The absence of an end date is a quiet but important issue. In research, timelines aren’t just administrative—they’re part of how you measure whether a project has done what it set out to do. Without a fixed end point or at least interim milestones, there’s no natural moment to stop, take stock, and decide if the original questions have been answered.

Australia treats this as evidence that JARPA II is less about specific research goals and more about keeping whaling going under a scientific label. That’s a strong claim, but the open-ended design does make it harder to separate scientific intent from policy motives.

The Court’s response is measured: it doesn’t accuse Japan of bad faith, but it does point out that a proper research programme should have a defined timeframe and checkpoints. That’s not just good scientific practice—it’s a safeguard against a project drifting on indefinitely without clear justification.

From a legal perspective, this matters because open-endedness weakens the connection between the design and the treaty’s object and purpose. A state could, in theory, meet the formal requirements of Article VIII while running a “research” programme that never actually answers its own questions. Setting time frames is one of the simplest ways to make sure that doesn’t happen.

This is essentially the Court pointing out a mismatch between what Japan planned to take and what it actually caught. If the target numbers were set because that’s what was scientifically necessary, then regularly falling short should have been a serious problem for the research design. But the programme continued largely unchanged, which makes those targets look less like strict scientific requirements and more like upper estimates.

Japan attributes the shortfall to operational issues—things like weather, ice, and protestor activity—while Australia sees it as proof the targets were overstated from the start. Either way, it raises a reasonable question: if your research can continue with smaller catches, how firm was the original claim that the higher numbers were essential?

From a legal point of view, this affects the strength of an Article VIII defence. The further your actual practice drifts from your stated “necessary” numbers, the harder it is to argue that those numbers were grounded in genuine research needs.

But it flags two problems:

The Research Plan never explains why a six-year period was chosen for minke whales.

Japan gave inconsistent explanations during the proceedings for that choice.

This points to a gap in transparency and consistency—both important for showing that the lethal sampling design genuinely matches the programme’s stated scientific goals under Article VIII. If a choice is genuinely based on science, there should be a clear record of the reasoning—especially when that choice is used to justify killing hundreds of whales a year. Instead, we get silence in the Research Plan and shifting explanations in the courtroom. That doesn’t look like careful research planning; it looks like someone working backwards from a number they already wanted.

Legally, this kind of inconsistency makes it hard to prove good faith. Article VIII gives states some leeway in designing their research, but that discretion isn’t a free pass—it comes with an expectation of transparency and consistency. Without those, the Court is left wondering whether the six-year period was about science at all, or whether it was simply the easiest fit for pre-set quotas and political convenience.

This is the Court saying, “We’re not here to tell scientists how to do science—but we are here to check whether your numbers actually make sense for what you say you’re trying to do.” It’s an important balance: they avoid the trap of becoming armchair marine biologists, while still holding Japan to a reasonableness test.

The problem for Japan is that this standard cuts right to the heart of their Article VIII defence. If you set a target of 850 minke whales a year, you need to show how that number logically follows from your research objectives—not just that it’s what you’ve always done, or that it’s convenient for your vessels. And “reasonable basis” here isn’t some fluffy diplomatic phrase; it means the Court will ask if the numbers match the aims without obvious inflation.

From a legal point of view, this is a smart move by the Court—it sidesteps accusations of second-guessing scientific expertise, but still demands that states show their working. The risk, though, is that “reasonable basis” is not a term grounded explicitly in the ICRW or past IWC resolutions, so it lives in a grey zone between legal interpretation and judicial policy-making. Japan could push back that the ICJ is importing a review standard that the treaty never agreed on.

Still, from a common-sense angle, it’s hard to fault the approach: if you can’t draw a straight line from your stated research goals to your lethal catch numbers, maybe those numbers aren’t really about the research. And in this case, the Court clearly suspects that’s exactly what’s going on.

Japan’s own paperwork says you need 131 whales a year to hit your accuracy target—but then they only plan to take 50. That’s not a rounding error; that’s less than half. And when asked why, Japan didn’t explain whether they were okay with fuzzier results or if they quietly changed what they were trying to measure. Either way, it’s a big hole in the research logic.

From a scientific perspective, it’s like saying you’re running a medical trial that needs 1,000 patients for reliable results, but you’ll settle for 400 without telling anyone what that does to your findings. From a legal perspective, it’s worse—it undermines the claim that the lethal take is strictly for science. If your own numbers say the plan won’t get you there, then why do it at all?

The Court doesn’t spell it out this bluntly, but the implication is clear: either the sample size was set for reasons other than the science, or the science was so badly compromised in design that it can’t justify the lethal take under Article VIII. And without tying the decision to state practice or external scientific guidelines, Japan leaves itself open to the charge that “scientific research” here is just the label on the jar, not the contents inside.

This is one of those points where the explanation sounds neat on paper but feels off the moment you think about it. Japan calls the twelve-year period a “precautionary approach” because it avoids the big spike in lethal catches that a six-year plan would require. But in environmental law, “precaution” usually means doing less harm, not quietly spreading it over a longer period. Here, it’s almost flipped on its head—stretching out the killing so it’s less visible year to year, but still adding up over time.

It also has the handy side effect of lowering the pressure to show results any time soon. A six-year plan would force more immediate scrutiny; a twelve-year plan buys breathing space. That might be convenient for a whaling programme under international criticism, but it doesn’t scream scientific necessity.

If the Court had really dug in, it could have asked: is this about what’s best for the science, or what’s easiest politically? And if you’re going to claim “precaution,” it should line up with how that principle works elsewhere—like in the CBD or the Rio Declaration—where it’s about stopping potential damage, not extending the timetable for it.

Instead, the Court pointed out the weak science behind the twelve-year choice, but let the “precautionary” label stand. That leaves the door open for other states to dress up strategic choices as conservation-minded planning. And honestly, once you see it this way, the whole twelve-year period feels less like a careful scientific design and more like a slow-burn version of the same problem.

From a legal standpoint, this is where good faith under Article 26 of the VCLT should bite hardest: the aggregation must be a genuine reconciliation of scientific needs, not a convenient way to align with predetermined catch ambitions. The Court was right to probe this, but it could have been more explicit in showing how this step interacts with the necessity principle implicit in Article VIII and visible in analogous cases like Pulp Mills, where the ICJ scrutinised whether claimed environmental measures were truly necessary for the stated goal.

The problem is that without clear subsequent practice from the IWC on how aggregated sample sizes should be calculated, or without systemic integration (Art. 31(3)(c)) with other marine science protocols that favour conservative sampling, the Court’s critique risks looking like second-guessing Japan’s science rather than applying an agreed international standard. In short, this step is the black box of the JARPA II methodology—opaque enough to allow manipulation, and yet framed as purely technical, which makes it one of the most legally delicate parts of the Court’s factual assessment.

Legally, the problem is that Article VIII does not require any international vetting of these methodological calculations, leaving the state free to design a formula that justifies almost any number. The Court’s willingness to scrutinise the application of these variables is a positive move toward an objective reasonableness test, but it lacks grounding in explicit ICRW text or in clearly articulated subsequent practice under VCLT Article 31(3)(b). Without such grounding, Japan could claim that the ICJ overstepped by substituting its own scientific judgment for that of the state.

A stronger approach would have been to couple this methodological review with reference to IWC-endorsed scientific guidelines, or to link it to analogous practices in international fisheries science, framing the critique as consistent with systemic integration (Art. 31(3)(c)) rather than as judicial second-guessing. As it stands, the Court’s scepticism is well-founded but vulnerable to attack as a policy-based intrusion into scientific discretion—especially since the treaty leaves sampling calculations entirely to national authorities.

The Court is on solid ground factually—Japan’s expansion of lethal sampling, especially to species like humpbacks that are otherwise fully protected, invites suspicion. From a legal standpoint, however, the Court is effectively reading a “least harmful means” principle into Article VIII that isn’t spelled out in the text. That’s a big interpretive step, and while defensible under Article 31(1) and good faith (Art. 26), it would have been stronger if tied explicitly to evolving ICRW practice favouring non-lethal research or to relevant environmental obligations under UNCLOS or the Convention on Biological Diversity (Art. 31(3)(c) systemic integration).

Right now, the reasoning feels intuitive—why kill more if you can learn the same with fewer deaths?—but also exposed to attack: Japan could argue the Court is imposing modern conservation preferences onto a 1946 treaty without clear state consent. The leap from “scientific research” to “non-lethal preference” is a policy move dressed as interpretation, and while it works for JARPA II, it risks being challenged as judicial activism in future disputes.

Japan claimed that Article VIII leaves method choice entirely to the state. The Court rejects this, making discretion reviewable for reasonableness. This transforms what could have been a self-judging clause into one with a judicially enforceable standard.

Opinion: This is one of the judgment’s most important guardrails against abuse—without it, “scientific research” could be a free pass for commercial whaling. But the Court underbuilt its foundation. Good faith (Art. 26) and object/purpose (Art. 31(1)) are fine, but they’re fragile alone. Imagine if the Court had grounded “reasonableness” in systemic integration under Art. 31(3)(c), drawing in UNCLOS’s Article 192 duty to protect the marine environment or the CBD’s sustainable use principle. That would have embedded the standard in a network of environmental norms, making it much harder for a future state to dilute it. As it stands, the Court’s stance is bold but vulnerable—like installing a lock on the door but forgetting the deadbolt.

While proportionality here feels like common sense, why kill hundreds more whales if it doesn’t produce correspondingly greater scientific returns? The Court is undeniably expanding Article VIII’s reach. The analogy to Pulp Mills is persuasive: there, too, the ICJ assessed whether harmful environmental conduct was necessary for the stated treaty purpose. But Pulp Mills dealt with obligations that explicitly required due diligence and environmental protection; Article VIII, by contrast, is drafted as a broad state discretion clause. The proportionality filter keeps Article VIII from becoming a blank cheque for disguised commercial whaling, but the Court never grounded it in shared state practice (VCLT 31(3)(b)) or supplementary materials (Art. 32) that might have turned this from a bold policy-driven reading into a firmly rooted interpretive one. As it stands, the Court’s proportionality test works well to catch JARPA II, but risks criticism for being a principled outcome in search of a textual hook.

The Court’s refusal to adopt Australia’s four-part test for “scientific research” was a deliberate choice to keep the term adaptable, relying on VCLT Article 31’s ordinary meaning, context, and object-and-purpose balance. It echoed the Namibia and Navigational and Related Rights approach, recognizing that treaty terms in environmental regimes must evolve with science and policy. That flexibility was especially relevant here, because JARPA II’s methods and stated aims needed assessment in a shifting conservation landscape.

But adaptability came at a cost. By avoiding even minimal guardrails from state practice (Article 31(3)(b)), related environmental treaties (Article 31(3)(c)), or supplementary means under Article 32, the Court left “scientific research” vulnerable to subjective self-definition by states. In Kasikili/Sedudu Island, the ICJ showed it can draw boundaries without making them rigid; a similar approach here, perhaps informed by Australia’s criteria that could have anchored the analysis while still leaving room for scientific innovation. Instead, when the Court later applied its loose standard to JARPA II, it had to rely on a highly fact-specific reasonableness test, which worked for this case but risks inconsistency in the future.

The Court describes the procedural framework for issuing special permits and notes that states have discretion in deciding whether to grant them. Here, VCLT Article 26 (pacta sunt servanda) intersects with the interpretive issue: discretion must still be exercised in good faith, in accordance with the treaty’s object and purpose. The Court hints that this discretion is reviewable, which aligns with Article 31(1) and 31(3)(b) when considering state practice on permit issuance.

New Zealand says that article VIII of the convention should have a strict interpretation, leaning in the favour of conservation and management of whale stock. Reading Article VIII without the bigger purpose in mind would extinguish the very reason the treaty exists. Its interpretation should flow from that shared commitment to conservation, so that every provision works toward protecting, not depleting, what it pledges to preserve. Vienna Convention on the Law of Treaties, Article 31(1) requires interpretation in light of a treaty’s object and purpose and its context, and Article 31(3)(c) further mandates that “any relevant rules of international law applicable between the parties” be taken into account. The case of Gabcíkovo Nagymaros Project (Hungary/Slovakia), ICJ (1997) also held that a provision or a clause of a treaty should be interpretated in the light of its goals and purpose.

This early framing is crucial because it effectively confines the interpretive exercise to the ICRW’s own text and context, setting aside broader environmental or biodiversity obligations that might otherwise be relevant. While this is procedurally standard in ICJ practice—resolving jurisdiction before merits—the choice of a narrow dispute definition has substantive implications. Under the Vienna Convention on the Law of Treaties, Article 31(1) requires interpretation in light of a treaty’s object and purpose and its context, and Article 31(3)(c) further mandates that “any relevant rules of international law applicable between the parties” be taken into account. By not explicitly incorporating other applicable treaties—such as the Convention on Biological Diversity or UNCLOS—into this initial framing, the Court arguably limits the interpretive scope from the outset, potentially overlooking systemic integration principles envisaged by Article 31(3)(c).