We need to talk about narcoanalysis

Sriram Lakshman
The Hindu, 2 May

Narcoanalysis is steadily being mainstreamed into investigations, court hearings, and laboratories in India. However, it raises serious scientific, legal, and ethical questions. These need to be addressed urgently before the practice spreads further.

A suspect was ` narcoanalysed' in Bangalore in a 2004 double murder case. In the drug-induced state, she spoke about a knife and purse allegedly involved in the crime but neither was recovered by the police. The outcome: acquittal owing to a lack of evidence. The judge also ruled that the narcoanalysis report and videograph could be used only for investigative purposes and not to convict suspects.

NARCOANALYSIS HAS become an increasingly, perhaps alarmingly, common term in India. It refers to the process of psychotherapy conducted on a subject by inducing a sleep-like state with the aid of barbiturates or other drugs. In a spate of high profile cases, such as those of the Nithari killers and the Mumbai train blasts, suspects have been whisked away to undergo an interview drugged with the barbiturate sodium pentothal.

This practice has also garnered support from certain State governments as well as the judiciary. Politicians have fallen into the habit of hurling the term `narcoanalysis' at opponents. In 2006, Karnataka Congress leader H. Vishwanath suggested that Chief Minister H.D. Kumaraswamy should undergo narcoanalysis in the Chenamma Trust bribery case. The Home Ministry's Directorate of Forensic Sciences plans to expand narcoanalysis facilities nationwide. It is not surprising then that there are about 300 people in the narcoanalysis queue at the Forensic Science Laboratory (FSL) in Bangalore alone.

It would appear that the narcoanalysis beast has acquired a life of its own. It is increasingly knocking at the doors of courts and finding ready acceptance as a device to get at the truth during police investigations, though its scientific basis and value are under strong challenge. It is for this reason that the scientific, legal, and evidentiary issues relevant to the narcoanalysis debate need to be critically discussed.

Narcoanalysis is rarely used for therapeutic purposes today. The reliability of the practice has been questioned by leading psychiatric and forensic experts. Dr. P. Chandra Sekharan, the highly regarded former Director of the Forensic Sciences Department of Tamil Nadu, has characterised the practice as an unscientific, third-degree method of investigation.

It is surely significant that while `truth serums' have been in use since the early part of the 20th century, they are not used in most developed countries today. During and after the War years, United States armed forces and intelligence agencies continued to experiment with truth drugs. The CIA has admitted to using these as part of its interrogation tactics. But a declassified CIA interrogation manual says that while truth drugs can be useful in overcoming resistance not dissolved by other methods, the actual content of what comes out during the interrogation can be "psychotic manifestations ... hallucinations, illusions, delusions or disorientation." At the 1977 U.S Senate hearings on its secret mind-control project, the CIA acknowledged that "no such magic brew as the popular notion of truth serum exists."

Studies have shown that persons who make truthful confessions are those who were likely to confess had interrogators persisted with regular methods; and that persons who lie can continue to manifest a lie even under the influence of a so-called truth serum. Moreover, the investigator can induce and communicate his own thoughts and feelings to the suspect. The scientific literature indicates that if narcoanalysis has any extra-therapeutic uses, it may be in making a suspect feel that he has revealed more than he actually did. With repeated questioning, it may be possible to reduce ambiguities although these cannot be eliminated.

Two objections

Scientific scepticism and the absence of controlled studies have not deterred Indian investigating agencies from running to the FSL in Gandhinagar or, more likely, Bangalore — the narcoanalysis hub for various police departments across the country. FSL, Bangalore, conducts sodium pentothal narcoanalysis in conjunction with three other tests — psychological profiling, polygraph (`lie-detector') tests, and brain mapping. Polygraph tests, which one can learn to `pass' or `fail,' are used for screening and confirmation purposes only. Brain mapping, a premature if promising technique not entirely free from controversy itself, indicates whether a subject's brain stores experiential knowledge about a certain object. Narcoanalysis is used when investigators need oral elicitations from a suspect. For instance, if brain mapping indicates that the suspect stores information about a blue getaway car allegedly used in the crime, the narcoanalysis, according to the FSL, Bangalore, is used to provide information such as the number of the car, where it is parked, and so on.

Dr. B.M. Mohan, Director of FSL, Bangalore, claims that he has data to prove his contention that narcoanalysis has a 96 to 97 per cent total success rate. Included in the definition of `total success rate' is the discovery of information that either triggers a relevant section of the law or may be cross verified with other tests (such as brain mapping). According to Dr. Mohan, findings that discredit narcoanalysis are usually based on studies of scopolamine and sodium amytal and are not applicable to sodium pentothal, which is used by the Indian laboratories. He adds that during narcoanalysis the tendency is to sleep if not questioned, rather than hallucinating or fantasising.

There are two problems with this argument. Using sodium pentothal is not a new advance in narcoanalysis. Two experts at the National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore interviewed by The Hindu pointed out that internationally the psychological fraternity has used sodium pentothal for decades; and discontinued its use in all but the rarest cases, partly because there is no guarantee that the drug will elicit factually accurate information. Secondly, Dr. Mohan's contention that it is difficult to manifest fantasies in narcoanalysis is questionable. False memory is an extremely well-researched area according to Dr. Chittaranjan Andrade, a professor of psychopharmacology at NIMHANS. While patients under narcoanalysis may find it difficult to lie consciously depending on the depth of the narcoanalysis, they can say things that are not true and on the surface of their minds. Dr. Andrade explains the case of a suspect who is repeatedly accused of a crime during regular interrogation: "The same thing goes on during the narcoanalysis. He remembers `you've done this, you've done this.' He says, `I have done that.'"

When science has outpaced the development of law or at` least the layperson's understanding of it, there are unavoidable complexities regarding what can be admitted as evidence in court. In the United States, where science often interfaces uncomfortably with the law, the Supreme Court offered four criteria, part of the Daubert Standard (1993), by which to judge the credibility of a scientific principle held by a minority of practitioners: hypothesis testing; peer review and publication; knowledge of error rates; and acceptability in the general scientific community.

Pseudo-science

We must give narcoanalysis its due and grant that it has provided valuable leads to the police in some instances. However, one swallow, or even many swallows in this case, do not a summer make. It is logically consistent for even a pseudo-science to produce reliable outcomes in particular cases. The overall reliability and science behind the practice can only be determined after statistical analysis of a sufficiently large sample.

The irony of the situation we face in India is that the science behind narcoanalysis, as we know it, has not leapfrogged the courts by any stretch of imagination. The Bangalore research results and methods have been neither peer-reviewed nor published. Regarding publication of the data, Dr. Mohan says he will go public with the FSL data in three to four months (from March 2007) and is willing to debate its implications at international forums. But it is unlikely that studies based on some 300 criminal investigations will yield controlled experimental data. The feedback that goes into defining the success of the analysis is provided in part by police questionnaires. Here lurks a conflict of interest.

Legal aspects

There are other significant legal aspects to the narcoanalysis debate. In a 2006 judgment (Dinesh Dalmia v State), the Madras High Court held that subjecting an accused to narcoanalysis is not tantamount to testimony by compulsion. The court said about the accused: "he may be taken to the laboratory for such tests against his will, but the revelation during such tests is quite voluntary." There are two fallacies in this reasoning. First, if narcoanalysis is all that it is made out to be by the Bangalore FSL, the accused will involuntarily answer questions posed to him during the interview. The second fallacy is that it is incorrect to say that the accused is merely taken to the lab against his will. He is then injected with substances. The breaking of one's silence, at the time it is broken, is always technically `voluntary.' Similarly, it can be argued that after being subject to electric shocks, a subject `quite voluntarily' divulges information. But the act or threat of violence is where the element of coercion is housed. In narcoanalysis, the drug contained in the syringe is the element of compulsion. The rest is technically voluntary.

In 2004, the Bombay High Court ruled in the multi-crore-rupee fake stamp paper case that subjecting an accused to certain tests like narcoanalysis does not violate the fundamental right against self-incrimination. Article 20(3) of the Constitution guarantees this: "No person accused of any offence shall be compelled to be a witness against himself." Statements made under narcoanalysis are not admissible in evidence. However, recoveries resulting from such drugged interviews are admissible as corroborative evidence. This is, arguably, a roundabout way to subverting the right to silence — acquiring the information on where to find the weapon from the subject when, in his right senses, he would not turn witness against himself.

Arguments have been made that narcoanalysis constitutes mental torture. It works by inhibiting the nervous system and thus lowering the subject's inhibitions. It is not difficult to interpret this as a physical violation of an individual's mind-space.

The State police departments are responsible for generating demand for the process. The decision to conduct narcoanalysis is usually made by the Superintendent of Police or the Deputy Inspector General handling a case. A high-ranking official in the Karnataka Police told The Hindu that police departments in India have poor skills when it comes to collection, collation, and presentation of evidence before the courts. Consequently, when there is enormous pressure on a police department to solve a case, sending suspects to narcoanalysis not only buys time but also gives the impression that something concrete has been done about the case.

Some officials connected to law enforcement argue that narcoanalysis can be of great use in instances where witnesses turned hostile; rape cases where issues of consent are being debated; and cases where the investigating officer is hard pressed for time or working to disrupt offences planned for the near future, including terrorist acts.

Scope for abuse

This ticking-bomb terrorist case argument has also cropped up frequently in the media after the 9/11 attacks. It has been championed by Harvard Law School professor Alan Dershowitz, who argues for legitimising torture in select scenarios, for example when a hypothetical bomb is waiting to explode. There are many arguments against the selective use of normally banned cruel practices. Authorities are likely to abuse the power to decide which situations will warrant such exceptions, even when such extraordinary situations are explicitly laid out by law. It will be difficult to find a fool-proof way to determine which suspect is concealing information about a hypothetical bomb. It will often be impossible to know if there is a bomb ticking in the first place. These questions of discretion aside, when a country claims to be committed to human rights and against torture, one may ask if there can ever be a situation that warrants a deviation from its commitment to such principles.

While the expert studies and court opinions available internationally have granted that there may be some use in narcoanalysis, the overwhelming evidence is that narcoanalysis is by no means a reliable science. In the face of a near-consensus internationally, one or two Indian forensic labs claim to have new evidence and studies claiming remarkable success rates for the process. They must now prove their claim that narcoanalysis is backed by sound science. In the absence of proof, narcoanalysis must necessarily be suspended, especially given its ethical and human rights implications.

State governments need to work with the central authorities to enhance the investigative capabilities of their police departments. The police now hand over one of the most crucial parts of the investigation to a clinical psychologist conducting narcoanalysis. Interrogation is an art as well as a science. It takes enormous amounts of training and patience — skills evidently lacking in much of the police force and increasingly outsourced to Bangalore. The central government must make a clear policy stand on narcoanalysis — because what is at stake is India's commitment to individual freedoms and a clean criminal justice system.