects of their ads on particular target audiences. The varieties of visual data / We can categorize visual data in two ways: according to the medium's properties and according to whether they were generated by the researcher or some other agent. When attending to the medium'properties, we are essentially asking what kind of visual materials are we looking at? As stated earlier, the 70 QUALITATIVE RESEARCH IN SOCIOLOGY preferred choice for most sociologists has been still photos. One explanation for this tendency is that pictures are easier to reproduce for the purpose of print publications. Technologically, only recendy have we begun to consider the possibility of multimedia presentations either on the internet or on compact disks utilizing moving images, stills, sound, and the written text all in the same context. It is reasonable to assume that the so-called electronic revolution could motivate sociologists to experiment with other visual media. Videos and cinematic productions are a common source of data for sociologists. Naturally, working with these media requires a broader analytical framework. First, there is the issue of analyzing the effect of sound. For example, Chion's Audio-Vision: Sound on Screen (1994) is entirely devoted to how sound! or its absence, complements imagery. Another consideration, when working wi£t moving pictures, is the internal logic and narrative coherence of the piece. That is, we could ask how different scenes are pieced together to tell the story. When considering who created the visual data and for what purpose, the categories of interest will be 'researcher-generated' and 'found' images (Ball' and Smith 2001). To put it simply, researchers either create their visual images of society, or they study works produced by someone else. Examples of researcher-generated materials include photos, films, or videos created by sociologists firsthand. The main advantage of producing your own visual images is that you can be in control of what, where, and how the work is shot or filmed. You pick the subject matter, the location and the style of the production. The drawbacks of this approach are that it is cosdy and very time-consuming. Alternatively, found images (e.g., movies and magazine adds) are widely accessible and inexpensive. The range of visual data is by no means limited to moving and still pictures. Other options include animations, comics, cartoons, book illustrations, oil paintings, maps, or generally anything that is depicted visually. However, photos, videos and films appear to be the dominant sources of visual data. How can we use visual data? This section surveys some of the ways visual data is used in qualitative research. In particular, we will focus on the following approaches- 1 'Researcher-generated' visual material as the primary source of daw and method of representing the research findings 2 The visual as a complement to the written text or as a way of telling ethnographic tales 3 The^isual, particularly photographs, as a way of eliciting interviews and other data 4 Using 'found" photographic or filmic materials as secondary data for research and analysis (Adapted from Harper 2000 and Ball and Smith 2001: 314) VISUAL SOCIOLOGY 71 This is not intended as an exhaustive list. Certainly, one can combine these media to create new ones. Furthermore, rapid technological advances foreshadow the creation of new visual media and modes of representation. With these points in mind, let us briefly review the items listed above. Some social scientists, particularly in the field of anthropology, use videos and films to represent their research findings. A popular version of this approach is referred to as ethnographic films or videos. While informed by theoretical and methodological debates, such productions are not purely pedantic or illustrative of formal concepts.They combine aesthetic priorities with the auspices of scientific research to create works that inform general audiences about various social issues. For example, feminist ethnographic filmmakers have used the medium to raise public awareness about the plight of women and minorities in general. The camera, if used consciously as a tool to understand different interpretations of a culture and not as a tool to reproduce objective exotic proofs, can assist not only in revealing to ourselves our own cognitive and cultural constructions, but it can also allow others to tell their stories in their own voice, with their own voices. (Kuehnast 1990' 26; as quoted inVered 1993:179) For these ethnographers their visual productions are not necessarily an addition or a supplement to the written text. Instead, for them, ethnographic film is used as the primary means of communicating field observations to audiences. Another use of the visual in the social sciences is as a complement to the written text, as discussed earlier in the case of Bateson and Mead's Balinese Character (1942). This type of work blends the visual with the textual. The visual in this sense becomes part data, part illustration, and part analysis. Establishing such a relationship between what the readers see and read is as much art as it is social science. Another example of this kind of work is Becker's Exploring Society Photographically, which uses photographs to examine social issues and realizes Becker's (1975) call for blurring the distinction between the art of photography and the science of sociology. Similarly, Quinney (1996) uses photographs from his father's trip to California in the 1920s to tell the intimate, nostalgic story of his relationship with his father. Although Quinney's photographs are interspersed with written text, he places greater importance on the aesthetic quality of the visual. As he puts it, 'photographs are not to be subjected to 'scientific' and 'professional' discourse. Photography resists a language of analysis. The image speaks in silence. We give ourselves up to that which is beyond language and rational thought' (p. 381). Thus, the practice of combining the written text and the visual in sociology varies on a continuum tanging from aesthetic preferences to theoretical concerns about objectivity and scientific rigor. The third item from our list at the start of this section has to do with the use of photographs as a way of eliciting interviews. In my research, I used photo-elicited interviews to examine college students' perceptions of race. QUALITATIVE RESEARCH IN SOCIOLOGY Four female respondents (two blacks and two whites) were provided with point-and-shoot cameras and instructed to take ten photographs from their everyday surroundings. The goal was for them to generate^a^visual answer to the question:'What does racism mean to you?'Three of the subjects brought the cameras back after they had taken the pictures. The fourth never returned the camera or my phone calls about the project (I suppose that was her way of answering the question). After developing and printing the photos, I interviewed the research participants individually, with the photos laid out on the table. Each respondent elaborated on the meaning of their pictures. The following is an excerpt from one of these interviews. The photo being discussed depicted a white waitress with her arm around a young black male (the manager of her restaurant). * I nt p r viewer: With this picture, you said that it shows a major improvement.... Do you think it shows that blacks are now upwardly mobile, that they can have good jobs? Or are there still racial barriers to their progress? Respondent: They [blacks] are getting job offers. ... It's really neat to see that he sees his potential [pointing to the black male in the photo] to be a restaurant manager. I get the impression that a lot of black people don't. First of all, maybe they don't want to go to school. Maybe, they don't want to reach for what they can do like working at McDonald's, which isn't a bad thing. If they are really smart they can go to school and do_what they want to do. And i am a person who's not going to stand in their way of doing that. Here the photo provided an avenue into the respondents taken-for-granted assumptions about blacks. In particular, she articulates the view that the United States is a meritocracy in which blacks have opportunities, which, according to her, they choose not to fully exploit. From this perspective, the responsibility for racial inequality falls on the shoulders of minority group members themselves. This research strategy allowed me to develop the interviews and the probing questions around the respondents' photos and what their stated views were about race. The last item from the list refers to sociologists' analysis of 'found' visual materials, such as Goffman's (1979) analysis of magazine ads or Denzin's (1991; 1995) research on cinematic representations, both of which were described earlier. This method of using visual data could focus on topics such as culture and identity. Found visual data are plentiful and very inexpensive to obtain (you need only turn on yourTV or flip through the pages of a magazine). Analyzing the visual The general topic of analyzing qualitative data (text, narrative, discourse, talk and so on) is addressed in Chapter 5, in this section, I offer a brief review of VISUAL SOCIOLOGY 73 three orientations (content analysis, semiotics, and conversation analysis) that are especially suited for analyzing visual data. Content analysis This type of analysis in some respects follows the conventions of quantitative research. It makes use of random sampling techniques, coding schemes, and possibly numerical representations of visual data. Content analysis, as applied to visual data, brings together the qualitative interest in the substance of social experience and the quantitative emphasis on objective, unbiased research. As Ball and Smith put it: The objectivity of content analysis resides in the devising of precisely and clearly denned categories to apply to the material analyzed in accordance with explicitly formulated rules and procedures. In principle, different analysts using the same categories and rules would obtain idendcal results from their analysis of any given body of data; therein lies the reliability of the method.The rules of procedure serve to minimize the influence of the individual analyst's disposition and preconceptions. (1992: 21). Typically, content analysis follows the steps listed below, 1 Define the research problem. 2 Select a source for the visual material to be used in the study. 3 Identify the categories or features chat will be the focus of your research. 4 Sample documents from the sources previously defined. 5 Measure or count the occurrence of the pre-established categories, (Adapted from Ball and Smith 1992:22-25) In addition to generating numerical data, content analysis could also produce qualitative findings. A good example of grouping quantitative and qualitative techniques in content analysis of visual data is Lutz and Collins' Reading the National Geographic (1993), which explores the representations of non-Western cultures in the magazine National Geographic. Their analysis aims to reveal how the magazine's photographs reflect Western assumptions about the lifestyles of exotic people from faraway places. Here is how the authors describe their niethod of investigation: Our method consisted of randomly sampling one photograph from each of the 594 articles featuring non-Western people published m that period [between 1950-1986] Each photo was coded independendy by two people for twenty-two characteristics.,..Although at first blush it might appear counterproductive to reduce the rich material in any photogtaph to a small number of codes, quantification does not preclude or substitute for qualitative analysis of the pictures. It does allow, however, discovery of patterns that are too sobde tobe visible on casual inspection and protection against an unconscious search through the magazine for only those which confirm one's initial sense of what the photos say or do. (1993:88-89) As Rose (2001: 55) notes, Lutz and Collins's work shows that content analysis rules and procedures, if properly followed, can provide protections against bias and reveal patterns in the data that would go undetected using a less structured approach. However, content analysis of visual data suffers the major shortcoming of ' primarily dealing with what is visible on the surface (Ball and Smith 1992). In Rose's words: 'Content analysis focuses on the image itself! aut there are two other sites at which an image's meanings are made: the site of its production, and the site of its audiencing. Content analysis simply ignores both of these' (2001: 67). An orientation that is better suited for dealing with the cultural context of a visual image is semiotics. Semiotics When you look at a photo, what do you see? Say it's an image of a famous pqjitician, like George W Bush or Tony Blair. Typically, the photos of these leíders show them standing tall, dressed in suits with their national flags promi-něndy displayed in the background. What information does such a picture convey on its surface? Perhaps that the leader is a man and he is dressed in professional attire. But as you look at the image of a middle-aged white male dressed in a suit, other meanings beyond the surface level might be evoked. For example, you might assume competence, honesty and professionalism about the person in the photo. Additionally, the flag might provoke a sense of patriotism in connection with the political figure, or the feeling that he is not just standing but standing/er something (e.g., sovereignty or nationhood).Thinking about an image in this way is what semiotics (or semiology) is about. Semiotics is about signs or objects that represent things other than themselves. In Silverman's words,'Semiotics is the study of signs'. It shows how signs relate to one another in order to create and exclude particular meanings' (2001: 198). While the study of signs from this perspective began with an interest in words, for our purposes, I will limit the discussion to the analysis of photos. Going back to the example of a famous politician's photo, a semiotic analysis would suggest that the photo itself is a 'signifier,' or a culturally meaningful visual stimulus. What it stands for or symbolizes might be called the 'signified.' Drawing on the works of Barthes (1967), Silverman would say that the photo of our political leader both 'denotes' and 'connotes' a certain meaning. The meaning denoted on the surface is that a man dressed professionally is going about doing his job. The connoted meaning, on the other hand, is patriotism and nationalism, that the man in the photo represents and protects superior moral values that 'we' all believe in. Thus, semiotics allows researchers to move conceprually berween what is evident on the surface of a photo and its deeper cultural symbolism and meaning. In contrast with content analysis's detached scientific mission, the proponents of semiotics are ostensibly concerned about social inequalities and how they are rationalized through visual imagery (Rose 2001:10).This emphasis is underlined in the following passage from an introductory book on social semiotics. 75 In contemporary capitalist societies as in most other social formations there axe inequalities in the distribution of powéŕand other goods.As a result there are divisions in unsocial fabric between rulers and rules, exploiters and exploited— In order to sustain these structures of domination, the dominant groups attempt to present the world in forms that reflect their own interests, the interests of their power. (Hodge and Kress 1988: 3; as quoted in Rose 2001: 70) This way of analyzing images is not without its critics. One criticism against semiotics is 'a certain density of terminology' (Rose 2001: 73).The approach tends to be heavy on disciplinary jargon. Words like 'diegesis,' 'polysemy', and 'punctum' make the orientation inaccessible to unseasoned readers.The other problem, according to Rose, is that semiotics lacks methodological rigor: 'preference for detailed readings of individual images raises questions about the representativeness and replicability of its analyses' (p. 97), Conversation analysis Another way of analyzing visual data is using the techniques of conversation analysis and ethnomethodology (the study of how social reality is constructed and/or negotiated through everyday interaction and talk) (for a more detailed discussion of conversation analysis see Chapter 5). For example, Heath (1997) analyzes videotaped medical consultations to show how patients use body movements in combination with spoken words to convey the nature of their illness. The following excerpt from Heath's analysis describes how a woman physically demonstrates her difficulties with climbing stairs: As she [the patient] begins to step up [to the physician's desk] for the second time, she swings her hips towards the doctor....As the hips move towards the doctor he looks up, turning to face the patient.lt is as if the patient's movement elicits the reorientation by the doctor, encouraging him to temporarily abandon writing the prescription and transform the ways in which he is participating in the delivery of the [patient's] story...(1997: 194) Heath supplements the above analysis with still images from his video-recording that show a woman physically enacting her complaints within a few feet of a physician sitting at a desk. For conversation analysts, visual images could further document how social reality is created through everyday interactions and talk. Again, as Heath puts it: the possibility of capturing aspects of audible and visual elements of in situ human conduct as it arises within its natural habitats provides researchers with unprecedented access to social actions and activities. With ethnor/ethodology and conversation analysis, the technology [videos and digital cameras] opens up the possibility of developing a sociology which begins to take visual as well as vocal aspects of human interaction and the physical environment seriously, as important topics for investigation and analysis. (1997: 198) 7G QUALITATIVE RESEARCH IN SOCIOLOGY Following a similar ethnometho do logical approach, Suchman (1987) uses video-recordings to examine problems associated with human-machine interactions. In particular, she videotaped how people interact with copy machines to underscore the many gaps between the users' orientations toward their tasks and the way artificial-intelligence devices are programmed to help them, Suchman argues that such machines assume their users follow a specific plan or predetermined course of action. In practice, however, users approach these devices more fl.ui.dly; they develop their plans for using them as they perform daily tasks. The fundamental problem is that the machines' step-by-step instructions are often incongruent with the dynamic ways in which people go about using tools to do their work. There are many other ways of analyzing visual data. Your choice of methods and analysis will most likely depend on your theoretical leanings, the type of material you are working with, and other logistical considerations, such as time and financial cost.The next section deals with the more practical side of doing visual work, particularly in regard to photography and videography. Technical considerations In doing photographic research I want to end this chapter by offering you several tips that might save you hours of agonizing over the question 'What went wrong?'This technical advice is intended to help you shoot better photos or videos. Of course, the idea of 'better' is relative. Sometimes, you might decide, for very good theoretical reasons, that distortions (e.g., the subject matter being blurred or not well-lit) are research findings in their own right and need not be corrected. With visual sociology, what constitutes a 'good' photo or video is determined by your theoretical and methodological parameters. For example, if you are doing photo-elicited interviews using pictures taken by respondents, everything about the way they view their social world is empirically relevant, including defects caused by the way they use the camera. Another example of meaningful distortion is when ethnographic filmmakers shoot their subjects from the neck down to hide their identity. Having said that, regardless of your particular visual strategy, you should be familiar with the technical aspects of the equipment, especially if you are generating your own visual data. As Harper states: Becoming a visual ethnographer means becoming conscious of the potential to make visual statements by knowing how the camera interprets social reality. This means learning how cameras work, making the technical decisions that, in fact, create the photograph self-consriousiy, and relegating the automatic camera to the wastebasket, (2000:724) ill While Harper's stand on automatic cameras might sound a bit radical, his overall message of'learning how cameras work' is sound advice for anyone interested in the field of visual sociology.This section helps you with this task by VISUAL SOCIOLOGY 77 describing 1. the different rýp'ís of photographic cameras (viewfmder, single lens reflex and digital) and 2. the factors that affect how the pictures come our (aperture control, shudder speed, composition, fighting, etc.). I end the section with a few temarks about the use of video equipment in qualitative research. Viewfmder, single lens reflex and digital cameras There are three basic types of cameras: viewfmder, single lens reflex (SLR) and digital. The first two work with 35mm film (a popular choice for most photographers named after the length of each frame in millimeters). With viewfmder cameras, a common variety of which is the automatic point-and-shoot, the image that you see through the viewfinde r before you take the picture is not what the camera lens sees.The problem with this system is even when you think you have framed the subject matter perfectly in the viewfmder, the lens might see the image differendy from what you want to record.The result is that important features of the subject matter are 'cut off' in the finished product. The solution is simple: when working with a point-and-shoot, always leave a litde extra space around the edges. Don't frame your subject too tighdy because you might lose a limb or a head, as it were. The other type of film-equipped cameras, SLRs, gives you more control over the production process. An SLR's most important advantage over its viewfmder counterpart is that what you see is what you get — the images you see through the opening on top of the camera is what the lens sees. Thus, the final product will appear as it was framed. Another advantage of this type of camera is that it often provides you a choice of lenses to work with. For example, if yon want to bring the subject matter closer, you can take the standard lens off and mount a telephoto one on the camera (a telephoto lens works like a telescope, it brings distant objects closer). Additionally, SLRs are more likely to give you control over other features of the camera such as shudder speed and aperture opening, two important camera controls which I will discuss in more detail later in this section. In addition to viewfinders and SLRs, digital cameras are becoming more widely available at lower prices and with more features. While the initial cost of a digital camera might be slightly higher than its counterparts, in the long run, it more than pays for itself in savings from the cost of film, developing, and prints. The more advanced digital models give you all the control and precision of SLRs. Furthermore, digitally taken photographs (or digitalized images in general) can be enhanced or altered considerably after the image is recorded. Many come with computer software that could correct lighting and color problems after the picture is taken. Digital images can also easily be shared with colleagues via the internet and email. However, this type of camera has at least one major drawback. Most digital images don't produce clear prints, especially when they are enlarged. Fortunately, this problem is being solved by the more advanced, high-resolution models. 78 QUALITATIVE RESEARCH IN SOCIOLOGY Camera controls and other considerations Photography literally means writing with light. The appearance of a photo is determined by the amount of light that is registered on film: the more light the brighter the picture, the less light the darker. Point-and-shoot cameras don't let you manipulate the amount of light that enters the camera. At best, you are provided with an automatic flash that solves the problem of taking pictures in dimly-lit places. Alternatively, SLP^s are packed with little knobs and buttons for controlling how much light enters the camera. The two main camera controls are aperture and shutter speed.The former determines the light coming in through the lens, and the latter sets the speed at which the shutter (a small curtain, inside the camera) opens and closes. The longer the shutter is opeS or the wider your aperture, the more light reaches the film. Suppose you want to take a picture of a building at night, and you don't want to use a flash because it distorts the natural lighting. The solution is to use a wide aperture opening and a slow shutter speed to allow enough light into the camera to expose the film. On the other hand, if you are taking a picture of a man standing outside on a sunny day, a smaller aperture and faster shutter speed are the logical choices because you need less light. Other basic components of a 'good' photo are composition and framing. Composition basically means arranging the subject matter in a way that best communicates your message. As semiologists would suggest, this is as much about cultural convention as it is about aesthetics. For example, family photos of Middle Easterners might show the father, or the patriarch, in the center. Be aware of the significance of composition and use it effectively. In regard to framing, the same principles apply. Make deliberate choices about how you frame a picture and what angle you shoot it from. For example, if you want to depict the perspective of a person who is in a wheelchair, you might want to stoop down and shoot your photos from the waist level to emphasize the visual angle of the disabled. Another factor to keep in mind when working with photographic equipment is your choice of film. There are three considerations here. First, decide if you want to work with color or black and white film. Second, consider whether you want to work with slide or print film. Slides can be projected on a large screen and thus shown to a larger audience, whereas prints cannot be displayed with the same ease. Lasdy, take into account the speed of the film. This is indicated on the film as ISO. The faster the film, the better you are able to capture or freeze an image in motion. Also, faster films work better in darker conditions and reduce the need for a flash, A few notes on videography Video technology has steadily improved in the past few decades.The trend has been in the direction of giving you more control over the recording process VISUAL SOCIOLOGY 79 and making the camera (or the camcorder as it is sometimes called) smaller in size. Working with video cameras is somewhat similar to the use of still cameras,The same concerns about manipulating the lighting, composition and framing apply here. As far as the choice of film is concerned, keep in mind that videotapes do not last forever, they degenerate with repetitive use.The metallic coating on the surface of a video tape wears off a bit each time it is shown. A good practice is to always make backup copies of your original tapes. If you are going to display your work repeatedly, use your backup, so that the original remains intact. It is also important to purchase quality film, cheaper video tapes wear out faster. Another important consideration with video tapes is sound. A standard camcorder microphone is 'omni-directional' That means within an approximately 360-degree range, the microphone picks up just about every sound.The problem is when you are doing an interview on the street, for example, in addition to your respondent's voice, you are recording all the background noises (e.g., cars driving by and other people speaking). This type of sound interference could mask the voice of the respondent, which means you could end up in the very frustrating situation of not being able to hear your best interview. If you plan to use a camcorder for interview purposes, you have at least three options. One is to conduct the interview in a studio, or at least a very quiet room. Alternatively, you can use special equipment such as a shotgun mike. This -is a microphone that can be pointed in specific direction and will pick up only the sound emanating from that source. Another option is a lapel microphone, which could be cordless or wired and attaches to the speaker's shirt or jacket. CHAPTER SUMMARY This chapter explored how qualitative sociologists have incorporated the visual into their research. I began by considering the idea of a 'visual culture' and its implications for research. From this perspective, the visual is approached as a powerful medium that constructs reality and ways of seeing for audiences in various cultural contexts. I then discussed the history of visual research in sociology beginning with the earlier interest in the late 1800s to the more recent work by Becker, Gofmian, and Denzin.The next part of the chapter focused on visual data and its analysis. We discussed the three dimensions of visual data (production, the image, and the audience) and went on to classify visual data as either researcher-generated or found. In addition, we examined how the visual could bo/ised as: primary data, secondary data, a complement to the written text, or a way of eliciting interviews. Next, 1 presented three ways of analyzing this type of data: content analysis, semiotics, and conversation analysis, I suggested that content analysis tends to be more concerned about sampling and measurement, whereas semiotics takes a mote reflexive approach, analytically moving 30 QUALITATIVE RESEARCH IN SOCIOLOGY between the qualities of the image and its cultural context. With conversattoi analysis, the goal is to use recorded images to understand how social reality i constructed in everyday interaction and through talk. The chapter ended with a review of some technical and practical considerations about shooting photographs and videos, the gist of which was that you should familiarize yourself with the equipment before embarking on your research. SUGGESTED READINGS A relatively accessible and thorough read on the use and interpretation o* visual media is Gillian Rose's Visusl Methodologies (2001). Bali and Smith's Analyzing Visual Data (2001) is another work that introduces the basic components of visual analysis. For a detailed discussion on the status of the visual in the discipline of sociology see Chaplin's Sociology and Visual Representation (1994). For those interested in an overview of the field of cultural studies, Evans and Hall's edited volume Visual Culture: The Reader (2000) is packed with classic and contemporary articles on the topic. EXERCISE 4.1 OBJECTIVE: To apply visual analysis to understanding how gender is portrayed in magazine ads. DESCRIPTION: The source of data for this study could be pictures from popular magazines in which men and women are shown in the same frame. If possible, instead of choosing your pictures at will, select every other suitable photo and photocopy it for future reference. Write a short paragraph in answer to each of the following questions about social patterns that could be detected in your photos: 1 Where are the women placed in the ads relative to the men? For example, are they standing behind the men or kneeling in front of them? 2 Are there any gender distinctions in the activities shown in the photos? Are the men and women involved in the same type of jobs or tasks? Is one gender more likely to be physically active and the other watching the performance? 3 Sociologically speaking, what can you infer from these differences in spatial positioning and activities? i/t 5 Data Analysis Theoretical considerations: objectivism and constructionism S3 Grounded theory 84 A constructionist grounded theory 85 Doing qualitative analysis: the basic steps SS Reducing your data 89 Displaying your data 89 Drawing conclusions 90 Content analysis 90 Text as a reflection of public opinion 9Ü Text as a cause of public opinion 92 Narrative analysis 94 What is a narrative and how can It be analyzed? 95 Narrative practice: how narratives are put together 3^ Self-presentation in narratives 100 Conversational analysis 102 Interactional sequence 104 Institutional interaction 105 Discourse analysis 107 Interactional discourse analysis 108 Foucauldlan discourse analysis 110 Evaluating findings 113 Validity 113 Reliability 115 Using software in qualitative analysis 115 Chapter summary 116 Suggested readings 117 Exercises 118 imagine a group of graduate students in a course seminar. A doctoral candidate is presenting data from his dissertation in progress. A field-interview extract, projected on a screen at the end of the room, is the focus of everyone's attention. It is about an exchange with a homeless man outside of an emergency shelter who is asking the researcher to relay to the staff his complaints about being unfairly treated.