Episode 69: Debate | Why systems need redundancy to survive
A debate on whether living systems and institutions survive through competitive pruning and optimization—or through redundancy, reserve capacity, protected variation, relational safety, and the freedom to adapt before crisis arrives.
This episode explores a central question:
Does redundancy provide the generative margin required for resilience and innovation, or does protection from competition allow biological and institutional systems to accumulate waste, dependency, and pathological complexity?
This episode accompanies the academic white paper:
Academic White Paper | The evolution of worlds: Natural drift, Inverse Darwinism, and the biology of love – From protected variation to life-coherent civilization
https://bsahely.com/2026/06/24/the-evolution-of-worlds-natural-drift-inverse-darwinism-and-the-biology-of-love-from-protected-variation-to-life-coherent-civilization-chatgpt-5-5-high-intelligence-and-notebooklm/
The debate begins beneath a suspension bridge.
On an ordinary calm day, the bridge’s extra cables, excess steel, safety factors, and overlapping supports may appear inefficient. An accountant concerned only with immediate cost might classify them as unnecessary weight.
To an engineer, however, the apparent waste is what prevents collapse when the wind intensifies, a cable fails, or a hurricane exceeds normal expectations.
The bridge becomes a metaphor for the central argument of the episode.
Modern institutions are often designed according to a selectionist grammar: eliminate slack, cut duplication, maximize utilization, reduce inventories, compress staffing, and remove anything that does not produce immediate measurable output.
The promise is efficiency.
The danger is brittleness.
One side of the debate argues that redundancy, excess capacity, and relational safety are the real engines of biological and institutional innovation.
The opposing side argues that without selective pressure, pruning, competition, and the willingness to remove failing components, redundancy becomes administrative bloat, parasitic dependency, and organizational decay.
The dispute is therefore not simply between efficiency and waste.
It is about how living systems distinguish generative reserve from pathological accumulation.
The case for generative reserve
The life-coherent position begins by challenging the popular interpretation of natural selection.
Natural selection remains an essential biological mechanism. It explains why some inherited variations persist while others disappear. It helps account for adaptation, differential reproduction, and the refinement of traits.
But selection is primarily subtractive.
It can preserve, refine, or eliminate possibilities that already exist. It cannot select a possibility that was never generated.
Novelty therefore requires more than pruning.
It requires a protected space in which variation can arise before its usefulness is known.
The paper draws upon inverse Darwinism: the idea that evolutionary innovation often begins not with scarcity and elimination, but with duplication, redundancy, and excess capacity.
When a vital gene exists in only one copy, it is tightly constrained. A damaging mutation may destroy an essential biological function.
When the gene is duplicated, the original copy can continue performing the necessary task while the duplicate varies.
The redundant copy is protected from immediate lethal consequences.
It can drift, mutate, lose part of its original function, or enter a new relationship with other components.
Most variations may lead nowhere.
Some may deteriorate into non-functional remnants.
But occasionally a duplicate develops a complementary function that enlarges the organism’s total capacity.
The evolution of trichromatic vision provides a central example.
An ancestral opsin gene involved in colour detection was duplicated. The original copy preserved existing visual capacity. The duplicate changed its wavelength sensitivity over time.
The brain eventually learned to compare the overlapping signals produced by the original and altered receptors.
A redundant copy became the basis of a new perceptual world.
The life-coherent side argues that selection did not plan colour vision. Selection could only stabilize the new arrangement after redundancy had already created the possibility.
This produces a general evolutionary sequence:
duplication → buffering → variation → divergence → complementarity → new capacity
Redundancy is therefore not automatically waste.
It may serve as evolutionary permission to explore.
The argument for selection
The selectionist position accepts that duplication can generate novelty but warns against romanticizing excess.
Most duplicated genes do not become transformative innovations. Many accumulate damaging mutations, lose function, or become pseudogenes.
Maintaining extra components also carries metabolic cost.
Every duplicated gene must be copied, regulated, and integrated into a living system whose energy and resources are finite.
The same problem appears in human institutions.
An organization may preserve overlapping departments, duplicate administrative functions, redundant procedures, and unused resources in the name of resilience.
But these components can gradually become mutually dependent without enlarging the institution’s life-serving capacity.
This resembles constructive neutral evolution.
Complexity accumulates because earlier alternatives disappear, not because the new arrangement offers a superior function.
The result can be a Rube Goldberg machine: an elaborate network of steps, rules, departments, forms, and committees required to perform a relatively simple task.
Removing one component becomes difficult because every layer has adapted to the presence of the others.
The system becomes complex, fragile, and expensive without becoming more capable.
The selectionist side therefore argues that redundancy must eventually justify its continued existence.
Without pruning, systems may conserve historical accidents indefinitely.
Selection acts as a discipline that forces variation to prove that it contributes to the whole.
Shannon information and the problem of noise
The debate uses information theory to sharpen both positions.
The life-coherent side compares redundancy to transmitting a message through a noisy communication channel.
If every word is spoken only once and static interrupts the transmission, the message may be lost.
Repetition, overlap, context, and error correction allow the signal to survive disturbance.
Biological systems use similar strategies.
Genetic duplication, paired organs, immune diversity, overlapping metabolic pathways, and distributed neural networks create protection against error and loss.
The selectionist side responds that a communication channel also has limited capacity.
Too much repetition consumes bandwidth.
If every word is repeated endlessly, the message becomes inefficient and eventually difficult to distinguish from noise.
Redundancy therefore has an optimum range.
Too little reserve produces fragility.
Too much unstructured reserve produces congestion, cost, and confusion.
The disagreement concerns how that range should be established.
Should competition and failure act as the primary regulators?
Or should systems use a broader life-coherence criterion that distinguishes useful reserve from harmful accumulation?
Natural drift and the absence of a predetermined optimum
The generative-drift position draws on Humberto Maturana, Francisco Varela, and Jorge Mpodozis.
Living systems are autopoietic: they continually produce and maintain the organization that enables them to remain living unities.
Evolution does not necessarily move toward a universal predetermined optimum.
Organisms undergo structural changes while conserving the organization required for continued living.
As long as the lineage remains viable, it continues drifting through possible forms.
The environment does not act as an engineer handing the organism a blueprint.
Organism and environment change together through structural coupling.
The life-coherent side therefore resists the assumption that every variation must immediately demonstrate competitive superiority.
A variation may initially be neutral.
Its eventual significance may emerge only through later changes in the organism, the environment, or the relationship between them.
Premature pruning can therefore destroy possibilities before their complementary function becomes visible.
The selectionist side responds that drift without constraint can become entropy.
Living systems operate under material limits. Energy, time, attention, space, and resources are finite.
Selection helps prevent systems from maintaining every possible variation indefinitely.
The issue is not whether variation should exist, but whether the system possesses a credible method for determining what should be conserved.
Hospitals and the hidden cost of lean efficiency
The debate then turns to healthcare.
For decades, many hospitals were reorganized around lean management, maximum utilization, just-in-time supply chains, reduced staffing margins, and the elimination of unused capacity.
An empty hospital bed appeared economically wasteful.
Stored protective equipment looked like idle inventory.
Overlapping staff expertise appeared inefficient.
The system was optimized for ordinary demand.
The COVID-19 pandemic exposed the hidden cost.
Demand exceeded the narrow operating range for which hospitals had been designed.
Beds, staff, oxygen, protective equipment, supply chains, laboratory capacity, and local manufacturing proved insufficient.
The apparently efficient system had removed its ability to absorb a major disturbance.
The life-coherent side argues that the empty bed was not waste.
It was reserve capacity.
Its value could not be measured solely by daily occupancy because its purpose was to protect the system when ordinary conditions failed.
A suspension bridge is not judged only by the amount of steel required on a windless day.
A health system should not be judged only by the minimum resources required during an average week.
The selectionist side agrees that pandemic preparedness matters but argues that the failure resulted from poorly chosen optimization criteria.
Hospitals optimized for short-term financial efficiency instead of including resilience, surge capacity, and disaster preparedness in their definition of performance.
The corrective, from this perspective, is not unlimited reserve.
It is better selection.
Systems should preserve redundancy whose value is demonstrated through rigorous stress testing and remove redundancy that merely reproduces administrative complexity.
Institutional autopoietization
Both sides agree on the danger of institutional autopoietization.
An institution may begin with a clear life-serving purpose.
A hospital exists to heal.
A university exists to educate.
A government agency exists to serve a public function.
Over time, the institution develops procedures, budgets, reporting systems, hierarchies, legal protections, performance metrics, and internal cultures.
These mechanisms may initially support the founding purpose.
Eventually, however, the institution may begin conserving them for their own sake.
The procedure becomes the product.
The metric becomes the mission.
The budget becomes the measure of success.
The institution becomes highly effective at reproducing itself while becoming less effective at serving the lives that justify its existence.
The selectionist side interprets this as the predictable outcome of protection from failure.
A bureaucracy insulated from competition and consequences can continue consuming public resources despite declining performance.
Without the possibility of elimination, restructuring, or loss of funding, the institution has little reason to correct itself.
The generative-drift side argues that competition alone does not solve institutional autopoietization.
An institution exposed to harsh market selection may simply optimize for the wrong proxy.
A hospital competing for revenue may increase billing efficiency while reducing relational care.
A school competing for rankings may narrow education to examinations.
A company competing for engagement may capture attention while damaging cognition.
Selection does not automatically connect the system to life.
It connects the system to whatever metric determines survival.
The deeper question is therefore:
What is doing the selecting, and according to which value?
Tumours and local success
Cancer becomes a shared warning.
A tumour can be highly adaptive.
It generates genetic variation, duplicates chromosomes, reroutes metabolism, recruits blood vessels, avoids immune detection, and responds rapidly to therapeutic pressure.
At the local cellular level, it may appear evolutionarily successful.
Its success is nevertheless life-incoherent because it destroys the organism that sustains it.
The tumour reveals that persistence, adaptability, complexity, and competitive success are not sufficient measures of value.
A corporation may dominate its market while exhausting workers and ecosystems.
A bureaucracy may expand its budget while degrading public service.
A technology platform may increase engagement while damaging mental health.
A financial system may become more sophisticated while destabilizing the real economy.
Local fitness can coexist with global destruction.
The life-coherent criterion must therefore evaluate the system’s effects across scales.
The biology of love
The debate then moves from material redundancy to relational safety.
Maturana’s biology of love is introduced not as romantic sentimentality but as a technical description of a relational domain.
Love is the dynamic in which another is accepted as a legitimate other in coexistence.
This does not require agreement.
It does not eliminate boundaries.
It does not prohibit accountability, restraint, or consequences.
It means that correction does not require denying the other’s basic dignity or right to exist.
The generative-drift side compares love to biological buffering.
A duplicated gene can vary because the original protects the essential function.
A person can question, experiment, dissent, admit error, or revise a belief when their basic legitimacy is protected.
Relational safety becomes the social equivalent of redundancy.
It prevents every mistake from becoming an existential threat.
This matters because institutions depend upon accurate feedback.
In a culture governed by fear, employees conceal errors.
Clinicians practise defensively.
Students repeat expected answers.
Administrators manipulate indicators.
Subordinates tell leaders what they want to hear.
The system appears orderly while losing access to local reality.
The biology of love preserves the possibility of truthful speech.
The selectionist position worries that unconditional legitimacy may be interpreted as protection from consequences.
Institutions contain corruption, incompetence, free-riding, exploitation, and deliberate harm.
A system that refuses to exclude destructive participants may allow them to consume the very reserve intended to protect everyone.
Fear, scarcity, and the possibility of failure can therefore perform an immune function.
They signal that actions have consequences and that shared resources cannot be consumed indefinitely without contribution.
The life-coherent response is that legitimacy and accountability are not opposites.
A person can remain a legitimate other while a harmful action is stopped.
A clinician can be supported through honest error reporting while reckless misconduct is addressed.
A student can retain dignity while being required to revise inadequate work.
An employee can receive fair process while remaining answerable for corruption or abuse.
Love protects the person’s non-disposability.
It does not protect every behaviour from evaluation.
Fear and feedback
The debate becomes especially sharp around fear.
The selectionist side argues that fear of failure motivates discipline, prevents complacency, and protects institutions from exploitation.
The life-coherent side argues that fear often corrupts the very information required for discipline.
A hospital dominated by blame may appear strict but become less safe because workers hide near misses.
A corporation dominated by dismissal threats may appear productive but lose innovation because employees avoid disagreement.
A government that punishes dissent may appear stable but become incapable of detecting errors before crisis.
Fear may increase obedience while decreasing truth.
Relational safety may therefore be more demanding than punitive control because it requires exposing reality, admitting failure, and revising conduct without using humiliation as the primary enforcement mechanism.
Civil commons as societal reserve
The episode then applies redundancy to public institutions.
Healthcare, education, social protection, libraries, ecological safeguards, public infrastructure, and other civil commons function as collective reserve capacity.
They allow people to experience illness, unemployment, failure, caregiving responsibilities, or economic transition without losing every condition required for continued participation.
The generative-drift position argues that safety nets enable innovation.
People can retrain, create enterprises, challenge institutions, conduct research, care for relatives, or attempt new forms of work when one failure does not result in hunger, homelessness, or untreated disease.
A society that demands innovation while eliminating all protective reserve creates a contradiction.
Only the already wealthy can afford to experiment.
The selectionist position accepts the value of civil commons but insists that public systems must remain subject to evaluation.
Programmes can become self-protective.
Administrative structures can expand without improving outcomes.
Resources devoted to inherited procedures may displace resources from genuine public needs.
The question becomes:
Which dependencies enlarge life-capacity, and which merely reproduce complexity?
Life-coherence as the selection criterion
The debate approaches synthesis through John McMurtry’s life-value framework.
A system should be evaluated by whether it protects, restores, or enlarges life-capacity without transferring disabling costs onto other lives, ecosystems, or future generations.
This standard supplies a criterion for distinguishing reserve from bloat.
Redundancy is life-coherent when it:
- protects essential functions;
- increases resilience;
- allows truthful feedback;
- supports experimentation;
- improves recovery after disturbance;
- enlarges future possibilities;
- preserves ecological and human life-capacity.
Redundancy becomes pathological when it:
- primarily reproduces administrative procedures;
- blocks correction;
- consumes resources without serving a living function;
- suppresses alternative pathways;
- protects entrenched power;
- externalizes costs onto patients, workers, citizens, ecosystems, or future generations.
Selection also becomes life-coherent only when its criterion is life-serving.
Pruning is justified when it removes structures that demonstrably diminish life-capacity and when the process preserves the dignity, participation, and future possibilities of those affected.
Selection becomes life-incoherent when it eliminates reserve solely to improve short-term financial or institutional metrics.
Where the debate converges
Both sides ultimately agree on several principles.
A living system must conserve enough organization to remain viable.
Variation without conservation can produce dissolution.
Redundancy without evaluation can generate pathological complexity.
Selection without reserve creates brittle systems incapable of adaptation.
Institutional survival is not equivalent to life-serving value.
Metrics must remain subordinate to the living purposes they are intended to represent.
Relational safety must not become immunity from responsibility.
Accountability must not become the destruction of truthful feedback.
The central disagreement concerns which danger should receive primary emphasis.
The selectionist position fears decay, parasitism, administrative bloat, and the indefinite protection of ineffective structures.
The generative-drift position fears brittleness, suppressed variation, hidden error, collapsed resilience, and the destruction of possibilities before their value can emerge.
A mature life-coherent institution may therefore require both:
a protected generative margin and a credible process of life-coherent evaluation.
The system must preserve enough reserve to experiment, absorb disturbance, tell the truth, and learn.
It must also remain capable of removing structures that consume the life-ground without contributing to its renewal.
The relevant opposition is not redundancy versus selection.
It is:
life-serving reserve versus pathological accumulation, and life-serving correction versus destructive pruning.
The guiding question is:
What must we conserve together if life is to remain capable of becoming otherwise—and what must we release when inherited complexity begins consuming the life it was meant to serve?
AI use and transparency
This episode is part of an AI-assisted audio pathway through the Life-Knowledge Commons. Some deep-dive conversations, debates, and critiques are generated or supported by tools such as NotebookLM and other large language model systems, using Dr. Bichara Sahely’s writings, papers, and source materials as grounding documents.
These tools are used to support reflection, accessibility, synthesis, dialogue, critique, and sharing. They do not replace human judgment, responsibility, authorship, scientific discernment, ethical accountability, institutional responsibility, or lived experience. The responsibility for what is curated and shared within this Commons remains with Dr. Bichara Sahely.
Host: Dr. Bichara Sahely
Podcast: Toward Life-Knowledge
Theme: Knowledge in service of life.