1. Taking Criminal Appeal No. 40 first, in this case 8 of the 24 accused were convicted by the acting Sessions Judge of Goodavari division, as to accused 2 and 4 to 8 under Section 147, as to accused 1 under Sections 326, 324 and 148 and as to accused 3 under Sections 324 and 148. The riot in which these persons are held to have taken part was a fight between the Panchamas of Voolapalli and the Panchamas of Biccavole, a village close by, and it took place about midday on 12th April 1922 at the boundary between the two villages. The story of the beginning of the fight is told by P.W. 2, whose credibility there is not the least reason to doubt. He is an Inspector of Co-operative Societies and, when on tour from Biccavole on the day in question, 9th accused told him that the Biccavole and Voolapalli Panchamas had quarrels and that the Voolapalli people were going to Biccavole to settle the quarrel. It appears that some Voolapalli men had married Biccavole wives and had refused to fetch them away to their houses. P.W. 2 was asked to mediate. He agreed, but thought it would be better if he had some support and went to fetch the schoolmaster, P.W. 9. He left the two parties of Panchamas, one on each side of the boundary. There was a panchayat formed, but according to the evidence of P.W. 2 a Biccavole man whistled and this was a signal for a general fight between the two factions. P.W. 2, seeing a fight was imminent, went away. He says there were about 70 of the Voolapalli people and about 30 or 40 of the Biccavole people. He saw no weapons in the hands of either party. P.W. 9 does not give any useful evidence as he. seeing the two factions sitting opposite each other on the boundary, apprehended that there would be a fight and ran away. That there was a fight cannot be disputed. The learned Sessions Judge held that accused 1 and 3 were the leaders of the Voolapalli party and that the fight was entirely due to their action.
2. The other case, Criminal Appeal No. 39, is a case in which certain members of the Biccavole party, accused 1, 3,10,11 and 22 there have been convicted, as to accused 1 and 3 under Sections 148 and 324 and as to accused 10, 11 and 22 under Section 147. The accused in case No. 39 are most of the witnesses in the other case, Criminal Appeal No. 40. Mr. Lakshmanna who appeared for the accused in criminal appeal No. 39 pleaded self-defence largely, I think, on the ground that the learned Sessions Judge, possibly rather inaccurately, says 'there is a certain element of self-defence to be considered,' though he adds that the doctrine is not, in a rioting case, to be discussed when the question is whether a man is guilty or innocent of the offence of rioting. It is admitted that there is no evidence of self-defence on the record. The statements of the prosecution witnesses in Criminal Appeal No. 40 were relied on but were not put in at the trial. It does not seem to me that the defence of self-defence which must be specific and individual in every case can be relied on here where there is good evidence to hold that there was a general fight between these two parties. The plea in 39 is that the Biccavole party went to the panchayat peacefully and they got the worst of it and that the Voolapalli party were the aggressors. There is no evidence of this whatever. In fact what little evidence there is (of P.W. 2) seems to show that the Biccavole party who are the accused in 39 began the fight as one of their numbers sounded a whistle which was the signal for the general engagement.
3. The case would present no subject of interest were it not for the fact that the witnesses in both the cases partly resiled in the Sessions Court from their statements made before the committing Magistrate. For instance, to take one example from the record in Criminal Appeal No. 40, P.W. 4 before the committing Magistrate's Court distinctly stated that the first accused cut him on his right wrist with a sword stick. Third accused with a sword stick hit him en the shoulder, seventh accused hit him with a stick on his back. In the Sessions Court he admits there was a fight and that he was injured on his right wrist and that it was caused with a knife, but he could not say who gave him the blow. He says the police forced him to tell the Magistrate that the seventh accused hit him on the back with a stick, and that he did not see the first accused strike anybody. To take another example, P.W. 5 in the Magistrate's Court says that the first accused stabbed him with a knife on his left palm. In the Sessions Court he said he received a knife cut on the palm of his left hand. He did not notice who gave him the blow. When he was examined by the committing Magistrate he says he was confused. P.W. 6 says there was a fight, but he does not know how it arose and that some one gave him a blow on the back with a knife. Before the committing Magistrate ho said that the first accused chased him and cut him with a knife on the back. In the Sessions Court he said that the police compelled him to implicate some one and that was why he said that the accused had struck him on the back with a knife. P.W. 7 said in the Sessions Court he received no injuries that day and that he went to the hospital to get the wound attended to, which is obviously false (of. Exhibit G). P.W. 8 said in the Sessions Court that he received three injuries of which two were caused with a knife and the other with a stick. He cannot say who struck him. In the committing Magistrate's Court he said the accused 1, 7 and 6 hit him with knives. He also says that the police frightened him. The same unwillingness in the Sessions Court to name the man who injured each witness is to be observed in Criminal Appeal No. 39. It will be tedious to go through the whole of the depositions, but from what I have quoted above it will be gathered that the witnesses on the whole admit there was a fight. Most of them admit they were injured but when it comes to naming the person who injured them they say they do not know. There is in my opinion ground for thinking that these two villagers have now composed their quarrel and that it was a condition of the settlement that none of them should inculpate the other when it came to the trial. On the best consideration that I can give to the facts of the case it appears to us that the Sessions Judge was perfectly justified, subject to what will be said hereafter, in relying on the statements before the committing Magistrate under Section 288, Criminal Procedure Code, and in believing those statements in preference to the statements made by the witnesses in the Sessions Court when they profess ignorance of their assailants.
4. The allegations made in the Sessions Court as will be seen are two: (1) that the witnesses were confused when they made their statements before the Magistrate and (2) that they were made under compulsion by the police. But the learned Sessions Judge has found that there could have been no confusion in their minds over three months after the occurrence as they were examined by the Magistrate in August whereas the occurrence was in April, and that the police compulsion, if it existed, would be just as strong in the Sessions Court, as it was when they were examined. The Sessions Judge therefore finds that these two allegations were entirely unfounded. This is the reason given by him for preferring the evidence under Section 288. But it is objected before us that the learned Sessions Judge was wrong and that the evidence under Section 288 ought not, under the circumstances, to have been believed by him, as there is no material corroboration of these statements. Several cases were quoted to us on this subject and I might premise by referring to a case to which I was a party reported in In re Velliah Kone : (1922)43MLJ222 where it was held that the effect of Section 288 is to place the deposition of a witness before the committing Magistrate on exactly the same footing with the deposition in the Sessions Court and that it is a testimony within the meaning of Section 157 of the Evidence Act. The learned vakils for the defence in this case quoted before us Queen-Empress v. Jeochi I.L.R. (1899) All. 111 and Queen-Empress v. Nirmal Das I.L R. (1900) All. 445. Queen-Empress v. Jeochi I.L.R. (1899) All. 111 is to the effect that conviction cannot be based solely on the evidence before the Magistrate there being no other evidence to corroborate it. In Queen-Empress v. Nirmal Dos I.L R.(1900) All. 445 it was said 'It is difficult to conceive that any responsible tribunal should permit the conviction of a person upon such evidence (i.e., Section 238), if it stood by itself.' In a later case in the same Court Emperor v. Dwarka Kurmi I.L.R. (1906) All. 683 in which one of the Judges was the Judge who, sitting alone, decided Queen-Empress v. Jeochi I.L.R. (1899) All. 111 it was held that when admitted the statements under Section 28S are exactly on the same footing as all the other evidence in the case, its value being of course a question for the tribunal which has to decide the case on the evidence. Two cases of our own Court were relied on Queen-Empress v. Bharmappa I.L.R.(1889) Mad. 123 which was a case of a retracted confession corroborated by depositions under Section 288, the statements made before the Magistrate being retracted in the Sessions. It is really an example of what the Bombay High Court calls in Queen-Empress v. Gharya I.L.R. (1895) 19 Bom. 728 'the Madras practice of requiring corroboration of a retracted confession' and which it may be noticed the Bombay High Court does not follow. In Queen-Empress v. Bharmappa I.L.R. (1889) Mad. 123 the learned Judges held the depositions read under Section 288 and retracted at the trial are not by themselves material corroboration of a retracted confession. But in Queen-Empress v. Dorasami Ayyar I.L.R(1901) . Mad. 414 it was held that under Section 288 the Court is not restricted to using the evidence before the committing Magistrate only for the purpose of contradicting the witness at the trial. The section was intended to enable a Court to read the previous evidence as substantive evidence at the trial where for the purposes of justice the adoption of such a course is found necessary by the Judge. The only other case that is necessary to notice is the one cited in King-Emperor v. Bhut Nath Ghose (1902) 7 C.W.N. 345. In that particular case it was held that the Sessions Judge did not exercise a proper discretion in acting on the statements under Section 288. It is perfectly clear from the later Madras ruling cited above that, in a case like the present where you clearly have a conspiracy to suppress the names of assailants in the Sessions Court but you have corroboration of the statements made under Section 288 in all material respects except this, the Sessions Judge was perfectly justified, in order to attain the ends of justice, in regarding the statements under Section 288 as substantive evidence at the trial. The contention on behalf of the accused that there must be corroboration on the very point on which the witnesses have resiled is clearly in my opinion wrong. If this is necessary, then I cannot see any scope for the statements under Section 288. In my opinion all that the earlier cases come to is that where you have two equally balanced stories, one story in the Sessions Court and the second the story as told before the committing Magistrate, it is as a matter of discretion unsafe to prefer the latter rather than the former. That is a case of an equal balance of evidence. When there is nothing to show the preponderance of one over the other that may be a perfectly good doctrine. Whereas here you have got general agreement in the two stories the only difference being in the name of the actual assailant of each witness, and such a doctrine is clearly untenable. It therefore seems to me that the legal point fails.
5. With regard to the facts of the two cases I am of opinion that they have been fairly found by the Sessions Judge on evidence which was available for him to consider and I am not disposed to differ from his findings. The first accused in Criminal Appeal 40 has been given consecutive sentences amounting to three years and the third accused 1 1/2 years. It is suggested that these are unduly severe. It must be pointed out that this was a serious riot in the course of which one man of the Biccavole party died as the result of injuries received. On consideration I am unable to see any reason to remit any part of the sentences. I would confirm the convictions and sentences and dismiss the appeals in each case.
6. I agree with my learned brother for the reasons given by him that these appeals should be dismissed.
7. I wish to say something on the point of law raised in these two cases. It is quite clear from a comparison of the depositions in the Sessions Court of each set of witnesses in these two cases with their depositions in the committing Court that each eye-witness supported his general account of the fight in every point except that of identifying particular assailants and that each, with one accord, in the Sessions Court professed his inability to make any such identification of assailants. None of them goes so far as to give a categorical denial of his former statement that such and such an accused person committed such and such an act, but all say merely that they cannot identify particular assailants or that they do not remember who they were. When it is noted that in S.C. No. 51 of 1922, 13 out of the 14 eye-witnesses are accused in S.C. No. 52 and that in S.C. No. 52 8 out of the 9 eye-witnesses are accused in S.C. No. 51, the conclusion cannot easily be resisted that between the hearing in the committing Court and the hearing in the Sessions Court both parties came to a mutual agreement that each would not in the Sessions Court identify as assailants the members of the other party. Such a conclusion could only be rebutted if there are strong grounds for thinking that the identification evidence in the committing Court was due to some coercion or undue influence and that it was so has been argued before us. It is true that in S.C. No. 51 seven out of the fourteen eye-witnesses say that police pressure to identify particular assailants was put on them, but three others of these witnesses merely say their identification in the committing Court was due to a confusion of mind, while the other four give no reason at all. In S.C. No. 52 only one eye-witness out of the nine complains of police pressure and the rest give no reason. The theory of police pressure is untenable, since that would have been tightened and not relaxed when the case came to Sessions. It is further very significant that in the Sessions Court in both the cases only two out of the twenty-three eye-witnesses were cross-examined at all and these two were only asked a few unimportant questions. The defence pleaders in each case evidently were aware from the beginning of the hearing of the cases there that no eye-witness was going to identify any of the accused and therefore it was not worth while cross-examining even on the general features of the fight. The common sense conclusion is that in order to save one another the members of each party had agreed not to identify their assailants in the Sessions Court, that is, they entered into a concerted conspiracy to defeat the ends of justice.
2. That being the reasonable deduction from the facts we have to decide whether in such circumstances the lower Court was justified in law in admitting under Section 288, Criminal Procedure Code, and using as substantive evidence against the accused in each case the evidence of identification given in the committing Court, such evidence of identification being lacking in the depositions given before it. It is hardly necessary to emphasise here that, once a Court has admitted on behalf of the prosecution evidence under Section 288, the depositions so admitted are just as much evidence for the prosecution in the case as the depositions recorded for the prosecution in the Sessions Court. There is nothing in Section 288 which indicates that there is any difference in the probative value of the former when compared with the latter. See In Re Velliah Kone : (1922)43MLJ222 , Emperor v. Dwarka Karmi I.L.R(1906) . All. 683 and Queen-Empress v. Dorasami Ayyar I.L.R. (1901) Mad. 414. Whether the Court is justified in admitting under Section 288, with mechanical regularity, depositions given in the committing Court of any and every witness who retracts in the Sessions Court is a, question to which I shall come later.
3. We have been referred in argument at the Bar to various cases from which it is sought to deduce a general proposition that a conviction by a Sessions Court based wholly on depositions admitted under Section 288 is bad in law. We have not been referred to any ruling which lays down any such broad proposition. The broad principle to be observed clearly is that where witnesses are so careless of the truth as to abandon readily on oath what they have previously sworn on oath, their statements on any point should not be accepted without great caution and sound judicial reason for accepting as true anything they have said. Sometimes this principle has been interpreted in particular cases to mean that sound judicial reasons are absent unless there is independent corroboration on material particulars; see Queen-Empress v. Jadub Das I.L.R.(1900) Calc. 295 Queen-Empress v. Jeochi I.L.R.(1899) All. 111 and Queen-Empress v. Nirmal Das I.L.R.(1900) All. 445 . That that is a rule of practice and not a rule of law may be readily seen by taking the extreme case of a conviction based on the uncorroborated evidence of an accomplice, which conviction is not bad in law.
4. An automatic test for the credibility of witnesses is not one that can be stated or ought to be stated in statutory terms. The test cannot be mechanical but emerges for each case from the circumstances and probabilities of that case and is the net result of the effect of these on the mind of the Judge trying the case. In cases where witnesses have retracted in the Sessions Court their statements before the committing Court, the proper test is 'are there reasonable grounds, grounds sufficient to satisfy a judicial mind, for holding that the former statements are false and the latter are true?' Such grounds might be corroboration by other witnesses but need not necessarily be so. Each case must rest on its own evidence.
5. In my view, however, such grounds must be found by the trying Court and put forward by it, as the lower Court has done, as a reason for the admission against the accused of former statements under Section 288, Criminal Procedure Code, that is, unless the Sessions Court is satisfied that there are judicial grounds for holding that the retraction in the Sessions Court is false testimony and that the statements in the committing Court are true testimony, it ought not to admit on behalf of the prosecution these latter statements under Section 288 at all. And here I would follow the principle laid down in Queen-Empress v. Jadub Das I.L.R. (1900) Calc. 295 in so far as it lays down that such statements are not to be mechanically placed on the record without some sufficient grounds for admission as I have indicated. The Court must be satisfied first that 'there is something to show the truth of the former statement.'
6. The cases which usually arise are where a few witnesses who speak in the committing Court to having seen the commission of a crime deny in the Sessions Court their whole story and allege that they did not see the commission of the crime at all; and in most of such cases it would be a sound conclusion that the evidence in neither Court is reliable, unless there is some outside corroboration of the witnesses in the committing Court. But the present cases are very different from these. Here is a large body of evidence in the Sessions Court to support the main story of the fight as given in the committing Court up to the point of the identification of the assailants. Then the witnesses retract, being themselves accused in the counter-case and expecting the benefit of a similar retraction of the eye-witnesses in the case in which they themselves are accused; and, as I have said, there is no other reasonable explanation for their retraction on that sole point than that they came to a compact with the other side that the other side would similarly exonerate them in the Sessions Court. It is impossible to resist the conclusion that their retraction on the one point of identification in the Sessions Court is a perjury designed to defeat justice and that the true evidence is that which they gave in the committing Court. In these cases, therefore, I find no difficulty in agreeing with the lower Court as to which are the reliable depositions; and on its conclusion on that point which was a judicial one, that Court was perfectly competent to act. I agree, therefore, that the learned Sessions Judge was amply justified in the circumstances of these cases in accepting as true and using the evidence of identification of assailants given in the committing Court by the eye-witnesses, and cannot accept the argument put forward before us that the conviction based on that evidence is bad or unjustifiable in law.