Mobilised but not democratised: Protest and political regimes in South Asia

Across South Asia, citizens continue to protest even as democratic institutions weaken. Some movements have forced leaders from office; others have won concessions or sustained opposition under repression. These outcomes raise a question: how far does mobilisation translate into democratic change?

The Bertelsmann Transformation Index (BTI) records uneven democratic trajectories in the region. India, which BTI rated a “democracy in consolidation” until the 2014 edition, has been a “defective democracy” since 2016. Its Democracy Status score fell from 8.1 in 2014 to 6.10 in 2024 and has not moved since. Pakistan, the region’s second-largest country, has not qualified as a democracy in any of the BTI editions since 2006. Specifically, Pakistan is a “hard-line autocracy”, with the country’s democracy score declining from 3.75 in 2020 and 2022 to 3.25 in 2026, indicating further authoritarian entrenchment. Bangladesh, after two turbulent decades oscillating between “defective democracy” and “highly defective democracy,” has been a “moderate autocracy” since 2018. Its democracy score has nonetheless improved modestly, from 4.25 in 2022 to 4.72 in 2026. Nepal, Sri Lanka, and Bhutan are the remaining South Asian countries¹ to have stayed on a relatively stable democratic path in recent years, yet all three are classified as defective democracies. Overall, in 2026, no South Asian country qualifies as a consolidating democracy.

Despite this regression, citizens across the region continue to exercise their right to assembly. They have mobilised against their governments on issues of public concern, such as the recent examination paper leaks in India in May 2026, and in defence of democracy itself, as in the opposition-led protests in Pakistan since 2022 and the 2024 uprising in Bangladesh that ended the Awami League’s government. Repressive governments in the region have repeatedly sought to suppress mass protest through intimidation and violence against civil society actors and ordinary citizens.² Yet protesters have also shown that attempts at autocratic consolidation, particularly in the absence of reliable social provision, will not go uncontested.

The scale of mass mobilisation in South Asia is substantial. According to the Armed Conflict Location and Event Data Project (ACLED), South Asia, defined here as Bangladesh, Bhutan, India, the Maldives, Nepal, Pakistan, and Sri Lanka, recorded one of the highest levels of mobilisational activity worldwide between January 2021 and 5 October 2025, second only to Europe over the period and the highest of any region in 2024 (Figure 1). Over this period, the region saw 139,483 protest events and 6,365 riot events. Figures 1 and 2 count demonstrations, combining ACLED’s protest and riot categories. Protest events involve demonstrators who do not use violence, whether or not they meet intervention or force; riot events include violent or destructive activity by demonstrators. A detailed account of individual protest events or the broader waves they belong to falls outside the scope of this article, but several broader trends can still be identified.

First, despite historically weak levels of democracy, mass mobilisation by citizens, organised groups, and civil society actors is nonetheless common across the region, Bhutan being the only exception to a trend of frequent protests in other South Asian countries (Figure 2). Although India leads the South Asia region in raw demonstration event counts, because event counts scale with population and with ACLED’s source coverage, the per-million panel of Figure 2 gives a more contextualised picture: Nepal and the Maldives record the highest demonstration prevalence adjusting for population, while India, despite accounting for 60 per cent of the region’s events, ranks near the bottom. Although the prevalence of protest is meaningful, aggregate levels of protest do not automatically indicate democratic vitality. Mobilisation can serve narrow, instrumental purposes, such as attempts by some actors to secure material concessions or policy accommodations from the regime, while leaving the structure of political power unquestioned. In such cases, protest may even strengthen the regime, since it relieves public pressure in the short term and lends the regime a measure of legitimacy by creating the false perception of an open and responsive system. Protests in China and Russia illustrate this logic. Their tolerated forms of protest and state-managed civil societies perform only a limited version of the accountability function that contentious politics performs in democracies, and leave aside demands for structural political change. Our analysis therefore considers the prevalence of mobilisation for democracy and autocracy across all South Asian countries over a 26-year period (2000-2025) and assesses major trends and changes for this particular form of protest.

Figure 3, based on V-Dem data, shows that mobilisation for democracy has outweighed mobilisation for autocracy across South Asia in almost every year since 2000, except in Bhutan, where there is little of either. On V-Dem’s 0 to 4 scale, pro-autocracy mobilisation stays below 1 in most countries and periods, while pro-democracy scores typically sit between 2 and 3. Nepal recorded the strongest pro-democracy mobilisation in the early 2000s (3.04 in the six-year period from 2000-2005). In 2006-2010, democracy mobilisation scores shifted markedly in the region: whereas Bangladesh’s fell to 1.72 during the 2007-2008 emergency, the Maldives rose to 2.88 and Pakistan to 2.62. Mobilisation for democracy then weakened in parts of the region during 2016-2020, with Bangladesh (1.92) and Nepal (2.01) recording markedly lower scores, whereas India registered its strongest pro-democracy score of the period (2.75). In 2021-2025, pro-democracy mobilisation rebounded in Sri Lanka, Pakistan, Bangladesh and, in 2025, Nepal, with Sri Lanka scoring the highest (3.79 in 2022), followed by Bangladesh (3.66 in 2024), Pakistan (3.50 in 2022, its highest annual score since 2000) and Nepal (3.12 in 2025, its highest since 2006), while India’s period average fell from 2.75 to 2.40. Pro-autocracy mobilisation is mostly limited. It exceeds democracy mobilisation only in Bangladesh in 2020 and the Maldives in 2013, 2015, and 2016, and in Bhutan where both are near zero. In India, autocracy mobilisation remains low throughout but has risen from 0.06 in the early 2000s to 0.38 in 2021-2025. India’s trend is therefore better described as rising pro-autocracy mobilisation alongside pro-democracy mobilisation that peaked in 2016-2020 and then declined, rather than as a one-sided shift.

Over the past two decades, South Asian countries have experienced varying forms and degrees of autocratisation, driven by both civilian political actors and military rulers. Yet authoritarian actors across these cases have also faced persistent resistance from civil society actors, opposition parties, and ordinary citizens.

In Bangladesh, Sri Lanka, and Nepal, sustained mass mobilisation has contributed to the fall of the dynastic and increasingly authoritarian governments of Sheikh Hasina and the Rajapaksas in 2024 and 2022, the end of direct monarchical rule in Nepal in 2006 and the removal of a Nepali government widely seen as corrupt and repressive in 2025. By contrast, authoritarian actors in India and Pakistan have thus far preserved their dominance.

In India, authoritarian consolidation has primarily occurred under a dominant political party since Narendra Modi first became prime minister in 2014. His Bharatiya Janata Party (BJP) has since won two further terms, most recently in 2024. In Pakistan, the military has increasingly shed the facade of hybrid rule since Asim Munir’s appointment as army chief in November 2022. Munir has since emerged as Pakistan’s de facto chief executive. The 26th Amendment (October 2024) and the 27th Amendment (November 2025) curtailed the superior judiciary, facilitated executive aggrandisement and, through the new post of Chief of Defence Forces, consolidated Munir’s control over Pakistan’s three military services.

In both cases, however, consolidation has involved the capture or weakening of key state and democratic institutions, primarily elections (see the Special Intensive Revision of electoral rolls controversy in India and the disputed results of the February 2024 election in Pakistan), but also the judiciary and the media. Despite these efforts, authoritarian actors have not been able to fully suppress dissent, which persists through political opposition, civil society mobilisation, and non-conventional forms of resistance.

In India, established opposition parties have challenged the Modi government’s consolidation of power, but one of the most notable challenges came from the social media phenomenon known as the Cockroach Janta Party (CJP). Originally a satirical Instagram account, the CJP gradually gained public traction and evolved into a broader political movement. Its most consequential intervention was a sit-in from late June to 25 July 2026 at Jantar Mantar in New Delhi triggered by the leak and cancellation of the NEET medical entrance exam. Buoyed by support from renowned education activist, Sonam Wangchuk, and several others who went onto hunger strike to protest the leaks, the CJP demanded the resignation of the education minister, Dharmendra Pradhan. The movement has since become a political party with a manifesto. The Indian National Congress (INC), the leading opposition party, endorsed the CJP’s demands and organised separate demonstrations, deliberately refraining from joining the sit-in because it seeks to be the central party representing India’s youth. While it also risks facing politically motivated criticism from the BJP over “weaponising protests” to compensate for electoral defeats, such allegations have not deterred the INC from engaging in other street mobilisations. Even though multiple regional and smaller ideological parties backed the cockroach agitation in June and July, the INC’s distance was still politically significant, as it allowed the CJP to frame its initial mobilisation as a civic protest rather than partisan opposition. In an established democracy, a ministerial resignation may appear routine. However, in the context of growing executive dominance and for a government that has resisted opposition pressure, a minister’s resignation following sustained public mobilisation should be read as a meaningful challenge to its authority.

Democratic breakdown has been more pronounced in Pakistan, where the military has expanded its influence through electoral manipulation, extensive censorship of legacy and social media, constitutional changes affecting the judiciary and armed forces command structure, and the tight constraint of civilian leadership. These measures have not eliminated dissent. The most sustained opposition has come from former Prime Minister Imran Khan and his Pakistan Tehreek-e-Insaf (PTI), which have continued to contest the political order established after Khan’s removal from office in April 2022. Khan has been imprisoned since August 2023, in conditions his family, his party and the UN Special Rapporteur on torture describe as prolonged solitary confinement, which the jail authorities deny. The PTI has faced extensive restrictions, arrests, and institutional pressure. Even so, the party has continued to mobilise supporters and demand the restoration of democracy, above all free and fair elections, arguing that despite an uneven electoral playing field, it was denied victory in the February 2024 elections through widespread manipulation of results. It also challenges the legitimacy of the legal cases against Imran Khan, whom it maintains, along with other party leaders, has been imprisoned on politically motivated charges.

While coercive institutional consolidation has narrowed the space for opposition in Pakistan, it has yet to eliminate the latter’s capacity for resistance.

Bangladesh offers one of the most striking recent cases of authoritarian breakdown in the region. In 2024, mass protests forced Sheikh Hasina, who had governed for fifteen years, out of office. The aftermath was deeply polarising. Politically motivated violence increased in the country and the Awami League’s founding role for Bangladeshi independence came under attack. Protesters burned down the Dhanmondi 32 home of Bangladesh’s founding president, Sheikh Mujibur Rahman, and vandalised his statue in Dhaka. In 2025, the interim government led by the Nobel laureate Muhammad Yunus, in office since August 2024, amended the country’s freedom fighters’ law to delete the “Father of the Nation” designation for Sheikh Mujibur Rahman.

The 2026 elections following the 2024 unrest were held amid widespread partisan violence, with frequent use of firearms. Notably, a prominent student activist, Sharif Osman Hadi, associated with the anti-Hasina movement, was assassinated in December 2025, just as the polls had been announced.

Bangladesh is not a consolidated democracy today, and none of the other South Asian countries is clearly on a path to becoming one. At the same time, sustained pressure from below for accountability, economic opportunity, social welfare, and representation may make stable, consolidated autocracy correspondingly harder to achieve. Research on civil resistance and authoritarian breakdown suggests that peaceful mobilisation can raise the costs of repression and encourage elite defection from some kinds of authoritarian setups. At the same time, threatening forms of contentious collective action can also provoke durable authoritarianism rooted in negative “protection pacts” among ruling elites aimed at securing their political survival and suppressing dissent. The descriptive data presented here cannot establish these causal mechanisms or predict regime outcomes; they show only that democracy mobilisation has been persistent and widespread across South Asia with divergent consequences.

Marzia Raza is a PhD candidate at the Institute of Political Science, Heidelberg University, currently on a research stay at the University of Oxford. Muhammad Salman holds a PhD in Political Science and is a guest researcher at the Institute of Political Science, Heidelberg University. This article is cross-published with the Bertelsmann Transformation Index (BTI) blog.

1. The Maldives is not covered by the BTI but is included in the ACLED and V-Dem analyses.
2. For example, see the annual US Department of State Country Reports on Human Rights Practices for India, Pakistan, Bangladesh and Sri Lanka, 2020 to 2024 (state.gov).