This paper examines the case for a structural syllabus in the light of second language acquisition research. It argues that a structural syllabus cannot easily serve as a basis for developing implicit knowledge of a second language because of the learnability problem-learners are often unable to learn the structural properties they are taught because the manner in which they are taught does not correspond to the way learners acquire them. It is possible, however, to envisage a role for a structural syllabus based on a weak interface model of L2 acquisition. This role consists of intake facilitation (i.e., causing learners to pay attention to specific formal features in the input and to notice the gap between these features and the ones they typically use in their own output). A structural syllabus can also serve as a basis for the construction of problem-solving tasks designed to develop learners' explicit knowledge of grammatical properties. It is argued that this knowledge may facilitate subsequent intake. The role proposed for a structural syllabus, therefore, is a substantial one. It is recognized, however, that such a syllabus will need to be used alongside some kind of meaning-based syllabus, which is designed to provide learners with opportunities for communicating in the second language.