I thank Allen for recommending that I read Quentin Meillassoux’s After Finitude: An Essay on the Necessity of Contingency. Meillassoux is an innovative thinker on a philosophical mission. His goal is to re-establish a secure foundation for our scientific knowledge, lacking in modern philosophy, without returning to an outmoded metaphysics of the past. The key for him will be taking the notion of contingency to the limit. If there is truly no reason for anything (including physical laws), then, rather than seeing this as a limitation, we should embrace this as the one positive absolute truth on which we can build our foundation.
Meillassoux (hereafter Q.M.) is a French philosopher; the English translation is provided by Ray Brassier. (A talk given in England by Q.M. which offers an overview of his work is here -- however, in my case I needed to read the book for things to sink in.) In some ways, Q.M.’s writing betrays a bit of what I think of as “continental style” – including some tendency toward the grandiose, and an unfortunate penchant for creating new terms. However, the content transcends any intra-academic boundaries – he is dealing with big philosophical questions of perennial interest, and indeed he doesn’t invoke the work of any postwar philosopher with the exception of a shout-out to Alain Badiou (who also wrote the preface).
Below are my notes on the book; I’ll add some further thoughts of my own in a follow-up post.
Chapter 1 (“Ancestrality”) presents the path which Q.M. took to apprehending what he sees as the biggest problem in post-Kantian philosophy. Science endeavors to describe, in mathematical terms, the phenomena of nature, and takes the stance that what is being described are the things in themselves. The whole arc of modern philosophy, however, has been variations on the theme that we can’t escape the circle of subjectivity: what science is about, at best, is inter-subjective agreement about the outcomes of its observations and valid inferences drawn from them. What is being discussed is not things-in-themselves, but, depending on the philosopher, some description of a subject-object interaction or interface: Q.M. calls this general stance “correlationism” -- it is opposed to naïve realism. Post-Kantian philosophy sees no way to describe phenomena from the outside, or “objectively”; there is no path to the real-in-itself. Consciousness and language are always directed to the outside, but they don’t reach it.
But here is what fascinates Q.M.: science manages to talk about dates and events which predate humans, and life itself! He calls this “ancestrality”. How do we think about this kind of fact? Absent positing an absolute God or Mind which was around to perceive things, modern philosophers would have to say that what scientists are describing is still not the real-in-itself, but how it is “for humans.” They can’t accept the facts outright, given correlationism. So it causes a conundrum if you ask the correlationist outright: “Did the earth form 4.5 billion years ago or not?” If the philosopher rejects the naïve realist stance, he/she must reject it completely. And this is a problem for Kantian and post-Kantian philosophy: the distinction between these philosophies and old-fashioned idealism vanishes upon interrogation. For the Kantian position, note that even transcendental idealism implies the existence of a point-of-view. It is positioned in the world. So we can still pose the problem of ancestrality. Q.M. says that modern philosophy shares this problem broadly, whether it is the focus on human consciousness of the continental phenomenologists, or the language focus which has dominated a good deal of analytic philosophy.
What is at stake here is scientific truth in general – the case of these ancestral facts, particularly in their mathematical formulation, just brings this into high relief. We use mathematical science to go beyond what is given to us in experience: how is this possible?
Chapter 2 “Metaphysics, Fideism, Speculation”
So, what to do? How do we think about a reality which exists even if we do not? We can’t go back to Descartes and other pre-Kantians: although by reexamining the turn from Descartes and the rationalists we might get some clues. Q.M. discusses the principle of sufficient reason and the ontological arguments, and concludes these paths to an absolute reality can’t work for us. We need an absolute foundation, but because the critiques of these earlier efforts were sound we can’t return to those kinds of necessarily existing entities – that ship has sailed.
Now, a “strong” correlationist, unlike Kant, maintains that we can’t be sure of any necessity – of laws, of logic, etc. We’re not sure they’re not necessary either, we just don’t know. Q.M. refers to this as having the property of “facticity”. But if we can’t even be sure of logical necessities, doesn’t this mean “anything goes”? Q.M thinks that this kind of philosophy, where even our reasoning power is ungrounded, leads to an inability to criticize the irrational – inadvertently opening the door for “faith-based” or “mystical” paths to the absolute.
The philosopher is left with truth as inter-subjective agreement as grounds for science at best, or if you deny even this possibility, you are left with an even emptier post-modernism. This has been a path to the marginalization of philosophy.
Many philosophers have thought criticizing metaphysics goes hand in hand with debunking religion, but this is wrong. Philosophers need to recover some modicum of the absolute in order to be in a position to criticize religion effectively.
Chapter 3 “The Principle of Factiality”
Q.M. asks: can we discover a truth underlying this very principle of facticity (whereby laws of nature, etc. are indeterminate)? But how could we turn an inability into a positive thing – a new absolute? Q.M. proposes that the defeat of the principle of sufficient reason can be turned into something positive. He proposes that the lack of grounding of our knowledge is not just a result of our limitation, our finitude; it is a governing truth of the world – that it is without reason. The PSR is not just wrong, it is backwards. (This is the central pivot point of the book.)
And, Q.M. argues, this conclusion cannot itself be subjected to the correlationist-deflationist circle—that it is just a truth “for us” -- that would elevate the correlationist principle to dogma. Our new principle says “it is absolutely necessary that every entity might not exist.” It is a principle of unreason. It is a greater contingency – for it applies to everything, not just material objects in our world, but all laws as well. Our absolute is a principle of hyper-chaos.
Wait now: what happened to our project from Chapter 1 of finding a way to ground the mathematical descriptions of science? We have replaced ignorance of the thing-in-itself with positive knowledge of chaos which seems no better. What can we do with this hyper-chaos to make it useful? What follows from it?
Q.M. believes we can derive important things. First restate the principle in two parts as follows: 1. a necessary entity is impossible; 2. the contingency of the entity is necessary. Now, Q.M. concludes that a contradictory entity cannot be contingent, because it can’t become anything else, and something has to be able to change to be contingent. So this allows us to derive from the principles above the principle of non-contradiction (not in thought, but ontologically). Also, we can derive the principle that the things-in-themselves do exist. For facticity to be true, there has to be something which exists for it to apply to. Q.M. calls “factiality” the speculative essence of facticity. It alone transcends being a (contingent) fact.
Now, having derived the (Kantian) principles of non-contradiction and the reality of the in-itself, the next step will be to try to derive the (Cartesian) reality of the mathematical descriptions of the in-itself.
Chapter 4 “Hume’s Problem”
In this chapter, Q.M. deals with what he sees as a major objection to his move. If everything is contingent, including laws, the world could go haywire at any time (a concern which follow if, like, Hume, we exclude causal laws – hence the chapter title). Can we explain the manifest stability of physical laws if they are contingent?
First, note that stability does not entail necessity. And why should contingency imply frequent instability? It is a mistake to invoke the concepts of chance and probability here, since this presupposes (indeterministic) physical laws. The contingency of laws cannot be subject to the laws of chance. But going beyond this point, we want a positive account of how stability can be manifest from chaos.
Now, chance implies some sense of a denominator – a totality (finite but large, or infinite) that we can conceive. But when considering a whole world, we cannot a priori conceive of the totality of the set of all possible worlds, according to Q.M.: “…we have no grounds from maintaining that the conceivable is necessarily totalizable. (p.103)” Next, Q.M. references Alain Badiou’s work on the transfinite (a concept from set theory). The transfinite gives us the ability to distinguish contingency from chance. The transfinite is untotalizable. And this would deflate Hume’s problem (note that Q.M. doesn’t think he has demonstrated this is the exclusively correct answer, but he thinks its viability serves to indicate the problem is defeasible).
Q.M. claims that his approach offers a path to answering deep questions like this one (and his derivations above) which contemporary philosophy has approached with a deflationary mindset or else claims are meaningless. Q.M. wants to show the questions can be answered without lapsing into what he sees (here agreeing with the consensus) as discredited metaphysics. Ironically, by realizing there is no reason for the way things are, we arrive at a way to answer metaphysical questions. He sees the potential for deriving further conclusions, including, he hopes, the foundation for mathematical science.
Chapter 5 “Ptolemy’s Revenge”
The challenge of “ancestrality” is really one for all of empirical science, particularly the mathematical account of nature, which seems to let us know what is even where we are not. This is a world separable from man: the creation of the Galilean-Copernican revolution. We want an alternative path to the (essentially Cartesian) thesis, “that whatever is mathematically conceivable is absolutely possible (p.117)”; that is, contingent, but able to be independent of our thought.
The next sections of chapter 5 are basically a good old rant. Q.M. says the real Copernican revolution was in opposition to Kant’s thought, despite his invoking it as an analogy for his system. Kant’s was really a “Ptolemaic counter-revolution” which put the subject back into a central role (p.118). The major result of this move over time was to promote science over metaphysics as the path to knowledge while “serious” philosophers, now sidelined, would tell themselves that of course the knowledge gained wasn’t really objective knowledge.
Q.M. believes philosophers would have been better served over the past 200 years thinking about the question: “how is empirical knowledge of a world anterior to all experience possible? (p.123)” But Kant and post-Kantian philosophy followed up on its rejection of metaphysics only with different forms of “correlationism” which had no hope of answering the question.
Now we need to bring back the absolutization of the mathematical truths of science without the PSR and the metaphysics of necessary beings. Q.M. believes he has started the process by deriving some conclusions from factiality – the principle of unreason. Can we do the same for these mathematical descriptions of nature: i.e. show they are not necessarily true, but are absolutely possible? Also, returning the topic of Chapter 4, can we derive the absolute and necessary scope of the theorem regarding the non-totalizability of the transfinite? The next step in the project is to derive these two things from the factial.