The meaning of open systems

Abstract

There has been considerable discussion lately of the concept of open systems, which has revealed that different participants are using the terms ‘openness’ and ‘closure’ in different ways. The purpose of this paper is to address issues of meaning that arise in this particular discourse, with a view to clarifying both conflicts in usage and the underlying issues involved. We explore the different meanings of openness and closure extant in the literature, as applied at the ontological and epistemological levels, focusing on our own use of the terms in relation to that which prevails in neoclassical economics on the one hand and to the use made of them by Tony Lawson and other critical realists on the other.

Keywords:

• open systems
• closed systems
• ontology
• epistemology
• theory construction
• abstraction

Acknowledgements

We are grateful for the comments of Andy Brown, John Davis, Geoff Hodgson, Jesper Jespersen, George Krimpas, S. G. (Fieke) van der Lecq, Tony Lawson and Menno Rol. They are not responsible for the errors and omissions that may remain.

Notes

1. Systems theory (Bertalanffy 1968 is the locus classicus) has found principles that are common to systems regardless of their content. It has been applied to economics by Boulding (see, for example, 1956).

2. One commentator pointed out that this is a definition from thermodynamics; and asks, ‘why go to the dictionary when you can tailor to needs’? Apart from the fact that, since the definition refers to systems of ideas as well as to the natural world, the scope of the definition goes beyond thermodynamics, there is an important issue of principle here. A similar point was made by Hicks (1965: 15–16, 23) concerning equilibrium. He maintained that economics should have its own definition, which he thought should be the meeting of preferences. First, such a tactic breaks continuity with the use of language across disciplines and cultural life in general. We do not favour a private language for economics, even if it were generally agreed within the subject. Our second objection is that, in the face of a lack of consensus, no single individual, even one with a considerable constituency, should have the right to ‘tailor to needs’, that is, to determine definitions for the subject as a whole.

3. The term ‘ambiguity’ is used here not in the special sense of the ambiguity literature, that something knowable is not known, but rather to refer to the unknowability of demonstrably true knowledge. It follows from the openness of social reality that knowledge about it is inevitably held with uncertainty.

4. The term ‘isolation’ is another that is given different meanings. It is most closely associated with the work of Mäki (1992), to which we refer in sections 2.3.1 and 2.4.2.

5. An additional source of confusion is the reference to closed‐economy and open‐economy models to indicate whether international trade and payments are, respectively, excluded or included.

6. These two approaches can be classed as partial representation and idealization (Rappaport 1996: 217), or legitimate and bogus abstraction (Lawson 1997: 232–7), respectively. Mäki's (1992) distinction between isolation by omission and isolation by idealization does not correspond directly to these two approaches, since they are discussed only in terms of isolated systems, and for him abstraction is a subset of isolation. We discuss the debate between Lawson and Mäki on the method of isolation further in section 2.4.2 below.

7. ‘To study an open system by using any form of closure…disfigures the landscape of an open system’ (Nash 2004: 76).

8. Expressed by him as a discussant to Chick's INEM (2002) paper, now Chick (2004).

9. The seeds of this idea were planted as early as the mid‐1970s (in a lecture given by Chick in Perugia).

10. In Chick and Dow (2001), we omitted to thank Jesper Jespersen for organizing the seminar in methodology of which these conversations were a by‐product. We make good that omission now, with apologies.

11. The resulting correspondence between theory and reality is however not direct, since reality is understood to be stratified, and there is no necessary synchronization between the different levels.