Visual Inquiry: Exploring Embodied Organizational Practices by Collaborative Film-Elicitation

From empirical studies on organizational practices we know that people provide best accounts of their work when they are in a work situation (Lave & Wenger, 1991), but being in a work situation might not always be the best time to give such accounts. Because organizational practices are inherently collective in nature, it requires a methodological approach enabling exploration of patterns of interrelatedness of practices among actors. The content of such an exploration and the process of gaining it are not clearly separable (Tsoukas, 2009). We are taking the anthropological stance, not approaching the visuals as text that has to be decoded, but rather we approach the visuals as something with which things are done. This allows us to investigate further the relationship and interaction between people, their actions, and the visual objects.

Our video research method was explored in the setting of an internationally recognized architectural firm with offices in Scandinavia and the United States. Since the establishment in 1989, they have had success in winning large competitions worldwide. At the time the study took place, they were approximately 120 architects, landscape architects, and interior architects from a number of different countries working at the Scandinavian headquarter office. The research process can be divided into three phases (see Table 1 for summary): (a) ethnographic fieldwork with (and without) camera in the architects’ office where the researchers recorded all team meetings throughout the project period, as well as individual interviews; (b) preliminary analysis by the researchers and preparation of analytical categories and videos describing architectural practices (recreation); and (c) facilitation of workshops where the researchers elicited the architects’ perceptions through group discussions and elaborations on the categories and videos presented (reenacting), contributing to a co-construction of the analysis.

Two researchers trained in methods of visual ethnography followed a project team of architects with a camera for 4 months at their weekly meetings, from start to end in a conceptual competition for a new national library. These meetings were the main meeting point for the whole team and the main collective arena for the team’s creative work where the architects presented, discussed, and generated new ideas toward a shared creative concept. The researchers had different roles; one was behind the camera and one was observing and taking notes. The latter concentrated on the content of the dialogue, whereas the researcher who filmed moved around mostly with a handheld camera, sometimes zooming in on details of the architectural sketches and models or people talking, sometimes getting an overview of people’s responses, other times placed in the background to capture the whole collective setting in one frame. The aim was not to make recordings for public screening, but rather to record thick data (Geertz, 1973) containing the necessary contextual information for it to be analytically accessible outside the setting. We recorded approximately 40 hr of footage in addition to handwritten notes and headnotes (Barnard & Spencer, 1996). In parallel, 20 individual semistructured interviews with key employees in the firm were conducted.

Being a “fly on the wall” is more easily described than accomplished. It calls for an established environment of trust and reliance for the collaboration to succeed. Before filming we presented the research objective to the team and emphasized our interest in their collective, rather than individual activities. Even though many of them had a natural skepticism in the beginning, fieldwork with camera seemed to be a raison d’être for our presence (Henley, 1998), making our intention appear evident to the participants. Our experience was that the participants soon saw us as a natural part of the team and to a degree forgot our presence, as well as being eager to tell us what progress had been made in our absence.

We talked regularly with the architects between meetings and it soon became clear to us that as outsiders we did not see the same or make the same interpretations as those directly involved in the project. As a participant observer, one has a selective ability to grasp the ongoing action as it unfolds. While recording, the ability to distinguish the irrelevant from the relevant in real time was a challenge, as the meetings were intense and rather disorganized from an outsider perspective. Consequently, all meetings were filmed in their entirety, leaving us with a massive amount of visual data to examine. We also experienced that looking through the lens of the camera created a distance to the field which was somehow comfortable, but on the contrary, it weakened the researcher’s ability to grasp the subtleties of the atmosphere. The feeling resembled the experience of looking at the footage and not actually “being there.” It was in this respect an advantage to be two researchers in the field simultaneously, because we could pay attention to different matters and share these experiences afterward to conduct a comprehensive analysis of our collective observations.

The observations of the architects’ work reinforced our assumption that filming was significant to tap into the nuances and elusive characters of the architects’ collective work practices. First of all, we observed how the architects were committed to continuous teamwork throughout the entire idea and design process. They argued for prolonged conversations as a central tool, especially in the beginning of the conceptual phase of an idea process, with the goal of reaching a collective understanding of the concept they wanted to develop. The company has built their philosophy on a culture of consensus and cooperation, inspired by egalitarian values prevalent in the work life of the Nordic countries. As a result, everyone is expected to voice their opinion, but also to be open for new ideas suggested by others in the team. Most of the time, their discussions revolve around visualizations and materializations, like models and sketches, produced in and between meetings. They would always have a visual representation of an idea, a sketch, or a model placed in front of them on the table and explained to us how they need these material objects to create a shared understanding of the imminent possibilities for further creative development and exploration. The visualizations were not intended to resemble buildings at an early stage in the concept process but were rather regarded and perceived as ideas until the project entered the design phase where actual drawings were made. The architects communicated ideas through speech, but their movements, body language, and gestures were significant to capture on video, as hands and arms, and sometimes the whole body was used to (sometimes soundlessly) convey arguments, meaning, and visualize design solutions or possible obstacles. Often, the suggested ideas would be conveyed as “air sketching” (Hagen, 2014): using their hands with or without a pen pretending to draw without touching the paper or leaving a mark. For instance, the architects in our project aspired for an environmental focus in their design. In one of the latest meetings, the project leader remade the model of a detailed building to look like a green puffy box by sprinkling the model with green moss solely with her hands, while the whole team sat in silence (Figure 1). Sometimes, parts of the idea sessions went on without a single word uttered.

Several scholars have emphasized the lack of attention given to the analysis of visual data (Bell & Davison, 2013; Mondada, 2006; Prosser & Loxley, 2008; Warren, 2005), especially in participatory visual methods (Ray & Smith, 2012; Vince & Warren, 2012). Some studies combine analysis of video material with other data sources like observations, interviews, and field notes (e.g., J. Collier & Collier, 1986; Knoblauch, 2006; Liu & Maitlis, 2014; Smets, Burke, Jarzabkowski & Spee, 2014). In other and more recent studies, video analysis has been conducted in multiple ways: by dividing data into fragments that are thoroughly transcribed using certain codes (Heath, Hindmarsh, & Luff, 2010), using software package designed for content analysis of verbal and behavioral matters (Iedema et al., 2006), single frame analysis of explicit micro-behaviors (Gylfe et al., 2016; Zundel et al., 2016), identifying “interesting” scenes for capturing the subject’s perspectives (Lahlou, 2011), or using analytical tools like Goffman’s frame analysis for selecting sequences for individual screening (Jarrett & Liu, 2016). Shortt and Warren (2019) argue for a combination of dialogical and archeological approaches in visual research for grounded visual patterned analysis (GVPA) for obtaining a broader level of meanings to be interpreted from the analysis of photo-collections. However, how to analyze video material containing patterned social action for the purpose of collective film-elicitation is underdeveloped in the literature. We will in what follows describe an analytical framework of how this can be accomplished and follow Guillemin and Drew (2010) in that the researchers are in the best position to provide an overall analysis at this stage of the visual research process. This is because of their distance to the field, and their previous research experience and training in identifying patterns and interpreting data within the context of other empirical work and theoretical frameworks. Yet we hold that the participants are in the best position to enrich the analysis in novel and unexpected ways in collaboration with the researchers.

Both Vince and Warren (2012) and Ray and Smith (2012) suggest three broad approaches to photographic analysis: (a) content analysis, (b) thematic analysis, and (c) a hybrid approach. When applying these in a collaborative video analysis, we found that a fourth approach should be added: the enacted approach, which specifically addresses elicitation of collaborative sensemaking processes. The four approaches applied on video analysis can be described as follows: Content analysis is in a narrow sense the cataloging of identified elements in the visual material (Banks, 2007; M. Collier, 2001), usually performed as the first step in a visual analysis. In a video, each still frame could possibly be subject to content analysis. Even though micro-behavior analysis is common in some areas of organizational research (Gylfe et al., 2016; Heath et al., 2010; Liu & Maitlis, 2014), it will not be the main object of interest in a qualitative research project studying the unfolding of creative practices over time. Rather, the way to conduct a content analysis for collaborative video screening is by mapping the social interaction in a sequential manner (Knoblauch, 2006; Lahlou, 2011), looking for sequences of interaction that are part of recurrent patterns of creative practices and stand out as noteworthy and engaging. The analytic selection process of deciding which sequences to be included from the flow of events, activities, interaction, and behavior is informed by the ethnographic knowledge from the field. By using this technique, it is an advantage to structure the viewing process according to a chronological principle to make sure that one does not miss out on these sequences. The selected sequences are then described and categorized, but not transcribed like in Heath et al.’s (2010) video analysis.

In our project, after the observation period ended and the competition proposal was handed in by the team, the researchers spent 2 days watching the footage following an open viewing procedure (J. Collier & Collier, 1986). The written field notes from the observations as well as transcripts of interviews were used as a guide to choose what footage to pay special attention to. This also reduced the problem of “data overload.” To organize further editing, we wrote down the timecodes of the chosen sequences within each take. We found that the videos provided us with details of the communication and interactions giving us the opportunity to somehow recreate the field experience and—although distant in space and time—allowed us to bring forth nuances and connections and make distinctions that we were not able to do when we were in the field. This process of analyzing the video content was a step toward finding patterns in the social interactions, and also a step toward creating meaning and new analytical categories because we could pause and discuss, rewind and replay, and write down keywords and interesting phenomena. For instance, we discovered that the architects’ final concept presented to the jury had its initial appearance as early as in one of the first meetings several months earlier. The main idea was introduced in the beginning of the project—a huge bookshelf being the main structure of the building—but was not followed up in team discussions before the end of the competition period.

Thematic analysis is about categorizing the empirical material and identifying patterns in the data material (J. Collier & Collier, 1986; Ray & Smith, 2012). In some visual research projects, the thematic analysis also encompasses frequency measures in the visual material, but for our project, finding statistically defined patterns were not an eligible solution for analyzing creative behavior. Our search for patterns was inspired by grounded theorizing (Charmaz, 2006; Suddaby, 2006), where we based on the content analysis examined field notes and interviews for samples of repeated sequential coordinated actions and interactions in the project meetings. As a result of the analytic process, we identified 12 categories of collective creative practices (see Table 2 for description). Each of the categories had a distinct core, yet they were interconnected and partly overlapping. They should also not be understood as exhaustive interpretations of the material, as they would be subject to further justifications and elaborations by the participants in the subsequent elicitation sessions.

Based on the thematic analysis, 22 video sequences ranging from 30 seconds to 1 minute were extracted. They were all one-takes and not edited within each sequence. We chose excerpts that displayed distinct characteristics of the architects’ creative practices and gave them titles inspired by the dialogue in the sequences (see Table 2). The videos corresponded to what Ambady, Bernieri, and Richeson (2000) call “thin slices”: “. . . brief excerpt of expressive behaviour sampled from a behavior stream” (p. 203) similar to Goffman’s (1974) “slices” of significant moments. Ambady and colleagues argue that “thin slices,” less than 5 min of video, provide a substantial amount of information because the viewers bring into the viewing process a vast set of previous understandings of context, history, culture, and social conditions that constitute their judgments of the videos. They found that expressive behavior is more accessible and reveal more than verbal communication in viewing videos because the viewer extracts most information from appearance, gestures, and manner of speaking. The sequences’ start and end point were decided based on the information needed to make sense of the interaction in the sequence and were guided and focused around momentary situations like (a) the start of a conversational theme, for instance, when someone said “What if, . . ..”; (b) moving things, for instance, when a person took the model they were discussing and put it on a light table; (c) shifting attention to other objects or topics, for instance, when the leader changed the model to look like a green box; or d) the change in intensity in the creative work, for instance, when some says “I don’t see that at all . . .” opposing against a suggested idea.

The enacted approach is an essential component to visual analysis when doing participatory video research. We define this approach as the analytic process in which the context of production and reception of the video is deliberately framed for enacting collective elicitation. This approach is connected to an awareness about the conditions for gathering the footage, the context of viewing, and the kind of audience the video material is being screened to. Following Banks (2001), the content of an image is first of all its internal narrative: the story that the image conveys for the viewer. The external narrative consists of the social context the image is produced and being viewed in (Banks, 2001). These are intertwined, but it is the external narrative, the images as a product of human action and the social relation entangled to the imagery, that is the analytic focus for collective film-elicitation. The camera does not film, humans do (Byers, 1966). The researchers’ intentions behind the filming as well as their field experience should therefore also be taken into account in the analysis because they impact the “organizational features of the recorded practices themselves, revealing their local order and intelligibility as reflexivity produced by their display to and for the camera” (Mondada, 2006, p. 52). Videos are reflexive, as they carry both the bodies in the video and the bodies behind the camera. MacDougall (2006) claims that “we see with our bodies, and any image we make carries the imprint of our bodies; that is to say, of our being as well as the meanings we intend to convey” (p. 3). Hence, both the image production and the image reception inform our understanding of the video material (Prosser & Schwartz, 1998). Therefore, this analytic approach goes beyond the image itself and integrates both background contextualization and anticipations of how the viewing process will unfold. With the architects as the main audience in mind, we searched for video sequences that would display discrepancies between what was said during interviews and observed during fieldwork and filming. The analysis is in this respect deliberately made to provoke opposition, elaboration, and confirmation. Overall, the videos extracted were aimed at displaying the elusiveness of creative work that is not easily communicated in words alone, and to stimulate the discussions and reflections in ways that would enrich understandings for both researchers and participants.

The hybrid approach is, according to Vince and Warren (2012), best achieved by an integration of the textual and the visual in a “holistic and forceful manner” (Vince & Warren, 2012, p. 13) and can take many forms, as the name suggests. There are several methods available for doing this, including involvement of participants, identification of photoset, or a photoscript with text and photo (Ray & Smith, 2012). In our project, the videos and the categories-as-text were intended to be mutually dependent sources of understanding, becoming reciprocal frames of reference. There is not much written about the contextualization in techniques like film-elicitation, although one agrees that it is needed when screening raw data to an audience (El Guindi, 2004). Textual information supporting the visual material appears to be necessary, because visuals as a form of data are “not capable of talking for themselves” (Ball, 1998, p. 187), remaining “mute” facts (El Guindi, 2004) as they are ambiguously polysemic (Goffman, 1979), beyond description (Barthes, 1977) and multimodal (Rose, 2007). They “make sense in relation to other things, including written texts and very often other images [but] they are not reducible to the meaning carried by those other things” (Rose, 2007, p. 11). They have to be interpreted and the visual availability has to be exposed both for the audience and for the purpose of research.

In our project, the written material of categories following our video sequences consisted of a title, a definition, quotes from interviews, as well as references to the organizational creativity literature. The categories’ titles were either inspired by expressions or metaphors from the architects’ daily language, or by relevant literature. With one exception, there were two video sequences for each category of practice. By preparing a set of printed A5 cards describing the 12 categories with still images from the videos for distribution in the workshops with the architects, we transformed our analysis into material objects (see Carlsen, Rudningen, & Mortensen, 2014). The intention behind this was to prompt dialogue on how physical models, sketches, and materials used by the architects enable creative exploration in the architects’ everyday work life.

There are many ways of conducting film-elicitation. For instance, one can involve participants in the beginning of the project by engaging the participants in filming themselves (Holliday, 2004; Worth & Adair, 1972), in the editing and analyzing process (Engeström, 1999; Turner, 1992), or at a later stage by getting reflections and opinions about the researcher’s filmed and edited material; turning the filmed subject into informants (P. Asch & Connor, 1994; Krebs, 1975), for cross-check of data (El Guindi, 2004), for stimulated recall (Dempsey, 2010; Stockton et al., 2004), as methods for problem solving (Iedema et al., 2006), or as a combination of these approaches (Jarrett & Liu, 2016; Lahlou, 2011). In this project, we have used film-elicitation as a way of collectively evoking culturally embedded, inarticulate and embodied aspects of situational interactions—making participants co-creators of the analysis. We will in the following give a detailed account of our way of using film-elicitation as co-creation and collaborative sensemaking.

The phase of film-elicitation consisted of two workshops where the videos and categories were presented by the researchers and explored in collaboration with the architects. The first session involved only the members of the filmed project team and took place at the architects’ office one afternoon after normal work hours. The second session included employees from the whole office, both from the Scandinavian and the U.S. office. In both workshops, the researchers presented the 12 categories of creative practices with the associated video sequences as a preliminary analysis of the architects’ creative practices, with the researchers taking turn in presenting. The presentation lasted for less than an hour, opening up for questions and clarification of content, rather than encouraging discussions at this point. To build rapport during the presentation, the researchers used the architects’ own metaphors and way of communicating when describing the various practices. Second, we attempted to create an atmosphere of relaxation and enjoyment by giving an informal presentation where everyone was included and could feel free to contribute. Laughter and comments from the audience were in this respect a good indicator of attention and recognition. The researchers deliberately highlighted the ambiguous nature of the practices to avoid closed interpretations (Martinez, 1992) where conflict, discrepancies, and contradictions are smoothed out, often leading to group conformity (S. E. Asch, 1956). Our facilitation of the subsequent discussion consisted of three strategies operating in parallel: (a) sustaining progression, (b) encourage elaboration, and (c) making connections. Sustaining progression implies maintaining conditions for a flourishing discussion to take place, for instance, when it gets sidetracked, becomes dominated by one or a few people, or several issues are discussed in parallel. The aim is to keep the balance between following up on interesting discussions and the need for progression to cover everything planned. Encourage elaboration is about evoking more details or examples, addressing or inviting others to voice their opinion on issues brought up by the participants. It also encompasses challenging an interpretation by suggesting alternative understandings or eliciting through making contested claims, in addition to pointing out possible contradictions in what has been said so far in the discussion. Making connections is a reflective strategy of observing, noticing potential contribution that is not followed up on in the discussion, and combining and suggesting for the group, when appropriate, aspects they should consider to discuss. It also includes summing up from time to time during the discussion, to get additional comments and confirmation of correct understanding. Overall, these strategies are used to obtain a multiplicity of voices, as it is the breadth and depth of the social, interactive, and co-constructed aspects of their practices that is the key concern for our investigation.

The overall goal for the visual and textual material presented to the architects was described the following way:

The categories should not be read as final answers. We hope to enrich the language of innovative practices and make people see new opportunities for developing and acting. It is also worth mentioning that the categories reflect a much noted quality of all creative practices—that of a paradox.

After the presentation, the participants were asked to individually choose three videos from the presentation (Figure 2) and answer four questions: (a) What did you see? (b) Why did you choose those particular films? (c) Why do you want more of these practices? (d) How can you make it happen? The architects were then asked to choose one of the categories describing the creative practice (see Table 2) they found most important to focus on in their shared creative work. Asking the participants to reflect and take individual notes before the group discussion begins was an attempt to avoid the discussion being dominated by the first speaker and his or her reflections. The discussion started with the participants taking turns presenting and arguing for their choice of categories to the rest of the group, before collectively deciding on one category they wanted to explore more thoroughly in smaller groups. This approach made sure the engagement in the discussion arose from the participants’ own interests. Finally, the groups were asked to discuss how to promote this practice in future projects to make sure that their reflections influence further work. In both sessions, they expressed that all of the categories were both useful and interconnected, giving them a hard time following the task given of deciding on one category to discuss and elaborate on. By keeping to our strategy of sustained progression, what happened next was not in the script of the researchers.

First of all, the categories were well received by the participants as they spontaneously started systematizing the categories, pointing to thematic clusters and conceptual relations between categories, for instance, placing them according to dimensions such as individual versus group or concrete versus abstract (see Figure 3). This happened in both sessions and was not part of the tasks given by the researchers. This could be interpreted as an attempt to avoid the predefined “rules” set by the researchers and aligned with the company axiom that “no one tells the architects what to do.” Architects’ autonomy is deemed a sacred value in their creative work.

The prepared tasks described above were more or less declined by the architects and the discussion took unexpected turns. Yet the intentions behind the tasks were still accounted for through the facilitation of the workshop.

Second, the architects interpreted the categories in new ways, reformulating them in their own words as well as adding more information to them. For example, the creative practice “Acquiring Uniqueness” initially defined as “using the availability and access to internal and external resources to find new possibilities” was reintroduced by one architect as a “smooth transition between doing the impossible possible and turning reality into dreams.” Another example from the plenary session provided another reinterpretation of our preliminary analysis. One group chose the category “Architectural Shamanism” (that we will account for more thoroughly below) and introduced it as:

An energy that is not possible to measure, not in money, it is not visible, you can’t touch it, but it is a kind of energy in the office and a spirit that can take different forms. It can be a person, like our leaders, and of course a lot of other people. It can be our mission, the feeling of being on a mission. It can be nature—architecture and nature are very closely combined. It can be the tool that we use and it can be the common atmosphere.

This explanation from the architects is a further specification of our more general definition: “Spiritual leadership in creative processes as the ability to use exceptionally sensory apparatus to connect to and draw upon cultural worlds and understandings that seem beyond entry.” By listening to our descriptions of their own practices first, and then justifying and elaborating them in their own words, the architects talked about their shared understanding of creative practices in ways they previously had not. To paraphrase Harper (2000), deconstruction of phenomenological assumptions happens when people overcome common belief by the introduction of new ways of framing. Film-elicitation is in this respect a generative approach that provides stimulus for eliciting reflections and emic elaborations leading to a co-created analysis.

Third, the sessions of elicitation not only gave the researchers and participants more details about the architects’ shared understandings but also revealed conflicting views and opinions. For instance, in the first session a discussion emerged on the concept of “Materializing the Idea” (Figure 4). While discussing which category to choose, two of the participants disagreed on how they perceived the use of tools in their work:

A: We are going to find out what is most interesting to discuss, and I don’t think (that is) “materializing,” because we are doing that. It is so physical and we do it all the time, so concrete and tool-based. I don’t think there is more to find, even though it is important. We materialize and materialize. We manage that.

B: I don’t think so at all. We are going in the wrong direction in how we command it—becoming more specialized. We have so advanced tools—tools that more and more people refuse to touch—and that kind of tool-thinking is dominant.

The recent introduction of 3D software, advanced printing machines, and robots in architectural practices challenges the presupposed ways of approaching tools and materials, without this being explicitly addressed in the everyday life of the organization. In the second session with the majority of the office, the same topic was raised again when one of the architects stated in plenary: “We don’t know how the new tools affect us.” He was putting into words what was left unsaid by managers and rarely reflected upon in daily life. The combination of videos and categories seemed to provoke discussions and reflections about contested understandings, friction, and divergent opinions among the architects that we as researchers and the architects themselves would not have discovered otherwise. This is particularly important in organizations characterized by strong communal values and an ethos of equality and autonomy. The videos’ multiple opportunities for interpretations and the way the videos display expressive behavior made it especially suitable for discussion purposes, as several and possibly alternative views among the architects came forth. The tasks given in the workshop spurred discussion on issues they seemed not to know they perceived differently.

In the literature on organizational creativity, artifacts are understood as repositories of mental structures supporting collective meaning construction, but how they actually affect the unfolding of the creative process is underexplored (Stigliani & Ravasi, 2012). Elicitation is well suited for bringing forth “silent” aspects of mental structures inherent in both conversational and material practices with artifacts bringing forth individuals’ different perceptions, and how these are negotiated and enacted during creative accomplishments. However, as the example indicates, studies of artifacts should be extended to also include the tools producing them. One senior architect told us during the workshop that when looking at a new building, he is able to see, based on shapes and curves of the façade, what kind of 3D software that was used to construct it. While some take them for granted and do not pay much attention to what impact they have on their work, others see tools as a critical component of how their creative work is performed and something they should be concerned about.

Another example from the first workshop with the project group describes different perspectives on leadership. The firm strives for egalitarianism and a flat structure, and this was highlighted in the interviews; every architect is required to voice their opinion as well as inspire others. Our observations, however, indicated that the leader of the firm had great influence over the ideation process and which direction it should take. In the workshops, we called this category “Architectural Shamanism” to extend the understanding of leadership as a practice not connected to only one person. We were not sure how the architects themselves understood this practice, as we had been given divergent signals about leadership. We also had a hard time in the editing process finding a relevant “thin slice” for the collective part of shamanism, and consequently this category had only one video whereas the rest had two. In the beginning of the selected video, the leader asks, “What if . . . ?” and the group of 13 is all listening. Transcript from the workshop shows that there is more to “architectural shamanism” than displayed in the video. Here is how they got a clearer sense of what this practice means to them in the discussion that followed:

“A: (. . .) Inspiration comes from around the table. The shamans are many, many.

B: I think also that the shaman has a professor role that isn’t involved in the daily work, but comes from the outside, with clear eyes or new eyes. And can tell—“oh, you’re forgetting the important thing—what about this, what about that.

C: It comes down to how you define the spiritual leadership, too. I mean, a spiritual leadership based on one spiritual leader or spiritual leadership based on a feeling within a group. And a group can be an entire office. And I think that maybe spiritual leadership like in an office is driven more by a process itself, and what happens with a process, and that is why one project is so different from another. It’s never the same spiritual type of leadership.

D: Well, it is, because I think it’s the ownership of the idea that makes a difference when you have a spiritual leader with very strong individual ideas. Like a master. Telling people what to do. Then there is the other . . . the transcendent spiritual leader, a master who turn the pupils into masters themselves. If you have this approach, then you accept that the ideas are not yours. But as long as you think of yourself as a master, you have ownership to the idea, because you strongly believe that you have the right and best ideas, better than your employees. That thinking makes a huge difference in our setting.

A: When the master is able to let go a little bit?

D: Yeah, but that’s the entire . . . shift. I mean, then you will become a true master, because you will turn people around you into masters instead.

B: Right.”

The architects told us that they get inspired by different people whom they see as shamans and that the “Spiritual shamanism” (renamed by the architects) can take different forms such as a person or a feeling. C and D in this transcript are senior architects who has worked at the firm for a long time, whereas A and B are new to the firm. In this discussion, it seems like the seniors are trying to inform the new employees on the firm’s ideology by telling them how this practice works. When we addressed our difficulties in finding “thin slices” displaying such shamanistic practices to the participants in the workshop, one of the architects responded that this practice might not be recognizable for other than the architects themselves, individually in the moment. This interaction shows how film-elicitation contributes to making the implicit, embodied, and situated nature sensible by “zooming with,” that is, incorporating participants’ understandings of the category in terms of meta-interpretations of their experience (Jarrett & Liu, 2016). Several scholars have pointed out that we do not know much about how teams select and build on each other’s ideas during momentary interactions (Hargadon & Bechky, 2006; Kohn, Paulus, & Choi, 2011), how they evaluate and make decisions on which ideas to follow up on and not (Harvey, 2014), and how ideas are reframed, tweaked, and synthesized into a coherent solution (Harrison & Rouse, 2014; Harvey, 2014). Film-elicitation represents a compound nexus of relational, affective, explicit, and tacit aspects of collective interaction and sensemaking that is not only brought to life, but also further co-created when they become voiced and connected during an elicitation session. As such, film-elicitation is a promising means for providing in-depth understanding of the situated dynamics of organizational creativity and how ideation processes actually progress.

Fourth, throughout the discussion the categories became part of the participants’ gesturing and body language similar to the practice of “air-sketching” referred to above. Excerpts from our field notes illustrate how the videos, titles, and categories of creative practices not only encouraged people to reflect and discuss verbally but also stimulated bodily interactions:

One of the architects is obviously disagreeing with the others about which practice to choose and playfully shouts: “Resistance!” (referring to the practice “generative resistance”) while raising her hand like an axe. Then another architect moves her hand back and forth horizontally while saying “shifting attention” (name of another category), followed up by the third architect uttering: “We need a shaman here” (“architectural shamanism”) while using her fist as a hammer towards the table.

The metaphoric and emic descriptions on the printed cards in combination with videos of expressive behavior effectively enacted lively communication and gesticulations. Carlsen et al. (2014) describe this event as tactile and sensory-motoric engagements with metaphors generating stories for collective sensemaking in a playful way. Metaphors are strongly linked to flows of experience and express an emotional reality beyond conscious awareness (Tsoukas, 1991) that can be accessed through elicitation.

In the beginning, there were some skepticism among some of the architects about “objectifying” and expressing their creative practices in words. They were afraid that the “magic” of their work would disappear (see Hagen, 2014, 2017). However, during the research process their critical stance changed. The norm in the company after completion of a competition phase is to rush off to the next competition proposal or contract assignment without debriefing or reflecting on what they have accomplished. The participants expressed a need for opportunities to reflect on their own practices and exclaimed by the end of the workshop that “we should do this after each project.” After the session, several commented that they found it inspiring to get access to their colleagues’ ways of thinking about work practices, discuss divergent views, and get new ideas for how they could work creatively together. We were told that the continuous replacement of team members on a competition project is part of the company’s policy for spurring a renewal of ideas (but more pragmatically it can be seen as a financial decision of allocating resources within the organization, due to shifting times and repeated processes of downsizing). Only two out of the total 20 team members had a stable role in the team throughout the project. All other team members were either temporarily or permanently taken off the project or put on the project again at a later date. In effect, very few in the group had a continuous perception of the process they had been part of. Our presence and later presentations of video recordings and observations thus represented an opportunity for all team members to be reintroduced to the process in hindsight—in other words, to reenact their creative process.