Holding

The conviction was overturned because the identification of the appellant was doubtful and the identification parade does not constitute substantive evidence under Section 9 of the Evidence Act.

Summary

Umesh Kamat was convicted under Section 396 IPC for a night dacoity in which an inmate was killed. The conviction rested solely on the testimony of three eyewitnesses (PW1, PW3, PW4) who identified him in a test identification parade held seven weeks after the incident, despite the dacoits wearing masks. The Supreme Court held that the witnesses did not know the appellant, that PW3 never identified him in court, and that the identification parades are investigatory, not substantive evidence under Section 9 of the Evidence Act. The Court found the eyewitness accounts unreliable given the darkness, masks, and the lapse of time, and noted the absence of any corroborative material. Consequently, the conviction was set aside and the appellant was released.

Issues considered

Whether identification of the accused based on a test identification parade and eyewitness testimony satisfies the requirements of Section 9 of the Evidence Act.
Whether eyewitnesses could reliably identify a masked dacoit after a seven‑week delay.
Whether conviction under Section 396 IPC can be sustained in the absence of corroborative evidence linking the accused to the crime.

Legislation cited

Indian Evidence Act, 1872s. 9
Indian Penal Code, 1860s. 396

Subjects

dacoityidentification paradeeyewitness identificationSection 396 IPCEvidence Act Section 9reasonable doubtmasked assailantscriminal conviction

Judgment

A UMESH KAMAT + I
V.

STATE OF BIHAR

JANUARY 13, 2005

B [P. VENKATARAMA REDDI AND P.P. NAOLEKAR, JJ.]

Penal Code, 1860-Section 396-Dacoity at night-By armed men with • 1

covered faces-Inmate of house killed by firing-Identification of accused- ....
Identification parade held after seven weeks-Appellant alone convicted based
c on the evidence of PWs /, 3 and 4-Appellant not a person known to the
prosecution witnesses-On facts, held, identification ofAppellant was doubtfal-
There being no identification of Appellant in the Court by PW3, results of
identification parade will be of little value-Evidence of PW/ not credible-
He belied his own version in cross-examination-There is also doubt if PW4
D could closely observe the identifiable features of Appellant in mask, that too
in the glow of dim lantern and light emitted by torches-Hence conviction of
y
Appellant set aside-Evidence Act, 1872-Section 9.
Appellant along with seven others was charged by the Additional
Sessions Judge under Section 396 IPC for committing dacoity at night in
E course of which one of them fired at and killed one of the inmates of a
house.
Four accused were acquitted by the Trial Court. On appeal by the
remaining four accused, three were acquitted by the High Court. Appellant
alone was convicted under Section 396 IPC and sentenced to life
imprisonment. High Court agreeing with the trial Court relied on the
F
evidence of PWl, PW3 & PW4 and held that the identification of appellant ~ .
could not be doubted. Hence the appeal.
Allowing the appeal, the Court
HELD: 1.1. Appellant is not a person known to the prosecution
G witnesses. PW3- the minor daughter of the deceased, did not identify him
in the Court as he was not present. Though the trial Court and the High
Court proceeded on the basis that the four accused including the appellant
were identified in the Court by PW3, in fact there was no such
identification, as is clear from her deposition. [480-8-CJ

H 474
UM ESH KAMAT v. STA TE OF BIHAR 475
' ~ 1.2. The identification parades belong to the stage of investigation A
and do not constitute substantive evidence. The substantive evidence is the
evidence of identification in Court because the facts which establish the
identity of the accused persons are relevant under Section 9 of the Evidence
Act. In the absence of identification in the Court at the time of tendering
evidence the results of test identification parade will be of little value. The B
High Court committed a factual error in observing that PW3 gave a
description of the general appearance of the appellant. Therefore the
testimony of PW3 does not advance the prosecution case. (480-C-El
I '

Malkhansingh and Ors. v. State of Madhya Pradesh, (20031 5 SCC 746,
relied on. c
2. Lt!aving apart the probabilities and the natural course of conduct,
the version of PWI- the neighbour living in the adjacent house, is belied
by his own version in the cross-examination. It is unbelievable that he
would go and remain at the place of occurrence even for a short-while
when the attack and dacoity by armed persons was going on and that he D
dared to flash the torch light on them more than once in order to get an
idea of the miscreants. On his own showing, he was concerned about his
own safety. Moreover, this witness stated that there was no electricity or
)
lantern light at the house of the deceased. On the face of it, one need not
say anything more to discredit this witness on the aspect of identification E
of the appellant which was done after a lapse of about seven weeks. As
the dacoits covered thefr faces, it was not reasonably possible for PWl to
identify each of the criminals, some of whom including the appellant were
unknown to him with the help of the light flashed by him intermittently,
even if that version is accepted. The High Court described PWl as an
'independent' and natural witness and believed him without testing the F
, J. veracity of evidence in the light of various circumstances. (481-8; D-F)

3.1. Even the evidence of PW4- the brother of the deceased does not
inspire confidence in ultimate analysis. There is any amount of doubt on
the point whether PW4, in the situation in which he was placed, could
closely observ·e the identifiable features of the appellant in mask that too G
in the glow of dim lantern and in the light emitted by the torches flashed
at him or other inmates of the house. This doubt has to be viewed in the
context of two things, firstly- there was no recovery of property, nor any
other corroborating evidence linking the appellant to the crime. The
second aspect is that the trial Court was not inclined to believe the evidence H
·+--
476 SUPREME COURT REPORTS [2005] I S.C.R.

A of the identification of three other accused at the same identification t '
parade and commented that they were falsely implicated. If so, the
evidence of PW4 should have been scrutinized with greater caution instead
of proceeding on a premise that he was a truthful witness. Further PW4
did not have the occasion to observe the dacoits' operations inside the
house. He would have noticed them only initially for a short-while before
B they entered the house. It is his case that he became unconscious a little
later as a result of injury inflicted on him. These are all the various doubts
which loom large over the prosecution story of identification of the
appellant. 1481-G; 482-D-GI < '

.....
c 3.2. The only reason given by the High Court in believing the
evidence of PW4 is that the incident must have left a deep impression in
his mind, especially in view of the injuries which he and his wife received
at the hands of the dacoits and such impression would not easily fade out
within a few weeks or months. This observation of the High Court was
based on the hypothesis that PW4 was in a position to clearly notice the
D physical features and appearance of the appellant. There was no warrant
for such ready assumption. The relevant aspects which give room for
reasonable doubt were not at all noticed by the trial court or the High ...
Court. Under these circumstances, interference with the finding recorded
\
by both the Courts is called for. Conviction and sentence against the
E appellant is accordingly set aside. (482-H; 483-A-B; GI

Tahir Mohammad v. State of MP., 119931 Supp. 2 SCC 697, relied
on.

CRIMINAL APPELLATE JURISDICTION : Criminal Appeal No.
F 23 of 2004.

From the Judgment and Order dated 6.8.2003 of the Patna High Court ~
in Crl. A. No. 60 of 1999.

K.B. Rohtagi, Ms. Apama Rohtagi and Mahesh Kasana for the Appellant.
G Kumar Rajesh Singh for B.B. Singh for the Respondent.

The Judgment of the Cou11 was delivered by

P. VEN KAT ARAMA RED DI, J. The appellant along with seven others
were charged by the Additional Sessions Judge, Madhubani under Section
H 396 IPC for committing dacoity on the night of 28.5. 1994 in the course of
UM ESH KAMAT v. STATE OF BIHAR (REDD!. J.] 477
.+ which one of them fired at and killed one of the inmates of the house, A
namely, Rajendra Thakur. Four were acquitted by the trial Court and on
appeal by the remaining four accused, three were acquitted by the High
Court. The appellant alone was convicted under Section 396 IPC and sentenced
to life imprisonment. It appears that the appellant has so far undergone
imprisonment for about six years.
B
PWl-a neighbor, PW3-the minor daughter of ti:e cJeceased, PW4-
the brother of the deceased who was also injured by the marauders and
I •
PW5-the sister-in-law of the deceased are the eye-witnesses. Though the
information in regard to the incident was supposed to have been conveyed to
the police station by PW!, the FIR was not recorded on that basis. However, C
the police arrived at the scene at about 2 a.m. and then recorded the statement
of PW5 i.e. the sister-in-law of the deceased, which was treated as First
Information Report.

• According to the statement of PW5 as incorporated in the FIR, at about
I I p.m., she and other inmates of the house were sleeping and on hearing the D
voice of his elder brother-PW4 who was sleeping beyond the main doorway,
Rajendra Thakur-the deceased opened the main door and the informant-
PW5 and PW3 also went behind him. She saw four persons in full pants and
half shirts standing at the gate and another wearing a black full pant and full
shirt with checks. Soon after Rajendra Thakur opened the door, the person
wearing the black full pant fired at him as a result of which Rajendra Thakur E
collapsed instantaneously. Thereafter, she beseeched the miscreants not to
harm and to take away whatever articles they wanted. Still, they inflicted
injuries with dagger on the body of Rajendra Thakur even after he fell down
and one of them also attacked her with a lathi. They also injured her husband
Laxman Thakur PW4 with lathi and rod as a result of which he became p
) f unconscious. Four/Five dacoits ente•ed the house and went on a looting spree
for about 15 minutes. On the alarm raised by the villagers, dacoits who were
20 in number fled away with looted articles. Rajendra Thakur succumbed to
the injuries then and there. She stated that the details of looted articles will
be furnished by the wife of Rajendra Thakur and other family members who
had gone to the hospital. According to her, the dacoits were young men G
wearing dhothi, full pant, half shirt etc. and they had fire arms, dagger, lathi
and torches with them and were speaking Hindi and Mithili languages. She
also stated that her husband would disclose the identity of the dacoits on
coming to senses and her other family members will identify the looted
articles if recovered. Informant also stated that three dacoits have covered H
478 SUPREME COURT REPORTS [2005] I S.C.R.

A their faces with 'galmocha'. The statement was recorded in the presence of +
her son-in-law and sambandhi.

PWlO is the main investigating officer. He stated that after coming to
know of the dacoity, he went to the place of occurrence with armed police
at about 12.15 a.m. He noticed the dead body of Rajendra Thakur and he
B could not record the statement of Laxman Thakur as he was senseless and
therefore he recorded the statement of the wife of Laxman ThakurPW5. He
found cash box and wooden almirah in broken condition and the articles
therein lying helter-skelter. He also found the Godrej almirah in broken
condition and found the articles therein on the ground in a disturbed condition.
C He held the inquest over the dead body and took steps to have the postmortem
conducted. The injured Laxman Thakur-PW4 was sent to hospital. On the
basis of information collected during investigation, he arrested four accused,
the appellant being one of them. Then he took steps to have the test
identification parade done by the judicial Magistrate. He submitted the charge-
sheet against the four persons while showing others as absconders. The further
D investigation was handed over to his successor after his retirement.

The factum of homicidal death of the deceased as a result of firing is );
not in dispute. It is not the case of the prosecution that the appellant herein
was a known person. The whole case of the prosecution rests on the credibility
of identification, said to have been made by the four witnesses in the course
E of test identification parade held by the Magistrate. In the case of the appellant
and three others, the identification parade was held after seven weeks i.e. on
19.7.1994 and in the case of others it was held much later i.e. after 6 to);)
months. In view of the long time gap, the High Court was not inclined to
believe the version as regards the identification of three appellants before it
p and therefore they were acquitted. As far as the appellant is concerned, the
High Court agreeing with the trial Court relied on the evidence of the
prosecution witnesses I, 3 & 4 and held that the identification of appellant
could not be doubted.

One important fact to be noticed at this juncture is that PWs 1 to 4
G claimed in the course of their evidence that they identified the three accused
(who were acquitted by the trial Court itself) at the time of occurrence because
they belonged to the same village but the Investigating Officer maintained
that none of the names of the accused were disclosed by the witnesses whom
he examined. A comment was made that the LO. did not record the statements
H properly with a view to help the accused but the trial Court did not accept
UM ESH KAMAT v. STA TE OF BI HAR [REDD!. J.] 479
I 1- this plea. The learned trial Judge commented that the evidence of PWs I to A
4 that they could identify the three accused (other than the appellant) was
"either an improvement or an embellishment and perhaps the aforesaid persons
have been made accused c!ue to previous enmity and the groups in the village".
The trial Court also referred to the statement of the l.0.-PWIO that initially
he was not willing to put the three accused who were the residents of the B
village in the test identification parade but on the direction of the Addi. S.P.,
the three accused persons were also presented for identification.

' . Another fact to be noticed at this stage is that there are discrepancies
in the evidence of prosecution witnesses regarding the number of persons
holding the gun. The Judicial Magistrate examined as PW6 stated that PW! C
pointed cut to him that the appellant Umesh Kamat was one of those having
gun in his hand. The Magistrate also stated that in the second identification
parade, PW! identified the suspect person Dinesh Mohato as the person who
had fired the shot on the deceased. However, PW4 attributed this role to the
suspect Rajeshwar Singh who was identified in the third identification parade.
As already noticed, both of them were acquitted. However, we need not D
dilate further on this aspect as it need not be proved by the prosecution that
the appellant himself caused death. Section 396 enjoins that if any one of the
five or more persons 'conjointly committing dacoity' commits murder in the
course of the same transaction, every one of the persons who participated in
the dacoity will be guilty of the offence of dacoity with murder. Each one of
the dacoits is liable to be punished under Section 396 irrespective of the fact E
whether he is the actual assailant or whether he had shared the common
intention to kill anyone.

Then there is a controversy on the question whether the number of
persons who committed the offence was five or more or less than that. It is F
pointed out that all the prosecution witnesses spoke about the presence and
participation of only four and there was only a vague statement by some of
the witnesses that a number of others (nearly 20) were also outside the house.
There is also a controversy on the question whether any property was plundered
at all, because no details of the properties lost were furnished and no recoveries
were made. In the view we are taking as regards the identification, there is G
no need to delve further into these aspects. However, one striking feature of
the case which we would like to mention is that investigation was most
perfunctory and inadequacies on the part of the prosecution are writ large in
the case.

We now turn our attention to the most crucial aspect of the case in H
480 SUPREME COURT REPORTS [20051 I S.C.R.

A regard to the identification of the appellant. The High Court relied on the
evidence of PWs I, 3 and 4. Neither PW5-the informant nor PW2 (who
identified three other accused) identify the appellant. Hence, their evidence
need not detain us. How far the two Courts were justified in acting on their
testimony on the point of identification is the question. The appellant, as
B already noticed, is not a person known to the prosecution witnesses. As far
as PW 3 is concerned, she did not identify the appellant in the Court as he
was not present. Though the trial Court and the High Court proceeded on the
basis that the four accused including the appellant were identified in the , '
Court by PW3, in fact there was no such identification, as is clear from her
deposition at Para 6. As pointed out in Ma/khansingh and Ors. v. State of
C Madhya Pradesh, (2003] 5 SCC 746 the identification parades belong to the
stage of investigat10n and they do not constitute substantive evidence. The
substantive evidence is the evidence of identification in Court because the
facts which establish the identity of the accused persons are relevant under
Section 9 of the Evidence Act. This Court further observed that failure to /

hold a test identification parade would not make inadmissible the evidence
D of identification in Court. Thus, in the absence of identification in the Court
at the time of tendering evidence the results of test identification parade will »
be of little value. With reference to the evidence of PW3, the High Court
committed another factual error in observing that PW3 gave a description of
the general appearance of the appellant. Therefore the testimony of PW3
E does not advance the prosecution case.

We may now turn our attention to the evidence of the other two witnesses
on which the High Court relied. It is seen from the evidence of PW5 that tht
"dacoits had covered their face with clothes at the time of incident". PW I
also stated that the dacoits were covering their faces except eyes and nose,
F with black cloth. PW 4 did not say specifically whether or not the marauders
were having masks on their face. Assuming that eyes and nose could be seen -t ·
to some extent despite the mask, the question is whether any of the crucial
witnesses could have identified the unknown masked dacoit. PWl-the
neighbour who was living in the adjacent house, came forward with the
G version that after hearing the noise, he put on his torch and in that light, he
saw the dacoits killing Rajendra Thakur at the courtyard of his house. PWI
further stated that thereafter, he went towards the doorway of the house of
Rajendra Thakur and when one of the dacoits flashed the torch on him, he
noticed Rajendra Thakur lying in an injured condition while one person was
attacking him with knife. The witness then claimed in the cross-examination
H that he had flashed the torch 5 to 7 times at the dacoits from a distance of
UMESH KAMAT v. STATE OF BIHAR [REDD!. J.] 481

20-30 feet. Therefore, his version is that he was able to identify the accused A
by flashing the torch light now and then. He claimed to have remained at the
place of occurrence for 2-3 minutes. It seems to us that the evidence of PW!
is not credible. Leaving apart the probabilities and the natural course of
conduct, the version of PW! is belied by his own version in the cross-
examination. While at one point of time he said that he did not get scared, B
at paragraph 24 of the deposition, he clearly stated as follows:

"When I went to the place of occurrence for the first time, then I saw
' . the assault. Accused had also run to assault me. I ran towards my
house in order to save my life. I was having an Eveready torch in my
hand".
c
Jn the next para, he stated that after the dacoits left the place of
occurrence, he and his family members went to the spot and stayed for about
10-15 minutes. It is unbelievable that he would go and remain at the place
of occurrence even for a short-while when the attack and dacoity by armed
persons was going on and that he dared to flash the torch light on them more D
.than once in order to get an idea of the miscreants. On his own showing, he
was concerned about his own safety. Moreover, this witness stated that there
was no electricity or lantern light at the house of the deceased. On the face
of it, we need not say anything more to discredit this witness on the aspect
of identification of the appellant which was done after a lapse of about seven
weeks. As the dacoits covered their faces, we do not think that it was E
reasonably possible for the witness (PW!) to identify each of the criminals,
some of whom including the appellant were unknown to him with the help
of the light flashed by him intermittently, even if that yersion is accepted.
The High Court described PW! as an 'independent' and natural w\tness and
believed him without testing the veracity of evidence in the light of various F
circumstances.
.• *' Amongst the eye-witnesses, it is the evidence of PW4 which assumes
more importance because he was the injured and he would have had the
opportunity to notice the offenders from close range and there was a reasonable
possibility of PW4 having in his mind the imprint of the image of criminals G
who attacked him and his brother. PW4 identified the accused in the course
of evidence. However, even his evidence does not inspire confidence in
ultimate analysis. Contrary to what PW I stated, he took the stand that there
was a lantern at the place (baithak) where he was sleeping. Of course, that
lantern which was described as 'old' was not even seized by the police.
Assuming there was a lantern, in all probability, it would have been quite H
482 SUPREME COURT REPORTS [2005] I S.C.R.

A dim as it is common knowledge that while going to sleep, normally the lamp
is kept at the minimum level in rural areas. PW4 further stated in the chief-
examination that he could identify the four dacoits in the light emanating
from the torch (flashed by the dacoits) and the moon-light. The High Court
observed that it was not a moon-light day as per the admission of some of
B the witnesses. According to PW4, the appellant herein is not the person who
aimed the gun at his brother. The question is whether at that juncture when
he was being subjected to blows soon after he woke up and his brother was
being simultaneously attacked by the armed miscreants, he would have really
observed each one of the four persons with covered faces so keenly and
minutely as to identify them by the uncovered portion of the nose and eyes.
C The answer could only be in the negative. It is pertinent to note that PW4 did
not spell out the distinctive features of the appellant (who was admittedly a
stranger to him) on the basis of which he could identify him despite the
mask.

Thus, there is any amount of doubt on the point whether PW4, in the ?
D situation in which he was placed, could closely observe the identifiable features
of the appellant in mask that too in the glow of dim lantern and in the light
emitted by the torches flashed at him or other inmates of the house. This
doubt has to be viewed in the context of two things, firstly-there was no
recovery of property, nor any other corroborating evidence linking the appellant
E to the crime. The second aspect is that the trial Court was not inclined to
believe the evidence of the identification of three other accused at the same
identification parade held on 19.7.1994 and commented that they were falsely
implicated. If so, the evidence of PW4 should have been scrutinized with
grea.~r caution instead of proceeding on a premise that he was a truthful
witness. One more aspect which deserves notice is that PW4 did not have the
F occasion to observe the dacoits' operations inside the house. He would have
noticed them only initially for a short-while before they entered the house. -ii .
It is his case that he became unconscious a little later as a result of injury
inflicted on him. These are all the various doubts which loom large over the
prosecution story of identification of the appellant. Unfortunately, the High
Court did not analyze the evidence of prosecution witnesses so as to test the
G credibility of their evidence in the light of admitted or undeniable facts apparent
from the record.

The only reason given by the High Court in believing the evidence of
PW4 is that the incident must have left a deep impression in his mind,
H especially in view of the injuries which he and his wife received at the hands
UM ESH KAMAT v. STATE OF BIHAR [REDD!, .I.] 483

' + of the dacoits and such impression would not be easily fade out within a few A
weeks or months. This observation of the High Court was based on the
hypothesis that PW 4 was in a position to clearly notice the physical features
and appearance of the appellant. There was no warrant for such ready
assumption. The trial Court as well as the High Court should not have taken
the version of the PW4 on its face value without testing its credibility. The
relevant aspects adverted to above which give room for reasonable doubt,
B
were not at all noticed by the trial Court or the High Court. Under these
circumstances, the interference with the finding recorded by the both the
I Courts, is called for.
.... We may before parting with the case refer to the decision Tahir
Mohammad v. State of M.P., [1993] Supp. 2 SCC 697. That was a case of
c
dacoity by armed men with covered faces. The passengers of a bus were
robbed at night time. The prosecution witnesses identified the accused in the
Test Identification Parade and in the court too. There were also recoveries of
the looted articles from two of the accused. This court set aside the conviction
under Section 395 to 397 while holding one of the accused guilty under D
Section 412 !PC. The main reason which weighed with this court in excluding
the evidence of identification was that the accused was placed in the Test
Identification Parade with fetters on their legs. This court gave additional
reason for not believing the witnesses on the point of identification in the
following words:
E
"In the instant case the witnesses who were the inmates of the bus
both in their earlier statements and in their oral evidence before the
court have not given any description of the dacoits whom they have
alleged to have identified in the dacoity, nor have they given any
identification marks such as the stature, complexion, height of the F
accused. Further under the stress and strain of such 1a serious incident
/ ' »- as the present one, it would have not been possible for the witnesses
to identify the culprits especially when the culprits were under masks."

The features pointed out by this court in the passage extracted above
are also present in the instant case. G
The appeal is therefore allowed and the conviction and sentence against
the appellant is set aside. He shall be set at liberty fo11hwith unless required
to be detained in any other case.

B.B.B. Appeal allowed. H