Consensus as Currency
Who governs science when no one is in charge: Part 3 of a five-part series
Editor’s Note: This weekly series explores how science is governed globally in the absence of a world authority, beginning with the myth and reality of coordination without control.
At a one-day session in Geneva last month, delegates to the United Nations process negotiating a global plastics treaty elected Chilean diplomat Julio Cordano as chair of the Intergovernmental Negotiating Committee after months of paralysis. The vote was held by secret ballot because consensus could not be reached.
Cordano prevailed in a second round, restoring leadership to a process that had stalled without agreed text. The session addressed governance without reopening treaty provisions on production caps, chemicals, or financing.
Addressing delegates after the vote, Cordano framed the crisis in universal terms. “Plastic pollution is a planetary problem that affects everyone: every country, every community and every individual,” he said. “If we don’t take concerted action, it will get much worse in the coming decades.”
The election kept the negotiations alive but did not resolve the divisions.
As David Azoulay of the Center for International Environmental Law put it, “While electing a chair keeps the process alive, it won’t fix what’s broken in and of itself.”
The episode illustrated a central feature of global science diplomacy. Most multilateral negotiations are designed to operate by consensus. But when consensus falters, the system has limited tools to prevent collapse. Voting becomes an exception — a procedural instrument used to preserve continuity when substantive agreement remains out of reach.
This is the tension at the heart of international science governance. Institutions without enforcement depend on agreement and the disciplined management of disagreement.
Three negotiations illuminate how this works: the plastics treaty process, the U.N. climate talks, and the pandemic agreement adopted at the World Health Assembly. Each relies on consensus. Each reveals different ways in which consensus holds, strains, or is engineered into being.
The Climate Summit That Ended in a Name
At the close of COP28 in Dubai in December 2023, the conference president, Sultan Al Jaber, framed the final package in collective terms. “It is a balanced plan that tackles emissions, bridges the gap on adaptation, reimagines global finance, and delivers on loss and damage,” he declared, calling it “the U.A.E. Consensus.”
The label did more than describe the text. It signaled that the outcome had cleared the procedural threshold that matters most in U.N. climate diplomacy: no delegation was willing to block it.
Under the U.N. Framework Convention on Climate Change, parties “shall make every effort to reach agreement … by consensus.” Voting is described as a last resort. In practice, consensus has become the operating norm. A single determined delegation can prevent adoption.
That architecture has preserved universality. Nearly every country remains at the table. It has also slowed decision-making.
When the scientific case for action is strong, as it has been for years, the disputes that remain are distributive: who moves first, who pays, what counts as fair. Consensus channels those disagreements into language.
Text is drafted so that governments with incompatible red lines can refrain from objecting. Timelines become conditional. Verbs soften. Phrases accommodate multiple readings. The same sentence can be celebrated as historic and criticized as insufficient because it carries different meanings for different audiences.
At climate summits, the gavel is a signal that no one will stop the outcome.
Plastics and the Visible Limits of Agreement
The plastics negotiations made the fragility of that signal visible.
Delegates had been directed to produce a legally binding instrument that addresses plastic pollution. When talks stalled without consensus on a text, the absence of agreement became the defining fact of the session.
In Geneva, the election of a new chair by secret ballot demonstrated both the flexibility and the limits of consensus systems. Voting preserved the process.
The chair’s role in such moments is partly rhetorical: to maintain the possibility of common ground. Cordano told delegates, “We are not far, either in time or in position, from a successful outcome. But no one in this room can achieve it alone. This must be a joint effort, with national delegations in the driving seat.”
The language was familiar to multilateral diplomacy. It emphasized shared responsibility while acknowledging divergence. Yet the fact that the chair himself was chosen by vote underscored the structural constraint: when consensus fails, there is no higher authority to impose a solution.
Deferral becomes the only option: another session, another round, another attempt to render disagreement non-blocking.
The Pandemic Agreement and the Art of Not Blocking
The pandemic agreement adopted at last year’s World Health Assembly offers a contrasting case.
Member countries voted overwhelmingly in favor. There were no votes against. Eleven countries abstained. The vote occurred because Slovakia requested it, breaking the traditional global consensus typically used to adopt international health instruments.
In consensus systems, abstention performs a specific function. It allows governments to register reservations without preventing adoption; it is a collective decision not to block.
The architecture of the agreement reinforced that logic. WHO emphasized a sovereignty clause stating: “Nothing in the WHO pandemic agreement shall be interpreted as providing the secretariat of the World Health Organization, including the director-general … any authority to direct, order, alter or otherwise prescribe” national laws or policies, or to “mandate or otherwise impose any requirements.”
Such language narrows the scope of objection, making the agreement adoptable.
WHO Director-General Tedros Adhanom Ghebreyesus framed the outcome in collective terms: “Governments from all over the world are making their countries, and our interconnected global community, more equitable, healthier and safer from the threats posed by pathogens and viruses of pandemic potential.”
Read together, these elements reveal how consensus is managed: affirmative claims about shared interest paired with explicit limitations on institutional authority.
Consensus as an Operating System
From the outside, consensus can appear as a mood, either cooperative or obstructive. In practice, it functions more like an operating system.
First, it creates usable outputs. Text adopted by consensus becomes a common reference point. Even when interpretations diverge, governments, industries, and institutions can cite the same baseline.
Second, it reduces the cost of participation. In systems governed by majority voting, countries with limited capacity risk being routinely outvoted. Consensus offers a guarantee that no state will be formally overruled.
Third, it channels power into procedure. Influence accrues to those who remain in the room, draft language, and shape what becomes acceptable. Legal expertise, technical depth, and negotiating stamina matter.
These dynamics were visible across the three cases. Climate negotiations preserved universality through compromise language. Plastics negotiations revealed how consensus can be used to block. The pandemic agreement demonstrated how abstentions, sovereignty assurances, and staged implementation can produce adoption without resolving every dispute.
Why the Currency is Harder to Spend
Consensus is becoming more difficult to achieve not because diplomats have changed temperament, but because the disputes themselves have shifted.
Science diplomacy increasingly touches distributional questions: intellectual property, industrial policy, supply chains, strategic technology, and financial transfers. These are more than technical disagreements. They implicate economic and geopolitical interests.
Consensus can still produce texts but cannot compel a convergence of incentives. In this environment, much of the work occurs at the margins: narrowing scope, sequencing obligations, deferring annexes, inserting clarifications that reduce perceived overreach.
The pandemic agreement illustrates this technique. It will fully enter into force only after a further annex on pathogen access and benefit-sharing is negotiated and ratified. Adoption was possible because the most contentious distributional question was structurally separated.
Consensus was preserved by staging conflict.
Looking Ahead
Consensus is the procedural substitute for authority in institutions that lack enforcement power.
It can function as a brake, as climate talks demonstrate. It can fail visibly, as in the plastics negotiations. It can also be engineered into an outcome through abstentions, limitation clauses, and careful drafting, as the pandemic agreement shows.
The question for science diplomacy is whether consensus systems can produce shared reference points quickly and credibly enough to matter.
Part 4 turns to the moments when they cannot: when scientific change outruns institutional capacity, and agreement becomes too slow, too narrow, or too contested to hold.
Next: Part 4 — When Science Outruns Institutions