Kaleidoscopy

Sensemaking for a plural world

Tension Thread

Who Gets to Decide?

Thirty-seven debates, one question. When new domains emerge with no settled authority, or old authorities get contested, the fight is rarely about facts. It's about legitimacy — which body has the right to make the call, and what makes that right real.

The European Union AI Act came into force in 2024 as the world's first comprehensive legal framework for artificial intelligence. Any company wanting to operate in Europe must comply — which in practice means every major AI developer in the world. The systems were built mostly in California. The training data was sourced globally. The communities most affected by algorithmic decisions span every continent. But it was Brussels that wrote the rules, because Brussels got there first and had the regulatory machinery to try.

No one elected the EU to govern global AI. The companies building the systems didn't ask for this jurisdiction. The users in Lagos and Manila and Jakarta whose lives will be shaped by these systems didn't participate in drafting it. And yet the Act is real, and compliance is not optional if you want access to 450 million European consumers. The EU asserted authority because the alternative was a governance vacuum — and vacuums don't stay empty.

That logic — someone has to decide, and if no one else will, we will — is running through thirty-seven debates on this site. The EU's AI jurisdiction is the explicit version. But the same scramble for legitimate authority animates debates about who governs the high seas, who controls the curriculum in public schools, who can override a state's pandemic response, and who holds authority over a pregnant person's body. In each case, the question isn't just what the right answer is. It's who has the standing to give one.

Three logics of legitimate authority

When people argue about who should decide something, they are almost always drawing on one of three underlying claims. Each captures something real about what makes authority legitimate. Each also fails in predictable ways.

The first is democratic mandate: authority is legitimate when it flows from the consent of those governed. The elected legislature, the popular referendum, the school board voted in by parents — these bodies have the right to decide because the people most affected chose them. This logic grounds everything from vaccine mandate debates (should the state have authority over what citizens put in their bodies?) to curriculum battles (should locally elected boards set educational content, or the state legislature, or the federal government?). Democratic mandate has enormous moral force. Its weakness is the question of which democracy. Federal authority and state authority are both democratically derived, but they can point in opposite directions. International bodies like the WHO or the ISA are formally democratic — built by national governments — but lack the direct accountability relationship that makes democratic authority feel legitimate to ordinary people. When the Supreme Court returned abortion authority to states in Dobbs v. Jackson (2022), it did so in the name of democratic legitimacy. The result was that the same decision became legal in one state and a criminal matter in the neighboring one.

The second logic is expertise-based stewardship: some decisions require specialized knowledge that democratic majorities don't have, and legitimate authority belongs to whoever can most accurately model the consequences. The case for the FDA deciding which drugs are safe is expertise logic. The case for the WHO coordinating pandemic response is expertise logic. The case for independent central banks setting interest rates rather than elected politicians is expertise logic. It doesn't require that experts be wise about everything — only that the specific decision domain is technical enough that uninformed majorities are more likely to get it catastrophically wrong than informed specialists. Its failure mode is the question of accountability: experts answer to their institutions and their fields, not to the people affected by their decisions. When the WHO's pandemic guidance proved partially incorrect in 2020, no democratic mechanism existed to hold it accountable. When nuclear regulators approve a waste site in a community that didn't choose the proximity, expertise logic provides no remedy. Stewardship without accountability generates resentment even when — sometimes especially when — the experts are right.

The third logic is proximity sovereignty: authority should rest with whoever lives most directly within the consequences of a decision. The pregnant person has authority over their body because no one else inhabits it. Indigenous nations have authority over their land because no one else has lived within its ecology for generations. The local community has authority over its school because they are the ones raising children in it. This logic is the grammar of bodily autonomy, indigenous sovereignty, and local control — it insists that legitimate authority is grounded in real stakes, not formal jurisdiction. Its failure mode is scale: some decisions have consequences that radiate far beyond the immediate proximate party. A landowner's water use affects downstream users who aren't proximate to the decision. A country's carbon emissions affect communities that had no voice in the industrial choices that produced them. Proximity logic works well when consequences are local; it struggles when they aren't, which is exactly when authority is hardest to establish.

When the three logics collide

Most of the thirty-seven debates in this thread are fights between these logics, usually with each side unable to see past the logic it finds most compelling. Pandemic governance pitched expertise logic (epidemiologists coordinating through WHO) against democratic mandate (state governors claiming authority over their populations) against proximity sovereignty (individuals claiming authority over their own bodies). Each position was internally coherent. The collision was not because one side was wrong about the facts — it was because they were using different theories of legitimacy that each captured something real.

AI governance is the same collision in slow motion. Expertise logic says that technically sophisticated oversight bodies should set the rules — people who understand alignment and inference and training data have the best chance of getting this right. Democratic mandate logic says that decisions about AI's role in society are too consequential to delegate to unelected technocrats, and elected legislatures should write the laws even if their members don't fully understand transformers. Proximity sovereignty says the communities most harmed by algorithmic discrimination — in hiring, in bail decisions, in credit — should have the most say over how these systems are constrained, not the companies building them or the governments funding them.

What makes these collisions genuinely hard is that each logic is protecting something that matters. Democratic mandate protects popular agency and accountability. Expertise stewardship protects against the particular danger of uninformed majorities making decisions that compound complex harms. Proximity sovereignty protects those with the most at stake from being overridden by those with the least. The problem is that in most real governance situations, you cannot maximize all three simultaneously.

The new commons problem

Several of the most contested debates in this thread involve what might be called new commons — domains where no existing authority has clear jurisdiction because the domain either didn't exist or was inaccessible when existing frameworks were built. The 1967 Outer Space Treaty and the 1959 Antarctic Treaty were designed for national space agencies and scientific expeditions; they are now governing commercial satellite constellations and potential mineral extraction in a world neither treaty anticipated. The 1982 UN Convention on the Law of the Sea created the International Seabed Authority to regulate deep-ocean mining under a "common heritage of mankind" principle that has been tested almost immediately by commercial pressure.

Digital commons present the same problem in a different form. The internet spans every jurisdiction while being incorporated in none. A platform headquartered in California makes speech decisions that affect elections in Brazil. An AI company in San Francisco trains on data from users who are subject to European privacy law. The mismatch between where decisions are made and where consequences fall is not a glitch in the system — it is the system. And it means that any legitimate authority claim will be under-inclusive: whoever governs will govern imperfectly, because no existing institution maps onto the actual shape of the domain.

The new commons problem generates a particular kind of authority vacuum — not a gap that more governance would fill, but a structural mismatch between the scale of the problem and the scale of any available institution. This is why debates about ocean governance, space governance, AI governance, and global health governance feel different from debates about abortion or curriculum: in those cases, there is a plausible institution to empower; in the new commons debates, every available institution is the wrong size or the wrong shape.

What the thread reveals

Reading across these thirty-seven maps, a clarifying distinction emerges between two kinds of authority disputes. The first is a dispute about which existing institution should hold authority — federal versus state, WHO versus national government, school board versus state legislature. In these disputes, there is a plausible answer, even if it's contested. The democratic mandate logic, or the proximity logic, can be applied carefully to reach a position. The second kind is a dispute about whether any existing institution has legitimate authority at all — over AI, over the deep ocean, over the predictive algorithm running inside a bail decision. In these disputes, the question is not which institution wins but whether legitimacy can be constructed in real time, under pressure, with incomplete knowledge.

The maps also reveal something about what makes authority claims accepted in practice, not just valid in theory. Authority requires three things to actually function: the right process (was this decided through legitimate procedures?), the right institution (was this the body with appropriate standing?), and acceptance by those governed. The third condition is the fragile one. The WHO's expertise was real during COVID-19. Its procedural legitimacy was largely intact. But its authority collapsed in many places because the acceptance condition failed — large numbers of people refused to treat it as the body with the right to tell them what to do. That refusal wasn't only irrational. It reflected a genuine gap: the WHO answers to member states, not to populations, and in a moment of high stakes and real uncertainty, the accountability gap became visible.

This is the harder version of the question. Not "who should decide?" but "what would make a decision actually feel legitimate to the people living inside it?" The two questions often have different answers. The institution that should decide is not always the institution that can build acceptance. The process that is most technically correct is not always the one that keeps people inside the framework. Most of the hardest debates in this thread are stuck at exactly this gap — between formal authority and felt legitimacy. Seeing that gap clearly is the beginning of being able to reason about how to close it.

Maps in this thread

Thirty-eight perspective maps tracing a single question — about democratic mandate, expert stewardship, and proximity sovereignty — through AI, digital governance, global commons, and contested personal authority.

  • AI Governance No existing institution has clear authority — the race is between national governments, international bodies, and the companies themselves.
  • AI Safety and Existential Risk Who should regulate transformative AI — and whether existing democratic institutions have the capacity to do it before it matters.
  • Platform Accountability and Content Moderation Private companies make consequential speech decisions for billions of people — under what mandate, and with what accountability?
  • Platform Moderation and Free Expression Values Competing philosophical frameworks — Mill, Waldron, Fish, Nyabola — each imply different institutional authorities over speech.
  • Digital Privacy and Surveillance Personal data flows through state, corporate, and individual control claims that are fundamentally incompatible.
  • Digital Identity and Biometrics Who decides who exists in a bureaucratic world — and what happens when the infrastructure that delivers belonging also enables tracking.
  • Surveillance Capitalism Whether corporations have the right to monetize behavioral data — and whether consent is meaningful when the alternative is opting out of the economy.
  • Social Media and Democracy Can private platforms subvert democratic processes, and who has authority to regulate them in ways that don't simply serve power?
  • AI and Democracy AI-generated content in electoral contexts — what oversight exists, and who is empowered to enforce it?
  • Global Health Governance WHO authority gaps, IHR compliance, and pandemic accord negotiations — when sovereign states and international bodies conflict over public health.
  • Space Governance and the Outer Space Treaty Property rights, resource extraction, orbital commons — a 1967 treaty governing what's become a commercial frontier.
  • Antarctic Governance Overlapping sovereignty claims, international stewardship, and increasing resource pressure on a continent frozen in a 1959 framework.
  • Ocean Governance and the High Seas The high seas belong to everyone and no one — governing them requires authority that no single state or institution holds.
  • Deep-Sea Mining The ISA governs the deep seabed on behalf of all humanity — while collecting royalties from the contractors it regulates; whether that structure can apply genuine precaution is the central governance question.
  • Nuclear Security and Nonproliferation Who has the right to nuclear technology, and who gets to enforce that line — the NPT's authority has always depended on an asymmetry most parties resent.
  • Bioweapons Governance A categorical prohibition with no verification mechanism — the Biological Weapons Convention relies entirely on self-certification by the parties most likely to defect, and the 2001 collapse of verification negotiations left the governance architecture thinner than for any other weapon of mass destruction.
  • Electoral Reform and Ranked Choice Voting How votes are counted shapes who can win — the debate is about which counting method best represents what voters actually want.
  • Vaccine Mandates The state claiming authority over what goes into bodies — where the line between public health authority and bodily autonomy falls.
  • Abortion Whether the state, the pregnant person, medical providers, or religious doctrine holds authority over this decision — and what each claim is protecting.
  • Reproductive Technology and IVF Creating and selecting embryos raises questions about who has authority over the beginning of life — families, clinics, the state, or religious ethics.
  • Free Speech on Campus University administrators, student bodies, courts, and legislatures each claim authority to set the terms of campus speech.
  • Education and Curriculum Whose history, whose literature, whose framework — curriculum battles are ultimately about who has authority to define shared knowledge.
  • Education and School Choice Parents, states, districts, and teachers unions each claim authority over where children go to school and how public funds follow them.
  • Compulsory Schooling and Educational Freedom Who has legitimate authority over children's formation — the state, the parent, or the child? The debate over compulsory attendance is really a debate about where parental rights end and public interest begins.
  • Early Childhood Development Policy Who is responsible for children's earliest years — the state, the family, or the market? The pre-K debate is really about who has authority to define what children need and who is obligated to provide it.
  • Childhood and Technology Parents, platforms, and states all claim authority over children's digital lives — the conflict is over whose protective instinct governs.
  • Parenting Whether child-rearing is a private matter of parental authority or a domain where state and community have legitimate claims.
  • Housing Supply and Zoning Reform Local community control over neighborhood character vs. state authority to override it in service of regional housing needs.
  • Land Ownership The philosophical grounding of property rights — Lockean labor-mixing, indigenous relationality, or state-granted title — each implies different legitimate authority.
  • Groundwater Governance Underground water crosses property lines and state boundaries — jurisdiction is ambiguous, and what's invisible is easy to deplete.
  • Prosecutorial Discretion Prosecutors hold enormous, largely unchecked power over charging decisions — the debate is about whether that discretion is a necessary feature or an unaccountable bug.
  • Predictive Policing and Surveillance Technology Algorithmic systems guiding law enforcement decisions — who audits the algorithm, and what accountability exists for its predictions?
  • Eminent Domain and Regulatory Takings Who defines "public use" and "just compensation" — courts, legislatures, or markets?
  • Algorithmic Hiring and Fairness Who decides what "qualified" means when an automated system processes your application before any human sees it — and who has standing to contest that decision?
  • Gig Economy and Worker Classification Whether courts, legislatures, or markets should determine who counts as an employee — a governance question with enormous material consequences for tens of millions of workers.
  • Charter Cities and Special Economic Zones When private entities are authorized to govern territory with their own courts and laws, the question of who legitimized that authority — and who can revoke it — becomes the whole debate.
  • Solar Geoengineering Who decides what temperature the planet should be? The governance void around stratospheric aerosol injection is the "who gets to decide" question at civilizational scale.
  • Geoengineering Governance Whether multilateral treaty, scientific self-governance, or emergency unilateralism should hold the key to planetary-scale interventions — and who gets to decide which framework applies.
  • Synthetic Biology and Gene Editing Gene drives spread through wild populations without anyone's consent; germline edits are inherited by all future descendants. Governance frameworks designed for individual clinical decisions are badly matched to technologies with population-level consequences.
  • Journalism and Media Trust Who decides what counts as credible news — and who loses standing in public discourse when the answer changes. The structural collapse of local journalism removes the watchdog from municipal government precisely where it is hardest to replace.
  • Psychedelic Medicine and Therapy Who holds the keys to psilocybin, MDMA, and other psychedelics as they move from Schedule I to clinical trials — whether the FDA approval model, medical gatekeeping, or decriminalization is the right governance frame for substances with contested therapeutic and ceremonial histories.
  • Global Trade and Industrial Policy Who sets the rules of the global trading system — and whether those rules protect open markets, developing-country policy space, or rich-country industrial strategy, depending on who's doing the deciding.
  • Energy Democracy and Utility Ownership Whether the grid should answer to shareholder-owned utilities, member boards, municipal publics, regulators, or competitive market rules is a governance fight disguised as infrastructure policy.
  • Social Trust and Institutional Legitimacy Why trust in democratic institutions collapsed across the wealthy world — and whether the fix is accountability for institutional failure, rebuilding social capital, better institutional design, or accepting that skepticism of concentrated expertise is healthy democracy.
  • Corporate Governance and the Purpose of the Firm Who governs the corporation — shareholders, managers, workers, or democratic publics — is the question underneath every debate about corporate social responsibility, ESG, and the political power of concentrated private institutions.
  • Big Tech and Antitrust Whether the companies that control digital infrastructure should be broken apart, constrained, required to open their systems to rivals, or changed in ownership — and whether any court or regulator has the tools to move fast enough to matter.
  • Foreign Aid and International Development Whether developing countries should be recipients of donor decisions or authors of their own development strategy — and who gets to define what "development" means — runs through every debate about conditionality, sovereignty, and structural adjustment.
  • Sovereign Debt and Austerity The IMF, creditor governments, and international courts decide what conditions a debtor country must accept — but the populations that live under those conditions had no seat at the table when the loans were made or when the terms were set.
  • Humanitarian Intervention and R2P When a state commits or fails to prevent mass atrocities, who has the authority to intervene — and who decides that the threshold has been crossed? The Security Council veto ensures that the answer depends on great-power alignment, not humanitarian need.
  • Gaza Aid Access and Ceasefire Bargaining When aid access itself remains part of ceasefire bargaining, the question is not only what should enter Gaza but who gets to decide when civilian relief becomes independent enough from military leverage to count as real protection.

References and further reading

  • Robert A. Dahl, Democracy and Its Critics (Yale University Press, 1989) — still one of the clearest accounts of democratic legitimacy, pluralism, and the boundary problem of who counts as the relevant public when authority is supposed to flow from "the people."
  • Allen Buchanan and Robert O. Keohane, "The Legitimacy of Global Governance Institutions", Ethics & International Affairs 20, no. 4 (2006) — a sharp guide to the problem this thread keeps hitting in AI, oceans, and global health: institutions can be necessary even when their accountability and representation are visibly incomplete.
  • Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge University Press, 1990) — essential for thinking past the false choice between pure state control and pure market allocation. Ostrom is especially useful here because many authority disputes are really fights over scale, jurisdiction, and whether governance can be polycentric instead of singular.
  • James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (Yale University Press, 1998) — a durable warning about what happens when centralized expertise treats local knowledge as noise. It helps explain why technically competent governance can still fail the legitimacy test for the people living inside its consequences.
  • World Health Organization, Constitution of the World Health Organization — the formal basis for one of the thread's recurring examples. Useful not because it resolves the WHO legitimacy question, but because it shows exactly what kind of mandate the institution does and does not have.
  • Dobbs v. Jackson Women's Health Organization, 597 U.S. ___ (2022) — the recent case this essay points to when democratic mandate and proximity sovereignty collide. Reading the opinion directly is clarifying because it makes visible how a claim about institutional authority can radically redistribute who gets to decide.
  • Algorithmic Governance and Automated Decisions — one of Ripple's clearest maps of what happens when public institutions, private vendors, and opaque models all exercise authority without a clean theory of standing or accountability.
  • Global Health Governance — a concrete extension of the WHO example in this thread: what authority a transnational health body can claim, what states retain, and why technical coordination does not automatically become felt legitimacy.
  • Indigenous Land Rights — a strong companion map for the proximity-sovereignty side of the argument. It shows why formal jurisdiction can feel morally thin when communities with durable ties to land are treated as stakeholders rather than governing peoples.