Promises, Threats, and the Thermodynamics of Power

ChatGPT Image Sep 26, 2025, 10_31_08 AM

In politics as in physics, energy without work becomes wasted heat. A statesman may radiate heat—boasts, ultimatums, and gusts of indignation—yet accomplish little if no force is actually applied along a clear line of action. Over time, the system notices. Entropy, in this case, takes the form of diminishing credibility.

Consider a leader fond of superlatives, negotiating with a hard, continental power on his frontier—an autocracy that redraws maps the way a glacier reshapes valleys: slowly when watched, abruptly when conditions permit. The leader brandishes tariffs today, sanctions tomorrow, and victory “by next week” the day after. He threatens, promises, revises, postpones. The spectacle is noisy; the ledger of costs paid is thin.

Does repeated bluffing render threats useless? Not entirely—mass and momentum still matter. Adversaries study capabilities and interests in the present tense. Yet each uncollected ultimatum drains a reservoir you cannot easily refill: the expectation that your next warning means what it says.

Game theory offers the parable of two cars racing toward a cliff’s edge. The classic trick is to throw your steering wheel out the window, converting uncertainty into conviction. Our leader rarely discards the wheel; he prefers options. Options are excellent for bargaining—until they are mistaken for vacillation. The rival power, skilled at probing, tests airspace here, cyber defenses there, and watches which lines are chalk and which are stone.

Ambiguity is not always vice. Deterrence sometimes rests on risk, not promises: raise the probability of consequences, and prudent adversaries pull back. But probabilistic threats still require visible machinery—deployed assets, automatic penalties, alliances that function as more than press releases. Otherwise, “risk” devolves into another word for “maybe.”

Permit me, then, three tidy rules—call them The (Quasi) Laws of Credibility:

  1. A threat must be coupled to a cost already paid or irreversibly set in motion. A battalion moved, a tariff triggered by statute, an air-defense umbrella switched from “exercise” to “on.” Without sunk costs, a threat is theater.
  2. Narrow beats grand. Promise less and enforce all of it. Grandiose red lines invite creative trespass; limited, well-patrolled boundaries build reputations that scale.
  3. Punish small violations swiftly to prevent large catastrophes. The world is an experimenter; if it learns that minor tests are ignored, it scales the experiment.

Balance requires we acknowledge the leader’s method has virtues. Pressing allies to spend more can raise the long-term price of aggression. Flexibility can reveal openings a rigid doctrine would miss; a ceasefire may hide where doctrine saw only stalemate. And there is moral arithmetic to ending a brutal war sooner, even at imperfect terms. But there is also arithmetic in precedent: rewards for conquest purchase future attempts at a discount.

What, then, should a serious power do? Convert declarations into mechanisms. Build sanctions that trigger automatically on specified acts, not on moods. Pair negotiation with deployments that would make a breach expensive tonight, not “after reviews are complete.” Define fewer red lines and wire them to circuit breakers the rival can see. Align rhetoric with timeframes you can actually meet.

A final image. Some say everything can be a cudgel if you swing hard enough. Perhaps; but a cudgel without mass is only air, and an arm that never follows through soon trembles from the effort. The adversary across the frontier understands mass, momentum, and patience. So should we. In the thermodynamics of power, credibility is the one quantity you cannot counterfeit for long.