“That the sun will not rise to-morrow is no less intelligible a proposition, and implies no more contradiction, than the affirmation, that it will rise.” – David Hume
My understanding of Hume is via Deleuze who has an idiosyncratic reading. Hume was the subject of Deleuze’s first book Empiricism and Subjectivity and pops up again later as the focus of Pure Immanence, although Deleuze is concerned with partiality in Hume’s moral philosophy, as well as empiricism. This article is an attempt by me to get to grips with contingency in Hume in relation to Meillassoux’s critique of Kant’s solution to the problem of induction.
As an empiricist, Hume begins his philosophical investigations with everyday observations about the world. So he privileges direct experience and seeks to ascertain the relations between things that provide the reality of our experience, rather than building an immovable philosophical edifice. In so doing, Hume seeks to emulate the method of scientific empiricism and physics of his time.
For Hume, the mind is just a radical set of ideas (not the origin of experiential thought). This amalgam of ideas can then be endlessly reconfigured and restructured into new associations, so it is never predetermined or preprogrammed. In essence, this amalgam of endlessly re-combinable and dynamic ideas is what constitutes the ‘self’ or ‘I’. Accordingly, the self for Hume is a fiction that is determined by practice (habits or custom). There is therefore never any ‘universal’ laws or constancy that governs the relation of these ideas. If there are any ‘rules’ these are merely contingent and impermanent, much as we observe in science, which allows us to ‘suppose’ a set of contingent rules until we encounter new evidence that refutes the model that we have espoused. Hume had a distinct influence on Karl Popper’s philosophy of science here, particulalry regarding falsification.
Deleuze appropriates Hume from the group of Enlightenment thinkers that he is usually lumped with (Descartes, Locke and Berkley) and reckons him to be a philosopher of the future, despite some of his more obvious flaws. He even claims that Kant owes something essential to Hume. Hume seems as relevant as ever at the moment, as the reality that he problematizes still lies at the heart of philosophical debate. These problems are confronted in After Finitude by French philosopher and speculative materialist Quinten Meillassoux. He embarks on resolving some of Hume’s questions, which Ray Brassier investigates in turn in Nihil Unbound in the chapter titled ‘The Enigma of Realism’; which is eminently useful for unpacking these core controversies.
The question that Meillassoux principally addresses is the controversy of the problem of causality and Hume’s ideas regarding contingency and the uniformity of nature. This is commonly known as the Problem of Induction. Which is all fascinating.
If we look at any causal relationship (A follwed by B) we know through experience and observation that B always follows A; it has done so in all past occurrances. Yet Hume maintains that A and B are not bound by any universal principle of causation. So the problem is characterised by the question: why do we infer that the regularity of B following A, which has been observed until now, will hold in the future? How do we know the sun will rise tomorrow?
Hume’s response is that of ‘inductive inference’, which is merely the association of ideas and hence a psychological habit formed over time. In so doing Hume casts doubt on inductive reasoning; which is presumed to be the core of scientific enterprise. This critique of induction casts doubt upon the verifiability of scientific theory. For Hume it is logically impossible to verify a universal proposition by reference to experience. In response to this argument Karl Popper develops his falsificationist philosophy of science, through which science’s law-like generalisations are not inductively verified, but deductively falsified.
Therefore, an experiment can never verify a scientific law; it can only falsify it, since a single counter-example allows us to deduce its falsity. This very much correlates to the importance of contingency in Hume. So, even the best corroborated scientific theories could be overthrown by one counter-instance. Part of the problem of Popper’s science philosophy is that it presumes a uniformity of nature, but Hume’s critique of causality undermines this.
Hume is fascinated by the discrepancy between the vast realm of logical possibility and narrow domain of empirical actuality. Brassier uses Hume’s example of billiard balls to illustrate this:
When I see for instance, a billiard-ball moving in a straight line towards another; even suppose the motion in the second ball should by accident be suggested to me, as a result of their contact or impulse; may I not conceive that a hundred different events might as well follow from this cause? May not 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 preference to one, which is no more consistent or conceivable as the rest? All our reasonings a priori will never be able to shew us any foundation for this preference. [David Hume, Enquiry Concerning Human Understanding]
Whilst there is no good ‘reason’ why we should find one outcome more plausible, our ‘belief’ in causality encourages our predicted outcome to be necessary. This is the uniformity of experience that Hume says we predict through habit; so the nature of belief is somewhat habitual. It is therefore merely habit that allows us to perceive nature as anchored in a uniformity of occurrences. Habits generate the appearance of uniformity. We have an irrational faith in the sun rising tomorrow through necessity; we believe it will through the constancy of appearances.
By confiscating the ground of uniformity from reason, Hume abandons it to faith. In so doing he paves the way for Kant’s natural theology, the transcendentalisation of uniformity and his legitimation of fideism (synonymous with christian apologetics and apposite to strong corelationism under Meillassoux’s definition). Kant completes the question of whether constancy should be a necessary feature of our experience of phenomena. This is the question that Hume begins by rooting uniformity in association and association in habit. Kant comes up with a ‘frequentialist’ solution to the problems of uniformity and constancy, which Meillassoux then counters with an ‘anti-frequentialist’ answer in which he deploys mathematics to demonstrate that Kant’s solution is untenable. If Hume denies the rationality of our belief in the reality of uniformity — without denying that uniformity itself — by identifying belief as originating in habit, he transforms the problem of induction into a question of how our experience of constancy is possible.
This is the aporia that Kant takes up in the The Critique of Pure Reason and it falls to him to complete it. Kant in turn transforms it from a metaphysical question about reality-in-itself, into a transcendental question about our experience of phenomena. Kant surmises that: If there is no constancy in appearances, then no representation of appearances (i.e. phenomena) would be possible.
So, in summary, since representation manifestly occurs, it suffices to refute a sceptical hypothesis of a world without constancy. Kant concludes that in a world in which causality and the principle of uniformity are suspended — a world in which not only billiard balls are unpredictable, but also their entire global context (the table, the walls, etc.) — such a world would be unrepresentable to ourselves, therefore unimaginable and an impossibilty. For Kant all such phenomena would be subject to endless chaotic transformations, which is inconceivable.
Kant therefore claims that uniformity is not a necessary feature of noumena (things-in-themselves), but that the possibility of consciousness and representation itself requires the constancy of phenomena. It is essential to it. One presupposes the other to produce a global stability of context. So in The Critique of Pure Reason Kant goes on to demonstrate that the law-like regularities which science finds in nature are essential and necessary features of phenomenal reality.
Meillasoux, in After Finitude, infers from Kant’s answer to the problem of induction a ‘transcendental legitimation’ for the necessity of uniformity. This is perhaps an unstated inference, but nonetheless detectable. Meillasoux observes that if representation presupposes constancy, and constancy requires a uniformity, one cannot conclude that uniformity is necessary. This relies on assumption and perhaps to legitimation via transcendental recourse.
Kant would have it that if phenomena were inconsistent and the laws of nature based in contingencies, it is the frequency of their transformations that would render them unrepresentable. This is why we could characterise Kant’s solution as ‘frequentialist’, as the rate of capricious transformation in an unstable global context would preclude the act of synthesis whereby the mind reproduces the properties of appearances from one moment to the next:
If cinnabar were sometimes red, sometimes black, sometimes light, sometimes heavy, if a man changed into this and sometimes into that animal form, if the country on the longest day were sometimes covered in fruit, sometimes with ice and snow, my empirical imagination would never find opportunity when representing red color to bring to mind heavy cinnabar. [Immanuel Kant, The Critique of Pure Reason]
So the ‘sometimes’ of Kant’s solution is a unambiguous reference to a ‘putative’ frequency that precludes representation. Meillasoux’s counter-argument calls this into question. Meillasoux then sets about constructing a ‘speculative’ solution to Hume’s problem in which he posits reality as a storm of hyper-chaos.
Ray Brassier, Nihil Unbound: Enlightenment and Extinction (2008)
Gilles Deleuze, Empiricism and Subjectivity (2001) and Pure Immanence (2001)
Quinten Meillassoux, After Finitude: An Essay on the Necessity of Contingency (2008)