Circuit Breakers: Preventing Red-State Failures
You wouldn't run production deploys when your system is overheating.
You've built a strategy (Post 12). You have operating rhythms, agreement experiments, and trust systems. But all of that collapses when you're overloaded. The same conversation that runs fine on Tuesday crashes on Friday. What's the variable?
State. Your nervous system's state determines what you can and can't do. Under load, your bandwidth degrades. Error rates spike. Logs become unreliable. That's not "bad communication skills"—it's capacity throttling.
You don't talk your way out of overload. You design circuit breakers that prevent predictable failure modes.
The Arousal Budget Model
Every day, you have a budget. It's determined by:
- Sleep debt: How rested are you?
- Work stress: What's your cognitive load from the day?
- Time pressure: Are you rushing?
- Unresolved resentment: Is there accumulated trust debt?
- Physical factors: Hunger, caffeine, alcohol, illness
When budget is low, conflicts flip into Red faster. The threshold drops. Topics that you'd handle easily at full capacity become triggers.
Arousal Budget Checklist
Before initiating a difficult conversation, run through:
| Factor | Score (0-2) |
|---|---|
| Sleep last night (7+ hrs = 2, 5-7 = 1, <5 = 0) | |
| Work stress today (light = 2, moderate = 1, heavy = 0) | |
| Time available (plenty = 2, some = 1, rushed = 0) | |
| Recent conflict (none this week = 2, resolved = 1, unresolved = 0) | |
| Physical state (fed/rested = 2, okay = 1, depleted = 0) |
Total: ___ / 10
7-10: Good conditions for difficult conversations
4-6: Proceed with caution; use shorter timeboxes
0-3: Reschedule unless urgent
The State Ladder as an Operating Model
Flexible, curious, can hold paradox. Good for: complex negotiations, exploring differences, building agreements. Capacity: full.
Narrowing attention, rising urgency, patience dropping. Good for: simple decisions only. Warning signs: interrupting, absolutes ("always," "never"), time pressure.
Threat system online, bandwidth collapsed. Good for: nothing. Risk: permanent damage to trust, saying things that can't be unsaid, decisions you'll regret.
Circuit Breakers
A circuit breaker is a pre-committed rule that triggers before the crash. You don't decide in the moment—you've already decided.
Define 2-3 circuit breakers. Review after 14 days.
Circuit Breaker Design
| Signal (If I notice...) | Action (Then I will...) |
|---|---|
| Voice speed increasing, interrupting | Call pause, state return time |
| Thinking about rebuttal instead of listening | Ask: "Can I reflect what you said first?" |
| Urge to use "always" or "never" | Switch to: "This time, I noticed..." |
| Partner uses agreed "pause" word | Stop immediately, set return time |
| Arousal budget score below 4 | Don't initiate; schedule for tomorrow |
My 2 Circuit Breakers (14-day trial)
1. If __________, then __________
2. If __________, then __________
Review date: __________
Post-Incident Integration
Every Red-state episode becomes data for your Incident Review (Post 10). Track:
- Trigger chain: What started it? What escalated it?
- First escalation point: Where did Amber tip into Red?
- Circuit breaker status: Which one should have fired? Why didn't it?
- Control to add: What would prevent this specific failure next time?
Incident: Argument about weekend plans escalated to yelling.
Trigger chain: Both tired (Friday night), one partner raised topic while other was cooking, no time budget check.
First escalation: When Partner A said "You always prioritise your friends."
Circuit breaker that should have fired: "Always/never" language → switch to specific instance.
Control added: New rule: no logistics conversations while cooking. Arousal budget check before weekend planning.
Assumption Testing
In Amber and Red, your brain generates false assumptions that feel true. Common ones:
- "If I don't resolve this now, it will never resolve."
- "If I pause, I'm losing the argument."
- "If they loved me, they'd calm down instantly."
- "Their tone proves their intent."
Convert these into testable hypotheses:
| Assumption | Test |
|---|---|
| "If I pause, it won't get resolved" | Pause for 30 min, return, measure outcome |
| "Pausing = losing" | Track: did pausing lead to worse outcomes? |
| "They should calm down immediately" | Test: what's their actual recovery time? Is it consistent? |
Metrics (Without Weaponising)
Track two metrics max. The goal is learning, not prosecution.
- Time-to-repair: How long from incident to functional collaboration?
- Red frequency: How many Red-state episodes per week?
- Successful circuit breaker fires: How often did the pre-committed rule work?
If you use these metrics against your partner ("See? You caused 3 Red episodes this week!"), you're weaponising the system. Metrics are for joint learning. Both partners contribute to state dynamics.
Can't consistently return after a reset?
If circuit breakers keep failing or Red-state episodes are frequent, the system may need redesign—not more intensity. A facilitated session can help identify what's driving the pattern.
Book an AssessmentEducational content. This material is for informational purposes and does not constitute professional advice.