Networks before the Internet

The story of the internet's creation usually goes something like this: In the early 1960s an engineer working for the RAND Corporation, Paul Baran, published an eleven-part memorandum titled On Distributed Communications, which outlined the possibility of "packet switching."¹ Baran claimed that distributed, networked telecommunications would allow military communication to persist in the wake of a catastrophic attack, something he stressed given the seemingly imminent threat of nuclear war. Influenced by Baran's models, the Department of Defense and universities across the United States collaborated to create ARPANET, a packet-switching network designed to share physical computer resources among universities, even if, unlike the model sketched in Baran's memos, it was not built to withstand the bomb. By the 1980s, ARPANET was linked with or replaced by other nonmilitary (but still predominately American) communications networks, like the National Science Foundation's NSFNet. Ten years later, the original government-sponsored "network of networks" had been privatized. In sum, as Janet Abbate tells us, the history of the internet "is a tale of collaboration and conflict among a remarkable variety of players," an ever-mutating patchwork of competing interests.²

The usual tales of the internet's development tend to sideline particular networks that shaped how the internet came to be in countries other than the United States. Narratives that privilege ARPANET inevitably position the United States—its military, universities, and corporations—as the locus of digitally networked power. They often ignore how various systems for digital networks existed in different configurations in different places and times, for different ends. They also overlook the alternatives not chosen for a particular technology. The specific protocol that made ARPANET central to the foundation of the internet, TCP/IP, existed and continues to exist alongside a range of alternatives, such as proprietary protocols like Apple's AppleTalk or Novell's IPX/SPX (Internetwork Packet Exchange / Sequenced Packet Exchange), or, most significantly, RINA (Recursive InterNetwork Architecture), designed as a replacement for TCP/IP that takes into account its many limitations. These alternatives could—probably subtly but perhaps significantly—transform the politics, economy, structure, and experience of digital networks.³ Thus, a broad range of internet histories have worked to challenge the privilege given to ARPANET, the United States, and the militaristic foundations of the internet, demonstrating that our current internet is one possibility among many, and that the technical and social institutions embedded in the material form of the internet are still marked by contingency rather than necessity.⁴

What I offer here is more oblique than the approaches taken by those working to globalize internet history or those who stress the alternatives and conflicts embedded in the development of technology, even though I follow their goal of rethinking the centrality or inevitability of the descendants of ARPANET. Challenging the particular political and social hegemony embedded in the internet today requires overturning a much wider cultural formation than that which is immediately conjoined with the internet's technical and material form. This involves a genealogy of beliefs about the social relations fostered, determined, or remade by the internet, beliefs about the necessity of "connection" or "connectivity" as that which grounds the social. The internet, I claim, is a material instantiation of a particular network imaginary, one that reduces the world to little more than a series of nodal points and the flows that move between them. Neither is this imaginary determined by the internet itself—at least directly—nor is it the only way of understanding the relations embedded within and perpetuated by the internet. It is, however, characteristic of countless popular and scholarly discussions of the internet.⁵ Like the protocols and technologies of the internet, this imaginary has a history; how we imagine the internet is the product of contestation and crisis, filled with forgotten minutiae, historical dead ends, and competing definitions of network, flow, and connectivity. Imagining a networked society was something that happened only through a series of dispersed historical precedents. These precedents include not just the direct technical predecessors of digital networks but also the networks figured in the histories of textiles, anatomy, branch banking, railroads, radio...