Driving together: Shared car journeys as research space

Abstract

This paper introduces driving around with people in private cars as a research space to which walking methods can be adapted and in which productive accidental ethnography can take place. Whether one is walking or driving together with research participant(s), one's shared mobility is key: the act and rhythm of moving together through land and sense-scapes provides prompts and insights and facilitates conversation and rapport. However, the coverage of larger distances at greater speeds in a car and the car's existence as a private space separate from the scenes and places passed through ensures that driving together is qualitatively different to walking together and that it can sometimes be more useful. The paper argues that driving together can be a productive research space depending on research focus, context, and ethical and security considerations.

Mobile methods – whereby researchers seek to accompany or observe research participants as they move – have become increasingly common especially in disciplines that focus on space and place such as anthropology, geography, and sociology. This rise is linked to a broader mobilities turn, which has seen scholars pay greater attention to how people and things move and the impact of mobilities on emotions, cultures, societies, and environments (Sheller, 2004; Sheller and Urry, 2006; Vannini, 2010). This dual shift is understandable. While one can study mobilities through stationary research methods – for example, by analysing how people talk about travel in focus group discussions (Guiver, 2007) or write about it in literary texts (Pearce, 2017) – there are clear advantages of observing and discussing mobility in real time (Brachet, 2012: 10). However, mobile methods and mobilities research should not be conflated: non-mobile methods can offer valuable insights on mobilities (Merriman, 2014), and mobile methods can be used to research much more than mobilities.

The versatility of mobile methods is now well recognised when it comes to walking research, which has been used for data collection on everything from people's daily routines (Kusenbach, 2003) and personal biographies (Holgersson, 2018) to job roles (Gilliat-Ray, 2011), activism (Anderson, 2004), and place-making (Pink, 2007). This diversity of research foci has borne a range of walking methods. This includes the walk-along where a researcher accompanies a research participant on a routine outing (Kusenbach, 2003: 463), walking interview where a researcher and participant engage with a route planned by either party (Kinney, 2017), bimbling where a researcher and participant walk for the general conversation that it can facilitate (Anderson, 2004), and docent method where researcher and participant(s) meet to plan and then walk and photograph a route, before meeting to discuss the route taken using the photos as prompts (Chang, 2017). Photography is thus built into the docent method, but recording devices – camera, voice, or film – are often incorporated into other walking methods (Pink, 2007), with GPS and GIS technology sometimes used to ‘connect what people say with where they say it’ (Jones et al., 2008: 2; emphasis in original).

The utility of driving research is more contested. On the one hand, much mobile methods literature follows John Urry (2004, 2006) in viewing cars as high-speed cocoons that separate drivers and passengers from external sense-scapes, societies, and environments – a distinction even being made between mobile methods and ‘sedentary methods in motion’ (Evans and Hones, 2011: 850). According to this literature, walking is more natural, enjoyable, sensory, and social than driving, and it constitutes a better way to engage with the multi-sensory world (Duarte, 2021; Kusenbach, 2003; Lee and Ingold, 2006). In turn, driving is largely limited to two principal methods: namely, passenger participant or recorded observations of what car drivers and passengers do while travelling (Ferguson, 2016; Laurier et al., 2008) and drive-alongs where a researcher accompanies a research participant on their scheduled journey(s) (Dahl and Tjora, 2023; Kusenbach, 2003; Yang et al., 2023).

On the other hand, the scholarship that does use driving methods, and the broader literature on automobility and car cultures, highlights how car journeys can be enjoyable and relaxing (Bijsterveld, 2010; Miller, 2003; Sheller, 2007), and how they can facilitate conversations within the car and provide insights into the worlds outside of it. More specifically, these literatures show how the rhythms of driving (Laurier et al., 2008), the prompts inside and outside the car (Goodwin and Goodwin, 2012; Pearce, 2017), the privacy provided (Ferguson, 2016), the shared experiences (Fenno, 1990) and achievements (Dahl and Tjora, 2023), and car journeys’ ‘contemplative space-time’ (Guzman, 2022: 325) and pause-full conversations can facilitate conversations between a driver and passenger(s) including on ‘very serious topics’ (Laurier et al., 2008: 17). This literature also shows how driving can help situate people in landscapes and societies (Bishara, 2015; Dawson, 2017) and bring people of different socio-economic backgrounds in segregated societies into close proximity with each other (Yazıcı, 2013).

Drawing on 20 years of research experience, this paper brings these literatures into conversation with each other to show how the differences between walking and driving have been exaggerated, and how the differences that do exist can sometimes render driving a more appropriate mobility than walking. In addition, the paper shows how walking methods can be adapted to driving, and how driving can be a particularly productive space for accidental ethnography. This is important as it means that driving has a much wider utility – in terms of disciplinary approach and research focus – than is commonly recognised. It also means that driving should be regarded not as a research method or activity, but as a research space in which various methods and accidental ethnography – or ‘the unplanned moments that take place outside … structured methods’ (Fujii, 2015: 525) – can take place. The paper concludes that whether driving is a potentially productive research space is determined by research focus, context, and ethical and security considerations.

This argument for the broad applicability of driving methods is strengthened by three factors. First, I write as someone who does not particularly like driving. My preference (weather, route, security, and time permitting) is to walk, whether this be to relax or get to a specific destination; and, while I own a car, I often walk or take public transport. Second, unlike most writings on driving research, which focus on car cultures from an anthropological or sociological perspective, I write as a political scientist who has never studied mobilities or car cultures. Third, while the majority of mobilities research focuses on case studies in the Global North, this paper draws on 20 years of driving around with people in Kenya – a country where only a small minority of the population own cars and the majority still journey by foot or public transport (Salon and Gulyani, 2019). In this regard, it is important to note that my Kenyan travelling companions have either been taxi drivers or research assistants, research participants, or friends for whom a drive is not an uncommon way of getting around.

In making an argument for the broad utility of driving research, the paper narrows in on driving with research participants or driving together. This focus is not to deny the utility of driving around by oneself (which can provide an opportunity for reflection and insights into research locations) or to downplay the insights that can be gained from recording in-car interactions (Laurier et al., 2008; Yang et al., 2023). It is also not to deny the utility of travelling with people via other modes of transport – from motorcycles and boats to buses and trains (Agbiboa, 2022; Mutongi, 2017). Instead, this focus simply recognises the particularities of driving together; namely, the controlled sense-scape within the car, a degree of control over routes and stops, and the sharing of a private space and journey with a travelling companion(s). My focus is also primarily on driving together with a single other as driver or front-seat passenger. This is significant, since, while drivers and front-seat passengers have different roles and practice different levels of concentration, their adjacent seating and partially shared visual field can facilitate conversation and a sense of being physically and socially close in ways that differ to backseat passengers. Front-seat passengers are also better positioned than backseat passengers to share in the accomplishment of driving – from helping to navigate to changing the music (Dahl and Tjora, 2023; Laurier et al., 2008).

The paper starts with an overview of the range of insights commonly associated with mobile methods before turning to the differences between walking and driving together, the driving versions of walking methods, and some security and ethical considerations.

Moving with people

The rise of mobile methods stems, at least in part, from a growing appreciation of ‘the transient, embodied and multisensual aspects of “the social”’ (Bates and Rhys-Taylor, 2018: 2), of how we live as embodied entities with multiple senses, and of how we perceive things ‘with the whole body’ and often by moving around them (Ingold, 2004: 330 and 331). This means that, to gain a more holistic understanding of people's lives and of their interactions with their environments and others, it can be insightful to move with people. Insights to be gained include the ‘pre-reflective knowledge and practices of the body, or the most trivial details of day-to-day environmental experience’ (Kusenbach, 2003: 462), which research participants may not consider sufficiently important to mention in a static interview. It can also include the under-reported, which researchers may not think to ask about, and the difficult, uncomfortable, or traumatic, which can sometimes be easier to discuss or observe when a researcher and research participant move adjacent to one another with a shared point of view than in a classic face-to-face interview, which, by its very nature, ‘is more confrontational and less companionable’ (Lee and Ingold, 2006: 80).

As people move, they pass through public spaces. The attendant sociability of mobile research is most obvious in urban environments where movement ‘inevitably involves social encounters’ (Pink, 2007: 244). However, it is also true of more remote locations where – even in the absence of others in the here and now – evidence remains of how others have engaged with the environment. In both contexts, people need to make (often minor) decisions about the environment that they want to leave for others. As a result, mobile research can facilitate observations and discussions of how people interact with ‘the people who live right here or over there; and about the particular person they just passed’ (Kusenbach, 2003: 474), and of the interactions that people have had in the past or hope to have in the future.

As people move, they are often also reminded about things that happened in the past in the same or similar settings. Movement through spaces provides important ‘context and stimulation to the mind’ (O’Neill and Roberts, 2020: 257), which, given the spatiality of memory, can aid with recall (Anderson, 2004: 256; Moles, 2008: 34). At the same time, mobile methods can encourage people to think of feared, likely, and/or aspirational futures. This temporal dimension means that moving together can be a useful way to discuss present realities, remembered or more distant pasts, and close or far off futures (Glass, 2016; Holgersson, 2018; Pink, 2007).

Mobile research can also provide observations and/or conversational prompts that help to narrate places. Mobility is ‘central to the production of place’ (Pink, 2007: 245), with places primarily becoming ‘what they are through everyday practices’ (Lund, 2012: 225). This means that one can gain a better understanding of space as an experienced place by accompanying people as they move through their world. Spaces also carry the scars and memories of the past and are often littered with insights as to possible futures, ensuring that mobile research – and the opportunity to discuss and survey people's interactions with space – can help to reveal the broader narratives that places have to tell. Such narratives can stem from background context about a case study location to revelations about the relationship between modernity and the individual (de Certeau, 1984); while movement can – in the way of Benjamin Walter's flâneur – provide a means to access ‘certain aspects of the truth’ about a place (Birkets, 1982: 170).

More generally, there is greater awareness of how ‘[t]elling a story and following a path [or road] are cognate activities’ (Turnbull, 2007: 142) and of how, just as ‘many narratives are about roads and journeys’, roads ‘elicit narratives’ (Argounova-Low, 2012: 195). Critical in this regard is the ‘flow, succession or development of one step to another, from one word to another’ (Argounova-Low, 2012: 197); the rhythm of movement facilitating thinking and talking (Anderson, 2004; Bazuń and Kwiatkowski, 2022; Evans and Hones, 2011). Particularly important are the moments of silence and reflection that mobility affords. While a long pause in a static interview can be uncomfortable and mark the end of a meeting, on a journey it can simply constitute a natural pause and a moment to reflect. This ‘pause-fullness’ (Laurier et al., 2008: 17) or ‘contemplative space-time’ (Guzman, 2022: 325) facilitates conversation, including on difficult topics (Bazuń and Kwiatkowski, 2022; Guzman, 2022; Laurier et al., 2008). This is important as it means that, in addition to conversational prompts and observational opportunities, mobile methods can be used to research things that have little or nothing to do with mobilities or the spaces and locations moved through.

Finally, there is something about the sociability of moving with others (Lee and Ingold, 2006), which, while it does not put ‘the researcher and researched on an equal footing’, disrupts the relationship between them (Holgersson, 2018: 74) and forefronts ‘the sharing of “moments” and rapport’ (O’Neill and Roberts, 2020: 19). When moving together, researcher and research participant tend to move side-by-side, share a similar rhythm and ‘virtually the same visual field’ (Lee and Ingold, 2006: 80), and, at least in a minimal sense, become companions for the journey (O’Neill and Roberts, 2020: 19). Together with the fact that the research participant can plan the route and lead or drive, this helps research participants to feel ‘more in “control” of the circumstance … and as more of an equal in the “exercise”’ (O’Neill and Roberts, 2020: 19). This unsettling can render conversations more natural; it can also help to build rapport and facilitate future access (Fenno, 1990: 71).

Of walking and driving together

While the advantages of mobile research are associated with mobility broadly speaking, many researchers prefer walking to driving. In part, this is a personal preference or reflection of practice with walking presented as a more ‘enjoyable activity’ (Holgersson, 2018: 82) and/or as ‘the most common’ (Duarte and Strasser, 2021: 1) and ‘democratic form of movement’ (Bazuń and Kwiatkowski, 2022: 570). At the same time, walkers are presented as engaging with multi-sensory environments to the fullest (Pink, 2007: 246). This is in contrast to car-dwellers, who – enclosed in a machine travelling at speed – are presented as ‘insulated’ and cocooned (Urry, 2006: 22) and as divorced from their senses, environments, and the rhythm of movement (Bazuń and Kwiatkowski, 2022; Evans and Hones, 2011; Lee and Ingold, 2006; Urry, 2006).

Certainly, feeling ‘through the car and with the car’ (Sheller, 2007: 181) can lead to a sensory disengagement. Some scholars then use this possibility to present walking as superior for mobile methods. In this vein, Carla Duarte and Anita Strasser posit that, while walking ‘encourages us to engage with our surroundings with all our senses, leading to new and different ways of knowing’ (2021: 2), the higher speed of cars alienates ‘the senses which feel confused with the stimuli that pass very quickly and are hardly understood by the brain’ (Duarte, 2021: 26; also Kusenbach, 2003: 465). In turn, Jo Lee and Tim Ingold stress how walkers seem ‘to have a real mobility, in terms of the ability to see in different directions and to discover the “little” things in his or her surroundings’, as compared to drivers who must remain focused ‘on the middle distance of the road ahead’ (2006: 70). James Evans and Phil Hones similarly argue that ‘researcher and participant are more exposed to the multi-sensory stimulation of the surrounding environment’ when walking, while drivers are ‘cocooned in a filtered “blandscape”’ (2011: 850).

Not only does one's encasement and speed affect one's interaction with sense-scapes, it also affects one's sociability. In this regard, John Urry has argued that car drivers lose ‘the ability to perceive local detail, to talk to strangers, to learn of local ways of life, to stop and sense each different place’ (2006: 23), leading ‘[c]ommunities of people [to] become anonymized flows of faceless ghostly machines’ (2006: 22). In contrast, walking is presented as inherently sociable (Lee and Ingold, 2006: 69) due to a different kinaesthetic rhythm. The argument is that, while ‘the rhythm of walking generates a rhythm of thinking’ (Bazuń and Kwiatkowski, 2022: 578), driving involves a ‘minimum of movement once one is strapped into the driving seat’ (Urry, 2006: 24) rendering it essentially a ‘sedentary [activity] from the bodily perspective’ (Evans and Hones, 2011: 850).

Clearly, the separation of a car's inhabitants from the places and locations passed through, the contained sense-scape, and the faster speeds alter the experience of driving together from walking together. However, this paper argues that the differences between the two have been exaggerated and that they can sometimes be beneficial to the research process.

In terms of exaggeration, it is sometimes more enjoyable and/or appropriate to drive than to walk. Many people find cars ‘comfortable, enjoyable, exciting, even enthralling’ (Sheller, 2004: 236) and appreciate the privacy, controlled soundscape, and status that it affords (Bijsterveld, 2010; Bull, 2003; Kent, 2015; Van der Geest, 2009). Many also drive as part of their everyday activities. The decision to do so stems in part from personal factors – from its affordability to journey lengths, physical abilities, and the value attached to privacy, control, comfort, and carbon footprints (Bijsterveld, 2010; Kent, 2015). The decision is also shaped by context. The quality of roads and footpaths, security, climate, automobility costs, and cultural associations help to determine whether driving is appealing and affordable and for whom. In Kenya, for example, walking is an everyday practice for pastoralists, hawkers, and for those who cannot afford public transport, whereas driving tends to be an everyday activity for the minority who can afford a private car and the associated trappings of car maintenance, insurance, and fuel (Salon and Gulyani, 2019). Furthermore, in a country in which walking often carries a ‘stigma of poverty’ (cf. Ingold, 2004: 322) and where there is an equatorial climate, a range of wild animals, no public footpath network, a relatively high crime rate, and much-maligned public transport (Mungai and Samper, 2006; Mutongi, 2017), driving tends to be the preferred mode of travelling even short distances for those that can afford it.

Moreover, while walking allows travelling companions to have a slower and more focused engagement with land and sense-scapes than is possible when driving, walkers are not necessarily fully engaged with their multi-sensory environments. Shoes and pavements can encourage a certain ‘groundlessness’ (Ingold, 2004: 329). Pedestrianised environments, such as malls, have become ‘progressively desensualized’ (Edensor, 2007). Poor weather can lead people to walk head down, while, even in good weather, walkers can find themselves focused inwards, oblivious to much of their surroundings, sometimes remembering only ‘episodes of contact’ (Lund, 2012: 232). So many people walk ‘wearing headphones, talking on a mobile phone, and/or looking down at an electronic device’ that ‘[t]echnology-related distracted behavior’ has become a recognised road-safety concern (Basch et al., 2015).

Similarly, cars do not necessarily constitute a ‘blandscape’ (Evans and Hones, 2011: 850) or cause a disassociation with ‘sensorial impressions’ (Duarte, 2021: 26). Cars have their own internal sense-scape, while car-dwellers engage with the landscapes and environments around them, albeit to varying degrees. Again, personal choices and context matters. Drivers can ride ‘slow and stop-start with (weather permitting) [their] window open to take in the sights, sounds, smells and tastes’ around them (Dawson, 2017: 17). Certain places are more likely to see cars sit in traffic, while certain roads and cars can lead the ground to literally reverberate through one's body. In the United Kingdom (UK) long stretches of (often banked) motorway render much long-distance driving devoid of many distractions besides the general countryside and occasional service station (Merriman, 2004). At the same time, relatively high speeds outside of residential areas undermine ‘the possibility of a fixing and penetrative look’ (Larsen, 2001: 88), while ‘the software revolution … has made possible an increasing disengagement of the driver from the work of doing driving’ (Featherstone, 2004: 11). The attention of drivers and passengers is therefore often focused on the soundscape within a car (Bull, 2003) and episodic disturbances to one's freedom due to traffic jams or parking problems (Hagman, 2006). In contrast, even short stretches of multi-lane highways in Kenya slice through densely populated areas are traversed by pedestrians and are bedecked by billboards (Manji, 2015). Indeed, a great deal of Kenya's economic and political activity takes place on and alongside roads, which are densely populated with markets, hawkers, adverts, proverbs, and political campaign paraphernalia (Klaeger, 2009; Mutongi, 2017; Quayson, 2010). At the same time, lower average speed limits and the greater ease of pulling over slows down the ‘travel glance’ (Larsen, 2001), which – together with the manual nature of most cars and commonality of rough roads and poorly maintained tarmac roads – ensures that car-dwellers are not entirely divorced from the multi-sensory environments around them.

Moreover, car-dwellers can sometimes engage with environments as well as or even better than walkers. Aspects of the built environment are designed ‘to make visual sense to the occupants of cars’ (Thrift, 2004: 46). Moreover, the fact that an increasing number of people experience and help to make places by driving – leading, at one extreme, to the emergence of ‘drive-through cities’ (Qamhaieh and Chakravarty, 2020) – means that an authentic engagement with place may be by car. As Mimi Sheller notes, ‘[i]n societies of automobility, the car is deeply entrenched in the ways in which we inhabit the physical world, but also the ways in which the world en-habits us’ (2007: 180).

Car journeys also do not necessarily lead to a disconnect between car-dwellers and the people and societies outside. On the contrary, state–society relations are often brought into particularly sharp relief by car journeys – from colonial and contemporary infrastructure projects to police stops and border controls (Bishara, 2015; Gewald et al., 2009; Masquelier, 2002; Melly, 2013). In turn, while walking provides much opportunity to interact with those that one meets, driving is not entirely devoid of the same. The use of horns, lights, and hand gestures as well as the elongated travel glances that can occur in traffic jams and interactions when one stops or pulls over are evidence of such. Indeed, in a highly segregated country such as Kenya in which people live, shop, eat, and access education and other social services in different spaces (Jimmy et al., 2020), no other part of daily life brings gated community-living car-dwellers closer to people of other socio-economic classes than does a traffic jam, where all sit side-by-side with bus and private minivan passengers and are entreated by hawkers and beggars (cf. Yazıcı, 2013). Even on Kenya's open roads, I have been struck by the frequency with which people stop to buy goods from hawkers, ask for directions, and greet people; by the frequency with which people (especially in rural areas) recognise cars; and by the extent to which cars engage with each other – from more regular use of the horn to conversations that arise when friends or neighbours meet on rural roads and stop, side-by-side, windows down.

Driving, like walking, also engenders a sociability with those that one rides with. In short, while drivers and passengers experience the car differently – the driver needs to concentrate on the road and the mechanics of driving – passengers often become involved in navigation and in watching the traffic such that ‘arriving at a destination can become a shared accomplishment’ (Laurier et al., 2008: 6; emphasis in original). Both the driver and passenger's positions are also fixed leading to a partially shared view and otherwise shared sense-scape. The fact that no-one can join or exit the conversation without stopping and the privacy that car journeys afford also allows ‘drivers and passengers [to] spend time together and have the flexibility to pursue conversations that are not predetermined by an explicit task’ (Guzman, 2022: 324; also, Laurier et al., 2008). The fact that driving takes varying levels of concentration (depending on the car, roads, and weather) but less physical energy, and that it involves higher average speeds than walking, also means that travel times and distances can be extended, which can expand the range of experienced events (Pearce, 2017). This means that car journeys can be a particularly good space in which to develop a sense of companionship and rapport (Fenno, 1990; Guzman, 2022).

Again, context matters. The relatively poor maintenance of many cars, roads, and signage in Kenya as compared to the UK – together with relatively low levels of adherence to traffic regulations, and high levels of corruption associated with police checkpoints (cf. Agbiboa, 2015) – ensures that car travel is often more eventful and can even have a ‘seemingly pathological capacity to cause injury and death’ (Klaeger, 2013: 360). These intertwined realities of occurrence and danger – from getting a car stuck in the mud with research participants on Mount Elgon to a car getting clipped by a lorry with shoddy breaks in Molo – have important implications for shared experiences and rapport (as well as for ethics and security). However, anywhere in the world, driving imbues a combination of privacy, a shared immediate environment and partially shared view, and range of often mundane experiences and interactions that can foster a sense of companionship and rapport that can equate with, and perhaps even surpass, that gained when walking (Guzman, 2022; Laurier et al., 2008).

Finally, car journeys are not devoid of rhythms. These can range from the slow stop-start of a city to the steady whirring of the open road (Dalakoglou and Harvey, 2012). Indeed, ‘everywhere where there is interaction between a place, a time and an expenditure of energy, there is rhythm’ (Henri Lefebvre cited in Middleton, 2009: 1956). The rhythm of walking and driving is shaped by one's gait and driving style, and by context – from the extent to which routes are punctuated by traffic lights to the density of traffic and crowds. Moreover, observations of in-car interactions between passengers (Laurier et al., 2008), ethnographic studies that have adopted driving together (Guzman, 2022), and personal experience all suggest that the rhythms of driving can facilitate thinking and talking.

It is thus an oversimplification to present walking and driving as sharply contrasting in terms of their enjoyability or naturalness, or in terms of people's engagement with multi-sensory environments, society, and travelling companions. The comparison instead depends on personal and contextual factors, the environments that one wants to engage with, and the research focus. However, while the differences between walking and driving have been exaggerated, important differences remain. The simple fact of being seated in a private mobile space allows (whatever the context) for a degree of separation between car passengers and the surrounding world; for greater privacy; for greater control over immediate environments; and for longer distances to be covered at greater speed. Or, to put it another way, while walkers share one space and sense-scape that they move through together, car passengers share a private space and sense-scape that then moves through and passed other spaces and sense-scapes. Moreover, while this central difference between driving and walking can undermine mobile research, it can also be beneficial to the research process.

Clearly, driving together is preferable if one is studying automobilities (Laurier and Lorimer, 2012) or if one is conducting participant observation or a go-along with people who tend to drive (Fenno, 1990). However, driving together may also be well suited to projects that focus not on ‘the “little” things’ (Lee and Ingold, 2006: 70) but on broader trends over larger distances, such as variations in levels of development and forms of policing between different neighbourhoods, areas, regions, and even countries (Bishara, 2015; Dalakoglou, 2010; Melly 2013).

More importantly, the aspects of driving that can dull one's engagement with immediate surroundings – the ability to leverage greater control over one's immediate environment, to travel at greater speeds and over longer distances, to pay less attention to other people (if not cars), and the knowledge that one cannot be overheard – can facilitate conversations that go beyond one's immediate surroundings. Indeed, from experience, walks around a particular neighbourhood tend to encourage conversations about that space and the memories and hopes that it evokes, while driving often encourages bigger picture conversations about life histories, experiences, potential futures, and/or more mundane conversations about local socio-economic, cultural, political, geographical, or historical contexts. Moreover, the privacy, constrained mobility, and slow and extended conversations that driving affords can be a particularly useful space in which to talk about more sensitive, difficult, or emotional topics (Ferguson, 2016; Guzman, 2022; Laurier et al., 2008).

Driving together can thus potentially prove useful at different stages of a research project – from revealing an initial puzzle or question to providing important context and/or primary data – and for a wide range of research projects and disciplines. Indeed, driving together can potentially be incorporated into any project that is focused on, or can glean useful insights from, the social settings or scenes and wider locations and places that cars can move through, and into any project that would benefit from the narrative storytelling or pause-full conversations that driving can facilitate.

The differences in driving together in different contexts should also not be taken to imply that driving together is generally more appropriate in certain contexts. Instead, it is simply to highlight how the appropriateness of driving together will vary depending on the intersection between research focus and contextually specific realities. For example, while the automobility of the UK renders driving together a good space in which to observe and discuss how people experience places such as suburbs and activities such as commuting, the elitist nature of car driving in Kenya would limit one's sample to the relatively wealthy. At the same time, the slower travel glance, relatively low penetration of new car technology, greater sociability of roads, and extent to which life and politics is lived on and along roads means that it is easier to engage with landscapes and sense-scapes while driving together in Kenya than in the UK. The fact that driving together can – depending on research focus and context – be as useful as, or more useful than, walking together leads to two final issues: the methods that can be adapted to that space and the ethical and security considerations.

Driving together

From walking interviews to driving interviews

Walking interviews – whereby a researcher and research participant walk-along a route (often pre-planned) and discuss the spaces and places that they move through – are increasingly common (O’Neill and Roberts, 2020). The idea is that ‘walking is an excellent method for entering into the biographical routes, mobilities, and experiences of others in a deeply engaged and “attuned” way’ (O’Neill and Roberts, 2020: 15). Walking interviews can be productively adapted to driving if much is visible from the road, research participants tend to travel the route in question by car, the route covers a large area, and/or selected sights are relatively spread out.

For example, following Kenya's post-election violence of 2007/8, I conducted several driving interviews with civil society leaders with the aim of discussing the displacement that had taken place and efforts to assist those affected. The routes were pre-planned by research participants so that we could pass areas with resettlement and rehabilitation programmes. At one level, these driving interviews allowed me to see the scars of displacement across a landscape and the new buildings being constructed, and to gain a sense of the scale of the problem. At another level, the driving interview helped to facilitate more detailed conversations about my travelling companions’ projects than had hitherto been possible in classic semi-structured interviews. This is because the tour provided important conversational prompts, ensured a better shared understanding of the relevant environment and political geography, facilitated moments of reflection and a periodic return to topics (for example, as we passed yet more evidence of arson or resettlement), and helped to build rapport. Regarding the latter, my request to see the impact of displacement and an organisation's projects was clearly appreciated, while the act of driving together unsettled the classic interview process. This was openly appreciated by one participant who reported being tired of speaking with researchers in the wake of this unprecedented crisis.

From participant observation to driving around, and walk-along to ride-along

Participant observation is often associated with static activities such as attending a meeting or a political rally. However, if one's research participants move as part of the activity observed, then participant observation will ideally involve moving with them; and, if their movement is by private car, then participant observation will involve driving around with them for short or extended periods depending on the individual(s) programme (Fenno, 1990: 81).

When the focus of the research and of the observations and discussions shifts away from a general observation of activities to a more focused attempt to ‘actively explore … [a] subjects’ stream of experiences and practices as they move through, and interact with, their physical and social environment’, then the research method adopted shifts from participant observation to a ‘go-along’ (Kusenbach, 2003: 463). As Margarethe Kusenbach explains, for ‘authenticity, it is crucial to conduct … “natural” go-alongs’ and to ‘follow informants into their familiar environments and track outings they would go on anyway as closely as possible’ (2003: 463). In turn, if the research participant's normal routine is to travel by car then a go-along becomes a ride-along (Kusenbach, 2003).

In practice, the boundary between driving around and a ride-along is blurry. Both participant observation and go-alongs combine observation and discussions. Both methods require that the mode of mobility be determined by the research participant's routine. However, in participant observation, the researcher tries to interfere as little possible. For example, during participant observation of Kenya's Truth, Justice and Reconciliation Commission (TJRC) (Lynch, 2018), I occasionally caught a lift with a commissioner or commission staff member from public hearings. During these drives, we sometimes sat in silence as my travelling companion caught up with the news or their driver, but when discussions did arise, they usually focused on their work and on the public hearing just held.

On a drive-along, in contrast, the researcher seeks to actively capture ‘the stream of perceptions, emotions and interpretations that informants usually keep to themselves’ about their environments (Kusenbach, 2003: 464), which requires an on-going semi-structured interview largely focused on the research participants experiences of, and interactions with, the physical and social environment passed through. In this vein, and as part of a project on elections in Ghana, Kenya, and Uganda (Cheeseman et al., 2020), I accompanied several politicians in their private car on the campaign trail. As we drove around, I was able to observe their strategies and interactions about where and when to stop, with whom to speak, and how they were greeted by local communities. I was also able to engage in extended interviews about their campaign strategy as it related to the areas that we moved through and the sections of society that we interacted with.

It is important to note that the TJRC commissioners, staff, and politicians that I have driven together with all had drivers, meaning that – in contrast to my experience of driving interviews and motoring – I sat in the back while the research participant either sat up front or in the back, depending on their routine. This experience echoes the findings of others on the effect of people's positioning within the car on social interactions – from leaning forward to speak to those in the front to the different visual field and lower likelihood of becoming involved in navigation (for more see Dahl and Tjora, 2023; Laurier et al., 2008).

From bimbling to motoring

While walking interviews are focused on local environments, and mobile participant observation and walk-alongs focus on routines or practices, bimbling is not primarily concerned with either. Instead, it is simply ‘the process of walking and talking that is important’ when bimbling. The idea is that walking ‘allows talking to flow naturally because the pressure of a face to face interview has been removed’ (Kinney, 2017: 2) and because the rhythmic relaxation of walking frees the imagination (Anderson, 2004: 258). The common association of walking as a more natural and enjoyable experience helps to explain why there is no clearly articulated driving equivalent to bimbling (or, at least, not one that I have come across). However, the fact that driving can be as enjoyable and relaxing as walking, while the privacy and pause-fullness of driving can encourage conversations on a wide range of topics, ensures that bimbling is clearly adaptable to driving. I coin bimbling-by-car as motoring – a reference to the recreational and/or pootling nature of these types of car journeys.

Motoring captures some of my drives around rural areas with research assistants to conduct interviews or attend meetings. There are a handful of research assistants that I regularly employ in Kenya who – with backgrounds in local civil society or media, and with lived experience of researched topics – are people that I first met as interviewees and who are incredibly well informed about local histories and politics. As a result, I have found that on our drives around research locations, the informal setting of the car, the hours spent together, and the pause-fullness of conversation have helped to bring new insights to light. Critically, while our car conversations have sometimes been shaped by prompts from inside or outside of the car (from news on the radio to billboards or accidents passed), they have ranged much more broadly – from aspects of my companion's personal biography to the current political context – in ways akin to the free flow of conversation hailed by bimbling's proponents (Anderson, 2004; Kinney, 2017).

Technology on the road

While I have not systematically used voice, photo, or video recording devices, or global positioning technology, when driving together with research participants, it is clearly possible to do so (Laurier et al., 2008; Yang et al., 2023). This means that the docent method where photos taken on a pre-planned route are discussed after a walk (Chang, 2017) can also be adapted to driving.

Accidental ethnography on the road

Finally, one's movement around research areas as part of a qualitative research project can provide invaluable opportunities for accidental ethnography, defined by Lee Ann Fujii as the paying of ‘systematic attention to the unplanned moments that take place outside … structured methods’ (2015: 525). A trip to a marketplace to buy vegetables, for example, can provide an opportunity to observe local political economy activities and interactions and to have short informal conversations about the same. While a trip to a bank or government office can provide an opportunity to observe local bureaucracy at work. Moreover, while such accidental ethnography is not usually recognised by researchers – with most giving little thought to pre-existing knowledge or to opportunistic data collected through everyday experience (Riemer, 1977) – it can play a critical role in understanding important research questions and puzzles, in considering what to ask people about and how, and in interpreting data (Fujii, 2015; Poulos, 2018).

Accidental ethnography is also possible when driving. Indeed, from experience, drives that one undertakes with others – for example, with a taxi driver from an airport to a hotel or between meetings – can provide a particularly valuable opportunity to benefit from such systematic attention to unplanned moments. The reason is simple: one has a travelling companion with whom one can discuss the people, scenes, places, and sense-scapes passed (from traffic levels and new infrastructure projects to roadside markets and vehicle slogans) as well as the car's own sense-scape (from news on the radio to popular local snacks). This enables the researcher to sense check their interpretations of such opportunistic data collection. It is for this reason that I always ask to sit in the front of a taxi and use prompts from passing sense-scapes to strike up a conversation. For example, while conducting research on elections in Ghana, Kenya, and Uganda, the campaign posters and billboards passed often aided conversations about the candidates, their electoral strategies and alliances, and constituencies. These conversations exposed various insights that I had not garnered through structured methods, but which I was then able to further investigate, unpack, and discuss in subsequent interviews. The fact that my experience of driving together has been in areas and countries that I am not from is significant as it means that many of the insights that I have gleaned are things that locals would already have known. This is not to say that accidental ethnography is only useful for outsiders, just that it may be particularly useful for those with limited knowledge of a case study area. For example, while a researcher who knows an area well will be well placed to discuss local changes over time and other context-specific insights, car journeys can be an excellent way for a visitor to discuss basic context – from local development projects to electoral candidates and aspirational lifestyles (Melly, 2013).

Driving together as a research space

The fact that driving together can involve such varied research methods and encourage accidental ethnography ensures that it is best considered not as a method but as a research space. Understanding driving together as a research space allows for the possibility that research methods might shift during a car journey. For example, a drive-along can shift into participant observation when one's research participant needs a break from an active discussion of their ‘perceptions, emotions and interpretations’ (Kusenbach, 2003: 464). Thinking of driving together in this way also allows for the possibility that one can stop a car and continue the same method beyond its confines. For example, when conducting participant observation, one may have periods of driving around that are interspersed with other forms of mobile or static participant observation (Fenno, 1990). Similarly, driving and walking interviews can easily be combined: cars can be stopped and parked, and passengers can opt to get out to look at something more closely or carefully, or to explore a particular area on foot.

The ability to adapt walking methods to driving raises the question of when it is appropriate to do so. As already noted, the appropriateness of driving together is determined in large part by the intersections between research focus and context. However, it is also determined by whether driving together can be conducted ethically and safely.

Research ethics and security

Many of the ethical considerations raised by driving together need to be considered in relation to the research method in question. For example, the practice of acquiring informed consent and need for participant information sheets varies between accidental ethnography (for which such paperwork is unnecessary and anyhow impossible) and semi-structured interviews with targeted research participants (for which such paperwork has become standard practice). This paper does not dwell on such methodologically specific considerations and instead focuses on more general ethical and security concerns.

Given that driving contributes to congestion and pollution, and there is always a degree of risk related to traffic accidents, it is critical that driving together has clear added value. The appropriateness of driving together is also shaped by how accustomed research participants are to private car travel and any sampling biases that may result. For example, if studying how Kenyan politicians campaign, then a sample of regular car users would directly correlate with one's targeted group and therefore be unproblematic. However, if studying the reception of such campaigns, driving together would unfavourably bias one's findings towards wealthy citizens.

If driving together is deemed appropriate, the next question is whether the researcher and research participant(s) will be safe and feel safe. At one level, this is a logistical question that requires attention be given to car maintenance, insurance, local traffic regulations, fuel, routes, and timings. In practice, many of the logistical concerns can be addressed by using legitimate car-hire or taxi companies, checking that a private vehicle has been recently serviced and is taxed and insured, careful planning of one's journey, and ensuring that one has the resources to facilitate a safe journey including a charged phone and money to pay for a mechanic, if need be.

At another level, safety relates to inter-personal relations and requires a base level of trust and respect. Things to consider include the way in which the researcher and research participant met, the length of time that they have known each other, the areas to be visited, the topics to be discussed, and the number of people travelling in the car and their relationship to each other. The fact that the researcher and research participant may be seen together as they get in or out of the car or sit in traffic may also raise context-specific reputational and security concerns. For example, it may be ill-advised for a man and woman to drive together in highly conservative societies. However, such concerns may also be less immediately obvious, requiring researchers to have a good understanding of local context. For example, when the International Criminal Court (ICC) had cases ongoing in Kenya, interactions with a visible outsider in the Rift Valley could prompt allegations that a research participant was an ICC witness, which could result in real threats to their life (de Brouwer, 2015). In contrast, for research participants who fear being overheard in common interview locations (such as their office or a restaurant), driving together may be regarded as safer than a static interview.

If driving together can be done safely then the next consideration is comfort. This includes meal and washroom breaks and due respect for any signs that one's travelling companion would like to be left in peace for a while to listen to the radio, read, make phone calls, stare out the window, sleep, or, when the research participant is the driver, simply concentrate on the road ahead. Since driving together can trigger memories, consideration should also be given to whether those memories are likely to prove traumatic or difficult for those involved, and to whether driving together is appropriate and, if so, how any re-traumatization might be minimised or mitigated against.

The fact that ‘in the car you cannot walk away from or walk into a conversation with another speaker’ (Laurier et al., 2008: 9) also raises more general challenges in terms of the research participant's right to opt out of the research, either temporarily or permanently. It is therefore essential that the researcher respects the research participant's right to change the subject, and, if the research participant wants to opt out entirely that passengers have alternative means of reaching their destination. If the researcher is the passenger this requires that they can get home independently of the research participant, and if they are the driver that they can support the research participant to reach their destination by another means.

Driving together also complicates informed consent. Given that driving together is an uncommon method, it may well take more time to explain what is involved. Agreement should also be reached on realistic timetables and about what should be done if the researcher and research participant interact with others (Kinney, 2017: 3). One of the benefits of driving together is that it can help to build rapport, which, as already noted, can facilitate conversations and future access. However, while driving together can unsettle power dynamics, it does not remove them (Holgersson, 2018: 74) since the researcher retains control of the research questions and usually ‘upholds a monopoly of interpretation’ (Kvale, 2006: 484). In turn, while an informal setting can serve ‘as a means to efficiently obtain a disclosure of the interview subjects’ world’ and can blur the lines between moments of research and everyday interaction, this can be problematic if the research participant ends up ‘laying their private lives open and disclosing information to a stranger, which they … later regret’ (Kvale, 2006: 482). In terms of practical solutions to this dilemma of wanting to make a research participant feel comfortable but simultaneously not lose sight of the nature of the interaction, one possibility is to adopt continuous consent (Klykken, 2022), whereby participants consent to participate knowing that, at the end of the journey, they can redact information from a transcript and notes. Alternatively, the researcher can fully anonymize all data collected even if informed consent is given to recognise the possibility that interviewees may let their guard down.

Conclusions

There has been a shift from viewing research as a static process that is conducted in particular places to recognising how mobile methods can bring a range of benefits – from providing particular moments of discussion and observation on people's mobilities, personal histories, and social interactions to facilitating more general conversations. However, while the versatility of mobile methods is now well recognised with respect to walking methods, driving is often limited to observations or recordings of in-car interactions or ride-along. Moreover, walking is often presented as superior to driving due to claimed differences in people's engagements with multi-sensory environments, societies, and other people, and due to different kinaesthetic rhythms.

This paper has shown how the differences between walking and driving together have been exaggerated and how these differences can sometimes be beneficial to the research process. Just as importantly, the paper has set out how walking methods can be adapted to driving, introduced the method of motoring as a driving equivalent of bimbling, and argued that driving together can facilitate accidental ethnography. The conceptualisation of driving together as a research space or, more specifically, as a constant space that moves through sequential spaces, also reveals how methods can shift within a car journey and how mobile methods can move in and out of cars. Finally, the paper has shown how driving together has much broader applicability than is usually recognised, and how the appropriateness of using this research space as part of a project is dependent on research foci and contexts and on ethical and security concerns.