Barney v R [2023] NSWCCA 85

Barney v R [2023] NSWCCA 85 · Read the judgment on AustLII

Catchwords: Assessment of witnesses – credibility versus reliability

In the course of considering this appeal, the Court of Criminal Appeal has discussed the concepts of ‘credibility’ and ‘reliability’ in a way that is likely to be useful for practitioners appearing in hearings and appeals at all levels.

The applicant was found guilty at trial of 12 sex offences relating to his stepdaughter. He was acquitted of a number of other counts. On appeal, he raised that the guilty verdicts were unreasonable, having regard to the asserted inconsistency between the guilty counts and the not guilty counts. This argument was levelled entirely at asserted factual inconsistency; that is, it was not suggested that the counts were legally or technically inconsistent, merely that there was not logical, reasonable basis to distinguish between them. Kirk JA (with whom N Adams and McNaughton JJ agreed) commenced by observing that it is difficult to succeed in an appeal asserting inconsistent verdicts (at [3]).

However, the more widely-applicable aspect of the appeal was the analysis of the differences between reliability and credibility (starting at [15]).

Credibility ordinarily involves notions of honesty and truthfulness in giving evidence, whereas reliability refers to a witness’ ability to recall events accurately, regardless of whether they are speaking honestly (at [17]).

Where a witness is found not to be credible (in the sense of not truthful), this will often taint the entirety of their evidence. However, it may not always taint all of their evidence. For example, they may have had a motive not to tell the truth on one particular issue. A case in which a witness exaggerated only part of their evidence to support their larger story might be very different to a case where there has been outright fabrication. In the example of fabrication, that might be so significant as to undermine the credibility of the whole of the witness’ evidence (at [19]).

In contrast, it is common for a witness to be reliable on some issues but less reliable on others, consistently with the fallibility of memory (at [20]). Just because a witness gives evidence which is undermined in some detail by a prior account they have given, it does not automatically follow that they are unreliable on all issues (at [21]). On the other hand, it is possible for a witness to be unreliable in all respects, for example where it is established that they typically have a bad memory, or where they have a neurological condition (at [22]).

In general, an adverse view about the credibility of a witness is likely to affect have more far-reaching consequences in deciding whether or not a witness can be accepted than adverse findings about unreliability in one respect or another. If a witness is prepared to be less than honest in a significant way, despite their serious obligations to be so, then it might be hard to decide what of their evidence can still be accepted; it may be necessary to set aside their evidence except to the extent that it is corroborated, unless that part of their evidence can be quarantined or discounted. In contrast, adverse findings in relation to reliability are more likely to allow some delineation, for example based on how long ago an event occurred, or how significant they would have been to the person (at [23]).

In the circumstances of this case (and consistently with the high bar set by an appeal asserting factual inconsistency of verdicts), and given that the different verdicts could not only be explained by a general doubt that the jury had about the complainant’s credibility or reliability, the appeal was dismissed.