Tor is not a “Fundamental Law of the Universe”
Editor's note: This is a guest post by David Golumbia, author of The Cultural Logic of Computation. Golumbia is Assistant Professor in the Department of English and the MATX PhD program at Virginia Commonwealth University. It originally appeared on Uncomputing.org.
In what I consider a very welcome act of journalistic open-mindedness, Pando Daily recently published a piece by Quinn Norton that responded both to Yasha Levine’s excellent and necessary piece on the US Government’s funding of the Tor Project, and perhaps even more so his even more necessary piece on the amazing attacks his piece received from some of the brightest stars in the encryption, “internet freedom,” and Tor universe.
I want to focus on a small part of Norton’s piece, in which she tries to explain the vicious attacks on Levine’s piece:
The incoherent frothing-at-the-mouth support for the fundamentals of Tor don’t arise from a set of politics, or money, or a particular arrangement of social trust like a statute or constitutional law. The support comes from an appeal to the fundamental laws of the universe, which not even the most vigorous of black budget ops can break.Yes, the Tor people somehow believe that Tor itself implements a “fundamental law of the universe,” and that their privileged technical knowledge grants them special access that the rest of us lack. That is false, breathless narcissism and arrogance at its most outrageous, and very typical of our digital age.
There are fundamental laws of the universe: that something with mass cannot move faster than the speed of light in a vacuum; that matter can neither be created nor destroyed, but only converted into energy, and vice versa. These are fundamental laws that DoD can’t change. All technologies dip into these fundamental laws, to greater and lesser degrees.
Tor is not a fundamental law of the universe. Math is *fairly* fundamental, but even the simplest math–say, 2 + 2 =4–is NOT a fundamental law. Addition obtains under some circumstances, and not under others (this is part of the point of the revolutions in mathematical theory of the 19th century, including non-Euclidean geometry).
Grammatically, the phrase “Tor is/is not a fundamental law of the universe”—which, to be clear, is my phrase, not Norton’s—makes no sense. But other than the vague notion of “mathematical laws,” which she does not even directly invoke, Norton’s statement that Tor advocates “appeal” to the “fundamental laws of the universe” is conceptually no clearer. There are not that many fundamental laws. Tor “appeals” to them no more or less than, say, the NSA does when it uses satellites that rely in part on relativistic physics for geolocation. Relativity itself is a strange candidate for a “fundamental” law, for lots of interesting philosophical and scientific reasons, which does not mean it is not fundamental either, but my point is to show that the belief of Tor advocates that they are tapping into something over and above what the rest of us have access to is misbegotten and hubristic in the extreme. If what the Tor people are trying to show is that their cryptographic procedures are sound, fine. But that is not what we are talking about. We are talking about the use of Tor in the world.
The math on which Tor is based appears solid, as far as I can tell not being a cryptographic mathematician myself, according to both the Tor developers and outside analysts. Yet the actions Tor exists to enable–the use of digital communications systems in our world, supposedly opaque to traffic analysis (the main purpose Tor’s development team claim for their product)–are governed by no parallel solidity. It is part of a huge social apparatus. No matter how perfect the math, as long as that apparatus is large and involves people, it cannot be governed entirely by an unbreakable fundamental law of the universe. One is not challenging the laws of motion by buying out the operator of a Tor relay or exit node. It would be more accurate to say that a fundamental law of the social world is that people–including many of the same hackers Quinn Norton defends and celebrates in other writings–will do everything they can to find their way in to systems like Tor.
Further, we have historical evidence that proves this. People keep breaking Tor (something I am personally happy about). Sometimes the Tor developers seem mystified. But I think people will keep breaking it (and so, apparently, do the Tor developers themselves). And I do think those who fund it are aware of its systematic vulnerabilities in a way those hypnotized by its math are not, and that’s why the story of its funding is actually extremely relevant. It is not something that can’t be controlled by the state: it can be, or at least degraded significantly, and as someone above says, in part because they could easily shut it down altogether through a program of disallowing a wide range of onion routing services, relays and nodes that Tor depends on to run, and continuing to disallow the new services that Tor sets up.
Tor has edges. It has layers. It has connections with other services. It has physical systems on which it sits, binaries that must be compiled, and hundreds of other points of vulnerability even granted that onion routing is mathematically robust to breaking. Oddly enough, Google Chief Engineer and life-extension supplementmagnate Ray Kurzweil, whom I consider a nut, believes Moore’s law is a “fundamental law of the universe” (or, more accurately, a corollary to the “law of accelerating returns” that he thinks is fundamental) which I think he is also wrong about, but whether it’s fundamental or not, Moore’s law suggests that computers will get fast enough eventually to break current encryption routines–perhaps breaking Tor, perhaps making historical records of Tor use much more subject to network analysis than they seem today. And who knows? Maybe there is some kind of non-Euclidean (or some other alternative) approach to cryptography nobody has stumbled on yet that will square the circle in a way none of us can foresee.
It is precisely the hubris and arrogance of Tor and its developers, the hubris of computationalists who believe their self-adjudicated superior knowledge of machines compared to everyone else, who believe that their access to fundamental computational/mathematical “laws” insulates them from society and politics–that makes this discussion so difficult. Tor’s amazing reactions to the Pando coverage are exactly the kind of petulant, arrogant, dismissive, power-hungry reaction computationalists typically have when anyone they deem unworthy dares actually to ask them to account for themselves. They think they are above us, in so many different ways. They are not, and they hate anyone who dares to suggest that they are just people too.