Abstract
In this paper we unpack how geographers have studied institutions, focusing specifically on institutional ethnography, often called ‘IE’. Sociologist Dorothy Smith is widely credited with developing institutional ethnography as an ‘embodied’ feminist approach. Smith studies the experiences of women in daily life, and the complex social relations in which these are embedded. Institutional ethnography offers the possibility to study up to understand the differential effects of institutions within and beyond institutional spaces and associated productions of subjectivities and material inequalities. We suggest that geographical scholarship on institutions can be enhanced and, in turn, has much to contribute to the broader interdisciplinary field on institutional ethnography, such as understandings of institutions that account for spatial differentiation. We argue that IE holds potential to enrich geographical research not only about a multitude of kinds of institutions, but about the many structures, effects, and identities working through institutions as territorial forces. In spite of recent interest by geographers, the broader literature on institutional ethnography remains under-engaged and under-cited by human geographers. Critical of this lack of engagement, we suggest that it has left a gap in geographical research on institutions. Our aim is to analyze and advance existing scholarship and offer this article as a tool for geographers thinking about employing IE. We develop a typology, categorized by methodological approach, to highlight ethnographic approaches to institutions undertaken by geographers.
Over the last two decades, geographers have expressed renewed interest in ethnography (Herbert, 2000) and institutions (e.g. Anderson, 1991; Herbert, 1997; Schoenberger, 1997; Del Casino et al., 2000; Hyndman, 2000; Nevins, 2002; Mountz, 2010), and growing interest in institutional ethnography. Whereas Steve Herbert (2000) argued ‘for ethnography’ in this journal 15 years ago, we argue for institutional ethnography in geography. Given geographers’ different ways of studying institutions and conducting ethnography, this article parses out approaches by unpacking how geographers go about studying institutions, with particular attention to institutional ethnography.
Institutional ethnography was originally conceptualized and coined by sociologist Dorothy Smith (1987, 2006), developed subsequently by other feminist social scientists (e.g. Campbell and Gregor, 2004; Devault, 2006), and operationalized recently by geographers using a range of methodological approaches. Institutional ethnography has explicit critical or liberatory goals in its exploration of processes of subordination. Rooted in Marxist and feminist scholarship, institutional ethnography encompasses an integrated approach. Smith (1987) developed institutional ethnography as an ‘embodied’ feminist approach. Her first analysis began from the example of single mothers constructed as ‘defective’ within dominant social narratives, including schooling and health care. The role of the researcher in this case is to uncover the ‘ruling relations’ that produce single mothers as outside the norm, exposing how teachers and other institutional actors are bound up in the production of dominant narratives and practices. Institutional ethnography brings to the fore these kinds of ‘problems’ in the system that discriminate against single mothers and maps paths to social change.
Institutional ethnography is valuable, useful, and productive for geographers as it holds potential to broaden their work on institutions – including their conceptualization, socio-spatial relations, effects in daily life – and potential contributions to social justice movements. In spite of recent interest among geographers, the broader literature on institutional ethnography (often called ‘IE’) remains under-engaged and under-cited by human geographers. Critical of this lack of engagement, we suggest that it has left a gap in geographical research on institutions. Rather than explore the reasons why geographers have not engaged existing scholarship more deeply, we explore its potential contributions to geography, and from geography, in turn, to the broader field of IE. Our aim is to analyze and advance existing scholarship and offer this article as a tool for geographers thinking about employing IE, which is practiced in a variety of ways.
We argue that IE holds potential to enrich geographical research not only about a multitude of kinds of institutions, but about the many structures, effects, and identities working through institutions as territorial forces. We advocate for continued development of institutional ethnography and find that geographic scholarship on institutions will be much enhanced through engagement with scholarship beyond the discipline. Furthermore, engagement with scholarship beyond the discipline’s borders will not only enhance scholarship on institutions, but potentially enhance contributions of research on institutions to social justice movements.
We begin with a brief overview of how geographers have studied institutions over time in order to situate the recent upswing. We then explain institutional ethnography in more detail as an interdisciplinary field developed by feminist scholars from the late 1980s to the present. In our penultimate section, we examine how geographers have recently used institutional ethnography with a typology of approaches. The section ends with an exploration of avenues to more sustained, critical engagement with other social scientists working on and in institutions. We conclude with a summary of contributions and new questions.
I Placing IE in geography: A necessarily incomplete genealogy of disciplinary engagement with institutions
Geographers have debated methods with which to research institutions (Flowerdew, 1982; Del Casino et al., 2000; Herbert, 2000) and epistemological frames through which to understand them (Philo and Parr, 2000). We aim here to situate geographers’ recent interests in institutions within a broader history within the discipline, without necessarily establishing direct movement or causality between one moment of institutional engagement and the next (as in the genealogical tradition). Underlying shifts in the discipline will be recognizable, including humanism and managerialism of the 1960s and 1970s, the cultural and institutional turns of the 1980s and 1990s, and the influence of postmodern, poststructural, and feminist thought in the 1990s and 2000s. Geographers have examined institutions in fits and starts over time, often lacking the history of disciplinary engagement and contemporary scholarship on institutions thriving in other disciplines (e.g. Iskander, 2010; Rodriguez, 2010). The result is a fragmented history of engagement.
Our review begins in the 1960s and 1970s, when geographers studying institutions were largely managerialists who, influenced by behavioralist geographers, tended to critique the idea of the institution as monolith and analyze the role of different government institutions, their effects on cities and populations, and their geographical patterns (Pahl, 1977; Flowerdew, 1982; Ley, 1983; Kariya, 1993; Philo and Parr, 2000: 515).
Humanist geographers challenged the work of managerialists, examining how managers shape social realities through the internal cultures of organizations. Ley, for example, contested a Weberian approach that conceptualized institutions as efficient and rational bodies with ‘perfect access to information’ (1983: 220). He observed a ‘lack of everyday empirical analysis’ (Ley, 1983: 220). This dearth of empirical evidence and the failure to examine quotidian practices are recurring themes in geographers’ approaches to institutions. Such an inductive approach would pay more attention to the everyday contexts out of which organizational actions emerge and to the meanings of events to organizational members who lie behind their initiatives and responses (Ley, 1983: 225). As precursor to contemporary forms of institutional ethnography, Ley (1983) explored internal subcultures of urban institutions and the rise of organizational consciousness that accompanied explosive growth in institutions in the 1960s and 1970s. He critiqued the ‘pure’ or ‘total’ organization of Weber and of Goffman (1961), analyzing instead how distinctly and unevenly institutions operated on the ground (see also Kariya, 1993).
Geographical analyses of institutions have shifted between macro-theoretical structures to micro-level theories. Some approached them at finer scales, focusing on spatial arrangements within institutions (Philo and Parr, 2000: 514). Others deemed this approach a ‘middle way’; following Giddens’ (1984) structuration theory and associated debates about structure and agency in the 1980s and 1990s, institutions offered an ‘interim’ level of social systems, or the social practices that regulate daily life. This meso-level approach proved significant among economic geographers who responded to the broader institutional turn within geography with an understanding that economic decisions were rooted in social institutions (Martin, 2000; Jessop, 2001).
Differing epistemological approaches contribute significantly to how institutions and organizations are defined and studied methodologically. 1 Del Casino and co-authors (2000) outline three frameworks for analysis: spatial science, critical realism, and poststructuralism. They argue that spatial scientists tend to believe that social relations can readily and literally be mapped onto the landscape through rules and patterns that govern and explain human behavior. Critical realists similarly believe in the material realities and effects of institutions. In contrast, poststructuralists tend to look beneath the surface to understand the underlying conditions, social relations and discourses that brought such material relations into existence.
Drawing on Foucault (1970, 1995, 1997) and exemplified by Gibson-Graham (1996), a constructivist approach focuses on discourses, such as economy, society, and politics, ‘that bring forth objects and events and determine their relationship to one another’ (Del Casino et al., 2000: 526). In this framework, institutions do not emerge out of preexisting conditions, but are produced and discursively constructed as entities with particular social significance and subjectivities (Del Casino et al., 2000). In the Foucauldian tradition, scholars mapped processes of deinstitutionalization and the entrance of mechanisms of social regulation into the realm of the social body. This approach understands subjectivity to be constituted through webs of legal, medical, and social relationships through which power shifts, operating in more dispersed, topological fashion (Martin and Secor, 2014) and highlighting the relationship between social and spatial relations. Constructivism fueled development of new approaches to the state and other institutions that grew more concerned with discourse and representation (e.g. Olsson, 1974; Jackson, 1989; Anderson, 1991; Driver, 1991) and critiqued spatially fixed and bounded notions of institutions. Constructivists understood institutions as actively produced through daily social contexts.
Rather than as a repressive, autonomous body that affects social relations, poststructuralists conceptualize institutions as themselves social constructions able to produce knowledge and identities. This approach – influenced by Foucauldian understandings of power and discourse and actor network theory, among other ideas – understands institutional powers and effects as dispersed, embedded, and entangled, with more permeable boundaries dividing what and who lies within and beyond the institution (Rutherford, 2007) and tends to focus on operations beyond the architecture of the institution, such as subjectivity formation and daily life (Ashutosh, 2010). This approach entails looking at institutions as sites where employees enact policies across time and space, where the everyday relations among those theoretically conceived of as ‘outside’ bleed into daily institutional life and vice versa. This geographical imagination of the institution posits boundaries as fluid in daily practice. Such an understanding requires a method that holds quotidian life as its main focus: ethnography.
Ethnography holds potential to address daily empirical knowledge on institutions and often reveals the unevenness of institutional practices and effects. As the study of daily life, ethnography has been important to the discipline since the early days of cultural geography. Duncan and Duncan (2009) note that Carl Sauer’s detailed observations of the landscape drawn from interviews, archives, and observations would today likely be considered ethnographic. The method came fully into practice through humanistic geographers (Ley, 1974) mapping detailed observation of daily interactions between individuals and their environment. The recent surge in ethnography once again examines the cultural dimensions of daily work of institutions (e.g. Herbert, 2000; Hyndman, 2000; Mountz, 2010; Belcher and Martin, 2013; Kuus, 2013; Delaney, 2014). Many ethnographies can be found among recent doctoral dissertations, evidence of renewed interest among a new generation of geographers (Ashutosh, 2010; Hiemstra, 2011; Houston, 2011; Lindner, 2012; Santiago, 2013; VandeBerg, 2013). Institutional ethnographies have grown in popularity as one incarnation of this trend (e.g. Perreault, 2003a, 2003b; King, 2009; Larner and Laurie, 2010; Billo, 2015).
II What is institutional ethnography?
Ethnography is the detailed study of everyday life, the ethnographer’s tools participant-observation, fieldnotes, and interviews (Emerson et al., 1995). Crucially, ethnography involves more than the conduct of interviews. Participant-observation and archival analysis enable the ethnographer to study how people interact and interpret meaning: ‘what people do as well as what they say’ (Herbert, 2000: 552). This need to observe is crucial to the workings of institutions. While interviews lend insight into actors and operations of institutions, participant-observation, fieldnotes, and detailed archival study enable spatial analysis and associated insights into power relations. Where, for example, are different workers located within a building and by social location? How are embodiments and encounters gendered, racialized, classed, and sexualized? The ethnographer unravels patterns of behavior and interaction, categories of identification, modes of management, exercises in power and interpretation in everyday life.
Ethnographic approaches to institutions have been popular since anthropologist Laura Nader published a seminal piece in 1972 where she argued that anthropologists ‘study up’ in order to better understand how institutions structure daily life. She suggested that through this effort, anthropologists could expand analyses beyond those marginalized peoples upon whom they had built the discipline. She emphasized how little most people knew about bureaucracies and organizations that had lasting material effects on them (1972: 294). She argued that people should have access to institutions and knowledge about how they function: A democratic framework implies that citizens should have access to decision-makers, institutions of government, and so on. This implies that citizens need to know something about major institutions, government or otherwise, that affect their lives. (Nader, 1972: 294).
Dorothy Smith’s conception of institutional ethnography relies on a dispersed model of the institution and its effects, with a focus on mapping the daily lived ‘relations of ruling’. Smith studies processes and practices that determine social relations from standpoints grounded in the everyday. Her approach to institutional ethnography begins with situated experiences of women in daily life, then explores relations in which these experiences are embedded: a complex of interactions between individuals, institutions, and society. Smith argues that sociology’s ways of knowing the world operate within the framework of dominant institutions that devalue women, differentially positioned by class, race, and other axes of difference. Social relations engender ‘relations of the ruling’ that guide, control, coordinate and regulate societies. Smith developed this approach to create ‘a sociology for women’, to explore how women are ‘organized and determined by social processes’ that extend beyond their immediate everyday worlds (Smith, 1987: 152). For DeVault (2006: 295), ‘ruling relations’ function not simply as ‘heuristic device’, but closely connect the contemporary everyday with historic, capitalist relations that privilege certain understandings of motherhood and women over others. These relations are represented visually in feminist geographer Isabel Dyck’s (1988) diagram in Figure 1.
Isabel Dyck’s (1988) p. 131 adaptation of Dorothy Smith’s institutional ethnography. Reproduced with kind permission of the author.
In feminist approaches to institutional ethnography, these processes are constructions of ‘text-based methodologies and practices of formal organization’ (Smith, 1987: 152–3; 2006). Methods include interviews, participant-observation, and textual analysis. The interview process includes not just a focus on the ‘subjective state’ of the interviewee, but a means to move onto next steps in an ongoing process of inquiry (DeVault and McCoy, 2002). The researcher must identify the key ‘problematic’ before moving to the next stage of analysis. Interviews prove insufficient to address ‘core power issues that impede social change’ (Winkelman and Halifax, 2007: 132). Therefore, institutional ethnography also often involves archival work, textual analysis, or recollection, as determined by the investigation (Smith, 1987).
In this tradition, analysis emerges deductively from empirical observations attentive to the role of policy documents such as medical charts and plans as key organizing sites. IE is designed to ‘reveal the organizing power of texts’ in order to link local and extralocal activities (DeVault, 2006: 295). Texts allow the researcher to ‘reach beyond the locally observable and discoverable into the translocal social relations and organization that permeate and control the local’ (Smith, 2006: 65).
Smith designed IE to facilitate collaboration among researchers (DeVault, 2006). The approach is considered an ongoing, evolving practice, rather than a clearly demarcated or completed project; the practice grows through networking, relationships, and group meetings among feminist scholars. The field has expanded beyond sociology, extending into other disciplines with both professional and scholarly applications (Campbell and Gregor, 2004; DeVault, 2008), the ideas especially influential in nursing (Winkelman and Halifax, 2007), social work, and education (DeVault, 2006).
An institutional ethnography begins with identification of an experience, followed by noting the institutional processes that produce the experience, and then investigation of processes identified (Dyck, 1997; DeVault and McCoy, 2002). Dyck (1988) exemplifies this approach in her adaptation in Figure 1. She explains in a subsequent publication how Smith’s conceptualization of institutional ethnography framed research with immigrant women: ‘we considered the way women talked of their experiences as a starting point to discovering the social relations organizing their day-to-day lives’ (1997: 189). What emerged out of these studies were complex and contextual understandings of identity, gendered and racialized subjectivities, and ‘more nuanced accounts of relations of oppression’ (Dyck, 1997: 198).
In their didactic text Mapping Social Relations, Campbell and Gregor (2004) show how the approach is rooted in feminist praxis and a politics of location, committing researchers to begin from their own socioeconomic locations and experiences. From there, researchers endeavor to address ‘problematics’ and focus on ‘puzzles emerging in everyday life’ (2004: 7). IE does not set out to develop generalized theory, but rather to explicate ‘experiential data’ (Campbell and Gregor, 2004: 8). Campbell and Gregor explain frequently how IE is distinct from ‘conventional research’ in its requirement that the researcher be ‘a knower located in the everyday world and find meaning there, in contrast to reliance on library research and the application of theories’ (2004: 11). Winkelman and Halifax (2007) distinguish between IE and ‘conventional ethnography’ by suggesting that IE takes both subjective and objective views of social relations, whereas conventional ethnography focuses more on participants’ or insiders’ views (Winkelman and Halifax, 2007). 2 As such, the approach is consistent with the politics of location, positionality, and self-reflexivity explored by feminist geographers (Katz, 1996; Mullings, 1999; Pratt, 2000; Moss, 2002).
Scholars in this tradition collaborate, engage with each other’s work to build upon previous work and advance the field. Building on this understanding and operationalization of IE, others have focused on neoliberalism and post-Fordist restructuring as an extension of Smith’s work (see Naples, 1997). Neoliberalism is analyzed as ‘meta-discourse’ that determines particular institutional relationships, making way for other IE studies to focus on the ‘intellectual institutions’ that develop the meta-discourses (DeVault, 2006). Ultimately, scholars doing IE connect through their scholarship with an ontological commitment to examining ‘ruling relations’ at work in historically-specific activities. Individual studies build toward meta-discourses that cross social arenas and create spaces for new discourses and political projects to emerge (DeVault, 2006).
While conceptually inspiring, we note two shortcomings that make space for geographical contributions to IE. First, IE produces somewhat aspatial understandings of institutions. The use of the term mapping in this approach is a figure of speech. Smith endeavors to write for the masses in ways that should be readily accessible as a map would be (Campbell and Gregor, 2004: 9). Mapping itself and spatial analysis more broadly remain largely absent from the approach. While deeply rooted in the everyday and feminist epistemology, the approach lacks spatial analysis of the institutional geographies and their effects. The method of IE does not place as much emphasis on differentiation between different spaces of the institution. As a result, the institution remains flat, leaving room for more complex analyses of the spaces of institutional productions of power. Second, IE is conceptually problematic in its quest for a ‘meta-narrative’ that risks losing sight of the messiness of institutional relationships in everyday contexts.
In contrast, geographical approaches to IE can and do account for the spatial differentiation by locating marginal spaces and spaces of exception, for example, within, through, and beyond the institution. Geographers often examine institutions as structures that influence society: asylums and hospitals, for example, seek to control and regulate, restrain or treat human minds and bodies (Philo and Parr, 2000: 514). Philo and Parr (2000: 515) argue that the geography of institutions and their relative location to people, land uses, and resources contribute to understanding social and spatial relationships that can challenge this disciplinary process. Institutional analyses address daily happenings within and between institutions and their relationships to larger economic, political and cultural processes (Philo and Parr, 2000: 514–15).
Methodologically, institutional ethnography enables location of the institution in the spatial relationships of multi-scalar everyday interactions to avoid characterizing it as a ‘repressive autonomous body that affects social relations’ (Mountz, 2010: xxiv). Ethnographic analysis can point to the ‘frustrations, subversions and networks’ amongst various actors, contributing to the ‘breaking points theorized as an institutional arrangement of social practices’ (Mountz, 2010: xxv). Ethnography and attention to the spatial relationships in these processes can link the particular histories of places with ongoing and overlapping processes of claims to territory and sovereignty.
Geographers’ ethnographies can document and produce more geographically textured understandings of institutions. In Herbert’s (1997) ethnography of local policing carried out by the Los Angeles Police Department (LAPD), for example, the method of conducting ride-alongs with officers draws into relief individual officers’ interpretations and productions of territoriality in local neighborhoods. His approach also highlights the nuances and intimacies of boundaries drawn between private domestic and public policing spheres.
Social scientists ‘study up’ by researching institutions of all kinds. Critical ethnography enables an approach to the state as ‘a set of social practices’ (Painter, 1995: 34) and diverse institutional actors exercising agency through quotidian bureaucratic arrangements (e.g. Herbert, 1997). Ethnographies of the state are another recent example emerging among geographers (Mountz, 2007, 2010; Houston, 2011) and other scholars (Nelson, 1999; Ferguson and Gupta, 2002; Sharma and Gupta, 2006; Iskander, 2010; Rodriguez, 2010). Hansen and Stepputat (2001) advocated localized, ethnographic approaches to the state centered in the field and relying heavily on participant-observation (e.g. Herbert, 1997). As Timothy Mitchell (2002) suggested, analysis of disciplinary power must occur at ‘the level of detail’, or the scale of the everyday. Related developments are anthropology of bureaucracy (Heyman, 1995; Gupta, 1995), ethnographies of modernity (Englund and Leach, 2000), globalization (Burawoy et al., 2000), neoliberalism (Colloredo-Mansfeld, 2002), and events (Brosius and Campbell, 2010).
Institutional ethnography can address the production of institutions and subjectivities in particular places and moments that become imbued with meaning. Institutions provide an important and necessary entry point into boundary-making, categorization, and subjectivity-making (Anderson, 1991, Ashutosh, 2010). To explore these possibilities more deeply, we examine recent work in geography characterized by authors as institutional ethnography. We also note, though, the significant omission of engagement with work by Smith and other IE scholars among some geographers doing institutional ethnography.
III How geographers have conducted institutional ethnography
This review provides a foundation for more cross-referencing and fuller engagement, offering potential to develop and expand ethnographic approaches to institutions within the discipline. In the first part of this section we explore studies by geographers who explicitly identify their own work as institutional ethnography, beginning with the earliest such references that we could find. Our findings are based on key term searches for ‘institutional ethnography’ and ‘geography’ in Google scholar, WorldCat dissertations database, and for ‘institutional ethnography’ in the following journals: Annals of the Association of American Geographers, Progress in Human Geography, Transactions of the Institute of British Geographers, The Canadian Geographer, Area, and The Professional Geographer. We include scholarship by anyone who framed writing as institutional ethnography and was either housed in a geography department, publishing in a geography journal, or trained as a geographer. Additionally, we presented the work at a national conference, workshopped the paper with graduate students in a large geography department in North America, and asked 15 additional geographers working in related fields to provide feedback on this paper and apprise us of any omissions.
In this section, we track uses of the term institutional ethnography by geographers (listed in Table 1), and then explain and illustrate a typology of ethnographic approaches to institutions more broadly (located in Table 2). Table 1 shows geographical scholarship where authors explicitly frame research as institutional ethnography. We note the specific methods (interviews, participant-observation, and textual analysis) used in each study. In every case, interviews proved central, but only half used all three methods. As a study of daily life, we suggest that ethnography must involve more than interviews, or what participants say, as Herbert (2000) notes. We ask what makes these studies ethnographic if, in some cases, everyday life in the institution is overlooked?
Scholarship characterized by geographers as institutional ethnography.
Typology of approaches to institutional ethnography within geography. 7
The earliest use of the term institutional ethnography that we could find was by feminist geographer Isabel Dyck (1988, 1997), followed by Jennifer Hyndman (1996, 2000). Both applied Smith’s approach to governmental and supra-governmental organizations. Both scholars began from experience, in Hyndman’s case prior employment with the agency studied. They locate the institution in daily life by following the quotidian experiences of people and policies. Dyck’s (1988, 1997) work in health geography tracked health-related institutions through people’s daily encounters, and she also studied the experiences of immigrant women. Hyndman (1996, 2000) researched the practices and effects of the United Nations High Commissioner for Refugees’ management of refugee camps in Kenya. She merges political economy with feminist analysis to understand how donor states and governments hosting refugees work through the organization across sites and scales to ultimately manage bodies, subjectivities, and resources on the ground.
The next usage of the term institutional ethnography occurs among geographers doing work related to environmental and development agencies (Perreault 2003a, 2003b; King, 2009; Larner and Laurie, 2010; Wolford, 2010a). Beginning less often with personal experience, this scholarship attends closely to the movement of ideas, discourses and policies about development, beginning within and emanating outward from institutions and often focused on the role of elites within institutions. King (2009) suggests that institutional ethnography, development ethnography, and network ethnography are interchangeable terms, linked by their ‘concern with penetrating organizations and social networks to understand how particular discourses and policies are created’ (King, 2009: 409). Scholars working in the field of environment and development (cf. Watts, 2001, 2002; Perreault, 2003a, 2003b; Wainwright, 2008; King, 2009) have asked how ideas about development become institutionalized, which gain currency in ‘battlefields over knowledge’ such as the World Bank (Bebbington et al., 2004), where and how conflict transpires and consent is produced, sometimes exploring the internal dynamics of large development institutions (Hart, 2004). In these studies, the researcher usually conducts interviews with or analyzes discussions of those on the inside (Perreault, 2003a, 2003b; Bebbington et al., 2004; Goldman, 2004, 2005; King, 2009; Wolford, 2010a).
Whereas Table 1 includes only those who frame their own work as IE, Table 2 broadens the analysis to include recent ethnographic approaches to institutions undertaken by geographers, whether explicitly labeled ‘institutional ethnography’ or not. We designed a typology to categorize these studies by methodological approach in order to discern different ethnographic approaches to institutions, as well as some associated insights and shortcomings. Our typology notes distinctions in methods used (participant-observation, interviews, textual and discourse analysis), and the spatial dimensions of each approach.
1. A typology of approaches to institutions
We delineate five ethnographic approaches to institutions in our typology. The first, ‘following’, refers to the tendency of researchers to follow institutional actors in their daily work. This following relies heavily on participant-observation to document power relations at a microscale, such as movements into and out of the material structure of the institution, and along territories constructed by institutional actors, such as neighborhoods, refugee camps, and border crossings. As noted earlier, Herbert’s (1997) ethnography with the LAPD involved riding with police officers to understand how constructions of space and renderings of the boundaries of neighborhoods enacted violence on residents. Like Herbert, Nevins (2002) did ride-alongs with US border patrol along the Mexico-US border to understand how readings of the landscape and daily transgressions therein enabled authorities to enact racialized violence that built on historical dispossession in the region. Although Herbert, Nevins, and others (e.g. Hiemstra, 2011; Bhungalia, 2013) do not label their work institutional ethnography, we find these to be contemporary and ethnographically rich accounts of institutions located along borders and carrying out bordering processes beyond office walls. Other geographers have conducted research that we find to be important to the development of ethnography in the discipline, if not labeled as institutional ethnography, including Anderson’s (1991) and Houston’s (2011) work on city government and Schuurman’s (2008) innovative database ethnographies.
The second approach, ‘time on the inside’, overlaps with the first, but has the researcher placing more emphasis on dwelling in the offices of the institution, and particularly in the bureaucracy. Whereas Herbert (1997) and Nevins (2002) spent more time in the field following actors in their work outside of the office (with occasional visits to the field beyond office), for example, Mountz (2010) and VandeBerg (2013) spent more time studying daily work within bureaucratic offices of federal and UN agencies (with occasional visits to beyond). Importantly, both of these first two categories involved participant-observation, which lent insights into the daily life of institutional spaces, whether in the office or the field. While time on the inside reveals much about the operation of power within the institution, less time is generally spent beyond the institution looking at institutional effects. This may relate to the topic itself, such as the difficulty of pursuing people involved in piracy (VandeBerg, 2013) or human smuggling at sea (Mountz, 2010).
As an illustrative example, we explore Gilmer’s (2014) work in greater depth. 3 Gilmer worked for the UN agency created to combat piracy in and around Somalia. Like previous scholarship on agencies that regulate mobility (Herbert, 1997; Hyndman, 2000), Gilmer makes extensive use of participant-observation. Hired by the agency to design and implement a public campaign to assist in this fight, Gilmer builds on her insider perspective to undertake a critical ethnography of the agency and its construction of pirates and piracy. She finds that many of the initiatives fail to thwart piracy, instead lining the coffers of development workers who themselves become the lucrative subjects of securitization as they compete for funding and become what she calls ‘piratized’ in the process. By ‘following’ and ‘spending time on the inside’, Gilmer’s analysis addresses distinct social locations of those who work across the hierarchy of the organization, with English-speaking employees from countries of the Global North performing the higher skilled and higher paid jobs, for example. She is also able to observe the minutiae of daily work, the power struggles over decisions made within the office and the corruption encountered on her trips into Somalia and island prisons where people arrested for piracy are held and interviewed.
One of the challenges of this approach involves the intimate entanglements that emerge through friendship and working ethnographic relationships in which researchers become embroiled, and the inevitable betrayal that ensues (Stacey, 1988; Visweswaran, 1994; Mountz, 2007). Herbert (2000) writes about the emotional and physical effects of participant-observation that may involve witnessing violence or discriminatory behavior or omitting information from the analysis. Such violence may be another reason why those who spend more time on the inside are sometimes less able to gain access to those who are biopolitically managed by institutions.
The third category, ‘getting at the inside: interviews with organizational actors’, is perhaps the most common approach among geographers identifying their work as ‘institutional ethnography’ over the last ten years. This approach relies less on participant-observation, and more heavily on analysis of interviews and documents to access the institutional structure, sometimes characterized as a ‘black box’ (Bebbington et al., 2004: 37). In these studies, the researcher is usually located outside of the institution and accesses information about the institution by conducting interviews and corresponding with institutional actors or accessing discourses of those on the inside through reports and publications (Perreault, 2003a, 2003b; Goldman, 2004, 2005; King, 2009; Wolford, 2010a). 4 Analysis may also examine how institutional discourses travel into the world beyond the institution (Bebbington et al., 2004). These studies have tended to adopt more bounded notions of the institution with more clearly defined populations, policies, and cultures (e.g. Goldman, 2004; 2005; Lewis and Mosse, 2006; King, 2009) and have drawn on discourse and debates among elites (e.g. Bebbington et al., 2004) to understand how institutions and individual actors within them shape development (e.g. Goldman 2004, 2005).
Perreault (2003a, 2003b), for example, draws on interviews with key informants to examine how ethnicity, territory, and identity intersect in an indigenous Kichwa organization in Ecuador, producing a discourse through which indigenous peoples participate in development processes. His analysis focuses on the ways in which indigenous organizations ‘resist, refract, and at times reproduce dominant narratives of development, modernization, and citizenship’ (Perreault, 2003a: 586). Perreault (2003a: 602) aims to uncover the organization’s political strategies to contest and negotiate processes of development and ‘social transformation’. King (2009) analyzes the centrality of a neoliberal commercialization discourse in a South African conservation organization, while Goldman (2004, 2005) focuses on the production of the World Bank’s ‘authoritative green knowledge’. These studies are linked by their ‘liberatory potential’ in negotiation of development processes (Perreault, 2003a: 603), but limited in their articulation of the daily operation of institutions and the social locations or embodiment of those who do (and do not) inhabit them. While acknowledging the partiality of their findings, these studies produce disembodied institutions, rather than what the anthropologist Clifford Geertz (1973) famously called ‘thick description’ of daily life of and in the institution.
While not working under the rubric of institutional ethnography, Bebbington and co-authors (Bebbington et al., 2004, 2006) study discursive debates over the concept of social capital transpiring in the World Bank. They explore the relationship between the institution’s inside and out, arguing ‘that “getting inside” development institutions is important for understanding how and why certain discourses emanate from them, for interpreting the significance of these discourses, and for understanding the indeterminate relationship between discourse and practice’ (Bebbington et al., 2004: 34). Through discourse analysis of publications and personal communications, the authors find: different arenas in which the contests are waged: internally among its staff (the battlefield we focus on here); externally with non-Bank actors and those encountered in the course of implementing projects; and – more intriguingly – cross-border battles in which different sub-communities within the Bank are linked to different communities outside the Bank, and where the battles engage larger communities whose memberships transcend institutional boundaries. (Bebbington et al., 2004: 38)
These studies did not involve participant-observation in the form of following actors or spending time on the inside, but tended to involve the conduct of interviews with employees both within and beyond institutional spaces. As Kuus (2013) argues, this limitation may reflect the challenge of actually conducting ethnographic study of policy issues. 5 Yet data drawn from interviews without the insights of participant-observation limit the claims a study can make to understand and interpret daily life in Geertz’s (1973) terms: detailed-oriented ‘thick description’.
In contrast, anthropologist Diane Nelson (1999) makes related arguments in her ‘ethnography of the state’, studying representations of indigenous and state identities in Guatemala with a distinct approach. She posits the state is imagined and lived through multiple ‘bodies politic’ of Mayan women, rooted in ‘manipulation and violence’ tied to indigenous rights and nation-building. She operates not only within a distinct discipline, but with a distinct set of epistemological frameworks from development ethnographies: feminist, postmodern, and poststructural approaches. The result is a more dispersed understanding of institutions and their embeddedness in daily life. Sawyer (2001, 2004), also an anthropologist, examines the daily social and environmental consequences of increased demand for oil via indigenous mobilizations in Ecuador. She employs participant-observation and interviews to explore social relationships between an indigenous organization, multinational oil companies, and the state that produce ‘indigenous opposition to economic globalization in its neoliberal guise’ (Sawyer, 2004: 7). She focuses specifically on the power inequalities that emerge in this ‘terrain of struggle’ over ‘identities, territories, and relations’ as indigenous peoples sought recognition and rights in a plurinational Ecuadorian state (2004: 222). In these analyses, people themselves embody, inhabit, and shape institutional structures at the same time that they are shaped by them.
While some geographers engage in discourse and ideas on the inside, others use ethnography to understand how institutions influence life on the outside. The fourth and fifth categories, ‘influencing life on the outside’ and ‘event ethnography’, involve even more distance from the physical spaces of the institution, deriving material from purviews outside of institutional spaces, either through interviews, archives, or observations that provided more limited glimpses and allow for less triangulation than sustained time on the inside. Although they do not characterize their work as institutional ethnography, for example, both Hiemstra (2011) and Bhungalia (2013) conducted ethnographic research on US agencies: detention and deportation systems run by the Department of Homeland Security and the work of US AID in Palestine, respectively. Their ethnographies were not in the headquarters in countries where these institutions are based, but rather in the remote locations where the effects of their institutional policies and practices were felt powerfully. Hiemstra worked with deportees and their families in Ecuador to learn about detention in and deportation from the United States. Bhungalia worked in Palestine with organizational actors securing funding from US AID. The result is a decentralized landscape of the institution – what some would call a governmentality approach (Foucault, 1991) or ‘state effects’ (Mitchell, 2002) – with effects that extend into daily life well beyond the borders of the institution or that reverberate transnationally (Hiemstra, 2011; Bhungalia, 2013). Research in the margins holds potential to shift understandings of power in and of hegemonic institutions that advance imperialism and confinement of populations.
While these more spatially removed approaches may not come as close to the daily rhythms of life inside the institution, including institutional oppression, they also result, we suggest, in a more geographically dispersed topography of the institutions being studied and a rich rendering of the effects of institutions in the daily lives of those whom they affect. Their analyses show production of subjectivities of aid workers in the West Bank and detainees and deportees in Cuenca, Ecuador, shaped by US-based national agencies, uncovering mundane acts of resistance in daily life. Both authors show the racialized, gendered, and classed dimensions of institutional effects beyond institutional walls and national boundaries. An obvious limitation of working ‘from the outside’ is the lack of direct observation of the daily operations of the institution, with reliance instead on narratives about institutional practices or policies from aid workers and deportees.
The final category, ‘event ethnographies’, encompasses a recent development undertaken by 17 scholars in environmental studies working collaboratively to do short, intensive periods of participant-observation to study events such as conferences and workshops of powerful institutional meetings (Brosius and Campbell, 2010; Corson and McDonald, 2012; Suarez and Corson, 2013). This approach draws on ethnographic traditions in development rather than the field of institutional ethnography. Event ethnographies differ from other approaches for their temporality: rather than pursue the tradition of long periods of time devoted to understanding the daily rhythms and relations in place, the event ethnography is premised on short-term, yet important (if fleeting), moments of coming together.
Similar to Wolford’s (2010a) characterization of ethnography of a site, yet differing in their use of participant-observation, Corson and MacDonald (2012) reconceptualize the field through the event by drawing on research at the UN Convention on Biological Diversity. We find this approach innovative for the characterization of the field as a place that brings together a number of institutions and actors in one time and place. The opportunity to see people interact in one setting is important, and yet at present under-developed in the findings reported – the potential insights of an ethnographic approach are not yet fulfilled in this incipient work. The innovative framing of a new field – the event as place – is at odds with the lack of description of that place and those within it. Missing from these analyses is Geertz’s (1973) ‘thick description’. In the decision to conceptualize the event as a place, the description of its characteristics, landscape, feel, or context has not yet been illustrated. The authors draw primarily on public statements made in side events, yet those who spoke them are not described, located, identified or embodied. The messy, everyday unfolding of institutional operations that might be constructed from fieldnotes would enhance the development of event ethnography. Participants in UN conventions generally agree that much of the ‘real’ conversation and work happens before the event, with the short event providing a public forum for the statement of positions and agendas. What kind of negotiations occurred before the more public performances quoted? Does the event ethnography include participant-observation in the more mundane, daily lead-up to the annual coming together? As a result of the decision to focus temporally and spatially on the event itself, this approach provides new insights while missing others.
‘Fast policy transfer’ (Peck and Theodore, 2010) and ‘collaborative event ethnography’ are part of the upsurge in ethnographic research, but not necessarily engaged with institutional ethnography. Peck and Theodore (2010) make brief reference to institutional ethnography in their writing (citing Larner and Laurie, 2010, who in turn cite Goldman, 2005). Yet this scholarship remains largely disconnected from earlier work on institutional ethnography by geographers and others. Still, there are important continuities to observe. As Delaney notes, contemporary trends in economic geography are influenced by the Deleuzian assemblage which ‘offers a frame in which to examine how institutions and policies “territorialize”’ and by actor network theory with its ‘focus on the act of translation in bringing policy and practice networks into coherent structures’ (Delaney, 2014: 19). These approaches render more dispersed ethnographic mappings of institutions that could draw from and advance practices of institutional ethnography. They also share thematic interests, such as advancing understandings of neoliberalism.
Our typology uncovers distinctions in the methodological approach and epistemology and ontology of institutions, with spatial differences tied to ethnographic methods employed by researchers. We found that studies that either omit participant-observation as a method or do not draw on these data in their analyses fail to observe the influences and embeddedness of the institution in everyday life. Participant-observation is attentive to emotion, subjectivity, power struggles, resistance, and proximity of the institution. IE is about accessing the everyday, as is ethnography generally. We still have much to learn from anthropologists such as Nelson, who is indiscriminate in engaging ephemera collected in daily life. Chance encounters, t-shirt slogans, cartoons, conversations – all become locations where the institution, its effects and productive capacity are readily evident. The result is Clifford Geertz’s (1973) ‘thick description’ of daily life. This more textured view of an institution differs from a more fixed notion of the institution whose politics, projects, subjects, and discourses are accessed primarily through interviews – in other words, the ways that institutional subjects or employees narrate the institution. The latter cannot readily account for the sociospatial differences in power operating within and across the daily productions of an institution. Institutional formation and operation across center and periphery lead to differential outcomes of discourse and practice, with unequal impacts and effects. More sustained, critical engagement with heterogeneous and interdisciplinary approaches to institutional ethnography will improve understandings of institutions within and beyond the discipline of geography.
2. Enhancing scholarship on institutions
Here, we explore institutional ethnography from geographical perspectives. With limited space, we offer two areas drawn from our own fields of research to demonstrate how critical engagement across approaches within and beyond the discipline of geography holds potential to enhance scholarship on institutions. The two kinds of institutions we discuss are corporate social responsibility (CSR) (Billo, 2012, 2015) and institutions where people are detained (specifically, detention centers and prisons) (Moran et al., 2012; Mountz et al., 2013; Loyd et al., 2012; Mitchelson, 2013; Mountz et al., 2013). Both are sets of institutions that geographers have shown more interest in of late (Moran et al., 2012).
CSR programs are the business response to social and environmental criticism of corporate operations (Sadler and Lloyd, 2009; The Economist, 2008; Watts, 2005). In Ecuador, these programs emerged out of state, corporate, and indigenous relationships, and are designed to ensure ongoing resource extraction. Mandated by the Ecuadorian state, CSR programs incorporate indigenous peoples in local development projects, such as drinking water systems and community infrastructure – things the state might normally provide. Billo’s (2012) research uncovers the disciplinary techniques of CSR programs through implementation of corporate programs for local development (cf. Foucault, 1991). An IE engages the everyday processes of implementation of CSR programs that can at once co-opt and legitimate discourses and practices of indigenous rights to resources through the material projects and programs of CSR. The subsequent presence of CSR programs in everyday interactions produces ambivalent indigenous subjects; indigenous entanglement with dominant power structures complicates notions of resistance and contributes to new forms of resource governance and development processes in Ecuador (Billo, 2012, 2015). In turn, rather than focus solely on narrowly defined corporate projects labeled CSR, IE seeks to understand the relationships that form within CSR programs and expand well beyond the local indigenous communities, raising questions about indigenous citizenship and about how the corporate and philanthropic come together to produce subjectivities in extractive industries (Billo, 2012).
Scholars have argued that prisons are not what Goffman (1961) labeled ‘the total institution’: enclosed facilities that contain everything and everyone therein (Moran et al., 2013: 110–12). Rather, they can be conceptualized as more fluid, ‘transcarceral’ entities (Moran et al., 2013). Although prisons immobilize and contain those imprisoned, they have surprisingly permeable boundaries. Many material things move across prison walls: food, supplies, medical services, information, capital, paperwork, statistics, workers, visitors, and detainees themselves. This movement proves helpful in studying prisons and detention facilities where people are held; these are difficult sites for researchers to access and where researcher access can put vulnerable populations at further risk. Penal institutions are therefore highly suited to study through institutional ethnography in Smith’s tradition. IE opens the institution to research in ways that do not necessarily require physical access. Interviews and participant-observation may fruitfully be conducted with workers, former detainees, and visitors such as family, friends, and lawyers. This opens a broader landscape through which to understand the prison. Focusing on what happens not only within but across boundaries also fosters research on sensitive topics and vulnerable populations in relatively ‘safe’ ways for those institutionalized. Conversely, research premised on entry into prisons would be riskier for participants and more likely prone to failure should access not be granted.
Feminist approaches to studying imprisonment have fruitfully pursued more dispersed understandings of the social relations of imprisonment with exciting outcomes. Mary Bosworth (2005), for example, conducted research on imprisonment and co-authored findings with four prisoners who reflected on the experience. The result is ‘a situated example of, as well as a call for, dialogue about research across prison walls’ (2005: 250). The authors aim to destabilize power relations and boundaries between researcher and researched, ‘make clear the fundamentally affective nature of qualitative research’ and show how emotions motivate participants in research and, in so doing, that prisoners – like researchers – are individuals with desires and emotions (2005: 251). Bosworth and co-authors bring to our discussion the alternative media and kinds of texts that can be part of IE, such as the role of letters, their potential to engage a broad emotional register through their play with time and space beyond and within the institution (as in the daily nature of mail call woven into the slow movement of mail). In so doing, they disrupt boundaries between inside and outside of prison and research project, and what counts as personal and professional interaction. Similarly, Moran and co-authors (Moran et al., 2013) conceptualize Russian prisons as ‘mobile, embodied and transformative’ transcarceral spaces that permeate prison walls. By recasting research as cooperation and intimacy, such approaches destabilize ownership of research agendas and outcomes and disrupt masculinist notions of penetrating institutions, while simultaneously reconfiguring geographical understandings of institutions. These approaches have tended to be feminist, premised on the project of analyzing women’s experiences (e.g. Dyck, 1997; Bosworth, 2005; Moran et al., 2013).
IV Conclusions
As we have shown in our typology and accompanying discussion, geographers have practiced institutional ethnography in a variety of ways. Yet the collective potential for geographers to deepen development of institutional ethnography and contribute to institutional analyses remains unrealized. Geographers stand to contribute more sophisticated socio-spatial understandings of institutions. Content analysis of references found shows that geographers often do not reference earlier ethnographic research on institutions in geography or the vibrant, interdisciplinary field of institutional ethnography. Our typology of institutions will aid in more direct engagement and discussion of the benefits and limitations of varied approaches, resulting in further development of the field. A deepening and diversity of approaches to institutions can advance understanding of micro- and macro-processes, as well as meta-discourses such as neoliberalism.
We have called for institutional ethnography and engagement with literature within and beyond geography. Indeed, pursuing the same method in distinct contexts illuminates parallels in the operation of states, capitalism, and the punitive and productive natures of institutions. In this context we note subject formation that emerges through institutions as mediators between state and subject, at times ‘erasing’ difference in the name of securing wealth and membership in the nation-state (cf. Nelson, 2001; Lindner, 2012). Institutional ethnography offers the possibility to study up to understand the differential effects of institutions within and beyond institutional spaces and associated productions of subjectivities and material inequalities. IE can function as an approach to look within, through, and beyond the architecture, policies, texts, and problematics of the institution to understand how, why, and for whom. Practiced in a variety of ways, IE holds potential to enrich geographical research, both in kinds of institutions studied and territorial dynamics rooted in institutional effects, structures, and identities.
Our research demonstrates that institutions are not uniform across times and places, although they often construct discourses and practices that produce a coherent, monolithic appearance. Through discussion of research on prison facilities and CSR programs, we show that IE can be attentive to where the institution is produced and in turn produces daily lives. Further engagement with subjectivity and intersectionality will deepen understandings of institutional power and help in the breaking down of barriers between those confined and those on the outside. By attending to gender, race, ethnicity, sexuality, and class, the approach can note differences within and among institutional actors and those affected by institutions. We argue that this approach is attentive to inequalities, relationships of power, and researchers’ social locations in conceptualizing institutions. A diversity of approaches will also potentially result in fewer masculinist constructions of knowledge predicated on accessing elites.
Our analysis opens a new set of questions yet unanswered. We are interested to learn more, for example, about how recent methodological developments in geography intersect with institutional ethnography, including fast policy transfer, event ethnography, actor network theory, and science and technology studies. While we have mentioned these approaches in this text, more work remains to be done as to their productive intersections and divergence from institutional ethnography as it has been practiced by geographers and other social scientists.
We have argued for institutional ethnography. Rather than suggest that there is one way of doing institutional ethnography, our goal has been to review the range of approaches pursued by geographers and other scholars and the potential insights and shortcomings of various approaches, with the ultimate objective of broader engagement. Institutions are fundamentally powerful and spatial in nature. As such, their ethnographic mapping across sites and scales is essential to advance understandings of political, economic, and social relations.
Footnotes
Acknowledgements
We are grateful to many insightful readers of earlier drafts of this article who shared feedback and their own scholarship: Ishan Ashutosh, Mat Coleman, Isabel Dyck, Roberta Hawkins, Nancy Hiemstra, Jennifer Hyndman, Vicky Lawson, Keith Lindner, Jenna Loyd, Jacob Miller, Dominique Moran, Lawrence Santiago, Nadine Schuurman, Margaret Walton-Roberts, Richard Wright, and graduate students at the School of Geography and Development at the University of Arizona. This material is based upon work supported by the National Science Foundation under Award #0847133 (Principal Investigator: Alison Mountz). The research in Ecuador was supported by a National Science Foundation Doctoral Dissertation Research Improvement Grant (#0825763) and by an Inter-American Foundation Grassroots Development PhD Fellowship.