By Dr. Bob Uttl
Legal maxim, qui tacet consentire videtur
He who is silent, when he ought to have spoken, and was able to, is taken to agree.
On June 18, 2026, under cross-examination before the British Columbia Human Rights Tribunal, Dr. John Braxton Suffield was asked, plainly, whether he was aware of any other errors in his reports or testimony concerning Ms. T. He said no. The next morning, on June 19, 2026, Dr. Suffield volunteered that he had, in fact, known about another error for five weeks.
The Question, and the Answer
On June 18, 2026, Dr. John Braxton Suffield was asked directly:
QUESTION: Now, Dr. Suffield, during your preparation over the last year or even before, have you become aware of any other errors you have made in your [expert] reports or in various graphs which you prepared to interpret Ms. T’s scores?
DR. JOHN BRAXTON SUFFIELD: I don’t believe so… I don’t remember saying anything in my reports, my written submissions.
Cross-examination of Dr. John Braxton Suffield
Pressed on whether he was aware of any false testimony given to the Tribunal, Dr. Suffield carved out exactly one exception — a possible mix-up between findings he attributed to Dr. Mary Westcott versus Dr. MK — and then said:
QUESTION: Are you aware of any false testimony you gave to the tribunal?
DR. JOHN BRAXTON SUFFIELD: No, other than that would be my concern. I think I raised that some time ago. But, no.
Cross-examination of Dr. John Braxton Suffield
“No, other than that.” One named exception. Everything else: denied.
The Next Morning
The very next day, on June 19, 2026, before cross-examination resumed, Dr. Suffield asked to address the Tribunal:
DR. JOHN BRAXTON SUFFIELD: The other error, if I may interrupt, Dr. Uttl asked if I had made any other errors in my direct or in my written submissions… I did consider that possibility when, in my direct testimony… I referred to a spreadsheet that Dr. L himself and Dr. [David] Crockett… had prepared… What I did not know at the time is that that spreadsheet contained a significant artifactual error.
Dr. John Braxton Suffield
Dr. Suffield characterized it as an honest, good-faith reliance on colleagues’ work, and told the Tribunal it “didn’t affect my opinion ten years previous.”
Later that same day, under further questioning about the exhibit itself (Exhibit CB166), he was asked when he first became aware of the error:
QUESTION: Okay. And when did you become aware of this error?
DR. JOHN BRAXTON SUFFIELD: Many weeks ago, probably five or so weeks ago.
Cross-examination of Dr. John Braxton Suffield
Many weeks ago, five weeks before this testimony. Not overnight. Not a fresh discovery made “in preparation for my testimony this morning” — a phrase he used elsewhere in this same proceeding to describe a different error (a data point of 78 that he said should have read 81 in his Figure on page 9 of his December 30, 2011 Report).
Many weeks weeks of prior knowledge, sitting behind a flat “no” given under oath the day before.
What the Spreadsheet Actually Did
The spreadsheet in question was not a minor supplementary exhibit. Dr. John Braxton Suffield relied on its output — a “relative profile” (RP) calculation comparing Ms. T’s WAIS-IV scores (index scores of 100, 92, 92, 105) against 13 different clinical/disorder groups, including a severe intellectual disability group with a mean IQ of approximately 55 — to support his sworn testimony given in 2024:
And there were 500 of those people, the aggregate of all the different control groups. And of all the comparisons that could be made, Ms. T’s data least resembled that control group of normal people… She is wildly different in her level of performance and her pattern of performance from the normal subgroups.
Dr. John Braxton Suffield, Direct Testimony Under Oath
Dr. Suffield’s claim sharply, irreconcilably contrasts with Dr. Suffield’s own original 2011 WAIS-IV computerized report printout for Ms. T, which he admitted was accurate. The index scores of 100, 92, 92, and 105 that he entered into the Crockett’s spreadsheet were, every single one, quite obviously, labelled “Average” by the test publisher’s own software.
Both cannot be true. A profile of four average index scores cannot “least resemble” a normal control group out of comparisons against thirteen groups including one centered nearly three standard deviations below the average. Dr. John Braxton Suffield’s testimony under oath was obviously patently false. The average cannot be “most wildly” different from the average.
By June 2026, Dr. Suffield conceded the point outright: Ms. T was, and always had been, of average intelligence, and he had “never said otherwise.”
On cross-examination, Dr. Suffield admitted to one specific defect: blank cells in the spreadsheet — corresponding to six subtests Ms. T was never administered — were converted by Excel spreadsheet into scores of “0” wherever the formula referenced them. A subtest score of zero is not a low score; it is an impossible one, outside the subtest’s 1–19 range entirely. Because each blank fed into multiple group comparisons, this alone injected more than eighty impossible data points into the analysis. Asked about one visible instance of this on the stand, Dr. Suffield’s own description was: “embarrassingly so.”
That is the error Dr. Suffield has personally admitted to, under oath, at his cross-examination.
It is not the only one on the record. In September 2025 — following disclosure of the spreadsheet on April 1, 2025 —I have published a detailed analysis of the Crockett’s spreadsheet used by Dr. Suffield, identifying additional defects independent of the blank-cell problem: the Cattell’s rp formula used throughout the spreadsheet is one that assumes sten scores (M = 5.5, SD = 2), and applied without adjustment to WAIS-IV subtest scores (SD = 3) and index scores (SD = 15) produces contradictory, nonsensical similarity coefficients even where the underlying scores are unremarkable, average. That analysis also identified that Ms. T’s actual Letter-Number Sequencing subtest score of 10 — an average score — was entered by Dr. John Braxton Suffield as missing rather than as 10, and was therefore also converted to an impossible zero by the same blank-cell defect. And there were more errors in Dr. Crockett’s spreadsheet — see the detailed analysis.
Although Dr. John Braxton Suffield admitted, in passing, to making “one typographical error, on one score on one of those tests” without any further explanation as to what the error was and how serious it was, Dr. Suffield has not, as of this writing, notified the Tribunal of these two additional defects, one in the spreadsheet itself and one of his own making. Perhaps he continues to be unaware of the Cattell’s rp formula error because he did not perform these analyses himself. Instead, he merely parroted the Crockett’s spreadsheet analyses as his own expert testimony. The fatally flawed spreadsheet became the eminence grise behind Dr. Suffield’s outrageous claim about Ms. T’s intellect. Perhaps Dr. Suffield continues to be unaware of incorrectly entering Ms. T’s Letter-Number Sequencing score of 10 because he is truly incapable of correctly transferring simple numbers from a computerized report to a spreadsheet, and thus, continues to be unaware of his own incompetence. Perhaps Dr. Suffield failure to enter Ms. T’s Letter-Number Sequencing score into the spreadsheet was what he referred to as the “typographical error, on one score on one of those tests.”
The analysis of the toxic spreadsheet demonstrates that Dr. Suffield’s outrageous testimony — that Ms. T, a Canadian woman of average intelligence, was “wildly different” from matched controls — was a stunning fabrication.
Yet, the toxic spreadsheet errors do not involve subtle statistics. A subtest score cannot fall outside 1–19. A formula built for one type of standardized score cannot be applied to another without producing a heap of garbage. These are the kinds of foundational errors that an essential working knowledge of psychometrics would catch immediately.
Dr. Suffield did not catch the errors. He testified he relied on the spreadsheet in “good faith” — which, in reality, means he adopted it blindly, without independent review, and without any actual understanding of how the toxic spreadsheet arrived at non-sensical Cattell’s rps.
Most critically, Dr. Suffield has never disclosed to the Tribunal that he was merely parroting the Crockett’s spreadsheet, had not performed the calculations himself, and was unable or unwilling to verify their mathematical integrity until his late disclosure on June 19, 2026.
Conclusion: A Fatal Breach of Expert Duty
Under Canadian law, established by the Supreme Court of Canada in White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23 (CanLII), an expert witness owes a paramount, continuous duty to the court to provide independent, objective, and non-partisan assistance. They cannot act as advocates for the party paying their bills.
[46] I have already described the duty owed by an expert witness to the court: the expert must be fair, objective and non-partisan. As I see it, the appropriate threshold for admissibility flows from this duty. I agree with Prof. (now Justice of the Ontario Court of Justice) Paciocco that “the common law has come to accept . . . that expert witnesses have a duty to assist the court that overrides their obligation to the party calling them. If a witness is unable or unwilling to fulfill that duty, they do not qualify to perform the role of an expert and should be excluded”…
White Burgess Langlille Inman v. Abbott and Haliburton Co., 2015 SCC 23 (CanLII)
When an expert becomes aware of an error in their evidence, their legal obligation is immediate disclosure. They are strictly forbidden from sitting quietly, withholding the truth for “many weeks,” and waiting to see if cross-examination will expose the flaw. The law has been succinctly summarized by the Alberta Court of Appeal Justice Slatter:
[96] An expert witness has obligations to his or her client, and his or her professional organization. The expert witness also has obligations to the court. If the expert discovers an error in the report, or if the expert believes the report is being used for a purpose for which it is was never intended, or is unsuitable, the expert has an obligation to advise the court [emphasis added]: Stanton v. Callaghan, [2000] 1 Q.B. 75 at pp. 101-2, 107-8, 110; National Justice Compania Naviera S.A. v. Prudential Assurance Co. Ltd. (“The Ikarian Reefer”), [1993] 2 Lloyd’s L. R. 68 (Q.B.) at pg. 99; Myers v. Elman, [1940] A.C. 282 at pp. 293-4. The expert cannot simply sit back and run the risk that the court or other third parties may be misled by relying on the report [emphasis added]. The expert is also entitled to protect its reputation by withdrawing reports that turn out to be flawed. The obligation and ability to withdraw a report, change an opinion, or correct an error in the report does not depend on the consent of the client; the client also has an obligation to ensure that the court is not misled by evidence it has filed. As Otton, L.J. said in Stanton at pg. 108: “The expert must be able to resile fearlessly and with dignity.”
Justice Slatter, in Deloitte & Touche LLP v. Institute of Chartered Accountants of Alberta, 2008 ABCA 162
Dr. Suffield’s timeline is legally indefensible. For five weeks, he harbored the knowledge that his dramatic 2024 testimony was built on impossible spreadsheet artifacts. When explicitly given the opportunity under oath on June 18, 2026 to correct the record, Dr. Suffield chose denial. Only when the closing moments of the hearing made exposure inevitable did he execute a sua sponte retreat the following morning.
By admitting to a multi-week campaign of silence while under a continuous oath, Dr. Suffield did not just correct his prior false testimony. He demonstrated a fundamental unwillingness or inability to fulfill his duty of candor to the British Columbia Human Rights Tribunal.