The possibility of naturalism Book

I was rather surprised to enjoy this book quite so much, since normally I am not particularly impressed by philosophy and philosophers. However, Hollis is even-handed, open-minded, undogmatic and very knowledgeable. He realizes all the approaches to the social sciences have their flaws, but many different ones have their uses. It is easy to run down the social sciences as being"fuzzy", but they're fuzzy because they deal with extremely complex assemblages of complex objects (such as people) where many of the important details of which can not be directly observed. But the subjects of the social sciences are vitally important for society to understand, even if we don't deal with them very well yet.

Physics, usually held up as the model of the sciences, is by comparison, simple (not that it is simple in the absolute sense). But as you go up from physics to chemistry to biology to historical sciences such as climatology, ecology or oceanography, not only do the amount of data to be considered at any one time increase dramatically, but so does the amount of emergent phenomena. By the time you get to the social sciences, the methods of the physical sciences don't really work very well any more, and you are left with fuzzy methods. But if you're going to be stuck with fuzzy methods, you better understand the assumptions and consequences of your methodologies. You need to understand their underlying philosophies.

So why did I read this book if I'm not normally a fan of philosophy? Well, as I get older I realized that there are some problems where maybe philosophy is the right tool to use. One of such problem, was the realization while reading Debt: The First 5,000 Years by David Graeber that possibly the whole social science of economics was potentially an invalid field of study as there was no such thing as "the economy" separate from society, and while there could be economic geography, economic sociology or economic anthropology, economics by itself didn't make any sense as a discipline. But what intellectual tools could you use to address such a problem? One really wants something better than "it's a matter of opinion" or "a matter of taste" (and one really wants something better than to say that how the social sciences are split up is just an arbitrary social construct of our society (although that might, alas, be true)). So I read this book. I'm not sure it helped with my problem (maybe a little), but reading it was its own reward -- a lot of smart people have spent a lot of effort trying to address these problems, and their solutions, while flawed, are well worth thinking about.

I will note that I was aided in understanding this book by a recent reading of Popper Selections, some previous exposure to the philosophy of the physical sciences, and a degree in a social science (Geography).