Thus we may discern within the judgment itself no less than three different resolutions to the question, deriving respectively from (a) CrPC 227 itself: “material (i.e. records and documents) and arguments disclosing sufficient ground”; (b) Niranjan Singh: “Facts prima facie disclosing all ingredients” ; and (c) Para 20 of Keshub Mahindra: “Material prima facie showing guilt”. Niranjan Singh differs from the others in that it emphasises facts rather than material. Since at this stage the evidence has not been examined, “facts” must mean alleged facts, i.e. the facts stated in the prosecution’s case, not established or substantiated facts. “Material” indicates indicate that which substantiates or establishes the facts. The question is, how does Niranjan Singh envisage the role played by material in S. 227.

The phrase “evaluate the material to find out if the facts emerging therefrom” (p. 642) is significant here. It seems to indicate that the judge must examine if the facts alleged in the prosecution’s case do emerge from the material furnished. But would this not entail examining the substantive aspects of the material? Perhaps this can be avoided by contending the judge is required to ascertain only if the material relates prima facie to the facts. It must be mentioned that the judgment’s inference itself derives from Union of India v. Prafulla Kumar Samal ((1979) 2 SCR 229). The four-part test proposed therein (at p. 234), which Niranjan Singh reproduces, is even more restrictive in its language. Its first aspect states, “[T]he Judge . . . has the undoubted power to sift and weigh the evidence for the limited purpose of finding out whether or not a prima facie case against the accused has been made out. (emphasis added)” This reconciles easily with our conjecture: Prafulla Kumar effectively restricts the judge’s perusal of the material to ascertaining if the material relates prima facie to the alleged facts. (Aside: the reasoning in Prafulla Kumar contains some further inferences with which I find myself unable to agree. However, as they play no role in the present matter, we may safely overlook them here.)

Let us now examine Keshub Mahindra in the light of the above. In Para 20, the Court states: “[T]he material on record must at least prima facie show that the accused is guilty of culpable homicide . . ..” This begs the question what this concept of prima facie guilt means. For law students trained to construe guilt in terms of proven beyond reasonable doubt, it seems especially difficult to comprehend. Let use then use a classic philosophers’ tactic here, and reverse the meanings of the terms involved. There exists a significant logical gap between “guilty” and “innocent”. Scots law, for example, gives the option of a “not proven” verdict when the accused’s guilt cannot be proved and yet suspicion upon her is too strong to merit an acquittal. Going by this logic, the inverse of “guilty” can only be “not proven” or “guilt not established”, within which “innocent”, “acquitted” and “not guilty” form specialised subsets. Hence the “material prima facie showing guilt” principle articulated in Para 20 of Keshub Mahindra inverts into “material not prima facie showing (or establishing) guilt”.

This device assumes greater significance when we examine the issue of burden of proof. The wording of CrPC 227 expressly requires the judge to hear either side. Which means that the onus rests on neither side fully, but partially on both the prosecution and the defence. The question is, what is the extent to which the defence needs to establish its case? If we go by Keshub Mahindra, the manifest answer lies in the inversion we derived in the previous paragraph: the defence must establish that the material does not prima facie show or establish guilt. But this is difficult to reconcile with what we discussed earlier. Going by our discussion of CrPC 227, the defence has to prove the charges are groundless in the light of the material. Similarly, as per the rule emerging from Niranjan Singh and Prafulla Kumar, the burden on the defence is to establish that the alleged facts of the prosecution’s case do not emerge from the material.

Clearly, proving the charges are groundless in the light of the material, or that the alleged facts do not emerge from the material, is much more difficult than establishing the material does not disclose prima facie guilt. It would appear, therefore, in construing CrPC 227 in the manner that it did, the Court in Keshub Mahindra served to relax somewhat the burden of proof on the defence. Given the extremely serious nature of the case (the judgment itself at Para 4 calls it “A grim tragedy of unprecedented nature”), I find myself unable to agree that this expedient was appropriate.

One may well argue that the consequences of all this is negligible – since the judgment based itself not on burden of proof but on what we saw was an incorrect appraisal of facts on record, the Court’s lowering the defence’s onus probandi amounts to obiter dicta. But obiter dicta means only that it will not be binding on lower courts in the future, something that does not concern us here. What we are trying to understand is how the Court responded to the “unprecedented nature” of the situation. It erred not once but twice – the first time in incorrectly construing the facts, and the second time in lowering the burden of proof when neither statutes nor precedents warranted it. And whatever be the reason behind it, the fact is that both errors tended to favour the accused, the perpetrators of the “grim tragedy”. This is what makes Keshub Mahindra verdict both tragic and inexplicable.

The story does not end here, though. As yet we have looked at only one aspect of the issue, that is, the role of the judiciary. It transpires that the prosecution’s handling of the case was not beyond reproach either. Both during the proceedings before the Court and subsequent to the verdict, it (the prosecution) had several options before it which it chose not to exercise. We shall examine these in the next part.