Twenty years ago, it was sometimes difficult to explain the importance of improving records management. Sceptics wondered whether it was worthwhile sorting through floor-to-ceiling stacks of old government files in musty Asian or African offices. They had to be persuaded that orderly sets of records could restore the memories of governments. However, effective records management facilitates government driven by precedent and not by the whims of politicians and bureaucrats. It promotes the rule of law. It prepares the ground for greater government transparency and accountability and for efforts to curb corruption. It enables governments – and citizens – to see which earlier and current policies and programmes worked, and which have misfired.

Moreover, times have changed. The old arguments about the virtues of managing paper-based records management are still valid, and it is still worth wading through heaps of old files, but remarkable advances in information technology (IT) have inspired euphoria about the immense promise of digital records, and many political leaders have embraced IT as a potent weapon in their efforts to promote development programmes and prevent corrupt bureaucrats and middlemen from plundering them. But so have politicians who seek tight top-down control of governments and populations.

Amid the excitement about IT, a reality check is in order. It is possible because solid evidence abounds demonstrating that well-managed digital records can contribute mightily to constructive programmes but that, if mismanaged, digital records can be the source of grave damage.

This chapter considers both phenomena by examining two gargantuan government initiatives in India. First, it briefly summarises key findings from a major study of the immense, positive contribution of well-managed digital records in the world’s largest poverty programme, the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA). Then, at greater length, it presents evidence from a diversity of reports on Aadhaar, a universal identification programme in which a government’s objectionable aims, haste and bullying have caused hardships, hunger and even deaths among many extremely poor people, compounded by poor digital records management.

The Mahatma Gandhi National Rural Employment Guarantee Act

Early in its decade in power (2004–14), the multiparty United Progressive Alliance led by the Congress Party passed the National Rural Employment Guarantee Act.¹ It gave every household in rural areas – where two-thirds of Indians live – the right to demand and receive up to 100 days employment per year, doing manual labour on public works sites at a reasonable wage. It was intended as a hedge against destitution, and the response was massive. At this writing in February 2018, 24.85 billion ‘person days’ have been worked under this programme. On 31 January 2018, for example, labour was performed on 947,824 sites.² It is thus the largest poverty reduction programme in history. Since this initiative has been exhaustively analysed elsewhere,³ what follows is a concise summary of the main relevant points.

The wages earned have made an immense material difference to poor families. One day’s earnings enable a family to purchase enough subsidised food to provide two decent meals per day for a month. Wages have also been spent on health care, education, livestock, small plots of land and microenterprises – all of which have eroded bonds of dependency on richer villagers and the resultant subordination of poorer people. Since many key decisions about the programme are made by elected local councils – which receive at least half of its substantial funds – poor people have become more active in the public sphere. That has enhanced their political awareness, confidence, skills and connections – which add up to greater ‘political capacity’ to defend their interests. Many women – who have performed most of the work – have been drawn into the public sphere for the first time. And since they receive wages directly to their bank accounts, their influence within and beyond their households has been bolstered.

The MGNREGA was shrewdly and effectively formulated by people with a realistic, sophisticated understanding of local power dynamics and of how pilfering has occurred in other government programmes. A formidable IT system records every event in each worker’s engagement with the programme: the date each demanded employment, the date on which it was first provided, compensation due to each worker for delayed provision of work, each day’s work performed, the wages due to each worker, and so on. Each worker is also given a ‘job card’, a small booklet on which the same information is recorded manually. It can be checked against the electronic record to detect efforts to cheat workers. The IT system is thus a potent transparency mechanism, and it is supplemented by other devices to promote transparency.

Enlightened, dynamic civil servants in New Delhi took steps to overcome logistical impediments such as poor internet connectivity in villages – problems that have greatly damaged the second programme analysed below, Aadhaar. For example, they overcame the problem of poor internet connectivity and power supplies in many villages by mandating the transfer of handwritten local records to IT centres for insertion into the electronic record. They also interrogated field researchers in order to discover flaws in the programme, and when these became apparent, they took corrective action.⁴ As a consequence, it became more difficult to steal funds from the MGNREGA than from almost any other Indian government programme.

Thanks in substantial measure to the well-managed use of IT in this enormous exercise in records management, this demand-driven programme has been immensely constructive in reducing poverty, promoting rights and village-level democracy, and undermining invidious local power dynamics across India.

Aadhaar

During the national election campaign in 2014, opposition leader Narendra Modi denounced Aadhaar as a ‘political gimmick’, but when he became prime minister in May of that year, he soon warmed to it. He is an enthusiast for technocratic approaches to government. He is also genuinely interested in reducing corruption.⁵ He regarded Aadhaar as a device to curb spending on poverty programmes that had soared under the previous government, by reducing the illicit diversion resources from them – in order to redirect funds to investments in infrastructure and other spheres that might accelerate economic growth. And crucially, as he radically centralised power in the prime minister’s office, he also saw Aadhaar as a means of achieving greater top-down control.

He wanted to show that he – unlike his staid predecessor – was a bold man of action. So, he pursued this initiative (and many others) particularly aggressively. To maximise the impact of Aadhaar, he sought to make it mandatory for all manner of things: access to subsidised food and fuels; pensions for vulnerable groups; work opportunities under the aforementioned MGNREGA; school enrolments and free midday meals for school children; mobile telephones; bank accounts; income tax payments, and much more.

The breadth of its coverage, the government’s refusal to make it voluntary, and the hasty, forceful pursuit of universal enrolment have raised serious concerns. We need to consider anxieties about possible leaks and the illicit penetration of the Aadhaar system; the government’s coercive approach; malfunctions and denials of services which have caused acute hardships; and the opportunities that it has created for new kinds of corruption.

Leaks and the system’s vulnerability to penetration

Worries about the security of Aadhaar data have long been expressed, and in 2017 they intensified when it emerged that ‘a simple Google search’ could give access to ‘thousands of databases, that contain Aadhaar numbers with other sundry personal data’. Another report listed nine government departments and educational institutions that had leaked data.⁶

In March 2017, India’s IT minister (echoing the prime minister) assured parliament that ‘there is no leak’ and that Aadhaar data are secure, but his statement was wildly inaccurate. In the three preceding months, at least 13 leaks had occurred. They included cases that affected (respectively) 12,000, 30,000, 500,000–600,000, and one million people. Twelve of those leaks were the fault of central or state government agencies or organisations associated with them.

The authority that oversees Aadhaar has exclusive powers to take legal action when such breaches occur. The Aadhaar Act⁷ explicitly denies citizens the right to sue for damages as a result of leaks. If a fraudster uses an individual’s details to steal government benefits, the authorities are not obligated to inform the injured party – and often fail to do so. If citizens learn of this, no procedure exists for them to obtain new numbers. Most attempts to use India’s strong Right to Information Act to uncover breaches in the Aadhaar system and the defrauding of individuals have been rejected – on the basis of a section of that Act which refers to ‘national security’.⁸

The Aadhaar Authority has initiated criminal proceedings against numerous private parties for leaks, but very few against government officials or agencies.⁹ This is a serious omission. Chandigarh’s Department of Food and Civil Supplies reportedly publicised the numbers of 490,000 Public Distribution System beneficiaries, quite vulnerable people. In Jharkhand state, personal details of 150,000 pensioners appeared on the Directorate of Social Security’s website. Another government website leaked data on 500,000 minors. A programme officer at the Aadhaar Authority said that several banks (some of them state-owned) were displaying customers’ data, which might enable someone to drain their accounts. Most spectacularly, he also found that India’s Ministry of Rural Development website revealed the Aadhaar details of 100 million MGNREGA workers – again, extremely poor people. Eventually, 210 government websites were found to have displayed ‘the list of beneficiaries along with their name, address, and other details and Aadhaar numbers’. When this was discovered, they were taken down, but they had been up for some time.¹⁰

There are good reasons for the authority to worry about abuses by private sector companies and banks, many of which saw opportunities for exploitation. Even before the Modi government’s Aadhaar law came into effect, a private company was advertising its capacity to use it ‘to verify your maid, driver, electrician, tutor … and everyone else instantly’. As the Bill was being discussed in parliament, the government revealed that ‘just about any person or company can draw on the Aadhaar system for its purposes. There are no qualifications on who may use it and why’.¹¹ This soon led the authority to investigate three firms – including a major bank, Axis – ‘for attempting unauthorised authentication and impersonation’ using stored Aadhaar biometrics.¹² A technology lawyer described parallel databases constructed by private firms like Jio, a formidable player, as ‘a goldmine for identity theft and fraud’.¹³

As late as December 2017, the authority was claiming that their database had never been hacked. But then it emerged that Airtel, a major telephone company, had compromised the system by misusing data. It took advantage of a government requirement that all SIM cards had to be linked to Aadhaar by 31 March 2018 – part of its excessively aggressive effort to force mass enrolment (see below). Airtel was opening bank accounts for users without their consent. That posed a danger that funds due to them under government programmes would go to these accounts so that genuine beneficiaries would never receive them.¹⁴

The system is clearly vulnerable to criminals. A gang of ten in Kanpur was arrested after the authority discovered that they had cloned Aadhaar client applications which supposedly can only be filed at an authorised centre. They obtained the centre’s employees’ fingerprints (needed to log in) by using ‘butter paper’ and sold replica application forms for Rs 5,000 (US$78.25) each. They also obtained the system’s source code and tampered with it to enable them to bypass other biometric protections like iris recognition. Some employees of Aadhaar centres were also fraudsters. The Authority ‘blacklisted 49,000 operators for corrupt practices’.¹⁵

It has also sought to hush up leaks and misdeeds – vowing to ‘take action against an individual for reporting a security vulnerability in Aadhaar’.¹⁶ As one commentator complained, it ‘should be rewarding those who find breaches – instead, we have attempts to intimidate them into silence through the abuse of the state’s police powers’. The Tribune newspaper paid a mere Rs 500 (US$7.83) to an agent who created a gateway with a login and a password and was thus able to get the full details of each of the one billion people enrolled. For a further Rs 300 (US$4.70), he could print official cards for any number of them, including photos and addresses. For another Rs 300, he supplied the newspaper with the tools to do likewise. The Aadhaar Authority initially responded by insisting – incorrectly – that no breach had been achieved. Then, instead of conducting an internal investigation, it initiated criminal proceedings against the reporter who wrote the story. The Editors’ Guild of India condemned this as an attempt to ‘browbeat’ the journalist.¹⁷ After a complaint from the authority, New Delhi police registered a criminal case against a man merely for saying on social media that the ‘Aadhaar ecosystem is flawed, vulnerable, has very poor security, and can be easily hacked’.¹⁸

Coercive action by a government in a hurry

Prime Minister Modi gives the impression he would like to be perceived as a dynamic leader who lets little stand in the way of the changes he forces through. That is how he has managed Aadhaar. His enabling law for the programme was passed by the lower house of parliament in just three hours, and all amendments were rejected. Then the speaker controversially designated it a money bill so that the upper house could not amend it. An attempt was made to revise a clause that allows any individual or private or public organisation to use Aadhaar data. Members – anticipating the problems noted above – tried to restrict this to government agencies, but all proposals from the upper house were rejected.¹⁹

Thereafter, Modi undertook a forceful drive to maximise enrolment. The government’s main tactic was to require citizens to register in order to obtain benefits from welfare programmes. This occurred despite two 2015 Supreme Court rulings – which in India are ‘law declared’, they have the force of law – that Aadhaar could not be made mandatory. In one case, the government asked the Court to require it for 88 social welfare programmes, but justices made it voluntary for only four.²⁰ In August 2015, the Court directed that a major publicity campaign be launched to explain that Aadhaar ‘is not mandatory’. It was ignored.²¹ In March 2016, the Court reiterated that officials could not make it mandatory for any subsidies, benefits or services, and stressed the point again in a ruling in October. In July, it sent letters to all state governments in this federal system – which were under heavy pressure from the national level to make Aadhaar mandatory – reminding them of its earlier judgments, and it re-emphasised the point in October. But those governments pressed ahead regardless, on many fronts.²²

In March 2016, parliament required enrolment for an array of government subsidies and benefits.²³ By early 2017, many schools – for example, 2,700 in Delhi – began demanding that both parents and prospective students have Aadhaar ID numbers and that students must open bank accounts before they could gain admission and receive educational benefits. That enabled schools to exclude slum dwelling and migrant children. In March 2017, the central government took a further, gravely damaging step: insisting that children would not receive free midday meals – on which a vast number of hungry families depend since pupils can take food home from school – unless they presented Aadhaar numbers.²⁴ This posed a grievous threat since Aadhaar enrolment among destitute families was very low – for example, only 17 per cent in Meerut District of Uttar Pradesh. Officials overseeing midday meals were exasperated but said that they were helpless to resist.²⁵

The government then announced that pregnant women would not receive maternity entitlements at state-run health centres unless they were registered with Aadhaar. Then, in its headlong drive for enrolment, the government began requiring newborn infants ‘within minutes of birth’ to sign on. In some cases, officials made this mandatory before issuing birth certificates.²⁶ The government also required Aadhaar certification for 11 other welfare programmes, including the Public Distribution System that provides subsidised food to over 800 million poor people, subsidised gas for homes, the National Social Assistance Programme, the massive MGNREGA and a programme for disabled people.

Education activists stressed that making Aadhaar mandatory for access to schools was a violation of the Right to Education Act. Others referred to Supreme Court orders covering many services. But the drive for enrolment continued apace. A legal scholar, Usha Ramanathan, stated that ‘They are making it clearer and clearer that the Unique Identification project is not about including or reaching one’s entitlements, but coercion and exclusion’.²⁷

The government responded to such concerns with blatant falsehoods. On 21 December 2017, an official reply to a question in the upper house of parliament stated that it was ‘not mandatory for a beneficiary to avail subsidies, benefits or services’.²⁸

Given the risks to destitute human beings and, not incidentally, to the government’s popularity, one commentator asked, ‘What is the extreme urgency?’ The rush was apparent from the silence of senior officials in state-level food departments who saw the damage being done but felt powerless, as a result of ‘tremendous pressure’ from New Delhi.²⁹ The commentator then asked: ‘Why the sneaky tactics?’ The latter included – again in violation of Supreme Court rulings – a last-minute amendment to the Finance Bill 2017, making Aadhaar mandatory for filing taxes. The commentator concluded that by requiring enrolment in so many areas, the government sought to force the Supreme Court to permit the mandatory use of Aadhaar because the system would become too big to fail, and to undo.³⁰

‘Inhuman and illegal’:³¹ malfunctions and denials of services cause hardships

The aggressive pursuit of mass enrolment has not been matched by efforts to simplify the process of ‘seeding’ – that is, linking the system with documents that demonstrate a person’s eligibility for services and benefits. Studies by the eminent economist Jean Dreze have shown the process to be ‘cumbersome’ and prone to error – for example, if a person’s name is spelled slightly differently in different databases. Even middle-class people have struggled with seeding when they tried to link Aadhaar numbers with numbers used to pay taxes. The vast majority of poor people – many of them illiterate and nearly all of them unskilled in navigating bureaucratic minefields – have found the process too complex and thus have been excluded.³²

To make matters worse, glitches in computer networks, inadequate training for those implementing the process, poor connectivity to internet signals, power failures (a particularly serious problem)³³ and more have prevented many from achieving ‘seeding’. Even people whose data have been ‘seeded’ have found that these problems stopped them from obtaining vital services and resources. The machines intended to identify beneficiaries by fingerprint – or much more unusually iris – recognition often fail to work, leading to people being turned away. The gnarled, cracked fingerprints of manual labourers and the elderly – who are often the people in the most urgent need – have been especially hard to recognise.³⁴ Even Nandan Nilekani, an iconic figure in India’s IT sector who stoutly defended Aadhaar during a long spell as head of its authority, eventually conceded that action was needed to address these problems.³⁵

As an exercise in records management, the Aadhaar system has been a woeful disappointment. It is expected to run before it can walk, but the feverish drive for universal implementation rolls on unabated. Shopkeepers who take pity on poor people who are excluded by malfunctions in the system, by using manual overrides, face retribution from higher officials. The grandson of one in Rajasthan said, ‘The signal is patchy and (when they get the signal) the internet stops working’. But when overrides were used, officials issued a ‘show cause’ notice, threatening punishment.³⁶

The prime minister’s pressure for quantifiable results has persuaded or forced some bureaucrats into drastic actions that have caused immense suffering. For example, the head of the Jharkhand state civil service claimed that ration cards and MGNREGA job cards that had not been successfully seeded were ‘fakes’ – even though the complexity and unreliability of the Aadhaar system caused much and probably most of this problem. He therefore barred those holding ‘fakes’ from receiving such benefits as subsidised food, pensions and work opportunities under MGNREGA. That caused ghastly suffering among huge numbers of poor people.³⁷

Prime Minister Modi has celebrated such ‘discoveries’ of ‘fakes’ as victories in the fight against corruption. In February 2017, he proudly told parliament that major gains had been made against ‘bogus’ ration cards. One of his ministers then informed the house that 700,000 ‘fake’ cards had been discovered in the state of Odisha. But an official reply by that state’s food department to a Right to Information petition indicated this was not the case. If they were excluded by the Aadhaar system, which seems very likely, another gross injustice had been done.

In a chapter of this length, it is only possible to refer to sources on the epidemic of glitches and malfunctions.³⁸ But to grasp the appalling scale of these problems, consider evidence provided from official sources in just two of India’s 29 states. In Andhra Pradesh, a state government study of access to subsidised grain found that ‘technical hiccups are depriving the poor of their access to food’. The problems identified included ‘glitches, lack of training (of staff in fair price shops in how to operate the point of sale devices) and mismatches’. As a result, 58.6 per cent of beneficiaries ‘couldn’t access their ration quota. The scale of the problems is astounding’.³⁹ Similarly, in Rajasthan, a government official stated that machines used to confirm beneficiaries’ eligibility worked for only 45 per cent – excluding the rest.⁴⁰ Many millions were excluded in these two states, and vastly more suffered across the rest of India.

The results of all this are alarming. Five people who were denied food from the Public Distribution System starved to death in Jharkhand state. The government responded by announcing that enrolment would not be mandatory for access to food, but in many parts of the country, nothing actually changed.⁴¹ An eminent analyst reported that in Rajasthan, people were ‘facing hunger and starvation’.⁴² Three deaths from starvation also occurred in Karnataka, where people denied rations were eating roots and leaves.⁴³

Officials’ reactions to these cases were astonishing. In Jharkhand, they claimed that one of the five deaths – of a young girl – was the result of malaria, even though ‘right to food’ campaigners found that she had not eaten for eight days. Her mother was pressed to say that malaria was the cause. Her refusal triggered a crass campaign of victim blaming. The state chief minister claimed that by saying her daughter had starved to death, she had given the state a bad name. Others claimed that she had brought shame to the nation. Her family was subjected to a social boycott. No one would employ them or sell them anything, and she was heckled when she appeared in public.⁴⁴

Other officials also behaved outrageously. After the three deaths from starvation in Karnataka, civil servants in the district where they had occurred, facing heavy pressure to push ahead with Aadhaar, cancelled 40,000 ration cards, claiming that they were ‘fakes’⁴⁵ even though they were mainly the result of failures in the Aadhaar process.

Reports of deaths from starvation are alarming, but a similarly ghastly outcome has been the malnutrition suffered by vast numbers of poor families. These are people who have been denied subsidised food, and/or whose children have been barred from school enrolments or, for those who attend school, from midday meals there. Even before Aadhaar, malnutrition was at appalling levels. India ranks 100th out of 119 countries on one hunger index⁴⁶ – below Myanmar, Nepal and Bangladesh. Denials of subsidised food grains under Aadhaar have compounded this ghastly problem.

The most shocking aspect of this is its impact upon children. Chronic malnutrition has caused 38.7 per cent of Indian children under five to suffer from stunting (low height-for-age) – a higher proportion than in most sub-Saharan African countries. If they suffer malnutrition during their first two years, their bodies and brains do not develop fully. The damage is irreversible. This means, to put it bluntly, that over a third of Indian children are less intelligent than they might be.⁴⁷ That is both a grotesque injustice and – to put it in terms that might attract Modi’s attention – the squandering of a major national resource. Here again, Aadhaar has made things worse. It did so not just by denying beneficiaries food, but also by denying young mothers access to ante- and post-natal care if they had not successfully enrolled.⁴⁸

Hundreds of leprosy patients in Andhra Pradesh have been impeded by Aadhaar from obtaining rations and modest government pensions. One who had no fingers had her iris scan rejected. Others who still have fingers failed fingerprint tests.⁴⁹ Across India, many HIV patients are dropping out of programmes supplying life-prolonging medicines because they fear that the system may leak information about their condition – leading to shaming and social exclusion.⁵⁰ The government claims that Aadhaar enrolment is not required in antiretroviral therapy centres, but in practice many demand it.⁵¹ In the drive to reduce ‘fake’ documents, many poor villagers have been denied the right to work under the MGNREGA – severing a lifeline for destitute people. And many who have performed manual labour under it have not received wages due to them as a result of Aadhaar malfunctions.

The Supreme Court has asked how homeless people – of whom there are 1.77 million in Uttar Pradesh state alone – will access Aadhaar since they cannot provide proof of address. One justice asked ‘Does this mean that they do not exist for the Government of India?’ The Aadhaar Act clearly states that an address is not required as proof of identity, but in practice, it is often demanded. The government has been reluctant to assist the homeless by clarifying this, because it might slow down the pace of enrolment.⁵²

An immense number of vulnerable people – the elderly, widows, the disabled – depend on modest government pensions. Many have had nightmarish experiences. In Rajasthan the government boasted of a major saving in expenditures after stopping pensions for a vast number of pensioners who could not achieve enrolment. Some who succeeded in linking Aadhaar numbers to bank account numbers (as is required) found that coding errors sent their pensions to others. Many were declared dead although they were very much alive. An investigation by a leading newspaper found that over a third of the state’s 297,000 pensioners had been incorrectly declared dead. A senior journalist said that ‘In village after village, we found that more than half of those declared dead were still living’. The state government paid arrears to people who showed that they were wrongly denied benefits in only 33 out of more than 29,000 cases.⁵³ Jean Dreze argues that pensions – together with ration cards and MGNREGA job cards – have been discontinued ‘just to meet the “100% seeding” targets’ set from on high.

Food provisions for impoverished families have been reduced in three other ways. First, if only three of five members of a family are registered, their provision is cut by 40 per cent. Second, in some states, a so-called ‘Direct Benefit Transfer’ programme disburses funds for food to bank accounts, but that requires the poor to buy grain at 32 times the subsidised rate. In many cases people are not told that those funds have been transferred to bank accounts.⁵⁴ Finally, glitches reduce the amounts of food provided. For example, one former bonded labourer who had previously received 35 kilograms of wheat per month for his family found the provision cut to only between five and ten.⁵⁵

Curbing – and enabling – corruption

Aadhaar is intended to reduce corruption, but its record is at best mixed. Some gains have been made. The authority’s blacklisting of 49,000 operatives for malfeasance was noted above.⁵⁶ The system also detected a scam in one sub-district of Karnataka which received 42,000 litres of kerosene for distribution, as against an average in others of 80 to 100 litres.⁵⁷

Ministers’ celebrations of gains in the struggle against corruption are based on a breathtaking misperception. They overlook the fact that identity fraud represents a tiny proportion of overall corruption in the Public Distribution System. The main problems are the provision of less than prescribed quantities of food (with the surpluses sold at market prices), and the substitution of poor-quality food (with better quality products being sold privately). The government has avoided the implementation of grievance procedures through social audits (local-level hearings) which are supposedly mandatory features of multiple social programmes, and which could reveal quantity and quality fraud.⁵⁸ Modi and his colleagues are either unaware that these are the main sources of corruption, or they fight shy of tackling them. Instead, we hear statements – based on the aggregate numbers of allegedly ‘bogus’ ration cards – that Aadhaar has saved the government US$9 billion ‘by eliminating fraud in beneficiary lists’.⁵⁹

As Jean Dreze’s analyses demonstrate, in the main, Aadhaar has increased corruption. Since many beneficiaries have been denied subsidised food because they have been unable to complete the complex process of linking ration cards to Aadhaar, because those linkages have been faulty, or because of widespread glitches in the system, proprietors of ration shops have been left with substantial surplus food. Some of them pass it to those excluded in exchange for bribes. Many of them sell it illegally for higher prices on the open market. He also found that many are able to distribute less than 100 per cent of the rations due to beneficiaries without being found out by Aadhaar. Some of them have told beneficiaries who are fully registered that it is ‘Modi’s wish’ that they undergo one ID exercise without receiving rations – and then they sell that food. In all of these cases, the system has provided new opportunities for corruption.

Dreze notes that the state government website in Jharkhand has shown that 10 per cent of beneficiaries have been unable to obtain food. He believes that the real number is ‘probably more’, but even if the figure is only 10 per cent, it means that 2.5 million mostly vulnerable people have been victimised. The number for the whole of India is far more massive – as evidence noted above from other states indicates – and so are the opportunities for shopkeepers to make illicit profits. He thus concludes that the new system has entailed a ‘revival of corruption’.⁶⁰

Under the Congress-led government (2004–14), the inept drafting of enabling laws for certain programmes prevented them from having the constructive impact that was intended.⁶¹ That was not true of the two Acts under discussion here. Both were carefully drafted and the aims of their architects were apparent in the two texts. But the aims differed, and so did the consequences that ensued.

The MGNREGA has demonstrated that a meticulously designed law that includes an IT system that is carefully crafted to enable poverty reduction can make an immensely constructive impact. The drafters had a realistic understanding of political, logistical and infrastructural constraints, and took steps to minimise them. They also knew that an IT system could not, on its own, ensure adequate transparency, a core element in the programme. So they supplemented it with other transparency mechanisms. When shortcomings were discovered, enlightened civil servants in New Delhi tackled them. This explains why it is harder to siphon funds from the MGNREGA than from nearly all other government programmes, and why it has enabled such vast numbers of poor people to earn funds that they badly need. Finally, the energy that drove the MGNREGA came mainly from below – from those poor people in this demand-driven programme – who were responding to important new opportunities and not to coercion from above.

The damage that Aadhaar has done is not the accidental result of implementation that diverged from its architects’ original aims. It has emerged from the intent of the Act which is apparent from its text: most notably, to promote radically centralised control. It is also a consequence of the forceful pursuit of those aims by Modi and his government, and their haste to maximise enrolment even before they had provided adequate training to implementers and adequate infrastructure – reliable machinery, internet connectivity and electricity supplies. If the MGNREGA is demand-driven from below, Aadhaar is command-driven from the apex of the political system.

Haste and the drive for control are apparent on several fronts. The Act was rushed through parliament after severely limited debate in which all amendments were brushed aside. Only the Aadhaar Authority is empowered to take action when violations occur, or when citizens experience injustices. The authority and the government often fail to reveal such violations and injustices – even to citizens who suffer them. Right to Information petitions have been ignored on dubious grounds of ‘national security’. Critics of the system have been threatened with criminal charges. Heavy pressure on civil servants and shopkeepers in the Public Distribution System to ensure swift implementation has forced or persuaded many to join in the headlong drive for enrolment. Others who see the damage being done have been prevented from stopping it – and some who have sought to help victims by disregarding Aadhaar procedures have been punished.

All of this has been done in patent violation of multiple Supreme Court rulings that Aadhaar must be voluntary. At this writing in February 2018, the court has not found the government in contempt. Such a finding would carry serious penalties, but it would also entail a constitutional confrontation. The court has threatened charges of contempt on one previous occasion under the Modi government. In September 2017, after over two years in which Hindu extremist ‘cow protection’ vigilantes had indulged in an epidemic of beatings and murders – mainly of Muslims, but also of government officials and policemen – the court gave state governments one week to crack down. They complied, and the atrocities largely ceased.⁶² It remains to be seen whether the Supreme Court will take similar action on Aadhaar.

Aadhaar has failed to achieve the goals of its architects in one sphere: curbing corruption. It has had that effect at times, but for the most part, it has facilitated an increase in familiar forms of thievery and enabled the emergence of new ones. But the government has either failed to recognise this or has chosen to tolerate it as a price worth paying.

These two cases – MGNREGA and Aadhaar – demonstrate that advances in IT have made digital records a more formidable force than ever before. But they also indicate that records can serve both constructive and destructive purposes. Excitement about its positive impact is plainly warranted, but so are anxieties about its Orwellian potential.