November, 2014

Still Against Networks

Against Networks was first writ­ten in 2004. It is a bit of an odd ar­ti­cle, one that is ap­pa­r­ent­ly more in­te­res­ting on the con­ti­nent (whe­re this is the se­cond time an ex­cer­pt will be pu­blis­hed) than it is in the UK or the US (whe­re it has been re­jec­ted twice, but has a small, cu­rious and fri­end­ly re­a­dership). I sus­pect this has so­me­thing to do with the re­la­ti­ve and in­scru­ta­ble boun­da­ry po­li­cing of Ac­tor-Net­work Theo­ry (ANT). The ar­ti­cle star­ted as a cri­tique of ANT – or more pre­cise­ly, a cri­tique of what ANT theo­rists were not do­ing. It de­ve­l­o­ped over time into a more ge­ne­ral at­tempt to think through net­works and in­fra­struc­tu­re – con­cepts that sha­re the pro­blem of being both ana­ly­ti­cal tools and very cle­ar­ly ma­te­ri­al things in the world. In­de­ed, the pie­ce was sub­stan­ti­al­ly writ­ten be­fo­re, or just about the same time as the ex­plo­si­on of ‘so­ci­al me­dia’ – a time just be­fo­re we star­ted to talk ea­si­ly and ever­yw­he­re about so­ci­al net­works and so­ci­al graphs. With hind­sight, I am even less sure that an­yo­ne to­day, aca­de­mic or other­wi­se, knows qui­te what they mean when they use the word network, or the word social, much less social network. So I am still against net­works.

I like to think that Against Networks re­pres­ents my own at­tempt, howe­ver awk­ward and un­fa­mi­li­ar, not to cri­tique but to com­bi­ne Ac­tor-Net­work Theo­ry with Ger­man Me­dia Theo­ry. The­se are two tra­di­ti­ons that now seem to more pas­sio­na­te­ly traf­fic with each other than they did ten ye­ars ago, when I star­ted wri­ting this ar­ti­cle. That re­la­ti­ons­hip is now pro­du­cing its first (and lets hope not ste­ri­le) hy­brids.1E.g. Tris­tan Thiel­mann and Er­hard Schütt­pelz (eds.), Akteur-Medien-Theorie, Bie­le­feld, Tran­script, 2013; Lo­renz En­gell and Bern­hard Sie­gert (eds.), ANT und die Medien, Zeitschrift für Medien- und Kulturforschung, 4(2), 2013. I think it is safe to say that, whi­le the­re has al­ways been an awa­ren­ess by one of the other, they have ra­re­ly ta­ken each other se­rious­ly un­til re­cent­ly. This no doubt also has so­me­thing to do with the mys­ti­cal mid­wi­fe­ry of Spe­cu­la­ti­ve Rea­lism and New Ma­te­ria­lism – but I am no clo­se ob­ser­ver of such trends.

My own for­ma­ti­on as a Sci­ence Stu­dies scho­lar in the US was as much stee­ped in the Ger­man Me­dia Theo­ry as in ANT – but that com­bi­na­ti­on was most­ly of my own ma­king. Rea­ding Kitt­ler in the US in the 1990s was cut­ting edge (if not just ob­scu­re) only in Li­te­ra­tu­re de­part­ments – al­most no one in his­to­ry or an­thro­po­lo­gy would have known or ca­red much about it. But as so­meo­ne with a foot in at least three di­sci­pli­nes at any one time, I have ne­ver been much for flag-wa­ving or edge-cut­ting (too many ed­ges, not en­ough loyal­ty), and so this pie­ce pro­bab­ly reads to peop­le more like theorie naïf than it does theo­ry pro­per. But the ques­ti­on of how to think about the In­ter­net was, for me, trap­ped bet­ween the in­junc­tions of the Kitt­le­ri­tes (“Me­dia de­ter­mi­ne our si­tua­ti­on”; “the­re is no soft­ware” etc.) and tho­se top­ping the AN­T­hill (“as­so­cia­ti­ons not so­cie­ty”; “ob­jects too have agen­cy” etc.). So at one and the same time, the pie­ce in­ten­ded to “open the brown box” as a cri­tique of ANT (net­works are what nee­ds to be ex­plai­ned, not the thing that does the ex­plai­ning), and an ap­p­li­ca­ti­on of Kulturtechnik to the de­sign and or­ga­niza­t­i­on of the In­ter­net.2Cp. Bern­hard Sie­gert, “Cul­tu­ral Tech­ni­ques: Or the End of the In­tel­lec­tu­al Post­war Era in Ger­man Me­dia Theo­ry”, Theory, Culture & Society, 30(6), 2013, pp. 48–65.

That is the theo­ry, per­haps; but the most strai­ght­for­ward way to read this pie­ce is as a pre­quel to or a draft of ide­as in Two Bits: The Cultural Significance of Free Software. I wan­ted that book to pro­vi­de a theo­ry that ma­kes sen­se of the In­ter­net – or at least some core as­pects of the struc­tu­re of the In­ter­net, if not its uses – much of the book dwells on the in­tert­wi­ned his­to­ry and de­ve­lop­ment of both the In­ter­net and Free Soft­ware. (Asi­de: the ori­gi­nal sub­tit­le was The Cultural Significance of Free Software and the Internet – the co­pu­la suc­cum­bed to press mar­ke­ting). For in­stan­ce, if pro­to­cols are im­portant, and if TCP/​IP is the most im­portant of them, then that de­ve­lop­ment cle­ar­ly sha­red far more with the kinds of prac­tices that oc­cu­p­ied free soft­ware pro­gramm­ers than it ever did with the kinds of things that oc­cu­p­ied pro­fes­sio­nal en­gi­neers, stan­dards bo­dies in Ge­ne­va, or big com­pu­ter com­pa­nies like IBM. Ra­ther, it is a Cin­de­rel­la sto­ry with the fai­ry god­mo­ther play­ed by Ri­chard Stall­man. I wan­ted to ex­plain how that his­to­ry was cen­tral to what the In­ter­net was be­co­m­ing. To­day, of cour­se, the In­ter­net is be­co­m­ing so­me­thing else – less Cin­de­rel­la and more God­zil­la, but that is, as we say, ano­ther sto­ry.

It was, the­re­fo­re, na­tu­ral to ask of the In­ter­net: “is it a net­work or is it an in­fra­struc­tu­re or is it so­me­thing else?” Through the 1990s and ear­ly 2000s, “theo­ries” of the In­ter­net in­clu­ded thin, feel-good works by peop­le like Howard Rhein­gold and Pier­re Lévy fo­cu­sed on a va­gue con­cept of com­mu­ni­ty. And in so­cio­lo­gy pro­per, Ma­nu­el Ca­s­tells’ enor­mous, me­an­de­ring books pro­vi­ded a per­fect ex­em­plar of the kind of thing ANT ha­tes – ex­plai­ning ever­y­thing with the con­cept of ‘so­cie­ty’ ra­ther than re­co­gni­zing that “net­work so­cie­ty” is pre­cise­ly what nee­ds to be ex­plai­ned. But the­re were many others try­ing to make bet­ter sen­se of the In­ter­net as a pro­blem and a pheno­me­non. Among them were Paul Ed­wards, who was con­cer­ned with the con­cept of in­fra­struc­tu­re, again, both as an ana­ly­ti­cal tool and as so­me­thing of which the In­ter­net was an in­stan­ce.3Cp. Paul N. Ed­wards, “In­fra­struc­tu­re and Mo­der­ni­ty: Force, Time, and So­ci­al Or­ga­niza­t­i­on in the His­to­ry of So­cio­tech­ni­cal Sys­tems”, in: Tho­mas J. Misa et al. (eds.), Modernity and Technology, Cam­bridge MA, MIT Press, 2003, pp. 185–225. And Alex Gal­lo­way pro­vo­ca­tive­ly, but not per­sis­tent­ly, fo­cu­sed on “pro­to­col” as the ba­sis of an un­der­stan­ding of the In­ter­net.4Cp. Alex­an­der R. Gal­lo­way, Protocol: How Control Exists after Decentralization, Cam­bridge MA, MIT Press, 2004. Still others fo­cu­sed on is­su­es of go­ver­nan­ce and stan­dar­di­za­t­i­on as the cen­tral ob­ject of ana­ly­sis.5Cp. Mil­ton Mu­el­ler, Ruling the Root: Internet Governance and the Taming of Cyberspace, Cam­bridge MA, 2002; Lau­ra DeNar­dis, Protocol Politics: The Globalization of Internet Governance, Cam­bridge MA, MIT Press, 2009; An­d­rew L. Rus­sell, Open Standards and the Digital Age, Cam­bridge MA, Cam­bridge Uni­ver­si­ty Press, 2014. But over­all, the­re have been few at­tempts in eit­her ANT or Ger­man Me­dia Theo­ry to ‘theo­ri­ze’ the In­ter­net’s sin­gu­la­ri­ty and si­gni­fi­can­ce.

For me, the ex­plo­ra­ti­on of the con­cepts of net­works and in­fra­struc­tu­res in Against Networks was in­ten­ded to lead so­mew­he­re: I wan­ted to un­der­stand when and how networks – ge­nea­lo­gi­cal­ly spea­king – as­cen­ded to the sta­tus they had in ANT. I wan­ted to ar­ti­cu­la­te whe­ther and how ANT’s network had any­thing to do with the de­ba­tes and ad­van­ces wi­t­hin en­gi­nee­ring that trans­for­med com­mu­ni­ca­ti­on net­works from a prac­tical en­gi­nee­ring pro­blem of power grids and te­le­pho­ne sys­tems into theo­ries of se­ven-lay­er stacks and pa­cket-swit­ched/​cir­cuit-swit­ched dis­tinc­tions, and “de­sign prin­ci­ples” of “end to end” neu­tra­li­ty.6Be­fo­re the­re was neu­tra­li­ty the­re was stu­pi­di­ty: Cp. Da­vid Isen­berg, “Rise of the Stu­pid Net­work”, Computer Telephony 5(8), 1997, pp. 16–26. Si­mi­lar­ly, the ques­ti­on of infrastructure has both an ana­ly­ti­cal ori­gin and a set of prac­tical en­gi­nee­ring pro­blems that re­veal not just a me­dia-spe­ci­fic set of ques­ti­ons, but pro­per­ly po­li­ti­cal ones as well.7On this ap­proach to in­fra­struc­tu­re, cp. Ste­phen Col­lier and An­d­rew La­koff, “The Vul­nera­bi­li­ty of Vi­tal Sys­tems: How ‘Cri­ti­cal In­fra­struc­tu­re’ Be­ca­me a Se­cu­ri­ty Pro­blem”, in My­ri­am Anna Dunn and Kris­ti­an Søby Kris­ten­sen (eds.), Securing ‘the Homeland’: Critical Infrastructure, Risk and (In)Security, Lon­don, New York NY, 2008, pp. 40–62; Col­lier and La­koff, ...continue

As a re­sult, Against Networks is an at­tempt to ‘do theo­ry’ in the way that my col­le­agues in me­dia theo­ry, in Bri­tish so­cio­lo­gy and some in Ac­tor-Net­work Theo­ry com­mon­ly do. I think I fai­led at this, though I hope the­re is no shame in that. What I nee­ded for the work of wri­ting Two Bits, was not a theo­ry of the In­ter­net, but a more use­ful con­cept that might cut through the si­mul­ta­neous­ly media-historical ques­ti­ons and the lar­ge­ly a-his­to­ri­cal ANT-in­spi­red me­thod. And that was how the con­cept of “re­cur­si­ve pu­blics” emer­ged. I wan­ted a way to cap­tu­re what made the me­dia-spe­ci­fic cha­rac­te­ris­tics of the In­ter­net so si­gni­fi­cant at the same time as try­ing to ex­plain the exis­tence of so­me­thing (pu­blics or a pu­blic sphe­re) ra­ther than using that con­cept to ex­plain so­me­thing. What is more, I wan­ted a way to si­gnal the dynamic pro­ces­ses ta­king place – the re­cent past and near fu­ture – of the In­ter­net, and for that I nee­ded so­me­thing other than a pu­ta­ti­ve uni­ver­sal like infrastructure or network, and ra­ther a de­scrip­ti­on of how a set of con­cre­te prac­tices were mo­du­la­ted, com­bi­ned and re­too­led du­ring a cru­ci­al pe­ri­od of the de­ve­lop­ment of the In­ter­net (1970-1990). So Against Networks now looks more like a mis­sing sce­ne from a mo­vie – one in which a nar­ra­tor tri­es awk­ward­ly to ex­plain what is go­ing on in the sto­ry but only com­pli­ca­tes it by do­ing so. Am I still against net­works? In­de­ed. Ever more so: I am up against them all the time, and they are still the things that need ex­plai­ning. Every day I see a new re­port, blog, ar­ti­cle, sto­ry dis­cus­sing how the ma­gic data of Twit­ter al­lows us to ‘see so­cie­ty’ or ana­ly­ze a ‘so­ci­al’ net­work. The terms are un­li­kely to il­lu­mi­na­te any­thing be­cau­se they are pro­jec­tions of va­rious di­sci­pli­na­ry com­mit­ments or sal­ves to com­pre­hen­si­bi­li­ty. But so­ci­al net­works are un­doub­ted­ly both Kulturtechnik and tools for ma­king and brea­king as­so­cia­ti­ons; ever­yo­ne seems to be stu­dy­ing them, but few peop­le seem in­te­rested in explaining them any­mo­re… or yet.

   [ + ]

1. E.g. Tris­tan Thiel­mann and Er­hard Schütt­pelz (eds.), Akteur-Medien-Theorie, Bie­le­feld, Tran­script, 2013; Lo­renz En­gell and Bern­hard Sie­gert (eds.), ANT und die Medien, Zeitschrift für Medien- und Kulturforschung, 4(2), 2013.
2. Cp. Bern­hard Sie­gert, “Cul­tu­ral Tech­ni­ques: Or the End of the In­tel­lec­tu­al Post­war Era in Ger­man Me­dia Theo­ry”, Theory, Culture & Society, 30(6), 2013, pp. 48–65.
3. Cp. Paul N. Ed­wards, “In­fra­struc­tu­re and Mo­der­ni­ty: Force, Time, and So­ci­al Or­ga­niza­t­i­on in the His­to­ry of So­cio­tech­ni­cal Sys­tems”, in: Tho­mas J. Misa et al. (eds.), Modernity and Technology, Cam­bridge MA, MIT Press, 2003, pp. 185–225.
4. Cp. Alex­an­der R. Gal­lo­way, Protocol: How Control Exists after Decentralization, Cam­bridge MA, MIT Press, 2004.
5. Cp. Mil­ton Mu­el­ler, Ruling the Root: Internet Governance and the Taming of Cyberspace, Cam­bridge MA, 2002; Lau­ra DeNar­dis, Protocol Politics: The Globalization of Internet Governance, Cam­bridge MA, MIT Press, 2009; An­d­rew L. Rus­sell, Open Standards and the Digital Age, Cam­bridge MA, Cam­bridge Uni­ver­si­ty Press, 2014.
6. Be­fo­re the­re was neu­tra­li­ty the­re was stu­pi­di­ty: Cp. Da­vid Isen­berg, “Rise of the Stu­pid Net­work”, Computer Telephony 5(8), 1997, pp. 16–26.
7. On this ap­proach to in­fra­struc­tu­re, cp. Ste­phen Col­lier and An­d­rew La­koff, “The Vul­nera­bi­li­ty of Vi­tal Sys­tems: How ‘Cri­ti­cal In­fra­struc­tu­re’ Be­ca­me a Se­cu­ri­ty Pro­blem”, in My­ri­am Anna Dunn and Kris­ti­an Søby Kris­ten­sen (eds.), Securing ‘the Homeland’: Critical Infrastructure, Risk and (In)Security, Lon­don, New York NY, 2008, pp. 40–62; Col­lier and La­koff, “Dis­tri­bu­ted Pre­pa­red­ness: The Spa­ti­al Lo­gic of Do­mestic Se­cu­ri­ty in the United Sta­tes”, Environment and Planning D, Society and Space, 26(1), 2008: 7–28.
Christopher M. Kelty

Christopher M. Kelty is an as­so­cia­te pro­fes­sor at the University of California, Los Angeles. He has a joint ap­point­ment in the Institute for Society and Genetics, the de­part­ment of Information Studies and the Department of Anthropology. His re­se­arch fo­cu­ses on the cul­tu­ral si­gni­fi­can­ce of in­for­ma­ti­on tech­no­lo­gy, es­pe­cial­ly in sci­ence and en­gi­nee­ring. He is the aut­hor most re­cent­ly of Two Bits: The Cultural Significance of Free Software (Duke Uni­ver­si­ty Press, 2008), as well as nu­merous ar­ti­cles on open sour­ce and free soft­ware, in­clu­ding its im­pact on edu­ca­ti­on, na­no­tech­no­lo­gy, the life sci­en­ces, and is­su­es of peer re­view and re­se­arch pro­cess in the sci­en­ces and in the hu­ma­nities. He is trai­ned in sci­ence stu­dies (his­to­ry and an­thro­po­lo­gy) and has also writ­ten about me­tho­do­lo­gi­cal is­su­es fa­c­ing an­thro­po­lo­gy to­day.