Power through the Lens of Relational Ontology — 6. Maintenance, Repair, and Failure: The Engineering of Stability

It is easy to imagine power as something exercised in moments of decision, conflict, or visible intervention.

But most of what sustains a world is not dramatic.

It is:

maintenance.

And maintenance is precisely where power becomes most operationally real.

From command to continuity

If earlier chapters emphasised power as:

• constraint modulation,
• architectural alignment,
• operational execution,
• and production of possibility,

then we now encounter a crucial refinement:

Power is not only what produces structure.

It is:

what prevents structure from falling apart.

This shifts attention from:

• exceptional acts of control
  to
• continuous engineering of stability.

What is maintenance?

Maintenance is:

the ongoing reproduction of the conditions under which constraint architectures remain coherent over time.

It includes:

• repairing broken procedural chains,
• updating institutional categories,
• restoring infrastructural functionality,
• recalibrating semantic systems,
• and re-synchronising temporal coordination.

Maintenance is not supplementary.

It is:

constitutive of worldhood.

Without it, no constraint architecture persists.

Stability is not a state — it is an activity

A central misunderstanding is to treat stability as a condition a system has.

Relationally, stability is:

an ongoing achievement produced through continuous corrective activity.

Worlds remain stable because:

• misalignments are corrected,
• disruptions are absorbed,
• inconsistencies are managed,
• and fractures are repaired.

Stability is therefore:

the outcome of persistent operational labour across distributed systems.

Repair as constraint re-alignment

Repair is not simply fixing broken components.

It is:

the re-alignment of constraint relations after disruption.

A breakdown in one layer rarely remains isolated.

For example:

• an infrastructural failure may disrupt economic flows,
• which alters institutional procedures,
• which shifts semantic categorisations,
• which modifies expectations and behaviours.

Repair therefore requires:

cross-layer coordination of constraint re-stabilisation.

It is architectural work, not local patching.

Why failure is structurally inevitable

No constraint architecture is perfectly stable.

Failure arises because:

• systems are heterogeneous,
• couplings are partial,
• environments change,
• and internal complexity generates drift.

Failure is not exceptional.

It is:

an intrinsic feature of distributed relational systems under continuous stress.

This means:

• instability is not a deviation from normality,
• but a condition that normality must actively manage.

Power as failure management

At this level, power is revealed as:

the capacity to manage, absorb, and redistribute failure without systemic collapse.

This includes:

• buffering disruptions,
• isolating breakdowns,
• rerouting coordination flows,
• and restoring functional coherence.

Power is therefore not only productive.

It is:

reparative.

Institutions as maintenance systems

Institutions are often described in terms of authority or function.

But operationally, they are better understood as:

distributed maintenance systems for constraint architectures.

Examples:

• legal systems manage normative consistency and dispute resolution,
• education systems maintain epistemic continuity and subject formation,
• financial systems maintain resource flow stability,
• media systems maintain narrative coherence and salience alignment,
• infrastructure systems maintain material coordination.

Each institution performs:

ongoing stabilisation work across its domain.

Bureaucracy as stabilised repair

Bureaucracy is frequently criticised as inertial or excessive.

But from a relational perspective, bureaucracy is:

a formalised maintenance technology.

It standardises repair through:

• procedures,
• classifications,
• documentation,
• and repeatable decision pathways.

This reduces dependence on individual discretion and ensures:

reproducibility of stabilisation under varying conditions.

Drift: the slow form of breakdown

Not all failure is sudden.

Much of it occurs through:

gradual drift across constraint layers.

Examples include:

• semantic categories slowly losing alignment with lived experience,
• institutions incrementally diverging from their founding assumptions,
• infrastructures adapting unevenly to new demands,
• or temporal rhythms becoming desynchronised.

Drift is particularly important because:

it often remains invisible until coherence thresholds are crossed.

Crisis as repair overload

A crisis occurs when:

the rate or scale of disruption exceeds the system’s capacity for repair.

At this point:

• maintenance systems become saturated,
• coordination breaks down,
• and repair itself becomes unstable.

Crises are not simply failures.

They are:

moments where maintenance systems can no longer sustain architectural coherence.

Why breakdown reveals structure

When maintenance fails, what becomes visible is not just dysfunction.

It is:

the underlying architecture that was previously invisible due to successful stabilisation.

Breakdown exposes:

• dependency chains,
• hidden couplings,
• procedural assumptions,
• and infrastructural constraints.

What was experienced as “reality” is revealed as:

a continuously maintained relational construction.

Repair as world re-production

Repair is not merely restoring what was.

It is:

re-producing the conditions under which a world continues to be intelligible and actionable.

This means repair always involves:

• selective restoration,
• modification of constraints,
• and adaptation to changed conditions.

No repair is neutral.

Every act of maintenance:

subtly reconfigures the system it preserves.

Maintenance and temporal continuity

One of the most important effects of maintenance is temporal continuity.

Maintenance ensures:

• yesterday’s categories still apply today,
• institutions remain recognisable over time,
• and expectations remain sufficiently stable for coordination.

Without maintenance:

time itself loses coherence as a shared relational structure.

Closing: the hidden labour of stability

Power is often imagined as force, decision, or domination.

But at its most fundamental operational level, power is:

the continuous labour of maintaining constraint architectures so that worlds remain coherent across time.

Maintenance, repair, and failure are therefore not peripheral concerns.

They are:

the engineering substrate of worldhood itself.

A world does not persist because it is stable.

It persists because:

it is continuously being kept from falling apart.