International Encyclopedia of Systems and Cybernetics

The ability of a system to maintain its identity, in spite of the progressive replacement of its elements.

This concept, practically equivalent to autopoiesis, was enounced in 1951 by J.W.S. PRINGLE in these terms:

(An open system) "has the following properties:

1. It is a steady-state equilibrium, each unit appearing to be constant only because it is being formed and destroyed at the same rate;

2. Either or both processes are subject to variations;

3. The rate of formation and the rate of destruction of each unit depend on different functions of the quantity present, and on the environmental conditions.

"These features imply a property of the units which may be called "replication"; that is, the presence of a unit in the system causes more similar units to appear, with an accompanying increase in order which is balanced at steady – state equilibrium by the decrease in order involved in their destruction" (1956, p.92).

Self-reproduction is one of those ambiguous terms in cybernetics and systemics. W.R. ASHBY, who died before the autopoiesis concept was fully explored, could write in 1962: "… before we start to consider the question of the self-reproducing system, we must recognize that no organism is self-reproducing" (1981, p.72).

His argument was that the reproduction of a system could never occur without a movilizing interaction with some other system or with a specific environment and stated that "the adjective "self-reproducing" is highly objectionable semantically and logically" (p.72).

This did not deter him to dedicate his paper to the "Self-reproducing system". He came to the clear conclusion that self-reproduction terminology should be used only for cyclical processes of the type "A reproduces B, then B reproduces C, and then C reproduces A" (p.78) and recognized that "Such a cycle is of course extremely common in the biological world" (p. 79)… where "All sufficiently large systems will become filled with self-reproducing forms" (p.79).

Thus, genuine self-reproduction is always an endogenous process, and the exogenous production of a whole copy or the original system, as for example in some of J.H. CONWAY's Games of Life, should not be called self-reproduction, as it depends of an algorithm external to the original configuration itself, which, by the way, is also the case in biological (not self-) reproduction.

ASHBY was, after all, on his own way toward organizational closure and the hypercycle.