Customer Reviews

Lectures on Boolean Algebras

5 star:		(2)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

As always, Halmos is an excellent expositor. The brief first chapter has been known to scare away physicists and other intelligent people who don't happen to know ring theory, but fortunately the first chapter can be skipped. The book is at the right level to be used by mathematics graduate students learning Boolean algebras and Stone spaces for the first time.

Some people draw a sharp distinction between the concepts of "Boolean space" (a totally disconnected compact Hausdorff space) and "Stone space", the difference being that a Stone space is the Stone space _of_ a Boolean algebra. A Boolean algebra's Stone space is the space of all of its 2-valued homomorphisms with the topology of pointwise convergence of nets of such homomorphisms. That every Boolean space is the Stone space of some Boolean algebra (namely, the Boolean algebra of all of its clopen subsets) is one of the important facts of "Stone's duality". Halmos never mentions the phrase "Stone space", but he proves the basic facts about "Stone's duality": that the category of Boolean algebras and Boolean homomorphisms is the opposite of the category of Boolean spaces and continuous functions.

The flaw is in Lecture 21. Much of that section is founded upon an error -- that a Boolean algebra may have various non-isomorphic completions, of which one is "minimal". A careful mathematician can reconstruct that lecture and get much that is of value, but some may be unfortunately misled.

One other thing irritates me: Halmos uses the word "non-atomic" rather than the much better term "atomless". The problem with "non-atomic" is that it may be mistaken for "not atomic", and that is a quite different thing.

This book is now out of print. I'd like to see Dover reprint it.

A very good book, going through the stuff in a very nice kind. Giving you intuitive insights, easy proofs, and written with a large sense of humor.