1. I agree that Exhibit G is a statement that should not have been taken into consideration against the 1st accused. He and the maker of the statement were being jointly tried for murder and Exhibit G was not a confession of murder jointly affecting both. The law on the point was correctly laid down by Straight, ]., many years ago in the Allahabad case Empress of India v. Ganraj (1879) I.L.R. 2 A. 444 cited by my learned brother. As regards the case Shivabhai v. Emperor (1926) I.L.R. 50 B. 683 a statement by one of the accused that he by himself had burnt the clothes of a murdered man and would show the place was treated as a confession of participation in the murder and admissible under Section 27 of the Evidence Act against him. To that extent, the decision seems to be correct, but when it goes on to put forward some circumstantial grounds on which the Judges held that the confession "indirectly affected" another accused not named in it and could, therefore, be vised against him under Section 30 of the Evidence Act, I find myself wholly unable to follow it.

2. Apart from the so-called confession, there is, I think, a strong circumstantial case against the 1st appellant. Against the 2nd, there is that and more, that is to say, his statements, Exhibits G and H. They show that he was present when the murder was committed. That they give a true account of what happened is incredible. The murder and disposal of the body could not have been the work of one man. Nor is it believable that the 1st appellant would have suddenly murdered Ramaswami in the presence of the 2nd appellant without the previous knowledge and consent of the latter. The scene of the murder was probably deliberately chosen. It was a place where the Periyar channel, which was then certain to be in flood, runs close to the road. Doubtless the intention was, as stated in Exhibit H, to throw the body into the channel. Apparently, in the darkness, the murderers mistook the water in the burrow-pit for the channel. I agree in the order proposed by my learned brother.

Reilly, J.

3. In this case the appellants, accused 1 and 2, have been convicted by the Sessions Judge of Madura of murdering one Ramaswami Moopan, with whom it has been found they were dealing in sheep as partners.

4. The learned Sessions Judge has taken into consideration against accused 1 a statement, Exhibit G, made by accused 2 before the Sub-Magistrate of Dindigul. In that statement accused 2 does not admit that he took any part in the murder of Ramaswami; he says that accused 1 killed Ramaswami and then compelled accused 2 by a threat to assist in disposing of the body. The learned Public Prosecutor has contended that Exhibit G is a statement which the Sessions Judge was at liberty to take into consideration against accused 1 under Section 30 of the Evidence Act because it was a confession, not indeed of murder, but of causing the evidence of murder to disappear, which is an offence punishable under Section 201, Indian Penal Code. In my opinion that contention is mistaken. Section 30 of the Evidence Act is a very exceptional, indeed an extraordinary, provision, by which something which is not evidence may be' used against an accused person at his trial. Such a provision must be used with the greatest caution and with care to make sure that we do not stretch it one line beyond its necessary intention. It is true that the section provides only that the confession of one accused person may be "taken into consideration" against his fellow-accused. As I understand the section, the confession cannot take the place of evidence against the co-accused; nor can it be added to supplement evidence otherwise insufficient. As I understand the matter, the provision goes no further than this- where there is evidence against the co-accused sufficient, if believed, to support his conviction, then the kind of confession described in Section 30 may be thrown into the scale as an additional reason for believing that evidence. But even for that limited purpose the confession must be of the kind intended by the Legislature when enacting Section 30. What kind of confession is intended? The words of the section are:

When more persons than one are being tried jointly for the same offence, and a confession made by one of such persons affecting himself and some other of such persons is proved.

5. In my opinion "confession" in that collocation of words cannot reasonably be interpreted to mean a confession of any offence in the world, nor even of any minor offence included in the offence for which the accused persons are being tried, nor of any offence connected with that offence, nor of any other offence which may be disclosed by the evidence, but only of the very offence for which they are being tried "offence" always including under the explanation to the section abetments and attempts. That appears to me to be the plain meaning of the words. To interpret the word "confession"' in the section in any wider sense is to accuse the Legislature of using loose language in a matter of great importance. In my opinion we must read the word "confession" as if it were followed immediately by the words "of that offence"; and that appears to me to be the plain meaning of the section as read by a reasonable man. And indeed, if that were not the meaning, why should the use of the word "confession" be confined to cases where the accused persons are being tried for the same offence? But, keeping our attention for the moment on the words themselves, if we speak of persons being tried for the same offence and in the same sentence speak of one of them making a confession, how can we mean by "confession" a confession of any other offence unless we are most careless in the use of language? To my mind the words of Section 30 of the Evidence Act leave no doubt that the confession mentioned is a confession of the offence for which the accused persons are being tried. If the words left us in any doubt, the nature and effect of the provision might be called in to help us. The provision allows a statement of an accused person, not made an oath, not tested by the cross-examination, to be used against a person tried with him. What can be the justification of a provision at first sight so startling? If, while confessing some minor offence, a man could implicate his fellow-accused in a far more serious offence for which they were being tried, and the confession could then be used against his fellow-accused, would not that be offering to the confessing accused a plain temptation to make himself useful by throwing the serious blame on his fellow-accused while saving his own skin or exposing himself only to a minor penalty? To my mind it is inconceivable that the Legislature had any idea of offering to any accused person a temptation to such mean and immoral conduct. But, if the confession which may be used is confined to a confession of the very offence for which the accused persons are being tried, the position is different. Then at least the confessing accused has nothing to gain by his confession; on the contrary he exposes himself to the very penalty which he helps to bring down on his fellow-accused. To such a confession it is not entirely unreasonable to allow some weight, and the danger that it may be untrue is at least very greatly reduced. If we examine the effect and nature of the provision, we arrive again at the same result, that it is only such a confession which the Legislature can have allowed by Section 30 to be taken into consideration. Lastly, if in thise respect two interpretations of the section were grammatically possible, the ordinary canons of interpretation would compel us to adopt in the interest of accused persons the stricter and narrower construction, the construction which gives the less scope to this very exceptional extension of what may be used against an accused person at his trial. Examining the section in any or all these ways, in my opinion, we must arrive at the same conclusion, namely, that "confession" in Section 30 of the Evidence Act means confession of the offence for which the accused persons are being tried.

6. It appears to me surprising that any other interpretation of the section should now be put forward. The interpretation I have indicated as proper was adopted by Straight, J., in Empress of India v. Ganraj (1879) I.L.R. 2 A. 444 and his view has been followed in a great many later cases. The learned Public Prosecutor quoted in his favour Shivabhai v. Emperor (1926) I.L.R. 50 B. 683. So far as that case deals with this point, I must say with very great respect that I neither agree with it nor understand it.

7. In Exhibit G, as I have mentioned, accused 2 does not admit that he took any part in the murder of Ramaswami Moopan: he admits only that he helped to dispose of the corpse. That is not a confession of the offence of murder for which accused 1 and 2 were tried, and therefore in my opinion the learned Sessions Judge in dealing with accused 1 should have excluded Exhibit G entirely from consideration. If that is done, what is the evidence against accused 1? There is evidence, which I see no reason to doubt, that for a few weeks accused 1 and 2 had been trading in sheep in partnership with Ramaswami Moopan and that they had taken some sheep hought in the Trichinopoly District south through Dindigul to Madura. There is evidence that on Thursday, 28th November, 1929, they arrived at Madura with 12 sheep and stayed that night with P.W. 27, who knew accused 2, as accused 2 had married and divorced a daughter of P.W. 27's neighbour; on the Thursday they sold 2 of the sheep, and on the Friday they sold the rest with the help of P.W. 27, who himself bought the last one for Rs. 11-8-0; the money for the sheep sold on the Friday was given to Ramaswami Moopan, who was carrying a money bag; about 4 p.m., on Friday, 29th November, accused 1 and 2 and Ramaswami left P.W. 27's shop at Madura saying that they were going to walk home. This evidence of what happened at Madura is given by P.W. 27, whom there is no reason to distrust. The villages of Ramaswami and of accused 1 and 2 are close to each other, a few miles north-west of Dindigul. Their way from Madura lay along the Madura-Dindigul trunk road, which from Madura runs first slightly north of west and then slightly west of north. On the morning of Saturday, 30th November, Ramaswami's dead body was found near Vadipatti with 9 cut wounds on it floating in water in a burrow-pit by the side of the road, between it and the Periyar channel close by, 17 miles from Madura, at a point which the accused and Ramaswami would have reached in their journey at some time in the previous night. Though an inquest on the body, on which the only money found was 2 1\2 annas, was held on 30th November and a photograph of it and prints of its fingers were taken, it was not identified until 12 days later. P.W. 11, Ramaswami's widow, gives evidence that on Sunday, 1st December, She met accused 2 at a shandy near her village and asked him where Ramaswami was. Accused 2 replied: "We went to Madura and sold sheep. He got his share in the sale and went to the east to buy sheep and would be returning in 2 or 3 days. Why are you searching for him?" P.W. 12, Ramaswami's brother and P.W. 14 say that they went on Monday, 2nd December, to the village of the accused and asked accused 1 where Ramaswami was. Accused 1 told them that they had sold the sheep at Madura, that Ramaswami had got his share of the proceeds and had gone to "eastern parts" to buy other sheep and would be back in 2 or 3 days. P.W. 12 adds that accused 1 told him that Ramaswami's share of the sale-proceeds was Rs. 57. That the accused gave this explanation to Ramaswami's relations when they became anxious because he did not return is proved not only by oral evidence. Exhibit K is a letter, dated 4th December, written by P.W. 20, for P.W. 21, who is the brother of Ramaswami's widow, to their cousin, P.W. 22, who lives in the Trichinopoly District, near the place where Ramaswami and the accused had bought the sheep which they sold at Madura; and that part of the Trichinopoly District is east of the north part of the Madura District, where the accused and Ramaswami lived. Exhibit K runs:

Our brother-in-law, Palaniamma's (P.W. ll's) husband, went there (i.e., to your place), purchased sheep and went to Madura to sell them. Those who went with him have returned about 7 days ago. When questioned, they say that Ramaswami has gone to eastern parts and I suspected it. So I request you to search for him in those parts and send him back at once and, if he is not., found, to inform me about it.

8. The genuineness of this letter cannot be disputed. Its history in the post is shown by the post-marks on its cover, Exhibit K-l; and Exhibit L is P.W. 22's answer to it, dated 12th December.

9. That is the evidence against accused 1. It shows that he left Madura on the afternoon of 29th November in company with Ramaswami and accused 2 to walk to their homes; Ramaswami was then carrying a considerable sum of money; next morning Ramaswami's body, with wounds upon it which showed that he had been murdered, was found by the roadside at a point which they had to pass on their journey and would have reached in the night: there was practically no money on the body: the accused went on to their village, and on 2nd December accused 1 gave to P.Ws. 12 and 14 a reassuring but false account of Ramaswami's whereabouts. Accused 1 gives no explanation of this. He does not explain when or where he left Ramaswami nor profess that he had any reason to suppose that what he told P.Ws. 12 and 14 was true. He denies that he ever traded in sheep with Ramaswami and denies that he ever went to Madura with Ramaswami and accused 2. These denials are no doubt false. From the facts that accused 1 left Madura with Ramaswami and accused 2, that on the way they were to travel Ramaswami was 'murdered and the money on his person removed, and that 3 days later accused 1 gave a deliberately false account of Ramaswami's movements, the proper inference in my opinion is that accused 1 took part in murdering Ramaswami.

10. Against accused 2 there is the same evidence as against accused 1 except that his statement about Ramaswami's movements was made a day earlier, on 1st December, and was made to Ramaswami's widow, P.W. 11. But besides this evidence there is against accused 2 the statement Exhibit G, which he made to the Sub-Magistrate of Dindigul on 16th December. In that statement he admits that he and accused 1 and Rama-swami after they had sold their sheep at Madura left P.W. 27's house in the afternoon; they went along the road, and at midnight, when they were near Vadipatti, accused 2 was walking in front, Ramaswami next and accused 1 last. Accused 1 caught Ramaswami by the feet and threw him down and stabbed him to death. Accused 2 took no part in that, but, when Ramaswami was dead, accused 1 by threatening to stab him made him help in throwing Ramaswami's body into a burrow-pit by the roadside, Tha.t statement accused 2 withdrew in the committing Magistrate's Court and at the trial, explaining that he made it because he had been beaten and was afraid. Oddly enough in his appeal petition he says that he did make the statement because he was afraid but that he stated what was true. However accused 2's admissions in Exhibit G are corroborated in a remarkable way. There is evidence that at his own village on 9th December accused 2 told the same story to P.Ws. 12, 14, 15, 16 and 18. They all say that accused 2 then gave them practically the same account of the murder as is given in Exhibit G except that P.Ws. 12 and 18 say that accused 2 told them that he refused to help in throwing the body away and that accused 1 did it by himself. In corroboration of the evidence of P.W. 12 about accused 2's admission on 9th December we have Exhibit H, the statement made by P.W. 12 to his Village Magistrate, P.W. 19, the same day, when he produced accused 2 before him; and the Village Magistrate adds that, when questioned by him, accused 2 admitted that he had told "the Pan-chayat" what is recorded in Exhibit H. When accused 2 told his story of the murder to P.W. 12 and the other witnesses on 9th December, none of them knew what had become of Ramaswami, still less that he had been murdered near Vadipatti, more than 20 miles away. Nor did the Police of Sholavandan Station, within the limits of which Ramaswami's body had been found, then know whose body it was. On 10th December the Sub-Inspector of Palakanuthu, the Station for the village of the accused, sent P.W. 12, Ramaswami's brother-in-law, to Vadipatti: there on 11th December P.W. 12 learnt that a body had been found, the description of which was like Ramaswami: next day he went to Sholavandan Station and saw Exhibit B, the photograph of the body, which he recognised as that of Ramaswami. I may add that the identification of the body is not disputed, as besides the photograph the finger prints taken from the body correspond with those of Ramaswami on a sale-deed executed by him, Exhibit F. But it was because accused 2 told his tale of the murder on 9th December that Rama-swami's fate was discovered and his body identified. Though he has withdrawn his statement, Exhibit G, there is no doubt that his admission in it that he was present when Ramaswami was murdered is true. That is what he admits in Ex. G. But taken with the other evidence that cannot be accepted as the limit of his guilt. In the circumstances it is very highly improbable that, as accused 2 represents, accused 1 fell upon Ramaswami without a word to accused 2 of his intention. Of the 9 wounds found on Ramaswami's body only one, on the right of the neck, cutting the carotid artery, was necessarily fatal. From the fact that so many wounds were inflicted it may be inferred that Ramaswami was not killed without a struggle. Accused 2, according to his own account, stood by and was not responsible for what was done. To my mind in the circumstances the only reasonable inference is that accused 2 took part in the murder, whether he actually had a hand in the struggle, as is probable, or merely stood by on guard.

11. In my opinion the convictions and sentences of both the accused should be confirmed and their appeals should be dismissed.