Book Talk: The Benefits and Costs of India’s Internal Security Strategy
India has faced a wide a range of internal security issues since independence, from complex insurgencies to terrorist attacks, communal violence, and electoral violence. In a new Talking Policy episode, Amit Ahuja and Devesh Kapur discuss their groundbreaking new volume, Internal Security in India: Violence, Order, and the State. Amit is an Associate Professor of Political Science at the University of California, Santa Barbara. Devesh is the Starr Foundation Professor of South Asian Studies at the Johns Hopkins University School of Advanced International Studies. In the interview, they explain who does what, the successes of the security apparatus, and troubling challenges and what they might mean for the future of India’s democracy. This interview was recorded on April 24, 2023. The transcript has been edited for length and clarity.
Subscribe to Talking Policy on Spotify, Apple Podcasts, Soundcloud, or wherever you get your podcasts.
Since independence from colonial rule, internal security challenges have claimed more lives in India than all of India’s external wars combined. Can you give our listeners some background on India’s security context? What have been the major threats over the years?
India is recognized as the largest country in terms of its population. It’s a large country, a subcontinent, masquerading as a country. Given its large population and immense diversity, it has faced, over the decades, all kinds of security challenges and these have claimed lots of lives. There has been communal violence, ethnic violence, riots, terrorist attacks, insurgencies, campus violence, labor strikes, gender violence, homicide, caste violence, election-related violence. The challenges across decades and across regions have been immense. And the Indian state has faced these using a large number of government departments, security forces, and sets of laws.
You mentioned the number of people who’ve died. Let me give you one example. Across four of India’s major insurgencies, whether it’s the left-wing insurgencies in Central India today, the North East India insurgency, the insurgency in Kashmir, or the insurgency that’s ended in Punjab—if you look at the total number of deaths across these four insurgencies, it comes to above 75,000. That’s three times the total number of deaths across India’s five wars. If you add to that the number of people who’ve been lost in ethnic violence and pogroms, the number goes up even more.
This violence threatens the Indian state. It threatens different groups and the relationships between different groups. And, of course, it’s also felt beyond the state by individuals because it threatens individual lives and livelihoods. And it’s been there from the beginning; [during Partition] the foundational moment of modern India, between 1 to 2 million people died. The killing was at an industrial scale, and 10 million people were displaced. The people who framed the Indian constitution were feeling this threat—they realized the enormity of the challenge of internal security threats, because they were seeing it right in front of them.
In the vast majority of countries in the last half-century or so, deaths from internal violence vastly exceed deaths in wars, because interstate wars have really died. Most conflict is intrastate and hence internal violence is high. If you think of countries like Brazil or Mexico or South Africa, homicide rates are 8-10 times higher than in India. In Africa, internal conflict in Ethiopia, Libya or Sudan and the Sahel more broadly, has led to very large casualties. But internal violence in East Asia is much less. In most countries, when we think about violence, it’s internal violence.
Even today, the number of people who’ve died in the Ethiopian conflict, which is a civil war, is around 600,000. And no number that’s been put forward for the war in Ukraine comes anywhere close to that.
A lot has been written on security in South Asia. Your book looks at the internal institutions in India, which makes it a unique contribution. You and your co-authors provide a detailed look at all aspects of the internal security apparatus, from who does the enforcement, to who sets the policies, to who provides oversight. So, who are the security players? Who does what?
India is a federal polity. And like most federal polities, law and order is a subnational subject. Local police forces provide day-to-day law and order—and these are controlled by individual states. Unlike in the United States, where you have something like 17,000-18,000 individual police forces (every county or town has its own), in India, it’s very centralized at the second tier [of government] at the state and only the large cities have their own police forces. Otherwise, the police forces are controlled by the states.
The central government has something called the Central Armed Police Forces. (Most of the police in India are actually not armed.) The Central Armed Police Forces are brought in when large-scale violence breaks out, when the local police forces are unable to resolve the situation. India has low numbers of police relative to its population. The budgetary resources devoted to them are pretty low. India also has one of the lowest rates of incarceration of any large country. The United States is orders of magnitude higher. The entire law and order machinery is understaffed, underfunded, and poorly trained. The surprise is [that] despite this, levels of violence in India, compared to say, Africa and Latin America, are substantially lower. They’re higher than in East Asia which has some of the lowest levels of violence in the developing world. One of the things we wanted to understand is why have overall levels of violence as measured by a variety of indicators gone down despite pretty modest amounts of investment in the security apparatus.
Amit, can you talk about the role of the military in internal security? How does the military relate to and interact with these other groups?
It’s really interesting, because in some ways, especially if you compare [India] to the United States, the contrast is really sharp in terms of the military’s use in internal security, because there are restrictions on that in the United States. But, if you think about developing countries, [military involvement in internal security] is not unusual at all. In fact, across the developing world, a lot of what militaries actually do is not fighting interstate wars, but providing internal security or disaster relief. In that sense, the Indian military is not very different.
The Indian military also inherited its internal security role from its colonial predecessor. If you think about the colonial army, one-third of that army at any given time was [involved in] internal security. The difference is that was a colonial army enforcing colonial rule. The Indian military answers to a democratic government.
The military gets called out regularly to assist civil authorities to quell riots when they break out; it gets called out when there are floods or earthquakes to provide disaster relief. Something that both I and co-author Srinath Raghavan focus on is [the military’s] role during insurgency, which becomes very prominent in the 1980s and 1990s, when the military was called out to deal with growing insurgencies in the North East, in Punjab, and in Jammu and Kashmir. [What is really interesting is that the military, especially the Army, creates a force—the Rashtriya Rifles—[that consists of] about 75,000 troops who are devoted to internal security issues in Jammu and Kashmir. It also expands to the paramilitaries—the Assam Rifles—a force that’s used specifically in the North East and expands its role and size to work with military units deployed in that region.]
What we find remarkable is how much learning goes on over a period of time. The Army uses different kinds of tactics in different regions to deal with insurgents. It remains under civilian control, but it has an expanded role [and] it’s protected by special laws. Some would argue [that involvement in internal security] distracts the Army from its primary role, which is interstate conflict. And, I want to emphasize this, given India’s challenge in its neighborhood: it is flanked by Pakistan on one end, and it’s fought four wars with that country and has ongoing border disputes there. On the other end, it’s flanked by China, which has resulted in one war and other military skirmishes. India’s interstate challenges are not trivial. But, the internal conflicts have made major claims on the military’s resources, and in some ways have taken it away from its primary task. As violence has gone down and been controlled, some of that is now being corrected. The Army is now returning to its regular role of confronting external challenges.
To add to Amit’s point, [there is] a recognition that the Army was not happy to be called into internal security, especially riots and so on. Civilians also realized that that’s bad for democracy. That was one of the reasons for the very substantial expansion of the paramilitary forces. The idea was that the Army’s core role should be protecting the country from external threats, not internal ones.
And, this is not a very popular role with the Army. I’ve surveyed Army officers, and a majority of them have said, “Look, this is not something that we’re supposed to do. But, you know, this is a choice made by the civilian authorities to put the Army in.” In that sense, they are obeying orders when those are given.
It’s also because the Army enjoys one of the highest legitimacy of any Indian institution, which is why if there is violence and the Army is called in, that’s something that gives a lot of reassurance to the civilian population. It’s seen as a relatively less biased and more competent force. But over time, its role has declined. And now for the most part, other than counterinsurgency operations, its internal role has been taken up increasingly by the central armed police forces.
What determines India’s government’s approach to security? In the absence of a formal doctrine, how has India’s government decided where to devote attention and resources?
The foremost principle at the highest level that drives the Indian state’s approach is order. The first responsibility any state has is to provide order and state building. And almost everywhere it has been accomplished, it has been accompanied by violence, because state building requires a centralization of violence in one institution, which is the state. Any secessionist movement has been met with violence by the state, but the state’s approach has always been twofold. One is to offer the party a chance to share power. [This approach] uses the democratic process to draw them in. But if the movement is driven on religious grounds, then it doesn’t make compromises. It is willing to compromise on most other things.
The Indian state has performed better in delivering order but its effectiveness in maintaining the rule of law and ensuring justice leaves much to be desired. That’s because the emphasis has been on building state capacity in the police machinery to maintain order. Everything else that pertains to rule of law and justice—be it the investigative arms of the police such as the quality of forensic labs, the training of investigators, the prosecutorial machinery, courts—has been starved of resources and reforms. Cases in courts are delayed by years, sometimes decades. Evidence vanishes, witnesses are bought off or die or are silenced. India’s record on providing justice and upholding the law is much weaker and this has been true for all governments, irrespective of their political stripes. There have been numerous official reports on these failures and how to address them for more than six decades. Every decade, there’s a major report pointing this out. The problems are discussed in parliament, in the media, but nothing happens because a weak investigative and legal system serves the partisan and personal interests of the political elite.
Devesh is absolutely right that if you look at some of these pathologies, both in terms of how the state thinks about problems and also the persistent gaps in capacity, they span a long period of time across governments from different political ideologies. That being said, there is also something interesting and noteworthy about the Indian approach, which is that it is persistent. That old joke out of Afghanistan—you may have the watch, but we have the time—the Indian state, as one of our contributor’s notes, has both the watch and the time. Once it gets involved, it stays involved. Some of the counterinsurgency campaigns have run for over three or four decades. Gradually, it figures out how to bring down electoral violence, it builds special forces to deal with ethnic violence. There’s a chapter, for example, in the book on the Rapid Action Force, which is specifically meant for riots. And so, there’s learning. It may be gradual learning, but there is a degree of persistence which eventually begins to wear down the opponent.
How diverse are security organizations in terms of gender, caste, and religion?
In some ways across some indicators, diversity is improving, and there is a deliberate effort by the government to increase diversity, but these changes have happened in some areas and not in others. So, for example, on gender diversity, there is now a push from the state to diversify the police forces. In some states, this process has gone further than others. Similarly, in armed police forces, there are quarters that have been set up for women’s participation. There’s a chapter in our book that looks at the challenges that come with increasing participation, both of inclusion as well as integration. We also see similar things happening in the Indian Army, where women officers are being allowed to progress. Similarly with caste, given changes in Indian society, those are reflected in the security forces. So, there is a higher number of intermediate backward castes as well as Adivasis, or the indigenous people, and the former untouchables, Dalits. Their participation is increasing across police forces and central armed police forces.
Where we do see issues in terms of diversity is in the representation of Muslims, who are generally underrepresented across the Indian state. This gap is also there in the Indian security forces. So Muslims who are 14 percent of the population remain underrepresented in the Indian police forces, in the army, and in the federal armed police forces. And, if you look at some of the elite agencies, they’re pretty much absent in those agencies.
The representation of women in the police forces is still barely 10 percent. And even though they’re trying to increase, women are not that eager to join, in part because of the working conditions, such that having a family life is much harder in a patriarchal society and that poses particular challenges. And, like Amit said, there is substantial underrepresentation of Muslims in the security forces, which adds to the mistrust of the security forces by this minority group.
Violence has declined overall in India, despite capacity and resource constraints. But you also show that the decline in violence is not the same thing as ensuring durable political settlements. So the conflicts die down, but they don’t go away.
The Indian state’s record across indicators is actually in some ways quite remarkable—violence has come down from high levels. If you just look at, for example, issues of ethnic insurgencies, violence is controlled. Numbers are down for casualties across civilians, military personnel, and insurgents. But, do we get a political settlement? Not always. There is a larger operational strategy here, which is both shooting and talking happen simultaneously so the government never shuts the door on negotiations. It encourages political participation. It encourages the inclusion of insurgents in the political process. And, these conversations, these negotiations sometimes take a long time. And they haven’t always resulted in resolution.
Amit is absolutely right. A really good example is Kashmir, where violence is now much lower than it was in the 1990s and early 2000s. Yet there hasn’t been a political settlement. So you can control violence through a strong coercive apparatus of the state. But that control of violence does not imply that if that coercive apparatus of the state was removed, it would not come back. And in the democracy, you have to think and work toward a political settlement. India has managed that better, especially in places like the Punjab where you had a major insurgency in the 1980s and early 1990s, and in some parts of the Northeast. There’s been a major left-wing insurgency among the indigenous peoples of central India, and violence has come down a lot there, too. And there, the answer has been to provide much more development, and a realization that indigenous peoples were badly treated, and that the state did not really care about their development problems. So there, the state has poured in a lot of resources to try and mitigate these gaps. But fundamentally, as Amit said, security forces can end violence for a period. But to sustain the absence of violence, you need a political settlement.
Trust is so essential for security—trust between the security players themselves and between society and the security apparatus. If trust is damaged, if there’s a perception that the security agents are being used for partisan political purposes, or that they are the purveyors of violence or the enablers of violence instead of the protectors, that damaged trust is a deep foundational problem. Some would argue that since Modi’s party came to power in 2014, that distrust is deepening. What do you think?
If we start with the police, because for most citizens their day-to-day law and order [experience] is the local police (their encounter with the Army is very rare; their encounter with the central paramilitary forces is also pretty uncommon, except in certain places like Kashmir, the North East, and the central tribal belt). There is no doubt that when polls are taken [asking] which institutions citizens trust, the police rank very low. One reason that all the commissions of inquiry that have called for giving greater autonomy to the police, and no political party has any interest, is because they want to use the police for partisan purposes. In the case of insurgencies, in any insurgency which goes on for a protracted period of time, the security forces will begin to be seen as an oppressive force. They are sometimes seen to be the cause of violence, rather than a way to bring down violence. The longer a security force stays in an area, because, once you stay in an area, you will do search operations, civilians will be stopped, they will feel humiliated—all of those things build distrust. So, in places where you had the security forces there for protracted periods of time, it almost inevitably does breed distrust.
Now, that’s not everywhere. A few of our colleagues have tried to work with the police force in one of the Indian states (the state of Madhya Pradesh), and the police have been very receptive, say, to try different ways to give greater security and trust among women, because women face a lot of day-to-day violence. So there are attempts [to build trust], but overall, I would say the way political parties use police forces for partisan ends almost always implies that there will be mistrust.
You note in the book that India’s constitution privileges national security over individual civil liberties “in ways that are uncommon for long-standing democracies.” Why is India different in this regard and what does this mean for the character and quality of the world’s largest democracy?
The challenge here is that on one hand, there is a very large set of challenges that India faces, and the Indian state, in many ways, if you look at its record on controlling violence, and managing violence, has done a remarkable job in just sheer numbers. It is remarkable not only compared to India’s neighborhood, but also compared to what has happened or what is currently happening in some of largest countries in the developing world in Latin America and Africa. Devesh pointed out the price India has paid is in terms of the legal infrastructure it has built to deal with these kinds of security threats—some of the laws are intrusive, they are harsh, and they come at the expense of civil liberties. This has been a difficult challenge to manage for India, because there have been violations of civil rights. And this is not peculiar to one government, it is not peculiar to one time period. This is something which has been there for decades.
It has always been one of the foremost challenges of the Hobbesian state: how do you control the Leviathan itself? You’ve given it so much coercive power, how do you ensure that that power is used solely and only for the purpose for which it was assigned? That’s why in democracies, we try and come up with various checks and balances.
What we’ve seen in recent years in India, is that the courts are much more willing to defer to the state. The check and balance on the executive either come from the legislative branch, which is Parliament or the state legislatures, or it comes from the courts and from civil society (like the media). And in India, all these checks and balances have weakened so that deference to the executive branch is given when individuals are picked up supposedly on matters of national security, there is less proof demanded of the state to prove its case. And you can’t get bail for long periods of time, you’re not brought before the court—the state has been given a lot of latitude in how long it can hold a person without bail. All of these things undoubtedly undermine civil liberties which are so essential for any democracy. As Amit said, this is not new—this has been there. For instance, India under British times had a law of sedition. And under that law, many of the leaders of India’s nationalist movement went to jail. Ironically, India retained that law in the new constitution after independence. And while even Britain abolished this law because it saw its potential for misuse, India has retained it all these years. And, of course, it’s been misused.
It brings us back to the question of order, and the emphasis on order. This is a strong preference of the Indian state. When it comes to that tradeoff between order and liberty, you can have order without liberty, but you can’t have liberty without order. The Indian state over a long period of time has taken a clear position in terms of its preference for order.
Insofar as the book was a scholarly endeavor to understand how the world works, and insofar as it was the endeavor of a group of engaged academics who want to change how the world works—what motivated you to write this book at this time in both of your careers? Why do you care about this beyond filling a gap in the literature?
I don’t know if I have a really good answer for you, Lindsay. We were doing some surveys on women’s workforce participation in India, which is low and has been declining. And one simple question we would ask parents—what is the latest time in the evening you would be comfortable with your daughter coming back home alone, if she went out for whatever reason—could tell you something about how we think about violence and the fear of violence. We see that at work in the United States. African American teens—the fear that if you press the wrong bell, what could happen to you? Violence is not just about what happens, it’s about the larger atmosphere of fear.
There have been major changes in Indian society relative to the past, when we look at the lower castes, for example, which were very oppressed, we see considerable improvements. But there’s a long way to go. When it comes to gender, it’s quite evident that this is rooted in society, it’s not rooted in the state. And it’s therefore much harder to change. Hindu-Muslim violence clearly has both a societal but also a political dimension. Student protests and violence, labor protests, and violence—we see sharp declines in both compared to the 1970s and 1980s. So, it’s a mixed bag. We were seeing these very different trends, and we began to wonder, what should we make of this as a whole, and that’s what prompted me.
My interest in this started when I was working on a book on the military. When I started looking at the military, I was struck by the significance of its internal security role, and how much it has defined its view of itself and how it operates. I’ve looked at the Indian state and internal security issues and general security issues across my work. A lot of my work deals with processes of inclusion and exclusion, and just like Devesh, I think about fear, I think about security, because these are central to feeling included, and they are central to the process of exclusion. That’s how you exclude, by increasing somebody’s insecurity, by increasing fear in them.
The picture of security in India has transformed. I grew up in India during the 1980s and 1990s. That moment in Indian history, when violence left such a deep impression on Indian society, is something I felt and experienced firsthand, and, so for me, at a personal level, the difference in today’s India is very, very noticeable. We wanted to explore, not just what the state was doing, but to understand this from the side of the people who are on the front lines, in the name of the state, dealing with security challenges. And you see that across our chapters—what are the challenges that those personnel actually face, how does the state actually manage to provide security, what does that entail, and what are the gaps? Those gaps needed to be highlighted, because in the provision of the security, the state is actually impacting not only itself, but it’s also impacting the political process. It’s impacting society. And some of those changes are going to have long-term effects.
Internal Security in India: Violence, Order and the State, edited by Amit Ahuja and Devesh Kapur—thank you both for being with us on Talking Policy.
Thank you for having us, Lindsay. Really appreciate it.
Thank you, Lindsay, for having us.