The Rupture Assumption

The chronicle so far has looked backward. Twenty-seven chapters have traced one mechanism (a boundary between the people to whom obligations are owed and the people to whom they are not) from the deep past to the Security Council chamber. The remaining chapters look forward, and they rest on a single planning assumption that this chapter must state plainly, defend honestly, and mark clearly as an assumption rather than a prophecy.

The assumption is this: the interlocking crises of climate, resources, and politics blow up into systemic rupture somewhere around 2040–2050 (well before the most severe physical effects of the climate crisis themselves arrive), and when they do, the international system's response will follow the pattern twenty-seven chapters have documented, not the pattern its press releases promise. Nothing serious will have been done to prevent it. Nobody will be coming. And core economies under stress will do what core economies under stress have always done in this chronicle: they will externalise the cost onto people outside the boundary of obligation.

If that assumption is right, a small country has roughly one generation to prepare. If it is wrong, the preparations it dictates (food security, energy independence, domestic manufacturing, water storage, regional integration) are things a poor, landlocked, densely settled country should be doing anyway. That asymmetry is the entire logic of this final movement. The cost of preparing for a rupture that never comes is development. The cost of not preparing for one that does is the end of sovereignty, and this chronicle has spent its whole length showing what that looks like.

The evidence that nothing will be done

Begin with the record, because the record is not ambiguous.

The first World Climate Conference met in 1979. The IPCC was founded in 1988. The Framework Convention was signed in 1992, Kyoto in 1997, Copenhagen collapsed in 2009, Paris was celebrated in 2015, and thirty consecutive Conferences of the Parties have now been held. Across that entire diplomatic half-century, the one curve that measures the outcome (the concentration of carbon dioxide in the atmosphere) has not merely risen; its rate of rise has accelerated, from roughly 1.5 parts per million a year in the 1980s to well over 2 ppm a year in the 2020s.¹ Global fossil emissions set a new record in 2023 and again in 2024. The Paris signatories' own stocktake concedes that current policies point far past the treaty's limits.

This is not a story of trying and failing. The infrastructure decisions are being made now, in the open: new oilfields licensed, new pipelines financed, coal plants commissioned with forty-year design lives, and the largest historical emitters expanding extraction while lecturing others about ambition. The chronicle does not read this as hypocrisy, which would be uninteresting. It reads it the way chapter 2 read Rome: as a system doing what it is built to do. An economy organised around growth and extraction cannot decide to stop extracting any more than the empire of chapter 2 could decide to stop expanding. The consumption is not a policy failure; it is the system operating normally.

The planning consequence follows directly. A state that builds its national strategy on the premise that the great powers will decarbonise in time, honour their adaptation-finance pledges, and manage the transition equitably is a state betting its survival on a promise from the same institutions whose promises chapters 24 through 27 examined in detail. Rwanda has already run that experiment once, in April 1994, at the highest possible stakes. The result is on the record.

Why 2040–2050, when the worst physics comes later

The counter-intuitive part of the assumption is the timing, so it deserves the most care. The severest physical effects of warming (multi-metre sea level rise, the full force of tipping cascades, large regions passing heat thresholds for outdoor labour) belong mostly to the second half of the century and beyond. So why plan for rupture in the 2040s?

Because complex systems fail politically and financially long before they fail physically, and the chronicle is, from beginning to end, a study of exactly that. Some of the evidence:

The system-dynamics literature has pointed at this window for fifty years. The 1972 Limits to Growth study (ridiculed at the time, and repeatedly vindicated since by data-recalibration studies in 2008, 2014, and 2021) projected that its business-as-usual scenarios begin their decline in output and welfare around the 2040s, not because the planet is exhausted, but because the cost of coping with converging constraints eats the surplus that everything else runs on.² The most recent recalibration, done inside a major accountancy firm rather than a protest movement, found the world tracking closest to the scenarios that peak and turn down around 2040.

Food systems break at weather-shock frequencies, not at climate averages. A global food market in which a handful of breadbaskets and four or five maritime chokepoints feed the import-dependent world is exquisitely sensitive to simultaneous regional shocks, the "multiple breadbasket failure" scenario that reinsurers and defence ministries now model as a matter of routine.³ The 2007–08 and 2010–11 price spikes each required only modest harvest failures plus export bans to double staple prices and destabilise a dozen governments; the probability of far larger simultaneous failures rises steeply through the 2030s. Rwanda imported food inflation from a European war in 2022 within months. That was a rehearsal.

Finance retreats decades ahead of the water. Insurers are already withdrawing from Florida, California, and swaths of Australia: not in 2080, now. When insurance goes, mortgages go; when mortgages go, asset values go; and the exposed asset base of the global financial system is measured in tens of trillions. A disorderly repricing of climate risk is the kind of event that arrives suddenly, and the financial system has shown (in 2008) how much rupture a mere housing-market repricing can transmit in eighteen months.

The political feedbacks are already visible, and they run ahead of the physics. The hardening of borders, the criminalisation of migration, the electoral yield of siege politics in Europe and North America: chapters 17 through 20 traced the machinery; it is now being retooled around climate-driven movement that has barely begun. The World Bank's own projections put internal climate migrants at over 200 million by 2050, with Sub-Saharan Africa the largest share.⁴ The rich world's observable response is not preparation to receive; it is preparation to repel. Systems that respond to stress by exclusion do not wait for the stress to peak before turning exclusion into policy.

None of these curves needs to reach its own maximum to break the system they run through. They need only to converge: a food-price crisis, a financial repricing, a migration panic, and a war or two, arriving in the same decade into polities that have spent that decade radicalising. The 2040–2050 window is not where the physics peaks. It is where the convergence becomes hard to avoid. And as chapter 14 showed with 1914 and chapter 24 showed with 1994, the step from "structurally primed" to "actually happening" can take a season.

The scramble comes back

Here is the part of the assumption that will strike some readers as paranoid, and that this chronicle regards as the closest thing to a law it has found: when core economies face resource stress, they do not tighten their own belts. They come back for other people's.

That is what chapter 2 found in Rome's grain provinces. It is what chapters 8 through 11 found in the Atlantic system. It is what chapter 16 found dressed as free trade and anti-communism, and what chapter 27 found dressed as humanitarianism. The form changes with the era's vocabulary; the function does not. So the question is not whether a decarbonising, destabilising rich world will seek to secure African land, minerals, and carbon sinks on extractive terms. It is already doing so, politely:

Under the rupture assumption, these instruments are the mild, rules-bound early form. The chronicle's expectation (stated as expectation, not certainty) is that as the 2040s approach and core polities radicalise, the instruments shed their politeness in the order the politeness was applied: first the finance hardens, then the trade terms, then the conditionality becomes overtly political, and eventually some stressed power, somewhere, re-derives the conclusion of 1884: that it is cheaper to control the territory than to pay its government. "Old-school" colonisation (flags and governors) is the least likely form, because chapter 16 showed the modern kit works better. But the function returning is, on this chronicle's evidence, close to certain.

What follows for a country of thirteen million

If the assumption holds, then the strategic problem for Rwanda in 2025 is precisely defined. It has roughly fifteen years of a functioning global system left to use, years in which technology, capital goods, expertise, and markets can still be imported, to make itself into a country that can withstand the system's failure. The preparation has four load-bearing walls, and each of the remaining chapters takes one or more of them in detail:

Know exactly what is coming to the region (chapter 32). Preparation begins with projection: what the climate models actually say about the Great Lakes (rainfall, heat, disease altitude, lake stability), and, just as important, what they say about the neighbours, because Rwanda's climate exposure is regional before it is meteorological.

Close the survival dependencies (chapter 33). Food, water, and energy must be domestically secured or regionally secured with partners who share the exposure. A blockade-proof calorie supply and a fuel-import-proof energy grid are the difference between a hard decade and a terminal one.

Domesticate the manufacturing that matters (chapter 34). Medicines, fertiliser, cement, tools, spare parts, and the repair economy. Not prestige industrialisation but survival industrialisation, chosen by asking of every import: what happens here if this stops arriving?

Make the region the unit of survival. A landlocked country cannot autark its way out; its ports, its market, and its buffer against shocks are its neighbours. The East African Community and the continental market are not trade preferences under this assumption; they are the only available substitute for the global system that is assumed to fail. That, and not sentiment, is why the integration agenda belongs in a survival plan.

Notice what is absent from the list: the bomb of chapter 30, and any fantasy of fortress isolation. The conclusion of the previous movement stands. What a small state can do is make itself expensive to coerce and pointless to starve: a country that feeds itself, powers itself, medicates itself, repairs its own machines, and is embedded in a region that would have to be dealt with as a whole.

The wager, stated honestly

Every plan is a bet, and the chronicle owes the reader the odds as it sees them. The bet here is asymmetric in the way good bets are. Suppose the assumption is wrong; suppose the system muddles through, decarbonisation accelerates, the 2040s arrive without rupture. What has Rwanda lost by preparing? It has "wasted" its effort on food security, universal energy access, water storage, domestic pharmaceutical production, a machine-repair economy, and deep regional integration. That is a description of successful development. There is no scenario, rupture or none, in which those investments are regretted.

Now suppose the assumption is right and the preparation was not made. The chronicle has already written that chapter several times: it is the Taíno in chapter 9, Haiti in chapter 11, and Rwanda itself in 1994: the discovery, at the worst possible moment, that the safety net was rhetorical. The asymmetry is total, and it is the whole argument.

The turn

The chronicle exists, its opening pages said, to keep a record straight. This final movement is the reason keeping it straight matters. A population that knows exactly how the boundary of obligation works (who has been left outside it, how the exclusion is dressed up, and how reliably the rescue fails to arrive) is a population that cannot be talked into waiting for rescue. That is the practical use of twenty-seven chapters of history: not grievance, but calibration.

The next three chapters are the calibrated response. They contain no heroes and no slogans, only projections, arithmetic, and the record of what a small, densely peopled, landlocked country has already shown it can build when it assumes, correctly, that nobody is coming.

Footnotes

1. Atmospheric CO₂ growth rates from the NOAA Global Monitoring Laboratory Mauna Loa and global marine surface records; decadal mean growth roughly 1.6 ppm/yr in the 1980s against roughly 2.4 ppm/yr in the 2010s–2020s. Emissions records: Global Carbon Project, Global Carbon Budget (2023, 2024 editions). ↩
2. Donella H. Meadows et al., The Limits to Growth (New York: Universe Books, 1972); Graham M. Turner, "A comparison of The Limits to Growth with 30 years of reality," Global Environmental Change 18:3 (2008), pp. 397–411, and his 2014 update, "Is Global Collapse Imminent?" (MSSI Research Paper 4, University of Melbourne); Gaya Herrington, "Update to limits to growth: Comparing the World3 model with empirical data," Journal of Industrial Ecology 25:3 (2021), pp. 614–626. ↩
3. On chokepoints and simultaneous breadbasket failure: Rob Bailey and Laura Wellesley, Chokepoints and Vulnerabilities in Global Food Trade (London: Chatham House, 2017); Lloyd's of London, Food System Shock (Emerging Risk Report, 2015). On the political transmission of the 2007–11 price spikes, see the literature on the food-price/unrest correlation, e.g. Marco Lagi, Karla Z. Bertrand and Yaneer Bar-Yam, "The Food Crises and Political Instability in North Africa and the Middle East" (New England Complex Systems Institute, 2011). ↩
4. Viviane Clement et al., Groundswell Part 2: Acting on Internal Climate Migration (Washington, DC: World Bank, 2021): up to 216 million internal climate migrants by 2050 across the modelled regions, with Sub-Saharan Africa up to 86 million. On broader systemic-risk framings, Luke Kemp et al., "Climate Endgame: Exploring catastrophic climate change scenarios," PNAS 119:34 (2022). ↩
5. On the 2022–23 wave of large offset concessions negotiated by Gulf-based firms over forest estate in Liberia, Zimbabwe, Zambia, Tanzania and elsewhere (in some reported drafts approaching 10–20% of national territory), see the investigative coverage collected under "Blue Carbon" agreements, e.g. Middle East Eye and The Guardian reporting, November–December 2023, and the analysis by Power Shift Africa, The Great Carbon Grab (2023). On post-2008 farmland acquisitions, Lorenzo Cotula, The Great African Land Grab? (London: Zed Books, 2013). ↩