There are some puzzles that, despite years of careful study and incremental pro- gress, leave one longing for a deeper understanding. In generative grammar and formal semantics, the relative order of adnominal modifiers has a claim to being foremost among them. To be sure, a great deal of real progress has been achieved over the years, and the topic has garnered sustained examination in many places. Among many others, this includes Cinque’s landmark 2010 volume on adjectives, which one might regard as a kind of intellectual ancestor of this work. And of course, it has been a topic of typological research and research outside of the formal linguistic tradition. And yet, after all this, I can’t really claim to know why—at least in some truly deep sense of ‘why’—language after language seems to agree that an adjective of shape should occur higher in its nominal than an adjective of color. A priori, it’s a baffling fact. Why would languages be designed this way and not another? Why, in all their diversity, would they agree on this, something so seemingly arbitrary?

In this work, Min-Joo Kim faces this problem and its cousins head-on, but in a particular instantiation that isn’t widely familiar. She approaches the issue from the vantage point of Korean. With a wealth of new data and important new observa- tions, Kim sheds new light on old puzzles and offers fresh new ones that will no doubt continue to vex us for years. She proposes a generalized universal conception of the nominal extended projection, working broadly in the cartographic tradition, but characterized in a way that should be relevant to researchers approaching the topic from a variety of perspectives. For example, she splits the DP into high, middle, and low fields, each of them associated with particular families of syntactic and semantic roles. The terms are pretheoretical, and the work done in each syn- tactic neighborhood (the resolution of reference, quantification, and predication)