My purpose in writing this article is to revisit the most important consideration a judge makes when evaluating the reliability and credibility of witnesses, prompted by a recent decision of Justice Robert Centa in McPherson v. Global Growth Assets Inc., 2025 ONSC 5226 (“McPherson“). While this decision and others referred to below are not particularly novel, they serve as a useful reminder to trial counsel. These cases are of assistance in persuading the court to prefer your witnesses’ testimony and, conversely, to cast doubt on those called by opposing parties.
It has often been said that the trial judge is in the best position to assess the credibility of a witness as he or she, as opposed to appeal judges, is able to see the witnesses and assess their demeanour as they testify. It follows that considerable deference is given to the trial judge’s findings of fact, as noted by Justice McDermot in Patton-Casse v. Casse, 2011 ONSC 4424:[1]
[…] Considerable deference must be given to the trier of fact as he or she is the person who heard the evidence and has experienced the parties and their witnesses firsthand.
A trier of fact is therefore presumed to be in a better position to make findings of fact than is an appellate court. The appellate court does not hear the evidence, see the demeanor of the witnesses or hear the cross-examination of the witnesses […].
In McPherson, Justice Centa cautioned against overreliance on demeanour, emphasizing that confidence and presentation are not always proxies for reliability. He wrote:[2]
Witnesses can sincerely believe their evidence is true, but that does not mean that what they are saying is reliable. Memory is fallible and becomes increasingly frail over time. Even an apparently convincing, confident, and credible witness may not be an accurate or reliable reporter. There is significant risk in placing too much emphasis on demeanour or the confidence with which a witness speaks where there are contradictions and inconsistencies inherent in their evidence or where that testimony is inconsistent with contemporaneous records.
This caution is especially relevant in commercial litigation, where witnesses such as executives or directors may appear confident and articulate but those accounts must still be tested against contemporaneous documents and probabilities.
However advantageous it might be to see a witness live, caution must be exercised by the court so as not to give an advantage to a witness who may be a better actor than others or who may have other skills that give an air of confidence in their recitation of events than other witnesses. There is also the possibility that a witness may believe what he or she is saying, even if his or her recollection may be faulty.
For this reason, other factors come into play as the court considers the reliability and credibility of each witness. Justice Centa explained in McPherson that:[3]
It is important to recall that reliability and credibility are different. Credibility has to do with the honesty, sincerity, or veracity of a witness. Reliability describes the other factors that can influence the accuracy of testimony, such as the witness’s ability to observe, recall, and recount events in issue.
Witnesses can sincerely believe their evidence is true, but that does not mean that what they are saying is reliable. Memory is fallible and becomes increasingly frail over time. Even an apparently convincing, confident, and credible witness may not be an accurate or reliable reporter […].
Justice Centa’s comments should guide both how counsel prepares their own witnesses and how counsel challenges opposing witnesses. These comments underscore that sincerity is not the same as accuracy.
One of the leading decisions on assessing credibility is Faryna v. Chorny, 1951 CanLII 252 (BC CA). The court in Faryna explained that:[4]
[t]he credibility of interested witnesses, particularly in cases of conflict of evidence, cannot be gauged solely by the test of whether the personal demeanour of the particular witness carried conviction of the truth. The test must reasonably subject his story to an examination of its consistency with the probabilities that surround the currently existing conditions. In short, the real test of the truth of the story of a witness in such a case must be its harmony with the preponderance of the probabilities which a practical and informed person would readily recognize as reasonable in that place and in those conditions.
The McPherson decision builds upon this principle. The court there preferred the plaintiff’s version of events in part because it was consistent with the contemporaneous documents, internally coherent, and aligned with the surrounding probabilities, and in part because certain defence evidence “was not credible on many important points” and appeared to be shaped by hindsight rather than contemporaneous conduct.
Justice Centa went on to summarize the framework for evaluating testimonial evidence, writing:[5]
I may accept some, none, or all of the evidence of a witness. Taking into account my assessment of reliability and credibility, I will assess the evidence before me according to many factors, including the following:
a. if the evidence makes sense by being logical or plausible;
b. if there are inconsistencies or weaknesses in the evidence of the witness, such as internal inconsistencies, prior inconsistent statements, or inconsistencies with the evidence of other witnesses;
c. if there is independent or documentary evidence to confirm or contradict the witness’s evidence, or a lack of such evidence;
d. the witness’s demeanour, including their sincerity and use of language, although this must be considered with caution; and
e. if the witness, particularly one that is a party in a case, may have a motive to fabricate.
Of the five factors mentioned, only one mentions the demeanour of the witness, and even that must be considered with caution.
Justice Newbould in Springer v. Aird & Berlis LLP, 2009 CanLII 15661 drew upon similar principles when reviewing the credibility findings.[6] He cited O’Halloran J.A. in R. v. Pressley, [1948] B.C.J. No. 63, 94Â C.C.C. 29 (C.A.):
The judge is not given a divine insight into the hearts and minds of the witnesses appearing before him. Justice does not descend automatically upon the best actor in the witness-box. The most satisfactory judicial test of truth lies in its harmony or lack of harmony with the preponderance of probabilities disclosed by the facts and circumstances in the conditions of the particular case.
Justice Newbould also referred to the observations of Farley J. in Bank of America Canada v. Mutual Trust Co., [1998] O.J. No 1525, 18 R.P.R. (3d) 213 (Gen. Div.):[7]
Frequently in cases judges will be called upon to make findings concerning credibility of witnesses. This usually is a most difficult task absent the most blatant of lying which is tripped up by confession, by self-contradictory evidence, by directly opposite material developed at the relevant time period or by evidence of an extremely reliable nature from third parties. One is always cognizant that people’s perceptions of the same event can sincerely differ, that memories fade with time, that witnesses may be innocently confused over minor (and even major) matters as well as the aspect of rationalization, a very human and understandable imperfection. A point that a witness may not be sure of initially becomes eventually a point that the witness is certain about because it fits the theory of his side. Rationalization will also affect some person’s views so that a certainty that a fact was “A” evolves into a confirmation that that fact was “not A”.
Justice Newbould went on to quote Justice Farley’s decision in Olympic Wholesale Co. v. 1084715 Ontario Ltd. (c.o.b. Lady Lin Foods), [1997] O.J. No. 5482 (Gen. Div.):[8]
[…] The evidence and the way it is given should be taken in context and in a balanced way. No one should expect perfection in testimony and it is often said that evidence which is too consistent may be a sign of it being artificially constructed. I also recognize that there can be an inadvertent rationalization of memory to fit what is afterwards said that must have happened as opposed to actually remembering what did happen. This usually increases over time…
Justice Newbould summarized these points by observing:[9]
Farley J. used the word “rationalization”. I take his comments to refer to what is often said to be “reconstruction” of evidence. Reconstruction can be either inadvertent or advertent. In either case, when it occurs, it is something that the trier of fact must consider in weighing evidence.
While the demeanour of the witnesses is a factor in assessing credibility and reliability, the harmony of their testimony with other facts and circumstances is what the court will look to and where counsel’s focus should be throughout, from witness preparation through closing submissions.
For commercial litigators, McPherson serves as a reminder that witness credibility disputes often determine outcomes in complex governance and regulatory cases. The decision reinforces that the most persuasive testimony is that which aligns with contemporaneous documents and objective probabilities, not the witness’s confidence on the stand. Counsel should therefore build examinations and submissions around logic, documentary corroboration, and consistency, rather than demeanour.
The author wishes to thank librarians Nairne Holtz and Harneet Kukreja, and colleague Alfred Pepushaj for their assistance in the preparation of this article.
The information and comments herein are for the general information of the reader and are not intended as advice or opinion to be relied upon in relation to any particular circumstances. For particular application of the law to specific situations, the reader should seek professional advice.
[1] Patton-Casse v. Casse, 2011 ONSC 4424 at paras 44, 45, aff’d 2012 ONCA 709.
[2] McPherson, ibid, at para 30.
[3] McPherson, ibid, at para 29.
[4] McPherson, ibid, at para 31, citing Faryna v. Chorny, 1951 CanLII 252 (BC CA).
[5] McPherson, ibid, at para 32.
[6] Springer v. Aird & Berlis LLP, 2009 CanLII 15661 (“Springer“) at para 14.
[7] Springer, ibid, at para 15.
