[1949] EACA 35

## COURT OF APPEAL FOR EASTERN AFRICA

## Before EDWARDS, C. J. (Uganda), SIR JOHN GRAY, C. J. (Zanzibar), and AINLEY, J. (Uganda)

REX, Respondent (Original Prosecutor)

## (1) KIPKERING ARAP KOSKE & (2) KIMURE ARAP MATATU, Appellants (Original Accused Nos. 1 and 2)

Criminal Appeals Nos. 40 and 41 of 1949 consolidated.

(Appeal from decision of H. M. Supreme Court of Kenya)

$\cdot\hspace{0.1cm}\cdot\hspace{0.1cm}$

$\sim \sqrt{10}$

Criminal Law—Murder—Circumstantial evidence—Evidence of accomplice-Evidence utterly unreliable—Corroboration—Practice—Application to call further defence evidence at appeal—Procedure to be followed.

The appellants were committed for trial on a charge of murder jointly with one named Chepkwain.

At the trial, before arraignment, the Crown entered a nolle prosequi against Chepkwain who was discharged, and later called as a witness for the prosecution.

The evidence against the appellants was purely circumstantial. They were convicted of murder.

Held (2-5-49) (1).—That in order to justify, on circumstantial evidence, the inference of guilt, the inculpatory facts must be incompatible with the innocence of the accused, and incapable of explanation upon any other reasonable hypothesis than that of his guilt/ and the burden of proving facts which justify, the drawing of this inference from the facts<br>to the exclusion of any reasonable hypothesis of innocence is always on the prosecution and never shifts to the accused.

Wills' Circumstantial Evidence, 6th Ed. p. 311 cited.

(2) (a) That in the present case, as witness Chepkwain was an accomplice it would be wholly unsafe to accept his evidence without corroboration, and corroboration could<br>only remove the taint of suspicion as to his credibility from an otherwise credible witness.

(b) That whether by law or practice corroboration is needed or not, the testimony of a witness may be so utterly unreliable that no reliance whatever can be placed on his a minimum may be used to account of corroboration can render it safe to rely upon<br>his evidence in support of a conviction. When a witness is of this character, a conviction can only be had on entirely independent and reliable evidence from another witness or other witnesses proving the accused guilty of the offence.

(c) That the record of Chepkwain's evidence went very near to showing that he was an utterly unreliable witness and the evidence of the other witnesses did not afford reliable corroboration.

R. v. Baskerville (1916) 12 C. A. R. 81 cited; R. v. Watson (1913) 8 C. A. R. 249 referred to. .ì

Appeals allowed.

The judgment sets out the procedure to be followed in making application to call further evidence for the defence at the hearing of an appeal.

Cases referred to: R. v. Thorne (1925) 18 C. A. R. 186; R. v. Hewson (1908) 1 C. A. R. 47; R. v. Marcus (1923) 17 C. A. R. 187; R. v. Hancox (1913) 8 C. A. R. 176.

Appellants in person.

Sheridan, Crown Counsel (Uganda), for the Crown.

JUDGMENT (delivered by SIR JOHN GRAY, C. J.).—The appellants have been convicted of the murder of one Murono at the Ziwa Estate in the Rift Valley Province on 17th October, 1948.

The deceased was a watchman on the estate. On the night on which he came by his death a cinema show was being given, which was attended by all the estate employees. Another night watchman named Matuyo left the cinema to go on his rounds. He reached the hut of one Anderea Lugando, a storekeeper, and found

it broken open and the deceased lying near by, unconscious, and suffering from a head injury. The deceased died from that injury a few hours later.

The appellants were committed for trial on a charge of murdering Murono along with another man named Chepkwain. At the trial, before arraignment of the three accused persons, the Crown entered a *nolle prosequi* against Chepkwain. who was accordingly discharged. Later he was called as a witness for the prosecution against the appellants. His evidence will be referred to later.

The evidence against the appellants was purely circumstantial. There was no eyewitness to the circumstances in which the deceased received his fatal injury. A fairly considerable quantity of property had been stolen from the house of Anderea Lugando, but none of it had been recovered—and still less, traced to the possession of either of the appellants, and there is no evidence of any bloodstains on the weapons which it is alleged that the appellants must have used.

As said in Wills on "Circumstantial Evidence" 6th edition p. 311, "in order to justify the inference of guilt, the inculpatory facts must be incompatible with the innocence of the accused, and incapable of explanation upon any other reasonable hypothesis than that of his guilt". The burden of proving facts which justify the drawing of this inference from the facts to the exclusion of any reasonable hypothesis of innocence is on the prosecution, and always remains with the prosecution. It is a burden which never shifts to the party accused.

Here, if complete reliance can be placed upon the evidence of the witness Chepkwain, that burden has been discharged by the prosecution, but as the learned trial Judge rightly observed, Chepkwain is an accomplice and it would be wholly unsafe to accept his evidence without corroboration which implicates the accused. As said in R. v. Baskerville. 1916, 12 Cr. App. R. 81 at p. 91, "evidence in corroboration must be independent testimony which affects the accused by connecting or tending to connect him with the crime. In other words, it must be evidence which implicates him, that is, which confirms in some material particular not only the evidence that the crime has been committed, but also that the prisoner committed it". No doubt, the corroboration need only be very slight corroboration, provided that it connects or tends to connect the accused person with the crime. But it must be corroboration, and, as said by the Court of Criminal Appeal in England in R. v. Watson (1913) 8 Cr. App. R. 249, evidence which is consistent either with the story of the accused or with that of the accomplice is not corroboration.

To take Chepkwain's evidence first, it would appear from the record not only that he was an accomplice, but also that he was a most unsatisfactory witness. First of all, he declined to be affirmed. Later when he consented to be affirmed, he contradicted himself in certain respects. Now corroboration can only remove the taint of suspicion as to his credibility from an otherwise credible witness. But whether by law or practice he needs corroboration or not, the testimony of a witness may be so utterly unreliable that no reliance whatever can be placed on his evidence. In such case no amount of corroboration can render it safe to rely upon his evidence in support of a conviction. When a witness is of this character, a conviction can only be had on entirely independent and reliable evidence from another witness or other witnesses which proves that the accused person is guilty of the offence.

The record shows that at first Chepkwain refused to be affirmed. Later, when he was affirmed, he gave evidence which contained a number of self-contradictory statements. The learned trial Judge, who saw and heard this witness give evidence, has treated him as an accomplice, whose evidence could be relied upon provided that the taint was removed therefrom by adequate corroboration, and we will proceed upon the assumption that he was a witness of this character. At the same time we feel that the record of his own evidence goes very near indeed to showing that he was an utterly unreliable witness, whom no amount of corroboration could convert into being a reliable witness.

The learned trial Judge has not dealt with Chepkwain's evidence in any detail and we do not think it necessary to do so for the purposes of this appeal. As said already, it contains some self-contradictory statements, which tend to obscure its general purport, but that purport is that he went with the two appellants to the cinema, that he was invited by them to come and steal clothes but declines to do so, and that later he saw each of the appellants with a bundle of clothes, which they subsequently took away and hid. It is to be observed that in his evidence he seeks to absolve himself completely from the slightest degree of complicity in both the theft and the murder.

The alleged corroboration of this witness is firstly, that of the watchman Matuvo. He says that on the night in question he was on his way to the labour lines when he met the two appellants and Chepkwain. They all three were carrying clubs, which admittedly are commonly carried by people of this tribe when going about their lawful occasions at night. He also alleged that the first appellant was carrying a piece of iron under his blanket. The first appellant denies that he was carrying this piece of iron and it appears to us to be a matter for comment that the witness professed to be able to see this piece of iron under the appellant's blanket at night time. To continue, however, with Matuyo's evidence, after meeting them proceeding in the direction of the labour lines, he took them to the cinema. He stayed at the doorway of the cinema and after some time saw the appellant and Chepkwain leave the cinema. Commenting on this piece of evidence, the learned trial Judge has said and it may be noted in passing that, if he was suspicious of the three individuals, he did not immediately follow on their tracks. We respectfully agree with this comment and would add that the delay in following them would appear all the more to call for comment inasmuch as only about half an hour before he had seen them walking towards the labour lines and had noticed that one of them was carrying a piece of iron.

Commenting upon Matuyo's subsequent evidence, the learned trial Judge says, "he, P. W. I saw them, as he says, 'slip out'. I accept that statement. He says they slipped out ten minutes before he himself left with his key and then ten minutes later he met the woman Jane (P.5). If P. W.1's computation of time was exact and accurate, then this would only have allowed the accused twenty minutes in which to go to Anderea's hut—a distance of some 450 yards—do their breaking in, stealing and make a getaway. I do not consider this impracticable, but the time was short. On the other hand, Jane Grace Paul (P. W.5) says that it was about half an hour between the time she saw Murono (sic the deceased) go to Anderea's hut and the time she called out to him, and a further half hour before P. W.1 arrived on the scene. Neither P. W.1 nor P. W.5 impressed me with their ability accurately to gauge the time, which is not surprising in this class of witnesses". Now it may be unfortunate that witnesses of this class are poor judges of time, but that does not alter the fact that in this case the question of time is one of vital importance and that in order to convict the appellants the Court must be satisfied beyond all reasonable doubt that between the time when they left the cinema and the time Matuyo met the woman Jane they had the opportunity to break open Anderea's hut, to steal his property, and to assault and fatally injure the deceased. There is a conflict of evidence between the two prosecution witnesses on this point, and they are described as witnesses labouring under a disability to gauge time accurately. In the circumstances one can only say that the prosecution have failed to prove opportunity beyond all reasonable doubt. That being so, it is impossible to hold that Matuyo's testimony affects the appellants by showing or tending to connect them with the crime. In other words, it does not afford the necessary corroboration of Chepkwain.

The learned trial Judge has also treated as corroboration of Chepkwain the evidence of Anderea Lugando, who deposed that "accused 1 came into the cinema through the window, looked at me, and went away". The first appellant gave evidence on oath and said, "I got into the cinema through the window. I

know Anderea. I did not see him there". Now if Anderea intended by the words which he used to convey that the first appellant entered by the window, looked to see if he (Anderea) was there and then at once left, this evidence is at variance with that of Matuyo, who says the appellants were in the cinema for some 15 minutes. If on the other hand all that Anderea intended to convey was that he saw the first appellant at the window looking at him, and that some time relatively later the appellant went away, then with very great respect to the learned trial Judge this is no more than equivocal evidence which does not amount to corroboration. It is true that the appellant has said on oath that when he entered by the window he did not see Anderea there, but he may well have looked in his direction and failed to recognize him in a crowded audience. In the circumstances it cannot be held that Anderea's evidence affords reliable corroboration of Chepkwain.

As already said, the conviction of the appellants must depend upon the credibility of Chepkwain. In the absence of corroboration of his evidence, it would obviously be most unsafe to sustain this conviction.

We have already intimated that these two appeals must be allowed and both appellants have already been released.

In the circumstances it is not necessary to deal with the application to call further evidence for the defence, but as in this and other cases at this session it would appear that there is some doubt as to the procedure to be followed in making application of this nature, it may be perhaps desirable to set out the procedure which should obviously be followed in regard to such applications.

In the first place, it is perhaps desirable to point out that an application to call further evidence at the hearing of an appeal is not granted as of right. It is a matter entirely for the discretion of the Appellate Court and an application of this nature will only be granted in special circumstances (R. v. Thorne (1925) 18 Cr. App. R. 186). In particular, before granting the application, the Court requires to know the reason why the particular evidence was not given at the trial. It also requires that the nature of the further evidence which it is desired to give should be disclosed. It is generally speaking desirable that the reasons for the failure to give the evidence at the trial should be set out in an affidavit, which should be sworn either by the appellant or his advocate or some other person qualified to state those reasons. But the nature of the further evidence should ordinarily be in the form of proofs of the proposed witnesses $(R, v, Hewson (1908)$ 1 Cr. App. R. 47; R. v. Marcus (1923) 17 Cr. App. R. 187). Such evidence should not be submitted in the form of an affidavit or affidavits. As said by the English Court of Criminal Appeal in R. v. Hancox (1913) 8 Cr. App. R. 176, "although the Court is entitled, if it thinks fit to do, to act upon evidence by affidavit, it will not as a rule pay much regard to such evidence unless an opportunity has been given to the other side to cross-examine the deponent". Therefore even if the Court looks at the affidavit, it will as a rule, as in Hancox, not receive it as additional evidence, but will direct that the witness in question shall give his evidence upon the *voir* dire either before itself or some other Court subordinate thereto and submit to cross-examination by the opposite party. Therefore, as already said, the proper way of disclosing to the Court the nature of the proposed additional evidence is by submitting proof of the witnesses, whom it is desired to call.

In making these observations we desire to guard ourselves from being held thereby to reflect on Mrs. Mott, Counsel for the appellants, to whom the Court is extremely grateful for the trouble she has taken in the presentation of the appellants' case to this Court. We also desire to express our appreciation of the public spirited action of the first appellant's employer, Mr. Glasson, in taking steps to bring to the notice of this Court that there was further evidence which in his opinion the ends of justice required to be brought to the Court's notice.