What the majority did in HML was to extend and affirm the law from Pfennig v The Queen (1995) 182 CLR 461 to all similar fact cases. There are two aspects of Pfennig: admissibility and jury directions. In terms of admissibility, the Pfennig test or “no rational view test” fashions a very strict test for the admissibility of similar fact evidence. The trial judge is to apply the same test as a jury must apply in dealing with circumstantial evidence; the evidence will only be admissible if there is no reasonable view of the evidence that is consistent with the innocence of the accused. Once admitted, in a similar vein, jurors are to be instructed that they can only use the similar fact evidence if they accept it beyond a reasonable doubt.
Both these aspects of Pfennig are questionable and the fact that they are now being applied to all similar fact cases is regrettable.
The Problem
Pfennig was an identity case. The case against Pfennig was entirely circumstantial. A young boy had disappeared. Pfennig was put at the scene where the boy was last seen alive. Similar fact evidence was introduced to show that Pfennig was a homosexual paedophile who had abducted another young boy. Coincident reasoning was then applied; how likely would it be for another homosexual paedophile abductor to be in the immediate area?
The common law has long held that for identity cases a high degree of probative value and similarity between incidents is warranted before similar fact evidence is admitted. Pfennig is consistent with this accepted authority. However, to take the Pfennig test and apply it to all similar fact evidence is misguided. We see this in the HML case.
HML and two other associated cases heard as part of the appeal all involved sexual assaults and the admissibility of prior sexual acts committed by the accused on the complainants. These were not circumstantial cases. Identity was not at issue. Credibility was the critical issue in each. The similar fact evidence was relevant on a number of levels. It helped to explain the reactions of the complainants to the alleged attacks. It provided a “context” from which the jury could better understand the complainant’s evidence. It showed the nature of the relationship between the complainants and the accused. And it showed that the accused had a sexual interest in the complainants to such an extent that on other occasions they sexually assaulted the complainants.
The problem is that the “no rational explanation” test just does not work for this type of similar fact evidence. The test is at once too strict and too unfair. In a sexual assault case the fact that the accused sexually assaulted the complainant on other occasions is only consistent with guilt. There is no other innocent explanation; keeping in mind that we are to accept the testimony of the complainant. In other words, it is accepted that the other assaults did occur. There is no weighing of the probative value of the evidence versus its potential prejudice. The prior assaults meet the formula and are admissible.
In sexual assault cases, therefore, prior sexual violations of the complainant will go in as a matter of course — no matter how egregious or numerous. Whereas in other cases, where the inference may not be as definitive the evidence will be excluded even if its probative value is high and the risk of prejudice low.1 Simply put the Pfennig test is too blunt a legal instrument.
What is troubling is that trial judges are straying into the jury’s domain. It may make sense to have a no rational view of the evidence other than guilt as a matter of final adjudication; it is not appropriate as an interim evidentiary test of admissibility. Using the test the evidence is admissible to prove guilt because the only rational explanation for the evidence is guilt. This is circular reasoning at its worst.
The beyond a reasonable doubt instruction, continues the circular reasoning. This high threshold requirement regarding an individual piece of evidence is unusual in our law. Normally we allow jurors to accept or reject individual pieces of evidence and only place the beyond a reasonable doubt standard on the final decision.
The Aftermath
A number of courts, faced with the task of applying HML, have complained about its complexity.2 The Queensland Court of Appeal applied HML in R v Rae [2008] QCA 385 (4 December 2008). The Court’s decision underscores the difficulty in application of HML. Rae too was a sexual assault case. The Court of Appeal reviewed HML and concluded that the beyond a reasonable doubt threshold only applied if the prosecution was using the similar fact evidence directly to infer guilt. For example, if the relevancy of other acts of sexual assault is to show the sexual passion of the accused towards the complainant, from which we can infer that he acted on his passion on this occasion, then it is directly going to guilt and a beyond a reasonable doubt instruction is required. However, if the purpose of the other sexual incidents is going to prove motive or to provide a context to better understand the evidence, then no beyond a reasonable doubt instruction is necessary. The standard would vary according to the identified relevancy of the evidence. Surely this is too complex and too confusing a process. Not surprising the Supreme and District Court Benchbook recommends a beyond a reasonable doubt instruction in all cases.3
The Need for Legislation
Pfennig is a common law test. As it now stands only South Australia, the Northern Territory and Queensland are bound by it. The other states have legislated to override Pfennig.4 Queensland should do the same. The Queensland legislature has reacted in the past to reform portions of the law on similar fact and it needs to do so again.5
We need to bring the law in Queensland into line with the common law world. The Pfennig test is very much an aberration from the High Court. Whereas the prevailing view is that the admissibility of similar fact evidence should be determined by a rigorous weighing of the probative value of the evidence against its potential prejudice. The law already accepts that introducing other discreditable acts or other bad acts committed by the accused is presumptively inadmissible because of its unfair prejudice to the accused. The accused is apt to be convicted because of who he is or because of what he has done in the past rather than based on the evidence respecting the present charge. The balancing of the probative value of the evidence against its potential prejudice is the best way of dealing with similar fact evidence. It may not be perfect, and application will continue to be difficult, but it is a fairer and more practical approach then presently under Pfennig and HML.
Professor Lee Stuesser
Faculty of Law Bond University
Editorial Note: On 27 April 2009, Hearsay wrote to the Attorney-General, the Honourable Cameron Dick. A copy of this article was forwarded to him and he was invited to comment. As at 15 June 2009, no response has been received.
Footnotes
- Spigelman CJ provides an insightful critique of Pfennig in R v Ellis [2003] NSWCCA 319 (5 November 2003).
- See for example GBT v The State of Western Australia [2009] WASCA 19 at para 57; R v Sweeney [2008] SASC 300 at para 23; R v Sadler [2008] VSCA 198.
- See No 66.
- Perhaps the clearest and best piece of legislation is found in Western Australia Evidence Act 1906 section 31A. Other legislation overriding Pfennig is as follows: Victoria (Crimes Act 1958, s 398A), the federal courts and the Australian Capital Territory (Evidence Act 1995 (Cth), ss 97 and 98), New South Wales (Evidence Act 1995, ss 97 and 98), and Tasmania (Evidence Act 2001, ss 97 and 98).
- See sections 132A and 132B of the Evidence Act 1977. To discuss this article, visit the Hearsay Forum.
