Skip to content

Destabilization Depth and Breadth Monitor

Monitor — instantiates Reopened Malleability Window

Tracks the induced labile interval in real time — how deeply the configuration has loosened and how far the loosening has spread — against a pre-set boundary.

Once a trigger has reopened the window, the system sits in a hazardous intermediate: soft enough to change, and therefore soft enough to be damaged. Destabilization Depth and Breadth Monitor watches that interval as it unfolds, tracking two things a single readout would blur — depth (how far the target configuration has actually loosened from its stable state) and breadth (how far that loosening has spread beyond the target into functions that were supposed to stay put). Its defining move is treating the labile state as a bounded envelope with dimensions to hold, not a binary "open," so that "opening too deep" and "opening too wide" are separate alarms with separate thresholds. It measures the destabilized target; it is the instrument that tells the stop authority when the envelope is being breached.

Example

A mid-size company is unfreezing a rigid regional-approval norm so it can be rewritten during a reorganization — a deliberate Lewin-style unfreeze of something that had hardened over a decade.[1] Leadership wants the approval norm loosened, but not the safety and compliance habits tangled up next to it. The monitor tracks depth by watching how far the target norm has actually come unstuck (are teams genuinely questioning it, or just nodding?), and breadth by sampling adjacent norms for collateral loosening (are people now also treating safety sign-offs as negotiable?). Early in the reorg the readout shows healthy target depth and contained breadth. Two weeks in, breadth starts climbing — the destabilization is bleeding into compliance behavior — and the monitor flags it well before the change lands, giving the stop authority a reason to narrow the intervention before a protected norm comes loose.

How it works

The monitor keeps the labile interval inside a declared envelope by tracking its dimensions separately:

  • Depth tracking — measure how far the target has actually loosened, distinguishing genuine lability from surface compliance or mere arousal, so induction success is not confused with agitation.
  • Breadth tracking — sample the target's neighbourhood for spread, so loosening that leaks into protected functions registers as a distinct, earlier warning than mere over-depth.
  • Envelope comparison — compare both against the pre-set onset, depth, breadth, and duration bounds, and raise a graded alarm as any dimension approaches its limit rather than after it is crossed.

Tuning parameters

  • Depth threshold — how loose is loose enough versus too loose. Set shallow, you may pair corrective input before the system can accept it; set deep, you risk a state too labile to re-stabilize cleanly.
  • Breadth sensitivity — how readily spread into adjacent functions trips an alarm. High sensitivity protects neighbours but generates false alarms from normal cross-talk.
  • Sampling cadence — how often the interval is read. Faster catches a fast-moving breach but is costly and noisier; slower can miss a transient over-shoot.
  • Alarm gradation — how many warning bands precede the hard limit. More bands give the stop authority room to act early; fewer make the signal cruder but harder to ignore.

When it helps, and when it misleads

Its strength is separating two hazards that a scalar "openness" reading fuses: a window can be open to the right depth yet spreading dangerously wide, or narrowly contained yet far too deep. By dimensioning the labile state it lets the pairing and the stop decision react to the right hazard.

Its failure modes come from measuring the wrong thing. Arousal, distress, or effort can masquerade as depth, so a monitor that indexes activation rather than genuine editability will report a window that isn't there. Breadth is easy to under-sample — the protected function that comes loose is often one nobody instrumented. And the readout tempts a false calm: staying inside the envelope means the state is bounded, not that the right thing is being learned. The discipline is to validate the depth signal against genuine editability, sample breadth wider than feels necessary, and treat the monitor as an input to the stop authority rather than a substitute for it.

How it implements the components

The monitor realizes the archetype's containment instrument — the component that governs the labile interval while it is open:

  • destabilized_state_boundary — its whole function: characterizing and tracking the labile interval by onset, depth, breadth, duration, and closure conditions, and signalling when the envelope is approached.

It does NOT decide induction has succeeded (Reopening-Signal Verification Panel), probe specific protected functions for lasting damage (Non-Target Change Probe Battery), or hold the authority to halt and reclose (Adaptive Stop, Reclosure, and Rescue Protocol) — it supplies those with the live picture they act on.

Notes

Depth and breadth are genuinely different alarms, and conflating them is the monitor's most common failure. A window that is open just deep enough can still be spreading too wide, and a tightly contained window can be far too deep to re-stabilize — the two dimensions must carry separate thresholds and separate responses, or the one that isn't being watched is the one that causes the harm.

References

[1] Lewin's unfreeze–change–refreeze model of planned change — the target is deliberately loosened from its settled state before being reshaped and re-set. It is the organizational anchor for reopening a hardened configuration on purpose, and it is why the monitor watches an induced loosening rather than a naturally receptive phase. ↩