|Gregory Bateson defines information, or the basic unit of information in communicational and mental processes, as "a difference which makes a difference" (e.g. Bateson, 1972:453). And in Niklas Luhmann's terminology, observation means nothing more than handling distinctions - making a difference in this context involves an observing system. Luhmann, following George Spencer Brown, defines 'observation' as indication by means of a distinction (Luhmann, 1998: 167ff; Luhmann, 1989: 144). In terms of the simple model of cognition in figure 2, observation involves both representation and action besides perception. An indication can be seen as a simplest kind of reference to the environment - a reference to this as distinct from that - by means of a distinction entailed in the action and perception process, in the form of for example the sense apparatus, attention, or conceptualisation. In accordance with the necessary relationship between the different aspects of cognition, Bateson pictures perception, representation and action in a circular cybernetic process, where the transform of a difference travels in circuit (Bateson, 1972: 458f). In more complex learning processes the three aspects gain a more independent existence, as in the models, experiments and observations of research - and this seeming independence makes the acknowledgement of their relationship all the more important.|
|Investigating the process of observation as part of cognition, we shall
look at Niklas Luhmann's formulation of a theory of cognition from a systems
theoretical approach, in the tradition of constructivism, see for example
his "Erkenntnis als Konstrktion" (Luhmann, 1998). In the tradition of cognitive
idealism, the cognitive problem of the unity in the difference between
cognition and the real object starts with the question: how is cognition
possible, even though it has not of itself independent access to outside
reality? Or: how can a subject know anything outside itself, when any such
cognition cannot take place independently of the subjects cognition? The
radical constructivist approach, on the other hand, starts with the empirical
statement: cognition is only possible because it has not access to realities
outside itself - because it is operationally closed (Luhmann, 1998: 164).
The subject theory of cognition has never addressed this question, Luhmann
says, because it has always wrestled with the paradoxical quest of inferring
the relation of others to the world by way of introspection. Therefore
it must presume a common world, or at least a commonly observed world.
In a constructivist approach, we are able to ask the contrary question
of subject theory: how is the disconnection (Abkopplung) of the cognitive
system possible (Luhmann, 1998: 165-66). An object theory of cognition,
on the other hand, also fails to address this question, because the reduction
of the description of cognition to processes within the object leaves out
the problem of disconnection.
The question of disconnection is approached by replacing the fundamental distinction between subject and object with a distinction between 'system' and 'environment'. This approach maintains the classical cognitive problems connected with subject-object theories, by making a distinction in cognition where one side of the distinction re-enters the other , but it also transgresses the classical problems by revising both subject and object theories. The question of disconnection is asked as a question of differentiation and operational closure of systems and the premise of a common world is replaced by a theory of observation of observing systems (Luhmann, 1998:164ff). Operational closure of systems is possible by way of the system producing and reproducing its own operations thereby maintaining a distinction between system and environment. This is what Humberto Maturana and Francisco Varela calls autopoiesis in their answer to the questions "what is life?" and "what is cognition?", using the concept in a description of organisms as cognitive systems. They consider first of all unicellular organisms as autopoietic systems. The neural system and metacellular organisims (e.g. multicellular organisms and colonies) are operationally closed systems, and meta-cellular organisms are considered second order qutopoietic systems (menaing no more than that they consist of cells), while they leave the question open as to whether mea-cellular organism are (proper) autopoietic systms (Maturana and Varela, 1987: 96-98, 156ff). Luhmann extends the concept of autopieses to other kinds of stystms - he distinguishes between biological, psychic, and social autopoietic systems.
Although Luhmann uses autopoiesis theory as one point of departure for his general systems theory, he does not adopt the conception of cognition originally connected with the theory. Maturana and Varela characterise cognition as effective action, an action that allows a living being to sustain its existence in a certain environment, as it reproduces its world - no more, no less (Maturana and Varela, 1987: 44-45) . Luhmann (1998: 167) presupposes a concept of cognition as observation based on distinctions and indications, and this allows him to extend the concept of autopoiesis beyond autopoietic living beings. But is also entails an approach quite different from the original autopoiesis theory. Maturana and Varela's concept of operational closure of a system's organisation (which is not quite clear from the sources I have) entails that the identity of the system is specified in a network of dynamic processes, the effects of which do not leave the network (Maturana and Varela, 1987: 98, 157). Luhmann's concept of closure allows for the disconnection of the cognitive system and thus for cognition, as discussed above. He asked: "How is closure possible? Probably only by this, that a system produces its own operations and reproduces them in its network of recursive progresses and regresses. The process itself creates the difference between system and environment."  (Luhmann, 1998: 167). That is, closure presupposes autopoiesis. And "the concept of autopoiesis can be extended to the social domain only when the elements of social systems are conceived as communicational acts (events) and not as persons, roles, subjects, individuals, etc." (John Bednarz's introduction to Luhmann, 1989: xi).
Taking this approach, Luhmann may be neglecting an important aspect of social systems, having to do with Gregory Bateson's 'analogue communication' [12, my note LB], since Luhmann's entire system is based on observations and, fundamentally, distinctions. It is important to keep this in mind, when drawing general conclusions from the insights of autopoietic theories of cognition. Where Maturana and Varela takes the position that all cognition is effective action and Luhmann only recognises observation as indication by means of a distinction, I shall leave the question open as to whether there is more to cognition than either one of these positions.
Luhmann's theory of cognition does, however, have a series of important implications. First and foremost, that all observation is dependent on distinction, and that the recognition of the environment therefore is dependent on the distinctions applied in observation. The distinction with which a cognitive system observes, is its "blind spot" or latent structure, because this distinction cannot be distinguished in the observation - if it were, it would be distinguished by means of another distinction, with its own blind spot. Any observation presumes and produces a spitting of unmarked space (Luhmann, 1998: 168f).