Universiteit Leiden

nl en


Workshop Anti-Humean metaphysics

Thursday 29 September 2016
Johan Huizinga
Doelensteeg 16
2311 VL Leiden

This workshop brings together five young philosophers interested in developing anti-Humean ideas. We will discuss possibility, causation, lawhood, dispositions and powers, and also take a look at the possibility of a more Kantian approach to contemporary problems.

Attendance is free. Prior reservation is not required, but if you know you are coming, a quick e-mail to the organiser (V.Gijsbers@hum.leidenuniv.nl) is appreciated. If you wish to join us for dinner afterwards – at your own expense, sadly – please let the organiser know before the 20th of September.


Although there has never been full consensus, analytic philosophy has always had a strong tendency towards a broadly Humean epistemology and metaphysics. The empiricist idea that all knowledge about the world is based on the observation of actual, singular, non-modal events and that, therefore, all metaphysical claims must be traced back to such observations, has been extremely influential. Yet the success of this approach has not been overwhelming. Concepts such a lawhood, causation, possibility, necessity, disposition and power have continued to resist analysis into Humean concepts, generating a vast literature of often baroque complexity. Recently, more and more philosophers have started to wonder whether it may not be the Humean presuppositions of the discussion that are at fault.


14:00 – 14:40    Victor Gijsbers, Possible Worlds and Real Possibilities
14:40 – 15:20    Irene van de Beld, A Kantian Approach to Laws of Nature


15:40 – 16:20    Frank van Caspel, Humean or Anti-Humean Causation as an Ontological Demarcation Criterion? Neither...
16:20 – 17:00    Jesse Mulder, The Limits of Humeanism

Break with drinks

17:20 – 18:00    Niels van Miltenburg, Manifestations as Processes

Final drinks

Victor Gijsbers (Leiden)- Possible Worlds and Real Possibilities

Analytic philosophers have learnt to understand the modal concepts of possibility and necessity in terms of possible worlds. Although often billed as a harmless semantic tool, possible worlds in fact perform a critical function for the broadly Humean metaphysics of David Lewis and his followers: they remove possibilities from the actual world. That it is possible for this vase to break does not mean, according to possible worlds semantics, that this vase has the modal property of fragility; instead it means that a relevantly similar vase in some other possible world does in fact break. Thus, by adopting talk about possible worlds, one at the same time abolishes talk about inner-worldly possibilities.

In this talk, I will argue against possible worlds and in favour of a world of possibilities. I will attempt to show that while possible worlds semantics can successfully make sense of statements of conditional possibility – statements such as “if we keep the laws of physics fixed, you can't move faster than light” – it cannot make sense of statements of unconditional, real possibility – statements such as “you can't move faster than light.” In particular, attempts to analyse such statements in terms of conditional probabilities fail to preserve their sense. And yet it is really true that I just cannot move faster than light, or flap my arms and fly, or write this e-mail in fluent Russian. This failure of possible worlds semantics suggests that a very important part of our thinking about possibilities cannot be given a Humean construal.

Irene van de BeldA Kantian Approach to Laws of Nature

Kant is well-known for his idea that we can only obtain knowledge about the world in an interplay between the understanding and the senses. Thus, he refutes the Humean or empiricist claim that knowledge is based on observation alone. In this paper I will especially focus on the consequences of Kant’s idea for the concept of laws of nature, starting from a famous passage in which Kant makes clear that “the understanding is itself the source of the laws of nature”. (CPR A127)

First, I will attempt to give a coherent interpretation of the passage cited above, since the idea that the understanding adds to or even ‘prefixes’ something in our knowledge of the laws of nature has been a source of much controversy. I will thereby contrast my interpretation with more empiricist readings given by e.g., Michael Friedman. My argument centers around the idea that only the general ‘form’, but not the laws of nature themselves, are prescribed by the understanding. Second, given that this workshop focuses on contemporary anti-Humean ideas, I will address the main question of this paper: to what extent is Kant’s work still valuable for current concepts of laws of nature? I will approach this question by discussing several historical accounts of scientific breakthroughs that in my view fit into a broadly Kantian scheme, and then explicate what principles were presupposed. I will point out that one of the main principles that have to be presupposed by the understanding is the idea that nature is uniform and regular in character. Lastly, I will argue that this Kantian approach gives a more satisfactory result than a Humean or empiricist view, especially because it seems much more fit to explain the nature of a scientific ‘paradigm’.

Frank van Caspel (Open Universiteit / Nijmegen) - Humean or Anti-Humean Causation as an Ontological Demarcation Criterion? Neither...

Processes should not have the same ontological status as whatever entities engage in them. 'To be is to be a cause', appears at first sight to be an attractive (non-Humean) demarcation criterion that can be used to argue for this point. However, a popular non-Humean theory of causation, interventionism, can accommodate processes as causes without issue, thus weakening the proposed demarcation criterion. In my presentation I will muse upon this predicament and argue that we need another notion than causality (be it Humean or non-Humean) to clarify the difference in ontological status between processes and their performers: performance.

Jesse Mulder (Utrecht) - The Limits of Humeanism

Humeans take reality to be devoid of ‘necessary connections’: things just happen. Causality and laws of nature are to be understood in terms of what ‘just happens’, not vice versa. Here the Humean needs some conception of what it is that ‘just happens’ – a conception of the Humean mosaic. By exploring different such conceptions, I delineate the limits of Humeanism in two ways: first, only conceptions of the mosaic of a very specific kind form a suitable basis for the Humean project; and second, taking the Humean way of thinking to its limit results in a rejection of the whole idea of such a mosaic – and hence of Humean mosaic-based accounts of anything. For concreteness, I develop my argument in dialogue with Humean accounts of laws of nature, but it applies to any Humean account involving a Humean mosaic.

Niels van Miltenburg (Utrecht)Manifestations as Processes

Traditionally analytic philosophers have thought that dispositions are problematic entities because they are unobservable (logical positivism) and/or modal (neo-humeanism). Hence these philosophers have long since tried to reduce the dispositional to the non-dispositional. But agreement about the proper way to reduce dispositional properties has never been reached. This is mainly due to notorious problem of manifestation prevention (finks, masks and antidotes). Perhaps because of the continued unavailability of a satisfactory reduction, more and more philosophers are starting to develop the idea that there are properties that are perfectly real and yet truly dispositional: powers. These so-called anti-Humeans about dispositions hold that a power is a modal property that necessitates its manifestation when it is triggered. Unfortunately, however anti-Humeanism is itself also susceptible to the problems of manifestation prevention, which seem to show that a power does not necessitate its manifestation because a power can be triggered but nevertheless fail to manifest.

In this paper, I will argue that this problem only arises because the anti-Humeans are not anti-Humean enough: While they have reintroduced modal connections into the world, they still think about the connected events – the trigger and the manifestation – in a broadly Humean way. It is my aim to present an alternative to this Humean view on triggers and manifestations by arguing for the existence of the metaphysically sui generis category of process. Finally I will argue that the subsequent strong anti-Humeanism about powers and processes is immune to the problem of manifestation prevention.


This website uses cookies.  More information.