I have to compare this book with another one which I recently bought: "Infinitesimal Calculus" by James M. Henle and Eugene M. Kleinberg. Both books are basically the same in that they use the concept of infinitesimals to provide a more intuitively satisfying basis for the concepts of calculus than the common, "delta/epsilon" limit approach. Yet the two could not be more different in the way they go about it.
Henle and Kleinberg's book uses a concept of infinitesimals developed by Abraham Robinson, known as "nonstandard analysis." In this system, an expanded number system, the "hyperreal number system," is created, which obeys almost all the same rules as the real number system but includes infinitesimals (numbers different from zero but smaller in absolute value than any other real number), as well as infinite numbers (larger than any real number) and finite but nonstandard numbers. By contrast, the "smooth infinitesimal analysis" used in this book has no infinite numbers, and does not obey the normal laws of logic (in particular, the law of the excluded middle). Bell is well aware of the difference between these two approaches, and gives detailed and valuable comparisons between them in this book.
Oddly, nothing could be further than infinitesimals from the ideas of the intuitionist mathematicians like L. E. J. Brouwer, yet Bell's logical system is based on the modifications to logic which Brouwer had to make so that his intuitionistic program could work. And Bell refers to his logical system as intuitionistic.
My own personal feeling is that nonstandard analysis has the merits of the logic being familiar and of its being based on the extension of the real number system in a compatible manner, but smooth infinitesimal analysis makes the mathematics easier to _do_ (as, in nonstandard analysis, it is continually necessary to extract the standard part of a nonstandard number, and a corresponding step is unnecessary in smooth infinitesimal analysis). So both have their merit.
Another contrast with Henle & Kleinberg's book is that the other book ignores applications, while this book is strongly oriented toward the use of calculus in physical applications.
I was tempted to give this book 5 stars, but I find the mathematics in some places rather dense and hard to follow, which was my reason for deducting one star. But I am glad to own both of the two books, this one and Henle & Kleinberg's.