The notion of a textbook is a historically contingent category.  
Treating textbooks aimed at the college market differently from other books  
may make sense nowadays, but in the past, as the examples of Smith, Mill and  
Marshall make clear, there was no clear demarcation. In parts of the  
humanities this is still the case.  
 
If students do not read their textbooks, why do they buy them, given the high  
prices at which they sell? Are students not economizing agents who purchase  
only if the utility exceeds the cost? 
 
Roger Backhouse