– Vasu Aggarwal*
In India, the confession of a co-accused has been provided in Section 30 of the Indian Evidence Act (‘IEA’). Section 30 provides that courts may consider the confession of a co-accused against the accused if the confession affects both of them when they are tried jointly. However, such confession may only be used as an additional support for the evidence against the accused, if it is sufficient to secure a conviction [Emperor v. Lalit Mohan Chuckerbutty ¶23]. More importantly, in cases where the judge is not willing to act upon the available evidence even though it is believed that the evidence is sufficient to secure conviction, the judge may use the confession of the co-accused to lend themself assurance and act upon the evidence [Kashmira Singh v. State of Madhya Pradesh ¶10].
It may appear that the probative value of a confession of the co-accused is low, and entails limited importance. However, even this low probative value has prompted judges to grant convictions in numerous cases. Since the confession of a co-accused can potentially make a difference between acquittal and conviction, there must be some guarantee for it to be true. However, there are no such safeguards under Indian law (or under UK and US law). Thus, this paper argues that the confession of a co-accused which implicates the accused must not be accorded any value as against the accused due to the absence of sufficient safeguards. In the first part, the article discusses the dismal status of confessions, in general, in India and how it induces confessions against the co-accused. The second part discusses how the absence of safeguards such as oath and cross-examination renders the confession against the co-accused highly unreliable. It also discusses how the supposed safeguard of self-incrimination is grossly insufficient. Finally, the article draws a comparison with the UK and the US to conclude that there are no such safeguards that may be borrowed to make the confession of the co-accused reliable.
General Status of Confessions in India
In India, a confession per se is not so voluntarily taken as it may appear. The safeguards related to confessions, such as the requirement of stay in judicial custody, explanation of consequences of the confession, and time to reflect on the consequences, are not religiously followed. Section 463 of the Code of Criminal Procedure itself gives leeway for the courts to consider confessions when the procedures are not complied with. Examples of safeguard relaxations/violations include non-compliance with the requirement of a stay in judicial custody before the recording of confession [Babu Singh v. State of Punjab 1962 Indlaw SC 499 ¶14-16], and not recording all answers [Ram Singh v Sonia2007 Indlaw SC 122 ¶31-32]. These are often not complied with by stating that these are merely ‘procedural’ compliances.
Even if it is assumed that the safeguards are followed religiously, the safeguards are grossly insufficient. Torture and coercion are commonly used as a tool to extort confessions. People are picked up by the police, treated with third degree methods, and made to confess to the crimes that they did not commit. This is particularly seen in cases involving those belonging to socio-economically backward classes. Other times, investigating agencies use Section 27 of the IEA to make statements of those accused admissible by planting evidence and then using it against the accused in the courts. Although the magistrate is expected to ask the accused whether the confession being rendered is voluntary, the accused may have been threatened by investigating agencies beforehand. Therefore, even though the confession may be recorded by following procedure, there is no guarantee that the confessional statement will be ‘voluntary’ or true.
Forced Confessions: Inducement of Confession Against Co-accused
It may be contended that ‘self-incrimination’ per se deters insincere implication of the co-accused. This is because an inference may be drawn against the co-accused for other major offences, and the confession may be taken as a substantial piece of evidence for the minor offences confessed. Thus, the co-accused will not be incentivized to jeopardize their own defence. However, as pointed out, the confession per se suffers from many shortcomings.
In cases where the accused are subject to torture and coercion, they might not think that they have any choice. Even when the accused are not subjected to torture, a mere image of torture may be enough to produce the effect. In the classic situation of prisoner’s dilemma, when two accused are given a choice between confessing (implicating both the accused themselves and the co-accused), and getting lower punishment, and on other hand, not confessing and getting a higher punishment, the accused is more likely to act in their own self-interest, and thus, confess to the crimes in anticipation of a lower punishment. When there is involvement of torture and coercion, it is considerably more likely that the co-accused may be incentivized to implicate themselves of a minor offence and implicate the accused of a major offence.
Absence of Oath, Cross-examination and Demeanor
The probative value of the confession of a co-accused is low because of the absence of safe-guards such as oath and cross-examination. Confessions are not made under oath. Oaths act as a safeguard for truth telling because: first, the act of swearing to their God’s name holds the witness (or the co-accused, here) to their religious faith. Although it may be in the name of God, the real engagement is with the moral consciousness. It lends a sense of credence in the eyes of others, and self-satisfaction in oneself. Second, a false statement made under oath is an offence under Section 191 of the Indian Penal Code (‘IPC’), and punishable up to seven years with imprisonment and fine. It creates a deterrence effect against the making of a false statement under oath.
The confession of a co-accused is not made in front of the accused, nor is it subject to any cross-examination. Cross-examination is a powerful means of testing the veracity and accuracy of a statement. Cross-examination is important to uncover biases, and personal motives that may lie beneath the statements. Particularly, in the case of the co-accused, this assumes more importance as the co-accused has an incentive to implicate the accused for the greater part in the crime in order to absolve themself. Even for the testimony which is protected by safeguards such cross-examination, the courts have recognized this shortcoming in evidence provided by the co-accused, by laying down an additional test for the testimony of the accomplice.
Additionally, the demeanour of the declarant may be taken into account by the court in determining the truth of the statements. A close observation of the one telling the story may instinctively tell us about the truth of the statement. However, in the case of confessions by the co-accused, the Magistrates are not specifically trained to evaluate the demeanour of the co-accused, while being cognizant that the confession may be used against the accused. Therefore, the utility of this safeguard cannot be effectively realized.
Prisoner’s Dilemma: An Incentive to Lie
Although cross-examination, demeanour, and oath are absent, an alternate ‘safeguard’ is provided under Section 30 of the IEA—that the confession of the co-accused must implicate themself too (self-implication). However, this supposed ‘safeguard’ is grossly insufficient.
The confession of the co-accused may have two parts to it—relation of the offence with both the accused, and the nature of the offence. The former means the extent of involvement in the crime. Multiple commentators, including Sarathi and Sarkar, have argued that the confession of the co-accused must implicate them to the fullest extent. The SC has also held that the confession of the co-accused must implicate them substantially to the same extent as the accused.
However, the co-accused may be tempted to differentiate between the nature of the offence committed by the accused as against the co-accused. The co-accused may implicate themself of abetment or attempt, or a minor offence of a major offence, while implicating the accused of commission, or the major offence.A two-judge bench of the Madras HC, in In re Periaswami Moopan ¶5, held that the confession of a co-accused, who confesses for any minor offence or a minor offence within the major offence for which the accused is tried, cannot fall within the ambit of Section 30 of the IEA. In holding so, the court observed that such a confession would not be genuine, and the legislature could not have intended to keep the provision so wide to include non-genuine confessions. However, the court in In re Periaswami Moopan did not consider previous authorities. A three-judge bench of the Madras HC, In re Manicka Padayachi, held that even if the co-accused is charged of a minor offence, their confession may implicate the accused of a greater offence under Section 30 of the IEA. It held that as long as the minor and the major offence form a part of the same transaction, the confession against the co-accused under Section 30 of the IEA may be considered. Since Manicka Padayachi is a three-judge bench as against Periaswami Moopan,which is a two-judge bench, Manicka Padayachi must be considered the correct position of law.
Mirza Zahid Beg v. Emperor ¶9 held that there is nothing in Section 30 that suggests that the confession must affect both the accused equally. Even if the confession minimises the guilt of the co-accused, and maximises the guilt of the accused, Section 30 does not bar the court from considering it against the accused. However, Mirza Zahid Beg does not address the part of the provision regarding the accused being tried jointly for ‘the same offence’. Thus, the textual meaning of the provision must be probed deeper.
Dichotomy of Section 30 of the IEA
The first requirement of the provision is that the co-accused (Accused-1) and the accused (Accused-2) are to be tried for the same offence. This requirement may be fulfilled by the co-accused implicating themselves of a minor offence, or abetment/attempt (O1 in the table). The second requirement is that it must affect both the accused and the co-accused. This requirement is also fulfilled by implicating oneself of a minor offence, or abetment/attempt. When both these requirements are fulfilled, the entire confession, not just the part related to ‘the same offence’, of the co-accused may be taken into consideration against the accused. Thus, the part of the confession of the co-accused relating to a major offence (O2) will also be considered against the accused although the co-accused did not implicate themselves. Therefore, the effect of the provision is that once the two requirements are fulfilled with respect to a minor offence (O1), it opens the confession of the co-accused with respect to any major offence (O2) to be considered against the accused. Therefore, this provides the co-accused the incentive to implicate themselves of a minor offence, or abetment/attempt, and the accused of a greater offence to absolve themselves of the guilt.
Offences (General)Example from Manicka Padayachi, Co-accused who gives confession Accused 2 (Accused against whom the confession is considered)O1 (Minor offence) or abetment/attemptHurtImplicationImplicationO2 (Major offence)MurderNo ImplicationImplication
Manicka Padayachi derives its effect from a similar understanding of the provision, where the co-accused implicated himself of ‘Hurt’, while implicating the accused of ‘Hurt’ and ‘Murder’. The confession of the co-accused with respect to ‘Murder’ was also considered by the Court under Section 30 of IEA.
Although abetment/attempt may open the door to the major offence as enunciated above, a closer look at the Explanation to Section 30 offers supportive reasoning. The explanation to the Section provides that ‘Offence’ in this provision includes abetment and attempt to commit the offence. Therefore, ‘Offence’ could be read as abetment or attempt or the commission itself [Please refer to the Venn Diagram]. Now, for the first requirement to be completed, the accused and the co-accused must be charged of the same offence. ‘Offence’ includes within itself both abetment and attempt. Therefore, the co-accused may implicate themselves of attempt, thereby, exposing themselves to lesser punishment; while implicating the accused of the commission of the offence, and thereby, exposing the accused to greater punishment.
Thus, not only are safeguards such as oath and cross-examination absent, but the provision provides for an incentive to implicate the other accused of a major crime, while limiting the self-implication to minor offences. This incentive may lead the co-accused to actively make false claims in order to exonerate themselves.
Even if it is assumed that self-implication is for only the same offence and to the same extent, it does not negative the possibility of the confession being false to the extent that it refers to other accused. Such false implication may arise from hatred, malice, or any other reason. This bias cannot be exposed in the court due to absence of cross-examination. In other words, it does not exclude the risks that arise out of absence of oath, cross-examination, and demeanour, but only affects those risks that arise due to the special nature of it being made by the co-accused, such as an incentive to implicate the accused of a greater part of the offence. Therefore, the supposed ‘safeguard’ of self-incrimination is not a replacement for oath, cross-examination, or demeanour.
Comparative Analysis: the UK and the US
Now that the shortcomings of the Indian position are clear, it may be useful to understand the position of law in the US and UK where the question of confessions of the co-accused has been considered in great detail. The purpose of understanding the position of law in other jurisdictions is to observe whether there are any safeguards that these jurisdictions have employed that may be borrowed to make the confession reliable.
In the UK, evidence law has been codified in the Police and Criminal Evidence Act 1984 (‘PACE’) and Criminal Justice Act 2003 (‘CJA’). Section 76A of PACE, also inserted in Section 128 CJA, makes the confession of the co-accused relevant for the accused and not against the accused. The Law Commission Report ¶8.96 that recommended the enactment of Section 76A clearly drew a distinction between providing evidence for and against the co-accused. It stated that while a confession of the co-accused may be used to exonerate the accused, it cannot be used to convict the accused who did not make the confession. This position was also affirmed in R v. Sliogeris ¶20-26, and Regina v. Hayter ¶49-50,that the confession of the co-accused cannot be used against another accused as it amounts to hearsay evidence being admissible without any safeguards. To that effect, the judge is expected to direct the jury to disregard the evidence against the non-declarant. Therefore, in the UK, there are no safeguards envisaged that could help improve the reliability of the confession of the co-accused.
In the United States, Bruton v. United States P.43 lays down that all portions of the confession of the co-accused which implicate any accused person other than the confessor are to be effectively deleted before producing before the jury. The reason for disregarding the confession against a non-declarant remains the same—that it is hearsay evidence. The reason for total redaction has been stated that post-facto instructions do not effectively negative the impressions generated in the minds of the jury. In the US too, there are no safeguards that are employed to improve the reliability of the confession of the co-accused.
Therefore, the position of law in the US and the UK is starkly different from that of India. These jurisdictions have also observed that the absence of safeguards such as cross-examination and oath render the confession of the co-accused highly unreliable, and that they must not be accorded any value. Thus, these jurisdictions do not suggest any such safeguards that can be imported into India to make the confessions against the co-accused reliable.
This paper has delved into the question of sufficiency of safeguards in ascribing probative value to the confession of the co-accused against the accused. The co-accused is not under oath, nor subject to any cross-examination. However, there is an alternate remedy—self-implication, which is built in the scheme of Section 30. This paper argues that this supposed ‘safeguard’ is not only incapable of replacing oath and cross-examination, but also incapable of impeding the incentive that the co-accused may have—implicating the accused in major offences, while limiting their own implication to minor ones.
It may be argued that self-incrimination in itself has the potential of creating a deterrent effect to insincere implication as even an implication for minor offence may be taken as a substantive evidence on which their conviction may be secured. However, in India, accused persons are regularly subjected to torture and coercion till they confess, and the safeguards are grossly insufficient to prevent such torture. It is highly likely that the confession rendered then would implicate the co-accused to a greater extent, thus, bringing back the shadow of insincerity. Law enforcement agencies may also want to use it as a tool to get confessions against the co-accused. Therefore, the confession of the co-accused must not be accorded any value as against the accused.
* Vasu Aggarwal is a third year student of law at the National Law School of India University, Bangalore. He would like to thank Professor Kunal Ambasta and the editors of NLSIR for their invaluable suggestions on the previous drafts of this article.
 Multiple reports substantiate the claim that the police and other investigating agencies operate with methods of third degree to extort confessions of those innocent. UNCAT, ‘India: Annual Report on Torture 2019’ http://www.uncat.org/wp-content/uploads/2020/06/INDIATORTURE2019.pdf accessed 1 August 2020 page 21-24. For specific cases recorded by the report, See page-25, 36, 61, 77, 86, 88-89 etc;
‘State of Policing in India Report: Policy Adequacy and Working Condition’ (Common Cause, 2019) https://www.commoncause.in/uploadimage/page/Status_of_Policing_in_India_Report_2019_by_Common_Cause_and_CSDS.pdf accessed 1 August 2020, 141. Reported, “Four out of five personnel believe that there is nothing wrong in the police beating up criminals to extract confessions”.
 Page 88-89 of this report mentions Yusuf’s case, an unreported case, where Yusuf was a poor school dropout, who was convicted solely on the confession of the co-accused.
 Jeremy Bentham, Rationale Of Judicial Evidence, Specially Applied To English Practice (Hunt And Clarke, London 1827) 369.
 Jeremy Bentham, Rationale Of Judicial Evidence, Specially Applied To English Practice (Hunt And Clarke, London 1827) 365.
 John H. Wigmore A Treatise on the Anglo-American System of Evidence in Trials at Common Law Volume 2 (Little, Brown, 1940) 143.
 John H. Wigmore A Treatise on the Anglo-American System of Evidence in Trials at Common Law Volume 2 (Little, Brown, 1940).
 Vepa P. Sarathi Law of Evidence (Eastern Book Company 2010) 276.
 Vepa P. Sarathi Law of Evidence (Eastern Book Company 2010) 29.
 Vepa P. Sarathi Law of Evidence (Eastern Book Company 2010) 29.
 Vepa P. Sarathi Law of Evidence (Eastern Book Company 2010) 164.
 Sarkar Law of Evidence (18th edn, Lexis Nexis) part I, ch 2, page 7.
 The co-accused is not expected to benefit themself by implicating to abetment because the general provision regarding abetment—Section 109 of the IPC—provides for the same punishment as commission of the offence.
 This provision provides that the punishment for attempt to commit a crime may be one-half of the longest term of imprisonment for that offence
 This position was reaffirmed in United States v. DeLeon 2020 WL 353856. Recent cases in which the Bruton’s Rule was directly applied are United States v. Burden 964 F.3d 339 [2 July 2020], Maple v. Clark 2020 WL 3566714 [30 June 2020]. Confrontation Clause [Sixth-Amendment] provides for the right of cross-examination, similar to S.138 of IEA. A.D. Almedia v. State 1994 SCC OnLine Mad 505  held that without the opportunity to cross-examine the witness, the evidence provided by the witness becomes inadmissible.