Nine cubic metres per second. The figure appears in a 2013 arbitration award as a requirement for water released below a hydroelectric dam. It is not a price placed on a river’s whole value, or a symbol of two countries dividing everything equally. It is an operating limit, with conditions that matter as much as the number.^[3]

The river is called the Kishenganga in India and the Neelum in Pakistan. Diverting water through a hydropower scheme changes what continues down the original channel. Two countries can make separate decisions about projects. The water cannot separate an upstream choice from its downstream consequences in quite the same way.

This was not a place without prior rules. India and Pakistan signed the Indus Waters Treaty in Karachi on 19 September 1960. It entered into force with the exchange of ratifications in January 1961, retroactively from April 1960. The World Bank signed for specified purposes. The later arbitration grew from an existing undertaking, not an improvised invitation to mediate.^[1]

An agreement does not make disputes disappear

Before the treaty, the World Bank participated in prolonged mediation and negotiation. Its archive places the problem after Partition: an established irrigation system met new political borders. Reorganizing political membership did not automatically reorganize the material connections. This was a practical difficulty, not merely a test of whether governments were willing to sound friendly.^[5]

The treaty allocated uses of the eastern and western rivers while allowing specified uses by each country on rivers allocated to the other. It was not an unconditional assertion that every drop belonged solely to one side. The Kishenganga scheme later raised questions about how those permissions and restrictions worked together.^[2]

The project would divert water through a tunnel for power generation and discharge it into another tributary. Pakistan disputed the effects on the original channel and its own downstream hydropower use. One use was changing the conditions of another, rather than simply creating an entirely new benefit waiting to be shared.^[3]

‘It is our project’ does not answer where its effects fall. ‘It affects downstream users’ does not establish that every upstream project must stop. Choose only between those statements and the argument easily becomes a competition to make one swallow the other. An agreed framework can keep the question from restarting with who is entitled to speak.

The treaty provided a Permanent Indus Commission and distinct routes involving a Neutral Expert and arbitration. These are not interchangeable names or a compulsory staircase of appeals for every issue. The kind of question and the treaty’s allocation of functions matter. A technical element does not itself settle which procedure applies.^[2][4]

The World Bank was not a water government above them all. It had helped the negotiations and held specified procedural functions, including certain appointments, rather than authority to determine every power station’s operation. Its 2018 fact sheet is used here to clarify that role, not to project its later disputes back into the 2013 case.^[4]

Pakistan initiated arbitration on 17 May 2010. A seven-member court was constituted under the treaty, with the Permanent Court of Arbitration serving as secretariat. The name did not mean a standing world water court had taken over the river. It was not the International Court of Justice either. The institutional address had to remain precise.^[3]

The 2011 interim order did not stop every aspect of construction. It restricted particular permanent works that could obstruct an effective later award while allowing other work under stated conditions. Preserving choices while a dispute is examined is different from deciding in advance which country will ultimately prevail.^[3]

Permission to divert leaves another question

In February 2013 the Partial Award recognized India’s right to divert water for the scheme under treaty conditions, together with a requirement to maintain a minimum flow. Permission and restriction arrived together. A separate dispute about lowering reservoir levels for sediment management also received its own ruling; one word, victory, cannot accurately describe every issue.^[3]

The minimum was not fixed that February. The parties were asked for further hydrological, generation and environmental material, with opportunities to comment on one another’s submissions. Knowing that water must remain did not entitle the court to pretend it already had enough information to determine exactly how much.^[3]

This step can vanish from a news account organized around winners. Recognizing a requirement and turning it into a workable boundary in a particular place are different tasks. Without the second, concern for downstream interests can remain agreeable language. Without the first, a convenient number can arrive before a reason for choosing it.

The environment supplied no single uncontested answer either. Models, limited local evidence, dry-season effects and power-generation needs entered the comparison. The award acknowledged uncertainty. A model could help explain consequences without becoming a statement that the river, through experts, had finished saying everything relevant to the decision.^[3]

On 20 December 2013, the Final Award specified nine cubic metres per second. Where the daily average inflow immediately above the dam met or exceeded nine, the minimum release below it was nine. Where that upstream daily average was lower, all of that inflow had to be released down the original channel.^[3]

The rule did not require water to be manufactured when less arrived. Nor did low inflow permit everything to be retained. It was not a single standard for every reach of the Indus or every river elsewhere. Evidence about a location and project cannot silently become a universally appropriate environmental prescription.

The court did not let environmental considerations cancel express treaty rights to generation, or let generation cancel environmental needs. A reader can dispute its balance while first identifying its authority: which agreement it interpreted, which rights it had to consider and which policymaking role it did not claim to possess.

This was not a scientific discovery automatically becoming a legal command. Evidence helped assess possible consequences; the treaty established rights and obligations that had to be considered together. The resulting figure included reasoned judgment. Hiding that judgment behind ‘the experts calculated it’ would conceal rather than explain the decision.

For the project, the number defined part of how it could operate. For a reader, it also demonstrates that permission need not spread without limit. A challenge need not end at ‘stop everything’: it can become a question about location, retained flow and verification. That is a more specific disagreement, not a smaller one.

Seven years later, another look may be needed

The award preserved a further route. Beginning seven years after diversion for power generation started, either party could seek reconsideration of the minimum through the Commission and treaty mechanisms. The clock did not simply start with the award in 2013. Reaching seven years would not automatically make the requirement expire.^[3]

The reason is ordinary enough. Actual river conditions, evidence and uses can change. An award’s existence should not indefinitely extend reasoning that no longer fits the river. Stability matters, but it cannot mean declaring all future evidence irrelevant before it arrives. Reconsideration protects attention to reality as well as an opportunity to object.

Binding does not mean incapable of any later reconsideration. The accepted arbitral arrangement made the decision more than advice to follow if convenient. Seeking review through a specified route does not treat the original award as nonexistent. The ability to ask again within agreed boundaries differs from unilateral permission to disregard them.^[2][3]

Pakistan also requested a special monitoring regime. The court pointed instead to the Permanent Indus Commission and existing treaty obligations for information and inspection. Verifying a release could not rest on trust alone, but the award did not thereby assume unlimited power to create new supervisory institutions.^[3]

The treaty already provided exchanges of gauge and flow data and reservoir withdrawals and releases, alongside inspection arrangements. Pointing to those provisions is not proof that every later day complied with them. The existence of a verification obligation, a tribunal’s reliance on it and verified subsequent performance remain distinct claims requiring distinct evidence.^[2]

Now separate the countries at the table from people beside the river. Pakistan brought the arbitration against India. Residents affected by the project did not thereby gain individual party status in that interstate case for their own relocation or damages demands. Two country names on an award cannot write the conclusion to every personal claim.

This limit must not be overstated. Absence of individual standing in this arbitration does not prove that every domestic administrative or judicial route was absent. Assessing a household’s compensation, relocation or livelihood would require further records. The essay invents neither composite villagers’ experiences nor receipts showing compensation that it has not verified.

An interstate procedure can therefore do important work without completing the personal layer. It placed limits on diversion and admitted evidence and reasons into adjudication. It was not a hearing every affected resident could initiate personally, and it did not convert every consequence for a life into a matter already resolved.

Neither half needs to erase the other. If only solving everything counts as cooperation, the boundary becomes worthless. If a national award counts as solving everything, people are compressed into country names. We can acknowledge both an effective agreed constraint and the fact that it has not answered for every person.

This history stops at the 2013 awards. Later operation, any request for reconsideration, and subsequent disputes about the treaty or procedures require a separate body of evidence. Participation in that arbitration cannot settle every question today. Later controversy does not erase the institutional work accomplished then either.

Read beside the shrimp dispute, it supplies no identical answer. One case concerns the implementation of a conservation aim; the other concerns a specific limit on a permitted use. Powers and access differ. The shared question is whether an apparently justified purpose has been allowed to stand in for the consequences someone else actually bears.

Nine cubic metres per second is not proof that the river’s difficulties ended. It is a specific boundary with conditions for another look. We can still ask whether releases can be checked, who can respond to changed conditions and how those without a national seat can keep their lives from becoming scenery behind the award.