The State has a legitimate and urgent interest in investigating crime, locating victims, preserving evidence, and identifying offenders. That interest, however, is exercised within a constitutional order that treats the individual as a bearer of dignity rather than as a source of information available for compulsory extraction. Narco-analysis brings this tension into unusually sharp focus. The procedure uses a psychoactive substance, commonly sodium thiopental, to place a person in a sedated and disinhibited condition while questions are asked. Its popular description as a “truth-serum” test implies that the drug opens a dependable route to concealed knowledge. The scientific and legal record supports no such confidence. A person whose inhibitions are reduced may speak more freely, but may also become confused, compliant, suggestible, or prone to filling gaps in memory. The procedure therefore alters both the subject’s capacity to choose whether to answer and the reliability of what is said.
The constitutional objection is deeper than the ordinary rule excluding an involuntary confession. Narco-analysis changes the mental condition under which communication occurs. It places chemical means between the suspect’s will and the State’s questions. Involuntary administration can compel personal knowledge, interfere with bodily integrity, penetrate mental privacy, and create medical risk. Even where a person formally agrees, detention, promises of release, fear of adverse inference, inadequate legal advice, or misunderstanding of the procedure may compromise consent. The subject’s capacity to withdraw also becomes uncertain once sedation begins. These features distinguish the technique from the collection of fingerprints, handwriting, voice samples, or other physical characteristics that exist independently of a suspect’s volitional testimony.
Indian constitutional doctrine now supplies a clear starting point. Article 20(3) protects an accused from being compelled to be a witness against himself. Article 21 requires any invasion of life or personal liberty to satisfy a fair, just, and reasonable procedure, and protects dignity, bodily integrity, and privacy. Article 22(1) protects access to legal counsel after arrest. In Selvi v. State of Karnataka, a three-judge bench of the Supreme Court held that involuntary narco-analysis, polygraph examination, and brain electrical activation profile testing violate Articles 20(3) and 21.1 Fifteen years later, Amlesh Kumar v. State of Bihar reaffirmed the prohibition, clarified the limited treatment of voluntary tests and derivative discoveries, and held that an accused has no indefeasible right to demand narco-analysis.2
The replacement of the Code of Criminal Procedure, 1973 and the Indian Evidence Act, 1872 by the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS) and the Bharatiya Sakshya Adhiniyam, 2023 (BSA) makes renewed analysis necessary. The new statutes modernise terminology and certain procedures, but do not confer a power to conduct narco-analysis or dilute constitutional guarantees. The BSA continues to exclude police confessions and custodial confessions made outside the immediate presence of a Magistrate, while preserving a narrow discovery rule. The BNSS continues to regulate police examination and Magistrate-recorded confessions. Neither framework converts chemically induced answers into admissible testimony.
This article addresses a gap that remains after the formal prohibition of forced testing: whether “voluntary” narco-analysis can be ethically meaningful in custody, and what evidentiary consequences may follow from it under the new criminal laws. It uses doctrinal legal research, reading constitutional text, legislation, judgments, professional medical standards, and relevant scientific material together. Its central argument is that voluntariness is necessary but not sufficient. Because the technique is scientifically unreliable and compromises continuing autonomy, it should never be treated as proof or as a routine investigative shortcut. At most, a narrowly confined, judicially supervised defence request may be considered, and any resulting discovery must be independently proved and corroborated.
Thiopental is an ultra-short-acting barbiturate used intravenously to induce anaesthesia or deep sedation.3 In a narco-analysis session, a controlled dose is intended to produce a state between full consciousness and unconsciousness. Questioners assume that reduced inhibition will weaken a subject’s ability to fabricate and thereby reveal information that conscious resistance would suppress. This assumption confuses lowered inhibition with enhanced truthfulness. The drug does not separate accurate recollection from fantasy, suggestion, error, or deliberate falsehood. It acts upon consciousness and responsiveness, not upon the truth-value of propositions.
Memory is reconstructive rather than a perfect recording. Recall can be influenced by the wording, sequence, and repetition of questions, by expectations communicated by the examiner, and by information supplied during the interview. Sedation may further impair attention and judgment. A subject may accept a suggestion embedded in a question, combine remembered facts with imagined details, or offer an answer to satisfy an insistent examiner. Confidence and fluency do not establish accuracy. The danger is amplified when investigators already possess a theory of the case: confirmatory questioning may generate apparent support for that theory while excluding alternative explanations.4
The technique also involves genuine medical intervention. Dosage must account for age, weight, health, medication, allergies, and individual sensitivity. Respiratory depression, cardiovascular complications, allergic reactions, and injury associated with deep sedation are foreseeable risks. A forensic setting does not remove the physician’s duty to protect the person undergoing the procedure. The medical professional is not merely a technician operating an investigative instrument; professional judgment and the obligation to avoid harm continue to apply.5
Legal analysis becomes clearer when three different objects are separated. The first is the audiovisual recording or written report of what the subject said under the drug. The second is information communicated during the session, for example an alleged location of an object. The third is physical or other evidence later discovered and proved through lawful investigative steps. Treating these as a single product obscures both constitutional and evidentiary limits.
The first object is the most problematic. A drug-induced response is testimonial in substance because it communicates personal knowledge, yet it is produced while mental control is altered. Selvi therefore rejected direct admission of the test results even where the procedure is voluntary.6 The BSA independently reinforces this conclusion through its rules governing inducement, coercion, police confessions, and custodial confessions.7 A recorded answer does not become reliable or admissible merely because it is technologically captured.
The second object, information supplied in the session, may guide investigators, but a lead is not proof. It can be inaccurate, contaminated by prior questioning, or impossible to distinguish from information already known to the police. Investigators who act on a lead must document what was known before the procedure, the exact information obtained, the steps taken afterwards, and the chain of custody of any recovered object. Without this discipline, the discovery exception may become a route for laundering an inadmissible statement into apparently objective evidence.
The third object may possess evidentiary value if it has an existence independent of the compelled or drug-induced response and is proved according to law. The proviso to section 23(2) of the BSA, like former section 27 of the Evidence Act, preserves only so much of the information received from a person in police custody as distinctly relates to the fact discovered.8 The proviso is narrow. It does not admit the entire narrative, establish the truth of every assertion, or dispense with proof that the discovery was genuine. In Amlesh Kumar, the Supreme Court further held that a voluntary test report, or information found as a result of it, cannot form the sole basis of conviction without other evidence.9 This corroboration requirement is essential because physical recovery may show knowledge or connection, but its meaning still depends upon context, lawful proof, and the exclusion of innocent explanations.
Article 20(3) declares that no person accused of an offence shall be compelled to be a witness against himself. The guarantee is not confined to testimony delivered from the witness box. In M.P. Sharma v. Satish Chandra, the Supreme Court described the protection as extending to the compelled production of personal knowledge by an accused.10 State of Bombay v. Kathi Kalu Oghad later distinguished testimonial compulsion from the collection of physical identifying material, explaining that being a witness ordinarily involves imparting knowledge derived from one’s mental faculties.11 Narco-analysis falls on the testimonial side of that distinction because its purpose is to obtain communicative responses based on memory, perception, and thought.
The guarantee also operates during investigation. Nandini Satpathy v. P.L. Dani recognised that the right to silence would be hollow if it protected only courtroom answers while permitting coercive interrogation beforehand.12 Compulsion is not limited to physical violence. Threats, pressure, deception, and circumstances that overbear meaningful choice can engage the privilege. Narco-analysis makes the problem structural: chemical alteration reduces the very mental control by which a person decides whether and how to answer.
The State might argue that the subject is not “compelled” when consent is recorded. That argument is incomplete. Voluntariness must be real, informed, specific, and continuing. A signature obtained in custody does not resolve whether the person understood the test, received independent advice, believed that refusal would harm bail or the investigation, or remained capable of withdrawing consent. Selvi specifically warned that threats, false promises, or deception could produce apparent consent.13 Accordingly, Article 20(3) requires an inquiry into conditions, not a formal document alone.
Nor can the constitutional defect be cured by describing the procedure as a “medical examination”. The ordinary collection of a biological sample or examination of an injury observes physical facts. Narco-analysis is administered for the purpose of eliciting answers. Its testimonial purpose and effect cannot be reclassified through terminology. The State cannot do chemically what it is constitutionally forbidden to do through coercive questioning.
Article 21 prohibits deprivation of life or personal liberty except according to procedure established by law. Since Maneka Gandhi v. Union of India, such procedure must be fair, just, and reasonable rather than arbitrary or oppressive.14 Involuntary narco-analysis lacks specific statutory authorisation, involves bodily intrusion and altered consciousness, and pursues information through a method of doubtful reliability. These features make it incompatible with substantive and procedural fairness.
The invasion affects both body and mind. Introducing an anaesthetic without valid consent intrudes upon bodily integrity and exposes the subject to medical risk. Questioning the sedated subject intrudes upon the mental domain. K.S. Puttaswamy (Retd.) v. Union of India located privacy in dignity, liberty, and autonomy, including the protection of personal choices and the inner sphere of thought.15 Mental privacy is especially important because freedom of thought precedes outward expression. A constitutional system that protects speech and silence cannot permit the State routinely to bypass both by chemically reducing a person’s capacity for self-control.
Proportionality supplies a further discipline. A rights-infringing measure must pursue a legitimate aim, have a rational connection to that aim, be necessary in the sense that less restrictive alternatives are inadequate, and maintain a proper balance between public benefit and individual harm.16 Crime detection is legitimate, but narco-analysis is weakly connected to accurate fact-finding because it does not reliably distinguish truth from confabulation. Less intrusive alternatives, such as scene reconstruction, digital forensics, financial tracing, witness interviewing, surveillance records, forensic biology, and conventional corroboration, are generally available. The severe interference with bodily and mental autonomy is therefore difficult to justify.
Fairness also protects the integrity of the criminal process. Investigators may become anchored to drug-induced statements and discount contrary evidence. Media disclosure of sensational answers may create public prejudice long before trial. An unreliable technique can thus harm not only the tested person but also victims, witnesses, and the accuracy of adjudication. Article 21’s concern with fair investigation and fair trial requires institutions to prefer methods that preserve, rather than distort, the search for truth.
Article 22(1) provides that an arrested person shall not be denied the right to consult, and to be defended by, a legal practitioner of his choice.17 Legal assistance in the narco-analysis context must be practical and timely. Counsel should explain that the subject may refuse, that the report is not direct substantive evidence, that derivative discoveries may be used if lawfully proved, and that medical risks exist. Advice delivered only after police persuasion or immediately before sedation cannot neutralise earlier pressure.
Independent counsel also serves an evidentiary function by helping to create a reliable record of voluntariness. Yet counsel’s presence does not solve every problem. Once the drug takes effect, the lawyer cannot restore the subject’s decisional capacity. The right to withdraw must therefore be operationalised through conservative medical stopping rules and continuous monitoring, not left as a theoretical entitlement exercisable only by a fully alert person.
Section 180 of the BNSS authorises a police officer making an investigation to examine orally any person supposed to be acquainted with the facts and circumstances of the case. The person must answer truly, except for questions whose answers would tend to expose that person to a criminal charge, penalty, or forfeiture.18 Section 181 restricts the use of statements made to the police during investigation, while section 183 regulates confessions and statements recorded by Magistrates.19 These provisions assume communication by a person capable of hearing, understanding, and choosing an answer. They do not authorise chemical impairment to defeat the statutory exception for incriminating answers.
The BNSS provisions permitting medical examination of an arrested person do not supply the missing authority. Such provisions are directed to ascertaining facts of which examination of the body may afford evidence of an offence, subject to statutory conditions and to reasonable force where expressly permitted.20 Narco-analysis is materially different. The injection is not used merely to observe the body or to secure an existing sample; it is used to alter the mind so that testimonial information may be elicited. Extending a general medical-examination power to that purpose would contradict Articles 20(3) and 21 and the binding rule in Selvi.
The BSA continues a layered exclusionary structure. Section 22 renders a confession irrelevant when it is caused by inducement, threat, coercion, or promise from a person in authority, connected to the charge in a manner that would reasonably give the accused grounds to suppose that confessing would produce an advantage or avoid an evil.21 Section 23(1) bars proof of a confession made to a police officer. Section 23(2) bars proof of a custodial confession unless it is made in the immediate presence of a Magistrate.22
A narco-analysis report cannot evade these rules by being labelled a forensic result. Its content consists of answers elicited through questioning, and it is ordinarily produced for the police while the subject is in custody or under investigative control. Even an apparently exculpatory answer is scientifically suspect and cannot bind the court. The prosecution must prove its case through admissible evidence; the accused ordinarily bears no burden to demonstrate innocence by undergoing an invasive procedure.
The discovery proviso to section 23(2) requires careful application. Only the portion of the information that distinctly relates to the discovered fact may become provable. The prosecution must establish that the fact was genuinely discovered because of the information, that it was not already known, and that the recovery and chain of custody are reliable. Where information emerged during a constitutionally prohibited involuntary test, Amlesh Kumar indicates that the report and the subsequent information are not per se admissible.23 Where a test is voluntary and judicially authorised, derivative evidence remains subject to the BSA, independent proof, and corroboration. The proviso is an evidentiary exception, not a licence to conduct the underlying procedure.
Earlier constitutional cases created the foundation for Selvi. Kathi Kalu Oghad distinguished physical identification from personal testimony. Nandini Satpathy carried the right against self-incrimination into police questioning and treated compulsion realistically. D.K. Basu v. State of West Bengal responded to custodial abuse by insisting upon arrest and detention safeguards.24 Together these decisions reject the idea that investigative efficiency can displace dignity and procedural control.
In Selvi, the Supreme Court considered involuntary narco-analysis, polygraph examination, and brain electrical activation profile testing. It held that the techniques elicit personal knowledge and therefore involve testimonial responses. Forced administration violates Article 20(3) when used against an accused, and Article 21 more broadly, because it intrudes upon bodily integrity, mental privacy, and personal liberty without fair procedure.25 The Court refused to create an exception based on the seriousness of the crime or the perceived utility of modern investigation.
The decision also addressed scientific reliability. Because a subject may speak in a suggestible state and cannot exercise full control, the results are not equivalent to voluntary testimony. The Court prohibited direct admission of test results, even where testing is voluntary. It nevertheless recognised that information leading to a subsequent discovery could be considered under the then section 27 of the Evidence Act, subject to ordinary evidentiary rules.26 For voluntary testing, it adopted the safeguards developed by the National Human Rights Commission for polygraph tests: consent must be recorded before a Judicial Magistrate; the person must have access to counsel; the physical, emotional, and legal implications must be explained; the Magistrate must consider the conditions of detention and the nature of the questioning; and the procedure must be conducted by an independent agency and recorded.27
Selvi is sometimes read as leaving a broad consensual pathway. That interpretation overlooks the judgment’s reasoning. Consent removes one threshold objection; it does not make the technique scientifically dependable, transform its responses into admissible evidence, or create an investigative entitlement. The safeguards are conditions of legality, not certificates of accuracy.
In Amlesh Kumar, the Patna High Court, while considering bail, accepted an investigating officer’s proposal to conduct narco-analysis on all the accused persons and other witnesses. The Supreme Court set that order aside. It stated categorically that an involuntary or forced test is impermissible, and that a court considering bail should not order specific investigative techniques or convert the proceeding into a mini-trial.28 The Court thereby reaffirmed that judicial enthusiasm cannot substitute for consent or statutory authority.
The judgment answered two questions that had produced uncertainty. First, a voluntary report cannot be directly admitted, and neither the report nor information found as a result of it can be the sole basis of conviction without supporting evidence.29 The principle protects against the false appearance of certainty that may attach to a laboratory procedure. Second, an accused may request narco-analysis at an appropriate stage, when exercising the right to lead defence evidence at trial, but has no indefeasible right to obtain it.30 The trial court must assess the request in light of the circumstances, the procedure’s suspect nature, and the Selvi safeguards.
This position properly avoids two extremes. It prevents the prosecution or the court from imposing the technique as an investigative shortcut. It also prevents an accused from presenting self-selected drug-induced answers as if they were scientifically privileged proof. A court retains control of relevance, reliability, fairness, and the integrity of the trial. Amlesh Kumar should therefore be understood not as endorsing voluntary narco-analysis, but as confining a residual possibility within judicial discretion.
Judicial recording of consent is indispensable but does not eliminate custodial pressure. A suspect may believe that refusal will prolong detention, weaken a bail application, anger investigators, or be treated as consciousness of guilt. A person seeking the test may also overestimate its exculpatory capacity because of the “truth-serum” myth. Genuine consent requires disclosure that the procedure cannot guarantee truth, may generate false or misleading statements, carries medical risks, and can produce derivative investigative consequences.
Capacity is dynamic. Consent obtained before sedation does not necessarily remain meaningful after the drug affects consciousness. Medical ethics ordinarily permits a competent patient to withdraw consent, but a sedated subject may be unable to communicate a stable choice. This makes narco-analysis different from a single, completed sample collection. The procedure involves continuing questioning during progressive impairment. A rights-respecting framework must therefore treat ambiguity, distress, unexpected depth of sedation, or resistance as grounds to stop.
Autonomy requires more than the absence of overt force. It demands adequate information, comprehension, voluntariness, and decision-making capacity. Each element is fragile in custodial narco-analysis. Information may be incomplete; comprehension may be distorted by fear; voluntariness may be undermined by dependence on the police and the courts; and capacity is deliberately diminished during the procedure. The ethical defect is not incidental but built into the technique’s design: investigators seek speech precisely because ordinary conscious control has been weakened.
The medical principle of non-maleficence requires the avoidance of unnecessary harm. Administering an anaesthetic without therapeutic benefit calls for especially strong justification. The World Medical Association’s Declaration of Tokyo prohibits physicians from participating in torture or cruel, inhuman, or degrading treatment, and from using medical knowledge to diminish a detainee’s ability to resist such treatment.31 Not every voluntary narco-analysis session is legally equivalent to torture, but the Declaration identifies the professional danger of placing clinical expertise at the service of coercive interrogation. The physician’s primary responsibility cannot be displaced by investigative objectives.
Role conflict is acute. The examining doctor may be asked to assess fitness, administer the drug, monitor vital signs, and facilitate questioning. If the same professional is answerable to investigators, independence may be compromised. A medically safe dose does not make coercive questioning ethical, and an ethically recorded consent does not remove medical risk. Separate medical and investigative responsibilities, independent clinical authority to terminate the session, and post-procedure care are minimum safeguards.
Narco-analysis recordings often contain allegations, intimate information, speculation, and statements about third parties. Disclosure can cause lasting reputational damage even when the material is false or irrelevant. Medical confidentiality and informational privacy therefore require strict control over recording, storage, access, transcription, and destruction. Investigators should not release recordings to the media, and courts should use sealed procedures where necessary to protect uninvolved persons.
Public fascination with the idea of a truth serum creates a further danger. Refusal may be portrayed as evidence of guilt, while ambiguous answers may be broadcast as a confession. Both responses invert the presumption of innocence. A suspect has no duty to assist the prosecution by surrendering mental autonomy. Courts should expressly prohibit any adverse inference from refusal and should sanction unauthorised disclosure.
Hard cases, such as missing persons, terrorism, organised crime, or imminent danger, are frequently invoked to justify intrusive techniques. Urgency deserves serious attention, but constitutional guarantees are most vulnerable when the facts are disturbing and public pressure is intense. Amlesh Kumar itself arose from a missing-person allegation; the Supreme Court nonetheless refused to sacrifice Articles 20(3) and 21.32 Rights are not obstacles external to criminal justice. They are conditions of legitimate investigation.
Narco-analysis may also impose opportunity costs. Time spent arranging, conducting, and interpreting an unreliable session may divert resources from verifiable evidence. False leads can expose innocent third parties, contaminate witness accounts, and delay rescue efforts. The technique’s theatrical appearance of scientific certainty can weaken disciplined investigation. Public safety is better served by methods whose error rates, chain of custody, and inferential limits can be examined in court.
The strongest policy response would be the statutory exclusion of narco-analysis from criminal investigation, while leaving courts to consider genuinely therapeutic or clinically indicated sedation under ordinary medical law. If the residual defence-request pathway recognised in Amlesh Kumar is retained, legislation or binding rules should define it narrowly. The following safeguards operate cumulatively; the failure of any material safeguard should terminate the request or exclude its products.
First, no police officer, prosecutor, or court should direct an involuntary test, and refusal must carry no adverse inference. A request should originate from the accused personally, through independently instructed counsel, and only at the defence-evidence stage identified in Amlesh Kumar. Investigators should not obtain “consent” by first proposing the test.
Second, a Judicial Magistrate should conduct a private, recorded examination after a cooling-off period. The Magistrate should explain the absence of scientific certainty; the right to refuse and to withdraw; the medical risks; the inadmissibility of direct responses; the potential use of independently discovered facts; the limits of confidentiality; and the absence of any promise concerning bail, discharge, or acquittal. The Magistrate should examine the conditions of custody and ask whether threats, inducements, media pressure, or family pressure influenced the request.
Third, an independent medical assessment should address physical and psychiatric fitness, current medication, substance use, allergies, and capacity to consent. The clinician responsible for safety must have authority to refuse or stop the procedure without institutional penalty. Emergency facilities, resuscitation equipment, dosage records, continuous physiological monitoring, and post-procedure observation should be mandatory. No procedure should be conducted upon a child, a person unable to give informed consent, a pregnant person, or a medically vulnerable person where the risk is materially elevated.
Fourth, a neutral forensic institution, not the investigating police unit, should conduct the session. Questions should be settled in advance under judicial supervision, framed in non-leading language, and confined to relevant issues. The entire process, including the pre-test briefing, drug administration, questioning, recovery, and post-test debriefing, should be audio-visually recorded with secure time stamps. Counsel should be able to observe, while clinical personnel retain the power to end questioning.
Fifth, consent must be continuing. Before each material phase, the subject should be asked whether the process may continue. Because sedation can impair communication, predetermined stopping criteria must favour autonomy: distress, inconsistent assent, unexpected sedation, or a request to stop ends the session. Silence or inability to respond can never be treated as consent.
Sixth, the report and the responses should be inadmissible as substantive evidence and unavailable for impeachment. Any claimed derivative discovery must be litigated in a focused admissibility hearing. The proponent should prove the exact information that distinctly led to the discovery, the absence of prior police knowledge, the legality of the recovery, an intact chain of custody, and independent corroboration. The court should give a reasoned ruling and exclude the evidence where an involuntary test, a material breach of the safeguards, or unreliable documentation tainted the process.
Seventh, confidentiality rules should restrict access to named officials, counsel, the court, and necessary experts. Copies should be logged, encrypted, and retained only as long as legally necessary. Unauthorised disclosure should attract disciplinary and legal consequences. Third-party and medically sensitive material unrelated to a lawful discovery should be sealed or destroyed under judicial direction.
Finally, every authorised request should be reported in anonymised form to an independent oversight body. The report should record the origin of the request, the court’s reasons, the consent process, medical events, the questions asked, claimed discoveries, admissibility rulings, and the outcome of the case. Periodic public statistics would allow evaluation of whether the residual procedure produces useful, reliable evidence or merely consumes resources and endangers rights. Without such data, exceptional use can expand invisibly.
These safeguards do not validate the “truth-serum” premise. They recognise that the Constitution places the burden on the State to justify intrusion, and on the courts to prevent unreliable science from acquiring evidentiary authority. The preferred investigative model remains the patient accumulation and testing of independently verifiable evidence.
Narco-analysis promises access to concealed truth but delivers, at best, chemically disinhibited speech of uncertain accuracy. Its scientific weakness and constitutional intrusiveness are connected: the same impairment that reduces a person’s ability to resist also increases the danger of suggestion, confusion, and confabulation. The procedure therefore cannot be treated as a reliable shortcut to proof.
Indian law establishes firm boundaries. Article 20(3) prohibits the compelled extraction of testimonial knowledge from an accused. Article 21 protects bodily integrity, mental privacy, dignity, and fair procedure, while Article 22 requires effective legal assistance. The BNSS does not create a power to conduct chemical interrogation, and the BSA’s discovery proviso neither authorises the method nor admits an entire drug-induced narrative. Selvi correctly prohibited involuntary testing and the direct evidentiary use of results. Amlesh Kumar reinforced that rule, required supporting evidence for any result or consequent information, and rejected an indefeasible right to demand testing.
The remaining category of voluntary testing must be approached with scepticism. Custody changes the meaning of choice; legal advice cannot restore capacity once sedation begins; and medical participation creates serious conflicts between care and investigation. The law should therefore exclude narco-analysis from routine investigation. If a narrowly confined defence request is retained, it should arise only at the defence-evidence stage and remain subject to judicial discretion, independent advice, robust medical screening, audiovisual documentation, continuing consent, confidentiality, and rigorous proof of any derivative discovery.
The central answer to the research problem is accordingly clear: investigative necessity cannot make involuntary narco-analysis constitutional, and consent alone cannot make it reliable. A criminal process earns legitimacy by proving guilt through lawful and dependable evidence, not by weakening a suspect’s control over body and mind.
*****
1. Selvi v. State of Karnataka, (2010) 7 SCC 263, 328–69 (India).
2. Amlesh Kumar v. State of Bihar, 2025 INSC 810, ¶¶ 8–23 (India).
3. Nat’l Ctr. for Biotechnology Info., PubChem Compound Summary for CID 23665410, Thiopental Sodium, PubChem, https://pubchem.ncbi.nlm.nih.gov/compound/23665410 (last visited Sept. 4, 2026) (describing thiopental as an intravenously administered barbiturate used to induce general anaesthesia).
4. Selvi, (2010) 7 SCC at 304–16.
5. See id. at 311–16; Nat’l Ctr. for Biotechnology Info., supra note 3.
6. Selvi, (2010) 7 SCC at 366–68.
7. The Bharatiya Sakshya Adhiniyam, No. 47 of 2023, India Code (2023), §§ 22–23.
8. Id. § 23(2) proviso.
9. Amlesh Kumar, 2025 INSC 810, ¶¶ 14–16.
10. M.P. Sharma v. Satish Chandra, AIR 1954 SC 300, 306–07 (India).
11. State of Bombay v. Kathi Kalu Oghad, AIR 1961 SC 1808, 1815–17 (India).
12. Nandini Satpathy v. P.L. Dani, (1978) 2 SCC 424, 435–48 (India).
13. Selvi, (2010) 7 SCC at 360–69.
14. Maneka Gandhi v. Union of India, (1978) 1 SCC 248, 281–84 (India).
15. K.S. Puttaswamy (Retd.) v. Union of India, (2017) 10 SCC 1, 263–65, 297–98 (India).
16. Modern Dental College & Research Centre v. State of Madhya Pradesh, (2016) 7 SCC 353, 411–13 (India).
17. India Const. art. 22(1).
18. The Bharatiya Nagarik Suraksha Sanhita, No. 46 of 2023, India Code (2023), § 180.
19. Id. §§ 181, 183.
20. Id. §§ 51–53.
21. The Bharatiya Sakshya Adhiniyam, No. 47 of 2023, India Code (2023), § 22.
22. Id. § 23(1)–(2).
23. Amlesh Kumar, 2025 INSC 810, ¶¶ 8–9.
24. D.K. Basu v. State of West Bengal, (1997) 1 SCC 416, 435–39 (India).
25. Selvi, (2010) 7 SCC at 328–69.
26. Id. at 366–68.
27. Id. at 368–69.
28. Amlesh Kumar, 2025 INSC 810, ¶¶ 8–13; see also Sangitaben Shaileshbhai Datanta v. State of Gujarat, (2019) 14 SCC 522, 524–25 (India).
29. Amlesh Kumar, 2025 INSC 810, ¶¶ 14–16; Vinobhai v. State of Kerala, 2025 SCC OnLine SC 178, ¶ 8 (India).
30. Amlesh Kumar, 2025 INSC 810, ¶¶ 17–21.
31. World Med. Ass’n, Declaration of Tokyo: Guidelines for Physicians Concerning Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment in Relation to Detention and Imprisonment ¶¶ 1–2, 5–7 (rev. Oct. 2016), https://www.wma.net/policies-post/wma-declaration-of-tokyo-guidelines-for-physicians-concerning-torture-and-other-cruel-inhuman-or-degrading-treatment-or-punishment-in-relation-to-detention-and-imprisonment/.
32. Amlesh Kumar, 2025 INSC 810, ¶¶ 2–13.