[1950] EACA 440

## H. M. COURT OF APPEAL FOR EASTERN AFRICA

# Before Sir Newnham Worley (President), Bacon, Justice of Appeal, and CORRIE, J. (Kenva)

## (1) GITHAE s/o GATHIGI and (2) NDIRITU s/o WAMITI, Appellants (Original Accused)

#### **REGINAM, Respondent**

### Criminal Appeals Nos. 475 and 476 of 1955

# (Appeals from the decision of H. M. Supreme Court of Kenya sitting in Emergency Assize, Seton, J.)

Murder—Accessory after the fact—Accomplice—Corroboration.

The trial Judge had found correctly that one of the Crown witnesses was an accessory after the fact and was also an accomplice whose evidence required corroboration. He found the corroboration in six items of evidence which he set out in his judgment.

Held $(3-1-56)$ .—(1) An accessory after the fact in a case of murder must always be an accomplice and any decisions prior to *Davies v. D. P. P.*, (1954) 1 E. A. R. 507 H. L. which suggest the contrary were no longer law.

(2) A witness's ability to point out the place where the deceased was buried and medical evidence that the killing took place in the manner described by the witness could not be corroboration of his evidence that the accused killed the deceased.

(3) Where only one of six items relied on by the Judge as corroboration had that effect in law the conviction was unsafe.

Appeals allowed. Convictions quashed and sentences set aside.

Cases referred to: Davies v. Director of Public Prosecutions, (1954); 1 A. E. R. 507 H. L.; Kinyangabwaru s/o Matasa v. The King, (1942) 9 E. A. C. A. 90; Mwanahawa d/o Saidi v. The King, (1944) 11 E. A. C. A. 118; Erunasani Sekoni s/o Eria v. The King, (1947) 14 E. A. C. A. 74; Ali Saleh v. The King, (1948) 23 Kenya L. R., Part I, 40; R. v. Okello Byenzi, 5 E. A. C. A. 140.

#### Appellants in person.

Brookes for respondent.

JUDGMENT (prepared by Bacon, J. A.).—The appellants were convicted on V 20th October, 1955, by the Supreme Court of Kenya of the murder of an African woman in July, 1954.

The leading Crown witness, Wambugu s/o Gathimba, was, according to his evidence, himself a principal in the first degree. His story at the trial was that he and the appellants (who all shared a hut) had together strangled the deceased with a length of rope while she lay on a bed. He also stated that the deceased was wearing a red coat on that occasion. Immediately after the woman died Wambugu went and summoned one Maina to come to the scene of her death. In a short time Maina arrived and helped the other three men to carry the body in a sack across a river and to conceal it in a hole in the ground where they left it covered with earth. Maina also was a Crown witness, and it is clear from his own evidence as well as from that of Wambugu that he was an accessory after the fact. A third witness for the Crown was one Eliud Kogi who said that he

remembered an evening when he was slightly drunk and the second appellant had arrived at his (the witness's) hut and told him that there was a red coat in his (the second appellant's) hut and asked him to take the coat away. Kogi then went to the appellant's hut and saw a red coat there.

Each of the appellants made a statement before trial and their statements were admitted in evidence. The first appellant's statement contained nothing admitting any connexion with the incident. The second appellant in his statement alleged that he had been told by Wambugu and the first appellant that they were going to kill the deceased whom he saw at that time in their hut, that he then left the hut, that he later returned to find the deceased dead with a rope round her neck and that Wambugu and the first appellant then obliged him and Maina, under threat of death, to carry away the body. At the trial each of the appellants gave evidence and denied having any connexion with the death of the deceased.

On the hearing of the appeal the appellants were present but unrepresented. Mr. Brookes for the Crown respondent drew our attention to the passages in the judgment in which the learned trial Judge dealt with the question of corroboration, and submitted that he was unable to support the convictions. We allowed the appeals, quashed the convictions and set aside the sentences. We now give our reasons for so doing.

Having stated that Wambugu was an accomplice and that his evidence needed corroboration, the learned Judge then set out six items each of which he designated as coming within that category.

Two of those items consisted of evidence given by Maina whom the learned Judge described as "not an accomplice although, by assisting in the disposal of the body he would seem to have made himself an accessory after the fact". This was an error in law. Maina, being—as he clearly was—an accessory after the fact on his own showing at the trial, was an accomplice. The rule of law has been put beyond doubt by the unanimous opinion of their Lordships in *Davies* $v$ . Director of Public Prosecutions, (1954) 1 A. E. R. 507, H. L., at p. 514, where it was said that the definition of the term accomplice "covers participes criminis in respect of the actual crime charged, whether as principals or as accessories before or after the fact". That statement of the law must be regarded as binding on this Court with the result that previous decisions to the contrary are no longer the law. Such previous decisions are to be found in Kinyangabwaru s/o Matasa v. The King, (1942) 9 E. A. C. A. 90, Mwanahawa $d/o$ Saidi v. The King, (1944) 11 E. A. C. A. 118, Erunasani Sekoni s/o Eria v. The King, (1947) 14 E. A. C. A. 74, and Ali Saleh v. The King, (1948) 23 Kenya L. R., Part I, 40. Accordingly, the witness Maina cannot be treated as corroborating Wambugu, since one accomplice cannot corroborate another.

The third item said to have corroborated Wambugu's evidence was "the fact of his ability to point out to the police where the deceased's body had been buried". This involved the erroneous proposition that a witness can corroborate himself. In truth the only result of Wambugu's ability to disclose the site was to establish his credibility as a witness who asserted that he took part in the burial; his disclosure of the site was no corroboration in the true sense of that term, namely, independent testimony which tends to connect the accused with the crime charged, for neither was it independent of Wambugu's testimony, nor did it implicate the appellants.

There was all the difference between the instant case and such a case as R. v. Okello Byenzi, 5 E. A. C. A. 140 where it was held that there was true corroboration in that the accused himself had disclosed the hiding-place of a spear, the alleged offence being murder by means of a weapon of that kind.

The fourth item of corroboration was said to be the pre-trial statement made by the second appellant, inasmuch as he had stated that he found the deceased already dead in the hut with a rope round her neck and assisted in carrying away the body. As the learned Judge observed, this could not be evidence as against the first appellant at all; and, since the statement expressly averred that its author took no part in the killing of the deceased it is difficult to see how it could amount to corroboration as against himself; the essence of his story was that he had the misfortune to be in the vicinity when the deceased was murdered by other persons who then forced him to help them to dispose of the body.

There are two further items which were relied on as corroboration of Wambugu, First, Dr. Rogoff had examined the deceased's remains more than a year after her death and testified that there seemed to be a piece of rope, and that there was some sacking, with the skeleton. This evidence certainly corroborated Wambugu's statement that a rope had been used to kill the deceased and a sack to remove and bury the body. But the implication of the appellants in the crime still rested on the sole word of Wambugu, for there was no independent and untainted evidence, nor anything admitted by either of the appellants themselves, as to the association of a rope or of a sack with the killing of the deceased by the appellants or either of them. Secondly, there is Eliud Kogi's evidence as to the presence of the deceased's red coat in the appellant's hut. This evidence is said in the judgment to relate to "the night of the murder". Kogi, testifying some 15 months after the alleged date of the incident, purported to identify the occasion on which he had seen the coat by reference to the visit to his (the witness's) hut by the second appellant and to the conversation between them, from which evidence it could properly be inferred that the occasion was the one on which the deceased died, and from which it might also be inferred either that the second appellant had been a party to the killing or that, having discovered the coat in the hut which he shared with Wambugu and the first appellant, he had taken fright and wanted the coat removed. It was thus open to the learned Judge to treat (as he did treat) this evidence of Kogi as corroborating Wambugu's evidence that the deceased was wearing a red coat on the evening of her death in the hut occupied by Wambugu and the appellants.

We think, however, that this last-mentioned item, the only one out of the six relied on in the judgment which could be in reality any corroboration of Wambugu in the true legal sense, was by itself not of such weight as against either appellant on the charge of murder that if it had stood as the sole corroboration of Wambugu, a self-confessed principal in the first degree, it would inevitably have led to the conviction of the appellants.

On the contrary, viewing the judgment as a whole, it clearly appears that the learned Judge placed great reliance on the five other items of supposed corroboration to which we have referred and must thus be taken substantially to have founded his decision on false premises. In our view plainly cogent evidence from a source independent of the suspect witness was required in such a case as this, and, when the record is analysed, it appears that no such evidence was before the trial Court.