A Theory of Neutrality Rights in War
Scott Wolford
Department of Government
University of Texas at Austin
21 February 2014
Scott Wolford (UT Austin)
Neutrality and War
Maryland
1 / 13
Question
How (if at all) can the laws of war affect state behavior?
Scott Wolford (UT Austin)
Neutrality and War
Maryland
2 / 13
Motivation
Laws govern behavior between
Co-belligerents (POWs, civilians, violence)
I
I
Threats of reciprocity
(Morrow 2002, 2007; Wallace 2012; Valentino et al. 2006)
Belligerents and third parties (neutrality)
I
I
Threats of intervention
Where this paper comes in
Scott Wolford (UT Austin)
Neutrality and War
Maryland
3 / 13
Motivation
Laws govern behavior between
Co-belligerents (POWs, civilians, violence)
I
I
Threats of reciprocity
(Morrow 2002, 2007; Wallace 2012; Valentino et al. 2006)
Belligerents and third parties (neutrality)
I
I
Threats of intervention
Where this paper comes in
Scott Wolford (UT Austin)
Neutrality and War
Maryland
3 / 13
Motivation
Laws govern behavior between
Co-belligerents (POWs, civilians, violence)
I
I
Threats of reciprocity
(Morrow 2002, 2007; Wallace 2012; Valentino et al. 2006)
Belligerents and third parties (neutrality)
I
I
Threats of intervention
Where this paper comes in
Scott Wolford (UT Austin)
Neutrality and War
Maryland
3 / 13
Theory
Theory of international law
I
I
Coordinates expectations on unacceptable behavior and
response (Morrow 2002)
May also reduce costs of intervention (cf. Voeten 2005)
Theory of war expansion
I
I
Intervention desirable against expansionist states
But belligerent’s type isn’t known ex ante
Scott Wolford (UT Austin)
Neutrality and War
Maryland
4 / 13
Theory
Theory of international law
I
I
Coordinates expectations on unacceptable behavior and
response (Morrow 2002)
May also reduce costs of intervention (cf. Voeten 2005)
Theory of war expansion
I
I
Intervention desirable against expansionist states
But belligerent’s type isn’t known ex ante
Scott Wolford (UT Austin)
Neutrality and War
Maryland
4 / 13
Theory
Theory of international law
I
I
Coordinates expectations on unacceptable behavior and
response (Morrow 2002)
May also reduce costs of intervention (cf. Voeten 2005)
Theory of war expansion
I
I
Intervention desirable against expansionist states
But belligerent’s type isn’t known ex ante
Scott Wolford (UT Austin)
Neutrality and War
Maryland
4 / 13
The Model
Players: belligerent (B), third party (A)
Belligerent honors or violates neutrality
I
I
I
military boost at some cost cB
expansionist c B values violation more than satiable type c B
cB < cB
Third party joins war or not
I
I
I
would like to fight expansionist but not satiable type
uncertain over belligerent’s type
expansionist w/ probability φ
Scott Wolford (UT Austin)
Neutrality and War
Maryland
5 / 13
The Model
Players: belligerent (B), third party (A)
Belligerent honors or violates neutrality
I
I
I
military boost at some cost cB
expansionist c B values violation more than satiable type c B
cB < cB
Third party joins war or not
I
I
I
would like to fight expansionist but not satiable type
uncertain over belligerent’s type
expansionist w/ probability φ
Scott Wolford (UT Austin)
Neutrality and War
Maryland
5 / 13
The Model
Players: belligerent (B), third party (A)
Belligerent honors or violates neutrality
I
I
I
military boost at some cost cB
expansionist c B values violation more than satiable type c B
cB < cB
Third party joins war or not
I
I
I
would like to fight expansionist but not satiable type
uncertain over belligerent’s type
expansionist w/ probability φ
Scott Wolford (UT Austin)
Neutrality and War
Maryland
5 / 13
The Model
for generic cB ,

(ρ + b)



(ρ + b − a)
uB =
(ρλ + b) − cB



(ρλ + b − a) − cB
if (h, n)
if (h, j)
if (v , n)
if (v , n)

(ρ + b) uA (cB )



(ρ + b − a) u (c ) − c
A B
A
uA =

(ρλ + b) uA (cB )



(ρλ + b − a) uA (cB ) − dcA
Scott Wolford (UT Austin)
Neutrality and War
if (h, n)
if (h, j)
if (v , n)
if (v , n)
Maryland
6 / 13
The Model
for generic cB ,

(ρ + b)



(ρ + b − a)
uB =
(ρλ + b) − cB



(ρλ + b − a) − cB
if (h, n)
if (h, j)
if (v , n)
if (v , n)

(ρ + b) uA (cB )



(ρ + b − a) u (c ) − c
A B
A
uA =

(ρλ + b) uA (cB )



(ρλ + b − a) uA (cB ) − dcA
Scott Wolford (UT Austin)
Neutrality and War
if (h, n)
if (h, j)
if (v , n)
if (v , n)
Maryland
6 / 13
The Model
ui (h, n|c B )
neutral
neutral
A
join
ui (h, j|c B )
honor
B
violate
ui (v , n|c B )
A
join
cB φ
ui (v , j|c B )
N
ui (h, n|c B )
cB 1 − φ
neutral
A
ui (h, j|c B )
Scott Wolford (UT Austin)
honor
B
violate
join
neutral
A
join
Neutrality and War
ui (h, n|c B )
ui (h, j|c B )
Maryland
7 / 13
The Model
ui (h, n|c B )
neutral
neutral
A
join
ui (h, j|c B )
honor
B
violate
ui (v , n|c B )
A
join
cB φ
ui (v , j|c B )
N
ui (h, n|c B )
cB 1 − φ
neutral
A
ui (h, j|c B )
Scott Wolford (UT Austin)
honor
B
violate
join
neutral
A
join
Neutrality and War
ui (h, n|c B )
ui (h, j|c B )
Maryland
7 / 13
The Model
ui (h, n|c B )
neutral
neutral
A
join
ui (h, j|c B )
honor
B
violate
ui (v , n|c B )
A
join
cB φ
ui (v , j|c B )
N
ui (h, n|c B )
cB 1 − φ
neutral
A
ui (h, j|c B )
Scott Wolford (UT Austin)
honor
B
violate
join
neutral
A
join
Neutrality and War
ui (h, n|c B )
ui (h, j|c B )
Maryland
7 / 13
The Model
ui (h, n|c B )
neutral
neutral
A
join
ui (h, j|c B )
honor
B
violate
ui (v , n|c B )
A
join
cB φ
ui (v , j|c B )
N
ui (h, n|c B )
cB 1 − φ
neutral
A
ui (h, j|c B )
Scott Wolford (UT Austin)
honor
B
violate
join
neutral
A
join
Neutrality and War
ui (h, n|c B )
ui (h, j|c B )
Maryland
7 / 13
The Model
ui (h, n|c B )
neutral
neutral
A
join
ui (h, j|c B )
honor
B
violate
ui (v , n|c B )
A
join
cB φ
ui (v , j|c B )
N
ui (h, n|c B )
cB 1 − φ
neutral
A
ui (h, j|c B )
Scott Wolford (UT Austin)
honor
B
violate
join
neutral
A
join
Neutrality and War
ui (h, n|c B )
ui (h, j|c B )
Maryland
7 / 13
Analysis
Three types of equilibrium
No law / opportunistic violation → both types violate*
Full compliance / pooling → no type violates
Separating → only expansionist violates
Two types of effect
Equilibrium selection
Equilibrium replacement
. . . but both depend on opportunistic defection existing as well
Scott Wolford (UT Austin)
Neutrality and War
Maryland
8 / 13
Analysis
Three types of equilibrium
No law / opportunistic violation → both types violate*
Full compliance / pooling → no type violates
Separating → only expansionist violates
Two types of effect
Equilibrium selection
Equilibrium replacement
. . . but both depend on opportunistic defection existing as well
Scott Wolford (UT Austin)
Neutrality and War
Maryland
8 / 13
Analysis
Three types of equilibrium
No law / opportunistic violation → both types violate*
Full compliance / pooling → no type violates
Separating → only expansionist violates
Two types of effect
Equilibrium selection
Equilibrium replacement
. . . but both depend on opportunistic defection existing as well
Scott Wolford (UT Austin)
Neutrality and War
Maryland
8 / 13
Analysis
Three types of equilibrium
No law / opportunistic violation → both types violate*
Full compliance / pooling → no type violates
Separating → only expansionist violates
Two types of effect
Equilibrium selection
Equilibrium replacement
. . . but both depend on opportunistic defection existing as well
Scott Wolford (UT Austin)
Neutrality and War
Maryland
8 / 13
Analysis
Three types of equilibrium
No law / opportunistic violation → both types violate*
Full compliance / pooling → no type violates
Separating → only expansionist violates
Two types of effect
Equilibrium selection
Equilibrium replacement
. . . but both depend on opportunistic defection existing as well
Scott Wolford (UT Austin)
Neutrality and War
Maryland
8 / 13
Equilibrium Selection
probability B is expansionist (φ)
1
separating
partial compliance
separating
full compliance
opportunistic violation
opportunistic violation
φ
ρ(λ − 1) − a
0
c B = ρ(λ − 1)
expansionist type’s cost for violating neutrality (c B )
Scott Wolford (UT Austin)
Neutrality and War
Maryland
9 / 13
Equilibrium Selection
probability B is expansionist (φ)
1
separating
partial compliance
separating
full compliance
opportunistic violation
opportunistic violation
φ
ρ(λ − 1) − a
0
c B = ρ(λ − 1)
expansionist type’s cost for violating neutrality (c B )
Scott Wolford (UT Austin)
Neutrality and War
Maryland
9 / 13
Equilibrium Replacement
probability B is expansionist (φ)
1
separating
partial compliance
separating
full compliance
separating
full compliance
opportunistic violation
opportunistic violation
φ
φ
ρ(λ − 1) − a
0
c B = ρ(λ − 1)
expansionist type’s cost for violating neutrality (c B )
Scott Wolford (UT Austin)
Neutrality and War
Maryland
10 / 13
Equilibrium Replacement
probability B is expansionist (φ)
1
separating
partial compliance
separating
full compliance
separating
full compliance
opportunistic violation
opportunistic violation
φ
φ
ρ(λ − 1) − a
0
c B = ρ(λ − 1)
expansionist type’s cost for violating neutrality (c B )
Scott Wolford (UT Austin)
Neutrality and War
Maryland
10 / 13
Implications
1. Effect greatest when expansionism believed rare.
2. Violations facilitate desirable interventions b/c only
expansionists violate.
- “punishment” spurious to desirable intervention
- compliance a poor standard?
3. More powerful 3rd parties more prone to regretted neutrality.
4. Violations independently increase chances of war expansion.
Scott Wolford (UT Austin)
Neutrality and War
Maryland
11 / 13
Implications
1. Effect greatest when expansionism believed rare.
2. Violations facilitate desirable interventions b/c only
expansionists violate.
- “punishment” spurious to desirable intervention
- compliance a poor standard?
3. More powerful 3rd parties more prone to regretted neutrality.
4. Violations independently increase chances of war expansion.
Scott Wolford (UT Austin)
Neutrality and War
Maryland
11 / 13
Implications
1. Effect greatest when expansionism believed rare.
2. Violations facilitate desirable interventions b/c only
expansionists violate.
- “punishment” spurious to desirable intervention
- compliance a poor standard?
3. More powerful 3rd parties more prone to regretted neutrality.
4. Violations independently increase chances of war expansion.
Scott Wolford (UT Austin)
Neutrality and War
Maryland
11 / 13
Implications
1. Effect greatest when expansionism believed rare.
2. Violations facilitate desirable interventions b/c only
expansionists violate.
- “punishment” spurious to desirable intervention
- compliance a poor standard?
3. More powerful 3rd parties more prone to regretted neutrality.
4. Violations independently increase chances of war expansion.
Scott Wolford (UT Austin)
Neutrality and War
Maryland
11 / 13
Implications
1. Effect greatest when expansionism believed rare.
2. Violations facilitate desirable interventions b/c only
expansionists violate.
- “punishment” spurious to desirable intervention
- compliance a poor standard?
3. More powerful 3rd parties more prone to regretted neutrality.
4. Violations independently increase chances of war expansion.
Scott Wolford (UT Austin)
Neutrality and War
Maryland
11 / 13
Conclusion
Law of neutrality are unique
I
govern behavior b/w belligerents and non-belligerents
Integrated theories of law and war expansion
Deterrence vs. solving information problems
Implications for
I
I
I
Judging “effectiveness”
Spuriousness of “punishment”
Effects on compliance and war expansion
Scott Wolford (UT Austin)
Neutrality and War
Maryland
12 / 13
Conclusion
Law of neutrality are unique
I
govern behavior b/w belligerents and non-belligerents
Integrated theories of law and war expansion
Deterrence vs. solving information problems
Implications for
I
I
I
Judging “effectiveness”
Spuriousness of “punishment”
Effects on compliance and war expansion
Scott Wolford (UT Austin)
Neutrality and War
Maryland
12 / 13
Conclusion
Law of neutrality are unique
I
govern behavior b/w belligerents and non-belligerents
Integrated theories of law and war expansion
Deterrence vs. solving information problems
Implications for
I
I
I
Judging “effectiveness”
Spuriousness of “punishment”
Effects on compliance and war expansion
Scott Wolford (UT Austin)
Neutrality and War
Maryland
12 / 13
Conclusion
Law of neutrality are unique
I
govern behavior b/w belligerents and non-belligerents
Integrated theories of law and war expansion
Deterrence vs. solving information problems
Implications for
I
I
I
Judging “effectiveness”
Spuriousness of “punishment”
Effects on compliance and war expansion
Scott Wolford (UT Austin)
Neutrality and War
Maryland
12 / 13
Conclusion
Questions? Comments?
Scott Wolford
[email protected]
http://www.scott-wolford.com/
Scott Wolford (UT Austin)
Neutrality and War
Maryland
13 / 13