Abstract

Defining life remains an unresolved problem because prevailing approaches rely on lists of functional traits—such as metabolism, reproduction, or evolution—without explaining why these features cohere uniquely in living systems. This limitation has become increasingly consequential with the rise of synthetic biology and artificial intelligence, which reproduce isolated life-like behaviors without clearly constituting life. Here, I propose an ontological framework inspired by Karl Popper’s three-world model, integrating physical embodiment (World 1), interpretive agency (World 2), and symbolic abstraction (World 3) into a single causal architecture. I argue that life is not defined by any particular biological function or functions, but by the reciprocal closure between these three domains: physical embodiment, interpretive processes and symbolic abstraction. This triadic ontology provides a substrate-independent definition of life, a logic to derive the functional traits of life, resolves longstanding ambiguities at the margins of biology—most notably in the case of viruses and artificial systems—and offers a unified conceptual foundation for life detection, synthetic life design, and the interpretation of emerging non-biological life-like systems.