4.1 Core Ideas

Research in an archaeological tradition is primarily concerned with the systematic reconstruction of socially shared meaning based on the analysis of pre-existing (multimodal) artefacts in which such meaning is embodied. Artefacts, or texts, in this regard, serve as a cultural memory—as the ‘storage’ or ‘crystallization’ of social knowledge. Whether or not embodied in tangible products, texts are therefore in the first place semantic units (Halliday & Hasan, 1976), meaning(-making) constructs. Actors in the field produce artefacts and other texts to organize and legitimate the social world. Texts, accordingly, constitute traces of such world-building and world-maintenance. In social semiotics, this is reflected in the concept of metafunctions. ‘Experientially’, texts represent social knowledge. ‘Interpersonally’, texts enact social communication, including the expression of personalities and personal feelings as well as the (inter)actions that organize the social world. ‘Textually’, texts structure ‘interpersonal’ and ‘experiential’ meanings in such a way that they make sense as plausible messages in a given context. All three are manifested by specific verbal, visual, and other traces in the texts (Halliday, 1978; Kress & van Leeuwen, 2006).

However, texts do not simply represent the social world in one ‘true’ way. Thus, a central topic for multimodal archaeological research is how exactly texts—and the different modes that they draw on—relate to the social world. For the visual mode, for instance, Preston, Wright, and Young (1996) contend that at least three distinct relationships exist (see also Davison, McLean, & Warren, 2012). First, visuals can be assumed to reflect (i.e., represent) social reality by transmitting unambiguous messages. Looking at visuals from this perspective leads to efforts at reconstructing their intended message(s). Second, accepting that visuals may also mask and/or pervert social reality by transporting ideological messages (Anderson & Imperia, 1992) provides the opportunity to go beyond authorial intent and reconstruct visual meaning on the more fundamental level of “society’s deep structures of social classification, institutional forms and relationships” (Preston et al., 1996, p. 113). Third, images may also be perceived as constituting social reality. This makes it possible to challenge the intended message(s) by exploring and constructing alternative realities. Most existing studies in organization research blend notions of representation, construction, and masking of reality. They regard multimodal text as produced, as well as interpreted, in a specific cultural and historical context, thus making use of shared cues and symbols to be comprehensible. A similar ‘unmasking’ approach is taken in critical multimodal discourse analysis (Djonov & Zhao, 2014; Machin, 2013; Machin & van Leeuwen, 2016), which combines the tradition of critical discourse analysis (Wodak & Meyer, 2016) with social semiotically grounded multimodal analysis.

Accordingly, research drawing on an archaeological approach does not necessarily only claim to reconstruct elements of organizational life, but also acknowledges multiplicities and ‘pockets’ of meaning, power relations, and the strategic use of text in negotiations over meaning.

4.2 Aspects of Organization

Archaeological research provides opportunities to investigate a vast variety of organizational elements and topics, as long as they leave enduring traces in artefacts and other texts. Whereas some descriptive analyses restrict themselves to identifying and counting such traces, the majority of studies in organization research focuses on their meanings.

A major research area in archaeological research is, accordingly, dedicated to the reconstruction of the prevalent norms, values, rules, and identities in and around organizations, and how they are established, expressed, and transmitted multimodally. Such studies are situated on various levels of analysis, from the whole organization to sub-communities, and to individuals within organizations. Some studies take an even broader lens and reconstruct meaning structures within whole professions or fields. Others apply an archaeological lens to the interfaces between organizations and their environments. Here, traces of the communication between organizations and other relevant actors come to the fore, therefore putting the focus on questions of image, reputation, identity, and/or legitimacy. Especially interesting for multimodal organization research are questions about which norms are expressed primarily through which mode(s), which communities rely on which mode(s), and how modes are combined to create and transmit specific meanings. A variety of empirical studies have employed visual and multimodal data to study identities (Jones & Svejenova, 2018; Schroeder & Zwick, 2004), categories (Croidieu, Soppe, & Powell, 2018; Hardy & Phillips, 1999), or actorhood (Halgin, Glynn, & Rockwell, 2018). Multimodal data enrich such studies by providing aspects of identities that go beyond the purely verbal, and include visual, material, and embodied elements and cues.

Studies with a broader focus take into account whole populations or fields of organizations, and how they interact through multimodal texts. It has, for instance, been suggested that multimodality enhances theories of how novel ideas become institutionalized (Meyer et al., 2018) and facilitates reconstructing the different institutional ‘logics’ that pervade management discourses (Höllerer et al., 2013). Further, an institutional lens to multimodality has been applied to the legitimation and de-legitimation of field-level issues (Christiansen, 2018; Lefsrud et al., 2018). Multimodal texts have been investigated regarding their potential for blending tradition and modernity (Kamla & Roberts, 2010) and translating generic rational myths into more specific and locally resonating ones (Zilber, 2006). Archaeological research may also look at how meanings transcend and travel through levels of analysis; for instance, how societal level norms and values are instantiated in organizations through multimodal texts. Similar studies have been undertaken by social semioticians, for example studies on the multi-modal expression of organizational identity by Aiello and Dickinson (2014) and Maier (2017).

Research from more critical traditions may utilize archaeological approaches to unveil the underlying power structures in the organization of the social—and organizational—world. Prominent questions in this tradition ask who is able to express themselves in particular discourses (i.e., who has ‘voice’), and who the benefactors of particular social constructions are. Naturally, power is not only expressed (and resisted) through the verbal mode. Although not all research reconstructing power relations expressly subscribes to the label of critical management research, some multimodal research has been dedicated to such questions, such as Hardy and Phillips’ (1999) investigation of stereotypes in media discourse, Davison’s (2007) deconstruction of images in the annual reports of an non-governmental organization, and Bell’s (2012) investigation into how employees visually resist official discourse on the ‘death’ of an organization.

What is common to all these topics is that archaeological approaches focus on crystallized traces of meanings that are available over longer periods of time. This, of course, suggests longitudinal studies that look at changes in meaning structures over time. Whereas archaeological approaches are strong with regard to the reconstruction of shared meanings and knowledge, they commonly lack insights on the interactive and situated construction of such meanings, that is, the interactions between actors and artefacts in their construction and reception. The emphasis within social semiotics (e.g., in the ‘discourse historical’ approach, Reisigl & Wodak, 2016) on combining the analysis of text and context—both the situational and the broader social-cultural and historical context—seeks to fill this gap (van Leeuwen, 2005).

4.3 Methods

Similar to the range of theories, research designs in the archaeological tradition cover a broad area and include both quantitative and qualitative studies. However, systematic quantitative designs are rather rare, with the majority of studies employing qualitative designs. Methods include content analysis, rhetorical analysis, different variants of discourse analysis, framing analysis, semiotics, critical and deconstructive designs, and a variety of hermeneutical approaches. Thus, researchers can draw from a variety of methods that are suitable for data with a certain durability. In contrast, modes that are more ephemeral and difficult to ‘freeze’ in form of durable texts (e.g., scent) are less conducive to archaeological analysis. Methods that aim at capturing the fleeting character of specific forms of communication (e.g., ethnographic designs) or aim at producing data (e.g., interviews) are more often found in other approaches.

A standard compendium often referred to by archaeological researchers of the visual mode is Rose’s (2012) book on visual methodologies. Although it covers more than just the archaeological approach, it is most often referenced within that community. Notably, Rose (2012) distinguishes between different traditions of analysing visual material, namely composition-based interpretation, content analysis, semiological analysis, and discourse analysis. Although the focus of the book is decidedly visual, it discusses some media that also contain additional modes, such as websites, videos, and video games. Rose (2012, p. xviii) is also an excellent choice to inform archaeological research, since she openly rejects any notion that visuals are a ‘mirror’ of the ‘real world’: “[M]y own preference—which is itself a theoretical position—is for understanding visual images as embedded in the social world and only comprehensible when that embedding is taken into account”. Another comprehensive compendium of visual research methods is Margolis and Pauwels (2011). For multimodal analysis, there are now compendia such as Jewitt (2014) and the 4-volume anthology edited by Norris (2016).

In a recent volume on critical discourse analysis (Wodak & Meyer, 2016), Jancsary, Höllerer, and Meyer (2016) suggest a potential research design for analysing visual and multimodal data critically. Their approach is more concerned with research design than concrete methodical tools but, although originally developed for the analysis of interactions between the verbal and the visual mode, it is, in principle, adaptable to other modes due to the rather general character of their suggestions. In a nutshell, they propose to start from very small units of meaning and a clear division of modes, which allows for acknowledging the specific ways in which different modes accomplish meaning construction. In subsequent steps, the micro-level meanings identified in the individual modes are abstracted more and more, constantly compared to each other, and finally, embedded in the broader context of the respective text(s).

A different route is taken by Jancsary, Meyer, Höllerer, and Boxenbaum (2018) who aim at a more structural analysis of meanings in a larger set of visual and multimodal texts. Building on systemic functional linguistics (Halliday, 1978; Halliday & Matthiessen, 2013) and a social semiotic perspective on multimodality (Kress, 2010; Kress & van Leeuwen, 2001), they take up the concept of ‘metafunctions’ as discussed in Chapter 2 and briefly repeated above, as a basis for developing more refined coding procedures that can be adapted for other modes. In their article, they demonstrate the development of such a coding procedure based on how the metafunctions work in visuals. They briefly illustrate this procedure with data on the Austrian CSR discourse. Since their conceptual starting point is relevant for all modes, their methodology can be understood as a ‘toolbox’ for archaeological research covering a large variety of modes as well as orchestrations of multiple modes.

Greenwood, Jack, and Haylock (2018) outline a methodology for analysing visual rhetoric in corporate reporting. Building on the work of Roland Barthes, they suggest (a) that linguistic theories developed for the study of verbal text are useful, within certain limits, for the study of visual text, (b) that images are reflective and constitutive of social realities, but that the historical and cultural nature of images is vulnerable to being erased in analysis, and (c) that sites of production and consumption of text are vital for meaning-making. Their three-phased approach starts with categorical analysis, that is, counting the number of occurrences of visual elements, as well as size, compositional features, or semiotic elements. Subsequent content analysis involves second-order abduction and both denotative and connotative readings of the material. Phase three, finally, is constituted by rhetorical analysis including photography, non-photographic images, text in relation to image, financial graphics and numbers, as well as typography and layout. These rhetorical elements are interpreted in their interaction and their embeddedness in cultural agreements about rhetorical power.

Methods for the analysis of other modes include van Leeuwen’s approach to the analysis of sound and music (1999, 2011), and approaches to the analysis of typography (van Leeuwen, 2006) and colour (van Leeuwen, 2011), as well as approaches to integrating the analysis of different modes (Baldry & Thibault, 2006; van Leeuwen, 2005, 2016).

4.4 Exemplary Studies

For illustrative purposes, we now introduce three empirical studies taking an archaeological approach in some more detail. At the end of this section, we provide a table with further readings.

First, one example of empirical research specifically focusing on the interaction of the verbal and the visual mode is the study of Höllerer, Jancsary, and Grafström (2018) on the multimodal construction of the Global Financial Crisis (GFC) in the financial media. Specifically, their article aims at examining the discursive mechanisms by which a variety of interrelated, but geographically dispersed phenomena become encapsulated in a single event with distinct boundaries. Their study is interesting for multimodal organization research primarily because they focus on how verbal and visual aspects of media articles together achieve such integration and encapsulation. The authors acknowledge that the verbal and the visual both have a distinct rhetorical potential, and that their combination may facilitate processes of sensemaking and sensegiving. Their data encompass news coverage of the GFC in the Financial Times between 2008 and 2012. Building on and extending the methodology suggested by Jancsary et al. (2016), they first analyse each mode separately, then capture the respective role of each mode in the construction of the multimodal meaning communicated by each media article, before aggregating these multimodal meanings into higher-level narrative types. In essence, they conclude that specific compositions of verbal and visual meanings (e.g., mutually extending, specifying, or contrasting) enhance both resonance and perceived validity of sensemaking efforts through strengthening theorization and representation of particular aspects of the GFC. In contrast to other research in the archaeological tradition, their study puts the interaction between modes, and the meanings emerging from such interaction, front and centre in their theory development.

Second, Jones et al. (2012) show how materiality can be integrated into research on categories. They focus on the puzzle of how de novo categories are created—in their case ‘modern architecture’. They criticize that existing research has mostly focused on established categories with discrete boundaries, not yet acknowledging the material aspects of categorization. They set out to examine the formation and theorization of a novel category, focusing particularly on the materials that actors use to create categories. Their study shows that the process of category formation was crucially sponsored by architects and their clientele, and strongly associated with the value spheres of business, state, religion, and family. Furthermore, they outline how the specific mix of clients is mirrored in particular ‘artefact codes’—that is, the materials used by these architects for their buildings. For instance, ‘modern functional architects’ act on the basis of a commercial logic and exhibit a restricted artefact code in their buildings, whereas ‘modern organic architects’ resist the traditional logic and strongly mix old and new materials in their artefact codes and buildings. Hence the struggle over what constitutes ‘modern architecture’ has both a discursive and a material dimension, and both are equally crucial in the debate. In the end, the authors find, modern architects resolve the creative tension by integrating aspects of both logics and materials in their buildings, thereby extending the scope of the category. Similar to Höllerer et al. (2013), Jones et al. (2012) aim at addressing the multimodality inherent in their subject matter. Both studies deal with the issue of making modes other than the verbal accessible for empirical research by eventually ‘verbalizing’ them and plotting them into structural networks.

Third, a study of sonic logos by van Leeuwen (2017) shows how the values of corporate identities can be expressed by sound and music. Analysing the sonic logos of IT companies such as Microsoft, AT&T, and Intel, he shows how logos serve both a ‘heraldic’ function, announcing a product, service, or organization, and an ‘expressive’ function, conveying the identity of that product, service, or organization. The heraldic function is expressed by the kind of musical ‘fanfare’ motifs that have long been associated with power, victory, and triumph (rising melodies, dotted rhythms, etc.). The expressive function is realized by the timbre of the sound, and here the triumphant trumpets of traditional fanfares are now replaced, for instance, by the wind chime-like sounds of Brian Eno’s famous startup sound for the Windows 95 operating system. The meanings of timbres can be analysed on the basis of the connotations of recognizable sound effects, instrumental timbres or musical genres, or on the basis of a set of aural attributes such as narrow versus wide pitch ranges, tense versus lax sounds, rough versus smooth sounds, etc. (see Chapter 2, Figure 2.3). Overall, IT logos seek to balance electronic sounds expressing technical perfection with sounds expressing a human touch, whether in the form of those wind chimes or in the form of the nostalgia of old pianos and the sweet retro sound of the Wurlitzer, as in the AT&T logo, which, as described in an account of the making of that logo, combine ‘warm and forward thinking’ and ‘make AT&T more human and expressive’ (Kessler, 2014).

4.5 Implications of Different Modes for Archaeological Research

As the aforementioned examples show, one of the main challenges of acknowledging different modes in the archaeological tradition is to do justice to their specific form of meaning construction. All modes need to fulfil the same basic functions, but each of them does so in its particular way (Kress & van Leeuwen, 2006). Research aiming at reconstructing the meaning potential and patterns of meaning from texts encompassing different modes therefore needs to be aware of, and able to capture, these differences.

For instance, Meyer et al. (2018) disentangle what they call the ‘constitutive features’ of the verbal and visual mode. They suggest that the verbal mode is characterized by primarily conventional forms of signification, a linear, additive, and temporal textual structure, and descriptive perspectives. The visual mode, on the other hand, signifies iconically, indexically, and conventionally; is characterized by a spatial, holistic, and simultaneous structure; and provides embodied perspectives. They further suggest that

the combination of these constitutive features imbues texts building on either mode with particular affordances, with distinct impact on perception and meaning construction. Additional modes can be characterized in similar ways. The aural mode, for instance, shares some features with the verbal, and some with the visual mode, while it is also characterized by a linear and temporal structure (in that sounds have a beginning and an end) and therefore better able to communicate indexically (e.g., by mimicking sounds from particular sources). Kress (2010) stresses both the similarities of the modes developed by a given society and the differences stemming from their different materialities. He also discusses the epistemological consequences of this, in the context of their use in science education. A visual representation of a cell, for instance, requires a different kind of ‘epistemological commitment’ (Kress, 2010) than a verbal representation—it commits the learner to position the nucleus in a particular spot within the cell, something that is not required in a verbal representation.

What this means is that archaeological research that takes different modes seriously needs to find specific ways of making their distinct meaning-making potentials useful for organizational inquiry. To date, such systematic engagement with the specific characteristics and features of different modes is still in its infancy in organization research. This is one of the areas where a more pronounced engagement with multimodality research could substantially bolster the impact and contribution of multimodal organization studies. At the same time, the field of organization and management research is not one that is commonly engaged with in social semiotic research, which means that this field has the potential to provide for new insights to this theory, for instance, insights into the functioning of modes in different organizational contexts.

In terms of methodology, the acknowledgement of multiple modes of meaning-making also requires more flexibility and innovation in terms of research design. Whereas most of the methods commonly used in organization research are tailored to the verbal mode, such methods are only partially useful for the investigation of other modes. One possibility is the eventual ‘verbalization’ of information gained from other modes (as in Höllerer et al., 2013 and Jones et al., 2012 presented above). Verbalization has the distinct advantage that it enables the application of a methodological toolbox that has been tested and gradually improved for a long time. Its downside, however, is that researchers risk losing precisely the kind of additional insights into the social and organizational world that motivated their turn towards multimodality in the first place.

4.6 Specific Challenges and Opportunities Regarding Multimodality

In addition to the challenges outlined above, the combination and orchestration of multiple modes within single texts (Kress, 2010) adds an additional layer of complexity. It is necessary not only to assess and capture each mode according to its very own characteristics, but also to find a way to model the interactions of multiple modes (see, for instance, the idea of ‘integration’ in Baldry & Thibault, 2006).

There is ample potential in organization research to realize the additional insights offered by a multimodal lens, since a substantial number of studies draw on data that are essentially multimodal, however without explicitly addressing the interplay of modes. The study of Zilber (2006) on the translation of rational myths in Israeli high-tech, for instance, draws on newspaper articles as well as ads of Israeli high-tech firms to explore how generic myths (national-religious, secular-humanistuniversalistic, and individualistic) become translated into high-tech-specific rational myths (informative, individualistic, nationalistic, and enchantment). While Zilber (2006) acknowledges the existence of visual aspects in her data and also provides examples containing visuals, it would be interesting to see how the visual and the verbal interact in such processes of translation. Similarly, Vaara, Tienari, and Irrmann (2007) study an ad campaign meant to create and communicate a new identity for a Nordic bank after a series of mergers and acquisitions (M&As). They show how multimodal advertisements support the construction of authenticity, distinctiveness, self-esteem, and future orientation. A more systematic engagement with the multimodality of these artefacts could provide additional insights into the ‘division of labour’ between modes, and the way in which such attributes of organizational identity are created through specific multimodal orchestrations.

Accordingly, to assess the effects of such orchestrations, in which the combined meaning is supposedly greater than the sum of all meanings expressed in each mode, modes need to be made comparable for analysis. A first step in this direction, as has been discussed under ‘methods’ above, is the idea of metafunctions. When a common analytical scheme is established, research can begin to capture the particular meanings created through the relationality of modes. Van Leeuwen (2005) details four possible approaches to analysing multimodal texts and artefacts: visual composition, which spatially integrates text, image and other graphic elements through the use of layout, colour, and typography; rhythm, which temporally integrates bodily movement and speech, and in audiovisual media, also sound and music and camera work; information linking, which details the semantic relations between different modes such as the visual and the verbal mode; and dialogue, which integrates different ‘voices’ that might also be expressed through different modes. Van Leeuwen (2016) also explores the common features between graphic form (in typography and decorative graphics), colour, texture, and sound quality.