# Sound and Fury

Signifying nothing

## Probabilities

Probability is something I am interested in. I’m interested in the philosophical aspects of probability. When I was a maths student, probability didn’t excite me at all. But I am interested in the philosophy of probability and I wanted to write something more about it. My philosophy of physics essay was about one aspect of probability in science. But I’d like to look more at probability in general. I thought maybe of making it the topic of my postgrad seminar talk. It would be an interesting diversion from writing my dissertation. On the walk back from the bookshop today (dammit. I forgot to by the book I meant to pick up…) I was planning out how the talk would look in my head. It would have three sections… I use bullet points quite a lot in this blog, so perhaps I’ll use the ennumerate thing this time…

1. Discuss some of the terminology of philosophical work on probability. Different writers distinguish different numbers of types of probability and give them different but sometimes overlapping names. So I’d talk about what people mean by; objective, subjective, epistemic, ontic, chance, credence and so on…
2. Look at which of the extensions of these terms overlap or are contained in one another. Come up with my own classification of probabilities and associated vocabulary. This would probably take the form of different terms on different axes, because I think probability is more than a one-dimensional concept. (Does that make any sense? I might have to think about how to explain that some more…)
3. As well as terminology and concepts, there is the interpretation of probabilty. So I would finish by looking at which interpretations cause what kind of collapsing of concepts. If you’re a determinist, objective probabilities are going to have to collapse somehow. If you believe in GRW you can’t be a frequentist. (I’ve mentioned this point before I am sure…)

I’d take a realist, indeterminist position for most of the talk, because I think that will allow you to differentiate the most types of probability. Then to avoid criticism of the “theory-ladenness” of my characterisation, in the third section I’d look at the problem through different philosophical goggles. Wearing different philosophical hats.

This got me thinking about another awkward point. If you’re a realist about objective, physical probabilities, how do you guarantee they satisfy the axioms of probability theory? I can’t see how you do it. Even if you replace Kolmogorov with some other axiomatisation (Luder’s rule conditionalisation or whatever) you still have to explain why the probabilities satisfy that system. Since the probabilities will be inherently independent how do you do this?

Right I have to stroll back up to the bookshop before it starts raining again. I need to pick up On the Shoulders of Giants. Because it will look good on my bookshelf next to God Created the Integers and The Road to Reality. Because it will be a good source for my history of science essay.