[2023] UGCA 239

## THE RIPUBLIC OF UGANDA

IN THE COURT OF APPEAL OF UGANDA AT KAMPALA

# CRIMINAL APPEALS NOS. 39 E L29 OF 2O2O

## I. MAZUKU JONATHAN

2. MPAMULUNGI JUSTINE APPELLANT

## VERSUS

UGANDA .... RESPONDEI{T

(Appeal from the convlction and sentence of the Htgh Court of Uganda at I{akawa before Hon. Mr. Oyuko Anthony OJok dated 3'd February 2O2O)

CORAM: HON. JUSTICE RICHARD BUTEERA' DC., HON. JUSTICE CHRISTOPHER GASHIRABAKE, JA HON. JUSTICE OSCAR KTHIKA, JA

## JUDGMENT OF COURT

The Appellants were indicted and convicted of the offence of Murder contrary to sections 188 and 189 of the Penal Code Act and sentenced to 25 years' imprisonment.

The Appellants were dissatished with the decision of the trial court and l-rled this appeal against both conviction and sentence on the following grounds;

- 1. That the L,earned Trial Judge erred in law and fact when he convicted the Appellants based on uncorroborated circumstantial /canine evidence. - 2. That the Learned Trial Judge erred in law and fact when he misconstrued the findings of the expert in regard to the samples of soil picked from the Appellants.


- 3. The Learned Trial Judge erred in law and fact when he failed to take into account principles governing admission of caline evidence. - 4. That the Learned Trial Judge erred in law and fact when he convicted the second Appellant for not testifying against the frrst Appellant. - 5. The Learned Trial Judge erred in law and fact when he disregarded the Appellants' defence of alibi. - 6. The Learned Trial Judge erred in iaw and fact when he sentenced the Appellants to 25 years' imprisonment, a sentence that is harsh and excessive in the circumstances of the case by not taking into account a.ll the mitigating factors.

### Background

The facts as accepted in evidence at the trial are as follows;

On 16th August 2Ol7 at about 6:00am a one Nakasi rang Kazibwe Charles, the defence secretar5/ LC1 of St. Francis zone, Kawempe Division and informed him that a dead body had been found dumped at a rubbish heap. Kazibwe then informed the o/c-Bwaise Police Post and thereafter rushed to the scene. shortly after, the O/C-Station Kawempe Police Division ASp Aanyr: Agnes, Detective Corporal Asindua the SOCO (scene of crime officer) together with other police Oflicers and a Police sniffer dog arrived at the scene. The sniffer dog was given access to the crime scene whereupon it commenced tracking of the scents which led to Katogo mne. It then entered the house of the 1"t and 2"d Appellants who were both inside the room and sat next to the lst Appellant who was seated in a chair. Both Appellants were arrested and taken to Kawempe police Station. The deceased was taken to Mulago Hospital mortuary for post mortem.

An autopsy was carried out on the deceased's body which revealed the cause of death as manual strangulation. On 16th August 2077, the scene of crime was visited and a sketch plan drawn. The following day a search was conducted in the Appellants' home in their presence. Exhibits were recovered including a faded black grey skirt with soil stains, a pair of closed shoes stained with soil, a pair

² UL #M

ofopen shoes stained with blood and soil inside it. The 2"d Appellant recorded a charge and caution statement in which she admitted that the recovered shirt arrd open shoes belonged to them.

#### Repreaentation

At the hearing of the appeal, Ms. Wa.kabala Susan on private brief, appeared for the Appellants while Ms. Caroline Marion Achio, Chief State Attorney, appeared for the Respondent. Both parties filed written submissions which were adopted by this court.

### Conslderatlon of the appeal

This being a l-rrst appeal, it should be recalled that the duty of a ltrst appellate court is to re-eva.luate the evidence, weigh conflicting evidence and reach its own conclusion on the evidence, bearing in mind that it did not see the witnesses testify. (See Pandya v R [195{ EA p.336 and Klfamunte v Uganda Supreme Court Crlmlnal Appeal No. 1O of 1997 and COA Crimlnal Appeal No. 39 of 1996. In the latter case, the Supreme Court held that;

"We agree tlnt on a first appeal, from a conuiction bg a Judge tle Appellant is entitled to haue the appellate Court's own consideration and uieuts of th.e euidence as a uhole and its outn decision thereon. The first appellate court ha.s a dutg to reuiew tle euidence ofthe case and to reconsider the materials before the trial judge. The appellate Court must then make up its own mind not di.sregarding the judgment appealed from but carefullg uteighing and considering it."

We have kept these principles in mind in resolving this appeal.

We shall resolve the grounds of appeal in the order in which the parties argued them.

W

koo M3

# Grounds 1,2 and 3

The appellants counsel argued grounds 7 and 2 together while the respondent's counsel argued grounds 7,2 ar,d 3 together. We shall resolve grounds 7,2 and. 3 together.

1. That the Learned Trlal Judge erred ln law and fact when he convlcted the Appellants based on uncorroborated clrcumstantlal /canlne evldence.

2. That the Learned Trlal Judge erred in law and fact when he misconstrued the flndings of the expert in regard to the samples of soll picked from the Appellants.

3. The Learned Trial Judge erred ln law and fact when he failed to take lnto account princlples governlng admlsslon ofcanine ewldence.

## Appellant's submlssions

Counsel submitted that the appellants were convicted solely based on the evidence of the sniffer dog and yet such evidence is circumstantial and therefore needs corroboration. Counsel relied on the decision in Godl Akbar Vs Uganda, S. C. C. A No. 03 of 2013 for the proposition that in a case depending exclusively upon circumstantial evidence, the court must find that the inculpatory facts are incompatible with the innocence of the accused and incapable of explanation upon any other reasonable hypothesis than that of guilt.

Counsel submitted that the witnesses presented by the prosecution were untruthful and their testimonies were filled with discrepancies, contradictions arrd irregularities. Counsel relied on the Supreme Court decision in Obwalatum Francis Vs Uganda, Crlmlnal Appeal No. 3O of 2O15 for the proposition that where contradictions and discrepancies are grave, this would ordinarily lead to the rejection of such testimony unless satisfactorily explained.

Counsel submitted that PW5 Detective Corporal Kayondo Saidi stated that there were 3 officers that entered the Appellants' house at the time of arrest including

4 st/ WL

the dog handler, the OC station ASP Asanyu and DC Kayondo Saidi and yet the evidence of PWl was that PW4 did not enter the Appellant's house. Counsel submitted that this cannot be a minor or trivial discrepancy because it goes to the root of one of the main exhibits that was discovered from the Appellants' house which is the faded black grey skirt that bore stained particles of soil. Counsel submitted that the evidence of PW4 ought to have been struck off the record for being inconsistent with that of the other prosecution witnesses.

Counsel argued that there was no direct evidence presented to the trial court to satisfy the ingredients of murder as presented by the prosecution and it relied solely on circumstantial evidence. Counsel relied on the decision in Bullla and

Anor Vs. Uganda, (Crlmlnal Appeal IYo.16 of 2O15) in which the Supreme Court held that before a conviction is upheld on the premise of circumstantial evidence, the Court has to consider and ensure that the inculpatory facts are incompatible with the accused person's innocence.

Counsel submitted that a-Ithough the leamed trial Judge rightly set out the test when dea-ling with evidence of sniffer dogs as set out in case of Wllson Kyekurugaha vs. Uganda Crlmlnal Appeal No. 51 of 2014, the evidence presented by the prosecution specifically, PWI did not fit the admissibility test' In addition, the prosecution did not lead evidence to prove the involvement of the second Appellant as a perpetrator of the murder and the dog never identil-led her as a suspect. Counsel argued that the assessors that sat in for the trial and observed the witnesses give their evidence, returned a NOT GUILTY verdict because they were not convinced that the prosecution based on the circumstantial evidence presented had proven their case beyond reasonable doubt.

#### Respondent's Submlsslon

In reply to the Appellant's arguments on canine evidence, counsel relied on the case of Wllson Kyakurugaha Vs. Uganda COA Crlmlnal Appeal No. 51 of 2O14, this Court approved six principles which should be applied as guidelines

uz C'to.n 5 b/

in dealing with canine evidence and these have a bearing on the exclusion /admissibility and the weight to be attached to canine evidence;

- The evidence must be treated with the utmost care by the Court and given the fullest sort of explanation by the prosecution. - 1l There must be material before the Court establishing the experience and qualihcation of the dog handler. - llt The reputation, skill and training of the tracker dog is required to be proved before the Court (by the handler/trainer who is familiar with characteristics of the dog). - lV. The circumstances relating to the actual trailing must be demonstrated. Preservation of the scene is crucial and the trail must not have become stale. - The human handler must not try to explore the inner workings of the animal's mind in relation to the conduct of the trailing. This reservation apart, he is free to describe the behavior of the dog and give an expert opinion as to inferences which might properly be drawn from a particular action of the dog. - VI The court should direct its attention to the conclusion which it is minded to reach on the basis of the tracker evidence and the perils in too quickly coming to that conclusion from material not subject to the truth eliciting process of cross-examination.

That the relevant pieces of evidence are found in the testimony of PWl, PW4, PW5 and PW7. That Mr. Okello Felix consistently and firmly told Court that he is the one who handled the dog on this occasion. That he was trained in basic dog handling from Nsambya Canine headquarters and he had experience of ¹¹ years (page 16 ofthe record of appeal).

Counsel for the Respondent submitted that the Appellants' claim that the scene was tampered with and stale for a credible dog tracking exercise was not backed by evidence and not in compliance with other principles. Counsel submitted that

F '.f h/"

there was corroboration of the Canine evidence. With regard to the inconsistencies, counsel submitted that although there was some inconsistency regarding time when the search was conducted in the house of the Appellants and the number of items recovered, it is not in dispute that a search was conducted and that zrmong the items recovered was the black skirt and a shoe which belong to the Appellants. Counsel further argued that the inconsistency regarding when the search took place should therefore, be considered minor and does not go to the root of the case.

# Conslderetlon of grounds 1,2 and 3

It is trite law that the prosecution has the duty to prove each element of al offence beyond reasonable doubt. For the Appellants to be convicted of murder, the prosecution must prove, beyond reasonable doubt, the following elements;

- 1. That there was death ofa human being. - 2. The death was caused by some unlawful act. - 3. The unlawful act was actuated by malice aforethought; - 4. It was the accused who caused the unlawful death.

In the present case, it is not in dispute that Namuwonge Jennifer is dead and that her killing was unlawful. The post mortem report marked PExh 1 indicated that the cause of death was strangulation. The manner of her death pointed to the fact that her demise was actuated by malice aforethought. What is in contention is the participation of the Appellants in the killing of the deceased.

The prosecution produced evidence of 7 witnesses, all being police officers and amongst them was the dog handler.

PWl, the dog handler attached to the canine unit department, testified that he was called by the OC Kawempe Police Station who informed him that there was a murder at St Francis Bwaise. When he arrived at the scene, he found it cordoned off by the scene of crime officer, DC Asindua using police tape. He also found other police men already at the scene including the investigating officer

W ⁷ br cts,l(

DC Kayondo', in charge police post Bwaise ASP Wangolo and sergeant Buyinza who were at the scene managing the gathering crowd.

At the scene, he went with the police dog to the body ofthe deceased which was lying face up and covered with rubbish right from the leg up to the neck. The rubbish consisted of banana peels, kavera, dry grass and there were bare footed marks of footsteps around the body.

PW1 testified that he introduced the police sniffer dog number 70 Neval, which picked the scent around the body and moved for a distance of half a kilometer leading him to the house of the two Appellants. When the dog arrived at the Appellalts' place, the door was open so it entered the house where Mazuku and Justine were. Mazuku was seated on the sofa chairs and was facing the opposite direction. The sofa chair as they entered was on the right, the bed in front of them and then the other side was where he had his music system. The woman, 2"d Appellant, was behind the curtain. The dog entered and went direct to A1 sniffed him and sat in front of him.

PW2 tendered in the post mortem report which indicated the cause of death being by strangulation and described the body as having no injuries but with numerous old scars aged 29 years and normal status.

PW4 was the investigating Officer ard he testified that he reached at the crime scene together with a team of detectives headed by Deputy OC CID, at around 8:00 am in the morning. That the crime scene was situated in between two houses, one was a residential house and the other one was a pit latrine with a trench that passes nearby the toilet. PW4 stated that they found people at the crime scene and after arriving at the scene, PW4 cordoned the scene to stop the crowd from accessing it. PW4 took photographs which were exhibited and collectively marked PEX. 3.

When the sniffer dog led them to the home of the Appellants, the Deputy OC CID conducted a search and one black faded and grey skirt was recovered stained with some particles of soil. One pair of closed shoes was also recovered and its

\$d'l u-

sole was stained with some particles of soil. From the crime scene, PW4 also picked soil samples for comparison and that the soil on the shoe and skirt was not a.ll that dry, it was like a stain. After recovering these exhibits, he submitted them to the Government Analytical Laboratory (GAL) on police form I7A.

During cross examination, PW4 testified that the dog handler was the first officer to enter into the house. Under further cross examination, PW4 testified that it would not be right to say that it is the dog handler who stepped on the skirt because a dog handler is a trained police officer, the moment a dog enters or sits in front of a house where you are tracking the dog handier will not enter, he will have to wait for PW4 as a Scene of Crime Officer (SOCO). When he entered and the dog sat in front of the l"t Appellant, the dog handler also stood and invited PW4 to enter with a team of detectives. The team that searched the house asked the l"t Appellant about the shoes and he said that the shoes that were recovered from his house were his shoes. PW4 also testified that he recovered blood samples from the body of the deceased and submitted all of them to the Government Analytical Laboratory.

PWS testified that he was €unong the team that was commanded by the OC station Kawempe together with the SOCO and the Dog handler. He testified further that a police dog was introduced to the body and it led them to the house of the Appellants which was a-lmost 500 meters away from the scene. PWS testified that the search was conducted with him among the team and they recovered, among others, faded black grey skirt that was stained with soil, a pair of closed shoes that was also stained with soil and they also recovered a pair of open shoes that had blood stains.

Whatever was recovered from the Appellant's home, was taken to police exhibited and marked. Thereafter it was submitted to the Directorate of Government Analytical Laboratories at Wandegeya together with blood samples and high virginal swabs that were retrieved from the deceased alongside the blood samples that were retrieved from both Appellants. PW5 testified during cross examination

P Aad\ M

that they did not bring the sandals to court because upon consultation with the state, the sandals were not part of the evidence they intended to rely on in this case and therefore they were left out. PW5 testifred that they went to the Appellant's house' twice on that day and also went there the next day to conduct a search in the presence of the 1"t Appellant.

PW7 was the chemist from the Government Analytical Laboratory and according to her testimony, the conclusion was that the soil traces from exhibit AK7 (skirt) and exhibit AK8 (shoe) had a high degree of comparability in terms of their structure, texture and manifestation of organic material in them. On a scale of 5 they were close to a scale of 3 to 4. When they compared each of these exhibits with the soil that was picked from the scene of crime bearing in mind that the quantities were far apart, the degree of comparability was not too strong. They could not adequately conclusively have a perfect match for them.

PW7 testified that on a scale of 1 to 5, the soil of the shoes basing on the quantities from the shoes and the quantity received from the scene, she would give it a 2. On a scale of 1 to 5, PW7 would give the skirt 2. PW7 also testified that soil may defer according to location. That as individuals walk, they interact with the soil on the ground so depending on the weather and how compact the soil is, some of it gets attached to the shoes and some of it is loose so soil can get in contact with that body and it is retained or not. All in all, PW7 testified that on a scale of 1 to 5, the possibility for the soil on the skirt and the shoe to have been soil from the scene is 2.

The Appellants on the other hand raised the defence of alibi. The testimony for both Appellants shall be delved into later.

The evidence relied upon by the prosecution in this case is largely canine evidence since there were no eye witnesses. The sniffer dog led the handler and a team of detectives to the home of the Appellants, who are husband and wife and this led to their arrest. A search was conducted on the 17th August 2017 after the Appellants had been arrested on the 16th August 2Ol7 and items were

W ¹⁰ s6/ recovered, which the prosecution relied upon in their evidence. The items recovered included a black faded grey skirt with soil on it and a pair of closed shoes which also had soil on them and were submitted to the Government Analytical Laboratory. There were also open shoes recovered which according to PW5, had soil and blood stains, but were never produced by the prosecution. PW5 stated that after consultation with the State, the open shoes were found not to be relevant to the case and left out.

Given that the major piece of evidence relied upon by the prosecution to pin the Appellants is that of the sniffer dog, it is important to note that such evidence ought to be admitted and treated with caution.

This Court, in Wilson Kyakurugaha Vs Uganda, C. A. C. A No. 51 of 2014 cited the Kenyan case of **Omondi & Anor vs R, [1967] E. A. 802,** where at p. 807 the High Court observed as follows with regard to sniffer dog evidence: -

"But we think it proper to sound a note of warning about what, without due levity, we may call the evidence of dogs. It is evidence which we think should be admitted with caution, and if admitted should be treated with great care. Before the evidence is admitted, the Court should, we think, ask for evidence as to how the dog has been trained, and for evidence as to the dog's reliability.

To say that a dog has a thousand arrests to its credit is clearly, by itself, quite unconvincing. Clear evidence that the dog had repeatedly and faultlessly followed a scent over difficult country would be required, we think, to render this kind of evidence admissible. But having received the evidence that the dog was, if we might so describe it, a reasonably reliable tracking machine, the Court must never forget that even a pack of hounds can change foxes and that this kind of evidence is quite obviously fallible."

The Court also cited, with approval, the decision by Gaswaga J., in the case of Uganda vs Muheirwe & Anor, Mbarara High Court Crim. Session Case No. 11 of 2012, where the learned judge recast and proposed the following

$\mathbb{R}$ Charlo $\mathbb{M}$ .

principles to guide trial Courts with regard to admissibility and reliance on dog evidence; as follows: -

"The evidence must be treated with utmost care (caution) by Court and given the fullest sort of explanation by the prosecution.

$2.$ There must be material before the Court establishing the experience and qualifications of the dog handler.

3. The reputation, skill and training of the tracker dog is required to be proved before the Court (of course by the handler/trainer who is familiar with the characteristics of the dog).

The circumstances relating to the actual trailing must be $4.$ demonstrated. Preservation of the scene is crucial. And the trail must not have become stale.

5. The human handler must not try to explore the inner workings of the animal's mind in relation to the conduct of the trailing. This reservation apart, he is free to describe the behaviour of the dog and give an expert opinion as to the inferences which might properly be drawn from a particular action by the dog.

6. The Court should direct its attention to the conclusion which it is minded to reach on the basis of the tracker evidence and the perils in too quickly coming to that conclusion from material not subject to the trutheliciting process of cross-examination."

After approving of these proposed principles, the Court then stated as follows: -

"We wish to add that there are two aspects that are important to be observed. Firstly, what is the threshold for such evidence to be received by the trial Court? Secondly after the reception or admissibility how is such evidence to be considered? In the first place, with regard to admissibility.

$\mathbf{R}$ 12

utc regard lt essentlal that the tralnlng and experlence ol the dog handler and hls assoclatlon wlth the dog ln questlon be estobllshed.

Secondlg, there mzst be establlshed ln evldence the naf,;,tre of tralnlng, sklll and perJorrnance oJ the dog ln questlon wlth rcgard to the pa;rtlcrzlar subJect at hand. be lt tracklng scents, or drugs, or uhatetnr speclallzed skllls lt allegedly possesses so crs to establTsh lts credentlals tor that sklll. The Joregolng are prerequlsltes before the admlsslbtlltg oJ such eridence. Itletertheless, once admltteL lt ls clear that such eutdence mast be treated tttlth cautlon cs lt ls posslblc that lt nloy be falllble."

In the instance case learned trial Judge (at page 50 of the record of appeal) held that;

"In the instant case the dog ltandler told court tlnt he receiued training in basic dog handling and acre (sic) ottaining from Nsambga Canine Headquarters and has tuorked in the dog section for 11 gears. That the dog in question no. 7O Neual, is trained in picking scent, that is a Germon Shephard that was trained for ⁶ months ond hns a certificate in canine headquarters. That the dog is trained to help police in detection of crime uthich CID cannot manage on ttteir own uithout the telp of the technical team. That in this ca.se the dog trained, sat...."

The learned trial Judge took the view that the prosecution had established the training and experience of the dog handler and that it had also established the nature of training, skill and performance of the dog in question. It was on that basis that the trial Judge admitted the canine evidence. We are satisfied that the experience of the dog handler was established by the prosecution. The dog in question, according to the evidence of PWl, had been trained for six months and had been in service for eight years. What was lacking however, was evidence of association between PWl and the dog which is one of the prerequisites to be satisfied before canine evidence can be admitted. We are of the view that the learned Judge ought to have inquired into PW1's association with the sniffer dog before admitting the evidence.

p-- C-ila'ft

Having admitted the canine evidence, the trial judge should have cautioned himself about relying on the said evidence. This he did not do. Whereas the learned trial Judge was alive to the principles set out in the case of Wlleon Kyakurugaha Vs Uganda (supra), nowhere in his judgement did he caution himself about reliance on the canine evidence before him, especially given the fact that this was the major piece of evidence that the prosecution relied on to pin the accused persons. The canine evidence would, as a matter of necessity, have to have been corroborated. This position of the law is echoed in the persuasive authority of THE STATE OF WASHINGTON v. ALLEN B. LOUCKS 98 Wn. 2d 563 (trIash. 1983f, where the Supreme Court of Washington held as follows;

oTracklng d.og euldence ls not sulficlent bg itself to convlct a crimlnal deJendant. Corroborating euldence tdentttying the defendant as the perpetrator oJ the crlme ls requlred.'

The Supreme Court of Washington went on to further observe as follows;

uThe dangers inherent ln the use of dog tracklng evldence can onlg be allevldted bg the presence of corroboratlng euldence tdentiJglng the accused. as the perpetrdtor of the crlme. Pollce dogs cannot be conclusloelg relled. on to follout the trall ol one lndluldual ff other human trcri ls cross this one, or euen corne nedr lt.'

We agree entirely with the holding in the case of THE STATE OF WASHINGTOIT v. ALLEN B. LOUCKS (supra). It is therefore our firm view that the evidence of the dog handler, could not, on its own, sustain a conviction of the appellants without corroborating evidence.

Although the learned trial judge may not have cautioned himself about reliance on the canine evidence before him, he nonetheless acknowledged the fact that the prosecution relied entirely on circumstantial evidence and that it needed to be corroborated. It is therefore necessary, as a first appellate court, to re-

F W 1,4 tr

appraise what the learned trial Judge considered as corroborating evidence to that of the sniffer dog.

There were two exhibits produced in court to wit a black faded grey skirt and closed shoes with soil which were picked from the house of the Appellants. PW4, the SOCO, is the one that recovered the exhibits and submitted them to the Government Analytical Laboratory. PW4 testihed that when he got to the crime scene, there were many people around the body and on arrival, he cordoned off the scene with tape. When the dog led the offrcers to the house of the Appellants, PW4 stated that the dog handler was the first to enter the house then the rest of them followed. when asked during cross examination whether the dog handler is the one that stepped on the skirt, which according to the Appellalts, was their mopper, he stated as follows;

(No, gou utould not be rtght to so:g tha;t it ls the dog handler utho stepped on the sklrt because a dog handler ls a tralned police officer, the rnomcnt a dog entcrs or sits ln lront of a house where gou are tracking the dog handler tttlll not enter, he utll hante to ualt fonr.c as ., SOCO, so uthen he entered and the dog sat ln Jront ol the accrtsed person the dog handler also stood and tnvlted us to enter ulth a teann of detectlues.,.'

PWs, the Investigating Officer, testified that in addition to the faded grey black skirt and the closed shoes, there was also a pair of open shoes, which was recovered with soil and a pair ofopen shoes that had blood stains. During cross examination, PW5 stated that the pair of open shoes with blood stains was, upon consultation with the state, not part of the evidence that was to be relied upon and therefore they were left out.

We are at a loss as to why the shoes with blood stains was not presented in evidence. PW7, a senior government analyst, testified that they did not receive any blood sample and only received the three exhibits; exhibit packages marked AK7, AK8 and AK9. AK7 was a faded black grey skirt stained with soil, the second

w ¹⁵

LMr/l br

one AK8 a pair of closed shoes, the third was AK9 which was soil sample collected from the scene of crime.

In our view, had the open shoes with blood stains on them, been taken to the Government Analytical Laboratory for analysis and comparison; especially given the fact that blood samples were also taken off the body of the deceased, this would have helped establish whether or not the 1^st appellant had anything to do with the murder of the deceased. It is worth noting that PW1 had earlier testified that when he got to the body, it was covered with banana peels, kavera, and dry grass, and that there were bear footed marks of footsteps around the body. PW7's evidence with regard to the laboratory tests carried out on AK7, AK8 and AK9 is therefore crucial and thus central to this case. She testified (see pages 30 and 31 of the record of appeal) as follows;

"In the stage of analysis we went ahead and first did a preliminary examination. We examined the 3 exhibits AK7 which is the skirt, had traces of soil and we managed to get a few of those, AK8 which is the shoe also had very many traces of soil on it and off course AK9 and we concluded that all the 3 had some soil on them.

We went ahead to characterize them, and analyze them first by visual to compare the soil. We did colour of the soil, the texture, the structure of the soil and we also examined it for presence of organic material and for that analysis the manifestation. We also subjected all these exhibits to an analytical technic which is the XRF it is semi quantitative because of the quantity that we had available and we were able to screen out all these elements that are naturally occurring in soil. From these findings we remarked. The conclusion was the traces from exhibit AK7 (skirt) and exhibit AK8 (shoe) had a high degree of had a high degree of comparability in terms of their structure, texture, manifestation of organic material in them. On a scale of 5 they were close to a scale of 3 to 4. $\frac{1}{2}$

When we compared each of these exhibits with the soil that was picked from the scene of crime bearing in mind that the quantities were far apart these were very traced and this was sufficient. The degree of comparability was not too strong. We could not adequately conclusively have a perfect match for them......... It would be a vague assumption to say that the soil on the skirt and the shoe tested positive for elements retrieved from the soil from the scene of crime. On a scale of 1 to 5 for the soil on the skirt and the shoe may have been soil from the scene is 2."

However, the learned trial Judge (see page 48 of the record) in evaluating PW7's afore-stated evidence, found as follows;

"In the testimony of PW7 Christine Zawede, a chemist from government analytical laboratory that the soil sample from the items recovered and that from the scene of crime had a high degree of comparability in terms of their structure, texture, manifestation of organic material in them. That on a scale of 5 they were close to a scale of between $1, 2, 3$ . With the above scale given it only means that the soil that were on the recovered items is soil from the scene of the crime."

The learned Judge further evaluated PW7's evidence (see page 51 of the record) and found as follows;

"It is the testimony of PW7 that the samples of the soil got from the scene

of the crime was highly comparable to the soil found on the recovered items belonging to the accused person on a scale of $3$ . According to the explanation of the government chemist that the soil on the skirt was so little that they could not do much (sic). Circumstantial evidence is the best kind of evidence since it does not leave room for doubts as to the honesty of the alleged eye witnesses and given the circumstances of this case as testified by the witnesses.

$I$ find that there is no logical conclusion other than the guilty of the accused persons. I therefore disagree with the opinion of the two assessors for the above reasons I find the two accused persons guilty as indicted and I therefore convict them."

With the greatest of respect to the learned Judge, we are of the view that he totally misconstrued the evidence of PW7. From the above excerpt, it appears to us that the learned trial Judge misinterpreted PW7's evidence when he held that the soil found on AK7 and AK8 was a perfect match with soil from the scene. This was clearly not the case, PW7's testimony was that the soil on the skirt (AK7) and that on the shoe (AK8) were not conclusively a perfect match to the soil from the scene.

Yet it was on that basis that he convicted the appellants. That was a misdirection on the part of the learned Judge. The evidence of PW7, if anything, was clearly inconclusive as to the issue whether the soil found on the appellants' skirt and shoes matched that of the crime scene, and could therefore not be relied upon as corroborative to that of the sniffer dog.

The East African Court of Appeal in **R v Manilal Purohit [1949] 9 EACA 58** held as follows:

"......corroboration is some additional evidence rendering it probable that the evidence of the witness is true and it is reasonably safe to act upon it, that it must be independent evidence which affixes the accused by connecting him or tending to connect him with the crime confirming in some material particular, not only the evidence that the crime has been committed, but also that the accused committed it."

The crucial question here is whether the appellants could with certainty be placed at the scene of crime. PW7's testimony was inconclusive in that regard.

The learned trial Judge mainly relied on the evidence of Pw4, PW5 and PW7 to corroborate the canine evidence. Much as sniffer dogs have played an important

role in police investigations for decades with their keen sense of smell being noticed and utilized, a lot of caution has been taken before relying on their evidence. But one fact which is clear is that, such evidence when admitted must be corroborated by some other evidence which gives strength to the canine evidence as presented through its handler or trainer. The Prosecution must provide answers to any lingering questions in the afhrmative before relying on police dog evidence. In this case, the evidence of PW4, PWS and PW7 cannot be treated as reliable corroborative evidence to sustain a conviction against the Appellartts.

We are of the firm view that the learned Judge erred in law and fact when he acted upon PW7's evidence to convict the appellants when it was quite clear that the findings did not connect them with the crime in question. PW7's evidence left crucial questions unresolved. Any lingering doubt must therefore be resolved in favour of the appellants. Had the learned trial judge applied the tests laid down for treatment of evidence that is entirely circumstantial, he would have found that the prosecution had failed to prove the case against the appellants beyond reasonable doubt.

For the reasons given above, grounds 1, 2 and 3 of the Memorandum of Appeal accordingly succeed; and having found as we have above, we find no reason to resolve grounds 4, 5 and 6 ofthe Memorandum of Appeal.

In the event, we find merit in this appeal and quash the Appellants' convictions. Accordingly, we set aside the sentences appealed against, and order for the Appellants' immediate release from prison; unless they are being held on any other lawful charge.

## We So Order.

...... day of ................................... Delivered and dated this ....................................

**RICHARD BUTEERA**

**Deputy Chief Justice**

$cBk$ $m(r)$

**CHRISTOPHER GASHIRABAKE**

**Justice of Appeal**

**OSCAR JOHN KIHIKA**

**Justice of Appeal**

$\overline{a}$