Alliances That Never Balance: The Territorial Settlement Treaty

Abstract

An extensive body of evidence has been accumulated showing that alliances are more often followed by war than by peace (Singer and Small, 1966b; Levy, 1981). It has also been convincingly demonstrated that alliances are associated with the expansion of war (Siverson and King, 1979; 1980; Siverson and Starr, 1990), and alliance polarization and the buildup of alliances in the system are associated with world wars (Kegley, 1994; Wayman, 1984; Midlarsky, 1983; 1986). Nevertheless, a significant number of alliances are not followed by war. Levy (1981) reports that 44 percent of neutrality and defense pacts in the 17th century, 33 percent in the 18th century, 72 percent in the 19th century, and 13 percent in the 20th century are not followed by a war involving an ally within five years. The variation across centuries, especially the large percentage of peaceful 19th century alliances, suggests that not all alliances are created equal. This article identifies a large portion of the peaceful alliances as territorial settlement treaties. By resolving territorial disputes between alliance members, these treaties remove one of the most contentious issues from the agenda of states and pacify some of the most traditionally belligerent states in the system. It is argued that the identification of these alliances is an important step toward re-conceptualizing the alliance variable since the territorial settlement treaties behave very differently from what realist theories would predict. Instead of trying to balance power in a region or in the system, these alliances exhibit the robust and multifaceted nature of diplomacy, demonstrating that a common practice having the same form (forming an alliance) can have very different motivations behind it. Therefore, it is strongly recommended that researchers remove these alliances from their data analyses (or, at the very least, control for them) if they are investigating the paths or correlates to war. The presence of these irenic alliances would tend to mute most findings.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 49 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.