Date of Award

7-11-2023

Author's School

Graduate School of Arts and Sciences

Author's Department

Political Science

Degree Name

Doctor of Philosophy (PhD)

Degree Type

Dissertation

Abstract

How do presidents persuade the public? How do lawmakers respond? And what do voters infer from these competing messages? This dissertation addresses these questions across three papers. The first paper, "How Presidents Persuade: Facts, Feelings, and the Language of Presidential Power," argues that presidents change minds with factual appeals and mobilize supporters with emotional speech. The rhetorical approach presidents pursue depends on their coalition strength: institutionally strong presidents mobilize allies with affective speech while weak presidents try to expand their coalition with informational appeals. To support this theory, I collect a corpus of over 18,000 presidential speeches and measure their factual and affective content using word-embedding methods. Consistent with my theory, I find that presidential speeches are relatively more affective when presidents are strong, including during conditions of unified government, high approval ratings, or conditional party government. I also find that affect in presidential speech changes in line with the theory following a key episode of party power change in the U.S. Senate. Of course, presidents do not go public in a vacuum. In the second paper "Presidential Cues and the Nationalization of Congressional Rhetoric, 1973-2016" I explore how lawmakers think about presidential leadership. I argue that members of Congress, especially out-partisans, invoke the president in speeches to nationalize legislative debate and polarize constituent opinion. Using a corpus of nearly two million floor speeches given by members of the House and Senate between 1973-2016, I show that lawmakers more frequently invoke the president when in the out-party. I complement my observational analyses with a survey experiment showing that when a republican senator references President Biden, republican respondents increase their approval of that Senator and are more likely to oppose compromise. Finally, in "Energy versus Safety: Unilateral Action, Voter Welfare, and Executive Accountability," I develop a formal model of interbranch policymaking, unilateral action, and accountability. In the game, the president chooses between working with a representative lawmaker to pass policy or acting unilaterally. The choice and policy outcome signal information to a voter, who makes inferences about the politicians' policy preferences before choosing to reelect or replace them. Ultimately, I find that the president's ability to use unilateral action increases accountability and voter welfare even in some cases where the legislator is more likely to share the voter's policy preferences. Unilateral action allows an ideologically congruent executive to overcome gridlock and reveal information about both politicians' preferences. This dissertation contributes to our understanding of presidential agenda-setting, going public, and how interbranch conflict shapes what politicians say and how they say it. My research highlights the institutional consequences of nationalization, polarization, and negative partisanship.

Language

English (en)

Chair and Committee

Andrew Reeves

Download

Available for download on Saturday, August 17, 2030

Share

COinS