Identity Crisis

Why does India make it so hard to prove who you are?

The Living Dead of India

Last year, a man named Baburam Bhil entered a local school near his home village in the state of Rajasthan, armed with a knife. He proceeded to stab a few teachers, take students hostage, and generally menace everyone until he was apprehended by the local police. This wasn’t a normal stabbing – it had an interesting motive. Apparently, Bhil found out that he had been declared legally dead. The government had a death certificate for him on file. This, naturally, made life slightly awkward. When Bhil’s attempts to fix the problem through the bureaucracy went nowhere, he resorted to committing a crime to force the state to recognize that he was still very much among the living; he hoped, perhaps naively, that they couldn’t actually prosecute him without declaring him alive first.

The case was bizarre enough that it made news across India, and even the world (assuming the New York Post counts). Interestingly enough, this is far from the first such incident in India. Cases like Bhil’s are common enough that they formed the plot of the 2021 Bollywood movie Kaagaz (literally ‘papers’). The causes vary: sometimes bureaucratic mistakes have been blamed, sometimes it’s supposedly relatives trying to get their hands on land or property. There have been enough cases – often in North India, where populations are large and the legal system can be especially sluggish – that victims even have their own advocacy group, the Association of the Dead. The group believes there to be thousands of cases in the state of Uttar Pradesh alone; the true number nationwide is unknown. With many of these people lacking the money, political reach, or publicity to level successful legal challenges against the government, cases often move at a glacial pace if at all – the founder of the group was ultimately declared ‘alive’ after eighteen years and countless publicity stunts.

Who says India isn’t a land of miracles? The dead still walk and talk here. They even hold protest marches.

Identity Crisis

There’s a mantra that’s familiar to any Indian who’s ever dealt with the country’s officialdom (which is every Indian), chanted over and over with all the reverence of a ritual:

Aadhar copy.

PAN copy.

Passport copy.

Birth certificate copy…

And so on. Sometimes they throw in interesting little twists, like “electricity bill copy” (even though many people won’t have their names on their bill.) But the same patterns always emerged. Identify yourself. Prove you’re somebody. And if you can’t – good luck getting anything done. It’s a rare Indian that has every single one of these documents. For a large chunk of independent India’s history, most citizens might not even have had any of these. Even today, 1 in 10 Indians – or less – has a passport (for comparison, half the US does, and over 8 in 10 Brits).

Identity documentation has two purposes. For the individual, it’s meant to be a key that opens the door to public services, civic rights, and everything else that a legal status is meant to confer. For the state, it’s a means of verification – proof that someone is who they say they are, that they deserve to walk through that door. The problem is that these two functions aren’t always aligned. For the individual, what’s important is convenience – it should be easy to get your hands on an ID document. For the state, it’s security – making individuals prove that they have the right to a particular ID document. The more difficult it is to get an ID document, the less likely it is that someone who doesn’t have the right to it will obtain it.

Therein lies the paradox. Ease of access is inversely proportional to security. The citizen wants ID documents to be very, very easy to get. The state has an interest in making them as hard to get as possible.

Enter Aadhaar, India’s way of trying to thread this needle. The Aadhaar card was developed by a new organization, the Unique Identification Authority of India (UIDAI) headed by the universally respected Nandan Nilekani, founder of tech services giant Infosys; the Aadhaar card was launched with a storm of publicity, and projected to be the world’s single most widely issued identity document. It was trumpeted as a visionary idea that would ensure every Indian could easily identify themselves and access the services to which they were entitled, at long last.

Did it work?

Image courtesy Wikimedia Commons, CC-BY-SA 4.0

Aadhaar: the Good

It would be unfair to dissect Aadhaar’s flaws without making an honest attempt at pointing out the good it has brought – so here are two things Aadhar genuinely has done right:

1. Give people ID.

As late as 2009, when Aadhaar was launched, hundreds of millions of Indians lacked ID documents like a birth certificate or ration card. Whatever Aadhaar’s faults, distribution has genuinely been wide – it took a while, but well over a billion Aadhaar cards have been issued. For millions of people, an Aadhaar card may well have been the first piece of ID documentation they ever possessed. There were a number of attempts, in fact, to make sure that Aadhaar cards would reach populations who needed them the most, like India’s multitudes of urban homeless, or residents of slums and other informal settlements. Theoretically, it’s possible to receive an Aadhaar card without presenting a single piece of ID documentation. ‘Proof of address’ and other requirements can be waived by the use of a ‘care of’ address, or through an ‘introducer’ – a government official, social worker, or similar. Aadhaar enrollment drives have been carried out in orphanages, homeless shelters, and charity/NGO-run institutions to specifically target underserved groups. Whatever else one might say, there have been genuine and sustained efforts to make Aadhaar enrollment inclusive.

2. Be useful.

Aadhaar was built on the promise of access – an ID that would let people in. In many ways, it has done so. People can now use Aadhaar to open bank accounts, get medical treatment, receive scholarships, pensions, subsidies, and cash transfers, verify income tax returns, enroll in schools or universities, and more. For Indians who already had forms of ID, this is simply convenient; for those who did not, it may well have been transformative.

These two cover the main goals of Aadhaar. Its creators were well aware that ID is access. It is the means by which citizens assert their existence in the eyes of the state, and demand access to the privileges, benefits, and protections to which they are entitled. Aadhaar was meant to make it easy for Indians to assert themselves – it was meant above all to improve inclusion and make sure nobody was incorrectly left out of the system.

At least, in theory. The problem is, the Indian state asserted in the beginning that Aadhaar wouldn’t be mandatory for any of these functions. It has largely failed to keep that promise.

And therein lies the problem.

Finally, the Entrée

Let’s talk about food. India’s Public Distribution System is the world’s largest distribution scheme, and possibly the largest single social welfare system in existence in terms of people covered. It provides 800 million people – the population of the United States and the entire European Union combined – with a monthly allotment of free or very heavily subsidized grains, oil, and sugar through a network of ~500,000 Fair Price Shops (better known as ‘ration shops’) across the country. Ours is a country still haunted by the spectre of famines that assailed us both before and after we became independent. The PDS is the Indian state’s guarantee that mass starvation is a thing of the past; a promise that in a free India, nobody will lack basic sustenance.

Access to the system was governed by the possession of a ration card, which are prevalent enough that they’re also a commonly accepted form of ID. Having the card on you, and presenting it to the shopkeeper at the Fair Price Shop for visual verification, was enough to get you your allotment for the month.

Then, in 2012, the government began requiring the PDS to integrate Aadhaar-based verification. Proponents laid out a number of justifications. It would reduce the vast leakage and fraud that permeate the system; it would allow for better data collection and targeting; it would, importantly, allow for the portability of ration cards – with a the security that came with Aadhaar, the same ration card could be used anywhere in the country, where India’s millions of migrant workers once had to cancel their old ration cards and apply for a new one if and when they moved across states (this was dubbed One Nation, One Ration Card).

Under the new system, just having a ration card, or an Aadhar card, was no longer sufficient. Even having both wasn’t enough. Now, ration cards had to be ‘seeded’ – i.e. linked to – Aadhar cards. And even then, customers at ration shops or other PDS locations became subject to biometric verification on site – they’d have to scan their fingerprints, and would be served only if their details matched with those on the central database.

How does this work in practice? Let’s put this narratively: you’re one of India’s millions of rural poor, and you’re trying to get your hands on the rations you’re entitled to from the Public Distribution Scheme. You get your Aadhaar Card, and you believe that, along with your ration card, should be sufficient. But then your state’s government tells everyone that needs to have their Aadhaar Card ‘seeded’ – i.e. linked to their ration card.

Easy enough. Except there are just so many things that can go wrong. Maybe there’s a database error. Or maybe there’s a spelling mismatch between your names on both documents – even just a single letter, accidentally mistyped by an overworked clerk. A computer system will then automatically compare the data, find the mismatch, and reject the ‘seeding’. So now you have to find a way to update your details, or no food for you.

Depending on what you need to change, this may be possible online. If not, you’ll have to queue up at an Aadhaar center in person by showing up in the morning to receive a ‘token’. The centers only give out a limited number of these each morning – if you don’t get one for that day, you’ll have to come back again, and again, until you get a slot. For someone who needs to work every day to feed themself and their family, that’s inconvenient at best and impossible at worst.

But in the end, you managed to fix your details. Now you can go claim your food – except your state’s government has instituted mandatory Aadhaar-Based Biometric Authentication (ABBA, who I assume wouldn’t approve of their musical reputation being invoked here). This requires you to match your fingerprints on a physical terminal at point-of-sale, or no food for you.

And it fails, often. Fingerprints can get worn away with time and hard labor. The scanners themselves degrade with heavy use and insufficient maintenance. Or the point-of-sale just doesn’t have good enough Internet connectivity. Any of these, and countless other possible glitches, will see you automatically rejected. No food for you.

You can argue that these problems are all unlikely. But that’s the danger of building a system that stacks multiple points of failure – the probability of something going wrong multiplies. Maybe the clerk noting down your name for your new ID document has a 90% chance of getting it exactly right so it matches all your other information; maybe the scanner is 90% likely to work; maybe there’s a 90% chance there’ll be reliable Internet; and so on. Each individual link in the chain is overwhelmingly likely to work. But if a system has just 4 checkpoints that are each 90% accurate – it will fail 1 out of every 3 times.

Proponents of mandatory Aadhaar integration argue that rejections due to technical failures account for a small minority of cases. The government’s own figures give the lie to that argument. In 2020, the Comptroller and Auditor General of India, the central government’s auditor, noted that the success rate for fingerprint authentication for ration card use hovered around 75% – in other words, 1 in 4 attempts failed.

To complicate things, the government also embarked on a huge drive to cancel ration cards that weren’t verified with Aadhaar, on the grounds that they must be illegitimate. Millions of ration cards were rendered invalid for this reason. And when it involves food supplies for India’s poorest, the costs can be all too human.

In 2016, an 11-year old girl in the state of Jharkhand died of starvation. Her family had had trouble linking or verifying their Aadhaar card to their ration cards. She was theoretically eligible for school meals under India’s ‘mid-day meal’ scheme, often hailed as a major victory against childhood malnutrition – but that, too, began to require an Aadhaar card linkage. Turned away, she ate nothing for eight days, and died.

She wasn’t the only one; NGOs and rights groups have attempted to track these deaths linked to Aadhaar exclusion, counting dozens just anecdotally, often in India’s most marginalized and vulnerable communities. After reports began emerging, the central government instructed state governments to ensure that nobody was denied food because of biometric failures. Investigations on the ground, however, found inconsistent implementation. The true toll remains unknown.

Doing the Job

These problems unfortunately carry over to another initiative central to Aadhaar – linking it to direct cash payments made under job guarantee schemes like NREGA, the largest public employment scheme in the world. NREGA guarantees 100 days of paid work a year to rural households that apply. 250 million people are registered with NREGA, over 80 million of whom actually do NREGA jobs every year. Along with the food distribution system, NREGA forms the other pillar of India’s expansive social safety net.

Starting in 2017, the government began encouraging job cards to be ‘seeded’ to Aadhar cards and to bank accounts so they could implement an Aadhaar-based Payment System (ABPS) that automatically linked all three together to pay wages quickly and efficiently. NREGA workers are entitled to receive their pay within 15 days of completing the job – this system was meant to make sure they got it.

It was a well-intentioned idea aimed at eliminating endemic payment delays. It also meant that the system was now dealing with three different data points – a bank account, an Aadhaar card, a job card – which multiplied the possible points of failure:

• Any information mismatch would cause registration to be rejected, including minor spelling errors, the presence or a lack of a middle name, etc.

• Not every bank account was ABPS-enabled, which would cause the system to reject the registration

• Biometric failures continued to be a problem

And so on. These aren’t edge cases. At the end of 2023, 1 in 3 NREGA registrants were still ineligible for the ABPS system; among those with active job cards, that figure was 1 in 8. Well over half of workers in large states like Maharashtra and Gujarat are ineligible for ABPS as of earlier this year. I think it hardly needs to be said that a system that excludes a large proportion of the people it’s meant to serve – most people in some states – is not a particularly good system.

As with the ration issue, the government went on mass drives to delete active job cards that failed Aadhaar verification. Some 15 million job cards were cancelled on the grounds that they were bogus.

Did mandatory Aadhaar integration at least reduce payment delays? Possibly not. As the last, bitter joke in that saga, a recent study argued that ABPS implementation had no significant effect on payment delays at all.

The fundamental promise of Aadhaar was that it would expand access – make it easier for people to access things, not harder. Let more people in, not keep them out. If it’s done the opposite, as the government’s own data suggests, then something has gone very badly wrong.

Why has all of this happened?

Seeing like an (Indian) State

James C. Scott was declared dead, but he isn’t leading any protest marches over it. The Yale University professor, one of the most fascinating characters in American academia, died last July. Scott was well known for his theories of the modern state and the way it interacts with and shapes the structures of people’s lives, with his arguments best summed up in the book Seeing Like a State. I highly recommend the book to essentially everyone, but a particular part of his thesis is relevant to our discussion here.

Basically, Scott believed that modern states have the urge to make things modern, rational, and above all legible; they want to use scientific principles, categories, and databases to turn the chaos of human life into order, a particular kind of order that makes society easy to govern and administer. Ideally, of course, the systems they create – ID numbers, legal documents, and so on – will make life more convenient for the people on the ground. In practice, though, these systems will often be designed to be convenient for the government – which is not always the same as making things easy for the people actually using these systems.

That’s how you get the living dead of India, and poor Baburam Bhil: a situation in which it’s much more convenient for the state to deny reality than for it to provide any form of redress to the individual. A state is not always interested in making what’s on paper conform to what’s real – it’s always more convenient to deny any reality that doesn’t match what’s on paper.

Now think about all these Aadhaar-linked systems from the perspective of the Indian state: they seem great! A computer can easily compare names and details and flag mismatches; biometric authentication will add another layer of security; cash transfers will be automatically initiated to the correct Aadhaar-linked bank accounts; the colossal fraud in India’s public welfare systems can finally be fought with the power of modern technology! From the state’s point of view, making everyone jump through these multiple hoops makes perfect sense. The system that all this mandatory integration creates will, ideally, function like a well-oiled automated machine. No adjustment required, no messy human involvement needed. It’s logical, it’s rational, it’s orderly – it’s good for everyone!

Now look at it from the perspective of someone actually using these systems. You know your name isn’t always going to match exactly across documents – an overworked clerk will mistype something, or a handwritten form will be inaccurately digitized. Maybe one document required a middle name, but another didn’t, and now there’s a discrepancy. And you know your fingerprints aren’t always going to be clear on a scanner, that the scanner itself won’t be in good working order, that the erratic Internet connection in your region can and will fail at the worst possible moments. You know all this, and from your point of view, the new system is a terrible idea because – a large chunk of the time – it will fail.

That’s the reality of India, a country of a billion people. It’s messy, indistinct, illegible. A system that attempted to force that reality into a neat, ‘rationalized’ pattern was always going to exclude large parts of the population. Every system makes tradeoffs – but a tradeoff that leaves a third or a quarter of people out of the system cannot and should not be acceptable.

The Aadhaar system was built to aid inclusion. By accepting these mass exclusions, it is undermining its own admirable intentions and hurting millions of people who’ve done nothing wrong.

And it doesn’t stop there.

Aadhaar Woes

The Indian state of Bihar is due to hold elections for the state legislature in a few months. Like every Indian election, it will be a grand exercise. Some 70 million voters are eligible to cast ballots, a number exceeding the total population of the UK. This has concerned the government, which believes there to be a significant number of people on that voter roll who shouldn’t be there – some deceased, some no longer resident in the state, some who are not even from India but may have acquired fradulent documentation. Not long ago, it announced a Special Intensive Revision (SIR) of Bihar’s voter rolls. Voters in Bihar who was not on the list as of the year 2003 must now prove that they deserve to be allowed to vote.

There are 11 documents that can be used to prove this. Aadhaar is not one of them. Nor, indeed, is a driver’s license or a taxpayer PAN card, which generally work for official purposes in other contexts. Some 30 million people must now present proof or be taken off the rolls. They were given one month to present a document on the list and have it verified by their local Booth-Level Officer, most of whom already work full-time jobs, and each of whom is now responsible for verifying hundreds of people’s documents and ‘enumeration’ forms.

This is hardly the first time the Indian state has made it clear that Aadhaar offers no protection. In recent weeks, a number of Indian citizens from the state of Bengal working in other states have been picked up by the police, transferred to the custody of the Border Security Force, and physically expelled across the border to Bangladesh. The justification for doing so was that they were suspected to be undocumented migrants from that country. Most of these people had valid Aadhaar cards (among other documents); the police largely refused to even try to verify them.

One might well ask: if Aadhaar won’t protect people’s basic citizenship rights – including the right not to be exiled without due process – what is it good for?

One More Paper on the Pile

Aadhaar has immense potential, whether it’s to replace the pile of ID documents one needs to get anything done in India, or simply to add a new, more convenient way of proving identity for official purposes. That being said, the current trends in implementation have undermined that goal immensely:

• Aadhaar is not sufficient for accessing core social welfare entitlements; ‘seeding’ and biometric authentication requirements have eroded the key promises behind the Aadhaar idea
  

• Aadhaar is, increasingly, not applicable when it comes to asserting civic rights; it won’t secure you a place on a voter roll, or even prevent your unlawful deportation from your own country

Some proponents of the system cite the Indian Supreme Court’s 2018 judgment in the Puttaswamy case, which challenged the Aadhaar Act that gives the system legal force, as forcing the inclusion of other identity documentation along with Aadhaar. At the same time, the government has continued efforts to make Aadhaar mandatory for a range of functions, such as Employee Provident Fund services, and to amend the rules in ways that are arguably in contravention of the Supreme Court’s decision.

It’s a strange contradiction: the Indian state continues to push for mandatory Aadhaar integration in a number of different systems, yet when it matters most. Aadhaar simply isn’t any protection at all. It is, officially, not proof of citizenship, residency, address, or even age.

In effect, we’re seeing the creation of a paradigm in which Aadhaar is required for everything, but is proof of nothing.

Taken to its logical conclusion, we’ll have a system in which, rather than replace the existing jumble of ID documents, Aadhaar simply adds one more to the substantial pile of photocopies you’ll need to submit to get anything done in this country.

Surely we can do better than this.