Quite a small book but pretty full of stuff. It's got a relatively long introduction by John Skorupski, which I had more trouble with than the book. Basically Skorupski seems to be trying to provide a summary of Russell's arguments, rather than introducing them. And I don't think it's easy to summarise without missing details, so I don't think you need to bother with the introduction (unless you're going to read it after the rest of the book).
Although the ideas require you to think about them, the writing is not too technical, and especially at the start, it's quite matter-of-fact. Mostly it is concerned with epistemology and metaphysics, not much on ethics. He emphasises the distinction between "knowledge by acquaintance" and "knowledge by description".
I did mark some of the passages in the book but I'd be making the same mistake as Skorupski if I tried to summarise the ideas here. One thing I did sketch was a tree representing different sources of knowledge, based on his text at pp62-63:
+-- immediate : self-evident truths | [intuitive] | +-- truths --+ | | | | | +-- derivative : deduced from self-evident truths | | knowledge--+ +-- universals (e.g. sensible of | | qualities) | +---- immediate ----+ | | [by acquaintance] | +-- things --+ +-- particulars (sense data) | | | +-- derivative : derives from (acquaintance [by description] + knowledge of truths)
The other thing which he emphasised a couple of times was the fact that we often don't know things for certain. He says that memory is an example of this: an event which just happened, I can be fairly certain of; but for something that happened ten years ago it's less easy for me to know. In one passage he says "self-evidence has degrees: it is not a quality which is simply present or absent, but a quality which may be more or less present, in gradations ranging from absolute certainty and down to an almost imperceptible faintness" [p67]
Also he talked about coherence theory, which I thought he started off by making sounding pretty plausible. But he also says "coherence as the definition of truth fails because there is no proof that there can be only one coherent system"
Completed : 10-Dec-2004
The text of this book is also available online (without the introduction!)