Witness Protection in India: A Mixed-Method Empirical and Comparative Analysis with Special Reference to the Post-2023 Criminal Law Reforms

Abstract

India occupies a unique and troubling paradox in the global common law world. It is the world's largest democracy, home to a constitutionally entrenched adversarial criminal justice system with an independent judiciary, a vibrant legal profession, and a Supreme Court that has over decades expanded the scope of fundamental rights in ways that are the envy of the developing world. Yet India remains, as of 2025, one of the very few major common law jurisdictions without a standalone parliamentary statute dedicated to the protection of witnesses in criminal proceedings. This absence is not merely a legislative oversight, it is, as this paper empirically demonstrates, a structural failure with measurable, quantifiable consequences for the administration of criminal justice in India. This paper undertakes a comprehensive mixed-method empirical and comparative legal analysis of witness protection in India, set against the backdrop of the most ambitious criminal law reform India has undertaken since independence: the replacement of the Indian Penal Code, 1860 by the Bharatiya Nyaya Sanhita, 2023 (BNS); the Code of Criminal Procedure, 1973 by the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS); and the Indian Evidence Act, 1872 by the Bharatiya Sakshya Adhiniyam, 2023 (BSA). These three statutes represent a fundamental reimagining of Indian criminal law. And yet, astonishingly, none of them contains a dedicated witness protection provision, and none codifies or supersedes the Witness Protection Scheme, 2018 (WPS 2018)- the only instrument that presently governs witness protection in India, and which itself is merely an administrative scheme approved by the Supreme Court rather than an Act of Parliament. The paper draws on three interconnected empirical methodologies: first, a quantitative analysis of conviction rate data from the National Crime Records Bureau (NCRB) Crime in India Reports spanning 2018 to 2023, which reveals a consistent and statistically significant underperformance in conviction rates across precisely those categories of offences, organised crime, terrorism, sexual violence, communal violence, and corruption, where witness intimidation is most prevalent; second, a case-based empirical analysis of high-profile hostile witness incidents drawn from reported Supreme Court and High Court decisions and corroborated institutional sources, including the Best Bakery case, the Sohrabuddin Sheikh encounter case, and the 2008 Malegaon blast trial; and third, a systematic nine-dimensional comparative benchmarking of the WPS 2018 against the statutory witness protection regimes of the United States (18 U.S.C. sections 3521 to 3528 the WITSEC programme), the United Kingdom (Youth Justice and Criminal Evidence Act 1999, Serious Organised Crime and Police Act 2005, and the Coroners and Justice Act 2009), and Australia (Witness Protection Act 1994, Cth). The paper establishes three central findings. First, the hostile witness phenomenon in India is not anecdotal it is empirically verifiable, structurally predictable, and causally linked to the absence of enforceable statutory protection. Second, the WPS 2018, while representing a necessary and commendable administrative step, suffers from five deep structural deficiencies, the absence of an enforceable legal right to protection, an independent institutional authority, dedicated statutory funding, an operational identity-change mechanism, and uniform national implementation standards, and the 2023 reforms have done nothing to cure any of these deficiencies. Third, the three comparator jurisdictions offer specific, constitutionally compatible, and legally adaptable legislative mechanisms that can be transposed into the Indian framework to address each identified deficiency. The paper concludes by proposing a detailed legislative roadmap for a dedicated Witness Protection and Anonymity Act for India, built around six structural pillars: a National Witness Protection Authority; a three-condition anonymity test; a National Witness Protection Fund; an inter-agency identity document framework; a hearsay exception for witnesses in fear; and criminal sanctions for unauthorised disclosure. The paper argues that such legislation is not merely a matter of policy preference but a constitutional necessity under Article 21 of the Constitution of India, an international legal obligation under the United Nations Convention against Transnational Organized Crime (UNTOC) and the International Covenant on Civil and Political Rights (ICCPR), and an institutional imperative for the credibility of India's criminal justice system.