It is well-known that languages vary in their constituent order possibilities; some have very fixed word order, others very flexible. Scrambling is the purported movement operation that derives the non-canonical but perfectly acceptable orders observed in flexible word order languages. Early generative work on Scrambling greatly enriched the theory of movement (esp in the discussion of A vs Aʹ-movement) as research in the 1990s on Scrambling focused on differences among Japanese, Hindi, Germanic, and a range of other languages).

In the first part of this 1-week course, we will review the major accomplishments of early generative research on Scrambling, especially the work of Saito 1992, Mahajan 1994, and Webelhuth 1995, and their possible integration into broader theories of movement. In the second part, we will look at how Scrambling interacts with more recent approaches to displacement, especially featural Relativized Minimality (Rizzi 2004) and phase theory (Chomsky 2001), and what it tells us about the overall organization of the grammar. Along the way, we will also try to address some of the radically non-derivational approaches to free word order within generative grammar which either do not consider scrambling a movement process at all or require significant deviation from standard approaches to known movement operations.