IN THE SUPREME COURT OF BRITISH COLUMBIA

Date: 20101027

Docket: 28460

Registry:
Terrace

Regina

Respondent

Simeon Daniel
Denholme

Appellant

Before:
The Honourable Mr. Justice Voith

Reasons for Judgment

[1]            
Following a seven day trial the appellant, Simeon Denholme, was
convicted of assault and unlawful confinement under sections 266 and 279 (2) of
the Criminal Code, R.S.C. 1985, c. C-46.

[2]            
Denholme now appeals that conviction and raises the following distinct
issues:

a)       Did the
learned trial judge err in making use of inadmissible statements from various
co-accused to arrive at a determination of guilt against the appellant?

b)       Did the
learned trial judge err in concluding that the central Crown witness and
complainant, Mr. Wilson, was credible?

[3]            
I have determined that the first of the issues appealed is dispositive of
this appeal, though I have also dealt with the second issue raised. I have also
concluded that section 686 (1)(b)(iii) of the Criminal Code does not
apply in the circumstances of this case.

Overview

[4]            
The appellant and three co-accused, Finn, Buick and Silva, were accused
of assaulting and unlawfully confining Wilson in the early morning hours of
January 25, 2008, at or near Kitimat, British Columbia.

[5]            
It was alleged that Wilson had made arrangements to meet Denholme that
night, but when Wilson arrived at the meeting place he was assaulted and forced
at knifepoint into the trunk of Silva’s car.

[6]            
The complainant said he called 9-1-1 on his cell phone from the trunk of
the car, then jumped out while the car was in motion and hid in a snow bank
until the police arrived.

[7]            
Denholme did not make a statement to the police. However, the three
co-accused Finn, Buick and Silva did. These statements were admitted into
evidence after a voir dire. None of the accused gave evidence at trial.

[8]            
Two additional matters, which are expressly recognized in the Oral Reasons
for Judgment of the trial judge are relevant: R. v. Denholme (28 August
2009), Kitimat 27073-2-C (B.C.P.C.). First, Wilson was a very difficult witness
for the Crown.

[9]            
Wilson not only contradicted himself repeatedly, but he also said
several times that he was lying, that he was prepared to face a contempt charge
and that he provided the police with a false statement. He further suggested
that he might frame his evidence so as not to incriminate himself. The trial
judge, who ultimately accepted Wilson’s evidence, made a number of comments
regarding Wilson and his credibility. I have reproduced these comments at some
length because the acute nature of this issue is relevant and underlies both grounds
raised in the appeal.

a)         … Wilson started his testimony with all the guise
of an institutionalized, hardened criminal. He did not wish to testify. He did
not even admit to knowing Simeon Denholme. He was in custody at the time. The
accused were out on bail. (para. 10)

b)         Wilson, as I say, has a long criminal record. His
seven-page statement given to the police shortly after the incident did not
refresh his memory. This kind of testimony is unfortunately all too common in
these courts, particularly where the witness is in jail. At the outset of his
evidence, he included such well-worn phrases as "Well, we all know what
happened. I just don’t know who did it"; "Some guys did this to me,
but I don’t remember much about it"; "I remember how I got my
injuries, but I want to forget"; "I don’t remember making [I’m sorry,
it was] a 27-page statement, and reviewing it does not refresh my memory.” (para. 11)

c)         As he testified in this matter, Wilson seemed to
observe one or more of the accused in the body of the courtroom either smiling
or smirking at his somewhat non-compelling testimony along the lines just
related. He completely reversed direction and said, "They did do this. I
remember they put me in the trunk. They forced me into it. Everything in my
world changed. I want to see them in reds," prison garb, I gather. (para. 13)

d)         … Wilson claimed he had used cocaine that night,
however said the effect of the drug would not cause him to hallucinate a
beating. The court must be cognizant throughout of this admission of drug use
when assessing his overall reliability as a witness. (para. 25)

e)         Wilson’s decision to implicate the accused men was
not, I am satisfied, driven by some sort of epiphany. He was clearly mad and
upset at their demeanour in court, at least one
or more of the four accused. The Crown finished his examination in chief on
that note. (para. 27)

f)          To complicate matters, Wilson reverted to form in
his cross-examination initially by Mr. Cranston and then the other defence
counsel. He said to Mr. Cranston, "Yes, he [Denholme] did put the
knife to my throat but for you I was lying." He was not easy to
cross-examine. His glib responses continued further under cross-examination
from Mr. Mulder on behalf of Mr. Buick, where he said words such as
"I admit I was lying. I am trying to get them off," or "No, I
did not have knives out and ready. Yeah, maybe I did. Yeah, I had the knife out
ready." These comments were made one after the other in rapid succession
and hardly convincing. (para. 28)

g)         With regard to drug usage, he said yes he had done
cocaine: "Sure I was high." Initially, he said five to six times he
had used it that day; then he immediately upped it to say probably five to 30
times. He said, "No, I was not trying to scare Denholme." Then he
said, "Sure I was. That’s why I took the knives. I was confused. I was
fucked up. I probably was up a couple of days before this happened and using as
much cocaine as I could." This flip-flopping testimonial background to the
court must be kept in mind when assessing, again, his overall credibility. (para. 29)

h)         Continuing on with the cross-examination by Mr. Mulder
regarding Mr. Silva, who was not even Mr. Mulder’s client, Wilson
said, "I’m not sure he was even there. I don’t know why he’s here. He’s
not guilty." In cross-examination by Mr. Crampton, on behalf of Messrs. Finn
and Silva, Wilson continued to do what he could to recant his earlier evidence
in chief that had implicated each accused. He did so in an unconvincing,
supercilious manner. He is a person whose evidence has to be, as I say,
scrutinized in its entirety very carefully. (para. 30)

i)          The three accused’s statements generally do
support the chronology of events as described by Wilson in his brief attempt, I
find, to be more or less truthful as to what occurred, in his examination in
chief. He did so for reasons he somewhat inarticulately explained in court. His
resolve to do so was brief, and he reverted, as I say, to form in
cross-examination, responding facetiously to any defence counsel’s
propositions. He wanted to get the boys off, he said, basically, and that was
the tenor of this evidence on cross-examination. (para. 46)

j)          Despite Wilson’s best efforts, initially in chief
then later in cross-examination, to upset the trial process, which he did for a
brief period, I conclude that he actually told the truth about what occurred
when one looks at the corroboration for his statements, brief though his
acquaintanceship with the truth in his examination in chief may have been. …
(para. 85)

k)         At trial, Wilson, after initially denying even
knowing Denholme, and with the Crown about to stay the proceedings, had a brief
interlude of clarity, I conclude. He obviously was upset by the actions in
court of one or more of the accused as he unconvincingly testified to his total
lack of recollection of how he came to be injured and what occurred that night.
(para. 91)

l)          The issue squarely
before the court is whether or not Wilson, although primarily a stranger to the
truth, did tell the truth for a short period of time during his examination in
chief. Can his testimony be relied upon as proof beyond a reasonable doubt? Is there
corroboration? (para. 93)

[10]        
Second, the trial judge was clearly aware that Denholme did not provide
a statement and that this distinguished him from his co-accused. Thus, the trial
judge correctly acknowledged that he could not use any of the statements made
by the co-accused to the police in considering the case against Denholme. Accordingly,
the trial judge said:

a)         The court must look to other evidence to see if
there is any corroboration of Wilson’s assertions made in chief as to having
been assaulted, particularly by Denholme and Finn. The court, after a voir
dire, admitted into evidence statements to police of Buick, Silva, and Finn.
They are admissible only as against each individual but not to implicate their co-accused,
including Denholme who did not provide a statement. (para. 45)

b)         The case against Simeon Denholme is different.
Like the others, he did not testify; however he made no statement to the
authorities either. … (para. 86)

c)         In order to find Denholme guilty as the principal
under 21 (a), the court can only look towards his evidence, Wilson’s evidence. (para. 87)

d)         I infer on all of the
evidence that is admissible as against Denholme that it was he who was the directing
force behind events. … (para. 98)

Issue 1:  Did the learned trial judge
err in making use of inadmissible statements from various co-accused against
the appellant?

a) The Difficulty
Arising from the Trial Judge’s Reasons

[11]        
Notwithstanding that the trial judge had alerted himself to the hazards
and impropriety of relying on the statements of the co-accused, in considering
the case against Denholme, he appears to have done just that. The trial judge
dealt with the evidence against each co-accused, made the relevant findings of
facts relating to that individual and expressed his conclusions in relation to
each such co-accused sequentially. In that portion of his reasons which dealt with
Denholme, the trial judge made the following findings of fact, reproduced below,
which I am satisfied could only be inferred from the statements of the
co-accused as opposed to evidence provided by Wilson, another witness, or from
inferences based on other permissible evidence:

a)         The four co-accused left from the appellant’s
residence on the evening in question (para. 88);

b)         One of the co-accused, Silva, "was awakened about
midnight" by the appellant to drive him "as he did not have a vehicle"
(para. 88);

c)         Another of the co-accused, Finn, was dropped off
prior to the meeting time with Wilson to reconnoitre (para. 89 and 95);

d)         “Denholme arranged the transportation from his
premises to the meeting” (para. 94); and

e)         “As Silva’s vehicle
was driving off, someone gave directions. Silva was told by somebody he did not
identify as to where to go. Finn and Buick each denied giving such direction to
Silva. Denholme was the only other person remaining who could have been in a
position to do so, directing Silva where to go.” (para. 97)

[12]        
Importantly, the trial judge said at para. 102:

… Wilson’s evidence, which
although not consistent and difficult at times to accept and even resiled from
to an extent on cross-examination, nevertheless was supported by corroborated
statements of the accused sufficient to satisfy me that the only sensible
conclusion is that Denholme was behind the events that overtook the pathetic Mr. Wilson
that night.

b) Analysis

[13]        
Counsel for both parties agree that the out-of-court statements of a co-accused
are not admissible against an accused. This proposition is clear to the extent
that it relates to the use of such statements to make a finding of fact, or to
establish one of the elements of an offense. The proposition, or at least its
practical application, is less clear when such statements are used to
corroborate the evidence of a witness with a credibility problem. This lack of clarity
stems from the interrelationship of the principles established by the Ontario
Court of Appeal in R. v. Perciballi, (2001), 54 O.R. (3d) 346 (C.A.),
154 C.C.C. (3d) 481, aff’d 2002 SCC 51, [2002] 2 S.C.R. 761, and the Supreme
Court of Canada decision of R. v. Rojas, 2008 SCC 56, [2008] 3 S.C.R.
111, respectively. These two cases explained the extent to which, if at all, out-of-court
statements of a co-accused can be used to corroborate the evidence of a witness
as against the accused.

[14]        
In Rojas, Madam Justice Charron, who also wrote for the majority in
Perciballi, held that the principles established in Perciballi
and Rojas were consistent. With great respect, it is not easy to glean
what principles emerge from these two decisions. I have quoted from the
relevant parts of these judgments at some length to confirm what took place in
each case, what specific issues arose and what principles gave rise to the
results in each of the decisions.

[15]        
In Rojas, before the British Columbia Court of Appeal (2006 BCCA
193, 225 B.C.A.C. 32), Madam Justice Ryan, for the court, summarized the facts
in Perciballi:

[82]      Counsel likened this
aspect of the charge to instructions given to a jury in the Ontario case of R.
v. Perciballi (2001), 154 C.C.C. (3d) 481 (Ont. C.A.), aff’d 164 C.C.C.
(3d) 480 (S.C.C.). In Perciballi it was alleged that a co-accused,
Antonio Portante, had placed one of two diversionary calls to the police to
facilitate an armoured car robbery. The only evidence against Antonio Portante
was a witness who placed him in a restaurant near the pay telephone from which
the call was placed, and the evidence of an accomplice, DeFrancesca, who
testified that he placed the other of the two calls. Portante’s brother,
Angelo, another co-accused, made an out-of-court statement that Antonio
Portante had placed one of the two calls. The statement made by Angelo Portante
was admissible against him to show that he, Angelo Portante, was aware of
details of the plot. It was not admissible against Antonio Portante. Against
Antonio Portante it was merely hearsay. Nonetheless the trial judge permitted
the Crown to argue to the jury that the statement of Angelo Portante confirmed
the evidence of DeFrancesco. The Ontario Court of Appeal quashed the conviction
of Antonio Portante holding that the effect of this use of the evidence was to
indirectly achieve what the Crown could not do directly, that is, to permit the
jury to use the Angelo Portante statement against Antonio Portante. Counsel for
Miguel Rojas says that this parallels what has happened in the case at bar with
Hugo Rojas’ statement.

[16]        
In Perciballi, Charron J.A., as she then was, provided the
following background to the case:

[69]      …What is at issue on this appeal is the reference
made by Crown counsel to this evidence in his closing address, indicating to
the jury that they could consider Angelo Portante’s statement to Coiro in their
assessment of DeFrancesca’s credibility. The question becomes whether this
constituted a permissible use of the evidence or whether it offended the rule
against hearsay.

[70]      This issue was canvassed at trial at the conclusion
of the evidence. Crown counsel sought leave to make reference to this statement
in his closing address as one of ten items of evidence which could be
considered by the jury in support of DeFrancesca’s testimony. Counsel for the
appellant was opposed to the Crown addressing the jury in this fashion arguing
that, since this evidence was inadmissible against Antonio Portante, it was
impermissible and highly prejudicial for the Crown to tell the jury that they
could nonetheless use it indirectly to support the Crown’s case against his
client.

[71]      The trial judge ultimately ruled against the
defence’s objection. He noted that DeFrancesca’s credibility was very much at
issue in the trial and that he intended to give a strong caution in his charge
against relying on the testimony of accomplices. In view of this anticipated
caution, the context in which the Crown intended to make reference to this
evidence simply as one of ten items to consider in assessing DeFrancesca’s
credibility, and the instruction given during the course of the trial that this
statement did not constitute evidence against Antonio Portante, the trial judge
saw no reason to direct Crown counsel not to make the intended reference.

[72]      Crown counsel referred to this item of evidence in
accordance with the trial judge’s ruling. The submission was made in the
portion of the address which dealt specifically with the evidence against the
appellant Antonio Portante. Crown counsel said:

So when looking at Mr. DeFrancesca, you can look to the
fact that everything that he’s told the police so far has been confirmed. You
can look to the evidence of Mr. Coiro, who indicated that Angelo Portante
had told him after the robbery that Tony Portante had made a diversionary call.
And you may recall that His Honour indicated that that is not evidence
against Tony Portante, the fact of what his brother said, and I want to
repeat that now as well. I do not need to rely on that information to prove
that Tony made the phone call but it is evidence you can look at in
assessing Mr. DeFrancesca’s testimony as to whether he’s telling the truth
about the fact that he only made one call. [Emphasis added.]

[17]        
Charron J.A., at para. 74, then framed the question raised on the
appeal as "whether the statement of a co-accused, introduced in a joint
trial against its maker, can be used to support evidence against another
accused in the trial."

[18]        
Charron J.A concluded, at para. 76, that "the use of Angelo
Portante’s statement to support the case against Antonio Portante constituted
an impermissible use of the evidence. It infringed basic principles of
evidence". To support her conclusions she quoted at length, at para. 78,
from the reasons for judgment of Mr. Justice Finlayson in R. v. C.(B.) (1993),
12 O.R. (3d) 608 at 614-617 (C.A.), 80 C.C.C. (3d) 467, leave ref’d, [1993] 3
S.C.R. viii, 15 O.R. (3d) xvi:

I think it is necessary to go back to first principles. The
tension that exists in this case is between the desirability of having
confirmatory evidence that makes more reliable the evidence of the complainant,
and the long-standing prohibition in criminal cases against the admission into
evidence of the confession of one co-accused against his fellow accused. An
out- of-court confession is admissible against the accused who made the
statement because it is a statement against interest and falls within that
recognized exception to the hearsay rule. However, the statement remains
hearsay against the co-accused and the trial judge is under a heavy burden to
instruct the jury as to the limited use it can make of it. In R. v.McFall, 48
C.C.C. (2d) 225, [1980] 1 S.C.R. 321 at pp. 338-39, Estey J. stated at
p. 240:

This burden is discharged only by
the clearest instruction to the jury that only the statement made by the
accused himself is admissible against him, and his statement in turn is
inadmissible against his co- accused. There is in these circumstances a
constant risk because any instruction short of the foregoing will jeopardize
and may destroy any chance of a fair and proper trial.

…

The Crown now submits that this long-standing rule has been
modified by the Supreme Court of Canada in R. v. Vetrovec, [1982] 1
S.C.R. 811, 67 C.C.C. (2d) 1, when it relaxed the rule relating to
corroboration. Corroboration is evidence which confirms a witness’s testimony
making the testimony more likely to be true. In R. v. Baskerville,
[1916] 2 KB. 658, 86 L.J.K.B. 28, the Court of Criminal Appeal in England gave
corroboration a more technical meaning. It required that such evidence be: (a)
independent; (b) evidence that a crime has been committed; and (c) evidence
that the accused committed the crime.

While most statutory requirements of corroboration have been
eliminated, there remained a common law requirement that a jury be warned
against the dangers of convicting without corroboration in three instances
where the prosecution’s case consisted of: (a) the evidence of an accomplice; (b)
the evidence of a sworn but still young child; or (c) the evidence of a witness
with a “disreputable” reputation.

In Vetrovec, the Supreme Court of Canada discarded the
technical meaning of corroboration adopted in Baskerville, in favour of
the more common meaning of the word, i.e., that it is evidence tending to
confirm other evidence. The operative word is still “evidence”, however.
Hearsay is not evidence, and where it is admitted as evidence as an exception
to the hearsay rule solely against one co-accused, it does not become evidence
against the other. It is difficult to see how this change in the law
with respect to corroboration affects the rule regarding the restricted use of
confessions which are hearsay. [Emphasis added.]

…

In the case on appeal, the co-appellants did not adopt in the
witness-box their earlier statements to the police. Therefore, those statements
were made out of court and if they were offered to establish that they were
true, they were hearsay. While counsel for the Crown acknowledged that
admissions made by a party are exceptions to the hearsay rule and can only be
used against the person who made them, she argued that the trial judge properly
used the statements of each appellant to confirm the evidence of the
complainant. As authority for this submission, she referred to Vetrovec,
supra, and R. v. D.(S.) (1992), 10 O.R. (3d) 402 (C.A.). . In
particular, she picked up on Dickson J.’s reference to Wigmore on
Evidence, Vol. 7 (1978), and submitted that the approach now is to
determine if the complainant is worthy of credit. In making this submission,
she overlooked the significance of the word “evidence”. We do not look
everywhere and anywhere for confirmation. We look to evidence. Evidence used to
confirm a complainant’s evidence against Brian C. must still be admissible
against Brian C.

…

Counsel for the Crown referred us to the trial judgment in D.(S.)
and to the trial judge’s instruction to himself. After stating the conventional
rule that one accused’s statement is not admissible against another accused,
the trial judge stated:

In a hypothetical situation, if the
complainant swears that B held a knife to her throat while A had intercourse
with her and that the rape thereby accomplished, it may be, as here, that the
court would be unwilling to convict based solely on her assertion. If A then
confessed the rape was accomplished in [the] very manner, A’s confession is not
admissible against B. However, A’s confession, if believed, is capable of
confirming or corroborating, to use a loaded word, the complainant’s evidence.
The court may then convict B, not on the confession but on the confirmed
evidence of the complainant.

The above appears to be an inaccurate paraphrase of an
illustration given by Dickson J. in Vetrovec at pp. 826-27 S.C.R.,
p. 14 C.C.C.:

The point can be illustrated with
the following simple example. The accomplice, “A”, testifies against the
accused “B” and “C”. There is evidence implicating “B” in the crime, but
no evidence implicating “C”. Nevertheless, since the supporting evidence
relates to a vital issue in the case (the guilt of one of the accused) it
bolsters the credibility of “A” and increases the probability that he is
telling the truth. We therefore believe his story and convict both “B” and “C”.
[Emphasis added in C.(B.).]

The Court of Appeal did state, by
way of obiter, that the trial judge’s hypothetical in D.(S.) was in
accord with that illustration. However, this quotation from Vetrovec,
clearly dealt with the testimony of an accomplice. The question of
hearsay was not in issue. It is difficult to see how Vetrovec could be
authority for a widening of one of the hearsay exceptions to include statements
made by one accused to implicate another. The trial judge’s distinction between
admitting evidence directly against a co-appellant and admitting evidence
simply to confirm a complainant’s allegations is untenable. [Emphasis in
original.]

[19]        
In Perciballi, Charron J.A. also addressed the dissent of Doherty
J.A.:

[84]      There is no doubt in this case that the statement
made by Angelo to Coiro is confirmatory of DeFrancesco’s testimony as defined
in Vetrovec. Angelo’s statement that his brother Antonio made a diversionary
call to the police on the night of the robbery strengthens the belief that
DeFrancesco’s testimony is truthful when he says that he, DeFrancesco, only
made one of the two calls. The statement must nonetheless be excluded from
consideration in the case as against Antonio, not because it lacks
corroborative value from a logical or a common sense standpoint, but for the
same policy considerations that define the scope of admissibility of an
accused’s out- of-court statement and limit its use as against its maker only.
The underlying principle is one of fairness to the party who cannot
cross-examine the maker of the statement. While the maker can hardly complain
about the inability to cross-examine himself, the same cannot be said of the
co- accused.

[85]      Hence, I do not agree
with my colleague that the analysis of this court in C.(B.) constitutes
a departure from the approach in Vetrovec. Nor, in my view, does it
conflict with any of the other decisions referred to by my colleague Doherty J.A.
There is no question that the corroborative nature of any item of evidence is
determined according to the approach in Vetrovec. Evidence that tends to
confirm other evidence is corroborative in nature. This does not mean however
that the admissibility or use of that evidence may not be limited by other
applicable rules of evidence.

[20]        
Doherty J.A., in Perciballi, had expressed the view that C.(B.)
had been overtaken by Vetrovec and said:

[116]    Two features of the criminal trial process intersect
at this ground of appeal: the admissibility of evidence to prove a fact in
issue, and the assessment of credibility. The former is question of law for the
trial judge, the latter is a question of fact and is quintessentially a matter
for the jury. While admissibility is governed by rules and principles which can
be complex and legalistic, credibility assessments turn on common sense and an
understanding of the human condition. We trust juries to assess credibility
because we respect the combined common sense and understanding of human nature
that twelve members of a jury bring to bear on questions of credibility.
Judicial instructions as to the assessment of credibility that run contrary to
common sense undermine the function of the jury in the criminal trial process. Such
instructions will either be ignored, in which case the process becomes
dishonest, or they will be applied, thereby detracting from the very strengths
which are said to make the jury a bulwark of our criminal justice system.

…

[123]    As I understand my colleague’s reasons, she would
require the jury to assess the credibility of DeFrancesca’s evidence that he
made only one call as against each accused separately. In making these
individual assessments of credibility, the jury could look only to other
evidence which was admissible against each accused. On this approach, the jury
could reach different conclusions as to the veracity of DeFrancesca’s assertion
that he made one call as regards each accused. For example, in so far as the
case against Angelo was concerned, the jury might find as a fact that the
appellant made one call, but in so far as the case against the appellant was
concerned, the jury might find that DeFrancesca made both calls.

[124]    An instruction to a jury that it could make
different findings as to the veracity of the same piece of evidence as against
different accused could only bewilder a jury. Certainly, such an instruction
seems far removed from a direction which calls upon the jury to apply its
common sense to the assessment of the credibility of witnesses and the veracity
of specific pieces of evidence.

[125]    A jury may believe
certain evidence, may disbelieve it, or may be unable to decide whether the
evidence is true. It makes that determination once. In my view, a jury cannot
decide on the one hand that part of a witness’s testimony is true when considering
the case against one accused and then decide that the same testimony is false
when considering the case against another accused. That does not mean that
credibility assessments have the same impact on all accused. Having made that
credibility determination and any findings of fact which flow from that
determination, the jury must still determine whether, based on evidence which
is admissible against each accused, the Crown has proved its case. As set out
above, a finding that DeFrancesca’s evidence that he made one call was credible
could support a finding that he made only one call. That finding was admissible
against the appellant and the other accused. It remained, however, for the
Crown to establish beyond a reasonable doubt based on evidence admissible
against the appellant that he made the other call.

[21]        
At bottom, Perciballi appears to express the principled view that
an out-of-court statement of a co-accused, by virtue of its hearsay status,
cannot be used to corroborate the evidence of a witness as against another
accused. The out-of-court statement of a co-accused, as a matter of trial fairness,
cannot be properly tested. Furthermore, the hearsay status of such a statement,
absent an exception, precludes the statement from constituting
"evidence". Absent the statement constituting "evidence",
it simply cannot serve to corroborate other evidence.

[22]        
In Rojas, two brothers, Hugo Rojas and Miguel Rojas, were charged
with second-degree murder. At their joint trial before a jury, the evidence
against them was largely circumstantial, with the exception of several
statements they made to the witness David Miranda in the days following the
murder. Both accused were convicted. On appeal, one of the grounds raised was
that the trial judge erred in permitting the jury to use evidence of an out-of-court
statement, admissible against only one accused, to bolster the credibility of
an unsavoury witness with respect to matters implicating the co-accused.

[23]        
Writing for the British Columbia Court of Appeal in Rojas, Ryan
J.A. said:

[83]      With respect, I am of
the view that what was at stake in each of the two cases is quite different. In
Perciballi the trial judge permitted the jury to use the out-of-court
statement of the co-accused as confirmation that the testimony of the
accomplice against the accused was true. The opposite occurred in the case at
bar – the jury was told that there was evidence in the case against Hugo Rojas
that tended to confirm David Miranda’s testimony that Hugo Rojas made certain out-of-court
statements to him.

[84]      I accept that it may
be the case that the jury would take from the trial judge’s charge that if
David Miranda was telling the truth about Hugo Rojas it could consider that
fact in assessing his general credibility and reliability when it came to
consider whether David Miranda was telling the truth when he said that Miguel
Rojas made certain admissions to him as well. I am not persuaded that this
would be an impermissible error. It conforms to what was said about the nature
of confirmatory, as opposed to corroborative evidence, in Vetrovec.

[85]      In Vetrovec
Dickson J. gave this illustration at pp. 826-827:

The point can be illustrated with the following example. The
accomplice, “A”, testifies against the accused “B” and “C”. There is evidence
implicating “B” in the crime, but no evidence implicating “C”. Nevertheless,
since the supporting evidence relates to a vital issue in the case (the guilt
of one of the accused) it bolsters the credibility of “A” and increases the
probability that he is telling the truth. We therefore believe his story and
convict ”B” and “C”. [Emphasis added in Rojas.]

[86]      The complaint of Miguel
Rojas is that David Miranda’s credibility is bolstered by evidence admissible
only as against Hugo Rojas. In the example above, the credibility of the
accomplice is bolstered by evidence only admissible against accused “B”. There
is no difference in substance.

[24]        
I note in addressing the relationship of Perciballi and Rojas,
that the very paragraph from Vetrovec, referred to by Ryan J.A. in Rojas,
was also considered by Finlayson J.A. in C.(B.), with the approval of Charron
J.A. in Perciballi, as being irrelevant to an analysis of the  permitted
use of an out-of-court statement of a co-accused to support the credibility of
a witness. 

[25]        
In Rojas, Charron J., writing for the court, said:

[3]        On the first ground, I
conclude that the trial judge did not err in his instructions on how to assess
the credibility of the unsavoury witness, David Miranda. The trial judge gave
the jury a strong caution against relying on this witness’s testimony without
independent confirmation in accordance with the principles set out in Vetrovec
v. The Queen, [1982] 1 S.C.R. 811. The most important aspects of Miranda’s
testimony were inculpatory statements allegedly made to him by each appellant.
The trial judge therefore properly suggested that the jury look to independent
evidence for confirmation of Miranda’s testimony about these statements. He
then highlighted for the jury certain aspects of the witness’s testimony for
which there was potentially confirmatory or contradictory evidence. At no point
did the trial judge tell the jury that they could use the statements of Hugo to
confirm Miranda’s evidence on matters involving Miguel, or vice versa. To the
contrary, he repeatedly instructed the jury not to use the statements of one
appellant against the other. In the circumstances, the appellants’ reliance on
this Court’s decision in R. v. Perciballi, [2002] 2 S.C.R. 761. 2002 SCC
51, is misguided, and their argument was properly dismissed in the court below.

[26]        
Charron J. further said:

[24]      I agree with Ryan J.A.’s
analysis on this point. It is true that the underlying logic to the majority
judgment in Perciballi could arguably be extended so as to require a
parsing of a single witness’s credibility in the manner advocated by the
appellants in this case. In other words, in deciding whether Miranda was
telling the truth about the alleged statements made by Miguel, the jury would
have to discount the fact that his overall credibility may have been bolstered
by the evidence that confirmed the truth of his testimony concerning Hugo’s
statements, and vice versa. Indeed, Doherty J.A. argued eloquently in his
dissenting reasons in Perciballi that if pushed to its limits, the
majority’s ruling would have that precise effect. I remain unpersuaded that the
principle in Perciballi leads inevitably to this result.

[27]        
The difficulty is that Charron J.A. in Perciballi at para. 68,
stated that “the trial judge instructed the jury in clear and strong terms that
the statement allegedly made by Angelo Portante was not admissible against
Antonio Portante. It was only admissible against Angelo Portante. …”. Given
this statement, what then properly distinguishes the conclusions in Perciballi
and Rojas respectively?

[28]        
Is it because in Perciballi the trial judge expressly allowed
Crown counsel to refer to the statement of a co-accused, together with other
evidence, to support a witness’ testimony, while in Rojas that specific
issue did not arise? Does Rojas reflect some easing of the principled
concerns expressed by the majority in Perciballi in favour of the more
pragmatic views of Doherty J.A. in dissent? Does Rojas reflect a
recognition that the use of the statements of a co-accused in the assessment of
credibility of a witness by a finder of fact is "inevitable"? Is Rojas
limited to the proposition that a trial judge who has given the jury a Vetrovec
caution need not provide a further explicit warning on the use of out-of-court
statements to confirm a witness’ credibility?

[29]        
The issues I raise are not academic or esoteric. These issues reflect a
very real divergence in the views of commentators on the extent to which, or
the manner in which, the out-of-court statements of a co-accused can properly
be used to confirm the evidence against another accused. These comments run the
gamut from endorsing and rendering proper the use of an out-of-court statement
of a co-accused against a different accused to "corroborate"
evidence, to recognizing the uncertainty surrounding the issue, to drawing a
distinction between a trial judge who endorses the use of an out-of-court
statement of a co-accused to corroborate the credibility of a witness to a jury
with a judge who does not address this matter expressly.

[30]        
In light of the express use made by the trial judge of the various out-of-court
statements made by Denholme’s co-accused to corroborate or confirm the evidence
of Wilson, the issue is of significant importance. The following statements
reflect the uncertainty I have referred to:

a)         Confirmatory evidence against both accused is
capable of arising from evidence admissible against one accused only. (Rojas)

S. Casey Hill et al., eds., McWilliams’
Canadian Criminal Evidence, 4th ed., looseleaf (Aurora, Ont.: Canada Law
Book, 2010) vol. 2 at 31-47.

b)         Where a disreputable witness testifies in a joint
trial and a Vetrovec caution is given in connection with his or her
evidence, the trial judge is not required to instruct the jurors to make
separate and distinct assessments of the witness’ credibility as it relates to
each accused, especially where the instructions make it clear that jurors are
to decide the case against each accused separately, based on evidence
admissible in relation to that accused. (Rojas)

David Watt, Watt’s Manual of Criminal
Evidence (Toronto: Carswell, 2009) at 33:01.

c)         The trial judge did not err in his instructions on
how to assess M’s credibility. He gave the jury a strong caution against
relying on M’s testimony without independent confirmation and highlighted for
the jury certain aspects of M’s testimony for which there was potentially
confirmatory or contradictory evidence. At no point did the trial judge tell
the jury that they could use the statements of H.R. to confirm M’s evidence on
matters involving M.R., or vice versa. On the contrary, he repeatedly
instructed the jury not to use the statements of one accused against the other
and to decide the case against each accused separately, based on a careful
consideration of the evidence admissible as against that accused. Given the
intangible nature of any credibility assessment, it is inevitable for the
jury’s assessment of the overall credibility of a witness at a joint trial to
be influenced in some way by the totality of the evidence that it has heard,
including evidence relating solely to one co-accused. This result does not
constitute an impermissible use of the out-of-court statement as alleged. (Rojas)

Kent Roach ed., “Criminal Appeals
in the Supreme Court of Canada and Federal Criminal Law Amendments” (2009) 54
Criminal Law Quarterly 283 at 289.

d)         [184] In Rojas, the SCC considered two
issues: (1) the use of out-of-court statements made by one of the accused to
bolster the general credibility of a witness who testifies against both of
them, and (2) whether and how judges should give a Duncan instruction with respect
to mixed out-of-court statements.

[185]    With respect to the first issue, the accused argued
that the decision of the OCA in Perciballi required that the jury be
instructed to make separate findings of credibility against the witness for
each accused. In Perciballi, Charron J.A. (as she then was), held that
the Crown could not invite the jury to use an inculpatory statement made by one
accused to support the testimony of a witness who also implicated the second
accused. This was “for the same policy considerations that define the scope of
admissibility of an accused’s out-of-court statement and limit its use as
against its maker only”. Dissenting in Perciballi, Doherty J.A. had
cautioned that this would logically require a jury to make different findings
of credibility with respect to the same witness on the same evidence-one each
with respect to each of the accused.

[186]    That precise result was now being advocated by the
accused in Rojas. Now at the Supreme Court, Charron J. restated her view
that while the jury could not be expressly directed to use an inculpatory
statement when assessing the case against the accused who did not make it,
neither did the jury have to be specifically cautioned against using it as part
of the totality of the evidence relating to the credibility of a witness.

“Year in Review: Developments in Canadian Law in 2008”
(2009) 67(2) U.T. Fac. L. Rev. 359.

e)         Where there are co-accused, the out-of-court
statement of one of the accused to an accomplice is capable of
constituting “confirmatory evidence” of the reliability of that witness
in the sense that the witness is telling the truth even though the statement is
not admissible against the other accused. (Rojas at para. 25)

E.G. Ewaschuk, Criminal Pleadings & Practice in
Canada, 2nd ed. vol. 2, looseleaf (Aurora, Ont.: Canada Law Book, 2010) at
16-469.

[31]        
Neither the issues raised in Perciballi and Rojas nor the
divergence of views expressed by the foregoing authors or commentators appears
to have been addressed or considered in any of the cases which postdate Rojas.

[32]        
In the present case, I am satisfied that the learned trial judge
improperly relied on and made use of the statements of the various co-accused
when considering the case against Denholme. I say this for several reasons.

[33]        
First, notwithstanding his expressed concern about using inadmissible evidence
when considering the case against Denholme, the trial judge repeatedly looked
to the evidence of the various co-accused in making findings of fact that
pertained to Denholme alone. I have detailed examples of this at paragraph 10
of these reasons for judgment. These examples are not instances where the
evidence of the co-accused was relied on to corroborate the evidence of Wilson.
Rather they are examples of findings of fact whose only source can be the
statements of the co-accused. In addition, I am not satisfied that these
finding of fact were only of incidental or peripheral relevance.

[34]        
Second, to the extent that the trial judge relied on the statements of
the various co-accused to corroborate the evidence of Wilson, I find that he
erred. In Rojas, Charron J. reconfirmed that the trial judge in Perciballi
had erred in expressly allowing the jury to consider an out-of-court statement
admissible against only one accused to corroborate the evidence of a witness. There
can be no difference in a trial judge providing such an express confirmation to
a jury and a trial judge, sitting as a trier of fact, expressly engaging in
that exercise or activity.

[35]        
Based on Perciballi and Rojas it seems clear that a trial
judge, if asked by a jury whether a statement of a co-accused can be used to
corroborate the evidence of a witness which relates to another accused, would
be required to inform the jury that it could not. The contents of the statement
would be hearsay. The statement would not constitute "evidence"
against the accused. It could not be tested or cross-examined on, and its use would
be unfair. Whatever practical difficulties or elements of
"inevitability" might arise from a jury inferentially or indirectly
using such evidence, as part of its overall assessment of credibility, are not
engaged when the trial judge purports to openly and directly make use of the
out-of-court statement of a co-accused in the manner I have described.

[36]        
Third, the hazards associated with allowing the statement of a co-accused
to corroborate the credibility of a witness are accentuated in a case such as
the present one where, as acknowledged by the trial judge, the assessment of Wilson’s
credibility was so fraught with difficulty. It is difficult to see how the
credibility of a weak or poor witness can be enhanced or corroborated by
inadmissible hearsay evidence.

[37]        
I also accept, as counsel for the Crown has noted, that the trial judge,
in considering the credibility of Wilson, also looked to the evidence of other
witnesses, physical evidence, and various reasonable inferences of fact to
corroborate the evidence of Wilson. I do not believe this detracts from the
fundamental nature of the error that was made. In Perciballi the
statement of a co-accused was only one of 10 pieces of evidence which the Crown
wished to have the jury consider when corroborating the evidence of a witness. Nevertheless,
this use of the evidence was impermissible.

Issue 2:        Did the learned trial
judge err in concluding that the central Crown witness, Wilson, was credible?

[38]        
The appellant argues that the evidence of Wilson, central to the Crown’s
case, was so inconsistent and so rife with difficulty that it could not
reasonably support the conclusions made to the standard required. In making
this argument, the appellant does not point to any particular finding of fact,
or group of facts, which are not reasonably supported by the evidence. Instead,
the proposition advanced by the appellant is more holistic in nature – there
are some witnesses whose evidence is so erratic and self-contradictory that it prevents
the trial judge from being able to properly engage in a fact finding exercise
and to make necessary findings. Wilson is said to be such a witness.

[39]        
The submissions made on this second issue are alternative to the issue
raised by the first ground of appeal. That is, they presuppose that the
findings of the trial judge were properly corroborated by the statements of the
co-accused and argue that notwithstanding this, those findings cannot be
supported.

[40]        
On any appeal the court must determine whether the trial judge could
reasonably have reached the conclusion that the appellant was guilty beyond a
reasonable doubt. The test on appeal is whether a judge or jury could have
convicted if properly instructed and acting reasonably: R. v. W.(R.), [1992]
2 S.C.R. 122 at 131, 74 C.C.C. (3d) 134.

[41]        
At the same time the court must show “great deference” to findings of
credibility made at trial, because of the finder of fact’s special position.
The original judge or jury has the significant advantage of seeing and hearing
the evidence.

[42]        
A difference in view between the trier of fact and an appellate court is
not a proper or sufficient basis to interfere with the judgment. In Fletcher
v. Manitoba Public Insurance Co., [1990] 3 S.C.R. 191 at 204, 74 D.L.R.
(4th) 636, the court said:

… appellate courts should only
interfere where the trial judge has made a “palpable and overriding error which
affected his assessment of the facts.” The very structure of our judicial
system requires this deference to the trier of fact. Substantial resources are
allocated to the process of adducing evidence at first instance and we entrust
the crucial task of sorting through and weighing that evidence to the person
best placed to accomplish it. As this Court and the House of Lords have
repeatedly emphasized, it is the trial Judge who is in the best position to
assess the credibility of testimony. An appellate court should not depart from
the trial judge’s conclusions concerning the evidence “merely on the result of
their own comparisons and criticisms of the witnesses” …

[43]        
Findings of fact made by a trial judge should always be approached with
deference by appellate courts. This is particularly so when those findings of
facts are based on the trial judge’s assessment of witness credibility. The
British Columbia Court of Appeal in R. v. A.A.M. (1996), 71 B.C.A.C. 258
at para. 50, 717 W.A.C. 258, said:

[50]      A trier of fact, when
considering the issue of credibility and distilling from the evidence the facts
of a particular case along the way to determine whether a case is or is not
proven in accordance with the applicable standard, must take into account not
only what a witness says, but also how the witness says it and against what
background as outlined by the other evidence. Credibility is based upon the
assessment the trier of fact makes of the witness as a truthful reliable
person, based on the trier of fact’s experience and common sense in such matters.
It is precisely that unique opportunity to observe a witness, which cannot be
replicated on appeal, which allows a trier of fact to determine whether to
accept or reject parts or all of a witness’ testimony. That is why appellate
courts show “a great deal of deference to findings of credibility made at
trial”… R. v. R.W., [1992] 2 S.C.R. 122 …74 C.C.C. (3d) 134.

[44]        
The standard of review on credibility matters therefore requires there
to be a palpable and overriding error in the facts, or an error in the
inference drawing process: R. v. Gaudaur, 2008 BCSC 981 at para. 29,
citing Housen v. Nikolaisen, 2002 SCC 33, [2002] 2 S.C.R. 235.

[45]        
In the present case, I do not accept that the trial judge erred in his assessment
of Wilson’s evidence. I have already referred to the fact that the trial judge
was acutely aware of the difficulties with Wilson’s evidence. Notwithstanding
those difficulties, the trial judge was satisfied that he could make the
necessary findings of fact in relation to Denholme.

[46]        
Furthermore, even though Wilson’s evidence was erratic and troublesome
the trial judge had substantial proper corroboration for his findings. The
following examples are illustrative of evidence, properly available to the
trial judge, which were consistent with the parts of Wilson’s evidence that the
trial judge accepted:

1.       The
cell phone histories showed that there were calls back and forth between
Wilson’s telephone and the phone seized from the appellant, thus corroborating
the planning of the meeting that Wilson said occurred.

2.       Wilson
said he had knives with him for protection and two knives were found on the
roof of the Shoppers Drug Mart, scene of the incident.

3.       Wilson said
he was in the trunk of Silva’s car. The trunk contained spots identified as
dried blood inside and on the edge of the truck. Wilson said there were large
speakers in that trunk. These are visible in the photos that are Exhibit 17.
The telephone call from the trunk was recorded.

4.       Wilson
said he called emergency twice. His cell phone history described by Cst.
Henning corroborated this as did the recordings of his calls.

5.       Wilson
said he threw his cell phone at the police officer at the Kitimat Hotel to get
his attention. The officer testified he saw a cell phone on the road and that
it was later identified by Wilson as his.

6.       Wilson
said there was a taxi behind him when he exited the trunk that he tried to wave
down. A taxi driver, Fred Martin, corroborated seeing a person at that location
who tried to flag him down.

7.       Wilson
testified his jacket was pulled off in the incident and he was only dressed in
a t-shirt and pants by the Kitimat Hotel. Wilson was only wearing a t-shirt and
pants when found by Cst. Henning and seen by Fred Martin. His jacket was found
near the hotel.

8.       Wilson
told Cst. Henning that Denholme and Finn may have gone into the Kitimat Hotel. Denholme
was found there very shortly thereafter.

9.       Wilson said he had been in the
trunk and gotten out when the car was still moving. The photos taken at the
hospital show injuries consistent with this evidence.

[47]        
Accordingly, I do not accept that the trial judge made any error, other
than as described in my discussion under the first ground of appeal, in his
assessment of Wilson’s evidence.

Section 686 of the Criminal
Code

[48]        
Section 686(1)(a)(ii) provides that a court:

(a)        may allow the appeal where it is of the opinion
that

(ii)    the judgment of the trial court should be set aside
on the ground of a wrong decision on a question of law

[49]        
Reliance on evidence that is inadmissible is an error of law. Similarly,
whether evidence is capable of being confirmatory is a question of law: R.
v. Smith, 2009 ABCA 230, 460 A.R. 288, citing R. v. B.(G.), [1990] 2
S.C.R. 57 at 71, 56 C.C.C. (3d) 181.

[50]        
In the present case, for the reasons I have already set out, I consider
that the use by the trial judge of the inadmissible statements of the various
co-accused, both in making findings of fact and in corroborating the evidence
of Wilson, was in error. The analysis of whether, or to what extent, an error of
law is prejudicial to an accused is a distinct consideration that falls within
s. 686(1)(b)(iii).

[51]        
Section 686(1)(b)(iii) provides that a court:

(b)        may dismiss the appeal where

(iii)       notwithstanding that the court is of the opinion
that on any ground mentioned in subparagraph (a)(ii) the appeal might be
decided in favour of the appellant, it is of the opinion that no substantial
wrong or miscarriage of justice has occurred

[52]        
Once an error in law has been found to have occurred at trial, the onus
is on the Crown to establish that the verdict would have been the same absent
the error: R. v. Haughton, [1994] 3 S.C.R. 516, 93 C.C.C. (3d) 99. In
this case, that onus has not been discharged.

[53]        
In Smith at para. 20, the Alberta Court of Appeal dealt with
the impermissible use of DNA evidence to confirm the complainant’s absence of
consent to a sexual encounter, and said:

[20]      We do not consider it
necessary to comment on the trial judge’s findings that Smith’s note and the
complainant’s emotional state, in this case, support the complainant’s evidence
that the sexual encounter with Smith was non-consensual. We do note that both
of these factors are not inherently supportive of non-consent. Nevertheless, on
the high standard of review regarding inferences to be drawn from evidence, we
will not interfere with the trial judge’s view of these two factors. However,
because in his concluding remarks the trial judge refers to all the
supporting evidence in deciding the complainant did not consent to the sexual
activity at issue, and given his error in relying on the DNA evidence as
confirmatory, it is impossible to know whether he would have reached the same
conclusion in the absence of its support, relying only on the note and the
complainants emotional state. We therefore decline to apply s. 686 of the Criminal
Code in the circumstances of this case.

[54]        
The foregoing comments are directly relevant to this case. It is
impossible to know how significant a component the inadmissible statements of
the appellant’s co-accused were in the trial judge’s findings of fact, and in
particular, his acceptance of Wilson’s evidence. Therefore, I cannot safely say
that the legal error had no impact on the verdict: R. v. Khan, 2001 SCC
86 at paras. 28-31, [2001] 3 S.C.R. 823; R. v. Van, 2009 SCC 22 at paras. 34-36,
[2009] 1 S.C.R. 716; R. v. Sarrazin, 2010 ONCA 577 at para. 68.

[55]        
I therefore allow the appeal and direct that a new trial take place.

“Voith J.”