November, 2015

Against Water – A Reply to Christopher M. Kelty

“One thing about which fish know absolutely nothing is water, since they have no anti-environment which would enable them to perceive the element they live in.”1Marshall McLuhan, War and Peace in the Global Village, New York, Bantam, 1968, p. 175.

Ha­ving Mar­shall McLu­han’s fa­mous quo­te in mind, it may seem odd that the first is­sue of a new web jour­nal for di­gi­tal cul­tu­res in­clu­des an ar­ti­cle that ar­gues “against net­works”. Not only is its aut­hor ob­vious­ly awa­re of the ele­ment that sur­rounds him, it seems that this one is dar­ing to achie­ve the im­pos­si­ble by en­cou­ra­ging re­sis­tan­ce against his own me­di­um of com­mu­ni­ca­ti­on. The net­work, howe­ver, does not seem to care much that one of its in­ha­bi­tants ac­tual­ly at­tempts to crea­te an anti-en­vi­ron­ment wi­t­hin its nodes and ed­ges: Wi­thout he­si­ta­ti­on, my brow­ser fet­ched the data that is now being dis­play­ed to me in the re­a­da­ble form of Chris­to­pher Kel­ty’s thought-pro­vo­king text. Kel­ty’s cri­tique of net­works has be­co­me a part of the very thing that he cri­ti­ci­zed in the first place. Is this trans­mis­si­on of a po­ten­ti­al­ly sub­ver­si­ve mes­sa­ge, one might ask, sim­ply an ex­amp­le of the ge­ne­ral in­dif­fe­rence of a net­work towards its con­tent? Or is it a pro­of of the om­ni­po­tence of the one net­work of net­works that will ul­ti­mate­ly as­si­mi­la­te ever­y­thing and ever­yo­ne? Even if Kel­ty de­fi­ant­ly sends a bunch of IP pa­ckets af­ter his first ones uphol­ding the po­si­ti­on that he is “still against net­works”, it seems that re­sis­tan­ce is fu­ti­le.

Of course, the original argument that Kelty made is much more complex than simply ‘being against networks’. His text is a skillful deconstruction of Ac­tor-Net­work Theo­ry’s (ANT) network-concept and an ambitious endeavor to find a more adequate analytical tool to describe and understand concrete contemporary networks like the Internet. It seems, however, that Kelty was caught in a dilemma when he wrote his text in 2004: By introducing his brown box-concept he substituted ANT’s network for yet another network. Hartmut Böhme has described this inescapability of the network episteme in the following way: “The reflexive emancipation from the entanglement with the net in turn generates the net, that we are enmeshed in … To wherever we might flee, the nets will already be there; since escaping means expanding them even further.”2“Die reflexive Emanzipation von der Verstrickung ins Netz erzeugt im selben Akt das Netz, in dem wir uns verstricken … Die Netze sind immer schon da, wohin wir vor ihnen fliehen mögen; denn indem wir vor ihnen fliehen, bauen wir sie weiter aus.” Hartmut Böhme, “Netzwerke. Zur Theorie und Geschichte einer Konstruktion”, in: Böhme et al. (eds.), Netzwerke: eine Kulturtechnik der Moderne, Köln, Böhlau, 2004, pp. 17-36, p. 17. Even if fish be­ca­me awa­re of the exis­tence of wa­ter and star­ted to re­flect on it, this would only mean one thing: more wa­ter!3Note that two years before McLuhan, Michel Foucault made the same point using a similar metaphor: „Marxism exists in nineteenth-century thought like a fish in the water: that is, it is unable to breath anywhere else“, Michel Foucault, The Order of Things. An Archeology of Human Sciences, New York, Random House, 1970 p. 274.

That being said, I think that it is still worth ha­ving a clo­ser look at Kel­ty’s pro­po­sal to con­ti­nue using the net­work-me­ta­phor in or­der “to un­der­stand the In­ter­net its­elf” – if even against La­tour’s own will. Kel­ty ar­gues that by the end of the 20th cen­tu­ry, ANT’s ana­ly­ti­cal tools had lost most of their “rhe­to­ri­cal force”. Sin­ce its heyday in the 1980s, ANT’s re­pre­sen­ta­ti­ves had been using an or­ga­nic ver­si­on of the net­work-me­ta­phor that stres­sed the dy­na­mics of relation and translation. But in the con­text of the In­ter­net’s gro­wing suc­cess they had to con­front a new do­mi­nant un­der­stan­ding of net­works that, in con­trast, high­ligh­ted the no­ti­on of accessibility and immutability. Whi­le for La­tour this was re­a­son en­ough to ab­olish the term net­work al­to­ge­ther, it seems that Kel­ty en­cou­ra­ges us to stick to ANT’s ori­gi­nal net­work con­cept and use its di­ver­ging mea­ning pro­duc­tive­ly. He wants to de­scri­be how the In­ter­net be­ca­me one par­ti­cu­lar kind of net­work. Using the ex­amp­le of the TCP/​IP pro­to­col, Kel­ty re­minds us that its ge­nea­lo­gy can­not be ex­plai­ned as a top-down im­ple­men­ta­ti­on of stan­dards or a choice de­ci­si­on bet­ween cle­ar-cut al­ter­na­ti­ves. To “un­der­stand the In­ter­net its­elf” me­ans to ack­now­ledge the mul­ti­pli­ci­ty and fle­xi­bi­li­ty of “po­li­ti­cal, eco­no­mic and me­tro lo­gi­cal fac­tors” that com­pe­te, com­ple­ment or rein­force each other and thus cont­ri­bu­te to the emer­gence of its struc­tu­re and or­der of va­lues.

By re­a­so­n­ing this way, howe­ver, Kel­ty fails to take ac­count of the fact that the bio­lo­gi­cal and the tech­no­lo­gi­cal net­work-mo­del sha­re a long com­mon his­to­ry of sci­en­ti­fic ex­ami­na­ti­on and tech­no­lo­gi­cal im­ple­men­ta­ti­on. The mo­dern fa­sci­na­ti­on with net­works, which can be tra­ced way back way to the 18th cen­tu­ry,4Sebastian Gießmann, Die Verbundenheit der Dinge. Eine Kulturgeschichte der Netze und Netzwerke, Berlin, kadmos, 2014, p. 17. has al­ways meant to en­ga­ge with both man-made and ‘na­tu­ral’ forms of or­ga­niza­t­i­on and com­mu­ni­ca­ti­on.5Laura Otis, Networking: Communicating with Bodies and Machines in the Nineteenth Century, Ann Arbor, The University of Michigan Press, 2001. And in con­trast to Kel­ty’s ar­gu­ment that “his­to­ri­cal­ly spea­king, tech­ni­cal net­works were ne­ver about trans­la­ti­on, trans­duc­tion or trans­for­ma­ti­on”, the­re are qui­te a few ex­am­ples of tech­ni­cal net­works that were built to mo­del pro­ces­ses of trans­for­ma­ti­on, dis­si­pa­ti­on and self–or­ga­niza­t­i­on. The se­cond wave of cy­ber­ne­tics of the 1960s and 1970s for ex­amp­le, went to gre­at ef­fort to ex­plo­re bio­lo­gi­cal com­pu­ters, nont­ri­vi­al ma­chi­nes and so cal­led ran­dom net­works.6Through the writings of the French biophysicist Henri Atlan both Michel Serres and Bruno Latour were influenced by the theories of Heinz von Foerster, an Austrian cybernetican and protagonist of this Second Wave of Cybernetics. Bruce Clarke, “Heinz von Foerster’s Demons. The Emergence of Second Order-Systems Theory”, in: Clarke and Mark Hansen (ed.), Emergence and Embodiment. New Essays on Second-Order Systems Theory, Durham, ...continue In fact, whi­le it is true that ANT was built on the con­cep­tu­al foun­da­ti­ons of the first wave of cy­ber­ne­tics, the­re are many in­di­ca­ti­ons that its net­work-con­cept was also great­ly in­flu­en­ced by the offspring of this era of Bio­cy­ber­ne­tics: Agent-ba­sed mo­dels, ob­ject-ori­en­ted pro­gramming and ar­ti­fi­ci­al neu­ral net­works for­med ANT’s me­dia ar­cheo­lo­gi­cal con­text.7This argument is made by Claus Pias in: Martina Leeker, “Technik und Wissensgeschichte der ANT im Kontext entfesselter technischer Objekte”, in: http://entfesselt.kaleidoskopien.de/?pid=2&psid=0#2. By ad­op­ting the new net­wor­ked com­pu­ter ter­mi­no­lo­gy and a cer­tain style of thin­king from this emer­ging epis­te­mo­lo­gy of com­pu­ter si­mu­la­ti­on its prot­ago­nists were well equip­ped to de­scri­be clas­si­cal sci­en­ti­fic prac­tices in a lo­cal­ly con­fi­ned set­ting as net­works of trans­la­ti­on and trans­for­ma­ti­on. At­tempts to ap­p­ly ANT’s ter­mi­no­lo­gy to non-sci­en­ti­fic fiel­ds howe­ver have ne­ver been that fruit­ful in com­pa­ri­son.8Florian Hoof, “Ist jetzt alles ‚Netzwerk’? Mediale ‚Schwellen- und Grenzobjekte’”, in: Hoof et al. (eds.), Jenseits des Labors. Transformationen von Wissen zwischen Entstehungs- und Anwendungskontext, Bielefeld, transcript, 2011, pp. 45-62. And in the face of high­ly col­la­bo­ra­ti­ve and in­cre­a­sin­gly opaque di­gi­tal re­se­arch prac­tices in con­tem­pora­ry sci­ence,9Michael Nielsen, Reinventing Discovery: The New Era of Networked Science, Princeton, Princeton University Press, 2011. ANT’s wea­pons seem to be­co­me in­cre­a­sin­gly in­ef­fec­tive even in re­gards to their ori­gi­nal area of in­ves­ti­ga­ti­on.

But what if we re­place the term net­work with so­me­thing else? Can we gain a hint of an epis­te­mo­lo­gi­cal lead by in­tro­du­cing new terms or con­cepts from other theo­ries or me­thods? Loo­king back at his ori­gi­nal ar­gu­ment, Kel­ty in 2014 sug­gests that the in­tro­duc­tion of his brown box-con­cept could be un­ders­tood as an ear­ly at­tempt to com­bi­ne ANT with Me­dia Theo­ry. It is in­de­ed re­mar­ka­ble that when in­tro­du­cing Akteur-Medien-Theorie (AMT), of all the four things La­tour had ques­tio­ned in his fa­mous re­call of ANT10„[T]here are four things that do not work with actor-network theory … the word actor, the word network, the word theory and the hyphen“, Bruno Latour, „On Recalling ANT“, in: John Law and John Hassard (eds.), Actor Network Theory and After, Oxford, UK and Malden MA, USA, pp. 15-25, p. 15., ad­vo­ca­tes of AMT cho­se to re­place the midd­le let­ter.11Tristan Thielmann and Erhard Schüttpelz (eds.), Akteur-Medien-Theorie, Bielefeld, Transcript, 2013. Is this de­no­mi­na­ti­on of the new hy­brid a con­se­quence of – or may­be even a so­lu­ti­on to – the epis­te­mo­lo­gi­cal di­lem­ma men­tio­ned above? Is AMT bet­ter sui­ted to de­scri­be ac­tu­al net­works ra­ther than ANT be­cau­se it avo­ids the pro­ble­ma­tic over­lap­ping of network-as-object and network-as-theory? Well, it seems that we should not grow too op­ti­mis­tic in this re­spect. Ins­tead of being a true hy­brid that mer­ges the best fea­tures from both theo­re­ti­cal schools, on clo­ser in­spec­tion, AMT turns out to be an in­tel­lec­tual­ly in­s­pi­ring ad­van­ce­ment of ANT that ne­ver­the­l­ess con­ti­nues to rely on the net­work-me­ta­phor.12Andrea Seier has argued that while recent attempts of combining ANT and Media Studies do succeed in discussing the use of ANT’s tools to research media beyond established analytical barriers (“Entgrenzung des Medialen“), they fail to integrate established models and concepts from media studies (such as Foucault’s concept of ‘dispositif’), cf. Andrea Seier, “Kollektive, Agenturen, Unmengen: Medienwissenschaftliche Anschlüsse and ...continue In fact the ‘M’ in AMT is slight­ly mis­lea­ding, sin­ce their re­pre­sen­ta­ti­ves usual­ly pre­fer the term ‘me­dia­tor’ in or­der to high­light the as­sump­ti­on that what we com­mon­ly call ‘me­dia’ are in fact the pro­duct of (but can­not be re­du­ced to) so­ci­al, se­mio­tic and na­tu­ral-tech­no­lo­gi­cal forms of agen­cy and de­le­ga­ti­on.13“Ein gewisser Teil der ANT ist immer schon Medientheorie gewesen, und insbesondere der ‘Mittler’ oder ‘médiateur’ markiert den historischen und systematischen Hebelpunkt einer ‘Akteur-Medien-Theorie’.“ Schüttpelz, “Elemente einer Akteur-Medien-Theorie”, in: Thielmann and Schüttpelz, pp. 9-67, p. 18. Thus, AMT might well func­tion as a cor­rec­tive by which we can test es­ta­blis­hed me­dia theo­ries or his­to­ries for dis­sym­me­tries and hi­d­den a prioris. Howe­ver, I doubt that AMT will bring us any clo­ser to an un­der­stan­ding of our on­to­lo­gi­cal and epis­te­mo­lo­gi­cal ent­an­gle­ment in con­tem­pora­ry net­works, sin­ce ana­ly­zing ‘me­dia­tors’ re­qui­res the framing and lo­ca­li­zing of a con­stel­la­ti­on in which both agen­cy and ar­ran­ge­ment (“Agent­schaft und ih­rer An­ord­nung”) must be ta­ken into con­side­ra­ti­on.14Ibid, p. 13. The con­cep­tua­liza­t­i­on of ‘me­dia­tors’ still pre­sup­po­ses a way of thin­king in net­works and that me­ans: more wa­ter!

To sum up, I think that we should con­ti­nue to use net­work con­cepts and theo­ries whe­re ap­pro­pria­te, but we should not for­get that they are one ana­ly­ti­cal tool among many. They can be hel­pful to test, ex­tend or im­pro­ve other exis­ting theo­ries and me­thods. And they still pro­vi­de an es­pe­cial­ly power­ful mo­del to de­scri­be lo­cal con­stel­la­ti­ons of power and clas­si­cal forms of know­ledge pro­duc­tion. But I do not think that we can ex­pect them to bring us clo­ser to being able “to un­der­stand the In­ter­net its­elf”, as Kel­ty sug­gests. Neit­her will they help us to grasp the epis­te­mic com­ple­xi­ties of com­pu­ter si­mu­la­ti­ons or re­tra­ce con­tem­pora­ry forms of mas­si­ve­ly col­la­bo­ra­ti­ve sci­ence.15See for example the Polymath Project described in Timothy Gowers and Michael Nielsen, “Massively Collaborative Mathematics”, in: Science 461 (2009), pp. 879-881. Ana­ly­zing net­works with net­work mo­dels, that draw their for­mal lan­gua­ge and mo­des of re­pre­sen­ta­ti­on from the very same me­dia-ar­cheo­lo­gi­cal con­text, re­mains an epis­te­mo­lo­gi­cal pro­blem. They might enable us to make sen­se out of “the bor­der of time that sur­rounds our pre­sence” but they will not bring us clo­ser to an un­der­stan­ding of our pre­sence its­elf.16Fou­cault, Archeology of Knowledge, Lon­don, Ta­vistock, 1972, p. 130. Thus, it seems that what McLu­han sta­ted for the emer­ging elec­tro­nic age six­ty ye­ars ago still holds true for us to­day:

“No­bo­dy yet knows the lan­gua­ge in­herent in the new tech­no­lo­gi­cal cul­tu­re; we are all deaf-blind mu­tes in terms of the new si­tua­ti­on. Our most im­pres­si­ve words and thoughts be­tray us by re­fer­ring to the pre­vious­ly exis­tent, not the pre­sent.”17Marshall McLuhan, Counterblast 1954, Berlin, transmediale, 2011, n. pag.

   [ + ]

1. Marshall McLuhan, War and Peace in the Global Village, New York, Bantam, 1968, p. 175.
2. “Die reflexive Emanzipation von der Verstrickung ins Netz erzeugt im selben Akt das Netz, in dem wir uns verstricken … Die Netze sind immer schon da, wohin wir vor ihnen fliehen mögen; denn indem wir vor ihnen fliehen, bauen wir sie weiter aus.” Hartmut Böhme, “Netzwerke. Zur Theorie und Geschichte einer Konstruktion”, in: Böhme et al. (eds.), Netzwerke: eine Kulturtechnik der Moderne, Köln, Böhlau, 2004, pp. 17-36, p. 17.
3. Note that two years before McLuhan, Michel Foucault made the same point using a similar metaphor: „Marxism exists in nineteenth-century thought like a fish in the water: that is, it is unable to breath anywhere else“, Michel Foucault, The Order of Things. An Archeology of Human Sciences, New York, Random House, 1970 p. 274.
4. Sebastian Gießmann, Die Verbundenheit der Dinge. Eine Kulturgeschichte der Netze und Netzwerke, Berlin, kadmos, 2014, p. 17.
5. Laura Otis, Networking: Communicating with Bodies and Machines in the Nineteenth Century, Ann Arbor, The University of Michigan Press, 2001.
6. Through the writings of the French biophysicist Henri Atlan both Michel Serres and Bruno Latour were influenced by the theories of Heinz von Foerster, an Austrian cybernetican and protagonist of this Second Wave of Cybernetics. Bruce Clarke, “Heinz von Foerster’s Demons. The Emergence of Second Order-Systems Theory”, in: Clarke and Mark Hansen (ed.), Emergence and Embodiment. New Essays on Second-Order Systems Theory, Durham, London 2009, S. 34-61.
7. This argument is made by Claus Pias in: Martina Leeker, “Technik und Wissensgeschichte der ANT im Kontext entfesselter technischer Objekte”, in: http://entfesselt.kaleidoskopien.de/?pid=2&psid=0#2.
8. Florian Hoof, “Ist jetzt alles ‚Netzwerk’? Mediale ‚Schwellen- und Grenzobjekte’”, in: Hoof et al. (eds.), Jenseits des Labors. Transformationen von Wissen zwischen Entstehungs- und Anwendungskontext, Bielefeld, transcript, 2011, pp. 45-62.
9. Michael Nielsen, Reinventing Discovery: The New Era of Networked Science, Princeton, Princeton University Press, 2011.
10. „[T]here are four things that do not work with actor-network theory … the word actor, the word network, the word theory and the hyphen“, Bruno Latour, „On Recalling ANT“, in: John Law and John Hassard (eds.), Actor Network Theory and After, Oxford, UK and Malden MA, USA, pp. 15-25, p. 15.
11. Tristan Thielmann and Erhard Schüttpelz (eds.), Akteur-Medien-Theorie, Bielefeld, Transcript, 2013.
12. Andrea Seier has argued that while recent attempts of combining ANT and Media Studies do succeed in discussing the use of ANT’s tools to research media beyond established analytical barriers (“Entgrenzung des Medialen“), they fail to integrate established models and concepts from media studies (such as Foucault’s concept of ‘dispositif’), cf. Andrea Seier, “Kollektive, Agenturen, Unmengen: Medienwissenschaftliche Anschlüsse and die Actor-Network-Theory“, in: Zeitschrift für Medienwissenschaft No. 1 (2009), pp. 132-135.
13. “Ein gewisser Teil der ANT ist immer schon Medientheorie gewesen, und insbesondere der ‘Mittler’ oder ‘médiateur’ markiert den historischen und systematischen Hebelpunkt einer ‘Akteur-Medien-Theorie’.“ Schüttpelz, “Elemente einer Akteur-Medien-Theorie”, in: Thielmann and Schüttpelz, pp. 9-67, p. 18.
14. Ibid, p. 13.
15. See for example the Polymath Project described in Timothy Gowers and Michael Nielsen, “Massively Collaborative Mathematics”, in: Science 461 (2009), pp. 879-881.
16. Fou­cault, Archeology of Knowledge, Lon­don, Ta­vistock, 1972, p. 130.
17. Marshall McLuhan, Counterblast 1954, Berlin, transmediale, 2011, n. pag.
Jan Müggenburg

Jan Müggenburg is currently working as a pre-doctoral research assistant at the Institute for Culture and Aesthetics of Digital Media at the Leuphana University in Lueneburg, writing his doctoral thesis on Heinz von Foersters Biological Computer Laboratory. Between 1998 and 2005 he studied media studies, philosophy and British cultural studies at the Ruhr-Universitaet Bochum (Germany) and the Edith Cowan University in Perth (Australia). From 2006 to 2010 Jan was a member of the graduate program ›The Sciences in Historical Context‹ and a pre-doctoral assistant at the Institute for Philosophy at the University of Vienna. Jan was a visiting scholar in the Department of Electrical and Computer Engineering at the University of Illinois in Urbana Champaign, IL and at the Max Planck Institute for the History of Science in Berlin.