Refining stratified reference

This is a reply to the comments by Corver, Doetjes, Link, Piñón, Schwarzschild, and Syrett on the target article in this volume, 'Stratified reference: The common core of distributivity, aspect, and measurement'. Stratified reference is designed to capture semantic oppositions involving atelicity, plurality and mass reference, extensive measure functions, distributivity, and collectivity. Following suggestions by some of the commentators, stratified reference is refined here in two ways: it is restricted to the parts of the event or individual in question, and its granularity parameter is instantiated with a predicate built around mereological proper parthood and degree ordering