The paper investigates a semiotic conception of life. As a notion or general idea of life it is seen as a member of a set of definitions bordering science proper and philosophy of nature, here called ontodefinitions. The received view of definitions in science (according to which definitions of life are virtually non-existent or meaningless to pursue) is criticised, and the semiotic notion of life is related to the emergent character of a simple living system. Defining life as biosemiotic processes seems to imply the emergence of functionality as a kind of “biological meaning” in the physical world. The relevance of definitions is context-dependent, and one such context is Artificial Life (AL) research. A “strong version” of Artificial Life claims it possible to synthesize and thus realize life computationally or by other means. If life should be defined in terms of semiotic processes intrinsic to nature, then semiosis must be required to take place in any system that realizes life.

In this note epistemological problems in general theories about living systems are considered; in particular, the question of hidden connections between different areas of experience, such as folk biology and scientific biology, and hidden connections between central concepts of theoretical biology, such as function, semiosis, closure, and life.

I will investigate the plausibility of three theses: (1) The Umwelt theory of Jakob von Uexküll, even though his theoretical biology was often characterized as being thoroughly vitalist, can in the context of contemporary science, more adequately be interpreted as a branch of qualitative organicism in theoretical biology. Qualitative organicism is a position which claims, first, a kind of middle road position, that is, on the one hand, there are no mysterious or non-material vital powers in organisms (non-vitalism), but on the other hand, the characteristic properties of living beings cannot be fully accounted for by physics and chemistry because these properties are nonreducible emergent properties (emergentism); second, that some of these emergent pro- perties have an experiential, phenomenal, or subjective character which plays a major role in the dynamics of the living system. Modern bio- semiotics (inspired by C. S. Peirce and Jakob von Uexküll, instituted by Thomas A. Sebeok) is a kind of qualitative organicism. (2) This position sheds light on recent discussions in cognitive science, artificial life, and robotics about the nature of representation and cognition – indeed genuine semiotic questions as they deal with the role of information and signs for any system that has the property of being ‘animal-like,’ that is, systems that move by themselves and seem to be guided by a kind of entelechy or, in modern but shallow terms, a behavioral program. (3) Particularly, qualitative organicism allows us to approach the question of whether a robot can have an Umwelt in the sense that Jakob von Uexküll used the term (a subjectively experienced phenom- enal world) The eventuality of a positive answer to this question, i.e., a claim that a robot indeed can have an Umwelt, seems counterintuitive to the extent that a robot may be seen as – to use a bewildering word – an incarnation of the mechanical and reductionist world picture to which Jakob von Uexküll was so strongly opposed. But certain ideas and concepts may sometimes lead us to unexpected consequences, which threaten our cherished metaphysical assumptions, and we should try to face such questions with an open mind.