Haste Makes Waste – Issue Discipline in Construction Arbitration

Construction cases can be tricky: the record is often voluminous, the disputes - modular, and the evidence - diverse in both type and substance. In my experience, the skill that matters most to fair and expeditious resolution of a typical construction case is issue discipline: identifying the crop of questions that truly decide the case – the trees within the wood, holding them in view as the factual undergrowth thickens, and then demonstrating in the award that those questions have been understood and answered.

That discipline is central to enforceability, and therefore central to the legitimacy of commercial arbitration as a dispute resolution method in the world of construction and engineering. Construction disputes can generate seemingly plausible side roads running in every direction. If the tribunal takes too many of them, it risks arriving at an outcome that is procedurally vulnerable, not because it is necessarily wrong on the merits, but because it does not visibly engage with what mattered most.

There is also a broader context: the current climate is unmistakably pro efficiency. The Arbitration Act 2025 (UK), for instance, confirms, among other things, an express power for summary disposal. In parallel, the profession is concretely debating the responsible use of new technology in arbitration, and guidance has emerged aimed at preserving due process and enforceability.

In my experience, that combination of compressed procedure and ever larger written records only increases the premium on issue discipline. An arbitrator can move quickly and still be careful. But the arbitrator cannot be careful if precisely what is to be decided is unknown.

It is tempting to think that new (AI) tools will solve the problem by making it easier to digest the record. They will not. They may accelerate retrieval, summarisation and cross-referencing, but they cannot perform the essentially normative task that keeps an award safe: deciding what the live issues are, policing when a party has shifted its case, and ensuring that any dispositive line of reasoning has been squarely put to both sides before it becomes determinative. Those are matters of judgment and fairness. I suspect they will remain, for the foreseeable future, impervious to both analytical and generative AI.

The following crop of recent decisions in construction and infrastructure disputes bring this into sharper focus. Each shows a different way in which the essential issue can slip out of the arbitrator’s grasp, or at least out of the arbitrator’s visible reasoning. Each suggests, in its own way, that a firmer grip on the material issues might have produced a better result, in the only practical sense that matters: a result that would have avoided (and if not withstood) challenge.

When the Map Matters Most – Wan Sern Metal Industries

Wan Sern Metal Industries Pte Ltd v Hua Tian Engineering Pte Ltd concerns an expedited documents-only arbitration arising out of a construction subcontract. The Singaporean Court of Appeal noted the remarkable pace of the procedure and the absence of an oral hearing, and it observed that the accelerated process resulted in a lack of clarity as to the parties’ positions which the arbitrator failed to appreciate and resolve (see para. 5).

The point of fracture was the respondent’s counterclaim. The pleaded case was directed to payment for work completed to date, yet in written submissions the respondent introduced a materially different basis of recovery, calculated by reference to the value of both completed and uncompleted work, which later measure the arbitrator adopted. The Court of Appeal held that the arbitrator breached natural justice by failing to apply her mind to the parties’ cases on that critical issue, and it set aside the award insofar as it concerned the affected counterclaim, declining remission (see paras. 46, 48, 58).

For arbitrators and counsel, the lesson is partly about what a documents-only procedure removes:   routine and informal moments when misunderstandings surface in a hearing room and are cured by a question from the tribunal. In that context, pleadings and the agreed issues list become the sole working map. Issue discipline here is the tribunal doing a small, early piece of procedural work: noticing that a new measure or new claim has emerged, asking whether it is being pursued formally, giving the other side a clear opportunity to meet it, and recording briefly why the point is properly within scope.

Scott Schedules and the Risk of Drift – DOM v DON

DOM v DON, another Singaporean case, arose out of construction works performed, and a schedule-driven defects and quantum dispute emerging, under the SIA industry form. The High Court articulated two somewhat familiar routes to a fair hearing breach – namely, whether (a) there had been a failure to apply the mind to essential issues, and a defect in the chain of reasoning where what the tribunal did was not reasonably foreseeable, or (b) a lack of a sufficient nexus to the parties’ arguments (see para. 77).

What stands out, from a construction arbitration perspective, is the court’s analysis of certain consultants’ fees. In that respect, the tribunal had expressed clear doubts about whether particular consultants’ services were necessary for the rectification works, yet had still awarded a percentage of those fees by applying a broad reduction for insufficient evidence. The court set aside discrete parts of the award in relation to those fees (see paras. 94-96).

This is a familiar hazard in construction arbitration. Scott Schedules and similar mechanisms are indispensable for managing complexity. Yet, they cannot overcome the risk of fragmentation. When the dispute comprises a package of dozens, hundreds, possibly thousands, of distinct items, it is inherently harder to avoid the decision-making succumbing to the drift from legal entitlement into arithmetic compromise. Issue discipline in that context must be rigorously mechanical. The tribunal ought to keep two questions separate and visible: first, is the cost recoverable in principle; second, if it is, how should it be quantified. If the tribunal regards a category of cost as unnecessary, it either rejects it or explains clearly why, notwithstanding that characterisation, some portion remains recoverable on a coherent basis. ²

The “Unpleaded” Route – Mare Nova v Zhangjiagang Jiushun

The English Commercial Court’s decision in Mare Nova Incorporated v Zhangjiagang Jiushun Ship Engineering Co., Ltd offers a recent illustration of how a case can go off the rails when the decisive issue is not properly identified as an issue at all. The claimant’s challenge under section 68 of the Arbitration Act 1996 (UK) for ‘serious irregularity’ succeeded because the tribunal decided the damages claim by reference to a contractual construction point which had not been raised in the arbitration, depriving the claimant of a fair opportunity to address it.

The dispute arose under a ship repair contract. Shortly after the vessel left the yard, damage was discovered and the owners commenced arbitration seeking damages for breach of contractual obligations and negligence (with an alternative claim under a guarantee clause). As with Wan Sern Metal Industries, the arbitration was conducted on paper, without a hearing. Also, the respondent did not participate beyond a jurisdiction objection. The tribunal found the yard had caused the damage, but dismissed the main damages claim on the basis that (on the proper construction of contractual provisions addressing quality/acceptance and the duration of the yard’s liability) liability was discharged when the vessel left the shipyard. The difficulty was that this “discharge on departure” construction was not an issue that either party had put forward.

The court treated this as a classic serious irregularity: the tribunal had, in substance, decided the case on a point not argued, so the claimant had no opportunity to meet it. That was enough for section 68. The court also considered (for the purposes of assessing substantial injustice) that the tribunal’s construction was “obviously wrong in law”, but it remained careful to present the decision as a due process correction, not a backdoor appeal. The award was remitted to the tribunal for reconsideration, on the footing that the reconsideration must proceed on the basis of the law as set out by the court.

The link back to issue definition is straightforward. Where a tribunal is minded to decide a point of law or contractual construction which the parties have not squarely advanced, it must align that point with an identifiable live issue before it relies on the point in the determinative reasoning. In practice that means:

1. issuing a short procedural direction that frames the point as a question, and inviting short written submissions on it;
2. in paper-only cases especially, maintaining an agreed list of “questions to be answered” (updated as the tribunal’s own concerns develop), because there is no hearing at which surprises are easily flushed out; and
3. in the award, signposting the issues addressed so it is obvious that the dispositive point was within the scope of what the parties were asked to meet.

Mare Nova is a reminder that procedural fairness can be undermined as much by an unheralded legal route as by ignoring a pleaded one.

Why Perhaps Construction Arbitration Needs This Discipline More Than Most

The common thread across these decisions is not hostility to efficiency but rather that efficiency without issue discipline is a false economy.

Construction arbitrations need this discipline more than many other sectoral type settings for commercial arbitration because the disputes are structurally prone to sprawl. A single case can contain questions on contract interpretation, delay and disruption, defects, valuation, and expert methodologies, all at once. That sprawl is not better grappled with by more documentation (to the contrary!) but rather by the earlier and more rigorous identification of the decisive questions.

If I try to reduce the discipline into three good habits, they are the following:

1. Name the decisive questions early, and keep returning to them;
2. Treat shifts in case as procedural events, especially shifts in the basis or measure of recovery, and regularise them;
3. Make the linkage visible, particularly between findings on entitlement and the quantified outcome.

None of these require excessive convolution. But they require better orientation and significant professional collaboration.

A Personal Closing Reflection

The best construction tribunals behave like able engineers – identifying what is genuinely load bearing, thus resisting the temptation to treat every contested point as equally important. Crucially, well-versed counsel assist by simplifying, narrowing, and conceding where appropriate, so that the tribunal’s attention is spent on decisive questions.

It is this preoccupation which sits behind my new book, Dispositive Issues in Construction and Engineering Arbitration – my attempt to convert (so far) three decades of practice entirely in construction arbitration into a practical account of how to identify the few issues that truly decide construction and engineering disputes, and how to organise procedure, evidence and award writing around them so that the result is both efficient and resilient. The three decisions above serve as stark reminders that when issue discipline is lost, the price is often paid after the award is handed down, and in the wrong forum.

* The author has not acted, and does not act, for any party in the cases discussed.

For further reading, the author’s new book, Dispositive Issues in Construction and Engineering Arbitration, is now available via Kluwer Arbitration and the Wolters Kluwer eStore.