Sunday 21 September 2014

Evolution, Natural Necessity and Meta-Ontology

Evolutionary theory is characterised by an attitude to causation which Hodgson calls 'causal determinacy': simply, every event has a cause which may be determined. This is a contrasting position to the hard-science 'causal determinism' where it is held that knowledge of causes can be used to predict future events. 'Causal determinacy' situates causes within the realm of a natural world characterised by event regularities which can be determined and become knowable. Nobody would assume that a knowledge of evolutionary mechanisms could be used to predict future evolved states, but it does seem reasonable to say that event regularities studied in the ordering of species in the natural world can have transfactual causes identified. That's basically what Darwin did on the Beagle.

However, the approach to natural necessity and its connection to causation has been polarised since Hume and Kant. Critical examination of the differences between positions has only recently emerged in disputes regarding ontology between Critical Realists and a growing group of 'speculative realists' who follow the lead of Badiou and others. The central problem, as Meillasoux has argued lies in a remarkable passage in Hume's "An Inquiry Concerning Human Understanding":
"When I see, for instance, a billiard ball moving in a straight line towards another; even suppose motion in the second ball should by accident be suggested to me, as the result of their contact or impulse, may I not conceive, that a hundred different events might as well follow from the cause? May not both these balls remain at absolute rest? May not the first ball return in a straight line, or leap off from the second in any line or direction? All these suppositions are consistent and conceivable. Why then should we give the preference to one, which is no more consistent or conceivable than the rest? All our reasonings a priori will never be able to show us any foundation for this preference."
The question is, if Hume's scepticism about natural necessity is upheld, what does this mean for the mechanisms of Darwinian evolution? How might it relate alternative evolutionary descriptions? Is there a relationship between ontological assumptions of Darwinian theory and characterisations of ontology which promote natural necessity? What then of our understanding of evolution in society or in organisations? Moreover, if alternative ontologies not grounded in natural necessity can be articulated, how can competing ontological descriptions be evaluated?

Hume's problem with billiard balls forms the basis of Meillasoux's attack on the doctrine of natural necessity. Hume argues that we cannot make any such assumptions about event regularities in the natural world: Meillasoux agrees. Bhaskar's 'common sense' argument against Hume's causal idealism revolves around the fact that event regularities under closed conditions produce laws which have efficacy in the open world: so closed conditions and event regularities are not necessary conditions for the discovery of physical laws, and by extension (transcendentally) there must be natural necessity in what Bhaskar terms the 'intransitive domain'. Meillasoux, supported by Badiou and others, presents a counter proposal. In reconsidering Humean regularity theory as a theory of human expectation and probability, Meillasoux suggests that the problem lies not with regularity and reproducibility, but in the domain of mathematical probability and the logical possibility of identifying totalities within which probabilities can be determined.

What opens up here is an ontological 'cloud', ranging from strong positions of natural necessity and transcendental reasoning (from Kant to Bhaskar) through to rejecting natural necessity in favour of an ontology of logic and truth (Meillasoux, Badiou). Alternative ontologies demand some kind of approach to meta-ontology: ways of negotiating competing ontologies. Evolutionary thought is at the heart of this meta-ontologising: it is in the transcendental descriptions of the ontogeny and phylogeny of the visible natural world that underpins transcendental arguments about what must or must not be the case with reality. We either assume the evidence of the ancestors indicates natural necessity and mechanisms of which Darwinism is one example, or we draw the conclusion that the problem of 'ancestrality' is really a problem of 'being' and the consequence of confusion among scientists who fail to make a distinction between primary qualities (i.e. mathematizable, logical qualities) and secondary (or relational) qualities (see Levi Bryant's excellent discussion on Meillasoux at http://larvalsubjects.wordpress.com/2009/04/02/meillassoux-i-primary-qualities-and-correlationism/)

The theories of Darwin, Lamarck, Wallace and Spencer sit in different places in this ontological cloud. For example, if we were to uphold the Lamarckian view of the inheritance of acquired characteristics (which Darwin himself didn't entirely reject), we would sit in a different space to Spencer's insistence on the 'survival of the fittest'. In each position, there is an aspect of ancestral regularity, but there is also an aspect of "being" and of primary and secondary qualities. The assumption of ancestral regularity in Spencer is most strong, with its connotations of the blind watchmaker who through mechanisms of chance and fitness establishes the present order of nature. Lamarck, by contrast, never loses sight of 'being': his is a work focused on explaining being, taking the ordering of nature as its starting point. Lamarck attempts to articulate a logic that connects being to nature. The primary qualities of logic and the secondary qualities of relations inherent in observations push Lamarck's ontology towards the 'being' side of the ontological cloud.

It is also worth considering more recent post-Darwinian accounts. For example, the systems theoretical views of Stuart Kauffman articulate mechanisms which with more detail and sophistication demonstrate how life might emerge from a molecular level upwards. Kauffman's model tends towards a 'causal determinism' rather that 'causal determinacy'. Dynamic modelling of processes highlights an ontological position which is somewhere in-between one of natural necessity and one of 'being'. There is some logical necessity in the algorithms and heuristics which are proposed although this is mixed in with the assumptions about natural necessity both in the processes of exploration of heuristics on the computer, and in their comparison with data from the natural world.

The real question concerns the social import of evolutionary theories and their ontological grounding. Evolutionary theories which evoke natural necessity tend towards social characterisations which are historicist. Politically, such articulations point to the 'inevitability' of a decision for something or other (our VC likes this kind of thing!). Evolutionary accounts which privilege being (like Lamarck) present multiplicities, not singularities of possibility: the power of being and the power of agency to transform not only itself, but also historical trajectories.  The deep question is whether an evolutionary account opens up or closes down possibilities. 

No comments: