The current debate over models of self-selection in Congress - whether
Congressmen by-and-large find themselves on committees which most clo
sely correspond to their constituents' interests - has implications fo
r theories of Congressional organization. Building on recent findings
which question a categorical self-selection process, in this paper we
present a theory of committee function based on loyalty to party leade
rs. As a rationale for leadership privilege, and to provide context fo
r our argument, we first present a theoretical framework based on a mo
dified model of cooperation. We then focus on certain specifics of our
leadership theory; that rank-and-file members vote leadership interes
ts in exchange for leader support in gaining choice committee assignme
nts and aid in passing legislation. This leads to predictions about vo
ting patterns across committees. Static tests of these relations are p
erformed, as well as those incorporating changes in voting patterns wi
th seniority.