Add/Remove Symmetry Audit¶
Diagnostic audit — instantiates Ratchet Control and Release Design
Measures, tooth by tooth, how much easier a system makes adding than removing — turning a vague sense of creep into a scored, side-by-side asymmetry map.
An Add/Remove Symmetry Audit takes each class of increment a system accumulates — a requirement, a control, an approval step, a feature flag — and compares, side by side, the real cost of adding one against the real cost of removing one. It does not set caps, assign owners, or build release paths; its single job is to make the direction-asymmetry visible and to score it, so a diffuse worry — "we keep piling things on and never take them off" — becomes a specific finding: "adding a requirement takes one committee vote; removing one takes a two-term review, three sign-offs, and has no owner." Its defining move is the paired comparison: every tooth-class is rated on both directions at once, because a ratchet is defined not by how hard removal is in absolute terms but by the gap between the two directions. The audit is where the ratchet's hidden geometry — easy one way, hard the other — is first named and measured, before any intervention is designed.
Example¶
A university's engineering faculty suspects its degree has quietly bloated: students now take 138 credit-hours where the accreditor requires 120, and nobody can say when it happened. The associate dean runs an Add/Remove Symmetry Audit. She enumerates the tooth-classes — required courses, prerequisite chains, lab hours, capstone deliverables, mandatory workshops — and for each documents the actual path to add versus remove. Adding a required course: one curriculum-committee memo, approved in a single meeting, sponsored by whichever faculty member wants it. Removing one: a full program-review cycle, a vote by every affected department, a teach-out plan for enrolled students, and — decisively — no standing owner who benefits from the removal. She scores each pair on effort, elapsed time, number of approvers, and whether a sponsor exists on each side.
The output is a one-page asymmetry map: for six of eight tooth-classes, adding is five-to-ten times cheaper than removing, and three have literally no removal path at all. The map fixes nothing, but it settles the argument. The faculty stops debating whether there is a ratchet and starts deciding what to do about the three ownerless tooth-classes the audit flagged as most asymmetric.
How it works¶
- Enumerate tooth-classes, not instances. The audit works at the level of kinds of increment (a "required course," a "compliance control"), so the finding generalizes rather than drowning in individual cases.
- Trace both real paths. For each class, document the add path and the remove path as they exist today — who approves, how long it takes, what evidence is demanded, who sponsors — the lived route, not the official policy.
- Score the gap. Rate add and remove on comparable axes (effort, latency, approver count, sponsor presence) and compute the asymmetry as the difference, so classes rank by how ratcheted they are.
- Flag the ownerless. The sharpest signal is a removal path with no owner and no sponsor; the audit surfaces those first, because they are where teeth accumulate silently.
Tuning parameters¶
- Granularity of tooth-classes — coarse buckets are fast but blur distinct dynamics; fine classes catch real differences but cost more to trace.
- Scoring axes — effort-only is quick; adding latency, approver count, and sponsor-presence gives a richer picture at more cost.
- Real-path vs. policy-path — audit the route people actually use (accurate, harder to gather) or the documented procedure (fast, often fiction).
- Sampling depth — trace every tooth-class exhaustively, or sample the highest-volume ones; deeper is more complete, shallower is repeatable more often.
When it helps, and when it misleads¶
Its strength is that it converts a contested feeling into a scored, checkable diagnosis and points precisely at the tooth-classes where the asymmetry is worst — the natural starting list for caps, owners, and release paths. It also names why removal is hard, which is often status quo bias[1] wearing the costume of prudence.
Its failure mode is that an audit describes and does not cure: a beautiful asymmetry map with no follow-through just documents the ratchet more precisely. It can also mislead by scoring only what is easy to measure — effort and approvals — while missing the political and reputational costs that make removal hard in ways no rubric captures, lending false symmetry to a comparison that is really about who gets blamed. The classic misuse is to run the audit, circulate the finding, and treat the awareness itself as the intervention. The discipline that guards against this is to hand every high-asymmetry finding to whoever owns the follow-through with a mandate to close the gap, rather than filing the report.
How it implements the components¶
ratchet_direction_map— the paired add-vs-remove trace for each tooth-class is the direction map: it names the easy-forward and hard-reverse path explicitly and per class.ratchet_tooth_definition— enumerating tooth-classes forces an explicit definition of what counts as one preserved increment, the precondition for governing it.
The audit only diagnoses. It does not keep a running record of events (cumulative_displacement_ledger — that's Ratchet Event Log) nor set a ceiling on total burden (displacement_cap_or_budget — that's Cumulative Impact Budget); it hands its findings to those mechanisms.
Related¶
- Instantiates: Ratchet Control and Release Design — the audit supplies the direction map the rest of the structure is built on.
- Sibling mechanisms: Cumulative Impact Budget · De-Escalation Gate · De-Ratcheting Sprint · One-In/One-Out or Cap Rule · Ratchet Event Log · Ratchet Threshold Dashboard · Stale Tooth Review · Sunset Clause Register · Rollback Runbook
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: The mechanism measures, tooth by tooth, how much easier a system makes adding than removing — turning a vague sense of creep into a scored, side-by-side asymmetry map, so its operative form is a bounded assessment of existing evidence or work.
Independent corroboration: The frozen evidence defines Add/Remove Symmetry Audit as 'Measures, tooth by tooth, how much easier a system makes adding than removing — turning a vague sense of creep into a scored, side-by-side asymmetry map', so its operative form is Assessment, Review & Assurance.
Nearest alternative: Analysis, Modeling & Optimization — It evaluates an existing system's add/remove asymmetry and returns an audit finding rather than deliberately perturbing the system to learn.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Organizational & Management Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Universal
Rationale: Process and governance audits compare the approvals, ownership, cost, and time required to add versus retire requirements, controls, and organizational commitments.
Related originating lineages:
- Accounting & Auditing — Paired evidence, scored criteria, and control-by-control findings contribute the audit form.
- Behavioral Economics — Status-quo bias, loss aversion, and endowment effects explain why removing an existing element is treated as riskier than declining to add it.
- Law & Governance — Rulemaking versus repeal and institutional ratchet effects supply a major application.
- Public Administration & Policy — Institutional ratchet and policy-accumulation research show how programs and rules acquire constituencies and asymmetric repeal procedures.
- Systems Thinking & Cybernetics — Irreversibility and one-way accumulation supply the ratchet model.
Review resolution: The paired comparison of organizational add and remove paths is primarily a management diagnostic. Audit, behavioral lock-in, legal and public rulemaking, and ratchet feedback materially shape the method; its logic applies to accretive systems broadly enough to retain universal reach.
Attribution caveat: The paired symmetry score is an encyclopedic diagnostic rather than a standard named audit, though each contributing tradition is established.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; medium confidence.
References¶
[1] Samuelson, W., and R. Zeckhauser. "Status Quo Bias in Decision Making". Journal of Risk and Uncertainty 1(1): 7–59 (1988). Documents a systematic tendency to retain the status quo when alternatives are available. registry ↩