IN THE COURT OF APPEAL

AT NAIROBI

(Coram:Nyarangi, Platt & Gachuhi JJA)

CRIMINAL APPEAL NO. 15 OF 1986

BETWEEN

PATRICK ASIRA..............................................................................APPELLANT

AND

REPUBLIC.....................................................................................RESPONDENT

(Appeal from the High Court at Nairobi, Mbaya & Butler-Sloss JJ)

JUDGMENT

This appeal concerns the alleged conspiracy of an employee of the National Housing Corporation, who was held to have stolen a cheque from the Corporation, which he successfully negotiated through the aid of an employee of the Standard Bank, and received the greater part in cash. It is said that to do all this, he had an accomplice in the Bank called Ndiva, who aided him to pay the cheque for Kshs 6,000 into a newly opened account, and then later allowed him to withdraw Kshs 35,000. It is said that the Appellant assumed an alias calling himself “Wilfred Kipchira Lagat” purportedly working in the Ministry of Lands. His real name is Patrick Asira, an employee of the Corporation. All this came to light when Reuben Mulievi, who had been granted a loan by the Corporation, asked what had happened to it, only to be told that the cheque had been paid through the Bank. Investigations led to the events which had occurred at Ndiva’s part of the Bank, and he was expected to find the culprit, whom he said he knew. Inquiry statements were also made by Ndiva. After some days, the Appellant was pointed out by Ndiva and they were both charged.

The nature of the charge is of interest. The first count alleged conspiracy between the Appellant and Ndiva. But the second, third and fourth interrelated counts concerned only the Appellant. He was alleged in count 2 to have stolen the cheque on August 1, 1984 made out to Reuben Mulievi; to have uttered it to the cashier at the Bank, Martin Gachehe on August 8, 1984; and to have obtained Kshs 35,000 from the Bank by the false pretence that the cheque had been drawn in his favour. That was also on August 8, 1984. It will be apparent that the identification of the Appellant rested on the Statement of Ndiva, unless other evidence was forthcoming. The prosecution alleged that such evidence was to be found in the opinion of the handwriting expert, and the purchase of a new motor cycle by the Appellant on August 15, 1984 for Kshs 27,000. He had been intending to buy this motor cycle since June 1984. Otherwise the Appellant had no official access to this cheque in the offices through which the cheque passed. He worked in another department. It was lost from the Managing Director’s office according to George Odera, the clerk in charge of the cheque from the time it was made out by him. He also steered it from one office to another, while the cheque was signed and presumably scrutinized. It is not possible, but not at all obvious that the Appellant would be able to steal the cheque. There was also a muddle over the next cheque registered for Kshs 27,000 in favour of Mr Mbaya, which was handled by Mr Njogu. It was said that that muddle was put right; but the fact remains that the cheque in question (No NHC 033963) was signed for and then cancelled. Several people may have stolen the cheque, and the Appellant was not an obvious suspect.

The evidence of identification against the Appellant was entirely that of Ndiva. Mrs Catherine Rasugu, one of Ndiva’s supervisors, and who allowed Ndiva to go ahead with the deposit of the cheque in question and withdrawal of the money, (not strictly in conformity with standing instructions) never saw the Appellant. Mr Gachehe the Cashier, to whom the Appellant was allegedly introduced by Ndiva, could not remember the payee. Indeed it was Ndiva who pointed out the payee. The payee did not speak. He was asked to sign by Mr Gachehe as payee and he did sign. The fourth count was not accurately laid. The Appellant, in silence could not have pretended that the cheque had been drawn in his favour. It was drawn in favour of Mulievi. It was endorsed over to “Lagat,” if that was the Appellant. In times past, that sort of mistake could have been fatal, because the accused person would not know what case it was that he had to meet.

Relying upon Ndiva’s position, the prosecution had to choose whether it would use Ndiva as a witness, (even if there was a risk of his being an accomplice), or accuse him of conspiracy. That was not an easy choice, but in the eyes of the trial court, Ndiva was not a conspirator. There is a good deal to be said for both sides. Ndiva had given the impression that he had known “Lagat” from their school days, if Mrs Rasugu is reliable. He had said that he knew “Lagat” and the place where he worked. He had put the case for allowing the cheque to be banked after being endorsed over, because Mulievi was the old father of “Lagat” and had no account. He had assisted “Lagat to withdraw most of the money for the old father.” On the other hand, Ndiva had taken the authority of his superiors at every step. He had not really known “Lagat”. Ndiva was only a trainee. His request should have been turned down. He had acted quite openly and although he had been mistaken where “Lagat” worked, and what his name really was, he had in the end found “Lagat” with the new motor-cycle. The prosecution chose to treat Ndiva as a conspirator and the trial court decided that he was not. There was no appeal against that acquittal to the High Court, and so Ndiva must be held to have been innocent. In those circumstances the remaining evidence was very narrow indeed.

Ndiva explained to the authorities that the Appellant was the person who called himself “Lagat” and was the man who had opened the account,paid in the cheque on behalf of his father, and withdrawn money from the account. Of this evidence, there were actions of Ndiva not in the Appellant’s presence and in his presence; and there were the statements made under caution. Finally there was the unsworn statement in Court. It will be accepted at once that his unsworn statement was not “evidence” in the case, as it would have been had it been given on oath, and was therefore not evidence against the Appellant. It only affected Ndiva himself (See Patrisi Ozia vs Reg [1957] EA 36; Republic v Ramji Hirji (1946) 13 EACA 127). There remained the actions of Ndiva to be considered under section 10 of the Evidence Act (cap 80). It provides:-

“10 Where there is reasonable ground to believe that two or more persons have conspired together to commit an offence….., anything said, done or written by anyone of such persons in reference to their common intention, after the time when such intention was first entertained by any one of them, is a relevant fact as against each of the persons believed to be so conspiring, as well for the purpose of proving the existence of the conspiracy as for the purpose of showing that any such person was a party to it.”

Thus, the conspiracy having been reasonably believed to exist between “Lagat” and Ndiva, what Ndiva did and wrote could be considered against “Lagat”, both for showing that there was a conspiracy and that “Lagat” was a party to it. But that aspect ceases to exist after the conspiracy is not found to exist. What Ndiva said about “Lagat,” in the latter’s absence was inadmissible as hearsay against “Lagat.” What Ndiva said both in the absence and presence of “Lagat” after the latter had been identified by Ndiva was the statement of a co-accused, and not being an admission was inadmissible against “Lagat” (section 32 of the Evidence Act; Mattaka vs Rep [1971] EA 495 at page 511F). It follows that the actions and Statements of Ndiva could not be considered against the Appellant.

In those circumstances, the prosecution would have been wise to call Ndiva as a prosecution witness, so that what he had to say would be evidence for the record. In an obscure passage the learned judges of the High Court felt constrained to hold that the evidence of a co-accused, was evidence and needed corroboration. They held that the “evidence” of Ndiva was corroborated in all material respects, especially by that of the document examiner. It is difficult to know what “evidence” it is that the learned judges were referring to. From what has been said above, it cannot have been the unsworn defence statement; and it cannot have been the other actions and statements of Ndiva as explained. We can only assume that the learned judges mis-directed themselves, and unfortunately this was a fundamental misdirection of law going to the root of conviction. For without the actions and statements of Ndiva, there remained only the opinion of the handwriting expert and the purchase of the motorcycle. Whether the courts below would have relied only on this evidence is not, of course, known to us.

The handwriting expert, Mr Mweu compared the alleged writing of the Appellant with his known handwriting, and his opinion was as follows:-

“The standard writing shows evidence of disguise, but there were features confirms individual ... which suggest that the writing came from one hand. Which means they were writings by one person but the writings were disguised.” (sic)

That remarkable opinion set out in a report without any demonstration of what was disguised writing and what were the tell-tale characteristics of the Appellant’s writing, gave rise to the following findings by the trial court:-

“He found that accused No 1 (ie the present Appellant) had written the opening deposit for Kshs 300 which opened Account No 11698 as per slip now Exhibit 9 as well as the withdrawal slip for Kshs 35,000 now Exhibit No 4 as well as the deposit for Kshs 36,000 Exhibit No 3 and the opening form of the same account Exhibit 8 in this case.”

Then later on -

“The document examiner found he wrote the opening form as well as deposit for Kshs 300 to open account and for withdrawal he said there was disguise of the writing and that what one will expect when he had was to defraud or steal by false pretences.” (sic) Hence – “It has been proved that accused No 1 wrote the form to open the fake account by means of “Lagat” and no doubt he conned the bank people and withdrew the money.”

It is difficult to say whether the trial xourt ever made a finding of its own that the writing had sufficient characteristics of a similar nature for the learned magistrate to find that the writing was that of the Appellant. He relied on the document examiner’s opinion, one formed after examination by microscope. It was not possible to demonstrate the features of similarity in Court. The learned magistrate was then quite right to this extent that he could not form his own opinion, on a matter which could not be demonstrated to him. It is questionable whether he could blindly rely on the expert’s opinion.

The High Court came to a similar conclusion. It was the handwriting expert who came to the conclusion that the Appellant wrote all the relevant documents referred to by the trial court. The learned judges had no better opportunity of forming an opinion themselves than the trial court. They also relied upon the expert’s opinion while they themselves were blind to the disguise and similarities. Nevertheless on what they termed the entire evidence, they were satisfied that the Appellant had written the relevant documents. That was their re-assessment of the evidence.

Mr Ombete challenges that approach. It is for the court to decide the question and not for the expert to decide it for the court. In this case the courts had no opportunity to decide the issue. In fact they never decided the issue for themselves but allowed the expert to decide it for them. Mr Hag for the State found himself in difficulties because the handwriting expert could not demonstrate what he had found. Nevertheless he thought that the magistrate could look at the reasons in the report and that was sufficient. He thought that it would be safe to convict if the opinion was taken with other evidence.

The most authoritative local statement on this topic is to be found in Wainaina v Republic (1978) KLR 11. The Court of Appeal accepted that Spry J had correctly set out the proper role of the expert in Hassan Salum v Republic [1964] EA 126 at page 128 to the effect that the most an expert on handwriting can properly say, in an appropriate case, is that he does not believe a particular writing was by a particular person or, positively, that the writings are so similar as to be undistinguishable. This Spry, J culled from the opinion of Lord Birkenhead in Wakeford vs Bishop of Lincoln (1921) 90 L J PC 174. It is of significance that the House said that the expert could , of course, comment on unusual features which make similarity the more remarkable. But that fell far short of saying that they were written by the same hand. Nevertheless in Wainaina it was held that there is no rule of law which requires corroboration of the opinion of a handwriting expert whose evidence is like the evidence of the whole class of experts which it is open to the court to accept or reject. On this point Onyango vs Republic [1969] EA 362 was approved. On glancing at Onyango’s case it will be seen that the class of experts includes doctors. But in that case the crucial fact was that a very confident opinion was put before the court with other evidence. It seems that that was the position in Wainaina’s case. But that is not the position in the present case. Here we have writing which is not similar. That is called disguised writing. In the disguised writing were some features which could not be demonstrated which were said to be similar to the Appellant’s handwriting. The court could not itself judge the matter, so it was not a case simply of rejecting or accepting the evidence. It could do either. In a case such as this, one has therefore to go back to first principles. The art of comparing handwriting is no doubt one in which time and thought are given to the formation of letters and words, and therefore expert status may be accorded to a person versed in such comparisons. But as has been accepted in Wainaina’s case such an expert is not able to say definitely that anybody wrote a particular thing. The reasoning is based upon the knowledge that handwriting can very easily be forged. Moreover a person may not write in the same style all the time. The expert is therefore faced with trying to analyse forged writing as well as disguised writing. In cases where there is a problem about the writing it is the duty of the court to satisfy itself after examination whether the expert’s opinion can be accepted and cannot blindly accept such opinion. In these areas of conflict it is prudent to look for other evidence so that forgery can be excluded on the one hand, and mistaken identification excluded on the other. That was the way the House of Lords approached the matter in Wakeford’s case where at page 180 they observed as follows :-

“Looking at the papers before them, Their Lordships, upon the evidence of their own eyes, have reached the conclusion that there can be no doubt upon the matter. If this were the only piece of evidence, Their Lordships, without doubt in their minds as to the authenticity of the writing would not willingly rest their judgment on a single fact as to which an error might be possible. But the only alternative to the genuineness of the writing is the supposition that it was a carefully planned forgery of the Appellant’s name as an integral part of the alleged conspiracy. For the reasons already given Their Lordships feel that the hypothesis of such conspiracy furnishes an overwhelming corroboration of the other evidence.”

As far as the practice in India is concerned, the value of expert opinion on handwriting is set our in Sarkar on Evidence, 12th edition page 511. In the present case, after the acquittal of Ndiva there is no evidence which can support the handwriting expert’s opinion and as there is a question whether the Appellant who is Patrick Asira posed as “Lagat”, there is a possibility of an error about disguised handwriting.

The remaining evidence concerns the Appellant’s defence that he bought his motor-cycle with the proceeds of the cheque. This evidence may or may not be right and cannot by itself strengthen the prosecution’s case.

In the end it is our view that the conviction of the Appellant was unsound. It is quashed and the sentence imposed set aside, consequential orders set aside and the Appellant is set at liberty unless otherwise lawfully held.

Dated and Delivered in Nairobi this 4th day of November,1986.

J.O.NYARANGI

...................................

JUDGE OF APPEAL

H.G.PLATT

..................................

JUDGE OF APPEAL

J.M.GACHUHI

..................................

JUDGE OF APPEAL

I certify that this is a true copy of the original

DEPUTY REGISTRAR