Metabolism and the problem of its universalization

Metabolism tends to be conceived either as an operationally closed network of production of components or as an autonomous apparatus of management of energy flows. Taking up some recent ideas that connect the concept of auton omy with thermodynamic requirements, we move further to defend the hypothes is that there must be a deep intertwinement between the relational-construc tive logic of a basic biological system and the logic of its thermodynamic implementation. Hence, we propose that metabolism should be universally def ined as the recursive self-maintenance of controls upon the energy flows ne cessary for the physical realization of a component production system opera tionally closed. Finally, being critical with some claims of the so-called 'strong' artificial life approach, we try to show that present 'computation al metabolisms' are necessarily different in their structure and functionin g from any real metabolic system, due to the distinct type of causal relati ons and mechanisms which are respectively established in them. (C) 1999 Els evier Science Ireland Ltd. All rights reserved.