When the Regulator Becomes the Respondent: The NORI and TOML Orders and the Reviewability Gap in Article 189 UNCLOS

[Tajra Smajic is a PhD candidate at the Chinese University of Hong Kong, specializing in the governance of deep seabed mining.]

On 18 July 2026, the Seabed Disputes Chamber (SDC) issued orders on requests for provisional measures in two parallel proceedings brought by Nauru Ocean Resources (NORI) (Case 34) and Tonga Offshore Mining Ltd (TOML) (Case 35). These are the first contentious cases ever brought before the SDC under Part XI, but their significance extends beyond their novelty. By inviting the SDC to review the conduct of the ISA, the proceedings place the institutional relationship between the two under scrutiny and the line between judicial accountability and institutional autonomy within Part XI of the UNCLOS framework thereby testing the “practical administration of the institutional architecture” established by UNCLOS and the 1994 Agreement (Declaration of Judge Kittichaisaree, para2).

The dispute and its background have already been examined from several angles (see Kang, Virgil,Lily, Pecoraro and Singh and the three Vinson & Elkins pieces (Part 1, Part 2 and Part 3)). What that coverage has not yet examined thoroughly is the reviewability gap in Article 189 UNCLOS which these cases bring to the forefront: the under-defined line between the procedural illegality the SDC may review and the regulatory discretion it may not.

Both companies are wholly-owned subsidiaries of The Metals Company (TMC), a Canadian deep-sea mining firm, and hold fifteen-year exploration contracts with the ISA for polymetallic nodules in the Clarion-Clipperton Zone. When the proceedings were initiated, both companies had a pending application to extend their contracts, with NORI’s exploration contract weeks from expiry. While NORI’s extension has since been granted, the dispute between the two contractors and ISA remains.

The applications challenge the procedural conduct of the ISA’s Legal and Technical Commission (LTC) in a compliance inquiry into their alleged “possible non-compliance” with their contracts and no determination on jurisdiction or the merits has been reached.

The ISA and the SDC are products of a single constitutional compromise embodied in Part XI of UNCLOS, and were established as distinct yet complementary institutions to facilitate the implementation of the common heritage of humankind principle. While the SDC may exercise jurisdiction over the ISA, this is limited to claims regarding whether the application of any rules, regulations and procedures of the ISA in individual cases would be in conflict with the contractual obligations of the parties to the dispute or their obligations under UNCLOS, and claims concerning excess of jurisdiction or misuse of power (Article 189 UNCLOS). UNCLOS explicitly sets out that the SDC has no jurisdiction with regard to the exercise by the ISA of its discretionary powers and that it cannot substitute its discretion for that of the ISA (Article 189 UNCLOS). While some may read Article 189 as subordinating the SDC to the ISA, the Chamber itself describes the provision as balancing the preservation of the Authority’s discretionary powers against judicial review to protect the rights of entities involved in activities in the Area (Case 34, para 129; Case 35, para 113).

The limitation of SDC jurisdiction is architecturally significant: it reflects the UNCLOS drafters’ recognition that the ISA’s trusteeship mandate requires regulatory autonomy that the SDC should not override. It also allows for oversight and a remedy for situations in which the ISA would misuse its power and act contrary to its obligations towards contractors.

Drawing the Line Between Procedure and Discretion

In the provisional measures orders, the SDC had to examine whether it had prima facie jurisdiction, whether the rights asserted are plausible, and whether those rights face a real and imminent risk of irreparable prejudice. Article 189 was considered at this stage only briefly, as part of the assessment of prima facie jurisdiction, and the SDC will return to the question in greater detail at the merits stage. Nonetheless, the approach the SDC adopted warrants several important observations (Case 34, paras 74-133; Case 35, paras 61-117).

The first concerns the reach of the SDC’s jurisdiction itself. Article 187(c) confines that jurisdiction to disputes with respect to activities in the Area, and the orders hold that “[a]ctivities in the Area entail substantive as well as procedural rights for the contractor”, among them the right to due process in any inquiry into possible non-compliance (Case 34, para 95; Case 35, para 82). The ISA had argued for a narrower reading, relying on Article 1(3) UNCLOS, “all activities of exploration for, and exploitation of, the resources of the Area”, to submit that complaints about a compliance inquiry concern the exercise of its regulatory powers rather than activities as defined (Case 34, paras 93-94; Case 35, paras 80-81). The interpretation adopted moves beyond what the 2011 Advisory Opinion had settled, where the term was construed along an operational axis, the question being which physical operations it covers: the recovery of minerals from the seabed and their lifting to the surface, together with directly connected operations, but not processing or transport to land (2011 Advisory Opinion, paras 82-97). The orders add a different dimension, not which operations fall within the term, but what legal entitlements attach to a contractor engaged in them. The procedural conduct of regulation is thereby brought within the SDC’s subject matter, and with it the question of how far the widening extends: whether any procedural grievance a contractor may have against the ISA now concerns activities in the Area.

The second observation of significant importance is how the SDC characterised the dispute before it. It found that a disagreement existed and classified it as one of procedural compliance, concerning “the question of observance of due process by the [ISA]” in its inquiry (Case 34, para 84; Case 35, para 71). Whether Article 189 stood in the way of that jurisdiction turned on precisely this characterisation. The provision limits the SDC in three respects: it withholds jurisdiction over the ISA’s exercise of its discretionary powers, forbids the SDC to substitute its own discretion for the ISA’s, and bars any ruling on whether the ISA’s rules, regulations and procedures conform to UNCLOS. None of these is a free-standing prohibition. Each is defined by reference to what a dispute is about, so whether any of them applies depends on how the subject matter of the dispute is identified. A disagreement about whether a procedure was observed is not, on its face, a disagreement about how a power was exercised. Once the dispute was described as one of procedural compliance, none of the three limitations had anything to attach to. The description adopted by the SDC follows the applicants’: they had put their case as concerning “the applicable requirements of due process” and expressly not “as to whether the Applicants are in breach” of their contracts (Case 34, para 79; Case 35, para 66).

What the SDC did not say at this point deserves as much attention as what it did. Alongside those limitations, Article 189 preserves three categories of claim: that the application of the ISA’s rules in an individual case conflicts with the parties’ contractual or UNCLOS obligations; excess of jurisdiction or misuse of power; and claims for damages or other remedy. The applicants invoked a claim that the ISA’s conduct was in conflict with contractual and Convention obligations, while the ISA had argued that none of the three was applicable here (Case 34, para 116; Case 35, para 102). The SDC set the categories out and then said nothing further about them (Case 34, paras 116, 126, 128; Case 35, paras 102, 110, 112). The silence is consequential. Had the SDC reasoned through one of the preserved categories, review would have been available only on the conditions those categories impose: excess of jurisdiction, or misuse of power, or a conflict between the application of a rule and a contractual obligation. By treating the observance of due process as falling outside the limitations altogether, the SDC made those conditions unnecessary. Of the two routes open to it, this is much the broader, and it was taken without being explained. Whether that route survives is a question for the merits, where Article 189 will be argued in full rather than at the prima facie threshold. 

That leaves the question the characterisation only postpones: why a dispute about procedure does not touch the exercise of discretion at all. The ISA’s answer was that it plainly does, the conduct of an inquiry being “by its very essence, a discretionary power” and the drafters having spared the SDC “the burden of micromanaging” the ISA’s decisions (Case 34, paras 122, 124; Case 35, paras 107, 109). The SDC accepted the premise and denied the inference: it would not pronounce on the merits of the ISA’s exercise of discretion, still less substitute its own, but the observance of due process, “being required by the relevant legal framework, is extraneous to the exercise by the [ISA] of its discretionary powers” (Case 34, paras 131-132; Case 35, paras 115-116).

The distinction is between a power and the conditions under which it must be exercised. The ISA remains free to decide whether to open an inquiry, whom to question, and what to conclude; what it is not free to do is depart from the procedures governing how those decisions are reached. This is a limited conception of oversight, and its limits come from the source of the procedures. Every requirement the SDC enforced was drawn from instruments the ISA had itself adopted, the Council’s mandate that the inquiry ensure due process “at every stage”, the LTC’s identification criteria, the procedures for extensions (Case 34, paras 96, 163-170; Case 35, paras 83, 146-153). The SDC did not supply a standard of its own but held the ISA to commitments it placed upon itself. That is a defensible position in a regime designed to protect regulatory autonomy. It depends, however, on being able to say which norms belong to the procedural framework and which to the discretion that framework governs, and the orders do not say. “Extraneous” announces the separation without explaining it, and the parties disagreed about precisely where it falls, to the point of disputing whether ISBA/29/LTC/5 applies at all, a document they could not agree how to name (Case 34, para 55; Case 35, para 42).

If procedure is reviewable and discretion is not, the remaining question is when review becomes available. The orders answer it by treating the injury as one of timing rather than of loss: due process “must be exercised timely in the decision-making process”, and the loss of the opportunity to participate “cannot ordinarily be remedied ex post facto” (Case 34, para 197; Case 35, para 174). A procedural right, on this reasoning, is injured in the formative stages of a process or not at all. The absence of a final decision is what makes the moment matter; it is not what makes the claim premature.

This is potentially the most far-reaching of the SDC’s holdings, and the one that gives the objections raised before the orders their sharpest application. If a procedural right can only be vindicated while the process is still forming, review must come early, and the threshold at which the ISA’s regulatory conduct becomes contestable falls correspondingly. That was the objection Lily, Pecoraro and Singh pressed before the orders, describing the applications as an attempt to short-circuit an inquiry rather than to vindicate due process, on the grounds that the challenge is to an unfinished inquiry rather than to an adverse decision, that the LTC is an advisory organ with decisions reserved to the Council, and that the remedies sought would draw the SDC into managing the ISA’s internal processes. The orders answer part of this, though in the remedy rather than in the reasoning: the inquiry was not suspended, its outputs were not embargoed, and the ISA was required only to act within its own framework, to clarify its procedures and the questions put, and to cooperate and refrain from aggravating the dispute. Other parts remain unanswered, the advisory character of the LTC’s outputs, and the ISA’s submission that plausibility requires colourable evidence of a violation and not merely of a right (Case 34, paras 149-150; Case 35, paras 132-133). What weighs on the other side is institutional design rather than doctrine: the same LTC conducts the inquiry and, for NORI, assesses the extension application, an overlap Kang identifies as a risk of procedural contamination. Where the formation of views and the assessment of their consequences lie in the same hands, review at the formative stage is not premature; it is the only point at which it can do any work.

Taken together, these holdings describe a relationship rather than a division of territory. An institution entrusted with organising and controlling activities in the Area on behalf of humankind is required to observe the procedural standards it has itself adopted, and the SDC supplies the means by which that requirement is enforced. Accountability of this kind strengthens the Part XI framework rather than eroding it, which is why the orders can describe the provisional measures power as serving not only the parties but “the effective implementation of the common heritage of humankind regime” (Case 34, para 60; Case 35, para 47) a point Judge Kittichaisaree put directly in observing that such oversight strengthens the ISA’s own institutional legitimacy (Declaration, para 3). The limit is equally clear: were the inquiry to conclude that a contractor has failed to comply, the SDC could not substitute its own assessment of that conclusion. Whether the ISA judged the matter rightly is exactly what Article 189 places beyond review.

Each holding, however, leaves a question. If the applicable limitation depends on how a dispute is characterised, what prevents a complaint about substance from being pleaded as one about procedure? If review extends only to the procedures the ISA has itself adopted, does its reach expand and contract with the Council’s legislative output, and does it not incentivise the ISA to promise less? If a procedural right is injured while a process is still forming, at what stage does an ongoing regulatory process cease to be open to challenge? And if the categories of claim Article 189 preserves were unnecessary to found jurisdiction here, what work is left for them to do? These are not questions a provisional measures order was required to answer. They are the questions the merits might have to.