Social Interaction

1. Introduction

In this article, we report and reflect on the different participant positions that we have adopted when collecting mobile video data. We collected the data discussed here in various outdoor settings in which groups of participants engaged in activities such as foraging, search-dog training or sightseeing on nature walks. In addition to placing cameras on tripods or giving them to other participants, one of us was typically present to record an event with a handheld video camera. Sometimes, we adopted the position of observers standing by with cameras, while at other times we drew and acted proactively on our participant knowledge to be able to record the activity. We are particularly interested in moments of adjustment evident in the data when we made decisions about what to record and how to participate in the ongoing activity. A look into such moments allows us to consider the choices that we as researchers of social interaction have available to us and subsequently make when collecting data, and, in this way, to develop our practices of both data collection and analysis.

To unpack the variety of participant positions available during data collection, we first describe some key concerns in the study of video data within ethnomethodology and conversation analysis (EMCA). We suggest that as researchers of social interaction increasingly draw on data that involve mobility, a more thorough discussion about the possible positions and practices a researcher may adopt in the collection and analysis of data becomes necessary. We then report on the two main kinds of positions that we have adopted as they reflect our spontaneous, negotiable and planned participation in the activities that we have video recorded. We present five extracts from our research data¹ by combining conversation analysis with our ethnographic notes, recollections and knowledge. We thus complement our analyses of the extracts, which show how the activities unfold moment by moment, by considering the extracts as examples of how we in effect turn recordings into data in the process of making and viewing them. Finally, we sum up our observations and discuss some general implications that they may have for EMCA.

2. Practices and challenges of recording mobile video data

Video recordings of mobile face-to-face interactions not only make available for analysis a range of aspects of social interaction on which participants rely, but they also present a challenge: how do researchers determine the relevance of particular aspects of bodily conduct for the participants’ ongoing interaction? To face this challenge, Hindmarsh and Llewellyn (2018) follow Schegloff’s (1991, p. 64) suggestion that analytical relevance should follow from what is demonstrably relevant to the participants at just that moment, rather than from any pre-described theoretical assumption or categorisation. However, how does one know, while recording, what is relevant to the participants at every moment? As Goodwin (1994, p. 607) states, ‘any camera position constitutes a theory about what is relevant within a scene, one that will have enormous consequences for what can be seen in it later, and what forms of subsequent analysis are possible’. For the researcher who records the data, this is indeed a practical ‘one-shot problem’ because the idea of what type of data one is seeking steers the recording practices through which the data in effect come into existence.

Mondada (2016, p. 340) emphasises the careful attention given to temporally and sequentially organised details of actions as a distinctive feature of EMCA 'theory’. Because of the moment-by-moment nature of social order, video data produced for CA purposes do not typically focus on single participants nor consist of discontinuous documentation of action. The logic of most CA video production differs thus from the ‘forward playing’ or ‘going forward through’ character (Pink, 2013, p. 107) of video-ethnographic studies where the footage is arranged and rearranged ‘for the purpose of telling a compelling, intelligible story about some aspect of human interactions’ (Shrum & Scott, 2016, p. xix). Nevertheless, in both approaches wide-angle is generally preferred over close-up frames in order to ‘include all participants in a social situation and the physical environment in which social activity transpires’ (Ochs et al., 2006, p. 389; see also Heath, 1997; Heath & Hindmarsh, 2002).

However, including all participants in the camera view is difficult to maintain as soon as the social situations recorded become mobile and the physical environment is a dirt track rather than a tidy indoor setting. When the activity studied can be anticipated and takes place in a bounded space, it is easier to use multiple cameras to capture shifting interactional spaces (Mondada, 2009) between participants. When the activity studied is less predictable, however, and takes place in a less constrained space, keeping track of ‘the interaction’ becomes more difficult. Participants may move individually or together as a mobile formation (McIlvenny et al., 2014), and they may display orientation to something emerging from the margins of the interaction. What is the value of recordings that have missed some relevant aspects of the wider context, for example, and how could one avoid such omissions in data collection?

An interest in video practices has turned this methodological issue into a topic of investigation in itself (e.g., Broth et al., 2014), with an analytical focus on the actions of the camera operator. Operating a camera is a powerful sense-making and categorising device. Mondada (2014), for instance, has identified a variety of situated micro-practices whereby a cameraperson manoeuvres a mobile camera: 1) adjusting to the participants’ visible social actions and projections of next actions, while 2) orienting to the contingencies within the material environment. In doing so, the cameraperson avoids zooming in on individuals and providing fragmentary views and, instead, favours views on a participation framework that attempts to capture its entirety and continuity. Furthermore, Monteiro, Mondada and Tekin (in preparation) focus on mobile groups of participants that are recorded by several mobile camerapersons. To address the problem of coordination in having both mobile participation frameworks and multiple mobile cameras, Monteiro et al. (in preparation) show how camerapersons orchestrate a v-formation, an embodied framework for collaboratively recording video in mobile settings, producing complementary camera views with equal maximal visibility of relevant details of participants’ conduct without redundancy and invisibility of the recording activity. Panning out or using participants or objects available in the setting to hide other camerapersons from view ultimately results in the production of scientifically exploitable video data.

In short, Monteiro et al. (in preparation) emphasise camera operators’ methodological skills that help them to remain ‘hidden’ while producing a general record of what happened. By contrast, Brown and Banks (2014; see also Luff & Heath, 2012) state that it may sometimes be more fruitful to give cameras to the participants, rather than have the researcher follow and record the action, because the mere presence of the researcher may undermine the social dynamics of the participants or raise issues regarding when and how to film without interrupting what the participants are trying to achieve. Head-mounted or body cameras, especially, have untapped the potential for investigating ‘highly mobile, fast, risky, spatially constrictive, expert or equipment-laden practices’ (Brown, Dilley & Marshall, 2008, paragraph 7.1; see also Edmonds, 2021/this issue). Indeed, Brown and others suggest that with a body-mounted camera the resulting recording offers a point of view that may be later ‘used to evoke a sense of subjective positions and experiences’ (Brown, Dilley & Marshall, 2008, paragraph 2.3; see also Edmonds, 2021/this issue). Moreover, video recording on the go blurs the participants’ role as a film director with the life roles that they are performing in the film because there is no need to make conscious decisions about where to position and point the camera, or when to start and stop recording (Brown, Dilley & Marshall, 2008, paragraph 5.4).

These explorations, however, do not reveal how camera operators or camera-equipped participants gain and exploit their local membership in the practices of video recording. In this study, we look at our research data from the point of view of how we as camerapersons build on and manage our changing positions. Presenting data recorded with body-mounted, handheld and fixed cameras, we show how we act as ‘pre-analysers’ by adopting various participant positions and drawing on relevant memberships. In addition to the video recordings of the events, possibly from multiple angles, we base our analyses on various off-screen scenes that we have access to in the form of observations and recollections. Moreover, we discuss extracts that make visible the pre-analyses carried out through the camera work by a researcher who has knowledge of particular, context-specific (mobile) practices: the researcher can be seen to monitor and take part in the ongoing activity in relevant ways. Finally, we consider whether our reflections might contribute to greater transparency in data collection and analysis in studies of social interaction.

3. Adopting and adjusting participant positions

3.1 Orientation to interactional and environmental constraints

Let us first discuss situations in which a researcher engages in the activity under examination as little as possible. Our data show that even if we set out to record social activity in outdoor settings without intending to participate in the activity ourselves, we need to adjust our own conduct to various social and environmental contingencies. In Extract 1, a mother (off-camera) and two children (Sarah and Aaron, Figure 1a) are picking blueberries in a forest. While recording the activity with a handheld video camera, the researcher is exposed to the same natural elements as the participants, in this case mosquitoes. Although no mosquitoes are visible or audible in the video, the consequences are visible in the researcher’s handling of the camera and in Aaron’s conduct. The data have been transcribed according to the conventions described in Jefferson (2004) and Mondada (2019).

Extract 1. 21 HANS Picking blueberries (00:47:30)

In the extract, the family members negotiate where to pick blueberries. Aaron has taken off his cap to chase away mosquitoes and now puts it back on (Figure 1a). The researcher also attempts to chase away a mosquito from her right hand, which is holding the camera, by audibly blowing on it (l. 4). When the attempt fails, the researcher lays the camera on the ground (l. 10), brushes the mosquito away with her left hand and scratches the back of her right hand. The researcher’s two latter actions are not visible, audible or otherwise inferential from the video extract, but they are available from the researcher’s reflections on what happened. In other words, without the researcher’s account of the event, we would not know why the camera is lowered onto the ground in the middle of recording. What is more, the conduct of the video-recording researcher may have consequences in situ too. As can be seen in the extract, the unusual movement of the camera catches Aaron’s interest: he gazes at the camera on the ground (l. 13 and Figure 1b), making visible his interpretation of what is unexpected and thus noteworthy in the researcher’s conduct. To put it differently, the researcher’s actions have an effect on how Aaron conducts himself at that moment.²

Extracts 2 and 3 present other types of challenges that relate to recording human activity in natural settings, namely challenges posed by the terrain and the participants’ moving away from one another. In Extract 2, Sarah, Aaron and their mother first negotiate whether they should move forward and then, having decided that they should, begin to walk up a slope over rocks and brushwood. The video-recording researcher moves along with the participants and is obliged to decide whom to focus on at different moments. Here our interest lies, in particular, in the movement and focus of the camera.

Extract 2. 21 HANS Picking blueberries (00:46:58)

In the extract, the camera frame cannot capture all of the participants at once due to the distance between the participants as they pick blueberries. While the camera focuses on the children (Figure 2a), Aaron addresses the mother with Mom, let’s go forward (l. 2). On the last word, forward, the researcher begins to turn the camera towards the mother. In so doing, she displays her orientation to the ongoing interaction and to the mother as the intended next speaker (see Mondada, 2006). However, it is not the mother but Sarah who takes the next turn-at-talk: her yeah, Mom (l. 4) aligns with Aaron’s proposal. During the first word, yeah, the camera still focuses on the mother (Figure 2b), but the researcher then turns the camera back towards the children. Consequently, the mother is out of the camera view when she finally takes a turn by initiating a repair (l. 6, Figure 2c). At the end of the repair sequence, Sarah straightens her back from a berry-picking posture (l. 9) and, after standing still for a couple of seconds, takes a step forward (l. 10). At this point, the researcher turns the camera from the children to the terrain. The camera movement makes visible the researcher’s interpretation of the ongoing activity: she anticipates that the participants are about to move forward and prepares to do likewise. From line 11 onward, the camera view then moves along the terrain (Figure 2d) as the researcher walks up the slope to get ahead of the children and be able to record their movement from the front.

In Extract 3, a researcher is on a nature trip with three other participants, a walk along the paths and bridges of a group of islets. The researcher recorded the walk with a handheld video camera, and the other participants used their smartphones and GoPro cameras to shoot videos and take still photos. The researcher’s approach to video recording activities was: 1) to walk backwards ahead of the others when moving along a predictable route and an even surface, such as a bridge; 2) to follow the others from behind when moving along a more or less predictable route but an uneven surface, such as a path; and 3) to stay stationary, and possibly pan the camera, when the others stood still or moved within a relatively bounded area. It is the last of these solutions that the researcher adopts in Extract 3.

Here, the participants have descended from a bridge and followed a path to the tip of an islet. Anette, William and Thomas stand on the rocky tip, looking around and using their smartphones or cameras. The researcher has stayed behind, so that all the others are, in principle, in camera view, except that a moment before, Thomas has stepped behind William (Figure 3a).

Extract 3. 45 HANS Trip to nature 2 (00:12:12)

The three participants in front of the researcher’s camera are individually engaged in the same activity of doing sightseeing. Their positions and movement are constrained by the uneven and possibly slippery surface of the rock, by the pools of water on the rock and the flowing river around the islet, as well as by the bushes and trees on the islet. If the researcher were to overcome these constraints in an attempt to have Thomas in camera view, she would risk losing Anette or William out of view instead (Figure 3a). Consequently, she remains where she is.

The researcher’s first challenge is resolved as William takes a couple of steps up the rock to the right, and Thomas comes into view from behind him. At the same time, however, Anette steps down from the rock on the left. The researcher reacts to the participants’ movement by panning the camera slightly to the right so that all three remain in view (Figures 3b–c). William then continues to take a few more steps up the rock behind the foliage of a tree, while Anette walks towards and past the researcher and the camera, with her gaze directed somewhere ahead of her (Figure 3d). The researcher evidently interprets Anette’s conduct as her leaving the scene because the researcher now first pans right and then moves the camera slightly to the right, so that Anette disappears out of view. This allows for a better view of William, who nonetheless remains partly obscured by the foliage (Figure 3e), and treats capturing the actions of William and Thomas on video as more important than those of Anette’s.

By remaining at some distance from the others and by attempting to capture them all, rather than the scenery, in the camera view, the researcher evidently aims at maintaining the position of a video-recording researcher, rather than another sightseeing participant. That is, she is conspicuously engaged in another activity than her co-participants. Views from Thomas’s camera (Figures 3f–g), however, reveal that the researcher occasionally also engages in sightseeing with the others by keeping the camera steady but turning her gaze towards the scenery. At the time of writing, the researcher recollects agreeing with the others beforehand that she would mainly operate the video camera to record the others' sightseeing activity, and she also recollects making decisions about whom to focus on as the activity unfolded, usually on the principle of least effort. What is more, she recollects that she found it a challenge to capture even most of the participants on video most of the time with the equipment available.³

Extracts 1–3 show that even if we consider ourselves as data collectors only, we are necessarily participants in the ongoing situation when recording mobile outdoor activities. Firstly, although we may not take part in talking or in carrying out the same activity as others, such as berry picking or sightseeing, we are nevertheless exposed to the same specifics of the setting as those whom we are recording (see also Hofstetter, 2021/this issue). More specifically, we cannot avoid mosquitoes any more than others in the situation and we must move along on the same terrain as everybody else, adjusting our conduct to the prevailing natural conditions. Secondly, the technologies that we use – and the ways in which we use them – influence the data that we collect. By interpreting the ongoing activity and choosing what and whom to record at a particular moment, we treat some parts of the activity and some participants as (more) relevant to our possible research interests and therefore as (more) relevant elements of our data in the making. Finally, by just being present during the recording – not to mention conducting ourselves in unexpected or conspicuous ways – we have an effect on the activity that we record, in terms of what the others orient to and how they conduct themselves in the situation (see also Chen, 2021/this issue; Goico, 2021/this issue).

3.2 Active and proactive participation

In CA-labelled research that aims to discover the systematics of the organisation of talk-in-interaction, the minimum membership criteria for researchers is often that they belong to the same language community or culture as the studied participants. Once a researcher enters a specific institutional, organisational or sub-cultural setting, the notion of membership needs to be re-specified and the relation between the researchers’ background knowledge and the participants’ enactment of the context for interaction becomes an analytic problem (McHoul, 2008; McHoul, Rapley, & Antaki, 2008). Researchers doing CA in particular membership-relevant settings often characterise their work as being ethnomethodological or including ethnographic elements to grasp the ways in which participants use the context in fulfilling their interactional tasks. The same holds true for a researcher recording context-specific interaction. For example, studying practices of cycling, McIlvenny (2015, p. 61) states that "the video ethnographer [is] forced to become actively mobile to track 'the action'. [She or he] needs to be a competent member of the mobility practice(s) being studied in order to anticipate and maintain an appropriate and safe position from which to record."

Our final two extracts include recording mobile practices that are particular to a specific group of human and non-human participants, namely search-dog trainers with their dogs. These practices are negotiated by all participants, including the researcher whose primary task on the studied occasions was to video-record the activity, rather than take part in the training with a dog. However, as will be shown, the researcher’s active participation in effect facilitated the progression of the activity (see also Hofstetter, 2021/this issue). Compared with the spontaneous but negotiable movement of groups of participants who tended to stay together in relatively limited areas at any particular time in Extracts 1–3, the search-dog trainers presented in Extracts 4 and 5 were mobile within a large but pre-agreed area in the forest. The human participants moved alone or in groups, depending on the phase and purpose of the training. Most of the time, the handler-dog team (HDT) was the central actor and thus the focus of recording. However, all the participants, including the researcher, had to relate their mobility to the demands of the HDT. In this way, the participants were engaged in collaborative practices of looking and seeing: the participants who were not training their own dogs at a particular moment became observers and either looked at what the HDT did or avoided disturbing the HDT unnecessarily. In this setting, participants consider video recording an ordinary, mundane activity; in fact, search-dog trainers themselves often record trainings for their own purposes. Nevertheless, whereas search-dog trainers typically focus on recording the dog’s performance, the researcher was interested in capturing the moment-by-moment unfolding of the activity as a joint accomplishment.

The researcher benefitted from having experience in search-dog training and being familiar with two basic rules: 1) one should not enter the running line of the dog, and 2) one should avoid abrupt movements when the dog is working. Being a competent member in the activity recorded did not alone solve the challenge of finding a fixed, safe recording position. The training plan was prepared separately for each HDT and had to be revised as the training proceeded to address any contingencies, which changed the ongoing course of action. Such a situation arose, for instance, when a dog found the target persons hiding in the training area but in the wrong order. This is the case in Extract 4, in which a handler (Veikko) returns to the starting point in order to re-send his dog (Zorro) to search for the last target person still hiding in the training area. To not be in the way, the other participants (search-dog trainers Ulla and Niina, and the researcher) begin to walk away from the training area. The transcript includes data from three cameras: a handheld Sony HDR operated by the researcher (cam1), a wide-angle GoPro fixed on a tripod (cam2), and Veikko’s chest-mounted GoPro (cam3).

Extract 4. 12 HANS Search-dog training (00:01:18)

At the beginning of the extract, Veikko and Zorro are followed by the researcher, Ulla, and Niina (l. 1). The walking formation changes when Ulla and Niina stop (l. 2) and start talking about Niina’s training plan (not shown in the transcript). Ulla and Niina thus become disengaged participants in relation to the main activity of Veikko preparing to send his dog on a search. At this point, the researcher has to decide which position to take: either to remain with Ulla and Niina and continue video recording from a distance, or to move to a position from which to better capture the HDT’s interaction. From the point of view of ensuring a smooth continuation of the training activity, both options are complex. On the one hand, while standing in the search area close to the dog’s running line is not an ideal location for observers, remaining in the group could diminish the chances that the dog would mistake any of the observers for the target. Following Veikko, on the other hand, would make Ulla and Niina’s disengagement stand out as not adhering entirely to expectations about the conduct of competent observing participants.

The researcher is able to acknowledge this dilemma on reflection, at the time of writing, due to his competence in search-dog training, but his decision-making-in-action is also observable on record as a change in his walking rhythm (slowing down) and a quick glance at Ulla and Niina right after Veikko has produced a similar glance at them (l. 2). The researcher continues walking some steps towards Veikko and then stops in a v-formation (Monteiro et al., in preparation), with the fixed camera forming the other vector (l. 3). While unleashing his dog, Veikko glances back again (l. 3). Discovering that Ulla and Niina are still standing in the search area, Veikko produces a directive turn, tuutteko vähän tännepäin (‘will you come this way a little’) and an account, ettei se pyörähä sinne (‘so that it won’t twirl over there’) (l. 4). In so doing, Veikko displays his preference for the others to stand closer to him so that his dog would not be confused about which way to run. The researcher is the first one to follow the directive: he starts to move right after Veikko’s turn (l. 5). At the same time, Ulla produces an aligning response joo ‘yeah’ (l. 5), and then both Ulla and Niina walk closer to the researcher (l. 6).

The researcher performs a double role, on the one hand, by avoiding being in the way of others and, on the other hand, by finding an appropriate location for video recording, which is an activity reserved exclusively for the researcher on this occasion. This enactment of participation is not determined by the researcher alone but is negotiated during the course of the activity. Extract 4 shows in detail the methodological argument made in the field of multimodal interaction analysis about the role of the researcher as a participant: ‘with hand-held, roving cameras, the person doing the recording can be considered a participant who is involved in the recorded situation [and who] can display involvement in a situation by moving the camera in particular ways’ (Haddington et al., 2013, pp. 26–27).

In Extract 5, the researcher takes a proactive role in the emerging situation by noticing and then informing the handler (Ritu) about her dog’s (Hip) whereabouts. The incident was recorded with two cameras: a handheld device (cam1) operated by the researcher, and a wide-angle GoPro fixed on a tripod (cam2; N.B. all figures have been cropped from the original frame).

Extract 5. 13 HANS Search-dog training (GOPR0015 00:00:53)

In line 2, Ritu sends Hip to the search area on the left side of the road that functions as a middle line, a line that, according to trial rules, a handler should not leave when their dog is performing a search. Hip is running forward, and the researcher follows the dog with his camera (l. 3). Once Hip disappears into the woods, the camera turns back to Ritu as the relevant participant to observe (l. 4). She makes visible efforts to gain a better view and to spot her dog by jumping and taking steps backward (l. 4). She then praises (hyvä ‘good’, l. 5) and calls out to Hip (täällä ‘over here’, l. 7), but the dog remains out of sight. Ritu continues to look for Hip by carrying out small bodily manoeuvres such as walking forward (l. 8). These bodily adjustments, movements and verbal cues indicate uncertainty over the dog’s performance. Ritu also produces an ‘online commentary’ (Heritage & Stivers, 1999) in a soft voice, ei nää (‘can’t see’) (l. 9), and finally runs to the top of a small hillock to get a better view of the area (l. 10).

Ritu’s displays of uncertainty are a common practice in search-dog training, and her comment could well be just self-talk. The camera on the tripod (cam2) shows, however, that the researcher is also actively looking for the dog and produces similar bodily movements and trajectories (esp. l. 4 and 5) as the handler. The researcher mainly observes the situation with his own eyes and only occasionally glances through the camera to check that the focus stays on the handler. In line 11, Ritu is again visibly looking for Hip by orienting to the search area in front of her where she expects the dog to be. The researcher, in turn, glances towards the middle line on the right as another possible direction from which Hip might approach (l. 12). Indeed, Hip runs out of the forest from that side, and the researcher informs Ritu about this with a noticing (täältä tulee, ‘here {she} comes’, l. 13). Consequently, Ritu turns and runs towards Hip, praising the dog (l. 15).

By producing a noticing about the dog, the researcher influences the flow of the event. If we look at the data from the stationary camera (cam2), we can see that the noticing is a closing turn in a sequence in which the handler and the researcher have jointly addressed and solved an emerging problem. After receiving the noticing by the researcher, the handler turns towards the approaching dog (l. 14) and is ready to praise and reward the dog at the moment when it reaches her (l. 15). The researcher, the handler and the dog thus jointly create the precise moment for positive reinforcement to take place (see Greenebaum, 2010, p. 133). Rather than occurring as a surprise, the dog’s return is played out as an ordinary, non-problematic sequence of actions in which all the participants observe the training as co-instructors. In this way, the dog is made unaware of any problems that the handler might have experienced in locating it. The fact that the handler does not respond verbally to the researcher’s noticing (cf. Goodwin & Goodwin, 2012) but praises the dog after seeing it herself, here shows how the researcher as a competent member is able to recognise the activity of ‘looking for and, on finding, producing a noticing’ as relevant and expectable.

To summarise, Extracts 4 and 5 indicate that participants may treat the researcher’s role in the event as fluctuating and, depending on the ongoing activity, orient to the researcher as an active or even proactive co-participant. The change in the researcher’s participation may not always be as obvious as above, but we claim that researchers always exploit their own membership as an analytical resource for understanding how membership-bound practices are enacted and negotiated. The benefit of having recordings that also show researchers’ embodied actions, together with ethnographic accounts of the event, is that this additional information provides a resource for analysing, for example, any changes in the participation framework from the members' point of view.

4. Conclusion

In this article, we have explored the different participant positions that have been available to us as researchers of social interaction during the collection of mobile video data outdoors. Our aim has been, on the one hand, to develop our own practices in view of future data collection and analysis and, on the other hand, to contribute to such methodological reflection that aids an individual researcher to both appreciate and see beyond the specifics of the data that they have collected themselves. We have shown that regardless of being seen or not on the video, a co-present researcher unavoidably participates in the interactional situation that they are recording. Firstly, they are exposed to the same social and environmental features as the other participants in the setting and have to adjust their conduct accordingly. Secondly, their presence during the recording influences the orientations and conduct of other participants: the researcher’s position is negotiated by all participants in the setting. We have also presented moments of adjustment in our data during which a video-recording researcher interprets the ongoing activity (by drawing on their participant knowledge) and makes decisions about what or whom to record. These moments make visible the role of the video-recording researcher as a pre-analyser of social interaction. Finally, we have shown that in some cases a researcher, as a competent member of a community organised around a particular recreational activity, may easily take an active or proactive part in the activity under examination.

What is distinctive about ethnomethodology is ‘the unique adequacy requirement of methods’ (Garfinkel & Wieder, 1992, p. 182). To meet the requirement in data collection, the researcher should have ‘vulgar competence’ and an ability to act as an ordinary member in the local setting under examination. This competence is often taken for granted, while sometimes it must be deliberately developed so as to recognise, identify and describe the specificities or ‘haecceities’ of the local setting (Garfinkel & Wieder, 1992, p. 175). We have argued and shown that ‘the notion of member is the heart of the matter’ (Garfinkel & Sacks, 1970, p. 342; ten Have, 2002), not only as a policy statement in ethnomethodology, but also because this position allows researchers to analyse their own participation, and orientation to their membership, in data collection. Hence the question is: how do we as researchers orient to our collection of video data, not as a technical problem but as a first-hand analysis of the social situation?

Because our competence is not only about what we learn through our research projects (including technical solutions or members’ methods) but also about what is embedded in our historical bodies (as members and as researchers), we may more easily be drawn to topics that allow us to benefit from our competences. The search-dog training data in Extracts 4 and 5, for example, were collected by a researcher who was familiar with the activity but not with the participants. Another researcher not familiar with the activity but with the participants could have drawn upon a different kind of membership. Different kinds of memberships allow us to record and construct the data differently, and some make members’ practices more readily observable than others. Depending on the context of study, displaying one’s competence as a member may even grant the researcher access to continue data collection because the researcher can in this way render quite ordinary the actions that participants carry out (see also Hofstetter, 2021/this issue). Moreover, whatever our competences as members may be, not only can we trace them in the ways in which we have recorded data, but we can and should also actively reflect on the various positions that we take, throughout the research process and in research reports (see Amundrud, 2011, on autoethnography; and Hester & Francis, 2003; Smith, 2019; 2020, on researcher-centred approaches such as the talk-aloud video method).

One way forward in developing our reflective research practices is to go with the Big Video manifesto, advocated by McIlvenny and Jacobsen (2017; see also McIlvenny, 2018). The use of 360° video cameras to capture mobile activity, for example, reduces the need to operate the camera while being mobile and video recording. This leads to better chances to capture on video something that (other) participants also treat as relevant in the situation, rather than relying on fragmented recordings and possible recollections, or dismissing possibly interesting events because they are not on record in their entirety. Had we used 360° video cameras instead of 2D on the occasions discussed in Extracts 1–5, we would not have needed to anticipate the participants’ next actions and movement, to attempt to keep them all in the camera view or to make decisions about what and whom to record. Nevertheless, we would have recordings in which participants might occasionally be obscured by foliage, have their backs towards the camera, or move beyond the reach of the camera view (see McIlvenny, 2018). Moreover, we would have recordings in which our own conduct as researchers and, perhaps more so, as participants would show in different ways: our embodied displays of being exposed to the natural elements of the environment and of engaging in participants’ ongoing activities would be on record too. That is, although not necessarily planned nor necessarily returned to in analyses or research reports, new technologies, such as 360° video, necessarily increase the possible visibility of the researcher in the data that they collect, and in this way compel the researcher to consider their own position as a researcher and a participant.

Acknowledgements

We are grateful for the many opportunities that we have had to reflect on our positions as researchers and participants with the two anonymous reviewers and the editors of this special issue as well as with participants in the COACT Grassroots sessions, at NorDisCo 2016 and at Keskusteluntutkimuksen päivät 2017. A special ‘kiitos’ for David Monteiro for helping us with the v-formation. During the long preparation of the article, we have received funding from the project HANS – Human Activity in Natural Settings (Academy of Finland, decision number 285393); the Finnish Cultural Foundation, North Ostrobothnia Regional Fund; Tampere University (Academy of Finland, decision number 297053); the Finnish Centre of Excellence in Research on Intersubjectivity in Interaction (Academy of Finland, decision number 284595); the Finnish Work Environment Fund and the Kone Foundation.

^{1 The data were collected in Finland, where all of the participants resided at the time. The participants were variously accustomed to taking part in nature-related activities in the local context. We do not have information about the linguistic background of all the participants, but in the recordings, Finnish, English or both are used as the participants’ shared language(s).↩}

^{2 There is also a challenge that concerns the transcription of the extract: as the video camera caught no mosquitoes and caught the researcher’s embodied actions only indirectly (the loud blowing sound in line 4 and the camera movement), what would be the most appropriate way of integrating these features in the transcript? Here we have provided transcriber’s comments on the sound quality in double brackets but refrained from providing other comments because we do not have video available of the researcher's conduct during the fragment, and thus no way of verifying the interpretation.↩}

^{3 We thank one of our reviewers for pointing out that here we touch on issues that are ‘much larger than we often want to admit’. We agree, and we are afraid that space does not allow us to consider the technical specifications of our recording equipment further here. For a recent discussion on the collection of data with new camera technologies, see McIlvenny (2019), and for a reflection on the potential effects of video recording on the social situation, see Tuncer (2016).↩}