Visual Management Studies:
Empirical and Theoretical Approaches
Emma Bell is Professor of Management and Organization Studies at
Keele University, UK
Jane Davison is Professor of Accounting at
Royal Holloway, University of London, UK.
Address for correspondence:
Prof Jane Davison
School of Management
University of London
Surrey TW20 OEX
Visual Management Studies: Empirical and Theoretical Approaches
The field of visual research in management studies is developing rapidly and has reached a point of maturity where it is useful to bring together and evaluate existing work in this area and to critically assess its current impact and future prospects. Visual research is broadly defined to encompass a variety of forms, including pictures, graphs, film, web pages and architecture. It also incorporates work from several sub-disciplines (organization studies, marketing, accounting, human resources, tourism and IT), and includes research based on pre-existing visual material and studies that use researcher-generated visual data. We begin by considering the growing recognition of the visual turn in management research, as a counterweight to the linguistic turn, while also discussing reasons for resistance to visual approaches. Next we review research that uses visual methods to study management and organization and suggest that visual management studies may be categorized according to whether methods used are empirically driven or theory based. This categorization highlights the philosophical, theoretical and interdisciplinary underpinnings of visual management studies. It additionally enables the visual to be accorded a status equivalent to linguistic meaning, through dispelling the realist assumptions that have impeded analytical development of visual management studies to date.
Contemporary society has seen an explosion in the prevalence of the visual (Baudrillard 1994; Debord 1992), that permeates our everyday lives, through photographs, film television, video, web pages, and whose dissemination has become ever easier from mechanical reproduction (Benjamin 1999) to current digital technology. Visual media communicate in different ways from verbal language (Pink 2001; Rose 2007; Spencer 2011), and are acknowledged to be powerful in cognition and in memory (Anderson 1980; Tversky 1974). As a corollary to this explosion of the visual, a number of academic disciplines have embraced a shift from the ‘linguistic turn’ (Rorty 1979) to the ‘pictorial turn’ (Mitchell 1994). In the humanities, the former predominance of art history and historicism (Crary 1990) has been displaced in the past three decades by more theoretical and ideological approaches (for example, Barthes 1977a; 2000; Berger 1972; Sontag 1977), and by broader interests in a wide variety of visual artefacts, including film (Hayward 1993) and everyday images such as those of advertising (Barthes 1972; Eagleton 2003; Williamson 1978). Similarly, in the social sciences the visual has become well established in anthropology (see, for example, Ruby 2006; Banks 2001; Pink 2001), in sociology (for example, Emmison and Smith 2000; and through the institution of journals such as Visual Studies, Visual Communication and Visual Methodologies), and by interdisciplinary studies in history, geography (for example, Rose 2007 and work published by journals such as the Journal of Visual Culture), and economics (Thrift 2008 and the Journal of Cultural Economy).
The social scientific trend towards the visual has also informed the study of organizational life. Given the ubiquity of the visual, together with its distinctive characteristics and power, it is imperative for researchers to investigate the many organizational manifestations of the visual and the methodological challenges they raise. Yet the field of management studies research has, in comparison to other disciplines, been slow to respond to the ‘visual turn’, which has been observed as a ‘blind spot’ (Guthey and Jackson 2005; Strangleman 2004). However in the past decade significant inroads have been made in establishing visual methods in management research. While organizational researchers can look to other disciplines for guidance on how to conduct visual research in their field, and a number of methods books published in recent years provide guidance (for example, Banks 2001; Emmison and Smith 2000; Lester 1998; Margolis and Pauwels (2011; Pink 2001; Spencer 2011; Stanczak 2007; Rose 2007; Van Leeuwen and Jewitt 2000), we suggest greater understanding is needed both in relation to the specific research contexts, opportunities and challenges faced by visual organizational researchers and concerning the specific kinds of knowledge that may be generated through visual organizational research. Thus, as this review will demonstrate, the field of visual research in management is diverse and rapidly expanding and has reached a point of maturity where it is useful to bring together and evaluate existing work in this area.
The definitions of the visual adopted in this article are intentionally broad. Visual media include: two-dimensional static pictures, cartoons, photographs, maps, graphs, logos, diagrams; two-dimensional moving film and video, interactive web pages and other multi-media; three-dimensional and lived media such as dress and architecture. Research into some of these forms (for example photographs and visual branding) is fast-developing, whereas research into other visual manifestations (such as web pages) is lacking. The means of dissemination and location of visual media might be within or under the control of organizations, such as through corporate annual reports and other documents, advertising, web pages, presentations, organizational dress/fashion industry and corporate architecture. Equally, dissemination might be through external means, such as newspapers, film, television, internet reporting or artistic representations of organizations. The visual research covered by this review includes work from several management studies sub-disciplines (organization studies, marketing, accounting, human resources, tourism and IT), and encompasses both work on pre-existing visual material and work that uses researcher-generated visual data. The many and diverse issues at stake in visual management studies range from corporate identity and brand management (Schroeder 2005; 2012), to visually constructed representations of corporate leadership (Davison 2010; Guthey and Jackson 2005); from ideological questions such as gender (Brewis 1998; Kuasirikun 2010), to fun at work (Warren 2002); memorialization of organizational death (Bell 2012), to trust and accountability (Cho et al. 2009; Davison 2007).
In conducting our review, we suggest a categorisation based on the distinction between empirically driven and theory based methods, thereby highlighting the philosophical and theoretical underpinnings (Macpherson and Jones 2010) that are essential to the development of an epistemology where the visual, in addition to the linguistic, is regarded as constitutive of social reality. By empirical approaches, we understand methods that are primarily data-driven, such as visual content analysis (where elements of visual media in large samples are counted, coded and subjected to statistical analysis), and visual elicitation (where visual media ignite interview and/or discussion). By theoretical approaches, we understand methods that are primarily theory-driven, taking theory from other disciplines (such as aesthetics, semiotics, and ethical philosophy), and applying this theory in an organizational context. We also discuss the reasons for our choice of framework, alongside possible alternatives such as by data type, by organizational sub-discipline, or by whether the visual media are pre-existing or research generated.
We begin by giving context to the importance of the visual turn in management studies, while also discussing the challenges it presents. This is followed by our review of the empirical and theoretical approaches, where we discuss the ‘myth of transparency’ often inherent in empirical methods which have a tendency to presume that visual media have an objective veracity. We argue for the development of theoretical approaches to visual management studies, informed by greater interdisciplinarity, as a means of overcoming the limitations associated with empirically driven visual methods and moving towards more reflexive methodological approaches. We conclude by identifying opportunities for further research that could enable the ongoing development of visual management studies.
CONTEXTUALISING VISUAL MANAGEMENT STUDIES
From the linguistic to the visual turn in management studies
Management studies has in recent years been preoccupied with the ‘linguistic turn’ (Rorty 1979), ‘a theoretico-practical framework underlining the constitutive nature of language in all social undertakings and endeavours’ (Styhre 2010, p.10). Based on Sassurian linguistics, and further developed through critical theory (Barthes 1967; Foucault 1979), the linguistic turn represented a major shift in twentieth century thought, seeing language as constituting meaning and reality, rather than as a transparent conveyor of independent ideas. Through the development of discourse, narrative and conversational methodologies, the linguistic turn in management studies encouraged a view of organizations as ‘socially constructed verbal systems’ which are ‘actively constructed through discursive activity’ (Rhodes and Brown 2005, p.178). While the linguistic turn has undoubtedly been useful in moving management studies away from positivistic research methodologies which invite preoccupation with facts and logic, we suggest it may have gone too far in asserting the primacy of language in the constitution of socially constructed reality. Consequently, visuality and vision remain under-explored and under-theorized in the management studies literature (Styhre 2010).
This is surprising given the close connections that exist between the historical development of contemporary visual culture, in the form of photographic technologies, and rational industrialization, which both date from the turn of the twentieth century. Brown (2005) argues that photography has long been used to document the labour process, initially by proponents of scientific management like Gilbreth (1911), who photographed and filmed industrial workers in order to isolate, document and reconfigure bodily movements as an aspect of time and motion study. Contemporary versions of these organizational visualizing practices are widespread today, in the form of technologies of surveillance and identity creation, such as employee webcams and websites. Moreover, the importance of organizational image creation over time has only increased, as distinctions between employee and consumer have become increasingly blurred and the identity construction practices that apply to products and companies, are increasingly expected of employees, who are ‘increasingly being induced to reinforce and represent the brand image both within the workplace and without’ through their aesthetic, embodied and emotional practices (Brannan, Parsons and Priola 2011, p. 3).
However, there are signs that we are witnessing the beginnings of a visual turn in management studies. This builds on growing awareness of the importance of aesthetics in organizational life, and the acceptance of the relevance of sensuality, symbolism and art as a necessary counterweight to the cognitive, rationalized dimension of organizing (Gagliardi 1990; Strati 1992; Carr and Hancock 2003; Taylor and Hansen 2005). A number of recent initiatives have brought together previously isolated researchers in fruitful cross-fertilisation and enhanced the consolidation, maturity and credibility of visual management studies. Since 2000, the EIASM (European Institute for Advanced Studies in Management) has supported three workshops on aesthetics, art and management, two workshops on the theme of Imag[in]ing business, three on architecture and a forthcoming workshop on fashion. The UK ESRC (Economic and Social Research Council) has supported the Building Capacity in Visual Methods programme (2006-2009) and the first international visual methods conference at Leeds University (2009), and, in conjunction with the foundation of the inVisio research network (International Network for Visual Studies in Organisations, www.in-visio.org) has supported a seminar series (2008-9) and a Researcher Development Initiative to advance visual methodologies in business and management (2010-12). Routledge have recently commissioned several books on the visual in organisations: (Schroeder 2002a; Styhre 2010; Puyou et al. 2012; Bell et al. forthcoming). Special issues have also been commissioned in Accounting, Organizations and Society (1996), Organization (2004), the Accounting, Auditing & Accountability Journal (2009) and Qualitative Research in Organizations and Management (forthcoming 2012).
Several arguments can be put forward to justify the nascent visual turn in management studies. The first, which can be labelled the ubiquity argument (Banks 2007; Jay 2002), suggests that due to the visual saturation and occularcentrism which characterises contemporary organisational life, management researchers must be prepared to adopt visual, as well as linguistic, methods of data collection and analysis (Meyer 1991; Warren 2009; Puyou et al. 2012). This includes the proliferation of visual practices and artefacts (Benjamin 1999), combined with the increased prevalence and complexity of visual technologies in organizational life. This rationale builds on the notion that visuality reflects the cultural logic of postmodernity (Debord 1992; Baudrillard 1994). It also relates to the notion that societies are shaped more by moving images than written words and hence videocy, or the ability to interpret visual data, is of increasing importance (Denzin 1991). Visual research is thus suggested to be driven by the changing nature of organizations and the need for researchers to develop new and more appropriate methods to study them.
The second, or way of seeing argument, implies that visual research is a necessary counterweight needed to redress the privileging of language which has historically dominated organizational research (Holliday 2001; Strangleman 2004). This argument invites researchers to take images seriously as legitimate objects of inquiry and not merely as an adjunct to linguistic meaning-making activities (Pink 2001; Rose 2007). It also implies that visual communication is fundamentally different from verbal communication through the immediate, multisensory impact that comes from viewing an image which combines rationality with emotionality (Spencer 2011). The way of seeing argument draws attention to the epistemological aspects of the visual turn through focusing on its potential for creating new forms of knowledge and understanding. It is thus suggested that visual research opens up areas that have been overlooked by management researchers, such as embodiment (Emmison and Smith 2000), and reveals insights relating to established topics, such as corporate branding, that cannot be accessed through studying language alone (Schroeder 2002a; Davison 2009).
Thirdly, arguments from psychology are important in demonstrating the role of visual cognition and memory. Tversky (1974) finds that subjects presented with linguistic material and pictures devote twice as much time to examining the images, and Anderson (1980) discusses papers which suggest that pictures have a more powerful place in cognitive memory than words, and assist in communicating complex messages with simplicity. Spoehr and Lehmkuhle (1982) observe the relationship between picture processing and memory, and the importance of coherent visual structure to good understanding (Chater 1999). The ‘Stroop effect’ (Lupker and Katz 1982) demonstrates the interdependence of verbal and visual communication noted by Barthes (1977a). Based in psychology, the visual is often linked to the concept of framing (Tversky and Kahneman 1986) or impression management, whether in external relations, marketing (Schroeder 2002b), branding (Davison 2009) and annual reporting (Bernardi et al. 2002), or in dress codes and identity construction (Rafaeli and Pratt 1993; Dellinger 2002).
Overcoming challenges in visual management studies
There has, however, been resistance to visual research in our field and it is therefore necessary to explore the reasons for this relative ‘blind spot’ in management studies (Guthey and Jackson 2005; Strangleman 2004). The first reason for resistance to the visual in management studies relates to definitions of scientific rigour that are used to evaluate research quality and the ‘deep mistrust’ of the visual image within social scientific disciplines (Holliday 2001). As has repeatedly been noted in disciplines such as sociology and anthropology, the rapid development of visual research since the turn of the twentieth century has arisen relatively separately from other methods of social scientific inquiry and is sometimes perceived as something of an eccentric specialism (Emmison and Smith 2000). Researchers have noted the difficulties in publishing visual research in conventional social scientific forums such as journals and books (Banks 2007), even though certain types of visual representation such as tables and graphs are well established in these contexts (Stanczak 2007; Tufte 1983; 1990). This reflects the lack of scientific legitimacy that accrues to alternative modes of dissemination such as hypermedia (Ruby 2006) and publication via websites (Papson et al. 2007). Some even suggest that doing visual research might not be a particularly good career strategy (Prosser 1998).
Second, visual management researchers face challenges in demonstrating the scientific nature of their research due to the inherently ambiguous and polysemic nature of the visual, which eludes quantification other than, for example, by measuring size and occurrence, or by descriptive counting of apparently objective representations (for example of gender or ethnicity in the portraits of annual reports). A further challenge arises in defining the visual. Art historians and media studies scholars have focused on external visual manifestations including objects and pictures, while psychoanalysts or psychologists have regarded the visual as originating internally in the subject, such as in dreams (Mitchell 1986). Hence visual studies includes approaches that focus on what comes ‘after the eye’ and is ‘external, mechanical, dead’, and those that see the visual image as arising ‘before the eye’, and thus being ‘internal, organic and living’ (Mitchell 1986, p. 25). Additionally, the visual often overlaps with other forms of communication, most commonly language (in books, reports, newspapers, or even within the visual image itself), but also music (in film, television and video). Furthermore, the visual can be defined in terms of the phenomenon under investigation, such as corporate branding, or by the use of the visual as a method of study, for example by taking photographs.
Third, the visual presents dual puzzles of representation and meaning (Roque 2005). In the case of language and Saussurian linguistics (Saussure 1995), the signifying sign consists of two elements, a signifiant or signifier, and a signifié or signified – a word represents, designates or contains a meaning that is recognisable, while acknowledging potential ambiguity. However in the case of visual manifestations, the visual is in itself the representation, the signifier and signfied are combined (Barthes 1977a) and there may frequently be no recognisable meaning beyond descriptive representation, leaving the signified open to ambiguous and subjective interpretations. Thus the visual often depends on interplay with verbal language, included within the visual manifestation itself, or associated with it as a caption or title, to reduce but never eliminate the zone of free-floating ambiguity (Barthes 1977a). Hence if the visual cannot be said to have a recognisable meaning, it may be concluded that its role is restricted to the purely decorative or the functionally informative.
Fourth, it is often said that the visual lacks theory. While language-based theories can be related to the visual, they need careful adaptation and many would argue that pictorial meaning cannot be either conceptualised or expressed in linguistic terms, leading to an acute need for ‘picture theory’ (Barthes 1977a; Mitchell 1994). In the absence of visual theory that matches that of linguistics or literary theory in the case of language-based studies, most visual organizational research is interdisciplinary, and borrows from arts disciplines (art theory, architecture, semiotics, rhetoric, philosophy, history) and the social sciences (social anthropology and sociology, gender studies and psychoanalysis). Hence the visual encompasses physics and the science of light and colour, fine art and architecture, psychology and cognition, geography, mapping and landscaping, and the study of commodities, branding and economic value. Interdisciplinarity has certain strengths, through enabling potential revitalisation and re-examination of phenomena through exploration of cross-connections between different fields of study. However, it also presents significant challenges, such as the skill and imagination to make leaps and connections, the need for adequate expertise in more than one discipline, with the accompanying risk of amateurism (Pink et al. 2004), and the resistance, discomfort or bewilderment of those in one discipline to the unfamiliar conventions and skills of another (Quattrone 2000).
Last, but not least, visual research presents a challenge in the published reproduction of visual material, as there are considerable legal and ethical considerations (Prosser 2000; Warren 2009; Lester 1999; Pink 2001), with regard to confidentiality, anonymity, copyright and obtaining permissions. The legal framework that underlies these considerations is complex, especially for international work, and authors face considerable challenges not only in achieving full awareness and understanding of the legalities involved, but also in the time-consuming practicalities, and often the costs, involved in negotiating with individuals and organisations over the right to reproduce visual media in publication. The importance attached to such reproduction serves to underline the perceived power of the medium vis-à-vis verbal language.
Yet there are compelling reasons why these challenges should be overcome in the field of management studies, as the remainder of this review will make explicit. Most importantly, expanding the methodological repertoire of management researchers to include visual methods is not simply a response to the increasing prevalence of visual representation and communication in organizational contexts. It is also a means of extending the epistemological foundations of management knowledge in order to generate insights into aspects of management and organizational life that have tended to remain under-explored within the field. In the next section we will review some exemplary studies which illustrate the potential of visual management studies to uncover new insights in our field. We begin by setting out various ways of categorising visual management studies as a means of justifying our preferred approach.
REVIEWING VISUAL MANAGEMENT STUDIES
Our preferred way of categorising visual management research is on the basis of the technologies used to collect data and the type of image-based data collected (empirically driven approaches such as content analysis or visual elicitation), or according to the theoretical framework used to analyse the data (theory based approaches taken, for example, from aesthetics, semiotics and rhetoric, or ethical philosophy) A table summarising the characteristics and epistemological implications associated with empirical and theoretical approaches to visual management studies is provided below. We have structured our review in this way because it provides the most effective means of reviewing the current status of visual management studies, while also enabling consideration of the philosophical and theoretical underpinnings which we believe are required for the ongoing development of this field. By first discussing alternative frameworks, we demonstrate their central limitation, which concerns their lack of distinction between visual data and the epistemological implications associated with particular modes of analysis, which our categorisation framework seeks to capture.
INSERT TABLE 1 HERE
One way of understanding visual management studies is by sub-dividing research according to the type of visual data used. Two-dimensional visual images include static images, such as pictures - including paintings, drawings, cartoons, photographs - maps, graphs, diagrams, words and numbers. Moving images include film, television, video which may be combined in multimodal spaces such as web-pages (Kress and Van Leeuwen 2001). Three-dimensional visual images constitute a further data type, and may be static, such as architecture and sculpture, moving, including dress and other aspects of the body, or multimodal (Emmison and Smith 2000). They also incorporate categories of lived visual data, such as analyses of organizational buildings, objects, artefacts and architecture (Dale and Burrell 2008; Gagliardi 1990), and living forms of visual data, which could include analysis of the constructed body in management and organization through dress codes, movement and posture (Hassard et al. 2000; Trethewey 1999; Rafaeli et al. 1997). This categorisation highlights the fact that while there is a growing body of management research based on analysis of pictures, photographs and diagrams, there is less work that examines architecture, film, video or multimodal media such as websites. However, it gives pre-eminence to the notion of the visual as an external manifestation.
Another possibility would be to classify according to the sub-disciplines within management studies. Advertising, as a subset of marketing, and consumer behaviour research both have strong streams of visual research, as might be intuitively expected. Additionally, accounting, generally a less qualitative discipline, has a surprisingly rich developing body of visual qualitative research. However, such a disciplinary approach does not highlight methods which cut across the sub-disciplines, which is our main interest here.
A third possibility involves dividing the field according to how and why visual material is produced. Methodological commentators suggest there are two main types of visual data (Bryman and Bell 2011; Warren 2009). Some visual research focuses on examination of pre-existing or ‘extant’ visual material, which may be created by people in organizations and communicates messages about organizations. The role of the researcher in this type of study is to collect and analyse existing visual images, whether moving or still, rather than to produce any new visual data of their own. Alternatively, research can be based on analysis of visual material that is research-generated, or ‘research-driven data’, which is created for the purpose of research either by the researcher or research participants. Approaches based on research-generated data encourage preoccupation with methods of visual elicitation, such as photography, drawing or video diaries (Banks 2001; Harper 2002; Rose 2007; Stanczak 2007), and can open up psychoanalytical approaches (Rose 2007; Sievers 2008; Van Leeuwen and Jewitt 2000). However, the problem with this categorisation is that the distinction between extant and research-generated data can be difficult to maintain as both types of visual data may be used in the same study. For example, Bell’s (2012) study of organizational memory and death based on analysis of a case study of the closure of a UK based Jaguar car plant owned by Ford Motors, used visual data generated by the researcher in addition to visual data produced by organization members as part of their everyday activities. Moreover, as demonstrated by the linguistic tripartite model of ‘sender-message-receiver’ (Saussure, 1995), or Barthes’ equivalent photographic operator-spectrum-spectator (Barthes 2000), meanings are generated by audiences as well as by creators, an issue which tends to be overlooked by this categorising approach.
Much of the focus of visual researchers in the social sciences to date has been on collection and analysis of photographs and video recordings (Knowles and Sweetman 2004), to the extent that Emmison and Smith (2000, p.8) suggest they ‘have become fixated on the collection of images’ as the primary means of defining research as visual. Empirically driven categorisations are based on the use of technologies such as photography, film, video, digital graphics and television to produce two or three-dimensional, still and moving image data including photographs, newspaper pictures, advertisements, film, cartoons, animations and artistic impressions. The visual material may be pre-existing or research-generated.
Visual content analysis
Visual content analysis is an empirically driven method that has been used by several management studies researchers. Typically such analyses count and/or code pictures and photographs, and thus frequently combine quantitative and qualitative techniques. For example, Hunter (2008) uses content analysis to examine photographic images of tourism in brochures and guidebooks. Photographs were coded by the researcher according to the type of physical environment that was represented and according to the type of people in the image, and this information was used to analyse the social effects of tourism. Content analyses have also been used in the examination of photographs in annual reports in the context of gender and diversity studies (Anderson and Imperia 1992; Benschop and Meihuizen 2002; Duff 2011, Kuasirikun 2010).
While content analyses have the benefits of enabling the analysis of a large number of images, conforming to the norms of quantification associated with natural science methodology (Rose 2007), and permitting quantitative analyses comparing variables, they also have weaknesses. These methodological difficulties concern decisions about the measurement and quantification of images: what constitutes a visual image; how to code mixed images that contain several pictures or photographs; how to deal with words presented as a visual image; and how to measure images, (for example, by occurrence or space occupied). Studies are frequently silent on the precise methodology that has been employed. Methodological difficulties are also encountered concerning coding, since even using two coders is insufficient to capture the possibility of inter-coder variability in interpretation, and analyses tend to remain descriptive. Realist analyses also tend to assign meaning to images based on the authors’ own interpretation of the visual, paying little attention either to the role of producers in creating the meaning of an image, or the role of audiences in interpreting them.
Visual elicitation methods are also empirically driven, because visual data is produced during the research, expressly for the purposes of research (Banks 2001; Harper 2002; Rose 2007; Stanczak 2007). This may arise in one of three ways: visual data may be generated by the researcher; it may be generated by research participants; or pre-existing visual data may be used as the basis for interview or focus group discussion. Studies where visual data, such as photographs, film, drawings and diagrams, is generated by the researcher primarily for the purpose of the study (Banks 2007; Rose 2007) are comparatively rare in management studies. However, some researchers have taken an active role in taking photographs which are subsequently used as the basis for interviewing research participants, for example in Buchanan’s (1998; 2001) study of business process re-engineering and patient management in an NHS hospital, where photographs taken by the researcher were subsequently introduced into focus group discussions, or work in consumer research by Heisley and Levy (1991), who photographed research participants at family dinners as a basis for later discussion. In general, it is much more common for management researchers to involve research participants in the process of generating and/or analysing visual data.
Methods of photoelicitation where research participants generate the visual data (Warren 2005) include studies where the researcher invites research participants to take photographs themselves, thereby giving them a high degree of control over their choice and selection of images. In Warren’s (2002) study of organizational aesthetics she gave cameras to employees of a website design department of an IT company and asked them to ‘show’ her how it ‘felt’ to work there; Bolton et al.’s (2001) study of child employment involved asking young people to photograph their workplaces. Other studies have involved asking research participants to draw pictures to represent their organizations, using this as a basis for focus group or individual interviews and subsequent analysis (Vince and Broussine 1996; Broussine and Vince 1996; Stiles 1998; 2011), a technique which is suggested to enable expression of powerful emotions and unconscious thoughts that could not be expressed using written or spoken words alone. Visual autoethnography, a method that fuses visual elicitation with the researcher as a primary participant in ethnographical research, has been used to analyse tourist experiences in Peru through sharing and discussion of tourist photographs (Scarles 2010).
However, while photoelicitation, free drawing methods (Stiles 1998) and other forms of visual elicitation can be understood as visual methods of data collection, the images produced are often used as an intermediary, a means of accessing linguistic interpretations of the visual (Ray and Smith, 2012). The visual thus plays a supporting role and can remain subordinated to linguistic data which often takes precedence in the final analysis (Rose 2007).
Implications of empirical approaches: Realism and the myth of transparency
However, we suggest empirically driven approaches are of limited value in visual management studies because they focus attention on visual data, rather than on the analytical frameworks used to interpret it. This results in a series of methodological difficulties. First, empirically driven methods such as content analysis encourage oversight of the relationship between images and words and the ways in which they are used together to construct meaning. Few visual organizational researchers would suggest that analysis should be based exclusively on the collection and analysis of images and would acknowledge that visual communication often works in parallel with linguistic messages. Second, empirical approaches tend to give limited guidance on how visual data should be analysed and interpreted, which is problematic because this is one of the biggest challenges that visual researchers often face.
Finally, and most importantly, empirical approaches encourage the ‘myth of transparency’ which we see as one of the major barriers to the ongoing development of visual management research. Hence, in their review of photographic methods in organizational research, Ray and Smith (2012) assert that these methods are ‘perhaps more accurate than other methods’ such as interviews. While we concur with many of the advantages identified by Ray and Smith, we are concerned that visual methods should not be understood simply as a technical means for data collection, or an additional tool in the management researcher’s ‘tool kit’ (Ray and Smith 2012, p. 289), but also as a mode of analysis that implies a distinctive worldview. Specifically, we are concerned that the presumed veracity of the visual can invite a naïve realist perspective within which image data is not interpreted but presented as a window on the truth (Pink 2001). This applies particularly to photographs, their perceived veracity or seductive realism leading to the presumption that they provide ‘incontrovertible proof that a given thing happened’ (Sontag 1977, p.5; Barthes 1977b). The myth of transparency thus encourages a realist epistemology which assumes that images capture something that is objectively observable and real. Several examples of this can be found in cultural analyses of organization, such as those that explore representations of management in feature film. These analysts recommend the use of film as a teaching resource in management education and argue that this can provide a substitute for direct experience through giving unmediated insight into management and organizational practices (Champoux 1999; Comer 2001; Holmes 2005). This, however, obscures complex issues relating to the cultural and historical context of film production and the role of producers as authors of film texts (Bell 2008). It also assigns a relatively passive role to audiences who are assumed to uncritically absorb messages communicated by films (Bell, forthcoming a). Empirical approaches thus encourage a realist epistemological orientation towards the subject of study by treating images as indexically linked to the concept or object they represent. This ontological privileging of the visual as a reflection of reality is particularly associated with empirical approaches to visual management studies.
Emmison and Smith (2000, p. ix) argue that visual inquiry relates to ‘the study of the seen and observable’. Hence a study may be deemed visual even if images are not presented in the published research, provided the methodological framework employed enables a focus on visuality or visualization (Emmison and Smith 2000); images may thus be invisible as well as visible (Pink et al. 2004). This approach highlights the importance of theoretical and analytical perspectives in determining research as visual. Theoretically based methods of visual organizational research are useful in focusing attention on the analytical approach adopted and the type of knowledge generated, rather than on the type of data collected. However, more importantly, they invite an understanding of visuality not just as a method of study, but as a theoretical lens and a philosophical perspective through which different forms of management knowledge may be generated.
According to this viewpoint, studies that adopt a Foucauldian approach to demonstrate the role of the visual as a means of exercising power, or analyses that take a semiotic perspective to reveal how meaning is created through sign-making, may be understood as visual in orientation because they focus on the ways in which specific ‘scopic regimes’ (Metz 1975) are culturally constructed (Rose 2007). For example, the study by Bell et al. (2002) which focused on the people management initiative Investors in People could be deemed visual because it entailed semiotic analysis of the ways in which organizational members involved with the initiative constructed and interpreted the sign-making associated with accreditation, which included wall plaques and flags displayed in company premises. This was interpreted in relation to a broader ideology of organizational badge collection, even though no images were presented or analysed in the published article.
Theoretical frameworks derived from the study of fine art, with its expertise in the visual (painting, photography or film) might immediately seem natural companions to researching the visual. However, one of the problems in establishing picture theory (Mitchell 1986) has been the traditional origins of the study of the visual image in art history, thus privileging historicism over structuralism or semiotics (Crary 1990), a model which is not easily adapted to visual organisational research. In addition to art history, art theory furnishes universal models of compositional interpretation based on colour, line and perspective, but these tend to remain in the domain of aesthetics rather than assisting in the analysis of organisational issues (Rose 2007). Similarly, film theory which focuses on the formal aesthetic qualities of film, and auteur theory which concentrates on the artistic intentions of the filmmaker are of limited value in interpreting the socio-historically contextualised meaning of the visual since they are primarily concerned with making evaluative judgements about cultural worth or merit (Bell 2008).
Nonetheless, some organisational work has benefited from engagement with the discipline of fine art. For example, Schroeder (2005) shows how the work of three artists (Warhol, Kruger and Sherman), is inextricably tied to notions of branding, consumption and identity and argues that art criticism, somewhat surprisingly outside the traditional realms of consumer and marketing research, is a necessary component in both theory and method for the analysis of branding, since ‘brands are inherently visual’ (Schroeder 2005, p. 1292). Also drawing on art theory, Davison (2010) constructs a model of visual portraiture from art theory (physical, dress, spatial and interpersonal codes) and uses it to analyse intangible aspects of business communicated by portraits of business leaders in corporate annual reports (for example, Reuters CEO Peter Job) and in the media (for example, Richard Branson).
Fashion and dress
Allied to notions of visual portraiture is a body of work on fashion and organizational dress. Building on prior work in fashion and psychology as well as organisation studies (de Marley 1986; Joseph 1986; Molloy 1975), Rafaeli and Pratt (1993) construct a framework to analyse dress attributes, homogeneity and conspicuousness to discuss organisational status and power, arguing that greater heterogeneity of dress is demonstrated by more creative organizations. They further pursue their analysis of dress in a study of hospital uniforms (Pratt and Rafaeli, 1997). Dress may alternatively be viewed through the use of methods from gender or cultural studies, as in Rafaeli et al.’s (1997) examination of the use of dress by female administrators, Dellinger’s (2002) study of dress as an expression of gender and sexuality in the workplace, or Humphreys and Brown’s (2002) analysis of the role of the Turkish headscarf in a university department. Others make connections between fashionable image, marketing and economic value by examining, for example, ‘chic’ as a consumer item (Finkelstein 1994) or the way in which aesthetics, and in particular intangible concepts such as glamour, add value in the marketplace (Thrift 2008).
One of the most widely applied, and in our view potentially productive, theoretical approaches to visual management research is critical semiotics (Schroeder, 2006), or the study of signs. This methodology focuses on the duality of signs, the relationship between the signifier, the word or image that is used to represent a signified concept or meaning which, together with other signs, form part of an overall system of meaning, such as language. Semioticians study how meanings are made and reality is represented through sign systems. This project becomes critical through exploration of the underlying ideological bases of images as connected to power and knowledge and involved in the construction of notions of truth (Chandler, 2002; Spencer 2011). Drawing on Barthes’ (1972) notion of ‘mythologies’ as narrative ideologies, semiological analysis involves the interpretation of visual signs in relation to broader structures of cultural meaning. Critical or social semiotic analysis thus seeks to ‘uncover the intentional arrangement of an image [and] the manipulation of conventional codes privileging a certain “reading” of the image’ (Spencer 2011, p.147). The paradox of the photograph, and the reality of its socially constructed nature, as opposed to its apparent ‘snapshot’ characteristics, has been explored in marketing (Schroeder 2012), and its authenticity analysed in a study of CEO portraits (Guthey and Jackson 2005), and taken further in Guthey, Clark and Jackson (2009).
This invites a focus on the role of the sign-maker in shaping meaning by producing signs in accordance with their interests and in a way which is shaped by the social context in which they operate. It also draws attention to the role of intertextuality (Barthes 1975; Kristeva 1969) or provenance (Kress and Van Leeuwen 2001) in image interpretation, and the way in which an individual interprets an image is shaped by their accumulated cultural knowledge and experience of other visual texts that are drawn on and recycled to create new images. This contributes to the establishment of particular genres (Frow 2006; Bell 2008) common symbolic conventions through which the generic characteristics of particular types of visual organizational communication, such as annual reports or websites, are established and maintained. Genres are important in establishing the veracity of visual texts, affecting what audiences consider to be plausible according to genre conventions. These issues are of particular relevance in interpreting the cultural, historical and representational patterns that shape consumption, such as in Buchanan-Oliver, Cruz and Schroeder’s (2010) analysis of the visual strategies used in marketing communications texts, such as TV and website advertisements, which shows how they draw on the metaphor of the body to represent technology products.
Semiotics makes no distinction between so-called high art and the everyday, opening up organisational images such as those of advertising for legitimate examination (Barthes 1972; Barthes 1977a) and underpinning, for example, the arguments of Williamson (1978) with regard to the ideology and meaning of advertising, or an analysis of the career mythologization discernible in the photographs of a graduate recruitment brochure (Hancock 2005). Notions of plurality, multiple interpretations and the importance of the reader or viewer in constructing meaning also originate in semiotics. For Barthes (1977c), the ‘death of the author’ gives way to the ‘birth of the reader’. Approaches based in Barthesian semiotics have analysed the studium and the punctum of an Oxfam front cover, uncovering its dual signalling to the developed and developing worlds (Davison 2007) and have traced the metamorphosing icon (denotation), iconography (connotation) and iconology in the changing bowler hat visual branding of a bank (Davison 2009). The semiotics of the paratext (Genette 1999) can be usefully adapted for the analysis of the physical and visual surround of organisational documents such as annual reports (Davison 2011).
Multimodal semiotic analysis
Multimodal analysis (Kress and Van Leeuwen 1996; Kress and Van Leeuwen 2001; Kress 2010) constitutes a branch of semiotic communication analysis that focuses on signs comprised of multiple modes such as sound, music, image, three-dimensional objects, speech and writing (Jewitt 2009a). Multimodality is based on four interconnected theoretical assumptions: first, that representation and communication ‘always draw on a multiplicity of modes, all of which have the potential to contribute equally to meaning’ (Jewitt 2009b, p.14); second, that each mode enables a different type of communication; third, ‘that people orchestrate meaning through their selection and configuration of modes’ (Jewitt 2009b, p.15); and fourth, that the meaning of multimodal signs is shaped by the norms and rules of the social context in which it is created. The growing interest in multimodal analysis in recent years, from a variety of disciplines including education and media studies, stems from its potential in understanding how new on-line technologies and social media ‘have transformed the ways in which image and other non-linguistic modes circulate and are mobilized by people in powerful ways’ (Jewitt 2009b, p.4). Multimodality thus has potentially significant value in management studies yet remains as yet unexplored. However, it can be criticised for being unnecessarily technical and jargonized and for using extensive terminological frameworks inconsistently (Rose 2007).
In the humanities it has been suggested that the ancient aesthetic devices of rhetoric, highly formalised in linguistic terms since classical times as figures of speech (Barthes 1977c), might usefully be extended to provide the bases of models of visual rhetoric (Barthes 1977c; Mitchell 1994). Barthes (1977c) suggests that rhetoric might be apprehended in visual terms as ‘formal relations of elements’. While a number of studies in the organizational arena have examined rhetoric in narratives, those that have considered visual rhetoric are few. Studies of rhetoric in photographs in annual reports include Graves et al.’s (1996) analysis of the influence of television rhetoric, McKinstry’s (1996) historical analysis of the changing rhetoric and design of Burton plc’s annual reports, and Preston et al.’s (1996) analysis of selected photographs in US annual reports; McKinstry (1997) refers to the use of architectural rhetoric in promoting the interests of the accounting profession. However, none of these studies closely defines rhetoric. In advertising, McQuarrie and Mick (1996; 1999) have provided some definition of visual rhetoric as a theoretical model, and in testing for consumer response to advertisements find that the use of rhetoric produces a favourable response. In examinations of Reuters and BT’s financial reporting during the ‘dot.com’ era of irrational exuberance, Davison (2002; 2008) explores specific rhetorical devices apparent in the visual images: antithesis (taking Barthes’ S/Z (1970) as the point of departure, and repetition (constructing a framework from elements of Barthes, Deleuze, Eliade and Jankélévitch). McQuarrie and Mick (1999) and Schroeder and Borgerson (2008) have incorporated consideration of rhetorical repetition in analyses of advertising. We believe that repetition in particular is a universal rhetorical pattern that could be fruitfully explored in a number of applications.
Interdisciplinary work that develops frameworks from ethical philosophy has resulted in some especially thought-provoking analyses that extend the boundaries of visual management studies and draw attention to the ethics of managerial practice. Campbell et al. (2009) observe the rise in photographic human representation in annual reports over a fourteen year period and interpret this using a Levinasian ethic of engagement with, and accountability towards, the Other. Kuasirikun (2010) explores the ethics of human representation in Thai annual reports from a gender perspective informed by Habermasian theory of communicative interaction. These two studies are also noteworthy for their successful combination of quantitative content analysis with philosophical theory and discussion. In a similar vein, Schroeder and Borgerson’s (2005) analysis of marketing communication draws on feminist ethics and critical race theory to explore how subjects are represented and how representations can exclude, stereotype and cause harm. Driven by a theoretical framework constructed of the duality of the (paternal) law versus the (maternal) body from the Stabat Mater of Kristeva’s Tales of Love (1987), Matilal and Höpfl (2009) compare and contrast official reports of the Bhopal disaster with the moving and tragic story of photographs taken at the time. In a similarly theological vein, Davison (2004) explores the cross-cultural notion of salvation through ascension, in the light of Eliade (1980), in a number of corporate annual report photographic depictions of ascension, from staircases to escalators and cliff-climbing.
Implications of theoretical approaches: Reflexive methodologies
In contrast to empirical approaches, theory based visual management studies encourage a more reflexive orientation towards data collection and analysis. The meaning of an image is thus understood to be derived from the interpretations that research participants make in relation to it (Pink 2001; Stanczak 2007). This is due to the greater acknowledgement of the polysemic nature of images, or their ability to enable multiple readings. For theory based researchers, ‘there is no essential truth awaiting discovery in an image, instead it is a matter of developing a convincing interpretation’ (Rose 2007). Researchers who adopt a reflexive methodological approach regard images as social constructions which must be explored as a product of the encounter between the researcher and research participants (Pink 2001; 2004; Kunter and Bell 2006). This involves ‘researchers being aware of the theories that inform their own photographic practice, of the relationships with their photographic subjects, and of the theories that inform their subjects’ approaches to photography’ (Pink 2001, p.54). By focusing on the socially embedded nature of images and their framing in cultural contexts of production and consumption, reflexive methodologies also seek to recognize the ambiguity of images and their fluidity of meaning over time as the cultural context in which they are located changes (Spencer 2011).
Rose’s (2007) critical approach to visual research is typical of reflexive methodologies which tend to be founded on a cultural, anthropological perspective. For Rose, ways of seeing are multiple and depend very much on who is doing the looking. Drawing on Saussurian linguistics and literary theory, she proposes a triumvirate model of analysis based on three interrelated sites of meaning: the site of the image, the site of image production and the site of the audience. This invites a focus on how images are used, individually and in conjunction with other images. This is particularly important in visual management studies where the intention that lies behind the production and use of visuals in organizations varies considerably; images may be strategic (produced by an official source within the organization), explanatory (such as charts and tables), employee or consumer generatedi.
The role of audiences is crucial in shaping the meaning of images and understanding how they may be contested or even rejected (Fiske 1994). This perspective draws on work in cultural studies to suggest that the meaning of images is not fixed but dynamic and open to continual interpretation as part of an ongoing circuit of communication involving the author, the reader, and the text (Hall 1980). It thereby highlights the possibilities for multiple readings including a preferred reading, when the reader accepts the interpretation intended by the author of the text, a negotiated reading, when the reader broadly accepts the message but modifies it in a way which fits with their social position, interests and experiences, or an oppositional reading where the reader rejects the preferred reading of the image in favour of an alternative frame of reference that runs directly counter to the message intended by the producer (Hall 1980). Hence, as audience analysis in the fields of film and cultural studies confirms, visual messages can ‘boomerang’ through being used to reinforce attitudes opposite to those intended by the author (Bell forthcoming). While these concepts can also be applied in the analysis and interpretation of linguistic data, due to the polyvocal nature of images, they are particularly important in visual management research. Exemplary studies include several of the chapters in Hassard and Holliday’s (1998) edited collection, including Brewis’ (1998) interpretation of gender relations in the film Disclosure.