[PoSt] Philsophy of Statistics

[bryangoodrich]

I thought this would be an interesting read. While it is not directly about PoSt, it is about PoS and the relationship between philosophy and the sciences as practiced. This is an important issue, as I put on my philosopher hat, because I find great value in philosophy. However, I also can sympathize with the arguments that philosophy has fallen behind by not keeping up with the sciences. I think a small subset of naturalist philosophy is really the only route to keeping philosophy alive with respect the sciences. Otherwise, it's really nothing but an adjunct to the literary and historical disciplines. While that can be valuable, I find it's greatest merit in its relationship to application, particularly to the sciences. Now, I am a bit biased, since my focus in philosophy began with ethics and ended with the philosophy of science, particularly philosophy of mathematics, philosophy of economics, and rationality. All of these have been partly with respect to abstract thinking, but largely with regard the fact it's looking at the forest instead of the trees that comprise these scientific disciplines I'm passionate about. While the discourse on PoSt is light, the consequences to a divorce of philosophy and science apply just as focused on PoSt as other areas that could be lumped into the philosophy of science.

Enjoy

http://www.guardian.co.uk/science/li...ggs-philosophy

[bryangoodrich]

Since I'm using this Feedly app on my phone to subscribe to the only RSS feed I read, I've been keeping up on some of the articles that get updated or added to the Stanford Encyclopedia of Philosophy (SEP). Today I read a very nice one that lays out a topic I'm very interested in that has direct application to PoSt. The author makes a point to distinguish this topic from issues in inductive logic. Instead, this requires an understanding of formal logic and probability theory, because the point of probability logic is to take classical (propositional) logic, at the very least, and include probability functions to track uncertainty (error) propagation. Very interesting, especially when you start looking at more complex logics that are available (modal, prepositional, relevant, and more).

http://plato.stanford.edu/entries/logic-probability/

If anybody can track down these articles from one of my top five favorite philosophers (he teaches just a few hours from me too v_v), I'd love to read them!

Van Fraassen, Bas. 1981. Probabilistic Semantics Objectified: I. Postulates and Logics. Journal of Philosophical Logic, 10: 371–391.
–––. 1983. Gentlemen's Wagers: Relevant Logic and Probability. Philosophical Studies, 43: 47–61.
–––. 1984. Belief and the Will. Journal of Philosophy, 81: 235–256.

If anybody is wondering who my other in the top 5 philosophers are, in no particular order: John Stuart Mills, Nancy Cartwright, Thomas Kuhn, and John Searle. I'd have to throw Amartya Sen in there, but I don't know if he's every been considered a philosopher, proper. He's an economist from that classical tradition that is highly theoretical and philosophical, which is why I love his work.

[noetsi]

Probably a good thing I never went to Stanford. I got lost on the 2nd sentence above

[noetsi]

Until the "smoking debate" attitude of statistics was not used to establish causality.

They still aren't People decided instead that it was likely that smoking caused cancer based on the odds. Only in the last decade or so has it become clear (well somewhat clearer perhaps) what the physical process of smoking causing cancer. Which was tied to medical research/discovery not statistics. If you can't 1) create a theory to explain something and 2) show some of the basic mechanisms involved (at least in theory) you really don't have a basis to claim a causal link regardless of statistics.

Just ask Justice Scalia who just last year argued statistics can never be proof because there could always be an alternative cause

[bryangoodrich]

I think the point is that people just didn't think of statistics as supporting causal claims. Their efficacy is another matter. But you cannot separate "medical research" from statistics considering the design of those experiments on many cases is a product of statistics! On a philosophical note, there is much in the way of how theoretical models can justify causal claims. The key here is that you can never prove them. You merely justify the claim. Theory itself just provides an explanation of how that causal relationship works. But it is entirely possible that this explanation is accurate in prediction but not how the mechanics actually work. Something else is required to make a theory (explanation) actually how the phenomena occurs.

[bryangoodrich]

Since I am God-Master of PoST now, I figured I should share this gem I was told about last week. Along with the Stanford Encyclopedia of Philosophy (SEP), this is going to become one of my major sources of relevant research. It's focused on the philosophy of science and contains many pre-prints. For somebody like me that doesn't have access to JSTOR or some of the other closed-access aggregators, this is a valuable resource. If I find stuff on PoST, I'll definitely pass it along from there. I've subscribed to the RSS for 2013 so I'll always see the latest additions. I do the same with SEP to see new entries and revisions made.

http://philsci-archive.pitt.edu/

[bryangoodrich]

I guess this properly would go here: http://philsci-archive.pitt.edu/9774...probchance.pdf

"Chance" crops up all over philosophy, and in many other areas. It is often assumed -- without argument -- that chances are probabilities. I explore the extent to which this assumption is really sanctioned by what we understand by the concept of chance.

I'll say a few words once I find the time to read it. The abstract is utterly lacking in content, so it makes me question how good the content will actually be. Nevertheless, there's your PoSt reading for this week!

[spunky]

i know this may sound weird... but are there any VLOGS like youtube stuff preferably about this? i've started to realize that cardio sessions are great learning moments for me if i can just find the right podcast or video to listen to...

[bryangoodrich]

Oh I know, that's how I've been getting my cardio all year. I bike at the gym watching Coursera lectures!

Personally, I prefer reading about philosophy. Often lectures don't do it justice and are better for discussions. I do enjoy when prominent philosophers give keynote speeches and such, but in the philosophy of science? Eh, that's not too common! I'll keep an eye open for any videos I come across, though.

[spunky]

i caught this documentary like WAAAAAAAY a few years back on 4am re-runs and found it fascinating. then i saw that youtube has it as well.

does it look/sound interesting? you're right, ideally one should be able to read and reflect on what is being read... unfortunately (at least for me) real life says otherwise so i need to make-do with whatever i can...

http://www.youtube.com/watch?v=cG3sfrK5B4E

[bryangoodrich]

Hey, I've done a lot of philosophy and economics reading on a treadmill! Of course, then you're not working out very hard. Tradeoffs!

[bryangoodrich]

And that series I found on YouTube out of order. I put it in correct order (to watch straight through) on my YouTube account. Hilary is a rare philosopher. He could do a good job explaining PoS. I might find random clips and make a playlist. Could be a useful resource on the future.

[spunky]

I might find random clips and make a playlist

O-M-G do it! and i'll become your first subscriber! and in exchange i can share with you the most epic lolcatz videos you've ever seen..