The Dependence of Cyberspace

I wrote this for The Maintainers conference .

Introduction: There’s Never Been Freedom of Anything in Cyberspace

I want to begin by discussing the context in which John Perry Barlow wrote his Declaration of the Independence of Cyberspace from Davos in 1996. For those unaware, Barlow was on a writing deadline, at a party, with the world’s business elite. He went back to his hotel, typed out the Declaration, emailed it out to some hundreds of friends, and the rest is (cyber-)history.

Caveat: I think Barlow’s thinking and work is important, and he was genuine and kind to me when I met him once. I think he’s been selfless in trying to make the Internet a safe place for freedom and democracy and I have nothing but respect for him. I understand the goals that he meant to further with his rhetorical strategy in the Declaration. That said, I’m here to problematize. From the Declaration:

Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather. [EFF link]

Governments are people, and cyberspace is a real place, independent of the computers on which it resides, computers that are, like the rest of us, enmeshed in a grab-bag of power relations, discursive miscellany, and the rest of it. There’s a lot going on here. Let’s take a look at the context.

In 1996, the US National Science Foundation had just finished privatizing the then- teenage Internet, with its approximately 36 million users. Until late ‘94 / early ‘95, the Internet backbone was run by the NSF, making it amenable to government regulation: for example, it had a policy regulating commercial use. Its NSFNET staff had, to put it gently, gotten ahead of congressional plans for gradual introduction of commercial interests to this infrastructure that had been in development since the late 1960s at public expense. So, just two years after the rapid privatization of government-created Internet resources and the creation of an online world in which constitutional guarantees to free speech (for example) no longer potentially applied, Barlow demanded that the government stay out. Meanwhile, he continued efforts with the EFF, an organization dedicated to marshalling state power in defense of speech and other freedoms online. Libertarianism is complicated.

The Internet was not, of course, the only game in town. The early Internet infrastructure to which I refer is, of course, the origins of the architecture of power and governance under which we live today. That’s why I’m talking about that same cast of players. It’s relevant because here we are. But Barlow was a WELL user, and I don’t know what other precedents he might have had in mind when telling everyone that the government had no sovereignty in cyberspace. Even if the Internet’s portion of cyberspace was no Eden, maybe there was a historical precedent somewhere else, a precedent for a world in which the state, quite naturally, had no sovereignty; let’s take a tour.

In late ‘94 the NSF consigned the Internet to the private sector. The Internet’s origins lie with a military network where admins monitored user metadata and explicitly reminded users that use was a privilege and not a right. This network, the Arpanet, was the (first) architectural backbone of the Internet from the Internet’s emergence as infrastructure in the late 1970s, until the NSFNET backbone came online in 1986 (it was the first Autonomous System). During that time the Defense Communications Agency (DISA) reminded users that use was a privilege, not a right, because they owned the place. During the NSF’s governance the DARPA-formed and influenced governance institutions, such as the IAB, continued, and the IETF was formed, staffed largely by defense-funded researchers at defense contractors and universities funded by defense. After the NSF privatization, the Internet moved to the private sector, beginning a trajectory that continues today in which services designed to mimic public forms are actually businesses, and as such, not even common carrier status for the backbone infrastructure does much for speech.

Speaking of privately held servers, that’s what Bulletin Board Systems (BBSs) were. (Yes, I’m especially including The Well.) Entering cyberspace on a BBS usually meant entering a computer in somebody else’s living room or bedroom. This includes BBS networks like Fidonet, obviously. It also includes larger Wide Area Networks like Compuserve and Prodigy, and, hey, AOL, except here the person owning the system is a private firm, and by “living room” I mean “office park.” Prodigy actively regulated speech on their systems and terminated accounts without warning. Courts found that Compuserve, too, was not a First Amendment actor. Efforts to introduce introduce speech protections on “the new home of the Mind” failed. On USENET and BITNET, it was the dreaded sysadmin that regulated posts and local subscriptions. The servers on which the messages were hosted were, all together now, private firms.

I am not taking a stance on whether there should be freedoms of whatever in cyberspace. But any political project based around that goal should probably start with better mythopoetics than totally indefensible historical claims. There is no historical basis for privacy, or for freedom of speech, or for freedom of anything, on the Internet, or in cyberspace broadly defined. That so much depends on claims otherwise is important.

It is also worth taking a look at what, in 1995, the year prior to Barlow’s Declaration, people were saying about cyberspace. One overwhelming theme is information war. In 1994, INFOWARCON kicked off, the conference dedicated to information war. And the publications were numerous. For example, “Dawn of Information Age Will Change Military More Than Cold War’s End.” Aerospace Daily 174 (1995): 325. “Digitized Zypher Lifting Fog From No Man’s Land: Army Pushes Information Warfare Transition.” National Defense  80, no. 510 (1995): 32-33. “Information Warriors Raze Enemy’s Vital Data Chains.” National Defense 79 (1995): 30-31. Adams, David, and others. “The Public Switched Network: an Overview and Vulnerability Assessment.” Air Command and Staff College, 1995. Alexander, David. “Information Warfare and the Digitized Battlefield.” Military Technology 19, no. 9 (1995): 57-59+. Ayers, Robert L. “DISA and Information Warfare.”InfoWar Con 1995, H3-H58. Bean, Mark H. “Fourth Generation Warfare?” Marine Corps Gazette 79, no. 3 (1995). Bender, Brian. “Navy Chief Commissions Fleet Information Warfare Center.” Defense Daily, 25 October 1995, pp. 108-9. Berkowitz, Bruce D. “Warfare in the Information Age.” Issues in Science and Technology  (1995): 59-66. Clapper, James R., and Eben H. Trevino. “Critical Security Dominates Information Warfare Moves.” Signal 49, no. 7 (1995): 71-72. Cohen, Frederick B. Protection and Security on the Information Superhighway. New York: John Wiley & Sons, 1995. Constance, Paul. “From Bombs to Bytes: Era of on-Line Weaponry Is Here.” Government Computer News 14 (1995): 51. Feigenbaum, Edward A. “The Intelligent Use of Machine Intelligence.” CrossTalk  (1995): 10-13.

At DEFCON #3, also in 1995, Barlow was slated to speak alongside Winn Schwartau, who was giving a “comedic” talk entitled “Information Warfare: The Year in Review.” It’s not clear in the final agenda that Barlow made the conference, but you get the idea. Information War was in the air. Commentators everywhere noticed that the increasing number of connected devices and their centrality in society meant that ‘cyber’ was quickly becoming a relevant domain for the military, and for terrorist groups, for that matter. My point? By 1996, not only was there no historical precedent for a naturally free cyber domain, but that domain was being rapidly converted to a domain for global conflict. (In the case of the Internet, it went from an infrastructure meant to facilitate more efficient warfighting to an actual domain in which the warfighting took place.)

So why would Barlow and the tens of thousands of people who re-hosted his essay insist that cyberspace is naturally free? What are the consequences of that insistence?

Reification and Context

The claim that cyberspace is i) a real space and that ii) it is naturally or properly free from state sovereignty is possible by, conceptually, removing it from society. It’s possible by decontextualizing it so hard that it doesn’t matter who owns the computers, who designed the protocols, what armies of labor maintain it, what other infrastructures make it possible to exist. So hard that even its history of ownership, control, and practices don’t matter either, so hard that it appears as an “act of nature” that “grows itself through our collective actions,” that is most certainly not anything like a public project (“Do not think that you can build it, as though it were a public construction project.”). This decontextualization creates its own objectivity, its own principles and fundamental properties through which, ironically, actual social context can be interpreted or ignored.

And such an objectivity comes with its own problems. Ignoring for a moment the fact that cyberspace is a metaphor composed of things that governments and corporations own, let’s recall how in the US, the government and its constitution is the only entity capable of guaranteeing free speech, or free assembly. It’s also the only entity that can sign trade deals and guarantee the right to private property (on which every server farm depends). What does it mean to tell the government, without any evidence whatsoever, that it has no right in this new dematerialized world? It means libertarianism, it means the broken thinking in which everyone is taught to forget that government is required to guarantee the freedoms they seek to extend. It means an association of tyranny with the state, and freedom with what has ultimately become the private sector. It’s telling that Barlow cites de Tocqueville in the Declaration, a thinker who very explicitly warned that private associations were becoming a powerful private sector that could destroy the equilibrium that kept democracy in place in America (in fairness, it was nearer to the end of Democracy in America). And, when it comes to technology, thought leaders have the option to sneak discredited social and political theory into ostensibly technical claims, as David Golumbia documents in his scholarship on bitcoin. (Reading Golumbia’s work inspired me to revisit political theory and libertarianism in my work on the Internet, as I do in early form here.)

What, then, does maintenance tell us about cyberspace, and what does innovation conceal? Here I will explicitly contrast the methodological requirements for investigations of maintenance, which I characterize as contextual, with scholarship that seeks to explain infrastructures and other social facts with recourse to innovation. I offer that innovation-speak has the same consequence as the kind of strategic decontextualization that we see in (cyber-)libertarianism and the well-known Declaration.

As reification, innovation has the character of collapsing the space-time of knowledge and practices, and reducing them to a tiny subset that is then said to have determined them — or, to have determined all that matters about them. For example, when ‘the Internet’ or TCP/IP or CSNET or the ARPANET (etc., etc.) is said to originate from a tiny subset of people and ideas, it closes off from discussion that ways in which those things in fact were formed, and in fact continue to exist, embedded in social relations, as a part of society. It is then possible to link fundamental characters to a whole (in this case) network infrastructure, based entirely on a reading of a bit of innovation at the beginning, and design policy prescriptions based on the imagined essence and its relationship with society.

The Internet was designed to be free, and is fundamentally free, and because of this fundamental quality, I’m going to need you to go ahead and implement Policy X.

It gives the author permission to abstract away that which is messy, or reflects parts about society that are not fit to print. It makes heroes, not for the sake of valorizing a few people but because of what those people might represent. The actual “inventor” of an Internet technology (or of the Internet itself!) does not matter, which might be why those debates are usually so short of engagement with their actual work. Here, they do not matter, because their function is to center a system constructed so that others can be made to disappear. All that is necessary, all that should be a starting-point of an analysis of how an infrastructure functions and the ways it conditions our interactions that it mediates, it all goes away: the laborer vanishes, so does the government regulator, as does, for that matter, a range of inconvenient infrastructures (e.g. electricity, primary education) that are necessary conditions for its existence. Innovation speak is a massive abstraction that lets its author destroy most of the world and leave the only parts that matter.

The Dependence of Cyberspace

Maintenance, in contrast, comes in many forms. It’s obvious that there is the maintenance of institutions, of infrastructures, of relationships. Maintenance requires a broader contextualizing through time and space, because it answers questions about what happens to the requirements that accrue to things on account of actually existing in some kind of totality, in our world. I suspect it requires a more empirical take, just as I suspect it invites a sober recognition of multiple standpoints — to reveal the world, not annihilate (most) of it. (This is also why new materialism might better be described as anti-materialism.) It, too, comes with its own political commitments (to labor, for example), but does not as its starting-point offer explanations of complex and embedded infrastructures with recourse to a quickie intellectual history.

I speak of the “dependence of cyberspace” because in order to function as a notional space, in order for the dematerialized metaphors to fool so many people, it requires constant upkeep. For every person that thinks of cyberspace as a distributed, decentralized, and self-regulating space — there’s a NOC graveyard shift to thank. Cyberspace requires communities and labor and, yes, even government oversight. As I argue above, I do not think that the Declaration makes any sense historically, and I note that it was written in a context when government (and military) involvement in cyberspace was not just growing, but conspicuous. The Declaration makes sense in the same way that we can understand a consequence of much innovation-speak, as a project that in order to accomplish certain political ends, seeks to abstract cyberspace from society and remove that embeddedness and its political consequences from reasonable consideration.