M.L. Jain, J.
1. The facts of this appeal are that on 7-11-1961 at about 1.30 in the night 12-13 persons came to village Lakhan pur within police station, Nadbai to commit planned dacoity at the house of Poorna, Sobhararn, Motiram, and Girdhari. They found three persons sleeping in the verandah of the house of Moti Ram and therefore, three of the dacoits stayed in the verandah to prevent sleeping persons from possible intervention, One or two of the dacoits reached the roof of the house and opened the main door from inside. The remaining dacoits entered the house and looted silver and gold ornaments and clothes from the houses of the said persons. Sobhararn went to Subedar Charanjit Singh in village Daya-wali for help. Charanjit Singh and nine ten persons came to the scene of occurrence and fired their guns. The dacoits were scared and fled away with their booty.
2. It was alleged that the dacoits had first assembled in Kiraoli and had from there started on bicycles. One Jhamman happened to meet them near Uchhain and on inquiry the accused told him that they were going to meet their relations. Upon a report of the occurrence having been lodged by Bhogi Ram at the police station, Nadbai, a case was registered and investigation was taken up mainly by S. H, O. Chatursal P. W. 14. The police recovered some empty cartridges and pellets from the scene of occurrence. The victims gave the police lists of the stolen articles. The police arrested the accused. On 22-12-1961 the houses of accused Data Ram and Salig Ram were searched in the presence of Bhogi Ram and Girdhari. Some looted articles were recovered from their houses, vide recovery memos Ex. P. 10 and Ex. P. 11. Upon his information and at his instance, one country made pistol was recovered from the possession of accused Daulta, vide recovery memo Ex. P. 2 dated 2-10-1962. Accused Ram Sanehi produced a gas lantern 'Prabhat' No. 426 on 3-1-1962, vide Ex. P 13 and later on became an approver. The police also searched the house of accused Bahori and Ahiram on 4-1-62 and 16-1-1962 respectively. In the presence of their wives, some looted articles were recovered vide Ex. P, .15 and Ex. P. 16. Upon the information given by Ramji Lal, one pair of silver wristlets was recovered from his box, vide recovery memo Ex. P. 19 of 12-3-1962.
3. After test identification of the articles and the accused, the police put up a challan against 10 persons, Data Ram and others. One of the accused since then died in an encounter and one is alleged to have absconded. Therefore, only 8 persons in all were tried.
4. The learned trial judge examined 19 prosecution witnesses. Pyarelal was examined Under Section 540 Cr. P. C. He also examined 6 defence witnesses. After conclusion of the vidence, the learned trial judge held that a dacoity did take place at the houses of Poorna Ram, Sobha Ram, Moti Ram etc. in village Lakhanpur on the night intervening 6th and 7th November, 1961. According to approver Ram Sanehi P, W. 8 all the 8 persons were members of the dacoity party. Some 8 days before Diwali, Data Ram and Biri Singh accused came and told him that the house, of a Bania of Lakhan pur deserved to be looted. He was told that Daulta and Ram Singh will also join them. Prior to Diwali all of them proceeded on bicycles to meet at the intersection of Kiraoli. Data Ram, Ahiram ; and Daulta were armed with countrymade pistols and Ramjilal with a double-barrelled gun, Data Ram, Saligram, and Ham Sanehi went as an advance party via Bharatpur. They again assembled at Uchhain where Jhaman P. W. 15 happened to meet and ask where they were about. At Pingora Railway Station, Saligram brought the informer who came armed with a sword. On their way, they cut some branches of a tree for using them in the dacoity, The informer told them that they were to loot the haveli which had a tinshed verandah. Ramjilal stayed outside with his gun. Three of the dacoits remained in the verandah. One of the persons sleeping there was awakened and questioned them who they were., Data Ram gave a cane on his head and made him keep quiet. Saligram and Vasudeo climbed the tinshed and went inside the house. When the doors were unlocked, Saligram, Birisingh, Vasudeo, Jaishiv and Ram Singh went inside, while Dataram and Ahi Ram remained under the tin-shed. Daulta planted himself in the way. Armed dacoits fired three or tour shots. Ram Sanehi also went inside the house and brought for himself a gas lantern. When fire was heard from the side of village, the dacoit party retreated and again assembled with their booty at the place where they had left their bicycles. Salig Ram, Behori and Vasudeo had the clothes, while the ornaments were with Biri Singh, Bahori, Ahi Ram and Data Ram. Bhaori also had a gas lantern. As regards a division of the looted property, it was decided that they would meet again on the 'Pulia kriasli' next day. Biri paid Ram Sanehi Rs, 120/- and he was given no share of the clothes and ornaments. They were told that each one shall retain what he had with him.
5. The learned trial judge observed that although the approver is a competent witness against the other accused but before his evidence is accepted, he must satisfy a double test of reliability and sufficient cor-roboration qua each accused. The learned trial judge then proceeded to find whether the accused satisfied the said two tests. He held that the approver's evidence was not wholly reliable. He however, found that his evidence regarding the occurrence is corroborated by the testimony of the other prosecution witnesses. The learned judge then proceeded to examine the merits of the corroborative value of the test identification. In the test identification accused Ram Singh, Jaishiv and Bahori were identified, though they have been telling the magistrate that they were shown to the witnesses by Chatursal S. H. O. Data Ram was identified in the second round by witness Girdhari Lal. Salig Ram was identified by Bhogi Ram and Harprasad. Ahi Ram and Biri were not identified at all. Ramji Lal was identified by Bhogi, Girdhari and Har Prasad. Ramji Lal has alleged that he was shown to the witnesses prior to identification. These complaints of the accused were recorded by the magistrate in the column No. 10 in the respective identification tnemos.
6. The night was dark. There was no light with the prosecution witnesses but they claimed to have identified the accused persons in the light of the torches flashed by the accused. In the F. I. R. Ex. P. 1 there was no mention of the torches. Ram Sanehi P. W. 8 has stated that none of them had torches. The witnesses were in fright. At the time of occurrence, Har Prasad P. W. 7 was a only lad of 11 years and was standing 8-10 yards away from the tinshed. Bhogi Ram P. W, 1 was covered with the quilt throughout, and therefore, he could not see the accused. Ramji Lal P. W. 3 was sleeping in his Pator from where it was not possible for him to identify the accused. Same was the case with Har Prasad. Gird hari Lal P. W. 10 was lying on the cot. The learned trial Judge therefore, held that the aforesaid evidence regarding identification of the accused persons by the prosecution witnesses is not reliable and cannot be used to corroborate the testimony of the approver regarding the part of the particular accused in the occurrence.
7. The learned trial Judge further found that the case against Vasudeo was doubt ful because he was not sent for test identification, nor was any recovery made from him.
8. As regards the case of accused Jaishiv, besides the identification being useless, no recovery was made from him. Ram Sanehi even admitted that he included the name of Jaishiv upon the suggestion of S. H, O. Chatursal.
9. Regarding Ramsingh also the learned trial judge found that there was no corrobo- ration.
10. The case against Ramjilal was also rejected by the learned trial Judge because he thought that S. H. O. Chatursal P. W. 14 may have suggested the name of Ramji Lal to approver Ram Sanehi in order to oblige Ram Vijay Singh the S. H. O. Kotwali Bharatpur. Ramjilal was suspected in the murder of the father of Ram Vijay Singh. Chatursal and Ram Vijay Singh had taken training together and another brother of Ram Vijay Singh is also in the police force.Chatursal Singh wanted to oblige Ram Vijay Singh and that is how Ramjilal's name was included.
11. The case against accused Biri Singh was also found by the learned trial Judge as unreliable.
12. As regards Ahi Ram the learned Judge observed that Chatursal Singh admitted that his name was not given out by Ram Sanehi when he examined him on 20-11-1961. Shri Chatursal Singh has taken search of his house in the presence of his wife and in that search three articles one Malmal turban and two ladies' Sarees were recovered. These articles were identified by Sobharam P. W. 2 and Girdhari Lal P. W. 10. Mst. Rami D, W. w/o Ahi Ram also claimed that these articles belonged to her. But her evidence was rejected. Evidence of the identification of stolen property was considered sufficient corroboration and Ahi Ram was held to have taken part in the dacoity,
13. As regards the case against Data Ram some articles were recovered from his possession upon a search of his house, vide memo Ex. P. 10 in the presence of Bhogi Ram P. W. 1 and Girdhari Singh P.W.10. Both of them identified these articles, Sobha Ram P. W. 2 has also identified some of the articles.
14. Mst. Baikunthi P. W. 6 wife of Data Ram claimed some of these articles belonging to her but she was unable to give any special mark of identification of these articles. She identified the articles which were not recovered from Data Ram but from Salig Ram, Her testimony was considered unreliable. The learned Judge held that the circumstances showed that this accused had taken part in the occurrence.
15. Regarding Saligram some articles were recovered from his house, vide Ex P. 11. These articles were identified by Bhogi Ram P. W, 1 and Sobharam P. W. 2. These articles were mentioned in the list of the stolen articles. Wife of Salig Ram Mst. Bhagwati D. W. 4 claimed some articles as belonging to her but the learned trial Judge rejected her testimony. The learned Judge therefore, held that the identification of the articles made in the court lent enough support to the testimony of Ram Sanehi.
16. The result was that the learned Judge acquitted Birisingh, Vasudeo, Jaishiv, Ramjilal and Ram Singh. He convicted however, Data Ram, Ahi Ram and Saligram Under Sections 395 I. P. C. and 397, I. P. C. and sentenced them each to seven years' rigorous imprisonment and a fine of Rs. 200/- in default whereof to two months' rigorous imprisonment. Hence this appeal by these three persons.
17. Now, the first question that falls for consideration is regarding the value of evidence of approver Ram Sanehi. The learned Counsel for the appellants contends that he is an unreliable witness and has also been so held by the learned trial Judge in respect of some accused persons. Five persons were acquitted on his very testimony. He has therefore, not passed the first test as laid down by the Supreme Court that the approver should first of all be a reliable witness. The recoveries of the articles do not provide corroboration as they were new and rather of common use and therefore unidentifiable. They were not recovered from the sole possession of the accvised appellants and therefore, no presumption Under Section 114, Indian Evidence Act can be drawn against the appellants It was pointed out that Ram Sanehi was not even an accomplice because his statement is exculpatory. There was no corroboration in material particulars and therefore he has not passed the second test as well. The learned trial Judge should have rejected his evidence.
18. In respect of the evidence of an approver the law as stated in King v. Basker ville, (1916) 2 KB 658 is that where on the trial of an accused, evidence is given against him by an accomplice, the corroboration which the common law requires, is corroboration in some material particulars tending to show that the accused committed the crime charged, It is not enough that corroboration shows the witness to have told the truth in matters not connected with the guilt of the accused. In Davies v. Director of Public Prosecutions, (1954) 1 All ER 507 it was observed that in a criminal trial where a person who was an accomplice gave evidence on behalf of the prosecution, although the jury might convict him on his evidence, it was dangerous to do so unless his evidence is corroborated ; this rule although a rule of practice now had the force of a rule of law.
19. In Srinivas Mall v. Emperor AIR 1947 PC 135 the Privy Council held that no doubt the evidence of accomplices ought as a rule to be regarded with suspicion. The degree of suspicion which will attach to it, must however, vary according to the extent and nature of the complicity ; sometimes the accomplice is not a willing participant in the offence but a victim of it. When the accomplices act under a form of pressure which it would require some firmness to resist, reliance can be placed on their un-corroborated evidence. In Bhuboni Sahu v. King AIR 1949 PC 257 :50 Cri LJ 872 the Privy Council observed ;
The courts should be slow to depart from the rule of prudence, based on long experience, which requires some independent evidence implicating the particular accused. The danger of acting upon accomplice evidence is not merely that the accomplice is on his own admission a man of bad character who took part in the offence and afterwards to save himself betrayed his former associates and who has placed himself in a position in which he can hardly fail to have a strong bias in favour of the prosecution ; the real danger is that he is telling a story which in its general outline is true ; and it is easy for him to work into the story matter which is untrue.
20. But in Sarwan Singh v. State of Punjab : 1957CriLJ1014 the Supreme Court laid down that an approver has to satisfy a double test. His evidence must first show that he is a reliable witness and that is a test which is common to all witnesses. If this test is satisfied, the second test which still remains to be applied is that the approver's evidence must receive sufficient corroboration. This test is speccial to the case of weak or tainted evidence like that of the approver. This rule of double test does not however, appear to have been followed in Bhiva v. State of Maharashtra : 3SCR830 , wherein it was stated that the combined effect of Sections 133 and 114 illustration (b) of the Indian Evidence Act was that though the conviction of an accused on the testimony of an accomplice cannot be said to be illegal, yet the courts will, as a matter of practice, not accept the evidence of such a witness without corroboration in material particulars. In Major E. G. Barsay v. State of Bombay : 1961CriLJ828 , the court directly referred to Sarwan Singh's case and observed that the court in that decision could not have intended to lay down that the evidence of an approver and the corroborating piece of evidence should be treated in different compartments i.e. to say the court shall have first considered the evidence of the approver de hors corroborating piece of evidence and reject it if it comes to the conclusion that his evidence is unreliable but if it comes to the conclusion that his evidence is reliable, then, it will have to consider whether that evidence is corroborated by any other , evidence. This Court did not lay down any such proposition. In most of the cases the said two aspects would be so inter-connected that it would not be possible to give a separate treatment, for as often as not, the reliability of an approver's evidence though not exclusively, would mostly depend upon the corroborative support it derives from other unimpeachable pieces of evidence.
21. In State of Andhra Pradesh v. Ganeshwara Rao : 3SCR297 the view explained in E.G. Barsay v. State of Bombay, (1961(2) Cri LJ 828) (SG) was reiterated. Whether the evidence of the approver should in any given case, be accepted or not will have to be determined by applying the usual tests, such as the probability of the truth of what he has deposed to, the circumstances in which he has come to give evidence, whether he has made a full and complete disclosure, whether his evidence is merely self exculpatory and so on and so forth, and in addition, whether his evidence has been corroborated sufficiently in material particulars. What is necessary to consider is whether applying all these tests the evidence of the approver should be acted upon.
22. In K. K. Jadav v. State of Gujarat : 1966CriLJ605 the rule was more succinctly stated when it was observed that what the law required in the case of an accomplice is that there should be such corroboration of the material parts of the story connecting the accused with the crime as will satisfy the reasonable minds that the approver can be regarded as a truthful witness. The corroboration need not be direct. If it is merely circumstantial, it will be sufficient. The nature of corroboration will depend on and vary with the circumstances of each case.
23. In Sarvanabhavanv.State of Madras : 1966CriLJ949 Wanchoo J., combined the views both of Sarwan Singh's case (1957 Cri LJ 1014(SC)) and Barsay's case (1961(2) Cri LJ 828) (SC) and observed that the evidence of an approver cannot be dealt into two water tight compartments. It must be considered as a whole along with other evidence. Even so the court has to consider whether the approver's evidence is credible in itself and in doing so it may refer to such corroborative piece of evidence, as may be available. But there may be cases where the evidence of the approver is so thoroughly discrepant and so inherently incredible that the court might consider him wholly unreliable.
24. Even in Haroom Haji v. Maharashtra : 2SCR641 it was observed that before the courts act on accomplice evidence, they must insist on corroboration in material respects as to the offence itself and also implicating, in some satisfactory way however small, each accused named by the accomplice. In spite the Supreme Court having reaffirmed the double test approach in Piara Singh v. State of Punjab : 1969CriLJ1435 , in other cases, most of them later in time, the double test as stated in Swaran Singh's case (1957 Cri LJ 1014(SC) appears in a somewhat modified form explaining mostly negatively, when a witness can be considered to have passed the first test of reliability. In Lachi Ram v. State of Punjab : 1967CriLJ671 , it was said that the first test obviously means that the court should find that there is nothing inherent or improbable in the evidence given by the approver, and that there is no finding that the approver has given false evidence. In Madan Mohan v. State of Punjab : 1970CriLJ898 it was observed that in considering whether the approver's evidence passed the test of reliability, the Court would have to consider whether taken as a whole and in the light of the facts and circumstances of the case it was a credible version or not. In Sheshanna Bhumana Yadav v. State of Maharashtra : 1970CriLJ1158 , it was observed that the first test of reliability of approver and accomplice evidence is for the court to be satisfied that there is nothing .inherently impossible in evidence. After that conclusion is reached as to reliablity, corroboration is required to connect the accused with the crime. According to Ram Narain v. State of Rajasthan : 1973CriLJ914 , if the approvers evidence is found uninspiring and unacceptable, then corroboration would be futile, and unnecessary. In Tribhuvan Nath v. State of Maharashtra AIR 1973 SC 450 : 1972 Cri LJ 1277 while applying, the double test it was observed that the corroboration must be adequate enough to afford the necessary assurance that the main story testified by the accomplice can be reasonable and safely be accepted as true.
25. In Ravindra Singh v. State of Haryana : 3SCR453 , the law was stated as follows :
In a rare case, taking into consideration all the factors, circumstances and situations governing a particular case, conviction based on the uncorroborated evidence of an approver confidently held to be true and reliable by the court, may be permissible. Ordinarily, however, an approver's statement has to be corroborated in material particulars bridging closely the distance between the crime and o the criminal. Certain clinching features of involvement disclosed by an approver appertaining directly to an accused if reliable by the touch-stone of other independent credible evidence, would give the needed assurance for acceptance of his testimony on which a conviction may be based.
26. In Hussain Umar v. Dalip Singhji : 1970CriLJ9 it was held that the witness concerned may not confess to his participating in the crime but it is for the court to decide on a consideration of the entire evidence whether he is an accomplice. Though a conviction based upon the accomplice's statement is legal, the court will not accept such evidence unless it is corroborated in material particulars. The corroboration must connect the accused with the crime.
27. Upon a survey of the aforesaid decisions of the Supreme Court, one will find that the double test pronounced in Swaran Singh's case (1957 Cri LJ 1014) (SC) cannot be applied in a manner as if the evidence of an approver has to be judged into two separate compartments. No one can perhaps give a better account of the occurrence than a person particeps criminis. But an approver is generally not a man in penitence, he is rather a deserter in hope of pardon. He is likely to weave into his story matter which is untrue. That is why the courts do generally start with a suspicion and look for confirmation. If he is found to be unreliable, he must be rejected then and there, corroboration notwithstanding. If he is reliable or for that matter wholly reliable, which will be rare, he can be believed with or without any corroboration. It is only when the court is not able at once to say whether he is wholly reliable that the court shall in order to rely upon him, demand corroboration direct or circumstantial in material particulars in respect of the part played by the accused in question. The guide lines that emerge in respect of the approver's evidence, appear to be (1) according to Section 133 of the Indian Evidence Act, an approver is a competent witness and conviction shall not be illegal if based upon an uncorroborated testimony of the approver ; (2) according to the illustration (b) of Section 114 of the Indian Evidence Act, the court may presume that an approver is a man of untrustworthiness until ho is corroborated in material particulars ; (3) court should not ordinarily convict unless the evidence of the approver is corroborated in material particulars qua the accused ; and if more than one, qua each accused ; (4) such corroboration can be direct and circumstantial ; (5) if the evidence of an approver is intrinsically or inherently impossible or is otherwise unreliable or unacceptable, it should be rejected straightway without caring to seek for corroboration ; (6) there are several circumstances by which the reliability of an approver has to be adjudged ; one important test of reliability is that he is corroborated by other evidence in material particulars ; and (7) the approver should be a particeps criminis on his own admission or appear to be so by evidence.
28. Applying these tests to the instant case we have to see whether upon the totality of his evidence and in the light of the corroboration provided, Ram Sanehi can be considered a reliable witness so as to justify the conviction of all or any of the appellants. The learned Counsel for the appellants contends that Ram Sanehi is a totally false witness. It will be noticed that out of the 8 accused persons, in case of a majority of 5, the story of the approver was considered unreliable by the learned trial judge. It will further be noticed that in case of two accused persons, namely, Ramji-lal and Jaishiv the approver included their names upon the suggestion of Charursal S. H. O. In case of accused Data Ram, the learned trial judge found that the recoveries of certain stolen articles from his house provided corroboration to the testimony of the approver. But these articles were not recovered upon the information and at the instance of the accused. His house was Searched in the presence of the two prosecution witnesses who were also the victims of the dacoity and therefore, their identification of the property in a test identification was of next to no value. Moreover, the articles which were recovered from his house, are most of them unsewn fresh pieces of cloth. I have also myself seen the exhibits and it is very true that they are so and therefore I agree that the test identification in the aforesaid circumstances does not provide any acceptable corroboration of the identification of clothes made before the court by the witnesses. As regards his personal identification in the test parade, the learned trial judge considered that the test identification was not reliable. There is no corroborative testimony of the approver in respect of the accused Data Ram which could prove beyond any manner of reasonable doubt that Data Ram was a participant along with the approver.
29. In case of Salig Ram his personal identification, as stated above, has already been disbelieved. The corroboration that the learned trial Judge found was the recovery of stolen articles. These recoveries were not made upon the information and at the instance of the accused, His house was searched in the presence of the prosecution witnesses and the clothes recovered were Dhotis of ladies and gents and unused fresh pieces of Boski, Chhint, Madrasi Check, Shirt pieces, pieces of markin, Gents' Dhoti, Turban and 2 coats. Their identification in court can also not be believed because the test identification was of no value as the recovery took place in the presence of the witnesses who later on came to identify. The articles were also mixed up with the articles which were recovered from the house of another accused Bahori who has been absconding. I therefore, hold that even in case of Salig Ram the corroboration by the said recovery is not sufficient.
30. As regards AM Ram again, the test identification of Ahi Ram has not been believed by the trial Court. The recoveries were also not made at his instance but a search of his house was taken and in that search one 'Malmal' turban and two ladies' sarees with marks of fresh print of the manufacturer were found in his house. The learned trial Judge remarked that these articles were mentioned in the list Ex. P. 20 which was hit by 162, Cr. P. C. and yet he considered that the same having been identified by Sobharam and Girdhari Lal in a test identification, such identification could provide corroboration to the testimony of the approver. Obviously this is a wrong approach. I therefore, hold that in this case as well the recovery and the test identification are worthless and cannot be used for corroboration of the testimony of the approver against the appellant.
31. Considering all the facts and circumstances of this case, I am satisfied to say that the appellants were convicted on the uncorroborated testimony of an unreliable accomplice.
32. I therefore, accept this appeal, set aside the convictions and sentences of the appellants, and acquit them of all of the offences they were convicted of. They are in jail and shall be released forthwith if not wanted in any other case.