This book presents a typology of subordination systems across the world's languages. Traditional definitions of subordination are based on morphosyntactic criteria, such as clausal embedding or non-finiteness. The book shows that these definitions are untenable in a cross-linguistic perspective, and provides a cognitively based definition of subordination. The analysis is based on a representative 80 language sample, and represents the broadest study so far conducted on the cross-linguistic coding of several types of complement, adverbial, and relative sentence. These sentence types display considerable structural variation across languages. However, this variation turns out to be constrained, and appears crucially related to the functional properties of individual sentence types. This work is the first systematic attempt to establish comprehensive implicational hierarchies describing the coding of complement, adverbial and relative sentences at a single stroke. Concepts from typological theory and cognitive linguistics are integrated to account for these hierarchies.