#245 On Protests

Navalny's Death, Farmer Protests in India, and On-Chip AI Governance

Course Advertisement: Intake for the 8th cohort of Takshashila’s 48-week Post-graduate Programme in Public Policy (PGP) is open. It’s a fantastic pathway for those who wish to make a career switch to public policy. Check all details on the website.

Global Policy Watch #1: Why Some Protests Succeed and Others Don’t?

Global policy issues relevant to India

— RSJ

Alexei Navalny, the face of political challenge to Putin, died yesterday while being held in a jail somewhere north of the Arctic Circle. There are fairly predictable statements from the Western leaders holding Putin directly responsible for the death. I’m almost certain the rest of the world, including India, wouldn’t want to comment on it.

It is unclear to me what the western condemnation means in terms of action. Russia already faces multiple economic sanctions since 2022 but seems to be managing quite well for now despite them. Putin has been indicted by the International Court of Justice already for his crimes against Ukrainian civilians. He seems to be doing swimmingly well if recent interviews are to go by. There are no real options to punish or restrain Putin, really. The pro-democracy movement was pretty much dead in Russia after the Putin regime jailed hundreds of dissidents in the past two years in the name of national security during the Ukraine war. With Navalny gone, it is difficult to see another face emerging as a challenge to the regime anytime soon.

Back in 2012-13, when Navalny first came into prominence, it was possible to hold mass protests in Moscow against Putin. The pictures from those times that have appeared in multiple news sites mourning Navalny’s death look a bit surreal today. It is easy to forget that back then, Navalny even entered the fray for Moscow mayoral elections. Or that Putin had once voluntarily stepped aside to adhere to constitutional norms. In the course of a decade, Russia has slid further on the scale of authoritarian rule and is closer now, more than ever, to the Stalinist reign of terror of the early 1930s. This ruthless streak might appear to few optimistic types as some lack of confidence that’s eating Putin. That Putin’s continued repression of his own people to strengthen his grip on power will have consequences, and one of these days, the people might rise against the state. But I’m not so sanguine. That’s because mass protests require a certain coming together of the protesters, their beliefs and the nature of the state that they are agitating against. It will be useful to parse this a bit more.

Talking of protests, back in India, there is fresh farmers’ agitation brewing in Haryana and Punjab that have intensified after police violence and a few rounds of negotiations with the union government. The farmers’ unions are demanding a legal guarantee on MSP, pension for farmers and farm labourers and good old farm loan waivers. We have written extensively on this topic back in late 2021 when the farmers repealed the new farm laws that were aimed at dismantling the MSP system. At this moment, I’m not sure about the proximate trigger for the protests. Possibly, it is tepid growth in farm income in the past year that hasn’t kept pace with inflation. The ban on exports of rice, sugar, etc., in 2023 must also have impacted the farmers' earnings. Or, it could just be timed to keep the pot boiling until the May Lok Sabha elections. These protesters have a track record of success, given what they did a couple of years back. Will they succeed this time around? Again, I’m not too sure.

So, what goes into making a successful protest?

By definition, a protest is an action in the present about a grievance that has a long-standing past to shape what ought to be the future in the eyes of the protesters. The action in the present usually involves transgressing a particular law that one is protesting against or another law(s) that might help raise awareness about their grievance in the wider society while being fully aware that such transgression could have consequences. In that sense, a non-violent protest (of the kind of Navalny and the farmers) is a civil action within the broader framework of the legal system to push against the state and its motives. Considering this constraint, its success depends as much on its ability to mobilise and act as on the nature of the state that’s contending with this challenge.

Rawls, in his book ‘A Theory of Justice’, defines civil disobedience as a public, nonviolent, conscientious and yet political act contrary to law usually done to bring about a change in law or policies of the government. Broadly, for Rawls, the justified use of civil disobedience requires three conditions to be met.

Firstly, it should target long-standing injustice while appealing to widely accepted principles of justice. The success of a protest movement depends on the clarity of injustice, i.e. it must be evident to most (despite some of them justifying it) that there’s a violation of the principle of justice and liberty. The more easily the protesters can link their cause to publicly shared principles of morality, the greater the resonance of the protest. During the farmers’ protest of 2021, there was a concerted effort to portray the farmer as annadaata who has been wronged for ages. Keeping factual arguments aside on the validity of MSP, it is not too difficult for ordinary people to accept that farmers have borne the brunt of injustice in India for centuries. Though the state tried to counter it by showing the protesters to be a minority of elite, land-owning farmers with extremist support, the conventional image of a poor, hardworking farmer that the movement tapped into held sway.

Secondly, Rawls believed that any act of civil disobedience goes against the moral duty of the citizens to obey the law, which forms the basis of the sovereignty of the state. Therefore, public protests that transgress existing laws must be used only as a last resort after exhausting all other options for redressal. Unless this stringent criteria is met, every single grievance can lead to the breaking of the law, which will then threaten to destabilise society. This is how a Gandhi or an MLK steered their movements. It is easy to find some justification for violence or for acting unlawfully against a repressive state instead of appealing to their sense of morality through dialogue and negotiations. But, if that were to succeed, the fear was you would replace one violent state with another. This is what Gandhi feared when he opposed the more militant strands within the Congress and the broader freedom movement. It is easy to venerate those strands today without pausing to think about how it would have played out. The history of most African nations that won their independence in the second half of the 20th century without this moral clarity is evident to all. The means you employ to protest and succeed will often be the means you will use to govern. It is a trap that has caught out almost every national movement and its heroes.

Thirdly, Rawls believed that most protests initially involve an aggrieved minority or a minority brave enough to protest to begin with. For them to succeed, it is crucial to find similar minority groups with grievances that can be aligned under a broader umbrella. This coordination between other minority groups creates momentum for their protests, which makes their case more compelling and, importantly, creates a sense of anxiety within the government of a domino effect of other groups coalescing with this and getting out of control. This allegiance, or lack of it, is why you find specific caste-based protests (Marathas, Gurjars, etc.) demanding special status or reservations in jobs fizzle out. It is difficult for an exclusionary idea like that to find more support. Contrast this with the reservation movement in the 90s, where the focus was on numerous castes and subcastes coming together to demand more rights. And I suspect the farmers this time around will fail to find these other allies. They will struggle to find momentum.

While these are necessary conditions for a protest to be successful, it assumes that the state that’s confronting it is rational, possibly moral and can weigh the merit of continuing to ignore the protest with the risk to its primary goal, that of self-perpetuation. Gandhi and MLK faced the kind of states that understood the long-term risks of continuing to ignore their demands that were rooted in natural justice. Navalny faced a different kind of state that’s insecure and fearful of its immediate survival. Regardless of how well he could have mobilised and led a protest, it is impossible to think of a different consequence than his unfortunate death, given the amoral kleptocracy that Putin runs. In that sense, the fate of Russia is sealed. The current violent regime can only ever be uprooted by forces that use violence to achieve their ends. And like I said before, the means you use to run a successful protest are the means you will be yoked to forever.

I will close with an extract from the seminal essay on civil disobedience by Henry David Thoreau that inspired, among others, Gandhi, the civil rights movement of the ‘60s, and many others. Thoreau, unlike Rawls, is quite clear about the moral responsibility to break an unjust law instead of waiting for the due process to run. There’s not enough time for a man to wait for the state to see reason. He makes a compelling case:

“Unjust laws exist; shall we be content to obey them, or shall we endeavor to amend them, and obey them until we have succeeded, or shall we transgress them at once? Men generally, under such a government as this, think that they ought to wait until they have persuaded the majority to alter them. They think that, if they should resist, the remedy would be worse than the evil. But it is the fault of the government itself that the remedy is worse than the evil. It makes it worse. Why is it not more apt to anticipate and provide for reform? Why does it not cherish its wise minority? Why does it cry and resist before it is hurt? Why does it not encourage its citizens to be on the alert to point out its faults, and do better than it would have them? Why does it always crucify Christ, and excommunicate Copernicus and Luther, and pronounce Washington and Franklin rebels?

One would think, that a deliberate and practical denial of its authority, was the only offense never contemplated by government; else, why has it not assigned its definite, its suitable and proportionate penalty?

If the injustice is part of the necessary friction of the machine of government, let it go, let it go; perchance it will wear smooth,— certainly the machine will wear out. If the injustice has a spring, or a pulley, or a rope, or a crank, exclusively for itself, then perhaps you may consider whether the remedy will not be worse than the evil; but if it is of such a nature that it requires you to be the agent of injustice to another, then, I say, break the law. Let your life be a counter friction to stop the machine. What I have to do is to see, at any rate, that I do not lend myself to the wrong which I condemn.

As for adopting the ways which the State has provided for remedying the evil, I know not of such ways. They take too much time, and a man’s life will be gone. I have other affairs to attend to. I came into this world, not chiefly to make this a good place to live in, but to live in it, be it good or bad. A man has not every thing to do, but something; and because he cannot do every thing, it is not necessary that he should do something wrong. It is not my business to be petitioning the governor or the legislature any more than it is theirs to petition me; and, if they should not hear my petition, what should I do then? But in this case the State has provided no way: its very Constitution is the evil. This may seem to be harsh and stubborn and unconciliatory; but it is to treat with the utmost kindness and consideration the only spirit that can appreciate or deserve it. So is all change for the better, like birth and death which convulse the body.”

---

India Policy Watch: Inglorious Prices

Policy issues relevant to India

— Pranay Kotasthane

Just two weeks ago, we discussed the farm protests in Europe. On cue, similar protests have returned to northern India. In that edition, we discussed some structural reasons that make roadblocking protests a time-tested method for farmers worldwide.

Returning to the current wave of protests, it seems quite likely that in some form or the other, the government will agree to the protestors’ demand for guaranteeing MSP. By agreeing to repeal a reasonable yet hastily pushed set of farm reforms in 2021 in the face of street protests, the government opened itself up to this form of recurring blackmail. We have written quite a bit on the last round of protests in editions 70, 90, 97, and 106, so I won’t repeat those points.

Nevertheless, three other thoughts on the Maximum Smog Price, better known as the Minimum Support Price (MSP), are relevant to Indian public policy.

First, the end state of any “indicative” price is a strict price floor or a price cap. There’s nothing like a weak price signal. The MSP began as a benign indicative price to help farmers adopt new varieties of high-yield seeds at a precarious period when India was struggling with a ship-to-mouth existence. Over time, this indicative price morphed into an implicit but unenforced price guarantee. And now, the protestors want the guarantee to be made legal. Price interventions develop a political life of their own, separate from the underlying reasons or economic conditions.

Second, such interventions exacerbate inequity. They always benefit powerful incumbents at the cost of the weaker sections and future generations. Even if the government were to “legalise” MSP—whatever that means—there is no way it can ensure the procurement of 23 crops across millions of transactions across a continent-sized country. Even today, the benefits of MSP go to a small set of large farmers in Punjab, Haryana, and Western UP. In many eastern states, rice procurement happens below the MSP, yet the farmers from these areas aren’t protesting on the streets.

Moreover, the budget constraints are real. The politics of defence pensions educates us that any unfunded guarantee like the OROP (One Rank One Pension) will inadvertently be followed by a rearguard action like the Agniveer scheme to contain expenditure. This sequence of events might also repeat in the case of a guaranteed MSP.

Third, the only way out in the long term is to make MSP immaterial. If market prices are consistently higher than MSP, it will cease to be a lightning rod for protests. That requires a trade policy which allows farmers to export their products freely. Better storage and agro-processing facilities, which are conspicuous by their absence in the farmers’ demands, are also necessary to stabilise prices for consumers.

Either way, the MSP saga isn’t ending anytime soon.

---

Global Policy Watch #2: Compute Governance is the New Buzzword in AI

Global policy issues relevant to India

— Pranay Kotasthane

The AI triad consists of data, algorithms, and compute. The first two are difficult to control and exclude others from using. So, the attention in the US has shifted to finding better ways to restrict others’ access to AI chips.

Export controls on AI chips are just the tip of the iceberg. Over the past couple of months, I have come across many papers arguing for on-chip governance mechanisms that will allow the US to remotely disable AI chips being used to train workloads that violate certain rules or thresholds. Consider a report by the Center for a New American Security, which argues:

“On-chip governance mechanisms can safeguard the development and deployment of broadly capable AI and supercomputing systems in a way that is complementary to American technology leadership.”

Another recent paper titled Computing Power and the Governance of Artificial Intelligence suggests:

“.. governments and companies have started to leverage compute as a means to govern AI. For example, governments are investing in domestic compute capacity, controlling the flow of compute to competing countries, and subsidizing compute access to certain sectors. However, these efforts only scratch the surface of how compute can be used to govern AI development and deployment. Relative to other key inputs to AI (data and algorithms), AI-relevant compute is a particularly effective point of intervention: it is detectable, excludable, and quantifiable, and is produced via an extremely concentrated supply chain. These characteristics, alongside the singular importance of compute for cutting-edge AI models, suggest that governing compute can contribute to achieving common policy objectives, such as ensuring the safety and beneficial use of AI. More precisely, policymakers could use compute to facilitate regulatory visibility of AI, allocate resources to promote beneficial outcomes, and enforce restrictions against irresponsible or malicious AI development and usage. However, while compute-based policies and technologies have the potential to assist in these areas, there is significant variation in their readiness for implementation. Some ideas are currently being piloted, while others are hindered by the need for fundamental research. Furthermore, naïve or poorly scoped approaches to compute governance carry significant risks in areas like privacy, economic impacts, and centralization of power. We end by suggesting guardrails to minimize these risks from compute governance.”

It appears that the US government is already moving in this direction. Executive Order 14110, issued in October 2023, mandates that large compute clusters must report back to the US government if any foreign entity uses cloud computing resources above a certain threshold. This is a move to possibly restrict another nation-state, particularly China, from developing alternate foundational models.

Even so, on-chip mechanisms to disable devices remotely are categorically different and need deeper reflection. These mechanisms interfere with post-acquisition ownership rights in a fundamental manner. The usage of such chips by a company or a country, even after making a legitimate purchase from an American entity, will be subject to the geopolitical blessing of the American government in perpetuity. It can also be seen as a violation of a country’s sovereignty.

This might all be justified on the grounds of the China threat. But it will surely spook partners of the American government that also seek to develop their own foundational models. Countries will try accelerating efforts to train models using decentralised computing resources instead of large centralised clusters. Every nation-state will want to develop its own AI compute ground-up.

As for India, these developments underline that its progress in AI over the next decade will hugely depend on India’s geopolitical choices. Build domestic capabilities over time by aligning with the West on technology governance. Let us not choose the losing side again.

---

HomeWork

1. [Post] Another way to understand why protests occur is the Gurr Framework. We explained it in edition #163.

2. [Podcast] Check out our Puliyabaazi with Rahul Verma on the broad trends in India’s voter preferences over the last decade.

3. [Video] In this OoruLabs video, we discuss tax devolution in detail.

---