Why you lose against your own resolutions even though both sides are right — and why the one instance that could arbitrate is always a party to the dispute.
"Do I contradict myself? / Very well then I contradict myself, / (I am large, I contain multitudes.)"
— Walt Whitman, Song of Myself (1855)
7. July 2026
Playground
This article is based on a hypothesis, not a case study. It asks whether PI has a lower bound—and finds none.
Paradoxical interactions require multiple rational actors. Until now, that has been the framework’s unspoken lower limit: no PI with fewer than two people. The hypothesis of this text: The lower limit does not exist. A single person over time ismultiple actors. The “I” that makes plans on Sunday and the “I” that breaks them on Friday both act rationally—and collide structurally, not in terms of character. What’s missing here is the only thing that could defuse the collision: feedback that doesn’t originate from the same loop. Where it’s missing, the system feeds back on itself. This has a common name. It’s called brooding.
Sunday evening, 9:40 p.m. Someone is sitting at the kitchen table planning the week. Go for a run at 6:30 a.m. on Monday. No alcohol until Friday. Two hours on the manuscript in the evening. The plan is good. It’s realistic, well-thought-out, and made with a clear head.
Friday evening, 10:10 p.m. The same person, on their third glass of wine, cancels Saturday’s workout. The week was tough, the body is tired; rest isn’t an excuse, but a physiological necessity. This argument is sound, too.
Two decisions, both reasonable, both made by the same person. Except the two have never met.
The Negotiation Table with Only One Chair
Imagine a negotiation in which the parties are never in the room at the same time. The “Monday Me” leaves behind decisions. The “Friday Me” leaves behind faits accomplis. They communicate exclusively through what they leave behind—the workout plan on the fridge, the empty bottle in the glass recycling bin. Neither side can question, persuade, or engage the other. Each rules absolutely as long as it rules. This is not a lack of willpower. A lack of willpower would presuppose that there is a will that is being weakened. Here there are two wills, both intact, both rational—they occupy
The Nobel Laureate and the Cigarette
Thomas Schelling helped shape the logic of deterrence during the Cold War. He understood negotiations, credible threats, and self-commitment—for which he was awarded the Nobel Prize in 2005. And he smoked. For decades, he tried to quit, and the game theorist did what game theorists do: he analyzed his opponent. The opponent was himself, a few hours later.
Schelling described people who hide their cigarettes from themselves. Who set the alarm clock across the room because the “self” at 6:30 a.m. doesn’t trust the “self” at 11:00 p.m. He called this “self-command”: strategies of one party against another party that just happens to inhabit the same body. George Ainslie later modeled this same structure economically—successive “selves” haggling over the same resources, each with a different assessment of the future, each right in its own moment.
The man who modeled superpowers failed against an opponent with exactly his own intelligence. That’s no coincidence. It’s the most precise formulation of the problem: Against oneself, every strategy is an insider leak.
The Closed Loop
Now for the structural shift. PI requires feedback—actors whose rational actions influence one another. When n=1, this feedback exists. It has only one distinctive feature: it never leaves the system.
The Monday-me responds to the Friday-me with stricter plans. Friday-Me responds to stricter plans with a greater need for rest. Each cycle intensifies the next. Luhmann described systems that close themselves off from their environment and reproduce only their own elements—he was referring to organizations, functional systems, and schools of thought. The hypothesis here is that the smallest autopoietic system is a single human being who feeds back on themselves.
And then what every closed loop does happens: it calls in a mediator. A third “I” emerges—the Sunday “I”—which observes the conflict “from the outside.” But there is no “outside.” The mediator is produced by the same structure as the disputing parties, with its own interests in the outcome—usually self-reassurance. Every self-diagnosis is made by one of the participants.
Psychology refers to the continuous operation of this loop as rumination: the same material, run through again and again, without anything new entering—because nothing new can enter. Brooding is not a thought disorder. It is a closed system that functions flawlessly.
The Monday-Friday PI
A person’s successive states each act rationally, but never share the same moment. The only available feedback comes from the system itself—including the entity tasked with mediating the conflict.
Everyone acts rationally:
- The Monday self—plans rigorously (rational: Nothing happens without commitment)
- The Friday self—breaks the plan (rational: Exhaustion is real; recovery is necessary)
- The Sunday self—mediates gently (rational: Without self-reassurance, Monday is unattainable)
- Result—the stumbling: structural, unintended
Everyone is guilty. No one is to blame.
And here, only here, comes the addition that no other PI allows: Everyone is you.
The Half-Second Idiot
Navigation, not a solution. If the diagnosis is correct, it leads to an uncomfortable conclusion: self-help—when taken literally as self-help—is a call for a closed loop to open itself up using its own resources. The loop will deliver what it always delivers—another round. Insight is not a way out. Nowhere is this statement more stark than here: Insight is produced by the very structure it is supposed to see through.
What the loop cannot generate on its own is feedback from the outside. Not because the other is smarter—he is just as incomplete. But because his mistakes are different mistakes. That is the silent core of the Two-Idiots Theorem: Two incomplete systems do not correct each other in spite of, but because of, their differences. The One-Person PI is proof by negation—what is missing when the second person is absent reveals what the second person contributes.
One loose end remains: the diary. Yesterday’s written self provides feedback—but it’s made of the same material. Half a second idiot. Research on expressive writing shows exactly this pattern: It helps measurably. And never completely. Writing externalizes the loop without leaving it.
That’s all there is to it. The Monday “I” will keep planning, the Friday “I” will keep breaking, the Sunday “I” will keep mediating. The question isn’t how to end the argument. The question is whether there’s occasionally someone sitting at the table who isn’t wearing the same jacket.
Related Posts
— When Understanding Doesn't Set You Free
How systems theorists reproduce the enclosure milieus Luhmann warned against
The Phenomenon of Coupled Oscillators
Why systems that analyze closure risk reproducing it — and why structural awareness does not automatically prevent structural hardening
On piinteract.org:
- ["Shared Blindspot"] — When all parties involved are the same self, all parties share the same blind spot. The arbitrator included.
- ["Antagonistic Sparring"] — The only correction the loop didn't produce itself comes from a counterpart with different errors.
- ["Perspective Switching"] — The loop simulates the switch of perspective. Every perspective originates in the same system.
- ["More of the Same"] — Stricter plans against broken plans: the anti-practice in its most private form.
See also (external links):
Thomas C. Schelling — Biographical (NobelPrize.org) — The institutional record of the man who modeled deterrence between superpowers — and self-command against his own later state.
Picoeconomics — George Ainslie — Primary source from the researcher who modeled successive selves as bargaining parties with competing interests.
Constructive and Unconstructive Repetitive Thought (Watkins, Psychological Bulletin) — Peer-reviewed: repetitive thought as a process that can run constructively or closed — the empirical side of the loop.
Emotional and physical health benefits of expressive writing (Baikie & Wilhelm, Cambridge) — The finding behind the half second idiot: writing helps measurably — and does not replace the outside.
Personal Identity (Stanford Encyclopedia of Philosophy) — The philosophical groundwork of the hypothesis: what makes a person over time one — or several.
Paradoxical Interactions (PI): When rational actors consistently produce collectively irrational outcomes — not through failure, but through structure.
All are guilty. None are at fault.
Peter Senner Thinking beyond the Tellerrand
contact@piinteract.org
https://piinteract.org
Co-created with Claude (Anthropic) — two incomplete systems making each other's gaps visible.