The Internet is an inescapable feature of the modern landscape. It would be trite to list the ways that it has become part of our everyday lives, be that in the context of work or leisure. One hardly need reiterate the speed at which it spread from the American academy to virtually every country in the world. One only need look at the sequence of Lawrence Landweber’s[1] maps of countries connected to the Internet to get the impression of its relentless march across the globe (see Appendix 2). However, this impression is misleading, not least because the Internet in and of itself is unable to “march” anywhere. The Internet has not spread across the globe in the same way that a liquid spreads out so as to reach a level. Instead, it has been diffused through a series of interactions and negotiations between individual and institutional actors–mostly local in nature–which have in turn impinged on how exactly the Internet was imported by each country and the subsequent infrastructural shape it took in various states around the world.

Two recurrent themes in the writing of Saskia Sassen deeply inform the analysis of the Internet’s arrival in Israel that is offered below.[2] First is the notion that global forms do not just emerge as if from thin air. Indeed, talking about the global economy, she notes that it is “something that has to be actively implemented, reproduced, serviced, and financed. It requires that a vast array of highly specialized functions be carried out, that infrastructures be secured, that legislative environments be made and kept hospitable” (Sassen, 2000b, p. 217). More generally, her point is that globalization does not just happen, but that it must be made to happen. The same is surely true of the Internet, conceived of both as a global phenomenon in its own right and as a central part of globalization more broadly understood. In this chapter I show in some detail how the Internet was “implemented” in Israel and discuss changes to the legislative environment that its arrival instigated.

The second theme in Sassen’s writing that I draw upon here is the idea that “[s]tudying the global […] entails not only a focus on that which is explicitly global in scale, but also a focus on locally scaled practices and conditions articulated with global dynamics” (Sassen, 2003, p. 3). Elsewhere Sassen expands on this:
Critical here is that processes that do not necessarily scale at the global level as such can be part of globalisation. These processes take place deep inside territories and institutional domains that have largely been constructed in national terms in much, though by no means all, of the world. What makes these processes part of globalisation, even though localised in national, indeed subnational settings, is that they involve transboundary networks and formations connecting or articulating multiple local or ‘national’ processes and actors (Sassen, 2005, p. 527).

This in turn implies a particular research approach, namely:
to study the global not only in terms of that which is explicitly global in scale, but also in terms of practices and institutions that scale at subnational levels. Further, it entails recognizing that many of the globally scaled dynamics, such as the global capital market, actually are partly embedded in subnational sites and move between these differently scaled practices and organisational forms (Sassen, 2005, p. 535).

This is similar to the approach suggested by John Street regarding global popular culture (Street, 1997). When discussing the global diffusion of popular cultural forms, such as films and music, Street urges us to remember that “the key issue is how the product arrived there: what material and institutional interests organized its production, distribution and consumption” (Street, 1997, p. 76). Street is concerned that “the rhetoric of global culture has become detached from the material and institutional conditions that underlie the appearance of globalization” (p. 79). In this chapter, therefore, I return the Internet to the “material and institutional conditions” that underlay its appearance in Israel.[3]

This chapter, then, is focused on the local mechanics of the import of the Internet to Israel. It describes the practices of “subnational sites” such as Israeli universities and the activities of certain of their employees. Moreover, by exposing the local/global interactions and articulations mentioned above, it brings up a number of additional theoretical concerns. First, it engages with the importance attributed to transnational or cosmopolitan cultures in processes of globalization (see, for instance, Featherstone, 2002; Hannerz, 1990, 1996). This raises questions such as: who is a cosmopolitan? Does globalization depend on cosmopolitans? Are the individuals who bore the Internet from the United States to Israel to be seen as cosmopolitan? I argue that the mere existence of transnational connections per se was not enough to get the Internet fully off the ground in the Israeli case; indeed, the activities of transnationally-connected though very locally-oriented actors were clearly limited in their ability to institutionalize the Internet in the Israeli setting.

Second, the chapter contributes to discussions of the place of the state in globalization. While not directly confronting the question whether globalization undermines the state or reinforces it (see, for instance, Guillen, 2001; Hobson & Ramesh, 2002; Holton, 1998; Meyer, 2000; O’Riain, 2000; Weiss, 2005), I suggest that the state is an extremely significant unit of analysis in relation to the global spread of the Internet. Thus, while it may seem counter-intuitive, I adopt a somewhat Westphalian view of the Internet. This is at least partly because the Internet is organized around countries in important ways, as suggested by the allocation of Country Code Top Level Domains (.fr for France, .sy for Syria, and so on). In this regard, the Internet might be compared to the postal or telephone systems: each country has their own internal system, but this system interfaces with other countries’ systems. In the field of telephony, for instance, the end user experiences this state-based quality through international dialing codes, with each country having its own unique code. In other words, there is a prima faciae case for supposing that the state should constitute a significant unit of analysis. As I show below, in the particular context of the Internet in Israel, the state played a key role in affecting the speed at which various processes advanced.

Finally, the chapter is driven by the perception that analyses of the diffusion of the Internet have ignored a crucial aspect of the phenomenon, namely, the micro-mechanics of that diffusion. Focusing on the way that the Internet actually reached Israel–rather than assuming that it would inevitably arrive–leads us to important theoretical insights regarding the role of locals in processes of globalization. The work of Sassen is very useful in this regard, as she constantly shifts from the local to the global while keeping her analytic focus on both simultaneously. This is because, as mentioned in the Introduction, for her, “the global–whether an institution, a process, a discursive practice, or an imaginary–simultaneously transcends the exclusive framing of national states, yet partly inhabits, and gets constituted inside, the national” (Sassen, 2007b, p. 1). This tension between the local and global provides the fulcrum of this chapter, and indeed for the other chapters in this dissertation.

This chapter deals with the diffusion of the Internet to Israel, focusing on two particular moments in that process. The first is the initial connections made to networks overseas, and the second is the licensing of commercial Internet Service Providers (ISPs). After presenting a survey of mainstream research into the diffusion of the Internet by way of background, and critiquing that research on the grounds that it rests on technologically determinist foundations, I describe the diffusion of the Internet to Israel and suggest an original way of analyzing this process.

1.2 Background: Research into the Diffusion of the Internet

Research into the diffusion of the Internet can be divided into three major lines of study: (1) The invention and the diffusion of the Internet in the US; (2) Quantitative investigations of the variables held to predict the speed of Internet uptake in countries outside the US; (3) Qualitative studies of political, cultural, and economic factors held to predict the speed of Internet uptake in countries outside the US. I shall briefly survey these studies, before exposing their theoretical assumptions and critiquing them, and offering an alternative way to look at the diffusion of the Internet. .

1.3 The invention and the diffusion of the Internet in the US

A number of books, academic and popular, that document the history of the Internet in the United States have been published. Rosenzweig (1998) critically reviewed a few of those and categorized them according to the type of historical account they offer.[4] For instance, he characterizes the bookWhere Wizards Stay up Late (Hafner & Lyon, 1996) as a “great men” account. The heroes of the book are all amiable and extremely bright men working together to find the best way of sharing computer resources. As Rosenzweig describes it, it is “a tale of adventurous young men motivated by technical curiosity and largely unaffected by larger ideological currents or even narrower motives of self-advancement or economic enrichment” (Rosenzweig, 1998, p. 1534). Hafner and Lyon begin their history by reconstructing a meeting in which computer scientist Bob Taylor, who wanted to get his three computers to talk to each other, asked for some funding and was immediately given a million dollars. The story is thus set in motion, and is told with no shortage of drama and narrative tension. As we know, of course, the heroes of the story overcome the various obstacles in their path, while always keeping up their good humor, and the Internet is invented.

On the other hand, Norberg and O’Neill’s (1996) book on the Information Processing Techniques Office in the American Department of Defense (DoD) is much more focused on bureaucracy, and highlights the DoD context of all the work on the Internet, which was ultimately intended to solve military problems, and especially the problem of how to prevent the collapse of the entire command and control system in the event of a nuclear attack on a major American city. Even if not all of the research referred directly to post-nuclear war scenarios, there certainly was a definite interest in computer communications in warfare, using computers to reach rational decisions, and so on. This raised the problem of how to get these networks to talk to each other, which brought about the development of TCP/IP, still the Internet’s protocol today, out of military funding and necessity. Also, it is noted, in 1980, the Department of Defense made TCP/IP its standard, effectively ensuring its victory over the rival x.25 European protocols.

Paul Edwards’ (1996) book, The Closed World , is quite the opposite of the previous two. While they lack context, argues Rosenzweig, his is all context. He argues that defense and computers go hand in hand; more specifically, he maintains that the Cold War and computer technology enabled and shaped one another. At the very least, the Internet would have not have enjoyed the funding that enabled its development if the Department of Defense did not think it would help their efforts. At the same time, these computer networks nurtured an image of the world as a closed system subject to technical control, which means you always need better and more computers. The problem with this account is that there was resistance: for instance, scientists at MIT in 1968 explicitly said that the government’s actions in Vietnam made them question its use of scientific knowledge; and at Stanford, six months before that university was linked via computers to UCLA, 8,000 students and faculty demonstrated against militarily-funded research being conducted in the university.

Fourthly, Rosenzweig refers to Michael and Ronda Hauben’s (1997) Netizens. This is a completely different kind of narrative, one that places ordinary users at the center, arguing that it is they who discovered the real use of the Internet and defined its nature. They discuss USENET, invented by two students at Duke in 1979, as an alternative to ARPANET, at that time still restricted in its use. The book’s main point is that messaging and mailing came from below, not from the directives of bureaucrats. Hauben and Hauben also firmly contextualize the idea of networking in the 1960s, which they characterize in terms of democratization, openness, and cooperation, referring to the method of Requests for Comments (RFCs), “a series of notes, started in 1969, about the Internet (originally the ARPANET). The notes discuss many aspects of computer communication, focusing on networking protocols, procedures, programs, and concepts but also including meeting notes, [and] opinion”.[5] This is closely related to the hacker culture written about by Steve Levy in Hackers(1984). In this culture, hackers wanted to bring computing to the masses and use it to make a better world.

Citing similar texts, Jon Guice (1998) performs a similar task to Rosenzweig. He starts with what he calls the “standard perspective” of the history of the Internet, before critiquing it. The standard perspective is quoted from the Encyclopedia Britannica Online:
The Internet had its origin in a U.S. Department of Defense program called ARPANET…, established in 1969 to provide a secure and survivable communications network for organizations engaged in defense-related research. Researchers and academics in other fields began to make use of the network, and at length the National Science Foundation (NSF), which had created a similar and parallel network called NSFNET, took over much of the TCP/IP technology from ARPANET and established a distributed network of networks capable of handling far greater traffic.

He then goes on to categorize historical accounts of the Internet’s development into those that focus on its architects, and those that place more emphasis on its users. “However”, he argues,
the standard accounts of the Internet’s past are mistaken, not because they have gotten their facts wrong […] but because they recognize a limited range of facts. A technological system that has not only covered the globe geographically, but has been shaped in many different ways by countless individuals, groups, organizations, and institutions, requires a wide range of analysis (p. 209).

The wider analysis that he suggests brings to bear four main critical insights from the Science and Technology Studies literature. The first recognizes the fact that technologies develop in a contingent way and not according to an internal logic of their own (Bijker & Law, 1992a; Rip, 1995); the second points to the fact that many individuals and institutions were involved in the technological development of the Internet, and that conflict between them was a latent aspect of its emergence (Elliott, 1988); thirdly, Guice reminds us that “technologies have meanings”, though no one inherent meaning, and that these meanings are socially contingent and may change over time (Muller, 1996); finally, he suggests that “all technologies are network technologies”, referring to the notion that “innovation results from complex, opportunistic interactions among a variety of different entities” (p. 208; see also, Hughes, 1986). These insights form the backbone of the analysis offered in this chapter.

These, then, are just two summaries of the many works written on the history of the Internet within the borders of the United States. These accounts provide the overall context for the remainder of this chapter, which looks at crucial moments in the early history of the Internet in Israel. However, precisely because we are interested here in the diffusion of the Internet to Israel, we must add to this background studies of the Internet in countries outside the United States. It is to these that I now turn.

1.3.1 Quantitative investigations of the variables held to predict the speed of Internet uptake in countries outside the US

Rather than providing historical accounts equivalent to those mentioned above, most studies of the Internet beyond the boundaries of the United States look at the variables that have been put forward to explain the differential rates of Internet diffusion around the world. The following is but a sample of such studies and their findings. However, it is most certainly sufficient to appreciate the kind of scholarship being undertaken in this regard.

To start, per capita income (Beilock & Dimitrova, 2003) and other economic measures have been shown to be strong predictors of Internet uptake, especially when analyzing differences between wealthy countries (Hargittai, 1999). Hargittai further attributes an important role to telecommunications policy among OECD countries, whereby a liberal regime is held to encourage Internet connectivity. However, the suggestion that telecommunications policy is a central variable has spurred much debate among researchers. For instance, Guillen and Suarez (2001) find “little evidence that competition and privatization of telecommunications services matters” (p. 349). Instead, they conclude that “Internet development is a complex phenomenon, shaped not so much by public policy in telecommunications as by income per capita, infrastructure, proficiency in English, and conditions for entrepreneurship” (p. 368). Kiiski and Pohjola (2002) concur, arguing that “competition in telecommunications markets does not seem to exert any independent influence on Internet penetration”, and pointing to GDP per capita and Internet access costs instead. Hawkins and Hawkins (2003), though, adopting the social shaping of technology approach, do think that government policy can have a critical effect in bolstering the influence of increases of GDP and telecoms infrastructure on Internet use. Furthermore, writing about Latin America, Gutierrez and Berg (2000) explain why relatively strict governance is actually required in developing countries in that it offers greater assurances for stability, which is a condition investors clearly look for. Indeed, government policies enforcing flat-rate dialing schemes have been shown to increase access (Hawkins & Hawkins, 2003). Robison and Crenshaw (2002), meanwhile, agree that “Internet capacity is fundamentally predicable using current macrosociological theories of development” (p. 355), but suggest that the relationship is not a linear one, and that Internet capacity is rather “driven by complex interactions that could aptly be termed ‘post-industrialism'”, in particular due to the interaction of education with political openness and the effect of a tertiary labor force on Internet adoption.

Despite their competing explanations for the relative uptake of the Internet in various countries, what these studies have in common is their starting point: they study countries that are already part of the Internet, asking how quickly the Internet grew there. This growth might be measured in terms of the number of surfers, the number of sites in that country’s language, the number of Internet servers or routers, the amount of data going in or out of the country, and so on, but the starting point is always the moment after the Internet has arrived. That is, the export of the Internet from the United States is taken for granted, and only the speed of its growth once it has arrived at other countries is studied. This is a point to which I shall return below.

1.3.2 Qualitative studies of political, cultural, and economic factors held to predict the speed of Internet uptake in countries outside the US

In addition to the quantitative research of the kind cited in the previous section, researchers have also taken an interest in the political, cultural, and economic factors that are held to predict the speed of Internet in countries outside the US. Studies of this type are exemplified by the Global Diffusion of the Internet (GDI) series, “a multi-year project undertaken by The Mosaic Group to measure and analyze the growth of the Internet throughout the world”.[6] The GDI series has produced reports on the Internet in over 30 countries,[7] including Israel (Ein-Dor et al., 1999), all of which measure the condition of the Internet in nation-states according to a unified set of criteria: pervasiveness, geographic dispersion, sectoral absorption, connectivity infrastructure, organizational infrastructure, and sophistication of use (L. Press et al., 1998). All the studies also share a set of “Internet Success Determinants” and “Government Policies of Interest” (see Table 1 below).

Table 1 – Factors and policies influencing Internet success within a nation (adapted from L. Press et al., 1998, p. 23)

Internet Success Determinants Government Policies
Telecommunication infrastructure
Personal computing and software
Financial resources
Human capital
Sectoral demand and awareness
Competitive environment
Markets and choice
Investment policy
National security
Cultural concerns
Social equity

The determinants and policies presented in each report are specific to the country under study. For instance, in the case of Cuba, one of the “determining factors” written about is “Difficulty attracting capital”, which most strongly affects the dimension of “Connectivity Infrastructure” in that it “cannot be improved without capital”.

Similar studies, focusing on developing nations, have been carried out by the International Telecommunication Union (ITU).[8] All the reports submitted by the Mosaic Group and the ITU–some of which have been published in academic journals, some of which have been published in-house, and some of which have been published only on the Internet–share a similar structure. They start with a brief political history of the nation-state under study, and offer some data to characterize it (population, GDP, telephones per capita, and the like). They sometimes offer a very brief history of the Internet in that country (usually just stating the university from which the first connection was made), while detailing the current state of affairs, especially in relation to telecoms policies. They try to account for this state of affairs in various ways, before offering recommendations that will speed up Internet diffusion to that country in the future.

1.3.3 Underlying assumptions and critique

Both quantitative and qualitative studies of Internet diffusion share an underlying technologically determinist assumption, namely that the Internet was bound to spread from the United States to other countries. This is the reason that they take up the story from the moment the Internet reached other places, and do not ask how the Internet got there in the first place. In other words, the very diffusion itself is unquestioned, as is the process of that diffusion. According to one definition, technologically determinist thinking includes “the idea that technology develops as the sole result of an internal dynamic” (Winner, 1999, p. 29). In relation to the diffusion of technologies, this refers to the notion that technologies somehow diffuse themselves, or that their tendency to diffuse is latent and inherent to the technology. Of course, this is to ignore the social aspects of such diffusion; moreover, by ignoring technologies that failed to diffuse, or by explaining their failure in terms of their lack of this inherent quality of “diffusability”, this kind of approach runs the risk of narrow-sightedness on the one hand or tautology on the other.

For research that is based on this assumption, it is reasonable to try and isolate the factors that will make the process faster or slower. Thus quantitative studies carry out factor analysis–some more sophisticated than others–while others point out more qualitative features, as exemplified by the GDI studies. However, they all share a perception of the Internet as an unstoppable tide. Quantitative studies tend to see the speed of the flow as determined by a range of variables, while qualitative studies tend to analyze the dams holding back the tide. However, Kling (1996d) cites Dholakia et al. as arguing that “[a] realistic view of how information technology spreads in a society has to draw upon the institutional matrix in which such technology is created, standardized, justified, deployed, modified, priced, and promoted” (Dholakia, Belk, & Dholakia, 1992). I attempt at least part of this here.

The rhetoric of reports of nation-state case studies is particularly illustrative in this regard. Let us take the GDI project’s study of Bangladesh as an example (L. Press, 1999). The tone is set by the report’s title: Against All Odds, The Internet in Bangladesh, and is strengthened by the final sentence of the introductory paragraph: “There are few nations in which it has been such an uphill battle, but Bangladesh is on the Internet”. According to Press, the report’s author, “[t]he Internet came late to Bangladesh”, a sentence which attributes agency to the Internet. According to this turn of phrase, the Internet comes almost of its own accord to countries; it is not brought there by people or institutions. This is fetishization, or at the very least anthropomorphization (Finlay, 1987), and it is exactly the kind of sentence that sociologically sensitive studies of technology are meant to avoid. In other words, in saying that the Internet “came late to Bangladesh”, the author is emphasizing thelateness of the Internet’s arrival to that country, and not the very fact that it arrived at all. Revealingly, in another article, the same author describes the Internet as “growing like a weed” (L. Press, 1997b)[9]–another metaphor giving agency to the Internet and attributing its diffusion to forces that lie within the Internet alone.

Moreover, by talking about “an uphill battle”, the author is leaving us in no doubt as to his affiliation in this battle. He does not approach the arrival of the Internet in Bangladesh as a dry fact, but rather lets us know that he is pleased about it, and that it is the result of a worthy struggle. A similar tone is adopted in other reports when authors describe a certain pro-Internet decision as “fortunate”, for instance.

The assertion that the Internet came late to Bangladesh also raises the question: why did it come so late? Why did it not come earlier? The answers provided by the report are several, including: the inefficiency of state-owned enterprises; the slow implementation of economic reforms; and the outmoded telephone system–there are only 0.24 telephone lines per 100 people, 95% of which are in the capital city, Dhaka. For all of these reasons, “[t]here are few nations in which it has been such an uphill battle, but Bangladesh is on the Internet”. Not only is it assumed that the Internet will arrive eventually, even if delayed somewhat, but it is implied that as soon as Bangladesh solves these problems, the Internet will diffuse much more quickly through the nation.

This impression is strengthened by the uniform use of a hexagonal diagram to depict the state of the Internet in each country studied. For instance, these are the diagrams for Cuba (Figure 1 (L. Press, 1997a)) and Israel (Figure 2 (Ein-Dor et al., 1999, p. 3)):

Figure 1 – Model of Internet Diffusion Dimensions: Cuba

Figure 2 – Model of Internet Diffusion Dimensions: Israel

The message of these diagrams is clear: they are not merely descriptive illustrative devices that make it easy to compare the state of the Internet across nations; more importantly, they make normative statements and set targets for the future. They also dictate the format of the reports in which they appear by implicitly raising the question, “Why has the Internet in this country not yet reached level 4 in every dimension?”

In other words, there is a technologically determinist bent to both quantitative and qualitative studies of the worldwide diffusion of the Internet for two main reasons. Firstly, both types of study unquestioningly accept the very diffusion itself and set out to explain its varying pace within the context of different countries. Secondly, qualitative studies analyze what they perceive to be barriers to the Internet’s inevitable spread within each country it has arrived at, with the outer rim of the diagrams above representing the inherent potential of the Internet if only those barriers could be removed.

A further assumption which motivates certain studies is that there are benefits to be accrued by countries attached to the Internet. That is, studies can be purely descriptive, aiming to characterize a certain process, or they can be driven by a normative concern about the digital divide. Within this context, researchers stress the importance of finding out which government policies encourage the spread of the Internet so as to recommend their implementation elsewhere. These studies are based on a belief that the Internet will bring social goods to the countries in which it arrives. This belief is quite often shared by leading figures in those countries too (see China’s aggressive Internet adoption policies; or the emphasis placed on information technologies by the Bangladeshi political cadre, as described by Press (1999)). These attitudes can be summed up by the widespread belief that it is crucial for nation-states to integrate themselves in the modern world of telecommunications.

These assumptions were also shared by certain of my interviewees, as expressed in the following exchange with Hank Nussbacher, a key figure in the import of computer networking expertise in the early 1980s:
Question: What if you hadn’t come? Who would’ve done it? […] I mean, I can’t imagine that these, that Israel would never have connected to the Internet, right?
Answer: At some point–. It would, it would’ve eventually happened, but I could send you also, I think Israel’s got the domain name, you know the ‘il’, I think we were the second or the third, the first was US, and I think UK got number two, and Israel was number three in the world. […] So Israel was number three. […] so instead of being number three we would’ve been, number fifty-seven.
[…]
Question: So you think that these things would’ve happened?
Answer: Absolutely. […] The same way that Romania now has […] a network, and it’s running and it’s fine and they have commercial providers and they have Internet cafes, and they have an academic network, so too it would’ve happened in Israel, it just would’ve happened later.

Abe Peled, another man closely associated with the Internet in Israel, albeit at the later stage of commercialization, made very similar comments. He is proud of his part in opening up the Internet to commercial entities in Israel–indeed, he wrote the letter to an acquaintance of his, the manager of the Internet backbone in the US, asking for permission to hook Israel up for commercial usage–but readily acknowledges that if it had not been him it would have been someone else.

For these people, the arrival of the Internet in Israel was inevitable. Even the authors of the Global Diffusion of the Internet report on Israel join in the chorus, writing: “In a sense, the Internet has become inevitable, and it is only a question of what the rate of penetration will be” (Ein-Dor et al., 1999, p. 17), following this up with anecdotal evidence about the diffusion of the Internet into even the most ultra-Orthodox homes and offices. Beyond this, however, the “sense” in which “the Internet has become inevitable” is unfortunately not explored.

These technologically determinist assumptions, appearing in the literature and in my interviews, however, would seem to have been realized, as there is a hardly a country in the world that does not today have some kind of Internet connectivity, and those that do have connectivity have more of it today than they did ten years ago. Therefore, my critique of current studies of the diffusion of the Internet does not pretend to entirely pull the rug out from under their feet. However, it does open up the possibility for a different kind of study of the diffusion of the Internet, one that tries not to assume that it would have arrived in Israel, or indeed anywhere, but rather looks closely at the process by which it did arrive.

1.4 The Diffusion of the Internet and Globalization Studies

Previous studies treat the Internet like water, as if it naturally wishes to reach a uniform level; or like Poly Toynbee’s evocative description of Americanization as “strawberry milkshake” spreading out all over the world (Toynbee, 2000), though without the negative connotations that that image implies. Most researchers of the globalization of culture are nowadays alert to the processes of adaptation and hybridization that cultural forms undergo when transported from one place to another, and agree that its meanings in one place or among one social group will not necessarily be the same as its meanings in another (see, for instance, Arnason, 1990; Garcia Canclini, 1995; Hannerz, 1992, 1996; Nederveen Pieterse, 1995; Robertson, 1992). This has been clearly shown in Liebes and Katz’s (1990) seminal study of cross-cultural readings of Dallas. This has also been the thrust of studies of McDonald’s in countries outside the US, with Watson’s (1997) much-cited anthology providing examples of local adaptations of McDonald’s restaurants in five East-Asian societies, the main point of which is to urge researchers “to pay more attention to the responses of local people before drawing grand conclusions about the impact of transnational corporations” (Yan, 1997, p. 75). Meanwhile, in his study of jazz in the Soviet Union, Starr has shown how the same cultural product has been given contradictory interpretations in the same place at the different times (F. Starr, 1983).

Yet, similarly to studies of the diffusion of the Internet, we can say that these critical works take for granted the spread of the cultural forms that they examine, or at least start their inquiries from the moment of their arrival. Thus, the chapter in Watson’s book on McDonald’s in Beijing opens with the sentence, “On April 23, 1992, the largest McDonald’s restaurant in the world opened in Beijing” (Yan, 1997, p. 39). Apart from a sentence to the effect that McDonald’s HQ in America had had its eye on the Chinese market since the 1980s and that negotiations had been ongoing since then (pp. 66-67), the process by which McDonald’s was brought to Beijing is not analyzed. I claim that this is typical of studies of McDonald’s and other global cultural forms. It is this tendency that this current study seeks to avoid.

Studies of the diffusion of the Internet, then, pick up the story from the moment of the Internet’s arrival in a new locale in order to analyze its speed of diffusion or the barriers to its full penetration; similarly, and on a more general level, analysis of cultural globalization also tends to start from the moment that a cultural form has successfully migrated to a new environment. In contrast, in the remainder of this chapter I look closely at the way the Internet arrived from the United States to Israel. In doing so I am making a methodological statement about how the diffusion of the Internet and other global phenomena might be studied differently, and I am suggesting a new way to engage with theories of globalization and technological diffusion.

1.5 What is being Diffused?

This section looks at the diffusion of two separate, though of course related, innovations. The first is Israel’s connection to overseas networks; the second is the commercialization of the Internet in Israel.

1.5.1 International computer networks

By November 1969, three nodes of the emergent ARAPNET had been connected along the west coast of the United States–in UCLA, Stanford, and the University of California Santa Barbara.[10] A year later, the first cross-country link was established. In 1973 the first international connection was made between ARPANET and the Norwegian NORSAR agency, “an independent research foundation specializing in commercial software solutions and research activities within applied geophysics and seismology”.[11] The University College of London was linked to NORSAR, and thus to ARPANET.[12] Additional European countries were added in subsequent years, though not at first using the TCP/IP protocols that would later become definitional of the Internet.

What were being diffused in this instance are protocols. Protocols are “computer rules that provide uniform specifications so that computer hardware and operating systems can communicate”.[13]For a computer to be able to “talk” to another computer, they must first agree on how to do so–the protocols they use serve this function. Talking about the “institutional conditions for diffusion”, Strang and Meyer (1993) assert that “common conventions or protocols are needed or advantageous at the boundaries where exchange occurs. Handshaking arrangements aid in interpersonal (and electronic) exchange” (p. 488). While they see protocols as aiding diffusion, here we are talking about the diffusion of the protocols themselves, or, put differently, the diffusion of a mechanism of diffusion. Noting that Israel was early to join the abovementioned international computing networks, I ask when Israel first joined them, and which specific mechanisms and processes were involved.

1.5.2 The commercialization of the Internet

The term Internet, as is widely known, refers to a “network of networks”. Originally, these networks were academic, and were mainly organized around the American National Science Foundation’s network, NSFNET, and its European counterpart, the European Academic Research Network (EARN). In order to keep unnecessary data transfer to as low a level as possible, there were Acceptable Use Policies for these networks which prevented their commercial exploitation; for instance, the NSFNET was specifically for “scientific research and other scholarly activities”. However, during the late 1980s the use of these networks was growing so fast that the managers of the NSFNET had to consider ways of relieving the pressure on the network. The path decided upon was that “industry [would begin] to supplant the government in supplying these network services”.[14] This was a process that took five years, during which time NSFNET allowed “research arms of for-profit firms” to use its network services “when engaged in open scholarly communication and research”,[15] and only in 1995 did NSFNET revert to being a purely research network.

As I explain below, the network in Israel underwent a similar process of commercialization as the university-run network overseen by Inter-University Computation Center (IUCC, or MACHBA, to give it its Hebrew acronym) could not and did not want to deal with the ever-increasing volume of traffic and growing number of requests to hook up. In this context, then, what was diffused was the concept of an Internet Service Provider as a commercial entity selling connectivity to the Internet.

1.6 How were they Diffused?

1.6.1 The diffusion of international computer networks

When does the story of international computer networks in Israel begin?[16] The first date in the GDI report’s timeline is August 1984, when MACHBA connected to EARN (Ein-Dor et al., 1999, p. 7). However, before then, as early as 1982, there was a UUCP[17] link to the United States from the Hebrew University’s Computer Science Department. Other universities in Israel were able to use this link by means of regular telephone lines. The main service provided was email, and, like today, files could be attached to them.[18] This link was initiated by Prof. Shmuel Peleg, an Israeli computer scientist who had studied for his PhD at the University of Maryland, receiving his degree in 1989. As Peleg wrote to me by email: “I joined the Hebrew University from the University of Maryland on 1981, and I was scared to be disconnected from the UUCP email network I used there”. So with the help of computing and telephony personnel at the Hebrew University, and his doctoral supervisor at Maryland, and by promising to pay the phone bill himself, Peleg set up a daily dial up connection, which was established in 1982.

However, this connection does not really constitute a network–it was only available to people within the Hebrew University, and certainly there was no network of Israeli universities. Moreover, the phone bill charged by Bezeq, then Israel’s state-owned telecoms monopoly, became prohibitively high, and so the service was discontinued.

Israel’s next network connection was to EARN in 1984. How, then, was that connection made? How did EARN diffuse to Israel? As mentioned, much of the work in bringing EARN to Israel was carried out by Hank Nussbacher. Hank Nussbacher is an American-born Israeli who moved to Israel for ideological Zionist reasons in 1982 aged 24 (he took Israeli citizenship about a year later). He was educated at Baruch College, part of the City University of New York (CUNY), where he majored in Statistics with a minor in Computer Science. According to Nussbacher, at the time Baruch College was providing top technical people to AOL and had staff conducting cutting edge research. Towards the end of his studies he was given three job offers, including two from IBM and Bell Labs to work in database and system administration. The third was from CUNY itself, to work on networking the nineteen colleges dispersed around New York. As Nussbacher explained in interview:[19]
They decided they wanted to build a network between them and have a common operating system on their hardware. And I was tasked with building this entire network. So over the course of one year I would travel, I would take my backpack with the tapes, these were big tapes, travel to each one of these colleges, install a whole new operating system, and connect up communication lines between the college and the central place that was located in Manhattan. And by the end of the first year we had a network of these nineteen colleges, and it worked very well and […] people were able to […] log in one computer to the other and […] send files back and forth.

When the CUNY network was connected with a similar network in Yale University in 1981, BITNET was formed;[20] by the end of 1982 over 70 more academic institutions had joined the network. Nussbacher, then, had firsthand experience of networking in an academic environment, and he brought this experience with him to Israel when he came as a visiting scientist to the Weizmann Institute, a leading research center. Nussbacher said to me:
[I] spent four years at the Weizmann Institute as a visiting scientist, playing around on various things but also bringing the internet to Israel, because I had suddenly, I already had the Internet addiction after being there and like seeing how great it was to be able to talk to people via communications.

In other words, Nussbacher brought with him to Israel from the United States both technological expertise that was very rare in Israel at the time, if not non-existent, and an enthusiasm for what computer networks enabled him to do, namely, “talk to people via communications”. As I shall show below, this combination of know-how and enthusiasm also played an important part in turning the Internet in Israel into a commercial institution.

According to Nussbacher, computer networking had yet to reach Israeli academia, and he found himself trying to push it wherever he could:
[I was] evangelizing the Internet. It wasn’t here, no one understood what it was, you have to go to people, start convincing every single person you know, listen you have to put in money, you know to, to, you have to put in routers, you have to put in lines. It doesn’t come cheaply. […] there wasn’t even a network between the universities themselves. There was nothing. […] and you have to then start going and explaining to people and it’s very hard sometimes.

Two years after he arrived in Israel, Nussbacher oversaw the connection made with the European Academic Research Network. EARN had been established in 1983 and worked in a very similar way to BITNET, to which it was also connected. Indeed, according to Doron Shikmoni, another figure from the very early days of the Israeli Internet, when Israel joined EARN on 13th August 1984, the physical link actually landed in the Weizmann Institute, before being moved to IBM’s research center in Haifa later on.[21] This was the first time that the Israeli academic community was linked to academic institutions outside of Israel.

While not wishing to reduce the history of the Internet in Israel to the activities of one man, it is clear that Nussbacher played a crucial role in the timing of developments in Israel. He imported with him expert knowledge, the deployment of which gave Israeli academics connections with their overseas counterparts much sooner than might otherwise have been the case. Indeed, he has been involved with MACHBA ever since its establishment in 1984, and today is its “Communications Consultant”.[22]

There is another connection competing for the status of first overseas network connection, and that is the link to the American CSNET,[23] made in February 1984, again under the auspices of Prof. Shmuel Peleg. As mentioned, the UUCP connection became too expensive to maintain, so Peleg requested, and received, funding from the Ministry of Science to create a link to the emerging CSNET network in the United States. This link was established in February 1984, making Israel CSNET’s first international node.[24]

Another person who ought to be mentioned in the context of Israel’s adoption of early networking technologies is Doron Shikmoni. Shikmoni is currently running a hi-tech start-up, though his career follows the Israeli Internet very closely. After a romantic relationship with computers as a child (sending punch cards by regular mail from his home in Eilat to a friend in Tel Aviv who would put them in a computer there), he found work at Bar Ilan University in the early 1980s, where he was very involved in hooking the university up to the emerging academic network. He has acted as a consultant for the government, specifically the treasury, on Internet related matters, and sat on two of MK Michael Eitan’s Internet sub-committees, those related to security and Hebrew (see Chapter 6 below for more on the subject of Hebrew on the Internet). Today he is the manager of the .il top-level domain and responsible for the Israeli Internet Exchange (IIX) (of which more in Chapter 4).

However, most significantly for our purposes, he was involved in securing for Israel the .il suffix, an interesting aspect of the early diffusion of the Internet. Country Code Top Level Domains (ccTLDs) are the two letter codes allocated to individual countries (for instance, .fr for France, or .de for Germany). The first two ccTLDs to be allocated were those of the United States–.us–and the United Kingdom–.uk–in 1985. Israel’s .il suffix was the third (also allocated in 1985). While this does not testify directly to the quantity or quality of Israel’s network connections with the rest of the world, the fact that Israel was so quick to claim its suffix does suggest a high level of awareness of trends and developments in the field. I asked Shikmoni how he explained this. He replied, “Because we were fast and pioneering and we were active, and other countries weren’t going in that direction at all”.

Israel was one of the first countries to connect to EARN; it was one of the first countries to be allocated a ccTLD (.il); and in 1995 it would become only the second country outside the US to set up a local chapter of the Internet Society (after Japan). Thus two things here require explanation. The first is the timing of Israel’s connection to EARN and the role of Hank Nussbacher as an immigrant with expert knowledge in bringing that about. The second is Israel’s alacrity in certain Internet matters, with the aforementioned institutional affiliations providing a good indication of that. I shall provide explanations below.

1.6.2 The diffusion of the commercialization of the Internet

We now leap forward a decade, which is the amount of time that passed between the first overseas Internet connections being made between Israel and Europe and the US, and the opening up of the Internet for commercial use. By the time that the Internet was opened up for general public usage in May 1994, a number of actors were already chomping at the bit. These actors were Israel’s Internet Service Providers (ISPs), who were waiting for the opportunity to sell Internet connectivity to the Israeli public, as was already happening in the United States and a number of other countries (L. Press, 1994).[25] This was a process that will also be discussed in a different context in the next chapter. Now, though, we are concerned with the way in which a process begun in America made its way to Israel, and particularly with the people involved.

Starting from 1984, Israel’s network grew constantly, though at first it was limited to academic institutions and their affiliated members. It was run by MACHBA, which after a number of years began to feel that it was not able to perform the huge task it found itself responsible for, as more and more people were finding their way online. This was especially so from 1991, the year in which Tim Berners-Lee and CERN went public with the World Wide Web, introducing hyperlinks for the first time (for Berners-Lee’s own account, see Berners-Lee & Fischetti, 1999). In 1992 the first Israeli site was published (that of the Hebrew University of Jerusalem), and in April 1993 the web browser Mosaic was released, introducing color and images to web pages. (The number of worldwide Internet hosts during the years under discussion can be seen in Figure 3.)

Figure 3 – Internet Hosts

In around 1990, word reached Israel of an American project known as NREN (National Research and Education Network[26]). More precisely, Jules Finkel, then Director of the Computing Department at the Weizmann Institute, an American immigrant to Israel, heard of the NREN project and told Rafi Hoyda about it. Hoyda was then vice Chief Scientist of the Ministry of Communications.[27] Hoyda was already aware of the Internet–in 1977 he had been to the USA on a work trip and had been taken to the offices of BBN, the company that built the hardware components of the very first ARPANET connections. He had also recently come back from a three year job posting in the States, where he had learnt a great deal about working with computers. Hoyda initiated a bureaucratic process whereby a committee made up of representatives from the Ministries of Science and Communications approved the formation of the Israeli Research and Education Network (IREN), and set up another committee to establish requirements among the targeted institutions–schools, colleges, and research wings in industry. This, Hoyda explained, was the first decision made in the direction of opening up the network to organizations outside the universities (see also Rahav, 1998).

And indeed, in a meeting on 6th June, 1991, representatives from three ministries–Industry and Trade, Science, and Communications–asked MACHBA to start connecting commercial companies to the Internet via its infrastructure. However, MACHBA had no interest in becoming what for all intents and purposes would be a nationwide ISP. As Nussbacher made clear in interview, MACHBA had been set up to oversee the creation of a research-based network, and did not want to “dirty its hands” with commercial dealings. According to Nussbacher’s timeline,[28] this is the point at which the idea of licensing commercial ISPs was first floated. Shortly afterwards, the first ISPs started working, though with all their traffic going through MACHBA, until MACHBA was entirely cut out of the commercial aspects of the Internet in June 1997.

In the period between 1992 and 1994–when non-academic organizations could use the Internet, but before full licenses were issued–there was a mechanism for deciding who would and would not be able to get on to the network. This mechanism was a committee in the Ministry of Communications to whom applications for network access was made. The main criterion was that the connection was required for research and development purposes, and although the criterion was properly applied, getting past the committee was not very difficult.[29] Indeed, Shikmoni and others were trying to persuade companies to apply for Internet access. He explained:
At a certain stage, the Ministry of Communications gave permission to connect organizations to the Internet that were dealing with research and development. Intentionally, or unintentionally, it’s not important, they left the definition of research and development a bit vague. Great. Research and development. Excellent. […] Who does research and development? All of a sudden, everyone is doing research and development. […] So the first thing was all the hi-tech companies here. We went to hi-tech companies and started to persuade them to hook up to the Internet. They didn’t really understand what we wanted from them.

Still, though, there was no way for the private individual to attain Internet access from his or her home. This was finally enabled in May 1994, when three companies, Elron, Darcom,[30] and NetManage, were given full Internet licenses by the Ministry of Communications (this was covered in the local press by, among others, Gabizon, 1994).

From MACHBA’s point of view of, then, they had been asked by the state to provide widespread Internet services, but had refused, thus creating the need for alternative sources of connectivity supply. This need was met by ISPs, licenses for which were issued by the Ministry of Communications. However, there are other points of view, and interesting among them are those of a very small group of Israelis who coincidentally came back to live in Israel after many years of living in the US and working in the hi-tech sector. These migrants brought with them to Israel expectations of how communications between computers should be, and found that the reality in Israel failed to live up to those expectations. They also claim a certain degree of influence in bringing the MoC to license their companies to provide Internet services on an entirely commercial basis.

Their importance can be seen if we compare the list of the ISPs given permission to connect research arms of commercial enterprises in 1992, and the ISPs given permission to connect all and sundry in 1994. In 1992, four ISPs provided dial-up services: Actcom, Dataserve, Kav Mancheh, and Bezeq Zahav.[31] Of these companies one is still an active but very minor player in the local ISP industry (Actcom); one has disappeared entirely (Dataserve);[32] and the other two never became fully fledged commercial providers for home usage (Bezeq Zahav and Kav Mancheh, which provided mainly business services, such as providing live stock exchange data). So while there were some companies working at this time, they did not really constitute the foundations of Israel’s future Internet industry. By contrast, of the three companies to receive the first full Internet licenses–Elronet, NetManage and Darcom–two went on to enjoy supremacy in the ISP market for a number of years. Darcom never actually provided services, and Elronet and NetManage merged to become NetVision. All three entities were founded by Israelis who were either immigrants or returning from long periods of time in the US, and it is they who should really be considered Israel’s first commercial ISPs.

Ruth Alon, the founder and former CEO of NetVision, and daughter of Uzia Galil (see below), is one of these migrating Israelis.[33] She moved to California from Israel in 1979, where she worked with her husband, Zvi Alon, in their hi-tech company, NetManage. The company was very successful, largely due to its sales of a version of TCP/IP, the Internet protocols, for Windows, known as WinSock.[34] In 1993, Ruth Alon decided to come back to Israel with her children and run a branch of NetManage from Haifa, the town in which she had grown up and lived before moving to the United States. It was clear to her that the Israeli and American branches of NetManage would need a way of communicating with one another, but at the time there were no commercial ISPs that could connect them to the Internet. So, relates Alon, she decided to set one up. Although a programmer in a senior management position in NetManage, Alon did not know very much about the Internet at the time. However, she had heard of the “Smart Valley” project, an effort to put schools and businesses online in Silicon Valley,[35] which she mentioned in interview as providing some kind of inspiration.

Another figure who regularly brought knowledge to Israel from the US was Uzia Galil, founder in 1962 of what is commonly referred to as Israel’s first hi-tech company, Elron.[36] Galil claims that he was motivated to set up Elron after spending time in the States, during which he became aware of the US hi-tech industry and thought that Israel, a country lacking natural resources and thus dependent on its human capital, needed something similar. One of the very first ISPs in Israel was a subsidiary of Elron, named Elronet, managed by another Israeli migrant, Dr. Abe Peled.

Peled was born in Romania in 1945, and moved to Israel with his family in 1958, where he served in the army and completed his BA at the prestigious Technion Institute of Technology. When he was 26 he moved to the United States, where he went to graduate school, eventually receiving a PhD from Princeton in the emerging field of digital signal processing. In 1974 he started working for IBM, and in 1985 he was made “vice president of systems and software at the Research Division, in charge of, among other things, all the communications-related research”, a very senior position in which he remained in until 1993.[37] In interview,[38] Peled told me that personal reasons made him start thinking about a move back to Israel, where his parents still lived. At the same time, Uzia Galil was looking for a new vice president for Elron, and Peled was given the position. Arriving in Israel in 1993, Peled, like Alon, was used to conducting business through email and spending time on computer networks, but in Israel he was unable to do so. Peled made contact with the CEO of the Ministry of Communications at that time, Shlomo Waxe, to discuss possibilities of opening the Internet for commercial use. At the same time, he commissioned Dr. Beth Erez, a former MoC employee and at the time a freelance technology analyst, to conduct a feasibility study into the possibility of setting up an ISP aimed at business customers.[39] Crucial to the commercialization of the Internet in Israel was getting permission to allow commercial traffic onto the backbone in America. To this end, Peled wrote a letter to a colleague and acquaintance from the committee that was in charge of the backbone, asking for this permission, which was duly granted. In the meantime, Peled was involved in writing the licenses that would shortly be issued to commercial ISPs (see also Chapter 4 for further implications of this involvement). When ISP licenses were eventually issued in 1994, Elronet was the first to receive one, claims Peled, a fact that is related to his close involvement in the licensing process (see Chapter 4 for an analysis of this subject).[40]

1.7 Analysis of these Processes of Diffusion

In the previous section I described two processes in the diffusion of the Internet to Israel. The first was the arrival of the networked connection in 1984 of Israeli academia to its European counterparts and through them to the USA. The second was the establishment of the first commercial Internet Service Providers in Israel. In the following, I discuss similarities and differences between these two processes.

1.7.1 Similarities

In both cases I resist the tendencies to technological determinism that are discernible in the types of study described in Section y’3.2 above. That is, rather than simply assuming the Internet’s arrival in an unproblematic way, I inquire into the quite specific mechanisms of its diffusion.

Of the many reports of the diffusion of the Internet to other countries that I read in writing this chapter, only two offer the kind of analysis that I argue successfully avoids determinism. First, Bellman, Tindimubona, & Arias (1993) describe political and cultural aspects of the introduction of networking technologies to African countries in the early 1990s, referring to matters such as the quality of communication lines and cooperation between NGOs and government. Second, the GDI report on the Internet in Ghana (Foster et al., 2004) is singular in that it emphasizes the role of identifiable individuals in the process of bringing the Internet to that country. The authors note that
TCP/IP connectivity was first provided by a commercial system integrator, Network Computer Systems (NCS). NCS was started as a Digital Equipment Corporation (DEC) integrator by Dr. Nii Quaynor in 1988. Dr. Quaynor attended Dartmouth College and earned a Ph.D in computer science from the State University of New York at Stony Brook. He came back to Ghana on a sabbatical from DEC, where he was a senior software engineering Manager (Foster et al., 2004, p. 6).

In addition, “Dr. Quaynor applied for, and in January 1995 received, permission from Internet Assigned Number Authority (IANA) to use and administer the Ghana Domain name ‘.gh'” (pp. 6-7). They also note that in 1996 a Ghanaian ISP was set up by “three Kenyans who had studied at M.I.T. and Harvard and returned to Kenya to found an Internet service for all of Africa” (p. 8). I find this account of how the Internet arrived in Ghana interesting for a number of reasons. Firstly, as mentioned, it does not simply assume the arrival of the Internet; rather, it shows how certain people brought it to Ghana. Secondly, although it does not theorize the process in this way, it shows how the cultural and social capital of the Internet importers in Ghana was crucial to the arrival of the Internet in that country. Thirdly, it highlights the importance of physical space and people’s movement through it in the diffusion process. I shall now expand on these issues with regard to the Israeli case.

The concept of diffusion through space would seem to be central to the diffusion of the Internet to Israel, and, by extension, to other places as well. In the introductory chapter to his seminal book, Ha”gerstrand (1967 [1953]) talks about spatial diffusion. He says:
Its importance ought to vary with the form of the observed population’s distribution and with the communications media involved. In a society where there are no appreciable time or cost obstacles preventing one individual from coming into contact with any other individual, relations within ‘social space’ cannot be appreciably modified by the constraints of geometrical space. If air travel and the ability to transmit as well as to receive television were within the reach of every individual, then we would approach the conditions of a ‘one point society,’ in which case the spatial interpretation of social phenomena would become quite uninteresting. So far, such conditions do not exist; therefore, spatial analysis has not completed the playing of its role (p. 7).

In other words, as long as there is meaning to space, there is a point in conducting spatial analyses of innovation diffusion.

It has been posited that one of the major characteristics of late modernity is the collapse of space-time (Giddens, 1990, 1991). This is a process that is clearly intensified by technologies of the Internet. However, just as it is true that a pre-condition for the Internet becoming a vehicle for the global distribution of, say, culture and commerce, is the global distribution of the protocols that enable computers to communicate with one another in the first place, I argue that even a technology that collapses time and space, at least in its ideal version, must first travel across space, a process which must necessarily take time. Any concept of travel must necessarily have some physical dimension, some body that is bearing the information–something which is equally true of the data that travels between computers on the Internet. Previous studies of the diffusion of the Internet, however, have quite ignored this dimension, and it is that which creates the impression of technological determinism.

In the previous section I identified a small number of people whose physical movement in space from one place to another had much to do with the diffusion of the Internet to Israel. These people, along with those mentioned by the GDI report cited above (Foster et al., 2004), are what Rogers–the seminal writer on processes of diffusion–would have called cosmopolitans. For him, “cosmopoliteness is the degree to which an individual’s orientation is external to a particular social system” (Rogers, 1962, p. 102). This orientation to a system other than the Israeli one is latent in immigrants (such as Nussbacher or Winer), or may be acquired by nationals sojourning overseas for an extended period of time (such as for graduate studies, as with Prof. Peleg and the Ghanaians mentioned above).

While not wishing to reduce the fact that Israel is connected to the Internet to these individuals, their role in the timing of events is worthy of notice, especially against the background of other studies of the Internet in which it almost magically appears at each new country. In particular, it should be noted that both of the processes under discussion followed a kind of “false start”: in the first instance, Israel had email connections with the USA before its connection to EARN in 1984, but that connection was disconnected for financial reasons; and in the second instance, the commercialization of the Internet in Israel was partly realized in 1992 when four companies were given limited ISP licenses, but fully commercialized in 1994 when three different companies were given full licenses. The involvement of migrants is notable in the second, more successful, phases of these two processes.

In other words, not only is there intrinsic value in identifying the people involved, if only for the sake of a richer historical record, there is more to it than that. The physical movement I noted was from the United States to Israel. I shall comment on both elements of this movement: from the US, andto Israel.

1.7.1.1 Movement from the United States

It is hardly surprising that the movement originated from the United States. This was the country that had “invented” the Internet, and virtually all the knowledge and expertise was located there. This was borne out by my interviewees, whom I asked how they kept themselves up to date. They all said that by far their main source of information was American-based websites and online journals. Not only was the relevant knowledge located in the US, but the necessary hardware came from there too, by which I am referring to the routers and modems that provide the physical base for the Internet (and if the equipment was actually manufactured elsewhere, the companies designing and selling it were certainly American).

1.7.1.2 Movement to Israel

The significance of the movement to Israel lies in the fact that the individuals referred to above are all Jewish, and that their movement to Israel was a kind of homecoming (see Lomsky-Feder & Rapoport, 2001), and not labor migration for the sake of an improved standard of living either for the migrant or his family back home. Even for Nussbacher, who had never previously lived in Israel, his migration to Israel did not make him an “expatriate”, as he immediately joined the majority ethnic group in that country and enjoyed rights and privileges not afforded to non-Jewish immigrants. For Peleg, his time in America was intended for studying, and his movement back to Israel was quite natural. In other words, the diffusion of the Internet to Israel should be located in the context of Jewish ethnicity and movement between Israel, understood as the homeland, and the Jewish diaspora, particularly its American chapter. It would seem that it was this kind of back and forth movement that brought the Internet to Ghana too.

This was a theme stressed by Dov Winer,[41] who has been involved in the Israeli Internet for over 20 years, particularly in aspects related to education.[42] Winer has served as an advisor to the Ministry of Communications and is currently working with the Jewish Agency on projects to do with linking Jewish diasporas to Israel through the Internet. For him the informal networks linking Jewish scientists in the United States with Israel play a key part in explaining the early diffusion of networks from the former to the latter. In general, he noted that a number of the scientists involved in the original ARPANET project were Jewish, though he does not point to any specific consequences of that. He does, however, cite Richard Mandelbaum as an example of a Jewish computer scientist whose ethnicity impacted on the speed of the diffusion of his network to Israel: Mandelbaum was a co-founder in 1985 of the New York State Educational Research Network (NYSERNET), a non-profit Internet Service Provider for government and education in the north-east of the US, and served as its president for a number of years. Mandelbaum is also an Orthodox Jew, and it is this fact that led to Israel being the second country (after Ireland) to be connected to the NYSERNET network. Winer also recalled that the person responsible for communications at the Weizmann Institute was, like Nussbacher, an American immigrant, which he believes is relevant to the fact that the connection to NYSERNET physically landed at the Weizmann Institute. This movement to Israel by hi-tech people is not restricted to the Internet alone. Indeed, Intel’s hugely successful Israeli branch was set up by an Israeli Jew returning from America, as were IBM’s Israeli offices.[43]

In short, space, and the movement between places, played a crucial role in the diffusion of Internet technologies and practices. Knowledge and cultural expectations traveled from the United States to Israel. Nowadays, experts in the field keep up to date with developments by reading information that comes almost exclusively from America. For this reason, hardly any of my ISP interviewees attended industry conferences or events overseas. However, the role of those who did make physical journeys is strongly felt in the diffusion of both initial network connections, and the move to the commercialization of the Internet.

1.7.2 Differences

The main difference between the two processes of diffusion being discussed here is their relative speed. Israel was very quick in claiming its Internet suffix, .il, getting it assigned even before there were such things as Internet sites. Israel was also one of the first countries to connect to EARN. In these regards, Israel was what Rogers would term an “early adopter” (Rogers, 1962). However, in terms of licensing ISPs, while certainly not a laggard, Israel was not so quick (see Figure 4).[44]This was the focus of criticism among interviewees, as well as the subject of newspaper articles from the time (for instance, Beller, 1994).

Figure 4 – Rogers’ categories of adoption

This is a finding that might surprise a student of diffusion, especially in the light of Strang and Meyer’s important article, Institutional Conditions for Diffusion (Strang & Meyer, 1993). Their fundamental point is that “institutional similarity” is a key factor in diffusion. They resist “relational models” of diffusion, saying that they “underspecify the variety of effects that may be induced by interaction and interdependence” (p. 490). Instead, they propose that “linkages may be cultural as well as relational. That is, the cultural understanding that social entities belong to a common social category constructs a tie between them. […] We argue that where actors are seen as falling into the same category, diffusion should be rapid” (p. 490). The puzzle, however, is this: processes of diffusion seemed to happen quicker during the early 1980s than the early 1990s, that is, when Israel was less culturally similar to the United States. We might also see this in terms of Sassen’s concept of temporal order (Sassen, 2000b): local and global processes have their own temporal orders, whereby the former is slow and the latter fast. The change implied by globalization is a shift to a faster temporal order. The processes described here, however, seem to have moved from a faster to a slower temporal order.

To sharpen the puzzle, I propose a closer reading of Strang and Meyer. Their general rule of institutional similarity is specified with regard to modernity:
the more social entities are constructed and legitimated as modern entities (and particularly as modern ‘actors’), the more social materials flow among them.

Here, modernity is said to connote
the organization of society and the nation-state around universalized notions of progress and justice, as built up of rationalized organizations and associations, and as composed of autonomous, rational, and purposive individual citizens. And it implies the integrated functioning of these elements so that collective goods are enhanced by individual and organizational progress and contribute to such progress (p. 501).

According to this characterization, Israel was certainly a “modern” state in the early-80s, and so would communist nations seem to be as well. However, the latter are excluded from this version of modernity by a subsequent sentence: “When state socialism collapsed in Eastern Europe and the Soviet Union, Western academics rushed to the rescue, bearing analyses of optimal economic and political arrangements” (p. 502). In other words, communist countries attained the kind of modernity that Strang and Meyer are referring to only after communism collapsed. This is made clearer by a refinement of modernity: “Modern perspectives locate much value and responsibility in social ‘actors,’ both individual human beings and purposive rationalized organizations. The sovereignty and competence of these actors is celebrated. Social structures that rest upon the self-interested choices of autonomous actors are formally demonstrated to generate optimal outcomes” (p. 503). Here they reveal that in fact their definition of modernity is restricted to neo-liberal nation-states, which Israel could not really be described as being in the early 1980s (see, for example, Shalev, 1992). In sum, then, according to Strang and Meyer’s definitions of modernity, and according to characterizations of Israeli society in the early 1980s as compared to the early 1990s, it would indeed seem to be surprising that aspects of the Internet diffused faster precisely at the time when institutional similarities were much weaker.

Let me elaborate: although it has been argued that “ever since statehood, Israel has been undergoing a process of becoming capitalist society” (Ben-Porat, 1993, p. x; and see also Hanieh, 2003), there can be little doubt that the years following 1985 were particularly central in this regard. Indeed, by the end of the 1990s, Israel was held up “as a model of economic liberalization and successful adaptation to globalization and technological change” (Shalev, 2000). For many years Israel was an extremely centralist country (Ben-Porat, 1992a; Kimmerling, 1982), with the economy quite subsumed within the political field. Indeed, writing about the first few decades of Israel’s existence, Shafir and Peled wrote, “No autonomous business sector could emerge, and business decisions were made in response to or as part of political decisions” (Shafir & Peled, 2000b, p. 246).

However, following a period of hyper-inflation in the early 1980s, the 1985 Emergency Stabilization Plan put Israel more firmly on the road to a neo-liberal economy run according to the principles of the market, a process that was “openly supported and legitimized by the political regime and its ideology” (Ben-Porat, 1993, p. 28). By the early- to mid-90s these processes were well under way. They were also closely tied to changes in the political field, namely, the signing of the Oslo Accords in 1993. As well as delineating a political solution to the Israeli-Palestinian conflict, the Accords also ensured that Israel would open up its economy. Thus the labor organizations were weakened, with the result that failing factories were allowed to close, while entrepreneurs could raise capital on the newly prominent Tel Aviv Stock Exchange. This was accompanied by a cultural change in the direction of conspicuous consumption, signified by the arrival of a host of American firms to Israel during the early 1990s.

Given this background, then, how should we explain the surprising speeds at which the two diffusion processes under discussion took place? Perhaps the first thing to note is that the two episodes of diffusion being discussed were located differently with regard the state. In the first instance, the state had no involvement, and in the second, it was very heavily involved indeed. This is suggested by Sassen when she describes the “bureaucratic temporal order” as “subject to the slow moving and rule-bound public accountability of governmental processes” (Sassen, 2000b, p. 224).

Indeed, in the early 1990s the telecoms sector in Israel was still fairly centralized, despite moves to liberalize it, moves which took much further steps during the rest of the 1990s, and its bureaucratic wheels turned slowly. Thus, for instance, it took a year and a half for the Ministry of Communications to write the license giving MACHBA permission (post facto) to legally run an academic and research network, after Bezeq, the telecoms monopoly, said that it saw the academic network as illegal (Ein-Dor et al., 1999, pp. 7-8). Also, it took nearly two and a half years from MACHBA’s announcement that it was not prepared to become a nationwide ISP for the MoC to issue licenses to commercial ISPs. As will be discussed in a later chapter, certain figures in the ISP industry found this latter delay infuriating. However, other interviewees expressed understanding that these things take time, and that the MoC was acting in good faith and to the best of its abilities, even if these abilities were slightly less than might be desired.

Be that as it may, it is clear that state involvement in this diffusion process hindered it, especially when compared with Israel’s connection to EARN, and its attainment of the .il suffix, processes which were characterized by small-scale initiatives by groups of highly-motivated individuals. As Doron Shikmoni said about his involvement in such processes: “[We did it] of our own free will, not because someone sent us, but because we were mad about the subject, we were just, you know, freaks for it, it was in our soul”. This is a kind of activity that takes place well below the radar of analyses a` la Strang and Meyer, but it could well be the model for the diffusion of the Internet to countries outside the US–of course, I cannot make a clear assertion on this given the lack of data regarding other countries. However, this does strengthen my assertion that there is much to be gained for analyzing diffusion on this “micro” level.

On the one hand, then, the Internet could diffuse quickly to Israel because of Jewish ethnic ties and the movement of Israelis between Israel and the United States, and because Israel has long been a country with a deep technological bias;[45] while on the other hand, aspects that required state involvement meant that Israel moved slower than other countries.

1.8 Conclusion

In this chapter I have looked at two key “firsts” in the Israeli Internet: namely, the first connection to an academic network overseas; and the first licenses issued to commercial companies permitting them to provide Internet access to all comers. By focusing on these “firsts” I have avoided the technologically determinist tendencies in previous studies of Internet diffusion. Indeed, I have focused on a different aspect of diffusion altogether: while the studies cited at the beginning of this chapter look at the diffusion of the Internet within countries, simply taking for granted the fact that the Internet reached them in the first place, I have looked at the diffusion of the Internet to a specific country. By so doing, I am implicitly suggesting that the question of how the Internet arrives in each country is no less pertinent and worthy of research than the question of how quickly the Internet diffuses within each country, notwithstanding the policy implications for such understandings in terms of closing the digital divide within and between countries.

My analysis was conducted at a very micro level, identifying specific actors and pinpointing their role in the processes discussed. Following Street, I closely showed “how the product arrived there” (Street, 1997, p. 76). This is an approach that resonates with Sassen’s view that globalization is something that must be produced (Sassen, 2002b). The contributions of the individuals I identified in “producing globalization” were also placed in their broader context. This context is very specific to Israel of the 1980s and 1990s. In particular, my analysis revealed the importance of informal Jewish networks and migratory movement through space in understanding the timing and mechanics of the diffusion of the Internet to Israel, and highlighted the institutional and bureaucratic contexts of two major moments in that process.

This has implications for future research of Internet diffusion: it would seem that if we want to avoid assumptions regarding the very fact of the Internet’s worldwide diffusion, researchers ought to carry out case studies, most probably at the nation-state level, to determine the dynamics of the Internet’s arrival in those nation-states. What makes these studies challenging from a sociological point of view, of course, is that because each process may be assumed to be different, and because each context will most certainly be different, the theoretical issues raised by each case study must necessarily be unknown before the start of the research.

As mentioned in the opening section of this chapter, the current case raises the question of the place of cosmopolitans in processes of globalization, especially given that Everett Rogers himself refers to the important role of cosmopolitans in processes of diffusion (see also Beck, 2000, 2002). Rogers offers an inclusive definition, whereby the cosmopolitan displays an “orientation [that] is external to a particular social system” (Rogers, 1962, p. 102). Hannerz’s definition, on the other hand, is much more exclusive:
The perspective of the cosmopolitan must entail relationships to a plurality of cultures understood as distinctive entities. […] But furthermore, cosmopolitanism in a stricter sense includes a stance toward diversity itself, toward the coexistence of cultures in the individual experience […] It is an intellectual and aesthetic stance of openness toward divergent cultural experiences, a search for contrasts rather than uniformity (Hannerz, 1990, p. 239).

Accordingly, for Hannerz, not all contact with other countries or cultures indicates cosmopolitanism: “much of that involvement with a wider world which is characteristic of contemporary lives is […] largely a matter of assimilating items of some distant provenience into a fundamentally local culture” (Hannerz, 1990, p. 238). This more accurately describes how the Internet arrived in Israel. Moreover, the individuals pointed out above as having played an important role in bringing the Internet to Israel and institutionalizing it could not really be said to have displayed an “openness to divergent cultural experiences”. Their orientation was primarily, if not exclusively, to one other country apart from Israel, namely, the United States. Nonetheless, they provided the necessary bridge for the transfer of knowledge and technology from the US to Israel.

As this chapter has shown, and as is demonstrated more clearly in the following chapter, not all of the actors introduced here were embedded equally deeply into global networks. Moreover, it was those more deeply embedded actors who came to assume dominance in the emerging field of the Internet (see following chapter).

As mentioned above, this chapter did not set out to answer the question whether globalization strengthens or weakens the state (see, for instance, Guillen, 2001; Hobson & Ramesh, 2002; Holton, 1998; Meyer, 2000; O’Riain, 2000; Weiss, 2005). However, it is clear that the state played a central role, thus justifying the somewhat Westphalian view adopted above. The arrival of the Internet to Israel was a subnational process (Sassen, 2007b), an event that occurred within the Israeli academia. As long as this was the case, attendant processes (such as receiving the .il suffix) took place relatively quickly. However, state involvement soon became unavoidable as the demand for connectivity outstripped MACHBA’s ability or willingness to provide it. The diffusion of the Internet thus entered a bureaucratic temporal order (Sassen, 2000b) as government offices drew up licenses and contracts. This was partly necessitated by the state’s ownership of Bezeq, the telecoms monopoly, which gave the state an economic interest in the development of the Internet.[46] For all that it is perhaps the defining technology of the global era, then, the Internet is a fundamentally state-based technology. State involvement was essential in bringing the Internet out of the academy and into the private sector, including into people’s homes.

In sum, this chapter has shown, therefore, that while it is a global technology par excellence, the Internet was brought to Israel by people with weak transnational links; and while the Internet appears to be supranational, its institutionalization was heavily dependent on state-level machinations. This would seem to accord with Sites’ notion that,
[i]n certain national contexts […] it is more appropriate to conceptualize the state as highly enfolded within processes of globalization. […] the state continues to play a very significant, if underappreciated, role in the articulation of global linkages; […] this state influence over globalization consists not in a coherent strategy of national response but in an ad hoc series of policy reactions (Sites, 2000, p. 130).

In the following chapter I examine the state in greater detail, this time as the field of power in the emerging Internet provision industry in Israel of the early 1990s.


[1] See http://pages.cs.wisc.edu/~lhl/, accessed 26/6/2007.
[2] I am referring here to Sassen’s writings on globalization, and particularly global cities. (See especially, Sassen, 2000a, 2002a, 2007a).
[3] Somewhat similarly, geographer Stephen Graham argues that cyberspace is demonstrably grounded in metropoles (S. Graham, 1998; S. Graham & Martin, 1996).
[4] Katz-Kimchi performs a similar task in Chapter 1 of her PhD dissertation (Katz-Kimchi, 2007).
[5] http://www.rfc-editor.org/overview.html, accessed 20/7/2006.
[6] http://mosaic.unomaha.edu/gdi.html, accessed 20/7/2006.
[7] Some of those can be accessed fromhttp://mosaic.unomaha.edu/Pages/GDI_Publications.html.
[8] These studies are available at http://www.itu.int/ITU-D/ict/cs/.
[9] This refers to the online version of the article –http://som.csudh.edu/fac/lpress/articles/worldacm.htm, accessed 20/7/2006 – as posted by Press himself. In the journal version the Internet is described as “growing rapidly”.
[10] Hobbes’ Internet Timeline v8.1, http://www.zakon.org/robert/internet/timeline/, accessed 23/7/2006.
[11] http://www.norsar.no/NORSAR/, accessed 23/7/2006.
[12] http://www.norsar.no/NORSAR/history/internet.html, accessed 23/7/2006.
[13] http://www.globalcrossing.com/xml/network/net_glossary.xml#p, accessed 24/7/2006.
[14] Federal Research Internet Coordinating Committee, Program Plan for the National Research and Education Network, May 23, 1989, pp. 4-5, cited in Request For Comments 1192 (http://www.rfc-archive.org/getrfc.php?rfc=1192, accessed 23/7/2006).
[15] The NSFNET Backbone Services Acceptable Use Policy, June 1992,http://www.merit.edu/mail.archives/mjts/1993-01/msg00000.html, accessed 23/7/2006.
[16] The following is based on interviews with people who were involved in the process I am describing, and email correspondence with some others. In particular, Henry (Hank) Nussbacher was a crucial informant and plays a central part in the construction of the narrative. My review is also based on the Global Diffusion of the Internet (GDI) report on the Internet in Israel (Ein-Dor et al., 1999), newspaper clippings from the time, and various websites.
[17] “Unix-to-Unix Copy Protocol. An older protocol for sending e-mail between different Unix machines via regularly scheduled modem and network connections.”http://www.websightsolutions.com/faq_gloss.html, accessed 24/7/2006.
[18] As related to me by email by Daniel Brannis, nowadays Manager of Computing Facilities, The Selim and Rachel Benin School of Engineering and Computer Science, at the Hebrew University of Jerusalem.
[19] I interviewed Nussbacher on two separate occasions, on the 19 and 25 August 2004. I corresponded with him via email to clarify several points.
[20] “The international BITNET network began in the spring of 1981 when Ira H. Fuchs and Greydon Freeman, of the City University of New York and Yale University, respectively, decided that IBM’s Network Job Entry (NJE) communications protocol made computer-based communication practical between their universities.” http://www.cren.net/cren/cren-hist-fut.html, accessed 24/7/2006.
[21] Shikmoni was interviewed on 10 November 2004.
[22] http://www.iucc.ac.il/html/framesets_eng/about_iucc.html, accessed 24/7/2006.
[23] The CS in CSNET stands for Computer Science.
[24] “In February, 1984, Israel became the first international node on the CSNET, soon followed by Korea, Australia, Canada, France, Germany, and Japan” (http://www.livinginternet.com/i/ii_csnet.htm, accessed 7/8/2006); and, “In February 1984, Israel became the first international member of CSNET” (http://museum.media.org/eti/RoundThree08.html, accessed 7/8/2006).
[25] For example, Australia’s first commercial ISP started business in 1992, the UK’s in 1991, South Africa’s in 1992, and Austria’s in 1991. See Request For Comments 1192 – Commercialization of the Internet summary report (http://www.faqs.org/rfcs/rfc1192.html, accessed 5/1/2008).
[26] See http://www.nren.nasa.gov/
[27] I interviewed Rafi Hoyda on 6th August 2006.
[28] There are a number of small discrepancies between Nussbacher’s timeline, the Rahav Report (Rahav, 1998), and interviewees’ memories. However, these discrepancies do not impact on the overall picture.
[29] Nussbacher recalls that 80% of applications were granted; Shikmoni’s recollection is that virtually all applications were approved. They both sat on the committee that discussed applications. The owner of an early ISP told me of a poet whose request to be granted permission to access the Internet was denied on the grounds that his application was not research-based.
[30] Darcom received a license, but never provided Internet services. The company had been set up by David Reichmann, a Canadian immigrant, mainly to provide a communications link between Israel and North America. However, Reichmann died aged 34 of a heart attack in August 1994.
[31] See http://shekel.jct.ac.il/faq/net/israel-answer.html, accessed 1/2/2004.
[32] I tried to find the founder of Dataserve, Gil Ben Noon, in order to interview him, but was unable to track him down.
[33] I interviewed Alon on 18th May, 2006.
[34] See http://www.netmanage.com/company/history.asp, accessed 21/5/2006.
[35] For more on Smart Valley see http://www.jointventure.org/programs-initiatives/smartvalley/overview.html, accessed 21/5/2006.
[36] I interviewed Galil on 18th May, 2006 in his offices in Haifa. The walls were decorated with honorary doctorates from various universities, though pride of place was given to the certificate of his Israel Prize, which was awarded to him in 1997. A detailed biography can be found athttp://www.uzia.com/management/, accessed 24/7/2006.
[37] These biographical details are extracted from an interview Peled gave to an industry website,http://www.itvt.com/abepeled03.html, accessed 22/5/2006.
[38] I interviewed Dr. Peled in London on 2nd June, 2006.
[39] I interviewed Dr. Erez on 4th May, 2006.
[40] Peled’s involvement in Elronet was short-lived, however. In 1995 he was head-hunted by Rupert Murdoch to manage a company of his, and Peled now resides in London.
[41] I interviewed Dov Winer, himself an immigrant to Israel from Argentina, on 16 December 2004.
[42] http://web.macam98.ac.il/~dovw/, accessed 24/7/2006.
[43] As explained to me in interview by Dr. Abe Peled.
[44] See note 46 above.
[45] For instance, one of the first computers to be built outside the US was constructed at the Weizmann institute in 1954 (just one example of early technological achievement cited in a feature on technology in Israel published by Globes (Yachin, 2001a)).
[46] If only because connections were dial-up, and thus reliant on Bezeq’s telephone infrastructure.