How the Theory Developed

What separates a living person from the same body moments after death, when most of the matter, DNA, cells, and visible structure are still present?

How can systems built from radically different materials—animal nervous systems, human brains, organizations, and artificial neural networks—produce partly similar high-level abilities?

And what would allow an observer to distinguish intelligence from coincidence, randomness, or a fixed response?

The research program began with questions like these, not with a plan to construct a theory of everything. They pointed toward a common difficulty: many important properties cannot be identified reliably from what a system is made of or from how it appears at one instant. They become visible only through organized change across time.

A Theory Can Only Study What It Can Distinguish

The project adopts a strict methodological boundary. It studies differences that can be distinguished through observation, compared across time, and attributed to changes in organization, history, environment, or observation conditions.

This boundary makes time essential. Without a before and an after, there is no observed change to compare. Without repeated or extended observation, it is difficult to decide whether an event was produced by the system, imposed from outside, generated by a fixed rule, or merely occurred by chance. Temporal order does not by itself prove causation, but it is necessary for testing causal and organizational explanations.

This is why a single surprising reaction from an animal is weak evidence of thought. It may be coincidence. Mere unpredictability is also insufficient, because random behaviour is unpredictable. What becomes more informative is repeated, organized variation: similar situations produce different responses in ways that correspond to memory, context, internal condition, prior outcomes, or current demands.

The same problem applies to artificial systems. One impressive answer may demonstrate a capability, but it does not by itself establish a persistent intelligence, a continuous identity, or a capacity to improve through use.

These limits forced the project to make broad labels more operational and adjudicable. The aim is not to declare a final definition by fiat, but to connect each attribution to observable differences, alternative explanations, and stated conditions.

Read the operational-concepts note

Why Artificial Intelligence Changed the Question

An early intuition connected intelligence closely to biological continuity. Intelligence seemed to belong to a living individual that persists, maintains itself, accumulates experience, and changes through an uninterrupted history.

Artificial intelligence broke that bundle apart. An artificial system can display substantial reasoning, prediction, generation, or problem-solving ability without biological metabolism, organism-level self-maintenance, or a single uninterrupted bodily trajectory. Its components may be copied, replaced, restarted, distributed, or invoked only when needed.

This did not prove that AI is alive or conscious. It showed something more methodologically important: life, continuity, intelligence, and consciousness cannot be treated as one undivided property.

The inquiry therefore changed. Instead of asking which material or object possesses a property by essence, it began asking which temporally extended organizational processes make persistence, adaptation, intelligence, or observed change distinguishable.

The current public research develops operational questions concerning life and intelligence. Consciousness remains a separate issue outside the present research claims.

How the Three Theories Took Shape

Connection Theory: From the Instantaneous Object to the Extended Organization

The first shift was away from treating a system as an object fully characterized at one moment.

A recently dead body makes the problem vivid. Much of its material composition and visible structure remains, and many local processes may continue briefly, yet the person is no longer functioning as a living whole. The crucial difference is not an immediate disappearance of matter. It is the loss of the integrated organizational process that maintained, coordinated, repaired, and continued the organism through time.

A snapshot can show components and relations, but it cannot by itself establish persistence, repair, adaptation, or identity through change. Connection Theory therefore moved the unit of analysis from the instantaneous object to the causally connected organizational trajectory.

Its central question became:

Under what conditions does an organized process remain connected to its own past while its components, states, and external relations change?

Material composition still matters, but composition alone is insufficient. The explanatory object is a temporally extended organizational history, not merely a momentary arrangement.

Flow Shaping Principle: From “Feedback Changes Systems” to Correct Attribution

The early intuition behind the Flow Shaping Principle was broad: feedback changes systems, and repeated feedback can shape future behaviour.

That statement was true but too weak. Harmful feedback changes a system. Delayed feedback changes a system. Feedback assigned to the wrong action or internal configuration also changes a system. None of these guarantees improvement in the intended direction.

The principle therefore narrowed around a more demanding question:

Can a consequence be connected to the configuration that actually produced it?

Adaptive shaping requires a usable correspondence between outcomes and the organized states, actions, or pathways to which credit or blame should be assigned. Without that correspondence, feedback may still produce change, but the change can be misdirected, unstable, or destructive.

FSP therefore moved from the general observation that feedback matters to a specific claim about the conditions under which consequences can correctly shape future organization.

Evolutionary Relativity: From Calendar Time and Endpoints to Organizational Path

Early descriptions of change relied too heavily on elapsed time and endpoint difference.

Two systems may begin and end in apparently similar states while having followed radically different organizational paths. Conversely, a large visible difference may result from a short reorganization under one frame and appear gradual under another. If only the start, finish, and calendar duration are recorded, the transformation itself can disappear from the analysis.

Evolutionary Relativity therefore shifted attention toward organizational path and observer reconstruction. What an observer identifies as change depends partly on what is preserved, what is measured, the temporal resolution used, and how the sequence between observations is reconstructed.

This does not make change arbitrary or merely subjective. It means that claims about organizational change must specify the frame and reconstruction procedure through which the path becomes observable.

How Reality Corrected the Theories

The theories were not produced in a single act of invention. They became clearer only after assumptions were written down, exposed to counterexamples, and allowed to fail.

AI Broke the Original Bundling of Intelligence and Biological Continuity

AI showed that observable cognitive capability could not be defined by importing all the continuity conditions of a biological organism. The correction did not make continuity irrelevant. It forced a cleaner separation between persistent identity, adaptive organization, task capability, life, and intelligence.

The FSP Sleep Hypothesis Was Withdrawn

An early FSP hypothesis treated sleep as a potentially general requirement of continuously adaptive systems. Comparative biological cases did not support that claim at the proposed level of generality.

The hypothesis was withdrawn rather than preserved through increasingly loose analogy. Sleep can still be studied as a biological mechanism of periodic reorganization, but it is not presented as a universal requirement shared by all adaptive systems, including artificial ones.

ER's Early Path Definition and Objective Function Were Rewritten

Early versions of Evolutionary Relativity used path definitions and a target function that unintentionally favoured certain reconstructions, including shorter or more compressed paths. The formalization was therefore partly deciding the result it was meant to evaluate.

The definitions and comparison procedure were rewritten to separate the observed path, the reconstructed path, and the rule used to compare them. This narrowed the claim but made it more testable.

These corrections are not embarrassing side notes. They are part of the research method. A theory becomes more useful when its assumptions are externalized clearly enough to be challenged, rejected, or reorganized.

What the Three Papers Currently Contribute

The three papers now have distinct roles:

• Connection Theory (CT) provides a framework for analyzing persistent organization, continuity, and identity across change.
• Flow Shaping Principle (FSP) identifies consequence–configuration correspondence as a necessary condition for correctly directed adaptive shaping.
• Evolutionary Relativity (ER) examines how organizational change is reconstructed and compared across paths, observation windows, and analytical frames.

Together, they separate three questions that are often blurred together:

1. What persists through change?
2. How do consequences shape what happens next?
3. How does an observer reconstruct and compare the path of change?

Public work on life and intelligence builds on these distinctions, but neither enlarges the empirical claims of the three core framework papers.

What Has Not Been Established

The research program remains at a pre-validation theoretical stage.

The papers do not establish a universal empirical law, a complete theory of intelligence, a test for consciousness, or a general explanation of every living, adaptive, or evolving system.

CT does not by itself explain how intelligence is produced. FSP does not prove that every adaptive mechanism implements one universal algorithm. ER does not supply a single observer-independent measure of all evolutionary change. Public applications and higher-level attributions still require edge-case analysis, adversarial testing, and empirical comparison with alternatives.

The current value of the program lies in the distinctions it makes, the failure conditions it exposes, and the research questions it makes possible.

This website presents a developing and deliberately bounded research program—not a finished doctrine and not a theory of everything.