In papers or books, citation to another book usually doesn't contain the specific page, section, chapter of where a result is borrowed. If the book is really thick, and the readers may have different knowledge levels and familiarity with the book, some readers may find it not easy to locate the borrowed result within the reference book. So why don't people specify the source of a citation as detailed as possible in books?
BTW, it is good to specify as detail as possible for citation to a paper. But since a paper is usually much shorter than a book and it is usually in a searchable electronic form, it may be much easier to find the source in a paper than in a book.
There is also often a good positive reason to cite a book without reference to specific chapters or subsections, and that is if one is pointing the reader to a source of review or introductory material. I personally often find it much better to point to a comprehensive survey (which is equally often a book rather than a journal paper), rather than a giant and certainly incomplete list of individual references, especially in formats where the number of pages or reference counts is limited.
This is especially the case when doing cross-disciplinary research. For example, I recently had a reviewer query how our paper could assert something that is such common knowledge in my field that I wouldn't have even thought to cite it. Thus, in the revised paper we cite an appropriate undergraduate textbook. Pointing to a specific element inside the book wouldn't have made sense, since you really need the whole foundation. It would be absolutely inappropriate, however, for us to attempt to reproduce an undergraduate class in the text of our paper.