Login
  • Bookmark
  • PDF
  • Share
  • CaseIQ

Rattan Dhanuk v. King-Emperor

Patna High Court
Jul 17, 1928
Smart Summary (Beta)

Judgment Summary — Jail Appeal Concerning Approver's Evidence

Factual and Procedural Background

Eleven accused persons were tried before the Additional Sessions Judge at Bhagalpur with assessors under section 396 of the Indian Penal Code for a dacoity alleged to have been committed at the house of Bachi Mandal on the evening of 28 January 1927. The principal testimony implicating the accused came from an approver, Lachmi Bantar, who confessed and gave evidence in court. Five of the accused had the approver's evidence corroborated; six did not. The eleven accused were unrepresented on appeal. The appeal raises the question whether the Sessions Judge was right to convict the six accused on the uncorroborated evidence of the approver.

Legal Issues Presented

  1. Whether the uncorroborated testimony of an approver (accomplice) can, as a matter of law, support a conviction under the Indian Evidence Act (in particular having regard to sections 133 and 134 and the illustrations to section 114).
  2. Whether the Sessions Judge was correct in treating the approver's testimony as truthful (so that corroboration was not required) in respect of six of the accused who lacked independent corroboration.
  3. What tests or safeguards a tribunal should apply when deciding whether to accept uncorroborated accomplice testimony (the role of intrinsic and extrinsic tests and the proper exercise of judicial discretion).

Arguments of the Parties

The opinion does not contain a detailed account of the parties' legal arguments.

Table of Precedents Cited

Precedent Rule or Principle Cited For Application by the Court
An Attorney, In re. (opinion of Jenkins, C.J.) Judicial limits: a court cannot alter or add to plain statutory provisions; discretion conferred by statute must remain free and not be crystallised by decisions. Relied upon (by Macpherson, J.) to support the proposition that the statutory discretion regarding presumptions and maxims (Illustration (b) to section 114) must not be fettered by rigid judicial rules requiring invariable corroboration of accomplice evidence.
Gardner v. Jay When a tribunal is vested with a discretion without statutory guidance, superior courts should not lay down fixed rules to restrict that discretion. Applied (by Macpherson, J.) to reinforce that courts should not crystallise a rule requiring corroboration of accomplice testimony but should exercise discretion based on facts of each case.

Court's Reasoning and Analysis

Two members of the court (Courtney Terrell, C.J. and Macpherson, J.) considered whether convictions based on the uncorroborated evidence of an approver were sustainable. Their reasoning can be summarised in the following structured steps, taking only the content expressed in the opinion:

  1. Legal framework and initial proposition (Courtney Terrell, C.J.):
    • The Evidence Act (sections 133 and 134) codifies that an accomplice is a competent witness and that a conviction is not illegal merely because it proceeds upon uncorroborated accomplice testimony.
    • From English and Indian authorities, Terrell distilled principles: (a) approver's evidence is like any other witness except that (b) it is regarded ab initio as open to grave suspicion; (c) if that suspicion is not removed corroboration is required; (d) if suspicion is removed, uncorroborated evidence may be acted upon.
  2. Practical rule and caution:
    • There is a well-established practice (almost a rule of law) requiring explicit warning to a jury about danger in accepting uncorroborated accomplice evidence; omission of such warning typically leads to reversal when conviction rests on such evidence.
    • Corroboration must relate to the guilt of each individual accused; corroboration as to some accused does not automatically corroborate allegations against others.
  3. Tests to determine whether suspicion is removed:
    • The court should apply both intrinsic and extrinsic tests to evaluate the approver's testimony.
    • Intrinsic tests focus on the internal consistency, detail and probability of the narrative; extrinsic tests look to corroborative facts and circumstances that confirm truthfulness.
    • If these tests show the approver to be truthful, the approver becomes an ordinary witness and section 134 permits conviction on his evidence alone.
  4. Policy and practical considerations:
    • To impose an absolute rule that accomplice evidence must always be corroborated would conflict with section 133 and would create impractical standards of proof, particularly harmful in offences like dacoity where detection is difficult.
    • The rules of evidence are tools of practical convenience and must be applied with an eye to practical safety and experience, not as immutable transcendental rules.
  5. Application to the facts and assessment of credibility (Court's factual findings):
    • The Sessions Judge examined intrinsic elements: the approver's account had inherent probability, consistent detail, accurate description of the site and rooms that would be hard to invent without seeing them.
    • Where approver's account conflicted with other witnesses, the judge gave reasons for preferring the approver's account.
    • There was no evidence of enmity between the approver and the accused; cross-examination did not shake his story.
    • Corroboration existed for five accused (though not for the six in question), and that fact could be taken as an indicator of the approver's general truthfulness (but not as direct corroboration for the other six).
    • Police had earlier held a different theory of events; the approver's account did not mirror a police-tutored script, which the court treated as an indicium against tutoring.
    • The defence called no evidence to contradict the approver's story, and the Sessions Judge formed a favourable view based on demeanour from seeing the approver give evidence in court.
    • On these intrinsic and extrinsic considerations the Sessions Judge concluded the approver was a witness of truth regarding all he stated.
  6. Additional analysis on statutory discretion (Macpherson, J.):
    • Sections 4, 114, 133 and 134 of the Evidence Act do not support a rule that uncorroborated accomplice testimony is necessarily insufficient; Illustration (b) to section 114 is a maxim and discretionary, not mandatory.
    • English doctrine that an accomplice is inherently unworthy of credit unless corroborated is not fully reproduced in the Evidence Act; the statutory framework contemplates judicial discretion whether to apply the maxim.
    • It is impermissible to crystallise judicial dicta into a binding rule that fetters the discretion the statute confers.
    • In practice courts may demand strong rebutting evidence before accepting uncorroborated testimony, but sometimes—on the facts and after applying intrinsic and extrinsic tests—declining to apply the maxim is justified.
  7. Conclusion of analysis:
    • Both judges agreed that, on the facts of this case, the Sessions Judge rightly concluded the approver was a truthful witness as to all he stated, including as to the six appellants lacking independent corroboration.
    • Accordingly, conviction on the approver's uncorroborated testimony was upheld as lawful in these circumstances.

Holding and Implications

APPEAL DISMISSED.

Direct consequences: The convictions and sentences imposed by the Additional Sessions Judge were sustained and the appeal was dismissed.

Broader implications (as expressed in the opinion): The court clarified and reaffirmed that under the Indian Evidence Act an accomplice's testimony is competent and that courts possess a statutory discretion (illustrations to section 114 and sections 133–134) to decide whether to apply the maxim requiring corroboration. If a tribunal, after applying intrinsic and extrinsic tests, is satisfied that the approver's testimony is truthful, it may convict on that testimony alone. The opinion emphasises caution and the exceptional nature of accepting uncorroborated accomplice evidence but does not lay down an absolute rule that corroboration is always required.

Note: This summary is prepared strictly from the content of the provided opinion and does not add facts or materials not contained in that text.

Show all summary ...

Courtney Terrell, C.J:— This is a jail appeal and the eleven accused persons are not represented. We have, however, carefully scrutinised the judgment and we find that there is a point of law to be considered on behalf of six of the accused.

The eleven persons were tried before the Additional Sessions Judge at Bhagalpur and assessors under section 396 of the Indian Penal Code. The dacoity in which they are said to have taken part was committed at the house of one Bachi Mandal on the evening of the 28th January, 1927. The principal evidence against the accused persons and the only evidence as regards six of them is the testimony of an approver one Lachmi Bantar who made a lengthy confession and gave evidence before the Court. As to five of the accused persons the evidence of the approver has been amply corroborated and the question arises whether the Sessions Judge rightly convicted the remaining six on the uncorroborated evidence of the approver. Section 134 of the evidence Act lays down that no particular number of witnesses shall in any case be required for the proof of any fact and section 133 provides that “An accomplice shall be a competent witness against an accused person and a conviction is not illegal merely because it proceeds upon the uncorroborated testimony of an accomplice.” With the aid of the learned Government Advocate we have considered the effect of these sections which merely state in codified form the English common law rule of evidence and have examined numerous cases English and Indian with a view to ascertain the principles upon which the Court applies these rules. In my opinion the principles may be stated as follows:—

(a) The evidence of an approver does not differ from the evidence of any other witness save in one particular respect, namely,

(b) that the evidence of an accomplice is regarded ab initio as open to grave suspicion. Accordingly,

(c) if the suspicion which attaches to the evidence of an accomplice be not removed that evidence should not be acted upon unless corroborated in some material particular, and,

(d) if the suspicion attaching to the accomplice's evidence be removed then that evidence may be acted upon even though uncorroborated, and the guilt of the accused may be established upon that evidence alone.

In view of the rule according to which the evidence of an accomplice must be regarded with grave suspicion it is the practice, amounting almost to a rule of law, that a jury must be warned expressly of the danger in accepting the uncorroborated evidence of an accomplice and if the warning is omitted a conviction based upon such uncorroborated evidence must be set aside. And in the case where there are several accused corroboration is required with respect to the guilt of each individual accused, that is to say the fact that the evidence of the accomplice is corroborated as to certain of the accused does not amount to corroboration of that evidence as regards the guilt of the other accused. It must be remembered, however, that in dealing with the requirement as to corroboration one is ex hypothesi dealing with the case in which the presumption of suspicion attaching the accomplice's evidence has not been removed. In cases where the tribunal is satisfied for good reason that the evidence of the accomplice is truthful the tribunal is under no obligation to demand corroboration.

Now in order to ascertain whether the evidence of the accomplice is truthful and therefore exempt from the requirement of corroboration the tribunal should apply intrinsic as well as extrinsic tests but if having applied these tests it comes to the conclusion that the accomplice is a truthful person the accomplice then becomes an ordinary witness, section 134 becomes operative and the tribunal may proceed to convict upon his evidence alone. In any circumstances it will be seen that the acceptance of the uncorroborated testimony of an accomplice must be an exceptional event and in India where special care is needed in scrutinising the evidence of witnesses in general and of accomplices in particular such a case will be even more exceptional.

To hold as a general principle that the evidence of an accomplice cannot be accepted without corroboration would be to ignore not only the precise words of section 133 but the whole course of established practice of which that section is a summarised statement. There is a tendency on the part of tribunals to imagine that the rules of evidence are of a transcendental character with some ulterior sanction other than the practical exigencies of the administration of justice. The rules of evidence are really nothing more than rules of practical convenience. It is true that there are some rules which have their origin in questions of policy and in such cases the rule takes the form of a specific enactment by the legislature forbidding the Court to take into consideration certain specified matters but the uncorroborated evidence of an approver does not fall in this class and indeed in this country by section 133 is expressly excluded from that class. This question may be again viewed from another standpoint. Courts of justice being concerned with practical decisions demand proof in conformity with practical standards. There is no such thing as absolute proof of the existence of any past event. The standard of proof is fixed by the requirements of practical safety in the light of common experience, of action founded upon the state of belief induced by the evidence. To require invariable corroboration of an approver would create a standard of proof which might plunge the work of the Court and the detection of crime into futility especially in the case of the crime of dacoity where detection of the individuals concerned is a matter of great difficulty and the crime itself is very common and very grave.

The learned Sessions Judge whose judgment we have carefully studied together with the confession of the accomplice and his evidence given in open Court appears to have been fully aware of the principles I have stated above and to have applied them with special care. He examines the intrinsic character of the evidence in order to ascertain whether it exhibits any discrepancies. The story as told by the approver has an air of inherent probability. His account of the dacoity is told in great detail and is consistent throughout. The description of the site is accurate as is also the description of the house and rooms entered and is such as would be difficult for a person who had not seen them to acquire from others; where the story is in conflict with the accounts given by other witnesses there are excellent reasons stated by the learned Judge for believing in every case of discrepancy that it is the other witness who is making a mistake or telling an untruth rather than the approver. There is no evidence of any enmity between the approver and the other accused and his story was not in the least shaken by cross-examination. There is moreover the fact of the corroboration of the approver's story with regard to the guilt of the five accused first mentioned. This fact while not amounting to corroboration of the approver's evidence as to the remaining six accused may nevertheless be properly taken into account as one of the considerations pointing to the conclusion that he is a witness of truth. There is a further confirmatory element of great importance. Until the approver was apprehended and gave his account the police had in mind a theory of the series of events which had taken place and of the persons concerned which the evidence of the approver failed to support. Had the approver been tutored by the police it is unlikely that the differences and omissions would not have been supplied by the police. Finally it must be remembered that the defence called no evidence to contradict the story told by the approver. The learned Additional Sessions Judge moreover saw the approver and heard him give evidence and obtained the impression from his demeanour that he ought to be believed. In my opinion the learned Additional Sessions Judge in applying these intrinsic and extrinsic tests to the evidence of the approver was acting with that degree of caution required by the general practice of all British Courts and moreover observed with care the special conditions presented by Indian conditions and by the circumstances of this case. He has rightly accepted the evidence of the approver and the convictions and sentences should not be disturbed. The appeal must be dismissed.

Macpherson, J.:— I agree generally. No doubt in actual practice the uncorroborated testimony of an accomplice will generally be insufficient to bring home an offence to an accused person. The law on the point, however, as laid down in sections 4, 114, 133 and 134 of the Indian Evidence Act gives no countenance to the contention that the uncorroborated testimony of an accomplice is necessarily insufficient to establish a charge against an accused. The high artificial value accorded by English law to the doctrine that an accomplice is unworthy of credit unless he be corroborated in material particulars, is not reproduced in that enactment. On the contrary, it is in section 114 to which it is appended as Illustration (b), definitely designated a maxim, so that the Courts are subject to no technical rule on the subject. Indeed the Court is there enjoined in considering the applicability of such maxims to the particular case under decision to “have regard to such facts as” are set out in examples cited in each of which it is at least suggested that the maxim is inapplicable. Of the nine presumptions of English law set out as illustrations and designated maxims Illustration (b) is the only one to which more than one such example is appended. Manifestly therefore there is no conclusive presumption against the testimony of an accomplice even when uncorroborated in material particulars. Indeed so far from being enjoined to make such a presumption in all cases the Court is not even empowered to do so. In considering whether to apply or not to apply the maxim it is incumbent on the Court to exercise a judicial discretion and to have regard to the facts of the particular case. Thereafter if it declines to make the presumption it will call for proof. If it makes, the presumption then under section 4 it is a rebuttable presumption, that is, the Court will regard the fact as proved unless and until it is disproved.

It is in practice that the difficulty arises. It is often met by simply treating the rebuttable presumption which it is discretionary to make, as conclusive and restricting the further enquiry to considering whether the testimony of the approver or other accomplice is corroborated in material particulars as to the occurrence and as to the participation therein of each individual accused. Such a course has not infrequently received approval in judicial expressions of opinion evoked by the peculiar circumstances of the case in which they arose and it doubtless secures the ‘safety’ at which it aims but it does so only by jettison of the statutory provisions of law and even by tilting the scales of justice. And of course the discretion of the Judge cannot be fettered by such dicta. As was well said by Jenkins, C.J in ‘An Attorney, In re.’ “Not one jot or one tittle can be taken away from or added to the plain and express provisions of the Legislature by any decision of the Court; nor can this discretion vested by the section in the Court be crystallized or restricted by any series of cases: it remains free and untrammelled to be fairly exercised according to the exigencies of each case.” And again, as was pointed out in Gardner v. Jay, when a tribunal is invested by statute with a discretion without any indication of the grounds upon which the discretion is to be exercised, it is a mistake for a superior tribunal to lay down any rules with a view to indicating the particular grooves in which the discretion should run. Here the dicta would actually defeat the intention of the Legislature as laid down in the provisions cited. In fact not only is there no conclusive presumption but it is not even imperative to make a rebuttable presumption. There is only a maxim which the Court is authorised in its discretion to apply or not to apply. If the maxim is not applied by the Court the defence has the right to prove that it should be applied; if the maxim is applied the prosecution has the right to disprove its applicability in the circumstances of the particular case.

No doubt the same result will often be reached in a case by following the course mentioned as by following the terms of the statute. For instance, in the latter case the Court may demand rebutting evidence of great cogency, before regarding as ‘disproved’ the ‘fact’ which in its discretion it judges it right to ‘presume’, that is to say, to regard as proved subject to rebuttal or disproof by the prosecution. But there are certainly cases, as the qualifications to Illustration (b) to section 114 abundantly demonstrate,—exceptional cases perhaps—in which the Court is warranted in declining to apply the maxim and will thereupon call for proof of the ‘fact’ embodied therein. The considerations to which the Court must have regard in determining whether to apply or not to apply the maxim certainly include the intrinsic character of the deposition of the accomplice as well as extrinsic circumstances which tend to establish that in important particulars where his tesitimony can be satisfactorily tested, he has shown himself a witness of truth. In such case the Court clearly does not err in law in declining to apply the maxim.

On a perusal of the record and judgment before us I am satisfied that in this particular instance the learned Additional Sessions Judge was right in holding that the approver was a witness of truth in respect of all he stated, including his testimony as to the complicity of the six appellants in respect of whom be is not corroborated in material particulars, in declining in the circumstances to apply the maxim in Illustration (b) to section 114 and, in the absence of any proof of the ‘fact’ set out in the maxim, in convicting the said appellants on the testimony of the approver. At the same time I desire to add that in a long experience of criminal cases in India I have found only a few other instances in which I inclined to convict or to maintain a conviction depending on such uncorroborated testimony.

Appeal dismissed.