Abstract
This text develops the notion of meantime of development to excavate a temporal orientation that underscores techno-optimist engagement with the future. The discussion bounces off Jharkhand’s experience with the new smart version of India’s Public Distribution System (PDS), as an instance of using aspirational technology to fix long-standing social ills. The system deploys biometric IDs (Aadhaar) for policing distribution and preventing fraud and leakage at the end point of the commodity chain, the fair price shop, where customers receive subsidised grain. Aadhaar was intended to improve fairness and protect the poor from corruption, yet from the beginning the new infrastructure was plagued with problems and required further tweaking to become somewhat functional. As all stakeholders become embroiled in complicated negotiations, the landscape of problems has shifted, producing new exclusions and novel forms of fraud. In techno-optimist readings, such complexity becomes simplified and broken down into failures to be learned from and there is the naive expectation that the next iteration of the reform will fix these problems. Because aspirational technology proves to be particularly malleable and flexible, the iterative remaking of workflows constitutes more than a technical fix; rather, it sustains an optimistic view of the future.
The making of a national biometric registry in India, called Aadhaar, and the related reform of bureaucratic processes, have been interpreted as an important step in India’s continuous process of nation-making. The Aadhaar infrastructure, the largest biometric database in the world, has earned international acclaim (Gelb & Metz, 2018) and is an object of pride, not least because the successful biometric registration of over one billion citizens in less than ten years impressively underlines the country’s claim to excellence in IT engineering. Moreover, there is the hope that the digital infrastructure will level out inequalities in India, make redistributive processes work better and thus leapfrog the whole country into the twenty-first century. This notion of Aadhaar as a ‘time machine’, an infrastructure capable of bringing about the future, is mirrored in a series of temporal imaginaries associated with the implementation of biometric IDs. Aadhaar is anticipated to speed up development and make efficient a traditionally slow and cumbersome bureaucracy; public–private partnerships will inject the dynamism of entrepreneurial undertakings into the state structure and thereby transform it. Nair (2019) contrasts this celebratory rhetoric of the friction-free start-up state with her ethnographic observations about the slowness of implementation and application. She describes clients queueing for their turn to be heard by the gatekeepers of the digital state, waiting for electricity to return so that they may sign up for a service or wondering why, months after they applied for it, the Aadhaar number has still not arrived. This contrast between exaggerated expectations and heart-breaking stories of individually experienced injustice have become a durable feature of the debate on digital governance, according to which new developments, alternatively, make the administration more efficient or more cumbersome, fast-track development or undermine social security, presage the future or cement structural injustice.
In this article, I study how policy makers and critical scholarship engage with the stark difference between promise and reality of Aadhaar and by doing so excavate a rhetorical figure that allows the promise of the technical fix to survive, and with it the promise that improvement is imminent. Beyond the truism that what counts as improvement will depend on one’s social position and specific value judgements, here I am particularly concerned with the hope that Aadhaar could improve social justice. The system was deployed to smoothen out redistributive systems, to make them efficient and corruption-free. However, when it was rolled out, Aadhaar produced countless frictions and new informal workarounds. These have been addressed not just by tinkering but through an interactive process of improving the application of the system. However, these improvements have in turn created a range of new problems which have needed fixing, so that stakeholders have been drawn into an unending iterative process of doing and failing, hoping and repairing (Rao, 2022; Veeraraghavan, 2021). The great flexibility of digital technology allows this endless web of small changes to continue at a rapid pace, maintaining the momentum of reform and allowing bureaucrats and critical observers to hold on to the hope that digital surveillance may indeed be a game changer.
One instance of the evolving utility of Aadhaar, the much-discussed Aadhaar-enabled Public Distribution System (AePDS), is an excellent example of how, despite disappointment with the results of the implementation of the new system, the near future of techno-reform continues to look optimistic. When Aadhaar was introduced, the architect of the digital network, Nandan Nilekani, promised nothing less than a revolution in which, according to his hyperbolic formulation, ‘101 people can fix all of India’s problems’ (Nilekani & Shah, 2015, p. xxi). From the first day of implementation, however, difficulties mounted. And while each new application presented a new set of challenges and required further optimisation, Aadhaar nevertheless became an integral part of the administration. Adherence to the new technology was not so much due to resounding success than to the expectation of rapid improvement and the eventual success that would follow those improvements. In the meantime, errors were unavoidable and had to be tolerated.
Take, for example, the process of rolling out AePDS in Jharkhand. In the early 2000s, many were looking at the young state, freshly carved out from Bihar, with hopeful anticipation. The new elite were given the benefit of the doubt; they were widely believed to be earnestly trying to improve governance, which included using the tools of digitisation. Yet, the first experiments with Aadhaar failed quickly, not least because many places were not connected to the telecom network, and people in remote areas faced many barriers to accessing the bureaucracy. Euphoria turned into disillusionment and angry allegations of mismanagement. Some critics accused the political elite of negligence and causing starvation, demanding that they relent and reform the reform. In this reformist spirit, an avalanche of studies has reflected on the events and the technology’s failure and provided numerous recommendations on how to mitigate current challenges. None of them suggested returning to the old system; rather, they recommended reforming ‘forward’ and introducing more or different technologies.
Despite the variety of conflicting proposals, commentators in India tend to subscribe to the spirit of reform, and many see Aadhaar or similar technologies as providing opportunities to create more inclusive governance. This hope is fuelled not by the resounding success of previous reforms, but by the anticipated benefits of further reforms. I call this orientation towards the future tense of improvement the meantime of development. The term borrows from recent debates on ‘meantime’ yet moves beyond treating it as a product or a period. To date, authors have used the term to capture the mood of a particular point in time that is characterised by waiting or enduring (Cloke et al., 2020; McKay, 2017). It could be a time of nostalgia or of activism that strives towards transformation (Jansen, 2015). For example, for Redfield (2019), ‘meantime technology’ indexes products resulting from frugal innovation or products that are cheap but functional and can fill a supply gap—in the meantime—until the infrastructural need is finally met. My usage of the term leans on a combination of these usages and describes a particular orientation in time. The term then captures the spirit of anticipation born from the confidence that improvement is possible and will come about soon—with the next software update, a hardware upgrade, a technical workaround or any other micro-reform.
My analysis of techno-optimism partakes in the trend of paying renewed attention to questions of time (Bear, 2016). It breaks away from a monolithic notion of states as being static and expands the study of the anthropology of the state beyond the focus on spatial arrangements, the role of social relations and material objects (Anjaria, 2008; Ghertner, 2017; Hull, 2012). At the same time, it avoids viewing the state exclusively through the temporal registers of a lack, such as unnerving waiting (Carswell et al., 2019), pathetic stasis (Gupta, 2012) or nostalgic longing for a better past (Jansen, 2015). Instead, I take seriously the will to improve (Li, 2007) as it is manifest in the belief that reform will make a difference yet without succumbing to the exaggerated promises of the ‘golden bullet’ tale. While reforms often come with bold promises, in day-to-day practice they manifest as an endless, often arduous sequence of micro-changes, and it is difficult to tell which of these changes are part of a particular reform and which are not. The lack of clarity from close-up is tackled by techno-optimistic narratives that link processual changes to notions of progress and thereby allow the linear time of the development imaginary to emerge.
To demonstrate this contextual linking of historical analysis, empirical observation and the development imaginary, I have chosen the much-discussed example of Jharkhand’s experience with AePDS. The next section introduces the most recent reform of India’s PDS and explains the rationale for using end-to-end tracking. What for the government is a huge success, proudly displayed on a public dashboard, has multiple layers. These layers are revealed when moving beyond the façade of numbers in the second section. They narrow the focus and introduce several complications caused by the roll-out of AePDS in Jharkhand as well as the fixes suggested by commenters evaluating the project. The third section reviews the tense debate from a distance, to illustrate a temporal orientation that saves techno-optimism by defending friction as a meantime phenomenon.
Making the Public Distribution System Forgery Free
AePDS, ABBA and ePOS are the three new acronyms floated after Aadhaar, the new digital ID, became an integral part of India’s PDS. Aadhaar literally means ‘foundation’ and is an identification system that provides every Indian with a unique number linked to their biometric data for instantaneous paperless identification via the cloud, (theoretically) anywhere and anytime. Most Indians have signed up for the service, but the technology works better for some than for others, and it is used not just for convenient identification but more often than not as a form of compulsion, and there are accusations that the government is using Aadhaar to impose a new regime of excessive, even dangerous, surveillance (Ramanathan, 2010). For welfare clients, an Aadhaar number has become mandatory and must be submitted alongside the application for any governmental service, including the PDS, which in its new iteration is called the Aadhaar-enabled Public Distribution System and functions as an extensive network for the procurement and redistribution of subsidised food to the large number of needy people.
According to the National Food Security Act of 2013 (NFS), every Indian citizen living below the poverty line, which accounts for roughly 50% of the urban population and 70% of the rural population, is entitled to receive 5 kilograms of grain every month at a highly subsidised rate (
AePDS is operational in most Indian states, and the government likes to call it a success. As proof of the claim, the Ministry of Food Security and Distribution maintains a dashboard, the new public face of the bureaucratic apparatus. In bold, colourful letters the internet page announces that via AePDS, every month, the state distributes 60 metric tons of grain to 797 million customers using the network of 543,658 ration shops. The viewer is enticed to follow these processes through real-time digital tracking and has a full view of dynamic developments. For example, on 6 March 2022 the dashboard informed readers that 2.1 metric tons of grain had been distributed so far that month, a figure that doubled two days later to 4.06 tons on 8 March, before gradually climbing to 16.12 tons by 17 March (
PDS is an expansive system and, due to its size and complexity, extremely friction prone. As in any distributive system, many things must line up for the PDS to work and for grain to move through a convoluted system from farmer to beneficiary in ways that are perceived as just and fair. The number of nodal points at which things can go wrong has increased with digitisation, since on top of physical logistics (made up of things such as produce, trucks, storages and shops) and a paper bureaucracy (made up of things such as ration cards, licenses and ledgers), another infrastructure has been imposed, one that requires servers, digital IDs, biometric scanners and databases. As all of these objects and their supporting infrastructures are being produced, serviced, maintained and inspected, people observe successes, but also many errors, failures and fraud. These generate the desire to improve the system and narrow the gap between an ideally imagined system and the really existing one. In the process, the goal posts of what is an ideal system are always shifting, as welfare ideologies change, governments alter, political fads pale and new technologies of rule are invented.
As an aspirational technology, Aadhaar comes with a promise. It awakens the hope that digitisation can foster improvements in social distribution, and the confidence that there is a fair chance of success, not just for the technology to function, eventually, but for the social itself to be transformed. It is said that AePDS will eventually make PDS corruption free, because the technology will enable tracking of grain along the entire distribution chain, from farm to consumer. To date, not all the components of this network have been installed. However, most states have placed devices in the ration shops, the end point of the extended commodity chain. Throughout the history of PDS, ration dealers have been a prime target of corruption accusations and are suspected of siphoning off large quantities of grain. EPOS machines are supposed to intervene here and stop ration dealers from performing two widespread tricks: over-reporting sales and identity fraud. Traditionally, each ration shop receives the full load of grain each month, regardless of whether clients pick up their quota or not. To keep the leftover grain for sale at a hefty profit on the private market, shop keepers would falsify their accounts to exaggerate their sales. They would also maximise this illegally retained amount by maintaining inconvenient opening hours that prevent beneficiaries from purchasing their entitlements, as well as acquiring ration cards for ‘ghost beneficiaries’—fake beneficiaries added to the roster for the sole purpose of siphoning off cheap produce. ABBA makes diverting grain more difficult because ghost beneficiaries do not have a digital ID and cannot verify their presence at the shop and because automated sales records cannot be forged. Thus, states distributing ePOS devices anticipated that point-of-sale surveillance would lead to significant fiscal savings and empower beneficiaries.
The following section traces the deployment of ePOS in Jharkhand, a place that is known for its reform affinity. The young state was in a rush to implement AePDS because of a widely felt need to emancipate the bureaucracy from the oppressive Bihari heritage. However, as was the case in most other states, the biometric verification process did not work out as planned, and in the end, the reform penalised beneficiaries at least as much as ration dealers, leading to vigorous debates about the usefulness of the reform. Despite their criticisms of the new system, most commenters recoil from demanding a return to the past, since fraud has been omnipresent and is universally seen as a problem. Instead, they suggest reforms of the reform, sometimes by reflecting on past reforms, their benefits and failures. The next section illustrates this recursive process of implementing, failing and fixing, a process that changes the equation without ever reaching a stable end point when the reform reaches its goal of a friction-free system. I am interested in how, from this maze of reforms, a development imaginary emerges that constructs time as linear and sees technical innovation not so much as a rearrangement of relations, but as evidence of progress.
Really Existing AePDS
If Jharkhand, a mountainous state with a patchy telecommunication network, was an unlikely contender for spearheading the introduction of ePOS machines, the urgent need for fiscal saving, as well as the desire to finally improve governance, explains the hasty introduction of a digitally augmented PDS. Jharkhand had emerged as an independent state in 2000, carved out from Bihar after a prolonged struggle for independence. Advocates for the separation emphasised that citizens, many of whom were of tribal decent, deserved better care than they had received from the landowning castes that dominate state politics in Bihar. To prove the point, the new political elite was in a hurry to improve state operations, and PDS was a good point to start: on the one hand, a huge proportion of the population depended on subsidised rations for survival, while on the other hand, the system had a bad reputation, and at its worst times lost as much as 85% of grain through illegal sales. Hence the appeal of Aadhaar.
The Right to Food Security Act brought a ray of hope to the state because the government reversed the burden of proof for issuing NFS cards. By default, every resident of Jharkhand was taken to be eligible for rations, unless a person’s family met one of the exclusion criteria, such as having a family member who worked for the government or owning luxury goods such as a car or air conditioner. On the downside, NFS applications had to be accompanied by the Aadhaar numbers of all family members, at a time when the infrastructure was barely functional (Drèze et al., 2017). As early as 2013, when only one-third of India’s population, or 440 million residents, had enrolled in the biometric database, the first ePOS machine went online in Ormanjhi block in Jharkhand’s capital Ranchi. Gradually, from the urban heartland to the countryside, the new surveillance apparatus spread, and by mid-2017, ABBA had become fully functional across the entire state (Menon, 2018).
Only one year later the press came down heavily on the elite. ‘Not having an Aadhaar could starve you to death’, announced The Hindu dramatically.
Around half-a-dozen people have allegedly died of starvation in Jharkhand in the last six months. Most of them were reportedly denied rations from the Public Distribution System shops for failing to have Aadhaar-based biometric authentication. (Tewary, 2018)
Why did things derail so badly? And did they really? Or were these potentially highly specific local events sensationalised by the news media, and rather than prove the blanket failure of AePDS, did they merely indicate the typical teething problems that would plague any ambitious new system? The answer is convoluted, and the question of the social effects of digitisation is at the centre of a growing body of literature that discusses Jharkhand’s pilot project, AePDS more generally, and Aadhaar and other digitisation projects in India and beyond (see, for example, Abraham, 2021; Allu et al., 2018; Chaudhuri, 2021; Das & Masiero, 2019; Drèze et al., 2017; Hundal & Chaudhuri, 2020; Masiero, 2020; Masiero & Bailur, 2022; Menon, 2018; Mudliar, 2020, 2021; Muralidharan et al., 2020a; Sridhar, 2022). To show why in some accounts the technology emerges as the star in linear narratives of development, it is useful to untangle the specific framings and methodological choices of some of these accounts. Such a dissection of academic arguments helps to lay bare the elements that organise widely popular techno-optimist narrations, which crucially inform the political economy of pro-digitisation measures.
With Aadhaar, PDS had become more and not less cumbersome for citizens. Since ePOS machines were intended to discipline shopkeepers, difficulties for citizens were not anticipated, owing to the naïve assumption that biometric identification is straightforward and easy to use. Yet, users felt penalised, not least because they had to present a biometrifiable finger every time they wanted to pick up their rations. This was easier said than done.
My mother who is around 62 years old or my son who is 14 years old undergo verification at ration shop every month. My son undergoes verification when the thumb does not work for my mother. Given the age of my mother, when the thumb does not work, the dealer sends her back and asks her to clean hands. It takes about half an hour for the thumb to work for her. (Nagaraj & Prakash, 2021, p. 342)
Witness accounts like these testify to the stress and nervousness that have arrived with the ePOS machines. People fear being rejected, and in Jharkhand, half the population faces issues with fingerprinting and accordingly experience time delays during the purchase of rations (Nagaraj & Prakash, 2021). Unbiometrifiablity is a condition of age, and it is linked to lifestyle, particularly affecting manual labourers and farming families. As the above quote suggests, there are remedies, and sometimes they work. If people have a mobile phone, and if their number is linked correctly to the Aadhaar account, they may use a One Time Password (OTP) to circumvent biometric identity verification. If none of these options work, and for the poorest and the elderly they rarely do, clients must forgo their quota of grain (Mudliar, 2021).
While clients found the new routines unnecessary and painful, shopkeepers hated the machines. Although the devices were provided for free, the electricity costs were high and maintenance tedious. The batteries had short lifespans, and shopkeepers had to align opening hours with the irregular rhythms of a capricious electricity supply. They also had to find a spot with a telecommunication signal, and when the device was broken, they had to wait for technicians to arrive. Aside from these issues, ration dealers hated the surveillance and felt they had been demoted from being patrons over the local distribution of grain to mere clerks of a machine, for which they were paid an appallingly low commission (Chaudhuri, 2019; Mudliar, 2021).
For the state, the Aadhaar-seeding of NFS cards brought few, if any, benefits; apparently the problem of ghost beneficiaries was less dramatic than anticipated. Thus, to complete the reform and harvest fiscal benefits, the second step of Jharkhand’s PDS reform had to be rolled out. From July 2017 onwards, ration dealers no longer received the full grain load but were reimbursed only for the amount of grain they had sold in the previous months. This process of downward payment is called reconciliation. In Jharkhand, reconciliation was introduced with retrospective effect. This meant that shopkeepers were made responsible for the entire amount of ration they had apparently saved since the beginning of digital tracking, a decision that acutely impacted those shopkeepers who had first received the ePOS devices. During the first month after reconciliation, they received little or no grain at all, since the computer records showed that they had sufficient savings to cover the upcoming distribution in July. Not suspecting that downward payment would kick in so soon or that it would be applied retrospectively, and following their usual habit, most dealers had long since sold the leftover grain from previous months and so they had little or nothing to distribute when the July delivery was cancelled. Ration dealers panicked, and many successfully lobbied for exceptions, causing chaos in the state and exacerbating inequality. Confusion prevailed in the following months, so that after four months, reconciliation was interrupted, not least because it produced unimaginable hardship for beneficiaries, leading to the aforementioned accusations that the state was responsible for causing starvation (Drèze et al., 2020b). Reconciliation was reintroduced one year later, after much deliberation, and when the state was better prepared.
In the public sphere, multiple interpretations of the system competed for attention and sometimes directly clashed. For example, an open controversy emerged between two teams of researchers—economists and sociologists, respectively—who presented their respective readings of Jharkhand’s real-life experiment in the Ideas for India (I4I) blog. 1 What on the surface appeared to be a typical social-justice-versus-fiscal-discipline dispute upon closer inspection reveals itself to be a competition between different renderings of the role of technology in development. Each of the teams crafted a linear narrative of a movement forward in time, a movement that must or should continue, not least with the help of reflections from scholars themselves. For the sociologists, the arrival of Aadhaar was a cesura, but not the most important one. They point instead to the rights movement at the beginning of the millennium as a first step towards improving the PDS and recommend replacing Aadhaar with more suitable offline technology. For their part, the economists celebrate Aadhaar as a liberation from corruption-prone paper bureaucracy and advocate for more concerted investment in digital surveillance. They admit that the first round of reform fell short of improving customer experience and demand further tweaking the system. Both teams are enthusiastic about change. They dissect the history of making PDS, reflect on past success and failures, as well as loose ends, and argue for a different reform or a reform of the reform, to fix the ills of the system or of past efforts to fix it. In both cases, success appears to be possible and is projected in the future. Yet, while one team holds on to the dream of the technological fix, the other engages in a broader analysis of the political economy of the rights to food movement and assigns technology a secondary role in stories of improving the state.
The I4I debate was kicked off by the Harvard-educated economists Muralidharan, Niehaus and Sukhtankar (Muralidharan et al., 2020b). Advocates of evidence-based policy, they had cooperated with the government in Jharkhand to study the roll-out of ePOS from day one. The government had paced the distribution of the machines to suit the methodology of their study, a large-scale randomised control trial. The results were out in 2020, and to attract attention and maximise readership, the team published summaries of their findings (Muralidharan et al., 2020a) in news outlets. In a celebratory tone, they state that AePDS had enabled Jharkhand to achieve significant fiscal savings, although at the cost of causing some hardships for beneficiaries. Their assessment privileges a perspective from the centre and assesses effects in numerical terms. Their objectification avoids or steers away from the stories of hardship that populate the newspapers, providing a seemingly accurate account of the overall impact.
Reconciliation … led to big reductions in the value of grain disbursed by the government, received by households, and the difference between the two (‘leakage’). The value of grain disbursed fell by 18% in the ‘control group’ (36% in ‘treatment’), with 22% (34%) of the drop representing reduced value received by beneficiaries and the remaining 78% (66%) representing a drop in leakage. The ‘treatment group’ declines were larger because dealers in this group had transaction records for a longer period and were expected to have larger stocks of undisbursed grains. Realising that their ability to divert grains was now reduced, FPS owners in treatment areas reported a 72% lower bribe price that they would expect to pay to obtain PDS licenses. (Muralidharan et al., 2020b, p. 4)
These figures impressively show that savings are possible and that they are a direct outcome of real-time tracking and reconciliation. Distribution to shops dropped by 18%, whereby shopkeepers passed on some of their losses to customers. However, according to this calculation, the entire remaining amount of 72% represents a real reduction in leakage, at the expense of shopkeepers, whereby the thesis of increased efficiency is further confirmed by the reduced bribe price paid by concession holders for PDS licenses. Aadhaar seemed to make the state more efficient, although the costs of the digital infrastructure are not factored into this calculation. However, like other commentators, the team is aware of human suffering and so demands that the system must be reformed and equipped with robust grievance mechanisms. In this regard, Andhra Pradesh and Telangana feature as models (Muralidharan et al., 2020b).
Five months after the intervention by Muralidharan et al. (2020a), I4I published a rebuttal by sociologists Drèze, Khera and Somanchi (Drèze et al., 2020a), who were outraged by the sterile focus on numbers and emphasis on fiscal saving. By contrast, they drew attention to suffering in the villages by reiterating a finding they had originally published in 2017—that AePDS in Jharkhand had created ‘pain without gain’ (Drèze et al., 2017). Known for their social engagement and pro-poor activism, the team had observed the roll-out of Aadhaar in the state during the same decisive years of 2016/2017
2
and offered a broadly negative assessment:
Aadhaar-based biometric authentication is now compulsory for most users of the public distribution system in Jharkhand. Based on a recent household survey, this paper examines various issues related to this measure, including exclusion problems, transaction costs, and its impact on corruption. The findings raise serious questions about the appropriateness of this technology for rural Jharkhand. (Drèze et al., 2017)
The team saw enhanced suffering and no impact on corruption, instead finding that AePDS fuelled new forms of cheating. Beneficiaries suffered in at least two ways: many families lost their access to subsidised food because they were unable to acquire an Aadhaar number or because their bodies were incompatible with going biometric technology; others, those who were able to use ABBA, suffered because the ration shopkeepers passed on the costs of the new system to them, by hoarding illegal grain not primarily through cheating the system but by stealing from customers:
In Beltoli village of Latehar district, we discovered [… that] the dealer there is telling people that the food grain entitlements of priority households—5 kg per person per months—are now restricted to the names (not just the cards) that have been seeded with Aadhaar. Judging from 25 testimonies collected in Beltoli on 1 November 2017, this enables the dealer to distribute just 60% of the prescribed quantity while entering the full amount in the PoS machine. Further inquiries in neighbouring villages suggest that this practice is not confined to Beltoli. We heard of other methods too, including one whereby dealers simply tell people that it is ‘Modi’s wish’ that they should undergo biometric authentication at least once without getting any rice. This enable them to record fake transactions with abandon. (Drèze et al., 2017, p. 58)
The sociologists concluded that technology cannot unmake injustice or easily uproot habituated traditions of domination and submission. This is not to say that the team was against digital technology per se, but they disputed that the state was prepared for real-time tracking. In the case of Jharkhand, they attributed recent positive developments in the local PDS system to the rights movement. Thus, shifting the frame of the historical analysis, Drèze and Khera told a different, much longer and convoluted story of PDS in Jharkhand, its successes and failures.
During its century of existence, food distribution has gone through many reforms, and economic liberalisation in the 1990s is remembered as one watershed moment. It is associated, on the one hand, with the undermining of a formerly generous PDS and, on the other hand, with the popularisation of the social rights movements. Until the mid-1990s, subsidised food rations were available to all citizens regardless of their status and income. Since the quality of FPS grain was generally lower than that sold on the free market, it was assumed that clients would self-select; rich citizens, and everyone who could afford it, would naturally drift towards the high-quality products on the open market. This argument came under attack by advocates of lean states, and in an attempt to save costs at the height of India’s foreign exchange crisis, the universal PDS was replaced by a targeted system (TPDS). A comprehensive below-poverty-line (BPL) survey determined the group of eligible beneficiaries, and from the mid-1990s onwards, only families with a BPL card could apply for rations. The BPL survey had to be repeated several times, and yet it never captured all needy families. Migrants and the poorest of the poor who lacked documentation were the hardest hit because they found it impossible to jump through all the necessary bureaucratic hoops (Mooij, 1999). Not only did poor customers struggle; the sudden massive reduction in client numbers and resulting steep drop in turnover forced many ration dealers into bankruptcy. Those who survived had often found ways to siphon off grain as a way to maintain an additional stream of income. The reform was heavily criticised (Dandurand, 2018, pp. 121–145), and ironically, the complaints resembled those raised against AePDS: the reform missed its target, excluded the most desperate citizens and did not effectively undercut corruption. Furthermore, by unilaterally targeting ration dealers and making the economic survival of their shops impossible, it created new incentives for stealing grain. The reform needed not one but several reforms to better identify needy families and ensure the benefit actually reached them.
An important step forward, in Drèze and Khera’s rendering, is the campaign for a right to food security in the early 2000s, which stretched out over more than a decade; promoted robust debate among farmers, citizens, politicians and bureaucrats; and brought to the fore many interpretations of what such a right might entail. These ranged from a broadest possible vision that included thinking about farmers’ need for secure income, adequate storage facilities for ensuring long-term food security and a nutritious meal for every citizen, to a narrower concern about a more effective PDS. The final compromise was closer to this latter vision, and with the food security act, the number of entitled beneficiaries rose again, though the actual allocation per person was rather slim. Despite the legislation’s limitations, activists celebrated it as a success since it changed the language around welfare. Food no longer figured as a gift handed out by the generous patron state (Dandurand, 2018, p. 16; Khera, 2020) but as a social right. During the long-drawn-out campaign, the word ‘entitlement’ had percolated into people’s everyday language, making everyone aware of their right and empowering them to demand their share (Drèze & Khera, 2015).
Responding to the critique of their approach, Muralidharan et al. (2020c) stood by their cautiously positive assessment of AePDS and once again articulated the need for a shift in perspective, whereby the state would become less fixated on fiscal saving and instead be concerned, primarily, with the well-being of clients. While Aadhaar can reduce corruption, systems must be put in place to ensure that customers’ grievances are attended to. In turn, in their own response to the response, the sociologists, although more sceptical of top-down reform and repeating their critique of a narrow economic perspective (Drèze et al., 2020b), are equally inspired by reforms in more techno-progressive states. Khera (2019) looks to Chhattisgarh and suggests moving to smart card–based solutions for fraud-free reporting. Both teams join a chorus of voices that argue for investment in technology en route to improving the state. By embedding their assessment of AePDS in a historicising narrative of gradual learning, they strengthen the association between technology and aspiration.
Despite the overlap, there is a significant difference between the two accounts. The economists assign problem-solving capacity to technocratic planning by singling out technology as the main driver of change. In turn, the sociologists emphasise the importance of breaking the patronage system as a precondition for achieving social justice. Other more cautious tales of technology leave more space for alternative temporal narratives to emerge, beyond the notion of linear time. A case in point is Veeraraghavan’s (2021) analysis of digital surveillance in Andhra Pradesh’s version of the National Rural Employment Guarantee Act (NREGA). 3 Therein, he uses the term ‘patching’ to describe ‘the process of replacing a problematic set of commands and rules with a new set of instructions’ (p. 4) and explains that this type of iterative repair is used by bureaucrats to plug ever-opening loopholes that undermine oversight of the proper accounting of benefits. Thus, rather than treating technological oversight as in and of itself effective, in this analysis digital surveillance figures as one component in a dynamic surveillance assemblage that not only provides evidence to state agents but should also empower beneficiaries to hold those in power to account. By using the notion of patching, Veeraraghavan both escapes the linear imaginary of the ‘failing forward’ ideology of techno-optimistic accounts and complicates the empowerment narrative by treating empowerment as a fragile achievement that must constantly be defended against new forms of capture. In the case of the AePDS discussion, Nagaraj and Prakash (2021) come closest to this view. They propose the use of ‘complexity theory’ because it enables tuning into local diversity, which they argue is a precondition for developing customised solutions and creating the necessary flexibility to deal with a broad variety of issues.
Conclusion
In this article, ePOS serves as a starting point, not to debate corruption, but to demonstrate the dynamic of state making and how it is interpreted in reflective accounts. ePOS was invented to make ration distribution corruption free, and it has many effects, for example, fiscal savings; it also creates new exclusions, and it has altered the forms and patterns of corruption. While many commentators are ready to see advantages, even the sternest advocates of digitisation demand further reform. By engaging with exemplary altercation between two opponents, this article uses AePDS to shine light on the idealistic forward motion of time in technocratic accounts. By re-casting critique as recommendations about the next necessary steps, hope is recovered, and the arrival of the fruits of reform becomes an anticipation that is expected to arrive in the near future. This orientation most strongly underscores the approach by Muralidharan et al. (2020a, 2020b, 2020c) and is a perspective commonly found in techno-optimist argumentations. I have described how linear time emerges by explaining away complexity and viewing development as the undertaking of steps of learning from failure, which creates a sense of progression and renews the anticipation of success, a success that ultimately never arrives. It is a technique by which political reform and adjustments, failure and fixing, become subsumed under a tale of improvement that justifies the intervention, explains its failure and declares it to be a precondition for improving the technology. When cast in such aspirational terms, this technique is in danger of depoliticising development (Ferguson, 1990) because problems can always be explained away as short-term setbacks that will be fixed by the next reform. Against this trend, more cautionary accounts remind the reader of the power relations that organise the dispersal and usage of technology. Drèze and Khera point to the role of popular agency in struggles for the right to food and treat empowerment as a necessary component of any attempt to curtail corruption. Acknowledging the irreducible complexity of social relations leads some to steer clear of linear accounts of time and treat empowerment and improvement as fragile, socially contested, achievements (Nagaraj & Prakash, 2021; Rao, 2022; Veeraraghavan, 2021). Thereby, they escape the one dimensionality of the failing-forward ideology that is so typically found in optimistic development accounts and often underscores the propaganda that justifies the purchase of expensive equipment and implementation of more surveillance.
Footnotes
Acknowledgements
I am grateful to the two anonymous readers and the editor of this volume for the careful reading of my text and the valuable comments, which helped me sharpen my argument significantly. I would also like to thank the members of my department at the Max Planck Institute for Social Anthropology for discussing an earlier version of this article and providing useful input.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author received no financial support for the research, authorship and/or publication of this article.