This is not a system. It is a set of distinctions that keep a student from saying more than the evidence has earned.

What we count as alive

Metabolism, boundary, heredity, evolution: textbooks list signs of life and then meet the awkward cases. A virus evolves and has heredity, and it has no metabolism of its own. A sterile mule is alive and will not found a lineage. A dormant seed looks like a stone and is a plant waiting for water. The useful move is not to force one essence. It is to say which criterion you are using for the decision in front of you, and to stop pretending the word “alive” does the same job in a hospital, a forest, and an origin-of-life laboratory.

The Miller–Urey result belongs here. It shows that some building blocks are chemically inexpensive under particular conditions. It does not show that a cell is an easy next step. The gap between a soup of amino acids and a system that copies itself is the actual philosophical problem, and it is still open.

Purpose is a description, not a cause

Hearts pump. That sentence can be a functional description — this is the effect that explains why this structure was kept — or a smuggled intention. Natural selection licenses the first reading and forbids the second. Nothing in development consults a future use and then builds toward it. The embryo follows chemistry that has been filtered by what ancestors managed to leave behind.

Teleological shorthand is hard to quit because it is short. “The kidney wants to save salt” is faster than the list of transporters. Use the short form only if you can expand it on demand into a mechanism plus a selective history. If you cannot, you do not yet have an explanation. You have a habit of speech.

The species problem

A species is sometimes a group that interbreeds, sometimes a branch on a tree, sometimes a cluster of traits a taxonomist can apply on a Tuesday. These definitions agree on oaks less often than they agree on tigers, and they struggle with bacteria, where gene exchange is not a family tree in the animal sense. That disagreement is not a scandal. It is what you should expect if lineages split gradually and if horizontal gene transfer is real.

When a paper says “the species,” ask which criterion was used and what would have been excluded by a different one. The blog on trees is the practical companion: tips are not ancestors of their neighbors, and “more evolved” is not a position on a diagram.

Parts and wholes

A protein’s shape is a fact about atoms, and a cell’s behaviour is not a fact you can read off one protein by being clever in a café. Levels of organization, on the concepts page, are a warning against two mirror mistakes. One mistake is to treat a molecular detail as if it were already the disease, the behaviour, or the ecosystem. The other is to treat “emergence” as a license to stop looking for a mechanism. A respectable higher-level claim still says which lower-level facts it depends on, and which it does not reduce to.

What an experiment can settle

An experiment settles a contrast: this factor against its absence, under these conditions, at this scale. Pasteur settled spontaneous generation for boiled broth, not the origin of the first cell. Meselson and Stahl settled the pattern of DNA strands, not the identity of the enzymes. A result can be rock-solid and narrow. Philosophers sometimes say “underdetermination” for the fact that more than one story can fit a single result. The practical reply is not despair. It is the next contrast.

Some questions are not waiting for a cleverer flask. Whether we should edit a human germline, or release a gene drive into a wild population, is a question about irreversibility, consent, and who bears the risk. Biology can say what the intervention would do, within the limits of the evidence. It cannot convert that description into permission. This site does not provide methods for doing either intervention. The omission is deliberate.