Control Society Expanded?

By Mark Poster, 13 July 2004

How does control exist after decentralisation? asks Alexander Galloway’s book Protocol. Mark Poster is still waiting for an answer

Protocol: How Control Exists After Decentralisation, Alexander Galloway, Cambridge: MIT Press, 2004

As the social world becomes more and more saturated with information machines, with media increasingly producing, transmitting and storing cultural objects of all sorts, rules are being developed not so much to shape human behaviour but to facilitate and constrain the relations of one information machine to another and the interface between humans and media. Alexander Galloway, in this insightful and helpful book, explores these rules under the name of protocols. The author restricts his subject to the internet and its communication standards, but we must understand his most important subject as part of a larger tendency in which social space is increasingly populated by apparatuses or complexes of humans and machines. How the emerging global scene of human-machine assemblages, as Gilles Deleuze and Félix Guattari term them, are constituted, arranged, institutionalised, fought over, and transformed is one of the pressing issues of our time.

Galloway makes an important contribution to this discussion. He inserts himself into the ongoing debate over the nature of the internet against those who would picture networked computing as a realm of pure freedom; the John Perry Barlows (‘A Declaration of Independence of Cyberspace’ of 1996) and Nicholas Negropontes (Being Digital, 1995) who attracted so much early attention in their manifestos of cyberspace as utopian worlds, composed of bits not atoms, places where the normal, territorial rules simply did not apply. These ideologues of what was called by some ‘the Californian ideology’ proclaimed the digital as equivalent to the free (‘information wants to be free’ in Barlow’s influential phrase) and mocked those who would dismiss the internet as simply another tool of communication and especially technophobes, like Paul Virilio, who feared in its dissemination the end-of-everything-worthwhile. Barlow and Negroponte dominated the ‘positive’ view of the internet, associating it, unfortunately, with neo-liberal resonances and, although not directly, with the increasing domination of the globe by the United States.

Galloway is equally critical, however, of those who would minimise or even overlook completely the novelty of the internet. I am referring to the large number of commentators, theorists and scholars who reduced the significance of networked computing to simply another tool of capitalism (both liberals and Marxists belong here), or yet one more example of patriarchy at work (some feminists), or still a further development of racist practices, or a new and dangerous tactic in western imperialism, and so on. For this very large group of observers the internet is understood as nothing but an extension of previously existing forms of social life, benefiting those in power, and adding strength to the dominant institutions of modern society. For these critics there was nothing special about what computer scientists call ‘distributed knowledge’, nothing special about ‘the virtual’, only a repetition of modern, western society’s project of total, global domination by human beings.

Against the latter group Galloway argues for the unique nature of the protocols of the network; against the former he shows convincingly that these protocols are new kinds of constraints, ones not without their own kind of danger. There are according to Galloway three architectures of constraint: hierarchical, decentralised, and distributed. The first pertains to pyramidal organisations exemplified by feudal societies where authority is hierarchical and centralised. These are familiar enough. The decentralised model, Galloway contends, is that analysed by Foucault as disciplinary structure, with each institution – prison, hospital, workplace, school – replicating in its physical architecture and practices the model of the Panopticon. Computer networks constitute a new, distributed pattern that is ‘the enemy of bureaucracy, of rigid hierarchy, and of centralisation.’ (p. 29) Key to such networks are protocols that allow each element to function autonomously but yet are connected to the whole. Galloway follows Deleuze’s argument that we have passed from Foucault’s model of decentralised, disciplinary society to ‘societies of control’.

The strength of Protocol is its analysis in Part 1 of the book of ‘How Control Exists after Decentralisation’. Those who are not computer scientists or electronic engineers will benefit greatly from the clarity Galloway brings to the understanding of the physical structure of the internet and more particularly to the social organisations that regulate the protocols by which it operates, a truly complex array of voluntary and professional organisations that by and large make decisions by consensus and enforce those decisions only by voluntary cooperation of the local parties. As Galloway emphasises, there is nothing comparable to this in modern history. The regulation of TCP/IP and DNS, the basic protocols that enable communication of information among networked computers, is beyond the control of nation states and corporations yet is not at all arbitrary but organised and reorganised in a systematic albeit very complex manner. I highly recommend these chapters, along with Part 2 which outlines the institutionalisation of protocol, to anyone who wants to get insight into the basic workings of the internet.

Galloway also shows how the protocols contain and embrace contradictory models: TCP/IP (Transmission Control Protocol/Internet Protocol) protocols are regulated on the model of consensus defined above. TCP/IP enables each machine to send and receive packets of data without approval or control by any ‘higher’ level, without any specified path, and with some certainty of accuracy in the transmission. This is a highly ‘democratic’ model of organisation. DNS (Domain Name System) protocols that enable alphabetic URLs (Uniform Resource Locators or web addresses) to be translated into numbers comprehensible to computers are on the contrary structured in a hierarchical pyramid, from the highest level of type (‘org’, ‘com’, ‘edu,’ etc.) to the lowest level specificity. From the most basic level of individual machine and optical fibre linkage to the software (such as a browser) to the content of the transmission, the rules of connectivity (Galloway’s protocols) combine the two different systems of control in order to enable a remarkable mass of communications.

Part 3 of the book explores what Galloway calls ‘failures’ of control: hackers, viruses, resistant groups such as cyberfeminists, and internet art. In each case, the controls built into the internet are subverted, by-passed, eluded, refuted, made evident, and so forth. These sections of the book illustrate well what Galloway means by ‘control’ and explore the dialectics of opposition to the model of distributed control. In themselves each chapter consists of familiar material but their connection with the main theme of the book renders them worthwhile.

The chief difficulty with the book, one that requires further debate, is the link Galloway draws between Deleuze’s notion of societies of control and the analysis of internet protocols. I do not have space to do justice to this question but suffice it to say that everything hinges on what one means by ‘control’. Galloway does not help us here, providing no definition of control and no discussion of its varieties such as the difference between a feudal king and TCP/IP. In the one case ‘control’ refers to the agency of a human being and in the other to the constraint of an information machine. Deleuze’s discussion of control societies is maddeningly brief and, to my mind, unconvincing. Galloway’s purpose – to illustrate Deleuze’s theory in relation to internet protocols – is burdened more than aided by the French theorist’s speculations. Deleuze feared that control society extended the authority of Foucault’s Panopticon beyond the limits of territorial institutions. But Galloway indicates effectively how the control society of the internet constitutes a ‘rhizomic’ structure far less imbued with authority and hierarchy than the 19th century prison. This stinging contradiction troubles an otherwise highly valuable study and points to urgent directions for further theorising and study.

Mark Poster <poster AT> teaches in the Departments of Film and Media Studies and History at the University of California, Irvine. His most recent book is What's the Matter with the Internet? (2001)