See Kevin Donovan's most recent contributions to Society & Space: Infrastructuring aid: materializing humanitarianism in northern Kenya

Across the world, humanitarian organizations and governments are investing considerably in so-called "cash transfers." These initiatives typically provide small cash grants to low-income populations. In Latin America, programs like Brazil’s Bolsa Família or Mexico’s Oportunidades (not Progresa) have pioneered the recent profusion of social protection schemes in the global South. In South Africa, a welfare scheme with roots in the 1920s has been expanded in the post-apartheid period to incorporate nearly 20 million individuals. In his newest book, Give a Man a Fish, the anthropologist James Ferguson suggests these are indicative of a new politics of distribution. Indeed, cash transfers have been growing enormously, often replacing food aid. One 2009 study documented 123 cash transfer programs in sub-Saharan Africa alone. As leading proponents of enthusiastically sum up, the idea is “just give money to the poor.”

Underlying the trend is a belief that cash transfers are a more feasible form of aid, less likely to result in failure. Technology is key to this vision, with the World Bank and others rushing to “new leapfrog technologies” such as biometric identification. Such enthusiasm is not new to the aid world, of course; from fertilizer to mobile phones, the industry betrays a proclivity for technological fixes to political problems.

In my paper, "Infrastructuring aid: Materializing humanism in Northern Kenya," I describe the type of sociotechnical labor necessary to enact these visions. The aid workers I observed deploying cash transfers in Turkana, in northern Kenya, were deeply concerned with what Stephen Collier calls the “intransigence of things,” working, negotiating, and struggling to create a functional circulatory system. I draw on the idea of infrastructuring to approach ethnographically the human and material components of infrastructure "in the making." While much of the literature on humanitarianism, human rights, and development has attended to politics, law, and ethics, an attention to infrastructuring moves past idealist concerns to the technicalities through which ideals become enacted on the ground.

Because humanitarian biopolitics requires infrastructuring, it manifests in divergent and negotiated ways—ways that are not reducible to the humanitarian actors or logics alone. The purpose of this essay is to briefly provide some supplementary insights and, particularly, photographs of the humble things that do so much to make abstract universals into lived inequalities.

My work focused on the Hunger Safety Net Program, an initiative funded by UKaid and implemented in 2013 in partnership with NGOs, Equity Bank, and the Kenyan government. HSNP is particularly notable for its use of biometrics. Numerous social protection professionals explained to me that the appeal of biometrics was their ability to replace paper—a technology seen as vulnerable to fraud and leakage. Other cash transfer initiatives, which had not adopted biometric technology, required paper forms, printed in triplicate, to be transported from Nairobi throughout the country and either signed or marked with an ink fingerprint. These paper volumes were then reconciled manually back in Nairobi, a tedious process of data entry that eventually resulted in a digital record of who was (supposedly) paid. Such a system conflicted with the audit imperative sweeping the aid industry.

The designers of HSNP wanted a different system so they pioneered two key innovations: biometric identification and banking agents. Recipients’ fingerprints were to be scanned at enrollment and then used to identify them during payment. Secondly, local shopkeepers would be appointed as HSNP payment agents and equipped with fingerprint readers. Such a system, however, required considerable infrastructuring.

Due to international rules around banking services, it would be necessary for recipients to present their ID cards; however, when the HSNP started, 15-20 percent of adults in the program areas lacked national IDs. As one program officer told me, “no one is going to stop you in the middle of the desert and ask where is your ID?” So, in addition to rolling out their system, HSNP needed to boost government IDs.

HSNP did not consider ID cards sufficient for authenticating recipients so, as with more and more initiatives in the global South, they adopted biometric identification. Biometrics, however, are prone to failure, especially in places like northern Kenya. Instead of biometrics, HSNP could have adopted personal identification numbers (PINs), but when I asked about this alternative, two reasons were offered, both revealing much about the technopolitical rationalities operating within this particular humanitarian intervention.

First, biometrics were desired due to their individualizing effect: unlike PINs, you cannot share your fingerprint. Secondly, I was repeatedly told that PIN authentication was not viable because recipients were “illiterate.” “Even writing down a PIN and keeping it secure,” I was told, “can be a challenge for pastoralists” who lack “desks or drawers or anywhere where they can keep some of these documents.” Biometrics, in contrast, affix identification to the body. As one person told me, you can’t “forget your thumb.” Thus, it is a particular understanding of intellectual (in)capacity and material culture (or lack thereof) that informed the adoption of biometric adoption by HSNP.

Affixing identity to the material body, however, caused complications. HSNP needed to allow beneficiaries to nominate a secondary recipient who would be accepted for the payment because some beneficiaries were unable to travel to payment sites. Another “fairly big problem” was alterations to fingerprints (from age, manual labor, smoking and other practices). Biometric technology, then, “scripts” reality (Akrich, 1992) by assuming mobility, corporeal stability, and the universality of biometrics algorithms.

In addition to identifying recipients, the infrastructure of enrollment was crucial. As in other humanitarian settings, Toyota Land Cruisers were crucial to HSNP’s infrastructuring. Without paved roads, these rugged vehicles allow aid workers to abstract themselves from certain realities of Turkana. But such vehicles are also structuring, hierarchical agents, producing an affect of speed, placelessness, and separation. They create manageable space but only for those with access to the infrastructure.

In other cases, the infrastructure of cash transfers became a stage for quiet contests between recipients and aid workers. HSNP pays recipients different amount based on certain socioeconomic variables. During enrollment, recipients were assigned to one of four income brackets (very low, low, middle, and high). The aid workers had previously used an informal system of colored stickers on ID cards, with pink corresponding to very low, yellow to low, and so on. Although this was never disclosed to the grant recipients, they evidently recognized the pattern and soon enough the aid workers noticed the cards had been reshuffled, in contradiction to the dictates of the program. The quick fix that they adopted was to discard color-coding and instead use letters with ‘TCW’ indicating ‘Turkana Central Wealthy’. Literacy, then, became a tactic in infrastructural contests.

The final infrastructural innovation discussed more fully in the paper is the use of banking agents. HSNP worked with Equity Bank to use local shopkeepers as payment agents. As with many humanitarian efforts today, HSNP mobilizes a range of public, private, and civil society actors in what Didier Fassin calls “nongovernmental government.” Shopkeepers were desirable because they often already existed, demonstrating once again that infrastructure is cumulative and recursive. Additionally, banking agents were useful because HSNP could delegate responsibilities to them, such as cash management or troubleshooting. Indeed, of five payments I witnessed one day, fingerprint authentication originally failed on two of them, requiring the agent to scrub fingers clean and place them on the device correctly, cajoling human and machine to successfully interact.

Agency banking has, today, grown enormously due to the expansion of mobile money in Kenya and elsewhere. In this industry, agents are sometimes referred to as “human ATMs,” a phrase that is revealing for the distilling the core purpose of agents: cash provision. As with Latour’s classic discussion of “gunman,” it also reveals the interlacing of human and machine that is crucial to infrastructural success.

But using "human ATMs" introduces subjectivity that is not always desired. In one case of impropriety, when the grant was valued at KSh 2,250, a number of agents were found to be dropping the final KSh 250 (around US $3) because of the difficulty of providing such small denominations in such volume; instead they offered free goods from their shop valued around KSh 250. In this case, it was the pragmatics of physical currency that shaped behavior. Delegating to banking agents was desirable because they were feasible and could help recipients, but such subjectivity was not always desired and HSNP sought to discipline them—a type of infrastructuring that will not, I suggest disappear.

An attention to the infrastructuring of aid is important not only for highlighting the mundanities of materiality in places like Turkana. It is important, as I argue in the paper, because the biopolitics of humanitarianism is, at the end of the day, dependent upon the articulation of political ideals and material things—whether fingers, paperwork or SUVs. 

References

Akrich M (1992) Beyond Social Construction of Technology: the Shaping of People and Things in the Innovation Process. In: New Technology at the Outset. Boulder: Campus/Westview, pp. 173-190.