Skip to content

Instantly share code, notes, and snippets.

@afomi
Created October 27, 2025 15:28
Show Gist options
  • Save afomi/0f59b572d5a8bb34c55118c6f597c566 to your computer and use it in GitHub Desktop.
Save afomi/0f59b572d5a8bb34c55118c6f597c566 to your computer and use it in GitHub Desktop.

A primer on Amotz Zahavi’s idea of “costly signals” (often called the handicap principle).

1) The core idea, in one sentence

Some messages are believable because they’re expensive or risky to send; only the genuinely strong/committed can afford them, so the signal stays honest.

2) Signals vs. cheap talk

  • Signal: any observable act meant to convey hidden qualities (strength, reliability, loyalty, competence).
  • Cheap talk: words or gestures that are easy to fake (“Trust me!”).
  • Costly signal: a message that hurts (energy, time, status, risk) in a way that low-quality or uncommitted senders can’t imitate without paying too high a price.

3) Zahavi’s insight (the “handicap”)

Zahavi argued that certain animal displays are handicaps—they reduce survival efficiency but credibly advertise quality. Classic example: the peacock’s tail. It’s huge, conspicuous, and burdensome. A weak male would be eaten or exhausted by such a tail; only a fit male survives while carrying that “handicap.” The cost keeps the message honest.

Other ethology examples you’ll see:

  • Gazelle “stotting” (high, showy jumps facing a predator). It wastes time when you should flee—unless you’re fast enough that advertising your vigor scares the predator off.
  • Birdsong complexity or long courtship dances—time- and energy-intensive, harder for the unfit to fake.

4) Why cost can make truth-telling rational (without economics)

Think of two kinds of senders:

  • High-quality (really strong/committed)
  • Low-quality (weak/uncommitted)

If a signal is cheap, both types can use it; receivers should ignore it. If a signal is costly in a way that hurts low-quality types more, then it “separates” the two groups: high-quality types can pay; low-quality types can’t. Receivers learn to believe it.

Philosophically: the cost is a filter—a practical test that aligns appearance with being.

5) Human examples (no economics required)

  • Rites of passage: endurance trials, fasting, pilgrimages. Costly in discomfort/time, so they credibly indicate commitment to a community or ideal.
  • Monastic vows: poverty, celibacy, obedience—durable constraints that are hard to fake casually.
  • Academic credibility: years of training, hard problems, replication-ready methods—costly preparation that (ideally) sorts sincere inquiry from pretense.
  • Military decorations & elite units: grueling selection and real risks make the symbol believable.
  • Apologies with amends: not just “sorry,” but repair (time, money, public admission). The sacrifice lends sincerity.

6) When is a signal really “costly”?

A useful checklist:

  1. Differential cost: Is it much harder for pretenders than for the real deal?
  2. Visibility/auditability: Can observers see or verify the cost (or its consequences)?
  3. Tied to the trait: Does paying the cost require the very quality being advertised (e.g., stamina for a marathon)?
  4. Hard to outsource: Can’t be faked by borrowing resources for a day.
  5. Repeated or durable: One-off stunts are easier to fake than sustained sacrifice.

7) How costly signals fail

  • Subsidized cost: If you lower the cost (shortcuts, loopholes), fakers flood in and trust collapses.
  • Mere spectacle: If the ostentatious thing is not actually hard for pretenders, it’s fashion, not proof.
  • Audience drift: If receivers stop checking evidence (or are dazzled by style), the ecology favors peacocks without fitness.

8) Ethical reflections

  • Virtue vs. performance: A costly signal can encourage real cultivation (e.g., training) or slide into empty pageantry. The difference is whether the cost nurtures the underlying excellence or merely buys appearances.
  • Community design: If we want sincerity, we may need structures that make faking expensive (time to contribute, peer review, accountability)—but we should also avoid perverse hardship that excludes worthy participants without improving honesty.

9) Simple thought experiments

  • The Bridge Test: Suppose people claim they can carry a 40-kg load across a narrow bridge. If anyone can claim it, the claim is worthless. If the rule is “carry it across in front of everyone,” pretenders self-select out.
  • The Oath with Stakes: “I’ll do X by Friday.” Empty if no consequence. Add a visible stake (e.g., meaningful forfeit donated to a cause you dislike). The cost disciplines speech into truth.

10) How to use the idea in the wild

  • When you encounter a bold claim, ask: What’s the handicap? What makes this hard to fake?
  • When you design a process (community, protocol, hiring, research), ask: What selective cost protects honesty without being cruel or elitist? You’re looking for costs aligned with the virtue you want (effort, time on task, public accountability), not arbitrary hurdles.

TL;DR: Zahavi’s costly signals are credible precisely because they’re hard to fake. The “handicap” aligns words with reality by imposing a selective burden that only the truly capable or committed can bear. In animals, that’s tails and leaps; in human life, vows, proofs, and sustained work. Use costs that cultivate virtue and filter pretense—then watch trust (reasonably) follow.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment