As an undergraduate sociology student, I'd say it's pretty common for upper-division sociology courses to depend largely (or even exclusively) on articles from the library's journal databases rather than on textbooks.
In my experience, sociology books tend to be either (A) shallow overviews of several schools of thought or (B) narrowly-focused arguments seeking to apply a particular theory to a specific social process. The former are so shallow that they should only be used at the introductory level. However, the second type of book is pretty limited in scope; reading one or two of these wouldn't familiarize students with all the different perspectives they ought to be aware of.
Articles are a good middle ground between these two extremes. Though they have the same limitations as narrowly-focused books, the shorter length of articles means that instructors can assign a number of different pieces, each of which highlights something different. Classes which use articles (or a combination of articles and more narrowly-focused books) encourage critical thinking and analysis, while those which rely on overview books come down to little more than memorizing some over-simplified definitions.
Other reasons I've heard for using articles include:
- Textbooks are expensive. Students don't like having to buy them.
- Using articles gives instructors more flexibility -- they can pick the best text for each topic they plan to cover instead of using one or two textbooks which may handle some topics poorly.
- Overview textbooks are dull and usually make vague generalizations about different schools of thought. Articles which apply those different perspectives provide more concrete examples which and more engaging and help students understand the practical meanings of various theories.
- Some courses have objectives like "introducing students to the standards of the discipline." Reading peer-reviewed articles helps students become familiar with the standards of sociological writing and gives them examples of how social research is performed.
It would definitely be possible for someone to write a book which had more depth than the usual overview textbooks and still covered a range of different schools of thought. I suspect that at least part of the reason that no one has is that most sociologists are pretty focused on specific schools of thought and/or subsections of society and/or methodologies -- they might be too specialized (or too devoted to their own preferred theories) to write a more general textbook.