Firstly, if you haven’t seen my previous post, go there now and leave a comment. I’m hoping to get some feedback about what people would like to see on this blog. In the meanwhile, here is another post in the ‘Work in Progress’ series. This time a survey article of sorts based on the lectures that I gave in Geneva last December, entitled ‘Varieties of Modality’. I was hoping to get this published somewhere like Philosophy Compass, but it seems that I entered the party a bit too late, as they are not intending to commission any more modality stuff at this time. It may be difficult to find a home for this, as it is really a survey article, although I do entertain some rather wild ideas towards the end of the paper…

The question that I pursue in the paper is how many different kinds of modality – different realms of possible worlds – are there? Philosophers commonly talk at least about metaphysical, conceptual, epistemic, logical, physical, mathematical, biological, technological, normative and natural modality. It is not always clear how these different types of modality are related, or whether some of them are more fundamental than others. The relationships between metaphysical, conceptual and logical necessity and possibility are particularly interesting. The paper is a survey of our options in this regard. We can distinguish four approaches which are currently widely discussed: the Kripkean approach, the conservative approach, the conceptualist approach, and the essentialist approach. The differences between these approaches are best described by comparing their takes on the distinction between metaphysical and conceptual modality. The Kripkean approach holds that this distinction is genuine and that we are dealing with two different kinds of modality. The conservative approach, which is familiar for instance from Bob Hale’s work, challenges the role of metaphysical modality and suggests that logical necessity is the most fundamental type of modality, it is absolute. The conceptualist approach, most forcefully argued for by Frank Jackson and David Chalmers, also questions the distinction and suggests that metaphysical modality can be fully accounted for in terms of conceptual modality. Finally, the essentialist approach, defended especially by Kit Fine, suggests that conceptual and logical modality can be seen as species of metaphysical modality. I will also consider an alternative approach based on the essentialist approach, which takes metaphysical modality to be absolute in Hale’s sense.

You can download the paper for the survey bits, but what’s this crazy alternative approach..? Well, if we take the cue from the essentialist approach and consider logical and conceptual necessity as subspecies of metaphysical necessity, as Kit Fine suggests in his ‘Varieties of Necessity’ (2002), then I think we already have the tools for something a bit more radical. Firstly, we can rule out all extra-metaphysical possibilities — that is, possibilities such as water being XYZ, when we consider water to be essentially H20 — as pseudo-possibilities. What this means in practice is that there is no stronger type of necessity than metaphysical necessity; in fact, metaphysical, conceptual and logical necessity would all seem to be equally strong. But I think that we can go even further, and indeed that we must go further if we wish to maintain that conceptual and logical necessity are useful notions at all: otherwise it seems that we might just as well talk only about metaphysical modality. But if we reserve the notion of metaphysical modality to those modal truths which are not true in virtue of either the definitions of concepts or the laws of logic, and similarly for conceptual and logical modality, we get a rather surprising picture about the relationships between different kinds of possibility:

Metaphysical, Logical, Conceptual, and Physical Possibility

What makes this interesting is that, according to the line suggested above, the picture for necessity is exactly the same. This is obviously a rather strange and seemingly contradictory result, but there may be a way to accommodate it. The idea is that only metaphysical modality is fundamental, but there is still use for the notions of conceptual and logical modality exactly in the same sense as there is use for the notions of physical or biological modality. So, according to this picture, different subspecies of metaphysical modality should be considered as concerning the natures of specific subsets of the set of all things. Hence, conceptual modality concerns things that are possible or necessary in virtue of the natures of concepts, and only them. Specifically, although it would commonly be considered that something like ‘It is possible to travel faster than light’ is conceptually possible, according to this picture this is not strictly correct: the possibility of travelling faster than light is not ruled out by the natures of concepts, but nor do the natures of concepts make it possible to travel faster than light. For something to qualify as a conceptual possibility, it has to be made possible by the nature of concepts in this positive sense. A similar analysis applies to logical possibility.

Well, this doesn’t really do justice to the idea, and I’m not quite sure that it even works, but the survey of other approaches that precedes this alternative picture in the paper might motivate the approach somewhat. There’s also a lot more to say about the status of logical modality here and I do go into it in some detail in the paper. If the idea is at all feasible, it would seem to require revamping modal logic as well — that’s something that I will not attempt.