|The Brides in the Baths|
REX v. SMITH
11 Cr. App. R. 229, 84 L.J.K.B. 2153 (1915)
Appeal on points of law against a conviction for murder before Scrutton, J. at the Central Criminal Court. The appellant was indicted for the murder of Bessie Munday, who was discovered dead in her bath at Herne Bay on the 12th July 1912. The appellant had gone through a ceremony of marriage with the deceased, his own wife being then alive. At the trial of the appellant on the charge of murder evidence was given that subsequent to the death of Bessie Munday two other women named Alice Burnham and Margaret Elizabeth Lofty had both died in their baths under nearly the same circumstances as those which occurred in the case of Bessie Munday. In both of these subsequent cases the appellant had gone through a ceremony of marriage. The appellant was convicted at the Central Criminal Court of the murder of Bessie Munday and sentenced to death. The contention on behalf of the appellant is that the evidence was not admissible on examination in chief. It was admitted as evidence of a system of murder.
LORD READING, C.J. The principles of law governing the admission of evidence of this nature have been often under the consideration of this court and depend chiefly on the statement of the law in the case of Makin v. Attorney-General for New South Wales (sup.), where Lord Herschell says:
"It is undoubtedly not competent for the prosecution to adduce evidence tending to show that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried. On the other hand, the mere fact that the evidence adduced tends to show the commission of other crimes does not render it inadmissible if it be relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in his indictment were designed or accidental, or to rebut a defence which would otherwise be open to the accused."
In the present case the prosecution tendered evidence relating to the other two women, and it was admitted by the judge as tending to show that the act charged was committed with design.
The second point taken is that even assuming that evidence of the other two women was admissible, the prosecution should not have been allowed to give evidence beyond the fact that the two women were found dead in their baths. Obviously for the reasons given in dealing with the first point, it would not have been of any assistance to cut short the evidence in this way. We think that the prosecution were entitled to give, and the judge rightly admitted, evidence of the circumstances relating to the deaths of the two women.
Evidence of other similar happenings that might be only marginally probative if considered in isolation can become highly probative when considered in conjunction. The logic that applies is called the "doctrine of chances." The Brides in the Baths is the classic case showing lack of accident under the doctrine of chances. Considering each episode in isolation, it is hard to say whether the defendant killed his bride or whether she slipped and fell. But when the same apparent accident occurs repeatedly, the likelihood is that it is being caused to happen.
Professor Imwinkelried, The Use of Evidence of an Accused's Uncharged Misconduct to Prove Mens Rea, 51 Ohio St. L.J. 575, 586-588 (1990), explains how the doctrine of chances supports admissibility of defendant's other crimes without violating the propensity rule:
United States v. Woods is the paradigmatic case. In Woods, the accused stood trial for infanticide. The victim had died of cyanosis. The accused claimed that the suffocation was accidental. To rebut the accused's claim, the prosecutor offered evidence that over a twenty-five year period, children in the accused's custody had experienced twenty cyanotic episodes. The defense objected to the admission of the testimony on the ground that the testimony amounted to impermissible evidence of the accused's bad character. However, the prosecution rejoined that the testimony was relevant on a noncharacter theory, that is, the doctrine of chances....
... Under both the doctrine [of chances] and the character theory [of the propensity rule], the trier of fact begins at the same starting point, the evidence of the accused's uncharged crimes. However, when the trier engages in character reasoning, the initial decision facing the trier is whether to infer from the evidence that the accused has a personal bad character. In contrast, under the doctrine of chances, the trier need not focus on the accused's subjective character. Under the doctrine of chances, the initial decision facing the trier is whether the uncharged incidents are so numerous that it is objectively improbable that so many accidents would befall the accused. The decision is akin to the determination the trier must make in a tort case when the plaintiff relies on res ipsa loquitur. In the tort setting, the trier must decide whether objectively the most likely cause of the plaintiff's injury is the defendant's negligent act. In the present setting, the trier must determine whether the more likely cause of the victim's injury is the act of another human being.
Assume arguendo that statistics compiled by the United States Public Health Service indicate that during a twenty-five year period, only two percent of American children experienced an accidental cyanotic episode. Contrast that figure with the incidence of cyanotic episodes experienced by the children in Ms. Woods' custody. Suppose, for example, that during the same twenty-five year period, twenty percent of those children had cyanotic episodes. The frequency of the episodes among those children far exceeds the national average for such episodes. The episodes are so recurrent among those children that it is objectively implausible to assume that all those episodes were accidental. Either one or some of those episodes were caused by human intervention, or Ms. Woods is one of the most unlucky people alive....
... [T]he doctrine is distinguishable from a character reasoning theory in terms of the pertinent policies. The probative dangers posed by the doctrine differ to a marked degree from the risks raised by a character theory.
One risk raised by a character theory is that at least at a subconscious level, the jury will be tempted to punish the accused for uncharged misdeeds. That risk is acute under a character theory because the theory forces the jury to concentrate on the accused's personal character or disposition. The jurors must consciously address the question of the type of person the accused is. There is no need for the jurors to grapple with that question under the doctrine of chances. There is an undeniable possibility that on their own motion, the jurors may advert to the question. However, unlike a character theory, the doctrine of chances does not compel the jurors to focus on the accused's subjective disposition. Consequently, the nature of the initial inferential step under the doctrine significantly reduces the risk of a decision on an improper basis.
The second probative danger raised by a character theory is that the jury will overvalue the probative worth of the item of evidence. Although general character has only slight or small relevancy to the issue of the accused's conduct on a specific occasion, we fear that the jurors will treat character as a reliable predictor of conduct. There is less risk of overestimation of probative value under the doctrine of chances. The doctrine invites the trier to compare the accused's experience with statistical data or the trier's knowledge of everyday, human experience. We commonly accept the trier's knowledge of "the ways of the world" as a trustworthy basis for legal reasoning. That knowledge is one of the bases for the res ipsa loquitur doctrine; and the jury instructions in many jurisdictions specifically encourage jurors to employ that knowledge as a basis for resolving factual disputes.
Since the theory of relevance [under the doctrine of chances] is distinguishable from the forbidden theory ..., prosecutors may properly rely on the doctrine of chances as a noncharacter theory for satisfying Rule 404(b). However, the courts should not admit uncharged misconduct evidence as a matter of course whenever the prosecutor asserts that the evidence is relevant under the doctrine of chances to prove the actus reus. Rather than accepting the prosecutor's argument as ipse dixit, the courts should carefully evaluate the evidence to ensure that the prosecutor has established the factual predicate for invoking the doctrine.
What is the "factual predicate for invoking the doctrine"?