The Tetrad Tour, Part II

Politics, Beauty, and Why We Don’t Get Nice Things

In Part I, we applied the tetrad — physics, math, CS, economics — to anthropology, psychology, law, and linguistics. The pattern: take a field, identify its standard framing, apply all four lenses, and watch the real structure emerge.

Today we go deeper. Political science, aesthetics, history — and then something important: what happens when the tetrad catches its own errors.

Political Science

Standard framing: Democracy vs. authoritarianism. Left vs. right. The big debates are about which values should guide society, who should rule, and how to balance freedom against equality.

The tetrad:

Physics: Politics operates on territory, bodies, and weapons. The monopoly on legitimate violence is the foundation. Everything else is built on that substrate.

Math: Voting theory is mathematics. Arrow’s impossibility theorem. Gibbard-Satterthwaite. Coalition math. These aren’t just academic curiosities — they’re constraints, like thermodynamics.

CS: Constitutions are source code. Precedent is version control. Bureaucracy is runtime — taking abstract rules and executing them on specific cases. Political parties are long-running processes that seek to replicate themselves.

Economics: Power is a scarce resource. Political actors are self-interested optimizers. Policy outcomes are the equilibria of games between competing interests.

What falls out:

The first thing that falls out is a famous impossibility result. Arrow’s theorem says: no voting system can simultaneously satisfy a small set of obviously desirable properties (non-dictatorship, Pareto efficiency, independence of irrelevant alternatives). Something has to give.

This seems to suggest that all political systems are hacks around fundamental impossibility — that we’re always trading off one form of unfairness for another.

But wait. There’s an escape hatch.

The Approval Voting Escape

Arrow’s theorem applies to ordinal voting systems — systems where voters rank candidates. But what about cardinal systems, where voters assign independent scores?

Approval voting works like this: for each candidate, vote “approve” or “don’t approve.” Most approvals wins. Simple.

This completely sidesteps Arrow. Why?

• Arrow requires voters to submit complete rankings

• Approval voting doesn’t use rankings — each candidate gets an independent binary score

• The “Independence of Irrelevant Alternatives” criterion is satisfied trivially: adding candidate C doesn’t change whether you approve of A or B

So Arrow doesn’t apply. Approval voting is a genuine escape from impossibility, not a hack around it.

What about Gibbard-Satterthwaite, the other big impossibility result? It says any deterministic voting system with 3+ candidates is either dictatorial or manipulable (strategic voting is possible).

Approval voting is manipulable — you can strategically withhold approval. But:

1. The manipulation is transparent. You know you’re doing it.

2. The equilibria are often good. Approval voting tends to elect Condorcet winners when they exist.

3. The strategy is simple. Approve candidates above your expected utility of the winner.

The math says: approval voting isn’t perfect, but it’s provably better than plurality voting (what most democracies use) and arguably better than ranked-choice voting (which doesn’t even satisfy monotonicity — voting for a candidate can cause them to lose).

So why don’t we have it?

Why We Don’t Have Approval Voting

Here’s where the economics lens becomes essential.

The people who would need to approve the change to approval voting are exactly the people who won under the current system.

Think about it:

• Plurality voting converges to two parties (Duverger’s Law)

• Those two parties then control ballot access, debate rules, redistricting, and — crucially — voting system reform

• They’re not going to vote themselves into competition

It’s a monopoly that gets to write the antitrust laws.

This explains the pattern we observe:

• Approval voting wins in low-stakes elections where parties aren’t paying attention (Fargo, St. Louis, some professional associations)

• It gets crushed or ignored at state and federal levels where parties fight back

• Ranked-choice voting — which still tends to favor major parties over approval — gets adopted instead, often with party support, because it looks like reform while preserving the duopoly

Revealed preference: politicians say they want fair elections. They do block approval voting. Believe the behavior, not the words.

The UN Exception

Here’s what makes this argument airtight: the one place where political parties don’t dominate, approval voting emerges naturally.

The UN Security Council selects the Secretary-General through “straw polls” where each member nation indicates “encourage,” “discourage,” or “no opinion” on each candidate. This is essentially three-level approval voting.

In 2006, when Ban Ki-moon was selected, 15 countries voted on 6 candidates. The average number of “approves” per ballot was 2.6 — nowhere near degenerating to plurality voting. Voters genuinely used the full range of the system.

Why does the UN use this system? Because it’s an environment where no single party controls the rules. The member states have sufficiently divergent interests that no one bloc can rig the system in its favor. (Yes, the five permanent members retain veto power — but the straw poll itself operates without party-line discipline, which is the point. The approval mechanism works because no domestic-style duopoly controls the ballot design.) So they ended up with something close to optimal.

Donald Saari, a mathematician, once worried that approval voting would “degenerate” to plurality in practice — everyone would just approve their first choice. The UN Secretary-General elections are a high-stakes refutation: when parties can’t capture the rule-making process, approval voting works exactly as the math predicts.

Aesthetics

Standard framing: Beauty is subjective. Art can’t be analyzed. De gustibus non est disputandum.

The tetrad:

Physics: Light waves, pigment, sound frequencies, neural processing. Art is physical stimuli hitting sensory organs.

Math: Symmetry. Proportion. The golden ratio. Harmonic series. Fractals and self-similarity. There’s obviously somestructure to what humans find beautiful.

CS: Art is compression. A painting encodes more information than it literally contains — it evokes a scene, an emotion, an idea that would take thousands of words to describe. Good art achieves high compression ratios.

Economics: Art is status signaling. It’s attention economics. It’s costly signaling of fitness (the peacock’s tail, the patron’s Vermeer).

What falls out:

Beauty is the subjective experience of high compression — a lot of information encoded cheaply.

When you look at a beautiful face, a sunset, an elegant mathematical proof — what’s happening is that your pattern-recognition systems are clicking into place efficiently. The stimulus carries rich information but matches your priors well enough that encoding it is cheap. The prediction errors are small. It feels good because you’re getting a lot for a little. (A blank page has low computational cost too, but it carries no information — that’s why it’s not beautiful. Beauty requires something worth compressing.)

(This is the same principle that explains why retail traders overpay for binary options. A binary option is the “beautiful” financial product — two outcomes, trivially easy to evaluate. The bull spread pays more but costs more to process. Traders pay a simplicity tax for the same reason we find symmetry beautiful: high information at low processing cost feels good.)

This explains a lot:

• Why average faces are beautiful: They’re easier to encode. Each feature is close to the statistical mean, requiring less information to represent.

• Why symmetry is beautiful: One half predicts the other. Massive compression.

• Why “interesting” art violates expectations just enough: Too predictable (kitsch) is boring — you’re not learning anything. Too unpredictable (noise) is exhausting — you can’t encode it at all. Great art sits on the edge: compressible, but not trivially so.

• Why mathematical proofs can be “elegant”: An elegant proof is a short path through logical space. It compresses a complex truth into a small number of steps.

This doesn’t reduce aesthetics to neuroscience. You still need to understand art history, cultural context, the specific traditions a work is responding to. But it does give you the right organizing question: what are the computational properties of stimuli that humans find rewarding to process?

History

Standard framing: The great debate is “Great Man” vs. “Historical Forces.” Did Napoleon shape history, or did history shape Napoleon? Is it individuals or systems that drive change?

The tetrad:

Physics: Geography, climate, disease, calories, metallurgy. Jared Diamond wasn’t wrong about everything — the physical substrate matters enormously.

Math: Network effects. Exponential growth. Power laws. The mathematics of growth and contagion applies to ideas, empires, and plagues alike.

CS: History is the output of a cellular automaton we can’t rewind. Contingency is sensitivity to initial conditions. Path dependence is hysteresis — where you are depends on where you’ve been, not just where you are.

Economics: Malthusian traps. Institutional lock-in. Conquest as hostile acquisition. Civilizations rise and fall based on whether they solve the cooperation problems of their scale.

What falls out:

“Great Man” vs. “Historical Forces” is a false dichotomy. Both are true, at different scales of resolution.

At the macro scale — why did agriculture emerge in the Fertile Crescent? Why did Eurasia dominate? — individual choices wash out. Geography determines. The “forces” win.

At the micro scale — why did WWI start in 1914 rather than 1912? Why did the Archduke’s driver make that wrong turn? — we’re in the chaotic regime. Small perturbations cascade. Individuals matter.

The right question isn’t “which one?” It’s “at what resolution are we predicting?”

This has practical implications. If you’re trying to understand century-scale trends, focus on constraints: resources, geography, technology, demographics. If you’re trying to understand specific events, focus on agents: leaders, decisions, contingency. The tetrad tells you which lens to apply at which magnification.

The Tetrad Turned on Itself

Here’s where intellectual honesty matters. The tetrad is a framework, and frameworks can mislead. What happens when we apply the tetrad to its own examples?

QWERTY keyboards are a classic “path dependence” story. The standard narrative: QWERTY was designed to slow typists down (to prevent jamming on mechanical typewriters). Dvorak is faster. But we’re stuck with QWERTY because of lock-in. Market failure!

This story is too clean.

Let’s apply the economics lens to the story itself:

• August Dvorak designed the Dvorak keyboard

• August Dvorak conducted the Navy study “proving” Dvorak was faster

• Later independent studies (Liebowitz and Margolis) found minimal or no advantage

• The “designed to slow typists down” origin story is probably false anyway — QWERTY was designed to reduce jamming, which is different from reducing speed

So we have: - Dvorak had an incentive to find Dvorak superior — and lo, he did - Economists and journalists have an incentive to tell a good “market failure” parable — QWERTY is such a satisfying story that people repeat it without checking

The lesson: if a story seems too satisfying, ask who benefits from telling it.

This doesn’t mean path dependence is fake. It doesn’t mean lock-in never happens. But this particular example — perhaps the most famous example in the economics of technology — doesn’t actually work.

The tetrad demands rigor. Even for its own favorite stories.

Why We Don’t Get Nice Things

Let’s pull back to the big picture.

The tetrad tells you what’s optimal: the mathematically superior voting system, the economically efficient policy, the scientifically correct answer.

The economics leg tells you why you won’t get it.

Wherever you see a clearly superior solution that isn’t adopted, ask: who chooses the rules, and do they benefit from the current ones?

• Approval voting: Superior. Blocked by the duopoly.

• US healthcare: Wildly inefficient. But efficient for insurers, hospitals, and pharma.

• Academic publishing: Broken. But the prestige cartel benefits incumbents.

• Municipal zoning: Creates housing shortages. But protects incumbent homeowners’ property values.

• The filibuster: Enables minority rule. But senators like having it when they’re in the minority.

In each case, the people who would need to change the system are the people who benefit from it. The math says it’s broken. The economics explains why it stays broken.

This is not cynicism. It’s the opposite of cynicism. Cynicism says “nothing can be changed.” The tetrad says: change requires changing who writes the rules, or changing their incentives.

That’s actionable. It tells you where to focus.

Conclusion

Three posts, eight fields, one framework.

The tetrad isn’t a theory of everything. It’s a method for generating theories. Take any domain, apply four lenses — substance, structure, process, selection — and watch the real questions emerge.

Sometimes you’ll confirm what the field already knows. Sometimes you’ll dissolve fake debates (Universal Grammar). Sometimes you’ll identify impossibility results that constrain what can be achieved (Arrow). Sometimes you’ll catch yourself repeating satisfying stories that don’t survive scrutiny (QWERTY).

The point isn’t to replace expertise with armchair philosophizing. You still need to know the details. But the tetrad tells you which details matter and why.

Physics gives you the substrate. Math gives you the structure. Computer science gives you the dynamics. Economics tells you who’s optimizing for what, and therefore what will actually happen versus what “should” happen.

Put them together, and the world becomes more legible. Not simpler — more legible. You can see why things are the way they are, and what would have to change for them to be different.

That’s the project.

---

This post concludes the series on the tetrad framework. “The Simplicity Tax“ introduced the anomaly that motivated it. “The Tetrad“ laid out the framework and applied it to theology. “The Tetrad Tour, Part I“ covered anthropology, psychology, law, and linguistics. For the full development of these ideas — including the relationship between determinism, computational irreducibility, and free will — see my book The Science of Free Will.