Building good theories is difficult under the best of circumstances. It is particularly troublesome when we agree neither on key terms nor on a database of stylized facts to be accounted for by proposed theories. Group solidarity as problem for scientific analysis is a classic example. Fararo and Doreian's introduction describes the various usages and meanings of the term in sociological theory, but other authors, Breiger and Roberts, Johnsen, and Markovsky, for instance, also testify to the multidimensional nature of the solidarity concept. Equally important is the absence of some agreed upon database. Much progress has been made, for instance, in the study of power in networks because there exists a set of stylized facts describing which positions in which networks can extract large or only modest advantages from exchanges with adjacent positions (Skvoretz and Willer, 1993). Were the equivalent available for group solidarity, my task of commenting on the contributions in this volume would be simplified. As it is, my task is similar to that of an anthropologist studying another culture - major dimensions of conceptual variation must be mapped out and, if possible, some order brought to the collection.