418 Wataru Hashimoto
CC: Creative Commons License, 2011.
Homeostasis, introduced by Walter Cannon, belongs to this generation. The second-generation
system is a self-organizing system, a typical of which can be found in crystal deposition in a
beaker. While the first-generation views the hierarchical relationship as stable, the second-
generation is concerned with the realization of the hierarchy itself. Despite this noteworthy differ-
ence in approaches to hierarchy, both the first-generation and the second-generation systems are
characterized by system openness. In addition, it is an observer that sets the boundary between a
self-organizing system and its environment.
The third-generation system, that is, autopoiesis can be distinguished from both the first-
generation and the second-generation system in terms of hierarchical relationship and the position
of observers in theorization.
From the original definition of autopoiesis that Maturana and Varela (1973/1980, pp. 78-79) give,
two notable features of an autopoietic system are obtained. First, autopoietic systems are autono-
mous, and furthermore, they have no inputs or outputs. Secondly, atutopoietic systems’ boundaries
are not drawn by an observer, but rather “their operations specify their own boundaries in the proc-
ess of self-production” (Ibid, p. 81). For an autopoietic organization to be realized, therefore, neither
hierarchical relationships nor an observer is required, or rather is relevant at all.
According to Kawamoto (1995), autopoiesis can be put at the latest generation. This contrasts
markedly with Hayles’ explanation of the history of cybernetics. Hayles (1999) divides the devel-
opment of cybernetics into three waves. The first wave, from 1945 to 1960, regarded homeostasis
as central. This is nearly identical to what Kawamoto calls the first-generation system. The second
wave, roughly from 1960 to 1980, developed the concept of reflexivity and put an emphasis on
cybernetics of observing systems, that is, second-order cybernetics. Heinz von Foerster, Humberto
Maturana and Francisco Varela belong to the second wave. The invention of autopoiesis by
Maturana and Varela constituted the mature point of the second wave. According to Hayles, the
third wave from 1980 to the present has developed on the basis of virtuality. Central to the third
wave is the development of artificial life and cellular automata.
Although this chronological explanation of cybernetics is easy to understand, her assessment of
the second wave is absolutely irrelevant. Clarke and Hansen (2009) criticize her discussion as an
attempt to delineate a “third-wave” cybernetics. On the contrary, Clarke and Hansen put an em-
phasis on the novelty and effectiveness of “neocybernetics”. The term “neocybernetics” is intended
to suggest that second-order cybernetics and autopoiesis are not to be surmounted. In this respect,
Kawamoto shares his view of autopoiesis with Clarke and Hansen, and we can assess Kawa-
moto’s explanation as valid.
It is necessary, however, to explain the difference between Kawamoto and Maturana with regard
to the notion of atupoiesis. Kawamoto (2000, 2002, 2006) reformulates the concept of autopoiesis
in his own original way. His redefinition of autopoiesis is aimed at expanding “kôi-sonzairon” (act-
ontology). Kôi-sonzairon suggests that as long as the operation of an autopoietic system is main-
tained, its boundary is self-determined, and the system (or an entity) is realized and comes into
being. Kawamoto sets koi-sonzairon apart from another type of ontology, that is, “ninshiki-
sonzairon” (recognition-ontology). Ninshiki-sonzairon means that existence or objects are con-
structed by the act of observation.
Note that Kawamoto’s approach is quite different from Maturana’s one. Maturana’s conception of
observation and Spencer-Brown’s calculus of indication are parallel. In his work Laws of From,
Spencer-Brown (1969) formulates the calculus of indication, whose implication is that “a universe
comes into being when a space is severed or taken apart” (Spencer-Brown, 1969, p. v). For
Spencer-Brown, drawing a distinction allows an entity to exist. Likewise, Maturana proposes the
constitutive ontology of observers. Maturana defines the notion of observer as follows:
An observer is a human being, a person, a living system who can make distinction and specify
that which he or she distinguishes as a unity, as an entity different from himself or herself that
can be used for manipulations or descriptions in interactions with other observers. An ob-
server can make distinctions in actions and thoughts, recursively…. Everything said is said by
an observer to another observer, who can be himself or herself (Maturana, 1978, p. 31).