For the most part this was pretty much what you'd expect --- a definition of democracy, a defence of the idea, and a description of various practical implementations.
My main disappointment is that the various practical implementations are given so little space, after so much space is devoted to the (IMHO far less interesting) defence of democracy.

The most interesting idea I learned from the book is that representative democracy is a fairly new (late 1700s) idea that is apparently not obvious, and that was considered (perhaps justifiably) to be one of the greatest inventions of the age.

Along similar lines was that it took the US constitutional congress two months to come up with the idea of an elected president to run the executive (as opposed to the British equivalent of a monarch); for example one idea floated was that of an elected monarch (ie, I guess, president-for-life).
It could certainly be argued that mistakes were made in this regard. The unitary US presidency, defined as both the (non-political) head of state and the (intensely political) head of government, and, in practice, also the head of a political party, does not seem to provide a mechanism in the system for a person generally respected, mostly apolitical, but able to act in moments of crisis in ways generally considered to be in the nation's best interests (for a recent example, consider the King of Thailand). Dahl does not discuss this sort of issue, but I think he should have.

Also interesting was a binning of democratic political systems into four main groups:
* European : parliament + proportional representation
* UK : parliament + first pass the post
* US : president + first pass the post (he doesn't like this)
* South American: president + proportional representation (he thinks this is really bad)
This is the sort of political "science" I want to learn more about, and I wish he'd concentrated more on this sort of thing. ( )