The Structure of Scientific Revolutions

I was interested in this book after I saw it referenced in a couple of other books I read. Perhaps it would have been better to leave it that way. The author has some good ideas; however, his delivery could use a lot of work. I found myself dozing off every time I picked up the book. The author tends to be extremely wordy, and goes out of his way to make sure he doesn't miss any minute detail. (There is even a long postscript, where he apologizes and corrects some of the details he did miss.)

The core thesis is that scientific revolutions are generally evolutionary changes to an existing paradigm. "Normal" science involves solving small problems in a narrowly defined scientific paradigm. Occasionally questions arise within the paradigm. These can eventually lead to challenges to aspects of the existing paradigm. New solutions may "improve" on it until a revolution produces a new paradigm. Old practitioners are often reluctant to adapt, but eventually they do.

Scientists working within a paradigm tend to look with the eyes of the existing system. They tend to color their results by what they expect to see. (This observational bias is present in all aspects of life and has been elucidated with a number of different studies.) A new paradigm can appear very revolutionary because it unleashes many observations that would not have been noted (or even "observed") in the prior paradigm.

When textbooks are written, the historical evolution of science is given fairly short treatment, thus making science appear to be a more logical progression than it actually is. (And conversely making the revolutionary events appear more significant than they really are.)

The arguments in this book are very convincing. However, you have to wonder if they are colored by the scientific paradigm in which they were written.