Judgment & Order (Cav):

Arun Dev Choudhury, J.

1. The death Reference No. 01/2023 and Crl Appeal (J) No.150/2023 arise out of judgment and order dated 21.08.2023 passed by the learned Sessions Judge, Sivasagar in Session Case No. 148 (S-S)/2017 corresponding to Demow PS Case No. 87/2015, whereby the accused appellants, namely, Mangal Proja and Pagha Proja were convicted under Section 302/34 IPC and sentenced to death.

2. Since both the Reference and Appeal arise from the same judgment, they are taken up together for disposal by this common judgment and order.

3. We have heard Ms B. Sarma, learned Amicus Curiae appearing for the appellants. Also heard Mr Z Kamar, learned Senior counsel/ Amicus Curiae in the Death Reference case, and Ms B Bhuyan, learned Senior counsel / Additional Public Prosecutor, Assam, for the State.

4. The prosecution case in brief is that on 06.04.2015, PW1, Tikmai Proja, lodged an FIR alleging that on the previous night, the accused persons had murdered Santosh Proja and his wife Lilawati Proja by assaulting them with a ‘Dao’.

5. On the basis of the Ejahar, Demow PS case No. 87/2015 under Section 302/34 IPC was registered, and an investigation commenced.

6. During the investigation, the Investigation Officer visited the place of occurrence, conducted inquests, seized certain articles, allegedly recovered one ‘Dao’ led by the accused, Mangal Proja, and eventually submitted a charge-sheet against both the accused persons under Section 302/34 IPC.

7. Upon committal, the charges under Section 302/34 IPC were framed against the accused persons, to which they pleaded not guilty and claimed to be tried, and accordingly, the trial proceeded

8. To establish the charge, the prosecution examined 18 (eighteen) witnesses. The defence adduced no evidence. The statements of the accused under Section 313 Cr.P.C. were recorded, to which the accused persons denied any involvement in the crime.

9. The learned trial court, upon appreciation of the evidence, convicted the accused persons, and the impugned sentence of death was passed.

10. Hence, the present appeal and reference.

11. Assailing the impugned judgment, Ms. B. Sharma, learned Amicus Curiae, appearing for the appellants, contends that the conviction recorded by the learned Trial Court is wholly unsustainable, both on facts and in law. According to the learned Amicus, the prosecution has failed to establish the charges beyond reasonable doubt, and the evidence on record falls substantially short of the standard required for sustaining a conviction, more particularly, in a case resulting in the imposition of capital punishment.

12. The learned Amicus submits that the prosecution case principally rests upon the testimony of child witnesses, namely PW2, PW3, and PW4. Referring to their evidence, it is argued that testimonies are mutually inconsistent on material particulars and are further contradicted by their own cross-examination. It is contended that PW2, though initially projected himself as an eyewitness, subsequently admitted that he had not seen the accused persons, and he woke up after the occurrence. Similar is the nature of the evidence of PW3 and PW4.

13. It is further argued that the occurrence allegedly took place during night hours in a house admittedly having no electricity connection, and the prosecution admittedly failed to establish the source, intensity and adequacy of light, enabling identification of the assailants by three child witnesses. And therefore, in such circumstances, the identification of the accused persons by the child witness is inherently unsafe.

14. The conduct attributed to the child witnesses is unnatural and inconsistent with ordinary human behaviour. The prosecution has failed to offer to explain where PW4 spent the night, who sheltered him, or why he did not immediately return to his nearby residence and inform PW1, who is his mother, Ms. Sarma argues

15. Learned Amicus further contends that evidence regarding apprehension of the accused persons is wholly uncertain, and the prosecution has failed to explain who apprehended the accused person, at what time they were apprehended, and under what circumstances they were allegedly tied by the villagers. Such omissions, according to the Learned Amicus, affect the prosecution's case regarding the alleged confession and recovery. According to her, the extra-judicial confession in such a circumstance is also wholly unreliable.

16. Learned Amicus has also assailed the alleged recovery of the weapon. It is submitted that the ‘Dao’ was allegedly recovered from an open pond accessible to all. It is argued that essential requirements governing the admissibility and reliability of recovery evidence are therefore absent.

17. Learned Amicus further argues that the forensic evidence does not advance the prosecution's case. Though human blood was detected on certain seized articles, the prosecution failed to establish the blood groups of either of the deceased persons or the accused persons, leading to the non-establishment of any scientific linkage between the seized articles and the crime.

18. Referring to the medical evidence, it is argued that no definite opinion regarding the probable time of death was reflected in the post-mortem report, and no attempt was made to correlate the medical findings with the prosecution version.

19. Summing up her arguments, learned Amicus contends that the prosecution's case suffers from serious inconsistencies, doubtful identification, weak recovery evidence, unsafe extra-judicial confession, inconclusive forensic evidence and significant investigative lapses. Accordingly, it is argued that conviction cannot be sustained, and the appellants are entitled to the benefit of doubt.

20. Per contra, Mr. Z. Kamar, learned Amicus arguing in the Death Reference, as well as Ms. B. Bhuyan, learned Amicus, Senior counsel/ Additional Public Prosecutor, Assam, have supported the impugned judgment and contended that the learned Trial Court has correctly appreciated the evidence on record.

21. The learned Senior counsels submit that prosecution has succeeded in establishing a consistent, coherent chain of circumstances pointing unerringly towards the guilt of the accused persons.

22. According to them, the testimonies of PW2, PW3, and PW4 clearly establish the participation of the accused persons in the occurrence, and the minor discrepancies pointed out by the defence are natural and inconsequential, particularly having regard to the age of the witnesses and the traumatic circumstances under which they witnessed the incident.

23. It is argued that the testimony of a child witness is not to be discarded merely on account of age and that evidence of PW2, PW3, and PW4 receives substantial corroboration from the surrounding circumstances, including the prompt disclosure made by the children on the following morning.

24. The learned Senior counsels further submit that the prosecution has established the motive for the crime arising out of a dispute concerning cattle and tethering of cows. The evidence of the child witnesses regarding the prior dispute between the deceased and the accused persons lends assurance to the prosecution's version.

25. It is further contended that evidence regarding the recovery of Dao and the forensic report constitutes additional incriminating circumstances. According to the learned Senior counsels, the recovery was affected, pursuant to information furnished by the accused Mangal Proja, and the weapon was found stained with human blood. Such circumstances, coupled with the testimony of the prosecution witnesses, materially strengthen the prosecution's case.

26. The learned Senior counsels have also placed reliance upon the extra-judicial confession allegedly made by the accused persons before the villagers, and submitted that there is no legal bar to acting upon such confession if the court finds the evidence trustworthy and voluntary.

27. It is further submitted that the homicidal nature of the deaths stands conclusively established by the medical evidence, and the brutality of injuries is wholly consistent with the prosecution's version of assault by a sharp-cutting weapon.

28. According to the learned Senior counsels, the evidence on record, when considered as a whole, establishes the guilt of the accused persons beyond all reasonable doubt, and the learned Trial Court committed no error in recording the conviction.

29. We have given anxious consideration to the submissions advanced by the learned Counsel for the parties. Also, perused the materials available on record.

30. The prosecution's case rests primarily upon:

                   I. The eyewitness accounts of PW2, PW3, and PW4, three children aged about 10, 12, and 14 years, respectively.

                   II. the forensic detection of human blood in the wearing apparel of the accused persons, Mangal Proja and Pagha Proja;

                   III. The discovery of the weapon used in the crime is projected to have been led by the accused, Mangal Proja, and

                   IV. The extra-judicial confessions made by both accused persons.

31. Thus, the prosecution seeks the conviction primarily on the eyewitness account. Secondly, medical and forensic evidence is expected to corroborate the eyewitness account, the alleged recovery, and the accused's confessional conduct.

32. The central question before the court is not whether the deceased died a homicidal death, a fact that stands clearly established, but as to whether the prosecution has succeeded in proving beyond all reasonable doubt that the two accused persons were responsible for the alleged act on the strength of the testimonies of the eyewitnesses, with the projected corroboration by the prosecution.

33. Since the conviction was based substantially upon the testimony of child witnesses coupled with alleged recovery and extra-judicial confession, it becomes necessary to scrutinise the evidence with grave caution.

34. Against the aforesaid backdrop, let us now scrutinise the testimonies of the eyewitnesses.

35. The PW2, PW3, and PW4 are the star witnesses for the prosecution. These three witnesses are children, aged about 6, 8, and 10 years, respectively, at the time of the crime.

36. Before evaluating the testimony of PW2, PW3 and PW4, it is necessary to bear in mind the settled principle governing the evidentiary value of a child witness.

37. The law does not regard the testimony of a child witness as inherently unreliable merely because of age. Once the court is satisfied that the witness possesses sufficient intelligence to understand the questions put and is capable of furnishing rational answers, the evidence becomes admissible.

38. The subsequent enquiry concerns credibility rather than competency.

39. There is equally no rule of law that the testimony of a child witness must invariably be corroborated before it can be acted upon. Corroboration is a rule of prudence and not of law. Where the testimony is found to be natural, consistent and free from taint of tutoring or external influence, conviction can lawfully rest upon such evidence alone.

40. At the same time, owing to the susceptibility of children to influence, tutoring or imaginative constructions, courts are required to subject such evidence to careful scrutiny.

41. In assessing the reliability of a child witness, the court may examine the spontaneity of the narration, the consistency of the account during cross-examination, the naturalness of the conduct attributed to the witness and its compatibility with the surrounding circumstances and other evidence on record.

42. The court must, therefore, guard equally against two errors: rejecting evidence solely on account of age and accepting it uncritically without testing its entrenching reliability.

43. Now, coming to the case in hand, PW2, Karam Proja, son of the deceased, is one of the principal witnesses relied upon by the prosecution. In examination-in-chief, PW2 stated that at about 9.30 pm, both accused persons came and called his father outside. According to him, when his father went outside, both accused assaulted him with dao blows and thereafter assaulted his mother. However, when his testimony is subjected to closer scrutiny, substantial inconsistencies emerge. During cross-examination, PW2 admitted that there was no electricity connection in their house and that they were sleeping at the relevant time. He woke up and found his father lying injured. He had not seen the two accused, Mangal and Pagha; he did not know what his father had done after they went to sleep. The aforesaid contradiction is not a minor discrepancy. It strikes at the very root of ocular identification. A witness who first claims to have witnessed the assault but later admits that he had not seen the accused persons and was sleeping cannot be treated as wholly reliable.

44. The occurrence admittedly took place outside the house during night hours. PW17, the Investigating Officer, admitted during cross-examination that the sketch map does not mention any lighting arrangement.

45. PW2 also admits the absence of electricity. Although PW2 referred to a lamp inside their room, there is absolutely no evidence regarding the intensity of light, the exact location of the lamp, whether the lamp could illuminate the exterior veranda and backyard, or whether visibility from inside the room to the place of occurrence was at all possible. Beyond that, PW-3, who was with PW-2 in the bed, testified that the lamp was put off.

46. It is by now well settled that the identification in darkness requires a careful scrutiny. The identification evidence becomes unsafe where conditions of visibility remain doubtful. In the present case, the prosecution has failed to establish the basic foundational facts regarding visibility.

47. Another important aspect of the matter is that though every witness may react differently to trauma, the conduct attributed to PW2 and PW-3 appears to be unnatural inasmuch as such conduct becomes relevant where testimony itself suffers from serious infirmities.

48. According to the prosecution, both the parents of the PW2 and PW3 were brutally hacked. Repeated cries occurred. Severe assault took place. Yet PW2 and PW3 remained hidden under the bed throughout the entire night; no neighbours arrived; no alarm was raised to nearby villagers. Such conduct appears inconsistent with ordinary human behaviour. Therefore, prudence demands a cautious evaluation of the testimonies of such a witness.

49. Coming to the other star witness, PW3, Jonmoni Proja, daughter of the deceased, who also claimed to be an eyewitness, in her examination-in-chief, stated that the accused persons hacked her father and mother, and she witnessed the occurrence through the window.

50. However, during cross-examination, she admitted that there was no electricity connection, that the lamp in the room was turned off, and that PW2, PW3, and PW4 were all on the same bed. After hearing the hue and cry, PW2 and PW3 went under the bed. In the morning, PW1 took them out from under the bed, and they saw their parents lying outside.

51. Thus, her own cross-examination materially undermines her claim of actual witnessing of the assault. Her testimony suggests a real possibility that she merely heard commotion during the night and subsequently reconstructed the incident after seeing the dead bodies in the morning.

52. It is by now well settled that where a witness is neither wholly reliable nor wholly unreliable, the court must seek dependable corroboration. In our opinion, the PW2 and PW3 are such witnesses.

53. PW4, Ram Proja, another child witness, stated that the accused persons quarrelled with Santosh Proja regarding cows and assaulted both the deceased persons with ‘kolom kotari’. According to PW4, after witnessing the incident, he fled away and remained outside throughout the entire night.

54. However, the prosecution has not explained where PW4 spent the night, who sheltered him, why he did not return to his nearby house, or why he did not immediately inform PW1, who is admittedly his mother. This omission assumes significance because it has come on record that PW1's house was nearby and accessible.

55. The complete lack of any exploration of the whereabouts of a young child witness throughout the night creates a serious improbability in the prosecution's narrative.

56. PW4 also materially differs regarding the weapons used. He referred to a ‘kolom kotari’, whereas the prosecution sought to prove the recovery of a ‘dao’.

57. Though, the PW2 and PW3 are most natural witnesses, being the residents of the house and childrens of the deceased, even PW4, being the relative of the deceased, the most striking feature of the case is the complete absence of any natural independent witness, like the neighbours/ villagers to the occurrence in the backdrop of the projection of the prosecution that there were violent and multiple assaults, repeated hacking, loud cries, brutal injuries, including the severance of a hand.

58. The PW5 lived at about one kilometre distance, PW6 within approximately half a kilometre, and PW13 within 300 metres. All these witnesses admittedly came only in the morning after hearing about the occurrence from the villagers. There is no explanation of whether any nearby villagers other than PW5, PW6, and PW13 were there.

59. In a village environment, the complete absence of any immediate response from neighbours despite alleged cries and brutal assault creates a serious improbability. The prosecution has offered no explanation for why no villagers heard the incident during the night. This circumstance assumes greater significance because the prosecution simultaneously claims that the villagers later gathered in large numbers and apprehended the accused person.

60. What is equally difficult to reconcile is that despite the prosecution version of the repeated hacking, cries for help and violent assault resulting in extensive injuries, none of the aforesaid witnesses claimed to have heard the occurrence during the night. Therefore, the subsequent large-scale gathering of the villagers sits uneasily with the prosecution's narrative that no one responded during the incident itself.

61. Upon a cumulative assessment of evidence of PW2, PW3 and PW4, we find it unsafe to treat them as wholly reliable witnesses for the reason that:

                   i. The PW2 materially diluted his claim of witnessing the occurrence by admitting that he had not seen the accused and that he woke up after the occurrence.

                   ii. PW3, while claiming to have witnessed the assault through a window, admitted that the lamp was put off and that she went under the bed after the hue and cry.

                   iii. On the other hand, PW4 also admitted that he was in the bed with PW2 and PW3, though he was not sleeping. He also introduced a version involving a ‘kolom kotari’ instead of a ‘dao’ inasmuch as he admitted that he knows the difference between ‘kolom kotari’ and ‘dao’. He witnessed the two deceased persons inside the kitchen and saw the body lying outside, and he fled away out of fear. He further claimed to have spent the entire night away from home, without explaining his whereabouts.

62. In our opinion, their testimonies are not merely affected by minor discrepancies but also suffer from infirmities that touch the very question of whether they actually witnessed the occurrence and/or identified the assailants. Viewed cumulatively, their evidence lacks the degree of consistency, certainty, and intrinsic reliability necessary to sustain conviction unless supported by strong independent corroboration.

63. Now, let us deal with the other evidence, which, according to the prosecution, corroborates the ocular testimonies.

64. PW1, Tikmai Proja, is the informant. She admittedly is not an eyewitness to the occurrence. According to her, on the next morning, the children of the deceased, namely PW2 and PW3 and her son PW4, informed her that the accused persons had murdered the deceased. PW1 further stated that thereafter she went to the place of occurrence and saw the dead bodies of Santosh Proja and Lilawati Proja lying with injuries. However, during crossexamination, PW1 admitted that she had no personal knowledge of the incident; upon reaching the place of occurrence, she found no one there. She learnt of the incident only in the morning.

65. The evidence of PW1 becomes important because several prosecution witnesses subsequently stated that the accused persons were apprehended at the place of occurrence itself, they were tied up by the public, they confessed before the villagers and the police personnel. If the accused persons were indeed apprehended and tied at the place of occurrence itself, as projected by PW7, PW10, PW13 and PW17, it is difficult to reconcile the statement of PW1 that she found nobody there when she reached the place of occurrence in the morning, more particularly in the absence of any disclosure of the timeline/sequence of events by any of the prosecution witnesses.

66. From the aforesaid testimonies of prosecution witnesses, what emerges is that the prosecution version proceeds on the assumption that the accused persons were found tied at the place of occurrence.

67. Yet, no witness explained the most fundamental facts, namely, who first identified them as perpetrators of the crime; who apprehended the accused persons; at what time they were apprehended; from where they were apprehended; from where they were brought; who tied them; and how they were prevented from leaving the place of occurrence after commission of the crime; whether they were apprehended after commission of the crime or whether they were apprehended before the arrival of the police or after the arrival of the police.

68. Beyond that, no explanation is forthcoming as to why no witness was examined to establish these facts. The silence of the prosecution on these foundational facts assumes significance because the alleged confession and recovery are presented as a consequence of such apprehension.

69. The unexplained contradictions regarding the apprehension of the accused raise serious doubts about the genesis of the prosecution case itself. Such inconsistency is important because the prosecution's case regarding the recovery itself is doubtful.

70. Let us deal with the further corroborative evidence projected by the prosecution.

71. The first one is the discovery of the weapons used in the commission of the crime led by the accused.

72. The law regarding the admissibility of statements of the accused recorded under Section 27 of the Evidence Act is well settled since the decision of the Privy Council in Pulukuri Kotayya v. King-Emperor reported in 1946 SCC Online PC 47, wherein it was held that as per Section 27 of the Indian Evidence Act, 1872, only the information leading to the actual discovery of facts in consequence of such information would be admissible in evidence, but the extent of information admissible must depend upon the exact nature of facts discovered, to which such information is required to be related.

73. The Constitution Bench of the Hon’ble Apex Court, while dealing with the provisions of Sections 24 to 27 of the Indian Evidence Act,1872 and Section 162 Cr.P.C. in State of U.P. Vs. Deoman Upadhyaya reported in (1960) SCC online SC 8. The propositions can be culled out in the following manner:

                   i. A confessional statement made by an accused to a police officer or procured by inducement, threat or promise having reference to the charge against him cannot be proved against him in a proceeding in which he is charged with the commission of an offence, whether such statement is made by such person in custody or not.

                   ii. A confession made to a person other than a police officer, while in custody of police, cannot be used as evidence against such person in a proceeding in which the said person is charged with the commission of an offence, subject to the condition that such confession is made in the immediate presence of a Magistrate.

                   iii. A statement made, whether it is a confession or not, to another person, when the maker of the confession is not in custody, the person before whom such statement is made may be proved if it is otherwise relevant; however, the person before whom such confession is made shall not be a police officer.

74. Having recorded the aforesaid proposition of law, now, let us deal with the prosecution's proposition.

75. The prosecution sought to prove that the accused, Mangal Proja, led the police in recovering one Dao from a pond. However, the evidence on this aspect is riddled with contradictions. The evidence of PW7 assumes particular significance because he was projected by the prosecution as an independent seizure witness. Yet, PW7 categorically stated that the ‘dao’ shown in the court was not the ‘dao’ recovered from the place of occurrence. If accepted, the statement strikes at the chain of custody of the alleged weapon. If rejected, the prosecution is left without an explanation for why its own independent witness would make such a statement. Either way, the prosecution does not emerge completely safe to rely on. On the other hand, PW8 identified the Dao shown in the court as the recovered weapon. PW17 admitted that the pond was not enclosed; it was accessible to all. PW6 also admitted that the pond was easily accessible and visible. Another important fact is that the accused's statement was recorded after the recovery, not beforehand.

76. Therefore, the alleged recovery completely loses exclusivity.

77. It is by now well settled that discovery evidence from an open and accessible place is inherently weak unless exclusive knowledge is established. No such exclusive knowledge stands proved in the present case. Moreover, the contradiction regarding the identity of the weapon itself further weakens the prosecution's case.

78. The prosecution has also relied upon an alleged extra-judicial confession said to have been made by the accused persons before certain villagers.

79. The evidentiary value of such a confession is well settled. An extra-judicial confession is relevant evidence and is not rendered inadmissible merely because it was not recorded before a magistrate.

80. Nevertheless, by its very nature, it is regarded as a comparatively weak form of evidence and must be examined with a high degree of caution before being accepted as the foundation of conviction.

81. The court must first be satisfied that the confession was voluntary, truthful and made in circumstances which inspire confidence. The person before whom it is allegedly made, the exact words attributed to the accused, the surrounding circumstances and the conduct of the parties before and after the disclosure are all relevant considerations.

82. Ordinarily, human conduct does not suggest that a person who has committed a grave offence would readily confess guilt before strangers or persons in whom he reposed no confidence.

83. The court must therefore examine whether the alleged confession accords with ordinary human probabilities.

84. Though there is no inflexible rule requiring corroboration, courts generally seek assurance from the independent circumstances before placing reliance upon such evidence, particularly where a confession constitutes the principal incriminating circumstance against the accused.

85. In the present case, the PW6 stated that the police interrogated the accused persons in public and that they admitted their guilt. PW7 stated that the accused persons confessed before the police and villagers. PW10 stated that the accused persons admitted their guilt after apprehension by the police. PW13 stated that the accused confessed after the arrival of the police, and it has come to evidence, as discussed hereinbefore, that the accused were tied up at the place of occurrence by the public.

86. Therefore, the alleged confession admittedly occurred in the presence of police after apprehension and while surrounded by the public. Such a confession becomes legally unsafe and inherently weak.

87. Here, this evidence suffers from another serious infirmity. None of the witnesses clearly explained the exact words used by the accused persons; whether both the accused persons confessed identically; whether the confession was voluntary; whether the accused were assaulted or coerced by the public in a case when, admittedly, the PW7, PW10, PW13 and PW17 found the two accused persons at the place of occurrence tied.

88. Therefore, the alleged extra-judicial confession does not inspire confidence. The prosecution has failed to establish the circumstances in which such a confession was allegedly made with the degree of certainty that the law requires.

89. The evidence on record is inconsistent regarding the stage at which the accused persons were apprehended, the circumstances in which they were allegedly tied by villagers, and the manner in which the confession came to be made.

90. Significantly, the prosecution has not established any relationship of confidence between the accused persons and the witnesses before whom the confession is alleged to have been made.

91. The conduct attributed to the accused is not wholly consistent with the ordinary human probabilities, and the evidence regarding the confession suffers from material infirmities and uncertainties. In the absence of reliable corroborative circumstances, it would be unsafe to place reliance upon the alleged confession. This circumstance is therefore excluded from consideration while assessing the guilt of the accused.

92. Now, as regards the forensic evidence, PW16, Scientific Officer, proved the forensic report showing the presence of human blood on the allegedly recovered “Dao”, in the alleged ‘wearing apparel' of the accused persons and the clothes of the deceased. However, the prosecution failed to establish the blood group of the deceased persons, the blood group of the accused persons, and the matching correlations between stains and the deceased persons.

93. Thus, the forensic report merely proves the presence of human blood, but does not establish the participation of the accused persons in the crime. It needs no reiteration that blood stain evidence without conclusive grouping linkage cannot conclusively incriminate the accused.

94. Coming to the post-mortem report, PW14, Dr. Brojen Saikia proved the post-mortem report and opined that the deaths were homicidal in nature.

95. There can be no dispute from the evidence on record that Santosh Proja and Lilawati Proja died homicidal deaths. However, medical evidence only establishes the nature of death and the nature of the injuries sustained and not the identity of the assailants.

96. Significantly, no clear estimation regarding the probable time of death was given in the post-mortem report; secondly, no effort was made to correlate medical findings with the prosecution timeline.

97. The prosecution sought to project the medical evidence, forensic report, recovery of the weapon and alleged confession as circumstances corroborating the testimonies of PW2, PW3 and PW4, the three child witnesses. However, none of these circumstances withstands our closure scrutiny for the reasons recorded, and the discussions made hereinbefore.

98. The medical evidence establishes only homicidal death; the forensic evidence does not establish any scientific linkage with the accused; the recovery suffers from serious infirmities; and the confession is found to be unsafe.

99. The cumulative result is that the prosecution story suffers from serious lapses i.e.,

                   i) absence of lighting analysis;

                   ii) absence of explanation regarding apprehension of the accused persons;

                   iii) absence of explanation why neighbours did not arrive;

                   iv) doubtful seizure procedure;

                   v) recovery from the accessible pond;

                   vi) contradiction regarding the identity of weapons;

                   vii) no explanation regarding the whereabouts of PW4 throughout the night;

                   viii) no scientific correlation of blood groups;

                   ix) No proper explanation of the timeline of occurrence.

100. Consequently, the purported collaborative circumstance failed to provide assurance to the ocular version.

101. In a case like the present one, such omissions assume substantial importance. The cumulative effect of these circumstances creates reasonable doubt regarding prosecution version.

102. Mr. Kamar, learned Senior counsel has strenuously argued that the court should not discard the evidence of the eye witnesses, by searching for perfection and that the court must separate “the grain from the chaff”.

103. There can be no quarrel with the proposition. It is true that criminal courts are not expected to reject otherwise reliable evidence on account of minor inconsistencies, embellishment or omissions which do not affect the substratum of the prosecution case. Human testimony is seldom a photographic reproduction of events, and some degree of variation is often the hallmark of truthful evidence.

104. The difficulty in the present case, however, is that infirmities noticed by us and recorded hereinabove do not relate to peripheral details but to the foundational aspect of the prosecution case itself.

105. The contradictions are not confined to the sequence of events or subsidiary circumstances. They extend to the very fact whether the witnesses actually witnessed the occurrence or could identify the assailants; whether there exists sufficient visibility for identification; whether the accused persons were apprehended in the manner alleged; whether the weapon produced in court was the weapon allegedly recovered, and whether the extra-judicial confession was made in circumstances inspiring confidence.

106. The principle of separating grain from the chaff presupposes the existence of a reliable core capable of being segregated from exaggeration or embellishment.

107. In the present case, the difficulty is not that the prosecution witnesses have added embellishment to an otherwise trustworthy narrative; the infirmities permeate the very foundation of the prosecution version.

108. The court is, therefore, not confronted with a case where the truth can be separated from exaggeration; it is confronted with a situation in which serious doubt arises as to whether the foundational facts themselves are proved.

109. We are, therefore, unable to accept the submission that the present case can be resolved by merely separating the grain from the chaff. What emerges from the evidence is not a trustworthy core surrounded by embellishment, but a prosecution case affected by doubts touching its essential features. In such circumstances, the benefit of the doubt must necessarily be given to the accused.

110. It is true that deaths were brutal and gruesome. However, the gravity of the crime cannot reduce the burden of proof resting upon the prosecution. Criminal Courts, including appellate courts, are duty-bound to ensure that moral suspicion does not replace legal proof.

111. Upon an overall appreciation of the evidence on record, we find that the prosecution has failed to establish a trustworthy ocular account free from reasonable doubt. The testimony of the child witnesses suffers from material infirmities touching the very factum of witnessing. The circumstances surrounding the apprehension of the accused remained unexplained. The alleged extra-judicial confession is unsafe. The recovery of the weapon used in the crime is doubtful. The forensic evidence is inconclusive.

112. The cumulative effect of these infirmities is not merely the existence of isolated doubt but the emergence of substantial and reasonable doubt regarding the prosecution version itself.

113. In such circumstances, it would be wholly unsafe to sustain the conviction of the accused persons, much less to confirm a sentence of death.

114. Consequently, in our opinion, the accused appellants are entitled to the benefit of doubt.

115. Accordingly, Crl A (J) No. 150/2023 stands allowed.

116. The judgment and order dated 21.08.2023, passed by the learned Sessions Judge, Sivasagar, in Sessions Case No. 148(S-S)/2017, is hereby set aside.

117. The appellants, namely Mangal Proja and Pagha Praja, are acquitted of the charges under Sections 302/34 IPC, thereby giving them the benefit of the doubt.

118. The Death Sentence Reference No. 01/23 consequently stands rejected.

119. Accused appellants shall be released forthwith, if not required, in any other case.

120. The trial court records be returned.

121. While parting with the records, it is observed that the learned Trial court has already invoked the provisions relating to victim compensation and directed the District Legal Services Authority, Sivasagar, to consider and process the claim of the victim’s family in accordance with the law applicable to the victim compensation scheme as envisaged under section 357-A Cr.P.C.

122. The direction issued by the trial court shall accordingly remain undisturbed. It is, however, expected that the concerned authority shall ensure expeditious compliance so that the intended rehabilitative support reaches the beneficiaries without undue delay, if such compensation has not already been paid.

123. While parting with this record, this Court puts on record its appreciation as regards able assistance rendered by Ms. B Sarma, learned Amicus Curiae. The Registry shall ensure that Ms. B Sarma, learned Amicus Curiae be paid the legal fee, as payable to a Legal Aid Counsel as per the existing norms fixed by the Legal Services Authority.