Multi-sited ethnography in STS: Capturing global complexities and the role of the researcher.

Published: 2019/12/10 Number of words: 3153


In this paper I review and examine how Science and Technology Studies (STS) (Hackett et al., 2008; Jasanoff et al., 2001) draw on and influence the method of ethnography.​ The question I ask is: how can STS ethnographies capture the complexity of global connections? To answer this question I will introduce the main ethnographic approach that became popular in STS studies, that is, George E. Marcus’s concept of ‘multi-sited ethnography’ (Fortun, 2003; Hine, 2007; Marcus, 1995; Tsing, 2011) and examine how various STS scholars used it in their work or reflected on it.

The main goal of multi-sited ethnography was to overcome various problems that emerged when ethnography was used in the examination of science and technology and, more importantly, when trying to give relevant accounts about global phenomena without falling back into social constructivism. This, at the same time, also had several consequences for the position of the STS ethnographer. Therefore, in my literature review I look for answers to the question: what results did the use of multi-sited ethnography have for the STS scholar, especially in terms of reflexivity, modes of intervention and responsibility?

Before I attempt to answer these questions, however, I will discuss some of the specificities of the relationship between STS and ethnography.

STS and Ethnography

STS’s engagement with ethnography has a long history (Akrich, 1992; de Laet and Mol, 2000; Helmreich, 2007; Latour, 1993; Law, 1994; Licoppe, 2010; Oppenheim, 2007). David Hess distinguishes at least two periods (Hess, 2001). According to him, STS initially was more concerned with the very processes by which scientific facts, the notion of objectivity, and science as such. Hess, with others, labelled this type of approach ‘laboratory studies’, because most well-known research was usually conducted in laboratories with a detailed ethnographic account about how the work had been done (Collins, 1975; Latour and Woolgar, 1979; Latour, 1983).

This research, even though highly influential due to the very controlled methods, could not report about issues outside the laboratory. Even more important, some of them explicitly refused to take into account any external phenomena, arguing that doing so would automatically require simplification and the use of wrong assumptions – an attitude going against the laboratory studies’ co-constructive approach (Hess, 2001). However, it quickly became clear that the issues around science and technology do not stop at the walls of the lab and that the lack of use of universal concepts describing global relationships did not mean that these relationships were non-existent and should not be described at all. Consequently, new approaches emerged. These were an attempt to grasp global connections and at the same time, avoid repeating the earlier failures of social science.

Using David Bloor’s criteria of the ‘strong programme’ of the sociology of science (Bloor, 1976), Hess characterised what he called ‘the second generation’ of science and technology using the terms ‘impartiality’ and ‘symmetry’. By this was meant that they avoided making distinctions in advance between true and false claims and that they used the same type of causes to explain the results of both ‘true’ and ‘false’ science. However, according to Hess, this approach only applied to the ‘methodological’ level which he distinguished from the ‘logical’ level of investigation. While second-generation STS was symmetrical and impartial in its cultural interpretation, it was asymmetrical and partial in its final ‘higher level’ analysis.[1] Accordingly, ethnographers start with a culturally relativist (impartial and symmetrical) description but come to their conclusions by means of a morally and epistemologically anti-relativist (partial and asymmetrical) analysis. This process, then, fluctuates between these two positions (Collins, 1975, pp. 236–238).

A consequence of this was that ethnographers stopped being seen as naive students, innocently observing and learning from the investigating culture, and became people holding a special position: one that they could not deny but had to consider during the whole process of research (Coe, 1991). This applied especially to the realm of STS where the scientists were more aware of what was being investigated and, furthermore, were more willing to ‘talk back’ to these findings.

In the following chapter I will briefly describe the main conceptual framework of multi-sited ethnography. After this, I will discuss the questions about the political and ethical consequences of this kind of scholarship based on articles reflecting Marcus’s concept. Finally, I will reach a conclusion and ask some questions for further consideration.

Multi-sited ethnography

George E. Marcus, in his essay ‘Ethnography in/of The World System’ differentiates his approach from the prevalent single-sited anthropology. His approach is ‘self-consciously embedded in a word system, [and] moves out from the single sites and local situations […] to examine the circulation of cultural meanings, objects, and identities in diffuse time-space.’ This approach ‘acknowledges macro-theoretical concepts and narratives but does not rely on them for the contextual architecture framing a set of subjects’ (Marcus, 1995). In a departure from a culturally relativist or, for instance, a Marxist version of anthropology, Marcus’s method takes ‘unexpected trajectories’ by which it destabilises the distinction between previously conceived concepts. Consequently, this mode constructs both the ‘lifeworld’ and ‘system’ of the situated subjects, which means that in relation to the question of global connections, ‘this mode comes circumstantially to be of the world system as well’ (Marcus, 1995).

A similar example is that of Latour’s who, building on Foucault’s notion of the panopticon (Foucault, 1991, 1984) – a metaphor for omnipresent power – proposes two other terms describing ‘globally relevant’ actors. The first is ‘oligopticon’, an actor who has a relatively big influence over a great distance but only in a very small area. The second, ‘panorama’, on the other hand, has a much broader influence, but its particular strength is constrained (Latour, 2005, pp. 175–190). The two concepts, according to Latour, would more adequately describe global connections while also leaving room for local multiplicities.

This way of doing anthropology, although multi-sited, is not holistic and does not depend on concepts of global−local contrast. Rather, it stipulates a total world-system ‘as long as the terms of any particular macro-construct of that system are not allowed to stand for the context of ethnographic work’. This, however, does not mean that with an increase in the number of sites introduced, the descriptive power of the analysis will decrease proportionally. Rather, just as in the case of cultural historiography, ‘principles of selection operate to bound the effective field in line with long-standing disciplinary perceptions about what the object of study should be’ (Marcus, 1995). Furthermore, Marcus stresses the importance of the function of translation in the ethnographic work. According to him this is:

…enhanced since it is no longer practiced in the primary, dualistic “them-us” frame of conventional ethnography but requires considerably more nuancing and shading. […] Indeed, the persuasiveness of the broader field that any such ethnography maps and constructs is in its capacity to make connections through translations and tracings among distinctive discourses from site to site. (Marcus, 1995, pp. 100–101)

To give an example, Heath et al.’s research studying both online and offline sites, show the ways they construct the different places interactively and how they define ‘online’ and ‘offline’ sites interdependently, moving back and forth between them continuously. During their research they concentrated on the ‘nodes’ where the different sites intersect, on the ‘location work’ by which they actively created their context which ‘give meaning to the life worlds’ studied by them and, finally, on the ‘modest interventions’; that is, the engagements that reveal the links between different places (Heath et al., 1999).

With regard to the role of translation, Marcus refers to Donna Haraway (Haraway, 1997) and denies the possibility of innocent identity politics, including the self-positioning of the researcher into the place of the subaltern (Haraway, 1991a, 1991b). This view, nonetheless, decentres ‘the resistance and accommodation framework’ and as a result, the questions of resistance became ‘subordinated to different sort of questions about the shape of systemic processes’ (Marcus, 1995). Marcus writes in summary:

The object of study is ultimately mobile and multiply situated, so any ethnography of such an object will have a comparative dimension that is integral to it, in the form of juxtapositions of phenomena that conventionally have appeared to be (or conceptually have been kept) “worlds apart”. Comparison re-enters the very act of ethnographic specification by a research design of juxtapositions in which the global is collapsed into and made an integral part of parallel, related local situations rather than something monolithic or external to them (Marcus, 1995: 102).

On the operational level, all this means is that the STS scholar follows the different connections through different places and sites and by this movement, constructs an ‘initial, baseline conceptual identity’ as a contingent and malleable topic (examples of issues Marcus proposes are: the people, a thing, a metaphor, a plot or story, a life, a conflict etc.) (Marcus, 1995).

Besides this, of course, Marcus retains the possibility of single-sited ethnography, but specifies certain criteria to prevent falling back into the earlier mistakes of ethnography. This so-called ‘strategically situated’ single-sited ethnography is local only circumstantially, but attempts a broad understanding of the system in ethnographic terms:

Within a single site, the crucial issue concerns the detectable system awareness in the everyday consciousness and actions of subjects’ lives. This is not an abstract theoretical awareness such as a social scientist might seek, but a sensed partially articulated awareness of specific other sites and agents to which particular subjects have (not always tangible). (Marcus, 1995, p. 111)

In the case of both multi-sited and situated single-sited ethnography, the position and role of the ethnographer become quite significant. While Marcus ascribes an activist character to the researcher, he does this in a specific and circumstantial sense because of the commitments and conflicts met during research. This, nonetheless, shapes the role of the ethnographer – a topic I will discuss in the last part of the essay.

From Ethnography towards Imaginary

Christine Hine compares multi-sited ethnography to Merton’s concept of middle-range theory. Although finding significant similarities, she argues that the most crucial difference is that in the former, the ethnographer ‘embodies the tensions of a middle range which attempts to remain relevant to diverse audiences whilst faithful to a complex and ultimately methodologically elusive experienced word’ (Hine, 2007). Accordingly the ethnographer does not accept the boundaries imposed on the field in question but is willing to pursue connections and by so doing, offering ‘the possibility of crafting a research object specifically designed to engage in a particular argument’. As she argues, Marcus, with his renewal of ethnography, proposes a shift from ethnography to the ‘imaginary’ (a conceptual term she cites from another of Marcus’s work: ‘Ethnography through thick and thin’ (Marcus, 1998)), a term that expresses a more active and interfering role of the researcher.

Of course, the question is: what does this ‘imaginary’ look like in practice? Kim Fortun, in examining the common conceptualizing framework of Marcus and his colleague, Michael M. J. Fischer, points to two possibilities for politically motivated intervention with the help of multi-sited ethnography. The first, called ‘negative’, is characterised by its avoidance of ‘orientalising’ its objects of study. The second is the so-called ‘affirmative’ tack, which ‘produces rich descriptions of the reflexive idioms and evaluative modalities employed by the people studied’ and, by doing so, unsettles the orientalist interpretations of these pluralities. Fortun emphasises that just the act of ‘choosing the topics of study [is] loaded with potential to unsettle conventional understandings’ (Fortun, 2003).

She also compares ethnography to an ‘experimenting machine’, a term used by Hans-Jorg Rheinberger (1997), since

[c]onceptualizing ethnography as devices that shape questions to be asked allows us to think of them as scientific research tools. Their significance lies less in what they conclude than in the discursive resources they provide and the pathways they open up. Their aim, in short, is to improve literacy, to enable people to read the world better. Some would consider this value-neutral. I think that this is what gives ethnography ethical charge. (Fortun, 2003: 187)

This, however, does not mean that the main aim is just to ‘show the complete randomness or subjectivity’ of given phenomena. Casper Bruun Jensen, in his research on electronic patient records (EPR), investigated these devices as ‘partially existing objects’ and tried to ‘de-familiarize […] the traditional ontological assumptions’, underlying their main concepts. He argues, however, that ‘the social researcher is not free to articulate versions of this entity: most often other actors would immediately jump on incredible claims and propositions as they encountered them’. Instead, he argues, the studied object and the researcher mutually articulate each other (Jensen, 2004).

As we can see, it is quite hard to tell, at a theoretical level, what is the exact criterion of a ‘good’ ethnographic inquiry motivated by political intervention. The final decision is always the researcher’s but, in fact, the decision is not really final, since it needs the continuous practice of reflection and re-evaluation. Perhaps this is the main reason that Fortun, at the end of her chapter, argues that ‘there is something of value – perhaps even of ethical value – in the new’ and, drawing on the work of Drucilla Corness, recognises ‘the good’ as being ‘always deferred, as always outside any known system of ideality’.


As I have tried to show, the concept of multi-sited ethnography tries to give an understanding of the global relevance of local multiplicities. In most of the cases this happens by following something or by moving between various connected sites, but it also can involve research in one single place while considering the global influences. This also has consequences regarding the role of the researcher: the ethnographer, in many cases, does not just try to reveal the connections between particular places, but can play the connecting role. This requires greater responsibility from the researcher and also opens up new possibilities for intervention.

We need to ask about examining universal phenomena, those that are not physically present. Are these universals important, or do they escape any meaningful scientific inquiry? Does following actors and things linearly and drawing up only directions and connections prevent the ethnographer from seeing complex effects or are these demands just for ‘state-mappers’? These are all questions which, in my opinion, are worth further investigation.


  1. Akrich, M., 1992. The de-scription of technical objects, in: Bijker, W.E., Law, J. (Eds.), MIT Press, Cambridge, Mass. ; London, pp. 205–224.
  2. Bloor, D., 1976. The Strong Programme in the Sociology of Knowledge. Routledge and Kegan Paul, London, pp. 3–24.
  3. Coe, D.E., 1991. Levels of knowing in ethnographic inquiry. Qual. Stud. Educ. 4, 313–331.
  4. Collins, H.M., 1975. Seven Sexes – Study in Sociology of a Phenomenon, or Replication of Experiments in Physics. Sociology 205–224.
  5. De Laet, M., Mol, A., 2000. The Zimbabwe Bush Pump: Mechanics of a fluid technology. Soc. Stud. Sci. 225–263.
  6. Fortun, K., 2003. Ethnography In/Of/As Open Systems. Rev. Anthropol. 32.
  7. Foucault, M., 1991. Discipline and Punish: The Birth of the Prison, New Ed. ed. Penguin.
  8. Foucault, M., 1984. The Foucault Reader. Pantheon.
  9. Hackett, E.J., Amsterdamska, O., Lynch, M., Wajcman, J., 2008. The handbook of science and technology studies. The MIT Press.
  10. Haraway, D.J., 1997. Modest-Witness@Second-Millennium.FemaleMan-Meets-OncoMouse : feminism and technoscience. Routledge, New York ; London.
  11. Haraway, D.J., 1991a. A Cyborg Manifesto: Science, Technology, and Socialist-Feminism in the Late Twentieth Century, in: Simians, Cyborgs, and Women: The Reinvention of Nature. Free Association, London, pp. 149–181.
  12. Haraway, D.J., 1991b. Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective, in: Simians, Cyborgs, and Women: The Reinvention of Nature. Free Association, London, pp. 183–202.
  13. Heath, D., Koch, E., Ley, B., Montoya, M., 1999. Nodes and queries – Linking locations in networked fields of inquiry. Am. Behav. Sci. 450–463.
  14. Helmreich, S., 2007. An anthropologist underwater: Immersive soundscapes, submarine cyborgs, and transductive ethnography. Am. Ethnol. 34, 621–641.
  15. Hess, D., 2001. Ethnography and the Development of Science and Technology Studies, in: Atkinson, P. (Ed.), SAGE, London, pp. 234–245.
  16. Hine, C., 2007. Multi-sited ethnography as a middle range methodology for contemporary STS. Sci. Technol. Hum. Values 32, 652–671.
  17. Jasanoff, S., Markle, G.E., Peterson, J.C., Pinch, T., 2001. Handbook of science and technology studies. Sage Publications.
  18. Jensen, C.B., 2004. Researching Partially Existing Objects: What is an Electronic Patient Record? Where do you find it? How do you study it?
  19. Latour, B., 2005. Reassembling the social: an introduction to actor-network-theory. Oxford University Press, Oxford.
  20. Latour, B., 1996. On actor-network theory: a few clarifications. Soz. Welt 369–381.
  21. Latour, B., 1993. We Have Never Been Modern. Harvard University Press.
  22. Latour, B., 1983. Give Me a Laboratory and I will Raise the World. Sci. Obs. Perspect. Soc. Study Sci. 141.
  23. Latour, B., Woolgar, S., 1979. Laboratory life : the social construction of scientific facts. Sage Publications, Beverly Hills ; London.
  24. Law, J., 1994. Organizing modernity. Blackwell, Oxford.
  25. Licoppe, C., 2010. The Performative Turn in Science and Technology Studies: Towards a linguistic anthropology of “technology in action”. J. Cult. Econ. 3, 181–188.
  26. Marcus, G., 1995. Ethnography in/of the world-system – The Emergence of Multi-sited Ethnography. Annu. Rev. Anthropol. 95–117.
  27. Marcus, G.E., 1998. Ethnography Through Thick and Thin. Princeton University Press.
  28. Oppenheim, R., 2007. Actor-network theory and anthropology after science, technology, and society. Anthropol. Theory 7, 471–493.
  29. Rheinberger, H.-J., 1997. Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Stanford University Press.
  30. Tsing, A.L., 2011. Friction: an ethnography of global connection. Princeton University Press.

[1] Hess gives the example of the investigation of witchcraft opposed to the sociology of ‘modern’ science. In the former, an ethnographer explains the strength of an ‘emic’ body of knowledge (that is, meaningful only in terms of the investigated culture) pointing to ‘social criteria’ and takes into account the emic explanation itself as an explanatory reason only on the level of its acceptance as a belief system. In the latter, one cannot deny the ‘explanatory goodness’ of the studied scientific theory itself (Hess 2001). Hess parallels this distinction with the difference between cultural relativism and epistemological relativism.

Cite this page

Choose cite format:
Online Chat Messenger Email
+44 800 520 0055