H Heuristics Navigating a Changing World

H Heuristics Research Report · HH-2026-37

The Architecture of Life

Chemistry, Information, Selection and the Recurring Organisation of Living Systems

How chemistry could have become life, how life then built its many levels of organisation, and whether the same principles of construction recur from molecules to the biosphere.

Hunter Hughes · · 92 pages

Read the PDF opens in a new tab Download 5.1 MB

Abstract

Roughly four billion years ago the surface of a young planet held water, rock, gases and a great variety of small carbon compounds, none of them alive. Today the same planet carries about 550 billion tonnes of carbon in living tissue, perhaps ten million or more species of eukaryotes and an uncounted diversity of microbes, ecosystems that span every surface, and at least one species that studies the rest. This report asks how chemistry could have become life, how life then built its many levels of organisation, and whether the same principles of construction recur from molecules to the biosphere.

The thesis is that life is a remarkable form of organised matter, but matter obeying the same physics and chemistry as everything else, and that living organisation is generated by the same combination the series has found elsewhere, with one decisive addition. Chemistry and physics act as constraints that define what molecular systems can do. Autocatalysis, replication, mutation, recombination and symbiosis act as generators that produce variants. Thermodynamic persistence, natural selection and ecological filtering act as filters that decide which variants spread. The decisive addition is heredity: once a system can copy the information that specifies it, filters act not only on what exists but on what is passed on, and the filter becomes cumulative. That turns the creation structure into an engine that can accumulate design-like complexity without a designer and without a goal.

The report keeps the origin of life separate from the evolution of life. The origin remains unsolved: no experiment has produced a self-sustaining, evolving chemical system from plausible starting materials, and the setting is contested, so the evidence for the transition to life is chemical and inferential while the evidence for later evolution is direct and abundant. Claims are graded throughout on the series’ four-level scale, and the report concludes that a coherent architecture of life exists but as a family of related mechanisms rather than one law — energy flow, feedback and emergence recur at every scale, while information, boundaries, selection and cooperation are shared mechanisms within organisms but analogies, sometimes weak ones, at ecological and planetary scales.

The practical payoff is a way of reading any living system. Before asking what a biological structure is for, ask which chemical or physical constraint makes it possible, which generator produced its variants, which filter keeps it, where its energy comes from, and what information it inherits. That checklist separates what life had to do from what it happened to do, and keeps purpose-language from substituting for mechanism.

Key findings

  1. The ingredients of life are ordinary and were available early. Water, carbon compounds, energy gradients and mineral surfaces were present on Earth by about 4.4 billion years ago, and amino acids, sugars and all five nucleobases form under plausible early-Earth conditions and are found in meteorites and returned asteroid samples.
  2. The origin of life is a separate problem from the evolution of life, and it remains unsolved. No experiment has produced a self-sustaining, evolving chemical system from plausible starting materials, and the setting is contested.
  3. Three subsystems had to be coupled — metabolism, information and compartments. Autocatalytic networks appear suddenly once molecules catalyse on average about one reaction each; replicators with sub-exponential growth coexist rather than competing to extinction; copying is limited by an error threshold that real genomes respect across five orders of magnitude of genome size; and compartments let cooperative systems persist against faster-copying parasites.
  4. The cell has a remarkably conserved core, and the genetic code is not arbitrary: tested against one million random alternatives, only about one in 7,600 was better at buffering the effects of point mutations.
  5. Evolution is a generative process, not merely a sieve. Mutation, recombination, horizontal gene transfer, gene duplication and symbiosis generate variation; selection and drift filter it; and laboratory evolution shows open-ended improvement for more than 75,000 generations.
  6. The largest steps in complexity were mergers and new forms of cooperation. The eukaryotic cell arose from a partnership between an archaeon and a bacterium, and multicellularity evolved at least 25 times — each major transition requiring mechanisms to suppress conflict among the parts.
  7. Ecosystems are organised by energy flow and constrained by stability. Only about a tenth of energy passes between trophic levels, which keeps food chains short, and because random complexity destabilises communities, real ecosystems must be structured.
  8. Intelligence is one of several nested information-processing systems, not the destination of evolution. Complex cognition evolved independently in mammals, birds and cephalopods on different neural architectures, and evolution shows no built-in trend toward intelligence or complexity — the maximum has increased while most life has remained microbial.
  9. Life is a planetary process. Photosynthesis oxygenated the atmosphere in two great steps, about 2.3 and 0.6 billion years ago, and life and Earth have co-evolved through feedbacks running through the oceans, atmosphere and crust — though whether those feedbacks stabilise climate remains contested.
  10. A coherent architecture of life exists, but it is a family of related mechanisms, not one law. The framework predicts the general features of life elsewhere but not its frequency, which a single example cannot establish.

Contents

  1. —About This Report
  2. —Executive Summary
  3. 1Central Thesis and Scope
  4. 2The Framework: Creation Structures in Living Systems
  5. 3The Architecture of Prebiotic Chemistry
  6. 4The Emergence of the First Living Systems
  7. 5The Architecture of the Cell
  8. 6Evolution as a Generative Process
  9. 7The Emergence of Multicellular Complexity
  10. 8The Architecture of Ecosystems
  11. 9Intelligence and the Evolution of Information Processing
  12. 10Life as a Planetary System
  13. 11The Recurring Architecture of Life
  14. 12Life Beyond Earth and Alternative Architectures
  15. 13Testable Propositions
  16. 14Conclusion: Is There a Coherent Architecture of Life?
  17. —References

Data and method

The report synthesises published research and presents original simulations and calculations, which are labelled as such. Claims are graded on the four-level scale used across the series — A established, B well supported, C plausible or contested, D speculative — and the origin of life is kept explicitly separate from later evolution, since the evidence for the first is chemical and inferential and for the second direct and abundant. The 43 figures are of three kinds. Original simulations and calculations of standard published models were made for this report: competition between replicators with different growth laws, the emergence of autocatalytic sets in random chemical networks, Eigen’s error threshold, a stochastic-corrector model of protocells, a Bayesian analysis of what life’s early appearance implies about its probability, regulatory feedback and a synthetic genetic oscillator, a test of the genetic code against one million random alternatives, Wright–Fisher simulations of selection and drift, rugged NK fitness landscapes, predator–prey cycles, May’s stability bound for complex communities, Daisyworld and Turing pattern formation, with each caption stating what was computed. Charts of published measurements are credited in their source lines. Scientific illustrations of molecules, protocells, cells, neurons, food webs and Earth’s layers were drawn for this report and are marked illustrative — schematic, and not to scale unless stated.