top of page

Section 27 of the Indian Evidence Act

Updated: Jan 19


There have been rising concerns about custodial violence in India, especially since the 2018-19 NHRC report stated there were almost 2000 cases of custodial death in India during the year. (NHRC Report, 2018-19) At times, the hope of extracting confessions is what tempts police officials to resort to unfair means. The Evidence Act provides a scheme for safeguarding persons against forced confessions by making them inadmissible. However, Section 27 of the Indian Evidence Act, 1872 (“IEA”) overrides the exclusionary rules with regard to confessions made by a person in custody and thereby puts a powerful tool in the hands of the police. (Law Commission, 152nd Report)

The scheme of the IEA with regard to confessions

As per Section 24, only a ‘voluntary’ confession is admissible. In order to ensure that statements admitted into evidence are voluntary, the law goes on to provide for exclusionary provisions. For instance, Section 25 provides that no confession made to a police officer shall be proved as against a person accused of any offence and cannot be used as substantive evidence to convict the accused. Similarly, Section 26 lays down that no confession made by a person in custody of a police officer shall be proved against said person unless it is made in the immediate presence of a Magistrate.

However, Section 27 states that when a fact is received as “information” from an accused person in the custody of the policy, then this information, as far as it relates to the facts discovered, is relevant irrespective of whether it amounts to a confession. This leads to an almost overriding effect on Section 26, however, that is still up to debate. (Law Commission, 152nd Report) As per judicial decisions, such as the case of Pulukuri Kottaya vs King-Emperor, when it comes to Section 27, it does not nullify the substance of Section 26. The court stated that the ban as accorded by Sections 25 and 26 was “presumably inspired by the fear of the Legislature that a person under police influence might be induced to confess under pressure.” It was further stated that “The condition necessary to bring Section 27 into operation is that the discovery of a fact in consequence of information received by a person in police custody of a police officer must be deposed to, and thereupon so much of the information as distinctly relates to the fact thereby discovered may be proved.”

The draftsman of the Evidence Act, James Fitzjames Stephen stated that the background of Sections 25 and 26 was the inefficiency and third-degree methods used by the Indian Police. (Sen, 1994) Additionally, in the case Queen Empress v Babu Lal and Anr, an observation was made on the scheme of confessions under the IEA that “These legislative procedures leave no doubt in my mind that the Legislature had in view the malpractices of police officers in extorting confessions from accused persons in order to gain credit by securing convictions, and that those malpractices went to the length of positive torture …”.

Custodial Violence

The matter of fact is, even if Section 27 doesn’t have an absolute overriding effect on Section 26, there are still consequences of the same. As per the Death Penalty Report published by Project 39A, when 92 prisoners who had confessed in police custody were interviewed, 72 admitted to being subjected to torture for the same. Despite the fact that the methods used for extracting these confessions make them clearly tainted, considering a person undergoing such torture would agree to anything in order to escape the pain, they’re still used to secure convictions. (Project 39A, 2016)

In a report by Amnesty International, it was noted that there was a discernible pattern in police killings where the victims were picked up at random, illegally confined and tortured to extract confessions till they died. They noted the account of Raju Mohite’s death, a Dalit man who was beaten until he confessed to an offence. (Amnesty, 1992)

The NHRC reports in the past have found troubling cases of custodial deaths. In one instance, an under-trial prisoner Tukmesh placed in the District Jail, Badaun, Uttar Pradesh, was found dead in custody. The authorities suspected he had died from consumption of a poisonous substance. However, the post-mortem report revealed bleeding from the nose, fresh injuries and no trace of a poisonous substance. The case was taken up by the NHRC in 2016 and it was only in 2020 that the case was disposed off. However, the only action taken was compensatory relief being granted to the family and it was held that the treatment provided to the Under-trial prisoner was inadequate. (NHRC, 2016) There was no disciplinary action taken against persons that may have been involved in the death.

In the case of State of Uttar Pradesh v Ram Sagar Yadav, an accused person was beaten to death by the police for not seceding to their demands. The Supreme Court, setting aside the order of acquittal and restoring the conviction by the Sessions Judge u/S 304 of the Indian Penal Code (“IPC”), said that the accused should have been convicted under Section 302 instead.

From 1985 to 1994, it was noted that only 3 out of 415 cases of custodial deaths where police officers had been accused of torturing people to death have seen convictions in courts. (Sen, 1994) In the present day, the disparity would be much higher. Suspension of these police officials is often used as a delaying tactic to curb public outcry. The families of the victims, as in the case of Tukmesh, receive monetary compensation instead of having a chance to see convictions of accused police officials.


There’s a common pattern of excuses that are given when police officials are questioned about custodial violence. For instance, the police force has inadequate strength as compared to the rising crime rates; hardened prisoners only understand the language of violence; there are virtually no scientific investigation facilities in police stations, and so on. (Saini, 1994) Most low-ranking police officials come from socio-economic backgrounds where they haven’t received adequate education or sensitisation in terms of the application of Human rights.

The police have been engaging in practices of custodial torture as a matter of routine. At the time, the police force of the country is in need of education and sensitisation. The long-standing training practices of these academies have been largely drill-based. However, it is apparent that there exists a need for reform. (Kaura, 2019)

In a study by the SVP National Police Academy, it was a point of majority agreement in the sample body including high ranking police officers and academicians that “Police do not have adequate skills of interrogation and utilise third-degree methods on persons taken in custody.” (Sen, 1994) It was further recommended that there must be training interventions for all police functionaries, focusing on sensitivity towards human rights, skills in interrogation and scientific investigation, and teaching upon provisions of the law. There were additional recommendations about how based on recommendations of the Supreme Court, a provision should be added to the IEA which introduces a rebuttable presumption that injuries sustained by a person in police custody may be presumed to have been caused by the police officer-in-charge. (Sen, 1994)


In conclusion, the authorities may have a long way to go in terms of education and sensitisation of police officials to ensure the protection of human rights. Setting up an elaborate infrastructure with effective educators will require time and resources. However, in the meantime, we must ensure accountability of accused police officials in cases of custodial violence and deaths. Due to a lack of sufficient evidence, these crimes tend to go unpunished. Therefore, there needs to be greater judicial and legislative intervention to curb custodial violence.


1. National Human Rights Commission, Annual Report, 2018-19 <> accessed 27 April 2022.

2. Law Commission of India, The 152nd Report on Custodial Crimes (152, 1994).

3. Pulkuri Kottaya and ors vs King-Emperor [1946] UKPC 52.

4. Sankar Sen et al, Custodial Deaths in India: A Research Study. (SVP National Police Academy, 1994)

5. Queen Empress vs Babu Lal (1899) ILR All 106.

6. National Law University, Delhi, Death Penalty Report (NLU Delhi Press 2016).

7. Amnesty International Report, India: Torture, Rapes and Deaths in Custody (1992).

8. NHRC Case No. 14047/24/7/2016-JCD.

9. State of Uttar Pradesh v Ram Sagar Yadav AIR 1985 SC 416.

10. RS Saini, Custodial Torture in Law and Practice with Reference to India, 36 JILI (1994) 166.

11. Vinay Kaura, 'Indian Police Academies Need An Upgrade, Must Go Beyond Old Drill-Based Training' (2019) <> accessed 26 April 2022.

31 views0 comments

Recent Posts

See All

The term Euthanasia has been derived from two Ancient Greek words: ‘eu’ meaning ‘good’, and ‘thanatos’ meaning ‘death’, which combined means “well-death” or “dying well”. It is also known as ‘mercy ki

Post: Blog2_Post
bottom of page