Child Witnesses In Evidence Law

Author: LawSphereX

Category: General Law

Published: 2026-08-23T18:47:42.745Z

Updated: 2026-08-30T07:09:00.072Z

CHILD WITNESSES IN EVIDENCE LAW

Khayal Shah, National Law University and Judicial Academy, Assam Varada S., National Law University and Judicial Academy Assam

ABSTRACT

The testimony of a child witness has always occupied a contested space in the law of evidence. Questions about competency, reliability and the conditions of deposition have preoccupied Indian courts since before independence. With the enactment of the Bharatiya Sakshya Adhiniyam 2023, Parliament has carried forward the essential framework of the Indian Evidence Act 1872 while creating certain procedural openings for childsensitive practices. This article examines the legal position of child witnesses under the BSA, tracing continuity and departure from the predecessor statute and critically engaging with persistent gaps. The analysis proceeds through the connected issues of competency, corroboration, in-court protection and procedural justice, concluding with a forward-looking assessment of what meaningful reform requires.

INTRODUCTION

A child who has witnessed or suffered an offence occupies a peculiar position in the Indian criminal justice system. The law asks that child to step forward, take an oath or make a solemn affirmation and then recount, often under the gaze of strangers and the accused, events that may have been deeply distressing. The stakes are high on every side: if the evidence is excluded too readily, real perpetrators escape accountability; if accepted without scrutiny, the risk of false conviction rises. The law of evidence must therefore navigate between these opposing dangers with care. The Bharatiya Sakshya Adhiniyam 2023, which came into force on 1 July 2024 in place of the Indian Evidence Act 1872, carries forward the essential structure of the predecessor statute on this question.1 The provision governing competency of witnesses, now contained in section

124 of the BSA, is in substance the same as section 118 of the IEA.2 The transition, while

symbolically significant as part of the broader decolonisation of the criminal law codes, has not introduced structural change to the rules governing child witnesses. This continuity makes critical analysis both possible and necessary. India has a large and growing body of child witness jurisprudence, developed principally by the Supreme Court and the various High Courts over seven decades. This jurisprudence has progressively moved away from the rigid common law rule requiring corroboration before a child’s testimony could ground conviction, towards a more flexible position treating corroboration as a matter of judicial discretion rather than legal mandate. Alongside this, the POCSO Act 2012 and the Bharatiya Nagarik Suraksha Sanhita 2023 have introduced procedural protections that sit alongside the BSA.

The picture that emerges is one of incremental progress hedged by significant implementation failures. COMPETENCY OF CHILD WITNESSES UNDER THE BHARATIYA SAKSHYA ADHINIYAM 2023 The starting point for any discussion of child witnesses in Indian evidence law is the concept of competency. Under the common law from which Indian evidence law partly derives, there was historically a strong presumption against the competency of children of tender years, 1Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 124 (India).

2 Indian Evidence Act, No. 1 of 1872, § 118 (India).

rooted in the assumption that young children could not understand the nature and obligation of the oath. 3 Indian evidence law broke with this tradition even in its colonial form. Section 118 of the IEA provided that all persons are competent witnesses unless the court considers them unable to understand questions put to them or to give rational answers, treating rationality of understanding as the governing criterion rather than age. 4 Section 124 of the BSA carries this standard forward verbatim. 5 The Explanation to section 124 clarifies that a lunatic is not incompetent to testify unless lunacy prevents rational understanding, reinforcing that the test is functional rather than categorical. 6 The Supreme Court addressed the content of this test authoritatively in Rameshwar v. State of Rajasthan,7 a decision that remains foundational.

The Court held that competency depends on the child’s capacity to understand the questions put to it and to give rational answers, and that the trial court must conduct a preliminary examination — a voir dire — to satisfy itself of this capacity. 8 This functional test has been consistently applied in later decisions. In Panchhi v. State of Uttar Pradesh,9 the Court reinforced that the trial court must record its satisfaction with the child’s capacity before proceeding to record substantive testimony, and the failure to conduct even a brief competency inquiry has in several cases led to appellate reduction of the weight accorded to such evidence. The absence of an age threshold in the BSA is deliberate and reflects a legislative choice to defer entirely to judicial assessment.

The Law Commission of India in its 269th Report noted the difficulty of fixing any bright-line age threshold and recommended that the rationality test remain the governing standard. 10 This aligns with the UN Guidelines on Justice in Matters Involving Child Victims and Witnesses, which caution against rigid age-based competency requirements that may exclude children perfectly capable of providing accurate accounts. 11 3John Henry Wigmore, A Treatise on the Anglo-American System of Evidence in Trials at Common Law vol. 2, at 634 (3d ed. 1940). 4Indian Evidence Act, No. 1 of 1872, § 118 (India). 5Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 124 (India). 6Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 124 Explanation (India). 7Rameshwar v. State of Rajasthan, A. I. R. 1952 S. C. 54 (India). 8Ibid. 9Panchhi v. State of Uttar Pradesh, A. I. R. 1998 S. C. 2726 (India).

10Law Comm’n of India, Report No. 269, Amendments to the Indian Evidence Act 1872 on Competency of Witnesses 5 (2017). 11G. A. Res. 2005/20, Guidelines on Justice in Matters Involving Child Victims and Witnesses of Crime, ¶ 9, U. N. Doc. E/2005/INF/2/Add. 1 (July 22, 2005).

The BSA has been read alongside the general provisions on oath to permit children who do not understand the nature of an oath to give unsworn evidence, provided the court records its satisfaction about their capacity for rational testimony. 12 In State of Maharashtra v. Prakash,13 the Supreme Court held that the mere omission to administer an oath does not render the evidence inadmissible if the court has otherwise satisfied itself of the child’s competency. A significant weakness of the current framework is the absence of any structured guidance on how the voir dire is to be conducted: no specification of appropriate questions, no minimum duration and no qualification requirements for the presiding judge.

14 The result is that a trial judge with no understanding of developmental psychology must make a determination with serious consequences for both the admissibility of evidence and the welfare of the child. By reproducing the IEA’s minimalist provision without supplementary rules, the BSA has not addressed this structural problem. The voir dire in many Indian courts remains cursory, and the significant inconsistency across jurisdictions that this generates is a reform imperative the new statute has left unaddressed. CORROBORATION AND THE WEIGHT OF CHILD TESTIMONY Even where a child witness has been found competent, the weight to be accorded to testimony remains contested. The common law historically required corroboration before conviction could rest on a child’s evidence, premised on an assumed unreliability arising from immaturity, fantasy and suggestibility.

15 Indian evidence law has never had a statutory corroboration requirement for child witnesses, but courts adopted a practice rule of requiring corroboration that for a long period operated with much the same force as a legal rule. The landmark decision in Rameshwar v. State of Rajasthan established that the rule of corroboration in child evidence cases is one of caution rather than law. 16 The judge has discretion to convict on the uncorroborated testimony of a child witness provided reasons are recorded. This formulation left room for conviction without corroboration while simultaneously endorsing the practice of seeking it. For several decades Indian courts treated 12Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 124 (India). 13State of Maharashtra v. Prakash, (1992) 4 S. C. C. 225 (India).. 14Sudipto Sarkar & V. R. Manohar, Sarkar on Evidence 1482 (17th ed. 2016).

15Glanville Williams, The Proof of Guilt: A Study of the English Criminal Trial 172 (3d ed. 1963). 16Rameshwar v. State of Rajasthan, A. I. R. 1952 S. C. 54. corroboration as a near-mandatory step in practice. The gradual dismantling of this near-mandatory approach is visible through subsequent decisions. In Harpal Singh v. State of Himachal Pradesh,17 the Court emphasised that reliability should be assessed on the totality of circumstances, including the consistency of the account, the absence of motive to falsely implicate and the natural quality of the narration. In Golla Yellayya v State of Andhra Pradesh,18 the Court reiterated that conviction can be based solely on child testimony if the evidence inspires confidence. Bharwada Bhoginbhai Hirjibhai v. State of Gujarat marks an important moment in this evolution.

19 Dealing with a sexual assault where the only available witness was the child victim, the Supreme Court held that there is no absolute rule requiring corroboration in sexual offence cases and that an Indian court must be alive to the social and cultural context in which women and children make complaints. The Court noted the reluctance of child victims to report and the absence of adult witnesses to such offences as reasons to treat uncorroborated testimony with respect rather than suspicion. This contextual reasoning has been endorsed in later decisions and represents a significant shift in judicial attitude. The BSA contains no provision specifically addressing corroboration of child witness testimony.

20 The governing framework is therefore the case law: corroboration is a matter of judicial prudence, its absence is not a bar to conviction, and the quality and consistency of the child’s account are the primary determinants of weight. This is a sensible position, broadly consistent with empirical research which confirms that the accuracy of child testimony depends far more on the manner in which it is elicited than on any inherent characteristic of the child as a category of witness. 21 More recent decisions have reinforced this trajectory. In State of Rajasthan v. Darshan Singh,22 the Supreme Court held that the testimony of a child witness, if truthful and consistent, should be given full weight without mechanical search for corroboration. In Virender v. State of NCT 17Harpal Singh v. State of Himachal Pradesh, (1981) 1 S. C. C. 560 (India). 18Golla Yellayya v.

State of Andhra Pradesh, A. I. R. 1997 S. C. 343 (India). 19Bharwada Bhoginbhai Hirjibhai v. State of Gujarat, (1983) 3 S. C. C. 217 (India). 20Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 124 (India). 21Maggie Bruck & Stephen J. Ceci, The Suggestibility of Children’s Memory, 50 Ann. Rev. Psychol. 419, 421 (1999). 22State of Rajasthan v. Darshan Singh, (2012) 5 S. C. C. 789 (India).

Delhi,23 the Court went further and stated that insisting on corroboration in POCSO cases would defeat the legislative purpose and send the wrong signal about the credibility of child complainants. Together these decisions have effectively shifted the default in child sexual offence cases from suspicion of child testimony to a cautious acceptance of it, absent positive reasons for disbelief. A concern that arises is the risk that the pendulum may have swung too far in the direction of automatic credence. Empirical research demonstrates that children are capable of highly accurate testimony under appropriate conditions but remain vulnerable to suggestion, particularly where subjected to repeated or leading questioning.

24 The move away from rigid corroboration requirements must therefore be accompanied by attention to the procedural conditions in which child evidence is gathered, a matter on which the BSA is entirely silent, and which demands reform from other sources. PROCEDURAL PROTECTIONS FOR CHILD WITNESSES: STATUTORY FRAMEWORK AND IMPLEMENTATION The substantive evidentiary rules tell only part of the story. The physical setting, the manner in which examination is conducted and the availability of support persons all determine whether a child’s testimony is meaningful. India has developed a patchwork of protections across several statutes, and the challenge is to understand how this patchwork fits together under the new codes. The most significant body of special procedural rules for child witnesses is found in POCSO

2012. Section 26 requires that statements of children be recorded in the presence of a trusted

adult or family member.25 Section 33 requires that the Special Court ensure the child is not exposed to the accused during testimony and may permit evidence to be given through a screen or video-link.26 Section 36 requires that the identity of the child victim shall not be disclosed.27 These provisions reflect legislative recognition that the standard adversarial model is incompatible with child-sensitive justice. 23Virender v. State of NCT of Delhi, (2009) 14 S.C.C. 637 (India). 24Bruck & Ceci, supra note 21, at 425. 25Protection of Children from Sexual Offences Act, No. 32 of 2012, § 26 (India). 26Protection of Children from Sexual Offences Act, No. 32 of 2012, §§ 33, 36 (India). 27Protection of Children from Sexual Offences Act, No. 32 of 2012, § 36 (India).

The BNSS 2023 supplements this framework. Section 319 allows courts to take the evidence of child victims of sexual offences through audio-visual electronic means. 28 Section 398 provides for witness protection, including holding proceedings in camera, prohibiting direct cross-examination of the child by the accused and appointing an intermediary. 29 Judicially, Sakshi v. Union of India30 established guidelines for examining child witnesses in sexual assault cases, including recording in a calm environment and carefully monitoring leading questions in cross-examination. In Nipun Saxena v. Union of India,31 the Supreme Court issued comprehensive directions on POCSO case management, including the duty to record testimony in a single session where possible and scrupulous observation of identity protection requirements.

The BSA itself interacts with this protective framework through general provisions. Section 143, which governs the order of examination,32 and section 146, which addresses leading questions,33 give a trial court considerable power to create child-sensitive conditions for testimony. The problem is not the absence of legal authority but the variable and often inadequate exercise of it. Implementation data reveals a sobering reality. The conviction rate in POCSO cases remains subject to delay and inconsistency, and the backlog in Special Courts is substantial. 34 Childfriendly courtrooms with one-way mirrors or video-link facilities are unavailable in the vast majority of district courts across India.

35 The intermediary system, which requires that questions in cross-examination be routed through a trained intermediary communicating them in age-appropriate language, has been introduced in principle but trained intermediaries are largely unavailable outside metropolitan courts. 36 A child who testifies in a district court far from metropolitan centres is likely to face conditions bearing little resemblance to the childsensitive environment envisaged by POCSO and the BNSS. 28Bharatiya Nagarik Suraksha Sanhita, No. 46 of 2023, § 319 (India). 29Bharatiya Nagarik Suraksha Sanhita, No. 46 of 2023, § 398 (India). 30Sakshi v. Union of India, (2004) 5 S. C. C. 518 (India). 31Nipun Saxena v. Union of India, (2019) 2 S. C. C. 703 (India). 32Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 143 (India). 33Bharatiya Sakshya Adhiniyam, No. 47 of 2023, § 146 (India). 34Nat’l Crime Recs.

Bureau, Ministry of Home Affairs, Gov’t of India, Crime in India 2022 115 (2023). 35Asha Bajpai, Child Rights in India: Law, Policy and Practice 298 (3d ed. 2017). 36Bajpai, supra note 35, at 301.

GAPS, CRITIQUES AND THE PATH TO REFORM

The foregoing analysis reveals a legal landscape that is, on paper, reasonably advanced, but in practice marked by persistent structural weaknesses. This chapter draws together the principal criticisms of the existing framework and advances reform proposals directed at making the BSA and the surrounding statutory regime more responsive to the realities of child testimony. The first and most fundamental gap is the absence of a mandatory, structured competency examination procedure. The BSA provides no guidance on how the voir dire is to be conducted, what questions are appropriate, what minimum duration the inquiry should last or what qualifications a judge should have to conduct it. 37 A trial judge with no understanding of developmental psychology must make a determination that has serious consequences for both admissibility and child welfare.

The introduction of statutory guidelines or court rules specifying the procedure, backed by mandatory judicial training, would be a straightforward and effective reform. The second gap relates to investigative interviewing standards. The BSA governs the admissibility of evidence in court but says nothing about the conditions under which statements of child witnesses are recorded before trial. Suggestive or repeated questioning at the investigation stage can permanently alter a child’s account, rendering even the most careful incourt procedure inadequate. 38 Several jurisdictions, including England and Wales and New Zealand, have introduced statutory interviewing protocols requiring that investigative interviews be conducted by trained officers using open-ended questioning and that such interviews be recorded and made available as evidence.

India has no equivalent statutory framework; the Ministry of Women and Child Development guidelines, while useful, are not legally binding and are poorly enforced. 39 The third gap is the absence of any provision in the BSA or BNSS for trained intermediaries in cases that fall outside POCSO. POCSO’s child-protective mechanisms apply only to offences under that Act. Where a child witnesses a murder, robbery or other offence outside its scope, the child faces ordinary examination and cross-examination without special procedural 37Law Comm’n of India, supra note 10, at 9. 38Julia Hershkowitz et al. , A Meta-Analytic Review of the Effects of Interviewing Strategies on Disclosure of Child Sexual Abuse, 41 Child Abuse & Neglect 28, 30 (2017). 39Ministry of Women & Child Development, Gov’t of India, Guidelines and Protocols: Medico-Legal Care for Survivors/Victims of Sexual Violence 22 (2014).

protection. 40 There is no principled basis for this distinction. The trauma of testifying and the susceptibility to leading questions are features of the child as a developmental stage, not features of the offence witnessed. The BNSS provision in section 398 on witness protection offers some basis for judicial intervention but is framed as a discretion rather than a duty. 41 The fourth gap concerns the oath. For very young children, insisting on formal oath-taking may increase the risk of inaccurate testimony: empirical research suggests that children told that lying is wrong may be more prone to assent to adult suggestions out of a desire to please. 42 The BSA has not engaged with this body of research and its provisions remain anchored to a formalistic understanding of the oath inconsistent with what developmental psychology has established.

The fifth gap is the absence of dedicated child witness courts beyond the limited Fast Track Special Court scheme under POCSO. In Manoj Kumar Pratap Singh v State of Rajasthan,43 the Supreme Court noted with concern the continuing delays in POCSO trials and the failure to provide child-friendly facilities in many Special Courts. Delay itself causes harm: a child required to recount events months or years after they occurred is both less likely to remember accurately and more vulnerable to the distress of revisiting trauma. The sixth issue is legislative fragmentation. The rules governing child witnesses are spread across the BSA, POCSO, the BNSS and various sets of judicial guidelines.

A dedicated chapter within the BSA, consolidating the competency standard, protective procedural requirements, corroboration guidance and oath provisions into a coherent whole, would represent a significant improvement. New Zealand’s Evidence Act 2006 provides a useful template for this approach. These reforms are necessary not merely from a child rights perspective but for the integrity of the criminal process itself. Evidence gathered through inappropriate methods or presented in conditions that overwhelm a child witness is evidence whose reliability is compromised. The demands of fair trial and child welfare are, in this domain, more complementary than they are in conflict. 40Sudipto Sarkar & V. R. Manohar, supra note 14, at 1485. 41Bharatiya Nagarik Suraksha Sanhita, No. 46 of 2023, § 398 (India). 42Thomas D.

Lyon, Child Witnesses and the Oath, in Finding the Truth in the Courtroom 110 (H. Otgaar & M. L. Howe eds. , 2018). 43Nat’l Crime Recs. Bureau, supra note 34, at

118 CONCLUSION

The Bharatiya Sakshya Adhiniyam 2023 inherits a rich and complex jurisprudential legacy on child witnesses. The rationality-based competency standard of section 124, the judicial movement away from mandatory corroboration, and the protective provisions developed through POCSO and the BNSS together constitute a framework that is, in its formal legal dimension, considerably more enlightened than the suspicious exclusionary approach of nineteenth-century common law. The Supreme Court, through decisions from Rameshwar to Nipun Saxena, has progressively developed a body of doctrine that treats the child witness with seriousness and sensitivity. Yet the analysis in this article reveals that the formal framework and the operational reality remain substantially misaligned.

The absence of mandatory competency examination procedures, the lack of statutory investigative interviewing standards, the uneven availability of child-friendly facilities, the limited reach of the intermediary system and the fragmentation of protective provisions across multiple statutes collectively represent a set of failures that no amount of judicial creativity can fully compensate for. The BSA, in retaining the IEA’s minimalist posture on child witnesses, has missed an opportunity to embed these reforms at the level of primary legislation. The reforms proposed in this article are systemic rather than merely textual. Legislative amendment to the BSA can consolidate and extend the protective framework. Statutory interviewing protocols can safeguard the quality of child evidence before it reaches court.

Judicial training can improve competency determinations and the management of child examinations. Investment in child-friendly infrastructure can make formal protections real rather than nominal. What is required is the will to treat the child witness not as an evidentiary problem to be managed but as a participant in the justice process whose dignity and welfare the system is obliged to protect. Evidence law does not exist in isolation from the society it serves. In a country where offences against children remain alarmingly prevalent and where the criminal justice system is often the only formal recourse available to child victims, the adequacy of child witness law is a matter of urgent practical importance.