
A recent decision of an Ontario Commercial List judge disqualifying an arbitrator from sitting in a private arbitration on the grounds of bias is getting significant attention. Brendan Bowles, Isa Dookie, Ben Sasges and Mahsa Gardaneh review Atkinsréalis and contrast it with a previous Superior Court decision that reached the opposite conclusion on similar facts. The authors also review a preliminary decision from the same matter which stands as the first reported decision arising under section 13(7) of the Arbitration Act dealing with the court’s ability to stay an arbitration while an application to remove an arbitrator is pending.
Paul Ivanoff and Michael Matwichuk review an interesting attempt at appealing an arbitrator’s decision in the face of a “no appeals” clause in the arbitration agreement. The court enforced this clause and dismissed the appeal, notwithstanding that the court would have agreed with the merits of the appeal.
Ryan Krushelnitzky and Paul Kolida consider the recent Supreme Court of Canada decision in SS&C Technologies, which, while not a construction law case, provides helpful guidance on the doctrine of spoliation, which is helpful in the context of construction disputes where documentary and physical evidence is abundant and preservation of evidence should be kept front-of-mind by practitioners and clients alike.
We conclude with a case comment from Corbin Devlin and Luke Van Bostelen reviewing a British Columbia Supreme Court decision dismissing a claim for damages as being barred by an exclusion clause. Interestingly, the exclusionary language clause was found in the warranty provision of the contract. The authors rightly warn that although such clauses are ostensibly for the owner’s protection, drafting matters and in some cases warranty clauses can limit, not expand a contractor’s liability.
The Legal Update Committee welcomes your submissions of relevant legislative or case updates to Catriona.Otto@RoseLLP.com, or brendanbowles@glaholt.com.
In Atkinsréalis Construction (Pacific) Inc. v. GIP Construction Ontario East Ltd., 2026 ONSC 5666 (“Atkinsréalis”), the Ontario Superior Court of Justice (Commercial List) considered the permissible scope of an arbitrator’s questioning of witnesses and the circumstances in which an arbitrator’s conduct may give rise to a reasonable apprehension of bias, warranting their removal from an arbitral proceeding.
Atkinsréalis involved similar facts, and the same arbitrator, as Dufferin v Morrison Hershfield, 2022 ONSC 3485 (“Dufferin”), yet the Court in Atkinsréalis reached the opposite result. The two decisions can appear hard to reconcile, but a close reading shows that the difference lies in how the arbitrator questioned witnesses in each case.
Justice Myers in Atkinsréalis also issued a preliminary decision, reported at Atkinsréalis Construction (Pacific) Inc. v. GIP Construction Ontario East Ltd., 2026 ONSC 4785 (the “Stay Decision”), on whether the arbitration should be stayed while the application to remove the arbitrator was pending. The Stay Decision appears to be the first reported case interpreting s. 13(7) of Ontario’s Arbitration Act, 1991.
The Atkinsréalis arbitration arose from the Trillium Line Extension Stage 2 Project in Ottawa. Atkinsréalis Construction (Pacific) Inc. (“ATRL”) was the prime contractor, and GIP Construction Ontario East Ltd. (“GIP”) was its roadworks subcontractor. The parties had competing claims arising from delays and changes to the scope and timing of the work.
The parties agreed to arbitrate their disputes and appointed an experienced construction arbitrator. Both knew the arbitrator preferred a somewhat inquisitorial approach to hearings and accepted it. His terms of appointment expressly permitted him to “question any witness, including Party-appointed experts, provided that the parties shall be provided the opportunity to ask additional questions on any matters addressed or raised by the Arbitrator’s questioning”.
The hearing commenced on December 8, 2025. On the eighth day of the hearing, after the examination of the fact witnesses had concluded, ATRL challenged the arbitrator on the basis that his questioning of its witnesses gave rise to a reasonable apprehension of bias.
The arbitrator dismissed the challenge. ATRL then applied to the Superior Court under s. 13(6) of the Arbitration Act, 1991 for an order removing the arbitrator.
The arbitrator ruled that the arbitration should continue while the challenge was before the court. ATRL successfully moved to stay the arbitration pending the outcome of the application.
In the Stay Decision, Justice Myers had to decide whether the arbitration should continue while the challenge was pending. Section 13(7) of the Arbitration Act, 1991 provides that, while an application is pending, the arbitral tribunal, including the challenged arbitrator, may continue the arbitration and make an award “unless the court orders otherwise”. Neither party could find case law interpreting the provision.
Justice Myers applied the common law test for stays, guided by the tripartite test in RJR-MacDonald Inc v Canada (Attorney General), but read in light of the statutory scheme. The Act’s goal of respecting the independence of the arbitral process favours deference to arbitrators and the completion of the arbitration before judicial review (Stay Decision, para. 25). The default position under s. 13(7) is therefore that the arbitration continues, and a challenge in court does not by itself shut the arbitration down (Stay Decision, para. 26).
At the threshold stage, the Court held that it is not enough to show a serious issue to be tried. Observing that bias is an “overused” allegation that is rarely made out, Justice Myers required the higher standard of a “strong prima facie case”, recognizing the strong presumption of independence enjoyed by arbitrators and judges alike. That standard was also appropriate because the evidentiary record, consisting of the arbitration transcripts, was already fixed. On the record before him, Justice Myers was satisfied that ATRL had met that standard.
On irreparable harm, Justice Myers held that requiring ATRL to continue appearing before the arbitrator in light of a strong prima facie case of bias would itself be irreparable harm, relying on the Court of Appeal’s statement in Vento that a reasonable apprehension of bias irreparably compromises the integrity of an adjudicative process. In such cases, irreparable harm and the balance of convenience are “nearly merged”, particularly when, as here, the stay would cause only a brief delay. As he put it, an award by a tribunal later found to have created a reasonable apprehension of bias “is of no value to anyone”. The Court also rejected the suggestion that the arbitrator could hear the expert evidence and then await the Court’s decision, as this would neither expedite the proceeding nor assist the Court. The stay was therefore granted.
Having stayed the underlying arbitration, Justice Myers then dealt with the bias challenge itself. He began with the statutory framework.
Section 19 of the Arbitration Act 1991 requires the parties to an arbitration to be treated equally and fairly, while s. 3 expressly prevents parties from contracting out of those requirements.
The Court recognized that arbitration allows parties considerable flexibility in designing the process by which their disputes will be resolved. Parties can select arbitrators with specialized knowledge and agree to procedures that differ from those ordinarily followed in court. That flexibility may include permitting an arbitrator to take a more inquisitorial approach to the evidence.
However, an agreement to an inquisitorial procedure does not displace the arbitrator’s duty of impartiality.
Justice Myers referred to the Supreme Court of Canada’s decision in Szilard v. Szasz, [1955] SCR 3 , and the Ontario Court of Appeal’s more recent decision in Vento Motorcycles, Inc. v. Mexico, 2025 ONCA 82, which confirm that the requirement of impartiality applies to arbitrators as it does to judges.
The reasonable apprehension of bias test is objective: whether an informed person, viewing the circumstances realistically and practically, would reasonably apprehend that the arbitrator might not decide the dispute fairly. Actual bias or an intention to favour one party need not be established.
This was important in Atkinsréalis, where the parties had expressly agreed to the arbitrator’s questioning of witnesses. Justice Myers stated that even though an arbitrator’s expertise and familiarity with the evidence may justify a more active role in questioning, party agreement cannot override the statutory requirement of impartiality.
The central issue in Atkinsréalis was whether the arbitrator’s questioning of witnesses had crossed the line into impermissible cross-examination. Justice Myers explained at paragraph 63, “subject to contextual factors”, arbitrators, like judges, “must not cross-examine” because questioning in that way and for those purposes is the role of adversarial counsel and is inconsistent with the neutrality required of a decision maker.
The concern, as Justice Myers explained at paragraphs 65–67, is that cross-examination is inherently not neutral: it involves controlling a witness through leading questions to elicit desired evidence, undermine evidence already given, or impair the witness’s credibility. The examiner seeks to put words in the witness’s mouth and knows in advance what they want the witness to say. In Justice Myers’ view, this form of questioning is not simply a means of clarifying evidence or searching for the truth; it is designed to advance the examiner’s position, creating a risk that an arbitrator who adopts this approach will appear to have departed from neutrality.
That does not mean that every leading question gives rise to a reasonable apprehension of bias. At paragraph 68, Justice Myers emphasized that the impugned conduct must be assessed “in its full context”, having regard to the strong presumption of neutrality and the high threshold set by the Supreme Court of Canada in Committee for Justice and Liberty et al. v. National Energy Board, [1978] 1 SCR 369. Nor is the arbitrator’s subjective intention determinative: a stated intention to conduct an inquisitorial search for the truth does not answer whether the conduct gives rise to a reasonable apprehension of bias.
Applying these principles, Justice Myers found that the arbitrator’s questioning of two ATRL witnesses went beyond permissible inquiry and, in its context, crossed into cross-examination.
In questioning the first ATRL witness, the arbitrator began with seemingly general questions about how contractors deal with disputed changes. Through leading questions, he narrowed the witness’s answers to a list of options that did not include termination, a key issue in the arbitration. He later returned to those answers, dismissed the witness’s attempts to explain ATRL’s position, and pressed him to admit that ATRL had not pursued any of the listed options before terminating GIP (paras. 74 and 78-79). Justice Myers characterized this sequence as “classic cross-examination technique”: pinning a witness to seemingly harmless points early on and confronting them with those admissions later (para. 75).
The arbitrator’s questioning of a second ATRL witness raised similar concerns. He questioned the witness about counsel’s role in drafting his witness statement and dismissed portions of it as “spin” (paras. 83 and 87). Justice Myers accepted that evidence extending beyond a witness’s personal knowledge may properly be given less weight. However, he found that the questioning went further by targeting the witness’s credibility because counsel had drafted the evidence, rather than focusing on its factual basis. Notably, a GIP witness whose evidence raised similar concerns was not subjected to the same questioning.
What proved decisive was the nature and context of these interventions, not the mere use of leading questions. Considering the questioning as a whole, Justice Myers concluded that a reasonable observer could conclude the arbitrator had prejudged the outcome and, as such, might decide the arbitration unfairly.
Atkinsréalis stands in contrast to Dufferin, where Justice Woodley of the Superior Court of Justice dismissed a similar challenge arising from a matter involving the same arbitrator.
In Dufferin, a party applied to remove the arbitrator alleging that he had shown prejudgment, pursued admissions adverse to its case, and failed to take a balanced approach to the witnesses called by each party. As in Atkinsréalis, the procedural order expressly allowed the arbitrator to question any witness and to interject at any time. He had also received extensive written evidence before the hearing.
Justice Woodley reviewed the complete transcripts and rejected the challenge. She found “no significant difference in the Arbitrator’s approach to any of the witnesses”. He asked difficult questions of witnesses on both sides and always allowed counsel to follow up (paras. 124, 135 and 151). Where both sides face equally hard questions, it is difficult to see how an informed observer, considering the “full context” that Atkinsréalis also requires, could conclude that the arbitrator had prejudged the issues in favour of one party.
The arbitrator had questioned the party’s witnesses about an apparent change in position. Justice Woodley found that these questions flowed logically from the evidence. They gave the witnesses a chance to explain the inconsistency before any decision was made (paras. 126 and 148). She concluded that the arbitrator had not become an advocate for either party but was advocating for the truth, wherever that led.
On their face, Atkinsréalis and Dufferin appear to present contrasting outcomes concerning similar allegations of bias. Justice Myers himself described the issues in Dufferin as “very similar” (para. 57). However, the decisions are not necessarily inconsistent. The difference is better explained by the lens each court applied to the arbitrator’s questioning.
In Dufferin, Justice Woodley’s focus was even-handedness. She reviewed the complete transcripts and found no significant difference in the Arbitrator’s approach to any of the witnesses (para. 124). That finding was central to her rejection of the advocacy allegation: the arbitrator “did ask difficult questions, but he asked them of all witnesses” (para. 135). She also appeared to give weight to the arbitrator’s purpose, finding that his interventions were intended to ensure “a fair and independent process” (para. 123). She accepted that questioning by an expert tribunal, even if pursued extensively and vigorously, is acceptable if it is aimed at clarifying the evidence (para. 161).
In Atkinsréalis, Justice Myers focused on the nature of the questioning. Like Justice Woodley, he accepted that the arbitrator was highly prepared. The question for him was whether the questioning had moved beyond clarification into adversarial cross-examination, used to secure desired admissions, undermine evidence or impair credibility (para. 65). Not every leading question crosses that line (para. 68), but questioning of that kind “helps one side and hurts the other by design” (para. 104). Unlike Justice Woodley, Justice Myers gave no weight to the arbitrator’s purpose, holding that “the intention of the arbitrator or judge is not relevant” (para. 64).
Unequal treatment also played a role in Atkinsréalis. Justice Myers found that only ATRL’s witness was cross-examined to undermine his credibility, while a GIP witness with similar issues was not (paras. 84 and 89). But he saw no value in the broader comparison that had been decisive in Dufferin. GIP pointed to “competing snippets” to show that its own witnesses were asked similar questions (para. 97), and Justice Myers declined to engage with them, holding that the cross-examinations of ATRL’s witnesses “so plainly cross the line” that nothing would be gained by examining “disembodied snippets of the transcripts” (para. 98). The reasons do not identify the questions GIP relied on; presumably, they were not of the same kind as those put to ATRL’s witnesses.
Taken together, the decisions draw a line between testing evidence to understand it and questioning that objectively appears to advance one party’s case. An arbitrator may put an apparent inconsistency to a witness and invite an explanation. This was a key feature of the questioning approved in Dufferin, where the arbitrator told the witness, “I am giving you an opportunity to explain how this works, from your perspective” (para. 155). However, Atkinsréalis holds that a series of leading questions designed to secure adverse admissions or attack credibility is “presumptively on the wrong side of the line” (para. 103). That holds even where the arbitrator is even-handed and acting in good faith.
The arbitrator, Justice Myers, and Justice Woodley all adopted Glaholt and Rotterdam’s article, Arbitrator Questioning: Sphinx or Skeptic?, published in the Journal of the Canadian College of Construction Lawyers in 2016.1 Justice Woodley agreed with the authors’ observation that an arbitrator selected for subject matter expertise is better in the role of skeptic than Sphinx, provided that active skepticism does not become advocacy. The arbitrator relied on the breadth of areas of questioning listed by Glaholt and Rotterdam, but Justice Myers drew attention to the qualification that questioning must not create an appearance of one-sidedness or unfairness and that they did not specifically mention cross-examination (paras. 53–55).
Both decisions also considered whether failing to object to an arbitrator’s questioning at the time it happens prevents a later challenge. In Dufferin, Justice Woodley held that a party need not object before bringing a challenge. This is because a reasonable apprehension of bias may arise from the cumulative effect of a series of questions or comments, even where no single intervention would justify a challenge (paras. 84–91). Justice Myers took a similar approach in Atkinsréalis. Because parties cannot consent to unfair proceedings, a failure to object is relevant only to the question of the bona fides of the timing of raising the bias issue (paras. 106 and 111).
The practical lesson is that counsel need not object to every question to preserve a challenge. They should, however, act promptly once sufficient grounds exist, and keep in mind any statutory or institutional time limits.
The attraction of arbitration, whether domestic or international, is not simply that parties can resolve their disputes privately. It is also that they can select a decision maker with experience and expertise relevant to their dispute.
In construction arbitrations, that expertise can be particularly valuable. Arbitrators may be required to assess complex delay analyses, competing engineering opinions, extensive contemporaneous project records, and contractual arrangements that are not always familiar to judges. An arbitrator with experience in these matters may identify inconsistencies or gaps in the evidence that warrant further explanation. In international arbitration, the parties and counsel may also come from different legal traditions, with very different expectations of how actively a tribunal should question witnesses.
The decision in Dufferin recognizes the value of that approach. Indeed, the Court placed considerable weight on the arbitrator’s preparation and ability to use his knowledge of the dispute to identify the important issues.
Atkinsréalis complements this. Its approach to the equal-treatment issue is open to debate, since Justice Myers declined to compare the questioning of each side’s witnesses (para. 98). Still, it rightly accepts that context matters: an expert arbitrator may come into the hearing “hot”, and not every leading question will create a reasonable apprehension of bias (paras. 68 and 90). But context cannot displace the “fundamental requirement that the Arbitrator must be neutral” (para. 105).
There are several practical implications.
First, parties should consider carefully the procedures they agree to when appointing an arbitrator. A provision authorizing an arbitrator to question witnesses may be appropriate, particularly where the parties expect the arbitrator to use specialized expertise to understand the evidence. However, such a provision does not displace the statutory requirements of neutrality and fairness.
Second, a party challenging an arbitrator should consider early whether to seek a stay. The default outcome is that the arbitration continues, and an arbitrator who has rejected a challenge can be expected to keep the process moving. A stay will be granted only where the applicant can show a strong prima facie case, so the record supporting the challenge should be developed carefully from the outset.
Third, an arbitrator’s familiarity with the evidence does not, by itself, determine whether a particular line of questioning is appropriate. The distinction lies in how the questions are framed and pursued.
Fourth, full context still matters. It carried less weight in Atkinsréalis, and we do not know which questions put to the GIP’s witnesses Justice Myers considered less difficult than those put to ATRL’s witnesses. In our view, Dufferin remains good law. It can be distinguished as a case where the full record showed that the arbitrator asked both sides equally difficult questions in pursuit of the truth, wherever it led.
Finally, agreeing to an inquisitorial process does not mean counsel must accept all questioning without objection. Concerns should be raised promptly.
In practice, the line between legitimate inquiry and advocacy is not always obvious during a lengthy hearing. A question meant to clarify evidence can grow into a more extensive examination, especially where the answers raise further questions. The test is whether the arbitrator’s interventions, taken as a whole, objectively create a reasonable apprehension that the arbitrator may not decide the matter impartially.
Key takeaways for counsel and arbitrator:
Neither decision sets a precise limit on how much questioning is permissible, and neither suggests that arbitrators should avoid challenging evidence or identifying inconsistencies. Both reinforce the guidance in Arbitrator Questioning: Sphinx or Skeptic?: an arbitrator chosen for subject matter expertise is better in the role of skeptic than Sphinx, provided that active skepticism does not become advocacy.
1 Glaholt & Rotterdam, “Arbitrator Questioning: Sphinx or Skeptic?” (2016) Journal of the Canadian College of Construction Lawyers 81, cited in Dufferin, paras. 137, 161 and 168, and Atkinsréalis, paras. 53–55.
In its recent decision in Tedescon Infrastructure Ltd. v. The County of Simcoe (Tedescon), the Ontario Superior Court of Justice upheld an arbitration award in a construction dispute despite finding the arbitrator had committed an error of law. Specifically, the Court found that the arbitrator erred in finding that the contractor’s contractually-permitted work stoppage was nevertheless a repudiation of the contract entitling the owner to damages for the cost of a replacement contractor. However, the Court found that a clause in the parties’ arbitration agreement that arbitration was to be “final and binding” meant that it had no jurisdiction to hear the contractor’s appeal of the arbitration award.
Tedescon concerned a 2020 contract from the County of Simcoe for reconstruction of a section of road, awarded to Tedescon Infrastructure Ltd. (Tedescon). The contract was a unit rate contract based on the Ontario OPSS.MUNI 100 standard form for public works construction contracts.
During the project, Tedescon submitted claims for compensation and extensions of time relating to change orders and alleged delays, a number of which had been rejected. Tedescon commenced arbitration of those claims in March 2022. The parties appointed an arbitrator and entered into an Arbitration Agreement based on the arbitrator’s standard form.
Nearly a year later, Tedescon served notice of default on the County alleging over 80 breaches of the contract and claiming that the contract administrator had failed to act impartially. Shortly after issuing the notice of default, Tedescon gave notice that it was exercising its right under GC 7.10 of the OPSS.MUNI 100 general conditions to suspend work until the County resolved the alleged breaches. The County responded with a letter stating that if Tedescon did not resume work, it would consider Tedescon to have abandoned the contract.
Tedescon did not return to work. Shortly afterward, the County awarded a contract to another contractor to complete Tedescon’s work. The parties amended their pleadings in the existing arbitration to include the issue of Tedescon’s alleged repudiation of the contract.
The arbitrator found that Tedescon’s work stoppage amounted to abandonment of the contract, and that Tedescon’s contractual right to stop work did not include a right to abandon the contract. The arbitrator held that Tedescon’s refusal to remobilize was therefore a repudiation of the contract, which the County accepted by replacing Tedescon with another contractor. The arbitrator awarded the County a net amount of $511,243.13.
Tedescon applied to the Ontario Superior Court of Justice for leave to appeal the arbitrator’s award, as well as to set aside the award on grounds of lack of jurisdiction and procedural unfairness. The Court dismissed Tedescon’s application in its entirety.
The threshold issue on Tedescon’s application for leave to appeal was whether a right of appeal was available at all. The County argued that the arbitration agreement, which provided that arbitration decisions would be “final and binding,” precluded any appeal.
Section 45(1) of the Arbitration Act, 1991 provides that “if the arbitration agreement does not deal with appeals on questions of law, a party may appeal an award to the court on a question of law with leave”. Section 3 of the Act permits parties to contract out of provisions of the Act, including the appeal rights in section 45(1). The Court noted that the Ontario Court of Appeal has interpreted sections 3 and 45(1) to mean that there is no right of appeal if the arbitration agreement expressly or impliedly precludes appeals.
The Court in Tedescon noted that the Ontario Court of Appeal has previously found that words similar to “final and binding” in an arbitration agreement can be sufficient to exclude rights of appeal. In Baffinland Iron Mines LP v. Tower-EBC G.P./S.E.N.C. (Baffinland), the Ontario Court of Appeal found that an arbitration agreement that provided for disputes to be “finally settled” by arbitration pursuant to the arbitration rules of the International Chamber of Commerce negated the right of appeal that would otherwise be available under section 45(1) of the Arbitration Act, 1991.
However, the Court in Tedescon also noted that in Denison Mines Ltd. v. Ontario Hydro (Denison Mines), a case in which the parties’ contract mandated arbitration but the parties nevertheless entered into a separate standalone arbitration agreement, the Ontario Court of Appeal found that similar “finally settled” language was not sufficient to exclude the right of appeal in section 45(1). In Denison Mines, the “finally settled” language was included in the parties’ contract, but not in the subsequent arbitration agreement. The Court found that the later arbitration agreement prevailed, with the result that the parties were free to apply for leave to appeal under section 45(1).
The Court in Tedescon applied Baffinland and distinguished Denison Mines, finding that the words “final and binding” in the arbitration agreement were sufficient to exclude any right of appeal. The Court held that the facts before it were “logically the reverse” of the facts of Denison Mines, because the mandatory arbitration clause in GC 3.14 of the OPSS.MUNI 100 general conditions expressly contemplates the existence of a separate arbitration agreement. As a result, the Court found that it was unnecessary to “agonize over the priority” of the two agreements, and that the “final and binding” language in the arbitration agreement prevailed.
The Court in Tedescon went on to say that if a right of appeal had been available, it would have granted leave to appeal and found an error of law in the arbitrator’s award. The Court held that the arbitrator made two incompatible findings. In the Court’s view, once the arbitrator had found that Tedescon was justified in exercising its contractual right to stop work, the arbitrator could not have found that Tedescon had repudiated the contract. The Court stated that justification for stopping work and abandonment are “two sides of the same coin”, and “it must land on one side or the other”. As a result, the Court found that Tedescon’s conduct could not have been an abandonment or repudiation of the contract.
The Court found that the case law relied upon by the arbitrator to find that Tedescon had abandoned the contract was inapplicable. The Court held that cases like D&M Steel Ltd. v. 51 Construction Ltd. stand for the principle that a contractor cannot refuse to perform work as leverage to force an owner to pay amounts that are not owed or have not yet become due. In contrast, the Court found that Tedescon’s exercise of its right to stop work was based on amounts that were already due, but not paid, by the County.
In the result, the Court dismissed Tedescon’s application and upheld the arbitration award, despite having found that the award was based on an error of law and that it would have otherwise replaced the $511,243.13 award in favour of the County with a $2,598,518.75 award in favour of Tedescon.
Tedescon is a reminder that contractual language about the finality of arbitration, such as “final and binding”, can have significant consequences for a party’s ability to appeal an unfavourable award. Although eliminating rights of appeal might be desirable for some parties who may want to avoid the additional expense and potential delays to enforcement that could arise from appeals to the Court, that choice can lead to situations like Tedescon, in which a party may find itself without recourse even when the Court would otherwise agree with an unsuccessful party’s position.
Spoliation is an old legal doctrine. It refers to the intentional destruction, alteration, mutilation or concealment of evidence with a view to subverting the truth-finding process during litigation. According to the Alberta Court of Appeal, the doctrine is so old that in ancient Rome, “business[people] were obligated to keep a written record of their affairs for a particular period of time” and if they failed to do so the doctrine of spoliation was applied “with much harshness” by Roman courts.1
This doctrine was last addressed by the Supreme Court of Canada in a 19th Century case about lost records on a construction project.2 Since then, the doctrine was left to develop in the trial and appellate courts throughout Canada. That is, until this year, when the Supreme Court of Canada ruled on another case about missing records.3
The Court’s ruling in SS&C Technologies provides useful guidance on the modern application of this ancient doctrine. In document intensive construction disputes, which also involve a great deal of physical evidence, practitioners should be mindful of the potential adverse inferences in situations involving reasonably contemplated or ongoing litigation. This is particularly the case, as the Court held that “there should be little difficulty in concluding that [destruction] was done to affect the litigation” in situations where a party “unjustifiably” destroys, alters, mutilates, or conceals evidence when it knows or ought to know litigation is on the horizon.4
In SS&C Technologies, the Supreme Court of Canada considered a dispute arising from a data-licensing agreement. Under the agreement, SS&C provided proprietary securities information to the BNY banking corporation. Under the agreement, BNY was prohibited from sharing, redistributing, or selling the information to third parties, including its subsidiaries and affiliates.
SS&C learned that BNY was allegedly sharing the information across its corporate family and demanded that BNY preserve records documenting its use and distribution. Despite that demand and the prospect of litigation, BNY failed to preserve the records.
A lawsuit ensued. At trial, the trial court found that BNY had breached the licensing agreement but made no express finding of spoliation. The trial judge nevertheless drew two adverse inferences:
The trial court awarded damages against BNY. On appeal, the Ontario Court of Appeal held that BNY had committed spoliation, describing its conduct as “smack[ing] of contempt for the justice system.”5 The Appeal Court upheld the trial judge’s adverse inferences and damages award.
Both parties appealed to the Supreme Court of Canada.
Spoliation “refers to the intentional destruction of relevant evidence when litigation is existing or pending.”6 The governing test in Canada originated in the Supreme Court of Canada’s 1896 St. Louis decision, which held that spoliation occurs when a party intentionally destroys evidence relevant to ongoing or contemplated litigation in circumstances supporting a reasonable inference that the destruction was intended to affect the proceeding.7 Accordingly, the alleging party must establish, on a balance of probabilities, both intentional destruction and a reasonable inference that it was undertaken to gain an advantage in the litigation.
Under St. Louis, once spoliation is established, the principal remedy is a rebuttable presumption that the destroyed evidence would have been unfavourable to the spoliator. The alleged spoliator may rebut the presumption by showing that the destruction was not intended to affect the litigation or by relying on other evidence that proves or disproves the claim.
Before SS&C Technologies, the Supreme Court of Canada had not considered spoliation in more than a century. The case therefore provided an opportunity to clarify the doctrine in the context of modern technology and address several spoliation related issues, including:
In clarifying the doctrine, the Court established a four-part test that the alleging party must prove on a balance of probabilities to show spoliation:
Although the test largely reflects the existing law, the Court clarified that once the first three elements are established, “there should be little difficulty in concluding that this was done to affect the litigation”9 (emphasis added).
In the writers’ opinion, the “little difficulty” language appears to lower the burden on parties alleging spoliation to prove an intent to affect the litigation process. This has always been a requirement of the spoliation test, but stood as a significant hurdle as proving that evidence was destroyed with the specific intention to affect a lawsuit (as opposed to innocent or negligent reasons for destroying evidence) was difficult, to say the least. Now, SS&C Technologies suggests that absent justification, there should be “little difficulty” if evidence is destroyed in the face of anticipated or actual litigation.
The Court also held that if the rebuttable presumption is not rebutted, the adverse inference is mandatory rather than permissive. A trial judge must draw an inference capable of “filling the gap in the evidence” created by the spoliation.10 The Court made it clear that “[p]arties who commit spoliation should not be rewarded for the evidentiary vacuums they create. Rather, as the authors of their own misfortune, spoliators must carry the presumption that the impugned evidence would be unfavourable to their case.”11
While the presumption itself is mandatory, the trial judge retains the discretion to determine how and to what extent the spoliated evidence would have been unfavourable. The Court rejected SS&C’s argument that the highest possible adverse inference should be the mandatory response to spoliation. The Court held that enforcing a one-size-fits-all mandatory remedy that imposes the “maximum penalty” could create disproportionate consequences.12 Instead, the Court recognized that trial judges are well placed to consider the context, scope, and impact of the destroyed evidence.13
As to remedies outside of the mandatory presumption, the Court recognized that courts across Canada, exercising their inherent power to remedy abuses of process, have levied a variety of other sanctions targeted at the distinct prejudice created by spoliation.14
Historical case law indicates that such sanctions or remedies could include (i) striking of claims; (ii) the exclusion of expert reports; (iii) remedies arising from the preservation and production of evidence (like orders that an expert be subject to examination for discoveries; and (iv) costs penalties (amongst others).
The Court pointed future courts, in crafting additional remedies, to consider factors such as: the level of culpability of the spoliator; the intention or reason behind the destruction of the evidence; the prejudice to the non-spoliating party; and the impact of the destruction of evidence on the court’s ability to fairly dispose of the issues.15
Accordingly, what specific remedy might be available appears to be a highly fact and context specific exercise. Accordingly, courts may be unwilling to grant significant pre-trial remedies, preferring to wait until trial, when the full factual matrix has been developed. In that regard, the Court’s holding on potential remedies simply maintained the status quo that had developed in the 125+ years since St. Louis.
SS&C Technologies provides the Supreme Court of Canada’s first statement on the doctrine of spoliation since the 19th Century. While the decision does not change the common law relating to spoliation in a significant manner, it provides an updated assessment of this doctrine.
Importantly, SS&C Technologies appears to make it easier for parties to seek an adverse inference in the event evidence is destroyed, altered, mutilated or concealed in the face of actual or anticipated litigation. Historically, litigants would have faced significant difficulties in showing that a counterparty destroyed evidence with the intention to affect the litigation. But, with SS&C Technologies, this burden has been made easier, as according to the Court “there should be little difficulty in concluding that [destruction] was done to affect the litigation” in situations where a party “unjustifiably” destroys, alters, mutilates, or conceals evidence when it knows or ought to know litigation is on the horizon.16
In the construction context, disputes can generate a substantial amount of evidence: both documentary and physical. Parties who know or ought to know that a litigious dispute might arise on a project are well advised to take positive steps to preserve documents and other physical evidence. Robust document retention policies, and litigation holds, which have always been important, take on a greater level of importance to avoid a spoliation inference.
For situations involving defective work or products that have to be repaired, parties should be mindful that physical evidence that is destroyed is just as much spoliation as documents that are deleted. For circumstances involving defective work or products, parties should ensure that they document the work at hand, and provide notice to any potential counter-parties prior to any repair or replacement. More thought may be required in terms of taking positive steps to preserve physical evidence, if there can now be “little difficulty” in a Court drawing an inference of spoliation. Such steps might include:
In sum, SS&C Technologies provides a strong message that parties who destroy relevant evidence should not benefit from the uncertainty they create. Given that message, those involved with construction projects should take steps to preserve and properly document what they can, to avoid future arguments about spoliation.
1 McDougall v. Black & Decker Canada Inc., 2008 ABCA 353 at para 15 [“McDougall”]
2 St. Louis v The Queen, (1896) 25 SCR 649 (SCC) [“St. Louis”].
3 SS&C Technologies Canada Corp v Bank of New York Mellon Corporation, 2026 SCC 29 [“SS&C Technologies”].
4 SS&C Technologies at para. 80.
5 SS&C Technologies Canada Corp. v. The Bank of New York Mellon Corporation, 2024 ONCA 675 at para 165.
6 McDougall at para 29.
7 McDougall at para 18; St. Louis.
8 SS&C Technologies at para 80.
9 SS&C Technologies at para 80.
10 SS&C Technologies at paras 79-82.
11 SS&C Technologies at para 1.
12 SS&C Technologies at para 95.
13 SS&C Technologies at para 89.
14 SS&C Technologies at para 91.
15 SS&C Technologies at para 94.
16 SS&C Technologies at para. 80.
Warranty clauses are usually considered to be a source of rights. Owners and contractors often view warranties as protective, and as a promise provided by a subcontractor or supplier to correct a problem. Warranties do often provide remedies, including the replacement of materials, and the remediation of deficiencies. However, a recent decision from the British Columbia Supreme Court illustrates that in addition to providing remedies, warranty clauses can also limit or exclude remedies that would otherwise be available at common law or in tort.
In Belfor (Canada) Inc. v BC Hardwood Floor Co. Ltd., 2026 BCSC 293, the Plaintiff was hired to be the general contractor on a flood restoration and repair project at a home in Vancouver. The Plaintiff hired the Defendant to supply and install flooring in the house. The parties entered into a contract which included a labour and materials warranty clause. This warranty stated the following:
After the Plaintiff completed their restoration work, the homeowner discovered deficiencies with the flooring work done by the Defendant. The Plaintiff notified the Defendant, and the Defendant completed the necessary remedial work. While the remedial work was being completed, the Plaintiff arranged for additional steps to be taken so that the flooring deficiencies could be addressed. These included packing and storing the homeowner’s furnishings in the house, providing the owner with alternative living arrangements while the remediation work took place, and cleaning the house at the conclusion of the remedial work. The Plaintiff invoiced the Defendant for these additional expenses, which amounted to $145,258.08 including GST.
The result of the case hinged on the interpretation of the “labour and materials” clause in the warranty. The Plaintiff argued that the “labour and materials” clause must mean something more than simply the Defendant’s costs to rectify the flooring deficiencies. Specifically, the Plaintiff contended that it was entitled to reimbursement for additional expenses were necessary to rectify the flooring deficiencies. In order for the remediation work to be done, for example, the dishwasher and furniture needed to be removed and stored elsewhere. The Plaintiff asserted that when the Defendant did not disagree with the Plaintiff that the floor needed to be free and clear of furniture before the remedial work could be done, they implicitly agreed to cover the additional costs.
The Defendant argued that the wording of the warranty was clear that the additional expenses sought by the Plaintiff did not fall under the scope of “labour and materials” and that the warranty specifically excluded the additional expenses. The Defendant also stated that there were additional clauses in the contract that specifically dealt with issues including the removal of furniture and final touch-up, and that the contract should be read as a whole to determine whether the Defendant was liable for the Plaintiff’s losses.
Ultimately, the Court agreed with the Defendant, and found that the additional expenses sought by the Plaintiff did not fall under the “labour and materials” clause in the warranty. In doing so, the Court emphasized that the law of contracts has moved towards a practical, common-sense approach. In a contract dispute, the Court will seek to determine the intent of the parties, largely by giving the words used their ordinary and grammatical meaning, while also examining the circumstances known to the parties at the time the contract was formed.
The Court found that in this case, the Plaintiff and Defendant were commercially experienced parties, and there was no inequality in bargaining positions at the time the contract was formed. At no time did the Defendant agree to pay for the additional expenses. The contract was clear that additional tasks such as final dust cleanup, removal of existing flooring and baseboards, and installation of new baseboards were to be done by others. As a result, the Court found that the labour and materials clause was a valid and applicable contractual exclusion clause that shielded the Defendant from liability. Ultimately, the Plaintiff was responsible for the additional costs incurred.
Belfor provides a useful reminder that warranty clauses are not merely promises of additional protection. Although contractors and owners often view warranty clauses as a promise of additional protection, Belfor demonstrates that warranty clauses can also do the opposite and exclude or limit subcontractors’ or suppliers’ liability. Warranty clauses that appear to provide remedies may, in practice, actually limit or exclude rights available to contractors and owners in common law or in tort.
Brendan D. Bowles (Ontario)
Catriona Otto-Johnston (Alberta)
British Columbia:
Dirk Laudan, Mike Preston, Christopher Hirst
Alberta:
Donald C.I. Lucky, KC, Corbin Devlin, Ryan Krushelnitzky
Saskatchewan:
Collin Hirschfeld, KC, Misty Alexandre
Manitoba:
David Marr
Ontario:
Paul Ivanoff, Jason Annibale
Quebec:
Patrice Morin
Atlantic Provinces:
Greg Moores, Conor O’Neil