Floating Cities

Floating cities means permanent human settlement on the water surface: dwellings, streets and services carried on buoyant structures that are moored rather than founded, and that rise and fall with the sea instead of resisting it. It spans three traditions that are usually blended and should not be: the engineering tradition of very large floating structures, a field three decades deep with a substantial literature; the political tradition of seasteading, which holds that the ocean surface offers an escape from the fully-claimed land surface and therefore a venue for institutional competition; and the adaptation tradition of floating neighbourhoods built inside national planning systems as a response to flood risk and land scarcity.

Scope boundary. Working the seabed and the water column — mining, thermal energy, alkalinity — belongs to ocean engineering. Keeping the sea off existing cities belongs to coastal defense systems. Moving people away from the water belongs to climate migration planning. Dense settlement as an architectural programme belongs to arcologies and underground cities. This brief is about the surface, and about what it takes to hold a position on it.

The thesis, stated at the top because the rest of the brief is its defence. The binding constraint on floating settlement is neither hydrodynamic nor financial-in-the-ordinary-sense. It is the absence of a specific legal instrument: a registrable, transferable, mortgageable interest in a mooring berth that survives the insolvency of the platform operator. A dwelling nobody can mortgage is a dwelling almost nobody can buy, and a berth held as a contract right against a single counterparty is not collateral. Everything else here — wave loads, class rules, insurance, appraisal comparables, the secondary market — is downstream of that one missing artefact, and the artefact is draftable. That is the unusual feature of this topic: the impossible version is impossible for reasons a competent legislative draftsman could address in a session.

Provenance warning, stated once and meant. The research pack behind this brief obtained zero full texts on this subject and read no primary legal instrument. Where it describes the law of the sea it gives the substance and names the instrument, and deliberately does not reproduce article numbers it cannot source. Where a famous number is unverified, the number is not printed.

Speculative Start with the seasteading case in its strongest form, because a brief that dismisses it is less useful than one that says exactly what stops it. Institutions improve under competition; competition requires exit; the land surface is fully partitioned among states, so leaving one jurisdiction means entering another on its terms; the ocean is not partitioned the same way; therefore a floating polity would be the first genuinely new entrant to the market for governance since the closing of the frontier. Speculative That is a serious political-economy position with real antecedents, and it depends on nothing exotic in the physics. Established The Seasteading Institute, founded in the late 2000s with libertarian venture funding, is a documented organisation with a documented programme. Frontier The sharpest counter is the critical-geography treatment published as Atlas Swam: the vision misreads the ocean as a frictionless outside-of-politics, when it is among the most jealously claimed spaces on Earth. Frontier This brief takes that critique as correct and the political argument as surviving it — the case for jurisdictional competition does not die when the high-seas route dies, it redirects to land.

Established The engineering is the solved part, and it has been solved for a while. The pack behind this brief verified twelve distinct works on very large floating structures spanning 1992 to 2023, concentrated with unusual singleness of purpose on one problem: hydroelasticity. Established A pontoon-type structure of this kind is not a ship: it is long enough relative to the incident wavelength that it deforms elastically rather than moving as a rigid body, so the design problem is a coupled fluid-structure calculation. Established Predicting that deformation and reducing it is what the literature does. Established Those works are cited here for the existence and shape of the literature and nothing else — the pack verified the records and never obtained a text, and this brief will not launder a record-level citation into a number.

Frontier The empirical anchor of the whole field is a Japanese programme from the 1990s, and this brief cannot verify its numbers. The Technological Research Association of Mega-Float, a consortium of Japanese shipbuilders, tested pontoon modules in Tokyo Bay through the late 1990s, culminating in a floating runway on which aircraft take-offs and landings were demonstrated before the structure was dismantled. Frontier The dimensions and dates in wide circulation could not be confirmed and are not stated here as facts; Suzuki’s overview of Megafloat in Marine Structures (2005) is the standard reference and the highest-priority unread source in the subject. Established What survives without the numbers is the point that matters: a large floating platform in a sheltered bay was demonstrated, then abandoned, for reasons that had nothing to do with engineering. The abandonment is the interesting part, not the runway.

Frontier Occupied floating urban development already exists, at neighbourhood scale, and it exists by declining every hard problem in the subject. The Dutch cases — floating homes at IJburg, the Schoonschip neighbourhood, the Rotterdam floating pavilion and office — are the working demonstrations; no household counts are given here because the pack could not verify them. Established What matters is the configuration: these dwellings are moored, connected to shore utilities, inside a municipal planning regime, and financed under national mortgage law in a jurisdiction that already had a legal category for houseboats. Established Every difficulty that defeats seasteading is one the Dutch examples solve by not being at sea and not being outside a state. Frontier That is not a criticism. It is the finding.

Established Now the law, stated as precisely as the sourcing permits and no further. The governing instrument is the United Nations Convention on the Law of the Sea, concluded in 1982. Established Its substance here, as consistently reported across the secondary legal literature, is fourfold. First, no state may validly subject any part of the high seas to its sovereignty; such claims are invalid as such, which is stronger than merely unrecognised. Second, artificial islands, installations and structures do not have the status of islands — an island in the Convention’s sense is a naturally formed area of land above water at high tide — so an artificial structure generates no territorial sea, no exclusive economic zone, and no effect on any delimitation. Third, within its exclusive economic zone the coastal state has the exclusive right to authorise and regulate artificial islands and installations, with exclusive jurisdiction over them, out to two hundred nautical miles, and the equivalent applies on the continental shelf. Fourth, the freedom to construct installations on the high seas exists but is qualified, exercised subject to the Convention and to other rules of international law.

Established This brief does not print the article numbers, and the omission is deliberate. The research pass behind it read no primary legal instrument, and article numbers are exactly the class of citation repeated confidently and wrongly. Established The substance above is robust across the secondary literature; the numbering belongs in an annotated edition. Established One further provision, equally load-bearing and equally unnumbered: everything at sea has a flag, ships have the nationality of the state whose flag they fly, and on the high seas a ship is under the exclusive jurisdiction of that state. Established That yields the seasteading dilemma in one line. If the structure is a ship, it sits inside a state’s exclusive jurisdiction and is not sovereign. If it is not a ship, it falls into a regulatory void — and a void is a liability, not a freedom, because it means no registry, no applicable law, no insurability and no financeability.

Frontier Statehood, separately, has criteria that a floating structure does not meet. The conventional statement is the Montevideo Convention on the Rights and Duties of States (1933): a permanent population, a defined territory, a government, and the capacity to enter into relations with other states. Frontier The orthodox reading of “defined territory” requires territory, and a floating structure is a chattel. Speculative An advocate can fairly answer that recognition is political and could be earned. Established Empirically it never has been.

Frontier The failure record is the most useful evidence set in the subject, and it is uniform. The pack could not verify these cases against primary sources, and this brief reports them as the record commonly given. The Republic of Minerva (1972) pumped sand onto reefs south of Fiji; Tonga asserted a claim and removed the settlement. Isola delle Rose (1968), a platform off Rimini declaring independence from beyond what were then Italian territorial waters, was seized and demolished by Italy. Frontier The Principality of Sealand, an occupied wartime sea fort off the Suffolk coast, is commonly said to have been found outside United Kingdom jurisdiction in 1968 because it lay beyond the then three-nautical-mile territorial sea; the United Kingdom later extended to twelve and enclosed it. Frontier In February 2020 a spar-type single-occupancy seastead placed off Phuket by Ocean Builders was removed by the Thai authorities, who pursued charges. Established Four attempts; three physical removals and one boundary that moved over the top of the claim. Not one failed on waves.

Frontier The most serious attempt abandoned the sovereignty premise entirely and failed anyway. In January 2017 the Seasteading Institute signed a memorandum of understanding with French Polynesia contemplating a floating island project within a special governing zone in sheltered waters; it lapsed in 2018 after local political opposition. Established It is the most informative datum in the topic: the most credible seasteading effort sought a host state, sheltered water and a delegated legal zone — giving up the high seas and the sovereignty claim in one move — and still died on the host state’s domestic politics. Established The binding link was never offshore. It was in a legislature.

Frontier And here is the finding this brief exists to state. A settlement is not bought by a consortium; it is bought dwelling by dwelling, by ordinary households, with long-tenor secured debt — which requires a security interest a lender can perfect and enforce against a defaulting borrower, in a forum whose judgment will be recognised. Established Land mortgages work because three things exist together: a state-maintained land register, territorial jurisdiction, and a foreclosure procedure. Established A floating berth has none of them. Frontier The strongest workaround is maritime. Ship mortgages are registered against a vessel on a state’s register, are internationally recognised, and are enforced by arrest of the vessel in rem, under a body of law that includes an International Convention on Maritime Liens and Mortgages concluded in 1993 — an instrument this brief names but could not verify as to parties or entry into force. Speculative If each dwelling module were a registered vessel under a functioning flag registry, and the city were a mooring-service provider rather than a landlord, a lender could in principle take a first preferred ship mortgage over it. That is the best version of the seasteading financial case.

Established It still fails, and the reason is specific. Ship mortgages are enforced by seizing and selling the vessel — but a dwelling module’s value is overwhelmingly in its berth and network position, not in its steel, so arrest realises the wrong asset and hands the lender a de-mated box worth a fraction of the loan. Established The berth is the actual collateral, and by default it is a contract right against a single platform operator rather than a property right — not bankable at twenty-five-year tenor. Established Ship finance runs at shorter tenor against a depreciating, redeployable asset with a liquid second-hand market; residential mortgage economics require decades against an appreciating asset in a deep market, and modules have neither a second-hand market nor an appraisal standard, because appraisal needs comparable sales and there are none. Established The pack reached this conclusion twice, by routes that share no premises — once forward from mortgage mechanics, once by asking why the Dutch neighbourhoods work, which they do because Dutch law already contains a registrable interest in a mooring. Established The obstacle to a floating city is that no legal system currently contains an instrument by which a person can own a place to float.

Frontier The live research frontier splits cleanly into a hydroelastic half and a jurisdictional half, and only one of them is on the critical path. On the engineering side, four lines are active in the verified record. Wave-energy extraction as response mitigation (Nguyen 2020) solves two problems with one device and has not been consolidated. Frontier Passive perimeter treatments — permeable plates and porous flexible barriers (Singla 2019; Pu 2022) — promise response reduction without the cost and environmental footprint of a full breakwater. Frontier Machine-learning surrogates for hydroelastic response (Tay 2023) matter more than they look: real-time response prediction is a precondition for an operable settlement — occupant comfort management, mooring load management, decisions about when to restrict movement across a deck — not merely for a designable one. Frontier And seabed compliance in the response model (Das 2017) is the shallow-water case, which is the only case that will ever be built.

Frontier The most useful frontier item in this subject is not about floating cities at all: it is floating nuclear power. A 2026 chapter by Fialkoff on floating nuclear power plants, non-proliferation norms and the law of the sea is the clearest signal available that the legal question is being forced by someone else. Established Floating nuclear plants are large, permanently moored, non-ship structures with people on and around them, high consequence, and backed by states and utilities with the resources to compel an answer on applicable law, liability and security. Speculative Floating cities will inherit whatever regime that process establishes. Speculative Even if the regime turns out to be nuclear-specific rather than general, it establishes the precedent that matters most: that a permanently moored non-ship inhabited structure can be legally characterised at all. That is the whole question in miniature.

Frontier A second body of law is being built that a floating settlement will need and that its advocates rarely cite. Barnes on energy sovereignty in marine spaces (2015) records that the offshore energy sector has already litigated most of the jurisdictional questions a settlement raises. Frontier Papanicolopulu on human rights at sea (2023) is where the hardest question gets worked: a permanently inhabited offshore structure creates a population whose rights must attach to some jurisdiction, and if none attaches, the residents are not free, they are unprotected. Frontier Cassar on airspace, sovereignty and ocean governance (2023) raises a constraint almost nobody discusses: an inhabited platform needs air access, and airspace sovereignty is a separate regime from the law of the sea. Frontier Young on systemic integration, the law of the sea and courts (2025) and Golitsyn on the role of the International Tribunal for the Law of the Sea (2018) are the forum question — if a dispute arises about a floating settlement, who hears it.

Frontier Piracy law is included here deliberately, and the reason is uncomfortable. Tuerk’s treatment of piracy and the law of the sea (2013) is the doctrinal home of the question of what happens to a structure or vessel outside state protection. Established A stateless structure on the high seas is not immune from interference; it is subject to it, by anyone, with no flag state to complain on its behalf. Speculative The security position of an unflagged seastead is therefore the exact inverse of the freedom its advocates describe, and the literature that governs it is the literature of piracy.

Frontier The sharpest open question in the field is one nobody is funding: what does a floating settlement cost once the legal and financial overhead is priced? Floating construction is often claimed to beat land reclamation per square metre of usable deck. Speculative That comparison is incomplete to the point of being useless, because reclamation delivers ordinary land with ordinary title, ordinary mortgages and ordinary insurance, and floating delivers a chattel with none of them. Speculative No published study prices that difference. Producing one is the single cheapest useful piece of research in the subject and it requires no hardware.

Frontier Finally, the adaptation branch is running independently and should never be blended with the exit branch. Oceanix Busan, announced as a partnership involving UN-Habitat and the City of Busan, and the Maldives floating-city proposal, are framed as adaptation for coastal and low-lying states. Frontier This brief does not state their dimensions, resident counts or current construction status, because the pack could not verify them and because this class of project has a long record of being announced and re-announced. Established Announcement is not construction. Established What is certain is the structural difference: these projects have a host state, a public sponsor and a development-finance path, and seasteading is defined by the absence of all three. They share hull technology and nothing else.

Established The workback plan for this subject has ten links, and the one that binds is fourth. The target is a permanently inhabited floating settlement financed by ordinary mortgage debt and covered by ordinary insurance. Link 1, siting in sheltered water inside a national jurisdiction, is achieved in principle. Frontier Link 2, a host state granting a defined regulatory regime by enacted instrument rather than memorandum, has never been achieved — French Polynesia failed at exactly this step. Frontier Link 3, legal characterisation of the structure such that some registry accepts the filing, is unachieved but is a known kind of problem.

Established Link 4 is the berth right converted from a contract into a registrable, transferable, mortgageable interest that survives the operator’s insolvency. This has not been achieved anywhere. Established The distinction that carries the whole subject is between a berth licence, which is worthless the moment the operator fails, and a berth interest, which is not. Established A resident holding a licence loses their home when the platform company goes under; a resident holding an interest does not, and only the second is something a bank will lend against for twenty-five years.

Frontier Links 5 and 6 — a class-society notation for a permanently moored inhabited structure, and placed hull and third-party liability insurance — are partly precedented by floating production units and accommodation flotels, and are achievable by known means. Established Link 7, an appraisal standard supported by enough arm’s-length transactions to establish comparables, cannot begin until link 4 exists, because until then there is nothing to transact in. Established Link 8, the first mortgage, is definitionally downstream of link 4. Established Link 9, a secondary market in which interests are assignable and the operator can fail without extinguishing residents’ rights, is the hardest legal design problem in the chain. Speculative Link 10, permanent habitation demonstrated through a design storm and a full design life, is the last and the slowest.

Established One physical bottleneck is not removed by fixing link 4, and it deserves separate statement. Insurance for a moored settlement is catastrophe-correlated in a way that marine insurance is structurally bad at: a fleet is spread across oceans and one storm takes one ship, whereas one storm takes an entire settlement. Established And a cruise ship can run from a storm, which is the single most important loss-mitigation behaviour in marine underwriting; a moored settlement cannot. Established That pushes the whole programme toward sheltered water, which pushes it inside national jurisdiction, which is precisely where the sovereignty premise dies. The insurance constraint and the legal constraint point the same way.

Established This subject depends on results that are mostly not scientific. The first dependency is the least glamorous and the most urgent: somebody has to read the primary instruments. The pack behind this brief read none, and the entire enthusiast literature rests on assertions about treaty text that are rarely checked against it. Established An annotated UNCLOS commentary, the Montevideo text, the 1993 Maritime Liens and Mortgages Convention, and the 1986 Convention on Conditions for Registration of Ships — commonly said never to have entered into force, which if true is itself a finding about the genuine-link requirement — are four documents, and reading them settles more of this subject than any experiment would.

Frontier The second dependency is maritime finance scholarship. The mortgage argument in this brief rests on general knowledge of ship-registry practice and protection-and-indemnity underwriting, and a specialist source would materially strengthen or refute it. Frontier The third is the built record from the Netherlands: occupancy, financing terms, insurance terms and the actual legal form of the mooring interest at Schoonschip and IJburg. Established These are documents held by municipalities and lenders, not experiments; the reason nobody has them is that nobody has asked.

Frontier The fourth is engineering data that only exists at full scale: long-run structural monitoring of a real moored settlement across a real storm season, which no programme currently publishes. Speculative The fifth is the outcome of the floating-nuclear regulatory process, which this subject will inherit and cannot influence. Established Note the shape of that list. Four of the five dependencies are documents and one is a monitoring campaign. None is a discovery. Frontier That is unusual on this map, and it changes what a funder should buy: the marginal dollar in this subject buys legal drafting and archival retrieval, not a test tank. Speculative It also means the subject can move quickly if anyone decides to move it, because none of the prerequisites has a laboratory timescale.

Established The decisive experiments in this subject are legal and administrative, and they are cheap. They can be ordered.

Frontier Experiment 1: the registry filing test. Take an existing Dutch or Scandinavian floating dwelling and attempt, in a cooperative jurisdiction, to register a security interest over both the module and its berth as a single unit. Record precisely which office refuses, and on what statutory ground. Established The output is a list of the exact statutory amendments required, which is what a draftsman needs and what nobody currently has.

Frontier Experiment 2: the insolvency simulation. Commission an opinion, in three jurisdictions, on what happens to residents’ occupancy rights when a floating-neighbourhood operator enters insolvency. Established The measurable output is binary per jurisdiction: does the interest survive, or does the administrator sell the moorings out from under the residents. Speculative The prediction from this brief is that no jurisdiction currently delivers survival without a bespoke structure.

Frontier Experiment 3: the priced comparison. Build the cost model nobody has built — floating versus reclamation versus piled construction for one specific site — with the legal and financial overhead priced explicitly: the mortgage rate premium, the insurance premium, the appraisal discount for absent comparables, and the residual value at end of design life. Established A single site study would settle an argument that has run on assertion for fifty years.

Frontier Experiment 4: instrumented storm survival. Fit an existing floating neighbourhood with mooring-load and hull-strain instrumentation and publish a full storm season. Established This is the missing input to a class notation and to the first insurance quote, and it is inexpensive.

Speculative Experiment 5: the flag-state chapter. Persuade one small maritime-registry state to draft a habitation-structure chapter into its maritime code, with a registrable mortgage regime for permanently moored inhabited units. Speculative This is not a thought experiment — small registry states have both the legislative flexibility and the commercial motive — and it is the most promising concrete path in the entire subject.

Frontier Experiment 6: the residential motion standard. Run an occupant-comfort study on existing floating dwellings across a full range of sea states, measuring accelerations at deck level against reported sleep quality, nausea and willingness to remain. Established Offshore and ship comfort criteria are written for rotating crews of working adults; nothing exists for permanent residents, children and the elderly. Speculative Without that standard, no designer can say what wave climate is habitable, and the siting question stays a matter of assertion.

Established The engineering requirements are demanding, well-characterised, and not the reason this does not exist. The central one is hydroelastic response: a pontoon platform long relative to the incident wavelength deforms rather than heaves, so the design problem is a coupled fluid-structure calculation and the acceptance criteria are deflection, curvature and acceleration limits rather than stability margins. Established Wave energy scales with the square of wave height, which is why sheltered siting is worth so much: halving the design wave height quarters the energy the structure has to absorb.

Frontier Perimeter treatment is the main design lever — breakwaters, permeable plates, porous flexible barriers, or energy-extracting devices — and the verified literature contains working approaches in each family without a settled answer about which dominates. Established Mooring is the second: a settlement must accommodate tide, surge and possibly decades of sea-level rise while transmitting horizontal loads into a compliant seabed, and the bed itself participates in the response in shallow water.

Established The unglamorous requirements are the ones that determine liveability. Potable water, wastewater, power and data must cross a moving interface between shore and structure, with connections that survive relative motion for decades. Established Corrosion protection and inspection access on a structure that cannot be dry-docked drive the maintenance regime and therefore the design life. Frontier Motion comfort is a habitability criterion with no agreed standard for permanent residence — ship and offshore criteria are written for crews on rotation, not for children, elderly residents and people who never leave. Speculative A residential motion-comfort standard does not exist and would have to be written.

Established Two requirements are specific to settlements rather than to platforms. The first is connection fatigue: a modular settlement grows by adding modules, and the joints between them cycle with every wave for decades, which is why fatigue analysis of multi-module structures appears in the literature at all. Frontier The second is repairability in place. A settlement cannot be taken out of service, so every system must be maintainable while occupied, which is closer to hospital engineering than to shipbuilding. Speculative Neither is a research problem. Both are cost drivers that a serious pro forma would have to carry and that no published proposal does.

Established Nearly every component of a floating settlement exists in commercial service somewhere else. Floating production storage and offloading units are permanently moored inhabited structures with class rules and insurance. Accommodation flotels house hundreds of people offshore for months. Cruise ships carry populations larger than most floating-city proposals. Established Marinas already licence berths commercially; houseboat regimes in several countries already contain a legal category for a dwelling on water. Established The pieces of the missing instrument exist and have never been assembled: strata title supplies the model for owning a defined volume inside a shared structure, marina berth licensing supplies the model for a transferable right to occupy a water space, and preferred ship mortgages supply the model for a registrable, internationally enforceable security interest.

Frontier On the technology side the adjacencies are dense. Offshore wind has driven down floating-platform and mooring costs and produced a supply chain. Small modular reactors in floating form are, as noted, the vanguard of the legal question. Ocean thermal energy conversion and desalination are the standard proposed utility stack for an off-grid settlement. Speculative Ports and shipping supply the logistics, and land reclamation remains the revealed-preference competitor: Kansai, Hong Kong’s airport platform, Singapore, Monaco and the Dutch Maasvlakte are the actual built answer to the demand for more urban area on water.

Speculative The most interesting adjacency is conceptual rather than technical. Space habitats face a structurally identical problem — a permanent human settlement on a constructed object outside any state’s territory, needing a property regime, an applicable law and an insurer — and the space-law community has spent decades on it. Speculative Space law and governance is the closest available analogue to the instrument this subject needs, and the two literatures barely speak. Speculative The comparison is instructive in one direction in particular: space lawyers have spent decades on registration, liability and jurisdiction for objects outside territory and have produced treaty machinery for it, while the floating-settlement literature has produced almost none, despite operating in a legal environment that is far more developed and far closer to hand.

Established Three institutional artefacts are missing, and they are the topic. The first is a registry of berth interests: a public register, maintained by an authority with territorial competence over the water in question, recording a defined, transferable interest in a mooring position, capable of taking a registered charge, and effective against third parties including the operator’s liquidator. Established That is exactly what a land register is, applied to water, and there is no technical obstacle to creating one.

Established The second is jurisdictional extension. Somebody must be the local authority. Streets need policing, children need schools, disputes need a court, buildings need an inspector, and rights need a forum. Established The cheapest route by far is for an existing host city to extend its ordinary municipal jurisdiction over a defined water area, so that the settlement inherits an entire legal system rather than inventing one. Frontier That is what the Dutch cases did and it is why they work.

Frontier The third is a market institution: class-society notation for permanently moored inhabited structures, an insurance market with capacity to write the first policies, and a lender willing to advance without comparables. Speculative Each of these has a first-mover problem, and the standard solution — a public guarantee or a development-bank first-loss tranche — is available and has not been tried here.

Frontier Note what is not on this list. No new international organisation is required. No treaty amendment is required. Established The instruments that block seasteading are not instruments that block floating settlement inside a state, and a great deal of energy in this subject has been spent attacking a barrier that a floating city does not have to cross.

Frontier One institution that would help is unglamorous and specific: a model law. Somebody — a law commission, a maritime institute, a development bank — should draft a model floating-habitation statute covering the registrable berth interest, the insolvency waterfall, the jurisdictional extension and the registry mechanics, and publish it for adoption. Established That is how strata title itself spread: one jurisdiction drafted it, it worked, and it was copied. Speculative The cost is a drafting committee, and the absence of such a document is the most conspicuous vacancy in the subject.

Frontier The exit argument has an ethical cost that its proponents state honestly and its critics usually overstate. If capable people can leave a badly governed jurisdiction, the jurisdiction faces pressure to improve — that is the case for exit. Frontier It is also true that the people able to buy a berth are not the people most harmed by bad governance, so exit as actually practised transfers the most mobile resources out of the places that need them. Established Both are true and the tension does not resolve. Speculative What can be said is that the two branches of this subject sit on opposite sides of it: a floating neighbourhood inside a host city is a housing policy, and a seastead outside one is an exit.

Established Rights at sea are the hardest ethical question and the least discussed. A permanently inhabited offshore structure creates a resident population — including children, who did not choose the arrangement — whose civil, criminal, family and succession law must attach somewhere. Frontier The human-rights-at-sea literature is where this is being worked, and it is not yet settled. Established A settlement with no applicable law is not a libertarian utopia; it is a place where a person can be harmed with no court to hear it.

Frontier Labour is the second. Every proposal for a floating city implies a service workforce who live where they work, at sea, potentially under a flag state chosen for its permissiveness. Speculative Maritime labour regimes were designed for crews on rotation; applying them to resident staff would be a novel and uncomfortable extension.

Frontier Third, catastrophe risk is borne unequally. A settlement that cannot run from a storm concentrates risk on its residents, and if it is uninsured or under-insured, the residual falls on them and on whoever mounts the rescue. Speculative For the adaptation branch there is a distributional question underneath: floating adaptation is proposed most often for atoll states, which are precisely the states least able to finance it, so the technology is offered to the people who cannot buy it.

Speculative If the berth interest existed, the consequence would be larger than a few neighbourhoods. Roughly seventy per cent of the planet’s surface is water, and the shallow, sheltered fraction inside national jurisdictions is not small. Speculative A registrable interest in a mooring would convert some of that surface from a regulated void into ordinary real property — financeable, taxable, transferable, insurable — which is the mechanism by which land itself became capital. Handwave The plausible long-run effect is not floating megacities but a slow thickening of the water margin of existing coastal cities, which is what a working property regime usually produces.

Speculative The second implication is about adaptation. Floating structures possess one property nothing on land has: they rise with sea level automatically, at zero marginal cost, indefinitely. Frontier Against a slow, monotonic rise this is a structural advantage over any defence that has to be raised. Speculative It competes, however, with piled construction, which is far cheaper in shallow water and yields real property, and with simply moving inland — so the combination that actually favours floating is water deep enough to make piles uneconomic and no inland to move to. That combination mostly describes atoll states.

Handwave The third is jurisdictional. If a state ever does write a habitation chapter into its maritime code and a floating settlement operates under it successfully, the resulting entity is something new: a resident population under a flag rather than under a territory. Handwave Whether that would be a laboratory for institutional innovation or a mechanism for regulatory arbitrage at residential scale is genuinely unknown, and the historical analogue — flags of convenience in commercial shipping — points at the second more strongly than the first. Speculative It is worth noticing that this is the seasteading thesis arriving through the back door, in a weaker and more plausible form: not a new sovereign, but a resident population choosing among the regulatory offerings of existing ones.

These horizons track the legal and financial instrument, not the hardware. The hardware is available now.

• 10 yr: Frontier More sheltered-water floating neighbourhoods inside national jurisdictions, in the Netherlands and probably in East Asia, at the scale of tens to low hundreds of households, financed the way the existing ones are financed — case by case, by lenders who know the borrower and the municipality. Frontier At least one adaptation-branch project reaching partial construction, and at least one more announced with fanfare and not built, which is the base rate this class of project has established over two decades. Speculative A first serious attempt at a registrable berth interest in one national law, most likely as an extension of an existing houseboat or marina regime rather than as fresh legislation, because extending a category the registry already understands is far easier than creating one.
• 25 yr: Speculative If the berth interest exists in two or three jurisdictions, the derivative chain follows quickly: class notation, insurance capacity, appraisal comparables, and a first mortgage book. Speculative A floating district of several thousand residents inside a host city’s municipal boundary is the realistic upper end, and it would look much more like a waterfront neighbourhood than like anything in the seasteading literature. Speculative The floating-nuclear regulatory question is probably settled by this point, and this subject inherits the answer.
• 50 yr: Speculative Open-coast siting becomes conceivable only if a storm-survival record accumulates and the catastrophe-correlation problem is priced rather than avoided. Handwave A settlement large enough to carry its own economy rather than commuting to a shore economy is possible by then, and would be the first genuine test of whether floating urbanism is a distinct settlement form or simply a suburb with an unusual foundation.
• 100 / 250+ yr: Handwave A permanently inhabited high-seas settlement remains excluded by the instrument rather than by the engineering, and instruments change on century timescales. Handwave The plausible route is not a new sovereign but a delegated-jurisdiction arrangement that becomes conventional through repetition — the way flag-state jurisdiction itself became conventional. Handwave An ocean-surface civilisation of the kind the seasteading literature imagines requires a change in the law of the sea that no state currently has an incentive to make, and predicting that is prediction about politics, not about technology.

• Depends on Depends on the ordinary offshore engineering stack — mooring systems, corrosion management, marine utilities, class-society rules — all of which exist in commercial service, and on the hydroelastic design methods developed by the very-large-floating-structure literature since the 1990s. It depends far more heavily on results that are not technological at all: a reading of the primary law-of-the-sea instruments, a maritime-finance analysis of whether a habitation module can carry a preferred ship mortgage, and the outcome of the floating-nuclear licensing process, which will characterise a permanently moored non-ship inhabited structure in law before anyone does it for housing.
• Requires (not on this map) None of the three is a scientific result and none is expensive. The first is a registrable, transferable interest in a mooring position that survives the platform operator’s insolvency — the difference between a berth licence, which is worthless in a liquidation, and a berth interest, which is not. Without it there is no collateral, and without collateral there is no mortgage, no appraisal standard and no household purchaser. The second follows from the first but has an independent physical obstacle: a moored settlement cannot run from a storm, so its exposure is catastrophe-correlated in the way marine insurance is structurally worst at pricing. The third is the cheapest and least discussed — an existing municipality extending its ordinary jurisdiction over a defined water area, so that the settlement inherits a whole legal system instead of inventing one. The Dutch floating neighbourhoods work because all three are quietly present. Nowhere else has assembled them.
• Enables Enables coastal cities to add usable area without reclamation, which is the only currently practised alternative and which consumes seabed permanently. Enables a class of adaptation for low-lying and atoll states that rises with sea level at no marginal cost. If the berth instrument is ever drafted, it enables an entire asset class — water-surface real property inside national jurisdictions — and the mortgage, insurance and appraisal machinery that follows any new form of registrable title.
• Adjacent Sits alongside coastal defense, which is the substitute good in shallow protected water, and alongside climate migration planning, which is the substitute good when neither defence nor flotation is affordable. Shares its central legal problem with space habitats, where a permanent settlement on a constructed object outside any state’s territory raises exactly the same property, jurisdiction and insurance questions, and where the literature is considerably more developed. Sits alongside floating nuclear power, which is not a neighbouring technology but a legal pathfinder: it will force a characterisation of the permanently moored non-ship structure, and floating settlement will inherit whatever that characterisation turns out to be, general or nuclear-specific.

Handwave “International waters are ungoverned, so a structure there is free of any state’s law.” This is the load-bearing error and every other enthusiast claim rests on it. Established It is wrong in three independent ways: claims of sovereignty over the high seas are invalid under the Convention as such, not merely unrecognised; the freedom to construct installations there is expressly qualified and exercised subject to the Convention and other international law; and everything at sea has a flag state with exclusive jurisdiction over it, so the absence of a flag is not freedom but statelessness, and a stateless structure is subject to boarding rather than immune from it. Frontier This brief states the substance and withholds the article numbers because it could not read the treaty.

Handwave “The exclusive economic zone is international waters.” Not for this purpose. Established The coastal state has the exclusive right to authorise artificial islands and installations throughout its exclusive economic zone, out to two hundred nautical miles, with exclusive jurisdiction over them. Frontier The 2020 Ocean Builders seastead off Phuket was in Thailand’s exclusive economic zone, not on the high seas, so the Thai response was legally routine rather than an overreach — and enthusiast accounts almost always omit the zone point, which is the entire legal content of the case.

Handwave “Sealand proves a micronation can survive.” Frontier The 1968 outcome depended on the United Kingdom’s then three-nautical-mile territorial sea; the United Kingdom later extended to twelve and enclosed the fort, and no state has ever recognised it. Established Sealand demonstrates that the legal boundary is a state’s to move, which is the opposite of the lesson usually drawn from it.

Handwave “Once it is built and people live there, recognition will follow.” Frontier Effectiveness has produced recognition in none of the recorded cases, and in three of them the structure was physically removed. Frontier The Montevideo criterion of a defined territory is a threshold that occupancy does not cross, because a floating structure is a chattel.

Handwave “Mega-Float shows we can already build floating cities.” Frontier Mega-Float was unoccupied, in a sheltered bay, for a trial period, and was dismantled. It is evidence about structural feasibility at large scale in sheltered water and about nothing else — not permanence, not habitation, not open-coast exposure, and not law.

Handwave “Kansai International Airport shows what goes wrong when you build on water.” Established Kansai is a reclaimed island on soft marine clay, not a floating structure, and its settlement into the seabed — which substantially exceeded design prediction — is a cautionary tale about filling. If anything it is an argument for floating. This confusion appears constantly and should not be let through.

Handwave “Floating is cheaper than reclamation.” Speculative Possibly true per square metre of deck, and irrelevant on its own, because reclamation delivers real property with title, mortgage and insurance while floating delivers a chattel with none of them. A comparison that omits the legal and financial overhead is not a comparison, and nobody has published one that includes it.

Speculative “A floating city just needs a flag of convenience.” Established Flagging places the structure inside a state’s exclusive jurisdiction, so it is the surrender of the sovereignty claim rather than a workaround for it. Speculative It is also, separately, the most promising practical route available, because it is the only one with an existing registry, an existing mortgage regime and existing international recognition. Frontier Both statements are true and the brief holds them together rather than choosing.

Handwave “Sea-level rise makes floating cities inevitable.” Speculative Sea-level rise makes elevation valuable, and floating is one of four ways to get it, competing with fill, piles and moving — and piled construction is far cheaper in shallow water and yields ordinary real property. Inevitability is not the right frame; a narrow site condition is.

Established “The obstacle is that we cannot build big enough structures.” Three decades of very-large-floating-structure literature and a demonstrated kilometre-class runway say otherwise. Established The obstacle is a missing registrable interest in a berth, and the engineering only becomes hard because the law pushes a settlement offshore.

Established Two structural errors about the subject itself, which matter more than any individual claim. The first is blending the exit branch with the adaptation branch. Oceanix Busan and the Maldives proposal have a host state, a public sponsor and a development-finance path; seasteading is defined by the absence of all three. Established Treating them as one movement is the commonest error in popular coverage and it makes both look worse than they are.

Frontier The second is a fact about how citations get corrupted in machine-assembled scholarship, and it is worth stating because it happened to the research behind this brief. Established The bibliographic record for a multi-module floating-structure fatigue paper in the International Journal of Fatigue returns a publication year of 1992, which is almost certainly wrong, since multi-module work postdates it by two decades — so either the title or the year is corrupt in the index. Established In the same research session a publisher’s article server returned a completely unrelated paper, twice, under a correctly addressed identifier, with no error and full confidence, and four independent research passes in this programme caught their own bibliographic tools substituting one document for another. Frontier The practical rule that follows: verify the returned title against the requested title before using any content, and treat “everyone knows X” as a reason to read the source rather than a licence to skip it. Established And its converse, which is equally important: could not locate is not does not exist. Several sources this brief marks unverified are probably fine and were simply unreachable.