Reviewer: Andrew Goffey

1) Is the subject matter relevant?

A properly historicised and contextualised account of bitcoin and the infrastructural functions of cyptography seems to me to be a really important issue for anyone wishing to grasp key drivers in the developing logic of relations between the public and the private, and the politics of digital infrastructures

2) Is the treatment of the subject matter intellectually interesting? Are there citations of bodies of literature you think are essential to which the author has not referred?

The article is interesting and demonstrates a pretty good technical understanding of the kinds of technical issues that it is addressing. It is a little difficult to recommend citations of bodies of literature not mentioned here, in part because it is not entirely clear to me what the author is trying to do in this essay

3) Are there any noticeable problems with the author’s mean of validating assumptions or making judgements?
It’s not really clear what kind of status the different elements of the author’s account have. What status does Shannon’s theory have, for example? How do we go from the Mathematical Theory of Communication to bitcoin, other than to talk about how it can be modelled in terms of Shannon’s ideas? Doesn’t this leave the status of a theory like Shannon’s rather unclarified? And given the author’s later references to eg science and technology studies work and even to Foucauldian epistemes, isn’t this rather problematic? It would be great to see Shannon’s MTC reworked in terms of Foucault’s epistemic shifts – that would at least make the later use of Deleuze make more precise sense, I think.

4) Is the article well written?

As mentioned in 2 above, the article demonstrates a pretty good technical understanding of key elements of the discussion. I felt rather unclear though about what the article was driving at, and was not entirely convinced that the structure served the author all that well. For example, in the final section, the reader is introduced to a whole new body of material which is not previously mentioned in the article, which suggests that the author is moving into new ground not previously covered in the article. I think that this is a bit problematic and it suggests that what is being done with the article has not been thought through with sufficient care. I was a little uncomfortable about the extensive levels of description in the sections of the article that offered primers. And it wasn’t really clear to me how the links were being made between the different bodies of material that the article was drawing on (see 3)

5) Are there portions of the article that you recommend to be shortened, excised or expanded?
The subject matter of this article is important and a discussion that can make connections between technical-scientific theories, socio-technical devices such as bitcoin, and critical-philosophical historiography like that of Foucault (and Deleuze, even if Deleuze’s control essay is suggestive rather than closely analytic) would be invaluable. I would suggest that a lot of careful restructuring and rethinking is required here. What does the author want to do and how does he or she want to make those links that are otherwise suggestively pointed at here? The structure of the argument here and the presentation of the evidence used to develop it is currently rather unclear.

Reviewer: Adrian MacKenzie

1) Is the subject matter relevant?

This is a timely paper, given the many disclosures about cryptography and the US NSA in recent months. But even before all that, I’ve long wanted to read a paper that help makes sense of the increasingly pervasive role of cryptographic techniques in digital cultures. The strength of the paper consists in its close attention to contemporary cryptographic practice, and its effort to move cryptography to a more central, perhaps fundamental position in the politics of information. The core issue here does not concern secrecy or secrets as such. Quinn DuPont argues that while State and corporate actors attempt to monopolise secrecy in their own interests, the real issue is a set of non-secret powers that generate social effects of diverse kinds. Only sometimes are these applied for the purposes of secrecy.

2) Is the treatment of the subject matter intellectually interesting? Are there citations of bodies of literature you think are essential to which the author has not referred?

The case in point in the paper is the cryptographically-validated non-governent-backed currency, bitcoin. Bitcoin has been in the news because of high-profile seizures of bitcoin by US Federal authorities (for instance, in the raid on the website Silk Road), also because of sudden change in the values of the currency, as DuPont describes, and more generally, because it offers a way to carry out commercial exchange without relying on national, government-backed currencies. But what bitcoin tells us relates less to the power of cryptography to secure information, than the power of cryptography to attract belief and desire. And this power to attract points to a different conception of information, an alternative to the Shannon/Weaver mathematical theory of communication (MTC) that has governed and continues to orient most understandings of contemporary media and communication.

The price to pay for this alternative theory of information is quite high: we have to learn about how contemporary cryptography works in general and how it works in particular in ‘bitcoin mining’ (for instance, via the role played by ‘hash digests’). Much of this information could be found easily in any standard textbook (for instance, Bruce Schneier’s compendium described these techniques in details in the mid-1990s) or gleaned from the copious pages on the topic in Wikipedia. We need this in order to understand how cryptographic techniques find themselves used in unexpected ways. The examples that DuPont provides here include ‘verifying large data structures.’ MTC still prevails, and in order to see how we think of information differently. We are also asked to think about cryptography more formally as a ‘notational system sustained by second-order metaphor.’

The key point for DuPoint is that secrecy is only one of many possible uses of cryptography. Bitcoin demonstrates this through the utter transparency of its block-chain treatment of transactions. Every transaction associated with a given bitcoin can be read and validated, at least in principle.

One important gap in DuPont’s alternative account of the information order is the work of Alan Turing. In his work on computable numbers and the universal computing machine, in his WWII code-breaking work and his work on artificial intelligence Turing is both a leading example of the entanglements of information and cryptography, and embodies a very different understanding of computation. There is a variety of scholarly work that might be relevant here, including Alan Hodge’s biography of Turing, Friedrich Kittler’s readings of Turing’s work, and, although I hesitate to mention it but relates directly to this topic, my own ‘Undecidability: the History and Time of the Universal Turing Machine’ (Mackenzie, 1997). Other scholarly literature on public key cryptography must also exist!

In terms of theory, I was a bit underwhelmed by the recourse to Gilles Deleuze’s ‘Postscript on Societies of Control’ as theoretical anchor point for the alternative account of information. Even taking into account possible slips on Deleuze’s part in his characterisation of control in terms of number and modulation, it is very hard to say anything new about this piece because it has been used so extensively for at least a decade now.

3) Are there any noticeable problems with the author’s mean of validating assumptions or making judgements?

I’d say the last section of the article suffers an attempt to move quickly over a terrain spanning Rousseau, Heidegger, Latour, Foucault, Deleuze, Hayles, etc. This section was too synoptic for my liking: it didn’t help all that much in making sense of what is happening to cryptography, although I did try to follow the single line of thought that seemed most important to me: the notion of ordering. But the paper does not offer any sustained analysis of what the material practices of order are, other than saying the algorithms like SHA-256 are doing ordering work or ordering logics. I really would like to see this judgment fleshed out more in the last section. There is, it seems to me, a really valuable intuition here, but at the moment it is rather swamped by the other material.
4) Is the article well written?

Overall, I’d say the article is well-written. I think the order of the sections could be improved. For me, the bitcoin praxis section would have been more effective near the beginning of the article. It would provide material that could be progressively explored and analyzed in later sections. I also found it slightly strange for a theoretical ‘point of departure’ to appear in the last couple of pages. I think that should come earlier, or at least by signposted earlier. Not too much, however, hinges in the overall argument on the current order of the sections.

There are some deficits in the evidence provided. For instance, ‘the anarchist hype’ is mentioned a couple of times without citing sources or examples.

5) Are there portions of the article that you recommend to be shortened, excised or expanded?

I found the section on ‘Conceptualizing Cryptography’ the least convincing because it relies a single philosophical position (Nelson Goodman’s work) that itself is not analyzed, contextualised or contrasted with alternatives. I think the paper could do without this section.

I wondered whether the author really meant ‘non-reputability of datum’ in the ‘Advanced Primer’ section. This was a new term for me, and it comes up a number of times. Is it meant to be ‘non-repeatability’? Either way, I think the term needs clarification.