Talking to the enemy: Explaining the emergence of peace talks in interstate war
| Published date | 01 July 2023 |
| DOI | http://doi.org/10.1177/09516298231185112 |
| Author | Oriana Skylar Mastro,David A Siegel |
| Date | 01 July 2023 |
Talking to the enemy:
Explaining the emergence of
peace talks in interstate war
Oriana Skylar Mastro
Freeman Spogli Institute for International Studies, Stanford University,
CA, USA
David A Siegel
Department of Political Science, Duke University, Durham, NC, USA
Abstract
Why are some states open to talking while fighting while others are not? We argue that a state
considering opening negotiations is concerned not only with the adverse inference that the oppos-
ing state will draw, but also the actions that the opposing state might take in responseto that infer-
ence. We use a formal model, with assumptions grounded in extensive historical evidence, to
highlight one particular response to opening negotiations—the escalation of war efforts—and
one particular characteristic of the state opening negotiations—its resilience to escalation. We
find that states are willing to open negotiations under two conditions: when their opponents
find escalation too costly, and when there is a signal of high resilience that only the highly resilient
care to use. To illustrate the dynamics of the second condition, we offer an extended case study
detailing North Vietnam’s changing approach to negotiations during the Vietnam War.
Keywords
Bargaining, diplomacy, formal theory, interstate conflict, negotiations
Diplomacy plays a critical role in the management and resolution of armed conflict in the
international system. After a war breaks out, decision-makers see the opening of talks as a
constructive step in the conflict resolution: dialog allows for deals to be brokered and
implemented among the relevant parties. Despite wartime diplomacy’s positive effects,
Corresponding author:
David A Siegel, Department of Political Science, Duke University, 140 Science Drive, 294H Gross Hall,
Durham, NC 27708-0187, USA. david.siegel@duke.edu
Article
Journal of Theoretical Politics
2023, Vol. 35(3) 182–203
© The Author(s) 2023
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/ 09516298231185112
journals.sagepub.com/home/jtp
however, history is replete with examples of leaders refusing at times to talk to the enemy.
In over half the wars since WWII, at least one belligerent refused to talk, privately or pub-
licly, throughout the conflict.
1
Treating war as a bargaining process in which the belliger-
ents are simultaneously talking and fighting throughout the whole conflict thus fails to
represent war accurately. A review of the empirical record also reveals interesting vari-
ation in states’willingness to talk: positions on wartime diplomacy vary across countries
and time, even within the same conflict.
Why do states often refuse to negotiate? One reason states may not want to talk is the
adverse inference problem that arises when states signal an openness to negotiations:
they may subsequently be viewed as weaker (Fearon, 1995, 2013; Ramsay, 2011).
Switching from diplomatic silence to a willingness to talk might indicate a reduction
in a state’s war aims or assessment of victory, or that the state is less willing to
absorb the costs of prolonging the conflict than previously thought. We argue, via a
formal model and an extended case study, that such an inference does more than just
provide information to the opposing state that may serve as a coordination device
(Ramsay, 2011). States also fear that it can encourage an opponent to escalate its
war efforts to secure victory. We define an effective escalation as one that increases
the likelihood of victory for the escalating state. As such, escalation is similar to the
endogenous choice of militarization or arming (e.g., Coe and Vaynman, 2020). Thus,
instead of paving the way for peace, states rationally fear a willingness to talk would
encourage longer, bloodier wars.
Not all states suffer from that fear equally, however. Escalation can occur in a number
of ways, such as by increasing the rate at which one harms the other side, involving other
countries, expanding where the fighting occurs, or dedicating more military power to the
war. If escalation were too costly for one’s opponent to employ, a state would have less
need to fear the consequences of any adverse inference arising from taking an open dip-
lomatic stance. As a consequence, it may be more willing to take such a stance. Further, a
state able to reduce the effects of those forms of escalation—or resist them entirely—
should also have less need to fear opening negotiations.
We denote a state’s ability to resist the effects of its opponent’s escalation its resili-
ence. Substantively, resilience may take many forms, such as quickly being able to recon-
stitute power after an attack or being able to continue fighting even after forces have been
destroyed. Resilience overlaps established concepts such as costs of war, power, and
resolve. We expect, all else equal, states with greater resilience to possess lower costs
of war and greater power and resolve, and our formal model assumes that a state’s
costs of war and resilience are correlated. But resilience is different from those concepts.
Two states might suffer identical costs of war, each losing 20 aircraft for instance, but the
more resilient state will suffer less loss of efficacy. That resilience might be due to the aid
of allies, an industrial base that can aid in rebuilding, or a military that has prepared to
operate effectively despite reduced airpower. Resilience is also different from power:
conventionally weaker actors, such as insurgents, can be highly resilient, whereas power-
ful states often win wars without resilience because their adversaries cannot destroy
enough of their power to make resilience a determining factor. Finally, resilience
differs from resolve: a state may display resolve to maintain the course (Kertzer, 2016:
3), yet it may lack the functional ability to do so.
Mastro and Siegel 183
To continue reading
Request your trial