A few of these essays need a longer-than-usual introduction, and this one one of them.
In the late 1990s, I worked closely with Drazen Pantic, a Serbian mathematician who, as a Yugoslav representative to the ITU, had been responsible for introducing the internet to former Yugoslavia. Drazen had gone on to found OpenNet, the internet-based operation of B92, a major broadcaster following the dissolution of Yugoslavia. It became the only major independent broadcaster in Serbia and a crucial node in organizing protests against the Milosevic regime. When the government raided B92’s transmitter and shut it down, within a day or so Drazen arranged for the Dutch internet provider XS4ALL to broadcast B92’s programming over the net. That was a remarkable technical achievement, but it was also hilarious: B92’s terrestrial radio signal was local to Belgrade, but its internet *broadcast was accessible throughout former Yugoslavia — and, indeed, throughout the world, making B92 an international cause célèbre. For his efforts, in 1999 Drazen was awarded an EFF Pioneer Award, an honor that until then had mainly gone to pantheon of networking heroes — for example, engineers like Vint Cerf and Jon Postel, and encryption researchers like Phil Zimmerman and Whit Diffey. Drazen’s account of this period can be found in his paper, “Radio B92 and its Internet Project: From Local to Global Media.”]
The Milosevic regime was aggressively persecuting B92’s principals, notably its head Veran Matić, so Drazen wisely relocated to New York, where I too was based. One area of our work together involved the potential for low-power FM radio (LPFM), a field that was showing new promise as the US FCC considered liberalizing the licensing process. Based in part on Radioqualia, a project by our friends Adam Hyde and Honor Harger to build what was arguably the first instance of something we take utterly for granted now — an internet “radio” aggregator. Drazen and I were looking at ways that local LPFM broadcasters scattered around the US (or, potentially, anywhere), could syndicate their programming, creating cooperative content-development networks whose offerings would better reflect and respond to trans-local needs — for example, in rural farming communities.
In 2002, the Soros Foundation (at the time, the Open Society Foundation) invited me to propose a project for funding. In the decades since, “Soros” has become an object of fear and loathing for rightists. Their paranoiac ravings have obscured the astonishing role the foundation played in promoting democratic values and institutions around the world — particularly in what, in my circles, we jokingly referred to as the “ex-East,” that is, Eastern Europe. One detail that’s particularly been lost: the foundation’s deep technical expertise in working with new technologies to circumvent the often-lumbering apparatus of state oppression — like getting official approval for boring old “faxes between offices to mask data signals that, in reality, were much livelier dialup bulletin boards or BBSes.
In those days, I had close friends who were Sri Lankan expatriates. The cessation of overt hostilities there — or, more plainly, the end of a “shooting war” — seemed like a promising context to explore ideas along the lines of my work with Drazen. With Soros’s institutional interests in mind, I proposed an exploratory study to see whether it was possible to work with wifi networks — which, at the time, were still pretty new — to create new social contexts that were less susceptible to surveillance. The OSF gave me a one-year renewable grant to go ahead.
Over the course of the next year and some, I worked with aid organizations like Canada’s IDRC and traveled around northern Sri Lanka extensively to consult with a wide variety of people and entities across civil (and “uncivil”) society and the commercial sector — government officials (including the Ministers of the Interior and Finance), the CEO of Maharaja Media (which owned the cellular phone networks in disputed areas of the country, among many other things), Catholic clergymen, youth groups, Muslim “IDPs” (i.e, internally displaced peoples, distinct from transnational “refugees”) in camps in the western Puttalam District, advocates of the “disappeared,” and more. Unfortunately, the short answer to my study was no, it wasn’t possible — at all. *After a decades-long civil war between Sri Lanka’s two main opposing forces — the widely feared insurgency called Liberation Tigers of Tamil Eelam (LTTE), or “Tamil Tigers” for short, and the heavily militarized government — there was *no “safe” space of any kind: no spaces or places free of surveillance or coercion.
This essay is a result of that work. It was solicited for a book, Jodi Dean, Jon W. Anderson, and Geert Lovink, eds.,Reformatting Politics: Information Technology and Global Civil Society (London: Routledge, 2006) . I wrote the essay, sent it off, and didn’t give it another thought until I saw the book announced. When I checked with the editors, one of them said they had never received the essay. ¯\(ツ)/¯
In many “Southern” countries with a pervasive agrarian sector, the deployment of information communication technologies (ICTs) in rural areas is widely viewed as an essential step in the grand, aspirational project of social advance.1 By far the most common current model is the telecenter — roughly speaking, a small institution that provides basic telephony and internet access, often in informal conjunction with other practical services like computer-skills training, support for civil services (educational, medical, agricultural), and functions that help people to respond to administrative needs such as photocopying, plastic lamination of important documents, and so on. Telecenters are, basically, copy shops for the poor.
A more precise description of telecenters is possible — for example, a typology based on distinctions between sponsoring organizations, relative levels of interconnection with other telecenters, and so on — but it wouldn’t be very useful. The more nuanced or specific approaches that such a typology could suggest would undermine the value that the telecenter approach offers to its national- and international-level advocates, a value that derives precisely from its generic, one-size-fits-all nature. So, instead of critiquing the telecenter approach from within, as it were, by fleshing out a more analytical model, I’ll critique it from without.
In a perfect (or maybe just better) world, the surprising dominance of the telecenter approach to rural ICT deployment might suggest that the model was developed in accordance with more or less rational methods, or at least on the basis of heuristics more critical than typically found through the twin lenses of originary models and retrospective program assessment. For example, more diverse approaches could be formulated and refined through iterative research, viewed (not necessarily assessed) in the context of other approaches to some or all of the problems supposedly being addressed. But it’s at this point that the benevolent imperatives of development conflict with scientific imperatives; and since the institutional context is that of development, its own imperative — the need to act — prevails.
Setting aside those kinds of high-minded debates, it is of course possible that the telecenter model really is the most flexible way to respond to local ICT needs in the South. But we’d do well to view such an exceptional proposal with skepticism — or, at least, to put the burden of demonstrating it on advocates of that approach. There is lots of literature on the subject, but most of it takes the form, give or take a little, of sales brochures: the annual reports of CBOs and NGOs recounting their quantitative achievements, heartwarming newsletters about the amazing progress of a few individuals in a village, casual reflections circulated among colleagues, governmental press releases announcing various “e-”initiatives, and so on.
Given how pervasive the telecenter approach is worldwide, one would expect a fair amount of critical discourse and debate. But, in fact, critiques are pretty scarce. In part, this scarcity itself is a mute testament to some of the structural difficulties inherent in the telecenter approach, understood as one kind of more or less reproducible approaches to addressing problems in the South. A rigorous or exhaustive analysis of these difficulties lies far beyond the scope of this note; but my research (in Sri Lanka during 2002–2003) directly and indirectly has pointed up a few key dynamics that tend to discourage critical debate:
-
Local expertise: Those most familiar with the peculiarities of a local context are unlikely to be nearly as familiar with other contexts; as a result, the conditions that give rise to expertise do not encourage comparative analyses.
-
Stakeholder conflict of interest: Those involved enough to provide well-informed assessments are likely to be involved or aligned either with funding- or technology-transfer donors or recipients, neither of whom is likely to be openly critical of the dominant model of ICT propagation. (Alternatively, in what should be funders’ worst nightmares, well-informed locals are intimately familiar with, and therefore probably involved in, local politics and/or belligerence.)
-
Anything is better than nothing: In the absence of viable and accessible alternatives, the vaguely defined nature or purpose of telecenters is seen by decision-makers, local and otherwise, as the most “open” form of ICT intervention. It certainly could be — but hope is not a plan.
-
“Disinterested” local interests: Local stakeholders who are in a position to materially advance development goals (for example, the local business community) often see development projects as misguided allocations of effort and resources. The needs that ICTs deal with, they say, are much more attractive to commercial interests than are other serious needs such as substantial and sustained health-care resources in rural areas. The result, in which parties talk past each other, can lead to lack of coordination and competition between non-local development forces and local commercial forces.
These and other dynamics tend to reinforce other tendencies in development circles (for example, the heavy reliance on models and program assessment hinted at above, as well as a growing reliance on sponsorship-oriented public–private partnerships) that ossify current thinking about what forms ICT intervention can or should take.
For example, a central tenet in the field is that the comparatively expensive “one user, one device” assumption typical of computer usage in developed circumstances isn’t sustainable in less-developed settings. It’s certainly true that the imputed assumption is itself based on settings — increasingly general-purpose devices in white-collar work environments and homes — that offer inappropriate analogies for ICT interventions in the rural South. Better analogies are available.
One example is the shared computing lab or specialized digital workshop common in academic settings. Not only do these multiuser environments sidestep red-herring issues such as “one user, one device”; but insofar as they often involve equipment that’s high-end or specialized even within a relatively advanced setting, in many respects they better mirror some of the main conditions of a telecenter — say, neophyte users confronting complex, exotic, fragile, and precious equipment, often under the watchful eye of an expert operators and instructors. The downside is that this kind of analogy, which stresses an institutional setting and an expert staff, tends to deemphasize “the end user” as such.
To the extent that computing skills are widely (and somewhat tautologically) seen both as a means and an end of ICT interventions such as telecenters, marginalizing the end user runs counter to the strongly individual and small-group orientation at the heart of the telecenter approach. Alternative and experimental approaches that place less emphasis on a popular or populist notion of end users — for example, the use of interstitial, ad hoc ICTs to build local communication links, to support special-purpose projects such as surveys and censuses of the assets of those facing war in order to lay the basis for future remunerations, or to encourage covert communication under repressive regimes — will not be seen as optimal by telecenter advocates, however useful or needed.
The telecenter approach is sometimes described as a “movement” because it is the project, at once collective and distributed, of thousands of deeply committed individuals who have developed local approaches, oftentimes over and against more centralized, big approaches to development. Through their efforts, the intimately pragmatic concerns of beneficiaries have indeed filtered up to the higher levels of development organizations. Thus, for example, in 2003 the staff of the International Programs group of Cornell University’s College of Agricultural and Life Sciences compiled an exhaustive “Handbook for Telecenter Staffs,” a 245-page “modular” guide (curiously, in “Microsoft® Word” format — a $200+ application) which sets forth the boggling range of issues involved in operating a best-practices-oriented telecenter, from a broad overview of the historical role that ICTs can play in developing areas to the intimate details of providing reliable technical services in unreliable circumstances.
It’s an amazing work, to be sure. And yet it embodies some of the basic contradictions that make the telecenter such a problematic proposition. Most obviously, the formal language and bureaucratic substance cannot be the most direct or effective method of speaking to those who struggle to find ways to make ICTs relevant to the immediate concerns of impoverished and largely agrarian people. Even among the most advanced ICT experts in the most developed circumstances, it is an accepted fact that few, if any, users actually “RTFM” (“read the f—ing manual”). Along similar lines, the computer industry’s endless rhetoric about “ease of use” and the like belies the fact that any structure greater than a few interconnected ICT devices will all but require specialized systems-administration knowledge. All of this expertise is, obviously, of great value; and yet each “layer” of knowledge supposedly needed to establish and maintain an effective telecenter — technical, administrative, and so on — itself become a formidable barrier to entry for many at the local level. And even that sets aside the question of how one understands “entry.” Individual progress in terms of measurable digital skills? Broad social advance? Regional-historical interpenetration?)
Moreoever, these barriers are not at all limited to locals. As the Cornell handbook notes, “The telecenter movement has gained substantial momentum during the past five years largely through the leadership of international organizations such as the International Telecommunications Union (ITU), the Food and Agriculture Organization (FAO), the World Bank, UNESCO, the International Development Research Center (IDRC) and the U.S. Agency for International Development.” That’s a formidable list in itself, and it was made that much more so when Microsoft joined the ranks with its own growing interest in implement thousands of telecenters in developing areas worldwide. Given the gravitas of these organizations, to say nothing of their collective bureaucracies, it seems unlikely that the telecenter model specifically, or the field of ICT intervention in the South more generally, will henceforth evolve fluidly on the basis of grass-roots experience and input. On the contrary; this clear success of the telecenter model all but guarantees that it will become further codified and ossified according to past practices — very probably at the expense of riskier or less “open” approaches.
In particular, the growing involvement of Microsoft (an organization with a well-deserved reputation for zealous pursuit of long-term profit-motivated strategies) suggests that the rough description of a telecenter offered above should be recast in more neutral terms — say, a functional facility for housing commodity ICT hardware and software in a less-developed setting. Even so, the question remains why this model would be so appealing to these organizations?
To begin with, the generic nature of this approach to deploying ICTs in the rural South is well suited to the generalist needs of the transnational entities that increasingly promote them as an effective model: national governments, multilateral treaty organizations, national development agencies, and interested vendors. From the more professionalized standpoint of such organizations, the virtues of a self-similar model are obvious: centralized planning, efficient planning at high levels of governments and large organizations, professionalized and scalable management structures, standardized procedures, streamlined procurement systems, increased cost controls, and so on. All well and good; but, clearly, these are not the street-level issues that telecenters will face.
There is little doubt that the support of these powerful national and transnational entities ought to be very effective in pursuing telecenter programs. Furthermore, certain new benefits will accrue. For example, the aforementioned disinterested local commercial interests, which often shy away from non-profit-oriented projects, may show greater interest in engaging with larger-scale projects supported by such prestigious and effective entities. However, these trends will likely entrench some of the more problematic aspects of telecenters, for example:
-
Opaque resource allocation: Because telecenter programs will more often be the byproduct of negotiations at the national and transnational level rather than the fruit of well-articulated proposals emanating from the grass roots within a coherent and meritocratic framework, they will increasingly be planned and supported through processes far removed from telecenters’ stakeholders and constituents. While it is possible that these processes will be oriented toward mitigating disparities and inequities at the local level, that possibility seems remote; instead, it seems much more likely that telecenters will be allocated though mechanisms “native to” the formal and informal governance systems that perpetuate many of the dysfunctional conditions these deployments claim to seek to ameliorate.
-
Vendor path-dependency: Microsoft is already, without virtually no exception, the public face of ICTs throughout the vast majority of the world. (In this respect less-developed regions lead, because alternatives are virtually unknown.) Alternatives to Microsoft (for example, open-source software such as Linux or BSD, OpenOffice, and so on) provide alternatives that, more than merely being financially or functionally viable, offer much more varied forms of engagement (most notably, the possibility of participation in development through the informal networked activities of local talent, for example, in the translation or adaptation of software into local languages). If Microsoft extends its market reach through the vector of rural telecenters, the negative consequences will range from hazy “missed opportunities” to clearly defined technical problems (e.g., crippling plagues of Microsoft-specific worms and viruses wreaking havoc with vulnerable rural networks, vulnerable low-bandwidth networks needlessly saturated with software whose licensing enforcement mechanisms constantly “phone home,” and so on).
-
Monopolistic ICT deployment: Rumors of major telecenter initiatives already have a centralizing effect. Development agencies, which are ever wary of duplicating effort, are reluctant to support small-scale efforts and/or steer them toward collaboration with larger partners. National governments, which typically prefer large-scale initiatives, make use of various regulatory options (e.g., suspension of import taxes, exclusive frequency licensing, etc.) with which to encourage large projects while discouraging smaller ones. And local commercial actors seeking advantage will tend either to wait-and-see or to affiliate themselves with projects seen as prestigious. Should telecenters become more associated with multilateral and multinational organizations, these negative outcomes will likely intensify.
In sum, the success of the telecenter approach to ICT interventions, particularly in the rural “South,” comes at the expense of less efficient or rational research and projects that are context-specific and, perhaps, irreproducible. As a model, its current formulation is becoming more responsive to the needs of funders, development agencies, and, possibly, investors, and — therefore, maybe — less responsive to the needs of local beneficiaries.
Note
Footnotes
-
The research on which this was based received generous support from the Open Society Institute’s Information Program. Many thanks are due as well to R.K. in Colombo, Sri Lanka. ↩