Why Police in India Use 'Third-Degree' Torture Methods for Interrogation

An excerpt from Jinee Lokaneeta’s The Truth Machines: Policing, Violence, and Scientific Interrogations in India, republished with permission from University of Michigan Press, Orient Blackswan.

Truth in a legal sense may not be accessible through the deployment of truth machines [lie detectors, brain scans and narcoanalysis], but discourse about these techniques can reveal the dynamics of custodial interrogation. Because these methods were seen as scientific, legitimate, and professional, the police readily explained their logic of operation during my interviews. Three themes mark the pragmatic logic of applying the third degree in investigations: duration of custody, distrust of police, and recovery. Here, perhaps my most important departure from previous scholarship (that focus on ideological reasons and/or colonial continuity) is that these are reasons that are internal to police functioning, and they are linked to the original meaning of the third degree that of questioning. The pragmatic logic of third-degree interrogation is thus an articulation of the internal structural contingencies affecting police violence.

According to police accounts, one of the major reasons to turn to third-degree interrogation is the duration of custody. Under the CrPC, the Indian police are supposed to bring a suspect in front of a magistrate within twenty-four hours. Unless the suspect gets bail or an exceptional reason requires continuing police remand, the suspect is then sent to judicial custody. As a police officer based in Ahmedabad pointed out, this short time frame makes the duration of custody crucial for conducting an interrogation:

Now, in the Indian system what happens is, once you arrest someone… And within 24 hours of his arrest, maximum, he has to be produced before a court… So after his arrest the police is expected to complete the investigation, his interrogation within a period of 24 hours. And that 24-hours window is very odd because if you arrest someone this evening or in the night, then… the court wouldn’t be functioning till around afternoon by and large, and courts also dislike the accused being brought in late or at odd hours… So practically they would get 12, 13, 14, 15 hours with the accused… For investigation you get hardly 3, 4 hours with the accused before he’s produced. So these are the things which make it difficult for the police to go by the rule book. So as to those 3 hours, they would rather choose to resort to third degree or abuse and try and get the things out of him under the presumption that next day once he’s produced, they are not going to obtain his custody, and once he goes to the judicial custody then he won’t speak anything.

Following rules thus poses difficulties that mediate the process of interrogation. 

The NPC [National Police Commission] report similarly attributed shortcut methods of interrogation to the brief period allowed in police custody, as police sought quick confessions or requested further remand to recover stolen materials. As a senior Delhi-based official explained:

See, torture is taking place essentially for two things in police. One is to extract information; another is to extract money; people also are afraid that they might be subjected to that kind of routine, so they’ve been parting money… At least the information part will go immediately [if torture disappears]. And once this need for torture will go, need means perceived need…I honestly feel torture is useless.

In the absence of a change to the provision regarding duration of custody, requests for narcoanalysis or brain scans can become another delaying tactic to extend the period of custody. That may be a principal reason for the increasing number of requests for these techniques, although in practice, very few labs are equipped to conduct them and they now require the consent of the accused. The pressure to deliver results quickly, within the duration of custody, is thus mentioned as a common reason for turning to the third degree.

The second major reason for using third-degree interrogation that was cited by police during my interviews was distrust of their profession. As a police officer in Hyderabad pointed out, whereas CrPC allows for “the constable… to investigate and arrest even the president of this country,” the system is based on “distrust,” because a statement “before a police officer has no evidentiary value, as it reflects a system designed for perpetuation of colonial power.” Indeed, one of the police officials I interviewed made the United States a constant point of comparison and suggested that “whereas in a Western country, the accused makes a statement, he signs it. It is verified. If it is found to be untrue, you go back to him and say, ‘Look, man, you are telling a lie.’ And if it is true, well, it is used.”

These accounts of police officers might be understood as justifications for torture or third-degree interrogation. A caution, however, appears in Upendra Baxi’s famous essay on torture, where he poses a need to address a lack of trust in the police in recording confessions:

“Paradoxically, this very attempt to protect the dignity of the accused tends to create a situation of loss of dignity for the police profession, real or apperceived.”111 Some police officials do indeed connect distrust directly with accountability. For example, the draft anti torture bill, as revised in 2010, appeared to attribute responsibility for torture not just to those who commit it but also to those in charge.

One official I interviewed stated that the bill was too harsh because in India, the version of the police is not believed in courts of law; …whatever admissions are made before the police are not admissible in the courts of law. If that is the situation, if you bring the antitorture bill also, it is very difficult for the police to function. And in that case our only humble request would be to make police versions acceptable in the courts.

When asked how making police testimony admissible would ensure accountability, this official responded, “Well, antitorture bill brings in that accountability. I’m accountable if I do… You do some mischief in investigation or something, use third degree or do some hanky-panky in investigation, I’m accountable for it.”

The theme of distrust was prominent among my respondents. A Delhi-based police officer predicated trust and accountability on independence:

See, if you trust the police, then you have to, at the same time you’ll have to take the necessary steps which go along with the trust. You cannot trust somebody who’s dependent on others. If you trust the police the way they are now, only under the thumb of politicians, and if you trust them, there will be havoc in society. So if you really want to trust them, you have to make them independent… But along with the trust, they should be given the independence so that they’re independently accountable for whatever they are doing. Then they cannot say, “I got an order from this quarter,” or that thing, or this and that.

Trust, accountability, and independence from political control are linked in this narrative, which emphasizes the risk of retaliation and political pressure, even for high-ranking officers. These police officers clearly convey their experience with both lack of trust and incentives to use third-degree interrogation, despite recognizing that anything suspects say in their interaction is inadmissible in court. Their accounts thus present a contradiction: if they perceive (or know) that they are not trusted and cannot use the statements they gain in custody, why use torture?

A third theme that emerged from my interviews—recovery—addresses this question. Torture in India is used in a range of cases, from terrorism and murder to theft, even though confessions and evidence acquired under torture are inadmissible in court and policy prevents police from legally recording confessions (a remnant of colonial practice that has often been a sore point for the Indian police). Yet Section 27 of the Indian Evidence Act allows material discovered as a result of a confession, along with the part of the confession that led to its discovery, to be admissible. A senior police official based in Mumbai expressed a sentiment echoed by many other police officials:

Now, here a case has come to you of theft, all right. Now I cannot say that “My lord, here is the thief. He has admitted before me.” The court is not going to accept that, all right. What the court will accept is, if that man says that, “Look here, I will show you where I have disposed or hidden that property.” . . . The law itself is saying that, “Look here, I’m not going to believe you unless you produce this” and that “produce this” is not going to come voluntarily. The law also knows. I also know. The judge also knows that no recovery, no discovery of fact as is known in the legal parlance, can take place voluntarily.

One primary incentive for torture in extremely routine cases, therefore, appears to be recovery of evidence. As another police official said, “Courts don’t insist on recovery, but recovery does ensure culpability.”

In my interviews, theft and property offenses are the main categories of crime to which third-degree interrogation is attributed in routine cases. As one officer noted, “Unless the accused tells us, we cannot go ahead with it.” This officer did claim that other methods, such as call detail analysis, help provide evidence, but, ultimately, recovered property is most often used to confront the accused. As the Hyderabad-based police officer explained, when the conviction rate in criminal cases is so low, success depends on recovery rates. A Bangalore-based police official elaborated: “If some person comes and gives a complaint of theft, . . . a lot of tricks are there by the police to find out. Nearly we succeed, only 80–90% we succeed in [getting] the truth of the accused, and we will get the properties.”

Ensuring recovery thus becomes a priority. Indeed, states’ rates of property recovery are mentioned separately by the National Crime Research Bureau, and even some stories in the media point directly to states boasting about their lead in recovering property. At least two police officials speculated that this trend would stop only if the Indian population, especially the rich and powerful who exert pressure, could insure their property so that they were not dependent on police recovery. My interviews, however, also revealed a lack of police training in interrogation, so much so that the National Police Academy lacked an interrogation module until the 1990s. Police accounts thus reveal the pragmatic logic of third-degree interrogation, which has otherwise been linked to a normalization of torture and receives close scrutiny only in cases of custodial death or “severe” torture.

Jinee Lokaneeta teaches political science at Drew University, New Jersey. She is the author of Transnational Torture: Law, Violence and State Power in the United States and India and has most recently published The Truth Machines: Policing, Violence, and Scientific Interrogations in India.