Thursday, October 14, 2004

Reasoning with similar fact evidence

The recent Canadian case, R v Perrier 2004 SCC 56 (30 September 2004), reminds us about the controversy concerning whether evidence about charges on which the jury acquits the accused can be used as similar fact evidence against him on other charges considered at the same trial: see Mahoney, "Basing Convictions on Acquittals: Conjuring by the Court of Appeal" in Robertson (ed) Essays on Criminal Law: A Tribute to Professor Gerald Orchard (2004, Brookers, Wellington). Mahoney argues that a juror who is unsure that the accused committed the other offences, and who therefore would find him not guilty of them, cannot then use the evidence of those when deciding whether the accused is guilty of the offence in respect of which the prosecution says the evidence is relevant on a similar fact basis.

The New Zealand Court of Appeal has held the contrary in R v Holtz [2002] NZCA] 323, [2003] 1 NZLR 667; (2002) 20 CRNZ 14. Care is needed, as the Court pointed out in para 47:

"The care with which evidence of similar acts is scrutinised is justified because of the prejudice that inevitably arises from the risk of guilt being improperly inferred from mere propensity or disposition evidenced by previous bad conduct. But, where the evidence is truly probative and cogent, admission is appropriate so long as the circumstances are such that, while allowing the probative value of the evidence to be availed of, the risk of improper use can be avoided by appropriate directions to the jury."

Essentially the differences in these points of view is attributable to the differing standards of proof for the admission of evidence on the one hand, and the proof of guilt on the other.
In Holtz, at para 38, the Court said:

" … Where [similar fact] evidence is the only evidence or substantially the only evidence available from which the jury is asked to infer that the accused was the person who committed the offence charged, since the offence must be proved beyond reasonable doubt, the accused’s responsibility for the pattern of events necessarily must be found established beyond reasonable doubt before a verdict of guilty is justified. But that is because the offence charged must be proved beyond reasonable doubt. It is not because the similar fact evidence on which a verdict of guilty is based must be established beyond reasonable doubt before it can be used. By contrast it may be that in a particular case the evidence of pattern may be just another strand of circumstantial evidence pointing toward the accused. In that event, to require proof beyond reasonable doubt [of the similar facts], would be quite inconsistent with the proper approach to circumstantial evidence. Accordingly we do not accept as of universal application that, where identity is in issue, similar fact evidence cannot be used unless it is found to the standard of beyond reasonable doubt that the accused was responsible for the past conduct and that the same person was responsible for the offence charged…."

Mahoney agrees (p 41) that similar fact evidence is a form of circumstantial evidence, but suggests that, on the question of admissibility, similar fact evidence needs special treatment because its admission is exceptional. That said, however, Mahoney focuses his argument not on the decision as to admissibility, but rather on the way the jury should use similar fact evidence. He accepts (p 43) that jurors may have initial doubts about circumstantial evidence, and still use it in coming to their decision; but, he continues, if they find the accused guilty of the main offence, that means they must, at that point, have no doubts about the similar fact evidence:

"It is impossible to declare the accused guilty beyond reasonable doubt if doubts about the circumstantial evidence remain at the end of the jury’s deliberations. Doubts about the evidence upon which a verdict is founded amount to doubts about the verdict."

The Supreme Court of Canada mentioned the topic obliquely in Perrier. As to the question of admissibility, the Court said:

"[18] … The point is that we must be cautious when using propensity evidence in the context of identity. We want to be sure, on a balance of probabilities, that the same person committed the acts in question such that we can safely say it is not a coincidence nor a case of mistaken identity."


"[23] Once the trial judge has determined that the crime charged and the similar act were likely committed by the same person, the judge must then consider whether there is evidence linking the accused to the similar act. …
If the similar facts relied upon were not in fact the acts of the accused, then they have no probative value.
"[24] The threshold is not particularly high. The trial judge must determine whether there is "some evidence" linking the accused to the similar acts. However, evidence of mere opportunity or possibility is not sufficient."

All well and good, as far as admissibility is concerned. How, then, should the jury reason if the evidence is ruled admissible? The SCC followed the English approach:

"[29] … ‘The issue for the jury, once they are satisfied that it is the same gang who have committed both groups of offences, will be whether the prosecution have established on all the admissible evidence that the particular defendant was a member of that gang and, where the prosecution say that that defendant was a member of the gang on both occasions, whether the totality of the evidence has established that beyond reasonable doubt.’ " (a quotation from R. v. Brown, [1996] E.W.J. No. 2403 (QL) (C.A.))

In Perrier it was held that the evidence should not have been admitted, and the Court did not need to examine in more detail how the jury might otherwise have properly used it. There does, however, appear to be consistency with the Mahoney approach.

It is not always necessary to go as far as Mahoney and require juries to convict on the similar fact charges if they also convict on the main charge. As was pointed out in Holtz, things that are, on the balance of probabilities, true, can combine with other evidence to establish something else beyond reasonable doubt. A jury may well be sure of the main charge and acquit on the others, as they did in Holtz. See also the entries under the heading Bayes' Theorem in the Index to this site.

[Update: see the discussion of R v Mitchell (Northern Ireland) on 21 October 2016.]

No comments: