Andrew L. Urban
A combination of bad law and poor judicial decisions results in wrongful convictions. Frank Valentine was found guilty by the judge at his trial. The judge’s decision and reasons raised over 500 points of objection at appeal but the three appeal judges shrugged it all off. How can the accused (and justice) be protected from such unconstrained judicial errors?
A Griffith University study found erroneous judicial instructions were contributing factors in 32% of wrongful convictions. In the case of Frank Valentine, the trial and appeal judges gave themselves grossly erroneous instructions. Their cumulative verdicts produced what in my upcoming book I set out to show are wrongful convictions of the historical sexual abuse of minors of which Frank Valentine was charged. I wouldn’t write the book, Frank Valentine and the Abuse of Lady Justice, if I had any doubt that Valentine was wrongfully convicted of the historical crimes allegedly committed in his 30s with which he was charged and tried when he was 78. I would not defend anyone who I believed had perpetrated such violent and hideous crimes on teenage girls whether yesterday or almost half a century ago.
My certainty is based on the absence of any primary evidence, leaving only the six claimants’ 21 extraordinary uncorroborated allegations – but also on the gross failures of the legal system, coupled with multiple judicial errors.
Bad law
In criminal cases, the prosecution must prove guilt beyond a reasonable doubt, and uncorroborated testimony from a single complainant is sufficient for conviction if deemed credible and reliable by the jury or judge. This applies equally to historical child and adult sexual abuse.
Convictions based solely on uncorroborated testimony are common in historical cases, provided the evidence is consistent and the complainant is credible. This is bad law, however well meaning, relying on the judge/jury acting rationally. That cannot be assumed.
Longman Warning (Longman v The Queen [1989] HCA 60): this is mandatory in cases with significant delay (common in historical abuse). Judges warn juries about the “danger” of convicting on uncorroborated evidence due to potential memory unreliability or lost defence opportunities. Applies to both child and adult cases. Nicole Noman SC DCJ doesn’t appear to have adequately warned herself.
Poor judgement
The litany of errors by the trial judge, Nicole Noman, as outlined in Valentine’s appeal against the convictions, are cause for deep concern. It is also concerning that judges hearing the appeal, Basten AJA, Button J and Wilson J, failed to accept the many errors.
Indeed, the appeals court seems to have made a virtue of the errors at trial when it took the view that “in assessing the reasonableness of verdicts, the Court is entitled to consider the cumulative effect of all the evidence, including tendency evidence, to support a conclusion beyond reasonable doubt. The Court must be entitled to give as much weight to the admitted evidence as it considers appropriate, and it is open to the Court to accept tendency evidence on the basis that it rebuts the natural propensity to dismiss complaints as inherently implausible.” The author argues that one inherently implausible claim cannot be relied upon as tendency evidence to support another inherently implausible claim.
“Establishing propensity for criminality” is the basis of the tendency provisions which are often used in Australian courts, especially in sexual abuse cases. One senior counsel says: “Tendency and coincidence is a way to convict someone in a case that does not stand the test of beyond reasonable doubt. It puts aside the relevance of all evidence.” In Valentine’s case, it fills in the void left by the absence of direct evidence.
Some points in the appeal document by Valentine’s legal team:
-The trial judge erred in making critical findings of fact which were not open in reasoning towards guilt.
-The guilty verdicts are unreasonable including by their inconsistency with the acquittals in respect of Counts 13 – 19.
-The trial judge erred in failing to have regard to, or acknowledge, what was uncontradicted in the evidence giving rise to a reasonable doubt.
-It is contended the compounding improbabilities caused by the unchallenged or competing evidence at trial required of the judge acting rationally to have entertained a doubt in respect of the allegations of each complainant.
-It is to be noted that in the Reasons the trial judge observed that she was not to be inappropriately influenced by palpable distress, yet the Reasons contain no analysis of why it was appropriate to be influenced by DW’s “palpable distress”.
-The trial judge evidently concluded that “she could tell truth from falsehood accurately on the basis of such appearances” including by the “manner” of a witness’ appearing to do their best. (See below)
-Mrs Valentine’s evidence was entirely rejected on critical issues because it was unpersuasive yet when capable of being set against objective evidence it was wholly persuasive. For example, Mrs Valentine was captured on intercept material referring to a camp at Daruk, at which her car was stolen, which measured against the documentary was clearly correct. That camp, the evidence established, was March 1974. She referred to what must have been DW in intercept material as a lovely boy, entirely consistent with her retaining a photo of him, yet her Honour found as much irreconcilable with the defence case and moreover in a way persuasive of guilt.
In conclusion the trial judge said at [394]:
-“I consider that the accused was an unimpressive and unreliable witness. His lack of reliability on key issues causes me to reject his denials of offending. I, therefore, disregard his evidence and am required to focus on the other evidence.” (Emphasis added)
The “key issues” which underpinned her Honour rejecting the appellant’s evidence in its entirety were not identified at this part of the Reasons [at 394].
-It is critical to observe that at no point in the Reasons did her Honour refer to an aspect of the appellant’s evidence which, measured against the available objective evidence, was dishonest.
-It is contended it was an error on the part of the trial judge to jettison the entirety of the appellant’s evidence for the Reasons identified and in so doing the trial judge has palpably misused her advantage failing to, “as far as possible, rely on objectively established facts.” The problems in the Reasons “underscore the highly subjective nature of demeanour-based judgments”
Judges can’t tell
Judges are among the last people you’d ask to accurately discern accuracy or truth telling in witnesses. No single study definitively quantifies judges’ lie detection accuracy across all contexts, but the consensus from available research (e.g., Ekman & O’Sullivan, Leach et al., Bond & Deaulo) indicates judges perform poorly, with accuracy rates near chance levels (50-55%).
Their reliance on traditional cues like demeanour and consistency is flawed, and courtroom constraints (e.g., positioning, note-taking) further limit their ability to observe behavioural indicators. Studies like Leach et al. (2016) suggest that reducing visual information might paradoxically improve accuracy by focusing attention on verbal content, but these findings are debated due to methodological issues. For practical applications, judges could benefit from training in evidence-based techniques, such as those emphasizing verifiable details or cognitive load.
Unlike some professionals (e.g., Secret Service), judges typically lack training in evidence-based deception detection techniques, such as analysing verbal content or micro-expressions.
For example, in the Ekman & O’Sullivan (1991) – “Who Can Catch a Liar?” study, published in American Psychologist, evaluated 509 individuals, including judges, law enforcement personnel (e.g., Secret Service, CIA, FBI), police, psychiatrists, and college students, to assess their ability to detect deception. Participants watched 10 videos of people describing feelings while watching a scene, with half lying about their emotions.
Judges performed poorly, with accuracy rates not significantly better than chance (around 50%). The Secret Service agents were the only group to exceed chance levels consistently.
The Denault et al (2019) study found that judges often hold misconceptions about deception cues, such as equating nervousness with lying, which can lead to false positives (judging truthful witnesses as deceptive) or false negatives (missing lies).
The study emphasized that cross-examination, while considered a powerful tool for uncovering truth, lacks scientific validation for its effectiveness in lie detection. This finding is especially relevant in the Frank Valentine trial.
Adding to the unreliability of judges (at trial and at appeal) in this regard, and despite the usual judicial protestations of not being influenced by media, judges’ minds would have been as poisoned as that of the jury pool by the extensive defamatory material published about Frank Valentine in the years leading up to the trial.
In conclusion, it has to be said that an accused given the option of a trial by jury or a judge alone …should take the jury option.