A matter of scale: multi-scale ethnographic research on education in the United States
Margaret Eisenhart
ABSTRACT
In recent years, cultural anthropologists conducting educational ethnographies in the US have pursued some new methodological approaches. These new approaches can be attributed to advances in cultural theory, evolving norms of research practice, and the affordances of new technologies. In this article, I review three such approaches under the rubric ‘multi-scale ethnography’. Each approach responds to the desire to understand cultural forms that extend beyond single sites and the capabilities of single researchers; each pursues cultural forms that ‘travel’ across spaces, levels, and times. The three approaches are multi-sited ethnography, meta-ethnography, and comparative case study. I view these approaches not as replacements for older forms of ethnography but as complements that extend the contextualisation and generalisability of traditional ethnography.
Introduction
Research on education in the US has taken new methodological directions in recent years (Moss and Haertel 2016). These new directions can be attributed to advances in theory, evolving norms of research practice, and the affordances of new technologies. Among cultural anthropologists studying education in the US, new directions in cultural theory have been a major source of methodological innovation in the practice of ethnography. Theories of culture have been critiqued and revised numerous times. Some scholars would abandon culture altogether as an irredeemable relic of anthropology’s colonialist origins (Abu-Lughod 1991). Others have made it virtually synonymous with political or social ‘identity’ (Anzaldúa 1987; Kondo 1990). Still others have recast it in terms of symbolic logics or representational technologies, their associated social practices, material resources, historical legacies, power dynamics, and local and global connections (Eisenhart 2001; Eisenhart forthcoming; Gershon 2011). Among those in the third group, such as myself, none has abandoned traditional methods of ethnography. But many have encouraged new approaches that expand its scope. In this article, I review three of these approaches – which I collectively label ‘multi-scale ethnography’. Each approach responds in one way or another to the desire to identify and understand cultural forms that travel across spaces, levels, and times. These approaches are not the only possible ones. Other scales of human life, for example, expressive modalities (Dicks et al. 2011), temporal contiguity (Maxwell 2012), may also serve as foci of ethnographic research that extends beyond a single site. I view these approaches not as replacements for or improved versions of traditional ethnography but as complements that extend the contextualisation and generalisability of ethnography. These complements are especially valuable in strengthening ethnographic claims for audiences of policy-makers and other influential stakeholders. In what follows, I first review the tradition of traditional ethnography as it has been practiced by US cultural anthropologists who study education. Then I turn to the three newer approaches.
Traditional ethnography
In US cultural anthropology, ‘culture’ has historically been theorised as a set of beliefs and practices characteristic of a bounded group’s way-of-life. In consequence, traditional ethnography consists of a set of flexible methods to learn about and make sense of how culture is constituted or produced in a group – in its families, neighbourhoods, villages, informal activities, and institutions such as schools. The foci of traditional educational ethnography are thus the practices and processes of socialisation and enculturation in (usually) singular contexts. Using methods of participant observation, in-depth interviewing, document collection, reflexive journaling, and iterative textual analysis, American ethnographers of education have attempted to understand how and why educational contexts are constituted as they are; what actions these contexts organise; what meanings are associated with these contexts; how individuals in these contexts make sense of what they are doing; how they are influenced by and themselves influence the structural, material and discursive artefacts of their lives; and how these processes accumulate to produce and reproduce knowledges, practices, identities, and artefacts.
Traditionally this work has been done by individual researchers through extended face-to-face fieldwork (months or years) with study participants in (or mostly in) one locale or ‘site’. Partly for this reason, traditional ethnography has earned a deserved reputation for detailed, in-depth, participant-oriented understandings of singular groups or institutions that other research methods miss or ignore. (See Eisenhart 2001 and Eisenhart forthcoming, for a more extensive discussion of changing conceptions of culture and its implications for ethnography in the US.) However, since the 1980s, many anthropologists have reframed ‘culture’ as interpretive logics or representational technologies that travel across time, space, and level, rather than as the characteristics or lifeways of bounded groups. In consequence, ethnographic research methods designed to intensively study singular groups are not necessarily sufficient or always appropriate. Interpretive logics, such as the meaning of ‘real school’ (Metz 1990), the ‘culture of romance’ (Holland and Eisenhart 1990), and the ‘educated person’ (Levinson, Foley, and Holland 1996), organise meaning, behaviour, communication, desires, and emotions. But they are not static, impermeable, geographically anchored, or necessarily coherent. They may originate in one place and time and move across group boundaries (Bartlett 2009). They may be produced during a specific moment-in-time and re-appear later in similar or different forms (Metz 2003). They may connect far-flung people linked only by mass media, textbooks, computers, or global economic networks (Nespor 1994). They may overlap and contradict each other (Luykx 1996).
Representational technologies – mobile productions of time and space, for example, textbooks, software programs, specialised equipment, discourses (Nespor 1994) – share some similarities with interpretive logics. They move (or are moved) across space–time dimensions, are taken up differentially, and establish linkages among disparate groups. This movement and variability inspires ethnographers to want to follow these cultural forms as they develop, move, are taken up, and change over time.
Some anthropologists of education have turned their attention to the entanglements of interpretive logics or representational technologies with forces of globalisation and how these entanglements affect life, education, and schooling in various parts of the world. This work focuses on how circulating flows of knowledge and resources – including those under government, corporate, and media control and those that are communicated in ‘real’ time and virtually – affect children and communities in local contexts. In this formulation, these entanglements often disrupt and unsettle local practices. Material and symbolic resources and processes originating both near and far are then (re) mediated in schools and other local contexts of learning (Bartlett 2009; Hall 2002; Levinson 2001; Lukose 2009; Stambach 2000). These processes produce uncertain results, sometimes including novel and liberatory possibilities of special interest to ethnographers and change agents.
For the most part, the movements of ideas, resources, and artefacts across disparate sites and across different scales of power and influence have not been a focus of traditional ethnography (but see Hannerz [2009] for a discussion of their elements in traditional ethnography). And although ethnographic fieldwork often takes a long time, traditional ethnographic accounts tend to describe a way-of-life at one moment-in-time, for example, the culture of Roadville and Trackton in the early 1970s (Heath 1983), the culture of the ‘lads’ in an English secondary school in the 1970s (Willis 1977), the culture of romance at two universities in the American South in 1979–1983 (Holland and Eisenhart 1990); they rarely take up change across time in any detail (see e.g. Holland and Lave 2001). As Lemke says about traditional network analyses, traditional ethnography does not tend to ‘ask how and why events widely separated in time and space seem to re-enact the same patterns;…[how] emergent phenomena unique to every level of organization [develop] in a complex dynamical system’ (2000, 274), or how signs and meanings are transformed across extended periods of time.
Multi-scale ethnography
‘Multi-scale ethnography’ consists of approaches to following ‘culture’ in the form of ideas, discourses, practices, and artefacts that move or are transported to multiple places, across various social or institutional levels (e.g. local, global; rural, urban: privileged, non-privileged; formal, informal), and through time. Here I focus on three approaches – multi-sited ethnography, meta-ethnography, and comparative (or vertical) case study.1
Multi-sited ethnography
George Marcus (1995) originally defined multi-sited ethnography as
designed around chains, paths, threads, conjunctions, or juxtapositions of locations in which the ethnographer establishes some form of literal, physical presence, with an explicit, posited logic of association or connection among sites that in fact defines the argument of the ethnography. (1995, 105; emphasis mine)
Multiple locations or ‘sites’ are purposively selected to study how connections across sites are made in and through the lives of participants. In a multi-sited design, cultural forms evidenced, produced or circulating in one site are emically explored in other sites, witht he intent of clarifying and understanding connections among them. ‘Strategies of quite literally following connections, associations, or putative relationships are … at the heart of designing multisited ethnographies’ (Marcus 1995, 97). The focus is on cultural connections across sites that are ‘substantively continuous but spatially non-contiguous’ (Falzon 2009, 2), that is, multi-sited ethnography implies some form of spatial decenteredness.
Marcus (1995) initially suggested several strategies for doing this kind of work. Ethnographers might follow the interpretive logic of one group of people as they physically (and socially) move from one site to another. Examples include Luykx’s study (1996) of the discourses of Bolivian teacher candidates as they moved back and forth between their rural homes and their school site at a teacher education college; Levinson’s study (1996) of messages about ‘the educated person’ that circulated in various spaces where groups of teenage Mexican students congregated; Rahm’s study (2010) of the understandings of ‘science’ as taken up by a group of Canadian students in school, at home, and in after-school activities; and Livingstone and Sefton-Green’s study (2016) of English secondary students’ ‘presentations of self’ at school, at home, and among peers.
Multi-sited ethnographers might also examine one form of cultural production, for example, the means of ‘becoming somebody’ (Wexler 1992), as it is formulated and put into practice in distinct but interrelated groups, for example, policy-makers, school administrators, teachers, parents, and students. They might follow the discourse about an innovation or a reform as it moves or is transported away from its source to new locations. Consider, for example, the discourse of science education reform as formulated by policy-makers, by the teachers who try and implement it, and by the students and parents who experience it. Multi-sited studies of cultural productions – such as policymakers designing an education reform, teachers involved in implementing it, young people experiencing it, and parents responding to its apparent effects on their children – offer a means of understanding how educational activities, values, and outcomes in one group in one place-time are constituted, constrained, and changed by connections or processes across groups, places, and times. In summing up this aspect of multi-sited ethnography, Marcus writes,
The past habit of Malinowskian ethnography has been to take subjects as you find them in natural units of difference … [T]he habit or impulse of multi-sited research is to see subjects as differently constituted, as not products of essential units of difference only, but to see them in development – displaced, recombined, hybrid…, alternatively imagined. Such research pushes beyond the situated subject of [traditional] ethnography toward the system of relations that defines them. (2009, 184)
In later work, Marcus has clarified that his main interest lies with the detailed situatedness of connections in the flow of events in multiple spaces, rather than with multiple locations in space per se (2009, 193). He has further emphasised that he considers multi-sited ethnography to be truly ethnographic only when it apprehends a system of connections from within participants’ actions and expressions. In this way, the key ethnographic act – consistent with traditional ethnography – is the commitment to develop an understanding (of connections) from an embedded perspective. As Marcus explains: ‘What is distinctive about anthropology and precious to preserve in the Malinowski ethos of ethnography is the pretense and aim to be able to work through subject positions, perspectives, and meanings in order to establish one’s own knowledge’ (2009, 188).
But this commitment entails a form of fieldwork that appears, at least at first glance, to be quite different from conventional ethnography. In multi-sited fieldwork, ethnographers do not depend on (or some might say ‘collaborate with’) participants for reconstructions of traditions; rather, they develop ‘paraethnographies’ that attempt to make sense of unfolding events – both locally and in other places – along with participants.
[A]t the core of ethnography in multi-sited research is the strategic engagement with para-ethnographic perspectives in research, epistemologically equivalent to one’s own, and working through them literally into other sites of fieldwork. (Marcus 2009, 188)
In other words, Marcus considers the joint production of insights by researchers and participants across sites (i.e. ‘in collaboration’) to be an ethnographic necessity. This is because ‘culture operates within a continuously unfolding contemporary … where everyone directly or indirectly is implicated in and constituted by complex technical systems of knowledge, power, health, politics, media, economy, andsoon’ (2009, 184, endnote 2). He proposes the concept of ‘paraethnography’ as the means of this collaboration. He describes paraethnography as an unfolding, largely uncharted process managed and negotiated by researchers and participants as they together attempt to discover and understand contemporary culture in terms of connections among disparate people’s lives and artefacts.
This approach to ethnography further entai...