When silence isn’t always golden: A rare successful challenge under Section 68

July 07, 2026

No doubt you'll all be familiar with the expression "Silence is golden", reflecting the idea that there are occasions when it is wiser to say nothing at all. Perhaps the more, shall we say, mature readers among you will also remember Silence Is Golden, the 1967 hit by The Tremeloes. While silence may sometimes be golden, it can prove decidedly less so in arbitration. Earlier this year, the Commercial Court demonstrated exactly that in Indus Powertech Inc. v. Echjay Industries Private Limited [2026] EWHC 827 (Comm). In one of the few successful challenges under section 68 of the Arbitration Act 1996, the Court held that an arbitral tribunal's silence on two material issues of causation amounted to a serious irregularity that caused substantial injustice.

Given the English courts' long-established pro-arbitration approach and their reluctance to interfere with arbitral awards, I think the decision is noteworthy and, more importantly, it provides useful guidance on the distinction between a tribunal making an error in its reasoning and failing altogether to determine an issue that was squarely before it.

The background in brief 

The dispute arose out of a Master Supply Agreement under which Echjay Industries Private Limited (“Echjay”) manufactured forged components for Indus Powertech Inc (“Indus”). The agreement contained a non-compete provision preventing Indus from sourcing certain components from competing suppliers during the term of the agreement. When the parties' commercial relationship broke down, Echjay commenced ICC arbitration in London, alleging that Indus had breached the non-compete provision by sourcing components from another Indian manufacturer. The tribunal agreed that Indus was in breach and, while limiting the recoverable damages period to two years, awarded Echjay approximately US$4.1 million in lost profits.

The challenge under Section 68

Indus did not challenge the tribunal's findings on liability, but instead the award under section 68(2)(d) of the Arbitration Act 1996, arguing that the tribunal had failed to determine two key issues of causation that went to the heart of the damages calculation:

• whether Echjay actually had sufficient manufacturing capacity to produce the relevant shafts had the alleged breach not occurred; and
• whether the necessary design and production validation processes could have been completed within the two-year damages period adopted by the Tribunal, such that Echjay would have been able to manufacture the gears within that timeframe.

The Commercial Court's Decision

Sean O'Sullivan KC sitting as a Deputy High Court Judge emphasised that section 68 establishes a deliberately high threshold. The provision is designed to address procedural injustice rather than errors of fact or law. Nevertheless, the Court concluded that the tribunal had failed to determine two genuine issues that had been fully argued by the parties. The Court also drew an important distinction between: (i) a tribunal giving inadequate or unpersuasive reasons; and (ii) a tribunal failing altogether to determine an issue. Only the latter engages section 68(2)(d). 

The judgment also distilled a number of useful principles from the authorities when considering whether a tribunal has actually dealt with an issue:

1. Parties should not be left to guess whether a material issue has been determined or simply overlooked.
2. If the award clearly demonstrates that the tribunal reached a conclusion on an issue, that is the end of the matter.
3. Where the position is less clear, the Court must look for indications that the tribunal engaged with the relevant evidence and submissions, reading the award as a whole.
4. It is not enough to infer that an issue must have been decided simply because particular relief was granted.

Applying those principles, the Court concluded that, even giving the tribunal "the benefit of every reasonable doubt" the two issues had simply not been determined. Serious irregularity means more than identifying an error; Section 68 also requires the applicant to establish substantial injustice, and this is often the most difficult hurdle to overcome.

Here, however, the unanswered causation issues had the potential to reduce the damages award significantly. The omission therefore mattered. Rather than setting the award aside, the Court directed that the outstanding issues, together with consequential questions relating to interest and costs, be determined by the original tribunal.

Practical lessons for arbitrators

The decision contains several practical reminders for tribunals, including:

• Know what you have to decide. Distinguish carefully between evidence, arguments and the material issues requiring determination.
• Use the award as a checklist. Before finalising an award, ask whether every pleaded issue has been addressed, even if only briefly.
• Don't confuse brevity with completeness. Tribunals are not expected to answer every submission advanced by counsel, but they must determine every issue that matters.
• Remember that silence has consequences. A carefully reasoned award may still be vulnerable if it leaves a material issue unresolved.

Practical lessons for parties

The case also contains some useful reminders for parties and their advisers:

• Frame the issues clearly. Well-structured pleadings, agreed lists of issues and focused closing submissions make it easier for the tribunal to identify what requires determination.
• Don't rely on post-award remedies. The scope for correcting an unfavourable outcome after the event is extremely limited not only in arbitration, but equally in adjudication (see Amy Bonczyk’s recent blog: “The slip rule is not a second bite at the cherry”).
• Know when section 68 may assist. If a tribunal has genuinely failed to determine a material issue, section 68 may provide a remedy.
• But remember its limits. Disagreeing with a tribunal's reasoning or factual conclusions will almost never be enough and the omission must relate to an issue, not simply an argument.

Could AI have helped I wonder?

One aspect of the judgment particularly caught my attention. The Court expressed … “The Tribunal certainly has my sympathy and my respect for their efforts” in navigating extensive evidential records and written submissions running to well over 100 pages. As many of us know, modern commercial arbitrations are rarely straightforward and tribunals are expected to digest thousands of pages of documents, multiple rounds of submissions, witness statements, expert reports and numerous alternative arguments. Although the reason why the tribunal overlooked the two issues was irrelevant for the purposes of section 68, the case nevertheless raises an interesting question: could AI have helped?

I have previously blogged about the potential roles of AI in decision-making (see Decision-making: Is it as easy as AI, B, C? ) and Indus v Echjay provides another practical example of where technology might have made a difference.  AI can help tribunals organise issues, cross-reference pleadings against witness and expert evidence, and check whether every pleaded issue appears to have been addressed in a draft award if used appropriately. In that sense, it would simply be acting as a sophisticated quality-control tool, helping to reduce the risk that a material issue slips through the cracks. Ultimately, however, accountability remains with the tribunal. AI cannot determine which issues are legally material, weigh competing evidence or exercise judgment and those responsibilities should remain entirely human.

Final thoughts

I don't think Indus v Echjay signals any greater judicial interference in arbitration. In fact, I think it’s the opposite as it reinforces the English courts' commitment to the finality of arbitration while confirming that there remains a narrow, but important safeguard where procedural fairness has broken down. Perhaps the old saying therefore needs a slightly less catchy qualification: “Silence is golden, but in arbitration, silence on a material issue can prove very costly indeed”.