1. The appellants in these two appeals were tried "before a Special Magistrate of Barakpore on charges under Section 19-A of the Arms Act as amended by-Bengal Act XXI of 1932 and Section 120-B of the Indian Penal Code. They have all been convicted under those sections and the appellant Aswini Kumar Ghose has also been convicted under Section 19-A of the Arms Act as amended by Bengal Act XXI of 1932. Bimal Krishna Biswas has been sentenced to undergo rigorous imprisonment for 7 years and each of the other two appellants has been sentenced to undergo rigorous imprisonment for 5 years. Besides these appellants four other persons were put on trial. Out of these three, namely Kalidas, Ghose, Lakshman Chandra Adhikary and Panehanan Samanta pleaded guilty and were convicted on that plea and each of them was sentenced to undergo rigorous imprisonment for 6 months. They hive been examined as witnesses for the prosecution. Kalidas, prosecution witness No. 37, Lakshman prosecution witness No. 38 and Panchanan, prosecution witness No. 39. Another accused Pankaj Kumar Mitra was granted pardon and examined as a witness, prosecution witness No. 34.

2. The case for the prosecution is as follows:--On February 18, 193-1, Bimal was arrested under Section 34 of the Criminal Procedure Cede. He was released the next day but was re-arrested. On March 19, as the result of certain information received, the Police searched the house of Kalidas and arrested him. He made a confession which was recorded on March 23. On the same day the house of one Bimala Bala Debi was searched and three revolvers which are marked as Exs. IV, V and VI, in the case were recovered. Thereupon a First Information Report, Exhibit 3, was drawn up. On March 20, there was a search in the houses of Lakshman, Panchanan and Gopi Mohan Daw but nothing incriminating was found. Gopi was arrested on March 21, and the other two on March 22. Lakshman and Panchanan made confessions which were recorded on March 22. On March 21, there was a' search in the house of the appellant Aswini. In a flower tub in the garden attached to his house were found two revolvers and a pistol which are marked as Exs. 7, 8, and 9,in the case. He was arrested and another First. Information Report Ex. 40 was drawn up. These two First Information Reports were amalgamated and the case was taken up as one case. On March 24, the library of the Bayam Samiti was searched and some articles including two proscribed books were found. On March 31, a tank close to the house of Aswini was searched and certain articles including some photographs of persons convicted of terrorist offences were found. On April 3, the approver Pankaj was arrested and he made a confession which was recorded on the following day. After further investigation charge-sheet was submitted on May 14, and the accused persons were put on their trial on June 11. On June 18, Pankaj was granted pardon and on the same day charges were drawn up against the remaining six out of whom the three persons named aboved pleaded guilty. No orders were pissed on that day, but on June 28, the three confessing persons were sentenced and on the same day were examined as witnesses.

3. The prosecution case rests on the find of two sets of fire-arms, one on March 19, 1931, in the house of Bimala Debi and the other on March 21, 1934, in the garden of Aswini. As regards the alleged conspiracy the charge runs as follows:

That you between January 1934 and March 1934 at Baranagore and Alambazar P.S. Baranagore, were party to a criminal conspiracy with others known and unknown, to possess fire-arms, viz., revolvers, pistols and cartridges in furtherance of terroristic movements, and thereby committed an offence punishable under Section 120-B of the Indian Penal Code read with Section 19-A of the Indian Arms Act (Act XI of 1878) as amended in Bengal Act, Act XXI, of 1932 and within my cognizance.

4. It has been pointed out that, in so far as the alleged offence under the Arms, Act is concerned, it was incumbent on the, prosecution to prove, that the fire-arms; were possessed in contravention of Sections 14' and 15 of the Act, in other words, that: they were possessed without license or other legal authority. As regards the weapons marked Rs. 7 and 8, there is the evidence of T. W. Hart, the Arms Expert, prosecution witness No. 16 brought out in answer to questions in cross-examination to the effect that the ownership of these weapons cannot be traced and the learned Magistrate therefore considers that they have been smuggled. But as regards the other set of fire-arms 5, there is no evidence formal or otherwise to show that they had been held without license. No doubt the accused do not claim them and the circumstances would also indicate that they had not been found, in the lawful possession of anybody but I only mention it in order to show that in a case of this nature, having regard to the charge, it is incumbent on the prosecution to lead some evidence which would, justify an inference that the possession was against the provisions of the arms act.

5. As is staled in the charge the prosecution case is that these fire-arms have been possessed in furtherance of terrorist movements. That in itself is not an offence under the Arms Act. It is necessary to say this, because the judgment of the learned Magistrate shows that there has been some confusion in his mind as to what the prosecution has got to prove. It may be that there is a conspiracy to commit terrorist crimes and it may be that there is a conspiracy to possess fire-arms in contravention of the Arms Act. But the two things are not the same. The prosecution has led some evidence to show that the accused in these cases are members of an organisation, the object of which is to commit terroristic offences. If this is proved, it would not follow, in the absence of other evidence, that the accused were also parties to a criminal conspiracy for the specific and definite purpose of possessing fire-arms in contravention of the Arms Act. Mr. Khundkar for the Crown has contended that the evidence as to terrorist organisation is relevant for the purpose of showing that it may lead to an inference that, if some members of the Organisation were in possession of firearms, the others were also in a conspiracy to possess fire-arms. Whether that is so or not would depend upon the nature of the evidence indicating the connection between the two different kinds of conspiracy. But, nevertheless, there are different conspiracies and it is a fallacy, into which the learned Magistrate seems to have fallen to suppose that merely because one was a member of the larger organisation, one was also a member of the smaller organisation although the immediate objects of the two were not one and the same.

6. In this case the prosecution evidence is as follows: There is the evidence of the approver and there is the evidence of the three accomplices. This evidence is specific evidence regarding the possession of the fire-arms. Then there is a certain amount of general evidence for the purpose of showing that there was an organisation which was at one time called Yuboksangha and subsequently called Bayam Samiti and some of the accused were members of this organisation. A. good, deal of this evidence is of a colourless character and indicates nothing more than that it is apparently a harmless club for the propagation of games and study. There is only this much that as the result of the Police search two proscribed books among other articles were found. But that does not carry us very far, so far as the present case is concerned. There is also the evidence of some Police Officers, particularly of Amarendra Kumar Sen prosecution witness No. 35 which is largely of a hearsay character. This evidence, to my mind, is not sufficient to establish that the accused were members of an organisation to commit terrorist crimes, far less would it establish that the accused were parties to the conspiracy which is mentioned, in the charge. Then there is the evidence of the approver and the accomplices. With regard to this it is important to remember the circumstances under which these persons were apprehended and brought to Court. Lakshman and Panchanan were arrested on March 20, Kalidas was arrested on March 19. Before they were brought to the Magistrate they were all kept in custody at Baranagore Thana. Then they were produced before a Magistrate and their confessions were recorded. It is also worthy of note that although they pleaded guilty on June 18, they were not convicted until June 28, and the sentences passed on them were disproportionately lenient, in spite of certain reasons recorded by the Magistrate and in view of the sentences which he thought fit to pass on the non-confessing accused. Immediate after their conviction the three prisoners were examined as witnesses. These circumstances lend support to the argument that there was some sort of arrangement lies (by?) which these three prisoners were made available as witnesses for the prosecution against these appellants. Now whether we consider the evidence of these three witnesses or the evidence of the approver, it is important to remember that they are all subject to the well known rule that it is unsafe to rely on the testimony of an accomplice without independent corroboration both as to the crime and as to the identity of the criminal. Mr. Khundkar has drawn our attention to a number of cases in which he has contended that this rule has not been followed: Aung Hla v. Emperor 9 R. 404 : 135 Ind. Cas. 849 : A.I.R. 1931 Rang. 235 : (1931) Cr. Cas. 875 : 33 Cr. L.J. 205 : Ind. Rul. (1932) Rang. 65, Naga Nyein v. Emperor 11 R. 4 : 142 Ind. Cas. 87 : Ind. Rul. (1933) Rang. 29 : 34 Cr. L.J. 286 : A.I.R. 1933 Rang. 57 : (1933) Cr. Cas. 452. Rattan Dhanuk v. Emperor 8 Pat. 235 : 113 Ind. Cas. 329 : 9 P.L.T. 672 : A.I.R. 1928 Pat. 630 : 30 Cr. L.J. 137. He has also referred to some unreported decisions of this Court, viz, Criminal Appeal No. 421 of 1933 Sarat Chandra Dhupi v. Emperor , dated January 9, 1934, and Death Sentence Reference No. 9 with Criminal Appeals Nos. 227 to 232 of 1934, Nirmal jeeban Ghosh v Emperor . I do not understand these cases to lay down anything more than this that it is not illegal to base a conviction on the uncorrborated testimony of an accomplice. It is also not illegal to base a conviction upon such testimony if it is corroborated by other accomplice evidence. But that is not tantamount to saying that independent corroboration is not necessary, or that the corroboration of one tainted evidence by another tainted evidence is independent corroboration. The aforesaid argument that it is not illegal does not take us very far; it is not illegal to do a great many other things, for instance it is not illegal to belive what is improbable, it is not illegal to impose the maximum penalty prescribed by law, and so forth. But merely because it is not illegal is not a reason for doing any of these things; in every case, the question is, what is just and proper. As has been pointed out by Rankin, C.J. in a judgment to which I was a party See Ambika Charan Ray v. Emperor 35 C.W.N. 1270 : 134 Ind. Cas. 1121 : A.I.R. 1931 Cal. 697 : 33 Cr. L.J. 19 : Ind. Rul. (1932) Cal. 33 : (1931) Cr. Cas. 977, we are here not merely to record a conviction that is not illegal, but we are here to record a conviction that is properly based. Now the rule of law requiring independent corroboration is a rule of caution which applies to all accomplice evidence. In England it is based on principles derived from the judicial experience of ages. In India it is expressly confirmed by Section 114 of the Evidence Act. The qualification to illustration (b) under Section 114 only points out that in a particular case there may be circumstances, for instance, circumstances showing previous concert among accomplices to be highly improbable, which may lessen the degree of independent corroboration required. But these circumstances themselves have to be proved by independent evidence and it does not make the general rule (if caution inapplicable, for without corroboration the risk remains and that is what the Court has to remember. Mr. Khundkar has also contended that an accomplice after conviction stands on a higher level than an approver: Queen-Empress v. Hussain Haji 25 B. 422, and Mohammad Yusaf v. Emperor 58 C. 1214 : 131 Ind. Cas. 142 : 32 Cr. L.J. 375 : A.I.R. 1931 Cal. 311 : 35 C.W.N. 490 : (1931) Cr. Cas. 405. That also is a circumstance which may not be left out of account, but the fact remains that an accomplice is an accomplice and more or less, having regard to the circumstances of each case, he ought to be corroborated by other evidence. Now, in the present case, as I. have pointed out, the conditions under which the accomplices were brought to Court as witnesses after their conviction are not free from suspicion, and I do not think that it will be at all safe to accept the evidence of these three persons and of the approver as corroborating one another to such an extent as to do away with the rule of caution and justify us in accepting their evidence without material corroboration from other and more independent witnesses. (The judgment then discussed the evidence and concluding that in tie case of Aswini the evidence of the accomplices was sufficiently corroborated by independent evidence but in the case of Bimal and Gopi, the evidence was not sufficient ,to justify their convictions and proceeded.) The result is that the appeal of Aswini is dismissed. The appeals of Bimal and of Gopi are allowed and each of these two appellants is acquitted of the offences under Section 120-B of the Indian Penal Code and 19-A of the Arms Act and they are directed to be set at liberty.

Henderson, J.-

7. The appellants were convicted of conspiracy to possess arms in contravention of the provisions of the Indian Aims Act. Although the prosecution adduced a great deal of evidence which, in my judgment, was entirely useless, they did not take care to see that the necessary formal evidence to show that the provisions of the Indian Arms Act were contravened was before the Court. If it had not been for the fact that a Police Officer in the course of his cross-examination made a statement from which it can be inferred that the arms recovered from the house of Aswini were not covered by any license, it would have been very difficult for us to avoid sending the case back and ordering the necessary evidence to be recorded, however harassing that course might have been to the appellants themselves.

8. The judgment of the learned Magistrate has not been of as much assistance to us as is desirable. He has really done little more than to make an abstract of what the prosecution evidence is. He then finds that it suggests that the accused and other persons were members of a conspiracy to commit acts of terrorism. Now, that is quite a reasonable view to take and indeed it is difficult to suppose that any persons would conspire to possess arms merely for the pleasure of contravening the provisions of the arms act and that they would not do so unless they were members of another conspiracy, the object of which could not be attained without the possession of firearms. Now it is quite obvious that it by no means follows that the members of the larger conspiracy must necessarily be considered in the subsidiary activities connected with the collection of arms. Mr. Khundkar, however, has explained to us that the prosecution case is that all the accused persons were, in fact connected with the subsidiary conspiracy. Unfortunately the learned Magistrate did not approach the case from that point of view at all and it is difficult to say whether he really brought his independent judgment to bear on the various matters which were placed before him. He says nothing about the liability or otherwise of the accomplices and he does not at all say what evidence of corroboration he finds with respect to the individual accused and he does not really discuss the evidence with regard to their credibility and so on. The result is that the case had to be argued before us at greater length than would otherwise have been necessary.

9. Before we can uphold the convictions of the appellants, we must be satisfied that they are proved to be members, not of a conspiracy to commit acts of terrorism, but of a conspiracy to possess arms in contravention of the provisions of the Indian Arms Act. The principal witnesses on whom the prosecution rely are the approver and the three accused persons who pleaded guilty. In this connection Mr. Khundkar raised the question whether the evidence of an accomplice could be corroborated by that of another accomplice. On this point I desire to say that I adhere to what was stated in the judgment delivered in the appeals arising out of the conspiracy to murder Mr. Burge to which I was a party [Death Reference No. 9 of 1934 and Criminal Appeals Nos. 227 to 233 of 1931 in Nirmal Jiban Ghose v. Emperor . A rule of law is not the same thing as a rule of prudence. The rule of law is contained in Section 114 of the Evidence Act which lays down that the Court may presume that an accomplice is unworthy of credit unless he is corroborated in material particulars. To say that an accomplice is to be corroborated in material particulars is not the same thing as to say that he is to be corroborated by a witness who is not an accomplice. No doubt it is a rule of prudence to say that ordinarily evidence which is itself tainted should not be accepted as corroboration of tainted evidence. But in my judgment it is opposed to all commonsense to lay down that in a case, the circumstances of which show that the rule of prudence does not apply, the Court is precluded from acting en evidence which it believes to be true. I cannot better illustrate this point than by quoting from the judgment of Sir Arthur Page in the case of Aung Hla v. Emperor 9 R. 404 : 135 Ind. Cas. 849 : A.I.R. 1931 Rang. 235 : (1931) Cr. Cas. 875 : 33 Cr. L.J. 205 : Ind. Rul. (1932) Rang. 65. The passage to which I refer begins at the bottom of p. 429 Page of 6 R[Ed.], and is in these terms:

War is waged and a battle fought against the forces of the Crown by certain rebels. The question is whether A. B. was a rebel who tock part in the battle. Suppose the only evidence against A. B. is that of two approvers, each of whom testify that A. B. was one of those who fought in the battle, and it is proved that both of these approvers are on bad terms with A. B. Obviously, the fact that unreliable and discredited testimony is given by two witnesses instead of one will not render the evidence of either witness the more worthy of credit. But suppose there are 30 rebels who have made statements whether as approvers or by way of confessions that have been made voluntarily, who come from different villages, and with respect to whose statements there is no reason to think that there was any colloboration or callusion between the parties making them; none of them, so far as it is impossible to judge, being actuated by malice towards A. B., and suppose that each of these 30 rebels should happen in his statement to implicate A. B. stating that he also took part in the battle and was a party to the rebellion, could it be suggested that the cumulative effect of so many statements, apparently independent and impartial, all implicating A. B. could not reasonably lead to the conclusion that A. B. was a rebel and took part in the battle? We think clearly not.

10. Applying this principle to the facts of the present case I am not prepared to say that the evidence of these four witnesses is of such a character as can safely be relied on as mutually corroborative. In the first place we have no guarantee that they were not acting in concert or that they had no opportunities of consultation. Then again the circumstances in which the three who pleaded guilty, were convicted on their plea and sentenced on June 28, suggest there might have been a bargain between them and the prosecution to the effect that they might be dealt with rightly if they agreed to give evidence. It is quite clear that before they were convicted and sentenced the prosecution knew that they were willing to depose and the sentences inflicted are not only inadequate but out of all proportion to the sentence inflicted upon the appellant Gopi whose case can hardly be distinguished from theirs. My learned brother has dealt with their evidence in detail and I am satisfied that it would not be safe to rely upon it in the absence of material corroboration.

11. On this aspect of the case I need only say that so far as Bimal and Gopi are concerned, I find no material corroboration at all, whereas in the case of Aswini it is conclusive. I agree that the appeal of Aswini should be dismissed and the appeals of Gopi and Bimai should be allowed.