You Can’t Always Get What You Plead: The DIFC Courts’ First Award Set-Aside

On 24 April 2026, the Dubai International Financial Centre (“DIFC”) Court of Appeal (“Court”) handed down Oheo Bank v Parker [2025] DIFC CA 006, understood to be the first occasion on which the DIFC Courts have set aside a DIFC-seated arbitral award.

The set-aside application was premised on procedural fairness objections: the arbitral tribunal upheld a claim that only crystallised in the claimant's post-hearing brief, and did so without inviting the respondent to respond. In reaching its decision, the Court also chose between two competing common law approaches to how the "scope of the submission to arbitration" should be tested.

The Claim

The underlying arbitration concerned Oheo Bank's role in a transaction to finance the purchase of a vessel by its customer, Parker. The arrangement involved liquidating existing assets, transferring EUR 1.4 million to a third party, and receiving two corporate bonds intended to serve as collateral in exchange for a loan (at [19]). Before the transfer, Oheo Bank's back office had discovered that the bonds were illiquid and untradeable, and required Parker to sign an indemnity agreement (“Indemnity”) before it would take them into custody. Crucially, Parker was never told that this was the reason for the Indemnity. One of the bonds later turned out to be worthless, resulting in Parker losing EUR 1 million.

Parker's pleaded case was that Oheo Bank's relationship manager had advised Parker into the transaction, and that the advice was "deliberately opaque and designed to mislead" (at [55]). The arbitral tribunal rejected Parker’s version of that case, unanimously dismissing claims in deceit, misrepresentation, Quincecare duty, and negligence.

In its post-hearing brief, Parker reframed its case: it was not that Oheo Bank had given misleading advice, but that in requesting the Indemnity, Oheo Bank had volunteered information about the bonds and done so misleadingly by omitting the reason it was really asking about them. This, thus, recast the regulatory-breach claim as a Crestsign duty case, a characterisation the pleadings had never engaged with. A majority of the arbitral tribunal accepted it and awarded EUR 1 million, while the dissenting arbitrator flagged the shift as "real procedural unfairness" (at [42]). Tellingly, the arbitral tribunal later agreed unanimously that the successful claim "was never pleaded" and "was articulated for the first time only in closing submissions" (at [49(a)], quoting Costs Award [28]). The majority did not deny the claim was unpleaded, but took the view that Oheo Bank nonetheless had "a fair opportunity to address the unpleaded claim" (Costs Award, at [28]). It was that finding of fair opportunity, not the finding that the claim was unpleaded, that the Court of Appeal went on to reject.

Oheo Bank's Response, and Why the Arbitral Tribunal's Silence Mattered

Oheo Bank's post-hearing brief listed some twenty points from Parker's brief that were not previously pleaded, including the paragraph containing the reframed claim. Oheo Bank submitted that it would not engage substantively with these ‘new’ arguments but that "if the Tribunal wishes to be addressed on them, the appropriate course would be for the Tribunal to direct the Respondent to file a Supplemental Post-Hearing Brief".  The arbitral tribunal never issued that direction and hence, Oheo Bank never made those submissions. The majority of the arbitral tribunal nevertheless found for Parker on the reframed claim.

In the setting-aside application the Court held that the tribunal's failure to direct Oheo Bank to file responsive submissions on the reframed claim, despite Oheo Bank's own request for exactly that direction, is what sank the award: "the real vice concerning the Successful Claim was not that it fell outside the scope of the submission to the Arbitration but rather that... the Bank was not given the opportunity to deal with it" (at [125]).

Two Ways an Objection Can Cut

Once Parker reformulated its case in the post-hearing brief, the arbitral tribunal faced a straightforward choice: stay with the pleaded case, or go with the new theory after hearing from Oheo Bank. It did neither. One point worth noting is that a request for a supplemental brief is not itself the opportunity to make one. The majority of the arbitral tribunal had reasoned that Oheo Bank had cross-examined Parker's principal witness on the Indemnity, and so had had its chance. The Court rejected that argument. It was of the view that at the time of cross-examination the successful claim was still unpleaded and "the Bank was not in a position to know the correct target at which to direct cross-examination". Far from curing the unfairness, Oheo Bank's written objection strengthened the case that it existed: Oheo Bank had put the arbitral tribunal on notice that it was deliberately not engaging with the unpleaded points and expected a direction if any were to be relied on, a stance the Court later described as one the Bank had "reasonably and clearly flagged".

What the arbitral tribunal should  have done to avoid breaching any procedural fairness in the proceedings, drawing on The Vimeira [1984] 2 Lloyd's Rep 66 and Zermalt Holdings v Nu-Life Upholstery Repairs Ltd [1985] 2 EGLR 14 (both cited at [92]-[93]), was to put the reformulated claim to Oheo Bank and invite submissions on it, "as the Bank had expressly requested in its Post-Hearing Brief" (at [116(c)]). Doing neither, and then relying on the reformulated claim, was procedural unfairness of the kind Article 41(2)(a)(ii) of DIFC Law No. 1 of 2008 (“DIFC Arbitration Law”) exists to remedy. The Court was careful to place its finding on "a very narrow, fact specific basis" and it emphasised the general distinction that Article 41(2)(a)(ii) of the DIFC Arbitration Law protects a party "from injustice, not from the consequences of its own decisions" (at [89]). But on these facts, Oheo Bank's inability to address the Crestsign duty case was the kind of "real unfairness or real practical injustice" the provision is designed to catch.

Notably, the classical authorities on an arbitral tribunal's duty to draw an unargued point to the parties (The Vimeira, Zermalt, and X v Y [2020] HKCFI 2782) all concern situations where the tribunal itself has spotted the point. Here, the departure originated with a party, in closing. Popplewell J's summary in Reliance Industries v Union of India [2018] EWHC 822 (Comm) draws only a binary line between having no opportunity and failing to take one. The Court extends the tribunal-duty line by analogy without directly confronting whether the two situations are truly equivalent.

Scope of Submission: A Choice of Common Law Approaches

Article 41(2)(a)(iii) of the DIFC Arbitration Law allows an award to be set aside where it "deals with a dispute not contemplated by or not falling within the terms of the submission to Arbitration." Oheo Bank's parallel argument was that, quite apart from the fairness point, the tribunal had exceeded what the parties had actually put before it.

The Court dismissed the parallel scope-of-submission challenge, but not before choosing between two rival approaches. Singapore, in CAJ v CAI [2021] SGCA 102, treats pleadings as the boundary of what has been submitted to arbitration (at [99]-[101]). Hong Kong, in C1 v IBS [2025] HKCFI 227, treats pleadings as an important starting point but asks, more broadly, what was "in play" across the whole record (at [105]).

The DIFC Court preferred Hong Kong, calling the Singapore approach "unduly rigid" and endorsing the Hong Kong test as "a more practical and surer guide" (at [108(c)]). The reasoning traces back to what arbitration actually is: the parties chose a process built for procedural flexibility, and Article 17.3 of the 2002 Rules of the Dubai International Arbitration Centre gives a tribunal "the discretion to adopt procedures it considers necessary," subject only to the core requirement that each party gets "a reasonable opportunity to present its case" (at [71]). Insisting on formal pleading amendments every time a case develops would place an "impossible burden" on tribunals and create a "trap" for expeditious proceedings (at [91]); doing so would also import the habits of domestic civil procedure into a process not meant to run on them. Although the Court called this discussion "academic" once the fairness ground succeeded (at [120]), it had already framed the analysis as "authoritative guidance" in the absence of prior DIFC appellate authority (at [76]).

Conclusion

None of this alters the established supervisory framework, which the Court set out in four propositions. First, it will not intervene simply to find fault with an award (at [86(a)]). Second, the starting point is minimal interference, in the interests of "arbitral autonomy and finality". Tellingly, the Court preferred a "high threshold" test to the more familiar language of "extreme cases", worried that the latter places too much weight on the label itself rather than the substance (at [86(b)]). Third, intervention depends on the applicant showing "real unfairness or real practical injustice", a standard well short of ordinary appellate review (at [86(c)]). Fourth, whether that standard is met always turns on the facts of the individual case (at [86(d)]). The approach echoes the Federal Court of Australia's insistence, in TCL Air Conditioner (Zhongshan) Co Ltd v Castel Electronics Pty Ltd [2014] FCAFC 83, that courts intervene only "prudently, sparingly and responsibly" (at [81]).

For counsel, the lesson is that unpleaded points are not automatically outside a tribunal's reach, but tribunals attracted to a late-emerging theory must give the parties a real chance to address it. Whether that principle extends comfortably beyond party-initiated pivots is a question for another judgment. Parties can't always get what they plead. But if they try sometimes, they get what they need: not necessarily victory, but a fair opportunity to address the case they have to meet.