Category Archives: Paul Baran

Three Visions of ARPANET

I’ve recently written about Larry Lewin’s 2001 guide Using the Internet to Strengthen Curriculum in regards to it’s argument that educators need to be “taming the web” for their students. Offering an approach to making research on the web palatable by trying to insure against serendipity. Such an idea seems bizarre in retrospect given that, rather than a bug, serendipity is proving to be the web’s most magical feature.

On the train trip back from NYC early Saturday morning I was reading the first chapter of Lewin’s book, “Why the Internet?” One particular bit about the history of the internet struck me:

The internet uses technology that has been around for over thirty years. Established in the 1960s, ARPANET was a computer network that functioned to keep U.S. military installations in constant communication. College and unversity researchers then started using the network to send data over telephone lines from campus computer to campus computer, and from there it grew. (Lewin 2)

Obviously this is an extremely condensed factoidal history, but it piqued my interest because I think the idea that “ARPANET was a computer network that functioned to keep U.S. military installations in constant communications” is an origin story few folks seem to question. It seems totally plausible that such a communications revolution would be spurred on by the Cold War rivalry for global supremacy—it was certainly true for the Space Race.  According to this version of history, ARPANET was a military creation that was then taken up by universities, commercial interests, etc. In fact, Lewin goes on to frame this chronology in the next sentence “College and university researchers then started using the network to send data…” This pervasive, almost commonplace, history that’s often repeated without question is quite different from the one told in Katie Hafner’s 1998 book Where the Wizards Stay Up Late.

This is not necessarily a knock on Lewin. He’s a teacher explaining how to integrate the web into curriculum for other educators. He’s not pretending his guide is a history of the web. In fact, it would be interesting to study the guides of the late ’90s and early ’00s to get a sense of how often this version of the internet’s beginnings gets reproduced. Anyway, the history of the internet is far more varied and complex than the progression from military to university researchers to corproate interests. In fact, from the very beginning all three sectors were deeply invloved in the development of the internet. Robert (Bob) Taylor, the director of the Department of Defense’s  Advanced Research Projects Agency from 1965-1969, got funding to create ARPANET, but the project wasn’t necessarily imagined as a military network for constant communciation between designated sites, rather….

Building a network as an end in itself wasn’t Taylor’s principal objective. He was trying to solve a problem he had seen grow worse with each round of funding. Researchers were duplicating, and isolating, costly computer resources. Not only were scientists at each site engaging in more, and more diverse, computer research, but their demands for computer resources were growing faster than Taylor’s budget. Every new project required setting up a new and costly computer operation….beyond cost-cutting, Taylor’s idea revealed something very profound. A machine’s ability to amplify human intellectual power was precisely what Licklider had in mind while writing his paper on human-machine symbiosis six years earlier. (Hafner 43-44)

Interestingly enough, the network was a way to have computers talk to one another and share resources so that the system would be more efficient and meet growing demand. Taylor was also a disciple of J.C.R. Licklider so the notion of computers as a way to harness the resources beyond time-sharing was the logical extension of his work. Rather than being dictated by Cold War bunker politics, this vision was very much inline with Vannever Bush’s post-WWII vision of the Memex in his 1945 essay “As we May Think.” Technology would make possible an entirely new paradigm for compiling, accessing, and sharing intellectual resources that woud transform our culture.

What’s more, according to Hafner, Bob Taylor’s vision of the ARPANET was not the only frame for the possibilities:

In the early 1960s …. two other researchers, Paul Baran and Donald Davies—completely unknown to each other and working continents apart towards different goals —arrived at virtually the same revolutionary idea for a new kind of communications network. The realization of their concepts came to be known as packetswitching. (Hafner 53)

Reading this blew my mind. The fact that Taylor, Baran, and Davies were all working towards the same idea between 1964 and 1965 is wild to me. What’s more, Baran and Davies had fairly distinct visions of how such a network might be used.

Baran was working on how to build communications structures whose surviving components could continue to fucntion as a cohesive entity after other pieces were destroyed. (Hafner 56)

Baran compares the structure of his distributed networks to neural networks, suggesting that when synapses in the brain breakdown the neural network remaps accordingly. He was very much thinking about the network in terms of the military explanantion of maintaining constant communication between military sites. So there is some basis for the Cold War military communciations explanation, but Baran was little more than a consultant for the building of ARPANET. He didn’t work for the U.S. military. What’s more, I’m not sure ARPANET, or the ensuing developements of the itnernet, were used for anything related to robust Military communications throughout the 60s and 70s.

The motivation that led Davies to conceive of a packet-switching network had nothing to do with the military concerns that had driven Baran. Davies simply wanted to create a new public communications network. (66)

Davies, who coined the term “packet-switching,” imagined this network as a way to rethink mail delivery and postal service more generally. Something Baran and Taylor weren’t necessarily even talking about, but with the creation of email in 1971. Davies vision would prove prescient, electronic mail being the most important application for popularizing the internet.

As for the progressions of ARPANET from Military -> Universities -> Businesses. In fact, Bolt, Beranek, and Newman Technologies (BBN)  was the engineering firm in Boston that ARPA turned to for building the network. What’s more, BBN was made up of a number of researchers who moved in and out of MIT—so much so it was called “the third university” in Cambridge. Military research, universities, and businesses were far more fluid and cooperative in creating the internet than the condensed history allows. What’s more, from 1970 through the 1980s universities developed the lion’s share of network protocols that defined the internet. And we haven’t even mentioned AT&T who partnered with all involved parties to provide dedicated phone lines for the experiment.

The details of the development of ARPANET are varied and truly compelling. What’s more, I’m thrilled a group from The Internet Course will be focusing their research for the final project on just this topic. They’ll be looking at the various visions that framed the development of ARPANET, as well as the various military, university, and commerical interests involved. I look forward to sharing out their work, and maybe they’ll make the soundbyte history of ARPANET that much more nuanced :)