Classification Markedness Audit¶
Taxonomy audit — instantiates Marked Default Audit
Takes a taxonomy's categories as given and traces how its default/special marking drives routing, reporting, and treatment, then restructures the categories and monitors that the asymmetry doesn't creep back.
Classification Markedness Audit operates one level below surface names — on category structure and its operational consequences. It treats the marked/unmarked pair as an existing feature of a taxonomy — eligibility classes, tags, routing codes, report dimensions — and asks how that marking flows into differential treatment: who gets routed to exception handling, who is invisible in reports, who waits longer. Its move is evidence-driven structural repair plus recurrence monitoring, not lexical reading. This is the one idea that separates it from its near-twin, the Naming Markedness Audit: the naming audit reads the words on labels and stops at interpretation; this audit ignores wording, takes the categories as given, follows their structure into consequence, and rebuilds it.
Example¶
A state benefits agency's case-management system tags applications "expedited," "complex," or "special-circumstance," and leaves everything else on an implicit, untagged standard track. Pulling routing logs and queue data (asymmetry evidence), the audit finds the untagged bucket silently absorbs everyone who fits no named tag — including applicants with disabilities whose accommodation needs match no existing tag, who then sit in the general queue with no flag at all (the affected group). The audit decides the tag schema itself is the revision target, then restructures it so accommodation need is a first-class parallel tag rather than an unnamed residue (label or category revision). Finally it wires a monitor that alarms when a rising share of cases fall into the untagged bucket (validation and monitoring check) — catching the archetype's subtlest failure, a new hidden default re-forming after the fix.
How it works¶
- Take the taxonomy as given. Don't re-litigate what categories are called; the words are not the object.
- Pull operational evidence. Routing paths, SLAs, report coverage, and treatment differences broken down by category.
- Find the silent residue. Where does the unmarked or default category absorb cases, and who does that advantage or harm?
- Choose the revision target — a tag, a class, a routing rule — and restructure it so parallel cases are marked on the same dimension.
- Monitor for recurrence, watching for both re-marking and the loss of a necessary distinction.
Tuning parameters¶
- Evidence depth — from a schema inspection to a full routing-and-outcome analysis. Deeper catches real consequence but costs analysis time.
- Restructure aggressiveness — add one parallel tag versus redesign the class system. More aggressive removes more hierarchy but disrupts records and legal references.
- Necessary-marker preservation — how strongly to protect operationally required distinctions (safety, legal, accessibility) against a symmetry push.
- Monitoring sensitivity — the threshold at which the drift alarm fires. Tighter catches recurrence earlier, at the cost of more false alerts.
When it helps, and when it misleads¶
Its strength is that it is the only sibling that follows marking all the way into real treatment and then closes the loop with monitoring, so it catches defaults that hide in behavior rather than in words — a pattern well described by the sociology of the unmarked[n1], where the dominant category goes untagged precisely because it is treated as generic. Its failure mode is overcorrection: flattening a distinction that actually protected people — an accommodation flag, a safety class — in the name of symmetry. The classic misuse is deleting a "special handling" tag that legal or accessibility processes silently depend on. The guarding discipline is to gate every restructure on whether the marking is justified — recording it via Style Guide Revision when it is — and to keep the monitor watching for both re-marking and lost-necessary-markers.
How it implements the components¶
asymmetry_evidence— gathers routing, report, and treatment data showing the marking's real operational effects.affected_case_or_group— identifies who the residual or marked category advantages or harms in practice.revision_target— decides that the taxonomy's structure (tags, classes, rules) is what changes.label_or_category_revision— restructures categories so parallel cases are marked on the same dimension.validation_and_monitoring_check— monitors that the old asymmetry, or a new hidden default, does not re-form.
It does not read surface marked_variant labels or infer the hidden_norm from wording — that's its near-twin Naming Markedness Audit, which stops at detecting and interpreting names; this audit takes the categories as given and instead traces their operational consequences and rebuilds the structure.
Related¶
- Instantiates: Marked Default Audit — carries the audit through structural repair and monitoring at the taxonomy level.
- Consumes: Search and Documentation Scan supplies the cross-artifact evidence the restructure rests on.
- Sibling mechanisms: Naming Markedness Audit · Default Label Review · Inclusive Language Markedness Review · Symmetry Labeling Matrix · Style Guide Revision · Search and Documentation Scan · Stakeholder Interpretation Review
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: The audit examines routing, service, reporting, and treatment evidence by category to determine where default or special marking creates asymmetric effects, so its defining form is a bounded taxonomy assessment.
Nearest alternative: Intervention, Treatment & Transformation — The audit may lead to restructuring a tag, class, or routing rule, but that correction follows from the evidence-based markedness finding rather than replacing the audit as the core form.
Review outcome: Adjudicated after independent review; medium confidence.
Origin Attribution¶
Primary origin: Linguistics & Semiotics
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Linguistic markedness established the default-versus-special asymmetry that this audit follows into taxonomy structure and treatment.
Related originating lineages:
- Gender Studies & Queer Theory — Feminist markedness analysis exposed how an apparently neutral default makes other identities exceptional.
- Library & Information Science — Taxonomy governance supplies the operational merge, split, and recurrence controls.
Review resolution: Both reviewers agree on linguistics_semiotics as primary. The source mechanism's defining operation supports that lineage; the reconciled record retains gender_studies, library_information_science only where it materially contributes the mechanism, and treats later application breadth separately from origin.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] The sociology of the unmarked (e.g., Wayne Brekhus's work) — dominant or default social categories tend to go unnamed and unexamined precisely because they are treated as the generic, neutral baseline, while non-default categories are heavily tagged. In a taxonomy this shows up as the untagged residue that absorbs whatever the named tags fail to capture. ↩