Why "freedom of speech" is an illusion

If you price speech as a constitutional absolute, you’ll misread the world. Price it as a controlled utility allocated by financial, technical, and reputational switches.

Control-Plane Capital

Oct 16, 2025

black and white cat sitting in front of grey and brown rock boulders during daytime — Photo by Marc A. Sporys on Unsplash

Why “free speech” is mostly a UX layer over control

1) Control lives in the plumbing

Financial rails: Banks, card networks, payment processors, brokerages, ad networks, social media platforms. If they de-risk you, your “speech” loses oxygen (funding, monetization, reach). No statute needed — just “risk” or “brand safety”.
Distribution choke-points: App stores, CDN/anti-DDoS providers, email/SMS gateways, domain registrars, DNS resolvers, certificate authorities, ISPs, cloud. Any one policy toggle ends your reach instantly.
Identity gates: Real-name/age/KYC rules bind speech to reputational and legal exposure; anonymity becomes exceptional and fragile.
Employer leverage: Contracts, “code of conduct”, morals clauses. Employment is a speech throttle hiding inside HR.
Insurance & legal risk: Defamation/PII/“harm” exposure priced by insurers. If you can’t get coverage, platforms won’t host you.
Revealed preference: When speech clashes with stability/brand risk, infrastructure moves first; constitutional rhetoric arrives later to rationalize the fait accompli.

2) The state prefers delegated enforcement

Perimeter governance: Governments “encourage” platforms/banks/clouds to write the rules into ToS/Acceptable Use Policies. Cheaper, faster, deniable.
Standards over statutes: Provenance/watermarking, “online safety”, AML/CTF, “harmful content” policies standardize enforcement knobs across multiple firms and countries.
Emergency elasticity: “Temporary” powers don’t sunset; they become defaults (ratchet effect).
Revealed preference: When legitimacy is strained, officials route censorship through private intermediaries to avoid court fights and headlines.

3) Algorithms launder editorial power

Ranking, demonetization, down-ranking, “quality” scores: Platform speech isn’t binary (on/off); it’s gradient control. A 90% reach cut feels like freedom and performs like a ban.
Model governance: Safety filters are policy encoded as weights. “Guardrails” are de facto editorial lines.
Revealed preference: Platforms posture as neutral while systematically tuning distribution to minimize regulator and advertiser costs.

4) Lawfare and process are the punishment

SLAPP (Strategic lawsuit against public participation) suits, discovery burdens, takedown regimes, compliance overhead: Even if you “win”, you drown in cost/time.
Strict liability shadows: Expanding categories (privacy, IP, “harms”) create permanent uncertainty → over-removal by default.
Revealed preference: Systems don’t need to outlaw speech; they just make it expensive to keep.

5) Social & professional sanctions carry the whip

Ostracism, rating systems, blacklist whispers, reference checks: Speech priced via career and status loss.
Gatekeeper guilds: Editors, moderators, accreditation bodies. Ideology maps to access.
Revealed preference: Belonging beats truth; self-censorship scales better than censorship.

6) “Free speech zones” are sandboxed

Tiny audiences, unstable monetization, brittle infra: Alternatives exist but get starved of payments, ads, distribution, or vendor support.
Revealed preference: The system tolerates enclaves that don’t change mass behavior.

How to read a speech environment in 5 steps

Follow the money: Who pays or can de-platform the payer? (Ads, payment processors, banks.)
Map the choke-points: Which two vendors — app store + cloud, bank + CDN — could end reach in 24 hours?
Find the standards verbs: Attest, verify, provenance, harm, safety, age-appropriate, brand-safe. Those are the dials.
Look for delegated power: “We updated our ToS…” right after a regulatory hearing? That’s the real law.
Check defaults: Is the default setting on for ID-linking, watermarking, automated filters? Defaults decide behavior.

Predictions (based on current incentives)

Provenance-by-default expands (images, audio, text, eventually live): signed content becomes “trustworthy”, unsigned becomes downgrade-by-default.
Identity binds tighten (age/ID + phone/SIM/device attestation): anonymous reach dwindles; anonymity survives for small cliques.
Payments become the master switch: “Harmful” or “non-compliant” content gets starved via card/bank/app-store policy before any courtroom sees it.
AI moderation centralizes: a few foundation models + safety layers set speech norms across most platforms; appeals exist, are slow, and infrequently successful.
Soft bans dominate: Your account exists, your reach does not; your monetization exists, your RPM does not.

Bottom line

If you price speech as a constitutional absolute, you’ll misread the world. If you price it as a controlled utility allocated by financial, technical, and reputational switches, your predictions improve. Incentives beat ideals, control beats fairness, stability beats truth. Act accordingly: build for persistence, not protests.

Other articles I’ve written on investing:

Other articles I’ve written on Bitcoin & Gold:

Subscribe:

Share: