The history of the Internet, the logical structure which links millions of computers worldwide, is both brief and recent. As a result, few historians have taken the Internet as their field of study. This is unfortunate, because the rapid and uncontrolled spread of linked computer networks across the globe has already had profound social and political effects. The Internet's history, although short, has exhibited a blistering pace of growth and change.
The Internet connected networks, not only of computers, but of people. Where there are groups of people there are politics. One of the concerns of politics is control. I was attracted to study the Internet because I had heard that it was uncontrollable, and I was sceptical. I found that the available literature on computer networks broadly divided into two groups: those who bubbled and enthused about the liberating effects of the Information Age, and those who gloomily pondered a future of global electronic control and regulation. Computer networks can go either way. The Internet, which is a network of networks, therefore encompasses both visions.
My thesis examines the historical evolution of the logical, social and (to a lesser extent) physical structures of the Internet, and the connections between these structures. I argue that the essence of the Internet is an American computer-to- computer 'protocol' called TCP/IP. I examine how this protocol affected the way people interacted with each other on the network and how, as the Internet expanded beyond the expectations of its designers, the decentralising aspects of this new medium affected social organisations and institutions and created a unique 'net' culture. I show that the 'democratic' Internet Protocol, combined with very rapid growth of worldwide networks, made Internet 'society' uncontrollable by and in some cases even a threat to the sovereignty of national governments.
My thesis is divided into four chapters. In the first chapter, 'Protokollon', I explore the development of the Internet Protocol (IP), the anti-hierarchical building- block of the Internet, and the diplomatic resonances of the name 'protocol'. In the second half of the chapter I discuss the Protocol Wars, an international standards rivalry between the United States and Europe. The Internet Protocol and its rival, called 'Open Systems Interconnection' differed in technical ways which reflected the cultural and political influences of their creators. But TCP/IP's non-hierarchical system made it more popular than the European protocol.
In Chapter 2, 'Bye-Bye, Backbone', I examine the issue of control of the Internet, and how it became impossible (or at least highly unlikely) in the 1980s. In the early eighties, the network was small and vulnerable to control by various groups. A group of computer operators called the Backbone Cabal attempted to control the flow of news and electronic mail in one network; they succeeded for a few years, but the growth of the TCP/IP network allowed users to "sidestep" the Cabal's computers. During this period, the physical infrastructure of the Internet - computers and phone lines - became cheaper and more distributed due to a combination of rapidly falling computer prices and the breakup of United States telecommunication monopolist AT&T. Cheaper infrastructure meant that the Internet grew fast, especially when aided by the National Science Foundation Network in 1986, which provided fast Internet access across the United States.
Chapter 3, 'The (almost) All-American Network' looks at the United States' historical influences on the Internet Protocol and on Internet 'culture'. There are similarities between the Internet's unregulated communications medium and American ideals of democracy, but these similarities were not planned by TCP/IP's designers. I examine the Internet's development using two theories of contemporary American history, the consensus model and the organisational synthesis. However, I find both lacking in explanatory power because neither deals adequately with the international context of the Internet.
In chapter 4, "IP on Everything" I look at the Internet in the 1990s, when it had grown beyond the borders of the United States. The Internet and Internet culture had transcended the 1980s problems of internal backbone politics and external competition, and connected people from almost every nation into a 'global civil society'. Internationalisation of the Internet brought its own problems. One was language; the Internet imposed English on its users. The other problem was the anti-hierarchical structure of TCP/IP itself, which increased the power of non-governmental organisations at the expense of traditional international actors such as governments and transnational corporations, creating a new framework for international relations.
As I mentioned earlier, few historians have tackled the problems of the Internet although others are starting to take an interest. None have examined the Internet from a structural-organisational point of view, as I do. Peter Salus' Casting the Net provides an excellent overview of the development of the Internet. Katie Hafner and Matthew Lyon's recent Where Wizards Stay Up Late is also useful, although it deals with the Internet's precursor, ARPANET. Apart from delving into communication theory and theories of contemporary United Sates history, I examined discussions of computer networking (very useful for the Protocol Wars). I have also legitimately drawn on various sources to be found on the Internet itself, ranging from technical documents created by the original protocol designers to news-posts to electronic mail. I have also had access via the Internet to several academic papers, including Henry Hardy's 'The History of the Net' , which contained very useful information on the Backbone Cabal.
Jargon and acronyms related to the Internet or little-known United States government departments appear frequently in this thesis. I have tried to keep their use to a minimum. An Appendix contains a glossary of these terms and they are explained when they first appear in the text.
Note for WWW version: footnotes in the chapters are generally not hotlinked at this time (26-Oct-1996). However, I have hotlinked as many electronic sources as possible in the Bibliography. Sorry for the inconvenience to surfers and grazers. - P.M.