Skip to content

Institutional Legibility, Problem Routing, and Capacity Feedback

Status: Future Topology Area — raw material only. Inclusion here does not mean this area is active, complete, or validated. Title is provisional and may be normalized when the area is developed.

Category: Cross-cutting substrate (parallel to System Resilience, Slack, and Adaptive Capacity — a mechanism that shows up inside many domain-specific topologies rather than belonging to one policy fight).

Parent-of: Carceral System Dynamics (future, not yet drafted) — the carceral system is the richest evidenced case of this mechanism, not its definition.

The Plain Version

Two employees each take about $3,000 from their employer over a year. One is a cashier skimming from the register. The other is a manager who shorts hourly workers’ paychecks by rounding down their hours. The cashier is far more likely to be arrested, charged, and convicted. The manager is far more likely to face nothing at all, or at most a back-pay settlement.

The interesting question isn’t “who’s more sympathetic” or “which is more illegal” — both are crimes, and wage theft is by some estimates a bigger dollar problem nationally than all property crime combined. The interesting question is why the same-sized theft gets routed through completely different machinery. The cashier’s theft is caught on camera, has one obvious suspect, fits a crime category police handle every day, and prosecuting it costs the DA’s office nothing politically. The manager’s theft requires payroll records, a labor investigator, possibly a civil attorney, and going after an employer who may have a lawyer, a lobbyist, or a relationship with the local business community. Nothing about the moral status of the two acts explains the gap. Something about how each one moves through institutions does.

This area is about that “something”: the checkpoints an ambiguous social problem has to pass before it becomes an institutional response at all, and what happens to an institution once it starts routinely clearing a particular kind of checkpoint.

Core Diagnostic Question

How does existing institutional structure affect which ambiguous social problems become legible, where they get routed among competing institutions, how they are processed and measured once routed, and whether that processing changes the odds that similar problems get routed there again?

Keeper sentence (candidate): A tool’s capacity to act changes which problems come to be seen as belonging to it.

The Checkpoints

Using the wage-theft-vs-shoplifting case as a walkthrough. Each checkpoint below is a separate claim with its own kind of evidence — a proposed factor doesn’t belong in this area’s issue-node list until someone can say which checkpoint it affects.

  1. Legibility — can the institution even perceive this as a case? Shoplifting: yes, instantly — there’s camera footage and a person walking out with merchandise. Wage theft: only if someone pulls and interprets payroll records across pay periods, which nobody does by default. (Source: Scott, Seeing Like a State — institutions can only act on what their existing paperwork and categories let them see.)
  2. Jurisdiction — assuming it’s visible, is this actually this institution’s problem? Shoplifting: unambiguously a police matter. Wage theft: could go to a state labor board, a civil court, or occasionally a prosecutor — three doors, no obvious front one.
  3. Tractability — is there something concrete enough to act on: an identifiable actor, usable evidence? Shoplifting: one person, one incident, one camera. Wage theft: a pattern spread across pay stubs, time clocks, and employment law, often requiring an accountant to even total up.
  4. Capacity — does the responsible institution actually have the staff and time to act right now? A police department has patrol officers who write up shoplifting as routine. Most states have a tiny wage-and-hour enforcement staff covering the entire workforce.
  5. Routing incentives / resistance — even where the institution could act, is it worth it to them? Net effect of political cost, organizational payoff, the target’s power to push back, legal exposure. Charging a shoplifter generates no blowback. Going after a well-connected employer might.
  6. Interpretive modifiers — how the act and the actor get framed, which mainly works by shifting checkpoint 5. Moralization, racialization, threat framing, who reads as sympathetic versus dangerous, whether the story told about the act is “crime” or “business practice.”
  7. Institutional processing — the domain-specific machinery once something clears the checkpoints: arrest, adjudication, sentencing; or licensing action, diagnosis, a civil judgment, depending on the domain.
  8. Outputs & measurement — what gets processed generates its own numbers: arrests, convictions, closed cases. What gets prevented, or never routed anywhere, generates no numbers at all — so it’s structurally invisible to whoever is deciding where to send next year’s budget.
  9. Environmental feedback — processing shapes the public story about what the “real” problem is, which shapes resourcing, which shapes what gets noticed as legible next time → loops back to checkpoint 1.

Note on 5–6: these aren’t “outside pressure vs. inside decision.” A prosecutor’s own office culture can supply exactly the same kind of pressure a donor or a news cycle would — see Interface Notes below. What matters is which checkpoint a given pressure is acting on, not whether it originated inside or outside the building.

Major Issue Nodes

  • Legibility Gap
  • Jurisdictional Ambiguity
  • Tractability Threshold
  • Processing Capacity Ceiling
  • Political / Organizational Routing Cost
  • Interpretive Framing Shift
  • Measurement Legibility Bias (countable outputs crowd out prevented or upstream outcomes in political attention and budgeting)
  • Net-Widening (new or expanded capacity pulls in marginal/ambiguous cases instead of only replacing harsher existing processing)
  • Path Dependence / Sunk Institutional Capacity
  • Institutional Narrative Production
  • Informal Constraint Formation (candidate interface node — see below)

Causal Chains

Chain 1 — Net-widening: An institution’s capacity expands → cases that used to be too marginal or ambiguous to process become processable → cases that would previously have gone unrouted (or routed elsewhere) get pulled in → the total processed population grows → the “problem” looks bigger than it did before → the growth gets read as justification for still more capacity.

Chain 2 — Measurement bias: Institutional outputs are countable (arrests, convictions, closed cases) → prevented or upstream outcomes aren’t → political attention and money follow the numbers that exist → capacity keeps building around the countable activity → whatever would have addressed the problem further upstream stays comparatively starved.

Chain 3 — Interpretive-modifier routing: An ambiguous case plus a framing (moralized, racialized, “dangerous” vs. “a business dispute”) → shifts the political cost of acting one way instead of another → routing changes even though no statute changed.

Chain 4 — Institutional narrative feedback (tentative, flagged unresolved): An institution’s capacity and output shape the public story about how big and dangerous the underlying problem is → that shapes informal pressure on legislators and administrators → that shapes future resourcing and routing → which can reinforce the original capacity. Status note: don’t default to calling this either “legitimate public response” or “manufactured demand.” A real underlying trend, selective emphasis by the institution, and outside amplification can all be true in the same case at once. This chain needs case-specific evidence before any link in it gets asserted rather than proposed.

Source Branches

Legibility Branch (Scott): institutions can only act on simplified, administratively usable representations of reality; whatever doesn’t fit that representation tends to become invisible to the institution regardless of how important it actually is.

Coupling Branch (Kingdon, Multiple Streams): problems and solutions don’t just match up in the obvious direction (problem appears, then a solution gets invented for it). Existing, available solutions get coupled to whatever problem opens a window — meaning the capacity that already exists shapes which problems get addressed and how.

Net-Widening Branch (criminology): interventions built as lighter alternatives to existing processing can end up expanding the total population caught in formal processing, instead of only replacing harsher processing that would have happened anyway.

Budget-Maximizing Bureaucracy Branch (Niskanen): administrators have a standing incentive to defend or grow their budget and scope, independent of the actual merits of any specific caseload.

Capture Branch (regulatory capture literature): institutions built to process or oversee something can end up oriented toward the interests of what they’re processing, regardless of their stated mandate.

Interface Notes (explicitly not established mechanisms — don’t treat these as causes yet)

CivMMI interface: a society’s metacognitive/civic capacity may limit its tolerance for distributed, delayed, probabilistic causal explanation compared with a simple “someone did something bad, punish them” explanation, which would in turn affect which mechanism above dominates. Keep this decoupled from any causal claim for now. Right now the only evidence for “low CivMMI” in a given case is the presence of exactly the punitive/simple-attribution behavior it’s being used to explain — that’s circular. Before this can be treated as a cause rather than a relabeling, it needs a measure that comes from somewhere else entirely — e.g., a society’s demonstrated tolerance for diffuse, delayed causal framing in epidemiology, climate attribution, or systemic financial risk.

Informal Constraints interface (Working Concepts): per the existing Working Concepts rule, this lens gets used only where formal authority, capacity, and visible material incentive don’t adequately explain what an institution actually did — not reached for by default. It’s a candidate carrier for how an interpretive modifier (checkpoint 6) actually turns into behavior at the routing-incentive checkpoint (5) — not the organizing theory of this whole area. Any specific Informal Constraint claim here needs its own detection-confidence and strength rating, stated as a hypothesis to check rather than something asserted outright. And naming a constraint out loud can itself change it once the naming reaches people who can act on it — it can weaken the constraint by exposing it, or strengthen it by giving someone a label to weaponize (“soft on crime”). Don’t assume a direction.

Evidentiary Discipline

To stop this area from becoming a drawer where any historical outcome can be explained after the fact by whichever factor is convenient:

  • State which checkpoint (1–9) a proposed factor affects before adding it to the issue-node list.
  • State what evidence would confirm or rule out a proposed factor, and write that down before checking the historical case, not after.
  • Tag causal claims at one of four tiers. This is independent of Informal Constraint confidence/strength — tier says what kind of claim is being made; confidence/strength says how well-supported a specific claim within that tier is.
    1. Observed outcome — a disparity or pattern exists.
    2. Mechanism existence — solid (often experimental) evidence that a proposed mechanism can operate in general.
    3. Episode-specific causation — evidence that a mechanism materially drove one particular historical routing decision.
    4. Cross-temporal / cross-case pattern — evidence that a mechanism recurs across several genuinely comparable episodes, with competing explanations ruled out and the effect predicted in advance rather than explained after the fact.

Related Topology Areas

Draft prepared outside the NeuroSaeculum CMS for review before entry. Terminology, including the working title, is provisional.