The circles of interest
A postulate, a measurement methodology, ten probes, and a way to scale the inquiry. Cross-listed in Politics.
1. The postulate
Around every individual lie circles of interest. Some are concentric — self, family, community, country, humanity — each larger, each more weakly held. Others are non-concentric: co-religionaries, coworkers, a diaspora — circles that overlap the rings without respecting their geometry. My postulate is that all politics is a trade-off of resources and interest across these circles: every law, budget and institution is an answer to the question “how much of what belongs to an inner circle must flow to an outer one, and who decides?”
Concentric circles of interest around the individual — crossed by non-concentric circles (faith, profession, diaspora) that do not respect geography. Politics operates on the slope of the decay.
The intuition is old. Hierocles the Stoic drew exactly these rings and made ethics the art of “drawing the circles inward”; Confucius defended graded love against Mozi's impartial care; Peter Singer turned the image into a program with the expanding circle. What is usually missing is measurement: the circles are treated as a moral exhortation, not as an empirical object. I want to treat them as an empirical object.
2. A methodology: revealed allocation
Do not ask people what they value — watch what they allocate. The method, in five steps:
- Fix the circles for the population studied. They are not universal: clan, sect, caste, class or platform may matter more than “community”. The circle structure is itself a finding, not an assumption.
- Enumerate boundary-crossing mechanisms. For each adjacent pair of circles, find the institutional dials and behaviours that move resources across the boundary: taxation crosses individual→country, aid crosses country→world, inheritance stays inside family, remittances cross borders without leaving the family.
- Measure flows in three currencies. Money is the most legible, but time (care, volunteering, service) and risk (war, organ donation, dangerous work) reveal weights money hides.
- Normalise into exchange rates. Per-capita flow toward a circle, divided by per-capita flow toward the innermost circle, gives the revealed weight of that circle — “how many units of outer welfare buy one unit of inner welfare”.
- Separate voluntary from compulsory. Voluntary flows (giving, remittances) reveal the individual gradient; compulsory flows (taxes, conscription, defaults) reveal the collective settlement. The difference between the two is precisely what politics adds — or forces.
A proxy is never pure: taxation buys insurance and public goods, not only solidarity; remittances repay debts as well as love. The method works by triangulating many imperfect proxies, not by trusting one.
3. Ten probes
Figures are order-of-magnitude anchors from well-known public data, used illustratively — the point is the pattern, not the decimals.
Taxation (compulsory levies)
France's compulsory levies run around 45% of GDP — roughly half of everything produced is redirected from private circles to the national one.
Inference. The state's defining act is flattening the inner gradient: citizens are forced to weight anonymous countrymen at roughly half the rate they weight themselves.
self : countryman ≈ 2 : 1 (forced)
Official development aid (ODA)
France spends ~55–57% of GDP publicly inside its border and ~0.5% of GNI on development aid outside it — a ratio of about one hundred to one.
Inference. At the border, the collective gradient falls off a cliff: a foreign life commands about 1% of the public resources a domestic one does.
countryman : foreigner ≈ 100 : 1
Charitable giving
Voluntary donations run from ~0.1% of income (France, declared) to ~2% (United States) — while essentially all the rest is consumed inside the household circle.
Inference. Left free, individuals allocate to strangers between one-thousandth and one-fiftieth of what they allocate to their own circle — the voluntary gradient is far steeper than the taxed one.
household : stranger ≈ 100 : 1 (voluntary)
Inheritance & the réserve héréditaire
The overwhelming share of estates passes to children; French law (réserve héréditaire) even forbids disinheriting them — family's claim is not just chosen but mandated.
Inference. At death, when nothing can be reciprocated, revealed preference is nearly pure: family absorbs almost everything, and the law ratifies the innermost circle rather than flattening it.
family : all others ≈ 10 : 1 at death
Remittances
Migrants routinely send ~10–15% of income to family at home; global remittances (~$850B/yr) are roughly four times all official aid combined.
Inference. The family circle is not geographic: it punches through borders, and privately moves more money across them than every state's organised solidarity together.
kin abroad ≫ neighbours at home
Organ donation rules
Living donation happens almost exclusively within the family (in France the law long restricted it to close ties); donation to strangers becomes common only after death — and only via an opt-out default set by the state.
Inference. For the body — the least fungible resource — the gradient is nearly vertical while alive; the state can flatten it only once the cost to the inner circle drops to zero.
living organ: family only, in practice
Conscription
Under existential threat, states have demanded the ultimate resource — risk of death — for the benefit of countrymen, overriding both the individual and the family circle.
Inference. The circle ordering is not fixed: in emergencies the outer circle can be made to outrank the inner ones, but only by compulsion, and only while the emergency lasts.
war inverts the gradient (by force)
Pandemic vaccine allocation
In 2021, rich countries pre-ordered several times their populations' needs and administered boosters while much of the world waited for first doses; COVAX was chronically short.
Inference. In an acute, symmetric crisis — the cleanest natural experiment — states revealed that a compatriot's marginal protection outweighs a foreigner's first protection.
compatriot booster > foreigner's 1st dose
Guilds, licences, strike funds
Licensed professions and unions secure concentrated benefits (entry barriers, protected fees, strike pay) whose diffuse costs fall on all consumers — and such rules persist across political systems.
Inference. Non-concentric circles are load-bearing: an organised professional circle routinely defeats the much larger but unorganised circle of countrymen (Olson's logic of collective action).
organised few > unorganised many
Discount rates in climate policy
Public cost-benefit analysis discounts the future at a few percent a year; at 3%, a life a century away counts for roughly one-twentieth of a life today — and effective carbon prices sit well below most estimates of the social cost of carbon.
Inference. Time is a circle too: future people are the outermost ring, with no mechanism — vote, strike, or remittance — to pull resources toward themselves.
today : in 100 yrs ≈ 20 : 1 (at 3%)
4. What the probes suggest
First: the gradient is steep and roughly geometric. Left to voluntary behaviour, each step outward costs about an order of magnitude: nearly everything stays in the household, ~1% of income reaches strangers, ~1% of public wealth crosses the border. Interest does not fade — it collapses.
Second: the state is a machine for flattening the gradient — inside its border. The deepest political act in the data is taxation: compulsion forces the self:countryman rate from ~100:1 (voluntary giving) to ~2:1. But the same state re-steepens the curve at its edge (aid at 0.5% of GNI). The nation-state, seen through this lens, is a plateau with a cliff: solidarity made flat within, vertical at the border.
Third: non-concentric circles are load-bearing. Remittances outweigh all official aid combined; organised professions defeat the unorganised many; faith and diaspora move resources across borders that states harden. Any politics that models people as points inside territorial rings only — ignoring the circles that cut across — will keep being surprised by them.
Fourth: political conflict is a dispute about the slope, not the circles. Almost every classic cleavage re-reads as an argument over one segment of the curve: left vs right is the steepness inside the border (taxation, redistribution); nationalism vs cosmopolitanism is the height of the cliff (aid, trade, migration); familialism vs statism is the innermost slope (inheritance tax); ecology is the temporal slope (discount rates). Politics is the collective bargaining of λ — the decay rate.
5. Scaling the method
To go from ten probes to real conclusions, make the decay rate a measurable object. Sketch:
- Fit a curve. For a person, a country or an era, fit w(d) ≈ e−λd where d is circle distance and w the revealed per-capita weight. λ is a parochialism coefficient — one number for how fast interest dies with distance; ε, the residuals, capture the non-concentric circles.
- Build the panel. Public data exists for every rung: tax codes and budgets, ODA, philanthropy statistics, remittance flows, time-use surveys, conscription and donation rules, carbon prices. Compute λ per country and per decade — a gradient atlas.
- Confront stated and revealed. Moral-circle psychometrics and the World Values Survey give stated weights; the atlas gives revealed ones. The gaps — where people claim wider circles than they fund — are where political rhetoric lives.
- Use natural experiments. Disasters, wars and pandemics shock the circles and reveal elasticities: how much did giving to distant victims rise after the 2004 tsunami? How fast did vaccine nationalism override treaties? Shocks identify what equilibrium hides.
- Make falsifiable predictions. If the model is right, λ should predict welfare-state size, aid budgets, migration attitudes and climate policy better than income or ideology alone; and politics should polarise wherever two populations with different λ share one budget. If it predicts nothing, discard it.