Structural Realism, Metaphysical Unification, and the Ontology and Epistemology of Patterns

ABSTRACT

Ladyman and Ross’s account of the metaphysical component of ontic structural realism was associated with a unificationist view of the connection between fundamental physics and special sciences. The aim of the present article is to assess the sense of unification that is at issue in Ladyman and Ross’s definition of metaphysics. Given the ontic core of Ladyman and Ross’s version of structural realism, it should be assumed that the unifying endeavour is worthwhile only if the connective links that underpin unification are metaphysically significant. Ladyman and Ross employed information-theoretic notions, e.g. ‘projectibility’, to account for the significance of real patterns, which underpin unification. I build upon McAllister’s engagements with the same topic, to argue that these notions fail to accomplish this objective.

Acknowledgements

The author has greatly benefited from the comments and suggestions of the three anonymous referees of this journal as well as the editorial team. This debt is gratefully acknowledged.

Notes

1 Despite the existence of subtle differences, in this paper, I use ‘metaphysical’, ‘ontological’, and ‘ontic’ interchangeably.

2 It should be noted that their emphasis on the fundamental role of physics does not commit them to foundationalism or micro-reduction. As I will explain, Ladyman and Ross bypassed Oppenheim and Putnam’s (1958) foundationalism about physics. Oppenheim and Putnam’s micro-reductionism was associated with a hierarchical conception of the organisation of sciences. The hierarchy was supposed to wind up at the level of particle physics and unification consisted in reducing different sciences (sociology, biology, molecular biology, etc.) to the atomic physics. Questioning Oppenheim and Putnam’s atomistic conception of the basic level of elementary particles, Ladyman and Ross rejected micro-reductionism as the philosophy of A-level chemistry at work. However, this did not push Ladyman and Ross to also forsake their view of the metaphysics of science as unification between difference scientific disciplines and domains with an eye to fundamental physics.

3 It is also worth mentioning that the idea of projectibility is technically connected to Bennett’s (1988) notion of logical depth. Building upon Collier’s previous use of the same notion, Ladyman and Ross defined logical depth as

a normalized quantitative index of the execution time required to generate the model of the real pattern in question by a near incompressible universal computer program, that is, one not itself computable as the output of a significantly more concise program. (Ladyman et al. 2007, 220)

4 At times, Ladyman and Ross’s theory finds a Kantian, epistemological undertone. For example, in clarifying their notion of information, they asserted that the world is not made of anything and information is a fundamental concept for understanding the objective modality of the world, for example, laws, causation, and kinds (Ladyman et al. 2007, 189). This assertion is in line with Kantian or even instrumentalist nature of information. But then again, there are numerous instances of paying homage to the ontic form of OSR in the book.

5 On another occasion, McAllister (2007) also considered other techniques for choosing among data models on the basis of being closest to the truth. The techniques consist of the minimum description length theory and the Akaike information criterion. While none of these corresponds to Bennett’s criterion of logical depth or Ladyman and Ross’s conception of projectibility, they can generally indicate that computational methods for choosing between data models cannot work in the expected objective and agent-independent way.

6 McAllister considered the techniques of algorithmic information theory, minimum description length theory, and the Akaike information criterion.

7 As I remarked before, Ladyman and Ross presumed that being counterfactual-supporting generalization is identifiable with being projectible by the maximally efficient computer that is physically capable of performing the projection in question (given the actual physical limitations on computers) (Ladyman et al. 2007, 225–226). But they did not explain why this characterization of counterfactual-supporting generalization should carry metaphysical weight.