This thesis explores the research question: to what extent does the idea of 'social science' persist in constructing IR textbooks in our contemporary context? In light of the considerable changes in the discipline's canon catalysed by the Third Debate, it is surprising that there has been no sustained engagement with a large body of textbooks in over a decade. The thesis uses a folklorist approach to review contemporary undergraduate IR textbooks by exploring their family resemblances to Donkeyskin and Bluebeard stories. The thesis finds that many contemporary textbooks resemble Donkeyskin stories, both because they employ a problem/choice structure that works to curtail how IR is defined and because they rely on a number of assumptions about what it means to write a textbook and study IR. However, the thesis also finds that there is a limited but notable body of textbooks that resembles Bluebeard stories in terms of how 'forbidden' assumptions about how IR is defined and what it means to write a textbook are confronted. These two readings of textbooks are complemented by a third aspect of the folklorist approach, a reading of framing gestures. While many textbook authors employ framing gestures that cast authors as curators of the field of IR, reinforcing strict boundaries for participation in the negotiation of the IR canon, there are notable exceptions that cast textbook authors as creators. The effect is to demonstrate and open up for participation the process of negotiating the boundaries of what gets to 'count' as IR.