Some types of semantic content have a potential or a requirement to project, i.e., to get interpreted outside the semantic scope of various operators such as negation, conditional and question operators, modals, etc., despite appearing to be in their syntactic scope. But what is the exact classification of projecting content, and how do we know which type of content projects in which way? In this dissertation I argue for a composition-driven approach to projection of pieces of content that have their own nodes in the morphosyntax, and I argue that this approach applies to both spoken and gestural content. Under this approach, how a given piece of content projects is uniformly determined by how it composes in the syntax/semantics, as opposed to a variety of alternative analyses for different types of content. I show that this approach is efficient in explaining the projection behavior of a range of content types, including spoken adjectives and appositives, hand gestures, facial expressions, "phi"-features on pronouns, and certain iconic properties of directional hand gestures. This dissertation thus has two main implications. First, it offers a general economical and explanatory approach to investigating projection of compositionally integrated content. Second, it establishes the need for treating content-bearing gestures as bona fide linguistic objects at all levels of representation. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]