7 March 2025

It's the principle of the thing

by Rachel Shu

Status: will revise
Meaningfulness: effort post

Co-written with Claude (transcript here).

Rachel Shu’s comments:

the integrity term of the equation doesn’t sit right to me. claude added that.
there’s some stuff about being worried about a larger class of similar situation that’s not being caught (or sorta caught, but tied up in the bs integrity term)

Austin Chen’s comments:

seems like this is tracking an important thing
haven’t experienced this too much in my own life
the math doesn’t seem right to me either
having the equation at all seems potentially distracting, as opposed to just explaining it in paragraphs

It’s the Principle of The Thing

People get mad about weird stuff. Your roommate moves your carefully arranged desk. Your coworker takes your pen and doesn’t return it. These are objectively trivial annoyances, yet you find yourself disproportionately angry, and when pressed you say: “It’s the principle of the thing.”

That phrase is a confession. You’re admitting the object-level harm doesn’t really matter. But you’re also saying something deeper. You’re gesturing at an implicit expected value calculation - one I think is worth making explicit.

The math you’re actually doing

When someone crosses a boundary, you’re not just reacting to the immediate harm (often tiny). Your brain is computing something roughly like:

\[P = D + ∑(p_i × F_i) + E + I\]

Where:

$D$ = Direct harm (usually negligible)
$∑(p_i × F_i)$ = Your estimation of probability-weighted harms from similar future interactions
$E$ = Energy you’ll spend maintaining this boundary
$I$ = Integrity cost (does letting this slide make you feel weak, inconsistent, or like a pushover?)

You object “on principle” when $P$ exceeds your personal threshold $T$.

Why this creates interpersonal nuclear war

These outbursts cause unnecessary social friction because:

They see $D$ (tiny), while you’re calculating the entire function
Different discount rates (they undervalue your future convenience)
You don’t know their intent or the probability of recurrence
People have radically different thresholds for what matters

If you don’t have common ground on what “it’s the principle of the thing” mentally stands for, it’s hard to find resolution.

When to just drop it

Install hearing yourself say or think “It’s the principle of the thing” as a trigger for explicitly estimating the above algorithm. A lot of times we feel that a particular value is sacred because either something needs to track a valuable part of it that we’re not explicitly tracking or able to articulate. (I might have more on this in a future post, for now Eliezer Yudkowsky’s post Feeling Moral covers a lot of what I mean to say).

This might serve as a good starting point for communicating what your needs are to the person who’s making you feel hurt in this situation. And at other times, you’ll run this calculation and realize:

\[D + ∑(p_i × F_i) + E + I ≈ 0\]

If the immediate harm is trivial AND you don’t expect recurrence AND standing your ground costs significant social capital AND it’s not core to your identity… then consider whether you’re standing on the right principle.

Better ways to handle principle violations

Instead of getting trapped in “why are you overreacting?” versus “it’s the principle!”:

Name the actual concern: “I’m worried this will become a pattern”
Get future commitments instead of assigning blame
Assess the enforcement cost honestly
Check whether you’re really protecting something important or just feeling neurotic

Principles exist as cognitive shortcuts. They’re cached decisions that save us from having to calculate complex social dynamics from scratch every time. But when defending a principle costs more than the violation itself, your cognitive shortcut has become inefficient.

tags: policy