Holding

Unexplained delay in conducting the test identification, coupled with the likelihood that the witnesses had prior familiarity with the accused, makes the identification evidence unreliable, and the convictions cannot be sustained.

Summary

In November 1973 a dacoity occurred at Gangaram Yadav's house, resulting in the death of two brothers. The accused were arrested on 9 November 1973, but a test identification parade was conducted only on 5 March 1974, nearly four months later. Three eye‑witnesses, who were relatives of the victims, identified the accused, while a fourth witness did not. The trial court convicted the accused under s.396 IPC; the High Court affirmed. The Supreme Court examined whether (i) the witnesses likely knew the accused beforehand and the omission of their names in the FIR undermined the prosecution, and (ii) whether a test identification after such a delay could be relied upon. It held that the unexplained delay and the probable prior familiarity rendered the identification evidence unreliable, creating reasonable doubt. Consequently, the convictions were set aside and the accused were released.

Issues considered

Whether the probable prior acquaintance of the eye‑witnesses with the accused and the omission of their names in the FIR defeats the prosecution's case.
Whether a test identification conducted after an unexplained four‑month delay is admissible and reliable under s.9 of the Evidence Act.
Whether the identification evidence, in the absence of promptitude, can sustain a conviction under s.396 IPC.

Legislation cited

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

Subjects

identification paradetest identificationeyewitness identificationdelay in identificationIndian Evidence ActSection 9reasonable doubtconviction overturned

Judgment

HARI NATH AND ANR. ETC.
A
v.
STATE OF U.P.

NOVEMBER 6, 1987

B [A.P. SEN AND M.N. VENKATACHALIAH, JJ.]

Indian Evidence Act, 1872: Section 9-Identification of accused
persons in dock after long lapse of time-Whether valid-Evidence of
test identification-Significance of.

C Criminal Trial: Promptitude in time with which suspected \
persons are put up for test identification-Necessity for. _.t-
i
It was alleged that on the nig~t between the 3rd and the 4th
November, 1973 at abont 1.30 a.m. while the two deceased brothers and
their relatives were sleeping, dacoits armed with weapons including fire
D arms raided the house. Apart from looting of property, both the
brothers were seriously injured. They snccnmbed to the injnries there-
after. In response to the commotion and the alarm raised by the house-
hold, a number of villagers assembled and challenged the dacoits, who,
while running away with the loot under cover of gunfire injnred one of
the villagers. "'(
E
An FIR was lodged at 3.45 p.m. that day. Though the identity of
the cnlprits was not disclosed, some general physiognomic features
which contd be of universal application were mentioned. .._

On 9.11. 73 the appellants and the non-appealing accused were
F arrested from two villages which were said to be in close vicinity of the _J
place of occurrence. They were put up for test identification in"'° l
indentification parade only on 5.3. 74, nearly four months after their
arrest. The accused persons were stated to have been picked out by PWs
I to 4.

G The appellants alongwith the non-appealing accused were
arraigned for the offence. Their defence was one of denial.

PWs I, 2 and 4, the son and two daughters of the deceased elder r
brother who were stated to be eye-witnesses and identified the appel·
lants as the culprits claimed to have been able to recall the identity of
H the appellants by perception made at the time of the occurrence with the
848
HARi NATH v. STATE OF U.P. 849

aid of the light of the lantern and the electric hulh which was burning at
A
the scene of occurrence. PW 3, who was also put up as an eye witness
did not support the prosecution.

During the cross-examination the appellants sought to prove that
they, being residents of a village close-by, the eye-witnesses, and
particularly PW 1, who was classmate of three of the appellants, had B
seen and known them earlier and the omission to mention their names
in the FIR detracted from the credibility of the indictment.

The trial court convicted all the accused under s. 396 IPC and
sentenced them to 10 years of rigorous imprisonment. In appeal, High
Court, affirmed the conviction and sentence.
c
On the questions: whether it is reasonably probable that the three
eye witnesses who had identified the culprits had known three of the
appellants and their omission to mention them in the FIR stultifies the
prosecution and, whether, even if there was no such familiarity or
because of it, the evidence of identification pursuant to a test identifica- D
tion conducted after a lapse of nearly four months after arrest, inspires
confidence.

Allowing the appeals,

HELD: 1.1 The one area of criminal evidence susceptible of E
miscarriage of criminal justice is the error in the identification of
the criminal. The evidence of test identification would call for a
careful scrntiny. [855F]

In a case where the eye witnesses, on their own admission, did not
know the accused persons before the occurrence, their identification for F
the first time in the dock after a long lapse of time would have been
improper. In suc.h cases, the test identification at an identification
parade to test the power of recollection of the witnesses assumed added
significance. [855A, C]

1.2 The evidence establishing the identity of the culprits assumes G
particular materiality in a case of decoity occurring in the darkness of
the night. [854H; 855A]

The evidence of identification merely corroborates the oral
testimony in court which alone is the primary and substantive
evidence as to identity. Absence of corroboration by test identifica- H
850 SUPREME COURT REPORTS [ 1988] I S.C.R.

A tion may not assume any materiality if either the witnesses had 'f.
known the accused earlier or where the reasons for gaining an enduring
impress of the identity on the mind and memory of the witness
are, otherwise brought-out. The test identification is itself, directed
for the purpose of reassuring that clear impression were gathered
at the time of the occurrence. [856G; 8570; Fl
B
1.3 The evidence of test identification is admissible under s. 9 of
the Evidence Act. But the valn.e of the test identification, apart
altogether from the other safeguards appropriate to a fair test of
identification depends on the promptitude in point of time with which
the suspected persons are put for the test identification. If there is
C unexplained and unreasonable delay in putting up the accused persons
for a test identification, the delay by itself detracts from the credibility -+-
(
of the test. There might conceivably be occasions when there could be
justificatioo, or acceptable explanation, for the delay. There are cases
where the delay was at the instance of the accused persons themselves,
as they expressed a choice as to the venue for the test-parade. There
D may be other causes which in the very nature of things cannot be ex-
haustively enumerated. 18550-E; 857H; 858A-B]

In the instant case, it is clear from the evidence that at least one of
the witnesses could reasonably be expected to have known some of the
appellants as residents of village in close vicinity and students of the
E same institution, is a statable probability and that his feigned ignorance
of any prior familiarity with their identity becomes irreconcilable with
what are incidents of ordinary human intercourse. The omission to
disclose their identify in the FIR would raise a reasonable doubt about
their complicity in the crime. Further, there is a serious lapse on the
part of the prosecution in putting up the suspected culprits for a test
F identification after a lapse of four months after they were arrested.
There is no explanation at all for the delay. 'The benefit of this regret-
able and wholly unexplained lack of promptitude' in holding a test
identification enures to the appellants. The evidence of test identifica-
tion lacks the requisite element of re-assurance to support the convic-
tion. A reasonable doubt arise. [857G-H; 858E]
G
Conviction and sentence of the appellants set aside and appellants
directed to be set at liberty. [858F]

Consistent with the findings in and the result of these appeals,
conviction and sentence of the non-appealing accused also set aside as
H the findings are inter-dependent and inextricably integrated and the
i
HARi NATII v. STATE OF U.P. [VENKATACHALIAH, J.) 851

~· . said accused directed to be set at liberty. [858F·Gl

Bali Ahir and Ors. v State of Bihar, AIR 1983 SC 289; Hasib v.
State of Bihar AIR 1972 SC 283, and Rameshwar Singh v. State of J &
K, (1972] AIR SC 102 referred to.

Halsbury's Laws of England IV Edn. Volume II, para 363; B
'Identification Parades //'-Criminal Law Review, [1963) page 546 and
. Glanville Williams in 'The Proof of Guilt' (Hamlyn Lectures) referred
. to,

CRIMINAL APPELLATE JURISDICTION: Criminal Appeal
No. 831, 635 and 649 of 1985.
·. '-· c
From the Judgment and order dated IO. u; 1984 of the Delhi
High Court in Criminal Appeal No. 14 of 1979.

Yogeshwar Prasad and U.S. Prasad (Not Present) for the
Appellant. ' . .D

Dalveer Bhandari for the Respondent. ·
.. i

The Judgme~t of th.e Co~rt was delivered by

VENKATACHALIAH, J. These appeals by special leave prefer- E
·red against ,the judgment dated 10/U/1984 of the High Court of
· Allahabad in Criminal Appeal No. 14 of 1979 on its file, dismissing the
appellant's appeal and affirming the conviction under Section 396, IPC
and sentence of rigorous imprisonment for ten years dated 22.12.1978
I passed against the appellants by the First Additional Sessions Judge,
! Jaunpurin S.T. No. 168 of 1974, raise the question of correctness of F
.

..,--- the principles applied to test the identification evidence.

• · 2. On the night between the 3rd and 4th November, 1973, at
about 1.30 A.M. there was a dacoity in the house of Gangaram Yadav
in Khuthana village in Jaunpur Disti;ict, Eight to ten· persons were
alleged to have participated in its commission. .· , . . . G
.. ,·
. Gangaram Yadav.was living in the house with his wife-Keshri,
his son Kamla Prasad (PW I); his daughters Amrawathi (~W 2).and
Chandravathi (PW 4);,his father·Mathura Yadav and his brother
Rajdeo. Tuat:day Gangaram Yadav and one Sunder were sleeping in
the Bythak of the house, while Mathura Yadav arid Kamla were sleep- , H
852 \ SUPREME COURT REPORTS
•
(1988] I S.C.R.
'
' of the house. Gangaram's wife and daughters
ing in another portion
A were asleep in an adjacent new construction put up by PW I. Rajdeo
was also sleeping there. ·

A !ante~ was lit in the Bythak. At the site of the pump house on
a side of the house, a 100 Watts electric bulb was burning.
B
At 1.30 A.M. the household was rudely woken-up by the barking
of dogs and the dacoits armed with weapons, including fire arms,
raided the house, in the course of which, apart from the looting of
property, both the brothers Gangaram Yadav and Rajdeo were seri-
ously injured: Both the brothers succumbed to the injuries soon there-
after. In response to the commotion and the alarm raised by the house-
C hold a number of villagers including Sanktha is stated to have assemb-
led at the spot with torches and lathis. They challenged the decoits,
who hastened away with the loot under cover of gun-fire, injuring,
Sanktha.

D Tue' F.I.R. (Exhibit 1) was lodged at the Station House at
Jalalpur at 3.45 p.m. that day. The identity of the culprits was not
disclosed; but some general physiognomic features which could be of
universal application, were mentioned. Post-mortem examinations of
the deceased-persons were conducted ori the followl!!g day i.e.
5. 11.1973.
E

· · 3. In t~e course of the investigation; on 9.11.1973 appellants and
_one Pheku Singh, who has not appealed aginst the confirmation of his
conviction by the High Court, were arrested from the villages of Mai
and Mathurapur which are said to be in close vicinity of the place of
: F _occurrence." But they were put up for test identification in an identifi- __
·cation parade only on 5.3.1974-nearly four months after their
arrest-:--held by the Executive Magistr~te (PW 6). The accused-
. persons were stated to have been picked-out by PW I to 4; Appellants,
alongwith the said Pheku Singh, were arraigned for.the offence. Their
defence was one of denial. In the trial Kamla (PW I), Amrawati (PW
G . 2), Chndravathi (PW 4) gave, what according to them, was an eye
witness account of the incident and identified the appellants as the
culprits. They claimed to have been able to recall the identify of the
·--appellants by the perception made at the time of the occurrence with
·the aid of the light of the lantern and the electric bulb. However, Sita!
(PW ·3) who _was also put up as an eye witness did not support the
· H prosecution.· ., ··"
HARi NATH v. STATE OF U.P. [VENKATACHALIAH, J.] 853

The trial culminated in a conviction and sentence of imprison-
ment for life which have, since, been affirmed in appeal by the High A
Court.

4. Shri Yogeshwar Prasad, Learned Se!]ior counsel, urged in
support of the appeals, a short, but not ineffective contention, touch-
ing the acceptability of the evidence as to identity of the culprits. B
Appellants, it transpired, were residents of and arrested from, the
villages of Mai and Mathurapur which, undisputedly, were situate in
close vicinity of the village Khuthana where the incident took place.

Appellants' endeavour in the cross-examination of P.W. l, 2 and
, 4 was directed to show that appellants being residents of a village
.~lose-by, the eye-witnesses had seen and known them earlier and the c
1 omission to mention their names in the F.I.R. detracted from the
credibility of the indictment. However, the eye-witnesses PW I, PW 2
and·PW 4 asserted ignorance on their part of any prior familiarity with
the appellants.
D
5. The question is, first, whether it is reasonably probable, that
PW I, 2 and 4 had known appellants Rajnath, Chabbinath, Ram Lach-
man and their omission to mention them in the F.I.R. stultifies the
prosecution and, secondly, whether even on the view that there was no
such prior familiarity-or more appropriately because of it-the
evidence of identification pursuant to a test identification conducted E
after a lapse of nearly four months after arrest, inspires confidence.

6. To probabalise their prior acquaintance with the eye witnes-
ses, appellants relied not only upon what, according to them, would be
a reasonable inference from the very circumstance of the parties being
residents of villages in close vicinity of each other; but also on the F
~ircumstances that PW 1 himself was a contemporary of three of the
l appellants at 'Kutir Inter College'. PW 1 admits that he studied at that
institution; but pleaded ignorance whether appellants Rajnath,
Chabbinath and Ram Lachman were his contemporaries at the col-
lege. The answers of PW 1 to the relevant questions in cross-
examination are these: G

" ..... There is a Kutir Inter College, Chakke in the
Khalispur village. This college is situated in Village Chakke

I have studied in Kutir Inter College for 3-4 years H
854 SUPREME COURT REPORTS [ 1988) I S.C.R.

I do not know either these three accused have received
A
their education in the Kutir Inter College and whether they ..,..
were studying thete at the time when I used to study there.
I do "not know as to whether these two accused Ram Lach-
man and Rajnath are the residents of village Mai and the
third accused named Chhavi .Nath in a resident of
B Mathurapur. I also do not know whether other two
accuseds named Feku arid Harinath are the residents of
village Mai ........ ". .,._

Suresh Chandra Mishra (OW 2), the Head Clerk of the Kutir
Inter College produced the "Scholar's Register" to show that appel-
C !ants Ram Lakhan and Rajnath Singh and Kamla Prasad (PW J),
deceased Rajdeo Yadav were all students at that institution. Exhibit,.+
Kha. 5, 6, 7, 8 are extracts of that Register. OW 2 also speaks to the ;
close proximity of these village to each other.

7. What emerges from the evidence is that, at least, Kamla
o Prasad Yadav (PW 1) could reasonably be expected to have known
some of the appellants, as residents of villages in close vicinity, and as
students of the same institution is a statable probability and that his
feigned ignQrance of any prior familiarity with their identity becomes
irreconciliable with what are incidents of ordinary human inter course.
The omission to disclose their identity in the F.I.R. would raise a
E reasonable doubt about their complicity in the crime. The following
observations of this Court in Bali Ahir and Ors. v. State of Bihar, AIR
1983 SC 289 are apposite:

" ......... This indicates that all was not well with the
identification. The fact that the appellants belong to the
F neighbouring village at a distance of less than a mile, that I
the witnesses who came to identify the appellants had seey
the accused from behild while escaping, that PW 2 had ,
known Bali Ahir from before, yet he did not name him in
the First Information Report and went to identify him
when he fully knew Bali Ahir, that the identification of the
G two of the appellants took place after a gap of four days
after their arrest, without explaining the cause for the
delay, speaks for itself ........ "

8. Even on the premise that there was no such prior acquaint-
ance, the evidence establishing the identity of the culprits assumes
H particular materiality in a case, as here, of a decoity occurring in the
HARi NATH v. STATE OF U.P. fVENKATACHALIAH, J.J 855

darkness of the night. The evidence of the test identification would call A
for a careful scrutiny. In a case of this kind where the eye witnesses, on
their own admission, did not know the appellants before the occurr·
ence, their identification of the accused persons for the first time in the
dock after a long lapse of time would have been improper. In Hals·
bury's Law of England (IV Edn. Volume II, para 363) this passage
occurs and is worth re-calling: B

"It is undesirable that witnesses should be asked to identify
a defendant for the first time in the dock at his trial; and as
a general practice it is preferable that he should have been
placed previously on a parade with other persons, so that
potential witness can be asked to pick him out."
c
In such cases, it is needless to say, the test identification at an
identification parade to test the power of recollection of the witnesses
assumes added significance. Prosecution has, therefore, relied upon
the result of the test identification (Exhibit 30) where the appellants
were picked out by P.W. l, P.W. 2, and P.W. 4. The conduct of an D
identification parade belongs to the realm, and is part of the investiga·
tion. The evidence of test identification is admissible under Section 9
of the Evidence Act. But the value of the test identification, apart
altogether from the other safe-guards appropriate to a fair test of
identification, depends on the promptitude in point of time with which
the suspected persons are put up for test identification. If there is E
unexplained and unreasonable delay in putting-up the accused persons

-- for a test identification, the delay by itself, detracts from the credibility
of the test.

9. The one area of criminal evidence susceptible of miscarriage
\.~f criminal justice is the error in the identification of the criminal. F
j Indeed Prof. Borchard 's "Convicting the Innocent" records several
criminal convictions in which the accused was subsequently proved
innocent. The major source of the error is to be found in the identifica·
tion of the accused by the victim of the crime. Indeed the learned
author refers to the source of mistaken identification thus:
G
" ........ The emotional balance of the victim or eye·
-..,-.; witness is so disturbed by his ext111·ordinary experience
that his powers of perception become distorted and his
' identification is frequently most untrustworthy. Into the
identification enter other motives µot necessarily stimu-
lated originally by the accused personally-the desire to H
856 SUPREME COURT REPORTS [1988) 1 S.C.R.

requite a crime, to exact vengeance upon the person "!<-
A believed guilty, to find a scape-goat, to support, con-
sciously or unconsciously, ah identification already made
by another. Thus, doubts are resolved ll{lainst the accused
"

B (See 'Identification Parades II'-Criminal Law Review, 1963-page
546) ;..-~

Glanville Williams in 'The Proof of Guilt' -(Hamlyn
Lectures )-refers to the errors of recognition breeding an invincible
assurance in the witnesses, highly deceptive for those who are not
fore~warne~ of s~ch possibilities, and excerpts Gorphe's results of a i
c continental mvest1gatton, thus: .___,-
. '
" ..... There is no difference from the subjective point of
view, between true and false recognition, so far as their
intrinsic qualities are concerned, and there are no objective
D signs to distinguish one from the other ..... The witness's
certainty may not be immediate, without this delay being
necessarily a sign of error. Nevertheless, error is more fre-
quent when recognition comes some time after seeing

E The act of recognition is very open to suggestion in all
is forms ..... .

Resemblance is a matter ·of relativity. For a white
person, all negroes are like each other, and conversely. A
person can much better distinguish those of his own age
F and condition than those of different ages and condition.
Uniform is a cause of fallacious resemblance, above all toi'-1
those who do not wear it."

(emphasis supplied)

G 10. The evidence of identification merely corroborates and
strengthens the oral testimony in Court which alone is the primary and
substantive evidence as to identity. In Hasib v. State of Bihar, (AIR
1972 SC 283 this Court observed:

" .... The purpose of test identification is to test that
H evidence, the safe rule being that the sworn testimony of
HARi NATH v. STATE OF U.P. (VENKATACHALIAH. J.I 857

the witness in Court as to the identity of the accused who is
a stranger to him, as a general rule, requires corroboration A
in the form of an earlier identification proceeding .... "

In Rameshwar Singh v. State of J & K ( 1972 AIR SC 102) this
Court observed:
B
" It may be remembered that the substantive evi-
~ dence of a witness is his evidence in court, but when the
I
accused person is not previously known to the witness con-
cerned then identification of the accused by the witness
soon after the farmer's arrest is of vital importance because
it furnishes to the investigating agency an assurance that
the investigation is proceeding on right lines in addition to c
furnishing corroboration of the evidence to be given by the
witness later in court at the trial ..... "

I I. It is, no doubt, true that absence of corroboration by test
identification may not assume any materiality if either the witness had D
known the accused earlier or where the reasons for gaining an endur-
ing impress of the identity on the mind and memory of the witness are, '
otherwise, brought-out. It is also rightly said that

"Courts ought not to increase the difficulties by magnifying
the critical possibilities. It is their province to deal with E
matters actual and material to promote order and not sur-
render it by excessive theorising or by magnifying what in
practice is really unimportant".

But what Shri Dalveer Bhandari contends for is that the test
identification in the case was itself unnecessary and superfluqus be- F
cause there was sufficient opportunity and occasion for the witnesses
to get a clear impression of the identity. This cannot be accepted.
Indeed, the test identification is itself, directed for the purpose of
reassuring that clear impressions were gathered at the time of the
occurrence.
G
12. We are afraid in the present case, there is a serious lapse on
the part of the prosecution in putting-up the suspected culprits for a
test identification after a lapse of four months after their arrest. There
is no explanation at all for the delay. There might conceivably, be
occasions when there could be justification, or acceptable explanation,
for the delay. There are cases where the delay was at the instance of H
858 SUPREME COURT REPORTS (1988] I S.C.R.

A the accused-persons themselves as they expressed a choice as to the
venue for the test-parade. There may be other causes, which in the
very nature of things cannot be exhaustively enumerated. Cases can
only be illustrative.

Under similar circumstances, this Court in Soni v. State of Uttar
B
Pradesh, [ 1982] 3 SCC 368 obseryed:

"2 ...... The conviction rests purely upon his identifica-
tion by five witnesses Smt. Koori, Pritam Singh, Kewal
Chaitoo and Sinru, but it cannot be forgotton that .the
identification parade itself was held after a lapse of 42 days
c from the date of arrest of the appellant. This delay in hold- ~
ing the identification parade throws a doubt on the genui-
neness thereof apart from the fact that it is difficult that
after lapse of such a long time the witnesses would be re-
membering the facial expression of the appellant. If this
evidence cannot be relied upon there is no other evidence
D which can sustain the conviction of the appellant. We,
therefore, allow the appeal and acquit the appellant."

The benefit of this regrettable and wholly unexplained lack of
promptitude in holding the test identification, we are construed to say,
E cnures to the appellants. The evidence of test identification lacks the
requisite element of re-assurance to support the conviction. A reason-
able doubt arises.

13. Accordingly, these appeals are allowed, the conviction and
sentence of the appellants in S.T. 168 of 1974 on the file of the First
Additional Sessions Judge, Jaunpur, and affirmed in the Criminal
F Appeal No. i4 of 1979 on the file of the Allahabad High Court are set ·
aside and the appellants are directed to be set at liberty forthwith. The
conviction and sentence of the non-appealing accused, namely, Pheku
Singh cannot also be sustained consistent with· the. findings in and the
result of these appeals as the findings are inter-dependant and inextric-
ably integrated. The conviction and sentence of Pheku Singh are also
G set aside and the said Pheku Singh, the accused No. 3 in S.T. No. 168
of 1974 is also directed to be set at liberty forthwith.

N.P.V. Appeals allowed.