‘Demi-regs’, probabilism and partly closed systems

ABSTRACT

In response to an article by Steve Fleetwood on the critical realist conception of open and closed system, this paper argues that we should differentiate non-closed systems according to their degree of openness or closure, rather than applying a dichotomy between closed and open system. Societies should be conceptualized in terms of partly closed (or partly open) systems, in addition to any fully open systems that might exist in the social world. The paper furthermore argues that demi-regularities represent probability distributions existing in the real world, as distinct from occurring merely in some epistemic domain. Such partial regularities at one level of the stratified reality are necessary for the emergence of new structures at a higher stratum, for example, the social structures of a given society. Finally, implications for social science research of accepting vs. rejecting the existence of probabilities and partly closed systems in the social world are discussed.

KEYWORDS:

• Open systems
• closed systems
• partly closed systems
• probabilities
• emergence

Acknowledgements

Many thanks to Doug Porpora for valuable critical comments on an earlier version of this paper, and to the two reviewers for comments on first submission of the paper.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributors

Petter Næss is Professor in Planning in Urban Regions at the Norwegian University of Life Sciences. He has contributed theoretically on urban sustainability, philosophy of science within the field of spatial planning, and methodology within infrastructure planning. A special interest in his research is relationships between urban spatial structures and travel, where he has made theoretical contributions concerning the causal status of the built environment. Based on critical realism, his studies on this topic differ from mainstream research by combining quantitative and qualitative methods, applying a multi-theoretical approach and explicitly emphasizing ontological and epistemological reflection.

ORCID

Petter Næss http://orcid.org/0000-0001-8571-8326

Notes

1 We of course should not claim a strong analogy between the distinction between open systems/partly closed systems and the distinction between possibilism and probabilism. Insofar as the probabilism/possibilism debate is analogous with the discussion of closed, partly closed and open systems, it is by virtue that the impossibility of certain activities represents some sort of closure, while the possible actions represent openness. And while the recognition of partly closed systems allows for different degrees of closure and the ensuing occurrence of demi-regs, the probabilist position acknowledges the different degrees to which possible actions are easy or difficult to carry out.

2 Here, the word ‘random’ is worth noticing. Fleetwood evidently uses the word in a particular meaning pertaining to statistical jargon, referring to processes ‘relating to, having, or being elements or events with definite probability of occurrence’, (Merriam-Webster Dictionary 2019), but very different from the synonyms of random listed in the same dictionary entry: aimless, arbitrary, catch-as-catch-can, desultory, erratic, haphazard, helter-skelter, hit-or-miss, scattered, slapdash, stray.

3 Here, the words in the parenthesis were directly copied from Roy Bhaskar’s comment on my initial draft of a book review (Næss 2008). At least, this suggests that Fleetwood’s rejection of well-grounded probability distributions in non-closed systems can hardly be claimed to be ‘the critical realist conception’, as suggested in the title of Fleetwood’s article.

4 However, Karlsson (2011, 159) presents two four-square tables according to which partly closed systems exist in situations where there is either external closure and internal openness (i.e. the extrinsic condition of closure is satisfied while the intrinsic condition is not satisfied), or internal closure and external openness (i.e. the intrinsic condition of closure is satisfied while the extrinsic condition is not satisfied). Instead of a conceptualizing of partly closed systems based on such a double dichotomy, I think both the external and internal closure can be partial, i.e. that there exist continuums between full closure and full openness both for internal and external conditions.

5 It should be noted that Porpora did not himself worry about any such ‘slippery slope’. He actually argued in favour of the use of regression analyses as an evidentiary tool, against some critical realist economists who seemed to fear that such acceptance would legitimize full-scale mathematical modelling.