# KASHMIRA SINGH v. STATE OF MADHYA PRADESH

- **Citation:** [1952] 1 S.C.R. 526
- **Court:** Supreme Court of India
- **Decided:** 1951-06-08
- **Case number:** Criminal Appeal No. 53 of 1951
- **Bench:** SAIYID FAZL Au, MuKHERJEA, v IVIAN BosE
- **Source:** https://unisonlegal.in/judgment/supreme-court-of-india/kashmira-singh-v-state-of-madhya-pradesh-47
- **Pages:** 18

## Headnote

Indian Evidence Act
(1 of 1872),
ss. 3, 30-Confession of
co-Occused-Evidentiary
value-Evidence
of
accomplice--Necessity
of ·corroboration-Confession-Practice of exan1ining magistrate whorecorded the confessi·on.
The confession of an accused person against a co-accused is
not evidence in the
ordinary sense of
the term.
It does not
come within the meaning of evidence contained in sec. 3 of the
Indian Evidence Act inasmuch as it is not required to be given
on oath, nor in the presence of the accused and cannot be tested
by cross-examination.
It is a
much
weaker type of evidence
than the evidence of an approver which is not subject to any of
these infirmities.
Such a confession can only be used to lend assurance to other
evidence against a co-accused. The proper way to approach a case
of this kind is, first!, to marshal the evidence against the accused
excluding
the
confession altogether from
consideration and see
whether, if tt is believed, a conviction could safely be based on
it. If it is capable of belief independently of the confession, then
it is not necessary to call the confession in aid. But cases may
arise where the judge is not prepared to act on the other evidence
as it stands even though, if believed,
it would be sufficient to
sustain a conviction.
In such an event the judge may call in
ald the confession and use it to lend assurance to the other
evidence and thus fortify himself in believing what without the
aid of the confession he would not be prepared to accept.
Bhuboni Sahu v. The King (76 I.A. 147) reiied upon. Emperor
v. Lalit Mohan Chukerbutty (38 _CaL 599 at 588) and In re Periyaswami Moopan (I.L.R. 54 Mad. 75) referred to.
A conviction
can be based
on the
uncorroborated testimony
of an accomplice provided the judge has the rule of caution,
which experience dictates, in mind.
Rameshwar v. State of Rajasthan [1952] S.C.R. 377 referred to,
The rule of caution is that save in exceptional circumstances
one accomplice cannot be used to corroborate another, nor can
he be used to corroborate a persoD. who though not an accomplice is no· more
~liable. than one. ·
.
~.
.! ..
...
t·
It is not proper or
desirable for the prosecution to. examine
as a witness the magistrate who recorded the confession.
).
-
-
-
' •
'
.-S.C.R.
SUPREME COURT REPORTS
527
CRIMINAL
APPELLATE
JURISDICTION: Criminal Appeal No. 53 of 1951.
Appeal by special leave from the
Judgment and Order dated the 8th June
1951 of
the
High Court of Judicature at Nagpur (Hemeon and
Rao JJ.) in Criminal Appeal No. 297 of 1950, arising
out of the Judgment and Order dated the 11th September 1950 of the
Court of the Additional Sessions
Judge of
Bhandara in Session Trial No. 25 of 195(1.
Bakshi Tek Chand, (Gopal Singh, with him) for the
appellant.
S. K. Kapoor, for the respondent.'
1952. March 4. The
Judgment of the Court was
delivered by
BosE J.-The appellant Kashmira Singh
has been
convicted of the murder of one Ramesh, a small boy
aged five,
and has been sentenced to death.
He was
granted special leave to appeal.
Three other persons
were tried along with him. They were his brother
Gurudayalsingh, his
nephew
Pritipalsingh
(son
of
Gurudayal), a boy of eleven, and one Gurubachansingh.
Gurudayal and Pritipal have been acquitted. Gurubachansingh confessed and was convicted. He
was also
sentenced to death. He has not appealed here.
The murder was a particularly cruel and revolting
one and for that reason it-will be nececssary to examine
the evidence with more
than ordinary care lest
the
shocking nature of the cnme induce an instinctive
reaction
against a dispassionate judicial scrutiny
of
the facts and law.
.
The prosecution case is this. The deceased Ramesh
was the son of P.W. 48 L.P. Tiwari who was the Food
Officer at Gondia at the relevant date.
The appellant
Kashmira Singh was an Assistant Food Procurement
Inspec~or there. On the 1st of July, 1949, Tiwari
found the appellant and Harbilas
(P.W. 31)
getting
rice polished at a certain rice mill. At that date the
poli

## Text

1952
March-4.
526
SUPREME COURT REPORTS
(1952}
KASHMIRA SINGH
v.
STATE OF MADHYA PRADESH
[SAIYID FAZL Au,
MuKHERJEA
and v IVIAN BosE, J.J.]
Indian Evidence Act
(1 of 1872),
ss. 3, 30-Confession of
co-Occused-Evidentiary
value-Evidence
of
accomplice--Necessity
of ·corroboration-Confession-Practice of exan1ining magistrate whorecorded the confessi·on.
The confession of an accused person against a co-accused is
not evidence in the
ordinary sense of
the term.
It does not
come within the meaning of evidence contained in sec. 3 of the
Indian Evidence Act inasmuch as it is not required to be given
on oath, nor in the presence of the accused and cannot be tested
by cross-examination.
It is a
much
weaker type of evidence
than the evidence of an approver which is not subject to any of
these infirmities.
Such a confession can only be used to lend assurance to other
evidence against a co-accused. The proper way to approach a case
of this kind is, first!, to marshal the evidence against the accused
excluding
the
confession altogether from
consideration and see
whether, if tt is believed, a conviction could safely be based on
it. If it is capable of belief independently of the confession, then
it is not necessary to call the confession in aid. But cases may
arise where the judge is not prepared to act on the other evidence
as it stands even though, if believed,
it would be sufficient to
sustain a conviction.
In such an event the judge may call in
ald the confession and use it to lend assurance to the other
evidence and thus fortify himself in believing what without the
aid of the confession he would not be prepared to accept.
Bhuboni Sahu v. The King (76 I.A. 147) reiied upon. Emperor
v. Lalit Mohan Chukerbutty (38 _CaL 599 at 588) and In re Periyaswami Moopan (I.L.R. 54 Mad. 75) referred to.
A conviction
can be based
on the
uncorroborated testimony
of an accomplice provided the judge has the rule of caution,
which experience dictates, in mind.
Rameshwar v. State of Rajasthan [1952] S.C.R. 377 referred to,
The rule of caution is that save in exceptional circumstances
one accomplice cannot be used to corroborate another, nor can
he be used to corroborate a persoD. who though not an accomplice is no· more
~liable. than one. ·
.
~.
.! ..
...
t·
It is not proper or
desirable for the prosecution to. examine
as a witness the magistrate who recorded the confession.
).
-
-
-
' •
'
.-S.C.R.
SUPREME COURT REPORTS
527
CRIMINAL
APPELLATE
JURISDICTION: Criminal Appeal No. 53 of 1951.
Appeal by special leave from the
Judgment and Order dated the 8th June
1951 of
the
High Court of Judicature at Nagpur (Hemeon and
Rao JJ.) in Criminal Appeal No. 297 of 1950, arising
out of the Judgment and Order dated the 11th September 1950 of the
Court of the Additional Sessions
Judge of
Bhandara in Session Trial No. 25 of 195(1.
Bakshi Tek Chand, (Gopal Singh, with him) for the
appellant.
S. K. Kapoor, for the respondent.'
1952. March 4. The
Judgment of the Court was
delivered by
BosE J.-The appellant Kashmira Singh
has been
convicted of the murder of one Ramesh, a small boy
aged five,
and has been sentenced to death.
He was
granted special leave to appeal.
Three other persons
were tried along with him. They were his brother
Gurudayalsingh, his
nephew
Pritipalsingh
(son
of
Gurudayal), a boy of eleven, and one Gurubachansingh.
Gurudayal and Pritipal have been acquitted. Gurubachansingh confessed and was convicted. He
was also
sentenced to death. He has not appealed here.
The murder was a particularly cruel and revolting
one and for that reason it-will be nececssary to examine
the evidence with more
than ordinary care lest
the
shocking nature of the cnme induce an instinctive
reaction
against a dispassionate judicial scrutiny
of
the facts and law.
.
The prosecution case is this. The deceased Ramesh
was the son of P.W. 48 L.P. Tiwari who was the Food
Officer at Gondia at the relevant date.
The appellant
Kashmira Singh was an Assistant Food Procurement
Inspec~or there. On the 1st of July, 1949, Tiwari
found the appellant and Harbilas
(P.W. 31)
getting
rice polished at a certain rice mill. At that date the
polishing of iice was prohibited by a State law.
Tiwari accordingly reported the matter to the
Deputy
Commissioner
of
Bhandara.
He
suspended
the
1952
Kashmira Singh
v.
State of Madhya
Pradesh.
1952
Kashmira Singh
v.
St•t•. of Madhya
Praduh.
Bose /.
528
SUPREME COURT REPORTS
[1952]
appeJlant and latter his services were terminated by an
order of the State Government with effect from the
7th of July.
The orders were communicated on the
17th of November. This embittered the appellant who
on at least two occasions was heard to express a determination to be revenged.
In pursuance of this determination he got into touch
with the
confessing
accused
Gurubachansingh
and
enlisted his services for murdering the boy Ramesh.
On the 26th of December,
1949, festivities and religious ceremonies were in progress all day in the Sikh
Gurudwara at Gandia. The boy Ramesh was there in
the morning and from there was enticed to the house
of the appellant's brother Gurudayalsingh
and
was
done to death in a shockingly revolting fashion by
the appeJlant, with the active assistance of Gurubachansingh, in the middle of the day at about 12 or
12-30. The body was then tied up in a gunny bag and
rolled up in a roll of bedding and allowed to lie in
Gurudayal's house till about 7 p.m.
At 7 p.m. the body wrapped as above was carried
by Gurubachan on his head to a chowkidar's hut near
the Sikh Gurudwara. The appe'lant accompanied him.
The map, Exhibit P-18A, shows that the distance along
the route indicated was about half a mile to three quarters of a mile. It was left there till about midnight.
Shortly before midnight the appellant and Gurubachan engaged the services of a rickshaw coolie Shambhu alias Sannatrao, P. W. 14. They took him to the
chowkidar's hut, recovered the bundle of bedding and
went in the rickshaw to a well which appears from the
map, Exhibit P-18A, to be about half a mile distant.
There the body was thrown into the well. That m
brief is the prosecution case.
Gurubachan's confession has played an important
part .in i~pli<lating the appetliant, and the question at
once arises, how far and in what way the confession
of an accused person can be used against a co-accused?
It 1s evident that it is not evidence m the ordinary
-
..
-
)
-
S.C.R.
SUPREME COURT REPORTS
529
sense of the term because, as the Privy Council say in
Bhuboni Sahu v. The King(1):-
1952
Kashmira Singh
"It does not indeed come within the definition of
v.
'evidence' contained in section 3 of the Evidence Act. State of Madhya
It is not required to be given on oath, nor in the prePradesh.
sence of the accused, and it cannot be tested by crossBose /.
examination."
Their Lordships also point out that it is
"obviously evidence of a very weak type ..... .It is
a much weaker type of evidence than the evidence of
an approver, which is not subject to any of those
infirmities."
They stated in addition that such a confesision cannot be made the
foundation of a
conviction and can
only be used in "support of other evidence." In view
of these remarks it would be pointless to cover the
same ground, but we feel it is necessary to expound
this further as misapprehension still exists. The question is, in what way can it be used in support of other
evidence? Can it be used to fill in missing gaps? Can
it be used to corroborate an accomplice or, as in the
present case, a witness who, though not an accomplice,
is placed in the same category regarding credibility
because the judge refuses to believe him except in so
far as he is corroborated?
In our opinion, the matter was put succinctly by
Sir Lawrence Jenkins in Emperor v. Lalit Mohan
Chuckerbutty(2) where he said that such a confession
can only be used to "lend assurance to other evidence
against a co-accused" or, to put it in another way, as
Reilly J. did in In re Periyaswami Moopan(3)·-
"the provision goes no further than this-where
there is evidence against the co-accused sufficient, if
believed, to support his conviction, then the kind of
confession described in section 30 may be thrown into
the scale as an additional reason for believing that
evidence."
(I) [1949] 76 I. A. 147 at 155.
(3J [1931] J. L. R. 54 Mad. 75 at 77.
(2) [ 19 l l] I. L. R. 38 Cal.559 at 588.
1952
Kashmira Singh
v.
. State of Madhya
Pradesh.
Bose/.
530
SUPREME COURT REPORTS
[1952]
Translating these observations into concrete terms
they come to this. The proper way to approach a case
of this kind is, first, to marshal the evidence against
the accused excluding the confession altiogether from
consideration and see whether, if it is believed, a conviction could safely be based on it. If it is capable of
belief independently of the confession, then of course it
is not necessary to call the confession in aid. But cases
may arise where the judge is not prepared to act 011
the other evidence as it stands even though, if believed,
it would be sufficient to sustain a conviction. In
such an event the judge may call in ail the confession
and use it to lend assurance to the other evidence and
thus fortify hi=elf in believing what without the aid
of the confession he would not be prepared to accept.
Then, as regards its use in the corroboration of
accomplices and
approvers. A co-accused who confesses is naturally an accomplice and the danger of
using the testimony of one accomplice to corroborate
another has repeatedly been pointed out. The danger
is in no way lessened when the "evidence" is not on
oath
and
cannot
be
tested
by
cross-examination.
Prudence will dictate the same rule of caution in the
case of a witness who though not an accomplice is regarded by the judge as having no greater probative
value. But all these are only rules of prudence. So far
as the law is concerned, a conviction can be based on·
the uncorroborated testimony of ·an accomplice provided the judge has the rule of caution, which experience dictates, in mind and gives reasons why he thinks
it would be safe in a given case to disregard it. Two
of us had occasion to examine this recently in Rameshwar v. The State of Rajasthan('). It follows that the
testimony of an accomplice can in law be used to
corroborate another though it ought not to be so used
save
in
exceptional
circumstances
and
for
reasons
disclosed. As the Privy Council observe in Bhuboni
Sahu v. The King(2):-
"The tendency to include the innocent with the
guJ'.lty is peculiarly preval'ent .in India, as judges have
ti) [1952] S.C.Jl.. 377.
(2) [1949] 76 I.A. 147 at 157.
-
.,,
-
-
}
-
'S.C.R ..
SUPREME COURT REPORTS
531
noted on innumerable occasions, and it is very difficult
for the court to guard against the danger ......•. The
-0nly real safeguard against the risk of condemning
. the innocent with the guilty lies in insisting on independent evidence which in some measure implicates
such accused."
Tur~ing now to the facts of the present case. The
evidence on which the prosecution relies, apart from
the confession, is this:-
(1) Previous association
between
Gurubachan and
the appellant.
The only evidence about t~ is P.W. 23 Upasrao,
a water carrier. He speaks of three meetings and is
curiously definite about days of the week and times
though he did not know on what day of the week
diwali fell nor could he give the names of anybody
else he met on those occasions. However, for what it
is worth, he says he saw them talkiing (1) three weeks
before the murder, (2) on the 24th and (3) on the 25th.
They SIPoke in Punjabi which he does not understand,
. but on the second occasion he heard them mention the
name of Ramesh. Two of these meetings, namely the
first and the third tally with two of the only three
meetings described in the confessiqn. It is proved
that the witness did not disclose these facts to the
police but despite that the Sessions Judge believed
him because of the confession. The High Court appear
to have disbelieved him, for in paragraph 37 of the
judgment the learned judges point out that he is contradicted by his own statement to the police. There
his story was that the three brothers met and not
Gurubachan and the
appellant. This evidence
can
therefore be disregarded and consequently the confession cannot be med to prove previous association.
It was argued however that if it is proved that the
appellant helped in disposing of the body after the
murder,
then their previous asssociation can be inferred
because one would
hardly seek the assistance of a
stranger for a task like that. That has some force but
the weakness of that in this case lies on the fact that,
1952
Kashmira Singh
v •
State of Madhya
Pradesh.
Bose].
1952
Kashmira Singh
v.
State ·of Madhya
Pradesh.
Bos~/.
532
SUPREME COURT REPORTS
[1952]
according to the prosecution case, as disclosed in the
confession, Gurubachan was
a stranger to Gondia.
He had come there only six weeks before the murder
and did not meet the appellant till three weeks later
and then only casually. Their second meeting, equally
casual, was on the 21st, that is, five days before the
murder, and on that date the appellant is said to have
disclosed his intention to this stranger whom he had
only met once before.
It is true this stranger knew the
appellant's brother, but how? The brother was a travelling ticket inspector on the railway and used to allow
Gurubachan to travel
without a ticket,
presumably
because he wa~ also a Sikh. If probabtlities are to be
calkd in aid, the story disclosed in the confession has
distinct
weaknesses,
particularly
as
Gurubachan's
· assi:stance was
wholl~ unnecesisaty). If tlie 1colnfession
is true there was a well thought out plot timed with
the precision almost o£ a minor military operation.
At a given moment the newphew Pritipal was to decoy
the deceased away from his companions and isolate
him. Then, after leading him several hundred yards
down the road, hand him over to Gurubachan. Gurubachan. was to take him down to point No. 6 on the
map well over half a mile from the spot where he took
over from Pritipal. In the meanwhile, the appellant
w~ to walk another half miEe at right angles to
Gurubachah's course to the point No. 15 to hire a
cycle. From there he was to cycle dose on a mile to
point No. 6 and meet Gurubachan and the boy. As
the learned High Court Judges,
who made a spot inspection, point out, the route would lie
through a
crowded bazaar locality. From point No. 6 Gurubachan was to hand over the child to the appellant who
_was to cycle with him close on a mile to his brother
Gurudayal' s quarters, point No. 16, through this same
crowded bazaar. In the meanwhile, Gurubachan was
to walk back to his house (No. 17) and pick up a
chisel and a piece of wire for the purpose of the murder
and rejoin the appellant at Gurudayal's house. As
will be seen, the timing would have to be within fairly
close tolerances. Then, at
the murder itself, what
\.
-
\.,
-
}
-
S.C.R.
SUPREME COURT REPORTS
533
assistance did Gurubachan give ?
Nothing which a
grown man could not easily have accomplished himself on a small helpless victim of five. The appeUan~
could have accomplished all 1tihis as easily without the
assistance of
Gurubachan, and
equally Gurubachan,
a mere hired assassin could have done it all himself
without the appellant running the risk of drawing
pointed attention to himself as having been last seen
in the company of the boy. We hold that previous
association of a type which would induce two persons
to associate together for the purposes of a murder is
not established.
(2) That the deceased Ramesh was in the Gurudwara about 9-30 or 10 in the morning of the 26th.
This is not disputed.
(3) That Kashmira Singh who had gone to the
Gurudwara in the morning was absent between 11 A.M.
and 12-45 P.M.
That the appellant was at the Gurudwara in the
morning is not disputed, in fact his case is that he
was there the entire day. The evidence to prove that
he left it between these hours consists of three persons: P.W. 30 Atmaram, P.W. 35 TiLakchand and
P.\V. 5 Bisan.
The prosecution story is that the appellant left the
Gurudwara about 11 A.M. to go to the shop of P.W.
5 Bisan to hire a cycle. He was first seen by P.W. 35
Tilakchand, a wood stall keeper, at point No. 13, just
near the Gurudwara. The witness places the time at
about 10-30 or 11 A.M. He says he saw him coming
from the direction of the railway station and going
past his stall. Fifteen minutes
later, he went past
his stall again in the opposite direction, that is to say,
towards the railway station which lies on his way to
the cycle shop.
Next comes P.W. 30 Atmaram. He keeps a bookstall on the broad gauge platform of the Gondia Railway Station. He says he saw the appellant coming
from the bridge and going towards the Railway Police
1952
Kashmira Singh
Y.
State of Madhye1
Pradesh.
Bose J.
534
SUPREME COURT REPORTS
[1952]
1952
Station of all places m the world. He came near
Kashmira Singh enough the witness to wish him good day. He places
y,
the time at about 10-30 or 11. The only comment
State of Madhya we make on this witness is that he says he used to
Pradesh.
see the appellant at ,the station almost every day and
Bose /.
they used to greet each other. The possibility that
the witness is mixing up this day with one of the
other days cannot be excluded. It is certainly a matter
for comment that a would be murderer on his way to
hire a cycle for the purpose and keep an assignment
with his accomplice and victim should go out of his
way and either go on to or very near the railway platform to gret;t a person he knows there and then walk
away towards the police station of all places where the
danger of recognition would be strong.
Next there is P.W. 5 Bisan, the man ii) charge of
the cycle shop. He speaks from his register and says
the appellant hired a cycle from him on that day at
11-20 A.M. and returned it at 12-45 P.M. The Sess10ns Judge and the High Court lay great stress on
this witness.
But as against this 1s the evidence of , Anupsingh
Bedi, D.W. 1,
a m;pectable disinterested witness, who
is a resident of Nagpur. He says he saw the appellant at
the Gurudwara at 11 and agam "about 11-45 A.M."
The Sessions
Judge thought he was interested because
he admits he reported a complaint he had received
from Gurudayalsingh, to the effect that the appellant
was being harassed by the police and that they threatened to arrest 1adie5 also, to the Inspector General of
Police and .the Home Minister. He explained that as
head of the Sikh community in that State he felt
bound to pass on these complaints to the highest
authorities. We are unable to regard this as disclosing interest. There is no suggestion that what he did
wa~ improper and we are of opinion he did nothing
more than any man of responsibility in his position
would have done. The High Court has not criticised
him. The learned Judges merely say that he may be
mistaken as to the time; nor of course does he suggest
that he is giving more than a mere estimate. All he
~.
-
)
S.C.R.
SUPREME COURT REPORTS
535
says is that, "It may have been about 11-45 A.M. by
this time."
We do not think there is much in all thisi. Nobody,
except P.W. 5 Bisan, pretends to be exact and when
one is guessing at the time several days
after the
event there
really is not much discrepancy between
11-20 and 11-45. Even if it was 11-45 there would
still have been sufficient time to commit the murder.
As two Courts have believed the evidence on this point
without calling in aid the confession, we are not prepared to depart from our usual rule regarding concurrent findings of fact. We will therefore accept the
position that the appellant was absent from the Gurudwara long enough to enable him to commit the
murder.
We will also take into consideration the fact
that he made a false statement on this point when he
said he was not away at all.
(4) Disposal of the body.
The rest of the · evidence relates to the disposal of
the body and the only direct evidence connecting the
appellant with this, apart from the confession, is that
of Sannatrao P.W. 14, the rickshaw .coolie. He does
not bring the appellant into the picture till about midnight. Now this coolie is a very shaky witness. We
cannot but note the remarkable series of coincidences
which emerge from his testimony. First, he is not a
rickshaw coolie at all. He merely happened to hire
a rickshaw that night, and he told the police that this
was the first time he had ever done that at night after
)'!,
a day's work. Next, he knew the appellant because
he happened to be a chowkidar in the Food Office at
Gondia at the same time that the appellant was there
·as a Food Inspector.
But at the date of the incident
neither was sitill in service, so by a . somewhat strange
:coincidence the appellant happens to hire, for the first
time, this old co-worker in the middle of the night
who, in his turn, happened to hire, also for the first
time at night, a rickshaw for which he had no licence.
N;ext comes a still stranger coincidence. He is taken
...4
to within a few paces of his own house and the body
1952
Kashmira Singh
v.
State of Madhya
Pradesh.
Bose /.
536
SUPREME COURT REPORTS
[1952]
is dumped, in his presence, into a well, a stone's
. throw from where he lives.
Gurubachan tells us that
Kashmira Singh
v.
earlier in the day, about 7 P.M. he, (Gurubachan) had
1952
State of Madhya carried, unaided, the "bedding" on his head for a
Pradesh.
distance which we know was half to three quarters of
Bose /.
a mile, namely from Gurudayal's house to the chowkidar's hut. Despite this, the two are sa.id to have
engaged this rickshaw coolie to carry it j\l.5/1: half a
mile (,i shorter distance) to the well and there they
threw it in in the man's presence; and none of this
was disclosed to the police till a month la,ter, namely
the 17th of January, though the witness was present
when the body was recovered and though he was
questioned on three previous occasions.
We do not doubt that a rickshaw was used because
rickshaw tracks were discovered by the well long before anybody had suggested that a rickshaw had been
used. But we find it difficult to resist the inference
that this witness was an accomplice so far as the disposal df
•th~
b~dy was concdrned. Consequent'J,y, he
is in much the same category so far as credibility is
concerned. That brings us at once to the rule that
save in exceptional circumstances one accomplice can- ·
not be used to corroborate another, nor can he be used
to corroborate a person who though not an accomplice
is no more reliable than one. We have therefore either
to seek corroboration of a kind which will implicate
the appellant apart from the confession or find strong
reasons for
using Gurubachan's confession for that
purpose. Of course, against Gurubachan there is
no
difficulty,
but against the appellant the pos,ition is not
as easy.
We will therefore examine the reliability of Gurubachan's confession against the appellant. Now there
are some glaring irregularities regarding t'ltls confession and though it was safe for the Se&',ions Judge
and the High Court to act on it as against Gurubachan because he adhered to it throughout the sessions
trial despite his pleader's efforts to show the contrary,
a very djffereqt position emerges when we come to the
•
..,
appella11t.
~
-
-
J
-
S.C.R.
SUPREME COURT REPORTS
537
1952
The first point which emerges regarding this is
that the confession was not made till the 25th of
February 1950, that is to say, not until two mo,nths
after the murder.
Kashmira Sing1'
v.
State of Madbyf!
We do not know when Gurubachan was first interrogated but P.W. 42 Narayandas tells us that when
he was taken to the police station house at Gondia
for interrogation about the 1st or 3rd January he saw
Gurubachan sitting in the police lock up. We do not
know how long he was kept there like this but it is
evident that he was not there voluntarily, at any rate
till the 1st or 3rd, because the Station Officer P.W. 44
says that "until Gurubachan Singh was arrested he
used to be allowed to go home." Also he says that
Gurubachan was interrogated
several times and was
confronted with Pritipal.
However, eventually Gurubachan was allowed to go
away and he went to Balaghat. Then, on the 16th of
February the Station Officer P.W. 44 went to Balaghat,
brought Gurubachan back with him to Gondia and
handed him over to the C.l.D. Inspector Guha. Guha
P.W. 50 tells us that from then till the 20th of
February, when he was arrested, he was kept · under
observation but was allowed to . go home at night. He
did not confess till the 25th and the Station Officer
P. W. 44 tells us that from the 20th to the 25th he
was kept in one of the rooms in Guha's quarters.
Then, after the confession on the 25th he was taken
back to Guha's custody for a couple of days and then
only was he sent to the magisterial lock up. (See
Guha's evidence). He was kept in this lock up till the
conclusion of the committal proceedings, that is, till
the 30th of June, instead of being sent to jail custody
in Bhandara where there is a jail. The other accused
including Pritipal who had by then confessed were
sent to Bhandara.
Now though Gurubachan was kept in the magisterial lock up the distinction between the magisterial
lock
up and
police
custody< 'in
Goncfi/a
is only
Pradesh.
Bose J.
1952
Kashmira Singh
...
State of Madhya
Pradesh.
Bose/.
538
SUPREME COURT REPORTS
[1952J
theoretical. In practice,
it is
no better than police
custody. Police constable
Lalbahadur P.W. 55 tells
us that-
"The Station House Officer ·Gondia deputes constables for duty m the lock up. The constables m
charge take the prisoners out to the latrine and also
arrange for their food .... The Head Constable in fact is
in charge."
Also, Guha admits that
he interrogated
Gurubachan in the lock up twice within the ten days which
succeeded the confession. This 1S in disregard of the
Rules and Orders (Criminal) of the Nagpur High
Court which enjoin at page 25, paragraph 84, of the
1948 edition that-
"After a prisoner has made a confession before a
a magistrate he should ordinarily be · committed to
jail and the magistrate should note on the warrant
for the iinformation of the Superintendent of the jail
that the prisoner has made a confession."
No explanation has
been given why these directions,
which were made for good reason, were disregarded in
Gurubachan' s case. As we have said, the other prisoners were all committed to jail custody in the usual
way, so there was no difficulty about observing th.e
rule. All this makes ·it unsafe to disregard the rule
about usmg accomplice testimony
as
corroboration
against a non-confessing accused.
None of· the judges
who have handled this case has given any reason why
this rule could safely be departed from in this particular case. In the circumstance$, we do not feel that
the confession by itself can be used to corroborate the
rickshaw coolie
Sannatrao, P.W. 14. But
there
IS
other corroboration. It consists of the . san border,
and this IS the next point on which the prosecution
relies.
There IS one arugment about this confession advanced on behalf of the appellant with which we shall
have· to deal. The prosecution were criticised for not
calling the magistrate who recorded the
confession as
a witness.
We wish to endorse the remarks of their
-
-
; .
-
S.C.R.
SUPREME COURT REPORTS
539
Lordshilps of
the
Privy Council
in Nazir Ahmad
v. King Emperor (1) regarding the undesirability of
such a practice. In our opinion, the magistrate was
rightly not called and it would have been improper
and undesirable for the prosecution to have acted
otherwise.
(5) Sari borders, Articles F, G, and T.
Articles F & G are two pieces of a sari border which
were used for tying up the mouth of the gunny bag
in which the body was placed. The evidence about
that is beyond doubt. Article T is another piece of a
sari border which was found in the appellant's hou~e
on the 30th of December, 1949. It is true the appellant was not present at the time but his mother was
there and it will be seen di.at it was scized on the
same day that the body was discovered. There is
strong proof that Articles F and G are a part of the
same border as Article T, and as there is a concurrent
finding regarding these facts we are not prepared to
to
take
a
different
view. That
therefore
affords
corroboration of Sannatrao's evidence and the confession can be called in aid to lend assurance to the inference which arises from these facts, namely that the
appellant did help to dispose of the body. The High
Court and the Sessions Judge were accordingly entitled
to act on thils evid~nc:e for establishing that panicular
fact and we are not prepared to disturb their concurrent conclusions.
But the matter cannot be carried
further because, not only are the sari borders not proved to have had any connection with the crime of
murder but the confession shows that they did not.
The only dondlusJon permi~ble on these facts is tl1at
the appellant, at some time which is unknown, subsequent to the murder assisted either actively or passively in tying up the gunny bag in which the corpse
was placed and that he then accompanied Gurubachan
in the rickshaw from the chowkidar's hut to the well
in the middle of the night.
(6) Coat, Article X, and Safa, Article Y.
(1) A.I.R. 1936 P.C. 253 at 258.
5--6 S. c. India/71
1952
Kashmira SitJgh
v.
State of Madhya
Pradesh.
Bose J.
1952
Ka;hmira Singh
v.
State of Madhya
Pradesh.
Bose /.
540
SUPREME COURT REPORTS
[1952]
These were seized on the 20th of January 1950
from a trunk in the house of the appellant's brother
Gurudayalsingh. The appellant's house is not in this
neighbourhood. It is some distance away in another
part of the town. The coat is a uniform coat of the
kind worn by a Travelling Ticket Inspector on the
Railways. Gurudayal is a travelling Ticket Inspector.
The appellant is not.
Here
again the appellant was
not present when the seizures were made.
This coat and safa were recovered in the fourth
search. The first search was on the 30th of December
1949.
The next on the 10th of January 1950.
The
third on the morning of the 20th and the fourth in the
afternoon of the 20th. These Articles were not found
in the first three searches.
The· Chemical Examiner reports that there is one
minute blood stain on the safa and some (the number
is not given), also minute, on the coat. The seizure
memo, Ex. P-55, picked out only five. Those stains
are not proved to be of human blood.
Now there is next no evidence to connect either
the coat or the safa with the appellant. The High
Court has relied on the evidence of Sannatrao (P.W.
14), Gokulprasad the Station Officer (P.W. 44) and
Tiwari (P.W. 48). Sannatrao does no more than say
that he noticed the appellant wearing a popat coloured safa and a black coat. But he was not able to
describe the clothes of the passenger he had carried
immediately before the appellant, nor was he able to
describe the appellant's coat
i!ri
detail. That therefore is no identification of this coat with the one the
appellant wore or owns. The Station Officer Gokulprasad said that he had seen the appellant wear this
very coat and safa and therefore he identified them as
his
clothes. In cross-examination
he
admitted
that
he had only seen the appellant on three occasions but
not to speak to. Consequently, that is not strong evidence of identification. But what in our opinion is
almost conclusive
against
this identification
is
that
Tiwari, P.W. 48, who is clearest on the point and who
of course
had the best opportunities. for observation,
. l
-
-
-
• ..,,
S.C.R.
SUPREME COURT REPORTS
541
gives
a distinctive feature of the appellant's coat,
namely that it had only one button. That is one of
his reasons for knowing what the appellant used to
wear. But the seizure memo, Ex. P. 55, shows that
the coat, Article X, had two buttons. In the circumstances we find it difficult to see how i~ can be the appellant's coat.
There is another strong point in the appellant's favour which the High Court has not noticed. P.W.
35 the wood stall keeper Tilakchand, who saw him on
his way to pick up his victim, is definite that the appellant was not wearing a coat at the time.
It is difficult to see why he should have donned a coat and got
it stained with blood just for murdering a child of
five. It our opinion, it would be unsafe to conch1de
on this evidence that any connection is established
between the coat and the safa and the appellant.
The furthest point to which this evidence can be pushed is to indicadte that the appellant possessed a coat
similar to Article X but which was not Article X.
We do not ordinarily interfere with a concurrent
finding of fact but when the finding omits to notice
these two very important points in the accused's favour which, in our opinion, swing the balance the other
way, we are unable to let the finding stand. In our
opinion, the nexus between the appellant and the coat
and the safa is not established.
(7) Motive.
This is the last piece of evidence on which the prosecution rely. Both courts hold that the motive is
established and there is strong evidence to prove it
We accordingly accept the finding that the appellant
had a motive for enmity against Tiwari and that he
had expressed
~ determination to be revenged. The
only comment we will make is that other persons who
were also dismissed from service had similar motives.
What then is the summary of the evidence ? In the
appellant's favour there are the facts that there is no
proof of his having been last seen in the company of
70
195Z
Kashmira Singh.
v.
State of Madhya
Pradesh
Bose J.
1952
Kaslimira Singh
v;,·
State of Madhya
Pradesh.
Bou·f.
542
SUPREME COURT REPORTS
[1952]
the deceased. The only evidence of the boy's movements is that of Krishna (alias Billa) P.W. 9, a boy
of seven years, and all he says is that Pritipal asked
him to bring Ramesh with him to the Gurudwara
that morning about 9 A.M. The boys played about
and had some tea and then Pritipal took Ramesh
away in the direction of the prostitute's house. Pritipal
later
returned
without
Ramesh. The
Sessions
Judge ·thought ·this witness had been tutored on at
least one point. Pritipal's so called confession has
been rejected because, in the first place, it is not a confession at all, for it is exculpatory, and, in the next,
the High Court was not able to trust it. Therefore,
the only evidence of the boy's
I~ movements 1s as
above.
The next point in the appellant's favour is that he
was seen without a coat shortly before the murder and
at a time when he was not in the vicinity of his own
house. According to the prosecution, the murderer
wore the coat, Article X, and the safa, Article Y.
The third point is that the appellant was not see11
by anyone in the vicinity of the place of occurrence.
The fourth point is that if the prosecution case is
true, then it i& remarkable that no one saw the appellant and the boy on a cycle through nearly a mile of
what the High Court, which made i spot inspection,
describes as a crowded locality.
The points against the appellant are ( 1) that he
had a motive and that he said he would be revenged,
(2) that he was absent from the Gurudwara about
the time of the murder long enough to enable him to
commit it, and denied the fact, (3) that some twelve
hours after the crime he assisted m removing the
body from a place between half to three quarters of
a mile distant from the scene of the crime, and ( 4)
that at some unknown point of time he assisted in
tying up the month of the gunny bag in which the
body
was
eventually placed.
In
our
opinion,
it
would be unsafe to convict of murder on these facts.
••
..
-
(
-
-
-
..
' ,.
,.,
S.C.R.
SUPREME COURT REPORTS
543
A number of rulings were cited, including one of the
Privy Council, and it was argued that in those cases
persons were convicted of murder on similar facts.
We do not intend to examine them because no decision can be a guide on facts. Each case has its own
special circumstances and must be decided on its own
facts. For example, m most of the cases cited the
accused was associated with the disposal of ·the body
very ~on after the occurrence and at the scene of the
cnme. Here, twelve hours had elapsed and the first
connection proved with the disposal is at a place over
half a mile distant from where the boy is said to have
been murdered. Next, the points we have shown m
favour of the appellant in this case were not present
there.
We allow the appeal on the charges of murder, conspiracy and kidnapping and reverse the findings and
sentences on those charges· and acquit ·the appellant
of them.
Vv e however convict the appellant of an offence under section 201.
Indian Penal Code, and
sentence him to seven years' rigorous imprisonment.
The learned Sessions
Judge omitted to record a
conviction under .section 201 because he was convicting the appellant of murder. He followed a Nagpur
decision which holds that in such a case it would
be
improper to convict
in the
alternative. We express
no opinion about that; the question does not arise as
we have acquitted the appellant of the murder and
the cognate charges.
The case now falls in line with
that of the Privy Council in Begu v. The King-Emper<>r (1) and the conviction and sentence are confined
to section 201.
Agent for the appellant: Ganpat Rai.
Agent for the respondent: P. A. Mehta.
(l) (1925) 52 I.A.191.
1952
Kasnmi'ra Singh
v.
State of Madhya
Pradesh.
Bose/.