The law and sexual offences against adults in Australia

ACSSA Issues No. 4 – June 2005

You are in an archived section of the AIFS website. Archived publications may be of interest for historical reasons. Because of their age, they may not reflect current research data or AIFS' current research methodologies.


There is no legal requirement for evidence that independently confirms the evidence of a single witness (corroborating evidence) in any Australian jurisdictions. An accused charged with any crime can be convicted of the offence on the evidence of a single witness, provided the jury accepts that his or her evidence proves the offence beyond reasonable doubt (Mack 1998). However, in the past, Australian jurisdictions required that juries should be warned about the dangers of convicting people accused of sexual offences on uncorroborated evidence (Sheehy 2002). A judge's failure to warn the jury about relying on uncorroborated evidence could result in a conviction being overturned on appeal, resulting in a retrial or an acquittal.

Victim-complainant witnesses in sexual offence trials have been, and to some extent continue to be, treated as potentially unreliable witnesses. The reasons judges have historically given for treating victim-complainants as potentially unreliable witnesses have included the ease of making unfounded allegations of sexual assault, the difficulty of refuting an accusation of sexual assault, jury sympathy for victim-complainants, and women's tendency to lie (and lie convincingly), especially where sex is concerned (Mack 1998). These "reasons" have long been discredited by extensive evidence that sexual offences have extremely low reporting rates, that there are extensive barriers to reporting, that cases are progressively and intensively filtered out prior to trial as they pass through the criminal justice system and that, consequently, the rate of convictions in sexual offences is very low (Naffine 1992; Mack 1998; Stubbs 2004).

Feminists have argued that warnings to the jury about the dangers of convicting an accused on uncorroborated evidence or requiring very careful scrutiny of uncorroborated evidence primarily function as barriers to reporting and prosecuting sexual offences, producing unwarranted acquittals and additional humiliation for victim-complainant witnesses.

Statutory reforms mean that there is now no requirement that a jury be warned that it is dangerous to convict on the uncorroborated evidence of the victim-complainant of a sexual offence (Mack 1998). The law of NSW, QLD, SA, TAS and WA states that there is no requirement for a judge to warn a jury that it is dangerous to convict an accused on uncorroborated evidence. NSW further states that, in cases where there is no corroborating evidence, judges are not required to direct the jury about the absence of corroboration.

However, during the process of reforming the corroboration rules relating to sexual offence trials in Australia, the High Court Australia handed down a critical decision.10 It simultaneously ruled that victim-complainants of sexual offences should no longer be regarded as a group of unreliable witnesses, and profoundly limited the effectiveness of legislative reforms abolishing the requirement for corroboration warnings in sexual offences. The court decided that, even in the face of legislative changes eliminating the requirement for a corroboration warning, judges retain a discretion to give a corroboration warning where it is called for (Sheehy 2002).11 In practice, this decision emphasised the necessity for trial judges either to warn, or to risk having their decisions overturned, sometimes even in cases where the defence had not requested a corroboration warning at trial (Heenan and McKelvie 1997; Mack 1998). More recently, the High Court has emphasised that a judge may be obliged to give a warning even if there is substantial corroborating evidence and the defence has not requested a warning. The High Court has also made it clear that a warning that the complainant's evidence should be ccarefully scrutinised may not be enough to discharge the judge's duty to warn. 12

In the ACT, NT and VIC, the law states that the judge must not warn the jury or suggest to them in any way that the law views victim-complainants in sexual offences as unreliable witnesses. However, courts retain the capacity to comment on evidence as appropriate to the circumstances in the interests of justice.

In practice, these provisions have allowed warnings that fall short of suggesting that victim-complainants of sexual offences in general constitute a class of unreliable witnesses. For example, corroboration warnings may direct the jury to scrutinise the victim-complainant's evidence with great care.

How does it work in practice?

Reforms to the corroboration warning have not been universally effective. Even after reforms abolishing any legal requirement for a corroboration warning, studies found continued references to a legal requirement to warn (Mack 1998). After the High Court stated definitively that sexual assault complainants were not to be seen as especially untrustworthy, judicial statements casting doubt on the reliability of victim-complainant witnesses continued (Mack 1998).

When the corroboration warning remained discretionary in VIC, case analysis showed that such warnings were given in circumstances where they were unnecessary, or where reasons given by the defence for requesting a corroboration warning were poor and unsupported by the evidence (Edwards and Heenan 1994). Evaluation of the application of corroboration warnings in NSW under a discretionary regime found that a warning that it was unsafe or dangerous to convict without corroborating evidence was given in 40 per cent of cases, despite the High Court having overruled warnings suggesting that sexual offence complainants should be regarded as a suspect class of witnesses.13 A less harsh warning suggesting that uncorroborated evidence should be evaluated or "scrutinised with great care" was given in 59 per cent of cases. Some cases received both types of corroboration warning. No corroboration warning was given in only 15 per cent of cases (NSW Department for Women 1996: 188-189). High rates of corroboration warnings were also found in SA and, to a lesser extent, WA, after reform (Senate Standing Committee on Legal and Constitutional Affairs 1994). As with sexual history reforms, "continued emphasis on judicial discretion has undermined the positive impact of the legislation" intended to eliminate unwarranted corroboration warnings in sexual offence cases (Mack 1998: 73).

Despite their demonstrated limitations, few jurisdictions have chosen to undertake further reform of corroboration rules. Proposals for further reform suggest legislative guidance about when corroboration warnings will be required, and propose mandatory instructions to the jury (Mack 1998). Mandatory jury instructions may be particularly desirable in view of the Victorian Law Reform Commission's finding that emphasis and degree of repetition make a significant difference to the way that directions on corroboration come across to the jury. The Victorian Law Reform Commission has proposed changes to the law to ensure that instructions are clear and concise, always address delay in prosecution appropriately and eliminate the harshest forms of the warning which state that it may be unsafe or dangerous to convict on uncorroborated evidence (VLRC 2004).


10 R v Longman (1989) 168 CLR 79.

11 R v Longman (1989) 168 CLR 79.

12 Doggett v R [2001] HCA 46.

13 R v Longman (1989) 168 CLR 79.