Three grams of sodium pentothal, three litres of distilled water and dextrose. That’s all it took to get Abdul Karim Telgi talking. Telgi, accused in the Rs 3,000-crore stamp paper scam, was administered a “truth serum” in 2003. So have others in high-profile cases such as Aarushi Talwar’s murder or the 2008 Malegaon bomb blasts. Their drug-induced bare-alls have called into question a powerful new tool in criminal investigations. Yesterday, in Selvi vs State of Karnataka, a three-judge bench of the Supreme Court provided an answer. The court held that lie detectors (polygraph tests), brain-mapping (P300) and truth serums (narco-analysis) are not compulsory. Suspects have a right to say no.
The key word here is “compulsory” because the judgment does not ban tests to which the suspect acquiesces. Besides, narco-analysis is inadmissible in court (though some high courts have held otherwise). Therefore convicting a suspect on the basis of a doped confession was not the SC’s primary worry. But can the police drug a suspect to aid their investigations, even without his permission?
The court said no, and in doing so, relied on two constitutional rights we all enjoy against the state. These are the right against self-incrimination and the right to a fair trial guaranteed under Articles 20(3) and 21 of the Constitution. The court held that truth serums are not merely physical tests like medical examinations that are routinely performed on suspects; they are “testimony” that suspects have a right not to give. The court went on to hold that such invasive tests violated the suspect’s right to privacy, amounted to “cruel, inhuman or degrading treatment” and might expose suspects to “non-penal consequences” such as beatings and torture. Besides, the dodgy science behind such tests made it too unreliable to ensure a “fair” trial.
It is difficult to argue with such a judgment, especially given how truth serums, taken to be instruments to gain the absolute truth, are insidiously used to taint. In September 2009, tapes of narcotic tests done on the two priests and nun accused of murdering Sister Abhaya in Kottayam were broadcast by a Kerala news channel. The video showed these three confessing, in drug-induced stupor, to their dark deeds. A magistrate soon ordered the channel to stop airing the video, but the damage had been done. What could not be admitted in a court of law was used to convince in the court of public opinion.
The SC verdict is also in line with liberal democracies elsewhere. Canada and the UK outlaw non-voluntary confessions induced through truth serums. In 1963, the US Supreme Court held that confessions induced through “truth serums” were inadmissible in court, and questioned their effectiveness. In the landmark 1966 case of Ernesto Miranda vs Arizona, the US Supreme Court emphasised the voluntary nature of pre-trial investigation. The famous Miranda warning — “You have the right to remain silent. Anything you say can and will be used against you in a court of law” — is etched in the memory of every Hollywood fan. This case was, in turn, referred to by the Indian Supreme Court in 1978. In Nandini Satpathy vs P.L. Dani, the court held that Article 20(3) protected suspects during the investigation stage itself. This week’s SC decision relies heavily on the Nandini Satpathy case.
The world might have come to outlaw truth serums, but India has something that these countries don’t: a pathetically low conviction rate. That’s the single biggest problem in our criminal justice system (if you don’t count years and years of delay, that is). This means that investigative agencies need every tool that aids them. As former Karnataka Director General of Police R. Srikumar says, “A criminal knows the crime more than anybody else, and tools that help us make him speak are always useful.” He should know. Srikumar was the investigating officer in the Telgi case. Srikumar agrees that these tests are not yet scientifically certain. “But as investigators we are always looking for new tools to get to the bottom of the crime.”
In his book Suspect Identities: A History of Fingerprinting and Criminal Identification (Harvard University Press, 2001), criminologist Simon A. Cole chronicles how long it took for fingerprinting classification systems, pioneered by the Argentinean and colonial Bengal police forces in the mid-1890s, to spread; it is now a ubiquitous tool in police investigations. Perhaps that is why we must draw limited inferences from the Supreme Court’s judgment. The judgment does not ban truth serums, polygraph tests and brain-mapping altogether. It merely forbids force and coercion, and holds these tests to have limited value. The science of crime detection is a work in progress, and a Luddite fear of technology must not prejudice us against new and improved tools.