Unpacking the Bangalore Ideology

Audio of this conversation is available via your favorite podcast service.

At the September G20 summit in Delhi, the government of prime minister Narendra Modi promoted the country’s digital public infrastructure (DPI) as a model for the world for how to develop digital systems that enable countries to deliver social services and provide access to infrastructure and economic opportunities to residents. Other world leaders were enthusiastic about the pitch, endorsing a common framework for DPI systems.

But even as an Indian vision for DPI appears to be attractive beyond that country’s borders, what are the ideas and events that shaped India’s approach? Today's guest is Mila Samdub, a researcher at the Information Society Project at Yale Law School who recently published an essay titled “The Bangalore Ideology: How an amoral technocracy powers Modi’s India,” looking at histories of technocratic ideas in India, and how they have combined with Modi’s particular brand of populism.

What follows is a lightly edited transcript of the discussion.

Mila Samdub:

So my name is Mila Samdub. I'm a resident fellow at the Information Society Project at Yale Law School. I research, I like to call it the aesthetics and ideologies of digital infrastructures in India today. I'm particularly interested in infrastructures of development. My interest in sort of aesthetics, and ideology specifically comes out of my background and training in art and architecture. Now I'm applying that more to domains of policy and political economy.

Justin Hendrix:

We've just come off the G20 in India, where Digital Public Infrastructure was a main topic of discussion, almost a main sales point from the Modi government. I reached out to you initially because I had read an essay that you wrote called “The Bangalore Ideology: How an Amoral Technocracy Powers Modi's India,” which really looks at the history, sort of strands of thinking and ideas that have gone into what has become this focus on DPI. Where did your inquiry start? What did you set out to do when you wrote this essay?

Mila Samdub:

I first got interested in this sort of set of Indian digital infrastructures around 2016, when there was a big movement against the Aadhaar Biometric Identification project in India. And over the years I've been tracking the ways in which that project has morphed, in which it's grown. The article that you mentioned was really trying to unpack the history of state constructed digital infrastructures in India, and as you mentioned, that now forms the backdrop of what India's trying to promote globally as digital public infrastructures.

Justin Hendrix:

This is a set of book reviews all smashed together.

Mila Samdub:

Yes, it is. So as I said, I'm interested in aesthetics and narratives, so I think also that technocracy is playing a really important narrative role in India today. And so, part of what I'm doing is not only telling the history of technocracy, but also showing us how other people are historicizing it today to make arguments about what should be done in the present.

Justin Hendrix:

You start with a quote, which comes from the first book you look into from Nandan Nilekani, the 2015 book, Rebooting India: Realizing a Billion Aspirations. And the quote you provide is, "The fundamental nature of government is a platform. We are talking about radically re-imagining government, its purpose, its role, and the way it carries out its functions with technology at its core." I suppose this is the premise that you're interrogating in this essay, this idea of government as platform, government as technocratic medium on which the state builds its function. Where did this idea come from? You say Nandan isn't just pulling this out of thin air.

Mila Samdub:

First. I think it's really important to understand what government as a platform means. For Nandan Nilekani, in that book, in rebooting India, Nilekani describes how a set of digital infrastructures have to be inserted into all aspects of a government function. So, we're talking about identification through the Aadhaar Biometric Project. We're talking about facilitating payments through what's called “unified payments interface.” We're talking about healthcare, we're talking about collecting tolls on highways, we're talking about agriculture, talking about the welfare apparatus.

At the center of all of these need to be a set of software infrastructures. It's important to note that these are software infrastructures. There's not a lot of hardware construction going on over here. And effectively, each of these domains is already functioning by using a set of application programming interfaces, or APIs, which make for a composite system that is interoperable, your payments and identity, and your healthcare can all be managed by the same set of software infrastructures. And so, that's what we're talking about, when in India that's taken to form of “India Stack,” and as it's being exported now, that's what we call digital public infrastructure.

Justin Hendrix:

So that's today. But the purpose of this essay is to take us back, of course, in time, and to look at the sort of strain of ideas that have led us here. You go quite a ways back. You go all the way back to even the 1930s. I suppose, for my listener, can you take them through the short version of the trajectory of the idea of technocracy in India?

Mila Samdub:

So, as you mentioned, this history begins in the 1930s, before India is independent, and already, technocracy and technocratic ideas are being used to imagine what kind of a future India will have as an independent nation. So, already, we can see that the narrative around technocracy is one that's about future aspirations. When India becomes independent in 1947, immediately a sort of large technocratic apparatus is set up, and this sort of consists of a few elements. India at independence is a socialist planned economy. The goal of planning is to provide economic growth in a self-reliant way. This also meant not falling into the trap of neocolonialism, and resulted in a broad policy of what's called import substitution industrialization, which effectively meant building up large scale industries, so that India would have to avoid depending on imports.

Now planning was also a very technical discipline, which is why we're thinking of it as a technocracy, and sort of a lot of its coming out of statistics at this moment. The Indian Statistical Institute in Kolkata is a really powerful institution. The sort of five-year plans are being written up by Prasanta Chandra Mahalanobis, who is the director of that institution. So, it's a kind of planned data based technocracy at this moment. I think it's also important to flag that it's incredibly elite. Mahalanobis was educated at Cambridge. His plans were really coming top down, onto the people of the country. The literacy rate at independence was 12%, and yet you had these set of really sophisticated plans that were planning to dramatically increase productivity, and eliminate poverty in a few decades.

Justin Hendrix:

So the goal here is almost to treat the economy, treat, question development almost as a kind of physics.

Mila Samdub:

I think that's a really useful thing that you point out there, because Mahalanobis was trained as a physicist, and he's often described his approach to planning as thinking about it as thermodynamic study. And that sort of appeal to physics is a classic kind of move that experts make, which allowed him really to sidestep the discipline of economics altogether. He would say that economists were engaged in politics, they couldn't come to objective conclusions about what needed to happen, and his training as a physicist let him arrive at the one right answer. And again, that's a disciplinary claim about technocracy that we see coming up over and over again.

So, technology is the answer for India in many domains, right? Scientists are being recruited to build dams, we're focusing on nuclear energy. Science is supposed to be applied across the board to India's developmental problems, and science here refers to a particular orientation towards the world. It signified... So teachers and doctors were also considered scientists in this sort of Nehruvian period. It was a sort of very broad definition of a sort of modern person that would apply any sort of technical skill to shape a bright future for the country. That was the weird, uniquely participatory moment which didn't really live up to its goals often.

So within planning, as I mentioned, statistics is the sort of particular form of techno science that is being pressed into service, and that's the story that story Nikhil Menon's book tells. It's called Planning Democracy. Nikhil Menon describes how the state under Nehru, and Mahalanobis built up what he calls India's first data infrastructure. That sort of happened off the back of a new technology at the time, which was randomized sampling, and which sort of allowed Mahalanobis to collect huge amounts of data about all sectors of Indian economy and society very quickly, and cheaply, and which formed the basis for the plan. So I think when we're talking about technology for the purposes of our project, we want to be thinking about data specifically, and what role that's playing at that time.

Justin Hendrix:

So it starts with math, starts with statistics, and this is right at the dawn of the computer age. What happens when computers come along?

Mila Samdub:

This is right at the dawn of the computer age. In fact, Menon has a sort of really nice chapter on how computers are imported into India first at the Indian Statistical Institute in order to compute some of these data on the economy. So, computers are intimately tied to this history. Computers proliferate in Indian technocracy, and in society in general, starting in the '60s and '70s. You have the application of computing within private corporations, but more importantly, you have it within the state. And at this point, the state is undergoing many shifts. So before we get to computing per se, I want to just talk about how technocracy itself shifted.

If the early promise of post-colonial planning under Nehru and Mahalanobis was to build up this bright future for India, that promise failed very quickly. The five-year plans didn't deliver on their data-based promises. Often that had to do with the fact that the plans simply just didn't compute the right things. They didn't account for the fact that so much of India was agrarian, and very poor, and couldn't provide the demand to sustain the kind of growth that they were hoping for.

As this sort of early optimism vanishes, the technocracy hardens into a bureaucracy, and it becomes this sort of Byzantine bureaucracy that imposes retail controls on everything. You can only export or import a very certain number of chips, or of computer parts, or of cars, or machinery, what have you. We call that the license permit Raj, return to the Raj, but in the form of licenses and permits. And that's the bête noire of today's capitalists, symbolizing everything that's wrong with state socialism, and that's a rhetorical foil for much of today's discourse, is around tech and the state.

Justin Hendrix:

Where does that take us to, from a timeline perspective? When does that period start to draw to its close?

Mila Samdub:

So, we're talking about the 1970s, right? That period starts to draw to close in the 1980s, which is also the moment when computing starts to take off in a big way in India. And so, there's this narrative, which kind of, a couple of the books that I review also really buy into, which is that computing brought the end of that kind of bureaucratic, corrupt state socialism. Computers enter within the state in the 1980s, and the 1990s in a big way. And if computers had once been in the headquarters, or in the big universities, in big cities, now they were suddenly being proliferated in more regional centers.

And in the state, this happens through the promise of what Nafis Hasan describes in an article in the excellent new book, Overload, Creep, Excess: An Internet From India. And Nafis' argument kind of hinges around the relational database management system, which is a new kind of database system, which is inserted into all aspects of state functioning, starting in the late '80s and early '90s, and suddenly, data input and output are happening at the local level. On the ground, government employees are becoming data workers, working with computerized systems on a day-to-day basis. That kind of collection of data also means that government services can now be privatized in new ways. That kind of data entry where it means that anybody could start fulfilling the rules of government work. Government work can start being contracted out in new ways. And so, there definitely is a sort of an interplay between authorized neoliberalism and computing.

Justin Hendrix:

Because this is also the moment where the state can begin to use these technologies to have a more direct interface with citizens?

Mila Samdub:

Absolutely. So now you're getting to the point where you might go to book a train ticket, and your Indian Railways booking clerk is typing into a computer over there, and suddenly computing, and data itself is becoming individualized. Right? In the 1950s and '60s, the sort of statistical data which was based on random samples was still at a population scale. And so, computers couldn't address individuals, the state couldn't address individuals. And now with relational databases, and individualized terminals, individual records start to matter. The other half of the story of computing in the '80s and the '90s is the rise of the India software services industry.

Justin Hendrix:

This is the great business process outsourcing economic miracle.

Mila Samdub:

This is, exactly, the great economic miracle. Business process outside sourcing starts a tiny bit later, but this is often told as a story that begins after the economic reforms of 1991. In 1991, the Indian economy leaves behind, planning opens up a lot of those controls, export controls are taken away. This sort of happens under pressure from the IMF. And so that's a major moment in recent Indian history. I think there's compelling reasons to just think that the story of software outsourcing actually began with state support in the 1970s. That's a sort of story that Jyoti Saraswati tells in the great book Dot.compradors: Power and Policy in the Development of the Indian Software Industry. But in any case, the software industry is becoming really powerful, and their main business is doing back office work for large corporate clients, and governments in the West.

Justin Hendrix:

This focus on still efficiency, these sort of notions perhaps still similar perhaps to Mahalanobis' day, a kind of paternalistic technocracy. All this is combining with what technology is actually making possible. For the regular citizen at this point, we're in the early aughts, what's going on? Is it working for them?

Mila Samdub:

So India's undergoing this economic miracle through computing, through sort of software services exports. That said, that's not a very labor-intensive kind of sector, and the people who are directly and indirectly employed by this sector are a very small fraction of the total Indian population. And so while you have these glossy images of a modernizing India, you also have an increasing divergence between the rich and the poor. Computing is really gated off at this point. Thomas Friedman, the New York Times columnist, claims that the world is flat after he visits the campus of Infosys, one of India's largest software companies, in a sort of software technology park. But that's basically like a gated zone where inside the roads work well, the electricity always works, everyone's fairly wealthy and well-dressed, and well-fed. That's not the case for most of the country.

Justin Hendrix:

What do you see as most significant in the period from around when Thomas Friedman's tromping around Infosys and having techno utopian thoughts, on through to say the last decade or so?

Mila Samdub:

I think there's one answer in the political economy off the tech sector itself. Software services in India are undergoing a crisis in the late aughts. Sort of these large Indian firms are losing their market share within the industry in India, and we have the entry of a lot of global players. So we're talking about companies like SAP and Accenture, that are effectively replicating the Indian model, poaching Indian engineers from the Indian companies, and taking advantage of all the cost advantages of working out of India.

And Jyoti Saraswati, what he describes in detail, how these global companies take over the industry body, the really powerful industry body which shaped much of the software sector's development, the National Association of Software Services Companies, or NASSCOM. So, that's one kind of crisis that's taking place. Another crisis that's taking place is that business models themselves are shifting. With the rise of cloud computing, you don't really need software services so much anymore, in the sense of having backend computer support working in-house for your company at all times.

And around 2010, you get a new shift in thinking in the Indian software industry towards products, towards software products. So, the great goal of the industry shifts somewhat to saying, "Well, what if we could build products? What if we could build truly innovative products that everybody could use? We would find new customers, and we would avoid sort of the problems that are currently plaguing the industry." And that search for new customers really meant looking at Indians as customers. And so in 2013 you have a group that breaks away from NASSCOM called the Indian Software Products Industry Roundtable or Iceberg. ISPERT. ISPERT has a lot of very important members of the tech community as working for it as its volunteers, and it really starts actively lobbying government in a way that we hadn't seen before.

Justin Hendrix:

And this is where the idea for Aadhaar is born?

Mila Samdub:

So the idea for Aadhaar is born in many places. I don't think this is a simple story, and I think we're still unpacking some of that complexity. The idea for Aadhaar is born from Nandan Nilekani. Nandan Nilekani is a sort of key figure also in the rise of ISPERT. He was also one of the founding members of NASSCOM, so he's one of our most celebrated technocrats today. And the idea of Aadhaar really is born around 2008, where you concurrently have a crisis in the state. Remember we talked about export... The license permit raj, the state still hasn't really been able to shake off that image, and it's no longer really attached to permits anymore, but now it's attached to corruption. It's attached to a really powerful bureaucratic, and political elite that is in cahoots with industry in a way that's excluding people, in a way that's sort of a massive drain on the public exchequer.

And so Aadhaar is pitched as a way to really plug corruption in the welfare system first. When ISPERT comes to Aadhaar, it sees in Aadhaar an infrastructure that can be reused for the purposes of private companies. A whole different strand of Aadhaar also is that the state sees it as a very useful tool for national security purposes. And I think the best way to understand Aadhaar and India Stack is actually as a set of platforms that allow all of these many uses, that allow the convergence of different interests, and different interest groups, and I think that's really their power.

Justin Hendrix:

But it's around this time that Indian politics starts to change.

Mila Samdub:

Yes, exactly. So, we have ISPERT taking up Aadhaar, and ISPERT sort of coins this new framework of India Stack, which is this sort of set of APIs that I mentioned already. So that's happening on the one hand. On the other hand, in 2014, Narendra Modi is elected. Narendra Modi is the leader of the Hindu Nationalist BJP party, and Modi's election is really interesting because it marks a break from previous BJP policies. Modi really campaigns, even though he is a Hindu nationalist, in favor of India as being a Hindu nation. His campaign is really on a development plank. He's promoting development, he's promoting sort of modern, high-tech images of India.

Justin Hendrix:

Is Modi interested in Aadhaar when he comes into office?

Mila Samdub:

Modi starts by campaigning against Aadhaar, because it was a sort of infrastructure that was championed by the opposition Congress. But very quickly after he comes into power, he meets Nandan Nilekani, and Nilekani demonstrates the uses of Aadhaar to someone like Modi. One of the key uses there is that he shows how Aadhaar could be used to... As the basis of attendance systems for government employees, to show that they're coming into work regularly, and on time. And so there's a promise of Aadhaar as beating efficiency and productivity in the state, that meshes very nicely with Modi's anti-corruption message, but also his sort of message of discipline. So, all of these different kind of uses of Aadhaar are getting layered onto each other, and as that happens, it's becoming more and more iron clad against criticism.

Justin Hendrix:

So we're seeing somehow the combination of technocracy and populism.

Mila Samdub:

Absolutely. Folks have been calling this managerial populism, or technocratic populism a core part of the narrative around technocracy like Aadhaar, which is so tangible in everyday life. People have to use Aadhaar to get their welfare payments, to get any sort of subsidy from the state, but also to register mobile phones, to use their bank accounts, and so on. So, it's this really tangible infrastructure, and as opposed to say the technocracy of old that we discussed earlier. And so part of that sort of everyday individual tangibility is really the claim that Aadhaar is a tool that works in favor of the people, and against corrupt, entrenched elites, no matter who those are, and that's the classic populist binary.

Justin Hendrix:

Is that what's really going on here?

Mila Samdub:

I think it's fair to say no. I think it's fair to say that Aadhaar has been incredibly exclusionary infrastructure. It's introduction into welfare has resulted in many deaths from hunger. Just recently there was a major movement around the introduction of digital Aadhaar related system into the payments of a guaranteed welfare work scheme in India called the National Rural Employment Guarantee Act. That sort of movement showed the amount of popular discontent that there is around Aadhaar in these infrastructures, and testified to the amount of exclusion that's taking place because of them. We should note that on the other side, this exclusion also results in less outlay, and in cost savings, so it can be spun as a benefit. Finally, I'll say that Aadhaar also has not had any observable role in decreasing corruption in welfare. A major study by the Jameel Abdul Latif Poverty Action Lab a couple of years ago showed that it had negligible effects on corruption in welfare.

Justin Hendrix:

How would you characterize at this point, and I don't know if this is the place to bring in this concept of the sort of slow violence, but how would you characterize what it's like for the typical Indian citizen to interact with the state at this point, through Aadhaar, with this sort of perpetual need to comport themselves to the database, comport themselves to this broader system?

Mila Samdub:

I think it's important to say, when I'm talking about Aadhaar, I'm actually talking about a larger set of infrastructures that were built on the Aadhaar model, and these infrastructures often just don't work very well. And Nafis Hasan sort of traces this back to the 1980s introduction of relational databases into the Indian state, and he argues that that creates a form of violence that he calls slow violence. You might show up for a welfare payment at a government office, try to use the Aadhaar fingerprint reader, and it doesn't recognize you. And so, then that sort of sets off an interminable process of trying to get your records fixed on the database, and Nafis contends that is what citizenship means in contemporary India. It's just an endless process of trying to get one's records fixed, of trying to fix the database to match the reality of one's circumstances.

Justin Hendrix:

And yet here we are in 2023, you've got apparently Bill Gates hailing Modi for his leadership on digital public infrastructure. You've got India basically saying, "This is the path forward," particularly for developing nations, "This is the way to go." A lot of folks, of course in India, Modi an incredibly popular politician in part because of the economic advances in the nation over the last several years. Is this working?

Mila Samdub:

As I mentioned, I think these infrastructures are incredibly powerful tools for generating alliances, and to the extent that they can generate a viable electoral alliance, or to the extent that they're one part of generating a viable electoral alliance, they're working very well. A lot of people who are benefiting from India's economic growth right now, that's still a small minority, but for those people, these infrastructures look great. They're working very well, they're showing how India is this modern, high-tech country. Digital payments are a part of everyday life for urban middle-class citizens, and have really revolutionized everyday life. I don't want to knock that. They have had an immense impact, and they are really palpable as a new form of using technology. And so, to the extent that one sees one's aspirations for India reflected in how easy it is to make a payment, to access your food delivery services, I think they're very successful. And that's what's, let's say, particularly dangerous about these systems.

Justin Hendrix:

In your piece, you say this Bangalore ideology, "Justifies the power and privilege of a new ruling coalition of Hindutva politicians, and digital capitalists. It allows digital India to retain all the extra institutional, and unaccountability of older technocracies, while facilitating a project of extraction cloaked in a rhetoric of populism." Them's fighting words, Mila.

Mila Samdub:

They are indeed.

Justin Hendrix:

What are you saying is really wrong with this DPI that's being hailed by world leaders?

Mila Samdub:

These systems are really being introduced into incredibly unequal societies. India is the major case here, and they're serving to exacerbate that inequality. They're serving to enable a small set of companies to suck up a lot of citizen data in the first place. They are also enabling those companies to build softwares, build apps and platforms on the basis of these infrastructures for which they've undertaken no risk at all. The government having built out these infrastructures takes the risk for private companies to innovate. I don't think that's how this should work. At the same time, they're allowing incredibly illiberal, and anti-democratic government to become even more so, and increasing its powers. And as I keep saying, I think it's the alliance between those two forces that is really troubling here to me. And I think because technocracy is so apolitical on the surface of it, because it signals apoliticism so strongly, I'm worried about the system and I'm worried about what happens when they're exported elsewhere as well.

Justin Hendrix:

So where do you see this headed? I suppose some would regard this focus on DPI, this focus on these public infrastructures as a kind of, potentially a sort of new approach to how to use tech. I don't know, when you think about it, how does it situate against other global models? It seems to me a little closer to China than perhaps anywhere else, although we do see that there are other democratic nation states that are also thinking about these kinds of unified approaches to digital public infrastructure. Where is this all headed? Is it possible to separate the future of DPI, of systems like Aadhaar, and the stack of technologies more generally built on it? Is it possible to separate that from the anti-democratic trajectory of Indian politics generally?

Mila Samdub:

I hope it is. I worry that as long as India is controlling the narrative on these, it may not be. I think until recently, digital public infrastructures were really an open question of what they meant. So, US scholars like Ethan Zuckerman for example, really promoting digital infrastructures as sort of infrastructure forms on the model of say, Wikipedia, or something like that. I'm all for that, but the kind of form of digital public infrastructure that we've seen with India, where they are public in the sense that they require state intervention, and a lot of money from the state, but their sort of benefits are not distributed equitably at all.

So, to the extent that we see that, I'm worried about it. I think part of why India has been so successful in pushing this narrative harks back to early Indian technocracy, right? It's a sort of narrative of moral leadership, and non-alignment. It's saying that the way that the internet has worked in the US, where you have a handful of extractive big tech companies running the show, and the way that you have it in China, where you have an authoritarian state running the show, those are not the only choices, that you can have a model that is built by a democratic government, that is low cost, and that empowers people, and serves development goals.

And that is a very seductive premise. To the development of Aadhaar and these infrastructures, there have been strong critical voices against them. As I mentioned recently, there's been a strong people's movement against the introduction of digital management systems into welfare. That continues, and I think the way forward has to mean foregrounding those voices of criticism, taking them seriously, as developing such infrastructures with actual input from people, as foregrounding what would actually serve people rather than creating extractive systems in the name of the people.

Justin Hendrix:

Well, we will see how, I suppose, the Bangalore ideology evolves, and I suppose that you'll be following it closely as well. What's next for you?

Mila Samdub:

I'm continuing to work and write on the Bangalore ideology, so I will have more out on that front shortly, I hope.

Justin Hendrix:

Well, hopefully we'll have you back on to discuss it the next time there is another major development.

Mila Samdub:

Great. Thank you Justin. It's been a pleasure.