Our Three-Year Analysis Reveals How India’s Anti-Terror Law Is Used To Jail People Without Clear Offences

LUBHYATHI RANGARAJAN
 
17 Feb 2026 4 min read  Share

Overcoming data hurdles, this investigation in Karnataka shows that the UAPA functions less as a narrowly defined security law and more as a mechanism of prolonged incarceration: most accused were booked during BJP rule, over 84% were Muslim, charges were frequently dropped before trial, and acquittals vastly outnumbered convictions. The study demonstrates how the law’s vagueness enables arrests for offences that are not clearly defined, allowing the law to be used as a narrative management system.

A memorial in Karnataka to 32-year-old BJP worker Praveen Nettaru, who was murdered in July 2022, one of five instances between 2021-23 during a BJP government, when the killers, all Muslim, faced charges of terrorism. All those murdered were members of the Rashtriya Swayamsevak Sangh, its affiliates, or BJP functionaries. The UAPA was not invoked during this period when Hindus similarly murdered Muslims/ MOHIT RAO

Recently, the Social Life of Law project at SOAS University of London and Article 14 collaborated over three years on a four-part investigative series examining the use of India’s principal anti-terror law, the 59-year-old Unlawful Activities (Prevention) Act (UAPA), in Karnataka between 2005 and 2025. 

Our study exposed long-standing fault lines in the UAPA’s use, tracing cases from the filing of a first information report (FIR) to final outcomes at trial.

Part 1, published on 2 February 2026, outlined the data, methods, and headline findings. Part 2, Part 3, and Part 4 examined how UAPA cases are constructed across political regimes and how trial outcomes expose structural flaws in the law’s design and application.

Among the key findings: 

- Nearly eight in 10 accused persons were booked during the 10.5 years the Bharatiya Janata Party (BJP) was in office 

- Of the 925 individuals accused charged in this period, 783—84.6%—were Muslim 

- Charges were dropped before trial in over 37% of cases 

- Acquittals outnumbered convictions by more than five to one (244 acquitted versus 46 convicted) 

- 80% of convictions were secured through guilty pleas 

Why study the UAPA at all? 

Three years ago, after publishing A Decade of Darkness, , India’s first digital repository of sedition cases from 2010 to 2020, we set out to build a similar dataset on UAPA prosecutions. 

Apart from the National Crime Records Bureau (NCRB), there is no nationwide data on the use of this law. In recent years, the UAPA has been stretched to keep people incarcerated, functioning as what is best described as a “narrative management system.” Though enacted by Parliament, it is deliberately framed so vaguely in the name of national security that arrests can be made with little clarity on what conduct actually constitutes an offence.

Narrative Control

Over time, the UAPA has increasingly functioned less as a narrowly tailored national security statute and more as a tool to keep people incarcerated, often operating as a mechanism of narrative control. 

We soon realised that a nationwide dataset was nearly impossible to build. The volume of cases—nearly 9,000 in a decade, according to NCRB data—was compounded by the absence of reliable judicial records online. Unlike sedition, the UAPA has almost no publicly available data in states such as Manipur and Jammu and Kashmir, where scholars, journalists, and civil society organisations have long documented its extensive use since its enactment in 1967.

Between April and June 2022, we undertook a manual, all-India data-mining exercise on the e-Courts Portal, reviewing records across more than 3,000 court complexes. In one sessions court in Jammu and Kashmir, the portal returned “data not found”. Nearly a year later, revisiting the same court revealed 77 UAPA cases suddenly listed as pending. FIRs dating back to 2009 now showed first appearances in 2023, spanning bail hearings, miscellaneous applications, and trial commencements.

Encountering this anomaly triggered three responses in me. 

First, unease: the discovery cast doubt on the reliability of data across India and raised questions about whether the exercise would need to be redone, with uncertain time and resources. Second, anger: at the instinct to blame myself rather than a public system that is expected to maintain accurate, real-time judicial records. Third, intrigue: what do such gaps mean for structuring empirical work on anti-terror laws in India? If a comparatively well-funded judicial platform like the e-Courts Portal produces such inconsistencies, mapping how the UAPA is used, and against whom, becomes far more uncertain for a researcher ‘outside’ the system.

These limitations forced us to shift from a national study to a state-specific one. Karnataka offered a rare comparative balance: nearly equal periods of rule by the Indian National Congress and the BJP, with the Janata Dal (Secular) as coalition partner over two decades. 

Overcoming Data Blocks

It still took two more years of fieldwork and intensive triangulation—verifying every case and every accused through court records, media reports, and FIRs, where available—to arrive at the findings presented in this four-part series.

Again, why build a dataset on the UAPA, or on crime more broadly? 

First, to get demographic information—religion, caste, education, socio-economic background—about those accused and imprisoned. This reveals how the State encounters its citizens at the ground level and whose lives are most deeply entangled in coercive legal processes.

Second, to move beyond a process-heavy focus that dominates official crime data. The NCRB’s emphasis on speed, pendency, and disposal reflects broader anxieties about backlog and efficiency, but obscures a more fundamental question: who is being arrested under these laws in the first place?

Data access has only grown more difficult. Since 2014, entry to public institutions has narrowed, the right-to-information regime has weakened, and routine accountability mechanisms have eroded. Researchers are forced to rely on creativity and triangulation—combining desk research with investigative fieldwork—to reconstruct what the State itself no longer reliably discloses.

This work demands considerable creativity and resourcefulness, a condition that characterises much research from the Global South. Our method of having academics work with journalists, is one way out of this bind. However the sustainability of such collaborations is uncertain, owing to time and resource constraints. 

Ultimately, this series aims to demonstrate, through evidence rather than rhetoric, that some laws cannot be applied fairly or reformed incrementally through legislative or judicial means. 

The UAPA is one such law. 

(Lubhyathi Rangarajan is a lawyer and now a research fellow at SOAS University of London, working on a project titled the Social Life of Authoritarian Legality. She is also Editor-Databases at Article 14. She led and conceptualised the sedition and UAPA projects.)

Get exclusive access to new databases, expert analyses, weekly newsletters, book excerpts and new ideas on democracy, law and society in India. Subscribe to Article 14.

   Share this story