DOI: to be assigned
John Swygert
June 30, 2026
Abstract
Recent discussion of the proposed “law of increasing functional information” argues that complexity may increase over time not only in living organisms, but also in nonliving systems such as minerals, elements, planets, stars, and chemical systems. Philip Ball’s Quanta Magazine article, “Why Everything in the Universe Turns More Complex,” presents this proposal as a possible expansion of evolutionary thinking beyond biology and into cosmic history. The article is important because it identifies a real pattern: under certain conditions, systems do not merely decay into disorder, but develop selected, persistent, functional structures over time. Its problem is that the word complexity becomes too large, too vague, and too easily mistaken for the actual mechanism.
This paper argues that TSTOEAO provides a cleaner lens. Complexity is not the primary law. Complexity is often the visible residue of a deeper process: gradients meeting boundary conditions, producing selection, correction, persistence, and encoded equilibrium. In TSTOEAO terms, “functional information” is better understood as encoded equilibrium under a defined boundary condition. This reframing matters because it applies not only to cosmic evolution and biology, but also to mineral evolution, cancer, artificial intelligence, social systems, language, scientific discourse, and any domain where systems must sort, preserve, correct, or stabilize themselves under pressure.
01
The Article And Its Central Claim
Philip Ball’s Quanta Magazine article, published April 2, 2025, summarizes a proposal that complexity increases over time not only in living organisms, but also in nonliving systems. The article frames this proposal as a possible new law of nature, comparable in ambition to the second law of thermodynamics, but oriented toward increasing complexity rather than increasing entropy.
The proposal discussed by Ball comes from work by Michael L. Wong, Robert Hazen, and colleagues, who argue for a “law of increasing functional information.” In their formulation, evolving systems are composed of many possible configurations, and some configurations are selected because they perform one or more functions. When that happens over time, functional information increases.
The article is valuable because it recognizes a pattern that many scientific fields keep rediscovering separately: matter, life, intelligence, and culture do not simply scatter randomly. Under the right conditions, they preserve certain structures, repeat certain solutions, and generate higher-order arrangements. Minerals diversify. Chemistry selects. Biology evolves. Minds model. Technologies accumulate. Language expands. Scientific fields organize around claims, evidence, objections, and revised theories.
But the article’s framing also reveals the central problem.
It asks, in effect:
Why does everything become more complex?
Through TSTOEAO, that is not the cleanest question.
The better question is:
Under what gradients, boundary conditions, and selection pressures do systems preserve increasingly structured equilibria?
That shift matters.
02
The Problem With “Complexity” As The Main Variable
The word complexity is seductive because it seems to describe everything at once. A living cell is complex. A brain is complex. A mineral history is complex. A galaxy is complex. A tumor is complex. A bureaucracy is complex. A social network is complex. A pile of wreckage is complex.
That is exactly the problem.
Complexity alone does not tell us whether a system is healthy, functional, stable, adaptive, meaningful, or merely tangled.
A collapsed city is complex.
A disease process is complex.
A misinformation cascade is complex.
A broken institution is complex.
A Facebook argument is complex.
But none of those examples automatically represents higher order in the useful sense.
TSTOEAO therefore does not treat complexity as the root variable. It treats complexity as a possible byproduct of deeper relational structure.
The root variables are:
Gradient — the pressure, imbalance, energy difference, need, threat, or opportunity driving motion.
Boundary Condition — the defined constraint, environment, membrane, rule, body, planet, platform, or system limit inside which outcomes must occur.
Selection / Correction — the filtering process by which some configurations persist and others fail.
Cost-Location — the place where failure, waste, misunderstanding, disease, disorder, or collapse is paid.
Equilibrium Target — the state toward which the system is being stabilized, optimized, restored, or transformed.
In this lens, the universe does not simply “turn more complex.”
It repeatedly tests configurations under pressure, and some of those configurations persist because they satisfy a boundary-conditioned equilibrium requirement.
That is a stronger statement.
03
Functional Information As Encoded Equilibrium
The law of increasing functional information is close to TSTOEAO, but it needs sharper language.
The Wong-Hazen formulation says that functional information increases when many configurations undergo selection for function. Their mineral-evolution case study defines functional information as the negative log base 2 of the fraction of possible configurations that achieve a degree of function, and it reports monotonic increase across nine chronological stages of Earth’s mineral evolution.
That is scientifically interesting. It suggests that nonliving systems can also show directional increase in selected structure. Minerals are not “evolving” in the Darwinian organismal sense, but their diversity and stability across planetary time can still be studied as a selected subset of a vastly larger possibility space.
TSTOEAO would translate the idea this way:
Functional information is encoded equilibrium under a boundary condition.
That phrase solves several problems.
First, it prevents “function” from floating loose. A function only matters relative to a system and a target.
Second, it prevents complexity from being treated as inherently good. Complexity matters only when it preserves, corrects, stabilizes, adapts, transmits, or opens a new functional layer.
Third, it makes the boundary condition explicit. There is no such thing as function in the abstract. There is function for something, within something, under some constraint, toward some target.
A mineral’s function may be static persistence.
A cell’s function may be dynamic persistence.
A nervous system’s function may be signal integration.
A scientific theory’s function may be predictive compression and falsifiable organization.
A social platform’s function may be discourse stabilization.
Without the boundary, “function” becomes opinion.
With the boundary, function becomes testable.
04
The Article’s Strongest Point
The article’s strongest point is that biology should not be sealed off as though evolution is a magical exception to the rest of nature.
Ball notes that the proposed law treats biological evolution as one case of a broader principle in which entities are selected because they contain information enabling function.
That matters.
TSTOEAO agrees that biology should not be isolated from mineral systems, chemical systems, planetary systems, artificial intelligence, social systems, and cognitive systems. The same general relational grammar appears repeatedly:
A system has possible states.
A gradient pressures those states.
A boundary condition filters them.
Some configurations fail.
Some configurations persist.
Persistent configurations become encoded.
Encoded configurations create new possibility space.
That pattern is visible across science.
The problem is not the intuition. The intuition is excellent.
The problem is the headline-level variable.
“Complexity increases” is too blunt.
“Selected configurations become encoded under boundary conditions” is cleaner.
05
The Article’s Weakness: Context And Testability
The article also identifies the main weakness of the functional-information proposal: functional information is contextual. It depends on what function is being measured and what environment the system occupies. Ball reports criticism that the theory may be difficult to test objectively, and he quotes skepticism about what a controlled experiment would even look for.
Hazen and Wong’s own mineral-evolution work acknowledges a related limitation: rigorous calculations of functional information are currently untenable for many natural systems, especially living systems, because one cannot enumerate every possible configuration of molecules, cells, organisms, and ecosystems, much less evaluate every degree of function.
That is not a small problem.
If the theory says functional information increases, but function must be chosen after the fact, then the theory risks becoming descriptive rather than predictive.
TSTOEAO helps here because it demands that the system, boundary, gradient, cost-location, and equilibrium target be named first.
A TSTOEAO analysis should ask:
What system is being measured?
What boundary condition defines it?
What gradient is acting on it?
What counts as persistence, correction, or failure?
Where is the cost paid?
What is the equilibrium target?
What would count as evidence against the claim?
That does not magically solve every measurement problem. But it prevents the theory from dissolving into vague admiration for complexity.
06
The TSTOEAO Correction
The Quanta article’s implied claim is:
Everything becomes more complex.
The TSTOEAO correction is:
Systems under sustained gradients and boundary conditions tend to test configurations; configurations that reduce cost, preserve function, or improve equilibrium are more likely to persist and become encoded. Complexity may increase as a result, but complexity is not the law.
That is the heart of the paper.
Complexity is not the engine.
Boundary-conditioned selection is the engine.
Correction is the engine.
Persistence is the engine.
Equilibrium is the engine.
Encoding is the memory of the engine.
This distinction matters because it keeps science from confusing the visible outcome with the deeper process.
A tree is complex, but its complexity is not random ornament. It is the encoded result of light gradients, gravity, water transport, atmospheric exchange, reproductive pressure, soil chemistry, genetic memory, and ecological boundary conditions.
A brain is complex, but its complexity is not merely more parts. It is selected signal-routing, memory, prediction, motor control, emotional regulation, and social modeling under biological cost constraints.
A scientific theory is complex only when necessary. The best theory is not the messiest theory. The best theory is the one that organizes the most phenomena with the least distortion while remaining falsifiable.
That is TSTOEAO.
07
Why This Applies To Other Scientific Problems
This lens applies to many scientific problems because science repeatedly runs into the same confusion: it observes a complex outcome, then struggles to identify the underlying organizing grammar.
Origin Of Life
The origin of life is often framed as a leap from chemistry to biology. TSTOEAO reframes it as a transition from unstable chemical configurations to boundary-conditioned, self-maintaining, self-correcting chemical cycles.
The key question is not “When did life magically appear?”
The key question is:
When did chemistry begin preserving encoded equilibrium across time?
That includes compartments, gradients, catalytic loops, persistence, reproduction, and error correction.
Cancer
Cancer is not merely “complex disease.” It is a system in which cellular boundary discipline fails. Cells that once served organism-level equilibrium begin optimizing for local replication. The cost-location shifts from cellular survival to organismal damage.
TSTOEAO would describe cancer as a boundary-condition rebellion: local cellular persistence becomes detached from whole-body equilibrium.
That is a much cleaner systems description than simply saying cancer evolves.
Artificial Intelligence
AI systems do not merely become more complex. They become more powerful when their outputs are trained, corrected, selected, reinforced, and aligned against defined targets.
An AI model without boundary conditions becomes dangerous or useless noise. An AI model with good boundary conditions becomes a correction engine.
This is directly relevant to the user’s platform idea for Piled Higher & Deeper: AI should not merely moderate speech. It should classify speech, identify gradients, detect category confusion, request references, track revisions, and guide discourse toward clearer equilibrium.
Social Networks
A social network like Facebook is complex, but not necessarily high-equilibrium. It contains enormous activity, but much of it is unclassified, unreferenced, emotionally amplified, and structurally confused.
A TSTOEAO-based social network would impose semantic boundary conditions:
Fact.
Opinion.
Satire.
Question.
Personal experience.
Speculation.
Correction.
Hypothesis.
Source addendum.
That one design change turns chaotic speech into categorized signal.
Scientific Discourse
Science itself is not merely the accumulation of facts. It is a correction system.
A hypothesis enters.
Evidence pressures it.
Objections test it.
Definitions constrain it.
Predictions expose it.
Corrections refine it.
Failed claims are discarded.
Surviving claims become encoded knowledge.
Science is TSTOEAO in practice, whether or not it has named the structure.
08
Why TSTOEAO Matters
TSTOEAO matters because it gives science a portable grammar for problems that appear unrelated on the surface but share the same underlying structure.
The same pattern appears in:
mineral evolution,
origin-of-life chemistry,
biological evolution,
cancer,
ecology,
neuroscience,
language evolution,
AI alignment,
social media disorder,
scientific publishing,
civilization collapse,
personal recovery,
and institutional reform.
That does not mean all systems are identical.
It means many systems can be analyzed by the same relational sequence:
Gradient → Boundary Condition → Correction → Cost-Location → Equilibrium Target → Encoded Persistence
This is why TSTOEAO is important. It does not merely say “things are connected.” That is too vague. It says systems can be studied according to how pressure meets boundary, how correction occurs, where cost is paid, and what equilibrium target is being approached or missed.
That is a usable scientific lens.
09
The Difference Between Complexity And Equilibrium
The deepest correction is this:
Complexity is not automatically progress.
Progress requires improved equilibrium.
A society can become more complex and more unstable.
A disease can become more complex and more destructive.
A theory can become more complex and less explanatory.
A platform can become more complex and more childish.
A bureaucracy can become more complex and less effective.
Therefore, any theory of increasing complexity must distinguish between:
mere complication
and
functional, corrective, boundary-conditioned order
TSTOEAO makes that distinction.
The goal is not more complexity.
The goal is better encoded equilibrium.
10
Functional Information Rewritten Through TSTOEAO
The proposed law of increasing functional information can be preserved, but clarified.
Original simplified formulation:
Functional information increases when many configurations undergo selection for function.
TSTOEAO reformulation:
Encoded equilibrium increases when a system under gradient pressure repeatedly tests possible configurations within boundary conditions, and configurations that reduce cost, preserve function, or improve system-target stability are retained across time.
That version is longer, but cleaner.
It names the missing parts.
It names the gradient.
It names the boundary.
It names correction.
It names cost.
It names persistence.
It names the equilibrium target.
Without those terms, “function” floats.
With those terms, “function” becomes scientifically disciplined.
11
Why The Article Should Be Looked At Through This Lens
The Quanta article should be looked at through TSTOEAO because it is wrestling with exactly the kind of boundary-crossing problem TSTOEAO was built to handle.
The article moves across:
physics,
biology,
chemistry,
minerals,
information theory,
astrobiology,
evolution,
complexity science,
and intelligence.
That is where ordinary disciplinary language starts to wobble.
Words like complexity, function, information, evolution, selection, and law begin meaning different things in different fields.
TSTOEAO is useful precisely because it asks for a deeper structural translation.
Instead of asking whether minerals “evolve” like organisms, it asks:
What is the system?
What configurations are possible?
What boundary condition filters them?
What counts as persistence?
What does the environment select?
What is encoded over time?
Instead of asking whether the universe “wants” complexity, it asks:
Where do gradients generate selected persistence?
Instead of asking whether intelligence is inevitable, it asks:
What boundary conditions make self-referential correction stable enough to persist?
This is the advantage of the lens.
It does not force every field into one metaphor.
It gives different fields a common diagnostic grammar.
12
Conclusion
The Quanta article is not ridiculous because the subject is unimportant. It is frustrating because the subject is important and the framing becomes too diffuse.
The real insight is not that everything becomes more complex.
The real insight is that systems under pressure preserve selected structures when those structures satisfy boundary-conditioned requirements for persistence, correction, function, or equilibrium.
That is TSTOEAO.
Complexity is what we see after the fact.
Functional information is one attempt to measure selected structure.
But encoded equilibrium is the deeper principle.
The universe does not merely pile things higher and deeper.
It tests, filters, preserves, corrects, and encodes.
Where gradients meet boundary conditions, systems either collapse, dissipate, or discover a structure that can hold.
That holding is the beginning of order.
That preserved order is encoded equilibrium.
And when encoded equilibrium compounds across time, we call it complexity.
References
Ball, P. (2025, April 2). “Why Everything in the Universe Turns More Complex.” Quanta Magazine.
Hazen, R. M., & Wong, M. L. (2024). “Open-ended versus bounded evolution: Mineral evolution as a case study.” PNAS Nexus.
Wong, M. L., Cleland, C. E., Arend, D., Bartlett, S., Cleaves, H. J., Demarest, H., Prabhu, A., Lunine, J. I., & Hazen, R. M. (2023). “On the roles of function and selection in evolving systems.” Proceedings of the National Academy of Sciences.