Where do we find the Internet? That is, where do we go to study it? I think a summation of the previous contributions to this session of Correspondences makes it clear that we can expect our accounts of the Internet to be fundamentally different based on how we answer these questions. When Daniel Miller says he doesn’t believe in The Internet, I take that to mean that the Internet is far from singular: there are more Internets than any of us can know or experience, as Miller himself showed forcefully in his coauthored book on the Internet in Trinidad (Miller and Slater 2000). I also take it to mean that an imagined boundary drawn around the online world as a kind of new frontier for ethnographic study—while it may have been theoretically generative for some time—is an idea that has become a conceptual impediment. The insight that lives are lived neither distinctively online nor offline, but betwixt and between, came early in ethnographic research on the Internet: certainly no later than Miller and Slater’s account, and perhaps even earlier. So why are we still talking about it? Why the need for a recent rebranding of this conceptual failure as “digital dualism” (Jurgenson 2012)?

Perhaps we continue trying to study the Internet in such a direct and focused fashion because of the intellectual safety of remaining a field of marginalized specialists with an unspecified (and thus unthreatening) relationship to the discipline of anthropology. Surely, by now, contemporary fieldworkers routinely account for emails, text messages, Facebook posts, and WhatsApp chat sessions, no matter what their topic or population of interest is. Yet ethnographers often get away with failing to contend with the theoretical developments that have come out of the focused ethnographic study of technologically mediated interaction. One tendency I have seen is that new forms of media (from web pages to Facebook posts) get folded in to all other utterances as one more part of the great flow of things said, which is the analyzable text of social discourse (Geertz 1973).

To my mind, interventions that are focused on materiality (within material culture studies, science and technology studies, media anthropology, and elsewhere) have the potential to get things back on track, to force ethnographic analysis to address and acknowledge not just the symbolic work of things said, but the significance of the mode of mediation. Can we now work toward reconciling and integrating this approach into the disciplinary mainstream of anthropology? Just as Miller and Jolynna Sinanan demonstrated in their contributions to this session, encounters with the Internet are part of the lives of almost any population that might be studied by traditional ethnographic means. These lives are not conducted in devotion to the Internet, but the Internet offers a number of interactive platforms that people engage with and through. As a consequence, so must their ethnographers.

Now I want to backtrack and make the other argument, the one in favor of a focused study of the Internet. As I mentioned, the work done by ethnographers often starts by asking the question “where?” In 2004, my answer to this question was the Internet cafes of Accra, Ghana (Burrell 2012). More recently, as I explain below, my answer has shifted to coastal Mendocino County, California. So to complement the idea of the Internet as part of ordinary, everyday human social and cultural practice, I would also agree with Julia Fleischhack that aspects of the Internet, such as its internal ordering or the distinctive communities it networks together, can still be the centering force of new ethnographic projects. Fleischhack gives credence to the idea of the Internet as a space that still has many unmarked and inaccessible territories. She alludes to a dark web in which a sharp disconnection from offline identity may be critical to particular social and political projects.

Another way of answering the question “where is the Internet?” is to tackle it in especially literal terms. There are a number of scholars—although not too many anthropologists so far—who have become enamored with uncovering the Internet’s ever evolving physical infrastructure. These scholars end up in data centers (Farman 2015; Holt and Vondreau 2015), at cable landing stations (Starosielski 2015) and Internet Exchange points (Dourish 2015), or, in my case, in Manchester, California, where a fiber optic cable (one of a dozen or so along the west coast of the United States) connects to Japan. In the vicinity of that cable are a series of tiny towns of a few hundred people, an area that is genuinely remote. I have sought to understand the relationship between the people in this area and the history of cable laying, along with largely unsuccessful efforts to gain local access to the cable.

While the initial description of this Correspondences series mentioned going behind the surface or interface (to consider developers, servers, or algorithms), this dimension of the Internet was little mentioned in the other contributions. Generally, such behind-the-scenes work poses major access problems and in more fundamental ways evades any answer to the question of “where?” I will end my contribution simply by promoting some of these starting points, such as data centers, cable landing stations, or the conferences of search engine optimization specialists (Zeiwitz 2013), sites that are starting to be settled by scholars from other scholarly traditions but that are likely to be enriched by anthropological attention as part of a still vibrant tradition of studying the Internet.